455 71 2MB
English Pages 491 Year 2009
The Second Automobile Revolution
Also by Michel Freyssenet ONE BEST WAY? TRAJECTORIES AND INDUSTRIAL MODELS OF THE WORLD’S AUTOMOBILE PRODUCERS (co-editor with Andrew Mair, Koichi Shimizu, Giuseppe Volpato) THE PRODUCTIVE MODELS: The Conditions of Profitability (with Robert Boyer) GLOBALIZATION OR REGIONALIZATION OF THE AMERICAN AND ASIAN CAR INDUSTRY? (co-editor with Koichi Shimizu, Giuseppe Volpato) GLOBALIZATION OR REGIONALIZATION OF THE EUROPEAN CAR INDUSTRY? (co-editor with Koichi Shimizu, Giuseppe Volpato)
Other GERPISA titles ONE BEST WAY? Trajectories and Industrial Models of the World’s Automobile Producers (Edited by Michel Freyssenet, Andrew Mair, Koichi Shimizu, Giuseppe Volpato) BETWEEN IMITATION AND INNOVATION: The Transfer and Hybridization of Productive Models in the International Automobile Industry (Edited by Robert Boyer, Elsie Charron, Ulrich Jürgens, Steven Tolliday) TEAMWORK IN THE AUTOMOBILE INDUSTRY: Radical Change or Passing Fashion? (Edited by Jean-Pierre Durand, Paul Stewart, Juan-José Castillo) COPING WITH VARIETY: Product Variety and Production Organization in the World Automobile Industry (Edited by Yannick Lung, Jean-Jacques Chanaron, Takahiro Fujimoto, Daniel Raff ) GLOBAL STRATEGIES AND LOCAL REALITIES: The Auto Industry in Emerging Markets (Edited by John Humphrey, Yveline Lecler, Mario Sergio Salerno) THE PRODUCTIVE MODELS: The Conditions of Profitability (Robert Boyer and Michel Freyssenet) GLOBALIZATION OR REGIONALIZATION OF THE AMERICAN AND ASIAN CAR INDUSTRY? (Edited by Michel Freyssenet, Koichi Shimizu, Giuseppe Volpato) GLOBALIZATION OR REGIONALIZATION OF THE EUROPEAN CAR INDUSTRY? (Edited by Michel Freyssenet, Koichi Shimizu, Giuseppe Volpato) CARS…CARRIERS OF REGIONALISM? (Edited by Jorge Carillo, Yannick Lung, Rob Van Tulder) WORK AND EMPLOYMENT RELATIONS IN THE AUTOMOBILE INDUSTRY (Edited by Elsie Charron, Paul Stewart) INDUSTRIES AND GLOBALIZATION: The Political Causality of Differences (Bernard Jullien and Andy Smith)
The Second Automobile Revolution Trajectories of the World Carmakers in the 21st Century Edited by
Michel Freyssenet
Department of Sociology, National Centre of Scientific Research (CNRS), Paris
In association with GERPISA Réseau International International Network Groupe d’Étude et de Recherche Permanent sur l’Industrie et les Salariés de l’Automobile, Permanent Group for the Study of the Automobile Industry and its Employees, École des Hautes Études en Sciences Sociales, Paris, Université d’Évry-Val d’Essonne.
Selection and editorial content © Michel Freyssenet 2009 Individual chapters © contributors 2009 Foreword © Bernard Jullien 2009 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2009 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN-13: 978–0–230–21971–7 hardback ISBN-10: 0–230–21971–3 hardback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. A catalog record for this book is available from the Library of Congress. 10 9 8 7 6 5 4 3 2 1 18 17 16 15 14 13 12 11 10 09 Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne
Contents viii x xii xvii xix
List of Tables List of Figures Notes on the Contributors Foreword by Bernard Jullien List of Abbreviations 1
Introduction: Ten Years On, What Have We Learnt? Michel Freyssenet
2
Wrong Forecasts and Unexpected Changes: the World that Changed the Machine Michel Freyssenet
3
Strategies of Internationalisation of Automobile Firms in the New Century: a New Leap Forward? Bruno Jetin
Part I: The Divergent Trajectories of Japanese and Korean Carmakers 4
The Uncertainty of Toyota as the New World Number One Carmaker Koichi Shimizu
1
7
38
67 69
5
Nissan: From the Brink of Bankruptcy Merieke Stevens and Takahiro Fujimoto
6
Honda: Serendipity or Strategy from 1997–2007? Denise J. Luethge and Philippe Byosière
112
7
The Rebirth of Mazda Under Ford’s Shadow Daniel Arturo Heller
129
8
Hyundai: Is it Possible to Realise the Dream of Becoming a Top Five Global Automaker by 2010? Myeong-Kee Chung
141
Part II: The Resistible Decline of the ‘Big Three’?
163
9
165
General Motors in an Age of Corporate Restructuring Richard Senter, Jr. and Walter McManus v
95
vi
Contents
10 Ford, 1993–2007: Losing its Way? Glenn Mercer
185
11 Can Chrysler Survive its Reinvention? Bruce Belzowski
206
Part III: The Resistance of Leading European Carmakers
223
12 The Final Chapter of the ‘VW Model’? The VW Trajectory, 1995–2005 Ulrich Jürgens
225
13 PSA: the Difficulties of a ‘Volume and Diversity’ Profit Strategy Michel Freyssenet
246
14 Renault, 1992–2007: Globalisation and Strategic Uncertainties Michel Freyssenet
267
15 Fiat Group Automobiles: an Arabian Phoenix in the International Auto Industry Giuseppe Volpato
287
16 From the Marriage in Heaven to the Divorce on Earth: the DaimlerChrysler Trajectory since the Merger Holm-Detlev Köhler 17 Driving with Engineers’ Professionalism and Family Values: the BMW Trajectory from a Regional Carmaker to a Global Premium Player Ludger Pries 18 A Break from the Past: Volvo and its Malcontents Matthias Holweg and Frits K. Pil
309
332 353
19 Production Counterfeits and Policy Collisions: the Rover Trajectory – a Salutary Tale Dan Coffey
366
Part IV: Newcomers and Global Suppliers
381
20 Made in China: Joint Ventures and Domestic Newcomers Hua Wang
383
Contents vii
21 Maruti-Suzuki’s Trajectory: From a National Champion to a Japanese-owned Subsidiary Florian Becker-Ritterspach 22 Winners and Losers in the Auto Parts Industry: Trajectories Followed by the Main First Tier Suppliers Over the Past Decade Vincent Frigant
404
419
23 Conclusion: the Second Automobile Revolution – Promises and Uncertainties Michel Freyssenet
443
Appendix: the GERPISA International Network Index
455 459
List of Tables 2.1 3.1 3.2 3.3 3.4 4.1 8.1 8.2 8.3 8.4 10.1 18.1 18.2 20.1 20.2 20.3 20.4 20.5 22.1 22.2 22.3 22.4
Compatibility between firms’ profit strategies and certain current trends in the automobile industry Comparison of the degree of internationalisation of automobile firms, 2000–2005, in % of total, and synthetical index Geographical breakdown of General Motors and Ford Automobiles net sales and financial revenues, in percentage Geographical structure of European firms’ net sales, in percentage Geographical breakdown of Asian firms’ sales and revenues, in percentage New grade system under the assistant-manager (rank 30) at Toyota from 2000 Hyundai, modularity by model and plant Hyundai’s J.D. Power and Associates IQS ranking Hyundai, modular advantage: reducing the number of jobs Hyundai, work conditions by plant Ford cumulative net income by operation, 1990–2006 Volvo’s product portfolio and platforms, 2007–2012 (projected) Volvo production by car model, 1994–2006 Contribution of Chinese firms and FDI to the Chinese automotive industry by type of products, 2005 Ranking of automobile sales in China by automobile firm and type of vehicle, 2006 Volume of production of major Chinese OEMs producing passenger cars, 2005 A summary of Geely’s international cooperation Geely’s productive model: ‘The low-cost, low-price car’ Top thirty worldwide original auto-parts suppliers, 1998 and 2006 Carmakers’ main supplier subsidiaries Facilities, production and research sites of FTSs as of end 2006 (units) Distribution of staff and plants of FTSs between domestic and nearby peripheral zones (year end 2006; %)
32 42 44 46 47 80 145 147 149 150 198 355 361 387 388 389 395 399 425 427 434 435
Appendix tables 4.1
Toyota, 1970–2006: worldwide production, production in Japan, export, foreign production, total workforce in Japan, gross operating profit, pre-tax profit. Consolidated accounts: 1999–2006 viii
89
List of Tables
4.2 5.1 6.1 6.2 6.3 7.1 8.1 9.1 10.1 11.1 11.2 12.1 13.1 14.1
15.1 16.1 16.2 16.3 17.1 19.1 21.1 22.1 22.2 22.3 22.4
Assembly plants of Toyota, outside of Japan Nissan, 1990–2006: production by region, exports by region, sales in Japan, workforce, financial results Vehicles sales, revenue, net income, employees of Honda, 1997–2006 Honda Automobile unit sales (in thousands), 1997–2006 Honda Automobile unit sales as a percentage of total sales, 1997–2006 Mazda, 1995–2006: domestic and overseas production, exports, employees, sales, debts, cash-flow, profits Hyundai, 1997–2006: domestic and overseas production, sales, exports, employment, gross sales, operating profit General Motors, 1995–2006: worldwide and NAFTA production, revenue, net income, ROA, labour Ford, 1990–2006: production (North America and rest of world), workers, sales, net income, assets Chrysler, 1995–2007: worldwide and NAFTA production, US market share, revenue, net income, operating profit, worldwide pre-tax profit per vehicle, labour Initial Quality Index of Chrysler marks, 1996–2007 Volkswagen, 1991–2006: domestic and abroad production, export, domestic and abroad revenue, domestic and abroad workforce, net earnings PSA, 1976–2007: worldwide production, group workforce, workforce in France, group revenues, group net income Renault, 1980–2007: worldwide and domestic production, exportation, workforce, turnover, net income and investments (group, automobile branch, France automobile branch) Fiat, 1990–2006: domestic and abroad production, workforce, turnover, net income, investments and R&D (group and Fiat Auto) Operating profit of DaimlerChrysler, 1998–2006 Employees of DaimlerChrysler, 1998–2006 Key figures of the new Daimler AG Group, 2005–2007 BMW, 1990–2006: sales by makes and region, investments, operating profit, workforce Rover, 1968–2003: total car production, sales outside UK, total sales, operating profit, total employees Maruti-Suzuki, 1984–2007: turnover, profit, production, market share, export, employees Employees of FTSs, 1998–2006 Sales of FTSs (group level), 1998–2006 (100 = 1998) Net incomes/sales (%) of FTSs, 1997–2006 EBIT/sales (%) of FTSs, 1997–2006
ix
92 110 126 126 126 138 160 182 202 220 221 243 264
284 305 328 329 330 350 376 416 437 438 439 439
List of Figures 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 3.14 3.15 4.1 4.2 4.3 8.1 8.2 8.3 10.1 10.2 12.1 13.1 13.2 13.3 14.1 14.2
Production share outside the country of origin, 1994–1995, 1995–1999, 2000–2005, percentages Production share outside the region of origin, 1996–1999, 2000–2005 and 2006 Geographical breakdown of GM’s net profit, 1953–2007 Geographical breakdown of GM’s net profit, 1977–2007 Geographical structure of Ford’s net profit, 1957–2002, and income before income tax, 2003–2007 Geographical breakdown of Ford’s net profit, 1975–2002, and income before income tax, 2003–2007 Operating income of the Japanese ‘big three’ in Japan Operating income of the Japanese ‘big three’ abroad Operating income of Japanese ‘five small’ firms in Japan Operating income of Japanese ‘five small’ firms abroad Geographical breakdown of Toyota’s operating profit Geographical breakdown of Honda’s operating profit Geographical breakdown of Nissan’s operating profit Geographical breakdown of Suzuki’s operating profit Geographical breakdown of Hyundai’s operating profit Production and exports of Toyota, 1960–2007 Movement of the average production allowance coefficient, Toyota, 1972–1989 Production, productivity and labour force of Toyota, 1985–1992 Working hours at Hyundai, 2000–2004 Strikes and wage increase at Hyundai, 1998–2006 Hyundai sales by region, 2003–2005 Ford share price history, 1993 to March 2008 Ford Motor Company units sold and financial results, 1990–2006 The operating margin of Volkswagen, 1991–2006 Operating margin/revenue rate of PSA (Group and Auto), 1976–2007 Worldwide production of Peugeot cars by model, 1945–2007 Worldwide production of Citroën cars, 1945–2007 Global production of Renault brand passenger cars, by model, 1945–2007 Financial results of Renault Group, 1976–2007, gross and net income/turnover ×100
48 49 52 53 54 55 56 56 57 58 59 60 60 61 61 70 73 76 152 153 157 196 199 239 248 250 251 268 271
List of Figures xi
15.1 15.2 15.3 16.1 16.2 16.3 17.1 18.1 18.2 18.3 20.1 22.1 22.2 22.3 23.1 23.2
Fiat makes’ market share in Italy, 2000–2007 Fiat makes’ market share in Western Europe, 2000–2007 Fiat Auto financial targets DaimlerChrysler Group profit development, 1998–2006 DaimlerChrysler Group unit sales, 1999–2006 Module strategy of Mercedes: from car line specific organisation to cross-modular organisation BMW Group world production, employment, sales, net profit, 1990–2006 Productivity of Volvo plants against European competitors and Ford plants in 2006 Volvo’s quarterly production by country, 2000–2006 European stock levels in days of demand Product line-up of Geely, 1998–2006 Net income/sales of FTSs (unweighted average for the sample, 1997–2006) Average income/sales and number of years of losses of FTSs (over 1997–2006) Net income/sales and R&D expenditures/sales of FTSs (%; average over 2002–2006) Worldwide automobile production, 1898–2007 Worldwide automobile production by continent and production of certain countries, 1898–2007
300 300 302 315 317 318 346 359 360 363 391 429 430 432 443 446
Notes on the Contributors Florian Becker-Ritterspach is Assistant Professor at the University of Groningen, Faculty of Management & Organisation (The Netherlands). His main topics are developments in the global automobile industry with a specific focus on multinationals operating in India; knowledge transfers in multinationals; the Indian business system and institutional context; and combining organisation theory and international business perspectives. A recent article, ‘The Social Constitution of Knowledge Integration in MNEs: a Theoretical Framework’, appeared in the Journal of International Management, 12(3) (2006): 358–77. Bruce M. Belzowski is Associate Director and Assistant Research Scientist at the University of Michigan (Transportation Research Institute, Automotive Analysis Division), Ann Arbor, Michigan (USA). He is member of the international steering committee of GERPISA. His research topics are globalisation of the automotive industry, manufacturer–supplier–dealer relations, information technology, powertrain strategies, and human resource strategies. One of his recent publications is Powertrain Strategies for the 21st Century. Philippe Byosière is Professor at the Doshisha Business School, Doshisha University, Kyoto (Japan) and Visiting Research Professor at the University of Michigan, Ann Arbor, Michigan (USA). His research topics are knowledge and innovation, and global leadership. One of his recent publications is ‘Microbursts of Knowledge and Creative Work in Japan’, in D. H. Whittaker and R. E. Cole (eds), Recovering from Success: Management of Technology in Japan (Oxford University Press, 2006), pp. 184–98. Myeong-Kee Chung is Professor of Economics at the School of Chinese Studies and Economics, Hannam University (Korea). He is currently involved in research on production systems and labour relations in the Korean automobile industry. A recent publication is ‘The Formation of International Production and Distribution Networks between Korea & China: a Case of the Automobile Parts Industry’, Journal of Labor Studies, 12(2) (2006). Dan Coffey is Senior Lecturer at Leeds University Business School, Leeds (United Kingdom). His main publication is The Myth of Japanese Efficiency: the World Car Industry in a Globalizing Age (Edward Elgar, 2006). Michel Freyssenet is Research Director at the Scientific Research National Centre (CNRS) in Paris (France). He is co-founder of GERPISA and currently a member of its International Steering Committee. His main topics are: productive models, national growth models, world productive recomposition, xii
Notes on the Contributors
xiii
history of work division, and social relationships theory. His main publication in English is The Productive Models: the Conditions of Profitability (Palgrave Macmillan, 2002) (with R. Boyer). His personal website is: http://freyssenet.com. Vincent Frigant is Assistant Professor, GREThA, UMR CNRS 5113, University of Bordeaux (France). His main topics are: industrial organisation, vertical inter-firm relationships, modular production, the auto industry, and the aerospace industry. His most recent publication in English is: ‘The Internationalisation of the French Aerospace Industry: To What Extent Were the 1990s a Break with the Past?’ Competition & Change, 11(3) (2007): 260–84 (with C. Carrincazeaux). Takahiro Fujimoto is Professor, Tokyo University (Japan). He is Executive Director of the Manufacturing Management Research Centre. His main topics are: evolution of manufacturing systems and strategic implications of product architecture. His main publication is: The Evolution of a Manufacturing System at Toyota (Oxford University Press, 1999). Daniel Arturo Heller is Associate Professor, International Graduate School of Social Sciences, Yokohama National University (Japan). His main topics are inter-organisational learning and strategic alliances. One of his recent publications is: ‘The Long-Term Value of M&A Activity to Enhance Learning Organizations’, International Journal of Automotive Technology and Management, 6(2) (2006): 157–76 (with G. Mercer and T. Fujimoto). Matthias Holweg is a Senior Lecturer in Operations Management and the Director of the Centre for Competitiveness and Innovation at the Judge Business School, University of Cambridge, UK. He is principal investigator on several research projects, including at MIT’s International Motor Vehicle Program (IMVP), where his research focuses on the dynamics of competition and patterns of evolution of the global automotive industry. He is the co-author, with Frits Pil, of The Second Century: Reconnecting Customer and Value Chain through Build-to-Order (MIT Press, 2004). Bruno Jetin is Assistant Professor of Economics at the University of Paris North (France), where he works at the Centre for Economics of Paris North (CEPN). He is a member of the GERPISA International Steering Committee. His main publication on the automobile industry in English is: ‘Internationalization of American and Asian Automobile Firms’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? (Palgrave Macmillan, 2003). Bernard Jullien is a Professor at the University of Bordeaux (Université Montesquieu-Bordeaux IV). He is interested in technological and regulatory changes that impact on industrial organisation. He has published numerous
xiv
Notes on the Contributors
articles concerning distribution and utilisation systems in the automobile industry. His work with the political scientist Andy Smith is summarised in B. Jullien and A. Smith, Industries and Globalization: the Political Causality of Differences (Palgrave Macmillan, 2008). Ulrich Jürgens is Leader of the WPA and Adjunct Professor at the Free University of Berlin (Germany). His research focuses on innovation processes, organisational and industry change, and social sustainability of work from an international comparative perspective. He took part in the first MIT programme on the ‘Future of the Automobile’ and is member of the International Steering Committee of GERPISA. His main recent publication is: Arbeiten am Auto der Zukunft. Produktinnovationen und Perspektiven der Beschäftigten (Working on the Car of the Future: Product Innovations and Perspectives for the Employees) (Sigma, 2005) (co-authored with Heinz-Rudolf Meißner). Holm-Detlev Köhler is Professor of Sociology, University of Oviedo (Spain). His main research interests are: industrial relations, internationalisation of firms and local and regional development. One of his recent publications is: ‘Consequences of Enlargement for the Old Periphery of Europe: Observations from the Spanish Experience with European Works Councils’ (with Sergio González Begega), in P. Leisink, B. Steijn and U. Veersma (eds), Industrial Relations in the New Europe (Edward Elgar, 2007), pp. 99–114. Denise J. Luethge is Professor of Management at Northern Kentucky University in Covington, Kentucky (USA). Her current research is in the area of knowledge transfer and conversion, network relationships, gender issues in management and cross-cultural leadership. A recent publication is: ‘An Empirical Examination of the Relationship between Knowledge Domains and Knowledge Conversion Processes’, Journal of Knowledge Management, 12(2) (2008): 67–78. Walter S. McManus is Research Scientist and Director of the Automotive Analysis Division (formerly known as OSAT) at the University of Michigan Transportation Research Institute. His research focuses on forecasting for the automotive industry, new automotive technologies and their effects on the environment, and the impact of government regulations on the automotive industry. A recent article is: ‘The Link between Gasoline Prices and Vehicle Sales’, Journal of the National Association for Business Economics, 2007 (January). Glenn Mercer is President of GM Automotive Consulting LLC (USA). His main activities are: private equity advisory in automotive, equity research in automotives, and teaching in automotive topics (e.g. Case Western, IMD). He was formerly a McKinsey & Company Consultant and President of Automotive Analysts. He has written many articles in periodical and refereed journals.
Notes on the Contributors
xv
Frits Pil is Associate Professor at the Katz Graduate School of Business, and Research Scientist at the Learning Research and Development Center, University of Pittsburgh, USA. He is also honorary Senior Research Associate at the Judge Business School (University of Cambridge, UK). He is the co-author, with Matthias Holweg, of The Second Century: Reconnecting Customer and Value Chain through Build-to-Order (MIT Press, 2004). Ludger Pries is Professor Dr, Department of Social Science, Ruhr-Universität Bochum (Germany) and Chair on Sociology of Organisations and Participation Studies (SOAPS). His research fields are: comparative sociology of work, organisations and migration. One of his recent automobile-related publications is: ‘Cost Competition or Innovation Competition? Lessons from the Case of the BMW Plant Location in Leipzig, Germany’, Transfer (Brussels), 12(1) (2006): 11–29. Richard Senter, Jr. is Visiting Scholar at the University of Michigan Transportation Research Institute and an Emeritus Professor at Central Michigan University. His research interests include the sociology of organisations and the sociology of work and occupations. His most recent articles include: ‘Organizing Supply Chains for Automotive Transplants in the United States’ (with Michael S. Flynn), Michigan Academician, 37 (2007): 93–115. Koichi Shimizu is Professor of Economics at Okayama University (Japan) and member of GERPISA’s International Steering Committee. His main topic is presently the study of the French 35-hours law and its application to carmakers. His main publication is: Le Toyotisme (Repère/Éditions La Découverte, 1999). Merieke Stevens is PhD Candidate, Judge Business School, Cambridge University (United Kingdom). Her main topics are supply chain management and changes in dyadic relations from a systems perspective. A recent publication is: ‘Foreign Influences on the Japanese Automobile Industry’, in R. Rasiah, Y. Sadoi and R. Busser (eds), Multinationals, Technology and Localization: Automotive and Electronics Firms in Asia (Routledge, 2007). Giuseppe Volpato is Full Professor of Management and Business Strategy at the Faculty of Economics of Ca’ Foscari University of Venice (Italy). He is Research Fellow of the IMVP (International Motor Vehicle Programme), member of the International Steering Commitee of GERPISA, and Senior Adviser for the ICDP (International Car Distribution Programme). His last book is: Fiat Group Automobiles (Il Mulino, 2008). Hua Wang is Assistant Professor of Managerial Economics and Innovation Management at Euromed Marseille, a management school in France. He is French director of the China Euro Mediterranean Centre for Diversity
xvi
Notes on the Contributors
(CEMCD). His research interests centre on foreign direct investment, transitional economies, innovation in product architecture, and the automotive industry in China. A recent publication is: ‘The Evolution of Strategy on Foreign Direct Investment: Case Study of Volkswagen in China’, in Haixiong Qiu (ed.), Regional Innovation and Firm Development (Economic Science Press, 2007), pp. 459–73.
Foreword At its Fifteenth International Colloquium in June 2006 in Paris, GERPISA launched its fifth international research programme for four years. Named ‘Sustainable Development and Automobile Industry’, this programme focuses on the question of trade-offs and synergies between the different dimensions of sustainable development (SD) in the automobile industry. It retains the predominant definition of SD, which considers sustainable development which guarantees compatibility between economic competitiveness, social responsibility and environment protection. This obligation to integrate the three dimensions of SD and to achieve synergies between them is increasingly significant at both the corporate and public policy level. Reference to this ‘new framework’ has grown in importance over the last decade to the point where it is now in the process of being translated into both discourse and practice. In this context of growing operationalisation of the framework, the fifth GERPISA programme is based on the belief that reference to the principles of SD alone is no longer sufficient and that an analysis of the practices is now required. In analytical terms, it is therefore important to contrast the notion of trade-off with the notion of synergy. This is necessary to understand the discrepancies between practices and discourse, to investigate the diversity of interpretations and to develop a realistic understanding of the changes related to this new framework. Two main research axes were adopted. The first focuses on the practices of the enterprises, and the second on the design of public policy. The Second Automobile Revolution is the first contribution to the practices of the enterprises axis. It seems useful and fruitful to begin with the analysis of the automobile firms’ trajectories since the middle of the last decade of the twentieth century, continuing the previous studies published in One Best Way? (1998), Globalization or Regionalization of the American and Asian Car Industry? and in Globalization or Regionalization of the European Car Industry? (2003). It will enable the reader to understand the SD discourses of the automobile firms in relation to the strategies that they pursue to be profitable in the international context before the crisis in 2008. So, the present book shows that the ‘productive model’ approach developed within GERPISA was in fact a way to analyse the social and economic sustainability of firms’ strategies. But the interest of the present book is more than that. It shows that SD discourses and practices must be replaced by a new global process combining the challenge of the clean and safe car, the market explosion of the BRICs and the activism of newcomers (firms and countries). The struggle to impose a new automobile standard started before and continues through the crisis.
xvii
xviii
Foreword
Alliances are preparing, with unusual actors. The automobile industry is entering a new era. Bernard Jullien Director of GERPISA International Network Professor of Economics, Bordeaux University
List of Abbreviations AP BLMC BMW BRIC BTO CAFÉ CASA CAW CBU CCFA CEO CFTs CGT CKD CNUCED
COFACE COMECON COO CVCC DB DC DCAC DFIs DFM DM DRG EBIT EC EEC EU FASA FAW FDI 4WD FRF FRG
Asia-Pacific British Leyland Motor Corporation Bayerische FlugzeugWerke Brazil, Russia, India, China Build-To-Order Corporate Average Fuel Economy Cessation Anticipée d’Activité des Salariés Âgés Canadian Auto Workers Completely Built Up Comité des Constructeurs Français d’Automobiles (French Chamber of Automobile Builders) Chief Executive Officer Cross Functional Teams Confédération Générale du Travail (France) Completely Knocked Down Conférence des Nations Unies sur le Commerce et le Développement (UNCTAD: United Nations Conference on Trade and Development) Compagnie Française d’Assurances pour le Commerce Extérieur Council for Mutual Economic Assistance (Eastern bloc) Chief Operational Officer Compound Vortex Controlled Combustion (Honda engine) Daimler-Benz DaimlerChrysler Dongfeng Citroën Automobile Company Direct Foreign Investments Dong Feng Motors Deutschmark Democatic Republic of Germany Earnings Before Interest and Taxes European Communities European Employee Committee European Union Fabrication de Automobiles, Sociedad Anonima First Automobile Works Foreign Direct Investment Four-Wheel Drive French Franc Federal Republic of Germany xix
xx
List of Abbreviations
FTSs FWD GATT GDR GERPISA GM GMAC HAM HCCs HRM ICDP IG IMF IMVP IQS IT JAMA JIT JV KD LCCs M&A Mercosur MIT MITI MPVs MVMA NAC NAFTA NICs NPW NRP NUMMI OEM OICA PAC PAG PE P&L PLC PSA QC
First Tier Suppliers Front-Wheel Drive General Agreement on Tariffs and Trade German Democratic Republic Groupe d’Études et de Recherche sur l’Industrie et les Salariés de l’Automobile (Permanent group for the study of the automobile industry and its employees) General Motors General Motors Acceptance Corporation Honda America Motor Company High Cost Countries Human Resource Management International Car Distribution Programme Industriegewerkschaft Metall International Monetary Fund International Motor Vehicle Programme Initial Quality Study Information Technology Japan Automobile Manufacturing Association Just-In-Time Joint Venture Knocked Down Low Cost Countries Mergers and Acquisitions Common Market of South American Countries Massachusetts Institute of Technology Ministry of International Trade and Industry Multi Purpose Vehicles Motor Vehicle Manufacturing Association Nanjing Automobile Corporation North America Free Trade Agreement Newly Industrialised Countries Nissan Production Way Nissan Revival Plan New United Motor Manufacturing Original Equipment Manufacturing Organisation Internationale des Constructeurs d’Automobiles Production Allowance Coefficient Premier Automobile Group (brand of Ford) Production Efficiency Profit and Loss Public Limited Company Peugeot Société Anonyme (holding of Peugeot-Citroën) Quality Control
List of Abbreviations
R&D RNPO ROCE ROS RSM RVs RWD SAIC SKD SME SOE SUVs TAB TEMA TL TMC TNI TPM TPS TTM TVAL UAW UETs UNCTAD UPH UTE UVs VE VEBA VLE VPCs VRS VTEC VW WEC WTO
xxi
Research and Development Renault–Nissan Purchasing Organisation Return On Capital Employment Return On Sales Renault–Samsung Motors Recreational Vehicles Rear-Wheel Drive Shanghai Automotive Industry Corporation Semi-Knocked Down Small and Medium Enterprise State Owned Enterprise Sport Utility Vehicles Toyota Auto Body Toyota Motor Engineering and Manufacturing North America Team Leader Toyota Motor Corporation Transnational Index Total Production Maintenance Toyota Production System Time-To-Market Toyota Verification of Assembly Line United Automobile Workers Unités Élémentaires de Travail United Nations Conference on Trade and Development Unit Per Hour Elementary Technology Units Utility Vehicles Value Engineering Voluntary Employees’ Beneficiary Association Vehicle Line Executive Vehicle Programme Centres Voluntary Retirement Scheme Variable Valve Timing and Lift Technology (Honda engine) Volkswagen World Employee Committee World Trade Organisation
This page intentionally left blank
1
Introduction: Ten Years On, What Have We Learnt? Michel Freyssenet
GERPISA’s first publication – One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers (Freyssenet et al., 1998) – was an analysis of automobile firm trajectories since the 1970s. Its aim was to test the received wisdom at the time that things would necessarily converge towards ‘lean production’, being one conceptualisation of the ‘Japanese productive model’ that some people expected to ‘change the world’ (Womack et al., 1990). The study ultimately showed that three and only three firms (Honda, Toyota and Volkswagen) were consistently profitable during slow growth years. Moreover, they pursued completely different profit strategies and productive models. Thus, instead of corroborating a convergence hypothesis, it offered a renewed differentiation of corporate trajectories and models. This collective research project, like others relating to the transfer of productive models from one country to another (Boyer et al., 1998) and to the organisation of production and of work (Durand et al., 1999; Lung et al., 1999) or offering a complete history of the automobile industry (Boyer and Freyssenet, 1999, 2006), enabled the compilation of a table comprising different automaker profit strategies, the societal conditions in which they operate, and the productive models they have historically implemented thanks to specific ‘company government compromises’ (Boyer and Freyssenet, 2002). This table became GERPISA’s main tool of analysis and was further enriched and transformed by subsequent research. A few years later, GERPISA published a new series derived from its second international research programme (1997–9) and aimed at determining whether automobile groups grow along global or regional lines. This was a topical question in the late 1990s given the ‘merger mania’ of the time. By re-examining the history of automakers along with suppliers and automobile distributors’ internationalisation since the 1950s, we tried to discover whether globalisation has become a new condition of profitability and whether groups are better described as ‘globalised’ or as firms whose products and organisation have been ‘regionalised’. Research indicated that a firm’s presence in the world’s leading markets is not a guarantee of profitability and 1
2 The Second Automobile Revolution
that firms who stayed in their region of origin have, under certain conditions, been very profitable. As for global product policies or organisations, these have all failed (Carillo et al., 2004; Charron and Stewart, 2004; Freyssenet et al., 2003a, 2003b; Humphrey et al., 2000). At the time, and since then, other actions have been presented as indispensable to the survival of all firms, with the idea that firms’ ratings would suffer in the absence of these actions. Two regular suggestions were that the automobile industry should convert to the ‘new economy’ (relying on the financial markets, new rules of governance, shareholder value as a criterion of profitability, development of and diversification towards more lucrative activities, etc.) and that all non-core automobile activities should be outsourced, a trend associated with product modularisation and assembly. Of course, there were other ideas as well, including platform policies, production innovations and production in low-cost countries. The present book reviews the trajectories of these firms from the mid-1990s to 2007, just before the financial crisis. It highlights the conditions in which each firm has approached the completely new period. It is also a chance to test our own analyses, and to ask what we have learnt over the past decade. It is noteworthy that not all authors in this book have come to the same conclusions, despite their joint participation over many years in GERPISA research projects. At the same time, there are a number of common threads. In the present as in the past, there is no universal model whose implementation can ensure a firm’s survival in all circumstances. Firms’ diversity is renewed within a limited number of possibilities. All action occurs under conditions of possibility and validity that will limit its diffusion in time and space. External conditions governing the adoption of a strategy, or internal conditions ensuring the coherence of the means adopted, are key to understanding the history of firms and their performance. Chapter 2, entitled ‘Wrong Forecasts and Unexpected Changes: the World that Changed the Machine’, specifically shows, using the trajectories of firms presented subsequently, the invalidation of all universalising predictions made over the course of the past decade (lean production, globalisation, ‘new economy’, outsourcing, disappearance of specialist automakers, increasingly restrictive oligopolies, commonalisation, innovation, transfer of production towards low-cost countries, suppliers’ parks, global sourcing, etc.). Where these trends have been effective, this was solely due to the actions of some firms, whose success was not always assured because they took these steps. Many major changes were barely if at all anticipated (modified structure of different markets, emergence of new automakers, collapse of certain emerging countries and the lift-off of several country-continents, reaffirmation of national automobile policies, skyrocketing oil prices, proliferation of restrictive environmental norms, life-threatening risks facing some of the leading automakers, etc.). If the crisis is going to modify the ranking of the automobile producers, one of the unanticipated changes will be still more important
Introduction: Ten Years On, What Have We Learnt?
3
for the future of the automobile industry – a veritable ‘second automobile revolution’, on which the book’s conclusion is focused. Chapter 2 ends by asking how forecasts can be improved. We show that the incorporation of productive models’ limited and renewed diversity has become an efficient way of developing even stronger ones. Chapter 3 offers statistical analysis of the firms’ performances over the course of the period in question, notably in light of the different regions where they run commercial and industrial operations. Using profitability and internationalisation (which, although not identical, are sufficiently close) enables a helpful comparison of firms. Automobile groups are more internationalised commercially and industrially than ten years ago but in fact they are merely catching up with the average for all industrial sectors. More than half of all automaker consolidation operations from the late 1990s have fallen apart since then, in particular the biggest deals (with respect to the size of the firms involved), with the exception of Renault–Nissan. The region of origin has until now remained crucial for firms’ profitability, including the more internationalised ones. Penetrating a new region is rarely profitable for the first ten years, given the sizeable industrial, commercial and human investments required. Later performance is far from guaranteed and can vary enormously. Understanding this is one of the aims of later chapters, which retrace at the same time the complete trajectory followed by given firms and, more unusually, a national industry (China) as well as major supplier groups. Questions have been raised about the order in which these trajectories are to be presented, thus about the criterion to be used when categorising them. Possibilities include ranking trajectories by firms’ region of origin, place in a global group, level of production, level of profitability, shareholder type or profit strategy. These criteria, which are quite different in nature, intimate that each of the firms being placed into a particular category shares contextual conditions or important attributes. In earlier books, we adopted the world region as an organising principle. The question is whether to continue in this vein, given the sharp rise in internationalisation witnessed in recent years. An entity’s belonging to a global group would appear most logical except for the fact that most alliances or mergers have been very unstable. Judging on the basis of the dominant shareholder perspective would intimate that the firms of each category behaved uniformly in time and space (i.e. pursuing a short- versus a long-term orientation), but this is not what happened. Size and average profitability are too descriptive and transitory. Judging on the basis of profit strategy would be more suitable except that not all authors characterise a firm’s strategy in the same way. Ultimately, in as much as the region of origin remains the framework within which firms adopt the profit strategy that they subsequently apply – for better or for worse – to the whole of their global group, this appears to be the most useful classification principle.
4 The Second Automobile Revolution
Our research has given us a vision of global groups that is totally different from the one people usually develop. We often read that so-called world firms are in the process of freeing themselves from important national constraints or that they are forcefully manipulating these to their benefit. It is true that firms can play on fiscal and social differences, but most continue to rely on the growth model found in their country or region of origin and then link into the growth strategies and models found in the countries where they are moving. Companies’ profit strategies and conceivably their productive models will have been constructed in a framework enabling the growth strategy and model typifying their country or region of origin. The only chance they have for lasting profitability is if these strategies and productive models are compatible with the growth strategies and models found in the countries where they move. One of the reasons why so many alliances or mergers have failed is that the firms involved pursued different profit strategies, undermining the anticipated synergies and complementarities. In addition, insofar as firms that are considered global do not all pursue the same profit strategies, they will have diverging views of which conditions they would like to see enforced in the different countries where they operate (Boyer and Freyssenet, 2000; Freyssenet, 2008). All in all, analysis of these firms can be divided into four parts: ‘The Divergent Trajectories of Japanese and Korean Carmakers’, ‘The Resistible Decline of the “Big Three”?’, ‘The Resistance of Leading European Carmakers’ and ‘Newcomers and Global Suppliers’. The range of automakers studied here is extended beyond our 2003 publications to include Mazda, Suzuki-Maruti and several Chinese automakers. Rover has been kept, despite disappearing as a company, to drive analysis about how a once major automaker can fade and be torn into pieces. Volvo and Mazda, which at the time of writing still belong to the Ford group, are also retained, despite future uncertainty. The conditions in which the Premier Automotive branch’s other brands (Land Rover, Jaguar and Aston Martin) have been sold are part of the analysis of Ford’s trajectory. General Motors and Fiat built an alliance during the period in question but it quickly fell apart. Dacia and Samsung are dealt with in the Renault chapter, while Nissan, despite being controlled by Renault – the only major ‘alliance’ to have flourished for the moment – is important enough to justify separate treatment. Another mega-merger that made the headlines but has collapsed since is DaimlerChrysler. These two automakers are treated separately. The only Indian company to make the list is Suzuki-Maruti, although a discussion is offered on the context within which it competes. Chinese automakers are studied successively, with special attention paid to Geely, an independent automaker which experienced extremely fast growth until the crisis. Leading suppliers are also dealt with in a single chapter, because of their number and our inability to mobilise researchers interested in their particular path.
Introduction: Ten Years On, What Have We Learnt?
5
The birth of a second automobile revolution becomes clear over time, and the book’s conclusion focuses on this one topic. After the first automobile revolution, which was the automobile itself and the internal combustion engine, we are witnessing a second revolution stemming from the combination of two rapid developments that this time around might actually last: new automobile markets (Brazil, Russia, India, China) with growth rates and size projections far exceeding past events witnessed in the USA, Europe and Japan; and the ongoing and probably irreversible move towards alternative energies and driving modes. The implications of these two changes, which despite appearances are interdependent, will turn the automobile upside down, at an industrial level but also in terms of utilisation. The question arises as to which old or new automakers will benefit from upcoming upheavals and what will this cost automobile employees and producer countries. We suggest a renewed analysis conducted in light of the transformations occurring in national or regional growth models and corporate strategies, so as to discern the effects on different ‘company governance compromises’ and productive models. Translation by Alan Sitkin Bibliography Boyer R. and Freyssenet, M. (1999, 2006) Le Monde qui a changé la machine, un siècle d’histoire automobile, http://freyssenet.com Boyer, R. and Freyssenet, M. (2000) ‘Fusions-acquisitions et stratégies de profit’, Revue française de gestion, 131 (November–December): 20–8. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Boyer, R., Charron, E., Jürgens, U. and Tolliday, S. (eds) (1998) Between Imitation and Innovation: the Transfer and Hybridization of Productive Models in the International Automobile Industry. Oxford and New York: Oxford University Press. Carillo, J., Lung, Y. and van Tulder R. (eds) (2004) Cars . . . Carriers of Regionalism? Basingstoke and New York: Palgrave Macmillan. Charron, E. and Stewart, P. (eds) (2004) Work and Employment Relations in the Automobile Industry. Basingstoke and New York: Palgrave Macmillan. Durand, J. P., Stewart, P. and Castillo, J. J. (eds) (1999) Teamwork in the Automobile Industry: Radical Change or Passing Fashion? Basingstoke: Palgrave Macmillan. Freyssenet, M. (2008) ‘Stratégies et modèles nationaux de croissance: Proposition d’une démarche d’analyse et esquisse d’un schéma d’analyse’, La Revue de la Régulation, 3: 21–42. Freyssenet, M., Mair, A., Shimizu, K. and Volpato, G. (eds) (1998) One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Freyssenet, M., Shimizu, K. and Volpato, G. (eds) (2003a) Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Freyssenet, M., Shimizu, K. and Volpato, G. (eds) (2003b) Gloablization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan.
6 The Second Automobile Revolution Humphrey, J., Lecler, Y. and Salerno, M. (eds) (2000) Global Strategies and Local Realities: the Auto Industry in Emerging Markets. Basingstoke and New York: Palgrave Macmillan and St. Martin’s Press. Lung, Y., Chanaron, J. J., Fujimoto, T. and Raff, D. (eds) (1999) Coping with Variety: Product Variety and Production Organization in the World Automobile Industry. Aldershot: Ashgate. Womack, J. P., Jones, D. T. and Roos, D. (1990) The Machine that Changed the World. Basingstoke and New York: Palgrave Macmillan.
2
Wrong Forecasts and Unexpected Changes: the World that Changed the Machine Michel Freyssenet
Since the 1990s, there has been an almost unprecedented proliferation and succession of forecasts and recommendations that automobile firms are supposed to take into account if they wish to survive and/or maintain their independence over the short term. The list is long: lean production, globalisation, cost-cutting, ‘new economy’, externalisation, modularisation, concentration, commonalisation, innovation, delocalisation, global sourcing, etc. At the same time, many developments and options have been underestimated and sometimes even ignored, despite the fact that the automobile world which they depict is very much at odds with the idea of a global convergence in strategies and practices: heterogeneous demands and markets; emergence of new carmakers; serious problems facing manufacturers that have long been at the top of the pile; the hiccups faced by some of the emerging countries and very rapid take-off of continent-sized countries; confirmation of a divergence in national automobile policies; skyrocketing petrol and raw material prices; increasingly stringent environmental standards; accelerated transition to alternative driving systems; the world financial crisis, etc. Moreover, it seems that some of these changes have paved the way for what might be called a ‘second automobile revolution’. The real questions are why so many forecasts have been wrong; whether current changes were foreseeable; and how to develop more reliable scenarios. A few initial responses can be found in some of GERPISA’s earlier studies. The chapters that follow will develop some of these themes. Here we use the profit strategy and productive model analytical schema that GERPISA developed (Boyer and Freyssenet, 2002) to debate and synthesise these responses.
The bursting of the property bubble in 1990 and the end of firms’ ‘Japanisation’ The monetary and oil crises of the 1970s pitted countries characterised by internally driven growth (i.e. propelled by internal consumption indexed to 7
8 The Second Automobile Revolution
higher national productivity) against other countries where growth had long been driven by exports and the indexation of national income distribution to external competitiveness. The need to offset higher oil bills and disadvantageous currency movements through rising export surpluses was detrimental to the former group (notably comprising the United States, France and Italy) and beneficial to the latter (led by Germany, Japan and Sweden). Despite their different growth drivers, these countries all had a similar form of national income distribution at the time, one based on the coordination of different sectors of activities and companies and featuring moderately hierarchised social and professional categories, a system that enabled as many people as possible to access mass consumption and benefit from satisfactory levels of social protection. Due in part to a slowdown in demand, the US, French and Italian carmakers discovered that they were not competitive enough compared to their German, Japanese and Swedish counterparts. The most notable members of the latter group were the Japanese carmakers, whose value for money competitiveness surprised everyone who recalled the mediocre quality of the vehicles that they used to export in the 1960s. This awareness triggered a major search for ways of explaining this unexpected superiority. The first explanation was automation, considered more precocious and advanced in Japan than elsewhere. This analysis not only motivated American and European companies to try to overcome their supposed shortcomings in this domain but also to seek long-term technical supremacy by designing equipment that would be both more flexible and integrated than anything seen before. The disappointing results of these efforts were the unavoidable effects of the problems faced in using unwieldy new equipment. Instead of providing a solution to firms’ financial crises, things got worse. This technological determinism was followed by an emphasis on cultural relativism. A collective spirit and a desire for consensus, presented as Japanese cultural characteristics originating in the mists of time, would supposedly help employers and employees progressively develop solutions of unparalleled efficiency that would be both original and adapted to the problems faced. The Japanese cultural model began to be portrayed as a benchmark for everyone. The net effect was to create a sense of desperation. After all, it was logically impossible for Americans and Europeans to acquire the qualities that were supposed to have engendered Japanese competitiveness if such qualities derived directly from the ‘culture’ of this one country. Out of this came the idea that getting the Japanese to establish operations in those countries whose markets they had penetrated so abruptly would be the best way to force them to produce in the same social conditions as the ones experienced by their competition, thus depriving them of the competitive advantage that they supposedly derived from their national culture. The success of these first ‘Japanese transplants’ in the United States was a death-blow both to the hopes of curtailing Japanese success and to the power of this particular explanation.
Wrong Forecasts and Unexpected Changes 9
Since it was neither technical nor ‘cultural’ in nature, people soon took to diagnosing the emergence of a new productive model, one that was destined to become universal and take over from a ‘Fordist’ model that was no longer adapted to an increasingly variable and diversified market. The new model, generally called the ‘Japanese model’, was given the more appropriate name of ‘lean production’ by researchers from MIT trying to characterise its spirit and emphasise its ability to be adopted elsewhere (Womack et al., 1990). The partisans of lean production attributed remarkable qualities to this construct, to such an extent that they viewed its rapid adoption as an absolute condition for firms’ survival. Lean production was meant to resolve the double crisis of the 1970s: an insufficiently diversified production that was not adapted to the demands of a customer base demanding better quality, price and delay conditions; and the crisis of a fragmented work system that employees no longer accepted. Small batch production, polyvalent operators and equipment, driven by the market production – all of this was supposed to increase the variety of the products on offer. Employees’ participation in efforts to eliminate defects and waste and in the search for the causes of dysfunctions – operating through ‘quality circles’ or in working teams under the supervision of a ‘leader’ – gave new meaning to assembly line work and offered everyone a chance to increase their competence with productivity and product quality continuing to improve. Employees would be remunerated for their day-to-day involvement by employment guarantees and good career prospects. Employer commitment to employees would also extend to subcontractors who would accept cost reduction and quality objectives and apply similar methods within their own organisations in exchange for the promise of sufficient volumes and profit margins. Thus, everyone with an interest in the firm would benefit: consumers would enjoy a greater variety of supply, innovations, better quality and cheaper products; shareholders would enjoy regular profits; and the state would benefit from the existence of prosperous firms that would be internationally competitive and sustain social harmony. In both professional and academic circles, this seemed like an open and shut case. A new productive model had been born and was destined to replace the old ‘Taylorian–Fordian’ model that had demonstrated, through its organisational rigidity and because it was now being socially rejected, an inability to respond to the new requirements of the marketplace and of society. An accelerated liberalisation of trade and globalisation of competition during the early 1990s seemed to confirm the need for marketresponsive companies that would constantly mobilise all concerned to ensure a more efficient use of inputs. In this view, the reorganisations undertaken by the American and European automobile firms could not help but converge towards lean production, in much the same way that the changes that followed the Second World War were analysed by some as having caused a generalisation of the ‘Taylorist–Fordist’ model. For many observers, it was
10
The Second Automobile Revolution
self-evident that lean production would change the world. But did this really happen? Carmakers who had been in dire straits during the first half of the 1980s began with a sharp reduction in their break-even points, rationalising their product ranges by closing many plants and cutting staff levels. They also reduced debt by selling off assets. Some of these firms, notably General Motors, Ford, Renault and Fiat, modified their organisations by adopting ostensibly Japanese systems (and letting everyone know what they were doing). The occasionally spectacular improvement in their results during the latter half of the 1980s was viewed as proof of the remarkable efficiency of these methods and the inevitable diffusion of the new ‘lean’ productive model. We now know that things turned out very differently. Seeking to improve the competitiveness of American goods, the US government estimated that the most direct path towards this end would be to deregulate the labour markets, massively support the electronics sector through ‘Star Wars’ military research and liberalise the capital markets. The effects of these measures, amplified by the counter-oil shock of 1986 and a falling dollar, were considerable, in the United States of course, but also in other countries following in its wake. A ‘speculative bubble’ took shape, notably in the property sector. In turn, this led to a strong resurgence in consumption. Many carmakers continued to follow a ‘volume and diversity’ strategy based on increased sales volumes and a commonalisation of invisible components, on the one hand, and on a diversity of bodies and visible equipment on the other (Boyer and Freyssenet, 2002). This temporarily helped them to regain viability and embark upon a recovery path. Many of the firms in this category formally adopted measures like working groups, lean flows, self-control, quality circles and total productive maintenance – but forgot that to be efficient, these kinds of systems had to be entirely geared towards staff reductions with constant volumes, and that these reductions would only be accepted by employees if they were based on a ‘social compromise’ guaranteeing employment and career prospects (Shimizu, 1999). In addition, lean production, in the way that it was theorised, amalgamated systems that had been observed at Toyota, Honda or Nissan, even though all of these firms pursued completely different profit strategies and productive models, as GERPISA studies later showed (Hanada, 1998; Freyssenet and Mair, 2000; Shimizu, 1998). For example, some people assimilated Toyota’s lean flows with Honda’s responsiveness, yet the former implied rigorous production planning whereas the latter was based on innovation rents derived from innovative models being launched before competitors could copy them. What followed was the construction of a number of systems that were cut off from their original objectives and even contradicted them on occasion, leading to some disappointing results. The sum total of best practices can only be transformed into profits for a firm if they are compatible with one another,
Wrong Forecasts and Unexpected Changes 11
i.e. if they all respond to the requirements of one and the same profit strategy by applying means that are coherent and which result from the ‘company government compromise’ that the firm’s main actors have agreed (Boyer and Freyssenet, 2002). This edifice was destroyed when the property bubble burst in 1990. At first, the effects were only manifested through lesser demand for automobiles, a situation for which a countermeasure was soon found in the United States, although not in Europe and Japan. The confrontation between internally oriented and extrovert economies did not cause the former to turn into the latter, as the theories of convergence had been predicting. France and Italy did try to go down this path but, as we have seen, the United States chose an entirely different road, maintaining its internally driven growth model grounded in the vast American market, even as the country abandoned its coordinated and moderately hierarchised mode of national income distribution and replaced it with a ‘competitive’ distribution mode based on ‘merit’, a local and categorial balance of power and financial opportunism. This led to a deep-seated change in the structure of income and demand, as well as in the conditions of employment and type of labour that could be mobilised. Massive injections of public monies in the 1980s, the counter-oil shock, exchange rate shifts, new venture finance facilities, adjustments in employment and wages and the mobilisation of capital – all of this contributed to a growth cycle that the United States had not seen for many years. The competition between economies also changed. Instead of pitting self-oriented against extrovert systems, liberalised regimes now vied with regulated ones. The chief weapon used by the former was the free movement of capital – a freedom that the regulated economies imprudently accepted. In Europe, the effects of the bursting bubble were amplified by the cost of German reunification and by the restrictive policies that the European governments felt they had to apply in consequence. Earnings fell as quickly as they had risen. The carmakers all reacted immediately through ad hoc cost-cutting exercises, albeit in different ways. Volkswagen and PSA accentuated their commonalisation policy and returned to the scale economies philosophy, with VW absorbing a fourth carmaker (Skoda) and PSA signing several cooperative agreements. Other carmakers placed their hopes in new market trends. Fiat began highlighting those countries that people were starting to refer to as ‘emerging’. Renault had the pleasant surprise of seeing the European market adopt its innovative models en masse. The specialist carmakers attempted to come up with top-of-the-range models for the different vehicle categories. Yet the new organisations failed in their original purpose, which was to offer a positive solution to continued demand stagnation. This raises questions about whether the model was being poorly or insufficiently applied; if it was ever transposable; whether it was adapted to different corporate contexts
12
The Second Automobile Revolution
and histories; and if it had ever done as well as some people said. On this score, surprising news began to filter out of Japan. In the 1990s, Japan found itself in a recession phase that would last for more than ten years. When the speculative bubble burst, the country found itself with a mountain of bad debt. The Japanese banking system was on the brink of collapse and sharply cut corporate loans after years of lax lending. Only Toyota and Honda, which for different reasons had long relied on selffunding, were unaffected. Both continued to cut costs – with the former adapting new methods and the latter implementing conceptual innovation and responsiveness – and they muddled through despite the fall in domestic demand and strong yen. The other Japanese carmakers suffered from the conjunction of a credit crunch, weak demand, increased competition and an overvalued currency. European manufacturers were pleasantly surprised to be able to fight off Japanese competition at a time when their own markets were depressed and when Japanese ‘transplants’ in Europe were operating in the red, unlike their counterparts in the United States (Pardi, 2005). What came out of this was the idea that lean production could not become the system that would change the world. Doubts were even being expressed about the construct’s ubiquitousness in Japan, given its inability to save its country of origin from the economic downturn. The question was, what had gone wrong? Back when the speculative bubble was swelling, automobile demand had skyrocketed in Japan more than elsewhere. Having considerably increased its overtime requirements, Toyota found it hard to recruit young workers willing to operate under such strenuous conditions. In other words, in a context of full employment, the Toyota production system began to suffer from limited social acceptability. A labour crisis ensued, requiring a thorough transformation of the system and reconstruction of the new ‘company–government compromise’. Toyota began looking overseas for ideas, turning to its foreign subsidiaries but also (and in a surprising turnaround) to European carmakers like Volvo or Mercedes, just as these firms were starting to question the sociotechnical principles underlying their long-standing work organisation. In Japan, production lines were resegmented and buffer stocks reintroduced, with each sub-assembly now being attributed to one team in an attempt to make the work slightly more meaningful, at this level at least. Employees were no longer expected to be responsible for reducing their own standard times and the wage system was reformed in a way that limited overtime and ensured greater planning of this kind of arrangement (Shimizu, 1999). The productive model that certain American or European firms had felt that they needed to copy was thoroughly modified by the very company that had invented it. Instead of creating doubts about the changes being made to the Toyota system, the new business climate in the early 1990s confirmed, if need be, that the company’s employees would no longer accept the old organisation and its underlying principles. The idea of guaranteed employment for
Wrong Forecasts and Unexpected Changes 13
everyone would later be abandoned and replaced by fixed-term employment contracts, reflecting Toyota’s inability to smooth out cyclical variations in demand (see Shimizu, Chapter 4, this volume). Honda, on the other hand, had not suffered any crisis of labour during this speculative bubble. Its employment relationship was more in sync with a young workforce’s new aspirations and the improved working conditions it offered protected it from wholesale rejection. Honda’s responsiveness and the flexibility of its productive apparatus – both requirements of a product innovation-based profit strategy – helped it to adapt more quickly. Thanks to its policy of self-funding and independence, the company was, unlike the other carmakers, not affected directly by the collapse of the Japanese banking system. The end result is that Honda became Japan’s second-largest manufacturer (Freyssenet and Mair, 2000; Luethge and Byosière, Chapter 6, this volume; Shimokawa, 2003). Nissan, Mitsubishi and Mazda, all heavily in debt, were unable to cope in a regime marked by more expensive (and less available) credit. International opinion was astounded by the weaknesses they manifested, in ways that were entirely at odds with the Japanese model: waste, duplication, overcapacity, paralysis, internal conflict, neglect of customer demand, wilful hiding of defects, etc. To avoid bankruptcy, the firms were forced to turn to European and American companies, respectively Renault, Daimler and Ford (Heller, Chapter 7, this volume; Kumon, 2003; Shimizu and Shimokawa, 1998; Stevens and Fujimoto, Chapter 5, this volume). The situation could be analysed, metaphorically, as an example of a master, drunk with success, forgetting basic principles and/or of students overtaking their master. In reality, there had never been a ‘Japanese productive model’ but instead a diversity of strategies and compromises between actors and socio-productive configurations subjected to internal and external constraints that forced them to evolve, similar to the way that things happened in other countries. As for the recovery of American and European firms and their comparatively good health, this was mainly caused by a change in markets and labour factors in their regions of origin.
A more ‘competitive’ distribution of national income and the heterogenisation of the demand for automobiles In the wake of great social and political tension, the ‘regulated’ economies progressively began to increase the flexibility of their labour markets, modifying their wage-setting rules and cutting social protections. Growth in these countries had been insufficient to pay for their high level of public spending and their ‘national compromises’ were breaking up. Company executives, shareholders and independent professionals were impatiently hoping to be remunerated at what they saw as their true value, in line with standards set in the United States and Great Britain.
14
The Second Automobile Revolution
To varying degrees and via different modalities, the national income distribution became more ‘competitive’ in the regulated economies. Some older and newer segments of the population benefited – others did not. The demand structure, for automobiles in particular, was significantly modified. The demand diversification that had followed the shocks of the 1970s was replaced by heterogeneous expectations reflecting shifting social trajectories. The first new social trajectory – and also the most far-reaching one – affected those segments of the population that benefited from better career prospects and faster wage hikes than they had previously experienced due to ‘merit-based’ remuneration systems and a categorial and local balance of power. The same segments (and others as well) could also take advantage of the financial opportunities created by lower taxes and sometimes by investment incentives. Of course, they were never sure of keeping their new affluence since there was every chance that their ‘merits’ would no longer be recognised if the balance of power shifted again. Hence the frequency of ostentatious, aggressive or defensive behaviours by the ‘new rich’, a situation perfectly illustrated by the general enthusiasm for urban sports utility vehicles (SUVs), vehicles of dubious utility (exemplified by how often their parking encroaches on pavements) but symbolically useful ones because of the way they stand out in traffic. The traditional and more durably affluent segments of society were also able to take advantage of the new situation. The past had taught them, however, that luxury is only socially acceptable if swathed in codes of ‘good taste’. Hence the success of top- and very top-of-the-range cars whose charms were both visible and discreet. Also benefiting de facto (albeit to a lesser extent) were all employees whose wage systems continued to be based on qualifications, seniority, employment guarantees and social protection. The slow but steady progression in some of their income sources meant that they developed a preference for vehicles that were more practical and user-friendly. This translated into a great success for minivans but also for recreational vehicles, even though they were much more expensive. On the other hand, many employees and independent workers lost confidence in their long-term prospects or else could no longer afford a new vehicle. Demand from this segment shifted to the used market. Hence the unexpected success in the mature markets of the entry-level vehicle, the Renault Logan, initially designed for markets in Central and Eastern Europe. Carmakers themselves contributed to the new vehicle market’s slow growth by failing to notice the desertion of customers lacking the financial wherewithal to purchase the models they were offering (Jullien, 2006). Lastly, we should note the rise of automobile ‘fleets’ purchased by companies for their employees or by short- and long-term car rental firms. Such purchases accounted for up to 50 per cent of all sales by carmakers in certain countries and became a significant outlet for certain more conventional models like saloons. But they were also less lucrative for manufacturers forced
Wrong Forecasts and Unexpected Changes 15
to offer major discounts. Companies and rental firms, who often sold off used fleet vehicles at slashed prices, also contributed to the fall in passenger car resale prices. The company car system was a cheaper way to ensure executive loyalty than wage hikes, due to the tax exemptions often associated with these vehicles and because the firms could always ask for the car to be returned. At the same time, this is a system whose beneficiaries often had a second and even a third vehicle to drive. In much the same way as the diversification of demand in the 1970s could be predicted and satisfied through a greater variety of bodies and equipment produced using shared platforms as per the traditional ‘volume and diversity’ strategy, now an increasing heterogenisation of demand was to be expected. This required new types of vehicles and sparked a renaissance of the ‘innovation and flexibility’ strategy, adopted de facto by Chrysler and Renault and also pursued by Honda (Boyer and Freyssenet, 2002). The first firm to feel the changing winds was Chrysler in the United States. Returning to its original ‘innovation and flexibility’ strategy (Boyer and Freyssenet, 1999, 2006), and after having long meditated on Honda’s experience (notably the success of the Honda Civic), Chrysler started to design and launch minivans, followed by SUVs and so-called fun pickups. A prime example of desperation in the American automobile industry, Chrysler was to everyone’s surprise reborn from its ashes (Belzowski, 1998 and Chapter 11, this volume). General Motors and Ford realised that they needed to copy its conceptually innovative vehicles. Over time, light trucks would amount to almost 50 per cent of the American new vehicle market. Very lucrative, this new kind of passenger vehicle sparked the recovery of the Big Three, especially once they were able to import foreign light trucks, mainly from Japan at the time, after convincing the government to keep import taxes down to 25 per cent. More than any other recovery initiative or the imitation of Japanese methods, this shift in product policy, associated with a sharply higher demand, is what allowed them to rebalance their books and increase earnings throughout the 1990s. In Europe (and particularly on the continent), the market also heterogenised but more slowly and less unequivocally. Renault was the first to propose a monospace in this part of the world (the Espace), at about the same time as Chrysler was launching its Voyager in the United States. Unlike Chrysler’s model, the Espace was slow to break through. Ultimately, however, this type of vehicle became a commercial success, especially once Renault had the idea of breaking it down into smaller models a few years later. It is the success of this monospace product range, and particularly of a lower midrange segment model called the Scénic, that allowed Renault not only to start a new cycle of profitability but also to successively acquire Nissan, Dacia and Samsung (Freyssenet, Chapter 14, this volume). One sign that the diffusion of a ‘competitive’ distribution of income was slow and perceived as uncertain in Europe is that European manufacturers long hesitated before
16
The Second Automobile Revolution
launching urban SUVs, despite the success of American or Japanese models amongst the European customer base. In Japan and the United States, Honda began in the late 1980s to highlight the burgeoning segment of financial ‘yuppies’ by offering expensive sporty coupés. However, demand for this kind of vehicle crumbled when the property bubble burst. Honda’s responsiveness meant that it was very quick to latch on to the much bigger and more lasting demand for minivans and compact SUVs. The heterogenisation of the market created a difficult dilemma for carmakers pursuing ‘volume and diversity’ and ‘reduction in costs with constant volumes’ strategies. Their vocation may have been to follow the leaders, i.e. letting the ‘innovation and flexibility’ firms take the risk of exploring new market segments and copying innovative models intended to sell high volumes over the long run, but they still had to be careful not to run too many platforms. In big markets like the United States, creating new platforms, for light trucks for example, did not stop companies like General Motors or Ford from achieving satisfactory economies of scale. The same did not apply in smaller markets, however. Hence the tendency, notably among European carmakers, to design new vehicle types as body variants based on the same platforms as traditional vehicles, or else to launch their own ‘innovative vehicles’ based on hybrids of existing body types: coupé-convertible, coupémonospace, monospace-SUV, station wagon-monospace, saloon-SUV, etc. (Freyssenet, Chapter 13, this volume). Europe’s lesser market segmentation and greater equality in terms of incomes and social trajectories meant that this strategy was commercially more acceptable, even if the vehicles involved often did not ‘hold the road’ as well. Hence the new vehicles’ need for a lower commonalisation rate, and the unmitigated success of ‘mixed’ vehicles. Demand heterogenisation was also a challenge for specialist manufacturers pursuing a strategy that has been called ‘quality-based’ (Boyer and Freyssenet, 2002). These players understood belatedly that affluent customers wanted top-of-the-range models for all sizes and types of vehicles, and not only large ‘road-tested’ saloon cars. Generalist carmakers were starting to appeal to their customers by launching top-of-the-range versions of their small and mediumsized models. Not all specialists were in a position to fund a considerable broadening of their product lines, however. The English makes ( Jaguar and Aston Martin) and the two Swedish carmakers (Saab and Volvo) saw no other solution than to sell out to generalist manufacturers. General Motors acquired Saab, and Ford created Premier Automotive by pooling its own Lincoln with Jaguar, Austin Martin, Volvo and Land Rover (Berggren, 1998; Bordenave, 1998; Bordenave and Lung, 2003; Matthias and Pil, Chapter 18, this volume; Mercer, Chapter 10, this volume). Volkswagen bought Bentley’s luxury brand and resuscitated Bugatti (Jürgens, Chapter 12, this volume). The two leading specialists, BMW and Mercedes, took different paths. BMW tried to make up for lost time by acquiring Rover, an English generalist with interesting
Wrong Forecasts and Unexpected Changes 17
models for a top-of-the-range customer base, like the Land Rover and the Mini Cooper (Coffey, Chapter 19, this volume; Eckardt and Klemm, 2003; Pries, Chapter 17, this volume). As for Mercedes, it believed that carmakers would henceforth need to produce all types of vehicles, ranging from the traditional to the innovative and from the very top-of-the-range to the very bottom. Hence the moves it made in the late 1990s: the acquisition of Smart, the absorption of Chrysler and takeover of Mitsubishi, all in an effort to cover as many global market segments as possible. The net effect was to accumulate all possible risks and challenges all at once. No automobile firm has ever achieved a successful cohabitation of strategies based on different profit sources. These initiatives all failed. General Motors was unable to get anywhere with Saab. Ford had to dissolve Premier Automotive and successively sell off Aston Martin, Jaguar and Land Rover. There is every likelihood that it will also have to sell Volvo in the near future. Daimler quickly got rid of Mitsubishi (Köhler, Chapter 16, this volume) and resold Chrysler after heavy losses. It has only kept Smart, which is just starting to turn profits more than ten years after its foundation. The creation of shared platforms for top-of-the-range models and mass diffusion models was either impossible or very damaging to the former. BMW ultimately muddled through without suffering too much, probably because it had not strayed too far from its ‘quality-based’ strategy. It sold Rover for a symbolic one pound sterling, while keeping its small top-of-therange vehicle (the Mini Cooper) and acquiring the Rolls-Royce brand. Out of all the specialists, BMW is the healthiest today. Will demand heterogenisation persist through self-renewal? The answer is very probably yes, if the ‘competitive’ distribution of income survives and especially if it replaces a coordinated and moderately hierarchised distribution of income in more and more countries. However, a second source of heterogeneous demand has now appeared with the emergence of new automobile markets. To understand this new situation and the uncertainties that characterise it, the second category of confrontations that many countries face will need to be analysed.
Carmakers were wrong at first about ‘emerging’ countries and the sorts of models that could be sold there In addition to the confrontation between more or less ‘liberalised’ and ‘regulated’ capitalist economies, there was also the confrontation between capitalist and communist countries. The fight with the ‘socialist bloc’ (before it imploded) strongly contributed to the emergence of new industrialised countries, particularly in Asia (South Korea, Thailand, Indonesia, Philippines and Malaysia), which were nurtured to keep them from falling into communist arms. Thanks to a more liberal trade regime and an indulgent attitude by the United States, these countries, which since the 1970s had adopted an
18
The Second Automobile Revolution
industrialisation trajectory based on exports of low value-added manufactured goods, were able to find the growing outlets that they needed as well as the (notably financial) resources that would allow them to migrate upstream towards higher value-added products even as they continued to protect their domestic markets. A ‘middle class’ appeared, one that wanted to consume things and acquire the attributes of modernity, notably automobiles. Latin America experienced an economic boom with certain economies like Argentina and Brazil being pressured by international financial institutions into liberalising their large-scale public sectors. Central and Eastern Europe’s abandonment of socialism created the hope in some quarters that they would quickly become part of the growing ranks of ‘emerging’ countries. Seeking lucrative investments, foreign capital flowed to these new destinations. Almost all of the carmakers developed plans for new facilities, some of which actually materialised, leading to sizeable investments. Companies were hoping to find the volume growth that they could no longer hope to achieve in the Triad countries, and which they had been unable to find during the 1970s in the world’s oil and raw material-producing countries. General Motors and Fiat were the most entrepreneurial in this respect. The former announced the creation of four ‘high-technology’ factories in Argentina, Poland, Thailand and China, each with a capacity of 180,000 vehicles annually. Fiat designed a special emerging country model called the Pallio/Siena (Volpato, Chapter 15, this volume), built a factory in Argentina for 150,000 vehicles annually and took control of Poland’s FSM. Renault, Volkswagen/Audi and Chrysler all built new factories at Curitiba in Brazil. And then a new crisis erupted, resulting once again from a speculative bubble. The impetuous growth in many of these markets was largely due to outof-control debt. When this reached worrying levels, international investors suddenly withdrew, with the crisis of confidence spreading from Southeast Asia in 1997 to Latin America and Eastern Europe. The automobile market in all these regions collapsed immediately and Japan, which was finding it difficult to put its own ‘property’ crisis behind it, again got caught in the storm, its banks having been unable to resist the temptation of investing capital in ‘emerging countries’ without sufficient guarantees. The constitution or consolidation of a middle class that would be likely to generate mass automobile demand in the ‘emerging’ countries of Latin America and Southeast Asia was regularly hampered by the financial or political shocks that the populations involved would experience. As for the former communist countries, starting with the regional leader (Russia), it was clear that in the absence of the kinds of institutions without which a capitalist economy cannot exist or prosper, they would be unable, at least in the short run, to launch a durable growth process creating a new vehicle market. As soon as they were finished, the factories built at this time would discover that they were suffering from over-capacities and that the models launched were unsuitable to local circumstances. Fiat suffered the most from the
Wrong Forecasts and Unexpected Changes 19
turnaround in these ‘emerging’ country markets, due to the sizeable financial and human investments it had made in them, possibly to the detriment of its European market (Volpato, Chapter 15, this volume). Few carmakers were backing India and China at the time (BeckerRitterspach, Chapter 21, this volume; Wang, Chapter 20, this volume). The two countries’ economic liberalisation path was considered uncertain, and too conditional, slow and politically manipulated (Kim, 2004; Thun, 2004). Only Volkswagen and PSA did the required work here, the former successfully, the latter much less so for many years. These continent-sized countries, soon rejoined by Brazil and Russia (having retaken control of their own destinies following the vagaries of earlier IMF and World Bank dictates), named BRIC, would provide the growth drivers that so many actors were looking for. Today, their markets exceed in volume and growth terms anything ever witnessed in the history of the automobile. Thanks especially but not only to its acquisition of parts of the Korean company Daewoo, General Motors has been able to penetrate the Chinese market just as it is starting to take off. Together with Volkswagen, Honda and Toyota, it became in 2007 one of the main producers there (Senter and McManus, Chapter 9, this volume). Despite this, it still took a long time until Triad carmakers’ local production broke even (Jetin, 2003 and Chapter 3, this volume). Producing in low-cost countries can be very costly due to the absence of good supplier, subcontractor and distribution networks, insufficient quality, deficient infrastructure, untrained labour, unstable legislation, over-active bureaucracies and a lack of knowledge about customer preferences due to the inapplicability of traditional marketing surveys. From dying models to models that succeeded in Africa and are now destined for China or India – without forgetting smaller urban models, monospaces, five-door saloon cars and downsized versions of models sold in mature markets – there have been countless examples of mistakes being made in the products offered in the BRIC countries. Clearly, these new markets have their own structures and constraints, as well as different practical and symbolic needs. Some platforms have had to be modified and others created, new models and body versions designed and design centres set up close to the intended market. National regulations, aimed inter alia at giving local manufacturers a chance to compete, are anything but uniform. Demand heterogeneity in mature countries also applies in the ‘emerging’ world. While investing in these new locations, Triad carmakers have discovered that automaking is a bigger business there than they first imagined, and above all that local carmakers or big industrial groups also intend to become actors, not only in their own countries but abroad as well. A number of joint ventures and cooperative agreements have been signed with a view towards acquiring the know-how needed to become autonomous one day. Others firms have tried, quite successfully so far, to develop independently by purchasing the technologies that they need (Geely and Chery in China)
20
The Second Automobile Revolution
or by acquiring Western brands like Jaguar and Land Rover, as India’s Tata did (Becker-Ritterspach, Chapter 21, this volume; Wang, Chapter 20, this volume). In reality, what these new players are doing is following the path that Japanese and subsequently Korean carmakers once took. But much more than merely copying other players, they have started to innovate themselves, notably in the low-cost vehicle niche. Examples include Tata’s Nana and Bajaj’s current plans for a $2,500 vehicle.
‘Globalisation’, if it ever happens, will have to wait: the failure of world cars and platforms, the collapse of mega mergers and the uncertainty of global sourcing The homogenisation of global automobile demand does not seem to be imminent. For this to happen, national growth models would have to become the same, national regulations would have to be unified and differences in driving conditions would have to either disappear or else be taken into consideration from vehicles’ design stage onwards. It can be demonstrated that the globalisation of trade and capital, which is far from complete, will generate different ‘growth and income distribution modes’, and thus automobile markets will be heterogeneous both with respect to one another and also internally (Boyer and Freyssenet, 1999, 2006). Yet it is the utopia of global homogenisation that animated the first theorists and practitioners of globalisation, who imagined that with trade liberalisation and increasingly homogeneous demand global firms could one day apply the same production organisation (design, manufacturing, distribution, services) everywhere, choosing the most profitable locations based on their competitive situation and freeing themselves not only from their country of origin but also from any territorial determinism. Ford was the first to try to live this dream by specialising its production regions (North America and Europe, and Japan, via its Mazda subsidiary) into one of the three main segments of what was supposed to become a global market: top-of-the-range, midrange and bottom-of-the-range. The first model with a global vocation that it launched was a midrange vehicle called the Mondeo. Because of poor sales everywhere except in Europe, this was the end of Ford’s globalist efforts (Bordenave and Lung, 2003; Mercer, Chapter 10, this volume). General Motors, as per its Sloanian strategy, did not try to design global models but instead built global platforms that were supposed to be used as a basis for designing regional models. This led to numerous conflicts within the group and ultimately prevented the creation of any global platforms at all (Bordenave and Lung, 2003; Flynn, 1998; Senter and McManus, Chapter 9, this volume). The same happened when Renault and Nissan tried to create a shared platform. Ten such platforms were announced but only two were built, painfully and after extensive redefinition. No other shared platforms would come afterwards.
Wrong Forecasts and Unexpected Changes 21
The heterogenisation of global demand has cast a shadow over everyone’s desire to use globalisation to reduce platform numbers. One example is the Logan, a five-door, low-cost saloon car requiring a new platform. Some of its variants (the MPV) or regional adaptations (the Sandero) have also needed to adapt strongly its platform (Freyssenet, Chapter 14, this volume). This has led to a change in the meaning of the term ‘globalisation’, which can be used more modestly to designate groups’ efforts to be present commercially and industrially in the world’s main regions (Jetin, 2003 and Chapter 3, this volume) and to offer all of the vehicles that different markets require. Being present everywhere and doing everything have become the two main criteria for globalised groups. However, this has not always been sustainable, especially the criterion of ‘doing everything’. As we have seen, it was for similar reasons that DaimlerChrysler failed, leading to the divorce of the new consolidated group. D. Piech’s departure from the Volkswagen group executive was an opportunity for his successor to return to the basics of the ‘volume and diversity’ strategy (Jürgens 1998 and Chapter 12, this volume). But even the criterion of ‘being present everywhere’ has become difficult to fulfil. As the automobile markets have started to awaken in continent-sized countries like China, India or Russia, there is much less need to maintain a presence elsewhere in the region. In 2007, only two carmakers produced more outside of their region of origin than they did at home: Honda (63 per cent) and Nissan (56 per cent). Apart from Renault, no carmaker has gained a foothold in Japan selling its own vehicles, i.e. the only possible entry has been to take over local manufacturers. Ford had to sell its Mazda shares. The mega consolidations that occurred in the automobile industry following the 1997 Asian crisis intimated a sudden acceleration in corporate globalisation and the formation of a global oligopoly comprising fewer than eight groups. Since then, it is notable that the three most durably profitable firms – Toyota, Honda and Volkswagen – are the ones that have voluntarily abstained from acquisitions or alliances with major manufacturers, even though they have the financial resources for this should they so desire. PSA, which has learned painfully what mergers really entail, has frequently proclaimed its preference for ad hoc cooperative arrangements instead of seeking a hypothetical critical mass (Freyssenet, Chapter 13, this volume). Out of the mega consolidations that occurred in the late 1990s, including Renault– Nissan, Ford–Mazda, DaimlerChrysler–Mitsubishi, General Motors–Fiat, BMW–Rover, only the first has survived. Taking over or acquiring controlling interests in smaller manufacturers has been a more sustainable path (Hyndai–Kia, General Motors–Daewoo, Renault–Dacia–Samsung, Toyota– Daihatsu–Fujy Heavy), although this too has not always been successful, since several of these operations have culminated in total or partial resales such as Ford–Premier Automotive with Jaguar, Land Rover and Aston Martin (see, in this volume, Chung, Chapter 8; Senter and McManus, Chapter 9; Shimizu, Chapter 4).
22
The Second Automobile Revolution
Mega consolidations would therefore appear to have had a smaller impact than expected. Certainly, the effects of these operations have been offset by the current or future advent of new market players. It is difficult to imagine that China, strongly motivated by a sense of historical injustice and which now possesses the financial, political, cultural and human wherewithal to achieve the autonomy it desires, will indefinitely accept industrial dependence on foreign firms. The same applies to India, with Tata, a highly ambitious firm, as an example. There are a number of reasons why the automobile market should expect new entrants in the future – not to mention the fact that major technical changes, notably relating to driving systems, could tempt leading suppliers or energy producers to make cars themselves, as discussed in the book’s conclusion. Analogous concentration and ‘globalisation’ trends affecting first-tier suppliers have resulted from the new compromises that these players and certain carmakers have agreed concerning the distribution of tasks. Not illogically, suppliers often accompany manufacturers when they move abroad, sharing both investments and risks (Frigant, Chapter 22, this volume; Humphrey and Salerno, 2000). Yet as discussed below, many manufacturers – starting with the most profitable, Toyota – do not delegate to suppliers responsibility for ‘module’ delivery and supply chain organisation (Shimizu, 2003 and Chapter 4, this volume). Additionally, a proliferation of subcontractors can be noted behind the top-tier suppliers, especially in the low-cost countries. But for many parts, logistics costs are already higher than manufacturing costs. Given the unpredictability of oil prices and the delays of overseas transportation, there is a possibility that companies will relocalise in the future, or at least rely to a lesser extent on distant producers. Future changes in the market will not necessarily involve convergence. Renewed heterogeneity is just as possible, an outcome that would betray the illusory nature of global product policies. If the aforementioned criteria were supplemented by the need for all foreign operations to have the same productive organisation (and even employment relationship), we would be hard-pressed to identify any global automobile firms at all (Charron and Stewart, 2004). What has happened instead is that certain carmakers are starting to regionalise their product policies, productive organisations and employment relationships. The future of this trend will depend on the future of the regions concerned, to wit, NAFTA, Europe, Mercosur and the continent-sized countries (Carrillo et al., 2004; Ramirez, 2007).
Is ‘cost-cutting’ a universal practice? Even as automakers (and suppliers) wait for the emerging world to deliver its much desired volumes and profits, there has been a broad consensus since the mid-1980s surrounding the need for cost-cutting. This has accredited the idea that the automobile industry has definitively changed its priorities, reflecting
Wrong Forecasts and Unexpected Changes 23
shifts in the global automobile market and the nature of competition. In this view, cost-cutting has replaced the search for economies of scale, in line with what the Japanese manufacturers have been doing for so long. Yet once again, the reality is quite different. A closer examination will show that the use of the term ‘cost-cutting’ has always covered – and continues to cover – very different objectives, practices and results. At least four uses and four realities can be distinguished. Cost-cutting is said to occur when a company takes measures to bolster a financial situation compromised by declining sales or out-of-control costs. This often involves redundancies, factory closures, asset divestments or temporary wage freezes. Since the late 1970s, many carmakers have followed this path at one time or the other (and often repeatedly), always trying to rebuild their existing profit strategy on a healthier financial basis. In this vision, cost-cutting is a temporary measure that does not imply the need to set up a specific and lasting organisation or employment relationship. Manufacturers also talk about cost-cutting when they constantly pressure suppliers to lower prices. This policy does not directly affect their organisation or force them to change their profit strategy. It does have one limitation, however, which is that manufacturers do need suppliers and must avoid making excessive demands if they wish to maintain their sourcing capabilities. There have been cases of manufacturers having to support suppliers financially after squeezing margins excessively. There is also the possibility that the balance of power between manufacturers and suppliers can be reversed. Where demand is strong, suppliers’ first priority is to support customers with long-term contracts, to the detriment of those who follow competitive sourcing practices. In 2005, for instance, Nissan’s Carlos Ghosn was forced to stop production in several plants due to a shortage of steel. Nor is it rare that quality and delay problems arise when suppliers come under too much pressure. It is clear that widespread ‘cost-cutting’ has little to do with the vision of lower costs as a priority source of profit in any and all circumstances – including in times of high demand – as is the case with the strategy referred to as a ‘permanent reduction in costs’. Toyota is the only carmaker to have succeeded in applying these latter principles, first with its patient construction of (and support for) an original productive model based on employees taking responsibility for reducing standard times – a mission they ultimately rejected in 1990 – before changing its ‘company government compromise’ to make cost-cutting acceptable again (Boyer and Freyssenet, 2002; Shimizu, 1999 and Chapter 4). For other carmakers, ‘cost-cutting’ has diverse connotations, encompassing the economies of scale achieved by increasing the percentage of parts shared by different car models. In this case, cutting unitary costs means extending the production runs of the vehicle’s constituent parts more than lowering each part’s return cost by reducing the time, materials, space and flows needed to produce it.
24
The Second Automobile Revolution
New accounting standards, which have the merit of forcing firms to break their operating margins down by profit and loss sources over the course of the year, seem nevertheless to maintain the confusion between cost-cutting with constant volumes and gains derived from commonalisation.
The disappearance of the ‘new economy’ shortly after its birth: the death of the ‘Anglo-Saxon model’ – how will the American Big Three fare? Alongside the enthusiasm for the first so-called ‘emerging’ countries, a new magnet for capital began taking shape in the United States before diffusing to the rest of the Triad. The new information and communications technologies (ICT) reached maturity and their generic nature seemed to offer considerable development prospects. Furthermore, the ICT sector seemed to offer a new productive paradigm, to wit, modular production, which would enable the modules’ externalised production as well as product customisation based on the modules being combined differently to match customer requirements, accompanied by a rapid home delivery of online orders. Capital flowed into the new sector, often in anticipation of double-digit returns. The majority of shareholders and the best-organised ones – pension funds, unit trusts, financial analysts and ratings agencies – began depicting these high returns as standards that all firms had to offer investors. This stringency was meant to force firms to adopt best practices and emphasise the most lucrative activities. Distributing stock options to executives and even to some employees was supposed to convince people of the advantages of the ‘new economy’. The ‘Anglo-Saxon model’ was born. The automobile industry was stigmatised as a symbol of the ‘old economy’, including within the sector itself. Some observers thought up a new ‘productive model’ for automobiles, one involving an externalisation of the least lucrative activities (particularly manufacturing); concentration on more profitable investments at both ends of the value chain (upstream on design and downstream on services like credit, leasing, rental, insurance, after-sales, maintenance, repair and accessories); reliance on market funding; viewing shareholder value as a strategic compass; and using acquisitions-divestments as a way to achieve price-maker status and generate capital gains. According to this model – one theorised before it ever existed – supply subsidiaries should be delegated most manufacturing functions and then sold off, with all capital ties being abandoned to force suppliers to compete with one another. This would supposedly help manufacturers distance themselves from the social problems that were recurring in their plants (conflicts, rising costs of social protection, contract wage bargaining, etc.). Few manufacturers ignored the new discourse, if only because they felt a need to reassure ‘markets’ and ratings agencies. Most were very careful,
Wrong Forecasts and Unexpected Changes 25
however. With the exception of Ford and General Motors, who made services their new strategic axis and sold shares in their respective supply subsidiaries (Visteon and Delphi) – or the Fiat group, which dreamt that becoming a European ‘General Electric’ would overcome its failures in the emerging market – manufacturers soon experienced the limitations, incoherence and dangers of applying the ‘Anglo-Saxon model’ to automobiles. All of the big players built up their customer services but few launched new activities like car insurance or consumer credit. E-business was no replacement for direct purchasing or sales relationships. As for shareholder value, this was portrayed for a while as a way of putting extra pressure on employees, although manufacturers generally did try to fund investments otherwise than through the stock market. Those carmakers that had stable core shareholders (founding families, state authorities, friendly firms, shareholder pacts, cross shareholding arrangements, etc.) would often integrate more closely with their partners to avoid hostile takeovers and raise their self-funding capabilities. This was the path followed by the three most profitable firms (Toyota, Honda and Volkswagen); although it is still uncertain whether the latter will be able to resist the combined onslaught of Porsche and the European Commission (Jürgens, Chapter 12, this volume). PSA and Renault-Nissan also hastened to secure their shareholder structure. Mitsubishi was ultimately refinanced by a group bearing the same name. General Motors and Ford, on the other hand, expended much more energy trying to remain masters of their own destinies. As for Chrysler, it was delivered to an investment fund by Daimler that, in legal terms, was supposedly an equal industrial partner (Belzowski, Chapter 11, this volume; Köhler, Chapter 16, this volume). In March 2000, prices of ‘new economy’ shares collapsed when the third big speculative bubble burst with the usual consequences for consumption and investment. The three carmakers that had put their faith in the AngloSaxon model, Ford, General Motors and Fiat, were seriously and durably damaged. The executives responsible for the strategic directions taken in the late 1990s were dismissed, sometimes after receiving golden handshakes. The new executives divested many financial and service subsidiaries to fund essential new models. They were also forced to bail out and restore ties to former supply subsidiaries on the brink of bankruptcy. The traditional series of plant closures and mass redundancies ensued (Mercer, Chapter 10, this volume; Senter and McManus, Chapter 9, this volume). Only Fiat has really recovered, thanks to support from its founding family, an orderly retreat to its core markets, a vigorous reshaping of its product range and substantial financial compensation from General Motors, unable to fulfil a contractual obligation to increase its equity stake in Fiat, as stipulated in the companies’ alliance agreement (Volpato, 2008). In 2008, there were major concerns about the future of the American Big Three, which only have a few months of cash left. Asset sales, which are
26
The Second Automobile Revolution
always very slow and would probably occur at knockdown prices in the current situation, are no panacea. General Motors and Ford are deploying considerable efforts to survive these dire straits, demanding and obtaining ever-greater sacrifices from employees. Chrysler is working all-out to define a more appropriate product policy. Yet the fallout from the fourth great financial crisis, the so-called subprime credit crunch, is overwhelming all three American firms. Excessive household debt in the US, mass foreclosures on housing owned by new owners wooed by easy credit, further devaluation of loan assets that banks have bought from the original mortgage lending institutions, credit restrictions, skyrocketing oil prices, a general economic slowdown and rising unemployment – all of these factors have led to a sharp fall in automobile demand, particularly for fuel-guzzling light trucks that have become too expensive for many households. This decline is especially serious for the Big Three, for whom light trucks had been a significant source of profit. Worst of all, GM, Ford and Chrysler, though they offer some small cars, lack the inexpensive and fuel-efficient compact models that are required to make an important impact in today’s markets.
The different forms of externalisation and manufacturer–supplier relationships Externalisation is a leitmotif that has convinced many people of its generality and uniformity. The basic principle is that manufacturer–supplier relationships are destined to undergo a number of radical changes as part of the shift from an era of prime contracting to one of ‘win–win’ partnerships. In reality, the integration rate of groups like Toyota, Volkswagen and PSA has barely changed since the mid-1980s. Thus, it is difficult to make the argument that increasing ‘externalisation’ is a condition for profitability. Conversely, there are some firms that are just as profitable as these three yet whose integration rates remain very low, like Honda since the beginning, or Renault since the 1990s. All in all, the different paths towards and objectives of externalisation deserve further analysis (Frigant, Chapter 22, this volume). The first instance of externalisation entailed troubled manufacturers in the 1980s trying to reduce debt levels by divesting ‘non-strategic’ subsidiaries, sometimes because this would also help them to disengage from social conflicts they had been unable to resolve. Then, in the 1990s, a number of companies tried to share with their suppliers the investments and risks they faced when extending their product ranges to encompass new vehicle types and/or when setting up operations in new countries. The latest example of externalisation occurred in the late 1990s, when this policy was often portrayed, as mentioned above, as an inherent trait of the ‘Anglo-Saxon’ model. In this recent version, externalisation meant disengaging from insufficiently profitable manufacturing activities and replacing
Wrong Forecasts and Unexpected Changes 27
them with design activities and above all marketing and services. At around the same time, some carmakers viewed this policy as a way to restructure the automobile branch around ‘modular’ vehicle design. First-tier suppliers were increasingly asked to specialise in the design and delivery of ‘modules’ assembled from components supplied by a host of subcontractors. Note, however, that modularisation was not always accompanied by externalisation. Indeed, on several occasions, the exact opposite occurred. For example, at a time when other carmakers were cutting all links to former subsidiaries, PSA used a series of acquisitions to build a major supply subsidiary (Faurecia) that would become world leader in six ‘modules’ (Fryessenet, Chapter 13, this volume). Nevertheless, manufacturers cannot afford to totally ignore their suppliers’ plight. To really work, a profit strategy must be pursued in a coherent manner not only by the manufacturer but also by all of the actors in its supply chain. It is impossible to pursue a ‘permanent reduction in costs’ strategy if suppliers are not committed to the same approach. This is why Toyota was so concerned with maintaining control over its main supply subsidiary, Denso, which had wanted to become independent. The same notion of supply chain coherence applies to other strategies. It is impossible to achieve economies of scale, ensure a product’s ‘social quality’ or launch innovative models without the assurance that one’s upstream partners are doing everything they can to commonalise products in the first case, ensure the ‘excellence’ and singularity of equipment and components in the second case, or be responsive to both the successes and failures of innovative models in the third case. Nissan, which sold almost all of its equity interests in suppliers under the direction of Carlos Ghosn, had to reverse its position on several occasions. Geo-political and technical secrecy considerations must also be taken into account. French carmakers feel the need, for instance, to ensure that local supply firm Valeo is in a position to resist the Anglo-American funds that have been trying to take it over and sell it off piece-by-piece. It remains that in many situations, the carmaker–supplier relationship is anything but partner-like. Competition remains a key variable, with certain manufacturers often setting a double-digit price cut as a precondition for any contract renewal. Some observers have hypothesised that the balance of power between large suppliers and manufacturers might switch if the former grew big enough. If carmakers were to delegate too much responsibility for their products, they could lose technical control – although nothing like that has happened yet. Manufacturers continue to be the only parties possessing sufficient knowledge to integrate components into an architecture that they alone are in a position to define. Of course, this does not mean that nothing will ever change. Electronisation and replacing internal combustion engines with electrical driving systems might upset this status quo.
28
The Second Automobile Revolution
Modularisation, electronisation and new paths towards commonalisation and innovation At present, ‘modularisation’ in the automobile industry involves more than just building large sub-assemblies. Unlike the modules used in electronic products, the ones that autos use are neither physically nor functionally unified. Thus, they do not feature the kinds of properties at a practical level (ease of assembly, interchangeability) or at an economic level (economies of scale versus variety of possible combinations) that would usually be part of this configuration. Automotive modules often fulfil several functions and have multiple and complex interfaces. So far in the automobile business, ‘modularisation’ can be interpreted as having been not an architectural choice for those who have adopted it but instead as a new ‘government compromise’ for the supply chain as a whole. This choice consists of delegating to firsttier suppliers, via such ‘modules’, the task of organising the supply chain, with all of the advantages that come with this (i.e. the ability to appropriate much of the value produced), in exchange for sharing the investments and the financial and social risks that the manufacturers either cannot or do not wish to assume. Of course, other firms (like Toyota) that have not faced the same problems, prefer to control their supply chains and still perform most assembly operations in-house. ‘Modularisation’, which remains a mainly metaphorical concept, might assume greater meaning one day. The electronised management of certain functions makes it possible to obtain very different levels of performance from a given piece of equipment, like an engine, gearbox or suspension unit. This enables a modulation of the power, consumption or responsiveness of the item in question, depending on the physical and commercial characteristics attributed to the model on which it is supposed to be fitted. This is a process still in its early stages but one that offers great possibilities, first in terms of commonalisation but further down the road in conceptual innovation terms. It could theoretically narrow the range of engine-propelled groups and equipment in general. Electronics can be used to compensate the degraded road performance of a given platform by turning it into a springboard for an innovative model that was not part of the original plans. The question is whether there will soon be, at least at some technical level, a degree of compatibility between commonalisation and conceptual innovation, something that General Motors was once able to achieve by associating volume and diversity – factors previously considered irreconcilable – when it invented the idea of a platform being shared by several models. Will carmakers pursuing a ‘volume and diversity’ strategy, whose current customer base of people with moderately hierarchised and foreseeable incomes has started to shrink, be able to use innovative models to attract customers with more variable incomes? For the moment, the dynamic constraints more than
Wrong Forecasts and Unexpected Changes 29
offset what might be done electronically. It is possible that certain innovative models, ones differing greatly from any of the others, will no longer be able to function using the existing platforms. This is not the most crucial consideration, however. In much the same way as the ‘volume and diversity’ strategy did not lead to the disappearance of the ‘volume’ and ‘diversity and flexibility’ strategies since each was based on different enabling conditions, a potential new ‘commonalisation and innovation’ strategy will not lead to the disappearance of the two strategies of which it is a hybrid (in the sense that a prototype has been created) – ‘volume and diversity’ and ‘innovation and flexibility’ – since there will always be countries where coordinated and moderately hierarchised incomes will not require manufacturers to design conceptually innovative models to be profitable. Inversely, there will also be countries where income distribution is based on merit and the balance of power, with this system being generalised to such an extent that the only useful and possible strategies will be ‘innovation and flexibility’ and ‘diversity and flexibility’. Lastly, even if electronically managed modules minimise the risks that an innovative model creates, the ability to cope with successes surpassing expectations, or else with total failure, will remain preconditions for profitability in such instances. Of course, this ability implies that productive organisations and employment relationships differ from the ones required by a ‘volume and diversity’ strategy. On the other hand, electronisation will facilitate the adaptation of models featuring certain specific conditions of usage in the new countries that companies are moving into, conditions that will not have been planned or considered when the project was first designed. This could reduce the negative impact of the markets’ geographic heterogenisation on the economies of scale. In the end, it can be said that electronisation reduces the rigidity of existing automobile architectures, without changing them. There is another more decisive development, however, that will modify the automobile architecture itself, to wit, the replacement of internal combustion engines by electric driving systems, be it using electricity stored in batteries or produced in the vehicle itself (fuel cells or another source). This will enable a functional miniaturisation and specialisation of engines. Once multiplexing has been mastered, this will reduce connectivity problems and enable the standardisation of interfaces. In turn, this will allow for a deployment of modularisation. It remains that a machine travelling at varying speeds on different road surfaces in changing traffic conditions in all kinds of weather and in different cultural and political contexts, when driven by people in very different ways, will never be as free of external constraints as a desktop computer might be. This means that modular combinations will always be limited. Nevertheless, electric driving systems constitute one of the ingredients of the pending second automobile revolution.
30
The Second Automobile Revolution
Towards a second automobile revolution The world is changing and so are automobiles and their industry. The rapid and apparently durable take-off of the BRIC continent-sized national economies, the increasing expense of extracting raw materials, the declining US dollar and market speculation – all of these factors have led to a nonstop rise in petrol prices since 2005 until the crisis. The unpredictability of oil prices is having market effects. The demand for automobiles has contracted significantly in certain countries and been displaced towards more fuel-efficient models in others. The conviction (widely shared now by most governments) that it is urgent to reduce greenhouse gases and all emissions creating public health problems seems to have finally convinced carmakers that they can no longer continue to fight yesterday’s battles by trying to stave off pollution standards. The race towards alternative driving systems has started. Moreover, China and India, which have no oil, will have a major role to play in the search for solutions, even if they still refuse to accept any limitations on their fuel consumption. A major redistribution of power amongst industrial and national actors may well be on the cards. The conclusion of this chapter is devoted to an analysis of the conditions and consequences of this revolution.
Increasing forecast accuracy and suggesting more suitable orientations The inaccuracy of many forecasts and recommendations, along with the upheavals created by so many unexpected changes, have devalued certain ways of analysing the past and reasoning about the future. The most common method of offering forecasts and suggestions consists in first identifying a few new contextual trends, extrapolating and generalising them and then seeking which practices provide good results in the alleged new context so that these practices can be offered to firms that are in trouble. This modus operandi can lead to veritable ‘theoretical bubbles’ (which are sometimes not unrelated to speculative bubbles as was the case in the ‘new economy’) when ‘best practices’ are amalgamated and systematised in a model that will supposedly guarantee prosperity for all and revolutionise the world. At least, this was the naïve belief of the promoters of lean production (Womack et al., 1990). Every trend has enabling conditions that restrict its diffusion in time and space. However, it is not sufficient to identify such trends; their conditions of emergence, generalisation and disappearance must also be established if they are to help people to think usefully about possible futures. Furthermore, their homogeneity must also be verified. Phenomena featuring an at best formal similarity at a given moment in time, and which actually possess a host
Wrong Forecasts and Unexpected Changes 31
of different meanings, are often inaccurately grouped in one and the same tendency. The so-called scenarios method has the merit of not assuming that the future is already mapped out. However, it often occurs that the scenarios on offer are ranked from worst to best in terms of their interest for the people ordering the study, and that the means publicised merely suggest how to avoid the worst and approximate the best! When the method is used more seriously, it can reveal links and possible interactions between facts and decisions in the absence of prior determinism. In this case, forecasting consists of possible and desirable actions in different instances, given the outlooks of the actors concerned. GERPISA’s studies have led to a similar method, except for the fact that the actors involved are not atoms subject to variable and contradictory forces but actors apprehended through the prism of the social relationship that connects them but also constitutes them (i.e. in the case of a capitalist company, making sufficient profits to ensure the survival both of capital and labour). They are also apprehended in light of their potentially contradictory perspectives: for some, expanding the basis upon which capital is accumulated by investing increasing amounts of human activities, in the name of efficiency; for others, freedom from a social relationship that puts them in a position of subordination to others, in the name of equality. The question then becomes under what conditions sufficient profits can be made without ruining each actor’s prospects. Studies of automaker trajectories have shown that there are several ways for capitalist firms to make profits; that not all can be used in every context; and that to be applied, each requires coherent means. Firms seem to be profitable under two major conditions. First, their chosen profit strategy (out of six possibilities: economies of scale, product diversity, quality, innovation, productive flexibility and reduction of costs with constant volumes) must be possible and viable in the countries where they are deploying their activities, given the strategy and growth models found in these settings. Second, the means employed (product policy, productive organisation and employment relationship) must be coherent with the strategy being pursued, compatible with one another and acceptable temporarily to the main actors in the company (Boyer and Freyssenet, 2002). Insofar as these are appropriate ways to envisage general conditions of profitability, thinking about the future must be rooted in two types of studies: one category that updates possible profit strategies depending on the differential evolution of strategies and of national or regional growth models; the other that outlines possible government compromises between the main actors in a firm (shareholders, banks, executives, employees, trade unions, suppliers, state authorities, etc.) as well as the means used to implement whichever of the chosen profit strategies is viable.
32 The Second Automobile Revolution Table 2.1
Compatibility between firms’ profit strategies and certain current trends in the
Profit strategies
Offering all types of vehicles
Shared platforms
Externalisation
Modularisation
Quality
No, only in the upper segments of the traditional product range
No, there is an absolute need for ‘distinction’, including in the non-visible aspects
Maintain the know-how that gives the project its image of ‘social quality’
No, commercial need for a highly integrated product
Diversity and flexibility
Yes, possible, but not indispensable
No, the demand is too ‘balkanised’
Yes, to have the necessary flexibility
Yes, possible for certain customer bases
Volume
No, only models in the main market segments
No, standard and ‘integrated’ model
No, need for continuous flows
No, insufficient diversity
Volume and diversity
No, only models whose platforms can be shared
Indispensable, diversity on ‘surface’ level
Yes, but by maintaining control over volumes and prices
Yes, in the form of sub-assemblies for the moment, and if externalisation conditions are fulfilled
Permanent reduction in costs
No, only models for which demand can be forecast relatively easily
Yes, if the demand is moderately hierarchised
Yes, but by maintaining technical control and if there is the same profit strategy and productive model
Yes, carefully, in the form of subassemblies, and if externalisation conditions are fulfilled
Innovation and flexibility
No, only conceptually innovative models
Difficult, when the aim is to respond to new expectations
Indispensable for rapid mass production if successful, or for the model’s withdrawal in case of failure
Yes, but only for research and innovative models
Source: R. Boyer and M. Freyssenet (GERPISA).
In theory, corporate actors can affect the strategy and national growth model, the choice of a firm’s ‘profit strategy’ and the contents of its ‘company government compromise’. However, their ability to act at these three levels is very unequal. This is because countries are not totally free to decide their strategy and growth model. Their choices depend on their resources, their history and their relations with other countries, particularly the hegemonic state, during the period in question. Growth based on consumption and on a
Wrong Forecasts and Unexpected Changes 33 automobile industry Clean engine, recyclable and silent car
Internationalisation
Merger-acquisitionalliance
Development of customer services
Yes, but not before finding a reliable and lasting solution
Indispensable, but only at a commercial level, production in the country of origin
Incompatibility with firms pursuing very different profit strategies
Yes, part of Possible, if indispensable and investors are highly customised seeking safety service
Yes, for certain customer bases
Not necessary, possible in countries with a competitive income distribution
Possible with a firm pursuing the same strategy but not indispensable
Yes, this is a condition for capturing certain customer bases
Yes, when there is a possibility of synthesising innovations and rapid price reductions
Indispensable, if national market is saturated, move towards countries with coordinated and egalitarian distribution
Very profitable if Standard service, there is a possibility minimum of adopting the indispensable same models
No, resistance to lower prices in case of strong demand
No, but very rapid copying if there is sufficient demand and economies of scale are possible
Indispensable, if the national market is saturated, towards countries with a coordinated and moderately hierarchised distribution
Vital where a second stage renewal market is involved; absolute need for shared platforms
Yes, if stable returns are expected over the long run
No, but quick copies becomes the new norm
Very profitable if possible, move towards countries with predictable demand and stable exchange rate, but not indispensable
Not advisable, since Yes, to ensure it is hard to find a customer loyalty firm with the same strategy and company government compromise
No, too costly, need for self-funding
Not necessary, possible with firm pursuing the same strategy
No, risk of losing financial independence, indispensable for innovation
Yes, at each stage Not necessary if located of innovation in a region marked by competitive distribution, indispensable elsewhere
Yes, when economies of scale can no longer be achieved with material products
No, unless it constitutes an innovative concept
Reliance on financial markets
Yes, possibility of high albeit variable returns
coordinated and moderately hierarchised distribution of national income is only feasible if customs barriers or some other structural advantage shelter the country from more competitive external products. At the same time, a nation has to accommodate other countries when defining its own customs regime. The determination of a growth mode, which conditions the suitability of different profit strategies, i.e. the first condition for corporate profitability, therefore occurs at a level where the actors of the firm have very little control. In fact, it might be much harder for them to preserve or change a national
34
The Second Automobile Revolution
growth mode to ensure that it becomes or stays appropriate for their profit strategy, than it would be for them to change their profit strategy to make it more suitable. At the same time, actors cannot afford to remain passive. Each of them, given their own objectives and vision of the future, may prefer the preservation or implementation of one or the other modes, depending on how they envision national independence or the distribution of wealth. There are historical circumstances where action is not only possible but necessary. One example is after a war or, as between 1974 and 1980 (and probably once again at the start of the twenty-first century), times when previous growth modes have been destabilised or when several options exist. Corporate actors can then theoretically choose amongst the different profit strategies enabled by the growth modes in the spaces where their firm operates. Of course, this choice is far from free and depends on the profit strategy already in effect within the company, and on the profit strategies that other manufacturers are pursuing. This is because a change in the profit strategy cannot simply be ordered. First there needs to be a reconstruction of the company government compromise in terms of the means employed, thus a commitment to a long and potentially conflictual process whose outcome is uncertain. Things also depend on how much space the competition gives the company to make its choices. It could be risky to adopt the same strategy as rivals who already occupy a certain square on the chessboard and have already built a solid company government compromise that creates synergy amongst the means they use. What remains is to find a profit strategy, amongst those enabled by the national growth mode, that is not completely taken. Of course, there are also circumstances where the invention of a new strategy is possible and even necessary, since this could ensure the compatibility of profit sources that used to be considered contradictory. One example is the way that General Motors created compatibility between volume and diversity during the interwar years. Clearly, it is in the construction of the company government compromise that corporate actors have the greatest room to manoeuvre. Generally, the corporate executives, sometimes operating under time constraints (notably during crisis periods), affirm that no solution exists outside of the one that they themselves have designated. Analysis of automobile firms’ trajectories over the past century allows us to say that this is the best way of never achieving long-term profitability. All of the company government compromises that have led to the invention or adoption of a productive model have required at least ten years – and highly intelligent partners – to create a situation in which no one is forced to deny his/her own convictions and objectives. Since the development of the aforementioned analytical schema, we have regularly tried to forecast how the different firms in this sector will be affected. Readers who have tracked our different publications since 1998 will have
Wrong Forecasts and Unexpected Changes 35
noted that these forecasts or questions have been reasonably good, or in any case more relevant than many rival analyses. Table 2.1 offers an evaluation of the suitability of certain recommendations made in light of different corporate profit strategies. It is one of the tools that we have developed to help us think about the future. Translation by Alan Sitkin. Acknowledgement to Richard Senter for his useful comments. Bibliography Belzowski, B. (1998) ‘Reinventing Chrysler’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Berggren, C. (1998) ‘A Second Come-back or a Final Farewell? The Volvo Trajectory’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Bordenave, G. (1998) ‘Globalization at the Heart of Organizational Change: Crisis and Recovery at the Ford Motor Company’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Bordenave, G. and Lung, Y. (2003) ‘The Twin Internationalization Strategy of US Automakers: GM and Ford’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Boyer, R. and Freyssenet, M. (1999, 2006) Le Monde qui a changé la machine, un siècle d’histoire automobile, http://freyssenet.com Boyer, R. and Freyssenet, M. (2000) ‘Fusions-acquisitions et stratégies de profit’, Revue française de gestion, 131 (November–December): 20–8. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Boyer, R., Charron, E., Jürgens, U. and Tolliday, S. (eds) (1998) Between Imitation and Innovation: the Transfer and Hybridization of Productive Models in the International Automobile Industry. Oxford and New York: Oxford University Press. Carillo, J., Lung, Y. and van Tulder, R. (eds) (2004) Cars . . . Carriers of Regionalism? Basingstoke and New York: Palgrave Macmillan. Charron, E. and Stewart, P. (eds) (2004) Work and Employment Relations in the Automobile Industry. Basingstoke and New York: Palgrave Macmillan. Eckardt, A. and Klemm, M. (2003) ‘The Internationalization of a Premium Automobile Producer: the BMW Group and the Case of Rover’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Flynn, M. (1998) ‘The General Motors Trajectory: Strategic Shift or Tactical Drift?’ in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Freyssenet, M. (2008) ‘Stratégies et modèles nationaux de croissance. Proposition d’une démarche d’analyse et esquisse d’un schéma d’analyse’, La Revue de la Régulation, 3: 21–42.
36
The Second Automobile Revolution
Freyssenet, M. and Mair, A. (2000) ‘Le modèle industriel inventé par Honda’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), Quel modèle productif? Trajectoires et modèles industriels des constructeurs automobiles mondiaux. Paris: La Découverte. Freyssenet, M., Mair, A., Shimizu, K. and Volpato, G. (eds) (1998) One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Freyssenet, M., Shimizu, K. and Volpato, G. (eds) (2003a) Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Freyssenet, M., Shimizu, K. and Volpato, G. (eds) (2003b) Gloablization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan. Hanada, M. (1998) ‘Nissan, Restructuring to Regain Competitiveness’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Humphrey, J. and Salerno, M. (2000) ‘Globalisation and Assembler–Supplier Relations: Brazil and India’, in J. Humphrey, Y. Lecler and M. Salerno (eds), Global Strategies and Local Realities: the Auto Industry in Emerging Markets. Basingstoke and New York: Palgrave Macmillan and St. Martin’s Press. Humphrey, J., Y. Lecler and M. Salerno (eds) (2000) Global Strategies and Local Realities: the Auto Industry in Emerging Markets. Basingstoke and New York: Palgrave Macmillan and St. Martin’s Press. Jetin, B. (2003) ‘The Internationalization of Automobile Firms: a Statistical Comparison with the European Companies’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Jullien, B. (2006) ‘Les inégalités de revenus et leurs effets sur les demandes automobiles’, in B. Jullien and Y. Lung (eds), Are Automobile Firms Market-Oriented Organisations? Myths and Realities, Fourteenth GERPISA International Colloquium, CD-Rom, Paris: GERPISA. Jürgens, U. (1998) ‘The Development of Volkswagen’s Industrial Model’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Kim, Y.-H. (2004) ‘The Indian Passenger Car Industry and the South Asia Market: Global Auto Companies’ Struggles in India’, in J. Carillo, Y. Lung and R. van Tulder (eds), Cars . . . Carriers of Regionalism? Basingstoke and New York: Palgrave Macmillan. Kumon, H. (2003) ‘Nissan: From a Precocious Export Policy to a Strategic Alliance with Renault’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Lung, Y., Chanaron, J. J., Fujimoto, T. and Raff, D. (eds) (1999) Coping with Variety: Product Variety and Production Organization in the World Automobile Industry. Aldershot: Ashgate. Pardi, T. (2005) ‘Where Did It Go Wrong? Hybridisation and Crisis of Toyota Motor Manufacturing UK, 1989–2001’, International Sociology, 20(1): 93–118. Ramirez, S. (2007) ‘Public Policies, European Integration and Multinational Corporations in the Automobile Sector: the French and Italian Cases in Comparative Perspective, 1945–1973’, unpublished thesis, European University Institute, Florence.
Wrong Forecasts and Unexpected Changes 37 Shimizu, K. (1998) ‘A New Toyotaism?’ in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Shimizu, K. (1999) Le toyotisme. Paris: La Découverte. Shimizu, K. (2003) ‘A Maverick in the Age of Mega-Mergers? Toyota’s Global Strategies’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Shimizu, K. and Shimokawa, K. (1998) ‘The Unique Trajectory of Mitsubishi Motors’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Shimokawa, K. (2003) ‘Honda, an Independent Global Automobile Company, Out of the “Four Million Units Club”’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Thun, E. (2004) ‘Going Local: Foreign Investment, Local Development and the Chinese Auto Sector’, in J. Carillo, Y. Lung and R. van Tulder (eds), Cars . . . Carriers of Regionalism? Basingstoke and New York: Palgrave Macmillan. Volpato, G. (2008) Fiat Group Automobile. Un’Araba Fenice nell’industria automobilistica internatiozionale. Bologna: Il Mulino. Womack, J. P., Jones, D. T. and Roos, D. (1990) The Machine that Changed the World. New York: Palgrave Macmillan.
3
Strategies of Internationalisation of Automobile Firms in the New Century: a New Leap Forward? Bruno Jetin
Introduction The automobile industry presents a paradoxical situation. It has a century-old tradition of internationalisation. Together with electronics/electrical equipment and petroleum exploration and distribution, it dominates the list of the top 100 largest transnational corporations ranked by their foreign assets established by UNCTAD, with twelve automobile firms in 1990, and still eleven firms in 2005. But despite this strong presence, these automobile firms were initially much less internationalised than the average. In 1990, the Transnational Index (TNI) calculated by UNCTAD1 was only 36 per cent in the motor vehicle industry against 51 per cent for all industries, well below the TNI of electrical and electronics, petroleum and chemicals. One of the reasons was that the largest firms such as GM and Ford, despite their strong presence in Europe and Latin America, realised at the time the bulk of their activities in the USA which was and still is the biggest automobile market in the world. Japanese firms had started to produce abroad but were still relying a lot on direct exports. At the other extreme, Swedish firms were long ago highly internationalised because of their very narrow domestic market but accounted for a small share of world production. Previous studies (Freyssenet et al., 2003; Lung and van Tulder, 2004) have acknowledged this situation. During the 1990s which saw the emergence of the globalisation process, no automobile firm had been able to build a significant presence on every continent at the same time, but only on one or two (Jetin, 2003). Firms lack financial capacity to invest everywhere; cannot extend their portfolio of products to fit every consumer’s needs; do not have the necessary manpower to organise, produce and sell efficiently; and sometimes lack the necessary political connections to develop profitable activities everywhere. They have to make choices between a determined set of countries and continents. Usually, the region of origin is at the top of the list of priorities to the point that at the end of the 1990s, regionalisation was better suited than globalisation to describe the process of internationalisation 38
Strategies of Internationalisation of Automobile Firms
39
at this stage. The question remained whether ‘regionalisation was a substitute for globalisation or an intermediate step towards its realisation’ (Lung and van Tulder, 2004: 4). Now the situation seems to be changing. By 2005, the TNI for the 11 automobile firms listed in the UNCTAD ranking reached 56 per cent, close to the all-industries average of 60 per cent. This means a 55 per cent increase from 1990 to 2005, against an 18 per cent increase for the all-industries average. These firms seem to have made a real leap towards globalisation. The first purpose of this chapter is to analyse this internationalisation push by extending the sample to twenty-two major automobile firms and carefully assessing their geographical reach to see whether there is a real globalisation process. We use a statistical approach that tries to measure the degree of internationalisation of automobile firms for the period 2000–5.2 In so doing, we refer to the analytical framework of the ‘productive models’ (Boyer and Freyssenet, 2002) according to which there are two conditions to sustainable profitability: (1) a relevant profit strategy and (2) the establishment of a ‘company–government compromise’ which establishes a coherency between certain requirements that the firm’s players (executives, shareholders, banks, employees, trade unions, suppliers, the state) must satisfy. Because this ‘company–government compromise’ is first established at the national level, and continues to play a vital role in the firm’s survival, we still consider the national level as the starting point of our investigation. We then turn to the regional level to assess the relative importance of intra-regional and extraregional activities of the firm. The implicit assumption is that a global firm must develop an important part of its activity outside its region of origin. The second purpose of this chapter is to study the profit that firms make abroad. Due to data availability this second part will be limited to American and Asian firms. But this is still sufficient to provide answers to some questions: Do firms succeed in making big profit abroad, or do they register losses ? Or in other words, is producing and selling abroad a bonanza or a curse? We do not pretend to judge definitely whether or not a profit strategy linked to a productive model is successful because a lot of different factors are involved in explaining the year-on-year profit and loss. The following chapters that analyse in-depth some of the firms under review will bring more comprehensive explanations. But at least we will provide some indications. Over the long term, a successful profit strategy should materialise in substantial profits. To start with, we present two features linked to the internationalisation process: mergers and acquisitions and the trend towards increasing concentration at the world level.
An unstable world oligopoly The automobile industry has long been a highly concentrated industry. But this does not mean there is a linear trend towards an ever-increasing
40
The Second Automobile Revolution
concentration of world production. Indeed the evolution is more complex. The world production share of the top twenty firms does show an increase from 91 per cent in 1985 to 96 per cent in 1999. But in 2006 it came back to 92 per cent thanks to the emergence of new competitors from China and India (author’s calculation based on CCFA data). The biggest firm’s share reveals a steady decline from about 20 per cent of world production in 1985 to 13 per cent in 2006. But the ten biggest firms’ share has increased from 73 per cent in 1985 to 82 per cent in 1999 and then decreased to 79 per cent in 2006. This contrasted evolution can be explained by two major facts: the impact of the wave of cross-border mergers and acquisitions by the end of the 1990s and the emergence of new competitors. In the 1990s, the search for economies of scale and scope at the world level, access to new markets and the need to diversify the portfolio of products led firms to increase their size either by giving priority to internal growth (e.g. Volkswagen and Toyota), or to mergers and acquisitions (e.g. BMW with Rover, Daimler with Chrysler in 1998, Ford with Volvo in 1999 and Rover in 2000), by building a network of partnerships supported by stake ownership (GM with Isuzu, Subaru and Suzuki), or by an unprecedented form of alliance (Renault group with Nissan in 1999). This resulted in a higher degree of concentration of world automobile production in the top ten firms noted above. Since 2005, most of these cross-border mergers and acquisitions have failed while less spectacular national acquisitions (for instance Toyota with Daihatsu and Hino, or Hyundai and Kia) and cross-border acquisitions (Renault with Dacia and Samsung motors, or GM with Daewoo) have proved successful together with pure internal growth strategies like Volkswagen, PSA and now Honda. This explains the small decline of the production share of the top ten firms from 82 per cent in 1999 to 79 per cent in 2006. It is expected that the sale of Jaguar, Land Rover and Volvo cars by Ford, and most of all, of Chrysler by Daimler, will reduce further the top ten’s share closer to its 1985 level (73 per cent). Among them, three firms (PSA, Honda and Hyundai) that have pursued successfully an endogenous growth have increased their weight. Whatever the strategy pursued, there is a clear tendency towards higher volumes of production. In 1985, there were three firms (GM, Ford, Toyota) producing between 4 and 9 million units per year at the world level, representing 42.2 per cent of world production that year (author’s calculation based on MVMA and CCFA data). Since 1999, there have been six firms in this volume range, two of them – GM and Toyota – producing around 9 million units. These six firms control 60 per cent of world production, down from 65 per cent in 1999. In 1985 and 1999, there were no firms producing from 3 to 4 million units. In 2006, there were three (PSA, Honda and Hyundai) with a share of 16 per cent of world production. Below the 3 million units, the number of firms has either stayed the same or has declined. The combined share of world production of these smaller firms
Strategies of Internationalisation of Automobile Firms
41
amounted to 16.1 per cent in 2006, down from 48.7 per cent in 1985. In the automobile industry where economies of scale give a decisive competitive advantage, this means that smaller firms will either have to specialise further in segmented products (BMW or Subaru) or increase dramatically their production volume. New competitors have chosen the second option: one Indian, Tata (Telco), and two Chinese firms (Chana Automobile and FAW Group) are now part of the top 20 world producers, along with one Russian, Avtovaz. Other Chinese producers are following closely. If the Indian, Chinese and Russian markets maintain a high growth or stabilise at a high level, these local producers will probably increase their volume of production and be able to contest the dominance of foreign producers in their home markets. Growth in their home market will probably not be enough to reach an efficient minimum level of production and this explains why Indian and most of all Chinese firms have already starting buying foreign firms and exporting massively as a first stage before direct investment abroad like Tata Motors in Thailand for instance. If these strategies prove successful, the consequence will be a further decentralisation of world production and a much more competitive market structure blocking the way to a stabilised world oligopoly that could share the world market and enjoy high profit margins.
A higher degree of internationalisation The largest automobile firms are now more internationalised. Table 3.1 describes this internationalisation push by presenting four indicators of internationalisation and our own calculation of the TNI for twenty-two major automobiles firms for which data are available. Automobile firms do not always produce cars only but also other manufactured products (motorbikes, marine and aircraft engines, trains) and services, in particular financial services. The process of internationalisation involves this whole set of activities and not only automobiles. In some cases like Fiat and Man, this aspect is especially important because these industrial groups are highly diversified and do not publish independent financial reports for their automobile branch. When automobile firms belong to an industrial conglomerate, like most Japanese producers, but publish independent financial reports, we have used them to trace more precisely the automobile segment. Our approach combines several indices that reflect different aspects of the internationalisation process. The first indicator measures the proportion of ‘total revenues’3 realised outside the country of origin. This makes it possible to assess the firms’ degree of what we call ‘commercial internationalisation’. It allows us to break world sales and revenues down by place of sale regardless of where the product or service was produced (with the exception of Hyundai where the location of production is the only criterion available). Three other indicators are related to different aspects of the productive process and define what we call the degree
42
The Second Automobile Revolution
Table 3.1 Comparison of the degree of internationalisation of automobile firms, 2000–2005, in % of total, and synthetical index Firms
Degree of internationalisation of automobile firms, 2000–2005
Synthetical index internationalisation
Commercial revenues
Production
Workforce
Total assets
Global
UNCTAD
AMERICAN FIRMS Ford General Motors Navistar Paccar Average
37 29 15 51 33
51 53 30 55 52
53 47 n.a. n.a. 50
39 32 14 56 36
45 40 n.a. n.a. 43
43 36 n.a. n.a. 40
EUROPEAN FIRMS BMW DaimlerChrysler Fiat Group Man Group PSA Renault Scania VW Volvo Average
75 84 67 73 66 63 94 72 92 76
30 74 46 38 46 44 79 62 84 56
25 50 53 35 37 39 58 48 63 46
64 75 59 33 43 35 30 50 63 50
48 71 56 45 48 45 65 58 76 57
55 70 60 47 48 46 61 57 73 57
JAPANESE FIRMS Honda Isuzu Mazda Mitsubishi Nissan Subaru Suzuki Toyota Average
79 60 65 71 70 54 58 65 65
57 53 27 51 52 19 43 39 43
52 10 8 36 37 52 36 35 33
69 18 18 41 47 23 34 50 38
64 35 35 50 52 37 43 47 45
67 30 30 49 51 43 43 50 45
KOREAN FIRM Hyundai
34
6
8
12
15
18
Notes: The Global Synthetical Index is an average of the Commercial Revenues, of Production, of Total Assets, and of the Workforce Indices. UNCTAD’s Synthetical Index is an average of the Commercial Revenues, Workforce and Total Assets Indexes. The American average is calculated with Ford and General Motors only. n.a. = not available. For details, see the statistical annex.
of ‘productive internationalisation’. The second focuses on automobile production in itself. It measures the proportion of complete vehicle production carried out outside the country of origin. The third indicator measures the proportion of workforce employed outside of the country of origin and the
Strategies of Internationalisation of Automobile Firms
43
fourth, the proportion of total assets owned outside of the country of origin. To give a synthetic view of the degree of internationalisation of automobile firms, we present UNCTAD’s transnational index, the average of commercial revenues, workforce and total assets indices and what we call the ‘global’ index which also includes the automobile production indices. One may consider the 50 per cent level as a qualitative threshold beyond which a firm can be considered as truly global in the sense that foreign activities have gained more importance than the national ones. With an average of 57 per cent, European firms appear as the most globalised, in front of Japanese firms (45 per cent) and American firms (43 per cent), which are on an equal footing. These results are strongly influenced by some structural features such as the weight of the country of origin in the world economy. Since the USA is by far the biggest market in the world, it is logical that its relative importance for American firms should be high and the relative importance of foreign markets low. The relatively low level of internationalisation of Ford and GM reflects also their incapacity to get a dominant share in Europe although they have been producing there for more than eighty years. The competition from local firms in Europe, Japan and in emergent countries is a natural limit to the level of internationalisation that a firm can reach. Nonetheless, internationalisation has progressed. The global index has increased from 36 per cent to 43 per cent for American firms, from 51 per cent to 57 per cent for European firms, and from 42 per cent to 45 per cent for Japanese firms between 1995–9 and 2000–5 (author’s calculations and Jetin, 2003). A closer look at commercial and productive internationalisation reveals contrasting situations.
Commercial internationalisation On average, the commercial internationalisation of American firms has increased rather from 26 per cent in 1995–9 (see Jetin, 2003: 15) to 33 per cent in 2000–5 (see Table 3.1, first column). Paccar, an American truck producer which has successfully bought Leyland trucks and DAF is the most internationalised (51 per cent) due to its strong presence in Europe, Mexico and Australia. GM and Ford are still very dependent on the US market but a closer look at the geographical breakdown of their sales and revenues over time shows that things are changing (Table 3.2). For both firms, the importance of the USA and even North America is clearly decreasing, especially during 2005–6 due to a decline in absolute numbers and market share to the benefit of foreign competitors. In GM’s case, this decline is compensated by a strong progress in emergent countries especially in Asia/Pacific where the share of sales and revenues has jumped from 1 per cent to 7 per cent from 1995–9 to 2005–6. The relative share of Europe has remained the same. In Ford’s case, the US and North American shares have plummeted. For the first time in its secular history Ford is on the verge of
44
The Second Automobile Revolution
Table 3.2 Geographical breakdown of General Motors and Ford Automobiles net sales and financial revenues, in percentage Period
USA
North America
Europe
Latin America
Asia/ Pacific & Africa
Latin America & Asia/Pacific & Africa
Total outside North America
GM 1995–1999 2000–2004 2005–2006
71 73 63
77 80 72
18 15 17
4 3 4
1 2 7
5 5 11
23 20 28
Ford 1995–1999 2000–2004 2005–2006
69 65 53
75* 71* 59
20 23 30
n.a. n.a. n.a.
n.a. n.a. n.a.
5* 5* 11
25* 29* 41
Note: * estimation based on a 6% share for Canada and Mexico. Source: Author’s calculations based on financial report data. Figures have been rounded.
making this decline partially offset by a progress in Europe whose turnover has increased from US$32 billion in 2000 to US$50 billion in 2006 thanks to both the Ford brand and Premier Automobile Group (PAG, Aston Martin, Jaguar, Volvo and Land Rover). The sale of these brands in 2006 and 2007 will be probably be followed by a return of Europe’s share to its historical level of around 20 per cent. The case of European firms is different. They were already realising more than half of their sales and revenues outside their country of origin. Still, their commercial internationalisation has increased from an average of 72 per cent in 1995–9 to 76 per cent in 2000–5. The detailed geographical structure of European firms for the most recent period is presented in Table 3.3. We can see that for all firms except DaimlerChrysler and Volvo, Europe still represents a very high share, in the range of 60 per cent (BMW) to 85 per cent (PSA) of total net sales and revenues in 2005–6 but with a downward trend. Without Chrysler, Europe’s share in Daimler’s turnover would have reached 48 per cent instead of 32 per cent, which is the lowest share of all European firms but stresses the ongoing importance of the regional market. This is in fact quite similar to the share of North America for Ford (59 per cent) and GM (72 per cent) for the same period. If we consider the region of origin as the new domestic space of the firm, and globalisation as expanding outside the region of origin, then we can conclude that globalisation is far from being achieved. For some firms like Man, Scania or Volkswagen, there is even a kind of stagnation as if a threshold was being reached. Fiat is a special case. As a diversified industrial, Fiat Group has increased its commercial
Strategies of Internationalisation of Automobile Firms
45
internationalisation from 63 per cent in 1995–9 to 72 per cent in 2005–6. But the automobile branch has suffered a severe crisis over the period and, as a consequence, its commercial internationalisation has decreased from 59 per cent in 1995–9 to 55 per cent in 2000–4. As far as geographical diversification is concerned, all firms, except the French ones and Scania, have a foothold in North America where they realise between 13 per cent (Fiat) and 29 per cent (Volvo Trucks). In the Fiat’s case again, this is mainly due to CNH-Case New Holland’s turnover, its agricultural and construction equipment branch. Since the split with Chrysler in 2006, Daimler returned to more common standards. North America now accounts for 28 per cent of Daimler’s turnover instead of 51 per cent for DaimlerChrysler (Table 3.3). Six among nine European firms are present in South America but only Fiat and Scania realise at least 10 per cent of their sales there. Likewise, seven European firms have a presence in Asia/Pacific but only three (if we include Volvo trucks) realise 10 per cent of their sales and revenues there. Africa accounts for 5 per cent (Man) or less. In general, the importance of the Asia/Pacific region is growing under the influence of China, while that of South America is declining. Japanese firms present an intermediate level of commercial internationalisation. On average, it has increased from 57 per cent in 1995–9 to 65 per cent in 2000–5. Even if it remains below the European level (77 per cent) Japanese firms are filling the gap with an increase of 14 per cent against 7 per cent for European firms. This is a big change for a number of Japanese firms, including Toyota, which sold around half of its production abroad at the end of the 1990s. Honda with 79 per cent is still the most internationalised of Japanese firms followed by Mitsubishi and Nissan which have reached the 70 per cent level after a leap between the two periods. In the case of Mitsubishi, the reason is due to a fall in Japanese sales after several quality scams shook consumers’ confidence, while in Nissan’s case it is due to a stabilisation in Japan and a strong progression of foreign sales. Toyota, which was one of the least internationalised firms by the commercial criteria at the end of the 1990s (54 per cent) now realises 65 per cent of its sales and revenues outside Japan in line with the increase in the Japanese average. The geographical breakdown (Table 3.4) reveals that only Isuzu (63 per cent) and Suzuki (55 per cent) are focused on Asia, which is explained by their stronghold in Thailand and India respectively. In contrast to American and European firms which are still very dependent on their region of origin, Japanese firms are usually more dependent on extra-regional markets, in particular North America. Five among nine sell more in value terms in North America than in Japan. Honda is the most extreme case which now realises 55 per cent of its sales and revenues in North America against 16 per cent in Japan. Only three firms (Mazda, Mitsubishi and Suzuki) realise more than 20 per cent of their turnover in Europe, although the European share is increasing for all Japanese firms.
46
The Second Automobile Revolution
Table 3.3
Geographical structure of European firms’ net sales, in percentage
Period
Country of origin
BMW 1995–9 2000–4 2005–6
28 26 23
Europe
North America
68 58 60
18 28 24
South America
Asia/ Pacific
Africa
9 10 12
5 3 4
Daimler-Benz (1990–96) DaimlerChrysler (1997–2006), Daimler ( ) 1995–6 37 65 19 4 8 1997–9 19 35 56 3 4 2000–4 16 32 56 2 5 2005–6 14 (22) 32 (48) 51 (28) 2 8 (12) Fiat Group, Fiat Auto ( ) 1995–9 37 (41) 76 2000–4 34 (45) 73 2005–6 28 68
7 13 13
12 7 10
Man Group, Commercial vehicles ( ) 1995–9 33 67 15 2000–4 27 68 14 2005–6 27 (30) 73 7 PSA 1995–9 2000–4 2005–6
39 34* 33*
93 89 85
Renault 1995–9 2000–4 2005–6
42 37 34
85 84 80
Scania 1995–9 2000–4 2005–6
10 7 6
75 79 76
Volkswagen 1995–9 35 2000–4 28 2005–6 27
74 70 72
Volvo trucks 1995–9 10 2000–4 8 2005–6 6
56 56 52
4 3 5 6 6 5 5
13 13 14
5 7 10
6 5 5 7 8 10
3 4 10
Other countries
4
1 16
5
6
16 9 12
5 6 6
10 18 14
10 5 6
4 6 6
29 28 29
6 3 5
6 8 9
9 5 5 6 2 1 2 3 5 5
Note: ‘other countries’ means all countries for which detailed data are not available. In the case of BMW, for instance, it means South America and Africa. For DaimlerChrysler it means Africa, and so on. Source: calculated by the author based on company reports.
Strategies of Internationalisation of Automobile Firms Table 3.4
47
Geographical breakdown of Asian firms’ sales and revenues, in percentage
Years
Japan
Asia
Total Asia
North America
Europe
Other countries
Honda 1995–1999 2000–2004 2005–2006
30 22 16
n.a n.a 11
n.a n.a 27
46 54 55
12 9 11
12 15 7
Isuzu 1997–1999 2000–2004 2005–2006
38 39 29
8 16 34
46 55 63
36 28 14
n.a n.a n.a
18 17 23
Mazda 1997–1999 2000–2004 2005–2006
37 36 29
n.a n.a n.a
n.a n.a n.a
32 33 30
17 17 24
13 15 17
Mitsubishi 1997–1999 2000–2004 2005–2006
43 30 23
8 10 10
51 40 33
22 26 19
14 21 29
13 12 18
Nissan 1997–1999 2000–2004 2005–2006
38 32 23
n.a n.a n.a
n.a n.a n.a
38 42 42
15 14 17
8 12 18
Subaru 1997–1999 2000–2004 2005–2006
44 47 40
n.a n.a n.a
n.a n.a n.a
44 42 42
7 6 9
5 6 9
Suzuki 1997–1999 2000–2004 2005–2006
50 43 32
10 16 23
60 59 55
13 17 15
21 19 23
6 5 7
Toyota 1997–1999 2000–2004 2005–2006
42 36 27
n.a n.a 9
n.a n.a 37
37 37 37
9 11 13
13 17 13
Note: When data for Asia are not available (n.a.) they are included in ‘other countries’, otherwise ‘other countries’ exclude Asian figures. Source: author’s calculations based on company reports.
A major step in productive internationalisation Productive internationalisation is decisive since it materialises foreign direct investment which has been a landmark feature of globalisation since 1985. The automobile industry has a long experience in international production starting in the 1920s either to bypass protectionism, to increase economies
48
The Second Automobile Revolution 90 80
1990–94
1995–99
2000–2005
Percentage
70 60 50 40 30 20 10 0
ai ru da W tar an ta ki ult iat A hi ar rd an M zu W da ler nia lvo G su V on ys a o nd ba az BM vis M oyo uzu na F PS bis acc Fo iss I u H hr Sc V a y u Su M T S Re N its P N H rC M e l m ai D
Figure 3.1 Production share outside the country of origin, 1994–1995, 1995–1999, 2000–2005, percentages Source: Own calculations with data from the CCFA and the OICA.
of scale or to be close to customers. But despite this historical tradition, international production registered modest progress in the 1990s and remained limited in scope (Jetin, 2003: 20). The years 2000–5 introduce a breakdown in this respect. Figure 3.1 reveals that in 1990–4, the (simple) average share of international production amounted to a modest 26 per cent and half of the sample produced up to 20 per cent abroad. Yet in the period 2000–5, the average share of international production had reached 47 per cent and the median 48 per cent. In 2006, the last year available, the international production has passed the symbolic threshold of 50 per cent and half of firms produce 56 per cent or more abroad. Of course, firms like Scania, Volvo and DaimlerChrysler push the scores upward for specific reasons. But even if we take them out of the sample, the trend is confirmed. The average share is 42 per cent and the median 46 per cent for the period 2000–5, up from 22 per cent and 18 per cent for the period 1990–4 respectively. The biggest firms are driving this increasing productive internationalisation: Toyota from 23 per cent in 1990–4 to 39 per cent in 2000–5, Honda from 35 per cent to 57 per cent, Nissan from 31 per cent to 52 per cent, Renault from 13 per cent to 44 per cent, PSA from 4 per cent to 46 per cent, VW from 42 per cent to 56 per cent and even GM from 44 per cent to 52 per cent. This does not mean that automobile firms have turned footloose, or that the majority of them have become fully global. As is well known, there is a debate about the real meaning of the term ‘global’. In the present case, does the push in production outside of the country of origin mean that automobile firms produce a significant share of their total
Strategies of Internationalisation of Automobile Firms
49
70 60
1996–99
2000–05
2006
Percentage
50 40 30 20 10
N Ma av n Su H ista zu yu r ki nd -M ai ar M uti a R zda en au PS lt Is A uz u M BM its W ub Su ish b i Sc aru an ia Vo lk F sw ia ag t en G To M yo Vo ta lv o Fo Pa rd c N car is D s ai m H an le on rC d hr a ys le r
0
Figure 3.2 Production share outside the region of origin, 1996–1999, 2000–2005 and 2006 Source: Own calculations with data from the OICA.
production in every continent where there is a viable demand? To answer this question, we have calculated the production share of firms outside their ‘region of origin’ in a broad sense.4 Figure 3.2 presents the results. On the whole, the production share outside the region of origin is much smaller although it has increased recently from an average of 19 per cent in 1996–9 to 23 per cent in the period 2000–5 and to 26 per cent in 2006. But the median shows that half the sample produces 19 per cent or less outside its region of origin in 2000–5 and 23 per cent in 2006. This means that with a few exceptions, automobile firms are still more regionalised than globalised as far as international production is concerned. This is especially the case of Hyundai (2 per cent outside of Asia), of the two French firms (around 10 per cent outside of Europe), and small Japanese producers (Suzuki, Mazda, Isuzu and Mitsubishi, Subaru) with less than 20 per cent outside of Asia. Among the top five firms, Volkswagen, GM and Toyota were producing around 30 per cent outside their region of origin in the period 2000–5, and Ford 38 per cent. Paccar, Nissan and Honda produce close to 50 per cent and DaimlerChrysler, 67 per cent until 2006. Daimler sold Chrysler in 2007. Had Daimler sold Chrysler in 2006, Daimler’s share of production outside Europe would have fallen to 33 per cent and Chrysler’s share outside the Americas would have amounted to a meagre 3.5 per cent. The average share of the sample falls to 23.5 per cent and the median remains constant at 19 per cent.
50
The Second Automobile Revolution
In sum, for the whole sample, the progress of international production outside the region of origin appears limited. For Japanese firms, this is explained by the fact that the most dynamic emergent markets where growth is concentrated are localised in Asia, their region of origin. Other promising markets, like Russia, Turkey and Iran, are just emerging now, and some (Iran) are difficult or impossible to access for some firms for political reasons. So their combined impact on international production is still reduced. In the meantime the most striking feature is the skyrocketing increase of foreign production in China that reached 3.6 million units in 2006, up from 529,000 units in 2000 (source: OICA, 2006). Foreign production in China is now superior to foreign production in South America (3 million units in 2006) which used to be the most important market for emerging countries (author’s calculations based on OICA). India and Turkey are following with a foreign production of around 1 million units each. But India is considered as the new bull market of the future, bound to reach the size of the Chinese market if growth promises are fulfilled. The foreign production in Russia and Iran is still very small and concentrated in a few hands, but again some firms have announced big investment plans in these two countries. Japanese firms have benefited the most from the Chinese market with a production of 1.2 million units, followed by American (976,000) and European (873,000) firms (author’s calculations based on OICA). The only Korean firm surviving, Hyundai-Kia, achieved an astonishing production of 521,000 units in 2006. These Chinese outcomes have a powerful impact on firms. For instance, in the GM case, which is in decline in its home market and where it has posted huge losses (see above for more details on profit), the Chinese market is an unexpected life-belt that will help it to survive. But not all firms present in China are as successful. In 2006, among nine European firms, only Fiat (51,000), PSA (202,000) and most of all Volkswagen (620,000) are producing in China. GM (including GM Daewoo) is producing 842,000 units but Ford only 134,000 units. The Japanese producers are more equally present: Honda leads with a production of 371,000 units, then Toyota (327,000), Nissan (218,000), Mazda (151,000) and Suzuki (134,000). With the exception of China, foreign production in the ‘other Asian countries’ and in India is overwhelmingly Japanese or Korean. In the ‘other Asian countries’ (Asia without China and India), Japanese firms produce 2.1 million units against 194,000 for GM and Ford and only 6,000 almost exclusively produced by Renault-Samsung (author’s calculation based on OICA). In India, Japanese and Hyundai-Kia produce 1.067 million of units against 79,000 for American and European firms together. This gives a clear picture of what we call the ‘regional effect’. In contrast, European (1.7 million units) and American firms (1 million units) are very strong in Latin America, which has traditional ties with Europe and the USA, while Asian firms are weak (280,000 units). European firms produce 8.4 per cent of their international production in Latin America, almost
Strategies of Internationalisation of Automobile Firms
51
twice the share of their production in China (4.3 per cent) while for American firms Latin America and China have now the same relative weight (6 per cent) which was not the case in 2000. There will probably be a new rebalance in favour of China and India as numerous new plants under construction in these two countries will be finished soon. Finally, foreign production in Turkey has doubled its size, and has a more balanced presence of European, American and Asian firms. Foreign production has also increased greatly in Russia although quantities are still small (from nearly 0 in 2000 to 182,000 in 2006) and has almost doubled in Africa. But up to 2006, there were no Asian producers and only one European producer (Renault) in Russia, and no longer any American producers in Africa. In Iran, there is only one European (PSA, and soon Renault) and one Japanese producer (Mazda). So there is still potential for growth in international production in these emergent countries provided that their economic growth is confirmed, which is far from certain. Just to give an idea of the impact of emergent countries on automobile firms, it is worth noting that the international production of all automobile producers in these countries has grown from 4.9 million in 2000 to 11.9 million units in 2006: an increase of 7 million, almost the size of the Chinese market that year.
The evolution of international profit Profit stemming from international operations can be an additional source of profit, and is sometimes vital for the firm’s survival when domestic loss soars; it is also valuable in the long-term to maintain sustainable foreign subsidiaries. In this section, we compare American and Japanese firms (European firms have stopped publishing data on the geographical breakdown of their global profit since the end of the 1990s thanks to the adoption in the EU of the IFRS standards). One should be cautious in the interpretation of these data. ‘Fiscal optimisation’ leads firms to underestimate their profit in countries where taxes are high and overestimate them in countries where taxes are low. ‘Creative accountancy’ is also used sometimes to hide losses. The evolution of account standards over time in each country and firms’ discretionary decisions to change the way they present their profit add to the difficulty of establishing long-term data series. Nevertheless, the data presented here fit broadly with the known facts about firms’ performance. It sketches more or less faithfully the evolution of domestic and foreign profits and their relative contribution to world profit. To start with, Figure 3.3 shows the evolution of GM’s national and international profit. GM’s world profit remained positive from the end of the Second World War until the end of the 1980s. It was not affected by the first post-war worldwide recession of 1974–5, and only briefly by the second one in 1980. Up to 1980, North American profit contributed typically in the range of 90 per cent and foreign profit the remaining 10 per cent. It changed thereafter,
52
The Second Automobile Revolution
US$ billions
8 7 6 5 4 3 2 1 0 ⫺1 ⫺2 ⫺3 ⫺4 ⫺5 ⫺6 ⫺7 ⫺8 ⫺9 ⫺10 ⫺11
North America
55 57 19 59 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07
Foreign countries
19
19
19
53
Global net income
Year
Figure 3.3
Geographical breakdown of GM’s net profit, 1953–2007
Source: Author’s calculations based on GM’s financial reports.
national and foreign profit evolving no longer in a complementary but in a symmetrical way. For instance, foreign subsidiaries made an average loss of US$300 million from 1980 to 1986 while GM in North America made an average profit of US$2.5 billion. The problem was that until 2005–6, since the bulk of production and sales was located in North America, foreign profits were unable to offset North American losses given their magnitude. Such was the case during the third worldwide recession of 1990–2 and again in 2005 when GM registered the worst world net loss of its long history: US$-10 billion. North American net loss was US$-8.2 billion and foreign net loss, US$1.8 billion. The highest net profit ever registered by foreign subsidiaries was US$2.66 billion in 1989, and again US$2.59 billion in 1995. Even if foreign subsidiaries had made such a record net profit in 2005, it still would not have been enough to offset North American net loss. In 2006 foreign net profit (US$1.4 billion) proved insufficient again, even if it offset in part a North American loss of US$4.6 billion which narrowed the world loss to US$-3.2 billion. The same scenario happened in 2007 although loss in North America narrowed to US$-3.3 billion and foreign net profit increased to US$1.5 billion thanks to huge profit in China and Latin America. It is noteworthy that since the 1980s foreign profits have been very unstable as if GM was unable to secure regular profit abroad, although not all economies are as cyclical as the USA. A closer look at the geographical breakdown of GM foreign net profit (Figure 3.4) shows that the period 1987–96 was exceptionally profitable not only in Europe (with profit ranging from US$1.3 to 1.8 billion) but also in Latin America at the beginning of the 1990s (with US$800 million in 1993–4)
Strategies of Internationalisation of Automobile Firms
53
1800
US$ millions
1300 800 300 ⫺200 ⫺700 Latin America
Asia/Pacific
1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Europe ⫺1200
Year
Figure 3.4
Geographical breakdown of GM’s net profit, 1977–2007
Source: Author’s calculations based on GM’s financial reports.
and to a lesser extent in Asia, Pacific and Africa (with profit close to US$300 million). By contrast, the years 2000–6 reveal a structural weakness of GM Europe with systematic losses of US$800 million on average. Only GM AsiaPacific-Africa5 shows a spectacular turnaround with growing profit reaching US$1.2 billion in 2006 and US$557 million in 2007. This mainly reflects the huge profits made in China where GM has ventured successfully. Added to the US$500 million of profit that GM made in Latin America in 2006 and the record US$1.3 billion in 2007, on average US$1.5 billion per year comes to the rescue from the emerging countries. Whether GM can sustain this performance in emerging countries and return to profitability in Europe is an open question, but in any case foreign profit can be of great help for a financially distressed company but it is not a solution in itself to the structural problems GM has to face in North America. Ford is in a worse situation (Figure 3.5). Until the end of the 1970s, it had enjoyed stable domestic and foreign profit, with a predominant share for domestic profit. Ford was relatively unscathed from the first worldwide recession (1974–5), its world profit staying positive to the tune of US$300 million, with a fairly equal share of domestic and foreign profit. But Ford’s North American operations went in the red during the second worldwide recession (1980–1) while foreign operations turned an exceptionally high profit in 1979 (US$1.4 billion) that more than offset North American losses (US$-208 million). But this was the only case of its kind and usually foreign profits have not been high enough to compensate for domestic losses when these are huge. Worse, after six golden years of combined growing domestic and foreign profits from 1983 to 1989, Ford’s foreign operations entered into
54
The Second Automobile Revolution
7000 4500 2000 US$ millions
⫺500 ⫺3000 ⫺5500 ⫺8000 TOTAL
⫺13000
North America
⫺15500
Outside North America
⫺18000
1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007
⫺10500
Year Figure 3.5 Geographical structure of Ford’s net profit, 1957–2002, and income before income tax, 2003–2007 Source: Author’s calculations based on Ford’s financial reports.
a long period of almost fifteen years of losses and meagre profits. World profit has depended almost exclusively on North American operations and Ford was able to accumulate huge profit only during the long period of growth of the US economy. This dependence on the home market for profit is a critical problem since Ford has never recovered from the bubble burst in 2000 and was unable to make any profit even when the US economy entered a new cycle of growth from 2001 to 2006. In 2006, Ford registered an incredible loss before income tax of US$-15.9 billion (a net loss of US$-17.2 billion) and had to set a financial rescue plan. To pay the interest of the debt contracted, Ford had to sell its Premier Automobile Group, which, incidentally, had never been profitable. During the crisis episode, foreign operations were also in the red and added to Ford’s problems. Figure 3.6 presents the geographical breakdown of Ford’s foreign profit and loss.6 Ford Europe was more profitable than GM at the end of the 1970s and like GM enjoyed a golden age during the second half of the 1980s with European net profit of US$1–1.5 billion. But this successful period was shorter than GM’s and since 1991, Ford Europe has lost money almost every year. The crisis deepened in the year 2000 with losses between US$ -1.5 and -2 billion, PAG being the main but not exclusive factor responsible. Latin American subsidiaries were also seldom profitable except for a short-lived period at the end of the 1980s. Except for the year 1995, Ford was not able to make profit
Strategies of Internationalisation of Automobile Firms
55
1500 1000
0 ⫺500 ⫺1000 ⫺1500 ⫺2000 ⫺2500
Europe Latin America Asia/Pacific and Africa Latin America ⫹ Asia/Pacific and Africa 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
US$ millions
500
Year Figure 3.6 Geographical breakdown of Ford’s net profit, 1975–2002, and income before income tax, 2003–2007 Source: Author’s calculations from Ford’s financial reports.
as GM did when growth returned in Latin America during the first half of the 1990s. Latin American subsidiaries have returned to profit only recently (in 2004) after almost a decade of loss. Like GM, Ford registered a record profit of US$1.2 billion in 2007. Over the long term, Ford in Asia/Pacific and Africa has performed slightly better, and this is mostly explained by the return of Mazda to profit since 2003 (in which Ford has a 33 per cent stake) that is reflected in Ford’s results. From 2001 to 2006, the Ford brand in Asia and Africa made only a marginal profit of US$42 million in 2005 and US$2 million in 2007, and losses for the rest of the period. Even combined with Mazda’s profit, the amount remains modest, the highest profit registered in the region being around US$300 million, far from GM’s record profit in the region (US$1.2 billion in 2006). Overall, foreign results in emerging countries have been disappointing for Ford unless the 2004 figures herald a new era of profitability. On balance, foreign operations have been more of a burden than a bonanza with an accumulated loss of around US$11 billion from 2000 to 2007. Such a gloomy scenario is not true of most Japanese firms, which have managed to turn foreign operations into a sustainable source of profits. Figure 3.7 presents the operating profit of Japanese firms in Japan from 1994 to 2006 and Figure 3.8 the operating profit realised by the same firms abroad.7
56
The Second Automobile Revolution
14,000 Honda
12,000
Nissan
Toyota
US$ millions
10,000 8,000 6,000 4,000 2,000 0
94
95
96
97
98
99
⫺2,000 Figure 3.7
2000 2001 2002 2003 2004 2005 2006 Year
Operating income of the Japanese ‘big three’ in Japan
Source: Author’s calculations from financial reports.
8,000 Honda
7,000
Nissan
Toyota
US$ millions
6,000 5,000 4,000 3,000 2,000 1,000 0
94
95
96
97
98
99
2000 2001 2002 2003 2004 2005 2006
⫺1,000 ⫺2,000 Figure 3.8
Year Operating income of the Japanese ‘big three’ abroad
Source: Author’s calculations from financial reports.
One can see that the Japanese big three (Toyota, Honda, Nissan) have been successful in the period on both markets. Toyota’s performance is astonishing. Its foreign operating profit reached US$6.67 billion in 2006, up from US$428 million in 1994, overtaking Honda in 2003 and increasing the gap with Nissan which from 2000 to 2002 had very close scores. It is also remarkable that Honda’s foreign operating profit already contributed to 56 per cent of its world profit in 1994–9, and yet increased to 64 per cent in 2000–6. This reflects both a successful internationalization but also an overdependence.
Strategies of Internationalisation of Automobile Firms
57
1200 1000
US$ millions
800 600 400 200 0 ᎐200
94
95
96
97
98
99
2000 2001 2002 2003 2004 2005 2006
᎐400 ᎐600 ᎐800
Year Isuzu
Figure 3.9
Mazda
Mitsubishi
Subaru
Suzuki
Operating income of Japanese ‘five small’ firms in Japan
Source: Author’s calculations from financial reports.
In Nissan’s case, the phenomenon is more recent and less pronounced. Yet during the period 2000–6, foreign operating profit contributed to 53 per cent of Nissan’s world profit and 60 per cent in the most recent period of 2003– 6. Toyota is even more more dependent on the Japanese market. In 1994–9, Japanese operating profit contributed to 80 per cent of its world profit and 20 per cent of its foreign profit. In 2000–6, its foreign profit share had increased to 32 per cent. The situation is more variable for the ‘five small’ Japanese producers. These firms have unstable results both at home and abroad and some of them, like Mitsubishi, had to be rescued from bankruptcy by a bail-out from its main shareholders. Among them, only Suzuki and to a lesser extent Mazda are really successful at home and in their foreign operations (Figures 3.9 and 3.10). Suzuki, which was making most of its profit in Japan at the start of the period, despite its long presence in Asian countries like India, has recently enjoyed a steady progress in its foreign profits which reached US$590 million in 2006, amounting to 50 per cent of its world profits, up from 14 per cent in 1995–9. Suzuki has benefited from its good positioning in the motorbike market and the cheap entry level small passenger car in Asian emerging countries. This success has persuaded Suzuki to embark on a new plan of investment in Asia, North America and Europe with the ambition of developing new models in the middle range of passenger cars. Mazda seems to have overcome its long-standing difficulties of the 1990s. With the exception of 2000, when it registered an operating loss of US$-120 million, Mazda has improved its profit both in Japan, where it amounted to
58
The Second Automobile Revolution
US$ millions
800 600 400 200 0 ⫺200 ⫺400 ⫺600 ⫺800 ⫺1000 ⫺1200
94
95
96
97
98
99
2000 2001 2002 2003 2004 2005 2006 Year
Isuzu Figure 3.10
Mazda
Mitsubishi
Subaru
Suzuki
Operating income of Japanese ‘five small’ firms abroad
Source: Author’s calculations from financial reports.
US$1 billion in 2006, and abroad, where it reached a record level of US$343 million the same year. Over the whole period 2000–5, Japanese operating profit contributed to 73 per cent of Mazda’s world profit and 27 per cent of its foreign profit. Mitsubishi is still struggling to recover from a 45 per cent drop in sales in 2003–4 in the USA that cost them US$-1.1 billion in 2003 and US$-691 million in 2004 and it is this that explains most of the abyssal fall of foreign profit that appears clearly in Figure 3.10. But it has also to regain Japanese consumers’ confidence after the 2004 announcement of a products defects cover-up that lasted 25 years and which cost Mitsubishi US$-0.685 million that year. Subaru is also in a precarious situation due to its specialisation in niche products (e.g. the all-wheel-drive car) sold mainly on the North American market. Its future lies in its capacity to launch new successful cars in this narrow segment of the market that is also supplied by competitors. The problem common to these smaller Japanese producers is that their productive internationalisation is more reduced than that of the big three. As a consequence they are more exposed to adverse currency fluctuations mainly between the yen and the US dollar. Their small size adds the problem of reduced economies of scale. The geographical breakdown of the ‘big three’ operating profit reveals that thanks to greater geographical diversification, the dependence on the North American market is still very high but decreasing. At the end of the 1990s, Toyota made almost all of its foreign profits in North America, with Europe making losses and ‘other countries’8 bringing a very small profit (Figure 3.11). Things have started to improve recently. Since 2002, profit in Europe reached US$1 billion in 2004 and 2006. In ‘other countries’, operating profit reached
Strategies of Internationalisation of Automobile Firms
59
4,500 4,000
North America Asia
US$ millions
3,500
Europe Other countries
3,000 2,500 2,000 1,500 1,000 500 0 ⫺500
97
Figure 3.11
98
99
2000
2001
2002
2003
2004
2005
2006
Year Geographical breakdown of Toyota’s operating profit
Source: Author’s calculations based on Toyota’s financial reports.
the stunning level of around US$1 billion in 2003, US$1.8 billion in 2005 and US$1.7 billion in 2006. One can see from Figure 3.11 that this is due primarily to the huge profit made in Asia, and in particular but not exclusively in China. In 2005, profit in Asia alone reached US$1.2 billion which leaves US$572 million in the rest of the world, Latin America and Africa. Over the period 2003–6, Toyota made 24 per cent of its world profit in North America, around 5.5 per cent in Europe and in Asia and 3.1 per cent in the rest of the world. There seems to be a process of rebalancing of world profit as a consequence of Toyota’s greater variety of commercial and productive internationalisation. Honda is also very dependent on its North American operating profit which amounted to 54 per cent of its world profit in 2000–6 (Figure 3.12). Honda also suffered operating losses in Europe at the end of the 1990s of even greater magnitude than Toyota before making profit in the year 2000 which nonetheless never reached US$500 million. Like Toyota, ‘other countries’ have turned a growing profit that exceeded the US$1 billion range in 2005 thanks to buoyant operating profit in Asia that accounts for half of the success. Honda made close to US$1 billion of operating profit in Asia in 2006 after US$1.2 billion in 2005. Nissan made a strong recovery on the North American market (Figure 3.13). Its operating profit reached a peak at US$3.9 billion in 2004, but in contrast to Toyota and Honda, Nissan was not able to secure these good results in subsequent years. Nissan’s operating profits in Europe rank between Honda’s and Toyota’s, but in ‘other countries’ Nissan lags behind Toyota and Honda. Nissan has been less successful in catching the Asian wave of growth until 2006. Suzuki (Figure 3.14)9 is representative of a company that makes the bulk of its operating profit in Asia, due to its strong presence and market share in the
60
The Second Automobile Revolution
US$ millions
4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 0 ⫺500 ⫺1,000
97
98
99
2000
2001
2002
2003
2004
2005
2006
Year North America Figure 3.12
Europe
Asia
Other countries
Geographical breakdown of Honda’s operating profit
Source: Author’s calculations based on Honda’s financial reports.
US$ millions
4,500 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 0 ⫺500 ⫺1,000
North America Europe Other countries
97
98
99
2000
2001
2002
2003
2004
2005
2006
Year Figure 3.13
Geographical breakdown of Nissan’s operating profit
Source: Author’s calculations based on Nissan’s financial reports.
small car market, in India and Pakistan, and in motorbikes throughout Asia. Suzuki was very well positioned to capture market growth in India although it will now have to face Tata’s Nano competition in the cheap mini-car segment. From 2003–6, Suzuki made a yearly average operating profit of US$335 million in Asia, compared to US$85 million in Europe and US$41 million in North America. It remains to be seen whether Suzuki’s decision to enlarge its model range in Europe and North America will reshuffle this geographical breakdown in favour of these two regions. Finally, we present Hyundai’s case, the last remaining big independent Korean manufacturer. During the period 2001–6, Hyundai made an accumulated operating profit of almost US$13 billion in Korea (US$2.1 billion per
Strategies of Internationalisation of Automobile Firms
61
450 400
North America Asia
US$ millions
350
Europe Other countries
300 250 200 150 100 50 0
97
⫺50
98
99
2000
2001
2002
2003
2004
2005
2006
Year Geographical breakdown of Suzuki’s operating profit
Figure 3.14
Source: Author’s calculations based on Suzuki’s financial reports.
300 200
04 20
02
03 20
⫺100
20
01
0
20
US$ millions
100 05
20
20
06
⫺200 ⫺300 ⫺400
North America Asia Europe Total foreign
⫺500 Year Figure 3.15
Geographical breakdown of Hyundai’s operating profit
Source: Author’s calculations based on Hyundai’s financial reports.
year on average) and a meagre profit of US$607 million abroad. The geographical breakdown (Figure 3.15) shows that Hyundai’s international operations were making a profit of more than US$200 million at the start of the period and ended with an operating loss of US$-29 million. The reason for this was its growing deficit in Europe which reached US$-452 million in 2006 partially offset by increasing profit in Asia due to Hyundai’s strong presence in China and India, and in North America for some years. This means that Hyundai’s plan for important investments in foreign subsidiaries may represent a risk
62
The Second Automobile Revolution
for its financial stability if it does not manage to turn a profit in Europe and stabilise its position in North America.
Conclusion This chapter started with the question: Has there been a leap forward in the process of internationalisation of automobile firms? The answer is undoubtedly affirmative. For the majority of firms, the degree of internationalisation on almost all criteria has made significant progress between 1995–9 and 2000–6. One of the main reasons for this is the development of the so-called emerging countries and in particular of China. If the pattern of growth in India follows the Chinese pattern, then having a presence in these two growing markets will be even more critical than it is today for old-established automobile firms from North America, Europe, Japan and Korea. In this endeavour, they will have to face the growing competition of Chinese and Indian firms which will not only defend their home market but will also increasingly try to develop their international operations. If Russia and to a lesser extent Iran also confirm their growth expectations, there will surely be an added impetus for more internationalisation. Does this mean that all automobile firms will reach and overtake the 64 per cent level already realised by Honda, the most internationalised producer? Probably not. The European and North American markets remain decisive for local firms because volume there is very important, which is less the case in the Japanese market where sales are declining. For Japanese firms, gaining foreign market shares is even more imperative than for their European and American competitors which must defend their home markets. But between the present level of internationalisation of most automobile producers and the level reached by Honda, there is room for another round of international expansion. The next question is what will be the impact of the new coming round of internationalisation on the productive models of the firms involved? The internationalisation process tends to question the productive model of a given firm because the profit strategy may not be relevant, or not as relevant, when the targeted foreign markets are very different from the market of origin, or when foreign expansion destabilises the existing government compromise. But the internationalisation process may also strengthen the existing productive model in two cases: when the targeted foreign markets are very similar to the market of origin, or when the targeted foreign market is very different but the volumes involved are so important and growing that almost all profit strategies are relevant or at least compatible. It is beyond the scope of this chapter to provide an exhaustive answer. The dedicated chapters of this volume will bring more insights on this topic. But it is our opinion that the growing trend of internationalisation promoted by the Chinese and Indian growth opens space for all kinds of productive models. This was not the case for Latin America, which was the first to be coined ‘emerging’ at the
Strategies of Internationalisation of Automobile Firms
63
beginning of the 1990s, because income inequality – a major hindrance to automobile demand – was not offset by the large population. Cheap-entry mini-cars and cars or pickups for less well-off consumers or farmers, standard passenger cars for the middle classes, and top-range and deluxe passenger cars, SUVs and sports cars for the richest are already common in China and India. Even innovative eco-cars using alternative fuels will be successful because the high price of petroleum and pollution are already big issues in Asian emerging countries. This means that the most innovative firms able to make a breakthrough in fuel cell or electric cars will find consumers there. The idea that big and growing volume makes all and even old productive models viable may look simplistic, but is appealing. Methodological appendix Sources Financial reports of automobile firms. OICA, Organisation international de la Construction Automobile. CCFA, Comité des Constructeurs Français d’Automobile. Profit definition in account standards Ford and GM have favoured until recently the publication of net profits and Japanese firms the publication of operating profits. Operating profit depicts more accurately the profitability of the production system and the marketing success of products. Deducting ‘net interest income’ and ‘equity in net income’ gives the ‘income before income taxes’ which is usually in the same magnitude. Deducting taxes, ‘income or loss from minority interest’ gives the ‘income (loss) from continuing operations’. In the USA, this precision is important because Ford and GM close automobile plants nearly every year and have bought or sold a lot of firms in other manufacturing sectors and in services since the 1950s. Deducting income (loss) from these ‘discontinued operations’ and the ‘effect of change in accounting principles’ gives the net profit (loss). For all these reasons, the difference between the operating profit and the net profit can be important, especially in the case of numerous plant closures and spin-offs (Visteon, Delphi). But in the long term, the evolution is the same, no loss is turned into profit and the geographical breakdown is not significantly affected. Statistical annex (Table 3.1) Ford: the share of the workforce refers to the workforce outside North America and not the USA and for the automobile segment only, which represents 95% of total workforce. The share of foreign assets is based on long-lived assets. GM: the share of foreign assets is based on long-lived assets.
64
The Second Automobile Revolution
DaimlerChrysler: the share of foreign assets is based on long-lived assets. Fiat: the share of foreign assets is the average of 2004–6. PSA: the share of commercial revenues is the average of 2000 and 2001 only. The share of assets is an estimation based the relative weight of France in Europe’s share of long-lived assets. Renault: the share of foreign assets is based on long-lived assets. Scania: the share of commercial revenues is the average of 2003–5. The share of assets is an estimation based the relative weight of Sweden in Europe’s share of total assets. Volvo: the share of assets is the average of 2001–5. Isuzu, Mazda, Mitsubishi, Suzuki: the share of employment is based on JAMA data for 2005.
Notes 1. The Transnational Index (TNI) is calculated as the average of three ratios: foreign assets to total assets, foreign sales to total sales, and foreign employment to total employment. 2. This statistical approach has been developed in a previous study (see Jetin, 2003). We use these previous results as a benchmark for the period 2000–5, although we have introduced some changes. 3. Total revenues include the revenues proceeding from manufactured sales to external customers, financing revenues from the captive financing arm when there are existing and other revenues. 4. For European firms, the region means the whole of Europe plus Turkey and Russia. For American firms the region encompasses North and South America. For Asian firms, the region includes Northeast, Southeast and South Asia. In each case, the country of origin is included in the region. 5. Africa is included by GM in the same division as Asia and Pacific but plays a marginal role in GM’s performance. 6. The breakdown data for Latin America and Asia/Pacific/Africa were missing for some years. So we put the aggregate for the whole set of these regions, which was available, to allow us to study the evolution of Ford’s profit in these emerging countries. 7. The data have been converted from Japanese yen into US dollars using the exchange rate at the Tokyo stock exchange at the end of the Japanese fiscal year, usually 31 March. The data cannot be compared directly with Ford’s and GM’s results, not only because of the different profit indicators but also because of the specific value of the exchange rate at the end of the fiscal year. Using the annual average exchange rate introduces a different bias but does not bring more accuracy. Hence the data are presented in dollars for the convenience of the reader to give only an order of magnitude in the same currency. 8. For the sake of simplicity of Figure 3.11, ‘other countries’ include Asia for the whole period. Starting in 2003, operating profit in Asia has been published by Toyota. Due to its importance, it is presented as such in Figure 3.11, although it is still included in the category ‘other countries’ that otherwise would have become inconsistent with previous years. The same holds for Honda in Figure 3.12 and for Nissan in Figure 3.13 which does not disclose operating profit data for Asia.
Strategies of Internationalisation of Automobile Firms
65
9. Suzuki has published data on its operating profit in Europe for 1997, in North America in 1999, and in Asia in 2001. ‘Other countries’ include Asia, Oceania and South America.
Bibliography Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Freyssenet, M., Shimizu, K. and Volpato G. (2003) ‘The Internationalization Strategies and Trajectories of Automobile Sector Firms’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Jetin, B. (2003) ‘Internationalization of American and Asian Automobile Firms’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Lung, Y. and van Tulder, R. (2004) ‘In Search of a Viable Automobile Space’, in J. Carillo, Y. Lung and R. van Tulder (eds), Cars . . . Carriers of Regionalism? Basingstoke and New York: Palgrave Macmillan. UNCTAD (1995) ‘World Investment Report 1995: Transnational Corporations and Competitiveness’. New York and Geneva: United Nations. UNCTAD (2000) ‘World Investment Report 2000: Cross-Border Mergers and Acquisitions and Development’. New York and Geneva: United Nations. UNCTAD (2006) ‘World Investment Report 2006: FDI from Developing and Transition Economies: Implications for Development’. New York and Geneva: United Nations. UNCTAD (2007) ‘World Investment Report 2007: Transnational Corporations, Extractive Industries and Development’. New York and Geneva: United Nations.
This page intentionally left blank
Part I The Divergent Trajectories of Japanese and Korean Carmakers
This page intentionally left blank
4
The Uncertainty of Toyota as the New World Number One Carmaker Koichi Shimizu
In 2007, the sales volume of Toyota group’s vehicles as a whole recorded 9,366,418 units, which is slightly behind General Motors’ sales, whereas its production volume reached 9,497,754 units, surpassing that of General Motors (GM). This was largely attained by a rapid growth of its overseas production and sales at the start of the twenty-first century, after production and sales stagnation for ten years (see Figure 4.1). It is as if Toyota deployed its explosive global strategy after patient preparation during the 1990s, a decade of stagnation for the Japanese economy and trade conflicts with the USA. In addition, Toyota successfully launched its first hybrid car, the Prius, with a reasonable price in 1997, and became an innovative leading company in the field of ecologically friendly cars. This new approach to Toyota’s product strategy has increased its share in the North American market, where the old ‘big three’ are suffering, and which is now the most important market for Toyota, accounting for 36 per cent of its global sales in 2006, whereas the share of its sales in the Japanese market reduced from 51 per cent in 1990 to 35 per cent in 2000, and to only 21 per cent in 2006. After such a striking success in its foreign markets, Toyota’s global competitive edge now seems solidly established and invincible, and its high production efficiency and products quality are assured by the famous Toyota Production System (TPS). However, during the 1990s Toyota revised its human resource and productivity management system, which constituted the core of TPS founded by Taiichi Ohno, as if Toyota had changed course in its evolutionary trajectory. Has Toyota changed its traditional Toyotaist paradigm to become a ‘truly global company’?
The Toyota Production System and its crisis The TPS is famous for its ‘just-in-time’ production (kanban system) and decision-making capability (jidoka) of machines and workers. As Spear and Bowen (1999) have argued, however, the TPS cannot be reduced to the tools 69
70
The Second Automobile Revolution
and practices observers can see when they visit a factory, or to those the business manuals describe as the techniques of the TPS. In fact, there exists a governance system, publicly unstated. Boyer and Freyssenet (2002) have characterised Toyota’s productive model as a system of ‘permanent reductions in costs’, and the TPS can be understood as a set of systematised ways to reduce the cost per vehicle (Shimizu, 1999a). In fact, this cost management system constituted the core management system of the TPS for forty years from 1950, assuring its high performance and competitive edge, but it encountered a labour crisis at the end of the 1980s.
Framework of the cost management system of the TPS Toyota’s cost management system, called ‘management by costing’ (yosan kanri), consisted of the two main phases in the value-chain: the ‘target costing’ in the design stage (genka kikaku), and the ‘kaizen costing’ (genka kaizen) in the production stage. When the development of a new or revised vehicle was decided, its sales price and target (or desired) profit per vehicle were fixed by the top management taking into consideration the market price and the new car’s features, which the sales division had proposed on the basis of marketing and consumer analysis. As a result, the desired cost per vehicle was given automatically: desired cost = sales price – target profit. Of course, the real production cost cannot be realised in such a way. But here is the specificity of Toyota’s cost management: by fixing this desired cost as ‘target cost’, it set about
4000
Thousand units
Production in Japan 3000
2000 Exports 1000 Overseas production 1960
Figure 4.1
1970
1980
1985
1990
1995
Production and exports of Toyota, 1960–2007
Sources: Toyota, financial reports, and company news about figures of 2007.
2000
2005
The Uncertainty of Toyota as a World Carmaker
71
organising activities to attain it, and then to reduce it further, throughout the product design stage and the production process. After the product manager had fixed the product plan, the target costing in the design stage was launched under the control of the Cost Council: the design engineers would draw the parts and components of the vehicle so as to meet their target cost, without of course neglecting their required quality. Following Toyota conventions, maintained until the mid-1990s, the cycle of drawing parts and their production trial on the real production line was repeated three times before the definitive designs were fixed; these activities belonged to the so-called value engineering (VE) stage. Meanwhile, the production engineering division, which conceived and prepared the production lines of parts and the final assembly line, fixed the ‘standard time’ for producing parts and a whole vehicle. Then the cost realised in the definitive drawing became the referential cost of the product. In other words, every production line had its own referential cost and standard time. These conventions were altered during the 1990s (see below), but the essence of the VE has been retained. The cost management in the production phase started after mass production had begun. Costs and standard times were not always met in production lines for the first few months. With the TPS, the learning effects appear in general three months after the launch of new production. If the referential cost was not reached despite the learning effects, kaizen was applied in order to reduce the cost or new drawings would be tried. In principle, when the target profit was not realised, efforts had to be devoted on the one hand to increasing sales, and on the other to reducing costs by kaizen. In any case, kaizen activities for cost reduction were imperative. But, even when referential cost and ‘standard time’ were satisfied, kaizen activities were pursued in factories not only to maintain the referential cost and standard time but also to reduce them further. This was an original feature of Toyota’s cost management. At Toyota, the costs to be managed in the production division consist of materials (including purchased parts, tools and energy) costs and labour costs, which were controlled in a different way. The former were supervised by the Cost Council organised at every level in the company, from top management to the shop floor. Until the end of the 1980s, the top management fixed the amount to be reduced by kaizen, and then distributed it as a kaizen norm to all production divisions, taking their feasibility into account. In a production division, the allocated sum was shared out among working groups (kumi) passing through sections (ka) and sub-sections or working units (kakari). The responsibility for this kaizen belonged to the director of the production division, but Cost Councils were held every month at every level in order to manage kaizen results as well as to discuss the various measures that might have to be taken. At the beginning of the 1990s, this imposed kaizen norm was abolished, so that every factory would fix for itself the objective to reduce the materials costs.
72
The Second Automobile Revolution
By contrast, the management of labour costs was very complex until the end of 1980s. Firstly, it was not so much labour cost but production efficiency that had to be managed, given the fact that the labour cost was calculated on the basis of production efficiency. Secondly, it was only the Production Allowance Council organised in the top management that watched over and controlled the movement of production efficiency. Thirdly, the management of the latter was related to the production allowance that constituted an important part of the monthly standard wage of workers (almost 60 per cent at the end of the 1980s). Though this production efficiency management changed at the end of the 1990s, it is worth explaining it further because of its importance as an aspect of the management system created by Taiichi Ohno, the founder of the TPS. Production efficiency management and production allowance In its simplified formula, the production efficiency of a working unit is its labour productivity per man-hour, measured against the standard time: Production Efficiency =
(Standard Time) × (Production Volume) Real Working Hours of Working Group
This efficiency was calculated for each working unit and every month. In order to achieve production efficiency, the production volume had to contain only the products without defect. When some products were defective and needed repairs, the production efficiency fell in part due to a decrease of production volume in a given time, and in part because the real working hours became longer for a given production volume. Therefore, the workers had to assure the quality of products in their operations in order to maintain or increase the production efficiency. Prolongation of real working hours always happened as workers stopped their production line when they had a problem in their operations such as the discovery of defective parts or when their operations were delayed with respect to cycle. When the production line stopped because of the delay arising from an inadequate organisation of tasks, it forced supervisory staff to revise standard tasks, and they were engaged in kaizen to improve the working process in question or the organisation of tasks for a whole production line. In fact, the line stop system was conceived by Taiichi Ohno so as to improve the production line precisely in this way. Until the end of the 1980s, the Production Allowance Council and the production management team supervised the movement of production efficiency of all working units as well as working groups. Furthermore, production efficiency determined the amount of the production allowance. Toyota’s wage formula for all employees, until the end of the 1980s (we neglect various parts such as family allowance) was: monthly wage = basic wage + production allowance + overtime payment. Here, the
The Uncertainty of Toyota as a World Carmaker
73
production allowance was determined as basic wage multiplied by the production allowance coefficient (PAC), which was determined on the basis of the production efficiency (PE) of each working unit. Although the calculation of the PAC is very complex, we simplify it here and show only that pertaining to blue-collar workers in production divisions (the details of Toyota’s wage system in this epoch have been explained by Nomura (1993); see also Shimizu (1999a)). After adjusting the PE in order to round off its fluctuations from month to month, this adjusted PE of all working units was ranked and classified into four levels A, B, C and D from top to bottom. Then the PAC was calculated as the average PE of each level. If a working unit was classified at a higher level, its workers received a higher production allowance. This means that the gains obtained by productivity increase were shared between the company and the employees in question. There was, however, a rule for cutting the standard times of the working units that marked the adjusted PE higher than the average PE of the best level group A. Their standard time was reduced so that their PE would become the average PE. This was called the ‘standard time cutting rule’. At the same time, however, management ordered the working units in question to reduce the number of their workers (perhaps one or two workers from the unit, for example). If they did nothing, they would be classified at an inferior level in the next month. Because of this convention, they were compelled to make kaizen in order to improve their production efficiency so as to reach at least the same production efficiency as before with a reduced number of workers (cf. the statement of an ex-administrator of Toyota – Ikebuchi (1990)). Thus,
1.6 1.5 1.4 1.3 1.2 1.1 1 1972
1975
1980
1985
1989
Figure 4.2 Movement of the average production allowance coefficient, Toyota, 1972–1989 Source: Toyota.
74
The Second Automobile Revolution
a kaizen–production efficiency–production allowance–kaizen chain was established (Shimizu, 1995a). As the result of such kaizen, the average PE of Toyota as a whole had a tendency to increase (Figure 4.2). However, this system was called into question in the 1990s, and radically altered in 2000. Like the kaizen for increasing production efficiency, those for lowering labour costs were not a matter of workers’ voluntary activities, carried out through quality control (QC) circles or through a suggestion system. In fact, the kaizen activities in question were controlled by the Production Allowance Council, and there was and still is a division of labour amongst group leaders (working group leader), chief leaders (working unit, or sub-section leader) and engineers, although managers (section chiefs) in the plant are promoters and coordinators of their activities. The group leaders were mainly in charge of reducing the cycle time of their workers by improving their operation process and shortening the standard times, because the establishment of standard work and standard time was their important function. The chief leaders, key-persons of the TPS according to Ikebuchi (1990), were also engaged in shortening the standard times of their working unit, but this occurred by means of process improvements for reducing the number of workers in their unit. In general, the reduction of workers from production lines constituted an important method at Toyota for increasing production efficiency as well as for decreasing labour costs (Monden, 1985; Ohno, 1988). The group leaders and chief leaders also organised occasionally an autonomous study group in order to reduce the real working hours by process improvements over several working groups and units. The engineers belonging to a plant were in charge of maintaining and improving the quality of products, productivity, cost and security of work from an engineering point of view. And as part of their usual activities, they provided mechanical solutions for the bottlenecks in production lines, discussing what was needed with group leaders and chief leaders. In general, capital-labour substitution constituted the main way of reducing workers, but in Toyota’s or Ohno’s philosophy, the replacement of workers by machines had to be done after improving the operation. It was such organized kaizen under the control of production efficiency and cost managements that constituted the source of Toyota’s high performance for more than forty years. The reason why this system worked well for so long lay in labour relations. As we saw above, the TPS was based on human capital, so that the involvement of all employees (white-collar and blue-collar) was crucial. The relationship between management and the union was and still is based on the principle of mutual trust, inscribed in the management–union joint declaration in 1962, which was signed twelve years after a major labour conflict in 1950, and eight years after the cooperative group became a majority force in the union. This declaration made official the cooperative relationship between management and the union and defined the company as
The Uncertainty of Toyota as a World Carmaker
75
a community composed of administrators and employees, where increasing productivity was considered to be a precondition for increasing both company profits and employees’ wages. This company governance compromise (Boyer and Freyssenet, 2002) prevented both the union using strong-arm collective bargaining, and management from taking decisions that might undermine this trust, such as redundancies. The radical revision of human resources management around the 1990s was realised on the basis of this compromise valuing mutual trust, which was also applied to Toyota’s relationship with dealers and suppliers (Shimizu, 1999a). Crisis of work at the end of the 1980s In the ‘bubble economy’ from 1987 to 1991, demand for cars was overheated, accelerating product diversification and upgrading quality. But the situation of general euphoria provoked a crisis of work: there was a labour force shortage in heavy manufacturing because new labour market entrants, whose numbers had decreased anyway due to the declining birth rate and the increasing popularity of higher education, were able to find high-paid jobs in the service sector, and regarded manufacturing work as kitanai (dirty), kitsui (hard) and kiken (dangerous). In response, carmakers tried to improve the appeal of assembly lines with the slogan ‘human friendly assembly line’ or to install ‘silver lines’ where older workers could work. Some of them also increased automation when they constructed new plants (for example, the Kyushu plant for Nissan, the Hofu plant for Mazda and the Tahara No. 4 plant for Toyota). At the same time, Toyota was engaged in a radical revision of its production and human resources managements in order to solve the crisis. Convinced of the necessity of making work more attractive, Toyota’s union and management formed a joint committee in 1990 in which their representatives re-evaluated the production efficiency management coupled with production allowance, human resources management, general working conditions and the organisation of tough assembly line work (Shimizu, 1995a, 1999a). The reason for this revision lay in the very characteristics of the Toyota Production System itself. As we explained above, the way to increase productivity or production efficiency was to reduce the number of workers by kaizen of the production process for a given production volume under the constraints of just-in-time production. Without necessarily intensifying work, the factories had a tendency to run with a minimum number of workers. In stable economic growth, this system operated very well, but it could not meet the explosive demand of the ‘bubble economy’. With diversification of parts and increased production volume, workers were under heavy pressure and Toyota lost many of its younger workers. For example, a quarter of newly hired young workers left Toyota during their first year in 1990. To solve the labour shortage problem, Toyota first employed more temporary workers than ever before (rising from 2,540 to 23,370 in 1990), who
76
The Second Automobile Revolution
1985⫽ 100 115 110 105 100 95 90 85 1985
1986
1987
Production Figure 4.3
1988
1989
1990
Productivity
1991
1992
Labour force
Production, productivity and labour force of Toyota, 1985–1992
Note: Productivity represents that of direct workers including logistic workers in the plants (labour force). All indices were calculated on the basis of Toyota’s data. Source: Shimizu, 1999a.
were mainly assigned to assembly plants. In some working groups, they constituted three-quarters of all workers. However, the lack of highly trained workers and an increased number of less skilled temporary workers unable to execute complex tasks on Toyota’s assembly lines, on which various car models with diversified parts were assembled, meant that the production lines were often interrupted. Group leaders and chief leaders then had to intervene on the line in order to solve problems and to help temporary workers. As a result, annual working hours rose (to 2,315 hours in 1990) but productivity fell sharply (see Figure 4.3). Even the group and chief leaders were exhausted, and the labour shortage turned into a crisis of collective working. In this state of confusion, traditional production efficiency management as well as the ‘rigid’ application of just-in-time production (which means one by one production without buffers in its ideal model) were called into question.
A decade of reconfiguration for the Toyota Production System The 1990s was a decade of reconfiguration for the Toyota Production System, which began with the revision of its human resource management. In the joint committee mentioned above, management and the union decided
The Uncertainty of Toyota as a World Carmaker
77
to modify the training system of blue-collar workers, the hierarchical positions of supervisory staff, the way of working on the assembly line, and the production efficiency management (Shimizu, 1995a, 1998, 1999a). As regards the training system, in 1993 Toyota modified the initial training of new recruits who had just finished high school, hoping that providing sufficient initial training would reduce the turnover rate of young workers. In the past they had been assigned to a factory after two weeks of training at company headquarters, whereas now they were assigned directly to a factory where they were trained over a longer period of time: nine weeks in the stamping, welding and assembly factories, and six weeks in the others. In addition, Toyota introduced in 1991 a new vocational training system, in which candidates received both off-the-job training and on-the-job training in order to qualify for various levels of technical competence: level C means the worker is able to do 20 per cent of the tasks in a working group; level B that the worker can do 60 per cent of the tasks and some repairs in a working group; level A that the worker can do 80 per cent of the tasks and some repairs in a working group; and level S means that the worker can do every task, repairs, and maintenance in a working group. In 1999, the number of qualified employees at the C level was 4,312, 18,976 at the B level, 5,743 at the A level, and 25 at the S level. This vocational training scheme had, in the beginning, the same objective as the ‘dream’ of ‘reflective production’ at Volvo’s Uddevalla plant (Ellegård et al., 1992), namely that through learning by doing, a worker would ultimately be able to assemble a whole car alone, or at least to have such competence that would give workers motivation for car production work. As for the hierarchical positions, Toyota introduced three expert positions into its grade system (see Table 4.1): an expert (EX) replaced team leader (TL); a senior expert (SX) was ranked to the same grade as group leader (GL), but did not execute any managerial task; a chief expert (CX) had the same grade as chief leader (CL, sub-section leader), but without doing managerial tasks. In 1993, the number of EX, SX and CX was respectively 9,580, 1,100 and 320, whereas the number of TL, GL and CL was respectively 1,930, 3,920 and 1,400. At that time, the post of TL had to be completely replaced by EX, and this replacement was achieved around the mid-1990s. Having high skills, they engage in special tasks in their plant, in kaizen of assembly lines for example. This change was made because TL without team members, and GL and CL with few subordinates had increased due to mechanisation and automation. Salary increases have traditionally been linked to hierarchical position and they are therefore promoted to such positions after evaluation of their ability. Incidentally, disappearance of TL, and then teams composed of about 5 workers, caused serious problems at the beginning of the twenty-first century, as a result of which Toyota recreated TL positions and teams in January 2007 (see below for more about this problem).
78
The Second Automobile Revolution
Kaizen for making assembly work more attractive Concerning assembly work, the Production Engineering Division developed a new concept of assembly line and work organisation. After trying an automatisation of assembly lines when the Tahara No. 4 plant was constructed, the new concept became concretised in the assembly line at Toyota Motor Kyushu (TMK – a subsidiary of Toyota) (Fujimoto, 1999; Shimizu, 1995a). An assembly line was divided into about ten segments per function (eleven in the case of TMK). It was permitted to have a buffer between two segments, the buffer corresponding to around five minutes of operations. Thus, when a segment stopped due to a problem, the others could continue to work. The production efficiency of the assembly line as a whole was thereby unaffected, and the line stop system did not oblige the workers to work as much overtime as in the past, because the loss of time caused by stopping a segment was absorbed by the buffer. This meant relaxing the just-in-time rule under which having any buffer in the production line was regarded as a bad thing. As a result workers experienced less stress when they stopped the line. Moreover, a working group was operating in a segment where its group leader had greater power over his or her work organisation than before. The group leader was allowed to stop the segment during the time their buffer permitted in order to examine a problem (this is called ‘planned stop’), whereas previously the right to do this kind of line stop had belonged only to the manager (section chief ). In this way, working groups attained a certain degree of autonomy in their work organisation. For example, at TMK, a group could hold a meeting during working hours and not disturb the others, i.e. so far as the buffer allowed. Each segment also had a quality checking post so that not only a worker but also a working group as a whole assured quality. In order to make the work less intensive, an ergonomic method developed by the Production Engineering Division (TVAL – Toyota Verification of Assembly Line) was applied to measure the load of all operations (about TVAL, see Fujimoto, 1999). All operations, of which TVAL value was higher than a certain level, were considered as too heavy operations, and then dissolved. This method instituted Toyota’s criteria on hygiene and security following which the production process was improved. On the basis of this method, many ergonomic means were taken in order to make assembly work easier. For example, a large conveyer installed on the ground level permitted workers to operate standing on it and without walking much while the car body was moving, and especially without walking backward. Tasks involving carrying heavy parts also disappeared by installing automatic or semi-automatic equipment. This assembly line concept was realised when reconstructing old assembly lines under the budget constraints of each plant. After TMK, it was applied to Motomachi No. 2 plant (1994, but now closed), Tahara No. 1 plant (1995)
The Uncertainty of Toyota as a World Carmaker
79
and Motomachi No.1 plant (1996) in Japan and to the new Kentucky No. 2 plant (1994) in the USA, and so on. In such a way, Toyota was engaging in improving assembly work and ‘humanising’ assembly work during the 1990s. New production efficiency management and wage system After the discussion about production efficiency management, management promised the union not to push too much cost reduction by increasing the production efficiency or reducing workers. Rather it was assured that it would make more effort to cut the costs of materials and parts in the product design stage to a greater extent than before. At the same time, management realised the modification of production efficiency management. First of all, management renounced the unilateral management of production efficiency. The objective of production efficiency improvement henceforth was not imposed on the plants every semester by the top management: the plants themselves set their annual objective of production efficiency increase. In general, it was a manager who established his section’s objective. Then the plant director arranged the objectives coming from the sections. The Production Divisions Council that checked the objective presented by the plants occasionally modified this, taking account of the company’s profit strategy. After the approval of the various objectives thus adjusted, these became a kaizen norm of each plant in terms of production efficiency. Thus, the kaizen for increasing production efficiency has become autonomic. Secondly, the method of determining production efficiency was also altered, and became less constrained. Whereas the standard time had been fixed on the basis of the best standard time marked in the past, it was then fixed by measuring the time really required for workers’ operations three months after launching mass production. Concerning working hours, management decided after negotiation with the union to reduce the long annual working hours; a successive two shifts without night shift system (06:30– 15:15 and 16:15–1:00) was applied in 1995 as in TMK. This would contribute to reduce long periods of overtime. Management and the union also encouraged employees to use up all paid holidays, while previously they had been discouraged from doing so in order to keep their attendance rate high. Thirdly, the method of calculating the production allowance, now called ‘productivity allowance’, was altered for blue-collar workers, and this allowance was abolished altogether for white-collar employees. After a minor change in 1990, which introduced a grade allowance (GA, 10 per cent of standard wage) and an age allowance (AA, 10 per cent), by reducing the share of production allowance (PA) from 60 per cent to 40 per cent, Toyota radically changed its wage system in 1993. It then had two systems (BW means basic wage): for blue-collar workers, standard wage = BW (40 per cent) + PA (20
80
The Second Automobile Revolution
per cent) + GA (20 per cent) + AA (20 per cent); for white-collar employees, standard wage = BW (40 per cent) + GA (40 per cent) + AA (20 per cent). The reason why the production allowance had been paid to white-collar employees is that by doing so, Ohno wanted to manage their overtime work, which would lower the production efficiency of the company as a whole and then increase the total labour cost. The union could not criticise this system under the authority of Ohnoism, although white-collar employees felt it strange. On the contrary, the productivity allowance continued to be paid to blue-collar workers, because the union recognised it as remuneration for workers’ efforts to increase productivity, and then as a share of profits, realised through kaizen. However, the calculation of productivity allowance began to be revised so as to be reasonable and equitable. The production efficiencies were then classified by a group of homogeneous factories: a group for foundries, forges, stamping and sheet metal shops, a group for mechanical components, a group for body welding, painting and plastic moulding, and a final assembly group, because there was a difference in mechanisation level among these groups, which had affected their production efficiency. Within a group, the production efficiency of all sections was classified into three levels A, B and C
Table 4.1 2000
New grade system under the assistant-manager (rank 30) at Toyota from
Ancient rank (required minimum service years) 30 (29) 40 (–) 50 (25) 60 (22) 7A (18) 7B (15) 80 (10) 9A (5) 9B (1) 9C (0)
Administrative & engineering staff Qualification Upper professional
Wage grade AE1 AE2 AE3
Professional AE4
Production staff Qualification (rank)
Wage grade
Managerial position
Expert position
CX (30)
P1
CL
CX
– SX (50)
– P2 P3 P4 P5 P6 P7 P8 P9
– GL
– SX
TL
EX
EX (60) PO1(70)
Operations personnel
AE5 AE6 AE7
PO2(80) PO3(F0)
–
Notes: AE means administrative and engineering staff; PO, production operators; P, production staff. TL is ‘team leader’, which has been recreated in 2007 after being abolished at the beginning of the 1990s. In this new grade system, the required minimum service years to be promoted to higher rank have been abolished so that younger but competent employees can be promoted rapidly.
The Uncertainty of Toyota as a World Carmaker
81
from top to bottom, in order to determine their production allowance coefficient. This coefficient was applied to the sum fixed for each worker’s grade (hierarchical rank – see Table 4.1), and not to the individualised basic wage as in the past. All workers of the same grade in the same section received the same productivity allowance. It is clearly an allowance given as remuneration for collective work, and the character of production allowance as a shared profit has been retained even in this new system. This change in the production efficiency management coupled with the wage systems, undertaken in the first half of 1990s, is remarkable when we consider the fact that the production efficiency management had remained sacrosanct for the union from the 1950s until 1993. At the very beginning of the twenty-first century, Toyota put in place a brand new and simplified wage system with a new grade system (Table 4.1) in order to revitalise employees’ motivation: for blue-collar workers, standard wage = AP (30 per cent) + PA (20 per cent) + GA (20 per cent) + PA (20 per cent); and for white-collar workers, standard wage = AP (50 per cent) + GA (50 per cent) (AP means pay for the employee’s gained ability, which has replaced the basic wage, and increases every year after the evaluation (satei); PA is productivity allowance; GA, allowance related to the grade position; and PA, proficiency allowance, which has replaced the age allowance in 2004). Here was the radical change concerning the productivity allowance. Even after 1993, the productivity allowance coefficient was fixed every month so that the amount of productivity allowance changed from one month to the next according to the fluctuation of the production efficiency. This system created by Ohno has been completely abolished. Now, the productivity allowance efficiency is fixed as the average coefficient of the company as a whole, and for one whole year, so that the productivity allowance is calculated in such a way that the coefficient is applied to the amount fixed by wage grade. In consequence, the monthly wage, except for the overtime pay, now remains constant for a year. The basic idea of this new wage policy is that the wage has to be determined according to the ability the workers have acquired in their job; whereas their contribution to the company profits has to be remunerated as their increased bonus, which reflects then the contribution of the working collective as well as of each employee. This modification recalls the wage system at TMK, introduced in 1992: monthly standard wage = basic salary + salary related to hierarchical position, which was fixed for one year; workers’ contribution to the company performance, made through kaizen, was remunerated as a second bonus evaluated by PIT, the performance incentive of TMK (Shimizu, 1995a). The current productivity allowance is an echo of Ohno’s earlier production allowance system. To outside observers it looks as if Toyota has taken almost ten years in order to bid farewell to Ohnoist production efficiency management-wage systems.
82
The Second Automobile Revolution
Toyota’s reconfiguration of the TPS can be considered as a new strategy, undertaken in preparation for new working methods in the twenty-first century; it has freed Toyota from the earlier routines that were specific to its operations in Japan which were not easily transplanted into its foreign factories.
Birth of the new world number one carmaker and its uncertainty The way to a global company At the dawn of the twenty-first century, Toyota aimed to ‘operate as a truly global company’ with the aim of conquering 15 per cent of the world market share (‘Toyota Global Vision 2010’), in contrast to its rather timid and prudent approach during the 1980s and 1990s when it wanted to avoid the appearance of an aggressive company. According to Toyota (1987), its objective for coming years was fixed as the ‘global 10’, which meant conquering 10 per cent of the world market share. However, Toyota never pronounced this goal during the 1990s because of trade conflicts between Japan and the USA. At the start of the conflicts, and as a consequence, with the voluntary restraint of Japanese passenger car exports to the USA from 1981 to 1993, as well as with increasing foreign exchange risk in keeping with its exportcentred strategy, Toyota decided to localise its production starting with the construction of a joint venture with GM, the NUMMI (New United Motor Manufacturing Inc.) in 1984, and then establishing TMMK (Toyota Motor Manufacturing in Kentucky) and TMMC (Toyota Motor Manufacturing in Canada) in 1988. In the second phase, characterised by aggravation of the trade conflicts and long stagnation of the Japanese economy, Toyota proposed the ‘New Global Business Plan’ in 1994 for the years 1995–8, in order to alleviate the conflicts. This plan consisted of promoting localisation of production, increasing local procurement of parts, materials and equipments, and augmenting imports from abroad and exports from overseas plants to other countries (Shimizu, 2003). By doing so, Toyota succeeded in expanding its global production capacity. It constructed eight assembly plants in eight countries during the 1990s, including TMMUK in the UK (1992 was the starting year of production), TMMT in Turkey (1994) and TMMI in the USA (1998), and then a further eight assembly plants in six countries after 2000: TMMF in France (2001); three joint-venture plants in China (TFTM (2002), GTMC (2006) and SFTM (2005)); TMMBC in Mexico (2004); TMMTX in the USA (2006); TMMR in Russia (2007); and a joint venture with Peugeot SA in Czechoslovakia (2005). As a result, Toyota produced, in 2006, around 3.9 million cars with 105,315 employees in 32 assembly plants deployed in 26 countries outside of Japan; the share of production in North America in its overseas productions was 39 per cent, that in Europe, 20 per cent, and that in Asia, 29 per cent (see Appendix Tables 4.1 and 4.2). In consequence, Toyota had achieved a global production network in 2007.
The Uncertainty of Toyota as a World Carmaker
83
As a result, its localisation rate (production over sales) per region attained 55.5 per cent for North America, 72 per cent for Europe and 102.8 per cent for Asia in 2006. The rate for North America in 2000 had attained 62.5 per cent (Shimizu, 2003), so the figure of 55.5 per cent means that the demand had progressed faster than the increasing production capacity: Toyota’s sales surpassed Ford’s in 2007. This breakthrough of Toyota in North America would have been highly unpopular if it had happened during the 1990s. Proceeding with the ‘New Global Business Plan’, Toyota has increased its production capacity in North America, creating many new jobs. By adopting the ‘open door policy’ about the procurement of parts, materials and equipment, Toyota has increased the mutual trust between its plants and their local suppliers, and Toyota has also contributed to the local communities where its plants are located in order to be accepted by them as a ‘corporate citizen’. In a word, Toyota has tried to localise not only the production, but also its business as a whole in North America. In the process, Toyota has modified the management of its North American plants. At the beginning, TMMC was a wholly owned subsidiary of Toyota, where 80 per cent of the capital for TMMK was financed by Toyota Motor Sales USA (TMS USA) and 20 per cent by Toyota. This was because Toyota wanted to localise TMM USA by TMS USA’s reinvesting profits gained there. This localisation policy in the USA, however, led it to establish a holding company in the USA, which would be Toyota’s regional headquarters in North America. Firstly, Toyota Motor Manufacturing North America (TMNA) was established in 1996 as the company overseeing Toyota’s manufacturing facilities. Secondly, Toyota Motor North America (TMA) was founded as the holding company for Toyota’s manufacturing, R&D, financing, sales, public relations and marketing operations in Canada, Mexico and the USA, to begin its operation in 2000. And thirdly, in 2006, Toyota Motor Engineering and Manufacturing North America (TEMA) was established by merging TMNA and the Toyota Technical Center USA, to become Toyota’s R&D and manufacturing support centre in North America. This arrangement of the way to manage its operations in North America was necessary because the expansion of its operations made decision and coordination matters complicated, and was a way of regionalising the management of overseas operations. In a word, Toyota established the regional headquarters in order to manage the increasing complexity that resulted from its rapid expansion. After North America, followed Europe. Toyota established Toyota Motor Europe NV/SAS (TME) as the holding company and regional headquarters in Europe by consolidating three affiliates: Toyota Motor Marketing Europe (TMME), which, established in 1990, coordinated sales and marketing; Toyota Motor Engineering and Manufacturing Europe (TMEM), which, founded in 1998, coordinated the production of vehicles and parts; and Toyota Motor Europe NV/SAS (TME), established in 2002 as the holding
84
The Second Automobile Revolution
company in Europe. In Asia, which constitutes the third biggest market and second biggest production base outside of Japan, Toyota is preparing to establish the regional headquarters in Thailand, which will coordinate and control Toyota’s operations in Asia, apart from its affiliates in China, where they are joint ventures with Chinese capital. In this way, Toyota is on the way to decentralising the management of its localised production by establishing a regional headquarters and an engineering and technical support company for each of the three strategic regions: North America, Europe and Asia.
Consolidation of production regime in Japan, and industrial reorganisation In order to expand its global production network, Toyota reconstructed the management of its production regime in Japan. Around 45 per cent of vehicles produced in Japan with the Toyota brand (40.5 per cent in 1985 and 44.7 per cent in 1999) were manufactured by its assembly companies: Toyota Auto Body (TAB), Toyota Industries Corporation (TIC: ex-Toyota Automatic Loom Works), Kanto Auto Works (KAW), Central Motor, Araco (absorbed by Toyota Auto Body and Toyota Boshoku Corporation in 2004), Gifu Auto Body, Toyota Motor Kyushu (TMK), Daihatsu and Hino. Among them, TAB, KAW, TIC, Hino and Daihatsu are member companies of the so-called ‘Toyota Group’ that Toyota organises with its closest 14 companies such as TIC, Denso and Aishin Seiki. In addition, TAB, KAW, Daihatsu and Hino have product development capabilities: TAB has developed and produces mini-vans such as Hiace and Estima of Toyota, KAW has developed sedans such as the Mark X and mini-vans like the Raum and Ractis of Toyota, whereas Daihatsu and Hino develop and produce vehicles of their own brand, along with subcontracted cars of Toyota. All member companies of the Toyota Group, except KAW, Daihatsu and Hino, are Toyota’s mother company (TIC) and daughter companies, spun off from Toyota or founded by Toyota or the Toyota Group. In contrast with these companies, KAW, Daihatsu and Hino have a specific history with Toyota. KAW, founded in 1946 to produce electrical vehicles using the production technology of the Nakajima Aircraft Company (from which were also derived Fuji Heavy Industry (Subaru) and the Prince Motor Company, absorbed into Nissan in 1966), began to produce Toyota’s cars in 1949 because of difficulty establishing car production. Beginning with aircraft manufacturing competencies, KAW then developed R&D capabilities in car production as Toyota’s car body maker, designing car bodies whose platform had been conceived by Toyota, and developing and manufacturing cars with the Toyota brand. Having R&D capabilities, and as Toyota’s subsidiary (with the capital participation of Toyota Motor Sales into KAW in 1951, and that of Toyota Motor
The Uncertainty of Toyota as a World Carmaker
85
Co. in 1952), KAW constitutes one of the key bases of the development and production of Toyota cars, producing about 522,000 units between April 2006 and March 2007 with 5,609 employees. Daihatsu, founded as Hatsudoki Seizo Co. in 1907 for producing internal combustion engines (hatsudoki) and renamed Daihatsu in 1951, began to produce three-wheel vehicles in 1930 and small passenger cars in 1937. After the Second World War, Daihatsu became renowned as one of the biggest manufacturers of three-wheel vehicles along with Mazda during the 1950s. From 1960, Daihatsu developed a small passenger car, the Compagno (sedans and cabriolets of 800cc and 1000cc), but its business fell into difficulties around the mid-1960s, so that Daihatsu decided to rely on a business alliance with Toyota in 1967. Though assembling Toyota’s low range vehicles (the Sterlet, the Corolla and the Townace, for example) from 1968 and accepting managers dispatched from Toyota, Daihatsu kept its own commercial strategy by developing its own vehicles in mini-cars (small cars of under 800cc) and low range segments. Toyota did not have the intention of making it a subsidiary, but increased its shareholding up to 51.19 per cent from 33.4 per cent (15.4 per cent before 1995) in 1998. This was done so as to eliminate internal competition among the Group’s companies, to reduce production costs by the common use of parts and equipment, and to promote further technological collaboration in order to enforce the competitive edge of the Toyota Group in the age of mega mergers, triggered by the merger of Daimler-Benz and Chrysler. Renouncing the production of sedans (Altis is produced by Toyota), it produces under its own brand the mini-cars Toyota does not produce, supplying also to Toyota small cars such as the Probox, the Passo, the Porte, the bB, the Sienta and the Rush (SUV), sold also under its brand name. In 2006, Daihatsu produced 1,078,000 units of which 263,000 units were assembled on consignment, 24,000 units being OEM supplies to Toyota, with 11,209 employees (12,572 employees in 2007). Making Daihatsu its subsidiary surely facilitated Toyota’s operations in Tianjin, where Tianjin Automotive Xiali (TAX, actually Tianjin FAW Xiali) had been producing a sedan of Daihatsu, the Charade, under its own brand name, ‘Xiali’, the translation in Chinese of ‘Charade’. Hino, founded as an automotive division of Tokyo Gas Industry, consolidated into Tokyo Automobile Industry in 1937 (renamed Diesel Motor Industry in 1941) producing trucks and buses, and concluded a business alliance agreement with Toyota in 1966, after a downturn in the passenger car business it had begun with the production of Renault 4CV. Since then, Hino develops and produces trucks and buses under the Hino brand, and produces Toyota’s vehicles (passenger cars and commercial vehicles). Toyota’s share in its equity was low until the end of 1990s (11.0 per cent until 1997, 20.1 per cent in 1998) but increased to 36.6 per cent in 2000, and to 50.1 per cent in 2001, so that Hino is now also a subsidiary of Toyota, specialising in trucks and bus manufacturing under the Hino brand (about 100,000 units
86
The Second Automobile Revolution
were produced in 2006), and producing commercial vehicles such as Hilux, Dyna and the FJ Cruiser under the Toyota brand (9,980 employees in March 2007). Making Daihatsu and Hino its subsidiaries, Toyota has clearly demarcated their role in the production regime of the Toyota Group: in principle, Daihatsu specialises in the production of mini-cars and low range SUV and 2 Box cars; Hino specialises in truck and bus manufacturing. Consequently, Toyota has become the manufacturer of a whole range of vehicles, from minicars to the high range Lexus series, together with trucks, buses and industrial vehicles, and operates a well-designed division of labour among the companies of the Toyota Group, not only in terms of production but also in the development of Toyota’s vehicles. Toyota has also become a shareholder of Fuji Heavy Industry (Subaru, producing 482,000 vehicles with 12,801 employees in 2006) and of Isuzu Motor (producing 230,000 trucks and buses with 7,750 employees in 2006), acquiring a part of their equity, possessed by GM in difficulty, so that new business alliances were born between Toyota and them. As regards Subaru, Toyota acquired 8.7 per cent of its equity in 2005, and became its top shareholder. However, their relationship is still vague as yet. As Subaru vehicles compete with those of Toyota and Daihatsu, it can be anticipated that they will promote common use of parts and equipment as well as of platforms jointly developed. Indeed, interaction of engineers has already begun, and the Subaru plant in the USA (SIA) began, in 2007, to produce the Camry with a production capacity of 100,000 units per year, following their business alliance agreement, concluded in 2006. As for Isuzu, although the context is different Toyota acquired 5.9 per cent of its equity in 2006 to become its third largest shareholder, after GM sold out the shares of Isuzu it possessed. Isuzu had specialised in truck and bus manufacturing since 2002 after renouncing the production of SUVs in Japan, and had already agreed with Hino about the integration of their bus manufacturing business in 2003, and founded a new company, the J-Bus. It was this business cooperation between Isuzu and Hino as well as the R&D capabilities of Isuzu concerning diesel engines that led Toyota to conclude a business collaboration agreement: the collaboration of the two companies in the field of diesel engines and alternative fuels development. Relations between Toyota and these two companies have only just begun, so that it is too early to tell whether Isuzu and Subaru will be integrated into the division of labour of Toyota Group. In fact, their collaboration agreements advocate the mutual respect of the autonomy in management of each company. As the long history of Daihatsu and Hino in their relationship with Toyota shows, it is hard to imagine that Toyota will rush into making them its subsidiaries, and their relations are likely to evolve gradually, as in the case of Daihatsu. Nevertheless, Toyota’s participation in terms of capital in
The Uncertainty of Toyota as a World Carmaker
87
the two companies is a sign of radical change in industrial organisation in the Japanese automobile indstry.
Problems, caused by success With the rapid growth of its production in Japan a serious problem arose. There has been a shocking increase in the number of recalled cars: 970,000 units in 2003, 1,880,000 units in 2004 as well as in 2005, and 1,098,000 just in the first seven months of 2006, and they are mainly the vehicles produced after 2000. The reason for this is not known, but it might involve human factors as well as technical factors. The abolition of TL positions and teams during the 1990s, and increasing employment of temporary workers from 2000 (only 760 of 22,710 direct workers in 1999, but about 8,000 of 26,000 in 2004) may well be one of the reasons for the increasing numbers of recalled cars. As was the case at the end of 1980s, but without TLs and teams who could take care of newly hired temporary workers, the rapid increase in production might generate troubles in assembly work and parts manufacturing operations. This employment policy – the mass hiring of temporary workers in order to adjust the labour force to the fluctuations of production volume as well as to reduce labour costs – must be regarded as a big mistake. In order to solve this second labour crisis, Toyota decided to change temporary employees to permanent employees by selection: the number of the temporary workers becoming permanent was 590 in 2005, and 943 in 2006, and rose to 1,200 in 2007 (from the union’s data and Toyota’s job advertisements). Trained as permanent workers, they had to acquire higher technical competence necessary to assure the required quality. In addition, Toyota recreated TLs and teams in January 2007, so that TLs are able to support their 4 or 5 team members. The mistake taught Toyota to recognise the importance of teamwork and human relations on the shop floor. As for the technical factors, Toyota mentions defects concerning parts design, computerised control devices and materials used, so that the development and design stage at Toyota as well as its suppliers was also one of the reasons for the increased returns. Rapid global deployment might have taxed the engineers in R&D departments too heavily to sufficiently assure the quality of products. The cost reduction method newly adopted for enhancing its competitiveness might be also be a cause of these defects. In 2000, Toyota launched a project called ‘CCC21’ (construction of cost competitiveness), which aimed at cutting the cost price of a newly designed car by 30 per cent, by reducing the production cost of 170 parts procured from suppliers. This method did not meet Toyota’s traditional ‘cost cutting rules’ (Shimizu, 1999a), even if the reduction was realised by cooperation between Toyota and suppliers (simultaneous engineering). The imperative of such a considerable cost reduction might have resulted in insufficient quality assurance.
88
The Second Automobile Revolution
In any case, Toyota decided in 2005 a new way to reduce the cost, which is called value innovation (VI): instead of improving the design and production process of each part, it promotes an integration of parts starting at the very beginning of the design stage, which contributes also to simplifying the production process, and then, to increasing productivity. As a matter of fact, the name is new, but the method is not. In fact, it is an advanced feature of what Toyota tried to realise during the 1990s after the discussion held in the joint committee (see above, and Shimizu, 1999a). In addition to the quality assurance at the design stage, in 2006 Toyota reinforced the quality control on the production lines by installing instruments to detect defects. In this way, through its cost and human resource managements Toyota has steered towards a position of external flexibility at the beginning of the twenty-first century. However, what is remarkable is that the remediation has been so rapid: following the changes to human resources management as well as the cost reduction method, the quantity of recalled cars has been considerably reduced since 2006. However, the quality problem is not restricted to Japan. The globalisation of production has caused problems also at foreign plants: even if the technical production system is the same at all Toyota’s plants, the labour relations and working practices are different from one country to another, so that workers are not always able to assure demanded quality. Toyota was aware of this problem and established the Global Production Center within the Motomachi plant in order to educate the workers into the ‘Toyota way’ and to train the employees of its overseas affiliates. As we saw above, Toyota’s production efficiency management and wage systems radically changed in 2000, but new cost and human resources management is not yet fixed. The same may be said of the governance of its localised production: to become a ‘truly global company’ the overall localising of development and design, production, sales and service has not yet been achieved, but is in the process of construction.
Production total
1,516,969 1,759,204 2,021,529 2,308,098 2,165,186 2,421,305 2,426,983 2,611,291 2,865,025 2,862,613 3,251,290 3,251,471 3,158,547 3,182,718 3,376,224 3,540,646 3,652,211 3,599,174 3,854,692
Year
1970 1971 1972 1973 1974 1975 1976c 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988
Appendix Table 4.1
978,345 1,219,647 1,456,660 1,631,940 1,489,738 1,768,736 1,741,342 1,813,017 1,961,703 2,033,003 2,270,198 2,266,765 2,230,157 2,305,424 2,435,226 2,458,671 2,624,038 2,643,236 2,904,071
Passenger cars
In Japana
Toyota, 1970–2006
538,624 539,557 564,869 676,158 675,448 652,569 685,641 798,274 903,322 829,610 981,092 984,706 928,390 877,294 940,998 1,081,975 1,028,173 955,938 950,621
Commercial vehicles 481,892 786,287 724,552 720,640 856,265 868,352 1,177,314 1,413,235 1,382,174 1,383,648 1,785,445 1,716,486 1,665,793 1,664,361 1,800,923 1,979,955 1,875,763 1,770,937 1,815,721
Export (volume) – – – – – – – – – – – – – – – 136,307 152,524 192,260 244,371
Foreign production 40,365 41,024 41,447 42,927 44,228 44,584 44,474 44,798 45,203 45,233 47,064 48,757 51,034 57,846 59,467 61,665 63,890 64,797 65,926
Total workforce in Japanb 784,471 923,068 1,083,321 1,325,860 1,473,852 1,809,871 1,995,742 2,288,069 2,617,407 2,802,469 3,310,181 3,506,412 3,849,544 4,892,664 5,472,682 6,064,420 6,304,859 6,024,910 6,691,299
Turnover c
73,521 68,249 86,124 100,015 46,487 57,385 148,561 167,678 153,082 158,289 233,232 140,183 230,513 304,543 406,482 505,891 329,387 248,364 369,087
Gross operating profit c
(Continued)
81,403 76,373 96,242 110,550 65,687 93,305 184,276 210,120 198,803 198,334 291,580 227,511 306,183 398,592 521,767 648,009 488,385 398,008 521,706
Pre-tax profit c
4,006,796 4,028,149 4,118,885 4,033,357 3,856,614 3,508,456 3,171,277 3,410,060 3,421,729 3,086,559 3,212,630 3,422,314 3,354,400 3,485,168 3,520,017 3,680,946 3,789,582 4,194,188
Year
1989 1990 1991 1992 1993 1994 1995d 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
3,026,947 3,181,919 3,232,391 3,179,080 3,124,973 2,771,550 2,557,174 2,889,573 2,850,442 2,628,294 2,792,422 2,984,253 2,956,593 3,070,456 3,082,044 3,302,583 3,374,526 3,826,819
Passenger cars
In Japana
Continued
Production total
Appendix Table 4.1
979,849 846,230 886,494 854,277 731,641 675,347 614,103 520,487 571,287 458,265 420,208 438,061 397,807 414,712 437,973 378,363 415,056 367,369
Commercial vehicles 1,669,130 1,677,127 1,703,589 1,698,236 1,539,005 1,504,472 1,202,420 1,276,662 1,494,308 1,462,776 1,548,041 1,706,210 1,665,700 1,816,800 1,836,000 1,951,700 2,043,200 2,529,290
Export (volume) 471,581 677,055 669,912 764,466 888,715 1,051,271 1,253,344 1,346,033 1,390,071 1,468,271 1,611,040 1,751,315 1,780,300 2,155,221 2,557,979 3,042,728 3,571,160 3,898,975
Foreign production 67,814 70,841 72,900 75,266 73,046 71,573 69,748 68,641 70,524 70,663 67,912 65,290 66,005 66,820 65,551 65,346 64,237 65,798
Total workforce in Japanb 7,190,590 7,998,050 8,564,040 8,940,898 9,030,857 8,154,750 8,218,500 7,957,152 9,104,792 7,769,486 7,408,010 7,903,580 8,284,968 8,739,310 8,963,712 9,218,386 10,191,838 11,571,834
Turnover c
400,522 538,677 338,787 124,864 103,629 76,780 155,126 235,271 517,058 560,360 491,943 506,806 748,924 861,323 833,791 701,390 847,998 1,150,921
Gross operating profit c
569,863 733,803 574,318 375,862 286,448 214,034 236,205 340,734 620,412 625,640 541,824 562,105 768,920 1,055,134 892,496 831,235 1,104,781 1,555,193
Pre-tax profit c
5,002,731 5,275,213 5,404,216 5,982,966 6,513,791 7,231,976 7,711,647 8,180,951
Total
3,863,760 4,097,704 4,029,259 4,162,291 4,283,943 4,534,838 4,684,956 5,100,823
Japan
651,130 675,390 674,859 677,123 763,926 853,892 895,374 906,635
Daihatsu, Hino 1,138,971 1,177,509 1,374,957 1,820,675 2,229,848 2,697,138 3,026,691 3,080,128
Foreign production 5,182,774 5,526,863 5,784,917 6,246,156 6,719,363 7,408,378 7,974,563 8,524,659
Total
Sales
2,177,524 2,322,838 2,217,002 2,217,739 2,303,078 2,381,325 2,364,484 2,273,152
Japan market 3,005,250 3,204,025 3,567,915 4,028,417 4,416,285 5,027,053 5,610,079 6,251,507
Foreign markets 183,879 210,709 215,648 232,928 264,410 265,753 285,977 299,394
Total workforce 12,879,561 13,424,423 15,106,297 16,054,290 17,294,760 18,551,526 21,036,909 23,948,091
Turnover c 775,982 870,131 1,123,470 1,363,679 1,666,890 1,672,187 1,878,342 2,238,683
Gross operating profit c 750,501 864,129 1,113,524 1,649,318 1,765,793 1,754,637 2,087,360 2,382,516
Pre-tax profit c
406,798 471,295 615,824 750,942 1,162,098 1,171,260 1,372,180 1,644,032
Netprofit c
Notes: a The volume of production is for the whole Toyota Group (including subcontracted assembly, but excluding the vehicles of Hino and Daihatsu brands). Figures for passenger cars include estate/station wagon versions of passenger cars, and minivans. b The increase of the workforce from 1982 to1983 is largely due to the merger of Toyota and Toyota Motor Sales in 1982. c Million yen. d The accounting period changes in 1976. Prior to 1976, financial results were published twice yearly: the first six months from June to November, the second from December to May. After 1976, annual financial results were published for the period from July to June. Accordingly, the figures (except for workforce) for the 1970–4 period represent the sum of the two six-monthly periods, making the year June to May. The figures for 1975 represent the thirteen months from June 1974 to March 1975. e The accounting period changes in 1995. Therefore the figures (except for production, export, foreign production and workforce, which are adjusted) for the year 1994 represent only nine months from July 1994 to March 1995, and the figures for the period 1995 represent the twelve months from April 1995 to March 1996. f The consolidated accounts of Toyota are disposable from 1999, but the accounting standard changes in 2002 from the Japanese standard to the American. Sources: Toyota, financial reports.
1999 2000 2001 2002 2003 2004 2005 2006
Year
Production
Toyota, Consolidated accountsf
France Portugal UK Czech Rep. Turkey
Kenya South Africa
Europe
Africa
AVA TSAM
TMMF SC TMUK TPCA TMMT
TA TDB FDA TVCA
– 27.80
TME 100 27 TME 100 50 50
99 99.99 17.50 90
Mexico
Argentina Brazil Colombia Venezuela
TMA 100 TMA 100 TMA 50 TMMNA 1
TMMI TMMTX NUMMI TMMBC
Latin America
TMA 100 TMA 100
TMMC TMMK
Canada USA
Share (%)
North America
Name
Country
Aug. 77 Jun. 62
Jan. 01 Aug. 68 Dec. 92 Feb. 05 Sep. 94
Mar. 97 May 59 Mar. 92 Nov. 81
Dec. 98 Nov. 06 Dec. 84 Sep. 04
Nov. 88 May. 88
Start of production
Assembly plants of Toyota, outside of Japan
Region
Appendix Table 4.2
1,000 147,000
250,000 4,000 282,000 100,000 177,000
65,000 61,000 7,000 23,000
324,000 3,000 370,000 23,700
317,000 504,000
Production in 2006
326 9,789
3,936 341 4,800 3,504 3,625
2,543 2,488 1,288 1,708
4,660 2,017 4,924 792
4,574 7,725
Employees (March 2006)
Hiace, LC Corolla, Dyna, Hiace, Hilux, Fortuner
Yaris(Vitz) Dyna, Hiace, Optimo Avensis, Corolla Aygo Corolla
Hilux, Fortune Corolla, Fielder LC Prado Corolla, Dyna, LC, Terios, Hilux
Corolla, Matrix, RX330 Avalon, Camry, Camry Hybrid, Solara Tundra, Sequoia , Sienna Tundra Corolla, Tacoma Tacoma
Products
92
TMCA
34 firms
Source: Toyota (2007), Toyota Internet.
Australia
26 countries
AA TKM
Bangladesh India
Total
TMV
Vietnam
Daihatsu 12.50 95 52
PMSB IMC TMP KM
Pakistan Philippines Taiwan
100
– 89
70
– 86.40
Daihatsu UMWT 100
ADM ASSB
Malaysia
TWA TMT
50 49
TFTM PTTAM
Indonesia
Thailand
50 45.50
GTMC SFTM
China
Oceania
Asia
4,610 105,315
112,000 3,874,000
Apr. 63
110 3,174
1,029
466 11,928
1,695 1,606 2,486
2,992
9,543 5,143
3,884 1,803
44,000
14,000
60,000 469,000
28,000 35,000 14,000 139,700
63,000 54,000
209,000 60,000
61,000 16,000
Jul. 82 Dec. 99
Aug. 96
May 88 Feb. 64
Mar. 93 Feb. 89 Jan. 86
Feb. 68
Oct. 02 May 70
May 06 Dec. 05
Camry
Camry Coaster, LC, LC Prado, Prius Corolla, Vios, Crown, Reiz Crown, Camry, Corolla, Soluna, Dyna, LC, AVANZA Camry, Corolla, Vios Hiace, Hilux, Innova AVANZA Corolla, Hilux Camry, Corolla, Innova Camry, Corolla, Vios, Zace, Dyna Fortuner, Hilux VIGO Camry, Corolla, Vios, Wish, Hilux, Yaris Camry, Corolla, Vios, Hiace, LC, Innova LC Prado, Hino bus Innova, Corolla
93
94
The Second Automobile Revolution
Bibliography Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Ellegård, K., Engström, T. and Nilsson, L. (1992) Reforming Industrial Work: Principles and Realities in the Planning of Volvo’s Car Assembly Plant in Uddevalla. Stockholm: Arbetsmiljöfonden. Fijimoto, T. (1999) The Evolution of a Manufacturing System at Toyota. New York: Oxford University Press. Ikebuchi, K. (1990) ‘Nyusha 2 Nen-me no Wakamono wo Tetteitekini Kitaeru’, Kojyo Kanri, 36(9): 30–40 (in Japanese). Monden, Y. (1985) The Toyota Production System, IIE. Monden, Y. (1989) Jidôsha Kigyo no Cost Management. Tokyo: Dobunkan (in Japanese). Nomura, M. (1993), Toyotaism. Kyoto: Minerva (in Japanese). Ohno, T. (1988) Toyota Production System: Beyond Large-Scale Production. New York: Productivity Press; Japanese version (1978). Shimizu, K. (1995a) ‘Humanization of the Production System and Work at Toyota Motor Co. and Toyota Motor Kyushu’, in Å. Sandberg (ed.), Enriching Production. Aldershot: Avebury, pp. 383–404. Shimizu, K. (1995b) ‘Kaïzen et gestion du travail chez Toyota Motor et Toyota Motor Kyushu’, Actes du GERPISA, No. 13, Université d’Evry-Val d’Essonne, pp. 13–42. Shimizu, K. (1998) ‘A New Toyotaism?’ in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 63–90. Shimizu, K. (1999a) Le toyotisme. Paris: Collection Repères/Editions La Découverte. Shimizu, K. (1999b) ‘ “We Will Construct Our Assembly Line”: the Case of the Tahara No.1 Plant of Toyota’, (I) and (II), Okayama Economic Review, 30(3): 125–62 and 30(4): 303–31 (in Japanese). Shimizu, K. (2003) ‘A Maverick in the Age of Mega-Mergers? Toyota’s Global Strategies’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Indsutry? Basingstoke and New York: Palgrave Macmillan, pp. 119–43. Spear, S. and Bowen, H. K. (1999) ‘Decoding the DNA of the Toyota Production System’, Harvard Business Review, September–October: 97–106. Tanaka, T. (1993) ‘Target Costing at Toyota’, Journal of Cost Management, Spring: 4–11. Toyota (1987) A History of the First 50 Years, Japanese version. Toyota (1988) A History of the First 50 Years, English version. Toyota, Annual Report, every year from 1974 to 2007. Toyota (2007) Toyota no Gaikyo (Outlook of Toyota), Toyota Internet Drive.
5
Nissan: From the Brink of Bankruptcy Merieke Stevens and Takahiro Fujimoto
Introduction Nissan was founded in 1933 and became a globally competitive automobile producer during the 1960s. In 2006 Nissan had a global workforce of 186,336 employees and sold close to 3.5 million vehicles worldwide, produced at production sites in eighteen different countries. This chapter provides an overview of the period 1994–2007 which arguably has been one of the most eventful episodes in Nissan’s history. Finding itself on the brink of bankruptcy, Nissan took the bold step of entering an alliance with Renault in 1998, and placed far-reaching managerial power in the hands of its European partner. In the first section of this chapter some roots of Nissan’s financial crisis of the 1990s are suggested. These are Nissan’s intertwinement with the governmental bureaucracy that resulted in an insufficient focus on profitability (Stevens, 2008), Nissan’s tendency to acquire technology and productive facilities as packages from outside the company (Fujimoto, 1999) and the ‘dual management system’ (Hanada, 1998) in which the strong voice of the Nissan union influenced middle management decision-making. The second section of this chapter investigates the 1999 alliance with Renault, and looks at the successes of the alliance as well as areas that could be improved. The section concludes with a discussion of the wider applicability of the lessons learned from the Renault–Nissan alliance. The third and concluding section provides a discussion of the Nissan ‘production way’ (NPW), in which special attention is paid to the implications of NPW for Nissan’s labour force and Nissan’s product policy.
Historical overview This brief overview of Nissan’s history focuses on three developments that had an important formative impact on Nissan’s corporate character. The first is Nissan’s intertwinement with the governmental bureaucracy which 95
96
The Second Automobile Revolution
ensured relatively steady financial support and resulted in an insufficient focus on profitability at Nissan (Stevens, 2008). The second topic is Nissan’s tendency to acquire technology and productive facilities as packages from outside the company which stands in contrast with Toyota’s approach of piecemeal acquisition of elements of technology and reverse-engineering, and resulted in Nissan’s image of being a ‘technology-driven’ company in contrast to Toyota’s image as a ‘sales-driven’ company (Fujimoto, 1999). The third is the existence of a ‘dual management system’ (Hanada, 1998) that enabled the Nissan labour union to interfere with middle management decision-making. These three developments all contributed to Nissan’s near-bankruptcy in the 1990s, which is the fourth topic discussed here. Attention is also paid to the general economic downturn in Japan during the 1990s. Intertwinement with the government Through family connections and an entrepreneurial interest in supporting the build-up to the Second World War, Nissan is one of the companies that more than any other became intertwined with the Japanese governmental bureaucracy and consequently profited from financial assistance after the war (Stevens, 2008). One result of this situation was that Nissan did not prioritise profitability but focused on expansion and manufacturing excellence; an orientation ill-aligned with the reality of the automotive industry at the commencement of the twenty-first century. Nissan was not the only Japanese automobile producer that profited from governmental support during its growth period. In the 1930s the car industry in Japan literally had to start from scratch, while American and European carmakers had already been strengthening their capabilities for a few decades. Because no major restrictions on imports existed at the time, the Japanese car market of the 1930s was mainly supplied by foreign manufacturers. None of the Japanese companies that attempted to manufacture automobiles during the 1920s and early 1930s was very successful, and most government officials saw this as a reason not to promote the domestic automobile industry (Cusumano, 1989). During the 1920s, however, the Kanto earthquake, deflation and a financial crisis helped military officials, led by wartime leader Nobusuke Kishi, a relative of Nissan’s founder Yoshisuke Ayukawa (1880–1967), to persuade the government to promote the domestic production of trucks in order to enhance Japan’s economic and military potential (Van Wolferen, 1989). Under considerable army pressure and behind the scenes pushing by Ayukawa the government in 1936 accepted a law drafted by Kishi that ensured the protection of the domestic automobile market (Van Wolferen, 1989). When the business climate created by the military officials forced Ford and General Motors to leave Japan in 1939, Ayukawa was given permission
Nissan: From the Brink of Bankruptcy
97
to confiscate most of the assembly machinery of Ford’s plant in Yokohama and ship it to Manchuria (Cusumano, 1989), which had been colonised by the Japanese army for its rich natural resources. Ayukawa was interested in the business opportunities presented to him by the build-up to war and he played an important role in the industrial project in Manchuria (Iguchi, 2003; Johnson, 1998). In the eyes of the American occupiers, however, Nissan’s involvement in Manchuria was proof of extreme nationalism at Nissan’s top and after the war Ayukawa, like Kishi, was imprisoned as a war criminal and his influence on Nissan came to an end. Soon, however, the Cold War became a serious threat in the eyes of the Americans and they searched for strong leaders that could keep Japan out of communist hands. Consequently, many Japanese wartime leaders were able to resume their political careers and several of them joined the Ministry of International Trade and Industry (MITI), which played a leading role in directing Japan’s high growth ( Johnson, 1998). Nissan’s assistance to the military during the war placed it in a favourable position in the post-war rebuilding period and it popularly became known as ‘MITI’s darling’. During this time Nissan complied with the national goal of exporting goods in order to increase capital and to finance the import of raw materials into Japan (Hanada, 1998). However, as a result of the focus on the post-war national goals of export, expansion and manufacturing excellence, Nissan did not gain sufficient experience in following a profit-oriented business approach. When considering the 1990s crisis at Nissan, this orientation, which was in line with national policies, should be kept in mind. In this sense Nissan cannot be seen as a failed company; its emphasis on production rather than profitability during most of its history, however, appeared out of sync with the increasingly globalising automobile industry.
Inorganic growth From its inception in the 1930s onwards a tendency existed at Nissan of acquiring firms, technologies and production facilities as packages from outside. This clearly contrasts with Toyota’s approach of piecemeal reverseengineering and the organic absorption of technological elements into its products and systems, and is often presented as one of the main sources of competitive disparity between the two carmakers (Cusumano, 1989; Fujimoto, 1999). When Japanese entrepreneurs ventured into the automotive industry in the 1920s and 1930s for the first time, technology in Japan had not yet reached a level where it would have enabled them to compete successfully with earlier established European and American automobile manufacturers. Nissan’s founder Ayukawa therefore decided to rely on technology acquired from outside the company. He bought an existing Japanese car manufacturer
98
The Second Automobile Revolution
(DAT Motors), imported designs, engineers and an entire truck company from Detroit and put American engineers in charge of most of Nissan’s operations (Cusumano, 1989). This approach enabled Nissan to become the first Japanese carmaker to produce automobiles in-house on a machine-paced assembly line, and gave it an initial advantage over Toyota. In 1952, Nissan, with support from the government, concluded an agreement with the British Austin Motor Company to start the knock-down production of Austin cars in Japan. Governmental protection continued after the war and during the 1960s imported cars declined to only 1 per cent of new vehicle sales, at which level it remained for more than twenty years. A concurrent rise in domestic demand ensured that Japanese producers could increase their production and their income. Between 1955 and 1964 Nissan grew 35 per cent, and was able to improve the quality and performance of its vehicles (Cusumano, 1989). The two energy crises of the 1970s triggered a rapid increase in the demand for small Japanese cars known for their excellent fuel economy, and contributed to further growth of the Japanese carmakers. From the 1960s until the 1980s Japanese automobile producers increased their technological strength tremendously and certain characteristics of the Japanese production system, like Toyota’s just-in-time production, became renowned worldwide. Nissan at this time was known for its technological strength and boldness. However, while Nissan quickly absorbed the imported technologies and established its indigenous car designs by the beginning of the 1960s, the tendency of acquiring firms, technologies and production facilities as packages from outside continued to affect Nissan’s company culture, while Toyota excelled in step-by-step reverse-engineering and absorption of technological elements into its indigenous products and systems (Fujimoto, 1999). Nissan’s focus on technological excellence and expansion proved an adequate approach during the decades of an expanding market and governmental support. However, when demand contracted, first after the second oil crisis and later at the end of Japan’s high growth period, Nissan’s strategy of focusing on technology instead of customer demand and sales proved to be ineffective and Nissan found itself on the brink of bankruptcy with an accumulated debt of approximately 2 trillion yen in 1999. Nissan’s domestic market share decreased from approximately 30 per cent before the 1990s, to between 10 and 20 per cent during the 1990s. Rival Toyota also had a domestic market share of approximately 30 per cent before the 1990s; this, however, went up to approximately 40 per cent, and even reached 46 per cent in 2006. Toyota’s global market share in 2006 was close to 13 per cent, while Nissan and Renault together held 9 per cent. While Nissan had focused on the acquisition of technology and on improving its operational performance, Toyota became the world’s largest and most profitable carmaker in 2007, renowned
Nissan: From the Brink of Bankruptcy
99
for its engineering strengths and its operational excellence in combination with a focus on actual customer demand and sales.
Dual management system Before 1945 employees at Nissan were not organised in a labour union. When trade unions were legalised by the American occupiers, however, 4,800 Nissan employees organised themselves and joined the nationwide ‘All Japan Automobile Industry Labour Union’ (known as Zenji) (Cusumano, 1989). The relations between the union and the Nissan management were quite tense for the first couple of years, and in 1953 the Nissan management and the union experienced the biggest crisis in their history, known as the ‘hundred days war’, during which several of Nissan’s production facilities had to be closed for approximately three months. This crisis ended when a moderate section of the union and a group of Nissan managers formed a new, more moderate union. This second union was led by ex-banker Katsuji Kawamata who became informally known as ‘the emperor’ at Nissan. From its inception onwards the second union had a significant, informal impact on the daily management of Nissan, which Nissan’s top management allegedly attempted to decrease over the years. Some managers at the middlemanagement level, however, used the constraints posed by the influence of the union as an excuse for their slow decision-making and indecisiveness. This tendency is said to have been more harmful to management than the union’s actual interventions. There are other reasons why the post-1953 labour union system at Nissan has often been criticised. It allegedly had unopposed union candidates chosen by open ballot elections that were closely monitored by the Nissan management, and for most of the post-war period the union was dominated by a small group of labour officials who often moved up to managerial posts. Hanada (1998), for example, mentions a powerful ‘union clan’ that existed within the Nissan management structure. An example is founding member Kawamata, ‘the emperor’, who served as Nissan’s president from 1957 to 1973. According to Cusumano (1989) the union’s Labour Affairs Department and Nissan’s Personnel Department coordinated their activities to the point where the two departments became indistinguishable. The irregular relationship that existed between Nissan and its union is said to have contributed to Nissan’s crisis in the 1990s because it discouraged employees from being critical of how the company was managed, while preventing management from executing necessary workforce reductions. However, when in the Nissan Revival Plan (NRP) a domestic workforce reduction of 14,000 employees was announced, Nissan union members publicly prided themselves on their cooperative stance towards the management based on their understanding of the seriousness of Nissan’s condition. In the
100 The Second Automobile Revolution
spring of 1999 during the annual wage negotiations, the Nissan management had informed their employees about Nissan’s huge debts (presented as debt per worker) and the critical situation Nissan found itself in, and announced that a very drastic restructuring programme would commence soon. In response, the union stated that ever since the Nissan Zama plant had been closed in 1995, Nissan employees realised the extent of the crisis at Nissan and would therefore support restructuring measures. Crisis in the 1990s The 1990s are widely known as Japan’s ‘lost decade’ and Nissan’s financial crisis should not be separated from this context. When Japan’s high growth period ended, most sectors went through a recession, and many Japanese companies found themselves in dire need of reinvention in order to stay at the leading edge of the world economy. Price competitiveness and profitability had become more important than ever, while post-war rebuilding goals of expansion and export had become unviable. Some companies, like Toyota, were quick to catch the new wave or had been developing in a compatible direction previously; others, however, like Nissan, were complacent in the face of the challenges of the new century. Kumon (2003) called the state Nissan was in at the beginning of the 1990s ‘a paradox’ because Nissan had successfully shaped a global network and had achieved the highest productivity in developed countries, but nevertheless suffered from a financial crisis. This section has identified several reasons for this crisis: Nissan’s protection by the governmental bureaucracy that caused a misguided focus on expansion and manufacturing excellence at the cost of profit, Nissan’s tendency to acquire technology and productive facilities as packages from outside resulting in inorganic growth, and the ‘dual management’ system. These three aspects of Nissan’s company culture all contributed to the financial crisis of the 1990s which cumulated in Nissan’s near bankruptcy in 1999. As was shown above, Nissan had accumulated approximately 2 trillion yen in debt in 1999 while its domestic market share had decreased from over 30 per cent to between 10 and 20 per cent.
Revival: alliance with Renault On the brink of bankruptcy in 1999, Nissan started to look for an affluent partner to help it alleviate its debt. It first negotiated with Ford and DaimlerChrysler before signing a deal with Renault on 28 May 1999. It was agreed that Renault would pay 5 billion euros to Nissan and thereby acquire 36.8 per cent of Nissan stock. Nissan’s problems that had surfaced during the 1990s were mostly strategic and managerial, and therefore the changes that occurred at Nissan after the alliance were mainly organisational. Nissan’s productive
Nissan: From the Brink of Bankruptcy
101
capabilities were not in need of a drastic overhaul and in this area Nissan was able to assist its partner Renault. The alliance between Nissan and Renault was not the first one between a Japanese and a Western automotive firm. In 1979 Ford became Mazda’s largest equity holder by acquiring 25 per cent of its stock, but in contrast to the Renault–Nissan alliance, in the Ford–Mazda case there was no partner that acquired a far-reaching managerial say in the other, as is the case with Renault and Nissan. Also on a global scale the Renault–Nissan alliance was the first example of a European car manufacturer taking on important managerial tasks in one of the show cases of Japan’s extraordinary post-war economic growth. This section first discusses the successes of the cooperation and the complementarity between the two carmakers and identifies their cooperation as a ‘mutual learning alliance’ (Fujimoto, 2004). Secondly, areas for further improvement are discussed. In conclusion, we examine the wider applicability of the lessons from the Renault–Nissan alliance. Successes of the Renault–Nissan alliance Overall, the alliance between Nissan and Renault is viewed as successful in most of the areas that were targeted (Fujimoto, 2004; Ghosn and Riès, 2005; Magee, 2003). Especially, Nissan is in a much better shape than it was in 1999. Between 1999 and 2006 Nissan’s net sales almost doubled, while its net income improved from a 684 billion yen loss to a 461 billion yen profit in 2006. Nissan’s debt of 2 trillion yen was alleviated in 2001. Global sales in the seven-year period between 1999 and 2006 increased by almost one million units, while Nissan’s consolidated operating profit margin increased from 1.4 per cent to 9.2 per cent (all financial data come from Nissan corporate data). While growth at Nissan slightly stalled in 2006, these achievements are impressive considering the state Nissan was in. As was noted by Jetin in Chapter 3, up to now there is no car manufacturer that on its own has gained a significant presence in all global regions. Nissan and Renault, however, both had a strong regional base and were able to profit markedly from the complementarity of their respective geographical presence. Renault had a strong presence in both Western and Eastern Europe as well as in Latin America, while Nissan’s strongholds are in Asia and North America. From a product perspective there was also a good match between Nissan and Renault. Both are volume producers under their own brands, while Renault makes entry-level vehicles under the Dacia brand and Nissan produces luxury vehicles under the Infiniti brand. In addition, Renault has a strong presence in the commercial vehicle market, whereas Nissan also produces SUVs and light trucks. Before 1999, Nissan’s managers realised that something had to be done in order to avert the pending bankruptcy, but they lacked the vigour and boldness to embark on a radical new course. By handing over far-reaching
102 The Second Automobile Revolution
managerial responsibility to its foreign partner Renault, Nissan hoped that the necessary changes, of which awareness did exist within the company, would now also be executed. In order to make clear to everyone inside and outside the company that a new turn had been taken, the 1999 Nissan Revival Plan that stipulated the necessary reforms for the survival of Nissan was announced not by a Japanese manager but by Carlos Ghosn, Renault-sent COO and later CEO of Nissan. Ghosn’s presentation of the Nissan Revival Plan made it clear that he was willing to divert radically from the path Nissan had been following for the past decades. The reform measures specified in the NRP were drastic, and mainly consisted of reduction measures. Nissan announced reductions of 50 per cent in the number of suppliers (which later appeared overstated), 20 per cent in overall purchasing costs, 50 per cent in lead times and number of platforms, 30 per cent reduction in capacity, 20 per cent in administration and sales cost, 20 per cent in distribution subsidiaries, and 10 per cent in retail outlets. Shares in all except four of 1,394 keiretsu companies were sold, and additional income was generated by the closure of three out of seven domestic manufacturing plants, a centralisation of global financial operations, and the disposing of land, securities and non-core assets. Nissan products of the 1990s had the image of being oldfashioned and Nissan designers were requested to design more distinctive vehicles; in support of this change Shiro Nakamura who was known for his remarkable style was recruited from Isuzu Motors. An example of the effective targeting of managerial complacency at Nissan was the establishment of nine cross-functional teams (CFTs), each assigned to a specific topic, and each consisting of Nissan employees from different departments and different levels. The goal of the CFTs was to improve cooperation between the different departments in finding solutions for each of the areas that were addressed. A successful example is the improved communication between the purchasing and R&D departments that before the NRP was clearly lacking (Sugiyama, 2002; Takeishi, 2003). Sugiyama (2002), for example, mentions that subcontracting proposals of the R&D department before 1999 often met with resistance from the purchasing department if these included foreign suppliers or suppliers that were not part of the Nissan keiretsu. After the NRP, purchasing managers were put in charge of product categories and not supplying companies as before, a change that contributed to the technical expertise and comprehensive product overview of Nissan’s purchasing managers, and also decreased the chances of favouritism developing between Nissan and any specific firm. It should be mentioned here, however, that while at the Nissan headquarters there may have been a lack of cross-functional and cross-sectional interaction before 1999, at Nissan’s technical centres this was not the case. Fujimoto (2004), for example, argues that at Nissan’s product development centre in Atsugi, cross-sectional and cross-functional teams existed long before the NRP.
Nissan: From the Brink of Bankruptcy
103
In 2001 the fulfilment of the NRP was announced one year ahead of schedule. However, most of the financial restructuring measures like selling assets and land could only be executed once, and the much more demanding task of regaining lasting competitiveness required fundamental changes and continuous commitment from both partners. Clearly, this commitment could only be expected when both parties perceived the cooperation as strategically important and more attractive than existing alternatives. The complementarity of Renault and Nissan’s geographical presence and product line-up has already been mentioned, and it was suggested that the effective leverage of these complementarities was an important source of success for the alliance. Another key to success appeared to be the complementarity of their skills. In 1999 Renault was known for its design and corporate management, while Nissan had operational skills that Renault could learn from. And while the two companies retained their individual identities and brand names, they successfully shared these respective skills. It has been argued that acquiring skills through a partnership is more effective than acquiring them through the market because skills are mostly organisationally embedded and not easily diffused across firm boundaries (Kogut, 1988). An alliance offers a medium through which embedded skills can be transferred because interaction is mostly planned to be frequent and long-term. In addition, international partnerships are seen as having great potential for learning and valuable knowledge creation, because the less similar partners are, the more opportunities there are for learning (Dussauge et al., 2000). In cases where knowledge is transferred between partners with distinctive skills, mutual influences mostly develop gradually into new models of industrial organisation, in this case, the ‘mutual learning alliance’ (Fujimoto, 2004). A successful example of technological cooperation between Nissan and Renault is the cost savings realised through the sharing of vehicle platforms (wheelbase systems). In 2000 Nissan and Renault had a combined number of thirty-four platforms and the alliance goal was to reduce this to ten shared platforms in 2010, the first of which was to be a common platform in the small/medium segment. However, once the project was started, Nissan and Renault realised that there were fewer shared vehicle components than initially anticipated and that performance expectations between the two engineering teams differed (Segrestin, 2005). While cultural differences and language barriers added to the initial difficulties, an attempt was made to maintain the equality between the two partners as much as possible, which sometimes meant that specific components were designed separately. But while the project did not achieve all of its initial targets, several managerial goals were reached, namely, the standardisation of purchasing methods, information sharing and development protocol synchronisation (Segrestin, 2005). On 2 March 2002 sales of the
104 The Second Automobile Revolution
Nissan March, the first vehicle built on the shared B-platform, started in Japan.1 Areas for further improvement Looking back at the years that have passed since the announcement of the NRP in 1999, two areas for improvement can be identified. One is the commitment to volume in the NRP, and second is the pressure that Nissan puts on its suppliers. The first point is fairly straightforward: if a company commits to both a financial goal and a volume goal, a rigid situation is created that requires knowledge of future markets that mostly is not available. The second problematic part of the restructuring at Nissan – supply chain reforms – calls for a more detailed explanation. After initial research by the alliance board, it was soon discovered that Renault paid approximately 20–25 per cent less for its parts than Nissan, and in the NRP this difference was taken as the benchmark for cost reductions at Nissan. Price reductions often amounting to 30 per cent were requested from suppliers, and it was announced that 50 per cent of Nissan suppliers would be cut off; an announcement that later appeared to be an overstatement and rather than the number of suppliers, the variety of parts purchased by Nissan per supplier was decreased in order to maximise economies of scale (Takeishi, 2003). In addition to the vertical keiretsu of suppliers that was targeted for a drastic overhaul by Nissan, the horizontal keiretsu practice of holding shares in main suppliers was earmarked for similarly sweeping changes. Holding shares in suppliers had led to high fixed costs at Nissan, and it had made the financial well-being of suppliers one of its own concerns, which prevented Nissan from demanding necessary price cuts, generating a negative impact on Nissan’s own financial situation (Itoh, 2001). Before the NRP Nissan did not have enough leeway to articulate a supply chain management approach that effectively combined cooperation and competition. In the NRP, however, Nissan made it very clear that only suppliers who delivered their best performance to Nissan and fulfilled the cost reduction requirements could stay on after the NRP. For each supplier it was decided how much cost had to be cut – mostly a reduction between 20 and 30 per cent – and suppliers were initially given three years to accomplish their requirement. When the financial goals of the NRP were reached a year ahead of schedule, the next reform plan, N180, was announced which requested further, unexpected cost cuts from suppliers. This caused much discontent in Nissan’s supply base as many suppliers felt treated unfairly by Nissan, considering that the new measures came in the context of enormous financial progress at Nissan. Many suppliers believed that it was unfair that Nissan did not let them share in its improved financial situation, or offer sufficient assistance in achieving further cost reductions (a strategy mastered by Toyota). The result
Nissan: From the Brink of Bankruptcy
105
was that overall Nissan suppliers became reluctant to invest in their relation with Nissan. The alliance team had not fully recognised the actual competitive content of the pre-NRP supply chain in all its intricacies, and the quality and innovativeness of purchased parts diminished. In executing the NRP, Nissan essentially attempted to get ‘the best of both worlds’ from its supply chain: low costs as Renault had had before 1999, combined with the high quality of parts Nissan was used to. The attempt to maintain the successful aspects of close relations with suppliers (for example via early cooperation in product development) while at the same time disbanding others (for example cross-shareholding and operational assistance) is an entirely understandable approach to building Renault–Nissan’s new supply chain strategy. However, the alliance team soon had to learn that such a hybrid strategy could not be ‘engineered’ one-directionally. The Nissan management recognised it had gone too far in severing its supplier relations, and in an attempt to re-embed ties with suppliers that had become wary of Nissan’s seemingly never-ending cost cutting requirements, Nissan started what was called the ‘New Group Enforcement’ process, for which Nissan’s MC-AFL (Administration for Affiliated Companies) was set up in 2003 and which became part of the Nissan production way (NPW). That Nissan was serious about re-embedding ties with its suppliers became clear in 2005 when it acquired 42 per cent of CalsonicKansei’s voting shares and made it a consolidated subsidiary. In April 2001 the Renault–Nissan Purchasing Organisation (RNPO) was set up in order to facilitate least-cost purchasing for the alliance partners on a worldwide scale. In 2007, logistics, computer services and all parts for powertrains, chassis and most of the interior of Nissan and Renault vehicles were purchased through RNPO. Starting with the initial aim of 30 per cent of commonly purchased parts, 85 per cent of the parts for Nissan and Renault were purchased through RNPO in 2007. Parts that Nissan and Renault still purchase individually include steel, exterior parts and some cockpit functions (Nissan corporate data). In 2004, RNPO’s scope of responsibility was extended beyond the alliance’s initial three geographical zones (Western Europe, Japan and the United States) to include all zones in which Renault and Nissan own industrial manufacturing facilities. All RNPO decision-making is done centrally, and ultimately it is attempted to have one supplier supplying the same part to both carmakers in multiple regions. In order to improve ties with suppliers and to secure access to new technologies, RNPO adopted the rebuilding of ties with its suppliers as one of its goals. This attempt to re-establish supplier ties was supported by the fact that the overall production volume at Nissan generally increased in the period between the founding of RNPO and 2007. While it can be said with hindsight that Nissan went too far in its rigorous keiretsu dissolving attempt, two qualifications should be made here.
106 The Second Automobile Revolution
One is that considering the financial state Nissan was in there were very few options besides demanding severe cost reductions from its suppliers, and secondly, supply chain reforms at Nissan also had a positive outcome. For some suppliers that remained after the NRP, the cost reductions and stricter quality requirements presented an impulse for both operational as well as managerial improvements, which consequently resulted in increased cost competitiveness, an aspect of keiretsu restructuring that Ghosn has focused on in particular (Ghosn and Riès, 2005). Wider applicability of the lessons learned from the Renault–Nissan alliance The sheer scope of the overhaul at Nissan and the number of people involved at all levels in the company makes it impossible to pinpoint the key that led to the success of this alliance. Moreover, the unprecedented approach of a smaller party taking the larger share of managerial responsibility for a bigger party, makes it difficult to use the Nissan–Renault partnership as a blueprint for other alliances. However, general themes can be distilled from this experience that entail important lessons for strategic alliance formation. Renault and Nissan themselves take special pride in how they approached the management of a cross-cultural alliance, namely, by mutual respect and equality, rather than adhering to one dominant culture (Moulonguet, 2007). The open and exploratory approach of both partners, characterised by equality and the mutual transfer of different but complementary skills, resulted in what has been called a ‘mutual learning alliance’ (Fujimoto, 2004). In this relationship there is no clearly identifiable partner that only teaches or a partner that only learns; both partners are simultaneously teaching and learning from each other, their driving force being building best practice at multiple hierarchical levels (Heller and Fujimoto, 2004). In this the mutual learning alliance differs substantially from old-style mergers (Fujimoto, 2004). Another important lesson from the revival of Nissan is the importance of the institutional context in which companies are embedded. The competitive difference between Nissan and Toyota, for example, clearly shows that carmakers that share a domestic setting and certain organisational characteristics, such as early governmental protection and efficient vertical supply structures, can still diverge markedly in their developments as a result of how they exploit these conditions (Stevens, 2008).
The Nissan production way In order to assess how the alliance with Renault has changed Nissan, this third and final section discusses the evolution of the Nissan production way from its formulation in 1998 to 2007. Topics that receive special attention
Nissan: From the Brink of Bankruptcy
107
in this section are its implications for Nissan’s labour force and for Nissan’s product policy. The NPW as it was articulated in 1998 entailed two main statements of principle: the never-ending synchronisation of manufacturing with customer needs, and the never-ending quest to identify problems and put solutions in place (according to Nissan corporate data). Complementary to the Nissan production way, the Nissan Integrated Manufacturing System was designed that consists of: (1) a global standard flexible line (multi-model random sequence production), (2) the introduction of an additional model every one and a half months, (3) integrated production with suppliers, (4) increased adoption of modular assembly and (5) an ergonomic production line that lightens the workload for operators. The Nissan production way partly leverages the lessons of Nissan’s experience with global cooperation in its attempt to globalise most areas of the company. In 2007 Nissan executed management, finance, marketing, production, purchasing and employment according to global standards as far as possible. Labour implications When the Nissan Revival Plan was announced, some Nissan employees felt resentment towards the management because in their opinion the managers had made the workers produce cars that did not sell, so they should be the ones taking responsibility for Nissan’s crisis rather than putting the burden of restructuring on Nissan’s employees by reducing the workforce by 14,000. One aspect that Nissan union members and employees found particularly difficult to accept was the opening of new Nissan plants overseas and the increase of production at existing overseas plants. However, it was agreed with the management that employees would support reform measures as long as Nissan would continue to produce 1 million cars in Japan each year. Other changes that followed Nissan’s alliance with Renault included the move away from the lifetime employment system that had existed for a selection of white-collar employees at Nissan, and which had as its most prominent elements pay increases and promotion based on seniority, and welfare provided by the company. As a consequence of the lifetime employment system, the mid-career job market is underdeveloped in Japan, and while Nissan set up natural attrition schemes for the 14,000 employees that had to leave Nissan after 1999, this reduction nevertheless sent a shock wave through Nissan. After the NRP Nissan embarked on a redefinition of its approach to employment in order to move away from its traditional promotion system based on seniority, to a reward system based on merit. An incentive to deliver was built into wages and they became partly based on performance. Moreover, the importance of female employees was stressed in response to the growing numbers of female customers.
108 The Second Automobile Revolution
Product policy In the last decades a paradigm shift has occurred as to what a successful R&D approach for technology-intensive companies entails. While traditionally companies’ success was believed to depend on their internal technology generating capacity, increasingly the origin of technology has become less crucial to an organisation’s success. We have seen above that domestic recession and insufficient experience with prioritising profitability brought huge losses to Nissan during the 1990s, but with the assistance of Renault managers, the new millennium brought a turnaround and also a relatively secure channel through which to transfer technology. Technology transfer often invokes the image of the transfer of skills and knowledge from an advanced partner to a developing partner, but as was mentioned above, in the case of Nissan and Renault technology transfer took place between equal partners with different but complementary skills. The huge financial losses Nissan suffered during the 1990s made its highgrowth era approach to retain its image of technological champion an unviable strategy, and R&D expenditure joined the ranks of areas targeted by the revival plan reductions. One decision that followed from this was to not develop hybrid engine technology in-house but rather to purchase it from outside the company. Hybrid engines represent a disruptive technology that only around 2005 started to approximate conventional ICE vehicles in cruising range, acceleration and costs – the most important concerns for mainstream customers. Nissan had many good reasons not to develop hybrid technology in-house: hybrid engine development is very expensive; currently the market is very small because of the high price of hybrids; Toyota has a major presence in this field due to its ‘first mover advantage’; and it is possible that more radical powertrains such as fuel cells will take over from internal combustion engines rather than hybrids (Daisho, 2003). Regardless of the strategic viability of Nissan’s decision not to expend large amounts of its R&D budget in order to be a leader in hybrid technologies, the fact that previous technological champion Nissan was not at the forefront of this technology had a negative impact at least on its domestic image.
Concluding remarks This chapter began by arguing that the last decade at Nissan has been one of the most eventful in its history. From being on the brink of bankruptcy, Nissan took the bold step of allying with Renault and placing much of its managerial responsibilities in the hands of its European partner. The crisis at Nissan clearly did not represent a superficial or temporary lapse; patterns that were deeply rooted in the company caused the problems that arose during the 1990s. These patterns were partly caused by path dependency, for example Nissan’s participation in the build-up to the Second World War
Nissan: From the Brink of Bankruptcy
109
through family connections. Other patterns at Nissan resulted from clearly delineated programmes, for example the aim to expand and export in support of governmental goals, which later proved to be an obstacle when these goals were changed. Some patterns resulted from entrepreneurial vision that later appeared ill-aligned with global requirements for attractive products, such as Nissan’s acquisition of technology as packages from outside the company. Other patterns that caused the crisis at Nissan were the result of poor management, like the tardiness with which decisions were taken and the weakness of solutions that were devised for critical problems. The leverage that the Nissan union held over management decisions, resulting in the so-called ‘dual management system’, was another sign of managerial complacency. In 1998 it was decided at Nissan that an alliance with another carmaker would be a way out of the crisis, and as we saw above, this decision proved to be successful. The attraction between Nissan and Renault resulted largely from the complementarity of their geographical presence, their product line-up and their respective skills. The effective leverage of the complementary skills, however, is not self-evident and this arguably presents the main obstacle to the alliance. Two areas for improvement identified in this chapter are the commitment to volume in the NRP, and the pressure that Nissan put on its suppliers. While the scope of the changes at Nissan and the number of people involved in the revival of the company make it impossible to pinpoint the exact source of the success of the Renault– Nissan alliance, the important general lessons that can be distilled from this experience are the importance of mutual respect and equality, and related to this the acknowledgement of the importance of institutional context.
3,063,064 3,084,851 2,963,282 2,743,298 2,679,769 2,731,061 2,742,640 2,754,598 2,465,863 2,404,650 2,613,948 2,428,279 2,586,602 2,883,409 3,293,339 3,340,827 3,428,981
Global
2,379,634 2,323,720 2,036,664 1,749,814 1,589,393 1,676,947 1,662,776 1,671,510 1,528,461 1,336,918 1,313,527 1,272,851 1,444,314 1,475,063 1,481,563 1,364,868 1,191,937
Domestic
683,430 761,131 926,618 993,484 1,090,376 1,054,114 1,079,864 1,083,088 937,402 1,067,732 1,300,421 1,155,428 1,142,288 1,408,346 1,811,776 1,975,959 2,237,044
Total overseas
237,279 266,492 331,749 401,576 452,800 446,674 409,958 369,887 279,392 348,214 352,927 363,366 392,458 619,665 803,556 808,586 716,211
Source: Annual reports of Nissan.
3. Return on assets in percentage.
2. Consolidated net income in millions of yen.
91,448 131,723 211,814 214,662 228,205 209,687 248,026 277,509 275,993 286,865 332,532 296,788 297,719 331,924 319,652 315,297 384,669
North England America
Production (passenger cars and light commercial vehicles)
Nissan, 1990–2006
1. Consolidated net sales in millions of yen.
Notes:
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Appendix Table 5.1
Asia
North America
Europe
32,968 24,554 30,506 35,156 32,335 26,477 30,417 43,889 57,406 48,966 51,822 43,325 55,662 65,314 69,136 54,349 49,804
South Oceania America
959,120 78,364 380,980 358,186 24,618 964,139 72,997 357,800 363,256 35,132 900,463 101,183 313,805 304,696 35,047 629,990 85,345 260,480 156,590 21,870 611,215 62,907 314,713 129,949 22,211 593,597 64,077 294,129 146,950 14,462 598,244 72,237 284,253 122,036 16,334 711,392 82,463 295,023 144,972 29,779 710,845 39,394 247,053 197,705 28,724 611,990 61,874 293,884 117,450 26,924 604,866 84,345 284,342 86,313 23,441 577,119 59,517 259,427 58,938 24,905 693,743 73,720 314,790 89,480 42,654 719,165 112,480 256,714 116,293 53,768 721,839 92,792 256,509 131,250 67,539 673,680 67,496 256,147 92,290 73,249 617,384 45,134 238,034 76,978 80,222
Total
Exports
Middle East
14,214 69,790 16,585 93,815 15,744 99,482 12,313 58,236 7,910 41,190 8,050 39,452 8,523 64,444 16,219 99,047 19,520 121,043 7,769 55,123 7,020 67,583 15,483 114,781 18,564 98,286 21,233 91,228 18,102 83,999 28,166 99,498 32,472 92,572
Africa
1,378,329 1,318,111 1,179,002 1,067,130 1,080,793 1,125,333 1,131,331 967,169 861,411 760,139 732,582 702,657 792,767 799,206 819,152 810,968 716,405
Japan
Sales
N/A N/A N/A N/A N/A N/A N/A 137,201 131,260 141,526 133,833 125,099 127,625 123,748 183,607 183,356 186,336
Global
Sales1
Income2
Financial results
ROA3
55,326 5,645,169 48,831 4.16 56,873 5,964,912 101,000 4.74 55,566 6,417,931 −55,998 2.37 53,071 6,197,599 −86,914 −0.73 51,398 5,800,857 166,000 0.12 49,177 5,834,123 −88,418 −0.19 44,782 6,039,107 77,742 1.03 41,256 6,564,637 −14,007 2.51 39,969 6,580,001 −27,714 1.58 32,707 5,977,075 −684,363 0.41 30,747 6,089,620 331,075 −1.01 30,365 6,196,241 372,262 3.80 31,128 6,828,588 495,165 5.06 31,389 7,429,219 503,667 7.45 32,177 8,576,277 512,281 5.62 32,180 9,428,292 518,050 5.12 32,489 10,468,583 460,796 4.50
Japan
Workforce
Nissan: From the Brink of Bankruptcy
111
Note 1. Other vehicles built on the B-platform are the Nissan March, Micra, Note, Cube, Platina, and the Renault Twingo, Modus, Logan, Kangoo, Clio III/Sport/V6 Sport/ Thalia/Symbol. Renault’s partner Dacia also uses the B-platform for its Logan model.
Bibliography Cusumano, M. A. (1989) The Japanese Automobile Manufacturing Industry: Technology and Management at Nissan and Toyota. Cambridge, MA: Harvard University Press. Daisho, Y. (2003) ‘Recent Developments of Fuel Cell Vehicles and Related Issues in Japan’, The 2003 RIETI-Hosei-MIT IMVP Meeting, Tokyo 12 September 2003. Dussauge, P., Garrette, B. and Mitchell, W. (2000) ‘Learning from Competing Partners: Outcomes and Durations of Scale and Link Alliances in Europe, North America and Asia’, Strategic Management Journal, 21: 99–126. Fujimoto, T. (1999) The Evolution of a Manufacturing System at Toyota. Oxford: Oxford University Press. Fujimoto, T. (2004) Nihon no Monotsukuri Tetsugaku. Tokyo: Nikkei Business Press. Ghosn, C. and Riès, P. (2005), Shift: Inside Nissan’s Historic Revival. New York: Random House. Hanada, M. (1998) ‘Nissan: Restructuring to Regain Competitiveness’, in M. Freyssenet, K. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 91–108. Heller, D. and Fujimoto, T. (2004) ‘Inter-Firm Learning in High-Commitment Horizontal Alliances: Findings from Two Cases in the World Auto Industry’, Annals of Business Administrative Science, 3(3): 35–52. Iguchi, H. (2003) Unfinished Business: Ayukawa Yoshisuke and US–Japan Relations, 1937– 1953. Cambridge, MA: Harvard University Press. Itoh, R. (2001) Goon ga Idomu Nanatsu no Byou: Nissan no Kigyou Kaikaku. Tokyo: Nikkei Business Press. Johnson, C. (1998 [1982]) MITI and the Japanese Miracle: the Growth of Industrial Policy, 1925–1975. Stanford: Stanford University Press. Kogut, B. (1988) ‘Joint Ventures: Theoretical and Empirical Perspectives’, Strategic Management Journal, 9: 319–32. Kumon, H. (2003) ‘Nissan: From a Precocious Export Policy to a Strategic Alliance with Renault’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Magee, D. (2003) Turnaround: How Carlos Ghosn Rescued Nissan. New York: HarperCollins. Moulonguet, T. (2007) ‘The Automotive Industry in the Global Economy’, Cambridge Leadership Seminar presentation, 14 May 2007, Cambridge, UK. Segrestin, B. (2005) ‘Partnering to Explore: the Renault–Nissan Alliance as a Forerunner of New Cooperative Patterns’, Research Policy 34: 657–72. Stevens, M. (2008) ‘Foreign Influences on the Japanese Automobile Industry: the Nissan–Renault Mutual Learning Alliance’, Asia Pacific Business Review, 14(1): 13–27. Sugiyama, Y. (2002) ‘Transforming Knowledge into Strategic Action: an Explorative Study on the Process of Nissan’s Recovery’, paper presented at the European Group for Organizational Studies (EGOS) 18th Colloquium, 4–6 July 2002, Barcelona, Spain. Takeishi, A. (2003) Bungyou to Kyousou: Kyousou Yuui no Autosooshingu Manejimento. Tokyo: Yuhikaku Business Press. Van Wolferen, K. G. (1989) The Enigma of Japanese Power. New York: Alfred A. Knopf.
6
Honda: Serendipity or Strategy from 1997–2007? Denise J. Luethge and Philippe Byosière
Introduction Serendipity or strategy? What Las Vegas bookie could have set the odds that one day in Detroit, the ‘motor city’, a robot called Asimo, developed by Japanese automobile giant Honda, would conduct the Detroit Symphony Orchestra to ‘The Impossible Dream’ from the Broadway musical Man of La Mancha? Yet, such an event took place on 15 May 2008 in order to stimulate interest in mathematics and science among America’s youth. It is a sign of the changing times that with Takeo Fukui at the helm, Honda is back on the track of its innovative, creative and risk-taking roots that exemplified the life of its founder, Soichiro Honda. But not all is rosy at Honda: February 2008 saw the announcement of the closing of its motorcycle plant, opened in 1973 in Marysville, Ohio, despite having a US market share of 25 per cent. Honda also announced the closing of its motorcycle plant in Hamamatsu in order to consolidate heavy motorcycle production under one roof in Kyushu. However, on the other hand, in February 2007 Honda decided to open its aircraft headquarters in Greensboro, North Carolina, close to the site where the Wright brothers pioneered at Kitty Hawk, to manufacture the Hondajet targeted at an increasingly wealthy customer base in North America and Europe. These two major strategic decisions are strong indications that current CEO and Honda visionary Takeo Fukui is taking the company back to what founder Soichiro Honda envisioned as a global technological innovator pursuing new frontiers. This chapter covers the decade between 1997 and 2007 when Honda was under the leadership of three different CEOs: Nobuhiko Kawamoto (1990–8), Hiroyuki Yoshino (1998–2003) and Takeo Fukui (2003–present), a situation unprecedented in the history of Honda. The leadership changes were most strongly reflected in the up-and-down movement of Honda automotive. We leave it to the reader to determine whether Honda’s automotive position is the result of a well-thought-out strategy or serendipitous in nature. 112
Honda: Serendipity or Strategy? 113
Honda entered the automotive industry because of its knowledge and strength in two-wheeled vehicles and its vision to manufacture its products close the customer base. Without exception, Honda automotive entered all markets, mature or emerging, on the strong reputation of two-wheelers. The motto for Honda market entry seemed to be ‘motorcycles first, followed by cars’. This latest strategic decision regarding motorcycles and aircraft seems to be an indication that after a decade of leadership under Kawamoto and Yoshino, current CEO Fukui is putting Honda back on track by exporting heavy motorcycles from Japan and at the same time investing in light aircraft manufacturing in the US. Furthermore, Honda is not putting all its eggs in the China basket, but is also investing heavily in alternative fuels and technological development in Brazil as well as investing in the alternative market of India. With Takeo Fukui at the helm, Honda is experiencing a very different type of leadership from the Kamamoto and Yoshino eras that made Honda less of a radical and more of a mainstream Japanese company.
Brief history In 1948, Soichiro Honda incorporated the Honda Motor Company. After the Second World War, the need for inexpensive yet efficient transportation in a country with limited resources provided an opportunity for Honda to market a motorised bicycle. Joined by Takeo Fujisawa, Honda grew rapidly, expanding its line of products to scooters and motorcycles. Soichiro’s focus on technologically innovative engines and individual employee efforts combined with Fujisawa’s concentration on an organisational model designed to be fast, flexible and creative, helped to build Honda into a world leader in motorcycle production in less than two decades. Either of these individuals alone would not have been able to create the organisational powerhouse that now exists. It was only the combination of Soichiro Honda’s technological brilliance and innovative approach to design with the organisational genius of Takeo Fujisawa, with each deferring to the other in their area of expertise, that allowed Honda to grow from a garage shop where a motor was attached to a bicycle, to a major player not only in motorcycles, but also in the automobile and power products markets, becoming the world’s largest producer of engines today. In the beginning, Honda broke out of the mould of established Japanese companies in a number of ways. Unlike many Japanese firms that operated under more traditional values of a strong seniority system, group harmony and consensus, Honda favoured values of individual effort, promotion based on performance and expertise and delegation of authority to younger staff members who inclined towards new and innovative solutions to problems (Brewster, 2004). Unlike companies that had traditional hierarchies, Honda was relatively flat, with no wage gap between engineers and mechanics. Honda also differentiated itself by assuming total control over its distribution.
114 The Second Automobile Revolution
These differences from established organisations made Honda a bit of an outcast in Japan, making it difficult to develop keiretsu linkages for either finance or distribution. As a result, it was necessary for Honda to retain command of its distribution channel rather than to use the more common method of relinquishing control to trading companies (Mair, 1998). By the mid-1960s, Honda had become a global player in the motorcycle market. At the same time, Honda began producing cars in small numbers for the Japanese market. The first vehicles produced by Honda were a sports car and a mini-truck, both produced at the motorcycle facility. Unable to mass-produce automobiles at that facility, Honda built a new production plant dedicated to automobiles. Unlike Toyota and Nissan, Honda did not have a network of large retail outlets in which to sell automobiles. Instead, they relied on small motorcycle dealers to sell automobiles while developing separate support facilities for service, sales and resale. Eventually Honda developed three distinct types of dealerships: Primo, for selling economy cars, such as the Civic; Clio, for selling higher end cars, such as the Accord; and Verno, for selling sporty cars, such as the Prelude. Further technological innovations with the CVCC (compound vortex controlled combustion) engine and later the VTEC engine would bolster sales of Honda products worldwide, enabling Honda to become a major player in the global automotive market. Honda was not a leading player in the Japanese automotive market, in large part due to the activities of Toyota, Nissan, Mitsubishi and their keiretsus that effectively blocked much of Honda’s way into developing this market. As a result, there were few options for Honda but to focus upon exports quite early in their history (Ando, 2005; Ehrenberg, 2007). Later, Honda would emphasise production facilities locally, in part as a philosophy and in part to avoid tariffs and transportation costs cutting significantly into their margins. They would continue with this local focus by developing local sources of supply (Ehrenberg, 2007). Further cost-cutting measures, both in North America and Europe, would be the establishment of production facilities with non-union labour.
The North American market in the twentieth century Honda America Motor Company (HAM) began operations in 1959. Originally established to sell larger motorcycles of at least 500cc, the American market was only a tenth of the size of that in Japan. Importing three motorcycle models into the US – the Dream, Benly and Supercub (called the Honda 50 in the US) – sales were meagre at first. About this time, both the Dream and Benly were faced with defects that were causing the engines to seize up. Rather than try to fix the problems by sending parts from Japan, Honda instead chose to recall all of these products, return them to Japan, and re-establish the brand with new, improved products. Having only one product to sell, a lightweight
Honda: Serendipity or Strategy? 115
50cc product in a market focused primarily on larger bikes, Honda started targeting younger consumers and college students who could afford an inexpensive bike. With the reputation of motorcycles as dirty and dark, common among Hell’s Angels but not among average consumers, Honda needed to change the image in the American motorcycle industry. A national advertising campaign focusing on fun, simple and socially acceptable motorbikes, coupled with additional distribution through sporting goods and outdoor stores, greatly increased distribution. Meanwhile, Honda worked with motorcycle dealers throughout the US to project a different image, one with a higher level of customer service and professionalism, providing valuable lessons when Honda eventually expanded distribution of motorcycles to automobiles. When the N600 automobile was introduced in 1969–70, sales were made through current motorcycle distributors. When Honda realised that US consumers were hesitant to purchase automobiles through motorcycle shops, a new dealer network was established. Thus, Honda used lessons learned from the motorcycle market in the development of the automobile channel. Honda’s reputation as a customeroriented provider of motorcycles may have been helpful in the establishment of the Honda brand name, but it is unclear how much a strong motorcycle brand assisted in building a brand image for an unknown automobile manufacturer, especially given that US automobile sales traditionally occurred through manufacturer-supported automobile dealers. Sales of imported Honda automobiles continued to grow with the introduction of the Civic in 1974 and the Accord in 1976, in part due to the gasoline shortages and price increases that resulted from the 1973 oil crisis. As these were two of the few small, energy-efficient automobiles being sold in the US, sales skyrocketed, especially with the use of the CVCC engine that did not require an expensive catalytic converter to meet new emissions standards. A further oil crisis in 1979 led to yet higher gasoline prices, resulting in the development of US manufacturing in Ohio and subsequent production in Canada to serve better the huge North American market. The late response of the ‘big three’ US automakers to provide their customers with a small-car alternative bolstered Honda’s market penetration. As Honda continued to upgrade its products in terms of style and features, US manufacturers were slow to realise that small size and luxury were not an incompatible combination. This delay helped both Honda and Toyota to make significant inroads into the market share of US manufacturers, setting up the battle that exists today. In the early 1990s, changes began to occur at both Honda headquarters in Tokyo and HAM. In June 1990, Nobuhiko Kawamoto took over as the president of Honda, and he immediately began reorganising the company. In the light of a poorly received redesign of the Accord, Prelude and Acura Legend, Kawamoto instituted a new, American-style management orientation which focused on communication and decentralisation, placing marketing on a
116 The Second Automobile Revolution
par with R&D for the first time in Honda’s history. In order to coordinate manufacturing with R&D, Kawamoto appointed Shoichiro Irimajiri, president of HAM and the individual widely credited with Honda’s success in the US (Sanger, 1992; Taylor, 1991), as the executive vice-president and chief of R&D. Within two years, Irimajiri would leave Honda, citing health reasons. With Honda since the early 1960s, Irimajiri was a wunderkind within the organisation and a strong supporter of innovation, risk-taking and the racing spirit. The growing bureaucracy and ‘big company’ orientation of Kawamoto’s Honda was problematic for Irimajiri (Sanger, 1992). It was not until Kawamoto stepped down in 1998 that Honda would begin to focus more on engineering than marketing. By that time, however, Honda had become more like a large Japanese organisation (Sato, 2006). One telling sign of this was Hiroyuki Yoshino’s 2001 acceptance of the vice-chairmanship of the Keidanren, Japan’s federation of top business organisations, something that would never have been an option for Soichiro Honda.
The European market in the twentieth century The European market has been much less of a focus for Honda than the US market. At only one-fifth of the size, it has shown virtually no growth in the past decade. Honda’s approach in Europe has been one of caution and small steps rather than risk and innovation. In order to understand why, one must look at the history of how Honda developed from its entry into Europe in the late 1950s and early 1960s to the last decade of the twentieth century. The entrance of Honda into Europe took a very different path from that in the US, and the lessons learned were much more difficult. Honda entered into the European mindset when it began competing in the Isle of Man TT motorcycle racing events in the UK in the late 1950s. Originally importing motorcycles into Europe, the formation of the EEC and the resulting common tariffs on motorcycles forced the prices of Honda’s products to rise considerably higher than European competitive products. Deciding to establish production facilities in Europe to service the large motorcycle market, production began in the Belgian facility in 1963 for both mopeds and small motorcycles. Language issues between Japanese and Belgian staff, and production quality problems, particularly with mopeds, caused initial sales to falter. However, determination to succeed, on the part of both the Japanese and the Belgians, provided a strong base to grow the market and to learn lessons that would eventually help the Belgian factories provide production expertise for Honda’s entry into the European automotive market. Honda’s entry into motorcycles was much easier than its initial entry into automobiles. First, growth in the European automobile market was not as rapid as it was in the United States, which at the time was a much larger market. In addition, Honda was investing heavily in the US, so investment in Europe had to be limited to some extent so that the company was not spread
Honda: Serendipity or Strategy? 117
too thin. In 1979, Honda began a decade-long relationship with Rover, as the British company was having difficulty with labour, quality and productivity. As a result, Rover was looking for a foreign partner, and Honda, with its innovative technology and production expertise, was particularly attractive (Sebenius, 2001). In addition, Honda was actively searching for ways to ease its movement into the EEC, so the match seemed beneficial to both. When the British government sold Rover to British Aerospace, the relationship continued, and to some degree blossomed. During this time both firms had cross-licensing agreements and cross-board memberships, and Honda eventually took a 20 per cent ownership in Rover. Much of Honda’s technology was transferred into Rover vehicles (Adams, 2004), with Rover owning 20 per cent of Honda’s UK manufacturing facility (Button, 2005). By 1994, faced with mounting losses in a business that was really outside of its scope of operations, British Aerospace cut a secret deal with BMW to purchase Rover (Sebenius, 2001). Clearly such a deal was not in Honda’s best interest, as Rover had a great deal of Honda’s small-car technology, and Honda was not eager to share this technology with a strong European competitor. In addition, Honda did not want to be in a minority ownership position with BMW, and as a result, quickly sold its 20 per cent interest in Rover. This sale put Honda in a very awkward and difficult position, losing face in the abruptly ending relationship, but also losing technology that would fill product line gaps in the small-car market for BMW (Button, 2005). In addition to boosting BMW’s competitive offering, Honda lost its foothold in Europe, effectively setting the company back several years when it was just beginning to grow on the continent (Brull, 1994; Button, 2005).
Emerging markets in China, India, Brazil and beyond Consistent with Honda’s four-wheel push on the back of the two-wheel success in the US and Europe, Honda undertook the same strategy in emerging economies. This strategy allowed Honda to build a strong reputation and begin to understand the cultures and the markets prior to entering with a four-wheel vehicle. As Honda was expanding motorcycle production and introducing automobiles via export in both Europe and the US, exports of motorcycles to developing countries were growing. Unlike previous markets, the use conditions in developing countries were such that motorcycles often carried two, three or four people, with motorcycles hauling much heavier loads than those in Japan. Furthermore, routine maintenance was virtually non-existent, as customers literally drove motorcycles until they stopped working, without oil, filter and other regular replacements. The innovative history at Honda once again provided a boon to these markets as engineers developed a motorcycle with both easier maintenance and higher performance and with a new frame that could accommodate multiple riders as well as cargo. When Honda decided to expand to the Latin American market with
118 The Second Automobile Revolution
the strategy of producing products in the markets where they are to be sold in the mid-1970s, the same CG125 that had been developed for Southeast Asia was also produced in Brazil. The success of Honda production in Brazil eventually led to the development of the Brazilian R&D centre, which has now become a major contributor to Honda’s environmental strategy of alternative fuel usage vehicles with the alcohol fuelled vehicles popular throughout Brazil. Although Honda’s reputation as a motorcycle manufacturer may have had some impact in Europe and North America, Honda’s experience with motorcycles may benefit them much more in the emerging markets. First, the motorcycle market is substantially larger than the automotive market in these developing countries. A strong brand image may well translate across product categories. As the automotive market is just now developing, Honda may encourage former Honda motorcycle customers to ‘trade-up’ to a Honda automobile. This was not the case in Western Europe and North America, where the automotive market was well developed prior to Honda’s entry. Second, Honda has a strong reputation in the small-car market, with tremendous experience in producing products that focus more on local needs than on one global car. Finally, Honda’s focus and success in the areas of fuel-efficient and alternative fuel vehicles will appeal to a market that is more sensitive to the cost of fossil fuel. With the market for automobiles showing nearly flat or declining growth in the mature markets of North America and Western Europe in the near future, most automobile manufacturers, including Honda, are looking to emerging markets for increased revenues and profits. For Honda, this means increasing production capabilities in many Asian and Latin American markets. Recently, Honda has begun or increased automotive production in China, India, Argentina, Vietnam, Brazil, Turkey and Malaysia while at the same time beginning operations in Russia and Ukraine (http://world.honda.com). The fastest growing segment in each of these markets is the small-car segment, with autos priced under $10,000 or 10,000 euros (Sommerville, 2007). Honda is well positioned in this segment, although other Western and Japanese manufacturers who produce in the small-car segment will be threatened by the growing Chinese and Indian manufacturers. In India, automotive sales are already over 1 million units, and although the market is not growing as quickly as in China, it seems likely that doubledigit growth will continue. As the fourth largest producer in India, Honda’s production volume of 100,000 units still has a long way to go to reach the same sales volume as Maruti, Hyundai or Tata, each of which is producing in the 500,000–600,000 unit range. Still, profitability is very strong in India, with the industry average profit margins in double digits (Bissinger et al., 2006). In China, although quality levels from firms such as Chery and Geely are not at levels which can compete with traditional manufacturers, these firms have made great strides, and that learning will continue in the future.
Honda: Serendipity or Strategy? 119
The current market share of Chinese automotive manufacturers in China stood at about 25 per cent of all sales in 2006 (Sommerville, 2007), with these firms focusing on the low end of the market. As the number of manufacturers with licences to produce cars has increased dramatically in the last decade (Lasserre and Zeng, 2002), competition will also increase in the future. As quality levels increase, as they are sure to do, this percentage will only increase, especially given the cost advantages of Chinese firms. In fact, it is estimated that demand in China will increase to 20 million units by 2020, making it the largest automotive market in the world (Xu, 2006). Honda may have made a tactical error by entering China first with the Accord rather than with a smaller vehicle, banking on taking a significant portion of the highend market, even though it has limited appeal. Still, one cannot discount Honda’s strong image from decades of production of motorcycles in China and other emerging markets. In Brazil, Honda is still a niche player behind Fiat, Volkswagen, GM and Ford, but sales increased in 2003 and 2004 (Kim, 2007), and Honda has completed development of a fuel-flexible vehicle capable of using fuels with various alcohol levels – a significant market in Brazil.
The impact of shrinking traditional markets When Honda began producing automobiles in 1963, the Japanese market was the sole market and sole source of production. Although growth occurred rapidly, with Honda motorcycle dealers handling the sales of vehicles and separate service and research networks supporting dealers, Honda was not a major player. Over time, the Japanese market grew, and expansion abroad began with knock-down production in Taiwan and export of vehicles to the United States in late 1969. As production expanded, sales growth outside of Japan increased dramatically. Sales within Japan became a smaller and smaller portion of total sales, and strong competition domestically caused more of Honda’s profits to be generated from foreign operations. Looking at Appendix Tables 6.1 and 6.2, one can see that in the last decade, Japanese unit sales of automobiles have dropped from a high of 35.7 per cent in 1997 to 20.5 per cent in 2006. Sales in the United States reached a high of 53.2 per cent in 2004, but with limited growth in that market vis-à-vis other Asian markets, it seems likely that future growth will come from Asia rather than North America, Europe or Japan. Honda’s revenue and profit centre is the US market, and loss of market share in that market has been a significant problem. Further, declining sales in Japan continue to be a concern. Older, tired product lines in Japan and North America in the early part of the current decade have had difficulty competing with new offerings from Toyota and Nissan, further exacerbating the problem (Kunii, 2003). Limited product lines relative to competitive manufacturers, particularly in the early part of this decade, gave buyers fewer options. As a result, in addition to having to deal with slow growth or even
120 The Second Automobile Revolution
declining markets in its traditionally strong geographic areas, Honda also had declining market shares within those markets. In Europe, North America and Japan, declining birthrates, ageing populations and slow growth have led to lower turnover of automobiles, as older customers tend to keep their autos longer, and slow growth economies require consumers to put off high value purchases. With all of these factors slowing growth prospects in the developed world, Honda, like most other automobile manufacturers, has turned towards Asia and other emerging markets for future growth. One of the high growth areas that remain in traditional markets is that of hybrid vehicles. Honda’s initial entry into the hybrid market in the US was with the two-seat Inspire. Although a top quality vehicle with the best fuel economy of hybrid entries in this market, the limited seating and small size have not been nearly as attractive as the more practical, five-passenger Toyota models. By the time Honda entered a larger Civic into the market, the Toyota Prius had taken considerable market share. Further entry with the larger Accord was a tactical mistake for Honda, as they entered the market with a V-6 engine designed more for performance than for fuel efficiency. This particular product could not compete with the four-cylinder Prius in terms of fuel economy, and that feature had higher value to the market than performance. As a result, the Accord was discontinued after several years of poor sales performance. Honda’s entry into Europe and North America with the larger and more practical Hybrid Civic has been much more successful than with the Inspire; however, Toyota preceded that entry in both regions, so Honda must now face strong, well-entrenched competition in addition to building awareness and preference.
Strategy beginning in the twenty-first century Continuing its past strategy of manufacturing products where they are sold, Honda has begun the twenty-first century with the introduction of a plan for commonalising all of its production plants worldwide (Kobe, 2000). The long-term goal, of course, is the ability to build all of its products on one platform. In order to do this, staff from all regions have met to discuss process changes that would require tooling modifications in order to increase the flexibility of production at all manufacturing locations. At the same time, R&D is now linked with both production and suppliers in what Honda calls a ‘Digital Manufacturing Circle’, which is an IT process designed to integrate processes for product development, thus speeding products to market. Further technological innovation can be seen in Honda’s effective use of robotics, where consistency in quality levels can be maintained at much higher levels than in human processes. Concurrent with Honda’s focus on process innovation is a trimodal approach to controlling the environmental impact of vehicle emissions. The three approaches which comprise the main focus of Honda’s energy policy at the turn of the century are (1) continued
Honda: Serendipity or Strategy? 121
development of highly efficient internal combustion powertrains, (2) development and expansion of technology that combines gasoline engines with electric motors (that is, hybrids), and (3) development and testing of fuel cell vehicles and bio-diesel. Given the differences in alternative fuel vehicles in various markets, Honda has chosen to work simultaneously on technological developments in hybrid, diesel, alcohol and fuel cell vehicles. In addition to focusing on the preferences in each market, Honda is not banking entirely on one technology, as there is great uncertainty in the type and timing of the technology that will prevail. As a result, Honda continues to build its expertise in all of these areas, thus avoiding the problem of ‘betting only on one horse’, as it is not clear whether there will be only one type of technology that prevails in the future. Despite its passenger car success, Honda has been slow to expand, and even to enter, both the truck and SUV markets, missing this window of opportunity in the mid and late 1990s in the US market. Even now, Honda continues to play catch-up. The Ridgeway truck, built on the same platform as the Odyssey and sporting a V-6 engine, lacks the heavy towing capacity of the V-8s made by many of its competitors in the US market. Recent gains with the Odyssey and Pilot SUV are helpful, but gains in light trucks in the US are not sufficient to offset relative losses in cars. Overall sales are growing, but its share drops Honda to third place for Japanese manufacturers in the US, behind Nissan and Toyota. The product line in Japan has been problematic since the early part of the decade, as designs appear plain in comparison to those of competitors (Zaun, 2005). Honda hopes to remedy this problem with the introduction of the new Elysion and Edix, but Toyota in particular has been quick to match product offerings. In Japan, Honda hopes to add value to its customers by integrating its three distribution channels (Primo, Clio and Verno) into one channel. Further additions to the value proposition can be seen in the introduction of the Acura brand in Japan, beginning in autumn 2008, which focuses on newer technology, an area of strength for Honda. Honda’s approach since 2000 in Japan, with flat design and limited risktaking, is a further sign of the movement away from the spirit of its founder, to a more traditional Japanese company. With Fukui at the helm now, the Honda pendulum may be swinging back to technological innovation and risk-taking.
Key steps ensuring Honda’s success in the past continues in the future: strategy or seredipity? In hindsight, it is relatively easy to see that several key steps have allowed Honda to become a dominant player in the auto industry. Whether or not these steps developed as a result of deliberate strategic planning or serendipity is, in some cases, a matter of debate. Clearly, Soichiro Honda’s commitment to technological innovation from the earliest beginnings of the company was
122 The Second Automobile Revolution
part of a larger plan, conceived by Honda and brought to fruition with the assistance of Fujisawa. This commitment, combined with an obsession with developing processes and equipment from within, has positioned Honda as a technological leader in the industry. Clear breakthroughs, such as the CVCC and VTEC technologies, occurred primarily because of the technological orientation of the founders. Further strategic foresights in flexible manufacturing processes to enable multiple product production on any line using the Honda Production System (HPS) gave Honda a tremendous advantage. This, coupled with a limited product line that allowed platforms to be refitted as demand changed, gave Honda both economies of scale as well as demand flexibility. Indeed, it was this limited product line coupled with flexibility that drove Honda’s low cost structure, enabling market penetration globally much more quickly than the level of competition, especially large OEMs, would have predicted. However, other processes that have come to be associated with Honda may be less of a plan than a solution to an unforeseen problem. Entrance into the US market with the 50cc motorcycle, which had virtually no competitive product and allowed rapid growth, was more of an accident than a plan. Whether or not Honda could have been as successful in the larger engine motorcycle market, which had much stronger competition and a much worse reputation among consumers, is unclear. Certainly, growth would have been slower and the transition from motorcycles to automobiles may have been delayed. Quality control problems that limited Honda to selling only the 50cc products could have been a blessing in disguise. Eventual rapid growth of the Honda automotive brand in the mid-1970s also benefited from the oil crisis of 1973. Rapid gains in market share occurred as US manufacturers were illprepared to compete in the small, energy-efficient car market. Their delay in developing smaller, more efficient cars, combined with a lax attitude towards quality caused further deterioration of US automakers’ positions, especially when the 1979 oil crisis fuelled demand in the small-car segment. Honda’s experience with Rover gave Honda an entrée into the European market as well as a strong reputation as the technological benefactor of Rover (Button, 2005). Unfortunately, the corporate divorce left Honda losing face and hesitant to enter into collaboration with other European firms. As a result, Honda was forced to develop a new and aggressive plan to regain its standing (Brull, 1994), and that may have actually benefited Honda in the long run, making it more flexible and competitive.
Challenges for the future: solid technological innovation as the company’s core competency Several trends in the automotive industry are evident, and Honda will need to address them to continue its success in the automotive market. Clearly Honda is focusing on solid technological innovation as its main core competency,
Honda: Serendipity or Strategy? 123
and that strategic initiative should give it a basis for competitive advantage. However, virtually all automotive companies are stressing new and innovative technologies as a means of increasing market share. In order for Honda to be successful in this bid, there are a number of trends which it must address. First, the environment will continue to be of top concern in the industry (Shimokawa, 2003), particularly with regard to the reduction of carbon dioxide (CO2 ) emissions globally. Beyond emissions, there will be added focus on environmental protection, from the use of recycled materials to waste disposal and green factories. This trend is prevalent now, and Honda, along with most, if not all, other automotive manufacturers, has begun extensive environmental action programmes. Vehicles that attain higher fuel efficiency, in the form of diesel, hybrid and natural gas-propelled engines will continue to be developed and marketed. Further, more and more companies are involved in research to switch from all forms of fossil fuels to recyclable forms of energy, such as hydrogen and alcohol. Emissions reduction will be increasingly important in emerging markets, particularly in China. In 2005, there were approximately 6 million vehicles on the road. By 2020, that number will increase to 140 million, making reductions in greenhouse gases even more imperative (Gold, 2007). Honda has been successful in developing technology that addresses this area, starting research on low-emission vehicles in the 1960s, and continuing this innovation with the development of the CVCC engine in 1972 and the HVEC engine in 1988. Honda also began the Green Factory Project in 1997 and the Green Dealer Project in 1998 in an attempt to increase environmental sustainability throughout the organisation. The first Civic natural gas vehicle was unveiled in 1997, the first hybrid Insight was sold in 1999, the first fuel cell vehicle was sold in 2001, and the 100 per cent alcohol vehicle (flexible fuel vehicle) was developed in 2006. The next-generation diesel fuel vehicle, with the i-DTEC diesel engine, was due to be introduced in mid-2008, which exceeds both the Euro5 and proposed Euro6 emissions standards. As of today, there is no clear leader in environmentally friendly technology, and it is not clear which type of technology will rule the market in the future. Honda is hedging its bets by undertaking development in hybrid, diesel, fuel cell, natural gas and alcohol technology, but it is unclear whether the cost of doing this in the future will be productive. With no alliance partner to share the cost of technological developments, questions arise as to the sustainability of continuing innovation. Another trend in the automotive industry is the increasing use of advanced electronics and information technology for both safety and communication purposes. In the area of safety, technology such as anti-lock braking systems and side-impact air bags are options in many vehicles. Increasingly, these technological advances will become standard features. Further technological safety measures, such as technology that assists the driver in avoiding frontal, rear, obstacle and pedestrian collisions using inter-vehicle and vehicle-object
124 The Second Automobile Revolution
detection, night vision assistance warnings, anti-skid assistance and other safety devices such as automatic seat belt tightening will become part of vehicle offerings. Currently, downloadable storage devices are now standard for MP3 and DVDs, as are global positioning systems and bluetooth connectivity. As the technology continues to develop, additional types of communication will be incorporated in vehicles. One can compare the trends in automotive communication and entertainment technology to that of cellular telephones in the past. In the early and mid-1990s, cellular telephones were used to make telephone calls. In 2008, they are used not only for telephone calls, but for texting, gaming, email, internet communication, videos and even for funds transfer when purchasing products. Automobiles are no longer only for getting from point A to point B. Automobiles are fast becoming communication and entertainment centres that also transport people more efficiently and safely than in the past. The cost of developing and applying this technology is very high, especially as new technologies continue to make huge leaps quickly, and the technology incorporated into a product today may well be outdated in a few years or even in months. Staying at the cutting edge of technological development requires huge outlays of resources, in both human and financial capital. Over the past three decades, Honda has been a leader in technological developments in the automotive and motorcycle industries. From the first CVCC engines to cutting-edge fuel cell technology, Honda has developed both products and processes in-house to compete as a technological leader. Current technology in the areas of collision mitigation brake systems, intelligent night vision, pedestrian avoidance systems, intelligent drive support systems, super-handling all wheel drive systems (SH-AWD), advanced safety vehicle (ASV-3) technology for inter-vehicle communication using onboard cameras and radar and numerous technologies for fuel efficiency and reduced vehicle emissions have been developed by Honda engineers. A number of other automotive manufacturers are developing similar types of technology in conjunction with OEM and supplier partnerships. Honda is one of the few automobile manufacturers without research and capital sharing partners (Shimokawa, 2003). As a result, it may be stretched financially more than other firms should they continue to pursue new technological developments independently. In addition to automotive technologies, Honda is also involved in technological development in motorcycles, power products, jet engines and robotics, which will require additional human and capital investments.
Conclusion Among automotive industry scholars and analysts, Honda is best known for its excellence in quality in the design, engineering and manufacturing of its products and the ‘Honda way’ of exceeding its customers’ highest
Honda: Serendipity or Strategy? 125
expectations. In addition to simply serving customers, the Honda way also embodies an entirely different approach to the way people work. Probably one of the most representative differences between the Honda way and any other way is through the use of waigaya, which refers to the sound people make when engaging in lively dialogue. However, in an organisational context, it refers to thought-provoking conversations leading to discovery beyond expectation, innovation and the creation of new knowledge all tied together in an atmosphere of equality, trust and mutual respect (Pascale, 1996). Essentially, waigaya allows hierarchy-defying dialogue where participants listen, disagree, debate and reach consensus in an atmosphere of trust and respect, enabling great creativity and innovation while maintaining harmony and strong relationships (Holford and Ebrahimi, 2007). The excellence found in Honda’s innovativeness stems from its fervent commitment to pushing the boundaries and frontiers in its research endeavours, and waigaya is a significant means to that end. With the performance of the robot Asimo conducting the Detroit Symphony Orchestra, Honda is on a new path of technological innovation. Much credit goes to the scientists and engineers of Honda, but its new leadership must be credited as well. All successors to the Honda throne have come from its famous Honda R&D group, and despite mixed commercial success, the most successful have been R&D gurus. In addition to its technological excellence, in this chapter we have focused upon the leadership of Honda during the past two decades. This leadership has had mixed success. Although Honda has grown tremendously and become a major global automotive player, it has failed to become a marketing giant like its rival Toyota. Honda has also not been able to capitalise on its hybrid technology, giving Toyota a huge advantage, and one that may be difficult to overcome. Further, Honda’s recent failings to design updated and attractive models, with the Civic looking like a mini Accord, have caused significant market share losses at home, but more importantly, in Honda’s most profitable US market. All is not completely negative. First, Honda has had a local market focus with its product policy decisions, avoiding the ethnocentric approach that caused problems for other manufacturers. Second, Honda’s manufacturing strategy in the US market was a novel approach, and it has become a model for future production facilities globally. Irimajiri developed policies and procedures on the factory floor that emphasised individualism, communication and egalitarianism (Shook, 1988), establishing manufacturing in the spirit of Soichiro Honda. This model is likely to be equally effective in developing markets as well, and that can only benefit Honda as emergent market production may soon serve developed as well as local markets. Finally, Honda’s patience in Europe has begun to pay dividends, especially combined with a reduction in the number of European automobile manufacturers from thirty-six in the 1970s to fourteen in 2003 ( John and Schwarzer, 2006), favouring Honda’s growth in Europe in the near future. Also, increasing trade liberalisation
126 The Second Automobile Revolution
in this region should make it easier for OEMs with production locations in low-cost labour areas to increase market share (Bradsher, 2005). Clearly Honda is an automotive success story. In an industry characterised by negative profits, flat or declining growth in traditional developed markets and a race for technological superiority, Honda has been a shining star. With Takeo Fukui heading Honda for the last three years, much has changed, and this bodes well for the company. Fukui, an engineer and a businessman who understands the essence of Honda’s existence as a technological innovator, may well be the best person to turn serendipity back into strategy. Vehicles sales, revenue, net income, employees of Honda,
Appendix Table 6.1 1997–2006
Vehicles sales Revenue Net income Employees
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2184 5293 221 101
2343 6000 260 109
2333 6231 305 112
2473 6099 262 112
2580 6464 232 114
2666 7362 363 121
2888 7971 427 127
2983 8163 464 132
3242 8650 486 138
3391 9908 597 145
Note: Vehicles sales and employees are in thousands. Revenue and net income in billion yen. Source: Honda annual reports, 1997–2006.
Appendix Table 6.2
Honda Automobile unit sales (in thousands), 1997–2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Japan 779 757 690 706 776 878 849 716 712 North America 958 1096 1183 1295 1346 1368 1522 1558 1575 Europe 211 261 250 249 191 176 207 231 267 Others 236 229 210 223 267 244 102 137 176 Asia 205 341 512 Total 2184 2343 2333 2473 2580 2666 2888 2983 3242
696 1682 291 201 521 3391
Source: Honda annual reports, 1997–2006.
Appendix Table 6.3 1997–2006 1997
Honda Automobile unit sales as a percentage of total sales,
1998
1999
2000
2001
2002
2003
2004
2005
2006
Japan 35.7 32.3 29.6 28.5 30.0 32.9 29.4 24 22.0 20.5 North 43.9 46.8 50.7 51.6 52.2 51.3 52.7 53.2 48.6 49.6 America Europe 9.7 11.1 10.7 10.1 7.4 6.6 7.2 7.7 8.2 8.6 Others 10.8 9.8 9.0 9.0 10.3 9.2 3.5 4.6 5.4 5.9 Asia 7.1 11.4 15.8 15.4 Total 2184 2343 2333 2473 2580 2666 2888 2983 3242 3391 Source: Honda annual reports, 1997–2006.
Honda: Serendipity or Strategy? 127
Bibliography Adams, K. (2004) ‘Come Fly with Me: Life under British Aerospace’, Rover: the Whole Story, retrieved from http://austin-rover.co.uk/wschapter7f.htm, 26 February 2008. Ando, K. (2005) Japanese Multinationals in Europe: a Comparison of the Automobile and Pharmaceutical Industries. Northampton, MA: Edward Elgar Publishing. Bissinger, M., Wang, Y. and Kordyukove, T. (2006) ‘Emerging Markets Beckon World Carmakers’, Business Week Online, 22 March, p. 2-2. Bradsher, K. (2005) ‘Next Wave from China: Exporting Cars to the West’, The New York Times, 25 June, retrieved 17 May 2008 from http://www.nytimes.com/2005/06/25/ business/worldbusiness/25honda.html?scp=1&sq=bradsher+honda+europe+jazz& st=nyt. Brewster, M. J. (2004) ‘Soichiro Honda: Uniquely Driven’, Business Week, 17 August. Brull, S. (1994) ‘UK Expansion Targets European Market: Honda’s Go-It-Alone Plan’, International Herald Tribune, retrieved 24 February 2008 from http://www. iht.com/articles/1004/06/11/honda.php. Button, L. (2005) ‘Why Did BMW Buy Rover?’, ARONline, 7 March, retrieved 24 February 2008 from http://www.austin-rover.co.uk/index.htm?whydbbrf.htm. Ehrenberg, R. (2007) ‘Honda Gets It Right’, Seeking Alpha, 26 May 2007, retrieved 8 March 2008 from http://seekingalpha.com/article/36567-honda-gets-it-right. Gold, H. (2007) ‘China: Kingdom of Bicycles No More’, thestar.com, 15 January, retrieved 27 February 2008 from http://www.thestar.com/printArticle/170671. Holford, W. D. and Ebrahimi, M. (2007) ‘The Case of Honda: a Dialectical Yet Coherent Firm’, Proceedings of the 40th Hawaii International Conference on System Sciences. John, S. and Schwarzer, D. (2006) ‘Industrial Lobbying Within the European Union: Actors, Strategies and Trends in the Multi-Level System’, AICSG Policy Report, Johns Hopkins University. Kim, C.-R. (2007) ‘Honda to Tread Warily in Latin America Car Market’, Reuters, retrieved 17 May 2007 from http://www.reuters.com/article/companyNewsAndPR/ idUST32831420070906. Kobe, G. (2000) ‘The Future of Honda Manufacturing – Strategy and Planning’, Automotive Industries, October, retrieved 17 May 2008 from http://findarticles. com/p/articles/mi_m3012/is_10_180/ai_66218717. Kunii, I. M. (2003) ‘Honda is Ready for a Tune-Up at Home’, Business Week Online, 7 July. Lassere, P. and Zeng, M. (2002) ‘Guangzhou Honda Automobile Co., Ltd: Honda’s Entry into the China Car Market’, INSEAD Case Study. Mair, A. (1998) ‘The Globalization of Honda’s Product-Led Flexible Mass Production System’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Pascale, R. T. (1996) ‘Reflections on Honda’, California Management Review, 38(4): 112–17. Sanger, D. E. (1992) ‘In New Jolt, Honda Loses’, The New York Times Online, 19 March, retrieved 26 February 2008 from http://query.nytimes.com/gst/fullpage.html?res= 9E0CE7DC1E3FF93AA25750C0A964958260. Sato, M. (2006) The Honda Myth: the Genius and His Wake. New York: Vertical Inc. Sebenius, J. (2001) ‘Negotiating the Spirit of the Deal: Economic and Social Contracts in Longer-Term Agreements’, Presentation to the Center for Gender in Organizations, 2000–2001 Seminar Series, 30 January.
128 The Second Automobile Revolution Shimokawa, K. (2003) ‘Honda, an Independent Global Car Company: Out of the “Four Million Units Club”’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan. Shook, R. L. (1988) Honda: an American Success Story. New York: Prentice Hall. Sommerville, Q. (2007) ‘China Car Firms Gear Up for Booming Sales’, BBC News, retrieved 17 May 2008 from http://news.bbc.co.uk/1/hi/business/6364195.stm. Taylor, A. (1991) ‘A US-style Shake-up at Honda, Fortune, 20 December, retrieved 18 January 2009 from http://moner.cnn.com/magazines/fortune/fortune_archive/ 1991/12/30/75911/index.htm. Xu, H. (2006) ‘The Chinese Automotive Industry: Feasible and Sustainable Development?’ CEIBS Knowledge, October, retrieved 27 February 2008 from 222.ceibs.edu/knowledge/strategy/16113._3.shtml. Zaun, T. (2005) ‘Honda Tries to Spruce Up a Stodgy Image’, The New York Times, 19 March.
7
The Rebirth of Mazda Under Ford’s Shadow Daniel Arturo Heller
Introduction Mazda has a storied tradition in the auto industry (Shimokawa, 1994). While Mazda’s production of three-wheel motor vehicles for the Japanese market began in 1931, and four-wheel vehicle production began in 1960, Mazda is probably most well known for its successful development of the rotary engine in the mid-to-late 1960s. Yet as of the early-to-mid-1990s, Mazda was a severely struggling automaker burdened with poor profit performance, a heavy debt load, an inconsistent product message and an over-extended product line (see Hino, 2006: 177, 200). In the early 1990s, Ford Motor Company was called in by Sumitomo Bank to rescue Mazda as a ‘white knight’. Sumitomo Bank, which has traditionally been very closely allied to Mazda (Pascale and Rohlen, 1983), itself began to struggle from the early 1990s with the collapse of the economic bubble in Japan. Closer involvement of Ford in Mazda’s business was viewed as a means of providing a measure of protection for the large amount of loans Sumitomo had extended to Mazda. Ford had cooperated with Mazda since 1969 and been Mazda’s largest equity holder (25 per cent) since 1979. On the eve of the start of Mazda’s corporate distress in December 1991, Sumitomo Bank dispatched Yoshihiro Wada to become the president of Mazda. The initial assistance Ford extended to Mazda under Wada’s leadership was to acquire half of the equity of Mazda’s manufacturing operation in Flat Rock, Michigan (US) in 1992.1 As Mazda’s struggles mounted, Sumitomo Bank sought greater assistance from Ford. Prior to making a substantially greater commitment to Mazda, Ford conducted detailed due diligence at Mazda. This review was apparently conducted with the very real possibility that Ford might decide not to strengthen its relationship with Mazda. However, following a positive result, in December 1993 Ford and Mazda issued a joint declaration of a closer strategic relationship. This cooperative agreement, which was cleared by anti-trust
129
130 The Second Automobile Revolution
authorities, opened the way for the two firms to discuss product pricing and other sensitive business planning issues for the first time. In early 1994, the number of Ford-appointed directors on Mazda’s board was doubled to six. The newly dispatched directors were also given specific managerial responsibilities at Mazda, including the position of executive vice-president of Mazda with the responsibility of assisting the company president. Ford selected Henry Wallace for this post. He had formerly held assignments such as Ford’s finance director of Europe and president of Ford’s Venezuelan subsidiary. The joint declaration in 1993 and the dispatching of Ford managers in 1994 to work as internal directors at Mazda marked the beginning of the increasingly close integration of Mazda within the Ford Group. The ensuing arduous process by which Mazda recast itself and found a way to prosper within the Ford Group is described in this chapter.
1994–1997: cost-reduction focus The primary thrust of the restructuring pushed by the estimated thirty to forty Ford-dispatched upper- and mid-level managers at Mazda in the mid1990s was lowering costs. Ford’s influence in the restructuring process at Mazda grew even stronger when Wallace was promoted to become the eighth president of Mazda in June 1996. This move coincided with Ford increasing its equity stake in Mazda to 33.4 per cent. Nevertheless, despite a Forddispatched executive assuming the presidency of Mazda, it is not accurate to say that the firm was somehow completely taken over by Ford or essentially remade into ‘Ford Japan’.2 Both in Japan and abroad, naturally a lot of attention was generated by the promotion of a gaijin (foreigner) for the first time to become the head of a major industrial corporation in Japan. Some even unflatteringly portrayed Wallace’s promotion as similar to the arrival in Japan of Admiral Perry’s Black Ships in 1853 when Japan was in effect forced to open its borders to trade with the US. Against this backdrop, by necessity the cost cutting at Mazda in the mid-1990s had to be pursued with some measure of discretion. Finance-led restructuring The contingent of Ford managers assigned to Mazda, particularly the first wave of internal directors, had especially deep expertise in corporate finance. In addition to Wallace, two of the longest serving directors dispatched by Ford to serve at Mazda, Gary Hexter (from 1994 to 2000) and Robert Shanks (from 1996 to 2001), both had strong backgrounds in finance at Ford. Through managers such as these, Ford sought to infuse its long tradition of a strong financial orientation to its business (Abernathy, 1978; Pascale, 1990) into Mazda, notably by bringing a greater cash flow and asset utilisation perspective to Mazda’s business (Heller, 2001).
The Rebirth of Mazda
131
Hisakazu Imaki, the current president of Mazda, whom Ford personnel have long referred to as ‘Mr Manufacturer’, was quoted in a 17 September 2003 Financial Times article saying that he would like to make his new nickname ‘Mr Cash Flow and Profit’. Appendix Table 7.1 shows the notable cash flow improvements at Mazda that occurred in the mid-1990s. For Forddispatched managers it was especially encouraging, for example, when they observed Mazda proper management initiating their own discussions about cash flow. Similarly, the author noted in a discussion he had with two production line managers after a plant tour in early 2004 how they began to mention without any prompting by the interviewers the cash flow impact of this or that decision that they or their division head had made. Mazda also sought to improve its asset utilisation and management by selling off equity stakes held by Mazda in affiliated suppliers and other companies. For example, in 1999 Mazda divested itself of its 35 per cent equity stake in Jatco, an automobile transmissions supplier headquartered near Mount Fuji in Japan. Mazda’s annual corporate report for fiscal year 1998 lists thirty-four subsidiaries and affiliated companies. The following year’s report lists only seventeen. The selling off of long-held equity stakes was strongly resisted at the time and is deeply regretted even today by some Mazda managers. Impact on productive capabilities Mazda’s productive capabilities are generally said to be particularly strong in low-volume, high-variety manufacturing and modularisation, among other areas (see, for example, Kinutani, 1997; Abegglen and Stalk, 1985). Manufacturing at Mazda has been strengthened over the years by Mazda’s careful learning from Toyota in the 1970s (Pascale and Rohlen, 1983) and by other internal efforts to cope with what Fujimoto (2007) calls the intense ‘capability-building competition’ that characterises the Japanese auto industry. Faced with the strong presence of Ford-dispatched personnel within Mazda management in the mid-1990s, Imaki, Mazda’s head of manufacturing at the time, stressed to his colleagues that Mazda needed to develop its productive strengths with even more vigour. In a 2003 television documentary produced by NHK-Hiroshima, Imaki says that in the face of Mazda’s close tie-up with Ford, Mazda had to make everything open to its partner and yet at the same time ‘stay one step ahead’. In other words, now that Ford would be able to learn from Mazda with unprecedented ease the only way for Mazda to stay relevant and maintain its organisational independence within the Ford Group would be for Mazda’s productive capabilities to continually remain more advanced than those of Ford. Imaki’s thinking was shared by Wallace who, according to Mazda’s website, pushed the automaker to launch Mazda Digital Innovation (MDI) in 1996. MDI is an initiative to use digital tools to speed up and raise the quality level
132 The Second Automobile Revolution
of Mazda’s vehicle development processes, including reducing the number of physical prototypes and improving the link between development and manufacturing. Under MDI, Mazda is said to have developed, for example, the 1999 Premacy, a small minivan derived from the Capella/626, in only twelve months from design freeze, and the 2004 Verisa, an upscale Japanesemarket derivative of the Demio/Mazda2, without the use of any full-scale engineering prototypes. The MDI initiative, which continues to date, builds on Mazda’s long-standing engineering excellence, exemplified by the automaker’s rotary-engine development (Heller, 2005) and the numerous highly successful vehicles the automaker has produced over the years (e.g. the RX-7, Familia/GLC/323 and the Roadster/MX-5/Miata). In the late 1990s, Mazda also opened in Thailand its second large-scale overseas plant, which like its first large-scale overseas plant is also a joint venture with Ford. Planning for the Thai plant began in 1994, construction started in 1996, and production of 1-ton pickup trucks began in May 1998. This production replaced Mazda’s domestic pickup truck production of about 50,000 vehicles/year, nearly all of which were exported out of Japan. Thus, the new plant helped reduce Mazda’s exposure to exchange rate risk due to the firm’s chronically high export ratio of its domestic production.3 There was also a strong push in the mid-1990s to increase sharing of components across Mazda products. At the time a similar trend towards increased parts sharing was observed at other Japanese automakers, with Toyota and Honda being the earlier movers (Cusumano and Nobeoka, 1998; Hino, 2006). In 1996, a new Mazda vehicle, the Demio, which was received very favourably in the Japanese market and which won the RJC car of the year award in Japan, reused the engine and chassis (i.e. platform) of another much less successful Mazda model, the Revue. In addition to the push for greater commonalisation among Mazda vehicles, greater commonalisation was also sought between Mazda vehicles and vehicles of other brands in the Ford Group (Heller, 2005). In May 1997 it was announced that Ford and Mazda had come to a basic agreement on ‘cycle plan coordination’ (i.e. the timing of future product introductions) based on vehicle platform, engine and transmission sharing.
1998–2001: reorientation of product development (R&D) activities Wallace’s tenure as president at Mazda was unexpectedly short, as he left Mazda in November 1997 to become the chief financial officer of Ford. Ford’s selection to replace Wallace was James Miller, who had been the head of Ford’s joint venture facility in South Africa. Ford’s choice of Miller, who has been described as an expert in sales, indicates that Ford seems to have believed that the major work of the financial restructuring at Mazda was largely finished and it was time for Mazda to switch to a period of sales-led growth.
The Rebirth of Mazda
133
One means by which Mazda sought to grow was by having a more strategic orientation to its business. In a keynote address given on 15 October 2005 by Seita Kanai, the head of product development at Mazda, to the annual gathering of the Japan Academy of International Business Studies held in Hiroshima, he said that Mazda learned from Ford that such a strategic orientation entailed having a profit-making business model based on an appropriate brand strategy and greater strategic thinking by managers at Mazda. According to Kanai, in order to create a profit-making business model at Mazda, one of the first actions pursued by managers dispatched from Ford was to introduce Mazda, particularly the product development organisation, to a basic schema rooted in Ford thinking called ABS (affordable business structure). This way of thinking required vehicle development project managers (i.e. chief engineers) to pay careful attention to achieving an appropriate profit ratio, whereas previously their responsibility was described by Kanai as focused more narrowly on achieving the right balance between product cost and product competitiveness in the market. Kanai described how Ford-dispatched directors would ask him, a chief engineer at the time, about the expected volume of his product in each of the major markets, the projected dealer margin, the expected revenue, the anticipated warranty costs, the situation for a comparable vehicle at Toyota and so on. All of these questions he said he was unable to answer as he thought that he was not responsible for such issues. He explained how his Ford-dispatched superiors said to him, ‘You can’t answer these questions, and yet you call yourself a chief engineer!’ Kanai told how these directors dispatched by Ford said that under ABS the cost target comes from subtracting the desired profit from the expected revenue, given the market price, which is dictated by the targeted vehicle segment. (Incidentally, Ford’s ABS model seems quite similar to what Hino (2006) describes as common practice at Toyota.) Kanai said that the cost target that results from the ABS schema was vastly lower than what would have been arrived at using the previous development schema at Mazda. Nevertheless, this lower cost target reflected the market reality of what was needed to be competitive. This new focus on cost minimisation required more resources at Mazda to be devoted to cost reduction, something not welcomed by many engineers who were accustomed to spending most of their time working on developing new innovative features rather than reducing costs. In order to strengthen the strategic thinking of Mazda managers a widespread internal managerial education effort – the Mazda Business Leadership Development – was initiated at Mazda in 2000 under the theme ‘change or die’. Taniguchi and Nobeoka (2003) explain that since the mid-1990s it has become necessary for managers at Mazda to supply much more detailed back-up data to support vehicle profitability projections and other forecasts. As a result, it has apparently become much more difficult to do, for example, after-the-fact manipulation of approved projected sales to cover, say, cost
134 The Second Automobile Revolution
overruns in development, or to allow cost targets to be loosened, a practice which was apparently not uncommon at Mazda in the past. Under Miller and his successor, Mark Fields, who became president of Mazda in December 1999, rationalisation of Mazda’s domestic sales channels was strongly pushed and the number of Mazda showrooms and dealers was reduced. Nakanishi (2006) shows that the average number of vehicles sold per outlet at Mazda has gradually increased since the mid-1990s, even as Mazda overall domestic sales volume has remained flat. Definition of brand identity From the mid-1990s, Mazda explicitly sought to strengthen its brand, including the creation of a new corporate emblem for its vehicles in June 1997. Efforts began with an attempt to understand the current state of Mazda’s brand and to determine how the firm would like the brand to be positioned in the world’s major auto markets vis-à-vis other brands. A large team of managers, including a few dispatched from Ford, was charged with pursuing this brand-based initiative. Mazda personnel in all three of the automaker’s major markets of Japan, the USA and Europe worked together and in close cooperation with Mazda’s Japanese advertising and research agency on this effort. The team undertook a rigorous analysis of Mazda’s corporate and brand heritage and its past and current brand positioning in the world’s major markets. Results of this investigation revealed that Mazda’s brand positions in its three main regional markets were actually quite similar. As such, the decision was made to unify them in one global brand strategy, as is outlined in the Mazda communication below. Mazda distributed a card containing a statement of this new brand positioning to all its Japanese employees in April 1999. In order to globally communicate Mazda’s unique value, Mazda established a ‘World Wide Brand Positioning’ (WWBP) in April 1998, as part of the overall brand management strategy. The WWBP incorporates the Brand Personality of ‘Stylish,’ ‘Insightful’ and ‘Spirited,’ and the Product attributes of ‘Distinctive Design,’ ‘Exceptional Functionality,’ and ‘Responsive Handling and Performance’ to create the Mazda Brand DNA. Mazda’s DNA is summed up in the new brand message ‘Zoom-Zoom’ (love of motion experienced as a child), which is promoted in the major markets around the world. (‘Mazda in Brief’ booklet, November 2002) Top management at Mazda sought to make this brand philosophy the central theme of the entire firm. As such, Mazda sought to reorientate itself so that ultimately it would build and sell only those products that would support its new brand position. For this to happen, Mazda’s sales, marketing and product development activities had to be closely integrated with its brand development activities.
The Rebirth of Mazda
135
Introducing, diffusing and solidifying this new orientation throughout the Mazda corporate entity and its sales and service affiliates qualifies as an attempt at large-scale organisational change, which Beer and Nohria (2000) among others stress is a difficult undertaking in any organisation. It was particularly so at Mazda due the automaker’s long-standing tradition of its engineering, manufacturing and sales organisations having a relatively heavy weight in corporate decision-making. In fact, according to Mazda’s website, the firm’s marketing division did not even come into existence until 1996 with the consolidation of the research and strategic planning functions of the former domestic and overseas marketing and sales divisions. Further complicating the transition to the new brand-based orientation was what can be considered a winnowing of Mazda’s corporate aspirations that the new brand reorientation required. While the new brand message emphasised the sporty aspects of Mazda’s tradition, which are rooted in the firm’s position as one of the largest producers of mass-market sports cars, such as the Roadster/MX-5/Miata, at the same time this brand message may be seen as de-emphasising the long-standing family-car tradition at Mazda. It would seem that the decision to focus on vehicles of a sporty nature was made under an overall framework which dictated the end result must be compatible (i.e. minimise cannibalisation) with the rest of the brands in the Ford Group. Mazda’s attempts to express this new brand philosophy can be seen in the new and refreshed products the automaker released into the market after 1998. Some unified styling themes, such as a five-point grill, can clearly be seen. Internally, Mazda designated twenty-four product attributes for which Mazda aims to equal or better rival vehicles. However, the combination of the long lead times of product development in the auto industry and the amount of time generally required for large-scale organisational change to occur meant that there would necessarily be a significant lag in the time between Mazda’s statement of its new brand message and the firm’s attempts to express the message fully in its vehicles released into the market. Due to this time lag, Takahiro Fujimoto remarked to the author at the 2001 Tokyo Motor Show the glaring mismatch between the production vehicles displayed on the venue floor and the brand message projected by Mazda’s concept cars and audio-visual effects at the show. As Mazda sought to infuse its new brand message into its new vehicles, the automaker experienced a slow-down in the renewal of its product line-up. From the end of 2000 to mid-2002, a span of about eighteen months, Mazda had no major new vehicle launches in Japan. Leading up to this period, Mazda experienced a drop in its operating profit ratio in 1999 and again in 2000, when the firm fell into the red. Mazda’s modest operating loss in 2000 was dwarfed by the massive ordinary loss the firm incurred in the same year, as it chose to take a single large up-front hit for a pension contribution shortfall rather than string the loss out over a number of years.
136 The Second Automobile Revolution
Following the dramatic arrival of Carlos ‘Le Cost Cutter’ Ghosn at Nissan in 1999, cost reduction efforts at Mazda also began to be pursued in a more visible way. Notably, in 2001, a production line at Mazda’s Hiroshima plant was closed and a large-scale voluntary retirement programme instituted in 2001 for indirect employees in Japan resulted in more than 2,200 people leaving the automaker. During this difficult period of no new product launches and headline-making restructuring moves, Mazda was strongly criticised by the media, particularly the Japanese press. At the same time Mazda was being criticised outside the firm, internally Mazda was increasingly committing more of its resources to develop new vehicles that would fully embody its new brand message. Evidence of the additional commitment of resources can be seen in the increase in Mazda’s debt position in 1999 and 2000. According to interviews conducted in the early 2000s by the author with product development managers at Mazda, beginning around 1998 there was a gradual rebound in morale and work motivation at Mazda, as more investment money was put into developing vehicles that would build up the Mazda brand, whereas from 1994 to 1998, the primary focus of product development had been on the much less exciting task of cost containment. Thus, it may be said that from the mid1990s until the early 2000s, there were dramatically different views of Mazda depending on whether people were looking at the firm from an internal or an external perspective.
2002–present: the ‘zoom-zoom’ era It was not until 2002 that products which had been developed from their inception under the new Mazda brand orientation began to be released into the market. The first of this new generation of vehicles was a C/D-segment passenger car, the Atenza/Mazda6 that began to be sold in Japan in May 2002. This car and some others that have followed, namely the RX-8 and Axela/Mazda3 in 2003, the Roadster/MX-5 in 2005, and the CX-9 in 2007, have won numerous awards in markets around the world and have been well received by the public. The new products that began to be marketed from 2002 have been sold under a worldwide marketing slogan ‘zoom-zoom’ that was first used by Mazda’s North American operations in 2000 to market the Tribute SUV. According to Mazda press material, the slogan seeks to invoke the ‘emotion of motion first felt as a child’. Coinciding more or less with the start of this new slogan, revenue growth and profitability at Mazda recovered. Under this improved backdrop, Imaki was promoted to become president of Mazda in 2003, which marked the first time a Japanese national had been president of the automaker since 1996, following four successive Mazda presidents who had been dispatched by Ford. Imaki, Mazda’s twelfth president since the automaker’s founding in 1920, was also the first president at Mazda in over fifteen years to have spent his entire career at Mazda. The string
The Rebirth of Mazda
137
of presidents dispatched from other organisations to head Mazda began in December 1987 with the sixth president of Mazda, Norimasa Furuta, who had previously served in the Japanese Ministry of International Trade and Industry. Needless to say, the promotion of Imaki was very warmly received at Mazda. Shortly after his promotion, Imaki indicated that he did not intend to introduce any major changes to how Mazda was being managed. In his first five years to date as the president of Mazda, the automaker has kept the same ‘zoom-zoom’ marketing message. During this time, profitability and overall sales growth at the firm, while not remarkable, have been solid, with the glaring exception of Mazda’s lack of any significant volume growth in its own stagnant domestic market. Production capacity at Mazda has been gradually increased, such as through the reopening of the domestic production line that was closed in 2001. Mazda has entered a critical phase as it begins to launch the next generation of the vehicles that were released at the beginning of 2002. The new Demio/Mazda2 was launched in Japan in July 2007 and has initially sold well and received good reviews, including winning the RJC car of the year award. This small car went to market shortly after Mazda issued a declaration committing itself to ‘sustainable zoom-zoom’, and this new vehicle is nearly 10 per cent lighter than the model it replaced, with a corresponding improvement in fuel economy. Time will tell whether good sales performance will continue and whether subsequent new vehicles, such as the new 2008 Atenza/Mazda6, will also sell well.
Conclusion Following an extended period of restructuring pursued at Mazda throughout most of the 1990s, since 2001 Mazda has shown steady improvements in its product line-up, sales and profitability. With a clearer profit orientation, Mazda could utilise more effectively its long-standing productive strengths to achieve better business results. Nevertheless, as measured by profitability at least, Mazda has not yet reached industry-leading levels. Within the Ford Group, it is clear that Mazda’s position over the last fifteen years has evolved from that of a minor and lagging player to one of a still small, but increasingly important, and perhaps even leading, player. Contributing to Mazda’s success is the fact that the automaker has been able to maintain its productive capabilities at industry-leading levels even as it went through a long period of restructuring. Were it not for such strong building of its competitive advantage, it is unlikely that Mazda’s position within the industry and within the Ford Group would have improved as it has. Mazda’s enhanced performance would seem to be due in part to the fusion of Ford’s managerial strengths with Mazda’s productive strengths. The first great challenge facing Mazda today is figuring out how the firm can contribute positively to bringing about a turnaround at its ailing partner,
138 The Second Automobile Revolution
Ford, without unduly burdening itself given that Ford is many times larger than Mazda. The second and ultimately more important challenge facing Mazda is for the automaker to continue to use its productive capabilities to release strategically sound, appealing and profitable vehicles into the market. Mazda must continue to do so in order to avoid the management crisis that always seems to have hit the company when times have started to look good. Appendix Table 7.1
Domestic production (vehicles) Overseas production (vehicles) Exports (vehicles) Employees (non-consolidated) Sales (million yen) Debt (loans & bonds, million yen) Cash flow (million yen) Operating profit (million yen) Ordinary profit (million yen)
Domestic production (vehicles) Overseas production (vehicles) Exports (vehicles) Employees (non-consolidated) Sales (million yen) Debt (loans & bonds, million yen) Cash flow (million yen) Operating profit (million yen) Ordinary profit (million yen)
Mazda, 1995–2006 1995
1996
1997
1998
1999
2000
770,567
781,363
873,127
818,476
804,891
737,943
132,160
151,244
144,603
N/A
N/A
N/A
427,443 26,072
466,039 24,891
556,296 23,873
546,618 24,076
507,847 21,876
450,623 19,478
1,842,891 711,407
1,894,199 689,395
2,041,428 735,554
2,057,097 728,717
2,161,572 770,610
2,015,812 777,292
57,442
63,115
92,679
90,488
117,959
84,351
−8,234
266
32,993
62,510
25,111
−14,937
−11,879
−17,548
−6,802
38,707
26,155
−155,243
2001
2002
2003
2004
2005
2006
729,951
776,682
811,333
812,772
904,220
967,261
128,409
192,239
254,358
307,918
306,831
335,605
484,632 18,698
530,532 18,191
555,248 18,077
560,986 18,359
650,625 18,995
749,490 19,772
2,094,914 686,318
2,364,512 678,205
2,916,130 630,360
2,695,564 528,145
2,919,823 455,409
3,247,485 474,684
91,512
89,668
92,379
133,735
114,598
116,358
28,553
50,656
70,147
82,947
123,435
158,532
8,830
24,134
33,901
45,772
66,711
73,744
Notes: Data correspond to Mazda’s fiscal year of 1 April to 31 March; N/A = not available; cash flow is total cash flow from operating activities. From fiscal year 1999, listed companies in Japan were required to report their cash flow position on a consolidated basis. Data on Mazda prior to this year are only available on a non-consolidated basis. Sources: Nikan Jidousha Shinbun (Daily Automotive Newspaper), Mazda Yuukashouken Houkokusho (Report of Fiscal Listed Stock Companies), FOURIN Sekai Jidousha Houkokusho (Report of World Automobiles), Mazda’s Official Website, eol DB Tower Service.
The Rebirth of Mazda
139
Notes 1. Since its inception in the late 1980s half of this plant’s volume has been designated for the production of Ford vehicles, so Ford obtaining half of the equity of this venture may be viewed as a natural evolution of this plant’s development. Furthermore, Fucci and Fucci (1992) describe labour relations difficulties experienced by Mazda in Flat Rock (see also, Babson, 1998), which may have pushed Mazda to sell half of its equity stake and transfer much managerial control over the facility to Ford. One of the first actions taken by the plant’s new leadership after the facility was changed into a joint venture (JV) was to send many of the Japanese expatriates at the plant back to Japan. In recent years, AAI has become more like a typical 50:50 JV with shared managerial control. However, for a period after Ford acquired an equity stake in the facility, the plant was largely run as a typical Ford manufacturing plant. 2. For the following two reasons, the author contends that the Ford–Mazda relationship may still be viewed as an alliance (see also Heller, 2001, 2003; Taniguchi, 1998; Taniguchi and Nobeoka, 2003). First, Ford’s ownership stake in Mazda has never exceeded 33.4 per cent, and a Mazda Advisory Board has remained an active body where senior management of Ford and Mazda interact as equals. Such a board has not been created for other automakers in the Ford Group (e.g. Volvo, Jaguar/Land-Rover) that are wholly owned by Ford. Second, Mazda and Ford continue to maintain distinct organisational identities. Interviews by the author with over 50 Mazda and over 30 Ford employees between 1999 and 2006 revealed that employees in the two automakers generally perceive the people in the other firm as belonging to a closely allied, but nonetheless separate firm. Furthermore, people in both Ford and Mazda generally agree without reservation that Ford-dispatched executives and managers at Mazda do indeed work for the sake of Mazda and have on numerous occasions taken positions that go against the wishes of Ford headquarters. 3. In recent years, Mazda has sought to increase its overseas production volume by achieving better capacity utilisation at its large-scale plant in the US and through its production of the Tribute (a small SUV) at a Ford plant in Kansas City (US) and the Demio/Mazda2 at a Ford plant in Valencia (Spain). Mazda also has its vehicles produced in China under licence and at Changan Ford Mazda Automobile Company plants in Chongqing and Nanjing. However, stagnant sales volume in a depressed Japanese market coupled with overseas sales growth has contributed to a steady increase in the ratio of vehicles exported out of Mazda’s domestic plants from 60 per cent in 2000 to nearly 75 per cent in 2006.
Bibliography Abegglen, J. C. and Stalk, G. J. (1985) Kaisha, the Japanese Corporation: How Marketing, Money and Manpower Strategy, Not Management Style, Make the Japanese World Pacesetters. New York: Basic Books. Abernathy, W. J. (1978) The Productivity Dilemma: Roadblock to Innovation in the Automobile Industry. Baltimore: Johns Hopkins University Press. Babson, S. (1998) ‘Mazda and Ford at Flat Rock: Transfer and Hybridization of the Japanese Model’, in R. Boyer, E. Charron, U. Jürgens and S. Tolliday (eds), Between Imitation and Innovation: the Transfer and Hybridization of Productive Models in the International Automobile Industry. Oxford: Oxford University Press, pp. 161–88.
140 The Second Automobile Revolution Beer, M. and Nohria, N. (2000) ‘Resolving the Tension between Theories E and O of Change’, in M. Beer and N. Nohria (eds), Breaking the Code of Change. Boston, MA: Harvard Business School Press, pp. 1–33. Cusumano, M. A. and Nobeoka, K. (1998) Thinking Beyond Lean: How Multi-Project Management is Transforming Product Development at Toyota and Other Companies. New York: Free Press. Fucci, J. J. and Fucci, S. (1992) Working for the Japanese: Inside Mazda’s American Auto Plant. New York: Free Press. Fujimoto, T. (2007) [trans. B. Miller] Competing to be Really, REALLY Good: the Behindthe-Scenes Drama of Capability-Building Competition in the Automobile Industry. Tokyo: International House of Japan. Heller, D. A. (2001) ‘Thirty Years of Ford–Mazda Cooperative Relations: Capability Learning and Interfirm Ties’, The Annual Bulletin of the Japan Academy of International Business Studies, 7: 47–55. Heller, D. A. (2003) ‘An Inquiry into the Role of Interfirm Relationships in Recent Organizational Change Initiatives in Japanese Automobile Firms’, Shinshu University Economic Review, 49: 45–88. Heller, D. A. (2005) ‘International Strategic Alliances and Technology Strategy: the Case of Rotary-Engine Development at Mazda’, Shinshu University Economic Review, 52: 31–56. Hino, S. (2006) [trans. A. Dillon] Inside the Mind of Toyota: Management Principles for Enduring Growth. New York: Productivity Press. Kinutani, H. (1997) ‘Modular Assembly in Mixed-Model Production at Mazda’, in K. Shimokawa, U. Jürgens and T. Fujimoto (eds), Transforming Automobile Assembly: Experience in Automation and Work Organization. Berlin: Springer-Verlag. Nakanishi, T. (2006) Autos. Tokyo: Nihon Keizai Shinbun Sha (in Japanese). Pascale, R. T. (1990) Managing on the Edge: How the Smartest Companies Use Conflict to Stay Ahead. New York: Simon & Schuster. Pascale, R. T. and Rohlen, T. P. (1983) ‘The Mazda Turnaround’, Journal of Japanese Studies, 9(2): 219–63. Shimokawa, K. (1994) The Japanese Automobile Industry. London: Athlone Press. Taniguchi, M. (1998) ‘Restructuring Human Resources Management Under a Foreign President: the Case of Mazda’, Hiroshima University of Economics Keizai Kenkyu Ronshu, 21(2): 51–73 (in Japanese). Taniguchi, M. and Nobeoka, K. (2003) ‘Fusion Process of Different Management Models: Organizational Changes of Mazda after Strategic Tie-up with Ford’, Kobe University Journal of Political Economy and Commercial Science, 187(3): 1–17 (in Japanese).
8
Hyundai: Is it Possible to Realise the Dream of Becoming a Top Five Global Automaker by 2010? Myeong-Kee Chung
Introduction Global competition in the automobile industry and changing consumer demand are resulting in numerous trends: greater product variety and innovation; shorter product life cycles; low unit costs; and higher product quality. It has been theorised that changing market conditions precipitated a crisis of Fordism that then changed in fundamental ways to provide a new form of production system. This system is known for achieving greater functional and numerical flexibility by computer-based automated procedures. For the carmakers themselves, radical restructuring was called for. Two alternative paths to renewal were hypothesised. One was euphemistically referred to as taking the low road. This strategy relies on intensifying the existing Fordist order, engaging in work relocation and other union avoidance moves, and ruthlessly cutting costs (especially labour costs). The alternative was a high-road strategy in which firms use advanced human resource management practices to create high-performance work organisations (Knauss, 1998: 273). In the late 1990s, Hyundai achieved the level of economies of scale with a yearly production capacity of around 2 million units. At the same time, the maturity of mass production exposed the limitation of the Taylorist– Fordist model of manufacturing in Hyundai. Management faced the problem of how to manufacture an increasing number of cars through the raised wage and traditional authoritarian and paternalistic control at work. Managerial unilateralism gave importance to the crisis of the mass production model in Hyundai. Hyundai has achieved stable growth by an authoritarian labour–management relationship policy in the last three decades. However, what seems to be clear is that the traditional authoritarian and paternalistic approach in Hyundai will face more serious structural problems in the coming decade. There is no doubt that Hyundai has experienced a major crisis since late 1997 which was caused by the Asian financial crisis. This crisis may indicate 141
142 The Second Automobile Revolution
the need for fundamental structural change and globalisation because it cannot continue to depend on the domestic market for sustained growth. Nowadays, Hyundai’s globalisation is characterised by a transition from a worldwide export strategy to a multi-domestic strategy based on manufacturing sites in different regions. This is also done with the aim of maintaining Hyundai’s competitive position in the market. This chapter aims to provide preliminary empirical results on the new trajectory of Hyundai after the Asian financial crisis: its adoption of a new managerial strategy, and modularisation of the work organisation and globalisation. The first section discusses the development of the company up to the restructuring strategy in the wake of the Asian financial crisis and the export expansion of the late 1990s. It then describes in detail the company’s production system and industrial relations which were based on the module production system. Finally, the chapter examines the shift towards the establishment of a new globalisation strategy.
Managerial strategies of business restructuring since the Asian financial crisis The Kia merger, consolidation of R&D centres and platform integration The Asian financial crisis has brought an oversupplied capacity from the Korean automobile industry and the final bankruptcy of Korean carmakers except Hyundai.1 The future of the Korean automobile industry is dependent on the result of restructuring in the industrial sector. Mergers and acquisitions (M&A) is one of the most important measures for the reduction of the oversupplied capacity of the automobile industry. Hyundai emerged as the winner of the third round of auctions for the bankrupt Kia Motors Corp. and its sister commercial vehicle maker, Asia Motors Co., in December 1998. The Hyundai auto group has been restructured into two units – Hyundai Motors and Kia Motors – and each unit will maintain its own brands for a long period of time. Hyundai Mobis (which produces four-wheel drive (4WD) vehicles and vans) and Hyundai Motor Service Co. Ltd. were both merged into Hyundai Motors in 1999 to upgrade services for both of their auto units. Meanwhile, Hyundai Motor Service has transferred its heavy machinery sales division to Hyundai Heavy Industry Co. At the same time, this group is likely to merge Kia and Asia Motors Sales, as well as Kia’s sister carmaker Asia Motors Co., with the acquired Kia Motors. Hyundai Motors will service 40 per cent of the total 1.1 trillion won ($880 million) needed to take over the bankrupt Kia and Asia Motors together, while other Hyundai subsidiaries including Hyundai Heavy Industry Co. will be responsible for servicing the remaining 60 per cent.
Hyundai: a Top Five Global Automaker? 143
After the acquisition of Kia, Hyundai outlined their three principles of a new management plan: performance management, global competitiveness with independent technology, and customer-oriented management processes. Part of the reason for their confidence is the decision to maintain the Kia and Hyundai brands in all markets. Although overlap of models will be reduced by the removal of some redundant products from both ranges, the two brands will continue to maintain their own identities. Significant savings are already being realised by the integration of research and development activities. Prior to the consolidation, Hyundai had four R&D centres – the Namyang and the Ulsan R&D centres to develop new passenger car models, the Mabukri R&D centre to develop engines and transmission, and the Junju R&D centre to develop commercial vehicles. Kia also had four R&D centres or departments – the Sohari Central R&D centre to develop passenger cars and core auto parts, the R&D department of the Kwangju plant to develop commercial vehicles, and the R&D departments of the Hwasung plant and the Shiwha plant to make and test pilot cars. The Joint R&D Division, formed in March 1999 and led by the CTO (chief technology officer), set out a reorganising plan to consolidate the R&D functions for developing passenger car models and core auto parts into the Namyang R&D centre, while integrating R&D units for commercial vehicles into the Junju R&D centre. The consolidation of R&D centres basically targets the integration of platforms between the two companies. Due to the desire among Korean manufacturers to be broad line companies, Korean companies have expanded into virtually all car segments. This has created a participation problem in the lower volume segments where Korea has been unable to gain sufficient volumes to achieve high productivity. Korea’s average production volume for large/premium cars and sports utility vehicles (SUVs) were respectively 16,000 and 35,000 units annually in 1996 – a volume that is significantly lower than the typical 100,000+ required for economic performance. This problem showed up operationally as lower capacity utilisation for these production assets. Korean auto manufacturers rapidly increased the number of available models through platform diversification during the 1990s. The result is that Korean automakers have introduced a large number of disparate platforms, rather than created model variations from an individual platform to leverage investment. This effectively means higher R&D and material costs, few common parts across different models, more complicated production processes, and low capital productivity from assembly assets. When a Korean manufacturer needs a new model, it imports a new platform from abroad (rather than developing or importing a model that is built off an existing platform). This results in a low model per platform ratio of roughly one model per platform compared to a roughly three model per platform ratio for Japanese and European manufacturers.
144 The Second Automobile Revolution
As a result of the acquisition of Kia, Hyundai expects synergy effects with the commonalisation of Kia’s platform. Hyundai produces eight models based on five platforms and Kia has eight models per eight platforms. Hyundai will be integrating five platforms and producing fifteen models. The volume of each platform will reach 500,000 units and each model will produce 150,000 units. Through the commonalisation of platforms Hyundai expects to gain a reduction of 2 trillion won in R&D and in manufacturing. Hyundai effected the integration of the production lines of Hyundai and Kia for parts by 2001 and that of main frames by 2005. Now, the Hyundai-Kia group has twentynine engine models (sixteen for Hyundai and thirteen for Kia) in total. It plans to reduce the number of engine models to fifteen, mainly by making many of Kia’s engine models obsolete and by sharing Hyundai’s, because most of the engine models (except the engines for RVs and large-sized commercial vehicles), developed by Kia, cannot meet the environmental standards of advanced countries. In addition, it plans to develop a new diesel engine for SUVs on the basis of Kia’s Carnival engine (Lee and Cho, 2001: 10–11). Commonalisation of parts and components and reshuffled supply chains Further savings will come from joint purchasing – which will also lead to efficiency gains for Korean-based suppliers and mergers in the Korean components industry prompted by this action. According to a company source the synergy benefits of the merger show that reducing individual vehicle platforms and components sharing will yield savings of more than US$4 billion over the next few years. We have already seen benefits from the merger as HMC gave Kia the Visto mini-car model to market as its own. In the past, Kia benchmarked each model Hyundai produced – now we will be able to develop joint models providing individual benefits for the two brands. (Korea Herald, 23 June 1999) This integration is also required for the reshuffled supply chains. Thousands of small-scale parts makers are extremely concerned about their fate in the aftermath of the ‘restructuring’. This change could, perhaps, more than halve some 1,500 first suppliers, which supply parts directly to the final assembler involved in the restructuring. This figure would increase greatly when secondary and tertiary subcontractors are taken into account. Hyundai set out to screen their suppliers first and then choose contract winners through bidding. Hyundai assessed the 466 then current primary suppliers to Hyundai and Kia in 1999. From those suppliers two or more qualified makers were identified for each group of component parts in late 1999. Parts contractors were determined later through bids from qualified firms. The capability of supplying integrated parts in modules was taken into account
Hyundai: a Top Five Global Automaker? 145 Table 8.1
Hyundai, modularity by model and plant
Model Modularity
Plant 1
Plant 2
Clik 27%
SantaFe 24%
Tucson 36%
Plant 3
Plant 4
Plant 5
Asan
Avante 26%
Stareks 12.5%
Tucson 36%
NF 36%
TG 36%
Source: Company internal report, Sept. 2004.
in the screening stage. Hence, at least one-third or up to nearly half of the 466 parts suppliers were eliminated in the process. The increased module purchasing policy of parts will most likely lead to the integration of the affiliated firm into the overall design and manufacturing process. The modular production system (in which Korean carmakers have been lagging behind American and European producers) carries several merits from the point of view of production control and reduced transaction costs when introduced in the assembly process. According to company sources, 36 per cent of the work activities on the assembly line have been modularised in the Hyundai Asan plant (Table 8.1). In order to further modularise production, Hyundai Mobis – an auto-parts making unit of the Hyundai Motor Group – built a new car module plant in Ulsan in 2003. The new plant, covering 62,700 square metres, was designed to produce 600,000 units of driver-seat modules and chassis modules a year. In addition, this supplier is building a new plant in Asan, which will supply 900,000 driver-seat modules, front end modules and chassis modules a year for the Asan assembly plant. Sixty-nine supplier units were located in the area of the Asan plant, and 40 suppliers were in Ulsan (Korea Automobile Industries Association, 2003). The restructuring has led to changes in contracting patterns and overall subcontracting relationships. This restructuring in the automotive industry is having a twofold effect. On the one hand, the restructuring through commonalisation of parts and dual sourcing will reduce the number of suppliers and achieve significant economies of scale and improved productivity or quality. Otherwise, M&A is often used to exit the business of small suppliers because their facilities often partially duplicate those of other subcontractors. Excess plant capacity and surplus workers have led many suppliers to increase their degree of vertical integration. On the other hand, affiliated firms and large and medium-sized firms have been pressured due to increased specialisation to improve their technical abilities because they will need them when competing with suppliers on whom they are heavily dependent. They will also be regrouped under the ‘big two’ system that stimulates M&A between internal parts suppliers based on the specialisation and commonalisation of particular parts. Hyundai has especially strengthened the support for small to mediumsized suppliers in order to improve their technological capabilities. In 2006
146 The Second Automobile Revolution
Hyundai set aside 2.5 trillion Korean won for this initiative. Along with these support funds, Hyundai continues to strengthen various technology support programmes such as guest engineering, advanced technology cooperative benchmarking, the transfer of new technology and patent related technologies. Moreover, Hyundai also offers training in technology, quality and management on a regular basis and increased the total number of trainees from 13,000 participants in 2005 to 20,000 in 2006 (Hyundai, 2005: 66). Work flexibility and labour relations To strengthen competitiveness, it has become inevitable to introduce flexibility into the labour market. In particular, the most important development has been legalising lay-offs, thus giving corporate leaders more control in personnel management affairs. Korean labour rules have had two primary impacts on the auto industry. The first is the inability to deploy workers in multi-tasked jobs. Korean unions have long focused on job categories, craft barriers and working hours as the basis for negotiations. This approach has inevitably led to the preservation of function-specific tasks that has prevented manufacturers from implementing the multi-tasking required for lean production. The second is the inability to reduce working hours. Since Korean unions have long been against lay-offs, Korean manufacturers have been forced to keep excess workers on the assembly line. The unions have continued to fight for minimum legal working hours – rebelling against reductions in actual working hours per shift while demanding more workers on line in the name of ‘better labour conditions’. It appears that the impact of unionism is one of the fundamental barriers to the improvement of productivity. Unions may have been right in the past to demand better wages and working conditions, but the unions’ confrontational attitude and rigid negotiating stance on job descriptions are detrimental to an efficient manufacturing process which could ultimately benefit the remaining workers. The labour unions finally consented to lay-offs in 1998 in a landmark agreement with the government and employers to help revive the nation’s battered economy through quick industrial restructuring. Legalising lay-offs was the core part of the agreement and marked a shift forward. Hyundai has since carried out large-scale organisational restructuring with extensive lay-offs affecting about 30 per cent of its officials. According to Hyundai, it has streamlined its organisations from fourteen divisions with 404 teams to seven divisions with 340 teams. The seven divisions, including production, sales and R&D, are now expected to act as an axle for smoother organisational operations. About thirty-six officials were also relieved of their positions or placed on a retirement list. After these executive level lay-offs, the company further planned to dismiss about 8,000 employees by 1998. All of the auto manufacturers had also limited production to eight hours a day, owing to the flagging demand for cars in the domestic market.
Hyundai: a Top Five Global Automaker? 147
Quality management and brand management Improved qualities were needed to remain competitive with overseas producers. Hyundai acquired a reputation for poor quality in its largest export markets until early in the twenty-first century. Hyundai began to renew its efforts in quality management initiatives from 1999. In 2002, Hyundai and Kia’s product quality divisions were merged to stimulate synergistic effects. For management of overseas quality standards, Hyundai developed a North American product quality division and an international product quality department in charge of all international affairs to further product enhancement efforts. Parallel to the reshuffled quality management organisation, they adopted the Six-Sigma initiative in 1999. Six-Sigma is a scientific product quality innovation system to enhance product quality management. This project was initiated in full force starting in 2000: 360 projects were named in 2000, and by December 2004 the number of cases totalled 5,092. The number of SixSigma experts (black belt and green belt) totalled 3,618 in all fields ranging from production to research and development (Hyundai, 2005: 32). Hyundai maintains a quality approval system at each stage from planning to sales. It is only after a vehicle passes quality control standards that it can proceed on to the next level. Also they have established a supplier product quality rating system to help improve overall quality standards. The company implemented a 5-star quality rating system to better determine a supplier’s quality and technological advancements. For second-tier suppliers, Hyundai introduced a supplier quality (SQ) marking system to ensure product enhancement and development. Through the implementation of quality rating systems, Hyundai is aiming to improve the working relationships of 150 domestic suppliers and to enhance supplier quality standard to a world-class level by 2010. As a result of strong quality management, Hyundai became the highest ranked non-premium nameplate and was ranked third among all nameplates in the 2006 Initial Quality Study (IQS) of J. D. Power and Associates. They have been steadily improving their nameplate ranking since 2002 with a ranking of 28th in 2002, 23rd in 2003, 7th in 2004, 10th in 2005 and 3rd in 2006 (Table 8.2). Table 8.2
Hyundai’s J. D. Power and Associates IQS ranking
Rating Non-premium nameplate average Ranking Source: Hyundai (2006).
2002
2003
2004
2005
2006
156 135
143 135
102 121
110 120
102 123
28
23
7
10
3
148 The Second Automobile Revolution
A brand is an enterprise’s integrated value comprising tangible and intangible factors. Not only does a strong brand result in higher profits, but it also plays an essential role in the gain of consumer loyalty. In January 2005, Hyundai announced the launch of its new global brand management strategy. This new brand strategy was designed to ensure that Hyundai reached industry leading levels, not only in terms of size, but also in terms of customer perception and overall brand value. As part of its brand strategy implementation, Hyundai has developed a Brand Committee, which includes a working level sub-committee. The committee regulates operations for brand consistency, and enhances the brand management system. 2005 was the first year that Hyundai was recognised by Business Week and Interbrand as one of the Global Top 100 brands (Hyundai, 2006: 16).
Is the module production system one of the low-road strategies? Modular production and numerical flexibility The maturity of mass production exposed the limitations of the Taylorist– Fordist model of manufacturing in Hyundai. Management faced the problems of how to increase labour productivity and engage in other costcutting measures, to improve quality control and, perhaps most importantly, increase the flexibility of production systems. The main elements of the increasing flexibility are: the introduction of microprocessor-controlled robotics, other computer numerically controlled machine tools and automated process control. These new forms of automation have dramatically increased the flexibility of the manufacturing system. The automation strategy is most meaningful for the understanding of the Hyundai strategy after the crisis in the late 1990s. At that time the company was pursuing the modular production strategy in its core operations. Now, Hyundai classifies auto parts into thirteen modules and plans to gradually upgrade the level of modularisation for new car models. Hyundai’s categories of modularised auto parts are as follows: cockpit, pedal, head-lining, doors, rear package tray, front suspension, rear suspension, fuel tank, muffler, tubes, cooling device, rear bumper and front bumper. The modular production concept has been undertaken primarily to reduce the number of jobs on the shop floor. Table 8.3 shows the reduction of jobs achieved by adopting the modular production system on the shop floor. Under the modular production system scheduled to be completed in 2004 in plant 1, the company’s workforce was theoretically slated to shrink to about 200–300 production workers. Instead of reductions of workers, the company has increased production volume and unit per hour (UPH). This production concept has no doubt required the elimination of job security and gives a greater opportunity to mix production on the line.
Hyundai: a Top Five Global Automaker? 149 Table 8.3
Hyundai, modular advantage: reducing the number of jobs
Module Front strut Rear strut Rear suspension Fuel tank Front end Crash pad Total
Before modularity (a) Number of jobs
After modularity (b) Number of jobs
Total jobs cut (a/b)
58.82 64.24 343.24 68.29 216.64 399.04 1150.27
0 0 0 53.57 74 0 127.57
58.82 64.24 343.24 14.72 142.64 399.04 1,022.70
Source: Hyundai Motors Workers Union.
In addition, the company has easily adopted the relocation of work to lower wage settings and increased the use of low-cost, part-time workers or subcontractors consistent with intensified Taylorism–Fordism. The utilisation of part-time workers or temporary workers has also been expanded. The company employed around 13.6 per cent of total product workers as irregular workers in 2001. These workers worked on the shop floor as employees of Hyundai. This increased to 37 per cent in 2006. The irregular workers on the same assembly line of Hyundai for the same period of time received only 60–70 per cent of the payment that regular workers get. These flexible personnel policies have a twofold effect: first of all, compensation for physical strains that can eliminate the conflict issue of labour–management relations; secondly, a potential reduction of labour costs. Under the modular production strategy, all of the work performed in a given factory for a particular product line is grouped together intensively. As far as teamwork is concerned, it is the group leader who organises the task rotation in his or her group, taking into account the skill level of the workers. Job rotation with the goal of training polyvalent workers is currently a prominent theme in the work organisation. However, the older plant in Hyundai has no special criteria for job rotation at the present time because workers have a negative view about it as a cause of work intensification, while rotation is frequent on the shop floor. Its goal is to compensate for work-related stress and strain among the group members. By contrast, job rotation in Asan is realised on the shop floor according to the management plan, but when one steps inside the product line to see how work has been organised on the shop floor, an indifferent picture emerges compared to the Taylorist–Fordist model of work organisation. Taylorist principles guide the division of labour and the content of individual jobs. Production jobs on the product line are narrowly defined. Direct production workers do not perform any maintenance functions. If repairs are needed or production processes are altered, the machines are shut down so
150 The Second Automobile Revolution
that specialised maintenance workers can adjust them. Obviously, assembly lines are organised to take advantage of production efficiencies typically associated with lean production concepts. Production tasks include integrated quality control, reduced inventory levels and a synchronised flow of production. This hybrid combination of modularisation and lean-style manufacturing is the distinguishing characteristic of the Hyundai production system. Flexible production systems require flexible work practices and worker commitment to quality control and maintenance. Hence the organisational hierarchy may shift from inflexible to flexible, which generally contributes to cooperation and horizontal communication. The most fundamental issue is the introduction of a new form of work and production organisation that is closely related to human resource management. There is considerable pressure to change the traditional work organisation. The capacity to mobilise workers in order to increase productivity is closely linked to a career system and individualised personnel assessment. For this reason, management intend to introduce a qualification system that will stimulate promotion of an employee’s status. This means that the work organisation will be divided by position and job classification. This system could solve the bottleneck of promotion of workers. This system also is characterised by the integration of the differential status system between white-collar workers and blue-collar workers. This differential status system is one of the grievances among bluecollar workers now facing personnel management.2 However, this reform plan is too hard to realise because wage reform is pre-required. Modularisation has resulted in an adjustment of working hours. Plant management has tried to shorten the cycle-time of production processes as a means of increasing the UPH of production lines because modularisation has led to a reduced number of jobs. From the worker’s point of view, modularisation has led to deterioration in working conditions. It not only increases the intensity of the work but also shortens cycle times. Table 8.4 shows the incidence of work intensity and time pressure at four plants. As is clear from the table, the Asan plant showed
Table 8.4
Hyundai, work conditions by plant
Plant
Work intensity
Time pressure
Plant 1 Plant 2 Plant 3 Plant Asan
46.99 (664) 45.83 (456) 39.86 (488) 49.52 (41.4)
41.62 (663) 39.50 (457) 25.67 (487) 43.68 (412)
Note: 100: very strong; 50: strong; 0: adequateness; ( ): number of interviews Source: Hyundai Motors Workers Union.
Hyundai: a Top Five Global Automaker? 151
the worst working conditions, while plant 3, one of the oldest manufacturing facilities, had the best result. The Asan plant is the most modern factory and has the highest level of modularisation among the Hyundai plants. The time pressure reflects the low degree of autonomy such as a traditional assembly line. Furthermore, although the task enlargement has been increased, modularisation has had an effect on enlarged job rotation and decreased job opportunity. Distorted relationships between labour and management Labour relations are a hot confrontational issue between top management and the union in Hyundai. The workers set up a labour union after the government’s declaration of 29 June 1987 that called for a relative relaxation of its tight, long-term suppression of labour activities. The struggle between labour and management began in 1987 with large damage to both sides. The major demands of the union were wage increases, improved working conditions and non-monetary rewards. The wage system at Hyundai was complicated, characterised by a combination of a traditional seniority wage system with a productivity wage system. According to the Hyundai labour union, wages were composed of basic wages (70.3 per cent) and extra wages (29.7 per cent). The extra wage depended on working hours and productivity. This efficiency pay was the most important component. In the formula to calculate efficiency that the union had to accept, only the number of vehicles actually produced was taken into consideration. The sharing of productivity gains continued to be the primary focus of conflict in collective bargaining up until the mid-1990s. Under this wage system workers were seriously concerned about working hours, because the reduction of working hours meant decreased wages. Therefore the labour union maintained maximum working hours in order to compensate for lost wages. While the standard working time per week was 40 hours, the annual working time for production workers with shift work was 2,602 hours in 2004 (Figure 8.1). It is definitely higher than for the average Japanese assembler. The financial crisis in 1997 changed the labour market rapidly. To overcome the economic crisis, it was essential to introduce flexibility into the labour market. The most important development was the legalising of lay-offs, thus giving corporate leaders more control over personnel and management affairs. Labour market flexibility was further enhanced in 1998 through the introduction of the manpower leasing business. This new option offers labour outsourcing as a new means of employment flexibility. The changed labour market has brought a new prominence to labour–management relations. The transformation of production at the company has also led to major changes in labour–management relations. Most noticeable is the shift from arm’s-length accommodation to adversarialism. The union has provided some protection for workers against arbitrary management decisions. While
152 The Second Automobile Revolution
2700 2672 2650 2600
2602 2572 2582
2550 2500 2465
2450 2400 2350 2000 Figure 8.1
2001
2002
2003
2004
Working hours at Hyundai, 2000–2004
Source: Korea Labour Education Institute (2006: 43).
the parties still speak of collective bargaining in terms of wages, working conditions and basic managerial strategy such as established transplant (e.g. the union proposed a bargaining demand for UPH co-determination at a collective bargaining meeting in 1999) the management has basically accepted this proposal. Moreover, the union, which experienced extensive downsizing in the crisis period of 1997–8, has put particular stress on the employment guarantee of their union members. In fact, the union implemented contractual clauses to enforce employment security for their members and, to some extent, constrain management’s action (e.g. lay-offs and outsourcing) in the adjustment of employment size through collective bargaining in 2003. After a long period of negotiation, Hyundai’s settlement with its union in August 2003 appears to have set a trend. Nowadays, the major issue of the union is job security which suffers from extended transplant capabilities. Growing distrust between workers and management is a major factor in chronic labour disputes. Despite its remarkable business performance from 2000, Hyundai failed to set up constructive labour relationships on the shop floor. A vicious circle of labour disputes with assemblers has surfaced regularly ever since. During 2000–6, the average number of hours of strikes and lockouts per year for Hyundai was over 185 (Figure 8.2). These labour disputes every year have prevented the full utilisation of the highly automated production facility and delayed improvements to product quality. As a result, Hyundai has suffered despite the expansion of its production volume. Almost all management regard their relationship with labour as problematic. They do not regard workers as a collective with co-determination rights
Hyundai: a Top Five Global Automaker? 153
350
12.9
12
12.9
10
231
8
12.2
300 250
227
8.0
314
9.0
8.6 200
183 7.0
6
154 150 5.1
4 88
100
2
100
Lost hours by strike
Rate of wage increase
14
50 0
0
1998
Figure 8.2
0 1999
0 2000
2001
2002 Year
2003
2004
2005
2006
Strikes and wage increase at Hyundai, 1998–2006
Source: Ministry of Labour (2006).
and there is a constant need to create a consensus for any change. Hence there is no guarantee that the company’s dream of becoming a top five global automaker by 2010 will be realised. The future of Hyundai as Korea’s top automaker is gloomy because chronic labour disputes might bring about its complete collapse. The latest strike at Hyundai ended once again without any progress in the relations between management and labour. If Hyundai hopes to strengthen global competitiveness, it must establish close partner relationships with the union. Consumers have criticised the two sides for ‘habitually resorting to strikes’ and for ‘colluding with union leaders’ respectively. Management and labour should learn the lesson from the collapse of the top three automakers in the US as a result of powerful union strikes and loss of customer loyalty. In this context, long-term commitment and participation is more necessary than ever and unions will also need to play an important role in workplace industrial relations to determine the socio-economic environment of the future.
Modularisation and reshaping of the supply chains The modular production system is basically a generalisation of the workflow principles of lean production. As such, it also envisages the elimination of sub-assembly work with modularisation of components and parts. In order
154 The Second Automobile Revolution
to supply modules promptly, suppliers are vigorously adopting a JIT (‘just-intime’) supply chain system. Module outsourcing requires that specialised suppliers should operate near the plant or in the surrounding areas. Therefore, large supplier groups need to establish themselves in the supplier-park near the assembly plant. A supplier then has the capacity to become responsible for the entire modules. Furthermore, this could work as a catalyst for the establishment of a new supplier hierarchy with a small number of module/system integrators at the top followed by many of the traditional suppliers in the second tier. We can see a clear concentration of first-tier suppliers to the Ulsan (31.8 per cent of total suppliers) and Asan region (41.1 per cent of total suppliers). As a result of modifying existing suppliers’ delivery routines under a synchronised production process based on outsourcing of modules, geographical proximity plays an important strategic role. The adoption of modularization has stimulated a synchronised delivery method. For example, materials and parts can be delivered to an assembly plant in Asan in two ways. One is a synchronised sequential delivery method, which is applied to function components such as engines, transmissions, instruments panels, radiators, brake drums, etc. The other is a scheduled batch delivery method, which is usually adopted for the other parts (Che, 1999: 8). Asan Hyundai delivers materials to suppliers seven days before production starts and places orders with suppliers three days before production starts. Consequently the company cannot hold inventories and cope with any fluctuation of production volume during the interval between material delivery and production order to the supplier. It has reached 95 per cent of the scheduled batch delivery plan and synchronised production process. Synchronised sequential delivery is conducted in three different ways. One is called MIP (manufacture in plant) sequencing, where components are produced in the plant and then sequentially fed to the assembly line according to the sequence of vehicles. The second method is sub-assembly sequencing. This is conducted by the supplier, who assembles units of components near the assembly line and delivers them by JIT. The last method is direct sequential delivery, where components are sequentially provided directly to the assembly line from suppliers. The supplier gets the information from Hyundai through the VAN network system, which indicates delivery volumes needed every day. For example, a cockpit module supplier gets an order for assembled products every day. The information for assembling sequences is finally determined just after vehicle comes out of PBS (painted body storage). As vehicles leave the PBS, the sequential order is electronically transmitted to the modular supplier. Some parts are assembled based on Hyundai’s two-hour schedule and delivered to the assembler in the lot. For example, Hyundai Mobis – a chassis and front-end module supplier – takes a production order by Hyundai 100 minutes before it needs to be assembled. The company takes 40 to 50 minutes (depending on the products) for assembly and needs around
Hyundai: a Top Five Global Automaker? 155
45 minutes for loading and unloading. This company therefore has 100 minutes to produce a chassis module product and deliver it to the assembler. It keeps just 10–20 units in the inventory of finished modules to sequence for deliveries. If they have problems with parts supply from subcontractors, they have a high risk of sequential delivery on the Hyundai assembly line. Otherwise, this module supplier has to guarantee an efficient management of second-tier supply flow. One interesting aspect of the restructuring process is the changes in the contracting patterns and the overall subcontracting relationships. This restructuring in the automotive industry is having a twofold effect. On the one hand, through commonalisation of parts and single sourcing it will reduce the number of suppliers and achieve significant economies of scale and improved productivity or quality. As we said earlier, M&A often results in loss of business for small suppliers, because their facilities often partially duplicate those of other subcontractors. Excess plant capacity and surplus workers have led many suppliers to increase their degree of vertical integration. And under the pressure of increased specialisation affiliated firms and large and medium-sized firms have needed to improve their technical abilities when competing with suppliers. They will also be regrouped under the ‘big one’ (Hyundai-Kia Group) system that stimulates M&A between internal parts suppliers based on the specialisation and commonalisation of particular parts. Module systems and synchronised production on Hyundai have radically changed the nature of the supply chain. Suppliers have polarised into three types: big mutli-technology suppliers like Hyundai Mobis; medium-sized suppliers of modules in the pre-modularisation stage; and small-sized suppliers of components. The first type is dominated by affiliated firms and international suppliers like Bosch. Assemblers have attempted to set up affiliated subsidiaries in order to secure a stable and adequate supply of parts or devolve R&D responsibility to the parts and components suppliers. The majority of these firms are relatively big in size and have high technological capabilities and the manufacturing of the core parts is complicated and capital-intensive. The portion of the parts supplied by these parts manufacturing companies is over 40 per cent of all parts that are supplied to the assembler. The module outsourcing requires close collabouration between large module suppliers and assemblers. New bilateral relations are required in terms of product design and cooperative development approaches with assemblers’ requests. In such cases we find cooperation and harmonious relations. Within this new process configuration, a system supplier such as Hyundai Mobis has been successful in managing an increasingly complex supply network. All production activities are closely linked with Hyundai’s production flows. The second type of supplier operation consists of sub-module suppliers who are mostly Tier 1 suppliers. Most of these suppliers are small in scale and lack R&D capability which inevitably leads to lower participation by
156 The Second Automobile Revolution
suppliers in the design process and a longer time for new parts development. Some of them have merged with large system suppliers or have been taken over by international suppliers (for example, Visteon bought out Duckyang Company in 1999). The third type comprises mostly common parts suppliers who have direct business transactions with the assembler. This pattern of closed transactions can be a source of monopoly rents for the automobile manufacturing companies by weakening the parts manufacturers’ bargaining power and by setting monopoly prices. This hierarchical subcontracting network is advantageous to the assembler, who has the power to drive down costs.
Transition from a worldwide export strategy to a multi-domestic globalisation strategy Globalisation is one of Hyundai’s core management initiatives to minimise risk in the external business environment and to ensure its future success. The early stages of a step-by-step implementation of globalisation exploited emerging markets with existing products and set up sales beach-heads that would ensure sufficient demand to generate the scale of economies. Since the middle of 2000 Hyundai has started to invest on a large scale in Europe and North America where there are protectionist measures against their products and a strengthening of the Korean currency against the US dollar. Two parallel patterns of Hyundai’s activities are likely to continue. On the one hand, they will continue to be heterogeneous in terms of enterprise, namely project-based collaborations, licensed manufacturing or joint ventures in periphery markets. On the other hand, Hyundai has founded transplants through direct investment in core markets such as BRIC, Europe and the US. The strategy of Hyundai is to access emerging markets that may require the overcoming of tariff and/or non-tariff barriers. Since 1990, Hyundai has moved from a worldwide export strategy to a multi-domestic structure based on manufacturing sites in the different regions. Therefore, the internationalisation of automobile manufacturing by Hyundai started mostly in peripheral regions. In Southeast Asia, the company started the production and sale of a specific vehicle destined for these markets on the basis of a platform shared with Korean products (the Accent and Atos models). This led Hyundai to utilise its components and platforms to the maximum. On the other hand, this platform strategy permitted the accelerated pace of product replacement in the home market. In late 1996, Hyundai began the construction of its largest manufacturing plant in Chennai, India. Hyundai had invested approximately US$1.1billion in the Hyundai Motor India Chennai Plant by the year 2001. The Indian project is divided into a two-phase construction plan. The first stage of this construction project was completed in 1998 with an annual production capacity of 120,000 units. In this phase, Hyundai
Hyundai: a Top Five Global Automaker? 157
manufactured 1300cc and 1500cc Accents and 800cc Atos. The second phase was completed in 2001, when Hyundai manufactured additional models in India. Hyundai’s Chennai plant consists of shops for engines, transmissions, press, body, paint, assembly and plastic injection moulding. Also in the pipeline are an R&D centre, a proving ground and a vehicle performance test centre. These facilities will enable Hyundai’s Chennai plant to become a self-sufficient manufacturing and production site for developing cars that meet the needs of its local market. Hyundai India has become the second top-selling company in India. In 2006, it held an 18.5 per cent market share in passenger cars. During the two-year period from 2004 to 2005, Hyundai India sold over 450,000 vehicles and continues to garner accolades for its remarkable progress in the Indian market (Figure 8.2). The Chinese market has been growing very fast since becoming a member of the WTO. Beijing Hyundai in China began operations in December 2002 with production of the new EF Sonata. Its production capacities expanded from 150,000 units to 300,000 units in 2007. Beijing Hyundai sold over 50,000 of the new EF Sonata in its first year of production, garnering a 10 per cent market share of the entry mid-size class in China. Beijing Hyundai has shown such remarkable progress in a short period of time that its success has been dubbed as ‘Hyundai speed’. It sold over 200,000 vehicles in 2005 and gained 8 per cent of the passenger car market share in 2006 (Figure 8.3). As the top foreign automobile manufacturer in terms of sales in the Chinese market, Beijing Hyundai is pushing forward to sustain its lofty status.
700,000 600,000 500,000
632,269
2003
2004
2005
569,721 551,226 455,012 418,615
400,000
400,221
339,767
350,406
280,641
300,000
215,630
200,000
252,861
150,724
100,000
233,868 144,090
52,123
0 Domestic
USA
Western Europe Regions of world
Figure 8.3
Hyundai sales by region, 2003–2005
Source: Hyundai (2006).
India
China
158 The Second Automobile Revolution
The price competition on sub-compact and smaller cars in China has been fierce in the past few years. Experts in the Chinese automobile industry predict that more people will start buying cars after the Beijing Olympics in 2008. Demand for super cheap small cars will skyrocket, the experts say, reaching more than 5 million cars a year. The total automobile market size is expected to be 9 million. This means that ‘super cheap’ cars represent the future for China, and Hyundai Motor Co. plans to join the fray, offering a vehicle costing less than $4,000. Hyundai will strategically develop a budget car aimed at the Chinese, Indian, Southeast Asian and Central and South American markets. Hyundai’s Chinese affiliate, Beijing Hyundai, has already sent official papers to its parts providers in China to prepare for the project (JoongAng Daily, 5 February 2007). The Turkish plant was scheduled to start producing 50,000 units of the Accent and the H-100 (minibus) in 1997. By the year 2000, the plant’s production capacities expanded to 120,000 units. Hyundai put the new Lantra into production in order to diversify its model line-up in the region. In the early stage of production, Hyundai concentrated on the Turkish domestic market, but it will eventually start exporting its vehicles to neighbouring countries. Hyundai will provide the joint venture company with key parts such as moulds, jigs, machinery and other production facilities along with core parts such as engines and transmission. Hyundai’s first US plant, which opened in 2005 in Alabama, can easily be expanded to an annual capacity of 400,000 vehicles from its initial goal of 300,000. The Montgomery factory is Hyundai’s most automated, with a parts-delivery system that reduces the time and cost of adding capacity. The plant makes Sonata sedans and Santa Fe sports-utility vehicles. The factory’s design included much consideration on how to expand the production capacity without big modifications and hassle. Hyundai spent 11 trillion won ($11 billion) since 2002 building the factory. It planned to produce 150,000 American-made Sonatas in 2005 and expand yearly production to 300,000 units with other models, such as a new version of the Santa Fe utility vehicle. Even though it hoped to increase its market share in the US from 2.6 per cent to 3 per cent, its share fell from 3.2 per cent July 2006 to 2.6 per cent in January 2007 (Chosun Ilbo, 2 February 2007). The effect of the strengthening of the Korea won against the US dollar has been to increase Hyundai’s efforts to ride out the storm by boosting its exports. Hyundai is expected to mainly target the European and North and South American regions. In order to maintain its market share in the European and American markets, which currently absorb about 70 per cent of its overseas exports, quality will have to be improved and marketing strengthened in the US and in West Europe. The success of Hyundai’s enhanced position in the world car market depends more on the competitive foundation of the company. Nevertheless, it faces critical challenges. The logical target for transplants is throughout
Hyundai: a Top Five Global Automaker? 159
Asia, but that means hard competition with Japanese and European automobile manufacturers in the periphery. In the emerging markets such as India, China, South Asia and Eastern Europe there is also competition from domestic manufacturing, frequently as a part of a joint venture between a local company and a global leader. Hyundai’s access to these regions will require an important strategic challenge. In response to the growth of the emerging markets, Hyundai will need modern technology and operation under modern management concepts in overseas plants in these regions. In spite of the pressures globalisation creates, the strategy of global sourcing and localisation is still in its infancy. Obviously, the major task is to stimulate the set-up of the international suppliers’ network. Hyundai may need to be secured through Korean sourcing from suppliers in Southeast Asia and Eastern Europe. To survive the fierce sales competitions in core markets, Hyundai should exert its best efforts to improve technology and quality. Quality and high technology are preconditions for survival. It must also improve its image, which is a reputation for poor quality cars in the main export market. In order to improve quality Hyundai will need appropriate human resource development. In summary, Hyundai as a major cost-based exporter will find itself under significant pressure in terms of globalisation. In adjusting to globalisation it faces a number of systemic impediments such as higher finance costs and lack of labour flexibility. Although it has been effective recently in responding to these challenges the game of survival in a world of global competition is not easy and therefore the future of Hyundai’s trajectory is hard to predict. It is at a critically important crossroads in the globalisation process.
Conclusion After the Asian financial crisis, Hyundai achieved remarkable success in domestic and international markets. Hyundai has adopted volume-conscious and labour-exclusionary polices based on a modular production system, and with this system an increasing degree of product standardisation was inevitable. The key competitive strategy remained one of lowering unit production costs through increased economies of scale and work intensity. The top management also stressed the need for further improvement in quality. There is no doubt that the efficient design and manufacture of vehicles of good quality in higher market segments was an appropriate strategic choice. Early in 2001, Hyundai announced its ambitious plan to become the fifth global carmaker by 2010. As a latecomer in the world auto industry it needs to improve the brand image of its vehicles in order to survive and grow in the intensified competition of the global auto market. Although Hyundai has achieved success under the relatively protected condition of the domestic
160 The Second Automobile Revolution
market and with the export of their low-priced cars to overseas markets it has reached a critically important crossroads. Under fierce competition in the global markets Hyundai’s survival will depend on enhancing its technological capability of R&D and manufacturing operations, as well as improving the quality and reliability of its products, which is a necessary condition for upgrading the competitiveness of its car brands. To improve quality, Hyundai must undertake appropriate human resource development. However, management does not appear to be pursuing a human-oriented manufacturing system but rather a man-machineoriented production system. Workers see this tendency as an expression of a lack of trust on the part of management. In addition, the union has not attempted to influence the technical-organisation design of work or to develop an independent work policy concept. They have only been interested in increased wages and job security and they have fully agreed to the introduction of new technologies like robotics, which can eliminate some difficult manual tasks. Therefore it is difficult to gain benefits from the kaizen effect from the Six-Sigma initiative. In conclusion, Hyundai’s strategy is fragile since it depends so heavily on maintaining global competitiveness. Nowadays the latter is eroding rapidly due to the strengthening of the Korean won against the US dollar and Japanese yen. Chronic labour disputes remain a major factor in the deterioration of competitiveness. It is a harsh reality that if a company fails to overcome these obstacles, it will fall in the ranks of its competitors. Hyundai’s road ahead remains unclear.
Appendix Table 8.1 Year
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Hyundai, 1997–2006
Production
Sales
Employment
Total (a)
Overseas
Domestic
Export (b)
a/b
1242 812 1269 1525 1513 1702 1646 1673 1689 1618
43 95 117 178 260 410 410 590 699 885
646 308 548 647 707 767 631 551 570 580
565 529 617 828 801 928 1012 1024 1131 1192
45.5 65.2 48.6 54.3 52.9 54.5 61.5 61.2 67.0 73.7
46683 45970 37533 50840 48506 48443 50019 51383 54400 54711
Performance Gross sales (c)
Operating profit (d)
116.6 87.0 142.4 182.3 225.1 245.7 250.0 274.7 273.8 273.4
8124 3671 9061 13133 20966 16062 22357 19814 13841 12344
Notes: (a), (b): 1000 units; (c): 100 billion won; (d): 100 million won Source: Korea Automobile Manufacturers Association (2006) and Korea Labour Education Institute (2006).
Hyundai: a Top Five Global Automaker? 161
Notes 1. To maintain a healthy corporate financial structure was of the foremost importance for Hyundai in 1998. With annual interest rates around 20 per cent, no firm could survive the financial pressure if its normal business operation heavily depended on debt. It had to review and postpone investment plans especially if they were to be financed by outside sources. Hyundai had also undergone intensive downsizing with extensive lay-offs affecting about 30 per cent of its officials and 8,000 workers. 2. In fact, the promotion of employees in the company is very important for their working life. Koreans have a strong sense of status or position. The opportunity for promotion from general worker to ‘chochang’ was limited. According to the new system, a general worker can be entitled ‘chochang’ without specific responsibility. A ‘chochang’ will be rotated among the members of the same qualification range. The new organisational innovation will boost voluntaristic behaviour in the production process. Obviously, improved quality and elimination of waste requires the cooperation of employees.
References Che, J. W. (1999) Introduction of Japanese Manufacturing System and its Limitations in the Hyundai Motor Company. Mimeo, Seoul (in Korean). Cho, H.-J. (2001) ‘Modular Production and Supplier System: a Case Study of Hyundai Motor Company’, Economy and Society, 50 (in Korean). Chung, M.-K. (1998) ‘Hyundai Tries Two Industrial Models to Penetrate Global Markets’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Chung, M.-K. (2000) ‘Greenfield Strategy in Hyundai Motor Asan Plant: New Production Technology and Labour Relations’, paper presented at the International Workshop on Lean Production and Labour Force in the Automobile Industry: the Forms of Implementation of an Epoch-Making Model, University of Calabria, Cosenza, Italy, 25–26 March. Chung, M.-K. (2001) ‘The Expanding Buyer–Supplier Partnership for Product and Process Technology Development in the Korean Automobile Industry’, International Journal of Automotive Technology and Management, 1(2/3). Chung, M.-K. (2007) ‘Modularization in the Auto Industry: Interlinked Multiple Hierarchies of Supplier System in Hyundai Asan Plant’, Journal of Economics, 25(2) (in Korean). Hyundai Motors, Sustainability Report, Seoul, various years. Hyundai Motors Workers Union (2001) Annual Report, Ulsan (in Korean). Hyundai Motors Workers Union (2001) Report on the Modularization in Hyundai, Ulsan (in Korean). Jeong, J. W. (1999) Trends and Prospects of Modularization in the Parts Industry, Hyundai Research Institute (in Korean). Kang, J.-Y. (2001) ‘A New Trend of Parts Supply System in the Korean Automobile Industry: the Case of the Modular Production System at Hyundai Motor Company’, KORUS. Knauss, J. (1998) ‘Modular Mass Production: High Performance on the Low Road’, Politics & Society, 26(2). Korea Automobile Manufacturers Association, Korean Automobile Industry, Seoul, various years (in Korean).
162 The Second Automobile Revolution Korea Automobile Manufacturers Association, Monthly Automobile Statistics, Seoul, various months (in Korean). Korea Labour Education Institute (2006) Diagnosis and Alternatives of Hyundai’s Labour Relations, Seoul (in Korean). Korean Institute for Industrial Economics and Trade (1994) Development Tendency towards the Twenty-First Century of the Korean Automobile Industry, Seoul (in Korean). Lee, B.-H. and Cho, S.-J. (2001) ‘Merger and Reconfiguring of Hyundai-Kia’, paper presented at the Ninth GERPISA International Conference, Paris. McKinsey Seoul Office (1998) Productivity-Led Growth for Korea. Seoul and Washington. Ministry of Commerce, Industry and Energy (2003) The Situation and Development Trend of the Parts and Componenst Industry, Seoul (in Korean). Ministry of Labour (2006) Analyses of Labour Disputes in Korean Automobile Industry, Seoul (in Korean). Verma, A., Kochan, T. A. and Lansbury, R. D. (1995) Employment Relations in the Growing Asian Economies. London: Routledge.
Part II The Resistible Decline of the ‘Big Three’?
This page intentionally left blank
9
General Motors in an Age of Corporate Restructuring Richard Senter, Jr. and Walter McManus
Introduction: General Motors and the organisation of large American firms General Motors’ achievement during much of the twentieth century was built on a foundation of some aspects of Fordist production and an organisational structure developed by Alfred P. Sloan, Jr. Sloan served forty-five years at GM, first as an executive and then on its Board of Directors (Sloan, 1990). Following unevenness in its early years, the company attained fairly consistent prosperity. This success continued until the early 1970s. After that, however, GM alternated periods of profitable operations with periods of serious financial losses. Furthermore, the company’s market share in its home market has been declining for years. We argue that General Motors’ problems stem partly from relying too long on Fordist production and also flow from not maintaining some of the valuable aspects of Sloan’s structure. We also contend that the company has been hampered somewhat by its slowness in modifying the Sloanist structure to match it more fully to GM’s growing international operations. In this introduction, we first discuss Fordist production and, especially, Sloan’s organisational structure. In subsequent sections of this chapter, we examine GM’s experience from 1995 to the present. Throughout, we emphasise GM’s record in the United States because its headquarters nation is also where it is having its greatest challenges. We focus particularly on the financial aspects of the company. For an account of GM prior to 1995, see Flynn (1998). Fordist production refers to assembly-line operations, an extensive division of labour and great economies of scale. The Sloanist structure means a pattern of multiple divisions, with each division in charge of development and manufacture of a distinct product line, and responsible for the marketing strategy for that line. Though these obligations are decentralised, the corporation maintains some control through centralising decisions about access to financial resources, and through measuring each division’s financial success 165
166 The Second Automobile Revolution
or failure. This in turn stimulates some competition among the divisions. This structure is termed the M-form, for multi-divisional. (One contrast to the Mform is the U-form, which lacks the pattern of separate divisions for separate product lines.) At GM, the divisions initially occupied different market strata, from lowest to highest1 (see Boyer and Freyssenet, 2002, chs. 5 and 6 for extended discussions of Fordist production and Sloan’s structure). It is important to note that in any firm, Fordist production, carried to an extreme, can conflict with a Sloanist structure. For example, at GM, Sloan’s multiple divisions reduced somewhat the economies of scale that a thoroughgoing Fordist model would have accomplished. Sloan tolerated this, for the sake of organisational decentralisation and for an overall marketing strategy that he believed would enhance demand for the company’s products. And at General Motors there has always been a degree of tension between Fordist production and Sloanist structure. A completely Fordist production scheme in a firm implies a strongly centralised administration. Sloan, by contrast, strove mightily to build and maintain a degree of decentralisation at General Motors while still preserving enough central control to prevent extreme inter-divisional conflict and to forestall foolish and unprofitable investments by divisions. Sloan thought the task of maintaining a good balance would be ongoing, that is, would require continuing administrative attention, and would never be complete. Sloan’s mechanisms included the rule-by-committee pattern and an emphasis on executives persuading, not simply ordering, their subordinates to carry out tasks. Sloan’s stress on consensus building foreshadowed a crucial principle of management practice in Japanese manufacturing companies. Sloan’s efforts were directed to establishing and upholding an organisational structure that was appropriate for a very big firm. This was both an onerous and a novel task in the early twentieth century. A tribute to the structure is that other giant firms in the United States, both in the automotive industry and in other industries, copied it from GM. We argue that GM’s Sloanist structure is a key to the company’s early dominance of the automotive industry. (Embedded within this structure were a number of aspects and innovations that Sloan helped institute, and that advantaged General Motors: a consumer financing arm, excellent inventory management, and vastly improved forecasts of demand that drove production schedules, among others; see O’Brien, 1997; Norton, 2004: 6–7.) GM’s apogee was the period from the end of the Second World War until about 1973 (Cray, 1980). During that time, GM’s share of the American market was very large, its profits year after year were robust, and its international reach expanded. From1973 through 1994, however, GM experienced considerable loss of market share in the United States. Underlying this were a number of problems. These included adversarial relations with the company’s production workers, short-term and arm’s-length relations with its suppliers, diminishing control over its retail outlets, and an oppositional
General Motors in an Age of Restructuring 167
stance towards federal government regulation. Additional problems derived from GM’s lack of progress in responding to the advantageous characteristics of foreign assemblers. The company retained an emphasis on product design and technology, while its overseas competitors were racing ahead in process techniques and manufacturing technology. GM undertook repeated reconstructions of its organisational pattern, at the same time that its foreign rivals maintained more consistent formats. Also, the firm did not seem to grasp that the era of relative product stability and industrial oligopoly would end as the industry globalised, and GM did not recognise the threats this entailed. The oil shocks of 1972 and 1979 should have triggered alarm bells at the company, but apparently did not. However, from 1984 to 1994, the firm’s US market share fell from 42 per cent to 32 per cent. Since 1995, the company’s trajectory has fluctuated considerably. Its market share in the United States has continued to diminish, though its overseas operations have met with some success. Its financial situation has shifted dramatically between years of strong profits and years of great losses (see Appendix Table 9.1). More seriously, its prospects for the future are uncertain.
Financial and governance issues The decade of the 1990s was a time of strong economic growth in the United States, and General Motors benefited considerably from this. However, by the end of that decade, economic storm clouds grew. GM postponed some of the damage of the recession that occurred at the beginning of the twenty-first century with no-interest vehicle loans. Nevertheless, in the latter years of the 1995–2007 period, the firm’s problems cumulated. In the discussion in this and the next two sections, we emphasise those latter years. In this section, we discuss a threat to GM that arose from the particular governance structure of corporations in the United States, as well as some of the financial problems that GM encountered. We also review the degree to which GM’s financial survival is dependent on weakness in its domestic rivals. Stockholder threats GM experienced a financial attack in its home country in part because of the open nature of stock ownership in the United States. This allows investors to attempt hostile takeovers of corporations. It also permits investors to buy large stakes in a company, even if the investors do not contemplate a takeover, in order to influence the firm’s operations or policies. Kirk Kerkorian subjected GM to this. He had previously been heavily involved in the Chrysler Corporation (Belzowski, 1998), buying large amounts of stock in 1990 and again in 1995, and later liquidating the shares for a sizeable gain. In 2006, Kerkorian made substantial purchases of GM stock, though not a controlling interest. Kerkorian’s public statements indicated he believed
168 The Second Automobile Revolution
GM needed to change much more to remain competitive in the automotive industry, and he attempted to broker an alliance between Renault/Nissan and GM. GM’s board agreed to study this possibility, but subsequently decided not to become tied to Renault/Nissan. GM’s board seemed to view Kerkorian with suspicion. The board instead supported its CEO, Rick Wagoner, in his attempts to change the firm in ways that would downsize it in North America, stem its losses, and return it to profitability. Soon thereafter, Kerkorian sold all his GM shares, obtaining a modest profit. The chain of events is instructive regarding the great extent to which large stockholders, independently of the interests of employees, executives, communities, governments or other stockholders, can try to influence a corporation’s policy in the United States. This reflects the system of corporate governance that prevails in the United States, and that differs considerably from the systems extant in most of Europe and Asia (see Gourevitch and Shinn, 2005, especially chs. 2 and 3, for a recent exposition). Hostile takeover attempts in the United States are currently not very common because of the policies that have been developed to oppose them. Nevertheless, the diminished stock prices of General Motors and of many domestic automotive suppliers make such attempts, as well as aggressive moves to assemble large holdings of stock in order to influence corporations, feasible in this industry.
Finances and restructuring: losses and GMAC The combination of GM’s vast losses in 2005, amounting to over US$10.4 billion, and the criticism by Kirk Kerkorian pressured the company to respond with a restructuring centred on its North American operations. This meant taking very big charges to accomplish the goal of downsizing the company. The restructuring also meant abandoning the strategy of continuing to build vehicles, even when it was known they would sell for little over cost, just to generate some sort of revenue to offset legacy costs and to keep production workers out of the jobs bank, discussed below. The implementation of the restructuring has had its own problems. GM began it with a large cash hoard of over $20 billion, although at the rate the company was losing money in 2005, that sum would not last long. In order to raise additional funds, GM in 2006 sold a majority stake of 51 per cent in General Motors Acceptance Corporation (GMAC), its highly profitable financial arm. The purchaser was Cerberus Capital Management, a giant private equity firm. GMAC makes money by financing loans to customers purchasing new GM vehicles. It also arranges financing for the many vehicles GM leases to customers. Furthermore, GMAC is increasingly active in financing both used-car purchases and also purchases of new vehicles of other OEMs. GMAC also offers residential mortgages; in 2007, GMAC suffered heavy losses on this part of its business.
General Motors in an Age of Restructuring 169
Observers have had mixed opinions of the sale. Clearly, it did mean a very substantial payment, about $14 billion, to GM for the majority stake. And it appeared that, by becoming separate from GM, GMAC would have a higher credit rating and be able to borrow money for its operations at more favourable rates, and hence earn more. The sale would also compel GM to be more attentive to developing and selling vehicles. Some industry analysts, pointing to the very large proportion of total profits at GM that are due to GM’s financial arm, worry that the company has been unduly reliant on GMAC. Such analysts voice concern over what they call the financialisation of the entire company. Thus, some analysts welcomed the dose of discipline that the sale of GMAC might bring to the company. At the same time, other opinion has been quite negative, arguing that, for a one-time gain, GM sold the controlling interest in its most solid profit source. In 2008, the recession in the United States badly damaged not only GM, but also GMAC, deeply cutting its value, and forcing it to seek federal loans while reducing its own lending.
Domestic competitors: Chrysler and Ford Early in 2007, the future of the Chrysler Division of DaimlerChrysler became uncertain. Dieter Zetsche, DaimlerChrysler’s head, surprised the industry with his remarks in mid-February that ‘all options are on the table’ regarding Chrysler. The questions became: what entity, if any, would buy Chrysler, and what would be the effects on the industry, including key competitors such as GM? Although General Motors has had discussions with DaimlerChrysler about a purchase, it is hard to see GM buying all, or even a part, of Chrysler. In the North American market, the Chrysler Division’s product portfolio overlaps its own in many ways. And the Chrysler Division’s problems duplicate GM’s in a number of respects. However, Magna, the very large Canadian-based supplier, and several private equity firms made serious inquiries. And Kirk Kerkorian again made an offer for the business. Kerkorian’s and the private equity firms’ inquiries were ominous. While the largest of the private equity firms could bring to the table far greater sums than even Kerkorian could muster, most of the private equity firms do not presently have the executives with the expertise to operate a large assembler, nor does Kerkorian. This leads to the question of whether such a firm or Kerkorian would prefer to break up Chrysler and resell it in pieces after purchasing it, hoping to turn a profit. While this could be disastrous for many Chrysler employees and dealers, GM would probably benefit. Taking apart Chrysler and auctioning it off in sections would likely lead to a further decline in market share for what was left of the company. GM might well lure consumers who had been Chrysler owners but were leery about buying another Chrysler vehicle from a company, or set of companies,
170 The Second Automobile Revolution
with an unclear future. In addition, if GM is successful in attracting former Chrysler customers, this would benefit not only the company, but GM dealerships as well. This in turn could lighten the necessary task of reducing GM’s dealership body. Cerberus Capital Management, the successful bidder for the Chrysler Division, has access to extremely large pools of capital and, unlike most of the other private-equity suitors, has considerable experience in the automotive industry. Thus, with Cerberus as owner, Chrysler probably has its best chance of remaining viable. Nonetheless, if Cerberus does end up selling Chrysler, GM will reap an advantage. In the meantime, Ford, like GM, anticipated a business downturn. Ford accumulated a vast hoard of cash and credit, proportionately larger than GM’s. Thus Ford was in a stronger position than GM in 2008, and did not initially seek federal loans. Like GM, however, Ford’s product portfolio has drawbacks.
Stakeholders: General Motors’ relations with its production labour, its suppliers and the federal government Among the stakeholders that have a strong interest in General Motors’ prospects are its production workers, the supplier companies for whom it is a customer, and the federal government, which both regulates industryspecific aspects of the company’s environment and also helps determine the governance structure that GM and all other corporations in the United States operate under. We discuss these three stakeholders in this section; we emphasise the financial aspects of their relationships to GM. Labour relations: labour cost issues GM’s labour relations during this period continued to be difficult. A major source of the friction between GM and its production workforce is legacy labour costs. The sums GM requires to fund defined benefit pensions for a large and growing number of retirees are substantial.2 The length of GM’s pension obligations is great. In addition, GM’s costs for health care insurance are large and continue to grow, with cost containment being very arduous. (The United States, unlike a number of other industrialised nations, does not have an all-encompassing national health plan.) The result is ongoing pressure to garner revenue every year, just to meet these gigantic expenses. In addition, the structure of the contracts GM negotiated over the years with the United Auto Workers labour union (UAW) built in other costs that became harder and harder for the company to fund. In 1984, GM and the UAW agreed to a ‘jobs bank’ provision that meant laid-off workers were maintained at 95 per cent of their base wages for a substantial period of time. This created
General Motors in an Age of Restructuring 171
pressure to keep employees working, even if they were assembling vehicles that sold with little profit. The UAW for its part has wanted to defend the jobs bank, defend health care coverage for both current employees and retirees, and defend defined benefit pensions. It is also very sensitive to potential downsizing of the company and outsourcing of jobs, as these would mean the loss of union membership. This is a serious problem given the union’s lack of success in organising non-union plants of suppliers and other OEMs. The appendix table’s data on employment (‘labour’) does not distinguish between unionised and nonunionised employees. Nevertheless, most of the unionised GM workers are in the NAFTA region. GM’s NAFTA employment has sharply declined over the years, shrinking from about 614,000 in 1977 to 173,000 in 2005, for example, and thus GM’s UAW-represented employees have diminished considerably. Labour relations: plant closings and labour attrition In 2001, with the US economy already in a downturn after the heady years of the 1990s, the terrorist attack of 9/11 threatened further the possibility of an economic recovery. GM responded with a ‘keep America rolling’ programme that entailed zero per cent, multi-year loans for vehicle purchases. This resulted in strong vehicle sales, though initially at the expense, primarily, of Ford and the Chrysler Division. These companies then felt compelled to respond with generous incentives of their own. The fear was that such incentives would merely ‘pull ahead’ sales that otherwise would have been made at some point in the future. Surprisingly, sales for the industry as a whole in the United States held up rather well for several years. However, for GM, as well as for Ford and the Chrysler Division, this merely allowed the postponement of what these companies had to do, that is, reduce their production capacity and reduce their workforces. The tactic of delay could not be extended beyond 2005–6. In those years, in addition to GM’s huge losses in 2005, massive losses at Ford (over $12 billion in 2006) and large losses at the Chrysler Division (over $1 billion in 2006) compelled restructuring at all these companies. These three assemblers are going through costly plant closings and expensive downsizings of their workforces. The three domestic manufacturers and the UAW had to negotiate new contracts in 2007. The question that faced all of them was whether they could create and agree to contracts that accommodated the reduced market share that GM, Ford and Chrysler were likely to have in the United States. The UAW chose GM as the firm it would negotiate with first. GM and the UAW, after a brief strike, signed a new agreement. This set the general pattern for agreements reached between the UAW and Ford and Chrysler, though each contract had unique features. In the new agreement, GM achieved considerable changes in such areas as health care benefits
172 The Second Automobile Revolution
for retirees and wage rates for future workers. It is too early to tell whether the cost savings from the changes will be sufficient for the company. The savings will be substantial, however. The two-tier wage system means that new, non-core employees who are in the UAW will be paid much less than current employees in comparable job categories. The changes in health insurance mean that, after GM transfers substantial assets to the UAW, the union will assume all responsibility for health care expenses of GM’s UAW retirees. The legal device to accomplish this is a VEBA, or voluntary employees’ beneficiary association. For it to succeed, GM will have to provide it with sufficient assets, and the UAW will have to administer it well.
Supplier relations: Delphi GM’s relations with its suppliers have been in flux during this period, as the company has alternated its approaches to dealing with supplier operations, both external and internal. The leading theme has been GM’s connection to its formerly internal supplier organisations. Historically, General Motors had been the most vertically integrated of all the world’s major assemblers. Most of its internal supplier facilities had been organised into a body called as of 1995 the Delphi Automotive Group. In 1998, GM announced that the group would become a publicly traded and independent company the following year. A driving force was a belief that GM could cut costs by spinning off Delphi. If that was the wish, the outcome did not match it. Delphi began with high hopes. Its head, J. T. Battenberg, had been a well-regarded GM executive. Its stock shares were accepted by the public. Yet it is hard to see where the savings would come from. Delphi hourly workers had been unionised GM production employees, and their union status, including their contracts, wages and benefits, carried over to the new firm. Delphi did reduce its workforce, and this had the potential for some long-term savings. However, the up-front cost of this was high. With some financial assistance from GM, Delphi downsized its employee headcount by offering generous buyouts. Moving Delphi outside of GM led to the possibility that Delphi would be compelled to compete more vigorously against other outside suppliers for GM contracts than it had in the past. In principle this might lead to somewhat lower costs on parts contracts for GM. But at the same time, GM needed Delphi not to fail, because it was such an important source of parts and components for the company, and, unlike most of GM’s external suppliers, it was also a major centre for research and development for GM. In any case, after several years Delphi went into bankruptcy. Not only did its original stock become worthless, but the firm became ensnared in embarrassment. Some of its top executives, finding it impossible to balance revenues and expenditures, had been mis-stating financial records. When
General Motors in an Age of Restructuring 173
this was discovered, it forced Delphi to restate its reports and created legal troubles for the executives. The effort to spin off Delphi now looks like merely an unsuccessful attempt by GM’s legal and financial executives to impress Wall Street and ‘unlock shareholder value’. The other important theme in GM’s relations with its suppliers is the vacillation between more long-term relations with outside suppliers and the pattern of arm’s-length, short-term relations with them. Linked to this appears to be some uncertainty about how much development work to grant to suppliers, and also, how much responsibility to award to the Tier 1 suppliers for supervising their sub-suppliers. General Motors could give more attention to other possible configurations of supply chains. A thoroughgoing imitation of Japanese OEM supply chains is not necessarily appropriate for GM. But there may be other models, as Whitford (2005) has discussed in a recent book and as Gereffi at al. (2005) described in an excellent comparative article.
Government relations GM, like the other domestic OEMs, focused its efforts in this area on the federal government. Its government relations at the state level have been primarily oriented to garnering incentives (tax reductions, infrastructure upgrades) for new plant construction and for factory expansion and improvement.3 GM did not face strongly increased federal government pressure during the 1995–2006 period with regard to safety or vehicle emissions. Some changes were made to fuel economy (that is, corporate average fuel economy, or CAFÉ) regulations, but they were prospective, not to take effect until 2008. (The federal government tests fuel economy of automotive vehicles sold in the United States, and mandates that the average fuel economy of all a company’s vehicles sold meets certain standards, one for passenger cars, and one for light trucks. Failure to reach the standards subjects the manufacturer to heavy fines.) Nonetheless, there was in 2007 a lot of activity in the US Congress regarding CAFÉ. This led to legislation requiring much higher standards for vehicle fuel economy including a 35 miles per gallon average for all the automotive vehicles a manufacturer sells. Such standards will not become fully effective until the year 2020. However, this would have a major impact on GM, given its reliance on large, low-mileage vehicles. GM has continued to lose market share, while foreign OEMs continue to gain. But the company has not been effective in acquiring federal government support to deal with this threat. The company has been hamstrung by its own ideology. For decades, the slogan of American manufacturers, including GM, had been ‘free trade!’ This meant advocating that other countries open their borders to American goods, and keep their tariffs low, but it also implied
174 The Second Automobile Revolution
keeping US borders accessible to imported products. As foreign assemblers mastered high-quality manufacturing at competitive cost, they have thus been able to penetrate the US market, through imports as well as through vehicles built in transplants. The competitive relationships among the domestic OEMs meant that Ford, GM and Chrysler (as of 1998–2007, Chrysler Division of DaimlerChrysler) did not work effectively together for assistance in Washington. At the same time, the administration of George W. Bush, preoccupied with overseas wars and in any case much more attentive to finance capital than to manufacturing capital, seemed indifferent to the plight of the domestic OEMs. It did not help the domestic OEMs that many of their offices and plants were in states that did not support President Bush in his election campaigns. The more general problem is that GM and the other domestic assemblers, compared to firms in some other industries, have not been able to build as beneficial relationships with the federal government in the United States.
General Motors’ operations in the United States and worldwide In this section, the focus is on General Motors’ operations. We speak to some changes in the way General Motors develops new vehicles, as well as to its general strategy for its product portfolio. Then we try to place the company in the context both of its worldwide presence and of its place in its home market. In these latter sub-sections, we discuss acquisitions, divestments and partnerships. Product development policy and processes This time period saw more attempts by GM’s central administration to reform the firm’s product development process. By 1996, the company had created the position of vehicle line executive, or VLE. There were thirteen VLEs, and they were to work in concert with almost three dozen brand managers. In principle, this should shorten the time the development process requires. Furthermore, the goal is to ensure that the vehicles thus developed fit into an overall brand strategy, so that the vehicles created (such as new mid-size cars for two or three of the divisions) do not overlap excessively in terms of the customers they target. The new process grew out of the recognition that GM’s product development was flawed. The VLEs were to have more authority compared to the managers in the past who were in charge of product development projects, and who were somewhat at the mercy of divisional executives. This tended to stretch out development timelines. Also, in the 1980s, product development sometimes led to vehicles that were differently badged for the various
General Motors in an Age of Restructuring 175
car divisions, but so similar that they failed to halt GM’s declining market share. The new process is intended to reinforce GM’s long-term commitment that different models will really appear different to the customer, in terms of what the customer can touch or see. At the same time, the commitment is that elements of the vehicles that the customer does not touch or see can be shared even more than in the past across platforms: brake systems, and to some extent powertrains and suspension components, for example. The challenge here is to achieve a good balance between product differentiation and economies of scale. There was also a recognition that GM needed to take more risks with its product development: identify particular areas where it might create new vehicles, or new generations of existing vehicles, and try new directions in styling and vehicle type. From this grew successes such as the Cadillac CTS, the Hummer2, and the Chevrolet Malibu. From this, unfortunately, also grew the badly received Pontiac Aztek. However, the entire initiative is problematic. It represents another attempt to make centralisation work at General Motors. While it tries to encourage a variety of employees to develop ideas for new vehicles, it is premised on having central administration closely guide the development of those vehicle concepts and their evaluation. It reinforces the effort to have the various vehicle divisions, such as Chevrolet, or Pontiac, or GMC Trucks, or Buick, continue to serve as marketing and assembly units, while cutting somewhat their previous role as organisations that also develop the vehicles they build and market. And the initiative assumes that central administration has a better strategic vision than the divisions of what a good portfolio for the entire company consists of. In addition, GM has taken many years to shift its product development policy. What the company needs is a process that can bring to market quickly a considerable portfolio of differentiated vehicles, each of which is expected to sell in moderate volumes. In spite of that, for a long time General Motors wanted to hold on to more of a Fordist process that assumed larger volumes assembled in factories dedicated to particular models. Such factories could not shift from producing a vehicle with falling demand to another vehicle with increasing demand. In the past, this meant GM had more assembly plants than it needed, some of them running at well below capacity. For the new product development process to be fully effective the underlying policy will have to include an ongoing effort to convert assembly plants to be more flexible in terms of what they can build. In addition, the policy will require moving further away from the Fordist emphasis on economies of scale while continuing to find ways to earn profits on a portfolio of smaller, but distinctive, model runs. Most importantly, full cooperation from the divisions will be necessary.
176 The Second Automobile Revolution
GM’s overseas operations A curious aspect of GM’s situation is that, despite its current losses in the United States, its overseas operations are less problem-laden at this time. GM has growing production and sales in China. This is based in part on a longterm and clever if unlikely initiative to build and retail Buicks there. While GM has made practically no headway in Japan, that is true for most other non-Japanese OEMs as well. In other parts of Asia, including India and Thailand, the company has experienced rapid growth, albeit from a base that is not yet really large. GM has a manufacturing as well as a sales presence in Australia. Its European operations, notably Opel, are satisfactory now. Its sales in South America have been growing quickly. Evidence of the overseas expansion can be seen in the appendix table, in terms of employment numbers outside NAFTA nearing those inside NAFTA for the most recent year. Part of this expansion has been accomplished through acquisitions. GM has a mixed record in this area. GM’s purchase of the auto division of Saab, completed in 1995, has not been particularly successful. General Motors’ involvement with Fiat led to a much more serious loss, compelling GM to pay about $2 billion to void most of its joint venture with the Italian assembler. GM has bought equity stakes in other OEMs, including Isuzu and Suzuki, but in several instances has sold all or most of such stakes. Nevertheless, General Motors’ purchase of a substantial amount of the Korean automaker, Daewoo, has been more fruitful. And GM’s stake in Fuji Heavy Industries, though sold recently, was not unreasonable. Of less visibility are the partnerships with other OEMs for specific purposes. Some of the partnerships are oriented to technical tasks, such as agreements with other OEMs to develop major elements of powertrains or complete engines. These more limited joint ventures and partnerships appear to be more successful than the equity stakes GM has bought. What distinguishes General Motors from a number of foreign OEMs is that the company is more concentrated in its home region (the United States and the adjacent nations of Canada and Mexico). To be sure, a solid majority of GM’s sales are now made outside the United States, but the proportion of non-home country sales for some of the foreign OEMs is rather higher. Furthermore, the profit rate of the vehicle mix GM sells in the United States is higher than the profit rate for its vehicles sold overseas. Thus unlike some of the other major assemblers, GM’s fate is more dependent on its success in its home market. The problem is that the United States market, though large, will not increase rapidly. The Japanese manufacturers face a similar dilemma in their home country, but are advantaged in being better positioned overseas. The question then becomes whether GM can grow its overseas operations sufficiently to overcome the limited potential of its home market. General Motors’ global presence is expanding, but, viewed as a worldwide entity, the firm continues to consist of semi-autonomous units, organised
General Motors in an Age of Restructuring 177
along regional lines, rather than forming a tightly integrated whole. This is not completely impractical. Successful global assemblers must adapt their presence to the local situation in each region and to some extent in each country where they have operations, and this requires a measure of local independence (Senter and Nyce, 2006). Nonetheless, in GM’s case, this runs the risk that the company both forgoes some economies of scale, and also misses the internal learning that rests on the units of a firm working more closely with each other. (For a much more detailed treatment of GM’s, as well as Ford’s, strategies for their overseas operations, see Bordenave and Lung, 2003.) However, we should point out that GM is more of a global company than Ford, and certainly is more of a global company than the newly independent Chrysler. This is not only a matter of having a larger international footprint. It is also due to GM’s leading its domestic competitors in the degree of international integration it actually has achieved. Though GM’s operations in various nations and regions may be semi-autonomous, the company has been fostering cross-national platforms as well as common powertrain components that are used in many nations, and is somewhat ahead of Ford and Chrysler in this. The company’s current level of success as a global business was preceded by internal conflict, however. In particular, tensions between the company’s headquarters and its European operations were real for a number of years. These tensions in part flowed from the sequence of internationalisation that GM followed. From being a company that initially was active in the United States and then added operations in other countries, GM shifted to a structure in which its operations in various countries were integrated into one or another region. The later process of trying to link the regions firmly into a single worldwide structure did not go well. In recent years, the company has improved by allowing more regional autonomy. This has permitted greater distinctiveness from one region to another in the vehicles each region develops, which in turn has increased sales. GM’s home market: uneven prospects General Motors’ position in its home country is threatened. The company seems finally to have accepted a diminished market share there, and to be adapting to that, in terms of shedding employees and closing unneeded factories. At the same time, it is less clear that GM has a well thought-out plan for defending the market share it still has in the United States. In the recent past, it has targeted specific opportunities. As passenger cars became less of a proportion of the domestic automotive market, GM emphasised vehicles in the light truck category. In addition, noting the money to be made in the luxury segment, GM invested very heavily, and successfully, in renewing its Cadillac division.
178 The Second Automobile Revolution
Nevertheless, GM participates in other areas of the market where the company has made much less of a mark. These areas include small cars as well as vehicles produced to sell in the United States by foreign assemblers, such as Isuzu, Suzuki and Saab, firms in which the company had or still has an equity stake. Despite downsizing its employee numbers and closing some of its factories in the United States, it is arguable that GM still has a product portfolio that is too large. There is considerable duplication of vehicle types across the divisions. To GM’s credit, however, it has shed operations that are not closely connected to automotive vehicles. These operations were outside the company’s primary area of emphasis and skill. GM is thus participating in the movement away from conglomerate and conglomerate-like modes of business organisation. At the same time, General Motors continues to need to improve the development process for the products on which it does concentrate.
Conclusions A decade ago, Michael Flynn (1998) titled his chapter on GM with a question: ‘The General Motors Trajectory: Strategic Shift or Tactical Drift?’ The implicit answer was that GM was not engaged in a strategic shift, though it was changing. That remains the case. In this section, we first discuss two characteristics of company structure that create limitations for General Motors. Then we close by referring to an aspect of the corporate governance form prevalent in the United States, pointing to the problems that poses for GM, and discussing how the company has partially solved these with its overseas operations. A central problem: reforming GM’s organisational structure It is important to emphasise another dimension of General Motors’ business environment. GM currently confronts a suite of competitors, including several that are very large and well financed, that contain some of the advantages of the General Motors structure, but that have also grappled successfully with or avoided some of the dilemmas GM now faces. Those dilemmas include two major internal tensions, as well as an external problem, discussed in the next sub-section. One of the tensions GM needs to resolve is between its finance leadership and its product leadership. The pre-eminence of finance leadership at the executive level has probably led to uncertainty regarding how to adapt the product development process and the manufacturing divisions to the competitive threats GM faces. Womack et al. (1990) offered an early example of this in their description and analysis of the very long time it took GM to bring to market a new generation of mid-size passenger cars. They suggest that
General Motors in an Age of Restructuring 179
the lack of authority given to product development teams, compared to the power of the manufacturing divisions themselves, compromised the effort. Underneath this was a structure that advantaged the divisions in comparison to the teams. The hope for GM is that the system of vehicle line executives, in existence for over a decade, will help remedy this problem without creating other problems. The other, and overlapping, tension is between the central administration and the divisions. Alfred Sloan’s ongoing project of bringing together divisional executives and executives from central administration in committees helped make for informed decisions. It also helped create the consent so that decisions, once taken, would actually be implemented rather than undermined. We suspect that a declining emphasis on this aspect of corporate governance has led to some of GM’s mistakes. Would a greater voice for divisional executives have helped win over central administration to the importance of more product differentiation in, for example, the company’s mid-size cars? Clearly, central control has been crucial over the decades in preventing the overspending that nearly destroyed General Motors more than once in its formative years. Nonetheless, in part because the central administration has not included many executives with engineering and manufacturing experience, ways must be found to use more effectively GM’s expertise in those areas, but without pre-empting and alienating the divisions. The VLE initiative brings to bear more of that engineering and manufacturing talent, but puts at risk the cooperation of the divisions. Robert Freeland’s (2001) book argues that centralisation more and more characterised GM after 1958, with GM increasingly taking on an M-form structure that abandoned the decentralisation that Alfred Sloan fostered. The resultant concentration of power in an administration that consisted primarily of finance executives led to ever-increasing levels of conflict between the top executives of the company and the divisional leaders. Thus we believe that a portion of General Motors’ problems stems importantly from discarding the adaptations of the M-form that Alfred Sloan continuously crafted. At the same time, it is uncertain whether even a structure at GM that brings back those adaptations will be sufficient to restore the company. In the conclusion to their analysis of the decline of conglomerates in the United States, Davis et al. (2004) raise the question of what the life span of the traditional structure of firms, such as the M-form, may be. The same question, in a different format, is raised by Powell (2001) in his chapter on new patterns of organisation in capitalist firms in Western nations. As businesses do more outsourcing and increasingly resemble networks of shifting alliances and partnerships, an open question is how competitive a big assembler can be if it retains a structure of multiple divisions each of which employs a large complement of staff and white-collar specialists and managers, in addition
180 The Second Automobile Revolution
to its production workers. GM is committed to downsizing its employee base, white-collar as well as blue-collar. And yet unless there is some reconceptualisation of how the divisions and departments of the company should be structured, it is unclear whether a smaller GM will be more competitive than the larger GM of the past. This is not to advocate a U-form for General Motors, because simply moving around the current employees within the company is unlikely to advantage it. Instead, the company could benefit by thinking about what staff functions it should retain, and whether there are any that it could shed. To be fair to Alfred Sloan, when he was creating his adaptation of the M-form, he could not have foreseen that GM would eventually have such a vast international presence, and be both assembling and selling more vehicles overseas than in the United States. The company is still trying to build a structure that is appropriate for this situation. The attempt to tie closely together the regional units of the company did not work well, but the company had the insight to grasp this, and to devolve some authority back to the regions.
A final corporate governance issue There is no question that GM is under threat. GM’s total revenue for its most recent fiscal year was $207 billion; Toyota’s, for its most recent year, was $179 billion. Yet Toyota’s market capitalisation as of early January 2008 was about fourteen times as great as GM’s. Clearly, the financial markets are far more impressed by Toyota than by GM. This leads to the external problem. To be sure, the financial side of the business environment in the United States is a demanding one for manufacturing firms set up as corporations. This is an ironic change from the nineteenth century, when the corporate form of organisation became available to manufacturing firms, as Roy (1997) discusses, and hugely benefited manufacturing businesses such as the OEMs in the early twentieth century (see also Prechel, 2000 and Fligstein, 2004). An important element of the contemporary American business environment, its corporate governance structure, is a very open, liquid equities market, undergirded by increased protections for minority shareholders4 (Gourevitch and Shinn, 2005). Such a market enhances the belief in shareholder value (Krier, 2005), which in turn has become more important in the conception of the firm. Corporations in the United States sometimes find themselves facing the risk of the type of onslaught that Kirk Kerkorian carried out against GM recently. Kerkorian’s strategy, as a corporate raider, is to use the protections afforded the minority shareholder by the American system of corporate governance while assembling blocks of stock that give him great, if not majority, influence in a company. The shareholder conception as it exists in the United States is linked to publicly traded businesses devoting great attention to their financial results for the next three months, perhaps at the expense of efforts that
General Motors in an Age of Restructuring 181
are more long term in their objectives. The huge size of the American consumer market and vast expanse of the American capital market lure firms to incorporate in the United States.5 As Europe’s and Asia’s consumer and capital markets expand, however, automotive firms, whether OEM or supplier, may have second thoughts, and choose to forgo incorporation in the United States, with its openness to shareholder assaults, for incorporation elsewhere. This option is scarcely available to General Motors. An alternative for such a firm is to devote increasing attention to the growing consumer and capital markets outside the United States, which GM is doing, as the data on production in the appendix table indicate. Indeed, in this respect GM resembles other very large, global OEMs. Each of these companies has come to terms with the particular corporate governance structure in its home country. As it expands overseas, it must be careful in how it exposes itself to the corporate governance structures of the nations where it constructs new operations. For example, Toyota and BMW, both of which are headquartered in nations with the blockholder model of corporate governance, have operations in the United States. However, their American presence is far less vulnerable than is GM’s to the attacks allowed by American corporate governance rules. And similarly, we wonder whether General Motors’ presence in nations such as Germany may be aided by the fact that at least it is not as subject, compared to BMW, Daimler and Volkswagen, for example, to Germany’s blockholding, with its particular limitations. The challenge for General Motors is to adapt, somehow, to the short-term pressures that are endemic to the American system of corporate governance, while the company constructs an organisation that is stronger at developing and implementing larger strategies, as distinguished from the more limited, though in some cases very profitable, initiatives that GM has been undertaking.
7,725,014 7,456,000 8,776,000 8,149,000 8,801,399 8,781,603 8,073,000 8,533,168 8,246,000 9,098,000 9,051,000 9,181,000
Total
5,390,809 5,110,000 N/A N/A 5,695,425 5,728,698 N/A 5,151,860 N/A N/A N/A N/A
Cars
Worldwide production
2,334,205 2,346,000 N/A N/A 3,105,974 3,052,905 N/A 3,381,308 N/A N/A N/A N/A
CV
5,373,368 5,045,032 5,472,251 4,973,982 5,739,413 5,630,124 5,003,149 5,507,412 5,301,902 4,991,924 4,576,080 4,393,585
Total 3,175,502 2,790,266 2,955,106 2,632,240 2,874,802 2,814,622 2,313,536 2,398,496 2,110,125 1,927,266 1,772,246 1,762,723
Cars
NAFTA production
General Motors, 1995–2006
2,197,866 2,254,766 2,517,145 2,341,742 2,864,611 2,815,502 2,689,613 3,108,916 3,191,777 3,064,658 2,803,834 2,630,862
Trucks 160,272 164,069 178,252 161,315 176,558 184,632 169,051 177,867 185,837 195,351 194,655 207,349
6,881 4,963 6,698 2,956 6,002 4,452 601 1,736 3,859 2,701 (10,417) (1,978)
Revenue Net income ($ millions) ($ millions) 0.03% 0.02% 0.03% 0.01% 0.02% 0.02% 0.00% 0.01% 0.01% 0.01% −0.02% −0.01%
ROA
649,000 605,000 608,000 594,000 391,000 390,000 366,000 338,000 326,000 324,000 335,000 N/A
434,000 245,000 237,000 226,000 173,000 213,000 202,000 198,000 190,000 181,000 173,000 N/A
Worldwide NAFTA
Labour
Notes: 1. Worldwide labour is average annual employees 2. ROA is calculated net income/average total assets 3. 1996 Worldwide CV number is actually light trucks 4. 1997, 1998, 2001: total worldwide production not available, so worldwide sales are given 5. ROW is Rest of the World (Outside of NAFTA). FTA Employment Sources: 1. Wards Automotive and World Motor Vehicle Data: Production data. 2. General Motors Annual Reports (1995–2006): Labour, numbers used to compute ROA; Total worldwide production (2003–2006); Substitute worldwide production total (sales).
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Appendix Table 9.1
215,000 360,000 371,000 368,000 218,000 177,000 164,000 140,000 136,000 143,000 162,000 N/A
ROW
182
General Motors in an Age of Restructuring 183
Notes 1. In practice, however, at the edges each division at GM had the potential to overlap, and hence compete to an extent, with the next higher- or lower-ranking division. After the middle of the twentieth century, this tendency became rather pronounced at the company. 2. Defined benefit pensions in the United States, like social security benefits funded by the federal government, are not as actuarially sound as well-managed defined contribution pensions. Defined benefit pension plans often require continuing contributions from the employer, in addition to the original contributions by the employer (and, where required, original contributions by the employees) to fund the payments to retirees. While a variety of government bodies, federal, state and local, still offer defined benefit pensions to employees, the trend has been away from that type to the defined contribution type. Private employers, such as corporations, are shifting to defined contribution pensions. Further, some companies, especially those with non-union workforces, may have no pension plan at all for their employees. 3. However, state governments are becoming more active regarding the automotive industry. California and some of the states in the northeastern part of the US have developed standards regarding vehicle emissions that are more stringent than the federal government’s. The assemblers generally prefer that regulations concerning automotive vehicles be made by the federal government, in order to avoid the burden of multiple and possibly conflicting state regulations in the same area, such as safety or emissions or fuel economy. However, the manufacturers are learning that they may not always have this benefit. Litigation is ongoing. 4. This is a considerable change from the nineteenth century, when minority shareholders had far fewer protections. Further, shareholders today have more power in that many of them are large mutual funds, insurance companies, pension funds and banks, whereas in the nineteenth century minority shareholders were typically individuals. 5. The high level of minority shareholder protection may also be an attraction for some firms from certain nations; however, this level varies depending on what state the firm chooses for incorporation. See Gourevitch and Shinn (2005: 288–90).
Bibliography Belzowski, B. (1998) ‘Reinventing Chrysler’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 242–69. Bordenave, G. and Lung, Y. (2003) ‘The Twin Internationalization Strategies of US Automakers: GM and Ford’, in M. Freyssenet, K. Shimuzu and G. Volpato (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan, pp. 53–94. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Cray, E. (1980) Chrome Colossus: General Motors and its Times. New York: McGraw-Hill. Davis, G. K., Diekmann, K. A. and Tinsley, C. H. (2004) ‘The Decline and Fall of the Conglomerate Firm in the 1980s: the Deinstitutionalization of an Organizational Form’, in F. Dobbins (ed.), The New Economic Sociology. Princeton, NJ: Princeton University Press, pp. 188–224. Fligstein, N. (2004) ‘The Transformation of Corporate Control’, in F. Dobbins (ed.), The New Economic Sociology. Princeton, NJ: Princeton University Press, pp. 407–32.
184 The Second Automobile Revolution Flynn, M. (1998) ‘The General Motors Trajectory: Strategic Shift or Tactical Drift?’ in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 179–210. Freeland, R. F. (2001) The Struggle for Control of the Modern Corporation: Organizational Change at General Motors, 1924–1970. Cambridge: Cambridge University Press. Gereffi, G., Humphrey, J. and Sturgeon, T. (2005) ‘The Governance of Global Value Chains’, Review of International Political Economy, 12(1): 78–104. Gourevitch, P. A. and Shinn, J. (2005) Political Power and Corporate Control: the New Global Politics of Corporate Governance. Princeton, NJ: Princeton University Press. Krier, D. (2005) Speculative Management. Albany, NY: SUNY. Norton, S. (2004) ‘The Schumpeterian Shock at General Motors in the 1920s’, paper presented at the International Industrial Organization Conference, April. O’Brien, A. P. (1997) ‘The Importance of Adjusting Production to Sales in the Early Automobile Industry’, Explorations in Economic History, 34: 195–219. Powell, W. W. (2001) ‘The Capitalist Firm in the Twenty-First Century: Emerging Patterns in Western Enterprise’, in P. DiMaggio (ed.), The Twenty-First-Century Firm: Changing Economic Organization in International Perspective. Princeton, NJ: Princeton University Press. Prechel, H. (2000) Big Business and the State: Historical Transitions and Corporate Transformations, 1880s–1990s. Albany, NY: SUNY. Roy, W. G. (1997) Socializing Capital: the Rise of the Industrial Corporation in America. Princeton, NJ: Princeton University Press. Senter, R. H., Jr. and Nyce, S. (2006) ‘Human Resource Issues for Globalizing Automotive Suppliers’, Automotive Component Manufacturers, Automotiveworld.com, 6 September, www.automotiveworld.com/ACM/content.asp?contentID=54332 Sloan, A. P., Jr. (1990 [1963]) My Years with General Motors. New York: Doubleday. Whitford, J. (2005) The New Old Economy: Networks, Institutions, and the Organizational Transformation of American Manufacturing. Oxford: Oxford University Press. Womack, J. P., Jones, D. T. and Roos, D. (1990) The Machine that Changed the World. New York: Macmillan.
10
Ford, 1993–2007: Losing its Way? Glenn Mercer
From the mid-1990s onward the Ford Motor Company entered a period of decline, continuing to the time of this writing, although for the first part of this downward trajectory it remained concealed by enormous profits earned from North American sales of trucks.1 The decline was driven by a loss of strategic direction, which this chapter will investigate. By 2000 or so the trajectory was clear and a series of turnaround efforts was launched, whose eventual results are as yet unknown. The arrangement of the chapter is as follows. First, we will set the stage by discussing developments leading up to the period in question, primarily by recapping Gerard Bordenave’s review of Ford’s history in the predecessor to this volume, One Best Way? (Freyssenet et al., 1998). Next we will review the course of the company under the guidance of the four individual men who served as CEOs during this time, as each one set a particular stamp on Ford. Finally, we will conclude by extracting and reviewing several crosscutting themes that perhaps explain the company’s trajectory, and point to a possibly brighter way forward for Ford.
Introduction: setting the stage The Ford Motor Company thrived throughout the 1960s and 1970s, safely tucked under the price- and product-umbrella that General Motors provided, maintaining a solid second place market share position to GM, and staying well ahead of third-place Chrysler/American Motors. The American market at this time was well suited to Fordist mass production, as individual model annual production runs reached almost to a million, and model updates were both widely spaced in time and generally superficial in nature (e.g. often limited to cosmetic changes in exterior sheet metal). Imported vehicles were present mostly in the form of the VW Beetle, which the Detroit makers hardly considered competition, as a trace of imported Japanese cars (mostly on the West Coast), and as some tens of thousands 185
186 The Second Automobile Revolution
of mostly upscale European imports (whose buyers Detroit considered well outside the American mainstream). Sales volumes might fluctuate with overall economic conditions, but these periodic recessions were well understood and not considered threatening to either GM or Ford, although third-place Chrysler was much more at risk due to its lower volumes. (This fact led Chrysler to constantly experiment with volume-building and cost-sharing alliances, such as the purchase of American Motors, a long-term JV with Mitsubishi Motors, and then of course the eventual sale to Daimler.) Each of the three main competitors had overseas operations, primarily in Europe, but the success or failure of each remained firmly dependent on the North American market. This peaceful oligopoly was overthrown by the so-called oil shocks of 1973 (related to the Yom Kippur War) and 1979 (related to the Iranian revolution), which drove up the low price of gasoline in the USA and thus opened the door to highly efficient Japanese cars.2 At about the same time the US government enacted the Clean Air Act and various amendments to it, which required much lower emissions of pollutants from cars. The Detroit OEMs were faced with the dual challenge of simultaneously converting their powertrains to be both more economical and cleaner. Uncounted billions were spent on doing this, which involved moving from carburetion to fuel injection, from rear-wheel to front-wheel drive, and from untreated exhaust to catalytic converters and exhaust gas recirculation. Detroit launched numerous import-fighting products along the way (including the Ford Pinto), but as these were rushed into production prematurely they fell short and did little to stem the Japanese tide. By 1985 or so the worst was over, in that the American carmakers had made the necessary conversions, but by then the Japanese had established a market-share beach-head which they would never thereafter give up. The game had changed. Ford fought back strongly in this period, on many fronts. It diligently studied and tried to adopt the principles of lean production that seemed to underpin the Japanese productivity advantage (aided by its acquisition of a minority share in Mazda in 1979); it implemented new systems (e.g. the Q1 quality management process) to help close the gap in perceived quality between its products and those of the leading Japanese firms; and in 1985 it regained momentum in the product arena with the launch of the Ford Taurus, acclaimed by many as ‘the car that saved Ford’. Ford would go on to sell over 6 million of this radically new design, which would become the best-selling car in the USA for several years in the 1990s. The company also launched, in 1990, the Ford Explorer, arguably one of the first modern (that is, car-like in terms of ride and features) SUVs3 : this vehicle would go on to become the best-selling SUV in the USA for fourteen consecutive years. While its mainstream business was now back on track, Ford realised it needed to supplement its premium vehicle lines (Mercury and Lincoln) with additional product, as the American market was steadily moving upscale, and
Ford, 1993–2007: Losing its Way?
187
determined that the best route to this goal was via acquisition. Accordingly, Ford purchased Aston Martin in 1987 and Jaguar in 1989. Additionally, the shock of the Japanese onslaught had made management realise that the company would have to compete globally, which meant making more use of both Mazda and Ford Europe. Accordingly, Ford increased its stake in the Japanese firm to a controlling interest on the one hand, and on the other attempted to jointly develop with Ford Europe so-called ‘world cars’ that could achieve global scale. The first result of the latter effort was the Mondeo of 1993, sold as the Contour and Mystique in the US. As one of the costliest product launches ever (reports estimated some $6 billion were invested), disappointment was all the greater when the car failed to meet its sales targets. Blame for the failure can be apportioned to various causes, including political infighting at Ford, misdesign of the product for certain major markets, and a failure to truly commonalise parts (this ‘world car’ was in fact made of very different components in different parts of the world). As we shall see, the first CEO we will study, Alex Trotman, would take as a main task rectifying this failure to leverage Ford’s product development4 resources globally. He would also take steps to improve the stand-alone profitability of Ford Europe, which had been part of the company for over half a century, but which was failing to deliver consistent profitability. One final development will bring this abbreviated history of Ford up to the start of our selected time period. This was the increasing ‘financialisation’ of the company, defined as a growing reliance on profits from services related to car transactions, rather than profits derived from car manufacturing. Ford Motor Credit of course provided income from car loans and leases, but Ford in the 1980s and early 1990s had very actively diversified further into financial services, not all of them even car-related. In 1985 Ford had acquired the large savings and loan institution First Nationwide Bank (to diversify into other consumer financial services). In 1994 it moved from a minority stake to 100 per cent ownership of Hertz (thus gaining a big share of car rental revenues). And in 1989 it spent billions to acquire the large consumer finance company The Associates. By the mid-1990s Ford’s profitability was highly dependent on these financial sources: in fact, financial operations including Hertz contributed 75 per cent of Ford’s cumulative net income between 1990 and 1995.5 To sum up then, as we approach the mid-1990s, we can characterise Ford in this way: generally the company had recovered from the Japanese invasion, especially thanks to two key products, Taurus and Explorer; it had armed itself with acquired upscale brands to buttress its domestic premium lines of Mercury and Lincoln; the company was not yet able to effectively leverage global scale and scope in product development; and Ford was increasingly dependent, for better or worse, on financial services as a source of income. As far as the broader American truck and car business was concerned at this time, it remained as exposed as ever to economic cycles in the US, and
188 The Second Automobile Revolution
accordingly the recession of 1990–2 had not spared either GM or Chrysler or Ford. Ford suffered large losses in 1991 and 1992. But by 1993, as Mr Trotman arrived, the recovery had begun, and Ford was already rebounding. It is here that we begin the main part of our story.
The Trotman era: 1993–1998 Baron Alexander James Trotman (he was awarded the peerage in 1999) was Ford Motor Company’s first foreign-born CEO, in this case hailing from the United Kingdom. Trotman got his start at Ford UK (an incubator for many Ford corporate executives over time), working his way up to become CEO in November 1993. His time in the office is marked by three key developments, in the arenas of products (e.g. SUVs), processes (e.g. product development) and M&A (e.g. Volvo). Following the introduction of the Ford Explorer in 1990 (and Chrysler’s Jeep Grand Cherokee in 1993), the age of the modern American sport utility vehicle began. The SUV was much more than a new model (as the 2007 Camry is a model of sedan): it defined a new segment of the market, and thus stimulated incremental demand; with the modern SUV on the market Americans either added additional vehicles to their driveways or purchased a new vehicle sooner than they had otherwise planned.6 It is almost impossible to overstate the financial impact of these vehicles on Ford, GM and Chrysler. An excerpt from Keith Bradsher’s history of the modern SUV, High and Mighty, will best make this point: The Michigan Truck Plant [which made the Explorer and other vehicles such as the even larger Expedition] had become the single most profitable factory in any industry anywhere in the world. It was cranking out 1040 full-size sport utilities every workday. The factory’s annual production was worth almost $11 billion – greater than the global sales for Fortune 500 companies like CBS, Texas Instruments, Honeywell, and Nike . . . The factory’s profits from those sales were even more spectacular: about $3.7 billion in pretax profits . . . while Ford had 53 assembly plants worldwide, the Michigan Truck Plant accounted for a third of the company’s entire profits. There were fewer than 100 companies in the world that earned more than this single factory did in 1998. (Bradsher, 2002: 89; emphasis added) Production of SUVs was so profitable due to reasons of both demand and supply. On the demand side, the virtual absence (at this time) of Japanese competition meant that, at least as far as SUVs were concerned, Detroit had recreated the profitable oligopoly of the 1960s. Additionally, the American public was eager to purchase the product, not just because cheap gasoline made it inexpensive to run, but because it offered real benefits. Essentially, the SUV provided large amounts of interior space (crucial for American
Ford, 1993–2007: Losing its Way?
189
families used to hauling bulk cargo such as groceries from Costco or children to hockey games) without the ‘soccer mom’ stigma of the minivan, and with a higher seating position that offered better visibility than a station wagon. Additionally, the SUV had a characteristic key to most prior very successful American vehicle launches: it was something new. The modern SUV was not just the latest iteration of a pre-existing product line: it was a new product category entirely, combining truck toughness and space with car-like ride and handling. American fervour for something new had previously ignited demand for the ‘pony car’ in the mid-1960s7 and the minivan in the 1980s: there was always significant pent-up demand for the next new thing. The SUV market share, less than 3 per cent in 1980, was over 15 per cent by 1995. The love affair was in full swing. On the supply side, Ford and the other Detroit makers were only too happy to offer up as many SUVs as Americans wanted. This was because the vehicle was fundamentally very profitable. Built on frame rails rather than a unibody, it was much easier to weld and assemble than a car. Expected to be big and brawny rather than sleek and smooth like a car, it did not require a sophisticated (expensive) engine or a highly insulated (expensive) interior. With no expectations for sports-car handling, suspensions could use primitive beam axles and heavy steel (versus costly aluminium) components. With styling basically reduced to a large box on wheels, expensive dies for stamping exotically curved body panels could be avoided. All these factors cut per-unit variable cost, and then per-unit fixed cost was driven down by adding in the enormous volumes of the pickup trucks on which the Detroit SUVs were based.8 Put high demand (and thus high price realisation) and low cost together and high margins result. Even as late as 2000 Ford was realising variable profit margins of about 50 per cent on the Explorer line: roughly $12,000 for each Explorer wholesaled to dealers at roughly $24,000. The comparable figure for cars like the Taurus was $4,000 or less, and after fixed costs were subtracted, the average Ford car in 2000 was probably only at break-even. By 1995, therefore, the SUV (and the pickup trucks on which it was based) were likely providing more than 100 per cent of Ford’s North American profits – a reality which would come back to haunt the company when the SUV boom subsided, and the company had to go back once more to relying on cars for profits. Even while the SUVs were bringing in enormous profits, Ford realised that its prosperity was concealing significant weaknesses. Yes, the Taurus had been a hit, but it had been the only one (on the car side) in years. The Mondeo, an attempt at a world car, had taken much too long to develop and had gone well over budget. The Ford development process was slow relative to Japanese benchmarks. These problems, Trotman’s team concluded, were the result of underlying organisational issues including: (a) lack of integration among Ford North America (NA), Ford Europe (EU), Ford Asia Pacific (AP) and Ford
190 The Second Automobile Revolution
Automotive Components (AC); and (b) functional organisation ‘chimneys’ that prevented synergistic interaction among manufacturing, engineering, marketing and other realms. The proposed solution to this problem was Ford 2000, a gigantic internal merger and restructuring of the entire car company. Ford North America and Ford Europe were combined with Ford Automotive Components into the umbrella group Ford Automotive Operations (Ford Asia Pacific was to follow later). Product development was wrenched 90 degrees around on the organisation chart, from a geographic focus (e.g. NA, EU, AP) into a product focus: five vehicle programme centres (VPCs) (each specialising in a vehicle type such as front-wheel drive (FWD), rear-wheel drive (RWD), light truck) were set up. And the entire organisation was matrixed along the two dimensions of the VPCs on the one hand and functional departments on the other: now every person would have two or more managers to report to. The merger was to create purchasing and engineering scale; the formation of the globally focused multi-disciplinary VPCs was to improve development performance (both time-to-market speed and development cost); and the matrix setup was to tear down the functional chimneys. While this would not become evident for a few years, Ford 2000 would turn out to be a failure. More charitable observers consider it the right idea at the wrong time, poorly executed; others believe it was an exercise in futility from the start. Blame can be apportioned to various factors. First, while the point was to be global, four of the VPCs were placed in Detroit, where they could hardly get an international perspective on product trends and needs, and where they were subject to the headquarters infighting that Ford was famous for. Next, while one goal was to gain scale, when each VPC tried to optimise its own results corporate scale was divided by five and as a result the expected cost savings were modest at best. Then, too, while the old Ford NA and Ford EU structures had their problems, they at least had a clear P&L (‘profit & loss’) focus, which seemed to become lost in the new structure. But finally and perhaps more importantly, Ford employees generally did not see the rationale for such an upheaval (after all, they reasoned, look at the profit numbers). Thus in ways big and small they undermined the implementation of Ford 2000 (despite heroic efforts at communication by Trotman), such that more effort seemed to be spent explaining why things couldn’t be done differently, than in doing them differently. The programme bogged down under its own paperwork (it was rumoured that by the time the Ford 2000 Powertrain Requirements planning document had made its way through all the editing rounds, it was over 1,000 pages long!) and in a few years was essentially abandoned. As we shall see, valuable time had been lost to competitors while the Ford 2000 wrangle was being sorted out. At the end of Trotman’s reign,9 Ford made one of the largest acquisitions of its history, picking up the car operations of Volvo for about $6.5 billion.10 Realising that the ailing Jaguar, Mercury and Lincoln brands could not carry
Ford, 1993–2007: Losing its Way?
191
Ford into the premium market segment on their own, Ford hoped the addition of Volvo would let it create a multi-brand premium-brand powerhouse, a plan which the next CEO, Jac Nasser, would begin to execute. The purchase was easily funded, both by light truck and SUV profits, and by the enormously profitable sale of The Associates. As to this latter transaction, it seemed that Ford had realised that financialisation could only go so far, and should be limited, in the case of a car company at least, to car-based transactions. Accordingly, while the company increased its stake in Hertz to 100 per cent in 1994 (floating a minority interest to the public a few years later), in the same year it sold off the struggling First Nationwide Bank, and then in 1998 the (very profitable) Associates group. Thus by 2000 or so the contribution to Ford’s total net profits from its financial operations had fallen from about 75 per cent to about 25 per cent, although this figure would later start to rise again.
The Nasser era: 1999–2001 While Ford was in good shape financially in the late 1990s there was growing concern that the underlying cost base was not sound, and this served as impetus to promote Jac Nasser into the CEO spot, famous as he was throughout his Ford career as a cost-cutter. An Australian of Lebanese origin who had spent time in Latin America and Europe for Ford, he was also thought to bring the international experience needed to knit the far-flung empire together, in a way Ford 2000 could not. But Nasser’s legacy would turn out to be very different from what one might have expected when he arrived in 1999. Nasser was impressed by the high profits (and stock prices) of consumer goods companies such as Nike. As he reportedly joked about the car business: ‘We make cars, everyone else makes money!’ He was also thrilled by the heady developments of the late-1990s internet boom, and the possibilities it offered. He determined to bring both themes to Ford. Accordingly, he repositioned Ford from being ‘a car company’ to being ‘a provider of automotive goods and services’, with a high degree of internet enablement. To make progress on the first front, he went on a multi-billion-dollar buying binge in the world of so-called ‘downstream’ automotive businesses (downstream, that is, of designing and building new cars), such as repair shops (Kwik-Fit), collision repair (Collision Team of America), parts recycling (Copher Brothers), extended warranties (APCO), limousine services, and even drivers’ schools. To further reconfigure the company away from manufacturing and into customer-facing businesses (and to emulate GM’s flotation of its own parts division, as Delphi), he accelerated the spin-off of Ford’s components group to 2000, when it became the independent company known as Visteon.11 He even attempted to enter the world of automotive retailing directly by forward integrating into dealerships in the US (a practice common in Europe
192 The Second Automobile Revolution
but unheard-of in North America), via the so-called Ford Retail Network. He hoped to further leverage advanced retailing and marketing techniques by forming the Premier Automotive Group (PAG), which would house under one roof Ford’s premium brands of Lincoln, Aston Martin, Jaguar, Volvo, and now (as of 2000) Land Rover. The goal was to offer a range of upscale brands to consumers in a one-stop multi-brand dealer shopping environment, much as a Nordstrom’s might offer multiple haute couture clothing lines within one store. This concept ran directly against American car retailing tradition and practice, which advocated instead focusing the salesforce on one or at most two brands within one store. On the second front, he elevated the IT department of Ford to a leading role, and advocated adopting the best practices of internet-enabled firms, both on the sell-side and the buy-side of a company. On the sell-side he sought to emulate PC builder Dell Computer, which was famous for building tailor-made computers to order, whereas Ford and other OEMs had traditionally built masses of similar cars to stock. On the buy-side he ordered the development of the revolutionary online purchasing portal, Covisint, which was to use online auctions to drive down the cost of purchased car parts. But then fate intervened, starting in 1999 and snowballing through 2000 and onward, with the catastrophe of the Ford/Firestone tyre scandal. While volumes have been written covering this story and assigning blame for the problem, in short form it can be said that Ford Explorer SUVs wearing Firestone tyres seemed to be rolling over and killing their occupants more often than might be expected. As the details emerged, both firms clumsily handled the public relations nightmare that resulted, leading to higher costs and greater ill will than otherwise might have been the case. Massive and costly recalls of tyres followed, huge brand equity losses were incurred, and hundreds of lawsuits were filed. Partly as a result of this debacle, and almost before he had begun to work on his grand plans, Nasser was out, retiring in the autumn of 2001. His legacy to the company was unfortunately not positive. As SUV profits started to erode (by now customer enthusiasm on the demand side was waning, and Japanese competition on the supply side was waxing), Ford’s record profits in the late 1990s had turned to losses by the early 2000s. There was a strong push within the firm to go back to basics in order to reverse these losses, and virtually all of Nasser’s innovative initiatives were overthrown, and quickly. The downstream automotive businesses were almost completely sold off, often at a loss. Covisint’s value proposition proved to be illusory, and the unit was downgraded from a purchasing portal to a communications service, and then sold to another IT firm entirely. The Ford Retail Network proved unmanageable (companies good at making cars are not generally good at retailing them) and was dissolved. And the Premier Automotive Group was steadily losing money, as Volvo profits were generally erased by losses at Jaguar, a brand Ford had never managed to turn around. In some respects this wholesale
Ford, 1993–2007: Losing its Way?
193
de-Nasserisation was a shame, as many of his ideas were strategically sound, if poorly executed: downstream businesses are more profitable than carmaking. Observers at the time asserted that Ford was willing to spend money to buy things, but not to run them, and so the diverse initiatives and their management challenges simply overwhelmed the company. Meanwhile, while management was focused on all these other matters, Ford’s product development process languished: most effort had been spent on the SUVs, and now that their star was fading, it seemed that the pipeline of attractive car products was nearly empty. In addition, neglected operations outside the US, especially in Europe, had moved into a loss position as well. It was into this difficult situation that the Ford family, who still had voting control if not a majority of financial ownership, inserted as CEO one of their own, William Clay Ford Junior, better known as Bill or ‘Billy’ Ford, in the autumn of 2001.12
The Ford era: 2001–2006 Ford was the great-grandson of the firm’s founder, Henry, and perhaps therefore was seen by the Ford family (who were the instigators of the management change) as someone who could indeed return the company to its roots: the profitable design and manufacture of passenger cars. As he arrived on the scene the company’s automotive operations had swung to a loss (in 2001 and 2002), and although the financial side of the house was still producing well, the overall results were inadequate to cover Ford’s cost of capital. The problem was, as noted above, that while the long American love affair with SUVs was ending, Ford had precious few competitive entries in the car market to offer drivers instead. The Ford Focus, a compact car on which the company had pinned its hopes of regaining some momentum in the American market, never caught on there as well as in Europe, in part due to enormous quality problems. The car was recalled a record fourteen times in its first two sales years, beating the sorry record of thirteen in two years set by the General Motors X-cars of 1980. Meanwhile, the once-mighty Taurus had aged so poorly in retail buyers’ eyes that over half of its sales were at wholesale prices to rental-car fleets. The Premier Automotive Group was no help, either, turning in a string of losses, and overseas operations such as Ford Europe and Ford Asia Pacific were also running red ink. Bill Ford’s response was perhaps predictable for a leader under such pressures: he launched the back-to-basics campaign and executed the aforementioned sale of the Nasser-era businesses. In addition, he began the restructuring of the core North American vehicle operations: factories were closed and headcount reduced. However, the company continued to lose market share: given the years between customers’ purchases of new cars, it takes a great deal of time to alter their buying preferences; and given the years
194 The Second Automobile Revolution
it takes to develop new cars, it also takes a great deal of time to offer them something new. The tragic events of 11 September 2001 occurred almost simultaneously with Ford’s accession as CEO, and with them came a seemingly permanent cap on Ford’s profits. To help reassure a badly battered country, General Motors launched its ‘keep America rolling’ discount and rebate campaign soon after the attacks. The programme, which offered massive rebates and cheap financing, worked as intended, and unit sales volumes in the US stayed remarkably firm. The problem, of course, was that these unit volumes were rolling out through showroom doors at sharply lower prices, thus slashing profitability. Ford and Chrysler had no option but to match GM, and so industry profits significantly eroded from late 2001 onward. The so-called ‘incentive wars’ had begun, escalating over the next few years to the point where it seemed an American consumer need not even consider a Detroit car or truck unless it carried a $3,000 or $4,000 discount. Given an average transaction price of some $25,000 (at retail), these rebates were destroying all hope of sustained profitability at the domestic OEMs. Ford now found itself in a series of vicious downward circles, in terms of its core North American vehicle business. Falling market share meant it had to downsize capacity. But costs did not easily adjust downward with lay-offs, as Ford – under UAW (United Auto Workers) contracts – was still responsible for dismissed workers’ retirement pensions and health care costs. To take the cost base down Ford had to turn to other sources, notably the outside suppliers of the parts that made up 60 per cent or more of its COGS (cost of goods sold). Pressuring suppliers for cost cuts resulted, however, in both key suppliers going bankrupt and healthier suppliers favouring more generous buyers, such as the Japanese. Accordingly, Ford’s relations with its suppliers deteriorated rapidly. Meanwhile, the Ford dealer body, protected by both regulation and contract from unilateral Ford action, remained sized for a company with a much larger share of the market. Seeing their volumes fall, individual dealers tried to regain profits by cutting prices, which lowered the overall financial health of the dealer group, reducing their ability to invest in the brand. As fixed labour costs stayed high, supplier health deteriorated, retail pricing and brand image suffered, and Ford’s market share dropped further, triggering another turn of the vicious circle. Ford was profitable in its automotive operations only two years out of the five that Bill Ford had been in charge, as 2005 ended. Something had to be done, and it was. Three key steps were taken. First, Hertz was sold off, raising $15 billion in badly needed cash. Second, a revised turnaround programme, dubbed ‘The Way Forward’, was launched in late 2005, with the goal of resizing the North American operations downward, to match the company’s shrinking market share: over 30,000 jobs were to be axed. And finally, the search was begun for a new CEO. Throughout his tenure there
Ford, 1993–2007: Losing its Way?
195
had been mutterings that Mr Ford was in over his head, or that only his name had got him the job – and in fact, he had been quite candid with the press that indeed he would hire a better man for the job if the situation seemed right. And in 2006 it seemed right. Ford, the family determined, needed another change: not necessarily to alter the direction of the company, but to accelerate the pace of the changes that were already under way. The search began for a new CEO. To truly shake up the company it was decided to look outside the industry – but not so far afield that the new leader would lack credibility on the factory floor. Aerospace executives were targeted, as ‘car guys’ had always felt that building airplanes offered similar challenges to building cars. In autumn 2006, therefore, Alan Mulally, CEO of Boeing Commercial Airplanes, was enticed to Dearborn to run Ford. Bill Ford would remain as executive chairman, but the real power would go to this new CEO.
The Mulally era: 2006 onward Some have said that the formula for success that formed the rationale for Mulally’s appointment was quite simple. At Boeing he had engineered massive downsizing (in response to Boeing’s own mis-steps and the competitive advances made by rival Airbus), followed by an equally massive product offensive, culminating in the 787 Dreamliner. He had been able to ask the organisation for significant sacrifice, in return for bringing the company new prosperity with the 787. The question now was, could he apply this formula now to Ford? He started out by attacking the cost side, where action to stop the ongoing losses could be taken fastest. The Way Forward plan was accelerated. Tens of thousands of blue- and white-collar buyouts were executed, more plants were closed, with more planned to be closed, and immense pressure was placed on suppliers. But all this could do was slow the rate of decline: Ford desperately needed still more funds to invest in the new product offensive, in order to boost the revenue side of the equation. Mulally took the unprecedented step of literally mortgaging the company, pledging assets from factories, to equipment, to even the Ford trademark itself to a consortium of banks, in return for about $23 billion in cash. This action meant rolling the dice in a big way, for now if Ford faltered it could find its entire legacy being delivered into the hands of the financiers. To raise even more cash and conserve more resources Mulally also sold Aston Martin, put Jaguar and Land Rover up for sale, and announced that he would seriously consider selling Volvo as well. In this way were both Ford’s globalisation and premium-brand strategies unwound, in a desperate attempt to rescue the North American core of the company. As his attention shifted to the product side of the P&L, Mulally moved rapidly to cut costs by rationalising part numbers (it was said that Ford had four times the number of different parts and components that Toyota
196 The Second Automobile Revolution
did, even though Toyota outsold Ford worldwide), standardising products and processes, and pulling forward key launches planned for 2009 through 2011, to 2008 and 2009. But this work would take years to bear fruit, and meanwhile the Ford product pipeline had run dry. The company would have to hang on until new key vehicles appeared, such as the redesigned F-150 pickup truck, in 2008. Meanwhile, in the marketplace, Ford’s domestic sales continued to decline: by early 2008 its US market share was lodged under 15 per cent, sharply down from the 20–25 per cent range of the 1990s. At the time of writing Ford’s future remains very much in question. Many positive steps have been taken: a take-charge outsider has been put in place, with carte blanche to do what needs to be done and enough money in the bank to fund his programme. Contract talks with the UAW, aimed at reducing both labour wages and benefits (i.e. health care for active and retired workers) produced a breakthrough settlement in autumn 2007, which will yield over a billion dollars of savings annually for Ford – once all the terms become active, starting in approximately 2010. And as of this writing, India’s Tata Motors is on the verge of closing a deal to acquire both Jaguar and Land Rover, sending another $2 billion to Ford’s coffers. But shareholders remain nervous that all the payoff from all these actions will be ‘too little too late’, and so have driven the stock price down, from historic highs of about $35 as recently as 1999, to about $6 in early March 2008 (Figure 10.1). The dividend has been suspended, so that the controlling Ford family has seen first its wealth dwindle and now its income stopped.
35 30
20 15 10
1993 1994
1996
1998
2000
2002
2004
Year Figure 10.1
Ford share price history, 1993 to March 2008
Source: Yahoo! Finance.
2006
2008
Dollar/share
25
Ford, 1993–2007: Losing its Way?
197
The question is whether Ford can pull out of this slide or, without the benefit of a segment-busting product such as the SUV, find itself unable to recover. It possesses a large enough cash hoard to keep afloat for a year or more, but the fundamental health of this iconic automotive manufacturer may be damaged beyond hope of recovery.
Four CEOs and three common themes While each of the four CEOs has left or is leaving a distinctive mark on Ford, three themes cut across their tenures, from which lessons can perhaps be learned, and which might help refine an optimal model of mass-market OEM strategy. These are the themes of globalisation, financialisation and distraction. Globalisation Every major OEM including Ford has recognised the need to deal with globalisation, on several dimensions. There is the opportunity to source components globally, in order to access the lowest possible cost of parts. There is the need to gain global scale, to drive down internal engineering and manufacturing costs per unit sold. And there is the opportunity to learn from different markets, in terms of discovering new segments, technologies and features that can be leveraged in other countries around the world. Ford, arguably, saw the global opportunity long before most OEMs. After all, Ford opened a sales office in the UK as long ago as 1909, and built a factory there in 1911. The company was present ahead of other rivals in Asia and Latin America as well. But these overseas units were run essentially as disconnected outposts for many years, arguably well into the 1990s. As we have seen, Trotman’s Ford 2000 initiative was an attempt to bring together the disparate pieces and forge them into a global whole, with all the benefits that this would entail. However, globalisation has been neither well-executed at Ford nor financially rewarding – and of course these two facts are interlinked. As Table 10.1 shows, from 1990 through 2006 Ford’s overseas automotive units (Ford Europe, Premier Automotive Group, Latin America, etc.) had together cumulatively lost over $3.5 billion.13 Obviously, Ford had not figured out how to design and market its products successfully in these markets14 (and in the case of PAG, how to find synergy across its multiple brands).15 Further, the company had great difficulty in bringing products from these markets back to the US, despite numerous experiments (e.g. the Capri of the 1970s, the XR4Ti of the 1980s, and the Mondeo of the 1990s). If there is any argument in favour of Ford’s approach to globalisation to date, it might be from the perspective of human resources: Ford’s non-NAFTA operations have long been a breeding ground of talent for executives who later rose very high in the
198 The Second Automobile Revolution Table 10.1 Ford cumulative net income by operation, 1990–2006 (billions of dollars) Numbers ignore gains or losses on sales of assets; rounding prevents exact footing. Activities and regions
Total
SUV boom (a) 1990–2000
Post-boom(a) 2001–2006
Vehicles Visteon (b)
21.2 2.4
24.7 2.4
−3.5 0
Total
23.6
27.1
−3.5
Europe PAG (c) ROW
−1.9 −1.7 −0.2
−1.8 N/A −0.4
−0.1 −1.7 0.2
Total
−3.8
−2.2
−1.6
Total
19.8
24.9
−5.1
Ford Credit Associates (d)
24.1 3.0
12.5 3.0
11.6 N/A
Total
27.1
15.5
11.6
Hertz
2.1
1.5
0.6
Total
49.0
41.9
7.1
Automotive operations
North America (NA) Outside NA
Finance
Notes: (a) To illustrate the impact of the declining profitability of Ford SUVs in North America, the period is split into SUV boom and a post-boom periods. The year 2000 is used as a somewhat arbitrary inflection point because, even though SUV unit sales rose slightly after 2000, in that year discounts on SUVs began to dramatically increase, and Detroit’s competition (Asian and European SUVs) passed the 1/3 market share point for the first time. While almost all other business units continued their prior trajectories, the collapse in NA vehicle profit as a result of the boom’s end is clear. (b) It should be noted that although Visteon was formally spun off from Ford in 1999, thus removing its profits and losses from the Ford income statement, it nevertheless became quite a drag on Ford in the 2000s, as the former parent found itself paying for Visteon’s weakening situation in various ways, from the collapse of the value of its investment in the firm (a balance sheet item for Ford), to costly employment guarantees for Visteon workers (who if laid off from the supplier were usually allowed to ‘flow back’ to Ford plants), to Ford’s reacquisition of various loss-making Visteon plants that the latter could not make economically viable, but on whose output the former remained dependent. (c) Premier Automotive Group (includes variously over time Jaguar, Aston Martin, Land Rover, Volvo, &c), formally created in 2000/1. (d) Other financial operations other than The Associates included in this line include the APCO extended warranty unit and the AMI leasing company.
Ford hierarchy: Trotman and Nasser themselves of course were ‘imports’ in this way. Thus, from the harsh perspective of financial results, especially with the sale or possible sale of non-US units Aston Martin, Jaguar, Land Rover and Volvo all in mind, it cannot be said that Ford has been able to solve the global puzzle. It is especially painful to note that, as the ‘Chinese century’ begins, Ford badly lags most rivals in China.
10
10
8
8
6
6 4
4
2
2
0 01 20 02 20 03 20 04 20 05 20 06
00
20
99
20
98
19
97
19
96
19
95
19
94
19
93
19
92
19
91
19
19
19
90
0 ⫺2
199
⫺2 ⫺4
⫺4
⫺6
⫺6
⫺8
⫺8
Figure 10.2
Units (millions)
Dollars (billions)
Ford, 1993–2007: Losing its Way?
⫺10
Année FMCC income
Net income
Units sold
Auto income
Ford Motor Company units sold and financial results, 1990–2006
Note: $16 billion gain on sale of The Associates in 1998 excluded from income. Source: Company financial statements.
Financialisation While Ford initially trailed General Motors in the US in this field (establishing its Ford Credit arm only in 1959, whereas GM’s lending group dates to 1919), the company soon caught up and in some ways surpassed its larger rival in the world of financing. Ford learned that the profits from issuing loans and leases to car buyers could be greater than those from building cars, and accordingly pursued these revenues aggressively. In fact, from 1990 through 2006 Ford’s financial services profits did indeed exceed those of the car operations, by billions of dollars (Figure 10.2). This occurred in part because financing is simply more profitable than carmaking (for one thing, price levels are less transparent to the customer), and in part because Ford concluded that it could leverage its financial skills beyond cars, to home mortgages and other services, via acquisitions such as The Associates and First Nationwide. However, financialisation has somewhat fallen out of favour at Ford, as most of the non-automotive finance units have been sold off, the company realising that these assets were worth more in the hands of others than in Ford’s, and the cash released has been redeployed back into the car business. Whether management attention paid to the finance businesses has also diverted key resources from automotive challenges is difficult to say,
200 The Second Automobile Revolution
but it is clear that they were managed quite independently, with a separate and specialised management team. A consensus view would be that for car companies, car-related financial services make sense, but non-car services do not. Distraction It is clear that Ford has paid a heavy price for its other (non-financial) excursions far afield of the core business of designing and making cars and trucks. Especially in the Nasser era, when dozens of downstream businesses were purchased, but also earlier, during forays into aerospace and other arenas, Ford has spent an undue amount of time and effort on distractions from the car business. Given that profit margins in automotive manufacturing are already slim, the company could not afford the diversion of money, time and human resources to these other pursuits. It is the author’s belief, bolstered by numerous interviews with Ford executives over the years, that the price paid for these distractions – in terms of eroding focus on and competitiveness in the core car business – was very high. The role of M&A must be pointed out in particular. It is instructive to note that some of the best performing OEMs over time have either avoided acquisitions altogether (e.g. Toyota and Honda), or reversed those they did pursue (e.g. BMW and Daimler). Ford, with the frantic pace of its buying and selling (in the case of Hertz even backtracking) could not help but pay less attention to its car business, thereby allowing its rivals to steal a march. Additionally, there seem to be some corporate cultures that handle mergers well and some that do not, and those that do not would be well advised to stay clear of M&A. Numerous observers have commented that Ford has a very distinctive, strong and aggressive corporate culture that does not pay a great deal of attention to outside influences, such as the managers or policies of an acquired company. This stubborn independence of thought can be an asset on many occasions (e.g. in motivating employees to focus on a key goal), but it is probably a liability when it comes to exploiting synergies with allied or acquired firms. Even in the case of Mazda, which has turned out to be quite useful to Ford,16 one has to say that it took a very long time, arguably two decades, before the alliance really began to bear fruit. Even within the vehicle business distraction was a problem. As noted, the money machine that was the 1990s SUV made Ford incredibly cash rich, for relatively little effort. However, as trucks took the driver’s seat at the company, they distracted attention from the car lines, which were allowed to languish. While GM was reinvigorating Cadillac, and Chrysler found its highend brand partner in Mercedes, Ford’s own domestic luxury line, Lincoln, was allowed to slip, such that today it is known mostly as a builder of outdated town cars for use by taxi and limousine firms. The company also failed to reinvest enough in its mid-market car products and production technologies. Then, when the oil price turned and the Japanese arrived in force, and as the
Ford, 1993–2007: Losing its Way?
201
high tide of the SUV receded, what were left were the previously hidden rocky flats of a weakened car division. No longer able to rely on trucks to buoy it, Ford found itself fighting for survival against Asian and European competitors who had never taken their eyes off cars. The SUV boom of the 1990s thus proved to be Ford’s undoing in the early 2000s. If there is any lesson from all this, it is that the car business is so difficult, that OEMs must remain laser-focused on that business if they are to survive at all: any distraction could be fatal, and any profits made in good years must be poured back into product development to ensure good health in the lean years. In this way, it can be said that in the modern car business execution trumps strategy: it is more important in making cars to execute consistently and well, than to pick the best market to enter or the right company to buy.
Conclusion As of 2008, Ford remains a troubled firm, as noted, and the impending recession in its home American market is threatening to scuttle even the modest progress its turnaround efforts have made. However, it is of course unwise to count this company out yet. It has started to address the issues our three themes raise. In terms of globalisation, Ford Europe seems to be returning at last to profitability, and Mr Mulally is bringing some European Fords to the American market. The company still lags years behind rival GM in terms of global product development systems and Chinese market penetration, but it is moving ahead. As for financialisation, all non-car-related services have been dropped. And as for distraction, it is clear all efforts are now turned to getting the basic car business fixed. In addition, Ford has taken steps to address two large liabilities that do not fit neatly within the themes mentioned. First, on the cost side, is the issue of employee retirement benefits, and second, on the revenue side, is the challenge of clean and efficient powertrains. As regards the retirement issue, Ford, like its two Detroit rivals, had been increasingly burdened over time with contractual obligations to pay virtually 100 per cent of the health care bills of its retired blue-collar US-based workers. Such pledges were inexpensive when the Ford workforce was younger and the company growing, but as more employees have retired and as the company has shrunk, the necessary payments have grown larger, and the ability to pay them has grown smaller: by 2007 there was almost $1,000 of future health care payments owed to retirees ‘embedded’ in every Ford vehicle made in the US. With the landmark labour negotiations of autumn 2007, Ford (and GM and Chrysler) have essentially solved the problem by placing all these obligations into company-funded trusts. Such a step entails enormous initial payments, but relieves ongoing operations of this particularly onerous burden. As to the powertrain issue, Ford like other OEMs has found itself overtaken in the world of ‘clean and green’ vehicles by high oil prices, tightening
202 The Second Automobile Revolution
emissions and fuel-economy regulations, and competitive initiatives such as the launch of the Toyota Prius (which as a single model has, over the course of 2007, outsold entire multi-model divisions of Ford, such as Lincoln, Mercury and Volvo!). The company is now working hard to regain its competitive footing here, launching its own hybrid models and stepping up research into alternative power sources of many kinds. However, it must be said that Ford is still a laggard in this regard, rather than a leader. Ford Motor Company is one of the oldest and most successful firms in the industry, and its Model T still reigns as the iconic original motor car for the masses. The company has abundant cash reserves and deep ranks of management talent, as well as tangible and intangible assets all around the world. Nevertheless, the legacy is at risk, and management clearly recognises this, as it sells off unneeded assets and mortgages the rest. Whether Ford will be here in another 100 years is hard to say, but it seems clear that at least for the present moment the painful lesson of distraction has been learned, and all hands on the Ford ship are rowing hard in the same direction: to rebuilding this once-mighty global institution.
Appendix Table 10.1 Year
Production (millions of units) N. America
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Ford, 1990–2006
3.6 3.2 3.7 4.1 4.6 4.3 4.3 4.4 4.4 4.6 4.7 4.0 4.1 3.7 3.5 3.3 3.0
Rest of world
Total
2.2 2.1 2.0 2.1 2.2 2.3 2.3 2.5 2.4 2.6 2.7 3.0 2.9 3.0 3.3 3.5 3.5
5.8 5.3 5.7 6.2 6.8 6.6 6.6 6.9 6.8 7.2 7.4 7.0 7.0 6.7 6.8 6.8 6.5
Workers (worldwide)
370,000 333,000 325,000 322,000 338,000 347,000 372,000 364,000 345,000 374,000 346,000 354,000 350,000 354,000 347,000 340,000 337,000
Financial results (billions of US dollars) Sales
Net income
Assets
98 89 101 109 129 138 148 155 146 162 172 163 163 165 172 178 162
0.9 −2.2 −0.1 2.5 5.3 4.1 4.4 6.9 6.0 6.5 5.4 −5.4 0.3 0.9 3.6 2.2 −12.6
60 62 67 74 82 86 95 102 106 111 108 101 122 134 130 122 133
Note: certain amounts, e.g. 2006 income, do not match amounts used in the text or figures, as the amounts here are SEC 10-K certified quantities, which include significant adjustments for write-offs and asset sales, deliberately excluded in the textual discussion. Sources: Ford; Union Bank of Switzerland.
Ford, 1993–2007: Losing its Way?
203
Notes 1. Throughout this chapter we will follow the American definition of a light truck: any vehicle traditionally built on or derived from a frame-rail chassis, which includes pickup trucks, SUVs and minivans. Over time more and more of these vehicles have migrated to unibody car frames (e.g. Honda Ridgeline pickup truck, Toyota RAV4 SUV, many minivans), so that this definition no longer holds. But through most of the 1990s the distinction was accurate, and so is appropriate in this historical context. 2. An often-posed but as-yet unanswered question is why the spike in gasoline prices did not trigger a similar invasion of European cars in the US, as European vehicles were as fuel-efficient as the Japanese models. Hypotheses on this topic include an unwillingness of European firms to invest in massive engine upgrades required to meet the US’s more stringent air quality rules, and a lack of European focus on export markets, given a historic emphasis on defending their own national home bases. 3. SUV is sport utility vehicle, as in, for example, a Land Rover, Jeep, or Ford Explorer. 4. Throughout this chapter we will use the American term for the process of designing and engineering new cars, ‘product development’. Our European readers may substitute if they wish the phrase ‘research and development’, which is an alternative term, used on the eastern side of the Atlantic, for the same thing. 5. It should be noted that Ford at this time was also selling off manufacturing operations not directly linked to carmaking: the Rouge Steel mill, Ford Aerospace and the Ford New Holland agricultural equipment company were all gone or on the way out of Ford by the mid-1990s. 6. Other ‘segment busters’ in the American market have included the Ford Mustang of the mid-1960s (which created a new middle ground between the traditional coupé and the outright sports car) and the Chrysler minivans of the 1980s (which invented a new kind of people mover). A European example might be the Renault Megane Scénic of the 1990s, which reinvented the minivan in a size and style specifically tailored to continental tastes. 7. The ‘pony car’ got its name from the first exemplar of the type, the Ford Mustang. 8. We must remember that pickup trucks are the true mass-market vehicle in the US. The Ford F-150 pickup truck has been the best-selling vehicle in the US every year for the last two or three decades, easily selling 800,000 units in a good year and vastly outstripping car sales leaders such as Accord or Taurus, which struggle to hit 400,000 units. The SUV boom of the 1990s greatly boosted profits for Detroit, but the foundation for those profits lay in the pickups on which the SUVs were based. 9. The deal was actually closed after he retired, in early 1999. 10. Ford did not purchase the other parts of Volvo, such as the heavy-duty truck division, which continued onward as the independent firm AB Volvo. 11. The story of Visteon is not central to this chapter, but the reader may be interested in some commentary. Both GM and Ford decided in the 1980s that their level of vertical integration was too high, yielding to Wall Street pressure to reduce their capital intensity, and seeking to emulate what seemed to be a higher level of outsourcing among the successful Japanese OEMs. They determined to spin off their large internal component-manufacturing operations into independent firms (respectively, Delphi and Visteon) – Chrysler had sold off much of its own parts divisions earlier, in its periods of financial distress. While in theory this might have been a good idea, neither entity was prepared to compete in the open market
204 The Second Automobile Revolution
12.
13. 14.
15.
16.
and both were too dependent on sales to their parent firms, which would prove disastrous as the two OEMs began to suffer sales declines, and thus cut back their parts purchases. Delphi declared bankruptcy in 2005, and while Visteon is still technically solvent, this is only because Ford in 2005 both took back seventeen of its worst money-losing plants, and absorbed about 18,000 of its excess workforce into its own headcount. At the time of this writing the future of Visteon remains very much in doubt. The impact on Ford of all this has been a modest but steady financial drain (either directly through Visteon losses when still owned by Ford, or indirectly through bailout costs later), and some diversion of managerial talent towards ‘fixing’ Visteon. The worst of this seems to be over, in that Visteon is likely to sink or swim on its own going forward: there will not be a further bailout. While Bill Ford would not formally adopt the title of CEO until 30 October 2001, it should be noted that even prior to Nasser’s departure the Ford heir had begun to assume a more active role in the company, working almost as a co-CEO with him. Thus some developments that might be ascribed to Mr Ford predate his actual assumption of the CEO title. Mazda, as a separate investment, is excluded from these numbers. The case of Ford Europe is especially troubling, given that, as noted, Ford has been in Europe for a long time, and has drawn many top executives from its European operations. Various observers have put forward various explanations for the steady slide in Ford Europe’s fortunes (see especially work by Bonin, 2003), but one factor that many accept is the ‘homeland’ hypothesis. In this reasoning, Ford Europe (and GM Europe for that matter) has done poorly since the UK opened its market wide (after the demise of the domestic OEM conglomerate British Leyland, and later, Rover). As the UK became an offshore European production site for the Japanese and others, Ford found itself with no safe home base, unlike VW (Germany), PSA and Renault (France), and Fiat (Italy). With no secure location in which to make profits to fund attacks elsewhere, Ford found itself on the defensive across Europe, and with no particular national image to leverage (such as, for example, ‘German engineering’ or ‘French style’). Such an across-the-board defence is expensive, and as a result Ford Europe profits are rare. As with Visteon, the story of PAG’s disappointment is one of an idea that had some inherent strategic merit foundering in the execution stage. Strategically, the idea was to bolster Ford’s weak global presence in the luxury market via acquisition of several brands, and then to leverage those brands jointly in both the upstream and downstream direction. Upstream, they would enjoy lower costs by sharing parts, and downstream they would gain market share by offering customers a broad menu of luxury choices in a ‘one stop shopping’ multi-brand luxury dealership. But the downstream savings never materialised as the dealerships proved much too hard to merge, and the upstream savings caused a disaster when customers realised (in the case of Jaguar) that they were paying luxury prices for cars made from the Ford mass-market parts bin. Ford eventually realised the problem but ran out of time and money to solve it, such that as of this writing PAG has been dissolved, with Aston Martin sold, Jaguar and Land Rover on the auction block, and only Volvo remaining. Once again, distraction was an issue: without sufficient resources (money and managers) to devote to PAG, it became a costly sideline that sucked up attention that should have been paid to the core North American car business. Mazda in fact is the happy exception to Ford’s recently difficult history with nonNorth American operations, in that the Japanese firm, for example, provides key
Ford, 1993–2007: Losing its Way?
205
product platforms on which various Ford cars are built. As noted, it has taken a long time for Mazda to bear fruit for Ford, but it is now more crucial to the company’s operations than other, wholly-owned divisions. Some observers assert that Mazda has done well because Ford has used a light touch with it, not trying to forcibly integrate it as it has done more aggressively with Jaguar and even Volvo. Another factor in Mazda’s favour is that, when Ford realised it had neglected its own small- and mid-sized car development in recent years, Ford saw that Mazda could provide a solution, by loaning its own platforms to the Detroit firm. Ford’s relationship with Mazda may be another indication that in the global automotive industry alliances often work better than complete takeovers and acquisitions.
Bibliography Bonin, H. et al. (2003) Ford, 1902–2003: the European History. Paris: PLAGE. Available online at: http://beagle.u-bordeaux4.fr/ifrede/Ford/ Bradsher, K. (2002) High and Mighty: SUVs – The World’s Most Dangerous Vehicles and How They Got That Way. New York: Public Affairs. Freyssenet, M., Mair, A., Shimizu, K. and Volpato, G. (eds) (1998) One Best Way? Trajectories and Individual Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Lacey, R. (1986) Ford: the Men and the Machine. New York: Little, Brown & Company. Magee, D. (2004) Ford Tough: Bill Ford and the Battle to Rebuild America’s Automaker. New York: John Wiley & Sons. Maynard, M. (2003) The End of Detroit: How the Big Three Lost Their Grip on the American Car Market. New York: Currency Books (Doubleday).
Journals Automotive News and Automotive News Europe, Communications.
various issues.
Detroit:
Research institutes (various personnel and proprietary reports consulted) Center for Automotive Research. Ann Arbor, Michigan, USA CSM Worldwide. Northville, Michigan, USA International Motor Vehicle Program. Cambridge, Massachusetts, USA PriceWaterhouseCoopers Autofacts. Available online at www.autofacts.com Union Bank of Switzerland, Zurich (numerous investment reports)
Crain
11
Can Chrysler Survive its Reinvention? Bruce Belzowski
Introduction Reinvention has been the key word in Chrysler’s history since the 1970s. The company has survived more near-death experiences than any other automotive manufacturer in the US. Over the past twelve years since the publication of One Best Way?, Chrysler has merged with Daimler, separated from Daimler, and begun life anew under the leadership of a private equity firm. It has been a stormy as well as a rewarding time for the company. In 1995, Chrysler was considered one of the more successful auto manufacturers after surviving the US recession in the early 1990s. Its noted strengths included its new products, supplier relations, product development organisation, and its management team. Its new products included mid-sized cars (e.g. Dodge Intrepid, Chrysler LHS) with a cab-forward design, its redesigned Ram pickup trucks, segment leading minivans, and sport utility vehicles (SUVs) (e.g. Jeep Grand Cherokee). Its supplier relations were modelled after the Japanese development model that focuses on supplier development rather than continually switching suppliers while scanning the market for the lowest price. Chrysler’s product development organisation was made up of platform teams that comprised representatives from design, engineering, marketing and manufacturing who brought a vehicle through the process from its inception rather than independently and sequentially. This product development organisation combined with a flat organisational design with fewer middle managers that provided easier access to decision-makers and quicker decision-making by executives. Finally, the decision-makers themselves were also part of Chrysler’s success. Its team of top managers included executives with broad international and cross-company experience. They showed a keen understanding of what makes an auto manufacturer successful borne from the company’s near-death experiences in the 1980s and 1990s. They understood the need to bring fresh innovative products to market quickly with the help of the 206
Can Chrysler Survive its Reinvention?
207
supply base rather than trying to engineer every component or system themselves. Despite its strengths, there were notable challenges facing the company, particularly the need to improve product quality and to develop a global strategy. Though Chrysler’s vehicle designs won awards and were sales successes, its initial quality, durability and reliability ratings were not at the levels of industry leaders such as Toyota or Honda. The company understood this weakness and was in the process of developing and implementing a Chrysler Manufacturing System that would standardise processes and provide a platform for continuous improvement in manufacturing. But becoming a global manufacturer was another matter altogether. To become a global manufacturer demanded large investments in plants and distribution systems in developed and developing economies. During the 1990s, Chrysler profited primarily from the popularity of its light truck vehicles (pickups, minivans and SUVs), as US consumers increasingly purchased more of these types of vehicles than passenger cars. But these vehicles were more unique to the US market than to any other major market in the world because of their poor fuel economy. If Chrysler were to become a global competitor, it would have to develop a fleet of vehicles that could be sold in countries where fuel prices were two or three times those of the US. In fact, of the ‘Big Three’ automakers (General Motors, Ford and Chrysler), Chrysler was the most exposed if fuel prices in the US suddenly increased as they had in the 1970s with nearly 70 per cent of its sales coming from the light truck segment. The cost of foreign plants and designing a new fleet of vehicles was more than Chrysler could afford, so its executives, particularly CEO Robert Eaton, began to explore options such as mergers and acquisitions that would provide the company with the resources for global expansion.1 Over the past twenty years, Chrysler was considered a takeover target as it fluctuated between a successful and a failing company. No one wanted the company when it was doing poorly, and it was not interested in merging or being acquired when it was doing well. The need to become a global company was one of the driving forces behind Chrysler’s thinking in the 1990s. The lack of potential growth in the US market, which was only growing at 1–2 per cent per year, meant that for the company to grow it would have to battle the strongest manufacturers in the world in the US or develop globally and enter the growing emerging markets. Chrysler’s executives understood that as soon as the company became successful, they would seriously consider merging with a global company, in order to meet their global ambitions.
The merger of Chrysler and Daimler The official merger of Chrysler and Daimler into DaimlerChrysler (DCX) in 1998 was considered a ‘marriage made in heaven’ (BBC News.com, 1998).
208 The Second Automobile Revolution
At the time, there seemed to be significant advantages and few disadvantages to the merger. First, the companies did not have overlapping products. Chrysler was considered a mass market brand with some near-luxury vehicles, and Daimler, with its Mercedes brand, was considered one of the ultimate luxury brands in the world. Because there was no product overlap, there was no need to sort out which products would continue and which would be cut. Also, because of the mass market and luxury vehicle differences between the brands, the dealer body for each company could remain intact because neither company considered its buyers cross-shopping between the two brands. Second, both companies wanted to become global players. Daimler already had some global reach with its Mercedes brand, but with a strong mass market brand such as Chrysler, they expected to expand their global presence. Chrysler understood the need to become a global brand due to the low growth potential in the US, its primary market. The main question became ‘How would the merged company reach its globalisation goals?’ This question was answered soon after the merger as DaimlerChrysler management expanded its network of manufacturers by purchasing shares of Hyundai Motors and Mitsubishi Motor Company. Third, both companies had money. In many mergers, one company is not as financially sound as the other, and therefore, the stronger company takes the lead in managing the resulting merged company. In the merger of Daimler and Chrysler, both companies were considered very successful at the time of the merger. Daimler was recovering from diversification growing pains from its investments in aerospace, heavy truck, and consumer and industrial electric companies, but its automotive division was still considered one of its strengths. Chrysler was reporting record profits from its auto sales due to the strength of its light truck vehicles, and also because it was lean in terms of diversification. During the late 1980s and early 1990s, it had divested itself of its in-house component company, Acustar, and it had also exited other non-automotive businesses. Finally, both had strong products and good managers. Daimler’s passenger car division, Mercedes, was not only strong in the high end of the luxury market, but it was also expanding into the lower end of the luxury market with the A- and C-class. Though Chrysler was making most of its money on its light truck lineup of pickup trucks, minivans and sport utility vehicles, its passenger cars were also holding their own, at least in the views of the media who found their overall designs very appealing. All of these products by both companies were developed under the auspices of teams of managers with deep automotive experience and a recent track record for running successful companies. What could go wrong? Integration vs. autonomy Though the phrase ‘mergers and acquisitions’ is commonly used to describe combining two or more companies, each part of the phrase offers a very
Can Chrysler Survive its Reinvention?
209
different perspective for the companies involved. When a company is acquired by another company, it is usually considered the weaker of the two, and therefore it is expected to follow the lead of the acquiring company in terms of management decisions. But when two companies merge, it is not always clear which company will take the lead, especially if both companies are financially sound. Though the Daimler architect of the merger, Juergen Schrempp, called it a ‘merger of equals’ (Kisiel, 1998b) it was an uneasy equality. At the beginning of the merger, both management groups were equally represented on the DaimlerChrysler Management Board, and there was a clear attempt to let the two groups continue to run their businesses as they had in the past, trying to find points of synergy that would make sense for both groups. The new company was trying to decide at which points a group should be autonomous, and when it should be integrated with the other group. The rise of information technology and e-business initiatives allowed for frequent transcontinental video conferencing between executives, as well as engineers in product development. Both groups thought there were integration and collaboration opportunities in the area of engineering DaimlerChrysler could leverage. The potential for global sourcing using new e-business tools was also considered a point of integration. Integration could also occur within the financial areas such as merging the financial services of both groups, as well as merging processes for financial reporting. Engineering integration But it became clear early in the relationship that there were significant differences in the engineering, purchasing and overall management areas that would create integration and collaboration difficulties. In the engineering area, the integration of the cultures of the two engineering organisations was uneven. Chrysler’s platform teams, which included representatives from product development, purchasing, manufacturing, and sales and marketing from the beginning of development until the sale of the vehicle, were noted for their flexibility in looking for the best solution given the time and cost constraints that were imposed on the programme. Each group in a platform team acted as both a check in cost containment and a partner in finding solutions to the complex challenges each vehicle programme faces. In trying to incorporate the Daimler engineers into their vehicle programmes, Chrysler ran into two major issues: how to integrate Daimler engineering expertise across all of the platform teams simultaneously, and how to indoctrinate the Daimler engineers to develop products that created the perfect balance between cost and function. Sharing learnings across the platform teams was always considered one of the weaknesses of the platform team concept. It was not unusual for members of a platform team to stay on the team exclusively for the duration of the development and rollout of a vehicle, which could range from two
210 The Second Automobile Revolution
to four years. As each team solved problems of engineering, manufacturing and sales and marketing for each vehicle, there was no formal way of transferring that knowledge to the other teams. Engineering vice-president François Castaing created ‘tech clubs’ which were informal meetings of engineers from the different teams to talk about challenges they faced and how they overcame them. He also instituted a ‘Book of Knowledge’, where members of teams could enter their lessons learned into an electronic database that members of other teams could refer to during the development of their vehicles. These methods for knowledge sharing had some success, but they were not completely formalised and required of each team member. The focus at Chrysler was always on developing new, exciting products quickly and within cost targets. Incorporating the expertise of Daimler engineers into this process was problematic. It almost demanded that the Daimler engineers become part of the platform team in order to understand all the different issues the team faced for that particular vehicle, or they would have to be brought in on an as-needed basis to provide their expertise on a particular problem or issue. The first method provided the best way of integrating Daimler engineers into Chrysler, but was not practical if the engineer was already working on a programme at Daimler. The engineer would not be able to dedicate the proper amount of time to both projects. The latter method did not lead to integrating the two staffs because the Daimler engineers would only be available for short periods of time. So, the challenges Chrysler faced with integrating its platform teams in the US became exacerbated when Chrysler considered incorporating its new colleagues from Daimler. Indoctrinating Daimler engineers, who focused on high quality even if it meant high cost, into the Chrysler culture of developing vehicles based on strict cost targets while at the same time providing very functional and exciting designs became one of the biggest challenges for the integration of engineering between the two companies. From an engineering perspective, the attempts by Daimler engineers to raise the level of technology of Chrysler vehicles sometimes helped differentiate Chrysler vehicles in the marketplace (such as active safety systems), and at other times conflicted with the goals of the programme. This conflict is one of the key themes that describe why the merger for the two companies failed: Daimler was not able to incorporate the thinking of a volume-oriented, mass market brand such as Chrysler into its corporate thinking that is based on developing expensive, high end vehicles. Purchasing integration In the area of purchasing, both companies expected there would be collaborative opportunities. The rise of e-business processes offered companies the opportunity to tap into the global supplier market through the use of online auctions. Though DaimlerChrysler began the process of setting up
Can Chrysler Survive its Reinvention?
211
these e-business processes, they soon found that many of the components the two companies planned to purchase were unique to each company, primarily because of the engineering requirements which were higher and consequently more expensive for Daimler components. There were some non-core components such as switches or brackets that could be sourced this way, but the majority of components were so highly engineered that only a few suppliers in the world were capable of bidding on them. E-business processes themselves conflicted with Chrysler’s purchasing strategy that many considered one of its competitive advantages. It was based on the Japanese manufacturer strategy of building a cadre of suppliers who are considered an integral part of its supply chain. This means that the supplier adheres to the manufacturer’s continuous improvement culture that includes the development of new technology as well as cost reductions and increasing levels of quality. In this relationship there is an element of trust between manufacturers and suppliers that means that both groups can rely on the other to fulfil the requirements of the relationship. Because of this philosophy, Chrysler was regarded by suppliers as the preferred US manufacturer to work with. Chrysler was able to leverage this preferred relationship in particular in the area of new products or technologies. It did not share its suppliers’ new products with other suppliers, trying to get a lower price, and it shared cost reductions provided by suppliers with the suppliers. E-business processes, especially reverse auctions, worked against Chrysler’s supplier model. They continually pushed suppliers to reduce the price of their components, so the supplier with the lowest bid would receive the business.2 Also, auctions did not work with complex systems that many of the global suppliers provided. It was very difficult for the manufacturer and supplier to come to an agreement about the final price of a complex system such as complete interiors, engine systems, or brake and steering systems. Though DaimlerChrysler saw these e-business processes as an opportunity to integrate Mercedes and Chrysler, the fundamental basis of e-business processes worked against an involved, secure supply base that fully collaborated with its manufacturer customers. Management integration Finally, at the management level, integration of the two companies did not yield the desired results, as key executives at Chrysler left the company before they could participate in the merger of the two companies. Twelve top executives left the company with large bonuses, stock and options. Those who left soon after the merger were able to take advantage of the record high stock price of $108 per share in 1998. But those who left after a year saw the stock price plummet to $38 per share by the end of 1999 (Kisiel, 1998b). But besides the large payouts to top Chrysler executives, it was the loss of autonomy within Chrysler that drove them out of the organisation. Tom Stallkamp, Chrysler’s purchasing head who was promoted to president
212 The Second Automobile Revolution
in 1999, best described the clash of company cultures in management meetings in Germany. He described the Daimler executives coming to the meetings with stacks of data and presentations to support their positions under discussion, and the Chrysler executives came with one presentation that laid out their position based on their group knowledge of the industry. The Chrysler executives were used to reaching consensus on key issues amongst themselves relatively quickly and then moving to implement the decisions. Because their organisation was relatively flat, it was easier to communicate decisions to directors and managers in the company. The Chrysler executives found the DCX management meetings interminable as the executives spent much more time in making decisions than was typical within Chrysler. In the end, it became clear to the Chrysler executives that they would not be able to run the company in the same way they had in the past, and that the path of least resistance was to leave the organisation. Stallkamp himself found an even quicker way to exit the company. He made several critical remarks about the process used at management meetings in the summer of 1999, and he was fired within a week (Chappell, 2003). When Robert Eaton retired in 2000 (with a retirement package of about $69 million), James Holden, formerly a marketing vice president, became CEO of the Chrysler group. It was clear by then that Chrysler would be able to maintain a certain amount of autonomy if it was profitable, and Holden tried to continue some of the successful strategies of previous Chrysler executives. In order to be successful, Chrysler needed to control its costs and to continue to be successful in its core vehicles such as the Grand Cherokee SUV and the minivan. In the spring of 2000, Holden was responsible for allowing Chrysler to produce and sell large numbers its 2000 model year minivans at reduced prices in anticipation of the redesigned minivans that would go on sale in September. But when people did not purchase the new minivans in the autumn because they were not sufficiently different from the discounted vans, Chrysler had a problem. Chrysler had to include incentives in order to sell the new minivans, and profits plunged. Holden was replaced by Dieter Zetsche, who had successfully restructured Mercedes’ truck division in Germany. The merger of Chrysler into Daimler was complete. But instead of taking the best of both companies and integrating them into one successful company, Daimler was forced to take over the American company that was in the throes of failure. Chrysler had gone through a number of reinventions since the mid-1950s; its new reinvention would be under the management of its German partner. The Daimler influence The two major influences on Chrysler during the early 2000s were the arrival of Dieter Zetsche and Wolfgang Bernhard to turn around Chrysler, and
Can Chrysler Survive its Reinvention?
213
Daimler’s attempt to develop a global presence through its relationship with Hyundai Motors and Mitsubishi Motors. By sending Dieter Zetsche (CEO) and Wolfgang Bernhard (COO) to Chrysler to restructure the company, Daimler was sending two of its top executives. Both men were considered very good managers, and they very quickly appraised the situation and developed a turnaround plan that they thought would save the company. At the core of the plan was the understanding that because Chrysler could not maintain the market share it maintained in the 1990s, it needed fewer workers. It also could not maintain the same relationship with its supply base. It needed to have continuous cost reductions internally and also from its supply chain: the main goals were to reduce employment by 26,000 and cut $4 billion from internal and external costs over three years (Bradsher, 2001). These were demanding goals, but the new management made it clear that they were necessary for survival. Internally, negotiations with the United Auto Workers began with the union agreeing to allow DCX to announce the closure of plants without the normal due process for closure, and externally, Chrysler began negotiating with its supply base to meet its goals. These discussions with suppliers completely changed the relationships that had developed between Chrysler and its suppliers from a collaborative relationship to a transactional relationship, based solely on the supplier’s ability to meet aggressive cost targets. The Daimler influence also became stronger in the engineering function as Zetsche remodelled the product development process, based on Mercedes’ process. He left the platform teams intact, but created Centres of Competence in order to formalise the sharing of learning coming out of each platform team. Finally, Zetsche aggressively integrated support of Daimler engineers into the introduction of new technology into Chrysler. Zetsche’s goal was not only to control costs, but to increase the level of technology and product sophistication in Chrysler vehicles. He hoped that increased technology and product sophistication would do for Chrysler what it did for Mercedes (Kisiel, 2001). While Dieter Zetsche was charged with turning Chrysler around, Juergen Schrempp became the architect of DaimlerChrysler’s global strategy. After the merger with Chrysler which he engineered, Schrempp’s main attempt to globalise the company was the decision to take a financial stake in both Hyundai Motors and Mitsubishi Motor Company in 2000. The Hyundai stake was based on a commercial vehicle venture between the two companies, while the Mitsubishi stake was to provide DCX access to small and mid-sized cars. These companies would also be potential partners in the growing Asian market, including protected markets like Korea and Japan. From a Chrysler perspective, these companies could create small and mid-sized vehicles for DCX’s US and European markets, where they needed support. One of the first combined efforts was the development of a new four cylinder world engine that would be used by all three companies worldwide and be built in
214 The Second Automobile Revolution
the US. This is the only surviving piece of the relationship between the three companies, and it provided all the four cylinder engines for Chrysler in 2008. Unfortunately, the Mitsubishi joint venture quickly turned sour as the company was accused of covering up defects in its vehicles from 1993 to 2000. Large recalls in Japan and the US damaged Mitsubishi’s reputation, and market shares in both countries plummeted. In the US it tried to recover by offering vehicle loans to large numbers of buyers who were high credit risks, but when these buyers defaulted on these loans, the company was again faced with another financial crisis. During the five years of its joint venture with Mitsubishi, DaimlerChrysler helped support the company financially, but eventually the DCX board rejected another injection of funds to support the company in 2004. Also in 2004, DaimlerChrysler divested itself of its stake in Hyundai Corporation, as board members decided to pull back from DaimlerChrysler’s globalisation efforts. These failed attempts at globalising DamilerChrysler also spelled the end of the tenure of Juergen Schrempp as CEO in 2005. The failure of Juergen Schrempp’s globalisation of DaimlerChrysler became a turning point for the Chrysler group because it left it with no small car development. The Chrysler group developed its future vehicle plans with the idea that Mitsubishi and Hyundai would take the lead in developing small and mid-sized cars, but the departure of Mitsubishi from the DaimlerChrysler global plan was a critical blow to its vehicle plans. Mercedes was unaffected by these changes in the organisation, but Chrysler was in an era when US purchasing patterns were changing as buyers began to purchase more passenger cars than light trucks and fuel economy became an important differentiator in vehicle purchases. Without more fuel-efficient passenger cars, Chrysler was at a serious disadvantage in the US marketplace. The continued development of the global four cylinder engine with Hyundai and Mitsubishi was one of the few positive effects for Chrysler that came from the DCX globalisation plan. The challenging US market When Dieter Zetsche initiated his turnaround plan of the Chrysler group in 2001, the US market was in the midst of a long record-setting sales cycle. Annual sales were breaking records, but the main US brands of General Motors, Ford and the Chrysler group were losing market share to the Japanese, Korean and European brands. Zetsche’s plans for ‘right-sizing’ the Chrysler group to fit its current market share which included reducing salaried and production workers and forced price reductions of its suppliers set the tone for his administration. But Zetsche combined these tough declarations with an honest demeanour professing he wanted to help Chrysler recover from its current situation. He was able to instill in the company a feeling that though times were tough, he would lead them to better times. He would not publicly denigrate any Chrysler employees or suppliers, but he would staunchly
Can Chrysler Survive its Reinvention?
215
defend his goals. He was just beginning to make some headway with his turnaround plan when 11 September 2001 arrived. The terrorist attacks on 11 September not only changed the political direction of the US, they also drastically changed vehicle sales. During the first two weeks after 11 September, vehicle sales collapsed. At no time in recent history had sales dropped and stayed as low. General Motors took the lead in reviving the industry with its ‘Keep America Rolling’ programme that offered low credit terms and cash incentives to stimulate sales. Chrysler was reluctant to follow GM’s strategy, but eventually it also had to follow suit and offer cash incentives and low credit terms as well. These sales incentives created the desired effect by increasing sales, but when GM refused to pull back on incentives, Ford and Chrysler had to continue their incentive programmes. This philosophy went against Zetsche’s goal of trying to raise the value of his vehicles in the eyes of buyers. The general feeling in the industry is that incentives degrade the brand, so GM’s strategy was a risky one. It may be that GM thought Ford or Chrysler would not be able to survive the cost of long-term incentives, or it may be that they wanted to flood the market with their vehicles with the intention of showing their improved design and quality. As GM continued the incentive war, it forced Ford and the Chrysler group to spend heavily in order to compete, especially in the light truck segment. The large incentives and the generally good US economy drove increases in vehicle sales after 2001, but Chrysler was not able to return to its 1999 market share peak of 17 per cent (Appendix Table 11.1). Though Chrysler’s market share was stable after 2001, its profitability per vehicle was a roller coaster ride of highs and lows. Some of the highs were because of their new products such as the PT Cruiser with its unique retro styling and the Chrysler 300C with its Bentley-like exterior and its powerful Hemi engine. Chrysler also eliminated the Plymouth brand during 2001, which reduced overall sales but allowed Chrysler to focus its development on creating unique brands for Chrysler, Dodge and Jeep. Eliminating long-time brands is not an easy process either internally within the company because of the emotional connections with the brand or externally with its dealer body which may have based its business decisions on supporting the brand. The incentive war among GM, Ford and the Chrysler group put pressure on all three companies’ profits, but during the early 2000s the legacy costs of these companies also began to take a toll. All three companies had ageing workforces with high wages, high pension support, and in recent years, double digit annual increases in health care costs. Because of the hypercompetitive US market, none of the companies was able to raise prices enough to cover these incentive and legacy costs, leading to low profitability and losses among the Big Three. Despite its profitability challenges, Chrysler was able to make some headway in its manufacturing operations. From 2001 to 2006, Chrysler reduced
216 The Second Automobile Revolution
the number of hours it took to build a vehicle by 25 per cent from 44.3 hours to 32.9 hours (Appendix Table 11.2). This reduction certainly helped reduce its internal manufacturing costs, but Chrysler continued to be confronted by its low initial quality scores which measure the problems buyers report within their first three months of ownership. Despite its best efforts, the Chrysler, Dodge and Jeep brands were all slightly below the average for the industry in terms of initial quality (Appendix Table 11.2). In a competitive market such as the US market, a company’s negative quality reputation creates a long-term image that companies can spend years trying to overcome. In the case of Chrysler, being average for many years made it difficult to compete against manufacturers who were above average in initial quality. Shareholder pressure By 2007, the pressure by DaimlerChrysler shareholders to sell the Chrysler group gained momentum. In 2006, Chrysler sales were affected by its lack of fuel-efficient vehicles as fuel prices rose and buyers began to place higher value on fuel economy in their purchase decision. Chrysler’s operating and pre-tax profits were again in the red after two consecutive years of profits, while the Mercedes brand rebounded from its first loss since the merger (Appendix Table 11.1). Gone were the days in the late 1990s when Chrysler’s operating profit doubled that of Mercedes. Gone also was Dieter Zetsche who was brought back to Germany to run DaimlerChrysler after Juergen Schrempp’s ouster as CEO in 2005. In 2006, Chrysler’s new leader, Tom LaSorda, fell prey to the same ‘sales bank’ strategy that damaged the company in the 1960s and 1970s when he allowed manufactured vehicles to be counted as sold, when, in fact, they were stored in parking lots throughout the Detroit area, waiting to be purchased by dealers (instead of being delivered to dealers directly from the factory). Because of Chrysler’s less than stellar financial performance since the turn of the century, some investors considered Chrysler a burden to DCX, calling for it to be sold. Chrysler was supported first by Juergen Schrempp who considered Chrysler as part of his plan for developing DaimlerChrysler into a global brand, and in 2006 by Dieter Zetsche who, as CEO, protected Chrysler after it had made small operating profits in 2004 and 2005. But after the loss in 2006, a second Chrysler recovery plan was developed that called for further job reductions, estimated at 13,000, and in February 2007 Zetsche announced that ‘all options are open’ concerning Chrysler’s future (Muller, 2007). It came as no surprise in May of 2007, that 80 per cent of Chrysler was sold to Cerberus Capital Management while Daimler maintained 20 per cent ownership. The final cost of the sale, US$7.45 billion, to Cerberus represented only a fraction of the US$36 billion Daimler paid for Chrysler in 1998.3 In his statement, Zetsche noted that executives had ‘overestimated the potential synergies’ of the two companies and that ‘the American volume customer
Can Chrysler Survive its Reinvention?
217
is not willing and is probably not able to pay premium prices for premium technologies’ (Muller, 2007).
Chrysler’s future under Cerberus Capital Management Chrysler made modern history with its merger with Daimler in 1998, and it made history again when it was purchased by Cerberus Capital Management in 2007, putting the company into the hands of private investors. Under this arrangement (and with a continued connection to Daimler technology) Chrysler could have several advantages. In terms of management thinking, private equity may be a better fit for an industry that has long development times and needs longer-term thinking instead of the being tied to quarterly financial reporting. There are many examples of managers in the industry making short-term decisions that make quarterly reports look good, but have a detrimental effect on the long-term prospects for the company. From a financial perspective, Chrysler’s private equity model may become a new way for companies with long development and return on investment timelines to manage their companies and free their managers from the pressure of producing short-term results. The continued financial link to Daimler AG will offer Chrysler access to new technology, especially alternative powertrain technology such as fuel cell research where it has little expertise. Daimler could offer advanced diesel technology that may also play an important role as US manufacturers strive to meet new corporate average fuel economy goals for 2015 and 2020. From a personnel perspective, Chrysler took the path of bringing in new executives that provided a fresh perspective to Chrysler. One report noted that Cerberus set aside 5 per cent of Chrysler’s equity to pay its top 75 managers (Automotive News, 2007). Bob Nardelli, the CEO, came with executive experience at the manufacturing level at General Electric and retail experience in mass market building supplies. After joining Chrysler in 2007 and reviewing the company he reported that it would take three years to turn around its finances and ten years to retool its product plan (Wernle and Leslie, 2007). Tom LaSorda remained from the DaimlerChrysler days, but he was joined as co-president by Jim Press who came to Chrysler after thirty-seven years at Toyota Motor Corporation, where he was Toyota’s top executive in North America and the only non-Japanese on Toyota’s Board of Directors. His knowledge of Toyota may prove invaluable in managing Chrysler’s new products and its globalisation plans. LaSorda’s expertise lay in manufacturing while Press’ strength lay in product and sales. One of Press’ first strategies was to convince the Chrysler dealer body to consolidate. His goal was to have one dealership sell all three brands rather than have separate dealerships selling the Dodge or Chrysler/Jeep brands.
218 The Second Automobile Revolution
Chrysler also lured other executives from Nissan, Toyota and Ford, but the hiring of Phil Murtaugh as CEO of Asian Operations from Shanghai Automotive Industry Corporation (SAIC) gave Chrysler much-needed expertise in building the brand in China, the world’s fastest growing automotive market. Before SAIC, Murtaugh designed General Motors’ China strategy, and was very well respected for his knowledge of doing business in China. He quickly helped Chrysler fill a hole in its product lineup by negotiating with Chery Automobile Company in China to rebadge one of its small cars for the US market. Heralding what can only be called a new way of doing business at Chrysler, when executives realised that the vehicles from Chery were not yet ready for the US market, they moved quickly to create a reciprocal manufacturing agreement with Nissan Motor Corporation in North America. Nissan would build small cars for Chrysler in Mexico for sale in the US and Chrysler would build Nissan full-sized trucks in its plants. The willingness of management to reconfigure its manufacturing footprint may be from a sense of urgency to fill product gaps, but it also may represent a new willingness to create alliances that fulfil mutual needs.
Conclusion After the ‘divorce’ from Daimler AG and the subsequent purchase by Cerberus Capital Management, Chrysler found itself in a similar but more precarious position to the one it was in before it merged with Daimler. In 1998, Chrysler had a strong management team that had worked together for over ten years, and had set up design, product development, purchasing and marketing systems that were envied by others in the industry. By 2008, the new Chrysler had assembled an impressive new management team with deep knowledge of the industry, but little experience working together and unproven in terms of actual results. In 1998, Chrysler’s products were selling well and profitability was high, especially for its light trucks (sport utility vehicles, pickup trucks and minivans). In 2008, Chrysler faced a problem of having too many light trucks and not enough passenger cars as buyers began to purchase vehicles with better fuel economy, and the government demanded a 40 per cent increase in fleet fuel economy by 2020. Chrysler faced an uncertain future if it could not develop new powertrain technology to meet consumer demand for greater fuel efficiency. This weakness would also affect Chrysler’s ability to globalise its vehicles, since most of the world has fuel prices two or three times higher than US prices. In 1998, Chrysler was an independent company looking to grow globally with a possible merger as a possibility. In 2008, Chrysler seems to be leaning more towards manufacturing agreements with other manufacturers to
Can Chrysler Survive its Reinvention?
219
provide immediate support for its product strategy, as well as provide other companies the use of its manufacturing excess capacity. What is unknown is whether these are temporary moves to fulfil immediate needs or examples of how flexible the company will be in trying to become a global manufacturer. If it is an example of flexible globalisation, then it may become a model for how smaller manufacturers can be global manufacturers. Unfortunately, the risks of flexible globalisation may be greater than the rewards. How well can Chrysler’s engineers develop products with companies from different parts of the world? Does Chrysler have enough depth of engineering and manufacturing talent to supervise the development and manufacture of products in a variety of global locations? Can Chrysler use this flexibility to develop the new powertrains that will be needed over the next ten to twenty years? What if one of the companies Chrysler relies on produces vehicles with extensive warranty claims? What if a company that provides a key part of Chrysler’s product portfolio is purchased by another manufacturer or goes out of business? The list of challenges of relying on companies that are not owned by Chrysler can be daunting, and gives one pause when talking about a new form of global development and manufacturing. In the end, one must also consider the company that owns Chrysler, Cerberus Management Capital. In 1998, Chrysler merged with another auto company. In 2008, Chrysler was purchased by a non-auto company. Private equity investors historically expect above-average returns on their investments, and the automotive industry has not been the type of industry that has provided such returns in many years. Though Cerberus is a large company with revenues in 2006 of its combined companies of approximately US$60 billion, Chrysler’s revenues alone for 2006 were US$62 billion. Chrysler doubles the size of Cerberus, and Cerberus’ other large major automotive ownership stake is in General Motors Acceptance Corporation (GMAC), General Motors’ financing arm. In the demanding world of private equity, if Chrysler is not able to remain profitable, will Cerberus sell off all or parts of the company? The key uncertainty is: How long will Cerberus wait? The chairman of Cerberus, John Snow, talked about the patience his company had for the automotive industry at the announcement of the purchase of Chrysler, ‘Our approach is fundamentally long term. We don’t think about the next quarter. We don’t think about what analysts have to say about us. Our capital is patience’ (Wilson, 2007). Based on the challenges facing Chrysler and the global automotive industry, that patience will be tested as the New Chrysler embarks on its latest reinvention in its long automotive history.
1,619,207 1,866,976 1,908,067 2,042,467 2,226,251 2,133,159 1,960,119 2,042,153 1,978,566 2,055,679 2,071,942 1,838,845 1,700,448
Trucks 14.7 16.2 15.2 15.9 16.7 14.5 13.3 13.1 12.8 13.1 13.6 13.0 12.9
$53,195 $61,397 $61,147 a130,122 a148,243 a160,278 a150,422 a147,408 a136,437 a142,059 a149,776 a151,589 N/A
US Revenue market millions share (%) $2,025 $3,529 $2,805 a4,820 a5,746 a7,894 (a593) a5,098 a488 a2,466 a2,846 a3,227 N/A
Net income millions N/A N/A a3,368 a4,212 a5,051 a501 (a5,281) a609 (a506) a1,427 a1,534 (a1,118) N/A a1,090 a1,868 a1,336 a1,470 a1,497 a170 (a1,679) a226 (a496) a1,014 a487 (a1,072) N/A
0.04 0.06 0.05 0.05 0.04 0.04 0.00 0.03 0.00 0.01 0.01 0.02 N/A
112,500 114,200 112,300 126,816 129,395 121,027 104,057 95,835 93,062 84,375 83,130 80,735 N/A
Operating Worldwide ROA Labour profit pre-tax (%) thousands profit Worldper vehicle wide ROW
91,000 21,500 93,700 20,500 94,300 18,000 122,602 4,214 125,549 3,846 118,024 3,003 101,027 3,030 94,903 932 92,034 1,028 83,542 833 82,321 809 79,568 1,167 N/A N/A
NAFTA
Notes: 1. Labour is average annual employees worldwide. 2. NAFTA labour only includes US for 1995–7. 3. ROA is calculated Net Income/Average Total Assets. 4. ROA is computed using DaimlerChrysler’s net income and total assets for 1998–2006 since Chrysler’s information could not be extracted from the annual reports. 5. ROW is Rest of World (outside of NAFTA). 6. N/A = Not Available because Chrysler was made private in August 2007. Sources: 1. World Motor Vehicle Data: Worldwide Car/CV (1995–6). 2. Chrysler Annual Reports (1995–7): Labour, Revenue, Net income, ROA variables, Vehicle sales. 3. DaimlerChrysler Annual Reports (1998–2006): Labour, Revenue, Net income, ROA variables, Vehicle sales, Worldwide production.
842,400 893,302 777,986 867,210 863,372 762,928 637,106 621,787 503,376 532,192 608,114 620,195 788,864
2,616,972 2,973,000 2,899,037 2,982,644 3,178,566 2,963,822 2,679,411 2,749,903 2,552,308 2,652,186 2,760,467 2,548,731 2,578,917
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
2,461,607 2,760,278 2,686,053 2,909,677 3,089,623 2,896,087 2,597,225 2,663,940 2,481,942 2,587,871 2,680,056 2,459,040 2,489,312
Worldwide NAFTA production production Total Cars
Chrysler, 1995–2007
Year
Appendix Table 11.1
220
Can Chrysler Survive its Reinvention? Appendix Table 11.2
221
Initial Quality Index of Chrysler marks, 1996–2007
Year
Industry average
Chrysler
Dodge
Jeep
Labour productivity (hours per vehicle)
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
110 86 176 167 154 147 133 133 119 118 124 125
124 105 188 148 151 137 133 136 120 121 120 151
125 109 190 175 174 170 145 137 121 130 132 156
134 92 207 234 188 155 151 146 136 120 153 161
40.5 38.7 44.3 43.6 44.8 44.3 40.6 37.4 35.9 33.7 32.9 N/A
Sources: Initial Quality Index from J. D. Power & Associates. Labour productivity from Harbour Reports.
Notes 1. Robert Eaton’s previous experience as a General Motors executive led him to believe that Chrysler needed to be bigger and global. 2. Some manufacturers even had their preferred suppliers enter the auctions only to get them to reduce their prices. They had no intention of shifting the business to another supplier. 3. Even the $7.45 billion Cerberus paid was invested back into Chrysler.
Bibliography Bradsher, K. (2001) ‘Chrysler Said to Map Plan for Recovery’, New York Times.com. 16 January http://query.nytimes.com/gst/fullpage.html?res= 9407E0D8133DF935A25752C0A9679C8B63 (accessed 14 May 2008). Chappell, L. (2003) ‘Stallkamp Was Fired After He Blasted DCX Bureaucracy’, Automotive News.com. 15 December. http://www.autonews.com/article/20031215/SUB/ 312150797 (accessed 14 May 2008). Kisiel, R. (1998) ‘Prime Movers Will Kick Off World Congress in January’, Automotive News.com. 14 September. http://www.autonews.com/article/19980914/ANA/ 809140778 (accessed 14 May 2008). Kisiel, R. (1998) ‘Merger Means Big Payday for Chrysler Execs’, Automotive News.com. 10 August. http://www.autonews.com/article/19980810/ANA/808100730&Search ID=7331762469745 (accessed 14 May 2008). Kisiel, R. (2001) ‘Zetsche’s Goal: Disciplined Pizzazz’, Automotive News.com. 7 May. http://www.autonews.com/article/20010507/ANA/105070779 (accessed 14 May 2008).
222 The Second Automobile Revolution ‘Merger is a Marriage Made in Heaven’, BBC News.com. 7 May 1998. http://news. bbc.co.uk/1/hi/business/88965.stm (accessed 13 May 2008). Muller, J. (2007) ‘Chrysler For Sale?’, Forbes.com. 14 February. http://www.forbes.com/ business/2007/02/14/chrysler-for-sale-biz-cz_jm_0214chrysler.html (accessed 13 May 2008). ‘Should We Spell it Pre$$?’ Automotive News.com. 24 September 2007. http://www. autonews.com/article/20070924/SUB/709240339 (accessed 13 May 2008). Wernle, B. and Leslie, A. (2007) ‘Nardelli: Chrysler Needs 3 Years to Turn Finances Around, 10 Years for Product’, Automotive News.com. 7 September. http://www. autonews.com/article/20070907/ANA/70907058 (accessed 13 May 2007). Wilson, A. (2007) ‘Cerberus Buys Chrysler Group for $7.45 Billion’, Automotive News.com. 14 May. http://www.autonews.com/apps/pbcs.dll/article?AID=/ 20070514/REG/70514001 (accessed 13 May 2008).
Part III The Resistance of Leading European Carmakers
This page intentionally left blank
12
The Final Chapter of the ‘VW Model’? The VW Trajectory, 1995–2005 Ulrich Jürgens
Introduction The period under study covers a phase in the Volkswagen (VW) trajectory marked by a particularly close succession of growth and crises, high-flying plans and bitter disappointments. At the end of the period, the VW Group was acquired by the family-owned niche carmaker Porsche. Our account describes the factors and processes which made this shift possible. The shift could have far-reaching effects on the productive model that had determined the company trajectory of VW for more than half a century. The concept of ‘productive model’ takes up the conceptual notions of GERPISA (Boyer and Freyssenet, 2002; Freyssenet et al., 1998). The account begins by recapitulating the specificities of the VW productive model and the course of development from the founding of the enterprise to the beginning of the 1990s. The following section describes the reconstitution of the VW productive model against the background of the crisis that hit the company in 1993/94. The new governance compromises concluded at that time and the structures and arrangements to which they gave rise had a decisive impact on developments in the period that followed. The discussion then deals with developments in the central dimensions in the sense of the productive model concept for the period 1995–2005. Changes in product policy are examined together with the production organisation, labour relations and capital market relations. The following section describes the critical phase of radical change in the crucial year 2005. Finally we offer conclusions and address the prospects of future development for VW.
The VW industrial model and its trajectory until the mid-1990s1 Boyer and Freyssenet describe VW as the only company apart from Ford to have adopted the Ford productive model after the Second World War (Boyer and Freyssenet, 2002: 58f.). The configuration that made up VW’s productive 225
226 The Second Automobile Revolution
model after the war corresponded, with some deviations, to the Ford model. There were clear parallels in product policy, based on a standard model, the Beetle. Secondly, the product policy was matched by a productive organisation that aimed to exploit economies of scale which made high wages and salaries possible to compensate for the adverse working conditions of mass production. With regard to the third characteristic, the presence of a strong trade union, the VW configuration corresponded to the Ford model, but went far beyond it in the influence wielded by the union. This strong influence was grounded firstly in the historical point of departure for the company and its corporate governance structures; and secondly, in the system of co-determination prevailing in Germany, installed by legislation in the 1950s and 1970s. The fourth characteristic of the VW model is specific to the company, and concerns the special role of the state as owner and at later stages the dominating block holder in VW’s ownership structure. The VW Works became a government-owned company in 1949. When it became a public limited company in 1960, the state kept 40 per cent of the shares, split equally between the State of Lower Saxony and the Federal Republic of Germany. The latter sold its shares in 1988, thereby reducing the government share to 20 per cent. Thanks to the VW Act 1960, however, state ownership remained dominant. The VW Act was passed when VW was privatised. Important clauses of the 1960 Act included: any increase in shareholder ownership beyond 20 per cent of total shares does not lead to further voting rights, thus the State of Lower Saxony with its 20 per cent share could not be outvoted by another block owner; decisions on new plants or plant relocation require a two-thirds majority on the supervisory board. Hence, when labour and government representatives on the supervisory board voted together they could block any decision concerning the construction or relocation of production sites. The European Union has criticised the ‘VW Act’ with growing urgency since the mid-1990s, and has declared it an impediment to free capital flow within the EU. EU pressure to rescind the VW Act and, finally, the formal ruling by the European Court of Justice that key elements of the Act contravened European law, played a key role in the later shift of the company into family ownership. For Lower Saxony, VW had been a tremendous stroke of luck in the postwar period. For the structurally weak, largely agricultural state VW became an engine of economic development in the years after the war. The central location of VW was the huge plant in Wolfsburg which was built in 1938 and became the company headquarters after the Second World War. The new plants which were hived off from the Wolfsburg plant to cope with rapid growth with one exception were all set up in Lower Saxony. Together
The VW Trajectory, 1995–2005
227
these six plants formed the core of the VW system and have retained a special status in the VW Group to this day. And within this core group, the Wolfsburg plant remained the undisputed centre of power for both management and the works council. The state refrained from direct influence on VW’s operations even during the 1950s when the company was fully state-owned (Wellhöner, 1996: 297ff.). After privatisation state and labour representatives could form a majority within VW’s supervisory board, a possibility which during periods of social democratic reign has given ground to the critique of VW as a quasi-socialist company. There is no evidence for such a boardroom coalition determining company policy at a certain stage, however. The most distinctive feature of the VW productive model is its special system of co-determination. The foundations of the traditional partnership between management and labour at VW were laid immediately after the Second World War, and its development was furthered by the legislation on company co-determination with the introduction of the works council system in 1950s and the Act on Co-Determination at the company level which provided equal representation of capital and labour on the supervisory boards of large companies. With its specific governance structure, co-determination, and the strong influence of the union on management, VW has so far differed strikingly from all other automotive firms. The result was that everything that happened at VW attracted great public attention. When the company went through a period of successful development, the VW model was held up as a perfect example of the German model. When the firm experienced times of crisis, however, it was regarded as a perfect example of the problems and weaknesses plaguing Germany. Fordist principles of productive organisation dominated throughout the VW system. In 1967 the huge plant in Wolfsburg was still dedicated to the Beetle. The effect of mechanisation and high volume production was enormously cost-saving. The product policy of model continuity also saved costs. These savings were reflected in the price of the Beetle. The price of the 1967 basic model of the Beetle was 7 per cent below the price the consumer paid for the standard model in 1950 in real terms. But the monoculture developing in the firm thanks to the enormous success of the Beetle was a major reason for the crisis in the mid-1970s. Against the background of the oil crisis and criticism of products in North America (Ralph Nader) directed particularly at the Beetle, sales fell, and the drive concept of the Beetle had reached its limits. Luckily, Audi, a new affiliate formed from two companies acquired in the 1960s, had a modern drive technology which was to become the basis for the new product generation. When the crisis hit the company in 1974/75, the new generation of VW products was still in the development stage or just being introduced on to the market. This helped to overcome the crisis rapidly but not to cope with the
228 The Second Automobile Revolution
direct personnel problem. When the situation intensified, there was a surplus of 30,000 employees that had to be shed as quickly as possible to save the enterprise. Mass dismissals of such dimensions were acceptable neither to the works councils and trade unions nor to the State of Lower Saxony. The essence of the governance compromise achieved was to utilise the potential for ‘voluntary’ departures through severance pay and early retirement programmes. The principle of carrying out personnel adjustments in a ‘painless’ and ‘socially acceptable’ manner remained a cornerstone of personnel policy in later years. On the basis of the new social consensus, the new product policy provided the point of departure for a new growth cycle, which lasted until the early 1990s. Product policy underwent a change, with VW gradually taking leave of the Ford model and developing a ‘volume and diversity’ productive model. The Beetle was replaced by three new models for different customer segments: the Golf as the founding product in the compact vehicle segment, and the Polo and Passat as small and mid-range cars. It did not take long before the Golf took over the Beetle’s role as the bread and butter car at VW. Dependence on a single product, however, was considerably reduced. In addition, during the 1980s, the number of models and engine displacement sizes of VW’s passenger car fleet increased more and more. The strategy of product upgrading and exploiting the lower price sensitivity of better-off customer groups increasingly superseded orientation on the mass market; this, too, was an important indication of the change in productive model (Boyer and Freyssenet, 2002: 103). The company fared well in most of its major markets (except North America, where VW was unable to repeat the success of the first decades) during the period from the late 1970s to the 1980s. The second oil crisis, which was strongly felt by many companies in North America and Europe, had little impact on VW. The company profited from strong sales of its product range, especially in Europe. This was the period in which VW gained its position as Europe’s largest automobile company, when it bought Seat and started its activities in China. Sales increased almost threefold within a decade and an expansionist mood took root.
Reconstitution of the VW industrial model under the impact of the crisis of 1993/94 In the early 1990s, the growth cycle peaked. Three and a half million vehicles were produced, the highest figure in the post-war period. Having taken over Seat in Spain in 1986 and Škoda in the Czech Republic in 1991, the company had gained new capacities in Southern and Eastern Europe. This was in addition to the takeover of East German production sites which were integrated into an affiliated company of the VW brand. With the construction of new assembly plants in Mosel, East Germany and Martorell, Spain, and the
The VW Trajectory, 1995–2005
229
investment in modernising Škoda, the company was set for further growth. But problems had been piling up since the late 1980s: poor productivity, negative returns on sales, and a break-even point exceeding 105 per cent of normal production capacity at some stage pointed to serious weaknesses in the organisation of production. In this situation a new CEO, Ferdinand Piëch, who had previously held this position at Audi, was appointed. He took over the job at the beginning of 1993. Piëch was expected to intervene strenuously, in keeping with the reputation he had earned at Audi. It was by no means self-evident therefore that the representatives of labour, the head of IG Metall and the leading works council representatives, and the representatives of the then Social-Democrat government of Lower Saxony would opt for Piëch when the decisions were made in VW’s supervisory board. But they did. Obviously they saw the need for a tough policy of rationalisation and a return to profitability. The hiring of Lopez from General Motors together with the forced exit of most of the management board’s old guard demonstrated Piëch’s determination to take tough action. Moreover, the German economy was now in recession. In late 1993, as in the mid-1970s, overstaffing accounted for 30,000 employees in a workforce of 108,000 in the six VW core plants. In this situation the strengths of the consensual VW system came to bear (see, for more details, Haipeter, 2000). Within only four weeks, an agreement between IG Metall and VW was struck to reduce working hours in order to share the work among existing employees. The reduction of weekly working hours by 20 per cent (from 36 to 28.8 hours) secured 20,000 jobs; the remaining 10,000 were to be secured by additional measures. The reduction of work time went along with a reduction of income in the order of 16 per cent. In return for this hefty concession, the company guaranteed an employment level of 100,000 at VW’s German core plants. Of the other group companies, only Audi followed with a similar agreement which reduced weekly working hours by 10 per cent. The work time agreement established a pattern of exchanging employment protection against concessions by the labour side. In the case of the 1993 agreement the reduction of income affected only indirect pay supplements with the effect that concessions were not felt by employees with respect to their monthly pay. The radical reduction of weekly working hours in the VW core plants was an important signal for other members of the VW Group, especially the new acquisitions in Southern and Eastern Europe. The newly constructed and modernised production capacities in these areas were thus given scope for growth while the core organisation took on itself the task of solving the problem of over-capacity. This signal was also important for acceptance of the dominant role played by the German works council leadership.
230 The Second Automobile Revolution
Despite similarities with crisis management in the mid-1970s, the causes of the crisis in 1993/94 went deeper. The high break-even point indicated more serious problems within VW’s productive organisation. What measures were taken after the foundations of social consensus had been consolidated at VW? When the actors put these questions to themselves, the lean production discussion was at its height in Germany. The German version of ‘The Machine that Changed the World’ appeared in 1990 (Womack et al., 1990). The key messages of the study found little applause at VW. This was particularly the case for the emphasis on team production which was praised by the MIT authors as the ‘heart of the lean factory’. In the area of supplier relations, priority was given to pressure on prices and supplier conditions first. Lopez as the head of ‘Purchasing and Production Optimisation’ was the symbolic figure in this context. It had become known that suppliers to VW had in the past demanded some 15 per cent higher prices for the same components than from competitors. Indignation about this and the bellicose manner of Lopez and his ‘warriors’ put the necessary pressure on suppliers to meet the demands of VW. But the real focus of VW’s new supplier policy was not in exerting price pressure or the like. It was concerned with reorganising purchasing policy by introducing global sourcing and a tendering system that included internal VW supplier divisions. And it was concerned with involving suppliers in the development of technical and organisational concepts for modularisation. Hence VW gave major stimulus to German suppliers, who accordingly appeared on the market with their own module concepts at an early date, enabling them to win new customers and open up new business areas (e.g. Hella for the front end, Brose for door modules, etc.). As far as outsourcing is concerned, VW was also out of step with the great lean production mainstream. Important component divisions were retained in-house; some, like tool and die making, which had been outsourced, were insourced again. The introduction of financial performance indicators and a controlling system on this basis, however, made the performance of internal supply divisions transparent, and they were integrated into the newly established global sourcing system. Like external suppliers, the in-house suppliers had to compete for tendered-out contracts, but had the advantage of last call vis-à-vis competing bids. On the whole, VW clearly distanced itself still further from the mainstream of vehicle manufacturers with the measures and arrangements it introduced in the aftermath of the 1993/94 crisis. More than ever, the firm was the focus of public attention, and was regarded as a pioneer into the era of postindustrialism with its reduction in weekly working hours. For critics, the company became even more of a provocation and symbol of the problems facing industrial production in Germany.
The VW Trajectory, 1995–2005
231
The development of VW’s productive model, 1995–2005 Product policy With the takeover of Seat and Škoda in Europe, the number of basic vehicle models produced in the VW Group rose rapidly; in the mid-1990s twenty different basic models were being manufactured in the group. These models often served the same market segments. Against this backdrop, the platform strategy pursued since 1993 had various functions. First, it served as a basis for organising product development in the VW Group. Second, it served as a basis for diversifying the product range, which now fully embraced the ‘volume and diversity’ productive model. The platform strategy was grounded in the difference between the platform, the parts of the vehicle invisible to customers but technologically demanding (engine, transmission, underbody, etc.) and which, from the point of view of development input and production costs, makes up 60 per cent of the entire vehicle, and the so-called ‘hat’, i.e. the upper bodywork, the vehicle interior and exterior, which the customer sees and which are accordingly more subject to fashion trends. The aim was now to reduce the number of platforms to four while increasing production runs to up to two million per year. It was important in regulating relations between the different brands of the VW Group for platform development to be reserved to the technical departments of VW and Audi in Germany. Seat and Škoda were left to develop distinctive ‘hats’ for their own models. VW and Audi were able to benefit most from the economies of scale offered by the development of platforms. At the same time, the brands were able to compete with one another on the markets. Škoda profited particularly from this, enjoying a considerable competitive advantage owing to the lower prices of its technically equivalent vehicles. The introduction of the platform system at VW was therefore in the interest of cutting costs as well as in the pursuit of group integration and governance goals. The first VW models to be built on the basis of the new platforms appeared on the market in 1996; in 1998 50 per cent of models were already based on platforms, and in 2004 95 per cent. As in modularisation, VW led the field in platform strategy. The energetic implementation of this course gave VW an edge on competitors in purchasing and in new product development and production costs at a much earlier stage. The fruits were apparent in rising cost effectiveness. The rich cash flow which the company enjoyed owing to a recovering economy and the platform strategy fed to a considerable extent into the acquisition of new vehicle brands and the development of the corresponding products: Lamborghini, Bentley and Bugatti. Particularly controversial was the development of the Phaeton as a new luxury model of the VW brand.
232 The Second Automobile Revolution
For the Phaeton, a special plant was built, where only the final assembly stages were carried out and which thus served also as a brand experience centre and showroom (the ‘Glass Factory’ in Dresden). The platform strategy created the possibility of expanding the range of models while exploiting scale advantages. The model policy aimed to cover more and more segments of the market. However, it remained controversial in the Group. The problem of ‘cannibalisation’ due to inter-brand competition has already been mentioned. To this were added the image problems higher quality models had with the platform approach. Thus the luxury model Phaeton was criticised because it differed little from the mid-range model Passat. Another point of criticism was that the strategy ran counter to claims, especially by the premium brand Audi, to be in a position to offer vehicles which are perfectly adapted to the segment strategy. After the experience of the first generation of platform vehicles, the platform approach was revised in 2000. Synergy effects were to be strengthened not only at the platform level but also across platforms. Considerable savings per vehicle were anticipated in production costs alone. In the period that followed, the further development of cross-platform modules relegated the actual platform approach more and more to the background. In 2005, a more ambitious platform approach was introduced throughout the Group under the heading ‘modular toolbox’, from which Group companies could help themselves with varying degrees of differentiation. According to company information, the degree of modularisation of the Golf V, which came on to the market in 2003, was 8 per cent; for the Passat in 2004 it was already 50 per cent, and for the Audi A5 in 2005, it had reached almost 70 per cent (VW AG Strategy Meeting: Status Report, 24 August 2005, p. 21). By applying the modular toolbox throughout the Group enormous savings and a productivity increase in the order of 30 per cent are expected (Freitag and Student, 2007: 49). Apart from cutting costs, the module box strategy offers more scope for product financing. The aim is to cover further niche segments and open up new segments. In sum, platform and module strategies have played a key role in the further development of the Group. The savings that could be achieved through this strategy were far greater than those that even the most stringent rationalisation measures could have brought. At the same time, this strategy offered far more favourable conditions for the growth of the company than a strategy relying chiefly on lean production, the approach preferred by other automotive firms at this time. But the savings attained on the basis of the platform and module approaches were not employed to lower prices but to upgrade products and introduce technological innovations, such as direct injection systems, the double clutch gear, etc.
The VW Trajectory, 1995–2005
233
Productive organisation As we have seen in the previous section, the organisation of production faced major problems in terms of low productivity and high structural costs. In the early 1990s, losses were made even when operating at full capacity, for which the Wolfsburg plant was blamed. Lopez had been brought in to deal with the productivity problem, and, as we have seen, became very active by mid-1993. His focus was on improvement activities. The idea was that improvements could be accomplished only with, and not against, the people on the shop floor. He also sought support in union and works council circles. His measures focused on ‘workshops’ in which all levels of the hierarchy, including members of the works council and shop stewards, sounded out the potential for improvement and developed appropriate measures. Lopez developed no new concepts in work and production organisation. Hence, there was no sweeping introduction of production teams and the like as propagated by the MIT study. In relations with suppliers, Lopez also failed to develop into the ‘Wolfsburg strangler’ as expected. His demands for hefty price reductions were accompanied by measures of supplier development and a support of improvement activities. Another focus was on the organisational implementation of the module strategy and, in this connection, the development of integrated plants with synchronised manufacturing structures. In the mid-1990s, implementation of the new concepts concentrated first of all on the new production sites, especially the East German works in Mosel, the former Škoda plant in Bratislava that had come to VW, and the main Škoda plant in Mlada Boleslav. These works became experimental plants for testing new supplier concepts and just-in-time production. In Mosel, satellite plants for the assembly of modules were set up in the region surrounding the VW assembly plant, but no supplier park was yet established. At Škoda, in contrast, the concept of locating suppliers under one roof with the manufacturer’s assembly operations was introduced. This form of supply was strictly rejected by the works council and union in Germany. In Bratislava two supplier parks in the formal sense of the term were established. In the German core works, comparable structures were introduced only at a much later date. Although VW was thus slow to outsource its internal component supply, it acted energetically when it came to tapping low cost resources in Eastern Europe. In the course of the 1990s, apart from the assembly plants set up or modernised in association with the development of Škoda, the entire production of Audi engines was relocated to Hungary, and VW built an engine plant in Poland. Seat manufacture and wiring harness production were entrusted to joint venture plants set up mainly in Poland (Blöcker and Jürgens, 2006).
234 The Second Automobile Revolution
Within a few years in the second half of the 1990s, VW thus developed intensive cross-border productions links between Western European highcost countries (HCCs), especially Germany, of course, and the low-cost countries (LCCs) in Central and Eastern Europe. Bratislava played a special role among assembly plants. In 2006, the range of products there included the small car Polo and the Audi sport utility Q7, as well as the twin models VW Touareg and Porsche Cayenne. From a cost point of view, these production networks permitted so-called ‘mixed calculation’. The favourable cost structures in low-wage locations could compensate high costs of activities carried out in high-cost locations. The combined costing option was expected to stabilise German ‘high road’ conditions (Widuckel, 2004). The flip-side of these low-cost locations was that they then served as benchmarks in comparing costs with German production sites. From the viewpoint of the German plants, especially the works council system, it was thus a shock when, in 1998, the Touareg became the first high-quality vehicle among of the VW brand to be awarded to an Eastern European works (Bratislava). After fierce discussion on the ‘rules of the game’ in allocating production among factories, it was agreed that the next model would be assembled in Wolfsburg on condition that this could be done at a comparable cost. The future of manufacturing in Germany was thus called into question. This condition was met, and the decision to build the van model Touran in Wolfsburg was made on the basis of a concept that combined far-reaching concessions with new forms of work and productive organisation. This new concept was labelled ‘Project 5000 × 5000’. Project 5000 × 5000 involved the founding of a new VW subsidiary, Auto 5000 GmbH, with the aim of creating 5000 new jobs at a standard wage (at the time) of DM 5000 per month. Alongside traditional production in the halls of the Wolfsburg parent plant, a new production system was to come into being subject to markedly different working and pay conditions. The labour and social policy objectives of the project are examined at a later point (see on Auto 5000 Hartz, 2007; Schumann et al., 2006). The specificities of productive organisation lay chiefly in the link between a team concept with comprehensive tasks and responsibilities, assigned to the teams and in a concomitant, very shallow hierarchy. A new approach was also adopted with respect to training. Setting up learning islands on the shop floor itself ensured that where training was needed it could be provided ad hoc when problems arose: that is to say directly and close to the job (in contrast to the procedures hitherto favoured in Germany). The emphasis on training was underlined by special regulation regarding the weekly working hours. At Auto 5000, a 38-hour week was introduced. While the regular working hours on the lines are 35, three hours per week are dedicated to training measures. Out of these three hours one and a half are paid by the company, and the remaining one and a half are contributed by the employee without
The VW Trajectory, 1995–2005
235
company pay. This regulation aims at developing Auto 5000 as a ‘learning organisation’. Another particularity was the pay system (‘Programmentgelt’) by which wage payments depended on fulfilling the requirements of the daily production programme. Product defects and assembly errors were to be fixed by the people responsible if necessary on overtime and – where it was the fault of the people – without payment. The Auto 5000 system exhibited no great peculiarities in the design of processes. Production took place in the old halls of the Wolfsburg plant. Assembly proceeded as before in the form of basically repetitive work, and compared to the main plant with an even shorter cycle time. After the founding of Auto 5000 in 1999, the first product, the Touran, went into series production in early 2003. Looking back, the technical director spoke of the economic success of the overall concept, which he attributed equally to the special structures and processes, pay arrangements and other factors. Auto 5000, he claimed, was a response to Toyota, adapted to the given cultural and societal environment in Germany (Ulbrich, 2006). The transfer of the special structures and arrangements at Auto 5000 to other VW plants – and to other firms in Germany – was a fiercely debated issue from the outset within management as well as within union circles. Within IG Metall this issue has led to bitter controversies. In 2008, it was decided to reintegrate VW 5000 into the production system of the main plant. At the same time, the introduction of a new companywide production system, the ‘Volkswagen Way’ modelled after the Toyota production system, was announced. Labour relations The crisis of 1993/94 and the way in which it was overcome not only consolidated and strengthened the basic consensus on the value of employment protection; it also demonstrated greater involvement of the works council system in the operative processes of company management. As far as cost and productivity problems are concerned, it was accompanied by sometimes spectacular projects and arrangements, in which, in one way or another, concessions on pay and working conditions were associated with job protection and job creation measures. Some of the most striking projects deserve more detailed mention. The socalled ‘site symposia’ (it was no accident that the name was borrowed from academe) are events organised at VW in each of the six core plants in the context of the annual ‘planning rounds’. They bring together the board of management and works council leaders with the management and works council of each plant. In a sort of ‘one-stop shop’, the event provides an opportunity to discuss alternatives for action with the people who make the decisions and, where possible, to attain consensus. The outcomes of the site symposia feed into the planning rounds, which adopt a five-year perspective
236 The Second Automobile Revolution
and which the group traditionally completes in the November of each year. In component plants, site symposia have played a particularly important role. They give the opportunity to stress the specific strengths of the plant in question with respect to the perennial issue of the ‘future of components’, and, in debate with central decision-makers, to develop strategies for mobilising existing resources instead of sinking into paralysis in the face of threats to sell or outsource (see, for more details, D’Alessio et al., 2000; Klitzke et al., 2000; Speidel, 2005). Project 5000 × 5000 has already been discussed in connection with the organisation of production. From the point of view of labour relations, it is even more noteworthy. The project is doubtless designed to address the problems of productivity and structural costs, especially in the Wolfsburg parent plant. Nevertheless, this appears to have been completely ignored. Despite the existing four-day week for the core workforce, personnel for the new firm Auto 5000 were newly recruited. Recruitment for Auto 5000 was given a specifically social thrust by the decision to engage only people who had been out of work, and thus to help reduce the high unemployment in the Wolfsburg area (17 per cent in the city), and in the territory of the former German Democratic Republic, the border of which is only 10 kilometres away where unemployment was much higher. Employees of Auto 5000 were to receive a 4 per cent lower monthly wage, but for a 38-hour week. In comparison with the core workforce still working a 28.8-hour week, this meant a ‘concession’ of about 27 per cent (about 20 per cent if we disregard the three hours dedicated to training). In compensation, profit-sharing was provided for Auto 5000 workers, in the hope that this would in the long run help close the gap. The 38-hour week for Auto 5000 workers out of which one and a half were unpaid was quite a bold step and would have been hard to swallow for the core workforce in Wolfsburg. The same held true for the ‘Programmentgelt’ and the principle of fixing one’s own faults in one’s own free time (which in practice rarely happens). For the works council system, the arrangements for Auto 5000 involved greater co-determination. For instance, performance regulation through a system of predetermined times was explicitly not introduced, and performance was made a subject of negotiation between works council and management. VW AutoVison and the Wolfsburg AG were another project skilfully combining concessions with social objectives. The point of departure was the sixtieth anniversary of the City of Wolfsburg. In view of the high unemployment in the area, the company offered the municipality the present of a ‘master plan’ for halving unemployment (Klobes, 2005). The main components of this concept were the recruitment of suppliers to the area, the setting up of an ‘innovation campus’ for spin-off firms, the construction of an ‘interaction discovery park’, and of the ‘auto city’ which combines a vehicle delivery centre for customers with exhibition centres for each brand of the VW Group, restaurants and other amenities. A company-owned temporary
The VW Trajectory, 1995–2005
237
work agency was founded for financing purposes, and a special wage agreement concluded for its employees between VW and the IG Metall. Returns from the temporary labour business with VW and external firms flowed into the financing of the projects mentioned. Agency labour was thus introduced at VW in a socially acceptable form. Euro works council and world works council With the takeover of Seat in 1986, VW had for the first time acquired a production site in a European low-cost country. Werner Widuckel, senior staff member of the general works council and group works council, claims that the development of parallel production in Spain was a key experience for the German employees’ representatives. In taking over Seat, management wanted ‘to create competition that was specifically directed against the traditional German production sites’ (Widuckel, 2004: 61). The plans for founding a European works council were developed already in the late 1980s. In 1990, the European VW Group works council was set up, initially without agreement with company management, which followed in 1992. The guiding principle was following up the experience of 1974/75, ‘in times of crisis to achieve a balance between volume and personnel to support production sites and distribute the burdens of the crisis’ (ibid.: 63). Relocation should not be undertaken at the cost of a location and needed to be compensated by alternative products or growth. No relocation plans were to be supported without having been discussed beforehand in the Euro works council, thus ensuring the coordination of interests (ibid.). The Eastern Central European locations were integrated into the Euro works council from early on. The Euro works council passed its first test in dealing with the crisis at Seat which was on the brink of bankruptcy in 1992. It also played a crucial role in dealing with major tensions with Škoda which emerged after VW had announced a drastic reduction and postponement of investment plans there. In both cases, the Euro works council played a key role in handling these critical situations. The same situation arose when management decided to close the plant in Brussels. Due to the interventions of the (Euro) works council a solution was found which guaranteed the continuation of the plant, however, at much lower employment levels. In 1999, the world group works council was founded with an agreement that was largely identical to that of the Euro works council. This integrated production sites in Brazil, Argentina, Mexico and South Africa but not the plants in China. In all, the development of working conditions in the study period have clearly been marked by an expansion and intensification of the codetermination system (Speidel, 2005; Widuckel, 2004). This is the case for the close cooperation between works council leaders and personnel management in developing the sometimes very ambitious projects described above, and it is also the case for the operative activities of works councils
238 The Second Automobile Revolution
and their involvement in questions of outsourcing, production relocation, performance regulation, etc. Capital market relations In view of the dismal record of profitability of the company, the new management team in 1993 had already stressed the need for higher profitability and financial performance. A return on sales target of 8 per cent to be realised ‘in the mid-term’ was defined. This target was later reduced to 6.5 per cent and complemented by other financial targets. Moreover, the introduction of cost centres decentralised responsibility for costs and results and defined appropriate controlling systems. These measures, however, have failed to impress financial analysts. ‘VW is Europe’s least transparent car maker,’ one investor said, and in the context of the critique of the investor community against preference shares at VW, a German funds manager complained: ‘VW is kind of a socialist company, it’s getting harder and harder to see why VW shares should be held at all’ (Wall Street Journal Europe, 8 June 2001). The scepticism of the capital market towards VW may have contributed to making mergers and acquisitions a marginal phenomenon as far as VW is concerned, and plans for takeovers had to be curbed by management in advance. In any case, VW, too, was infected with takeover fever at the high point of the New Economy bubble. Focal developments were the purchase of Rolls-Royce and Bugatti in 1998. In 2000 VW rounded up its ownership over the rental car firm EuropCar but sold the firm altogether in 2006. Strategically most far-reaching was the acquisition of an 18.7 per cent stake in the capital of Scania in 2000, which gave it a 34 per cent share of votes. In 2006, a strategic interest in Man was announced, which by early 2007 had been raised to just under 30 per cent of shares. It is likely that these investments had to do with far-reaching plans to create a leading world group in the truck sector. In any case, it is clear that VW has not taken the path that famous role models of the New Economy era had traced. Despite the expansion of financial services, VW has not followed Ford and other companies in officially announcing a strategy of ‘value migration’ away from automobile manufacturing as a low-margin activity towards higher-margin activities downstream the value chain. The official policy is that financial services support VW’s core business activities of auto manufacturing. At the same time that VW discovered the usefulness of the stock market to provide ‘acquisition money’ it became aware of its own weakness as a target for hostile takeovers. Rumours of Ford being interested in acquiring VW, the EU competition commissioner scrutinising the VW Act, and the low market capitalisation of VW – all this contributed to rising fears of a hostile takeover, which almost became obsessive at the beginning of the new decade. With a market value of about a12 billion (late 2002 and also, after a brief upturn
The VW Trajectory, 1995–2005
239
in 2004 again), VW was, after all, a ‘bargain’ for competitors or institutional investors. In 2001 various possibilities to build a protective wall around VW were played through with the help of consultants. Nevertheless, the danger of a hostile takeover hung over the company like the sword of Damocles from the late 1990s. Determined lobbying in the European Parliament succeeded in parrying the first initiatives of EU Commissioner Bolkenstein, but the parties involved were aware that it was only a question of time before the protection afforded by the VW Act was removed. This danger was also recognised by the works council, and has done much to further the willingness of the works council to step up efforts to reduce structural costs.
Struggles about the future direction of VW 2001 marked a turning point in the successful development of VW, also as regards profitability. Figure 12.1 shows that, after the sharp downturn in the operating margin in 1993, economic performance improved greatly until 1998 and, after a brief downswing in the following year, remained at a high level until 2001. In 2002, after the New Economy bubble burst, there was a renewed, major downturn. Profitability stabilised again in 2003, although at a lower level.
7 6.01
6 4.98
5
4.68
4 3
0
4.97
3.40 2.34
1.97
2 1
4.47
3.36 1.60
1.26 0.71
1.22
1.72
1.71
0.58
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
⫺1 ⫺2 ⫺2.13 ⫺3 Year Figure 12.1 The operating margin of Volkswagen, 1991–2006 Source: Own calculations, based on Volkswagen Annual Reports.
240 The Second Automobile Revolution
The downswing in 2002 was accompanied by growing criticism of Piëch and his policy. A phase of boardroom conflicts set in. In 2002 Piëch resigned as CEO to become chairman of the supervisory board. His successor Pischetsrieder cautiously set new priorities, especially concerning VW’s premium strategy. Differences increased when the Social-Democrat government in Lower Saxony was replaced by a coalition between conservatives and liberals. At this point in time, the decline in profitability put the question of the low productivity of VW works on the agenda once again. A Group-wide rationalisation programme (For Motion) achieved considerable savings, but more radical measures seemed to be needed. In 2005, Wolfgang Bernhard, who had made himself a name as a tough reorganiser at DaimlerChrysler, was called in. Bernhard remained true to his reputation. With reference to the findings of a Europe-wide productivity study (Harbour Report), he announced drastic intervention. Citing the report, he claimed that the average European competitor in the volume segment needed 25 hours or fewer to assemble a car while VW required up to 50 hours. According to Bernhard, the German part of VW could survive only because VW works abroad or subsidiaries helped with cross-subsidies. Once again, the fateful VW figure of a surplus of 30,000 jobs was at issue. But, like Lopez in the 1993 crisis, Bernhard did not blame the works council system for VW’s cost problems. In his critique he focused rather on issues of product policy. In his view, VW cars were over-engineered with too costly options and with a fundamental neglect of production requirements. Thus, product design solutions made it difficult and costly to produce the cars. Bernhard now began to redesign and de-content the upcoming models, revising decisions taken by his predecessor. As a result, tensions grew between him and the former CEO, Piëch, now chairman of the supervisory board. The dissension between the two positions obviously could have had farreaching consequences for VW’s productive model. But there was no explicit statement about an alternative product policy. Bernhard’s measures focused on costs and affordability of VW’s products while Piëch defended the policy of product upgrading and a market positioning in the premium segment. Bernhard’s position found the support of the representatives of the Lower Saxony government representatives. This critical situation was compounded by the scandal involving the central protagonists of VWs special system of co-determination. The scandal had all the ingredients beloved of the tabloid press: bribery, sex affairs, connivance, and finally the resignation of the heads of personnel and the works council, Hartz and Volkert. Two key people, who had played a major role in setting the tone for government compromises in the company during the period, were suddenly no longer there. The scandal was particularly shattering for the VW co-determination system.
The VW Trajectory, 1995–2005
241
The role of VW as an example of modern, innovative co-determination was compromised. In this situation, Porsche announced the acquisition of a 20 per cent stake in VW in September 2005, arriving on cue as the proverbial white knight. Piëch’s position on the supervisory board stabilised; and with his confidant Winterkorn, his successor at Audi, replacing Pischetsrieder as CEO, and Wolfgang Bernhard bowing out, the old model appeared to have regained its full power. The impression of stabilisation was also confirmed when the scandal involving Klaus Volkert proved to have less impact on political support for and confidence in works council activities among the workforce than had initially been feared. Indeed, in the regular election to the works council held a few months later, the candidate to succeed Volkert as chief works council representative even improved his showing. The turbulence and negative headlines provoked by the boardroom struggles concealed a development that first became apparent in 2006 and which suddenly caught even the severest critics among the financial analysts on the wrong foot. In 2006 and 2007, performance was phenomenal. Analysts wrote of the ‘VW turnaround’. A study by UBS Investment Research (Warburton et al., 2007) named VW and Fiat as front runners in a ‘European autosector turnaround’. The authors wrote of the ‘transformation of VW into a volume hungry price-setter’ (ibid.: 15). The VW brand, and the German core plants in particular, played a key role in the turnaround. Once again, as in the 1993/94 crisis, a ‘strong man’ had been needed. But again it was also the special co-determination conditions that made rapid and efficient implementation possible. On the basis of renewed security, VW announced far-reaching plans for growth in late 2007 in the context of ‘Strategy 2018’. Annual growth rates of over 10 per cent are to increase volume to over 11 million vehicles by that date (compared with 5.7 million in 2006). The new production system, the ‘Volkswagen Way’, was regarded as the basis for achieving this growth by increasing productivity, i.e. without further growth of the workforce.
Conclusions and outlook As this account has shown, the period under study was a phase of strong growth for the Group. The measures and concepts adopted in the face of the crisis in 1993/94 reinforced the specific VW trajectory. In many areas, solutions were found that deviated from the sectoral mainstream. VW’s product policy with modularisation and the platform approach established the crucial basis for growth during the period. At the end, it became a matter of controversy and boardroom struggles which were resolved by the entry of Porsche. In the field of productive organisation, the problem of lagging productivity remained an unresolved issue. Typical of VW are periodic rationalisation
242 The Second Automobile Revolution
campaigns driven by a ‘strong man’ called in with the agreement of all parties in the company. The introduction of the ‘Volkswagen Way’ could change this pattern and provide the basis for a future, self-sustaining system of continuous improvement in productivity and costs. The special system of corporate governance and co-determination also played a vital role during the period under study. Ultimately, the future of the VW model lies in the hands of the Porsche holding and the Austro-German family behind it. It is clear that this development has not been caused by a failure of the VW model to cope with challenges of the product markets and competition. As in the preceding phases of weakness and crisis, the model has, as we have seen, been able to stabilise itself by its own efforts. This is particularly true of the old core of the VW Group, to which even the financial press, not over-friendly towards VW, has conceded a key role in the turnaround. Government, union and the works council system, hitherto strong stakeholders, look like the losers in this development. Their influence appears to have markedly diminished in the new corporate governance structures the specific design of which is currently the subject of fierce discussion. It is not clear what strategic interests the new owners, the Porsche–Piëch clan, have in the company and how great is the risk of internal family conflicts. In view of the danger of growing conflicts in the hitherto consensual culture the future of the VW model remains an open question at the end of this account. Crucial will be whether actors will be able to balance out their positions of influence and arrive at a governance compromise which reconfirms the basic orientations of VW’s productive model. Another crucial factor will be the impact the world economic crisis has on the future of the VW model – but this is a question for another study.
3,238 3,500 3,019 3,042 3,595 3,977 4,291 4,823 4,853 5,156 5,108 5,023 5,021 5,093 5,219 5,660
1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1,814 1,929 1,411 1,425 1,526 1,591 1,619 1,938 1,879 1,830 1,886 1,781 1,740 1,832 1,913 1,935
Production domestic b
1,424 1,571 1,608 1,617 2,069 2,386 2,672 2,885 2,974 3,326 3,222 3,242 3,281 3,261 3,306 3,725
Production abroad c
Volkswagen, 1991–2006
874 1,051 762 799 921 957 952 1,302 1,316 1,216 1,312 1,220 1,224 1,226 1,219 1,307
Export d 39,019 43,666 39,158 40,924 45,055 51,192 57,901 68,637 75,167 83,127 88,540 85,293 84,813 88,963 93,996 104,875
Sales revenues totale 18,591 20,200 17,551 16,825 17,642 18,621 20,038 23,899 22,874 24,526 24,484 23,874 26,163 24,504 26,297 28,544
Sales revenues domestic e 39,955 45,895 42,260 47,134 53,615 63,704 37,863 44,738 52,293 58,601 64,056 61,419 58,650 64,459 67,699 76,331
Sales revenues abroad e 277 273 253 238 257 261 275 294 306 322 324 324 335 343 345 329
Workforce totalf 167 164 150 141 143 139 144 153 159 163 167 168 174 179 179 174
Workforce domestic f 110 109 103 97 114 122 131 141 147 159 157 156 161 164 166 155
Workforce abroad f 570 75 −992 77 172 347 696 1,147 844 2,062 2,926 2,597 1,003 697 1,120 2,750
Net earningsg
Notes (a) VW group; VW, Audi, Seat since 1991, Škoda since 1992, in thousand units. (b) From domestic VW and Audi operations, in thousand units. (c) In thousand units. (d) Source: VDA, Tatsachen und Zahlen, various editions. (e) Adjusted for internal transactions between the individual consolidated Group companies; Škoda was consolidated into the Group in 1992 (in million a). (f) Year’s average in thousand employees. (g) Gross sales minus taxes, interest, depreciation, and other expenses; i.e. the ‘bottom line’. Figures for the years 1991–2000 are taken from the Annual Report 2000 and were converted into euros. Source: VW Annual Reports, various editions.
Productiona
Year
Appendix Table 12.1
243
244 The Second Automobile Revolution
Note 1. This period has been dealt with in detail in Freyssenet et al. (1986) (cf. Jürgens 1998).
Bibliography Blöcker, A. and Jürgens, U. (2006) ‘The Restructuring of Value Chains by Multinational Companies in the European Automotive Industry and the Impact on Labour’, in B. Galgóczi, M. Keune and A Watt (eds), Jobs on the Move, special issue of Transfer: European Review of Labour and Research. Boyer, R. and Freyssenet, M. (eds) (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. D’Alessio, N., Overbeck, H. and Seitz, D. (2000) Rationalisierung in Eigenregie. Ansatzpunkte für den Bruch mit dem Taylorismus bei VW. Hamburg: VSA Verlag. Freitag, M. and Student, D. (2007) ‘Die Achsenmacht’, Manager Magazin, 12: 45–56. Freyssenet, M., Mair, A., Shimizu, K. and Volpato, G. (eds) (1998) One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Haipeter, T. (2000) Mitbestimmung bei VW. Neue Chancen für die betriebliche Interessenvertretung? Münster: Westfälisches Dampfboot. Hartz, P. (2007) Macht und Ohnmacht. Hamburg: Hoffmann und Campe. Jürgens, U. (1998) ‘The Development of VW’s Industrial Model’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 273–310. Klitzke, U., Betz, H. and Möreke, M. (eds) (2000) Vom Klassenkampf zum Co-Management? Perspektiven gewerkschaftlicher Betriebspolitik. Hamburg: VSA Verlag. Klobes, F. (2005) Produktionsstrategien und Organisationsmodi. Internationale Arbeitsteilung am Beispiel von zwei Standorten der VW AG. Hamburg: VSA Verlag. Schumann, M., Kuhlmann, M., Sanders, F. and Sperling, H. (eds) (2006) VW Auto 5000: ein neues Produktionskonzept. Die deutsche Antwort auf den Toyota-Weg? Hamburg: VSA Verlag. Speidel, F. (2005) Mitbestimmte versus managementbestimmte Globalisierung in der Automobilindustrie. Ein Vergleich der Internationalisierungsstrategien und ihrer Verarbeitung durch die Akteure der industriellen Beziehungen am Beispiel VWs und Renaults. München: Rainer Hampp Verlag. Ulbrich, T. (2006) ‘Eine Alternative zum Toyota-Konzept’, in M. Schumann, M. Kuhlmann, F. Sanders and H. Sperling (eds), VW Auto 5000: ein neues Produktionskonzept. Die deutsche Antwort auf den Toyota-Weg? Hamburg: VSA Verlag, pp. 137–41. VW AG Strategy Meeting: Status Report 24. August 2005, URL: http://www. volkswagenag.com / vwag/vwcorp/info_center/en/talks_and_presentations/2005/08/strategy_ presentation.-bin.acq/qualBinaryStorageItem.Single.File/2005- 08-24_London_TOP_ COPY.pdf. Warburton, M., Acquilla, A. and Hauser, E. (2007) ‘Which of the Emperors Has No Clothes?’, UBS Investment Research European Automobile Manufacturers, 15 October.
The VW Trajectory, 1995–2005
245
Wellhöner, V. (1996), ‘Wirtschaftswunder’ – Weltmarkt – Westdeutscher Fordismus. Der Fall VW. Münster: Westfälisches Dampfboot. Widuckel, W. (2004) Paradigmenentwicklung der Mitbestimmung bei VW. Reihe Forschungen, Positionen, Dokumente Bd. 1, Wolfsburg: Historische Kommunikation der VW AG. Womack, J. P., Jones, D. T. and Roose, D. (1990) Die zweite Revolution in der Autoindustrie. Konsequenzen aus der weltweiten Studie aus dem Massachusetts Institute of Technology. Frankfurt am Main and New York: Campus.
13
PSA: the Difficulties of a ‘Volume and Diversity’ Profit Strategy Michel Freyssenet
With its 1976 takeover of Citroën, followed in 1979 by Chrysler’s European subsidiaries, PSA was trying to quickly acquire the means to implement a new approach for it – a Sloanian ‘volume and diversity’ strategy. The constitution of a multi-brand group would have given the company a chance to design shared platforms for its Peugeot, Citroën and Talbot models (the latter being the new name given to Chrysler Europe) while hoping for economies of scale and a broad market coverage far superior to the one enjoyed by its direct competitors, Renault, Fiat and Volkswagen. Changing a profit strategy is, however, easier said than done, especially when the newly absorbed carmakers had previously followed quite different profit strategies. Sharply lower automobile demand and higher manufacturing costs following the monetary and oil crises of the time also made this a particularly risky operation. In reality, to avoid bankruptcy the Group needed to lower its break-even point as soon as possible. This meant that cost cutting became even more of a priority than the goal of commonalising mechanical parts, platforms and the industrial apparatus. It also meant that the new entity could take full advantage of the sudden rise in sales caused by the economic bubble of the late 1980s and the remarkable success of two of its car models. PSA achieved record profits that it has never equalled since. However, once the first speculative bubble burst in 1991, profitability fell as quickly as it had risen. Executives were particularly dismayed by this reversal because they were unable to use the good old recipe of cost cutting to stave off its effects. There was an evident incompatibility between the innovation strategy that had made Citroën’s national and international reputation and the ‘volume and diversity’ strategy that the new Group was supposed to pursue. Things were complicated by the way that the demand structure had also evolved to favour conceptually innovative car models. A final problem was the fact that rising demand from Mercosur, India and China did not yet offset lower demand elsewhere. It seemed possible to resolve this dilemma via a new approach where platforms would be used as a foundation for both traditional and innovative 246
PSA: a ‘Volume and Diversity’ Profit Strategy 247
models. PSA may have become technically and stylistically more innovative but it was still not fully capable of launching models that would create new market segments and generate innovation rents. It basically continued to derive its profits from volume and diversity, and this was enough to provide it with enviable profitability through 2002. It was also able to achieve economies of scale, thanks not only to its platform policy but also to the rising number of ad hoc cooperative agreements that it signed with other carmakers, both to manufacture mechanical parts and to produce models with too small a sales potential for a single carmaker. The Peugeot family, which still has a controlling interest in the PSA holding, yielded neither to the late 1990s fashion of mega mergers nor to the sirens of the ‘New Economy’. Instead, it stayed very close to its original automobile business. It remains that the particularity of a ‘volume and diversity’ strategy is its constant need for increasing economies of scale, especially when the demand for traditional models (or models that are viewed as traditional) slumps compared to the demand for new vehicle types. Ad hoc cooperative agreements cannot be reproduced ad infinitum. At a certain point, companies need to invest in growth markets and/or merge with other carmakers that accept the same profit strategy. Whereas PSA has been one of the first carmakers to try to move into China and India in the 1980s and 1990s, it missed out on these countries’ skyrocketing demand in the early 2000s, having preferred in the meantime to concentrate on Mercosur and Central Europe. Since then, it has tried to make good on lost time in China and Russia. It no longer excludes the possibility of an alliance or merger and still dreams of reconciling its shared platform concept with innovative models, despite its disappointments up to now with this strategy. PSA, like many carmakers, has followed a trajectory that has impacted its employees quite negatively when no compromise has been found with them; less so when group management has been able and willing to negotiate. We will see, however, that the variability of its earnings cannot be attributed to employees but derives instead from strategic choices that turned out to be unsuitable or else were not implemented in a sufficiently coherent manner.
PSA’s rescue and the effects on its subsequent strategy, 1993–1997 When the property bubble burst in 1990, West European governments felt the need to adopt certain restrictive economic policies that led in turn to lower automobile sales: 2.6 million fewer vehicles were sold in 1993 compared to the highs of 1990. Buyers were more interested than ever in mid- and lower range vehicles and less so in optional extras. Carmaker margins also fell, especially when they started to wage a price war.
248 The Second Automobile Revolution
The wrong policies PSA sales fell basically in the same proportions (−14.7 per cent) as West European new sales (−16.7 per cent), with sales elsewhere not offsetting this downward trend. PSA was still an entirely West European manufacturer, with 87.6 per cent of total sales being made in this region. Profitability started to decline in 1991 and by 1993 the company was in the red. The pill was especially bitter to swallow given PSA’s belief that its troubles were behind it, having previously come close to bankruptcy. West European demand slowly picked up over the next three years, reaching 14.8 million vehicles in 1996. Earnings remained mediocre, however, and the Group’s automobile branch lost money in both 1996 and 1997 (Figure 13.1). PSA cut staff numbers (a loss of 4,200 in 1993 and a further 6,000 in 1994), reduced working hours, squeezed suppliers, restructured its debt, optimised its investments, improved quality and gave assembly line workers more responsibility – but nothing worked. Profitability seemed unattainable. From that point onwards, Jacques Calvet, PSA’s chairman, began attributing his problems to the incoherency of European governments’ macroeconomic policies and the absurdity of regulations developed by ‘technocrats in Brussels’. He proclaimed this view loudly in the press, breaking with the legendary discretion that had been a characteristic of the Peugeot family. Calvet vigorously denounced the competitive devaluations taking place in Europe, the continent’s high real interest rates, the organised opposition to
25
Peugeot
Peugeot ⫹ Citroën
20
PSA Auto
15 10
Group
5
06
04
02
20
20
00
20
98
20
96
19
94
19
92
19
90
19
88
19
86
19
84
19
82
19
80
19
78
19
76
19
74
19
72
19
70
19
68
19
⫺5
19
66
0 19
Operating margin/revenue
IFRS standard
⫺10 ⫺15 ⫺20 ⫺25 Year
Figure 13.1
Operating margin/revenue rate of PSA (Group and Auto), 1976–2007
Source: Jetin (1999) and own calculations, based on PSA Annual Reports.
PSA: a ‘Volume and Diversity’ Profit Strategy 249
diesel (PSA’s pride and joy), the absence of sufficient protection from Japanese carmakers, new taxes, etc. He waged war against the Maastricht Treaty and lambasted the socialist policy of a 35-hour working week along with many other things. Several members of the Peugeot family were upset when Calvet threatened to resign and stand in France’s 1995 presidential election. The family head, Pierre Peugeot, even as he reminded Jacques Calvet that Peugeot and Citroën buyers were just as left-wing as right-wing politically, felt that he could not shorten the mandate of someone who had saved the Group (and the fortune of the Peugeot family). The question is why the spectacular recovery of the late 1980s had not laid the foundations for lasting profitability. To answer this, the strategic dilemma in which PSA found itself after its acquisition of Citroën and Chrysler Europe needs to be analysed. PSA’s historical dilemma The aim of having a multi-brand group had already been clearly formulated in 1976 when Citroën was acquired. It was confirmed during the 1979 takeover of Chrysler’s European subsidiaries (Simca in France, Rootes in the UK and Bareiros in Spain, later combined under the Talbot brand name). The idea was to get economies of scale on vehicles’ invisible parts and to offer diversity on visible parts, thus implementing a strategy very similar to the one that General Motors invented during the interwar years under the tutelage of its CEO of the time, Alfred Sloan (Freyssenet, 1998). Nor was there anything crazy about the successive acquisition of two other carmakers. Volkswagen had just had a good deal of success with the same approach, although there were differences between the two companies’ situations. VW had absorbed small, declining carmakers that were forced to accept its strategy, whereas PSA was integrating Citroën and Chrysler Europe, each of which produced at least as many autos as Peugeot did and enjoyed enviable brand images. Citroën, in particular, was nationally and internationally recognised and appreciated for vehicles that were technically, stylistically and conceptually innovative (front-wheel drive, the famous 2CV model, the DS, etc.). It had a loyal customer base that PSA could not afford to lose. In actual fact, this was the merger’s main challenge – how to conduct a commonalisation policy while preserving each carmaker’s strengths. How to reconcile the ‘innovation and flexibility’ strategy characterising Citroën with the ‘volume and diversity’ strategy desired for the whole of the new Group (Boyer and Freyssenet, 2002). It is this dilemma that PSA wanted to resolve. Many years later, it has yet to find the answer. Prioritising profit margins In truth, PSA’s executives had little time to figure things out. The second oil shock set world growth back for years. PSA’s financial situation worsened very
250 The Second Automobile Revolution
quickly (see Figure 13.1). The priority was to lower the break-even point from 2.2 to 1.2 million vehicles. Jacques Calvet cut staff numbers and closed twelve plants, despite industrial conflicts and government reluctance. The number of people working for the Group fell from 263,000 in 1979 to 165,000 in 1986. Calvet sold non-automobile activities (food outlets, tool businesses, gardening materials, etc.) and got rid of Talbot’s in-house sales network because it did not have enough new models to sell, before eliminating the brand name itself. PSA could not afford a third range. Ironically, after this latter decision falling oil prices in 1986 followed by the late 1980s property bubble led in most countries to resurgent automobile demand. PSA’s earnings were spectacular from 1987 to 1990, due in large part to its lower break-even point and much narrower product range. Moreover, two of the models that Jacques Calvet had inherited were big successes: the Citroën BX and Peugeot 205 (Figures 13.2 and 13.3). The turnaround meant that PSA could afford to resume its initial project, although people remained vigilant due to the trauma of its near bankruptcy. Once again, cutting costs to guarantee margins on each model was seen as the way to go. Slumping demand after the property bubble burst in the early 1990s seemed to justify this position. The number of automobile branch employees continued downwards, worldwide from 128,300 in 1993 to 121,100 in 1997, and from 102,400 to 93,100 in France. PSA made extensive use of severance facilities that were largely funded by the French state, involving either early retirement packages
2,000,000
807
1,800,000 607
407
1,600,000 1,400,000
605
308
Cars
1,200,000
806
1,000,000
405 604
406
800,000
Partner
309
600,000
307
306
504
505 206
207
400,000 305
304 404
200,000 202
203
205 106
204
104
1007 107
19 4 19 5 4 19 7 4 19 9 5 19 1 5 19 3 5 19 5 5 19 7 5 19 9 6 19 1 6 19 3 6 19 5 67 19 6 19 9 7 19 1 7 19 3 7 19 5 7 19 7 7 19 9 8 19 1 8 19 3 8 19 5 8 19 7 89 19 9 19 1 9 19 3 9 19 5 9 19 7 9 20 9 0 20 1 0 20 3 0 20 5 07
0
403
Year
Figure 13.2
Worldwide production of Peugeot cars by model, 1945–2007
Source: Peugeot Annual Reports, CCFA, J. J. Chanaron (1987).
PSA: a ‘Volume and Diversity’ Profit Strategy 251
or the repatriation of immigrant workers. In 1996, Jacques Calvet and Louis Schweitzer, CEO at Renault (which had just been privatised), suggested that the government adopt an Older Employees Activity Termination scheme (CASA), their excuse being that the two companies’ older workforces put them at a disadvantage to their Japanese rivals. The right-wing government of the time refused this project because of the cost to the wider community. Other cost-cutting measures that were just as severe ended up being adopted, however, particularly affecting wages. Labour relations, which were already poor, became execrable, despite a few belated improvements in employees’ skills rankings and in the application of labour law. The focus on cost cutting also impacted product policy. A refusal of risks relating to conceptual innovation Carmakers generally need to make technical improvements and introduce new equipment but things can be different when stylistic and above all conceptual innovations are involved. These two types of innovation can provoke total commercial failure and weaken a company just as easily as they can become a key factor of success and maximise profitability. Citroën had experienced both situations in the past. Jacques Calvet was unwilling to risk PSA’s survival in a game that he equated to Russian roulette. The one constant in his product policy was his refusal of conceptual innovation. Upon arriving at PSA he severed its ties to Matra Automobile, which the Group had inherited when it acquired Simca (Loubet, 2001: 426). He also stopped selling
1,400,000 C8
1,200,000
C6
C5
Cars
1,000,000
C4 Evasion
800,000 XM
SM
600,000
CX GS
400,000 200,000
Xantia
ZX
DS/ID Ami Traction avant
Méhari Dyane
Xsara
BX
C15 break
Visa LN/LNA
Axel
2CV
AX
C3 Berlingo C3 Saxo
C2 C1
19 1945 1947 1949 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 2099 2001 2003 2005 07
0 Year Figure 13.3
Worldwide production of Citroën cars, 1945–2007
Source: Peugeot Annual Reports, CCFA, J. J. Chanaron (1987).
252 The Second Automobile Revolution
the Matra-Simca Rancho, a forerunner of future fun vehicles like the SUV, instead of correcting its defects to increase sales. Calvet also refused plans for a monocorps vehicle by the same manufacturer, whereupon Matra turned to Renault. The product of this cooperation was the Renault Espace, whose well-known success derives from its creation of a new market segment. Calvet also rejected plans for a small utility vehicle, anticipating, fifteen years in advance, what would later become the Citroën Berlingo and Peugeot Partner. He opted instead for a more conventional model, the Citroën C15, derived from the Peugeot 104 platform. The entry-level model for this segment, the Citroën AX (1986–97), was the result of judgements that emphasised marketing surveys rather than the intuitions of Citroën design engineers, who had imagined a very fuel-efficient monocorps with a sharply sloping bonnet, a forerunner of the trends of the 1990s. A banal and relatively incoherent style was preferred, the idea being that anything else would have frightened customers (Broustail and Greggio, 2000: 155–6). It was only towards the end of this period that PSA began to act a bit more innovatively, launching a dual version of the Berlingo/Partner, now designed as a utility vehicle and fun passenger car. Renault did the same thing at the same time, however, depriving PSA of a possible innovation rent. Moreover, to save money, the models were not fitted with sliding side doors, unlike Renault’s Kangoo, which was one of the reasons for its better sales. The one area where PSA showed a desire to take risks before anyone else was in electric vehicles. Several experiments were conducted to determine under which conditions an electric car could become commercially attractive. However, instead of rethinking the vehicle’s architecture and engine in light of this energy source, PSA simply added cadmium-nickel then lithium-ion batteries to existing models. Thus, by the time Jacques Calvet left, nothing concrete had been achieved.
Laborious and hesitant commonalisation The refusal of conceptual innovation would have made sense had PSA been very committed to a ‘volume and diversity’ strategy in which the product policy targets customers whose needs are known and predictable, and who accept paying different prices for different models featuring a high proportion of shared components. The practical and symbolic needs of purchasers attracted by innovative models must be anticipated in so far as they are never formulated clearly. This customer base often requires a specific architecture and platform or else a highly adapted platform, all of which reduces the possibility of economies of scale (Boyer and Freyssenet, 2002). Commonalisation was the goal but costs were supposed to be minimised, which explains PSA’s frequent hesitations. The idea was less to build a multi-brand group that systematically uses shared platforms than to sustain two generalist carmakers, each of which would run a complete range
PSA: a ‘Volume and Diversity’ Profit Strategy 253
of passenger vehicles and small utility vehicles. Both sides would have very distinctive personalities and separate in-house distribution networks yet be able to benefit from the industrial, technological and financial synergies that would be possible in a group arrangement. The competition between the two carmakers was supposed to be an optimal lightning rod for increased competitiveness as well as a good way of ensuring that their models would not cannibalise one another. Peugeot and Citroën ranges were entirely renewed, rationalised and supplemented. At the end of the Calvet era in 1997, Peugeot and Citroën each had at least one model for each major segment in the passenger vehicle market (A-B, M1 and M2, H), as well as one model in each of the three light utility vehicle segments. The number of platforms (over the whole of the year) was cut from sixteen to ten between 1983 and 1997, excluding CKD kits and limited stocks of parts from old models that were sent to local producers over whom PSA had no control. The diversity grew from twenty models to twenty-four. The average volume per platform almost doubled to 204,000 from 117,000. The rate at which new models were launched (passenger and utility vehicles, excluding different body versions) accelerated from three to two years for each brand, and even to one year by end of the period. Despite these major developments, results were disappointing in comparison with PSA’s great rival Volkswagen, or even Fiat and Opel (Jetin, 1999). In part, this was due to (ostensibly) financial constraints. A second problem was the autonomy that the new organisation gave each brand to develop projects and manage plants. A series of unfortunate choices that can best be described as ‘unusual’ meant that the commonalisation policy did not produce the expected benefits: some model series were cut short, including ones competing in highly strategic segments; inversely, other models coexisted with their replacements in the European market for a period of several years; models positioned between two segments ran the risk of taking business away from existing customers; model denominations were confusing in terms of the series they belonged to; and launches involving two brands were not harmonised. For example, neither Peugeot nor Citroën had any new launches in 1988, 1990 and 1992. Surprisingly, the lifespan of certain low volume vehicles was extended; despite the official commonalisation doctrine, some of the models launched had their own platforms; inversely, other models launched were veritable clones and only differentiated because of their label. In addition, the Group staggered the commercialisation of models using the same platform. This meant a needless modification of the wheelbase on several occasions. Moreover, to satisfy customer expectations, the percentage of shared parts had to be lowered; several models were launched simultaneously on the same segments, notably top-of-the-range ones like the Citroën XM and Peugeot 605. Theoretically they shared the same platform but in reality far too many different components were needed.
254 The Second Automobile Revolution
The early 1990s were a real opportunity to launch a ‘volume and diversity’ strategy. Fiat suggested a merger to PSA, which refused, realising all too well the complexity of external growth between equals. At the same time, it was open to further cooperation in manufacturing models whose sales potential was too small for a single carmaker. Since 1978, Fiat and PSA’s Sevelsud joint venture in Val di Sangro (Italy) had been producing several light utility vehicles (Peugeot J5, Citroën C25, Fiat Ducato I) based on a shared platform. In 1992, a second joint venture was set up, Sevelnord, the purpose being to produce large monospaces (Peugeot 806, Citroën Evasion, Fiat Ulysse I and Lancia Zeta) using the same platform, as well as average sized light utility vehicles like the Peugeot Expert, Citroën Jumpy and Fiat Scudo. The plant, located near Valenciennes in France, was inaugurated in 1994. An inconclusive search for international volume The percentage of sales outside Western Europe only rose by three points between 1985 and 1997, from 13.7 to 16.7 per cent. There was an awareness that future growth in volumes could only come from non-Triad countries. PSA manifested its desire to penetrate the markets in some of these countries. Implementation was cautious, however, not without reason in certain cases (Argentina, Malaysia, Iran, India). In other countries, particularly China, the company tended to improvise. The outlook was good in China in the early 1990s. The Eighth Plan (1990–5) had opened up new partnership possibilities, albeit under strict conditions, including a requirement of higher integration rates, technology and know-how transfers, minority equity interests and planning authorisation for all investments. In addition to a rapid rise in its taxi fleet, the Chinese government decided in 1994 to allow private parties to own automobiles and in 1998 to take out car loans (Archambeau and Garcier, 2001: 93). PSA signed an agreement on Citroën’s behalf in 1992 with Dongfeng, a company whose main production up until that point had been trucks for the Chinese National Army. A Dongfeng Citroën Automobile Company (DCAC) joint venture was created, not only to assemble Citroën models in a plant at Wuhan (Hubei province) but also to produce mechanical parts at Xiangfang. An investment of a1.1 billion was needed, but with Dongfeng unable to find the 70 per cent in capital corresponding to its equity interest, Calvet got the French state to make up the difference. The first model launched in 1995 was the ZX and Citroën counted on this all-new model to steal a march on the VW Santana, corresponding to an old generation Passat. However, PSA’s second attempt to enter China, after a first joint-venture between Peugeot and Guangzhou Automobile Factory created in 1985, also ended in failure. The production of DCAC did not exceed 30/35,000 ZX. Many different factors were to blame. Firstly, PSA began to launch the five-door version of ZX with a rear hatch. The four-door version with a boot, preferred by the Chinese purchasers, was designed and industrialised
PSA: a ‘Volume and Diversity’ Profit Strategy 255
some years later. The local steel industry was unable to provide the sheet metal and cast iron that PSA needed. Connections between the two plants, located at a distance of 400 km from one another, were poor and expensive. There was no subcontractor network, meaning that parts had to be imported. Dongfeng had no commercial network to sell to private customers and lacked the corresponding know-how. Staff numbers were bloated and personnel management, initially run by the local labour union, i.e. by the Chinese Communist Party, had to be quickly taken over by PSA. The passenger vehicle market, which had started to trend upwards, unexpectedly began to stagnate. Regulations also started to change, notably concerning competition law and import duties. Disenchanted, a Citroën manager declared in late 1997, ‘we haven’t been selling advanced technology to the Chinese market but to the Chinese authorities’ (Archambeau and Garnier, 2001: 141).
Attempts to resolve PSA’s historical dilemma, 1998–2006 The new regime: brilliant results at first followed by stagnation After entering PSA in 1995, Jean-Martin Folz took over from Jacques Calvet on 1 October 1997. Folz had had the time to observe, consult thoroughly and analyse which activities he wanted to keep. As soon as he assumed power, he immediately broke with many of his predecessor’s strategic, organisational, social and environmental principles. From two carmakers lodged under umbrella holding, now there was supposed to be a single carmaker, PSA, featuring two brands, Peugeot and Citroën, each with a distinct identity (Gallard, 2004: 87). The two makes’ areas of competency were now restricted to styling and distribution alone, with the Group deciding upon everything else: platforms, mechanical parts, plants, launch planning, supplier relations, purchasing policy and human resources. Lasting profitability was to be mainly derived from maximum commonalisation: reduced number of platforms, increased number of ad hoc cooperative agreements with other carmakers and ultimately the development of a coherent chain from components to transportation. The idea was that subsequently, by innovating not only technically but also stylistically and conceptually, the Group would be responsive to an increasingly segmented market, thanks to a new type of platform that would create more space to manoeuvre. PSA would also try to spread beyond Western Europe with all its limitations and enter growth markets where people use and trade ‘European-type vehicles . . . thus optimising platform use’ (1999 Annual Report). The sum total of these measures was supposed to help PSA resolve its historical dilemma. On the other hand, the decision was made to maintain Calvet’s rejection of external growth as a complement to a ‘volume and diversity’ strategy. Whereas many of its competitors engaged in spectacular M&A operations in
256 The Second Automobile Revolution
1998–9, notably the Renault–Nissan alliance (see Chapter 14), PSA stressed the originality of its approach, for which Folz found a new justification. Now, the refusal of external growth was no longer a result of the painful experiences that had presided over the birth of the PSA Group: a lack of opportunities or a desire to maintain family control over the group. Instead, it was justified by external growth’s supposed lack of relevancy to an industry and a product characterised by increasing differentiation. This was based on the idea that mergers and acquisitions work when it is possible to rationalise very quickly so as to produce a standardised product. According to Folz, ‘Size for size’s sake is not the most intelligent response to the challenge of globalisation’ (Gallard, 2004: 98). This is like saying that a bachelor who has many lasting relationships is doing better than a married person. Folz benefited at first from circumstances that were exceptionally propitious for the changes he announced. He inherited a financially healthy situation along with models that would ultimately be very successful (the Peugeot 406 coupé, Peugeot 206 and Peugeot Partner and Citroën Berlingo twin models). Above all, the economy improved from mid-1997 onwards. Demand shot up, with total sales in 2000 being up 2 million in Western Europe and up 1 million in Central and Eastern Europe over 1990’s highs, and up 5 million and 1.7 million, respectively, versus 1994’s lows. The markets in Asia, Africa, the Middle East and Latin America, once victims of the ‘Asian crisis’, had renewed their rapid upwards trajectory. PSA’s European and global sales rose respectively by 36.9 per cent and 40.4 per cent between 1996 and 2001, reaching 2.4 and 2.8 million vehicles and stabilising around the 2.5 and 3.4 million mark. PSA’s market share in Europe went from 11.9 to 13.7 per cent between 1997 and 2003, inching ever closer to the leader, Volkswagen. With operating margins of 5 per cent, by early 2001 PSA’s four-year projections were more ambitious than ever. Of course, in 2001, 2002 and 2003, once the ‘new economy’ bubble had burst, new car sales plummeted in Western Europe, with 1 million fewer units sold compared to the 2000 high of 17.7 million. By 2004, sales had returned to this level but rose no further subsequently. In this context, carmakers all had to offer more for the same price to have any hope of keeping customers. Raw material prices also began to rise, as did the euro vs. the dollar and the pound. Under these economic conditions, PSA resisted relatively well in terms of sales, but operating margins fell steadily and were well below target. PSA’s worldwide output stayed steady and even rose slightly, with falling Western European volumes offset not by higher sales in Central Europe but instead by the take-off of Iran, Mercosur and China. Operating margins remained flat, however, falling from 5.4 per cent of turnover in 2002 to 2.0 per cent in 2006 at the Group level, and from 5.0 to 0.6 per cent at the automobile branch level. The same thing happened with net earnings, which went from 3.2 per cent in 2002 to 0.1 per cent in 2006. The question is whether PSA has
PSA: a ‘Volume and Diversity’ Profit Strategy 257
failed in its attempt to reconcile volume and innovation with the move into growth markets.
The need for a new ‘company governance compromise’ To complete its announced reorganisation, it was urgent that PSA first be reconciled with those ‘corporate governance’ stakeholders who had suffered under Jacques Calvet’s regime: labour unions, employees, politicians, Europeans and ecologists. To signify change and calm things down, Jean-Marin Folz immediately took two symbolic initiatives: he indemnified and reintegrated 169 union activists from the Sochaux plant, employees who had had an unusual career pathway; and he created a ‘carbon sink’ in Brazil by planting several thousand hectares of forest (10 million trees from thirty different native species). In June 1998, a labour law agreement was signed with all of PSA’s unions. Management said that it wanted to discuss and negotiate everything, and employees were offered a new compromise: higher staffing levels and shorter working hours in exchange for greater flexibility in terms of scheduling, status and remuneration. This compromise did nothing other than reproduce the principles underlying the 35-hour working week law enacted by the leftwing government that had come to power in 1997. PSA was the first major group to sign a 35-hour agreement. Jean-Martin Folz, like Renault’s Louis Schweitzer, quickly got the agreements signed in exchange for state-funded early retirement packages affecting 23,000 employees aged 57 years or more (and exceptionally, even 55 and 56) and the hiring of 8,000 young persons (Loubet, 2001: 479). Many union agreements covered issues like recruitment, gender balance, access for young people from disadvantaged neighbourhoods to positions of responsibility and job search schemes for people coming to the end of a fixed-term contract or for temporary staff members. In exchange, PSA was able to implement several labour schemes allowing it both to run plants to their full capacity and also to use internal mobility as a way of reorganising work, ensuring greater employee involvement and keeping the total wage bill under control. The number of persons working in the automobile branch rose between 1998 and 2004 from 90,700 to 100,400 in France, and from 120,600 to 139,300 worldwide. After 2005, however, things changed. Whereas Jacques Calvet was constantly denouncing European regulations – like the obligation to put catalytic converters on diesel engine cars – Folz tried from the very outset to make PSA into an environmentally friendly company. He started out in 1999 by targeting by 2008 an average CO2 emission rate of 140g/km for new vehicles sold in Europe. More concretely, he hastened the development of self-cleaning dust filters. The new HDi diesel engines reduced fuel consumption by 25 per cent and cut CO2 emissions substantially but had not resolved the problem of black soot generated by burning diesel. Folz found a solution with the particle emission filter. The German specialist press
258 The Second Automobile Revolution
and Green movement were strongly in favour of PSA’s new innovation and Group sales increased straight away in Germany.
A reduced number of platforms and the market heterogenisation Having disarmed the most pressing social and political problems, an active search for economies of scale could begin. In terms of mechanical parts, PSA got together with Ford in 1998 to design and manufacture a new generation of high pressure direct injection diesel engines. Since then, the two companies have become the world’s leading diesel engine producers. In 2002, PSA did the same with BMW but for small petrol engines. In 2005, 1,149,000 PSA engines were produced under cooperative arrangements. Plans are for the number to rise to 2,750,000 by 2008 (Frèrejean, 2006: 344). It was with a similar idea in mind that PSA decided to develop a supply subsidiary that would be sized to serve the global market, at the same time as many other automakers, like Ford and GM, were unloading any equity interest in the components industry. After taking over Bertrand Faure in 1998 and Sommer Allibert’s automobile activity in 2001, PSA’s Faurecia subsidiary became one of the world’s leading automotive suppliers. Note that PSA accounted for no more than one-third of total Faurecia sales. However, scale economies achieved by reducing the number of platforms were more limited than PSA had hoped. A four-year plan had envisaged a cut from seven platforms (or, according to our calculations, ten including the utility vehicle platforms) to three. The shared parts ratio, expressed in percentage of the total manufacturing cost, was to rise from 30 to 60 per cent. Technical systems built on several platforms were supposed to account for up to 30 per cent of all manufacturing costs. By year end 2002, one year behind schedule, PSA was running three platforms. Yet not all of the old platforms had been eliminated. Some coexisted with the newer ones, since their models still featured in the catalogue. They disappeared in 2006, but the same problem cropped up again. PSA’s habit of staggering its successive models’ positioning in their respective segments meant that several models would coexist for a few years, with each appealing to its own distinct customer base. Another problem related to platform definition and design. Increasing the flexibility of production lines and synchronising a larger number of sub-assemblies meant that it was now possible to assemble, on one and the same line, models that were based on the same platform, even as the platform wheelbase was, within certain limitations, being modified. In turn, this enabled a more intensive use of assembly lines, thus higher economies of scale across the production apparatus. The same did not apply, however, to the percentage of consumable parts, whose percentage had to be reduced, for reasons of roadworthiness, if the wheelbase width was too great. Certain models that were supposed to share the same 15–20 cm platform – the Xsara
PSA: a ‘Volume and Diversity’ Profit Strategy 259
and Xsara Picasso, C4/307 and C4 Picasso/Partner-Berlingo II, C2/1007 and C3/207, 407/C5 and C6 – were in fact very different. In the technicians’ opinions, they were based on distinct platforms. Lastly, several platforms were shared with other carmakers. An increasingly heterogeneous market led PSA to sign further cooperative agreements involving the manufacturing of new vehicle types that it felt incapable of making by itself: small city cars, small compact utility vehicles and SUVs. The first were made together with Toyota. An agreement was signed in 2001 covering a platform designed for three models: the Citroën C1, Peugeot 107 and Toyota Aygo. A plant was built at Kolin in the Czech Republic, with operations destined to start in 2005. In addition to renewing contracts covering bigger light utility vehicles and large monospaces, PSA and Fiat signed a new contract in 2005 covering the design and production in Turkey of three small compact utility vehicles models based on the same platform: the Citroën Nemo, Peugeot Bipper and Fiat Fiorino. Lastly, that same year, PSA concluded an exchange with Mitsubishi, swapping direct injection diesel engines equipped with dust filters for Mitsubishi SUVs, relabelled Peugeot (4007) or Citroën (C-Crosser). All in all, in 2006 and even before compact and normal SUV sales, according to our calculations PSA was not running three platforms (or six, including the utility vehicles) but fourteen, as many as it had operated in 1985! As a result, the average vehicle volume per platform was only 233,000. These figures clearly underestimate the economies of scale, since the percentage of components shared by each platform was undoubtedly higher than before and models sold to Toyota and Fiat were not included in the calculation. Nevertheless, the results were far below target. Conceptual innovations that were limited and incomplete PSA tried, as it had announced, to innovate not only technically and stylistically but also conceptually, using the three platforms it had planned. With 40 per cent of all components not being commonalised on models manufactured using shared platforms, the Group hoped to be able to offer innovative products that would respond to emerging market customers’ new expectations (Frèrejean, 2006: 332ff.). In the end, all it had was three limited and incomplete innovations, precisely because of this platform constraint. The innovation consisted in all three case of a hybrid of existing body versions. By year end 2000, Peugeot was selling a new version of the 206, one that was both a coupé and a convertible, called the CC, thanks to a hard roof that could be electrically retracted into the boot within twenty seconds. PSA ensured that this innovation, taken from a Mercedes luxury model called the SLK that dated from 1997, was financially accessible to small vehicle purchasers by limiting the surcharge to 25 per cent. 360,000 units were sold by late 2006, thus within six years, making the 206 CC the best-selling model
260 The Second Automobile Revolution
in its category in the world. It remains that this innovation still suffers from its inconvenient elimination of the car boot. The 206 CC is a ‘niche’ vehicle that has not really created a new market segment. PSA’s SW (i.e. station wagon) tried to offer some of the advantages of the monospace while maintaining the dynamic qualities of a saloon. The first model to have been equipped with this version in 2002 was the Peugeot 206, chosen by 366,000 clients between 2002 and 2005. This is a decent performance, although in the end the SW could not really compete with monospaces. The Citroën C3 Pluriel, launched in 2003, was a model that could be transformed into convertible, pickup and spider versions. The convertible was equipped with a canvas roof that folded electrically. The pickup version turned out to be practically unusable, however, with roll bars that were heavy, cumbersome and could not be stored on the vehicle itself. They were also difficult to install. All in all, this was a failure. It is therefore difficult to say whether PSA’s innovation/commonalisation dilemma was ever truly resolved. The Group was very active in copying other carmakers’ conceptual innovations that had already been validated by the market. It also innovated actively in areas like driving, equipment and active and passive security, as would be logical with a ‘volume and diversity’ strategy (Boyer and Freyssenet, 2002). PSA entered the average monospace segment with the Citroën Picasso; the small monospace segment with the Peugeot 107; the small city car segment with the twin models Peugeot 107 and Citroën C1; and the SUV segment with the C-Crosser and Peugeot 4007. At the same time, it failed with its top-ofthe-range Peugeot 607, which did less well than the 605 that was also no great success. Citroën only renewed it top-of-the-range XM model (1989–2000) six years later when it came out with the C6. In terms of clean engines, PSA had been working in two directions: diesel engines with improved environmental performance and electric cars. Folz decided to stop research and experiments on electric cars, despite PSA’s advance in this area. The immediate priority was to improve diesel and petrol engines. Aside from HDi engines and dust filters, PSA adapted a stop and start system where the engine cuts out when the car has stopped and starts up again when the driver simply presses on the pedal. This technology makes PSA the least polluting automaker in the world in terms of quantity of grammes of CO2 per km travelled, yet it does not enjoy as environmentally friendly an image as Toyota has been able to build based on its Prius alone. Launches disturbed by difficulties in implementing the new organisation These problems were not caused by the plants’ specialisation on a platform basis but by design issues. Peugeot and Citroën spent two and three years,
PSA: a ‘Volume and Diversity’ Profit Strategy 261
respectively, without launching any new models, aside from new body versions of previous models (Xsara coupé and an estate in 1998, with several models simply being restyled in 1999). There were other delays as well around this time, for at least two different reasons. The new organisational scheme adopted in April 1998 created a Project and a Business Management group and situated them at one and the same level, with the latter providing staff to the former. Conflicts were frequent, especially when model launches began accelerating. Disputes always had to be resolved at a higher level, and in the end the two management teams were merged in 2001. The Group also found itself having to hire many young engineers, who necessarily lacked experience. From 1999 to 2006, the number of engineers and managers working for Peugeot Citroën Automobiles in France rose from 11,774 to 17,210. Automobile R&D spending jumped from 3.9 per cent of turnover in 2000 to 5.2 per cent in 2003. Investments stabilised subsequently at a3 billion per annum – a level that it has been impossible to reduce since. Having plants specialised by platform and more flexibility in working hours meant that plant utilisation rates rose to more than 100 per cent (using the American definition of the annual capacity of a plant operating sixteen hours a day for 235 days a year or 3,760 hours annually). Suppliers – and not only PSA’s parts subsidiary – started to build plants or warehouses close to the Group’s terminal plants to enable synchronous just-in-time deliveries. A ‘convergence plan’ was implemented, with every site adopting ‘best practices’, receiving high performance tools and seeing an improvement in vehicles’ assembly possibilities and employees’ working conditions. An Elementary Production Units break-down was generalised to all sites. During this period, PSA created three plants and closed a fourth in Europe. In 2003, a new production plant for mechanical parts (gearboxes) was opened in Valenciennes. The joint PSA–Toyota plant at Kolin in the Czech Republic started production in 2005. In 2006, a plant dedicated to Segment A and B models began operations at Trnva in Slovakia, justified by the saturation of other sites, the need for an industrial presence in the new Central European markets and the advantageous production costs there. The same year saw the announced closure of the UK’s Ryton plant, which had previously produced the same models. This decision was a forerunner to recruitment policy changes that would become official after the departure of Jean-Martin Folz. Higher emerging markets sales did not improve operating margins immediately Sales outside Western Europe rose from 18.8 per cent in 2001 to 31.7 per cent in 2006. At first, this shift in the ‘geographic mix’ led to lower operating margins. Emerging market sales were not very lucrative at first and sometimes involved losses. It was only in 2007 that they would begin to contribute significantly to PSA’s operating margin (see Figure 13.1; Jetin, Chapter 3, this
262 The Second Automobile Revolution
volume). The costs of establishing an industrial and commercial infrastructure, adapting products, training workers and organising transportation had a very negative effect on margins. This initial period lasted longer than PSA had predicted. The emphasis was on those emerging markets (Mercosur, Central Europe, Iran and China) where the Group expected greater receptiveness to the kinds of vehicles that its two brands were offering. In 2000, PSA took full control of Sevel Argentina, its local assembler, which became Peugeot Citroën Argentina. A 70,000 vehicle/annum plant was inaugurated in 2001 at Porto Real near Rio de Janeiro (Brazil). These early efforts in Mercosur were difficult. The economies in this part of the world were undergoing a brutal recession following the ‘Asian crisis’ and the bursting of the ‘new economy’ bubble. Carmakers had begun to wage a price war. Local currency devaluations made imported goods more expensive. Even worse, PSA had neglected to envisage engines capable of the petrol and ethanol fuel mixture that Brazilians prefer. In the end, the models it offered got a poor reception and it was forced to adapt its engines and reorient its product range. PSA reinforced its presence in Iran. The Group had inherited the connections that Chrysler Europe’s British subsidiary had in this country providing a platform for the largest local carmaker, Iran Khodro. In 2001, Citroën signed an agreement to produce the Xantia with the country’s number two carmaker, SAIPA. Output rose but there was also a great deal of volatility due to political and economic uncertainty. Production hit 300,000 in 2004 before falling back to 250,000 in 2006. In November 2001, a new agreement was signed in China with Dongfeng involving the launch of six Citroën and Peugeot models over the next few years plus a twofold rise in capacities at the Wuhan plant. In 2003 and for the first time, PSA’s vehicle output exceeded the 100,000 mark in China. However, whereas the domestic market rose by 14.3 per cent in 2004 (from 4.2 million to 4.8 units), PSA’s Chinese production fell to 73,000, versus its sales target of 140,000 vehicles. The models that Citroën was turning out (not only the Xsara and the Xsara Picasso) were either too old or did not really satisfy local expectations. Moreover, no new Peugeot model launches were on the cards. This was a missed opportunity and when PSA finally realised that its models had to be adapted it launched a brand new version of the Peugeot 307 in 2004, plus the C-Triomphe in 2006. By so doing, it raised sales to 200,000 units by 2006 – but still fell from second to seventh place amongst all carmakers with a plant in China.
Conclusion and outlook As agreed, Jean-Martin Folz left PSA when he turned 60 on 6 February 2007, to be replaced by Christian Streiff, a former vice-president at Saint-Gobain. This event was followed by four major changes. The existing social compromise was turned upside down by an announcement that to increase
PSA: a ‘Volume and Diversity’ Profit Strategy 263
competitiveness, there would be staff cuts and a greater use of low wage countries. The product policy was reoriented towards growth market segments to the detriment of traditional saloons, whose sales had fallen off. There would be more launches involving innovative models and range extensions, using more flexible and modular platforms. In terms of cooperative agreements, PSA would no longer exclude any alliances or, indeed, mergers. The new imperative was a rapid increase in capacities and immediate entry into growth markets like Mercosur and China but also Russia, India and Mexico, zones from which it had been absent up until now. On 10 June 2008, for example, work started on a new plant at Kalouga in Russia. There are a number of continuities, however, notably the desire to build innovative models on shared platforms (despite the poor performance of this strategy so far). Efforts are also being made to maintain the Group’s leading position in clean engines. Group and automobile branch revenues were up again in 2007. Operating margins were up 2.9 per cent, and despite large extraordinary charges, net earnings were up 1.5 per cent. But it is difficult to see in these results a new departure, given the consequences of the credit crunch crisis. Effectively, PSA made financial losses in 2008. In a context marked by stagnating demand and fewer customers for traditional vehicles, a ‘volume and diversity’ strategy means that there is a constant need to find new sources for economies of scale. Three possibilities are: cooperative agreements or alliances with other carmakers; penetrating new market segments; and moving into growth markets. The example of PSA has shown that it is impossible to extend ad infinitum cooperative agreements with other carmakers for the joint production of mechanical parts and of vehicles whose sale potential is too small for a single carmaker. It is for this reason that PSA says that it is now open to the idea of an alliance. The question is with whom. Fiat would be the most logical partner, since the two carmakers pursue the same product policy and have long worked together on a cooperative basis. Fiat’s recovery has put the two potential allies back on equal footing, however, immediately raising the question of leadership. The same problem exists with BMW or Mercedes, assuming that these specialist carmakers would want to return to a strategy in which they manufacture all kinds of vehicles, despite the serious setbacks they suffered the last time they tried. PSA could acquire Volvo, which might help it to re-enter the top-of-the-range segment – that is, if Ford were to lower its asking price and no other bidders come in. An alliance with Mitsubishi, with whom PSA has started to cooperate, would have the advantage of opening new doors in Asia and the US. Unfortunately, the two companies have different product policies. Moreover, the Peugeot family is too prudent, probably rightly so, to risk getting together with one of the big three Americans, who have been spiralling downwards for years now – that is, unless a worst-case scenario hits General Motors and/or Ford and they are forced to sell off their European subsidiaries. What remains are the emerging carmakers from China
264 The Second Automobile Revolution
and India. These firms may have already established joint ventures with the historical big name carmakers in a bid to acquire the technology that they lack but their overt ambition is to become independent. They should be able to do this, given the outlook in their home markets and their governments’ policies. The two other possibilities for achieving economies of scale involve entering high growth national markets and new market segments. In both cases, however, experience has shown that economies of scale are harder to achieve than at first appears. In the first instance, not only does the company need to find countries attracted by the type of vehicles it is producing, but these vehicles must also be adapted to specific constraints, regulations and tastes. In the second instance, it has not been proven yet that it is possible to design platforms with sufficient flexibility and modularity to be used to manufacture truly conceptually innovative vehicles. PSA is now betting that its technique will allow it to resolve the first dilemma. It is also possible that clean vehicles will radically change the architecture of automobiles, hence the framework of PSA’s dilemma. Translation by Alan Sitkin
Appendix Table 13.1
PSA, 1976–2007
Year
Worldwide production
Group workforce
1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
1,498,694 1,518,141 1,633,987 2,052,079 1,703,638 1,515,381 1,459,754 1,470,077 1,562,329 1,629,796 1,724,215 1,904,936 2,129,178 2,219,821 2,234,911 2,066,928 2,021,474 1,744,145 1,986,381
176,500 184,500 190,170 263,000 245,000 218,000 206,000 203,000 187,500 176,800 165,000 160,600 158,100 159,100 159,100 156,800 150,800 143,900 139,800
Workforce in France
Group revenue (a)
Group net income (a)
145,800 135,000 130,300 126,300 126,700 127,700 127,100 122,650 116,800 112,100
5,345 6,385 7,288 10,828 10,839 11,035 11,473 12,989 13,889 15,289 15,998 18,013 21,105 23,316 24,387 24,416 23,694 22,834 26,053
218 191 211 274 −229 −304 −594 −411 −66 50 361 787 1,401 1,610 1,411 842 514 −191 506 (Continued)
PSA: a ‘Volume and Diversity’ Profit Strategy 265 Appendix Table 13.1
Continued
Year
Worldwide production
Group workforce
Workforce in France
Group revenue (a)
Group net income (a)
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
1,887,596 1,990,528 2,077,955 2,263,502 2,494,834 2,870,125 3,133,694 3,271,272 3,309,409 3,406,307 3,373,545 3,357,100 3,457,400
139,900 139,100 156,500 157,300 165,800 172,400 192,500 198,600 199,900 207,600 208,500 211,750 207,800
112,200 110,700 113,200 114,000 115,500 117,900 123,700 123,700 124,700 128,300 126,100 121,900 113,700
25,749 26,960 28,473 33,756 36,740 42,978 50,288 52,906 52,683 55,297 56,267 56,594 60,613
291 153 −381 494 774 1,325 1,681 1,690 1,497 1,357 998 70 826
Note: (a) IFRS standard since 2005. Sources: PSA Annual Reports; Jetin (1998).
Bibliography Archambeau, O. and Garcier, R. (2001) Une géographie de l’automobile. Paris: PUF. Beaux, S. and Pialloux, M. (1999) Sur la condition ouvrière. Enquêtes aux usines Peugeot de Sochaux. Paris: Fayard. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Broustail, J. and Greggio, R. (2000) Citroën. Essai sur 80 ans d’antistratégie. Paris: Vuibert. Calvet, J. (1998) La grande Faillite. Comment l’éviter. Paris: Plon. CCFA (every two years), Répertoire mondial des activités de production et d’assemblage des véhicules automobiles. Paris: CCFA. Célérier, S. (1994) ‘Gestion des contraintes et emplois. Les transformations d’une usine du groupe PSA’, Actes du GERPISA, 12 (December). Clot, Y., Rochex, J.-Y. and Schwartz, Y. (1990) Les caprices du flux. Paris: Éditions Matrice. Durand, J.-P. and Hatzfeld, N. (1999) ‘The Effectiveness of Tradition: Peugeot’s Sochaux factory’, in J.-P. Durand, P. Stewart and J. J. Castillo (eds), Teamwork in the Automobile Industry: Radical Change or Passing Fashion? Basingstoke: Macmillan. Durand, J.-P. and Hatzfeld, N. (2003) Living Labour: Life on the Line at Peugeot France. Basingstoke and New York: Palgrave Macmillan. Frèrejean, A. (2006) Les Peugeot: deux siècles d’aventure. Paris: Flammarion. Freyssenet, M. (1998) ‘Intersecting Trajectories and Model Changes’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Gallard, P. (2004) À l’assaut du monde. L’aventure Peugeot-Citroën. Paris: Bourin Éditeur. Harwit, E. (1995) China’s Automobile Industry: Policies, Problems and Prospects. New York and London: M. E. Sharpe.
266 The Second Automobile Revolution Hatzfeld, N. (2002) Les gens d’usine. 50 ans d’histoire à Peugeot-Sochaux. Paris: Éditions de l’Atelier. Jetin, B. (1999) ‘The Historical Evolution of Product Variety in the Auto Industry: An International Comparative Study’, in Y. Lung et al. (eds), Coping With Variety: Flexible Productive Systems for Product Variety in the Auto Industry. London: Ashgate. Loubet, J.-L. (1998) ‘Peugeot Meets Ford, Sloan and Toyota’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Loubet, J.-L. (1999) Citroën, Peugeot, Renault et les autres. Histoire de stratégies d’entreprise. Boulogne-Billancourt: ETAI. Loubet J.-L. (2001) Histoire de l’automobile française. Paris: Seuil. Loubet J.-L. (2003) ‘The Cautious and Progressive Internationalization of PSA Peugeot Citroën’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan. Satinet, C. (2008) ‘Citroën: une renaissance par le design’, École de Paris du management, 16 April, compte-rendu, http://www.ecole.org.
14
Renault, 1992–2007: Globalisation and Strategic Uncertainties Michel Freyssenet
After a period when everything seemed to come up trumps, Renault has fallen on hard times again. Since the 1980s, the company has run into serious difficulties once every decade: 1983–6, 1992–7 and 2005 to the present. Hard times have sometimes been followed by a spectacular recovery, as when Renault first set up its alliance with Nissan, or made Dacia into the world’s first truly ‘low-cost’ automobile maker, or penetrated the Korean market by taking over Samsung. Of course, there were dissimilarities between these three times of trouble. The first was caused by a crisis endangering Renault’s very financial survival. The second resulted from a performance crisis masking the decision it had made to pursue highly profitable product policies in the favourable context of the late 1990s. The gains from this move allowed Renault to ‘globalise’ in one fell swoop. Lastly, the problems encountered since 2005 seem to be rooted in a crisis of coherency that, although it may not yet be apprehended in this light, could ultimately turn into something more serious than mere ‘turbulence’.
1992–1997: from ‘quality’ to ‘conceptual innovation’ Wrong-footing the ‘quality’ strategy Having reduced debt levels by divesting a host of assets, Renault’s late 1980s recovery came from its adoption of a ‘quality’ strategy. Unable to continue counting on high volumes and relative diversity, it had to seek bigger margins on each vehicle sale by attracting customers through ‘quality’ and convincing them to pay higher prices. Hence the company’s decision to reorient commercial policy towards Northern European markets that were more apt to pay the price of ‘quality’. Quite logically, this culminated in a 1990 agreement with Volvo that lent credibility to the new orientation. Renault maintained its slogan of ‘Cars for Living’ to provide a modicum of differentiation. It was trying to respond to the rising demand for ‘user-friendly’ cars that would be a pleasure for passengers, drivers and other road users alike. The ‘Cars for 267
268 The Second Automobile Revolution
Living’ initiative was echoed in the ‘Agreements for Living’ conventions that Renault signed with almost all of its trade unions in 1989 – the one notable exception being France’s largest union, the CGT. In fact, the expression ‘Cars for Living’ came to be associated with an industrial dispute that the CGT Metalworkers’ Federation had been waging since 1986 after an announcement that Renault’s historical Billancourt plant would be closed and union activists fired. The end came in 1992 when the French judiciary confirmed the redundancy programme and the definitive closure of the Billancourt site, which all through the twentieth century had been a holy ground for industrial disputes in France. The ‘Agreements for Living’ gave birth, inter alia, to Elementary Work Units (EWU) that codified and spread throughout Renault the idea of ‘group work’, seen as a stepping stone to ‘total quality’ (Freyssenet, 1998). This laid the foundation for a huge human resource reclassification – following an earlier decision made under union pressure – in which ‘unskilled workers’ became ‘professional manufacturing workers’. Workers’ new competences were restricted to a few workstations, however, involving a handful of simple ancillary control and maintenance tasks. Renault seemed to have adopted the right strategy. The new models it launched at the time (the Renault 19 in 1988, Clio in 1990, Safrane in 1992 and Twingo and Laguna in 1993, see Figure 14.1) sold well or at least well enough. The cooperation with Volvo progressed so quickly that people immediately began calling this an alliance, and thought that a merger was
2,500,000 Espace
2,000,000
Safrane
R30
1,500,000
“niches”
1,000,000
R16
500,000
e auphin
D
4cv
R14 R12
Mégane Logan
R19 R9-R11
Kango
Clio
R10
R8-
Frégate Juva
R25 R21
R5
R6 R4
Twingo
Modus
19 1945 1947 1949 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 2099 2001 2004 06
0
R18 R20
Laguna
Year Figure 14.1
Global production of Renault brand passenger cars, by model, 1945–2007
Note: The models are classified by market segment: lower; lower midrange; upper midrange; upper; and ‘niche’. Source: Annual Reports of Renault.
Renault, 1992–2007
269
imminent. Renault benefited from this turn of events, mainly because of its quality efforts and due to the boom in automobile demand accompanying the era’s property bubble. The rapid enrichment of certain social categories translated, in the automobile sector, into increased demand for the best equipped and most stylised versions of the models on offer. However, given the restrictive budgetary policies that European governments felt obliged to pursue when the financial bubble burst and the considerable cost of German reunification, Western European automobile demand plummeted by 2.6 million vehicles in 1993, with Renault alone experiencing a fall of 300,000 units. Once again, customers were becoming increasingly price-sensitive. The problems in Europe and, conversely, Volvo’s resurgence thanks to renewed demand in the United States (its main market), reinforced the growing defiance of Volvo shareholders and some of its executives, exasperated by Renault and the French government’s attitudes. The merger announced on 6 September 1993 was rejected four months later at the annual shareholders meeting. These events raised doubts about the profit strategy that the company was pursuing. ‘Quality isn’t enough . . .’ Europe’s shift towards a more ‘competitive’ mode of income distribution had led to the emergence, alongside employees’ traditionally hierarchised desire for stable incomes, of a demand from those segments of the population that had benefited from the new forms of wages and incomes and who aspired to vehicles whose utilisation differed in practical and symbolic terms. Shortly after taking over Renault in 1992, Louis Schweitzer announced a clear change in strategy, with a new emphasis on innovation. Reassured by the proven and lasting success of the Espace, Europe’s first minivan, and of the Twingo, a small, remarkably comfortable non-conformist vehicle, Renault decided to be more proactively innovative, starting with its lower mid-range. The Mégane family was innovative in two ways: in terms of the number of body versions on offer (six); and because one of these versions – the Scénic minivan – was totally original in nature. The new orientation did not meet with unanimous approval within Renault, which had not yet become aware of the organisational, social and economic implications of conceptual innovation. To be profitable, the carmaker would have to show extreme responsiveness to both success and failure. When things were going well, it would have to benefit from the innovation rent for as long as possible. This could be achieved by saturating the demand. If things went poorly, it would have to minimise losses (Boyer and Freyssenet, 2002). Some choices reflected this thinking (notably the lower integration rate and more flexible working times) but others were based on a different logic. One example was the Technocentre in Guyancourt, where building began in 1993. The idea here was to get product and process centres – along with suppliers – to work on a ‘concurrent engineering’ basis
270 The Second Automobile Revolution
(Hirt, 2004). The centre had been designed according to a ‘Sloanian’ matrix scheme pursuing a ‘volume and diversity’ strategy based on a commonalisation of the vehicle’s invisible parts and a differentiation of its visible parts (Boyer and Freyssenet, 2002). This was preferred to an ‘innovation and flexibility’ strategy that was more geared towards an individualisation of models and the creation of integrated design teams. ‘Finding growth where it lives’ . . . as long as the financial and legal circumstances are right In the early 1990s, many observers viewed commercial and productive ‘globalisation’ as a new imperative. Developed country markets plummeted after the first speculative bubble burst, making it seem as if automobile demand in this part of the world might be definitively limited. In parallel, growth patterns in many so-called ‘emerging’ countries intimated that their economic lift-off was a done deal. More and more companies began setting up facilities abroad. Even as it continued to Europeanise its production apparatus and consolidate market share, Renault decided to create new bridgeheads outside Europe. In 1996, it started building an assembly plant in Brazil, and in 1997 assumed full control of its Argentine subsidiary. It also began to examine the possibility of an alliance with a carmaker in Asia. First, however, Renault needed to settle the question of its status. The failure of the merger with Volvo had convinced many stakeholders of the need for a real privatisation (Schweitzer, 2007). The French state sold some of its equity, reducing its capital interest to 46 per cent in 1996. Uproar at Vilvorde While awaiting the launch of innovative models and their potentially positive effect on sales, the financial situation took a turn for the worse. The French government did not want to abandon its restrictive budgetary policy and limited its interventions to ad hoc measures supporting automobile demand. These measures did not prevent carmakers from waging a price war, however, followed in 1997 by a spectacular fall in automobile sales. Renault’s financial performance worsened (Figure 14.2) despite the closure of small assembly plants like Creil in France and Setubal in Portugal. Management concluded that Renault’s vehicles were too expensive. Louis Schweitzer hired Carlos Ghosn, who had already shown his costcutting prowess at Michelin. Having become Renault’s second in command by autumn 1996, Ghosn set a target of reducing, before year end 1997, the return cost of all existing and future models by 3,000 francs (c. a457) per vehicle. He convinced Renault’s executive to close a third assembly plant, Belgium’s Vilvorde factory, whose 3,100 employees were considered Renault’s ‘good pupils’, having efficiently and uncomplainingly applied all previous
Renault, 1992–2007
271
15 10
Gross income/turnover
5
19 7 19 6 7 19 7 7 19 8 7 19 9 8 19 0 8 19 1 8 19 2 8 19 3 8 19 4 8 19 5 8 19 6 8 19 7 8 19 8 8 19 9 9 19 0 9 19 1 9 19 2 9 19 3 9 19 4 9 19 5 9 19 6 9 19 7 9 19 8 9 20 9 0 20 0 0 20 1 0 20 2 0 20 3 0 20 4 0 20 5 0 20 6 07
0 ⫺5 Net income/turnover ⫺10 ⫺15
Annêe
Figure 14.2 Financial results of Renault Group, 1976–2007, gross and net income/ turnover × 100 Source: Annual Reports of Renault.
corporate performance improvement plans. The announcement of the closure was therefore a political and social shock and triggered the first ever ‘Euro-strike’. Widely criticised, Renault’s management team refused to budge with regards to the imminent closure, although it did agree to improve its post-redundancy reclassification package and to adopt new European-wide negotiating procedures (Charron, 2004). Vilvorde’s closure was part of a plan aimed at specialising factories in a single model assembled on a single line, the one exception being Sandouville, which was to be dedicated to three top-ofthe-range models. Renault had not foreseen that this productive organisation might fall foul of its innovative model policy, whose profitability required a productive apparatus that could be converted quickly, depending on whether a certain production line was a success or a failure.
A time for audacity, 1998–2003 The success of innovative models in a rapid growth European market, together with the ‘Asian crisis’, gave Renault an opportunity to ‘globalise’ in one fell swoop. Heavenly surprise: strong recovery of the European market and successful innovative models . . . yet the right lessons are not learnt Focused on cost-cutting, Renault was surprised by the sharp rise in European automobile demand, caused by the third speculative bubble (rooted in the
272 The Second Automobile Revolution
so-called ‘new economy’) and especially by the success of innovative models like the Scénic, Kangoo and second-generation Espace. Launched in October 1996, Scénic sales skyrocketed, accounting for 44 per cent of all Méganes produced in 1997. Afraid that this could not last and still at war in Vilvorde, Renault opted to considerably lengthen the Scénic’s delivery times and raised throughput rates at the Douai plant production line that was dedicated to this model, instead of immediately adapting a second production line – a job that took until 1998–9 to complete. These hesitations gave Renault’s rivals time to launch their own minivan models. Squeezed by the Opel Zafira and Citroën Picasso, the Scénic topped out with 6 per cent of the European lower mid-range market, versus 8 per cent for its two rivals combined. In short, the Scénic was still leader but the others were catching up. This is a textbook example demonstrating that a ‘quality, cost, delays’ strategy is not the be all and end all of profitability. The general enthusiasm for the Scénic – which was, after all, the Mégane family’s most expensive version, the one whose deliveries took the longest and which offered the lowest quality – highlights the fact that many customers prefer an automobile offering the practical and symbolic uses to which they aspire. The third generation of the Espace, also sold in October 1996 and followed the next year by a seven-seat version, did even better than its predecessors, with annual volumes ranging from 60,000 to 70,000 units. Lastly, a new and conceptually innovative model was launched in October 1997, the Kangoo ‘Ludospace’, a sporty passenger vehicle version of the Kangoo Express utility vehicle. More than 100,000 Kangoos sold quickly in 1998, which was better than the utility vehicle version. Adding to these three models the tireless Twingo – an inimitable car with few if any modifications – Renault’s innovative models sold about 800,000 units in 2000, accounting for 37.7 per cent of the make’s global passenger vehicle production. Profits skyrocketed. For the first time, Renault exceeded the 2.5 million vehicle mark in 2000. Having reimbursed all its debts, it had the financial resource to undertake a number of initiatives (Figure 14.2). The ‘Asian crisis’ provided a window of opportunity.
The ‘Asian crisis’, viewed as an opportunity to achieve global scale quickly: Nissan, Dacia, Samsung The emerging country boom of the 1990s turned out to be largely speculative in nature. By mid-1997, investors and lenders had taken fright at the accumulation of credit. Starting in Thailand, the crisis steadily spread, like dominos, to a number of countries in Southeast Asia and Korea, then on to Latin America before finally reaching Russia. It plunged Japan back into the economic quagmire that had beset this country ever since the first financial bubble. The global automobile market’s recovery, propelled by emerging country demand, was postponed.
Renault, 1992–2007
273
At the same time, the ‘Asian crisis’ was an opportunity that Renault could and did seize. The company acquired Nissan in March 1999, the Romanian carmaker Dacia in July 1999 and the Korean Samsung in September 2000, even as it continued to refocus completely on passenger vehicles and small utility vehicles. In January 2001, Renault sold its truck subsidiaries, RVI and Mack Trucks, to Volvo Global Trucks, receiving in exchange 20 per cent of Volvo’s capital and voting rights and becoming the Swedish carmaker’s main shareholder, without assuming full responsibility for its operations. Combined, Renault–Nissan accounted for 9.2 per cent of the world automobile market in 2000, becoming the branch’s fifth largest group. Renault set a sales target of 4 million vehicles by 2010, including 50 per cent outside Western Europe. In two years, Renault had spent FFR 36.3 billion (a5.54 billion) taking equity stakes. The Renault–Nissan alliance: savings, complementarities and transferring ‘best practices’ Portrayed as an alliance, Renault’s takeover of Nissan would only be credible if Renault could rebuild Nissan right away. Louis Schweitzer delegated this task to Carlos Ghosn. 18 October 1999 saw the launch of a three-year Nissan Revival Plan, covering the period 2000–2. Its objectives were reached one year ahead of schedule, not only due to the measures’ suitability and the speed with which they were implemented, but also because the environment was particularly conducive. Nissan’s management had lost legitimacy following two earlier failed recovery plans, and above all because the company was almost bankrupt. On top of this favourable local context, there was the American market, which experienced renewed growth in 1999, before stabilising at higher levels in the following years. Note that a large proportion of Nissan’s sales had been – and were still – in the United States. A second three-year plan, called the ‘180’, was launched straight away. Most significantly, the two carmakers decided to accelerate the group’s construction along bi-national lines, with each partner keeping its own identity and operational autonomy but sharing one and the same strategic vision and pragmatism with the other. Three precautions were taken, however: each carmaker took a financial stake in the other’s performance by reinforced crossshareholdings; takeover attempts by third parties were discouraged through the creation of a shared foundation incorporated under Dutch law (Renault, whose own shareholder structure was highly dispersed, was particularly afraid of a hostile takeover); and it was stipulated that henceforth any decisions would have to benefit both parties. In 2002, Renault raised its stake in Nissan from 36.8 per cent to 44.4 per cent. Nissan acquired 15 per cent of Renault, without receiving – as per French regulations in this area – the associated voting rights. The French state’s share fell below 16 per cent and has not changed much since. At the same time, the two carmakers created a 50/50 strategic
274 The Second Automobile Revolution
management company, Renault–Nissan bv, a light structure domiciled in the Netherlands and tied to the aforementioned foundation. Its function has been to define strategy for the alliance. Immediately following the 1999 agreement, joint Renault–Nissan workgroups began to determine, at the behest of the Global Alliance Committee, which synergies, complementarities and savings the two carmakers could achieve in functions like purchasing, products, mechanical systems, vehicle engineering, manufacturing, production sites and sales networks. It was notably decided to reduce to ten by 2010 the number of platforms operated by the two makes, while reducing to eight the number of engines and to seven the number of gearboxes. The construction of the first shared platform, the B, started right away (Segrestin, 2005), as did work on a small direct injection diesel engine called ‘common rail’. Both were meant for Renault and Nissan’s 1.7 million lower range models. In 2001, work started on a second platform, the C, aimed at lower mid-range vehicles and representing a potential market of 2 million vehicles. In time, these two platforms would account for more than half of Renault and Nissan’s volumes. In April 2001, the two carmakers created a 50/50 subsidiary called RNPO (Renault Nissan Purchasing Organisation), with a mission of raising the percentage of shared purchases to 70 per cent in a bid to improve the group’s bargaining strength and obtain better quality. In 2006, RNPO was responsible for 75 per cent of the alliance’s global purchases. The two partners relied on one another to accelerate their international commercial deployment and optimise plants’ throughput rates by playing on their geographic complementarities and industrial logic. Knowledge transfers may have been less quantifiable but they were also significant, at least according to the two partners. Renault said it benefited from Nissan’s expertise in areas like industrial processes (quality, productivity, problemsolving) and logistics (delays). Inversely, Nissan benefited from Renault’s expertise in workstation ergonomics, cost control, marketing, design, platform strategy and sales finance. One domain that produced more quantifiable outcomes was Renault’s adoption of Nissan’s ‘serial-lots’ vehicle development method, which consisted of replacing a series of physically synthesised prototypes with digital simulations and partial prototypes. Many time savings were achieved, for example, the ‘entry ticket’ for the Clio III was cut to a953 million, including a630 million for industrial investments. Nissan helped Renault to design a sporty 4 × 4 that Renault Samsung would produce. In Renault’s opinion, significant progress was being made in the field of quality when Nissan’s expertise in this domain was integrated into the Renault Production System. All in all, however, it would seem that the main cost-cutting attributed to the alliance lay in its ability to get suppliers to sharply lower their purchasing
Renault, 1992–2007
275
prices; and in its development of a shared pool of mechanical systems like engines, gearboxes and transmissions.
Start-ups in Brazil and Russia, return to Romania and entry in South Korea In addition to the work done as part of the alliance, Renault developed into a multi-brand group with its own international strategy. Whereas Renault experienced disappointing performance in Brazil and Russia, things went much better in Romania and Korea. The Curitiba plant in Brazil, opened in the middle of the ‘Asian crisis’, was producing models that were not very adapted to the local market and suffered from quality and sourcing problems. Unable to break through the 70,000 vehicle threshold, it turned out to have major over-capacities. In 1998, Renault founded a company in Russia named Avtoframos. Half owned by Renault and half by the Moscow mayor’s office, Avtoframos provided access to facilities operated by Moskvitch, a Russian carmaker. The repercussions of the ‘Asian crisis’ and the political chaos in Russia at the time persuaded Renault to wait before committing. In 1998, Romania’s State Treasury chose Renault to take over the national carmaker, Dacia. Plans for a $6,000 car targeting ‘emerging’ countries were supposed to be implemented. Ultimately, however, Renault’s forecasts were toned down to an annual output of 200,000 vehicles, including 80,000 in the export markets by the year 2010. The five years that had been deemed necessary to design and launch a low-cost car, the future Logan, were used to resize and update production tools and methods, and to ensure proper staffing levels. A fabric of suppliers was also put together. An entity called CESAR (Romanian Automobile R&D Centre), employing 700 engineers and technicians, was given responsibility for developing products, prototyping and trials. Most notably, CESAR was from the very outset organised as a ‘project’ conducted on a concurrent basis. Existing models started to improve right away, and Dacia’s image changed dramatically with the launch of the Solenza, a model that enabled the company to reconquer its domestic market, build a sales and after-sales network and ultimately improve its finances. In reality, the Solenza was a ‘school-project’ that would pave the way for the Logan (Angelescu, 2007). Renault had to cover five years of losses at Dacia – 2005 was the first year that the company would turn a profit (a57 million). Ultimate success would depend on the Logan programme. In 2000, Renault acquired the South Korean carmaker, Samsung Motors (Jun, 2002). First year sales were around 70,000 units of the one model on offer. The second year turned a profit and total production hit 117,000 cars. These early performances convinced Renault to plan for 500,000 units in 2010, half as exports. Renault–Samsung Motors (RSM) was viewed at the time
276 The Second Automobile Revolution
as a decisive element in Renault’s international deployment in Asia. However, the Korean market shrivelled in 2003 and 2004, with RSM production falling 80,000. Renault had to drop its 2010 objectives to 300,000 vehicles, including half overseas. By late 2004, Renault had already been investing significant amounts in its own international development for more than six years, without any great benefit. It is also true that these new entities, and the alliance with Nissan, gave it a larger commercial presence in many markets and offset poor performance in Western Europe.
‘Turbulences’ in 2001: three new models failed but the company remained profitable thanks to the Scénic II . . . and Nissan When the ‘new economy’ bubble burst in 2001, the automobile market stagnated in the developed countries and completely collapsed in Central and Eastern Europe, Turkey and Argentina. Renault only had one new model to offer: the second-generation Laguna, positioned in the upper mid-segment. Operating margins plummeted that year, with a break-even point almost identical to the company’s value added threshold (Figure 14.2). The Laguna II barely sold any better than the Laguna I, already deemed insufficient. Even worse, after 2003 they plummeted. 2002 saw the marketing of three new models, all of them top-of-therange cars portrayed as innovative models. Instead of copying classic German sedans, Renault tried to stand out by appealing to a customer base seeking freedom from the stylistic standards that had characterised this segment. The first model, called the Avantime, was a hybrid coupé-minivan, targeting Espace customers whose children had left home and who were therefore looking for an anti-conformist sporty vehicle. The second model, the Vel Satis, a four-door sedan, targeted customers who ‘prefer Apple to Dell and Bang & Olufsen to Daewoo’. The third model was a fourth-generation Espace seven-seater. Despite being destined for very different practical and symbolic utilisations, the Laguna II, Vel Satis and Espace IV were designed using the same platform. Whereas the Espace IV sold as well as earlier generations had done and remained leader in its segment, the Avantime and Vel Satis failed badly, pulverising Renault’s worst scores ever. The Avantime was shut down after a mere 8,158 units were produced in two years. The Vel Satis was maintained, but within five years (by late 2006) it had only sold 56,137 units, or five times less than its objective. Sales would in fact have been even worse had there been less discounting. These models’ story has not been written yet. They all experienced quality problems during their launch phase, weighing on subsequent sales. Moreover, their lower purchasing prices meant that lower quality supplies were sourced for them. It seems that the initial Vel Satis project was progressively reconfigured in an effort to operate within the constraints of a
Renault, 1992–2007
277
shared platform. Conceptual innovation should constitute more than a few episodes of stylistic audacity. Despite the complete failure of the Avantime and Vel Satis and the relative failure of the Laguna, operating margins had returned to 4–5 per cent within three years, with net earnings reaching 6–7 per cent of revenues. The explanation is that the Mégane and Kangoo second generations did better than their predecessors, particularly the former, thanks to the Scénic II. In 2004, 906,876 Méganes were produced, more than half of which were Scénics. Nissan was starting to make an increasingly important contribution.
More flexible work, cost-cutting, plant specialisation, outsourcing, asset sales: a patchwork of actions that were not always coherent Hesitations between a policy of innovative models and a policy of classic models were reflected in contradictory decisions taken in areas like productive organisation and the employment relationship. Some changes, like increasingly flexible work, were coherent with a strategy of conceptual innovation. Others were less cogent, or even not at all, such as having hyper-specialised factories. Increased work flexibility was achieved in different ways. After 1999, the percentage of external workers (temporary and contractors) reached about a quarter of all persons employed on Renault premises, not including supplier representatives, notably those working at the Technocentre. A multi-annual calculation of working times, negotiated with a French legislative framework cutting weekly working hours to thirty-five, also enabled higher output in factories producing Scénics and Kangoos, as well as a sharp fall in working hours at plants assembling more commercially disappointing models. Staff numbers were stabilised and new industrial disputes avoided. The replacement of older employees by younger ones, thanks to a mechanism negotiated with the French state, contributed not only to lower labour costs but also to greater flexibility. Lastly, production and service externalisation reached its highest level ever, equalling 90 per cent of Renault holding company revenues and 77 per cent of total Renault group sales (outside the alliance). Purchasing revolved around sub-assemblies, including some that were delivered synchronously thanks to the generalisation of ‘supplier parks’ adjoining the factories. Still, it was in complete contradiction with the need to have a flexible and responsive production tool when producing conceptually innovative vehicles that the decision was made at the same time – in the name of economies of scale – to specialise factories in one model (and to have one platform for the top-of-the-range) whilst reducing the number of assembly lines to one per factory. In all likelihood, it was also due to the two vehicles types’ contradictory commercialisation requirements that the 1999 ‘new distribution’ programme never achieved its targets. This programme was supposed
278 The Second Automobile Revolution
to reduce inventory levels by ensuring that 70 per cent of all vehicles could be produced on command. In the end, an estimated 30–50 per cent were manufactured in this way. On the other hand, the Technocentre, which had been inaugurated in May 1998, and the use of new design techniques made it possible to run four to five projects simultaneously, causing a sharp fall in model design and development times. The Mégane II, delivered in 2002–3, was designed in 29 months, versus 45 for the Mégane I. The Modus, delivered in 2004, was designed in 29 months, against 50 for the Clio II. All in all, Renault was thought to have cut costs by a6 billion over two successive plans (1997–2000 and 2001–3). Fifty-five per cent of these savings were obtained in the purchasing function, 20 per cent in distribution, 10 per cent in production and 15 per cent in design and overheads. What wasn’t quantified, however, was the opportunity cost of Renault’s insufficient responsiveness to the demand for its innovative models, be they brilliant successes or terrible failures.
The return of certain unresolved problems: 2005 to the present. New ‘turbulences’ or symptoms of an uncertain product strategy? A change of era? Carlos Ghosn, Louis Schweitzer’s official successor, took over in 2005, becoming president of Renault and then of Renault–Nissan, whilst remaining president of Nissan. Renault had no debt and cash flow was satisfactory. The firm’s image had completely changed, there seemed to be no social tensions and employees were motivated. Yet everyone clearly felt that a number of weaknesses and problems remained. Firstly, there was the fact that improved results were only due to the good performance of Nissan and Volvo Trucks. Operating margins were down. Sales were down in Western Europe, which was supposed to be the group’s safety net, given the risks associated with expansion in emerging countries. Global sales were also down. There were many arguments with suppliers and quality could be sub-standard. Launch schedules were sometimes missed. Carlos Ghosn announced the long awaited ‘Contract 2009’ on 9 February 2006. His enthusiasm may have been reassuring, but the plan itself was not entirely convincing. Three commitments were made: the future Laguna, scheduled for 2007, would rank amongst the top three in the world in its product and service quality segment; operating margins would hit 6 per cent by 2009; and sales would rise by 800,000 vehicles between 2005 and 2009. To do this, twenty-six products would be launched during the plan, twice as many as over 1998–2005. Cost cutting would continue and the average production capacity throughput rate would rise from 60 per cent to 75 per cent.
Renault, 1992–2007
279
It remains that all these good efforts were unable to dissipate the concerns caused by a relatively unclear product policy. New ‘turbulence’ hitting the Renault brand and its method of management The Laguna II’s unwaveringly rapid decline translated into over-capacities at the Sandouville plant, dedicated to top-of-the-range production. Mégane sales began to drop off sooner than expected, due to the poor sales of the fivedoor sedan and the coupé. Renault wanted to count on the Modus, a new and innovative minivan model designed using the alliance platform B. Launched in September 2004, it started very strongly but fell off after one year, far from its sales targets. There were also over-capacities at Spain’s Valladolid plant, entirely dedicated to this one model – as per the principle of ‘one model–one factory–one line’ – and this led to a cut in working hours and staff numbers. Carlos Ghosn concluded that Renault needed to design and produce ‘cars that sell themselves’. He delayed by one year the launch of the new Twingo and Laguna III to review the first car’s styling and sales objectives, and to increase the latter’s quality and performance levels. Clearly, Ghosn was running the risk that, despite the successful September 2005 launch of the Clio III in the market’s lower segment, further ‘turbulence’ could happen in the meantime. The challenge under these circumstances was to avoid unduly compromising financial performance at a time when the company was publicly saying that profitability had become its be all and end all. Having decided that the answer was to embrace ‘turbulence’ by preserving margins rather than market share, Renault stressed sales to individuals and companies instead of less lucrative sales to rental companies. It also abandoned discounts that caused lower resale prices and made it hard to detect what customers really wanted. The end result was lower sales month after month. This dismayed observers, but from the company’s perspective, at least the operating margins were still positive. So what did it mean, ‘cars that sell themselves’? Why was conceptual innovation no longer a priority? The many models launched since mid-2007 are all characterised by a change in direction, putting greater pressure on Technocentre engineers and technicians. The new Twingo no longer has its own platform but was designed on the one used for the Clio II. It is a more mundane vehicle, the aim being to expand its target audience by removing everything that once made this model an emblem of Renault’s spirit of innovation. The Laguna III is still a conventional model, despite its more elitist style and finishing. The New Modus is a redesigned Modus, and the Grand Modus is the same with an extended wheelbase. The new Kangoo has improved on preceding versions’ performance. In addition, a sporty 4 × 4 called Koleos, engineered by Renault, developed by Nissan and manufactured by Renault–Samsung, was expected in 2008. Its goal is to combine the advantages of sedans, minivans and 4 × 4s, but it will be in direct competition with
280 The Second Automobile Revolution
models that are already famous. Renault has said that in the future, it would try to focus its products more on customers’ needs and expectations, wherever they are in the world, so they can benefit from the alliance’s ‘continuous technological efforts’. For the moment, once again it is one of the group’s more innovative models, decided and designed at an earlier date without any marketing or clinical studies, that offers Renault its only chance to achieve its volume objectives. The Logan is already part of automobile folklore. The Logan revelation Renault’s ‘a5,000’ model is the epitome of conceptual innovation. Lacking any mechanical or stylistic novelty, it has revealed a blind spot in carmakers’ vision of the market (Jullien, 2006): people who want to buy a new car but who cannot do it – or who can no longer do it – due to insufficient income. During a visit to Russia, Louis Schweitzer first came up with the idea of offering a true low-cost family car to the new middle classes taking shape in the emerging countries, people who up until that point had been condemned to buying imported used cars that were costly to run and sometimes expensive to acquire (Schweitzer, 2007). To meet the target price, design and manufacturing processes had to be rethought and a specific platform engineered. The vehicle’s twenty-four functions were conceived of in a way that responded to the dual requirement of a particular target cost along with just-in-time delivery. There were fewer components and the ones used were simplified, standardised and produced in low-cost countries. Assembly was done without recourse to robots or automatons. Commercialisation costs were divided by a factor of five. Launched in October 2004 by Dacia, the Logan was an immediate success in Romania and neighbouring ‘emerging’ markets (Eastern Europe, the Middle East and North Africa). The biggest surprise was to discover that there was also a demand for this type of vehicle in Western Europe. Renault and its rivals found out that the stagnation of this regional market was partially due to the fact that a growing section of the population had not been able to buy new cars due to low purchasing power or unstable income. Thus, sales were well above expectations eight months after launch, i.e. the Logan was already in the black. It was then decided to increase the number of versions planned, so as to respond to the vehicle’s various potential uses and to adapt to its various markets. For example, 2007 saw the launch of a station wagon/estate (the Logan MCV), a Logan utility van and a five-door sedan called the Sandero, designed first and foremost for the South American market and featuring a less austere body as well as a flexi-fuel driving system. In the future, we can expect to see a new pickup in 2008, a 4 × 4 in 2009 and three or four other variants by 2010–12. Forecast sales are up from 500,000 to 1 million by 2010. The Logan is meant to account for 90 per cent of Renault’s sales growth in 2009.
Renault, 1992–2007
281
Aside from Romania, the Logan has been manufactured since 2005 in Moscow by Avtoframos, Renault’s Russian subsidiary; in Casablanca (Morocco) by Somaca; and in Envigado (Colombia) by Sofasa. It production was also extended to three new countries in 2007: Iran, Brazil and India. South Africa was on the cards for 2008. However, by 2007, more than three years after its launch, only 368,000 Logans had been manufactured. Customers’ willingness to wait quite a while for model deliveries means that it has a much higher potential. Another reason why the Logan has not hit its volume targets is because of insufficient production capacities. Once again, Renault has been taken aback by the success of one of its vehicles. Belatedly and hastily, it raised the capacity of its Pitesti plant in Romania from 235,000 to 350,000 vehicles/year. There were also problems with its Iranian (Iran Khodro) and Indian (Mahindra & Mahindra) partners, delaying volume increases in these regions. As for the mega-factory of 400,000 vehicles being built in Tangiers (Morocco), this will only enter service after 2010. Even more serious is the fact that Tata, Volkswagen, Fiat, Toyota, General Motors and Ford have all announced lowcost cars now. Renault risks losing out on its innovation rent – an essential source of profit for a company that makes innovative vehicles – just when it finally acquires the capacities it lacks today. What happened to the Scénic could happen again. At the same time, Renault has been preparing the second generation of the Logan, and above all, an even less expensive vehicle, to be sold for $2,500 on the Indian market with a new partner, Bajaj Auto. Renault–Samsung, slower progress than expected Dacia and Logan may have succeeded beyond Renault’s wildest dreams, but Renault–Samsung Motors (RSM) has struggled to become the bridgehead in Asia that the company had been hoping for. Sales have stagnated since 2002 at around 100,000–120,000 vehicles. Targets for 2010 have been lowered to 350,000 units, half of them abroad. Exports only really started taking off in 2006, notably with 40,000 branded vehicles sent to Russia. Production hit the 161,000 mark in 2006, all concentrated in the Busan plant, which has a capacity of 240,000 vehicles per annum. The 2007 product range, comprising three models that have all been renewed once already, has been enriched with a sporty 4 × 4. Half of this line is meant for the Korean market. The other half is to be sold in the rest of the world under a Renault label. Plans are for annual sales of 100,000 units. What are the problems facing Nissan? A third plan (2005–7) called ‘Nissan Value Up’ was launched in early 2005 (see Stevens and Fujimoto, Chapter 5, this volume). It contained three objectives: remain one of the world’s most profitable carmakers for three years; sell 4.2 million vehicles in 2008; and average a return on capital invested of 20 per cent over the course of the plan. To achieve this, Nissan decided to
282 The Second Automobile Revolution
turn its Infiniti make into a globally recognised top-of-the-range brand. This involved launching twenty-eight new models (including ten that were supposed to be ‘new automobile concepts’); offering and selling a large number of light utility vehicles worldwide; using suppliers from ‘leading competitive countries’; and extending and building Nissan’s presence in China, India, Thailand, Russia, Eastern Europe, the Gulf and Egypt. The third plan materialised in a less favourable context than its predecessor due to rising raw material and energy prices, higher interest rates and low or even zero growth in the more mature markets. Nissan also had to increase promotional spending and was unable to pass on the extra costs in its retail price. These new circumstances coincided with a quiet period in terms of new product launches. Nissan’s world sales fell in 2006, as did its operating margin (which nevertheless remained an enviable 7 per cent). Management recognised that it would be hard to reach all of its targets and called for a more sustained and quicker implementation of the plan. Faced with these problems, executives also said that they wanted to be open and make transparency one of the solutions for their problems. Other aims were also given. The plan was broken down into different objectives at every level of the company, even at the level of individual employees, who would only be allowed to exercise their preferential stock purchase options if they reached their targets.
Goodbye GM, hello Avtovaz . . . when should Renault move into China and build a clean car? Since 2007, Renault has continued to act unpredictably, showing an appetite for any challenge it faces, however large. Carlos Ghosn surprisingly did not say no in 2006 when GM shareholder Kirk Kerkorian suggested an alliance with Renault–Nissan as a means of rebuilding the world’s largest carmaker (see Senter and McManus, Chapter 9, this volume). When General Motors’ Board of Directors decided to renew its trust in CEO Rick Wagoner, Renault had to look elsewhere. Already present in Russia – which is the Logan’s second biggest market – in late 2007 the government in Russia authorised Renault to take a 25 per cent stake in the leading carmaker, Avtovaz, and to manage its industrial and commercial recovery. Renault had won rights to Avtovaz by outbidding rival carmakers who were earlier favourites. This acquisition was no small affair, given the in-depth changes that the Russian firm required and in light of the country’s unstable political and social context. The expectation was that considerable economies of scale could be achieved by giving Avtovaz access to Renault’s mechanical systems and platforms (900,000 vehicles by 2007). Renault also counted, more surprisingly, on being able to access inputs that could only be sourced from the Russian carmaker’s upstream suppliers. Regarding China, in early 2008 Renault began analysing FDI opportunities in this country. For the moment, it has no presence here, although its ally Nissan
Renault, 1992–2007
283
does. Lastly, Renault seems to have chosen with Nissan the development of an electric car. A model is announced for 2010 in several countries: especially in Israel, Denmark and Portugal.
Conclusion Is the alliance creating compatibility between conceptual innovation and commonalisation, or merely inventing a new way of combining volume and diversity? At the time of writing (in late 2008), it is still difficult to ascertain whether the new models will satisfy customer expectations. Renault sales are up again, but only due to Dacia and Samsung’s contributions. The Renault–Nissan alliance and the partnership with Avtovaz are clearly part of a ‘volume and diversity’ approach that sees the commonalisation of platforms, engines and gearboxes as a key tool for achieving economies of scale. Experience shows, however, that shared platforms B and C were hard to define and that actual commonality was less than expected. The alliance was always trying to cut the number of platforms to ten, but none of these were supposed to be shared. The notion of ‘interchangeable components’ tended to take over from an earlier focus on platforms in Renault and Nissan’s discourse. Commonalisation efforts tended to make components interchangeable amongst several platforms, thanks to the standardisation of interfaces. Quite logically, the two carmakers were simultaneously attempting to design processes that would enable the production, on one and the same line, of models that did not share the same platform. The real question is whether Renault and Nissan are on the way towards making conceptual innovation compatible with commonalisation, or whether they are actually inventing a different way of combining volume and diversity? Are they about to overcome the contradictions that have stymied them for fifteen years? Everything depends on the extent to which the ‘interchangeable components policy’ will help designers and factories to fulfil the occasionally (terminally) vague expectations of a broad and shifting range of automobile users. Translation by Alan Sitkin
Renault, 1980–2007
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
Turnover(3)
1,999,591 1,764,702 1,921,307 2,035,133 1,740,737 1,637,634 1,754,332 1,831,390 1,850,667 1,966,724 1,776,717 1,790,709 2,041,829 1,713,633 1,850,267
1,659,099 1,479,691 1,674,416 1,842,801 1,607,441 1,499,979 1,537,123 1,612,146 1,630,786 1,717,279 1,571,264 1,587,787 1,777,401 1,459,188 1,618,831
760,879 640,156 757,954 979,425 887,177 881,149 779,867 809,589 807,739 837,608 784,112 829,298 987,932 817,788 923,485
340 492 285 011 246,891 192,332 138,264 137,636 217,209 219,244 219,876 249,445 205,184 202,922 264,428 254,445 231,436
223,450 215,844 217,269 219,805 213,725 196,414 196,731 188,936 178,665 174,573 157,378 147,185 146,604 139,932 138,279
164,461 157,402 152,202 161,643 157,696 144,961 139,313 136,646 135,010 129,699 114,516 106,232 106,912 103,148 102,358
105,319 103,613 103,759 102,528 98,153 86,122 79,191 75,911 71,898 70,720 68,713 63,644 61,075 60,608 59,346
80,118 87,971 104,145 110,274 117,584 122,138 131,060 147,510 161,438 174,477 163,620 171,502 184,252 169,789 178,537
58,006 63,669 76,272 82,271 85,379 89,634 101,824 114,375 123,495 135,717 129,230 133,206 143,387 130,179 135,506
49,864 53,620 65,752 73,560 72,105 72,644 82,992 93,333 99,802 113,731 110,694 112,297 129,972 116,776 130,875
638 −690 −1,281 −1,576 −12,555 −10,897 −5,847 3,254 8,834 9,289 1,210 3,078 5,680 1,071 3,636
Investments (3)
303 −875 −2,563 −1,875 −11,324 −11,241 −7,355 2,314 7,316 6,932 1,223 2,467 3,251 −5,225 1,463
5,442 6,345 6,331 8,043 7,872 6,209 4,141 5,170 6,197 8 703 8,847 8,303 10,347 9,173 12,188
4,733 4,555 3,524 3,941 3,938 2,762 1,786 3,090 4,002 5,146 6,013 6,624 5,540 5,378 7,337
France (4), World France (4), automobile automobile automobile branch branch branch
Net income(3)
Worldwide Domestic Of which Production World World France (4), World World France (4), World exports abroad Renault automobile automobile Renault automobile automobile Renault (1) (1) Group branch branch Group branch branch Group
Year Production Workforce (2) (passenger cars and light commercial vehicles)
Appendix Table 14.1
1,761,643 1,741,161 1,867,619 2,197,395 2,257,918 2,427,178 2,375,084 2,343,954 2,385,079 2,471,676 2,515,728 2,385,090 2,635,753
1,610,216 900,077 151,427 1,602,632 961,940 138,529 1,121,970 745,649 1,373,936 823,459 1,376,707 881,211 1,407,717 1,019,461 1,412,577 962,566 1,344,847 999,107 1,283,115 1,101,964 1,310,497 1,107,179 1,268,259 1,247,469 1,062,851 1,322,239 1,007,933 1,627,820
139,950 140,905 141,315 138,321 159,608 166,114 140,417 132 351 130,740 124,277 126,584 128,893 130,179
102,213 111,523 112,178 109,409 131,261 136,574 133,111 128,934 127,531
59,264 58,528 47,773 45,758 44,584 45,942 47,515 48,086 49,005 43,929 44,143 42,018 41,774
184,065 184,078 207,912 243,934 249,483 263,534 36,351 36 336 37,525 40,292 41,338 41,528 40,682
136,444 145,962 207,912 2 43,934 249,483 263,534 33,841 34,456 35,535 38,426 39,458 39,605 38,679
132,050 135,658 165,788 195,077 197,062 206,548
2,139 −5,266 159,232 182,096 193,753 202,495 1,051 1,956 2,480 2,903 3,453 2,869 2,669
944 −190 5,427 8,847 3,506 7,082
12,610 13,550 4,135 5,990 4,434 9,540 2,915 2,628 2,533 3,199 2,879 3,585 3,565 12,875 12,366 13,619 15,606
7,900
Note: (1) Exports include the CKD (Completely Knocked Down), with the exception of parts ‘small series’, corresponding to the vehicles ‘manufactured abroad’. From 1997, Renault discontinued publishing its exports statistics, providing only information on vehicles assembled in France or abroad, regardless of the parts’ origin. From 2000, the productions of Dacia and of Renault Samsung are included. (2) Staff numbers are as of 31 December of the year concerned, including temporary contract workers. (3) Turnover, net income and investments are in millions of French francs until 2000, in euros after. (4) Workforce, turnover, net income and investments in France are for the RNUR (Régie Nationale des Usines Renault), then for Renault S.A., and finally for Renault s.a.s. Source: Renault annual management reports.
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
286 The Second Automobile Revolution
Bibliography A bibliography on Renault from 1898 to 2007 can be accessed at: http://freyssenet.com/ ?q=fr/node/1015. All numbers and facts not referenced in this chapter can be found at: http://freyssenet.com/?q=node/876. Andria, J. F. de (2004) ‘Renault–Volvo: Des fiançailles rompues’, Renault-Histoire, June: 31–50. Angelescu, G. (2007) ‘Une analyse du processus d’apprentissage organisationnel dans un contexte spécifique: la gestion de projet. Étude de cas sur l’entreprise Dacia, groupe Renault’, Thèse en gestion, Université de Paris 1 Panthéon-Sorbonne. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York, Palgrave Macmillan. Bruner, R. F. (1999), ‘An Analysis of Value Destruction and Recovery in the Alliance and Proposed Merger of Volvo and Renault’, Journal of Financial Economics, 51: 125–66. Charron, E. (2004) ‘Making Renaults for Europe: Workers from Renault’s Plants Outside France’, in E. Charron and P. Stewart (eds), Work and Employment Relations in the Automobile Industry. New York and London: Palgrave Macmillan, pp. 220–39. Douin, G. (2002) ‘Renault–Nissan, les coulisses de l’exploit’, Journal de l’École de Paris du Management (November–December): 7–14. Freyssenet, M. (1998) ‘Renault, from Diversified Mass Production to Innovative Flexible Production’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 365–94. Digital publication, freyssenet.com, 2007, 555 Ko, ISSN 7116–0941. Freyssenet, M. (2003) ‘Renault: Globalization, But For What Purpose?’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan, pp. 103–31. Hirt, O. (2004) ‘La relation design–ingénierie dans les nouvelles organisations de la conception. La démarche des fondamentaux en design de Renault’, Actes du GERPISA, 37 (March). Jullien, B. (2006) ‘Les inégalités de revenus et leurs effets sur les demandes automobile’, paper presented at the 14th international colloquium of GERPISA, ‘Are automobile firms market-oriented organisations? Myths and realities’, Paris, 12–13 June. Jun, H. (2002) ‘Renault–Samsung’s New Strategic Choices: From Scale to Scope’, paper presented at the 10th international colloquium of GERPISA, ‘La coordination des compétences et des connaissances dans l’industrie automobile’, Paris, 6–8 June. Lauer, S. (2005) Renault, une révolution française. Paris: Éditions Jean-Claude Lattès. Le Quément P. (2006) ‘Créateurs d’automobiles: la gestion des designers chez Renault’, Le Journal de l’École de Paris du Management (January–February): 32–8. Loubet, J.-L. (2006) Carnet de route, 1898–2005: l’histoire de Renault, année après année. Boulogne-Billancourt: ETAI. Pelata, P. (2003) ‘Nissan: le nouveau et le renouveau’, Journal de l’École de Paris du Management (September–October): 16–23. Riès, P. and Ghosn, C. (2003) Citoyen du monde. Paris: Grasset. Schweitzer, L. (2007) Mes années Renault. Entre Billancourt et le marché mondial. Paris: Gallimard. Segrestin, B. (2005) ‘Partnering to Explore: the Renault–Nissan Alliance as a Forerunner of New Cooperative Patterns’, paper presented at the 13th international colloquium of GERPISA, ‘Organisation productive–relation salariale–financiarisation: les spécificités de l’industrie automobile’, Paris 16–17 June.
15
Fiat Group Automobiles: an Arabian Phoenix in the International Auto Industry Giuseppe Volpato
Crises and reorganisation in the evolution of Fiat Auto Fiat Group Automobiles,1 the group that controls automotive manufacturing activities within Gruppo Fiat SpA, has gone through many crises in more than a century of existence. The crisis that arose dramatically in 2002, albeit rooted in the previous decade, presents however some peculiar aspects. On the one hand it was certainly the most serious ever, since only the support by other Fiat Group businesses and a team of banks has prevented the ‘virtual bankruptcy’ of Fiat Automobiles turning into an actual one. On the other hand, however, the crisis has been followed by a recovery that has astonished scholars and observers for the speed and character of the turnaround. We could liken the recovery to a ‘rebirth’ from the ashes like the Arabian phoenix, given the many elements of novelty and change that the management of Sergio Marchionne, CEO of Fiat Group, has achieved since 2004. To describe and to interpret the new strategic trajectory requires, considering its peculiarities, outlining the situation facing Fiat Automobiles in 2004, after a long stage of managerial and competitive involution.2 In the second half of the 1980s Fiat Auto reached the end of the decade with brilliant results: both in terms of market performance, with a market share in 1989 equal to 12.30 per cent in Western Europe and 57.18 per cent in Italy, and in terms of profits that in 1989 totalled 28,424 billion lire, equal to a14,679 million, corresponding to a return on sales (ROS) of 8.3 per cent. But the end of the 1980s also marked the exit of Fiat Auto CEO, Vittorio Ghidella, due to disagreements with Fiat Group CEO, Cesare Romiti. The wish of the latter not to name the new CEO of the Auto Group from among Ghidella’s collaborators induced him to lead the group ad interim, without having the technical competence or the necessary commitment.3 So a stage of slow renovation in models and process technologies began, just when both nationally and internationally new conditions were maturing, conditions that would push Fiat Automobiles to a significant redefinition of its own strategies and of its positioning in the global competitive landscape. 287
288 The Second Automobile Revolution
The elements of change in the competitive scenario from 1990 In order to highlight the challenge involved in maintaining the competitiveness level that Fiat Auto achieved in the late 1980s throughout the following decade, it is important to underline that there were profound transformations both in domestic and international automotive demand, and in the reorganisation of supply where pressures by Japanese manufacturers were growing stronger, and there were hints of future competitive aggressiveness by Korean brands. Among the elements of this new scenario the following should be mentioned.
• The change in the Italian domestic demand. The safety net that eased Fiat Automobiles’ life in the alternating sequence of crisis and expansion is represented by domestic demand where the brands controlled by Fiat operated in a quasi-monopolistic position and, even eighteen years after the European common market, it still had a market share over 60 per cent. However, it was easy to see that such a landscape was going to change. In fact, thanks to the growth in motorisation that in Italy had reached particularly high levels after the mid-1980s, Italian automotive demand was gradually moving towards a ‘preference for variety’ that led to grow the share of cars in niches where Fiat was hardly present. • The downward expansion of product lines both for volume and for specialist automakers. With the transition in all major markets of developed countries from the stage of first purchase to replacement demand, and the growth in multi-motorisation,4 a wide process of gradual segmentation of the market began, into smaller and smaller portions and niches, with the need to develop targeted offers for specific consumer clusters.5 Volume automakers had begun to pay more attention to smaller car segments, previously believed to be less important,6 since the second car purchased by a same family was often a city-car model. Such evolution also had a significant impact on the room for manoeuvre for Fiat Auto that, pushed by a fiscal policy that penalised large engines,7 had historically developed by focusing on smaller segments. • The emergence of Korean automakers in Europe. The manufacturing growth achieved in the 1980s by Korean brands was a clear signal of their interest in entering the Triad markets. This clearly was true also for Italy, with the issue, however, that these new competitors would have started their race with models at the lower end of their range, i.e. segments that were in more direct competition with Fiat Automobiles’ range, and they seemed oriented to practise quite aggressive price policies. The Korean company that played the role of forerunner in Italy was Hyundai which started imports in 1982. In 1993 Kia followed, then in 1994 Daewoo.
Fiat Group Automobiles 289
Fiat Automobiles’ priority objectives for the 1990s Already in the late 1980s it was evident that a stage of growing competition in Europe was emerging, mainly for brands operating at the lower end of the range, such as Fiat Automobiles, which luckily could rely on a particularly favourable end of decade, but which nonetheless had to carry out profound transformations if it was to maintain its chances for success.8 In particular, the necessary objectives were the following.
• More balanced sales mix in the different markets. At the end of the 1980s the geographical distribution of Fiat sales was too dependent on the domestic and Western European markets, mainly after the exit from the US market in 1982. It was then necessary to adopt a new and strong policy of internationalisation aimed at newly industrialised countries (NICs) that were moving into a promising motorisation process. • Clearer definition of brand positioning. The purchase of the Alfa Romeo brand that took place in 1986 required an overall repositioning since in previous years the Lancia brand was progressing on two fronts, partly towards classic, comfortable and elegant cars with the larger models, partly towards sports cars, with smaller models, and in particular with the ‘Delta’ model that was awarded many prizes in the world rally championships. • Gradual product upgrading. The fact that automotive activities were largely leaning towards smaller segments necessarily had to be correct. If up to that time the specialisation on low end cars had allowed Fiat to fully exploit domestic demand (also thanks to lower salaries in Italy versus other European and US competitors), it was evident that this position would become much harder to maintain with the entry in the competitive arena of automakers having even lower salary levels: Eastern European countries, Latin America and Korea. • Overall quality enhancement. Another must for the Turin group was represented by the rationalisation of manufacturing processes, both internally and externally, to elevate in a consistent and lasting way vehicle quality and reliability. Without an upgrading of the quality–reliability combination, also in smaller cars, the gradual erosion of market shares would have manifested to a worrying degree. It is worth underlining that such a programme would not have taken place without a massive investment policy, so it would have been legitimate to wonder whether it was worth it. We should note that in the late 1980s the path sketched here entailed considerable risks for shareholders. However, the available alternatives were just two: either to adopt a strong policy of commitment in the industry trying to activate a ‘slow but systematic’ expansion (possibly also through cooperation agreements with other important
290 The Second Automobile Revolution
automakers), or to exit the industry by selling the automotive brands and switching to other industries. Ultimately there was a situation of tertium non datur – there is no alternative.
The 1990s decade: the years of inadequateness Changes at the top Unfortunately the changes that took place at the top of Fiat Automobiles in 1989 would turn out to be seriously inappropriate for the competitive challenge that was lying ahead, both because of previous mistakes and the inadequacy of the ‘new’ top management that, even when it moved in the direction required by the emerging competitive scenario, did so with delays and mistakes. The most relevant aspect is that, if strategic deficits were considerable, the operational management was also largely faulty. Cesare Romiti, CEO of Fiat Group, who at the end of 1988 also became CEO of Fiat Auto, was mainly anxious to reduce the weight of the automotive sector within the Group in order to avoid a too ‘auto-centric’ character that would subtract power from his management. This translated into a shift of profits generated by automotive activities into other sectors. In 1989 the investments dedicated to cars began falling below 50 per cent of total investments by the Fiat Group, while those directed to ‘diversified’ activities grew from 14.6 per cent in 1988 to 24.7 per cent in 1990, also due to new acquisitions in the insurance and distribution arenas. The overall outcome was a considerable reduction in market shares both in Italy and in Western Europe that alarmed Gianni Agnelli and in December 1990 induced Romiti to leave the position of CEO of Fiat Auto to Paolo Cantarella, who was his assistant in the past. Cantarella, in contrast to Romiti, was a ‘car guy’ and initially gave the impression of being able to reinvigorate the automotive group through model renovation and to lead it towards a more adequate business model: more open internationally and structured into a wider range of segments. In 1993 there was the important success of the launch of the Punto model, that replaced the Uno, followed by a promising Fiat range expansion with the Coupé and Barchetta models, and the Spider and Coupé for Alfa Romeo, adding to models traditionally offered by Fiat Automobiles in mid-size cars.9 However, probably due also to the workload following the period of inertia in product launches between 1989 and 1992, not all models were up to the task, mainly from the quality standpoint. For the above-mentioned niche models there were relevant reliability issues, while other cars were inadequate on the design side, such as the new Lancia Delta (1993), incapable of replacing the previous model. Similar troubles affected the Alfa Romeo 145 and 146 models, and total sales for this brand in 1994 were at half the levels achieved in 1990.
Fiat Group Automobiles 291
The deficits in the range/product plan The decision by Paolo Cantarella in the mid-1990s to change the business model of Fiat Automobiles towards a ‘volume and variety’ approach could not be maintained with the quite unsatisfactory results achieved during the following years, where once again the design stage was deficient. Cantarella tried to change course with the launch of the Stilo model in 2001, but results were once again unsatisfactory, and the continuous worsening in the image of the three major Fiat brands (Fiat, Alfa Romeo and Lancia) had repercussions for the whole range with significant losses in market share both in Italy (below 35 per cent share in 2001) and in Europe (below 10 per cent). The modest level of such results becomes even more evident if one considers that at the same time Fiat Automobiles exerted great pressure to support sales and to limit market share erosion. This drove the rise in the practice of cars being directly registered by dealers (or Fiat’s wholly owned dealerships), then sold into the market with very high rebates as ‘zero-km used cars’. The Turin-based company at the same time was considerably increasing its direct sales with high rebates to rental companies and other business customers. It was a disastrous policy both for dealers and for the automaker itself. Consequently, the achieved results, albeit lower than targets, were met through sales drugged by rebates that were largely greater than the competition.10 The fact that even in 2001, the year when total registrations in Italy peaked, Fiat Automobiles had an operational loss of a549 million, suggests the magnitude of the costs borne to support sales, basically by selling below cost.11 The internationalisation strategy With respect to the internationalisation process, the 1990s were a time of strong commitment for Fiat Automobiles, unfortunately partially affected by a chain of unfavourable events and some mistakes. The internationalisation envisioned by Avvocato Agnelli became a reality in the mid-1990s with the ‘178 Project’. It consisted in a family of cars starting from the same platform, to be manufactured in a range of countries, but in a highly standardised way, in order to exploit significant scale and learning economies both for plant construction and in manufacturing, with the possibility of transferrring parts between the different manufacturing locations. The globalisation of the car was aimed at new-motorisation countries such as Brazil, Argentina, Poland, Russia, Turkey, India and China. The objective was to reach a combined volume of about 1 million cars in 2003, to be manufactured in about a dozen countries.12 The idea made sense even if it emerged that inside these markets there were significant differences in customer preferences. After a first encouraging stage in Latin America the international financial crisis of 1998 led to a slowdown in the project.13 However, even though the results were partly affected by the negative context, serious mistakes were undoubtedly made both in the programme planned for India, with the development of a
292 The Second Automobile Revolution
plant that was then never utilised and with an excessive degree of product standardisation that was inadequate for that market, and in Russia and China with delays and quite unsatisfactory results. The reduction of internal competences and resources and the state of crisis The gradual worsening of the crisis of Fiat Automobiles began already in 1997 to trigger a vicious circle in which the deficit arising from operations led on the one hand towards greater debt and on the other hand towards a reduction in internal resources in terms of employees, investments and R&D expenses. The most evident indicator of the outsourcing of activities in Fiat Automobiles is represented by the share of personnel costs in the income statement. The reduction in employee levels had started after Ghidella’s exit. From the peak recorded by the whole of automotive activities in 1989 with 134,270 employees, numbers fell to 114,386 in 1995 and to 74,292 in 2000. A comparison between the share of personnel costs in turnover of Fiat Automobiles compared to the Renault and PSA groups, in the automotive fields, clearly shows the dramatic move to forms of outsourcing carried out by Fiat.14 In 1995 the share of personnel costs in sales was 14.44 per cent for Fiat, already lower than Renault (18.74 per cent), but in 2000 the share fell to 9.07 per cent for Fiat versus 14.98 per cent for Renault and 13.44 per cent for PSA. A highly relevant aspect is that outsourcing involved not only component manufacturing and sub-group assembly activities, but also design and coordination activities of suppliers’ work.15 Between 1991 and 2000 the parts within Fiat’s own area of responsibility decreased from 76 per cent to 28 per cent.16 This development, which began in the early 1990s, and apparently eased in the mid-1990s for a short period of time, returned in 1998 and started to accelerate at the beginning of 2000: Fiat’s market share, notwithstanding the policy of strong rebates and direct sales to rental companies, continued to fall, causing a reduction in sales and a collapse in operational results: a loss of a549 million in 2001, and a loss of a1,343 million in 2002. The crisis situation manifested in December 2001 with a shocking declaration of a a4 billion overall loss for Fiat Group of which 2.7 billion was due to Fiat Automobiles. In 2003 Gianni Agnelli died and his brother Umberto took his place, the latter dying in May 2004. The automobile firm then faced a stage of continuous changes in governance that did not facilitate recovery. The turnaround took place only in 2004 with Luca Cordero di Montezemolo becoming chairman of the whole Fiat Group, and Sergio Marchionne becoming CEO, the latter also becoming CEO of Fiat Automobiles in February 2005 to manage the recovery of this sector more directly.
The reorganisation operated by Sergio Marchionne Shaking up Fiat Automobiles’ management The steps taken by Sergio Marchionne to lead Fiat Group (and in particular Fiat Automobiles) out of the tunnel are many. Some are initiatives with
Fiat Group Automobiles 293
short-term outcomes that have already generated significant effects, while others will produce their effects over the medium to long term. Overall, these initiatives constitute a dense web that cannot be summarised without some simplifications, although they do not affect the larger picture. The priority initiative is the most delicate and relates to the replacement of some Fiat Automobiles figures located in high-level decision-making posts. The marked crisis of Fiat and the disappearance of key figures such as Gianni Agnelli in January 2003 and Umberto Agnelli in May 2004 had triggered many rounds of changes at the top of the group that, as often happens in such cases, considerably slowed down the initiative of managers and cadres in the face of ever-changing scenarios without clear guidelines. Moreover, the tradition of the Turin-based company with respect to management turnover had always featured a preference for internal solutions, and this translated into modest changes in ideas, managerial styles and approaches. Marchionne profoundly impacted on this overly bureaucratic and routineoriented structure with initiatives that can be defined as management by stress, based on hard work and strong commitment. On the one hand, he demanded clear commitment in terms of objectives and times, while on the other hand he made room for young managers, more oriented to fight for their ideas and willing to work hard to carry out more with less resources than their predecessors had available. Moreover, Marchionne did not bring along with him a staff when he became CEO of Fiat. He relied on his own ability to choose the right people, from inside the company if they seemed suitable, but also from outside and abroad if he identified the person he needed. If one looks at the number of employees in Fiat Automobiles one can see that their number decreased dramatically under Marchionne’s management, but he showed great leadership abilities, dedicating himself to extremely hardworking schedules and asking all Fiat Automobiles personnel to do the same, applying a very high personal drive in achieving objectives. In addition, Marchionne initiated in spring 2004 a process of assessment for individual managers, evaluating both their leadership abilities and their achieved performance. In little more than a year the whole of Fiat Automobiles’ managers reporting directly to him underwent a profound transformation. In mid-2006, 40 per cent of the twenty-seven first-level top managers were foreign. This would have been unthinkable just a few years before. The international cooperation agreements In the field of international agreements Sergio Marchionne initially had to untie the knot represented by the agreement with General Motors defined in 2000. Such agreement did generate positive outcomes, but it entailed a put option by which Fiat could have asked General Motors to purchase Fiat Automobiles. This possibility was a sword of Damocles hanging over General Motors since the auto activities of Fiat in 2003 were on the verge of bankruptcy and the American company was itself facing an acute crisis that the absorption of the Italian company would have worsened.
294 The Second Automobile Revolution
This put option was a time bomb and negatively affected the opportunities for cooperation between the two automakers, who had in place a shared structure for component purchasing and for the development of shared platforms, and also stood in the way of other international agreements that Marchionne was interested in developing.17 Marchionne managed to cancel the agreement, through the price of a1.55 million to be paid by GM. Immediately after this he moved on to finalise a set of agreements:18
• with the Indian group Tata for wide-ranging cooperation, including shared manufacturing of some models in India and in Argentina, but also joint development of models, component design and manufacturing, joint marketing and distribution in India, through Tata networks; • with Ford to use the new city-car platform designed for the new Fiat 500 also for the Ford Ka model, to be manufactured in the Thichy plant in Poland; • with Severstal Auto to assemble in Russia the Palio and Albea models manufactured in kits at the Turkish Tofas plant, owned by the jointventure between Fiat and the Koc group; • and with Chery to strengthen vehicle manufacturing in China. The new dynamism on internationalisation has been supported by the comeback in the major markets where Fiat Automobiles was involved. Latin America has been first, where Fiat operates with the new mega Betim plant in the Brazilian state of Minas Gerais. With the recovery in automobile demand that took place in the area the Betim plant is manufacturing at full capacity and has moved to a third shift in final assembly by employing new personnel, with a daily production of 2,500 vehicles. Fiat Auto do Brasil (FIASA) has contributed significantly to the economic comeback of the group. Other growth areas are Turkey and Poland, where the manufacturing of the 178 project models (Palio, Siena, Palio Weekend, etc.), adequately renovated, was augmented by new models. The ‘Linea’ model stands out among others, a three-volume car developed on the Grande Punto platform in the Turkish Tofas plant and launched in mid-2007 with an initial production of 60,000 units, but with the objective of reaching 280,000 units through manufacturing also in Brazil, Russia, India and China. Range renovation and sparkling communication Clearly the central initiative in Marchionne’s turnaround has been the refreshing of the range of models and of the brands offered. The new CEO could in part benefit from the work of some of his predecessors who had correctly started the renewal of some important models. The most obvious case is that of the new Panda launched in 2004. This small car, manufactured in Poland, gained the European ‘Car of the Year’ award. This city-car has
Fiat Group Automobiles 295
a significant value for money ratio, and an undisputed quality. The recovery of the image and quality of the product from this well-engineered model has generated interesting profit margins both for the automaker and for the distribution network. This model was followed with significant results for the new Croma in the third quarter of 2005 followed in 2006 by the Grande Punto, the Alfa Romeo 159, the Alfa Romeo Brera and in 2007 by Bravo, Linea and the 500. Overall, Fiat Group has scheduled the launch of twenty-three new models and sixteen major facelifts between 2007 and 2010. The strengths of the present range are mainly Panda and Grande Punto with ambitious volume plans that have even been exceeded, bearing in mind that manufacturing of the previous Punto is still running at about 100,000 units per year. Among the models wholly developed under Marchionne’s management there are two that are particularly important, namely the Bravo and the 500. The first model is in the C segment, the most important in Europe, with competitors such as Volkswagen Golf, Opel Astra and Peugeot 307. It is thus a very important model for the comeback of Fiat Automobiles both for its contribution to profits, and for the consolidation of the company image, where in the past the firm had suffered from the failure of a model like the Stilo. Marchionne has decided to start with quite a cautious policy, declaring for Bravo a target of 120,000 cars per year, a policy which the whole specialised press has judged too prudent. A significant part of the favourable reaction among customers for the new product range of Fiat Automobiles comes from the considerable enhancement in quality of product engineering, which has generated both higher perceived quality and higher satisfaction, and lower development and manufacturing costs. There is wide agreement that the new managers chosen by Marchionne are doing a good job, notwithstanding the need to control costs, through better coordination with suppliers. If one looks at the future programme of new model launches for the three brands, it is not the number of scheduled launches that surprises (albeit the innovative effort will undoubtedly be heavy), but rather the level of care adopted in preparing the launches and the new way in which the company communicates with customers. For example, the launch of the new 500 marked quite a break from the past. This model is a modern reinterpretation of the old car launched in 1957 and manufactured in 3.7 million units; the new one is already a media icon. For the new 500, Fiat Automobiles activated a website many months before launch (1 July 2007) to allow potential customers to express their views on many aspects of the car: a new and involving way to establish direct contact with customers and to exploit their suggestions. This kind of initiative drew much interest from the public, and generated a lot of goodwill towards the specific product and the overall Fiat range in general. Only three months after launch the orders recorded for the 500 have reached the whole target for 2007 and led to an increase in production capacity in the Tychy Polish plant, which by the end of 2008 was due to manufacture 530,000 vehicles between Panda, 500 and Ka.
296 The Second Automobile Revolution
The communication adopted for the Lancia brand has also been refreshed and, while still short of new models (in particular the new Delta HPE that was launched in 2008), managed to grow sales both for the Ypsilon and Musa model, both in Italy and abroad. A modern marketing strategy and effective communication represent a key step in the Group’s turnaround, and this has been accompanied by a range of innovations in other areas. Turning Fiat Automobiles towards the customer On the customer care front there was also a marked acceleration in change. Fiat Automobiles did not have a solid tradition in this area due to the chronic tendency of the Turin company to subordinate marketing policies to manufacturing and design needs, so that the relationship with customers had to be fundamentally turned upside down. The beginning was the establishment of a new customer care position assigned to a manager from Iveco who reports directly to Marchionne. This move marked a dramatic change of perspective from the past and has made room for a multiplicity of initiatives with high value for the end customer. Among the most significant elements of novelty is the establishment of a call centre, directly managed by Fiat Group, which all European consumers can access through a single toll-free number, capable of managing inbound calls in nine different languages. The importance and the innovative nature of the call centre consists in its capability to track customer needs from the time of the order throughout its delivery, monitoring the behaviour of all the different subjects involved in supporting the consumers’ needs. Through this initiative the average lead time required for trouble shooting has decreased from twenty-three days in the middle of 2006 to seven and a half days by the middle of 2007. Another relevant aspect from the commercial standpoint is the renewal of the website of the auto brands in order to interact with potential customers. The customers showing interest in the products when surfing the website of one of the three brands are then contacted by the Customer Experience Management group set up by the customer care system, and passed on to the distribution network, especially when new models are launched. The leads generated by this initiative are constantly growing and already contribute to 3 per cent of sales for the distribution network. Such an initiative is also important for the high level of commitment among dealers: this adds positive effects both in terms of customer satisfaction and for the network’s ability to attract new customers. According to data produced by the customer service team, the closing ratio of leads generated by the manufacturer has increased from 25 per cent to 31 per cent with a marked increase in the share of sales recorded locally by dealers, and an increase in customer satisfaction indices. Another important element of this new strategy of attention towards the customer lies in the strengthening of the after-sales care service, which during the crisis years experienced a dramatic loss of confidence. Nowadays
Fiat Group Automobiles 297
the after-sales care service is managed in an integrated way with the ambitious objective of eliminating car-off-the-road situations due to the lack of spare parts. The development of fourteen parts distribution centres in Europe is envisaged, to ensure twice-daily delivery of parts to the service network.
Doing more with less Choosing routes that differ from the competition As will be shown later, the objectives of the company relaunch in 2004 by Marchionne have been achieved on time; however, it is appropriate to note that the comeback has been favourably influenced by some factors that could change in the future. One of these is the current Italian government policy of incentives towards the scrapping of obsolete cars (Euro ‘0’ and Euro ‘1’) that supports automobile demand in the domestic market. The second factor is the slowdown in market penetration in Europe for the two direct competitors of the Fiat brands – the Peugeot–Citroën and Renault groups. This means that over the next few years one can expect a considerable increase in investments by these two groups, as well as by other manufacturers, particularly Volkswagen. As a consequence, the chance for Fiat to consolidate its recent comeback is related to the possibility of developing something ‘special’, i.e. initiatives capable of eliminating the competitive gap that has accumulated during the decade preceding the turnaround strategy implemented by Marchionne. On the basis of interviews carried out with the management we may define the new strategy as an effort aiming at ‘achieving more with less’. As Marchionne has repeatedly declared, Fiat Automobiles cannot limit itself to imitating competitors, but must pursue innovative routes and anticipate competitors’ moves. This philosophy of doing more with less is apparent in a wide range of company activities. In part the capability to improve manufacturing quality, and reduce costs at the same time, derives from the policy of agreements such as for the Sedici model, an off-road vehicle manufactured by Suzuki and customised for Fiat, or for the platform sharing for the Panda and 500 model in Fiat with the Ka model in Ford, both assembled in the same manufacturing line. Another innovative example is the new Bravo model which has been developed through a focused carryover based upon the previous Stilo model. The body has been completely revised, while the mechanical part has only been fine-tuned with a dramatic saving in investments. It was Marchionne himself who during the presentation of the new model to the press underlined that the investment required for the new model amounted to just a350 million, in contrast to investment levels that are much higher for the competition, with the effect of a dramatic reduction in the break-even point, to the advantage of profitability. In the case of the Bravo the manufacturer will start to make a profit starting from an annual production volume of 75,000 units, and this marks a true record.
298 The Second Automobile Revolution
Time-to-market, platforms and standardisation Some important aspects of doing more for less are: (a) the reduction in the time-to-market (TTM) of new products; (b) the implementation of a strategy based on platform sharing; (c) the standardisation of modules. The Grande Punto already marked a good performance with twenty months of TTM. With Bravo the TTM was reduced to eighteen months, which represents the best performance in Fiat Automobiles and one of the best in the world. The new standard will be achieved with the launch of the Alfa Romeo MiTo which is targeted with a TTM of fifteen months. These results have been achieved through an intelligent implementation of carryover from the most advanced components already optimised in previous models, and through relevant technological innovations. Among these, an important contribution came from the introduction of a system of digital mock-up (DMU) that during the development stage for a new model avoids physical mock-ups19 that require long lead times and much higher costs.20 All this has translated into a higher quality level. It is not by chance that the new Bravo was the first Fiat model to be sold with a five-year warranty. Fiat Automobiles can regain much competitiveness in manufacturing costs through the strengthening of the platform sharing policy where its direct competitors (such as Peugeot–Citroën, Renault and Volkswagen) are enjoying some advantage. According to this programme the number of architectures, nineteen in 2006, will be reduced to six in 2012. In 2010 the number of models based on a single architecture will increase from 1.7 to 3.7 and the number of vehicles manufactured per architecture will increase from 150,000 to 450,000 units. Finally, a wide process of standardisation of modules that constitute the vehicle sub-systems is in place. For example, for the HVAC module (heating, ventilation and air conditioning) the product development managers have programmed a standardisation that will allow it to move from the eighteen modules used in 2005 to five modules in 2012, with significant cost savings, higher quality and reliability. World class manufacturing A particularly important contribution to the company’s performance will have to be achieved through an improvement in output quality in manufacturing and final assembly plants. In this respect, Fiat Automobiles is carrying out a wide programme of training initiatives at the various levels of each plant, and of definition of procedures aimed at eliminating any form of waste and activity that does not add value for the end customer. This programme, the Fiat Group Automobiles Production System (FAPS), is inspired by the Toyota Production System in its most advanced form, and will allow it to achieve the excellence standards required in the certification system of world class manufacturing (WCM). This is a structured set of methods and
Fiat Group Automobiles 299
tools applied throughout the company, involving all employees in order to promote a radical improvement in production system performance, to ensure that the product is delivered to the customer in the required lead time and with the required quality levels, and eliminating non-value adding activities and any other type of loss or waste of human resources, equipment, materials and energy. This is carried out by involving people at all levels in problemsolving activities, providing focused training to line personnel so that they know and implement a wide range of organisational tools21 aimed at quality enhancement and waste elimination. The programme is based upon the greatest possible involvement of operational personnel, who are expected to come up with suggestions and innovative solutions. These organisational tools are quite well known, but their integral application represents a great challenge particularly for the large personnel training effort that their implementation requires. Clearly the system must also be monitored through a set of key performance indicators (KPI) capable of recording the quality levels of work teams, which in Fiat are called elementary technological units (ETUs) in the most complex areas (stamping, welding, painting and final assembly) and in the whole of the plant. The importance of such a programme clearly emerges from the trend recorded in some KPIs. For example, on the personnel involvement side improvement proposals in the Polish Tychy plant have increased from 0.5 suggestions per employee per year in early 2006, to four suggestions per employee per year in late 2006, with an implementation rate of these suggestions of 50 per cent. In the Melfi plant between the end of 2005 and the end of 2006 the quality for electric defects has improved by 50 per cent with significant savings in rework activities, and in the same plant total personnel hourly productivity recorded in 2006 a greater than 12 per cent increase.
Trends in performance Market share recovery The launch of new models, helped also by the growth in registrations in the Italian market during recent years, and a strengthening of the government incentives in early 2007 for the scrapping of non-catalytic cars, has generated a significant recovery in market share for the brands of Fiat Automobiles. The low point for all three brands was recorded in 2003 in Italy, while in Europe as a whole the fall continued until 2005. However, there were some signs of recovery, mainly for the Fiat brand, whose weight is much higher than the other two, in the second half of 2004. Then the comeback accelerated in the second half of 2005, with the arrival of the Grande Punto. The year 2006 has confirmed the recovery for the brands, and in 2007 this has consolidated further (Figures 15.1 and 15.2).
300 The Second Automobile Revolution
105 100 A.R.
95 90 Fiat
85 80 75
Lancia
70 65 60 2000
2001
2002
2003
2004
2005
2006
2007
Fiat makes’ market share in Italy, 2000–2007 (2000 = 100)
Figure 15.1 Source: ANFIA.
120 110 A.R.
100 90
Fiat
80 70
Lancia
60 50 2000 Figure 15.2
2001
2002
2003
2004
2005
2006
2007
Fiat makes’ market share in Western Europe, 2000–2007 (2000 = 100)
Source: ACEA.
Financial performance If one looks at the recovery of Fiat Automobiles by considering economic and financial indicators then the magnitude of the turnaround appears much higher than the market shares in Italy and in Europe would suggest. The reasons are twofold. On the one hand the fast comeback from the huge losses recorded in 2002 with net losses equal to a2,739 million, benefited from the contribution of profits gained internationally, and mainly in Brazil.
Fiat Group Automobiles 301
After the crisis at the end of the 1990s the Brazilian market showed a strong growth trend and Fiat Automobiles, which fights head-to-head for first place with the Volkswagen Group, could benefit from this situation through the development of models and variants consistent with that market.22 In particular, Fiat introduced very early on the flex-fuel system, internally developed by Magneti Marelli. On the other hand, the return to profitability was obtained through the Group’s speedy reduction of its debt levels, and its improved ranking among international ratings agencies. If one looks at the main indicators one can see that Sergio Marchionne has maintained the previous policy of employee reduction and of control of R&D and fixed capital expenditures (Appendix Table 15.1). He acted so that the reduction in resources could be compensated by greater leadership abilities among his managers and by a stronger commitment from all employees. As a result, the number of Fiat Automobiles employees remains at minimal levels and the increase in the number of high-level professionals (managers and engineers) has been offset by a reduction in the number of low-skill profiles, with a further reduction of the share of employee costs in total turnover.23 Also on the investment front, where approval procedures require fast lead times and significant payback rates, the maximum was extracted from available resources through carryover actions from previous models. In November 2006 Sergio Marchionne and some of his staff met financial analysts in Turin. Detailed information was given both on achieved results and on the most significant objectives to be reached before 2010. Among these are the details of manufacturing and sales of cars and of commercial vehicles derived from cars. According to the CEO of Fiat Automobiles, sales in 2010 should reach a level of 2.8 million units for the brands falling within the consolidated group (Fiat, Fiat Professional, Alfa Romeo, Lancia, Maserati and Abarth) and of 3.5 million including production developed in joint-ventures (mainly in Turkey). Such growth, clearly ambitious, will also require a very high increase in the registrations for the Alfa Romeo and Lancia brands that in the 2007–10 period should achieve an 87.5 per cent and 150 per cent increase respectively (compared to 2006 sales levels). On the economic and financial front Fiat Automobiles targets are set for the period leading up to 2010 (Figure 15.3) to sales of about a32.5 billion, versus the a23.7 billion recorded in 2006, and a trading margin on sales growing from the range 0.8–1.3 per cent on sales in 2006, to the estimated range 4.5–5.3 per cent in 2010 (Figure 15.3). For the time being the most up-to-date results available relate to the first nine months of 2007. They indicate that the objectives stated in the company turnaround plan have been met and indeed exceeded, both in terms of vehicle sales and in terms of financial results. During the first nine months of 2007 vehicle deliveries grew by 13.8 per cent from the figure for 2006, for a total of 1.66 million units and revenues of a19.6 billion. During the same time frame the trading profit of Fiat Automobiles grew from a374 million to
302 The Second Automobile Revolution (€bn) 40.0
5.3% 4.4%
30.0
10.0
3.6% 1.3% 19.5 0.8%
0.0 2005A
2006E
2.1% ˜25.5 2007E
˜28.0
3.9% ˜30.5
4.0% 3.0%
˜32.5
2.0% 1.0% 0.0%
2008E
2009E
2010E
⫺10.0
⫺1.0% ⫺2.0%
⫺1.4%
⫺3.0%
⫺20.0 Revenues
Figure 15.3
6.0% 5.0%
4.5%
2.9%
20.0
4.7%
Trading margin range - Low
Trading margin range - High
Fiat Auto financial targets
Source: Fiat Auto, November 2006
a570 million with a trading margin at the end of the period of 2.91 per cent. In other words, Fiat Automobiles has achieved at the end of the third quarter the highest value within the range assumed by Sergio Marchionne for the end of 2007, as shown in Figure 15.3.
The other objectives under Marchionne’s lead In the face of the expanding presence across all segments by premium brands as already mentioned, the equally strong current expansion of Korean brands, and the entry of Chinese and Indian brands, the achievement of Fiat Automobiles’ objectives implies the need for Fiat to reposition its offering upwards for all three brands, but especially for Lancia and Alfa Romeo, which were given extremely ambitious targets. Unless it is possible to command a premium price for a significant share of domestic production it will become harder and harder to maintain manufacturing operations in Italy. Besides, it must be reiterated that the success of this programme does not depend solely on the initiatives promoted by Marchionne, who undoubtedly has proved himself able to lead the group with great determination, but also on competitors’ actions. All major automakers are very committed to heavy programmes of investment and model range renovation, with growing R&D expenses. For example, the Audi division in the Volkswagen Group announced in 2006 the launch for the 2007–11 period of an investment programme of a8.4 billion to expand its range to forty models. The recovery of Fiat’s market share in the European markets might also benefit from the current difficulties of two of its most direct competitors, Renault and PSA. However, those two French groups are in turn developing massive recovery plans. Carlos Ghosn, CEO of Renault, announced in March 2007 the launch of twenty-six novelties over
Fiat Group Automobiles 303
the next three years (novelties include ‘new models’ and ‘silhouettes’) and Christian Streiff, new CEO of PSA, has announced an objective of 5 million vehicles sold by 2015, and twenty-nine new models in Europe between 2007 and 2010, including renovation of existing models and range extension. For these reasons the upgrading of the Fiat brands appears an uphill task that will have to be pursued gradually, without leaps. For example, the launch of the new Lancia model Thesis in 2001 represented a mistake not to be repeated. It would have been much better to invest in a renewal of the Lybra model, whose late introduction had compromised its success, than to propose a model such as the Thesis, which was inadequate to play the over-ambitious role it had been given. To develop in a gradual and fine-tuned but systematic way, the strategy of upgrading requires not just industrial agreements, such as those recently signed with Ford and with the Indian company Tata, which focus on cost reduction and investment sharing, but also a strong commitment to research and development aimed at product innovation, since it is clear now that in the most advanced markets only innovation is capable of allowing the acquisition of a premium price. Fiat has demonstrated, mainly in the design of diesel engines, the necessary innovative capabilities, but the task is to widen these capabilities. Over recent years, as already said, Fiat Automobiles has been required to reduce the scope of its own research and development initiative, acquiring more innovation than in the past through outsourcing. This approach, although useful in achieving a speedy economic and financial equilibrium, must however be replaced with a stage of image upgrading through a greater emphasis on internal innovation. It is all too clear that to rely on the marketplace to acquire innovations will not enable Fiat to overtake its competitors. Currently the innovation expenses are still too limited (less than half those of its main competitors), but if the R&D spending were to double, as it should, the share on the cost per manufactured unit would increase, thus eroding margins. Facing this dilemma, the most reasonable solution would be to carry out, besides industrial agreements, a financial agreement to source the additional resources necessary to fight competition simultaneously on two fronts: greater product innovation (mainly aimed towards medium-high segments) and volume growth investing in networks abroad. In the years of the crisis Fiat Automobiles followed a policy of dramatic reduction in its distribution networks, both in Italy and abroad. In part this choice was conditioned by the need to concentrate sales on a smaller number of dealers in order to support their profitability, a situation that was exacerbated by the loss of competitiveness of the product from the quality standpoint, and by very costly sales support policies (fleet sales, zero-km used cars, etc.). Fiat must then manage to obtain greater financial resources through: (a) The development of new models. In the 2007–10 Product Plan, disclosed in November 2006, twenty-three new models and sixteen model upgrades
304 The Second Automobile Revolution
are envisaged. This is undoubtedly an ambitious project, but the fact that it does not include either the substitution of the Lancia Lybra (the new Delta model aims at a completely different customer segment) or the launch of an SUV for the Alfa Romeo brand, represents two weaknesses that should be overcome. (b) The development of more powerful engines, but at the same time more fuel efficiency and lower CO2 emissions. (c) The expansion of distribution networks, mainly for the Alfa Romeo and Lancia brands that are currently under-represented in Europe. In the meeting with financial analysts in November 2006 the manager in charge of dealer development announced that the total number of dealers in Europe (4,400 in 2006) will increase to 4,800 in 2010. This is a step forward, but that appears small in terms of the need for a complete consolidation of the company over the long term. It is not by chance that network sizes for its major competitors are much larger, starting from Renault24 and PSA. (d) The strengthening of marketing policies abroad, where the number of Fiat vehicles in circulation is much smaller than in Italy and therefore does not provide the same level of visibility to the new offering of the Turin company. Marchionne’s management has been capable of carrying out miracles with a very limited set of resources, particularly thanks to the ‘doing more with less’ philosophy that is working very well, but over the medium to long term it seems necessary to move towards forms of growth that are supported by a broadening of investments and innovation initiatives.
2,162,900 1,963,500 1,829,800 1,786,600 2,201,100 2,235,000 2,336,200 2,786,000 2,467,000 2,414,000 2,437,000 2,193,000 1,954,000 1,875,000 1,881,000 1,847,747 2,113,165
1,949,600 1,719,500 1,546,400 1,164,600 1,413,600 1,531,100 1,430,500 1,658,846 1,494,467 1,499,258 1,488,116 1,323,212 1,185,788 1,080,915 913,358 826,704 996,809
Worldwide Domestic
213,300 244,000 283,400 622,000 787,500 703,900 905,700 1,127,154 972,533 914,742 948,884 869,788 768,212 794,085 967,642 1,021,043 1,116,356
Production abroad
303,238 287,957 284,102 261,021 248,179 237,426 238,135 242,322 220,549 221,319 223,953 198,764 186,492 162,237 161,066 173,695 172,012
World Fiat Group
World Fiat Auto Group
Italy Auto Branch
World Fiat Group
Net income2
Investments2
R&D2
551 701 688 594 482 493 502 535 608 711 776 870 861 939 952 665 675
World World World World World World World Fiat Fiat Fiat Fiat Fiat Fiat Fiat Auto Group Auto Group Auto Group Auto Group Group Group Group
Turnover 2
133,431 116,274 29,546 14,293 1611 468 2,175 1,032 1,162 128,925 111,168 29,174 14,206 817 83 2,160 1,115 1,291 125,378 104,289 30,526 14,175 285 −281 3,061 2,027 1,343 120,338 95,627 28,176 12,937 −921 −858 3,439 2,616 1,160 119,618 84,698 34,005 17,148 522 53 2,351 1,589 996 114,386 77,704 39,092 20,190 1109 301 2,919 1,700 562 116,144 75,518 40,244 21,950 1225 −100 2,746 1,677 1,129 118,109 72,894 46,257 26,202 1248 384 2,398 1,341 1,166 93,514 65,009 45,769 24,859 621 −151 2,418 1,373 1,160 82,553 55,778 48,122 24,101 353 −261 2,709 1,464 1,406 74,292 49,350 57,555 25,361 664 −304 3,236 1 412 1,725 55,174 36,715 58,006 24,440 −445 −1418 3,438 1,331 1,817 49,544 32,786 55,649 22,147 −3948 −2739 2,771 1,115 1,748 44,563 28,413 47,271 20,010 −1900 −2038 2,011 1,100 1,747 45,121 28,543 45,637 20,539 −1579 −2019 2915 1,792 1,791 46,099 29,136 46,544 21,275 1420 −1191 3,520 1,582 1,558 44,691 28,720 51,832 25,577 1151 363 4,312 2,163 1,598
Production Passenger cars and light commercial vehicles Workforce1
Fiat, 1990–2006
Notes: 1. Staff members are as at 31 December of the year concerned. 2. In million of a. Sources: Fiat Annual Reports.
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Appendix Table 15.1
305
306 The Second Automobile Revolution
Notes 1. Fiat Auto, a division of Fiat Group, in charge of automobile activities with the brands: Fiat, Alfa Romeo, Lancia, Maserati, Abarth and Fiat Professional (light commercial vehicles), became Fiat Group Automobiles on 1 February 2007. Here we use the name ‘Fiat Automobiles’ even for the time before the change or simply Fiat. 2. In the years of strong economic development following the Second World War Fiat had undergone a stage featuring typical elements of the Fordist model. For a description of the conceptual structure of this model see Boyer and Freyssenet (2002). On the economic and industrial history of Fiat there is a large bibliography. Among the contributions we recommend are Castronovo (1999), Camuffo and Volpato (1998) and Volpato (2003, 2004). For a more comprehensive view of the recent evolution and its reorganisation see Volpato (2008). 3. A very interesting analysis, from an insider, of the power struggles that took place in the management of Gruppo Fiat over those years can be found in Garuzzo (2006). 4. The presence of two or more cars within the same family. 5. A good example of such a policy is represented by Renault’s choice in 1996 to manufacture five variants of the Mégane model on a single platform. 6. Among the most significant cases of such ‘trading-down’ towards lower segments where Fiat was particularly present are: the launch of Ford Fiesta in 1976, followed by Ka in 1996; the launch of Citroën Ax in 1986; the launch of Opel Corsa in 1982; the launch of Renault Twingo and Nissan Micra in 1992; and Volkswagen Lupo in 1998. 7. This type of politics, that sees car purchasing as a sign of wealth to be heavily taxed, arose at the start-up of the automotive industry, when probably it was meaningful, but then it survived indefinitely, both during fascism, and into the present stage. This led to a preference for smaller cars among Italians, less affected by the fiscal regime. Volpato (2000). 8. For more details on this analysis see Volpato and Zirpoli (2006). 9. Models: Bravo and Marea for Fiat, models 145 and 146 for Alfa Romeo and renewal of the Delta model for Lancia, but lacking the renewal of the Croma model for Fiat. 10. Total costs of these rebates were estimated by Giancarlo Boschetti, late CEO of Fiat Auto, at about 23.7 per cent of turnover, plus the costs of warranty work at about 7.3 per cent: Boschetti (2002). 11. Volpato (2004). 12. Volpato (2000). 13. Balcet and Enrietti (2002). 14. On the organisational forms of model development in Fiat Auto see Calabrese (1997). 15. On the development during the 1990s of simultaneous activities of concentration in the number of supplier firms and upgrading and greater delegation to those confirmed, see Enrietti and Lanzetti (2002). 16. Volpato (2004). 17. On organisational aspects of the Fiat–GM agreement and on the outcome of such cooperation see: Camuffo and Volpato (2002); Enrietti and Barichello (2006). 18. During the two years following his entry as CEO of Fiat Spa, Sergio Marchionne has sealed thirteen international agreements, of which ten involve Fiat Auto.
Fiat Group Automobiles 307 19. To illustrate the economies generated by the digital mock-up we could note that the first physical mock-up of a new model costs between 500,000 and 1 million euros. 20. All new models sold by Fiat Auto in Europe have gained the Euro NCAP’s five star rating. 21. Among the best-known tools implemented in an integrated approach by the Fiat Group Automobiles Production System are the Kanban, Failure Mode and Effects Analysis (FMEA), Poka Yoke, Six Sigma, etc. See Massone (2007). 22. In city-cars Fiat do Brasil produces the Mille, Palio and Idea models. The Idea was awarded ‘Carro do Ano’ (Car of the Year) in 2006. 23. At the end of 2006, the employees numbered 44,691 and the share was 4.82 per cent for Gruppo Fiat Auto. The same data were respectively 125,827 and 14.08 per cent for Renault Automobile and 139,500 and 13.79 per cent for PSA Automobile. 24. Renault declared a network of 9,764 point of sales in Europe as a sum of direct dealerships, independent dealers and agents (2006 Annual Report).
Bibliography Balcet, G. and Enrietti, A. (2002) ‘The Impact of Focused Globalisation in the Italian Automotive Industry’, Journal of Interdisciplinary Economics, 13(1–3). Boschetti, G. (2002) ‘Interview’, in Professional: Periodico di informazione manageriale del gruppo Fiat, no.12. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Calabrese, G. (1997) Fare auto: La comunicazione e la cooperazione nel processo di sviluppo prodotto. Milan: F. Angeli. Camuffo, A. and Volpato, G. (1998) ‘Making Manufacturing Lean in the Italian Automobile Industry: the Trajectory of Fiat’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Camuffo, A. and Volpato, G. (2002) ‘Partnering in the Global Auto Industry: the Fiat– GM Strategic Alliance’, International Journal of Automotive Technology and Management, 2(3–4): 335–52. Castronovo, V (1999) Fiat 1899–1999: Un secolo di storia italiana. Milan: Rizzoli. Garuzzo, G. (2006) Fiat: I segreti di un’epoca, Rome: Fazi Editore. Enrietti, A. and Barichello, F. (2006) ‘Fiat–GM Powertrain: chi ha guadagnato dall’alleanza?’ AA.VV., Annali di Storia dell’impresa, no.17, Marsilio Editore, Venezia. Enrietti, A. and Lanzetti, R. (2002) ‘Fiat Auto: le ragioni della crisi e gli effetti a livello locale’, Quaderni del Dipartimento di Scienze Economiche Hyman P. Minsky, Bergamo, No. 7. Available at: www.westerni.unibg.it/dse/Quaderni/2002/200207.pdf. Massone, L. (2007) Methods and Tools for the Fiat Auto Production System. Turin: Fiat Group Automobiles. Volpato, G. (2000) ‘La filière automobile italienne vers la globalization’, in G. Dupuy and F. Bost (eds), L’automobile et son monde. La Tour d’Aigues: Éditions de l’Aube. Volpato, G. (2002) ‘Una crisi che viene da lontano – Il marketing strategico di Fiat Auto’, Economia e politica industriale, 29(116): 63–86.
308 The Second Automobile Revolution Volpato, G. (2003) ‘Fiat Auto: From “Forced” Internationalization Towards Intentional Globalization’, in M. Freyssenet, K. Shimizu K. and G. Volpato (eds), Globalization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan. Volpato, G. (2004) Fiat Auto: Crisi e riorganiszazioni strategiche di un’impresa simbolo. Torino: Isedi. Volpato, G. (2008) Fiat Auto: Un’araba fenice nell’industria automobilistica internazionale. Bologna: Il Mulino. Volpato, G. and Zirpoli, F. (2006) ‘Fiat Auto: Crisis and Resurrection? A Critical Analysis’, Finanza, Marketing, Produzione, 24(4): 106–22.
16
From the Marriage in Heaven to the Divorce on Earth: the DaimlerChrysler Trajectory since the Merger Holm-Detlev Köhler
‘We invented the automobile. Finally we will be at the top again.’ (CEO Dieter Zetsche at the end of his inaugural speech, September 2005)
Introduction This chapter outlines the conditions and objectives of the merger between Daimler-Benz and Chrysler in 1998 as the point of departure for a turbulent restructuring story which ended up in the sale of Chrysler to the private equity group Cerberus Capital in May 2007. The company never became really integrated and the German management had to take over the responsibility for the whole group with Chrysler, Smart, Freightliner, Mitsubishi and Hyundai in deep crisis. The nine years of DaimlerChrysler are mainly the story of the failure of former CEO Jürgen Schrempps’ ‘world company’ strategy. The permanent turnaround plans and restructuring programmes slowed down the ambitious internationalisation and synergy strategies and even forced the abandonment of some core projects. The DaimlerChrysler case shows the difficulties of transnational and intercultural mergers and of global corporate governance. The unintended outcome of the story is two new car groups, Daimler and Chrysler, clearly different from the pre-merger companies and with uncertain future. The analytical perspective adopted is a meso-political one concerning firms as complexes of power networks and interest constellations with a variety of internal and external actors involved. From this point of view, the DaimlerChrysler trajectory is a story of dynamic and contradictory interests, actors and alliances far away from neo-classical worlds of one-best-way optimisations and perfect competition in free markets. 309
310 The Second Automobile Revolution
The merger of two different companies On 6 May 1998 the ‘gamma-project’, one of the best guarded secrets in industrial history, was announced. A new star was born by the merger of the Mercedes-star and the Pentastar, the first ‘world company’, and similar attributes were awarded. The protagonists of the merger, Jürgen Schrempp (Daimler-Benz) and Robert Eaton (Chrysler), spoke of the birth of the leading twenty-first-century automotive company and The Economist (9 May 1998) wrote: ‘Once it was the Japanese who turned the world’s car industry upside down. Now it is the Germans.’ However, the following nine years of the merged company failed to fulfil any of the exaggerated expectations and turned out to be a turbulent story of crisis and repair management, abandoning step by step the ambitious world company strategy until reaching the final stage of divorce. Whereas Schrempp and Eaton spoke of a ‘marriage made in heaven’, others compared it to the match between Prince Charles and Lady Diana, a German aristocratic noble brand asking for the hand of a beautiful bride from the new world (Schneider, 2001). The former trajectories1 of the two companies had followed quite different dynamics. Daimler-Benz (DB) had been a pioneer of the German automobile sector, whose commercial vehicle business was internationalised soon after the Second World War, whereas the passenger car division followed the German export of premium articles model until the 1990s. In the 1980s, influenced by the oil crisis and the rising ecological movement, a new corporate vision named ‘Integrated Technology Corporation’ emerged in the DB headquarters under CEO Edzard Reuter (1987–95) to diversify its activities towards new potential growth sectors like financial services, electronics and aerospace. Several companies were purchased: AEG, Telefunken (consumer electronics), MTU (engines), Dornier, Fokker and MBB (aerospace, armaments); the Interservices AG (Debis) was founded and the automotive divisions (passenger cars and commercial vehicles) were converted into an autonomous company (Mercedes-Benz AG). In the mid-1990s the company passed through its most severe crisis in history with all non-automotive businesses incurring losses and abandoned the disastrous strategy. The new CEO Jürgen Schrempp (1995) designed the globalisation and restructuring strategy with a reconcentration on the automotive divisions and the sales or closures of other businesses. In the passenger car division the company pursued a globalisation and rationalisation strategy with four main elements: conversion into a full-sortiment producer, (re-)opening of the US market and the emerging markets, globalisation, and productivity and process improvement. In 1998, the small city vehicle, the Smart Compact Car, co-developed with the Swiss clock producer Swatch, was launched. Together with the new A-class, this model aims at the lower volume segments. Niche markets are to be supplied by the M-class (All Activity Vehicle) and the SLK roadster, the
The DaimlerChrysler Trajectory since the Merger
311
super-luxury segment with its own high-value model revitalising the old Maybach brand. The new greenfield plants in Tuscaloosa, USA (1994), production site for the new M-class, and in Juiz de Fora, Brazil (1999), designed for the A-class, were intended to strengthen the presence in both parts of the American continent. With regard to other emerging markets, mainly in Asia, Mercedes-Benz adopted the completely knocked-down (CKD) assembly as the main market access strategy. Mercedes assembles C-, E- and S-models in Vietnam, Mexico, Indonesia, India and South Africa for the local and regional demand. The pre-merger period was full of activities to leave the traditional German export model behind by building up production facilities (Tuscaloosa, Juiz de Fora, Hambach) and CKD-assembly plants abroad, internationalising the management (through expat-programmes and the ‘Corporate University’), looking for transnational alliances and adopting a new ‘global player’ image (from ‘made in Germany’ to ‘made by Mercedes-Benz’). Besides the ‘product’ and the ‘globalisation’ offensives, the Daimler-Benz report from 1995 announced two more strategic objectives: the ‘learning’ and the ‘productivity’ offensives. The new foreign union-free greenfield plants were not only market openers and new model producers but also laboratories for new production processes with very lean designs, flat hierarchies, teamwork and modular supplier organisations. The new corporate philosophy, involving value-added orientation, centre-organisation, project management, continuous improvement, benchmarking, strategic recentralisation/operative decentralisation, flexible work and time management, performance-based reward systems etc., was implemented in ways which transformed the organisational dynamic of the firm. The Chrysler trajectory was much more turbulent than the DaimlerBenz story and passed through several existential crises. ‘In a broad sense, Chrysler’s leaders reinvented the company each time economic conditions forced change’ (Belzowski, 1998: 243). The first reinvention occurred in the early 1980s after failed internationalisation attempts (the purchase of Simca and the British Rootes Group, and an alliance with Mitsubishi) and the impact of the oil shocks. The masterstroke of legendary Lee A. Iacocca revitalised Chrysler from virtual bankruptcy with the support of state loans and a tough downsizing strategy: 75,000 workers were laid off and twenty-one plants closed down. Chrysler’s acquisition and diversification adventure (Gulfstream Aerospace, Electrospace Systems, several financial services companies, Maserati, Lamborghini and the Jeep brand holder American Motors Corporation) ended in a second near-death financial crisis at the end of the decade. The second reinvention was based on a new product line, the minivans and light trucks, which gave Chrysler market leadership in the fastest growing segment during the 1990s, and on a restructuring in cross-functional project teams with a reconcentration on core competences. The 1990s converted the Chrysler
312 The Second Automobile Revolution
trajectory into an impressive success story (Womack and Jones, 1994: 96). From 1994 to 1997 Chrysler beat one historical record after another: several models were elected ‘car of the year’, Forbes crowned it ‘company of the year’ in 1996, the new CEO Bob Eaton was named ‘executive of the year’ by Automotive Industries, and former Chrysler chief executive Robert Lutz sold the success story entitled The Seven Laws of Business That Made Chrysler the World’s Hottest Car Company. However, the Chrysler strategy was risky, still limited to the North American market and not sustainable in the long run. The interest for the merger can be located in the acceleration of the ambitious ‘globalisation strategy’ in the case of Daimler-Benz and the assurance and consolidation of the risky ‘innovation and flexibility strategy’ in the case of Chrysler. Together they wanted to overcome their traditional development constraints in the globalisation race, and to open new overseas markets and market segments. The combination of Chrysler’s permanent innovation and Daimler-Benz’s high-tech quality profit strategies were to be the key to global leadership in the automotive industry. The main pillars of DaimlerChrysler’s world company strategy may be summarised as follows: integration of the two groups into one worldwide production and innovation network achieving technological, design and quality leadership in a variety of markets; strengthening the position of Mercedes in North America and opening up the European market for Chrysler; the presence in all relevant market segments from small city vehicles to super luxury cars and heavy trucks; entering the main Asian and other emerging markets; and concentration on the automotive business with complementary services. In the first weeks after the merger DaimlerChrysler looked extremely promising and powerful and the share price climbed to $108 in January 1999. But one year later, the merger had already been engulfed by an accumulation of problems, crises and tough turnaround plans with constant rumours of divorce for the very different automobile couple. Managing DaimlerChrysler in the twenty-first century became a hard battle against market trends, inherited strategic defects, intercultural conflicts and shareholder critics.
A three-world company: eight years of failed integration The merger created a completely new micropolitical situation which required new leadership, new stakeholder compromises and a new governance concept in order to implement a coherent profit strategy. This new micropolitical situation, however, became far too complex and the balance of the DaimlerChrysler decade shows a nearly complete failure of all main pillars of the ‘world company’ strategy replaced by ad hoc crisis management with very uncertain results.
The DaimlerChrysler Trajectory since the Merger
313
From the merger of equals to the German takeover (management problems) First of all, the announced ‘merger of equals’ turned out to be a new DaimlerBenz group governed by a German management team. The ‘new’ refers to the global player in a broad variety of markets and segments which DaimlerChrysler clearly contrasted to the former German luxury vehicle producer. But the two groups never became integrated,2 hardly any synergies could be materialised, and a lot of conflicts and problems emerged. The postmerger integration process mainly involved the departure of the Chrysler managers who had to be replaced by a German-led team. Bob Eaton and co. made an excellent deal: they gained huge extra profits from the merger, resigned, sold their stakes and left their marooned company in the hands of the DaimlerChrysler management. US union leader Steve Yokich confirmed later the impression of many analysts that Chrysler would be dead without the Mercedes millions (Schneider, 2001). The group never developed an adequate organisational structure. Schrempp was forced to restructure his management board several times which was not only reduced and Germanised but also became a battleground for the different German top managers with Wolfgang Bernhard and Eckhard Cordes as the most prominent victims. In 2006 a new leaner management model was implemented once again, looking for tighter integration and more cooperation. ‘Shorter, faster and leaner reporting channels and decision making’ (Company Report 2005) were to promote competitiveness and profitable growth. Thirty per cent of the management positions (about 6,000 jobs worldwide) were eliminated. The management board was reduced from twelve to nine members. R&D activities were centralised and directly linked to the car group. The Commercial Vehicles division was slimmed down to the truck group with the buses and vans added to ‘other activities’. The crash of the world company strategy (strategic problems) Several of the new DaimlerChrysler projects (Brazil, Smart) and strategies turned out to be complete and costly failures. Probably the clearest sign of strategy failure was the sale of the Mitsubishi and Hyundai stakes in 2005 after five years of expensive, unsuccessful refloating attempts. The alliances were once presented as the decisive step towards the world company: ‘With these brands we dispose of a highly attractive product portfolio in all markets and in all segments, from the micro car to heavy trucks. Only one company in the world is able to offer this: Daimler-Chrysler’ (DC CEO Jürgen Schrempp, Frankfurter Allgemeine Zeitung, 9 September 2000). Failed alliances in Asia: Mitsubishi and Hyundai In these five years Mitsubishi passed through a severe crisis, losing more than 20,000 of 62,887 jobs, and absorbing huge capital and German management resources without producing the expected results achieved by its competitor,
314 The Second Automobile Revolution
the Renault–Nissan alliance.3 When Mitsubishi needed another capital input in 2005, after a series of management scandals, recall cover-ups and strategic market failures, the DC management rejected this option under shareholder pressure and later on sold its stake for a970 million. The Mitsubishi crisis had bound the main management and financial resources for Asia and left the alliance with Hyundai without a future. The cooperation projects in the passenger car sector with Mercedes, Chrysler, Mitsubishi and Hyundai were soon abandoned whereas the promising truck alliances with Mercedes and Hyundai (common diesel engine developments and productions, share of distribution networks) failed after years of negotiation because of different strategic priorities. In 2004, DC decided to sell the 10.5 per cent Hyundai stake and to stop all alliance projects. The DC strategy for Asia crashed and there was no sign of a substitute strategy to tackle this weakness. The break with Mitsubishi also meant the failure of the most advanced synergy efforts as Mitsubishi shared components and platforms with Smart and Chrysler. One consequence is that Chrysler still lacks a fuel-efficient small car. The attempts to commonalise Mercedes and Chrysler models were unsuccessful due to the different image and model range. Mercedes components are too expensive for Chrysler cars and the models are mainly rear-wheel drive whereas all Chrysler cars have front-wheel powertrains. The only project that was undertaken, the Chrysler Crossfire, a model built on previous-generation Mercedes SLK technology by German contract coachbuilder Karmann, did not live up to expectations. The Mitsubishi and Hyundai deals crashed for different reasons, the former because of the inability of the group management to reinvent the struggling Japanese partner, the latter because of the maintenance of opposing strategic orientations; together they proved the inoperability of the world company project. The losses and troublemaker: Chrysler The Chrysler brand experienced a steady crisis immediately after the merger. It suffered huge losses, due to shrinking market shares and discount battles in the US market,4 ran out of steam in the high-margin minivan boom, responsible for the Chrysler success story in the 1990s, and failed to open up new markets which called for a reinvention of the brand. After 2001 the rumours about a possible Chrysler spin-off never ceased. As the American managers resigned, Chrysler’s turnaround became the masterstroke of later CEO Dieter Zetsche. A third of the workforce (40,000 of 122,000) was cut and six plants were closed down. In 2007 another three-year restructuring plan was announced, reducing the annual output by 400,000 units, closing an assembly plant and cutting shifts at other plants, with the ultimate loss of 11,000 hourly-paid jobs. Chrysler will also ‘reduce and optimise’ its dealer network and outsource non-core activities, axing a further 2,000 salaried posts. The tough turnaround plans brought Chrysler back to profit
The DaimlerChrysler Trajectory since the Merger
315
12 10 8 DaimlerChrysler Group
Million €
6 4
Mercedes
2 0 1998
Chrysler 1999
2000
2001
2002
2003
2004
2005
2006
⫺2 ⫺4 Year Figure 16.1
DaimlerChrysler Group profit development, 1998–2006 (a million)
Source: DaimlerChrysler Annual Reports.
in 2004 and 2005, when its main competitors Ford and GM struggled with severe problems. The US market share, which had fallen from 16.1 per cent (1998) to 12.8 per cent (2003) went up to 13.9 per cent at the beginning of 2005, but has been falling again since then. The turnaround was mainly the result of severe cost-cutting and offered no coherent brand strategy for the future. Chrysler has become cost-efficient and still needs a new successful reinvention (Figure 16.1). The management of the product portfolio renewal was chaotic and didn’t meet the market trends. In 2001, copying Mercedes, Chrysler implemented so-called product innovation teams for the development of new models. Three new brand directors were posted the following year and two years later again a new product-development organisation with new vice-presidents was introduced. In 2006 ten new vehicles were launched to consolidate the renewed profitable Chrysler brand based on a new minivan generation and hybrid vehicles, but the results tell another story: 2006 saw Chrysler back into the red. The constant struggles of the brand reveal a misled product strategy based on high fuel-consuming and overpriced cars such as the sporty van Pacifica or the Jeep Commander, whereas the market demanded more economical and fuel-efficient models. The renewed minivans and light trucks were also criticised for not offering a differentiated profile against the strong Asian competitors. Chrysler’s model and image upgrading strategy failed against the customer trends and fierce Japanese and Korean competition.
316 The Second Automobile Revolution
‘It’s like they’re selling cars by the pound,’ Gerald Meyers, a professor at the University of Michigan who once ran American Motors, now part of Chrysler, told the New York Times. ‘The heavier it is, the more money they can get. That’s not the way it works, fellows. I don’t think Chrysler has the strength or the talent to keep their future from eroding, because of the imports’ (justauto.com, 6 June 2003). Chrysler is still a North American brand with only 6 per cent of its vehicles sold outside the NAFTA region. Several attempts to introduce Chrysler models in Europe (with the relative exception of the PT Cruiser) failed. In 2007 a new attempt was announced for the Caliber, Avenger and Sebring models and the Nitro SUV, but all these middle-class segments are highly competitive and Chrysler still lacks a brand image. A real new reinvention of Chrysler requires huge efforts and investments, not just cost reduction, and the DC management and shareholders are unwilling to do this. Low profits but good performance: commercial vehicles The commercial vehicle division (Mercedes-Benz, Freightliner, Sterling Trucks, Western Star Trucks, Fuso), traditionally a low profit segment of the group, suffered a crisis at the beginning of the century with US Freightliner particularly affected. More than 11,000 jobs were cut in the US truck producer and the Mercedes truck division in Germany also reduced employment. The truck crisis revealed a structural management problem within DaimlerChrysler that had existed since the merger. The group management had to restructure Mitsubishi, Chrysler and Freightliner at the same time with German top managers sent overseas which slowed down other necessary modernisation processes. Cooperation with Hyundai, strategically very interesting for the commercial vehicle division, came to a standstill and the China plans were also delayed. In 2003 the truck joint venture negotiations with the Chinese First Automotive Works (FAW) failed and DC had to reorient the China strategy, starting negotiations with Beijing Automotive Industry Corporation (BAIC). ‘We can’t do everything at once,’ stated DC board member Eckhard Cordes (Frankfurter Allgemeine Zeitung, 31 May 2001). In 2007 DC achieved a joint venture contract with Taiwan’s China Motor Corporation and China’s Fujian Motor Industry Group on the production of minivans and light trucks in Fuzhou (Fujian). In 2004 the different truck divisions (Europe/Latin America, NAFTA, Powersystems, and with some months’ delay Fuso) were integrated into one worldwide truck group to share parts, developments and even platforms (Freightliner and Fuso). The spin-off of Mitsubishi Fuso Truck and Bus Corporation from Mitsubishi Motor Corporation and its subsequent integration in the DC truck group is the only positive outcome of the Mitsubishi adventure. After cost-cutting recovery programmes, the truck business currently benefits from worldwide growth in transportation and has increased its results. Daimler Trucks is well positioned in the main emerging markets with its own
The DaimlerChrysler Trajectory since the Merger
317
6
554,929
548,955
Million €
4
492,851
485,408
500,981
712,166
824,867
830,043
3,229,270
3,045,233
2,755,919
2,822,659
2,637,867
2,779,895
2,812,993
2,654,710
1,080,267
1,154,861
1,229,688
1,232,334
1,216,938
1,226,773
1,216,838
1,251,797
1999
2000
2001
2002
2003
2004
2005
2006
2
0 Year Mercedes car group Figure 16.2
Chrysler
Commercial vehicles
DaimlerChrysler Group unit sales, 1999–2006 (million)
Source: DaimlerChrysler Annual Reports.
production sites (Brazil, targeted for Russia) or alliances (India, China) and is market leader in Europe. The bus division (Mercedes-Benz, Setra, Thomas Built Buses, Orion) also matched favourable market conditions, particularly in the new EU member states (Figure 16.2). However, these positive results stem from very unstable economic situations and hide a low capital profitability, far below the shareholder value objectives. In the framework of the new management system implemented in 2005, the bus and van units were integrated in the ‘other activities’ division to give a clearer focus to the truck group. The commercial vehicle division, early internationalised in the Daimler-Benz group, is currently a more globalised and integrated division than the car groups (Hack, 2005: 268) but with smaller profit margins. The solid basement: Mercedes passenger cars The group performance depends to a large extent on upper-class passenger cars, a quite small Mercedes heart for the huge and diversified DaimlerChrysler body. The Mercedes Car Group stays quite solid with its core models (C-, E-, S-class) and the consolidated M- and A-class, although in 2005 there were some problems mainly due to model changes, the strong euro and quality issues (a costly 1.3 million recall action had to be undertaken). The quality problems in Mercedes cars were at least partly the result of the outsourcing strategy and the loss of competences. Mercedes had to contract electronics experts to recover know-how in several critical vehicle systems. Another Mercedes quality problem source was the huge number of engineers
318 The Second Automobile Revolution
sent to America to solve the Chrysler problems, but who were then missed in Germany. New models like the M- and A-class never achieved the usual Mercedes quality (Die Zeit, 2004 and 2005). A large part of the problems is due to electronics. It was wrong to rely completely on the system suppliers who supply a complete navigation system or engine management system and to trust that the individual components, when they are interlaced with one another in the car also function. We have strengthened our competences in the electronics sector, because otherwise we cannot guarantee the quality, for which MercedesBenz clearly stands. (Jürgen Schrempp, Der Spiegel 14/2005) However, the group has structural problems with the A-class in Brazil, the new Maybach and particularly with the micro car brand Smart. In 2005, facing fading market shares, the Mercedes Car Group started the CORE programme (‘cost down, revenues up, execution!’) to improve efficiency and increase earnings. Cost savings of a7.1 million per year and the loss of 9,700 jobs were reported as successful results of the programme (Frankfurter Allgemeine Zeitung, 26 September 2007). Apart from cost-cutting measures, a module strategy was announced for Mercedes to develop common parts and components across the model range. With the new R-class (Grand Sports Tourer) Mercedes is attempting to gain a unique profile in the North American market. It is striking that the US is a booming market for Mercedes (although the strong euro constrains profit margins) but a fading ground for Chrysler (Figure 16.3).
A/B
Car lines C/E S M/R Target:
Module management
Power train E/E, telematies Chassis
Reduced complexity Increased commonality
Exterior
Top quality
Interior
Competitive cost position
Body in white
Figure 16.3 Module strategy of Mercedes: from car line specific organisation to cross-modular organisation Source: Annual Press Conference, Auburn Hills, 14 February 2007 (www.daimlerchrysler.com).
The DaimlerChrysler Trajectory since the Merger
319
Whereas the sport utility vehicle model M-class, assembled in the first Mercedes plant in the USA (Tuscaloosa, Alabama) turned out to be a success, two other Mercedes projects resulted in clear failures and heavy burdens on the group results. The micro car brand Smart, produced in the new modular factory in Hambach, France, has been a huge failure and problem generator from the beginning in 1998. Presented as an innovative urban mobility project, it accumulated some a4 billion losses up to 2006.5 In spite of steady shareholder protests, the management persisted with the project with a new turnaround plan focusing on the new Smart Fortwo which will be available also in the USA, distributed by the big retailer United Auto Group. In 2005 the Smart roadster production was stopped, the offroad project cancelled and the whole management and sales structure was integrated into Mercedes, eliminating 700 jobs in the process. In 2006, the SmartForfour (four seats) was also stopped after two years of disappointing sales. The Forfour axe was extremely expensive as the company had to raise huge compensation payouts for the Dutch Mitsubishi/Nedcar joint venture plant where the model was assembled. The other failure is situated in the Brazilian state of Minas Gerais where Mercedes built a new plant, designed for 70,000 A-class vehicles per year and which should serve as entrance to the emergent South American passenger car market. The Juiz de Fora plant, opened in 1999 and announced as the most advanced automobile plant in South America (Eckardt et al., 2000), had to stop the A-class production and survives assembling other models without seeing a clear future. The middle-class model A-class could not find a demand in the South American market, centred in smaller and cheaper mass models and in 2001 Mercedes started the C-class CKD assembly for the North American export market in order to make a better use of the installed overcapacity, an operation requiring an additional $40 million investment. Finally, in 2005 the A-class assembly in Brazil was stopped and in 2007 the C-class Sports Coupé started leaving the assembly lines. Juiz de Fora will be the only plant of the group producing this model mainly destined for European and US markets. At the same time the Chrysler Dodge Dakota plant in Campo Largo, Brazil, had a similar fate, inaugurated in 1998 for 12,000 vehicles per year but producing only 5,000. On the one hand, the traditional upper-class Mercedes segment benefits from increasing demands in a world where products for rich and poor are booming and the classic middle-class segments are struggling. On the other hand, this segment faces increased competition from more and more brands and producers. The Japanese and Korean models gain market shares and traditional mass producers like VW, Ford and GM are trying to enter this segment with new acquired brands. The sales of the high-end luxury Maybach, for instance, are far behind the expectations.
320 The Second Automobile Revolution
Technological leadership and high profit margins: R&D and financial services Technological leadership is focused on considerable R&D investments in powertrain technologies, alternative propulsion systems (fuel cells, hybrid drive, lithium-ion batteries) and electronic vehicle systems. One result is the new E320 BlueTec, the ‘world’s cleanest diesel passenger car’, launched in 2006 in North America. The BlueTech diesel technology was first developed for the Atego and Axor truck series. Near Stuttgart, a new Van Technology Centre was opened to enhance the global van business. Another future technology project is the modular drive concept for electric vehicles BlueZERO, presented at the North American International Auto Show in January 2009. The supplier rooted quality problems have also motivated a programme to build integrated development teams with suppliers to maintain better control and central competences in the firm (Hack and Hack, 2005: 354). The Financial Services Division is the residual of the former diversified technology corporation. These divisions were sold out step by step, often with considerable losses, or closed down altogether (AEG). The latest sales were Debis Air Finance in June 2005 and parts of the 32.9 per cent stake of the European Aeronautic Defence & Space Co. (EADS), the parent of planemaker Airbus which ran into crisis in 2006. DaimlerChrysler thus returned to its origins as a company focused on the automotive business leaving behind the former dreams of a diversified technology corporation and integrated mobility and transport systems company. The current record earnings of this division are mainly due to business expansion and interest rate increases in the financial markets. In 2005 DC Automotive Finance started business providing vehicle financing in China. The division also holds a 45 per cent stake in the Toll Collect consortium, which in January 2005 launched an electronic toll system for trucks on autobahns in Germany, marked by several technical and political problems and conflicts. The German government demands millions in compensation for the delay. However, the financial and other services around the car business (leasing, e-business, second-hand retailing6 ) may increase in the future with higher profit margins than the core business.
The contested terrain: labour politics Labour politics are an important part of corporate governance in the big automobile groups in Germany and the US, although embedded in different institutional settings and capitalist models. The DC management has achieved quite consensual labour relations although very tough restructuring and job cutting plans had to be negotiated with the German IG Metall union and the American United Autoworkers (UAW); the latter were for the first time represented in a supervisory board as the IG Metall reserves one of its seats for the US colleagues, in accordance with the German co-determination law. Employee representation is organised in a quite hierarchical manner
The DaimlerChrysler Trajectory since the Merger
321
around Erich Klemm, since 1999 head of the works councils of the Sindelfingen plant (the biggest of the group), the German group (Gesamtbetriebsrat), the European works council (set up in 1996) and the World Employee Committee and at the same time vice-president of the supervisory board. The backing by UAW president Ron Gettelfinger of Zetsche’s turnaround plans at Chrysler, which implied severe job losses and concessions in wages and health care costs, is particularly striking. The US union made concessions in the health care and pension fund costs issue (in the US a company task lacking public social security) and the German IG Metall in flexible working time regimes that allow working shift adjustment to demand needs over the course of the year and among different plants. The unions seem to have accepted the capital logic of surviving in a competitive global market by putting profits and shareholder value first. In Germany a ‘Safeguarding the Future 2012’ agreement was signed in 2004 (after job cut and relocation threats from the management and worker strike actions), in which the works council made significant concessions in wage moderation and flexibility (resulting in a500 million labour cost savings per year) to get the guarantee that no German employee would be dismissed involuntarily until the end of 2011. All job reduction programmes have to be organised by voluntary leave and early retirement measures. The ‘Safeguarding the Future 2012’ agreement is related to the new uniform collective framework agreement for hourly and salaried employees in Germany (ERA) introduced in 2007. The work activities of approximately 125,000 employees have been re-evaluated in cooperation with human resources units, managers and works councils. In ‘Safeguarding the Future 2012’ it was agreed that after the introduction of ERA at DaimlerChrysler, the wage increase stipulated in the collective bargaining agreement would not be implemented in full. It was also agreed that the existing workforce at the group’s German locations would receive wage and salary guarantees and company compensation components if necessary. Another consequence of the agreement is that the wages of DaimlerChrysler employees covered by the collective wage bargaining agreement were reduced by 2.79 per cent in 2006. For senior managers in Germany, variable compensation for 2006 was reduced by 10 per cent in addition to the reduction of their regular monthly salaries. In a similar fashion to the group’s management system, the transnational labour relations are dominated by the strong German works council and the labour representatives in the supervisory board. As a reaction to the merger, a transatlantic working group composed of German (IG Metall) and North American (United Auto Workers and Canadian Auto Workers (CAW)) representatives was set up which significantly improved the traditionally hostile relations between these unions. In 2002 the World Employee Committee (WEC) was founded with European (six German, one Spanish), North American (three American, one Canadian), South African (one) and Brazilian
322 The Second Automobile Revolution
(one) union representatives. Whereas the European Works Council, a legally backed organ in the EU, has difficulties setting an operative agenda, the WEC has at least agreed with the group management on a code of conduct on social responsibility immediately after its foundation. The company agreed to enforce the agreement not only in-house, but also with its suppliers (Greer and Hauptmeier, 2006). Additionally, a labour committee was established comprising the employee members of the supervisory board (seven German employees, two from IG Metall and one from UAW) plus further delegates from UAW and CAW, who meet before the board meetings (four to six times a year) (Müller et al., 2004, 2005).7 The break up of the merger may well be followed by a failure of the new transatlantic union cooperation in DaimlerChrysler, a cooperation far from usual that failed in several other companies (e.g. General Motors). In Germany, which still has the bulk of Mercedes’ production facilities, there are well-established, fluent and professional industrial relations between the management and the works council. The resourceful and powerful German works council, with secretaries and academic employee support, also serves as an intermediary for non-German European worker representatives, because the EWC is not accepted as a bargaining body by the management. However, at least two signs question the success of the consensus bargaining model. In Germany there are increasing protests against a labour cost reduction strategy at the expense of the public welfare system. After cutting thousands of jobs, the company hired unemployed line workers whose wages were subsidised by the state (Der Spiegel, 17 May 2006). This has to be seen against the background that DaimlerChrysler is famous for avoiding any company taxes in Germany by certain accountability manoeuvres. Another public protest emerged as a result of the cutting of 3,500 apprentice jobs in 2005. The apprenticeship system is one of the cornerstones of the German industrial relations system. These conflicts have to be seen in the context of the union and works councils’ strategy to maintain high-skilled German workers even in the Taylorised line jobs, a battle which in Chrysler was lost in the recent job cut plans mainly oriented to older qualified salaried workers. In the German Mercedes plants IG Metall is proud to keep the principle of contracting mainly workers with vocational training degrees even in the assembly jobs. However, the continuous pressure on standardisation of the production system with strong neo-Taylorist elements (Köhler, 2003: 90ff.) leaves the high-skill union strategy in a somewhat paradoxical situation. On the other hand, there are a lot of grassroots groups and activities all over the group which try to organise transnational worker solidarity from the bottom-up in order to counteract the top-level politics between the German-led works council and the top management. This alternative DaimlerChrysler network receives support from churches and social foundations
The DaimlerChrysler Trajectory since the Merger
323
like the Transnationals Information Exchange (TIE), but not from the unions with the exception of the Brazilian CUT.
The end Eight years after the merger the main strategic objectives mentioned at the beginning had failed either partly or completely. DC never became the promised worldwide integrated production and innovation network and could not realise the goal of entering all important emerging markets with particular difficulties in Asia and Eastern Europe. The proposed concentration on the automotive sector was achieved mainly through the sale of failed ventures in other businesses. Although it achieved a presence with products in all relevant market segments through the broad model range, there remained important weaknesses such as the absence of Chrysler in smaller and fuel-efficient passenger cars or the Smart flop in the micro car segment. The Mercedes brand was significantly strengthened in North America but all attempts to open European and other markets for Chrysler failed. In sum, the world company project of Jürgen Schrempp had gone to pot and the happy marriage ended up in an unhappy divorce. On 14 May 2007, the sale of an 80.1 per cent majority stake of Chrysler for a5.5 billion ($7.4 billion) to the private equity group Cerberus Capital was announced as the final stage of its accumulated problems and failed integration attempts. Other interested equity funds (Blackstone, Centerbridge, Kirk Kerkorian’s Tracinda) and automotive groups (the main supplier of Chrysler Magna, GM) remained out of the deal. Daimler keeps a minority stake in order to develop several common projects started in the past. The bulk of the sales price flew into Chrysler in order to face the need for capital and to secure the huge employee pension and health care obligations that Chrysler had to take over which amounted to $17.5 billion at the end of 2006. The remaining German group adopted the name Daimler AG whereas Chrysler formed a new group named Chrysler Holding LLC. The financial conditions of the deal confirm the merger as a huge strategic failure and capital bleeding operation for Daimler. The remaining German group got $0.65 billion for a company acquired for $36 billion nine years before and the operating profit dropped down nearly a3 billion in 2007. The operation had to be additionally supported by a $1.5 billion credit line for the operative business of Chrysler and a $1 billion guarantee for the Chrysler pension fund. Winners of the deal are again managers like La Sorda and Ridenour, responsible for the Chrysler losses, but awarded huge extra bonus payments for the Chrysler sale. During these nine years Chrysler consumed a lot of technical, financial and human resources support from the Mercedes division. But Daimler has at least a solid industrial basis in the Mercedes car and truck groups while the future of Chrysler looks more than uncertain. The
324 The Second Automobile Revolution
company has to be reinvented again not just with union concessions and cost cutting but with a new model range and branding. Whether Cerberus is well prepared for this job as a financial investment company which needs to earn money in the short run, may be questioned. The private equity group Cerberus Capital, founded in 1992 in New York and headed by the former US Treasury secretary John W. Snow, may be a typical player in the new global automotive industry. Since the hiring of former VW and DC executive Wolfgang Bernhard in 2007, the investment group has acquired huge pieces of the crisis-ridden American auto firms like Tower Automotive and GMAC and also presented a bid for Delphi. Private equity companies are already strong players in retailing, steel and airline sectors and Cerberus may open the car industry for them with its iconic companies like Chrysler. In the shadow of these equity groups, investment assessment companies like J.P. Morgan, who shepherded the deal, arise as important industrial players. The UAW union, together with its Canadian counterpart CAW, tried to keep Chrysler as much as possible in the Daimler Group, fearing a break up by equity funds and even thought of organising an employee stock ownership plan. Immediately after the deal they had to change their attitude and start negotiations with the new owner on future recovery plans and particularly on the employee pension and health care liabilities. In October 2007, after a six-hour nationwide strike and against a strong inter-union opposition (demanding more guarantees on future production in the US), a four-year agreement was signed between Chrysler and UAW, covering about 45,000 active and 55,000 retired workers and 23,000 surviving spouses. The contract establishes a union-run trust to cover retirees’ health care and allows the company to pay lower wages to about 11,000 non-core, non-assembly workers. Whereas the American union leaders and workers fear a new wave of job cuts and plant closures, UAW president Gettelfinger, who had the seat on the twenty-member supervisory board at DaimlerChrysler, defended the deal as the best solution, which allows Chrysler to focus on its own future. From the Chrysler point of view the merger can be seen as a survival without reinvention. Without the German capital and engagement, Chrysler would be dead, absorbed by another partner or reinvented with a new profit strategy. Now it is still alive but with an urgent need for strategic and organisational restructuring. Daimler without the world company adventure, of which the merger was a central part, would be possibly a high performance and high profit company, selling premium cars, trucks and buses all over the world, a world market leader in upper-class car and truck segments. If the huge resources lost in Smart, Chrysler, Mitsubishi and Hyundai had been invested in the further strengthening and internationalisation of the Mercedes brand, it is easy to imagine Daimler-Benz as a model enterprise for successful business stories.
The DaimlerChrysler Trajectory since the Merger
325
To think of the future trajectories of the two groups immediately after the split requires a high degree of speculation, but at least some indicators may be considered. CEO Zetsche called the transaction a ‘new start for both companies’ (New York Times, 14 May 2007). Chrysler is again in a near-death situation requiring a new reinvention. The first measures adopted by the new owner Cerberus consisted of a change of top management, accelerated cost cutting and downsizing in the US, where market shares keep falling, and a strategic reorientation towards emergent markets, particularly China where new alliances were signed, but also Russia. The agreement with the Chinese Chery Automobile Co., an export-oriented, fast-growing car producer specialising in small, low-cost vehicles, aims at the development of new low-cost models for emerging markets in Eastern Europe and Latin America. If this strategic reorientation is confirmed, there will be hard times ahead for American dealers, workers and unions. Additional 12,000 job cuts to the 13,000 foreseen in the latest turnaround plan were announced at the end of 2007 and the model range will be slimmed down before newly developed fuel-efficient models bring Chrysler again on track. The new electric car unit ENVI is another attempt to meet the growing demand for fuel-efficient vehicles through the development of electric propulsion technologies. The surprising appointment of the controversial former Home Depot chief executive Robert Nardelli as new CEO in August 2007 leaves the reinvention of Chrysler in the hands of an automobile outsider, although assisted by the new vice-president Jim Press, an experienced Toyota executive. ‘The New Chrysler: Get Ready for the Next Hundred Years’, is the new advertising campaign; whether it becomes a reality, is more than an open question. The new Daimler Group is clearly focused on the solid Mercedes cars and trucks divisions and started its reorientation in a much calmer climate. Its share price nearly doubled after the announcement of the Chrysler sale. ‘We are able to do everything alone, although we remain open to alliances in different areas,’ is Zetsche’s new conviction (Frankfurter Rundschau, 15 September 2007). The growing demand for upper-class cars in emerging markets with huge social inequalities represents a considerable growth potential for Mercedes. After the separation from Chrysler, Mitsubishi and Hyundai, the two divisions have to reorganise their Chinese engagement looking for new partners in this crucial emergent market. In India, Mercedes already holds a leadership position in the luxury segment and plans to increase its production facilities. The clean bluetec diesel and BlueZERO electric motors should further increase Mercedes car sales in the traditional, more ecologically conscious European and North American markets. Together with competitors BMW and GM the hybrid drive technology will be improved and an alliance with Ford for the development of fuel-cell systems has been signed. Daimler and VW have each acquired a stake in the biofuel company Chorento in order to develop a market for second-generation biofuels.
326 The Second Automobile Revolution
Daimler also announced a breakthrough in lithium-ion battery technology for future automotive applications. A new A-class platform will be developed, perhaps with new alliance partners, to increase the model variety in the compact class segment in combination with the aforementioned modularisation strategy. So there are a lot of promising elements for a solid profit strategy based on the strong Mercedes brand, but there are at least two question marks: will the dispersed shareholder structure in an ever more volatile capital market allow a solid long-term Mercedes strategy and will the experience of twenty years of failed ambitious castle-in-the-air management projects really prevent current top managers from embarking on new adventures?
Discussion: the end of the ‘Deutschland AG’? In 2004 I adopted a political perspective to analyse firms as power networks and not just input–output machines for profit strategies. A firm complex may be defined as a historically formed and consolidated, i.e. trajectory bounded, set of relations between internal and external interests/interest groups. By their activities, ownership structure, management practices, conflict regulation forms and political relations to the stakeholders a firm develops a specific field of action, a corridor of not only possible but likely strategies in certain contexts. These strategies are not individual forms of interest persecution but temporary outcomes of interest and power relations, i.e. politics. (Köhler, 2004: 127) From this perspective DaimlerChrysler represents the end of a German governance model based on long-term cooperation among capital investors, firm management and works councils with active support from other stakeholders like governments and suppliers; a trend that has been dubbed ‘the end of the Deutschland AG’8 in academic debates. The post-merger difficulties and the retirement from the traditional Daimler-Benz financial partner Deutsche Bank, thus confirming the abandonment of Rhine-capitalist industry banking in favour of modern shareholder value financial management, left a fragmented group without clear strategic control. The biggest Daimler shareholder is now the Kuwait Investment Office with a 7 per cent stake and no strategic company interests. Top management has become as volatile as financial stock investment and the group lacks a leadership coalition able to impose a coherent long-term strategy. In this sense, the Cerberus deal marks a new step in the increasing role of financial investment players in leading global industries, a worrisome development for workers, unions and society. The micropolitical constellations in firm complexes come under pressure because of new actors with short-term financial profit interests undermining
The DaimlerChrysler Trajectory since the Merger
327
long-term governance compromises among owners, management, labour and other stakeholders. The profit strategy of DaimlerChrysler was much more a patchwork of different brand and regional policies than a coherent and coordinated corporate strategy. Perhaps the split was necessary for both parts to find a new coherent profit strategy and leadership. However, the margins to develop a coherent mid-term strategy are decreasing due to increasing shareholder pressure. The annual shareholder assemblies turned into protest meetings against the mismanagement which had halved the share price and multiplied the crisis brands since the merger.9 One reason for the merger was the fear of hostile takeovers by financial investors. The result, however, is the ownership of Chrysler by an equity group and a small Daimler AG floating in the stock markets without strategic shareholders. The DaimlerChrysler deal illustrates the fact that size is not the critical factor for global competitiveness and that mergers have a huge conflict and failure potential. After nine years of world company dreams, CEO Dieter Zetsche seems to have learned the lesson and answers a question on future acquisitions: For Mercedes, the passenger car business, we are at the top of the industry. Any integration of another brand would mean downward pressure. There is no need for an additional part to be added. For the brand and its profitability nothing is to be won by this. It seems different in the business with trucks and buses. There we are moving in different regions, as for the moment in China with Photon. Also in India I can imagine cooperation. (Frankfurter Allgemeine Zeitung, 20 May 2007) The small German competitors like Porsche (now main shareholder of VW) and BMW show much better performance than the big DaimlerChrysler group and in the case of BMW the breakdown of its merger with Rover was a necessary element of sustained success. However, these companies have an operative advantage over Daimler: their shareholder composition is controlled by families, which reduces the danger of takeovers and allows more strategic long-term planning. A Mercedes-centred Daimler AG could be a very successful automotive company after the sale of the non-automotive businesses and without the Chrysler, Mitsubishi and Hyundai adventures. The new Daimler AG is trying to get back on track, after nine years wasted and twenty-two years of strategic turbulence beginning with Reuters’ integrated technology corporation project in the 1980s.
0,39 0,62 −0,15
1,99 4,25 0,95 1,02 0,73 −0,22
2,70 5,19 1,06
10,316 6,230 6,21
1999
0,64 0,45 −0,28
2,87 0,53 1,15
5,213 3,480 3,47
2000
0,64 0,45 −0,28
2,96 −2,18 0,51
1,345 730 0,73
2001
1,240 1,329
7,47
3,126 −506 811
5,686 448 0,44
2003
9,64
3,020 1,317 1,76
5,829 3,329 3,30
2002
456
1,250
1,666 1,427 1,332
5,754 2,466 2,43
2004
591
−505 1,534(3) 2,093 (1,606) 1,468
5,185 2,846 2,80
2005
913(4)
1,714
2,415 −1,118 2,020
5,517 3,227 3,16
2006
Notes: (1) Adjusted, excl. one-time effects. (2) In 2006 the Commercial Vehicle Division was divided in the Truck Group and Van, Bus & Others. The figures in () for 2005 and for 2006 refer to the Truck Group. (3) The Chrysler profits in 2005 were impacted by the a240 million gain on the sale of the Arizona Proving Grounds vehicle testing facility. (4) From 2006 on, Others include Vans and Buses. Source: Annual Reports of DaimlerChrysler.
Services Aerospace Others
By segments Mercedes-Benz P.C. & Smart Chrysler Group Commercial Vehicles(2)
8,583 5,221 5,58
1998
Operating profit of DaimlerChrysler, 1998–2006 (a million)
Operating profit(1) Net profit(1) Earnings per share(1) in a
Appendix Table 16.1
328
99,459 129,395 90,082 34,133 26,240 46,107 34,297 7,225
466,938 241,233 123,928
1999
47,108
100,893 121,027 101,027 36,857 9,5891
416,501 196,861 123,633
2000
21,101
102,223 104,057 96,644 38,733 9,712
372,470 191,158 104,871
2001
Notes: 1. Decrease attributable to the sale of Temic. * Buses and vans passed from commercial vehicles (now: truck group) to other activities. Source: Annual Reports of DaimlerChrysler.
32.581
95,158 126,816 89.711 31.280 20.211 45.858
441,502 228,000 137,000
1998
Employees of DaimlerChrysler, 1998–2006 (31 December)
By segments Mercedes-Benz P.C. & Smart Chrysler Group MB Commercial Vehicles MB Sales & Marketing Services Aerospace Other Business Units Headquarters/Other Other Activities
Total Germany USA
Appendix Table 16.2
21,184
101,778 95,835 94,111 42,142 10,521
365,571 191,574 101,437
2002
20,636
11,224
11,035
20,192
105,857 84,375 114,602
384,723 185,154 98,119
2004
104,151 93,062 88,014
362,063 182,739 102,391
2003
18,164
11,129
104,345 83,130 117,183
382,724 182,060 97,480
2005
39,400*
10,718
99,343 80,735 83,237*
360,385 166,617 94,792
2006
329
330 The Second Automobile Revolution Appendix Table 16.3 (a million)
EBIT Net profit Employees (31 Dec) Unit sales Mercedes cars Daimler trucks Vans, buses, other
Key figures of the new Daimler AG Group, 2005–2007 2005
2006
2,873 4,215 293,839 2,041,704 1,216,838 509,299 315,567
4,992 3,783 274,024 2,072,885 1,251,797 516,087 305,001
2007 8,710 3,985 272,382 2,088,973 1,293,184 467,667 328,122
Source: Annual Reports of Daimler AG Group.
Notes 1. For an overview of the trajectory and internationalisation strategies of Daimler-Benz and Chrysler and the detailed merger process, see Köhler (2003). 2. Every year after publishing the bad Chrysler results, the same strategy was announced and never realised: closer collaboration between Chrysler and Mercedes in order to produce synergies and reduce costs. 3. It is an irony of history that Schrempp wanted to invest in Nissan in 1999 but was overruled by his management board, which argued against two big acquisitions at once. Probably the Nissan deal would have taken Daimler-Benz much further towards a global company than the Chrysler and Mitsubishi disasters. 4. In 2007 the medium discount for a Chrysler car in the US market was $3,962, $3,187 for Ford and $2,830 for GM. Honda and Toyota models had to be sold with $1,397 and $1,308 discount respectively (Reuters, 11 July 2007). 5. Recent reports speak of even higher figures: ‘DaimlerChrysler does not publish separate results for its Smart division, but according to a report, it has cost DaimlerChrysler a3.9bn in losses between 2003 and 2006, much more than analysts had estimated’ (just-auto.com, 12 June 2007). 6. In the USA and Germany DC ran its own second-hand car retailer brands called Starmark and Motormeile respectively. 7. The labour committee was a compensation for the shareholder committee established after the merger on the demand of the Chrysler management which wanted an exclusive management body preparing the meetings with the co-determined supervisory board. The German management had to negotiate this with the works council which accepted against the concession of a parallel employee body. 8. ‘The end of the Deutschland AG’ refers to a trend which undermines the main pillars of the German model of coordinated capitalism based on stable long-term stakes of the big banks in the main industrial enterprises and a consensus strategy with the main stakeholders (management, public authorities, financial entities and trade unions). 9. In 2003, the shareholder representative Ekkehard Wenger criticised Schrempp as the biggest capital destroyer in the company’s history and showed that the same investment (a511) in competitor BMW’s shares increased the value to a2,129 whereas in the case of DC it decreased its value to a409 (DPA, 9 April 2003).
The DaimlerChrysler Trajectory since the Merger
331
Bibliography Belzowski, B. M. (1998) ‘Reinventing Chrysler’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 242–69. Eckardt, A., Köhler, H.-D. and Pries, L. (2000) ‘Auf dem Weg zu global operierenden Konzernen?’ Erlangen: Institut für praxisorientierte Forschung und Beratung (IPRAS). European Foundation for the Improvement of Living and Working Conditions (2005) EWC Case Studies: the DaimlerChrysler Group. Dublin. Greer, I. and Hauptmeier, M. (2006) ‘Political Entrepreneurs and Co-Managers: Labour Transnationalism at Four Multinational Auto Companies’, paper, Cornell University (under review at BJIR). Hack, L. and Hack, I. (2005) Wissen, Macht und Organisation. Berlin: Sigma. Köhler, H.-D. (2003) ‘The DaimlerChrysler Deal: a Nice Marriage or a Nightmare?’ in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan, pp. 73–102. Köhler, H.-D. (2004) ‘The TNC as a Transnational Political Complex: Research Questions Stemming from the DaimlerChrysler and BMW-Rover Deals’, Actes du GERPISA, 36: 127–42. Lutz, R. (1998) The Seven Laws of Business That Made Chrysler the World’s Hottest Car Company. New York: John Wiley. Müller, T., Platzer, H.-W. and Rüb, S. (2004), Globale Arbeitsbeziehungen in globalen Konzernen? Wiesbaden: VS Verlag für Sozialwissenschaften. Müller, T., Platzer, H.-W. and Rüb, S. (2005) ‘Global Company Strategies: Global Employee Interest Representation?’, in F. Garibaldo and A. Bardi (eds), Company Strategies and Organizational Evolution in the Automotive Sector: a Worldwide Perspective. Frankfurt am Main: Peter Lang, pp. 57–82. Schneider, P. (2001), ‘Sieg der Sterne’, Die Zeit 36. Womack, J. P. and Jones, D. T. (1994), ‘From Lean Production to the Lean Enterprise’, Harvard Business Review, 72: 93–10.
17
Driving with Engineers’ Professionalism and Family Values: the BMW Trajectory from a Regional Carmaker to a Global Premium Player Ludger Pries
BMW is the world’s second most profitable carmaker in terms of the money earned per produced car (after Porsche). BMW experienced a worldwide growth in production and sales between 2000 and 2005 of more than 50 per cent up to 1.3 million cars – only Toyota has a similar high growth rate, but at an even higher level of about 7.7 million cars in 2005. Together with Mercedes-Benz, Lexus and Audi, BMW is one of the most important premium carmakers. BMW changed dramatically over only ten to fifteen years from a regionally producing and thinking company to a global player. How was this possible? What have been the driving forces of these fundamental changes of the consortium since the 1990s? The main argument to be developed in the following is that the successful BMW trajectory relies mainly on the impact of two specific social institutions: an engineer-driven professionalism and professional thinking on the one hand and family and networking-based commitments and values, on the other hand. The first argument stresses the technical approach and premium technical image of BMW as relying on the special weight of engineers. This could be shown looking at the predominance of engineers in CEO positions, the specific role of the FIZ in Munich, the genuine R&D strategies with suppliers, and the technical perfectionism in design and technical development. The second argument, the family and network-driven institutional embeddedness of the company, is based on analysing the family influence in the shareholders (Quandt), the network and long-term cooperation structures with suppliers due to historically very low added value, the management careers structured like a family (quite complete internal recruitment of CEOs as in Japanese carmakers etc.), the HRM strategies and the corresponding value and mission statements, and management–workers relations as family structured.1 To understand these two essential elements of the BMW 332
The BMW Trajectory
333
trajectory, the latter could be divided into three periods: the previous history of the automobile related activities of the Quandt family since the 1920s and until the existential crisis at the end of the 1950s, the stabilisation period of BMW as a German premium carmaker during the 1960s up to the 1980s, and the new phase of BMW becoming a global player since the 1990s.
Understanding the beginnings of the BMW trajectory To understand and explain the success of BMW in the last fifteen years or so it is necessary to sketch out its historical trajectory, mainly the existential crisis of the company during the twentieth century. In 1916, the Bavarian Aerospace Factories was founded as a manufacturer of airplane engines. In 1923 the company began producing motorcycles and in 1929 it began assembling cars (developed by BMW) after purchasing a car factory in Eisenach, Thuringia. As this factory supplied the Nazi regime with engines, it was dismantled in 1945.2 After re-establishing motorcycle production in 1948, BMW resumed car production in 1951, focusing on high-income consumers, which – following the war – was a very unsuccessful marketing strategy. Therefore, BMW changed its product structure and developed a minicar powered by a motorcycle engine, the Isetta. The Isetta turned out to be a very successful model, due to the increasing mass motorisation in Germany. Nevertheless, the losses from the sedan business could be only partially recovered by the production of small cars, and, as sales in this low-end segment decreased, BMW found itself once again in a deep financial crisis in 1959. The climax of the crisis was a takeover bid by Daimler-Benz supported by the Deutsche Bank (offers from General Electric, Ford and AMC had been previously rejected). This takeover bid failed, due to the resistance of the workers, their works council, retailers and small shareholders. As a consequence of this failed takeover strategy, the Quandt family bought the greater part of new shares that were sold to finance the development of new car models. The Quandt family represented an industrial and engineering tradition as they had made their fortune, for example, through developing batteries. Therefore, from the end of the Second World War the specific factors of the BMW trajectory were fixed. The finance capital-driven strategy of DaimlerBenz and Deutsche Bank was defended and an engineering and industrial family influenced company strategy. The Quandt clan hold about 60 per cent of the shares and therefore had a decisive influence on management decisions. Since then, BMW has strived to follow a long-term capital strategy, in contrast to, for example, Daimler-Benz. After entering mid-range vehicle production in 1963, the shareholders were paid a dividend for the first time in twenty years (Rosellen, 1983). In 1967, through the acquisition of the Hans Glass Company in Dingolfing, Bavaria, famous producer of the small car Gogomobil, production facilities in Dingolfing and Landshut were added to the main production lines and headquarters in Munich. These facilities had
334 The Second Automobile Revolution
previously been some of the main competitors in the small vehicle market and were also oriented towards the upper end of the market. However, they were unsuccessful due to the high customer loyalty within this market range. Under the leadership of chairman von Kuehnheim (formerly the deputy chairman of the Quandt-owned industrial plants in Karlsruhe), the company increased its annual production from 147,000 units in 1969, to over 500,000 units in 1989 (BMW, 1991). Primarily in the 1970s, the enterprise developed its characteristic product structure and the BMW brand image, which continues to be an important part of BMW today. Again, it was the logic of social networks based on the principles of family ties and of industrial engineering as a profession and a vocation that were decisive for the recovery during the 1970s and 1980s. The basic company identity and image was sharpened in this period and condensed in the product lines of the 3-, 5- and 7-Series. It built a reputation and self-understanding as a provider of innovative, sporty and comfortable vehicles. This was underlined with the gradual introduction of a new and successful product range: Series 5 (1972), Series 3 (1975), Series 6 Coupé (niche product, sports model) and the first Series 7 generation in 1977 (Lewandowski, 1998). Due to its fast-growing entrepreneurial success, BMW extended its production facilities. It is interesting that, until the 1980s, the new plants were opened exclusively in Bavaria: the first German greenfield plant in Regensburg in 1984, the new FIZ (the research and innovation centre) in Munich in 1986, an automobile parts production facility and a supplier park in Wackersdorf in 1989. Throughout the 1980s, BMW’s business activities expanded rapidly. At the same time, there was a comprehensive reorganisation within company management. First, there was a concentration on the most important development areas (development, technical planning, finishing technique, value analysis, cost control, purchasing, logistics, patent and personnel department) in the FIZ. Meanwhile the basic brand image and product lines and also the production facility structure in Germany were stamped until the 1980s; some really fundamental reorientations of the company strategy and structure took place afterwards in the 1990s. Until then, BMW had been a regionally (in Bavaria) producing company with an increasingly international reputation as a sportive premium cars producer. But during the last decade the company has changed dramatically in three main aspects. Firstly, the company reorganised its production and organisational structures towards flexibility and ‘lean’ production long before the ‘hype’ of Japanese production system strategies arrived in Europe. Secondly, maintaining and strengthening its premium segment brand image BMW managed successfully to differentiate its product structure. Last but not least, the company became more and more a global player in the sense of establishing a global production network. These three elements will be sketched out in the following section, and the two strategies of product innovation and of production internationalisation will be treated
The BMW Trajectory
335
in more detail afterwards, taking the example of the BMW Spartanburg plant in the USA.
The revolution of the 1990s: from a regional producer to a global player BMW was one of the first automobile companies in Europe to reorganise supplier relationships and production. Since the beginning of car production after the Second World War and due to the lack of a longer automobile producing tradition, BMW always bought a much higher share of a car’s value from suppliers – about two-thirds – than all other German carmakers. The first international purchasing office was created in 1983 for global procurement. By the late 1970s, BMW had begun to secure long-term business relationships with suppliers. This was followed by the development of system suppliers in the mid-1980s, as well as the systematic introduction of just-in-time deliveries in the late 1980s. During the 1990s the co-development of modules and sub-systems with selected suppliers was strengthened by investing in suppliers’ technical capabilities and by organising working groups of BMW and supplier engineers at a very early stage of product development (Eckardt et al., 2000; Pries et al., 2005). Today, products and manufacturing methods are developed simultaneously through close cooperation with system suppliers. BMW increasingly defines itself as a ‘system integrator’, which, in the future, will integrate model and comprehensive brand component concept development in strong and direct cooperation with first-tier suppliers. The portion of in-house production, and/or manufacturing, has been decreasing accordingly in the past years and represents about one-quarter of production value at present. This very low added-value share gives BMW a high level of product and production flexibility which is complemented by the very high flexibility of the BMW production itself. Since the 1980s, BMW has undergone many changes in order to achieve the goal of attaining comprehensive flexibility within organisations, production and work structures, in the context of increased standards for product complexity and temporal reaction. Reorganised company and activity structures should simplify the leadership of complex processes. Even at the end of the 1980s, sub-sections of the company were already combined into cost and profit centres. The large production structures, which were developed during the period of growth, were to a large extent reorganised into independent cost and business units, as well as areas not connected to a particular location, which carried full responsibility for their own results. Fields of activity were also transformed according to new process and project principles. Trans-sectoral sub-tasks were integrated into process chains – in the production department, for example, secondary functions were integrated (manufacturing, maintenance, material provision and quality assurance). Employee activities were also increasingly
336 The Second Automobile Revolution
aligned into horizontal process chains. The development department with its interdisciplinary project groups represents the beginning of this change. These reorganisations went hand-in-hand with the concentration on central processes. Thus, all necessary requirements and activities for the production of prototypes were integrated into the research and innovation centre in Munich (FIZ). Today, alongside the fields of automotive and motorcycle production, the BMW group also offers financial services (BMW Financial Services). Concerning the spatial distribution of its production facilities, BMW had been characterised by the most regionally concentrated location of its plants. In terms of productive capacity distribution, it was more a Bavarian than a German carmaker. BMW had already gained its first international experience in the late 1950s in Portugal: in order to avoid considerably high tariffs an assembly plant was built there. Since 1973 there has also been a BMW plant in Rosslyn, South Africa, which was originally supposed to produce leather parts for the whole consortium in addition to the assembly of CKD (completely knocked down) sets for the local market. This plan failed initially because of persistent quality problems. A further step was the decision to set up an engine production site in Steyr, Austria, in 1980. The opening of an R&D centre for diesel engines there was followed by the start-up of a plant for engine manufacturing with a capacity of 150,000 units per year. In spite of these international activities, BMW could be referred to as an almost exclusively regional car manufacturer with a clear strategy of exportation of cars manufactured in Germany. It was not until 1990 after the unification of the GDR and FRG that BMW bought a plant outside Bavaria in Eisenach, Thuringia. This production site – the Fahrzeugfabrik Eisenach – had been bought in 1928 by BMW already, but then had been expropriated after the Second World War due to the BMW concentration on aircraft engine production during the war. The production structure was further spread and decentralised inside Germany with the newly opened production plant in Leipzig, Saxony, in 2005. But more impressive is the internationalisation of production since the 1990s which was narrowly combined with a strategy to diversify the product structure. Since the 1990s the BMW consortium became aware that it had reached a growth and production level that was not sustainable in the long term. During the intensive international waves of mergers and acquisitions the company either would have to grow or would have to look for an ally or merger. As competition in the car markets turned increasingly global and international competitors such as the Japanese and South Korean carmakers entered the European markets, BMW could not just follow the traditional trails and strategies. Even in the premium segment competition intensified with the growing strength of the premium brands of Honda and Lexus (Toyota) as well as Audi (Volkswagen). For BMW it was obvious that it had to differentiate its (premium) products and to turn from a ‘made in Germany’
The BMW Trajectory
337
slogan to a ‘made by a German company’ motto. With the advent of Bernhard Pischetsrieder as CEO since 1993, both the production internationalisation and the product diversification strategies were intensified. A crucial and very visible step of internationalising BMW’s trajectory was the acquisition of the British Rover group in 1994. This former British Aerospace affiliated company consisted of four full production sites and seven assembly plants; it was purchased for about 2 billion Deutschmark plus the debts of about £900 million.3 This acquisition (and the strong competition for the Rolls-Royce brand with Volkswagen) reflected the strategy of becoming a full range producer (from small up to luxury cars) at an international level. But the acquisition of the British Rover group turned out to be a complete failure in both senses: as product diversification and as production internationalisation. From the standpoint of BMW, the Rover brand did not prove itself as a premium brand, and the British production sites were not as productive as expected. In 1998 the losses of the Rover group had not disappeared but reached a record of 1.87 billion Deutschmark, reducing the very successful consortium net profit results by almost a third. Contested strategic alternatives for the Rover group were represented by Wolfgang Reitzle and Joachim Milberg at the end of the 1990s. Whereas the former was going to reduce the Rover engagement to the successful brands of Land Rover and Mini, the latter promised to safeguard the Rover group as a whole. Joachim Milberg was elected as the new CEO in February 1999, mainly due to the resistance of the workers’ representatives in the supervisory board against Reitzle and his strategy to ‘fillet’ the Rover group and maintain only the profitable units (whereas Milberg represented the continuity of Pischetsrieder’s strategy of maintaining Rover as a ‘full sortiment producer’). Nevertheless, with Milberg the direct BMW management intervention into the Rover company increased, and the Rover group was restructured according to the global brand and production strategy. Due to continuous losses (the pound–mark currency relation has increased by almost 50 per cent since the takeover) the Rover company (including MG and later on also Land Rover) was sold in 2000. BMW only maintained the Mini group (Eckardt and Klemm, 2003). After the Rover disaster and its sellout in 2000 (for a symbolic £10) the BMW group concentrated exclusively on the premium segment strategy. Based on the negative internationalisation experiences, the consortium sharpened its product innovation and product differentiation strategy based on the established 3-Series structure and, at the same time, recovered its own tradition of sportive car production, entering the growing market of sport activity vehicles (SAV; the term was chosen to distinguish the premium segment from the traditional sport utility vehicle type of cars). The sportive Z 3 car was developed in Germany and production began in an all new greenfield plant in Spartanburg, USA, in 1994. In 1998, in the same plant the production of the new SAV X 5-Series began. When opening formally the new production plant
338 The Second Automobile Revolution
in Spartanburg, CEO Pieschetsrieder announced programmatically: ‘We are not going to build a new production plant in Germany, and Spartanburg is probably not the last we are going to build abroad’ (Wirtschaftswoche 1/2, 5 January 1995, p. 28). After some initial and severe quality problems the Spartanburg plant and the new model lines were very successful, and the Z- and the X-Series later on were differentiated, e.g. between Z 3 and Z 4, as well as X 3 and X 5 types. In the case of Spartanburg, the ‘double strike’ of product innovation and production internationalisation – new Z 3 and new X 5 in a new Spartanburg plant – proved to be a very good strategic choice. With additional net investments BMW also expanded its assembly plant in Rosslyn, South Africa, into a full production site and integrated it into the international production network (annual goal for the end of the 1990s: 45,000 3-Series BMWs; 4,000 Land Rover Defender). The strategy to become a real global producer consisted of three main features: (a) integrated full vehicle production in the main sales markets of Europe, North America and Japan and a long-term sales potential of at least 50,000 units/year; (b) CKD assembly plants for preparing market entry in countries where CKD vehicles are 20 per cent cheaper than imported vehicles because of higher tariffs and high shares of local manufacture of components; and (c) a double strategy for all production sites abroad: manufacturing of at least one world model and one for the specific region (Antrecht, 1995: S.52). According to this strategy, BMW pushed its CKD activities in regions outside the traditional Triad markets with predicted economic growth. In 1996 there were reports of a CKD assembly plant for the models of the 5Series in Egypt in cooperation with a local partner and 40 per cent of locally added value. In Thailand BMW decided to incorporate a subsidiary despite the Asian crisis and set up an assembly plant in an industrial park near Bangkok, with a capacity of 10,000 units a year. Also in Asia new assembly lines for CKD were erected, namely in Hanoi (Vietnam), Singapore, Thailand and Indonesia. Furthermore, there were investment activities in Malaysia. The concentration of FDI in Southeast Asia points to the regional balance point of BMW’s globalisation strategy. In the late 1990s BMW also conducted its business in Russia by installing a CKD assembly line (and a sales company) with a goal of 10,000 BMWs (Series 5) and Land Rovers per year in Kaliningrad. Besides new production and CKD assembly plants, the further internationalisation of procurement became a strong part of BMW’s strategy. The company improved and internationalised its organisation of procurement (global sourcing) and simultaneously strived to reduce the number of its suppliers (single sourcing). During the 1990s, therefore, the seven BMW offices in Spain, Italy, Great Britain, Australia, Canada and Mexico and the international procurement organisations in North America, South Africa and Steyr, Austria, were linked more closely with the department for supplier development.
The BMW Trajectory
339
Summarising, since the 1990s the BMW consortium turned from a regional Bavarian producer and exporter of mainly three series of high tech and high quality cars to a global producer of a broader range of premium series and niche models. The Rover deal and the corresponding strategy of expansion towards mass market segments failed. The BMW group recovered from this adventure and focused more closely on the premium segment and on offering a wide range of car models and niche types. Mainly since the second half of the 1990s the company was able to invest in new car models and, at the same time, in new production facilities abroad. This was an unexpected success given the fact that the company had almost no experience in international production. The intertwined processes of product innovation and production internationalisation and the high risks the company was prepared to take are best revealed in the case of the new Spartanburg plant.
BMW Spartanburg: a greenfield plant with an all new car4 The Spartanburg plant in the USA is the first fully integrated production plant of BMW outside Germany. This reflects the major shift in the overall firm philosophy that was based on a made in Germany production strategy (BMW means Bavarian Engine Factories) and a market strategy oriented towards the premium segments of the sportive middle- and upper-class automobiles. The company’s overall strategy was to build high quality with high product flexibility and high production flexibility, in both time and scope. In Germany, the high production flexibility had been guaranteed by a combination of a medium level of production technologies, an open and flexible work organisation and highly committed semi-craftsmen with high employment stability. In Spartanburg the conditions were quite different. The decision and announcement to open the plant was made in June 1992 (therefore globalisation of the consortium began at the beginning of the 1990s and not until the Rover acquisition). Production of some BMW cars of the 3-Series began in September 1994, and exactly one year later the first Z 3 sportive roadster model was produced. The factory was an all new plant: new country, new facility site, new product and new production system. The latter was an explicit aim of the new plant: to build experience with new and Japan-oriented production systems. This distinguishes the starting point of the BMW Spartanburg plant from the Japanese transplants in the USA of the 1980s. Meanwhile the latter came with a clear and successful production model (and products already produced in Japan or elsewhere); BMW aimed just at learning from other production experiences (and making a completely new car). In terms of the production system as a whole, the BMW headquarters did not make many prescriptions. The guideline was to use the new plant as a type of laboratory for new production methods. From the beginning there was a certain orientation towards the Honda-production model focusing on
340 The Second Automobile Revolution
high product and production flexibility and high innovation.5 This led to the recruitment of a great number of managers mainly from the Honda and other Japanese transplants. The strategic decisions on location, products and investment were taken in the headquarters in Munich; the operative design of the production model and system was developed in a mix between the headquarters and the recruited American managers with experience in Japanese transplants. The Spartanburg plant is the only plant worldwide producing all the different models and series of the sportive BMW roadster and, since 1999, the new X 5 sports activity vehicle. Taking into account the overall annual production of about 60,000 cars, the broad scope of models and variations reveals the core of the consortium-wide BMW product structure and market strategy and one of the secrets of BMW’s success: to produce highly individualised, high quality sportive cars. From the beginning, the technical production system included body shop, painting and an assembly line. Similar to the situation in Germany, BMW’s added value was low at about 25–28 per cent. Therefore the cooperation with experienced suppliers was crucial. Eighteen first-tier suppliers were located in an industrial park nearby, most of them global players with headquarters in Germany. Thereby a mutual trust-based bonding took place: BMW relied on the cooperation of critical suppliers, and these suppliers depended on the OEM’s promise to develop long-term cooperation. In the plant, there were only eighteen industrial robots, indicating the overall regular or low level of automation. Daily production was of about 220 units (of an installed theoretical capacity of 300 units per day). No buffers were planned; at the end of the 1990s there were few between body shop and painting and between painting and assembly; mainly the suppliers were forced to recompense the rigidity of the production system with their own and extensive buffers (see Martin, 1999). Organisation hierarchy was divided into four levels (managers, coordinators, supervisors and team-members). A work team consisted of an average of sixty team-members and was divided into three to five pots. As the technical equipment is rigid, the work organisation itself has to absorb the flexibility requirements of the production system. Tact-times were relatively long (almost three minutes in the roadster production line) to compensate the different time sequences needed for the variety of models. Although in Germany all BMW plant have a recognised works council as the workers’ interest representation body, in BMW-Spartanburg there is no union representation, although the local UAW 5841 had organised some regional suppliers since the 1990s. The explicit aim of the plant management was to avoid becoming a ‘bargaining unit’ in the sense of right-to-work legislation. Therefore the management tried to control and reduce to a minimum the rate of work accidents, which often is a good ‘entry port’ for a union. The plant also paid relatively high wages and impeded all formal acts that could lead to a union campaign (such as workers’ injuries, giving the working teams union-like authority, etc.).
The BMW Trajectory
341
The outlined plant profile of BMW-Spartanburg combines the traditional BMW company philosophy of individualised high quality and high scope production with some radically new elements such as the strategic role of an exclusive world production platform for roadster and X 5 models. German product, quality and technical standards were merged with Japanese work organisation mechanisms and American-style job-oriented management and work. The mixture of these different ‘philosophies’ seems successful: in a short time this all new plant reached considerable levels of production and later on also of productivity. At the end of the 1990s there were severe concerns and tensions as to how to obtain ‘economies of scope’ (product and production variety) and high productivity and quality at the same time. The Spartanburg plant reflects the dominant strategy of the BMW group during the 1990s, i.e. to rapidly differentiate and broaden up the product and model range and, at the same time, to internationalise the production structure of the consortium. Despite the concerns and defects of this new plant in some of its details, it strengthened its position in the overall international production network. Nevertheless, the most important new plant opening in the BMW consortium during the first decade of the twenty-first century was not abroad, but in Germany. Whereas the new Spartanburg plant stands for the internationalisation strategy of the 1990s, the new Leipzig plant in Germany reflects the focus on high production flexibility and the high institutional embeddedness of the BMW group in general.
A new plant in a high cost country when new EU member states are available? The decision to build a new production plant was announced in 2000, and must be interpreted in the context of steady growth in units produced and sales (see the following section). Years before this decision, the company had begun to develop more new models and product lines. In concentrating on the classic Series 3, 5 and 7, the BMW group had developed a very successful product structure and marketing strategy for new niche cars like the Mini, the Z 3 and the X 5. Because of the ever-increasing total production volume the strategic decision to build a new plant was made. The annual capacity was calculated to be 160,000 units per year. In the first stage, a set of desired criteria regarding the choice of production site was established. Based on these location criteria, 250 sites around the world were considered. Following various selection rounds, five options remained: Augsburg, Schwerin and Leipzig in Germany, Arras in France, and Kolin in the Czech Republic. In the final stage, Leipzig won out over Kolin, although the Czech site offered a 30 per cent overall labour cost advantage. In July 2001 the then BMW CEO Milberg estimated that approximately one-fifth of the total costs related to
342 The Second Automobile Revolution
automobile production stem from labour costs; therefore, institutional location criteria, such as qualified workers and managers, logistics infrastructure and market proximity are very important factors. The significance of qualified workers in reaching the expected quality levels was learned mainly from the initial difficulties in Spartanburg.6 One final and decisive advantage in Leipzig was labour and production flexibility. This point fits quite well with the overall BMW strategy and philosophy of high external employment stability and high internal labour and working time flexibility. Even before the decision and recruitment process for the Leipzig plant began, the management and workers’ representatives in the BMW consortium met with representatives from the regional Leipzig union and employers’ association at the headquarters in Munich to define the rough outline for what was called a ‘formula for work’. The most important compromise between management and labour was to agree on creating a win–win situation: management obtains innovative flexible models of working hours and shifts, and at the same time work and employment will be maintained according to the regional level of collective agreements between the IG Metall union and the corresponding employers’ association. This negotiated and regulated production and labour flexibility is a decisive aspect in the Leipzig plant model, which the competing sites in France (Arras) and Kolin (Czech Republic) could not offer. In the case of the Arras plant the working time scheme is based on the national state law on the regulation of the weekly working time which impedes flexible working time arrangements at the local or plant level between unions and employers. According to interviews with experts, in the Czech case the corresponding unions were opposed as a matter of principle to any flexible working time schedule as discussed and negotiated in the BMW consortium since the 1980s. From the very beginning, and as a crucial part of the Leipzig offer, the weekly factory running time was designed with a wide range of working hours (from 60 to 140 hours). The possibility of a working scheme with two and three shifts was agreed upon, and even the two-shift scheme allowed for a wide range of working hours (from a model with 8 shifts of 8 hours, or 64 hours weekly; up to a model with 11 shifts of 9 hours, or 135 to 144 hours weekly). Based on different systems of short, medium and long working time accounts, this system allows the flexible adaptation of the working and factory running rhythms to the market conditions (with up to 30 Saturday shifts per year without additional payments). As far as this production flexibility facilitates a mobile break-even point, the higher labour costs at the Leipzig production site in Germany could compensate, or even undercut, a rigid working and production scheme with nominally lower labour costs but an inflexible and high break-even point in Kolin. This demonstrates the significance of institutional company embedding, namely the role of labour relations and the works council in Germany. As
The BMW Trajectory
343
mentioned above, part of the BMW business model is a long-term orientation with a focus on stability and accountability in working and employment conditions. This philosophy towards professionalism and labour stability is considered a basic condition for developing a broad scope of high-tech, emotive and passionate products. In this framework, cooperation between capital and labour is important at the plant level, between management and works councils; and at the company level, in the supervisory board and between company management and the corporate works council. Strategic decisions regarding the location of new plants, as well as long-term decisions related to models (which models should be produced in which plants), and the annual decisions on production volume distribution between plants (how many units of each model should be produced and in which plant) are negotiated between management and workers by the works councils and by the advisory board, i.e. based on legal definitions and on specific customs and practices. In the case of the new Leipzig plant, the corporate works council and the advising IG Metall union were both in very difficult negotiating positions. As much as BMW was interested in long-term success and employment stability, they were also interested in setting up a competitive and successful new production site, and not in creating a new subsidiary that would not be financially successful. The company and its works council still had in mind the recent Rover venture, which cost the company billions of euros in losses. On the one hand, this fact could have influenced the works council to promote, or possibly accept, building a production plant in an Eastern European or even an Asian country. On the other hand, the works council was not interested in widening the divide between work and employment conditions in the three existing plants in Germany, and the proposed future plant. At a future time of only medium returns, this would have exerted substantial pressure on (reducing) salaries, (raising) working hours and other issues in Germany, where management would have pointed out the aforementioned possible competitive advantages of a new plant in Germany. Therefore, the works council was ready to look for innovative and competitive models that would not endanger the German and the BMW specific labour regulation model. Not only did this lead to complex negotiations between management and labour, but also between the works council and union at the corporate level, and between the corporate works council and the local–regional Leipzig union. Leipzig representatives – namely BMW managers, local public administration, local union leaders and works council members – were mainly focused on setting up a new plant, whereas part of the BMW management and the works council majority wanted to preserve BMW’s working and employment conditions. For the works council and IG Metall, it was a necessary condition that labour regulation would not fall below the regionally valid collective agreements – even if this meant
344 The Second Automobile Revolution
a worsening of working and employment conditions in comparison with the BMW company standards in Munich, Dingolfing and Regensburg. The standards in these cities were, and continue to be, higher than in the corresponding regional collective agreements in Leipzig. Even throughout the negotiation process surrounding the decision regarding the Leipzig site, the coalition had to offer competitive costs compared to the Kolin site in the Czech Republic, in addition to controlling the gap between the new Leipzig plant and the other German plants in Munich, Regensburg and Dingolfing. From a general standpoint, the localisation decision as a negotiation process seems to have led to a substantial win–win outcome for all major interest groups (but only inside Germany). The company works council ensured that the new factory was not opened under a completely different legal regime (for instance, as a limited liability company (GmbH), as was the case for Volkswagen Auto 5000 GmbH; see Pries, 2002 and Schumann et al., 2006). Regional collective bargaining agreements were not undercut; the total factory running time was high, but not higher in comparison to Regensburg plus the press shop and the Munich plant. Most of the existing BMW labour conventions were also applied in Leipzig (successbased bonus payments, Christmas bonus, holiday bonus, company pension fund, reduced car prices, etc.). Some innovative elements of the Leipzig labour regulation regime are related to the payment system. New output-oriented salary schemes were introduced in certain areas, with a basic salary and bonus payments of up to 21 per cent of the basic salary if production targets are reached. There are also individual bonuses according to performance and goal attainment (with 3 per cent of the basic salary as a general target step, and 6 per cent and 9 per cent as additional steps). The annual bonus payment is oriented around plant seniority – from 25 per cent (less than a year working at BMW) up to 100 per cent (three or more years of BMW service), with the start of production in 2005 as the starting date. The working time regime at the beginning of production in 2005 was a two eight-hour shift system, with five working days per week and 245 working days per year. The factory began running in September 2004. Production began in March 2005, with the product line two (Series 1 and 3) and a total of 2,000 workers and employees in the body shop, paint department and assembly. Average daily production is aimed at 650 units, with a possible expansion of up to a maximum of 960 units per day and a planned workforce of 5,000 (with an optional additional press shop). At the beginning of 2007, a very high number (of about 1,000) of temporary workers were employed, earning only about half the amount paid to formally employed BMW workers. In summary, the Leipzig site won out due to its meeting a complex set of criteria defined by BMW. Last but not least, language and culture were also considered important factors, which could have raised risks and costs had the
The BMW Trajectory
345
new plant been built in Kolin. According to Leipzig CEO Claussen, BMW’s business model is oriented to the long term; therefore, the location decision for a new plant is not only important for the lifespan of one car model, but for half a century. Keeping this long-term orientation in mind, it is probable that the direct labour cost differences between Kolin and Leipzig will be reduced more quickly than would be the cultural and language differences between the different locations. Space, logistics and the soft factors of the labour market and labour regulation are also important. The global, or at least translocal, management of knowledge, the complex value chains and fragile supply chains strengthen aspects of liability, which include the level of general economic development, infrastructure security, the scope of product and production flexibility, efficiency and speed of public administration. The Leipzig plant location process therefore reveals the strong long-term strategy and commitment of the BMW group in Germany – although the company has become a real global player during the last ten years. It also shows the strong impact of institutional factors such as the labour regulation and especially the works council system – economic globalisation did not result in the abandonment of the general German business model with strong workers’ participation but to a certain downgrading and flexibilisation of working and employment conditions in the newly opened Leipzig plant.
Present situation and future prospects of the BMW business model BMW today is a globally thinking and operating full-sortiment premium manufacturer. Although smaller than Volkswagen and Daimler-Benz, BMW was the first German automobile manufacturer able to offer the whole range of products from compacts up to high-end luxury cars. This broad scope of products ranges from the Mini and 1-Series over the 3-, 5- and 7-Series up to the 8-Series, the sportive cars of the Z-Series and the SUVs of the X-Series. The BMW trajectory since the 1990s has to be interpreted as a result of three interrelated processes of change. First, concerning the reorganisation of the production system two important features were developing the cooperation with the suppliers and the interdisciplinary R&D groups as well as improving the general flow of knowledge and experiences via the research and innovation centre FIZ. By this, BMW focused much more than other carmakers on the initial steps in the product development curve: optimisation potential was not mainly situated in the direct execution of body shop, painting and assembly tasks. But the most important rationalisation promise was located in the beginning of the life cycle of a product: in the knowledge and R&D management within the company and with (system) suppliers.
346 The Second Automobile Revolution
Second, after having settled a well proved and tested product range of the 3-, 5- and 7-Series, the company invested a lot of attention and money in the substantial differentiation of its products. Combining well-established ‘volume’ series with highly specialised niche models while, at the same time, maintaining high technological and quality standards helped to expand the number of customers and the production volume. The third important step was – in the very short period since the middle of the 1990s – shifting from a regional Bavarian producer to a really global player not only in terms of sales, but also in production and procurement. The establishment of an all new greenfield plant in Spartanburg and a large number of new CKD plants, mainly in Asia, and the strengthening of a worldwide procurement and logistics network reveal this strategy. In 2006, the consortium had twenty-three major production facilities in thirteen countries, expanding strongly in emergent markets. In China, since 2003 the joint venture with Brilliance China Automotive Holdings Limited is developing very successfully. In 2006, 36,430 automobiles were sold, an increase of more than 50 per cent as compared to 2005 (BMW, 2007: 6). In relative terms, the profit from this engagement is very high compared to other BMW plants. These three lines of strategic action since the 1990s were strongly interrelated: without having restructured the knowledge and R&D activities and
160000 140000 120000 100000 80000 60000 40000 20000
Production (units/10) Sales (Mio. U) Figure 17.1
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
0
Employment Net profit
BMW Group world production, employment, sales, net profit, 1990–2006
Note: the production in units has been divided by 10, sales and net profit in million £, the net profit has been multiplied by 100. Source: BMW Annual Reports.
The BMW Trajectory
347
developed cooperation with suppliers it would not have been possible to light this ‘firework of new products’; without broadening substantially the scope of products and entering or even creating new market niches it would not have been possible to almost triple the volume of the overall production in about fifteen years (Figure 17.1). In the same period, the overall direct employment only increased nearly half, which reflects the success of the efforts to reorganise the production system. Without this direct increase in production efficiency the company would not have had the money needed for substantially expanded R&D activities and the development of new products. Ultimately, the internationalisation of production activities allowed BMW to participate in the most rapidly growing markets during that period, namely the SUV segment in the USA and the premium car segment in Asia. The fundamental shift from a regional carmaker to a global premium player can be observed also when comparing the overall structure of sales and of car production of the BMW group in 1990 and in 2006. At the beginning of the 1990s, production abroad represented almost 3 per cent, and the sales outside Germany counted for about one-third. According to the Annual Report for 2006 of the BMW group (pp. 36 and 54), sales abroad represented 73 per cent of total sales, and the production in plants outside Germany represented 36 per cent of total unit production of the group. These figures best reflect the fundamental shift in the overall company strategy and structure. But this crucial move since the 1990s did not coincide with a rupture in the overall business model of the BMW group. As mentioned in the introduction and described in this chapter, the expansion of the product line was possible due to a firm basis of engineering capacities and the prevalence of engineers in management and skilled and highly motivated workers in the lines. All decisive CEO positions were filled by internal recruitment, and most of the leading managers were and are engineers by profession. The professional and engineering approach is reflected also in the strategic role of the research and innovation centre (FIZ) in Munich; this R&D and innovation centre is not only responsible for development of new models, but also designs and samples the production start of new models. By this means, problems and errors of model design that are revealed in starting running new products, are relayed at once to the product developing areas and can be improved for all groupwide plants at once. Besides the function of the FIZ the case of the new plant location and opening in Leipzig also reveals this approach towards professional and qualified work, being the one decisive aspect of the site selection decision. Besides the professionalism argument, the Leipzig case is also illustrative of the family-oriented system of social and labour relations. Because the works council of the BMW group was influential strategically in the location decision by developing and proposing a work schedule for the new plant it permitted a range of production flexibility that was not possible at the competing sites. The family- and network-driven institutional embeddedness of
348 The Second Automobile Revolution
the company is also reflected in the long-term cooperation structures with employees and suppliers. Labour turnover is one of the lowest in the German economy; workers’ satisfaction with the company and BMW’s attractiveness as an employer ranks high above the average (BMW, 2007: 25f.). The index of supplier satisfaction with their customers (the carmakers) always gives BMW (and Porsche) the best satisfaction degrees of their suppliers.7 The human resource management strategies and the corresponding value and mission statements also echo the family, qualified work and long-term orientation of the company’s business model. Last but not least, this is reinforced by the strong influence of the Quandt family as shareholders. In 2002, the Quandt family represented 46.6 per cent of the direct voting rights of the shareholders (BMW, 2007: 52). Besides the 8.8 per cent of ‘other investors’ the 44.6 per cent of ‘institutional investors’ are distributed quite equally in Germany, the rest of continental Europe, the UK, the USA, and the rest of the world. This gives management a certain independence from short-term driven shareholder thinking. Asked if he has a problem with short-term thinking in trimesters of shareholders, the BMW CEO Reithofer answered: ‘A reasonable time to value a company for me is a twelve-month period. More reasonable is a three to four year period’ (Süddeutsche Zeitung, 30 December 2006). From a general viewpoint, the BMW group trajectory of the last fifteen years reflects the strength of what elsewhere (Pries, 2004a) was coined the German automobile business model and that consists of five main elements: (1) a mixed system of corporate structure and profit strategies, whereby shareholder value and stakeholder value interests are balanced through a system of executive boards and supervisory boards and of ‘conflictive cooperation’ between management and labour; it includes the tangible influence of family networks, important banks and/or the state in the corporate governance of companies as well as public and collective interests of labour, represented through a mix of party and union representatives in the different committees; (2) hand-in-hand with this goes a shareholder–stakeholder configuration, with a long-term perspective and the building of identities as car companies that are genuinely dedicated to advancing engineering, as is reflected in the ongoing dominance of car guys as executive leaders; (3) without the former aspect, it would not have been possible to successfully put in practice a ‘firework of new products’ based on a historically proven, strong orientation towards research and development, engineering and knowledge development, in general; (4) the aforementioned enlargement of the products’ scope was not only technology-based but also oriented towards innovative, emotionally appealing, high-quality, sustainable products; these shifts in the product structure and market strategies reflect the recursive integration of social and political sensibility into the products, which was made possible in part by workers’ participation; (5) this workers’ participation was not simply a shortterm management strategy designed to exploit knowledge: it was part of a decades-long labour and personnel policy of promoting stable, long-term
The BMW Trajectory
349
employment, low external and high internal labour flexibility, and workers’ qualification as an important investment. These elements of its specific automobile business model could explain a large part of BMW’s success in the 1990s and the beginning of the new century. However, there are at least two challenges for this business model and for the BMW group in the future. The first is the possible bifurcation of labour between a core crew and a marginalised manpower working in precarious conditions in outsourced services and supplier companies. The second challenge involves the possible ‘market insensibility’ of the engineer-driven and professional car builders’ thinking that sets the technical solution and the highly technologically sophisticated products at the top of the priority list and not the customers’ preferences. This could be observed since the beginning of 2007, when in the context of carbon emissions debates in Europe, the USA and later on all over the world the German premium carmakers were hit by a kind of mental shock. For the future, building sustainable cars is much more important than producing only power cars – for the ‘car guys’ and their engineers at BMW, but also at Daimler-Benz, Porsche and Audi, this will be a strong test for their own sustainability. Although German premium carmakers may be more challenged due to the high average carbon emissions of their high performance cars, it is likely that they will find specific opportunities in their environment of highly qualified suppliers, highly motivated and skilled workforce, a reliable labour regulation system and intensive cooperation with knowledge producing universities to meet the challenge.
350 The Second Automobile Revolution Appendix Table 17.1
BMW, 1990–2006 1990
1991
1992
1993
1994
1995
1996
1997
1
Automobile sales 514.705 552.103 588.657 534.397 1.049.919 1.073.194 1.151.364 1.196.096 – BMW 514.705 552.103 588.657 534.397 573.953 590.072 644.107 675.076 – Rover / / / / 362 030 345 820 351 479 363 834 – Land Rover / / / / 90 050 115 590 125 222 128 048 – MG / / / / 773 1 667 13 106 14 721 – Mini / / / / 23 113 20 045 17 450 14 417 – Rolls-Royce / / / / / / / / Motorcycles sales 31 310 32 092 34 800 35 150 46 667 50 246 50 465 54.014 3 Auto. sales by region – Germany 223 500 250 200 258 800 – Great Britain 305 100 291 100 294 300 – Rest of W. Europe 265 900 305 700 307 200 – North America 119 400 135 600 154 500 – Asia / other regions 159 300 168 700 181 200 2.311 Investment 4 Operating profit 5 6 Workforce 70 948 74 385 73 562 71 034 109 362 115 763 116 112 117.624
Notes: 1. In units; until 9 May 2000 incl. Rover Cars and until 30 June 2000 incl. Land Rover. 2. In 2000 BMW sold the brands Rover and Land Rover of the Rover Group. Since 2001 the Mini in the Annual Reports. 3. In units; since 2000 without Rover and Land Rover. 4. In million euros; until 1999 reporting according to HGB, since 2000 according to IAS/IFRS. 5. In million euros; operating profit before taxes for cars and motorcycles. 6. At the end of year; since 1998 values without inactive work contracts, without employees in the Sources: BMW Annual Reports.
Notes 1. I acknowledge the research assistance of Philip Rebel and Sophie Rosenbohm. 2. Until now, the direct relations between the Nazi regime and those companies that were controlled by the Quandt family and represent the precursors of the BMW activities after the Second World War have not been studied in detail. After 1945 the Quandt family refused to reveal these relations. It was not until 2007 (after a detailed and in-depth TV report) that the Quandt family admitted to financing scientific research on their forefathers’ role during the Nazi regime. 3. BMW got the brand name rights of Mini and of Rolls-Royce (although the latter after a dispute with Volkswagen only since 2003); for details see Eckardt and Klemm (2003). 4. For details see Eckardt et al. (2000); Eckardt and Klemm (2003); Martin (1999). See also http://de.wikipedia.org/wiki/BMW for the BMW group in general. 5. As underlined by Freyssenet (1998) and Boyer and Freyssenet (2002) there is an important difference between Toyota and Honda in relation to their focus on standardisation and flexibilisation. 6. Besides ‘soft factors’ explaining the decision for Leipzig there was also an important ‘hard factor’: BMW received the maximum amount of financial support from the EU fund for disadvantaged regions. Initially, the EU agreed to pay 32 per cent of the total investment. Although this amount was later reduced in 2002, BMW ended up receiving approximately £360 million in EU regional development aid (in this
The BMW Trajectory
1998
1999
2000
1.187.115 1.180.429 1.011.874 699.378 751.272 822 181 303 805 227 743 n.a. 2 153 495 178 000 n.a. 2 14 415 11 719 n.a. 2 16 022 11 695 n.a. 2 / / / 60.308 65.168 74.614 280 100 270 000 316 000 161 700 159 300 2.179 118.489
282 700 219 900 329 000 194 600 154 300 2.155 917 114.952
240 600 68 300 208 500 200 400 104 400 2.781 1 645 93.624
2001
2002
2003
2004
2005
351
2006
905.657 1.057.344 1.104.916 .208.732 1.327.992 1.373.970 880 677 913 225 928 151 1 023 583 1 126 768 1 185 088 / / / / / / / / / / / / / / / / / / 24 980 144 119 176 465 184 357 200 428 188 077 / / 300 792 796 805 84.713 92.599 92.962 92.266 97.474 100 064 245 800 90 900 225 400 225 800 117 800 3.516 2 851 97.275
258 200 120 900 260 700 273 200 144 300 4.042 2 943 101.395
255 800 134 500 263 600 294 900 156 200 4.245 2 811 104.342
283 600 145 300 298 100 315 900 165 800 4.347 3 190 105.972
295 900 156 200 350 800 329 000 196 100 3.993 3 036 105.798
287 700 154 100 375 500 337 400 219 800 4.313 3 078 106.575
was produced as BMW brand. Since then, the former Rover brand Mini was not listed separately
free time period of their partial retirement, without low income earner.
context it is worth underlining that BMW would have received a similar portion of investment had it invested in the Czech Republic). 7. See, for example, the SSI (Supplier Satisfaction Index) raised biannually at the ‘Forschungsstelle Automobilwirtschaft’ of the Universität Bamberg: http://www. faw-bamberg.de/.
Bibliography Birkmann, K. (2005) ‘Der Weg nach Leipzig aus Sicht des Betriebsrats, Vortrag gehalten auf dem 2. Automobilkongress der Ruhr-Universität Bochum’ (www.ruhruni-bochum. de/autokongress2/download/beitraege/Praesentation_Birkmann.pdf). BMW AG (1991) Stationen einer Entwicklung. Die ersten 75 Jahre BMW. München: BMW. BMW AG (2001) Flexible Arbeitszeiten bei der BMW Group. München: BMW. BMW AG (2003) Das Konzept des neuen BMW Werkes in Leipzig. Entwicklung nachhaltiger Strukturen, Vortrag, Januar 2003, BMW Werk Leipzig. München/Leipzig: BMW. BMW AG (2005a) Rede von Dr. Helmut Panke, Vorsitzender des Vorstands der BMW AG, 85. ordentliche Hauptversammlung der BMW AG, gehalten am 12. Mai 2005 in München. München: BMW. BMW AG (2005b) Sustainable Value Report 2005/2006. München: BMW. BMW AG (2007) Annual Report 2006. München: BMW. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan.
352 The Second Automobile Revolution Eckardt, A. and Klemm, M. (2003) ‘The Internationalization of a Premium Automobile Producer: the BMW Group and the Case of Rover’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan, pp. 170–97. Eckardt, A., Köhler, H.-D. and Pries, L. (2000) Auf dem Weg zu global operierenden Konzernen? Fallstudien zu den Internationalisierungsverläufen deutscher Automobilkonzerne in den 90er Jahren (Forschungsbericht). Erlangen: IPRAS. Freyssenet, M. (1998) ‘Intersecting Trajectories and Model Changes’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. New York: Oxford University Press, pp. 8–48. Harbauer, J. (2003) ‘ “Standortwahl Leipzig”: Vortrag gehalten auf der Veranstaltung “Inländische und ausländische Standorte richtig bewerten”’, Fachtagung am 7 May 2003. Stuttgart: Fraunhofer Institut ISI/BMBF. Kinkel, S. and Lay, G (2005) ‘Automobilzulieferer in der Klemme’, in L. Pries and M. Hertwig (eds), Deutsche Autoproduktion im globalen Wandel. Altindustrie im Rückwärtsgang oder Hightech-Branche mit Zukunft? Berlin: Sigma, pp. 59–74. Lewandowski, J. (1998) BMW. Typen und Geschichte. Augsburg: Delius Clasing. Martin, S. B. (1999) ‘Transplants oder “Lernorte”: Beschaffungsformen und Zuliefererbeziehungen in den deuen deutschen Automobilnetzwerken im Süden der USA’, in H. Kilper and L. Pries (eds), Die Globalisierungsspirale in der deutschen Automobilindustrie. Hersteller-Zulieferer-Beziehungen als Herausforderung für Wirtschaft und Politik. Munich: Rainer Hampp Verlag, pp. 57–91. Pries, L. (2002) ‘5000 × 5000: Ende gewerkschaftlicher Tarifpolitik oder innovativer betrieblich-tariflicher Sozialpakt?’ Industrialle Beziehungen, 9(2): 222–35. Pries, L. (2004a) ‘Renaissance of the German Carmakers during the 1990s: Successful Japanization or the Development of a Genuine Business Model?’ in M. Faust, U. Voskamp and V. Wittke (eds), European Industrial Restructuring in a Global Economy: Fragmentation and Relocation of the Value Chains. Göttingen: SOFI, pp. 131–55. Pries, L. (2004b) ‘New Production Systems and Workers’ Participation: a Contradiction? Some Lessons from German Automobile Companies’, in E. Charron and P. Stewart (eds), Work and Employment Relations in the Automobile Industry. Basingstoke and New York: Palgrave Macmillan, pp. 76–102. Pries, L., Sandig, C. and Schweer, O. (2005) ‘Der Produktentstehungsprozess als Indikator für Unternehmensinternationalisierung: Das Beispiel der deutschen Automobilunternehmen’, in G. Schmidt, H. Bungsche, T. Heyder and M. Klemm (eds), Und es fährt und fährt . . . Automobilindustrie und Automobil-kultur am Beginn des 21. Jahrhunderts. Berlin: Sigma, pp. 201–27. Rosellen, H.-P. (1983) Das weiß-blaue Wunder. BMW – Geschichte und Typen. Stuttgart: Seewald. Schumann, M., Kuhlmann, M., Sanders, F. and Sperling, J. (eds) (2006) Auto 5000: ein neues Produktionskonzept. Hamburg: VSA.
18
A Break from the Past: Volvo and its Malcontents Matthias Holweg and Frits K. Pil
Introduction By the start of the 1990s Volvo had managed to establish itself with a clear and consistent brand identity, a unique manufacturing strategy and a loyal customer following. The world around it has changed drastically since the start of the second automotive century: shifting economies of scale due to an expansion of model range and shortened product life cycles, new efficiency imperatives and a changing competitive landscape in the near luxury segment have placed Volvo’s modus operandi in peril. In this chapter we will briefly touch on the failed merger with Renault in 1993, and leap forward in our discussion to start with Ford’s purchase of Volvo in the spring of 1999. Over the following eight years, Volvo has become an integral part of Ford’s Premier Automotive Group, a move which has resulted in a major expansion of its product range, through product sharing with family partners. We will comment on the success of this merger and – given Ford’s recurring yet conflicting statements of intent to sell Volvo on – also comment on the long-term future for Volvo. Since the 1990s, the changes in the motor industry have led to a fundamental rethinking of Volvo’s manufacturing strategy, its independence on the product design front, and broader strategic tensions between Volvo’s customer-driven approach and Ford’s volume-driven wholesale production and distribution model, which may ultimately translate into Volvo being divested from the Ford family. We will examine each of these evolutions in turn in an effort to highlight the extent and impact that these changes have had, and will continue to have, on a firm that was once known primarily for safe ‘boxy’ cars in the eyes of consumers, and a somewhat quirky if lovable firm at the forefront of socio-technical innovation in the eyes of academics.
Ownership dilemmas Volvo Cars, or ‘Volvo Personvagnar’ as it is known in Swedish, was formed as a subsidiary to bearing-maker SKF in 1924 by Assar Gabrielsson and Gustaf 353
354 The Second Automobile Revolution
Larson (allegedly during a meal consisting of crayfish), and only became an independent company in 1935 as Volvo AB. SKF allowed it to use one of its nameplates (‘Volvo’, Latin for ‘I roll’, as SKF was producing bearings). Volvo AB soon diversified into trucks and boat engines, with Volvo Cars as a division of its operations. The company produced its first car in 1927, by 1932 had its own factory, and managed to maintain production during the Second World War. In the post-war period, Volvo Cars launched a range of new models, including sports cars, and is credited with inventing the threepoint seat belt, which contributed to its image as the ‘safe family car’. The Torslanda plant was opened in 1964, and with expanding sales in the US, car production was a profitable part of Volvo AB. Fortunes changed, however, with currency fluctuations and periods of recession in key markets. With its strong dependence on exports to the US market, Volvo was feeling the pinch from either a weak dollar or a recession, and by the early 1990s saw itself in the position of not being able to fund new vehicle development programmes. After some deliberation, Volvo’s chairman at the time, Per Gyllenhammar, sought a partner to increase the scale and scope of the car operations. Renault and Volvo had already reached a general collaboration agreement in 1979, so Renault was identified as a suitable partner, yet the merger was stopped due to political resistance in Sweden (for more details on the Renault–Volvo merger see Williams et al., 1995 and Berggren, 1998). Thus, Volvo was forced to continue its own way – lacking the resources to expand its models range into the increasingly popular SUV and MPV markets, as well as the growing small car segment in Europe. Instead it was offering sedans and estate cars in a declining market segment. Hence, by 1998, Volvo AB reached the decision that it was no longer able to sustain Volvo Cars, and sought a large corporate home for its car division that was able to fund Volvo’s next round of vehicle designs. Ford emerged as the main bidder for Volvo, seeing Volvo as a strong complement to ‘Ford’s social values’, as CEO Jacques Nasser was cited as saying in the press. Ford Motor Company bought Volvo Cars for $6.45 billion in 1999, taking yet another step towards becoming a large-scale seller of highprofit luxury cars as it had already bought Jaguar and Aston Martin. Jacques Nasser and Bill Ford were quoted saying that ‘they liked Volvo’s current line-up of cars and were not planning immediate changes’, and that they admired Volvo’s tradition of paying attention to safety and environmental issues. Henceforth Volvo was integrated in the Premier Automotive Group (PAG) within Ford, and found itself in the position to operationally integrate with Jaguar, Land Rover and Aston Martin in terms of technology, platforms and production, which in many ways raised issues about destroying Volvo’s core values, which we will discuss in the following.
Volvo and its Malcontents 355
Rethinking product value: what is Volvo’s core strength? With the integration of Volvo into Ford, have come efforts to both extend Volvo’s product offerings and rationalise product design by leveraging product platforms. The result is that Volvo has now moved beyond a limited set of offerings of medium to luxury sedans and estate cars, to a full spectrum, ranging from a compact hatchback like the C30, to large sedans (e.g. S80), and even sport-utility offerings (XC90). To extend the vehicle range, and mitigate some of the development expenses, Volvo has increasingly become reliant upon Ford platforms for its products: platform sharing takes place from the smallest (the C30 is based on Ford’s Compact C1 platform – as are the S40 and V50), to the largest based on the Ford D3 platform (Ford Five Hundred, now Taurus, and Volvo S60, V70, S80, XC90). Table 18.1 lists the past and present models of Volvo, as well as the platforms these are based on. It is worth noting that the Ghent factory has secured Table 18.1
Volvo’s product portfolio and platforms, 2007–2012 (projected) Model
Compact
480 C30 V30
Internal platform name P1 tbc (Focus platform)
Year introduced
Year discontinued
1986 2007
1995
Mid-size
S40/V40 S40/V50 Mk II V50 MkIII
P2 P1 tbc
1995 2004 2009
2004 2009
Full-size
S60 S60 II 850/V70 V70 II V70 III V70 IV
P2 P24 N/A P2 P24 tbc
2001 2009 (planned) 1992 1999 2007 2012 (planned)
2009
700/900 S80 S80 II V100
P2 P24 tbc
1990 1997 2006 2012 (planned)
1998 2006
Convertible
C70 C70 II
N/A P1
1997 2005
2005
SUV
XC60 (planned) XC90 XC90 II (planned)
P24 P2 P24
2009 (planned) 2002 2010 (planned)
Executive
Source: Automotive News, Global Insight, Volvo Press Releases.
2000 2007 2012
2010
356 The Second Automobile Revolution
the XC60, which is based on the large vehicle platform, whereas these large vehicles were previously assigned to Torslanda (with the exception of the swing models, such as the V70, that were produced in both factories to adjust volumes in case of demand fluctuations). Two observations follow from considering the dynamics in Volvo’s product line-up: first of all, Volvo has leveraged Ford’s platforms to move out of its traditional segments of mid-size to luxury sedans and estates. Instead it has been expanding into SUVs, cross-overs, and even a C-segment vehicle that shares the platform with the Ford Focus. In addition, producing a B-segment car (codenamed V10) on the basis of the Mazda 2 is in discussion, which would mean further extension into the mainstream volume car market. On the upscale models, the traditional limousine (S80) is likely to be discontinued, as the S60 has been growing in size, and a large estate (V100) might be launched in response to BMW’s PAS and Mercedes’ R-Class vehicles. The V30 will replace the current S40 and V50 models. Overall, there are considerable dynamics at play that will that drive Volvo to compete in the mainstream volume market. With regards to its supply chain, and despite its platform commonality with Ford, Volvo’s component purchases are still predominantly located near its car factory in Sweden, Belgium and immediately neighbouring countries. Platform sharing provides new unique challenges to Volvo’s factories. As Volvo expands its product line, and derives more products from other arenas, it faces significant pressures on the flexibility front. At one level, there is the challenge of incorporating fundamentally different designs from a welding standpoint, as well as from an assembly standpoint. One outcome of those differences is that Torslanda has started specialising in larger vehicles, while Volvo Ghent has picked up the smaller end of the product range. A related outcome, however, is some reduction in flexibility in moving products between factories. For example, with the second-generation V70 (which seized production in 2007), a product based on Volvo’s P2 platform, both Ghent and Torslanda had the physical ability to produce the product, and volume could be shifted between the two facilities to offset volume shifts on other products. The new V70 introduced in February 2007 is based on Volvo’s P24 platform, and can only be produced in the Torslanda facility. One positive outcome of the overlap between Ford and Volvo product design is that it is now feasible to build the products on shared platforms, such as the S40, in Ford facilities in locations where Volvo has none. Changan Ford Mazda Automobile Corporation (owned by Changan, Ford and Mazda) is now producing the S40 for the Chinese market. Although Volvo Ghent won the rights to build the S40, it had to compete with Ford manufacturing facilities for the right to produce it. While it won this round, the fact that the product architecture is based on Ford and Mazda conceptualisation has implications beyond design, through to manufacturing, and if platform sharing increases, the outcome may be more Volvo products manufactured
Volvo and its Malcontents 357
outside of traditionally Volvo production facilities. It will also mean further pressure to increase alignment between Volvo’s manufacturing technology and practices, and Ford’s – a topic to which we now turn.
Work redesign Volvo’s efforts at work redesign date back to the 1970s, when Per Gyllenhammar, then CEO, pushed the company to rethink the role of workers in the assembly process. This led to a rethinking of some of the fundamental precepts of vehicle production, culminating in the establishment of the Volvo Uddevalla factory. Volvo is often credited with having developed a completely new way of producing vehicles – the so-called holistic or ‘reflective’ production model (Berggren, 1993, 1998; Ellegard et al., 1992; Sandberg, 1995). Uddevalla’s assembly system is reminiscent of craft work and has been labelled by some observers as ‘neocraft’ (Berggren, 1994c). One of the hallmarks of the Volvo craft model was its reliance on teamwork, along with extensive individual skill development. At the Uddevalla plant, vehicles were built in docking stations, where cars remained stationary while they were being worked on, with long work cycles of up to three hours per worker. The groups worked independently of each other and assembled complete cars, in a craft system. The resulting craft model is thus characterised by the lack of an assembly line, extensive skill development and professionalisation of workers, low levels of automation, naturally grouped assembly work, ergonomically sound production tasks and flexible production levels. These were supported by long cycle times, special parts-delivery and order systems, self-managed team based activities, and extensive on-the job training. It was argued that autonomy in work design in the socio-technical model provides adaptability, individual motivation and opportunities to leverage unique abilities, and fosters individual innovation. However, on the downside, the model does not sustain knowledge transfer and limits organisational learning. Equally daunting was an apparent lack of efficiency, although this has been subject to vivid debates in the academic literature (Adler and Cole, 1994; Berggren, 1993, 1994a, 1994b). In our assessment, Uddevalla was indeed competitive in terms of direct labour productivity; however, once overhead functions such as materials handling are factored in, its small scale and labour-intensive logistics meant that it was not competitive overall. In fact, even at its launch, Uddevalla’s assembly area productivity was reported as being half of Volvo Ghent’s overall efficiency (Prokesch, 1991). More specifically, in 1991 Uddevalla took fifty hours of labour to produce a vehicle (including direct labour in assembly, as well as indirect labour in material handling, quality control, but excluding weld and paint). By comparison, Volvo Ghent reported twenty-five hours per vehicle for the whole process,
358 The Second Automobile Revolution
including body and paint. Even at the start there was a 2:1 differential in productivity, which meant that the experiment was not competitive on the motor industry’s key measure, labour hours per vehicle, and hence unlikely to be rolled out further. The Uddevalla plant abandoned this approach only in January 2002, and instituted a traditional assembly line, by breaking down the assembly process into shorter cycle times. The stated reason behind the change was a desire to achieve increased process standardisation. The end of dockbuild at Uddevalla brings to an end an extensive debate around issues such as the professionalisation versus deprofessionalisation of workers, the role of ‘intangibles’ that people bring to the system (Cutcher-Gershenfeld et al., 1998), and the scripting versus non-scripting of work procedures and implications for individual and organisational learning (Adler and Cole, 1993). The lean production model emphasises cooperation between departments and among workers, extensive information flows within and across teams, minimisation of buffers, and multi-skilling of workers. It differs significantly from the socio-technical model in the degree to which work is scripted. Individual and team level learning and its diffusion across individuals or teams is related to the degree to which tasks and procedures are scripted and the associated standardisation in production processes (Adler and Cole, 1993; Argote, 1999). Scripting work procedures and the associated documentation are also important to ensure that superior routines identified by one team are maintained, enhanced, transferred and replicated across teams (Nelson and Winter, 1982). Scripting of procedures makes the processes more understandable and ensures retention of tacit knowledge, by transforming tacit knowledge held by individual workers into explicit knowledge (Polanyi, 1966; Szulanski, 1996).
Production strategy Volvo has embraced scripting at its two remaining full assembly factories – Volvo Torslanda and Volvo Ghent – as well as at Uddevalla (the latter is now a cooperative venture majority owned by Pininfarina). While the focus of socio-technical systems work design is worker needs (it is a ‘humancentred’ model), the main focus of the lean production model is continuous improvement in work operations. The socio-technical approaches embodied in Volvo’s philosophy advocate a better balance between the social and the technical aspects of production systems, and propose the introduction of semi-autonomous work groups as a key element for achieving high engagement. Volvo continues to draw on its ‘reflective production’ roots by embracing a recent trend we have identified at leading lean factories – the creation of a group-level identity through careful consideration of task allocation on-line (see Pil and Fujimoto, 2007 for a comprehensive discussion).
Volvo and its Malcontents 359
Comparative plant productivity 40 30 20 10 0 Genk Ghent (S40, GM Europe (Signum, (Mondeo) V50, S60, Vectra) V70)
Toyota (Avensis)
Renault (Espace, Laguna, Vel Satis)
PSA (407, C5, Xsara)
Torslanda (V70, XC70, S80, XC90)
Figure 18.1 Productivity of Volvo plants against European competitors and Ford plants in 2006 Note: Productivity = labour hours per vehicle. Source: Data from Harbour Associates http://www.mtmnorden.com/Flikar/Evenemang/Tvarkontakt_ 2006/A1.pdf (accessed 11 February 2008).
While takt time is short, workers are grouped so that collectively, they are responsible for coherent elements of the vehicle. This facilitates process improvement, quality control and allocation. It also generates a degree of group identity and motivation in ways that traditional decomposition of assembly tasks based purely on efficiency cannot attain. The result is an ongoing positive relationship with the workforce. Despite tremendous pressure of new product introductions and the adding of a third shift, Volvo Ghent, for example, has been declared one of Belgium’s top twenty-five employers to work for twice in the last five years. Likewise, it was nominated twice for the European award for lifelong learning. Both the Ghent and Torslanda Volvo factories are facing much more variety today than they did fifteen years ago. There is a broader range of products built at each facility. But also at the component level, and at the architecture level, the vehicles differ dramatically because some are based on Ford/Mazda platforms, while others have had much more Volvo design input. Despite this increased variety, the factories have been able to hold their own. Figure 18.1 shows that Volvo Ghent’s uncorrected productivity, at just under twenty-five hours per vehicle in 2006, was among the best in Europe, and while Torslanda’s thirty-four hours per vehicle placed it among the worst in Harbour’s sample, it was significantly more productive than some of the other Ford facilities producing high-end products. On the quality front, Ghent experienced an increase on its in-house measure of defects in 2004 with the
360 The Second Automobile Revolution
160000 140000 120000
Units
100000 Netherlands (Nedcar)
80000 60000
Sweden (Torslanda and Uddevalla)
40000 20000
Belgium (Ghent)
0 Jan-00
Figure 18.2
Jan-01
Jan-02
Jan-03 Jan-04 Dates
Jan-05
Jan-06
Volvo’s quarterly production by country, 2000–2006
Source: Ward’s Automotive Yearbooks, 2001–2007.
launching of the S40 and the V50, from 51 demerits/car in 2003 to 71 in 2004. However, here too, the factory learned rapidly, and its 2005 quality figures were below the pre-introduction figures, and by 2006, the factory was at 40 demerits per vehicle (Volvo, 2007). In terms of productivity, among the Volvo factories, Ghent is clearly on par with its European competitors, while Torslanda is slightly behind Renault and PSA. In comparison to other Ford plants, and in particular PAG plants, Volvo fares very well in the Ford empire, with a level of productivity comparable to Ford’s volume operations. In terms of overall production volume, Figure 18.2 and Table 18.2 clearly show a high level of stability up until the purchase by Ford, at which point the production volume increased from previously just below 400,000 units, to an average and projected output of 450,000 units. The long-term target of 600,000, set by Ford, seems unlikely to be met in the near future despite the increase in product variety. Overall, if one considers the production figures at Volvo from 1990–2006 it is clear that Ford has helped only slightly in volume (+10 per cent increase), but overall oversaw a fragmentation of product range, which meant that the basic issue of insufficient volume by model has not been alleviated. Ford has overseen an expansion into the SUV segment, as well as the compact car segment, yet overall these have not resulted in the projected increases in sales. In fact, considering Volvo’s trajectory since 1994, its production volume fell by 10.8 per cent in relation to the overall growth of the world passenger car production!
1994
1995
1996
1997
1998
Volvo production by car model, 1994–2006
Source: Volvo Annual Reports.
Volvo C30 Volvo S40/ 94,900 81,100 99,050 114,150 151,260 V40, 400 S60 Volvo 163,900 186,900 196,640 203,670 201,620 S70/V70 Volvo C70 – 880 9,210 Volvo 92,200 106,600 33,960 28,290 9,100 S90/V90 Volvo 940 Incl. Incl. 38,600 39,450 6,720 above above Volvo S80 21,770 Volvo XC90 Total 351,000 374,600 368,250 386,440 399,680
Table 18.2 2004
2005
N/A
N/A
N/A
N/A N/A
44,274
9,221
40,725 11,346
7,773
38,185 79,524
7,386
30,450 88,749
7,158
27,178 91,107
1,365
29,801 82,019
15,110
53,384 95,920
426,705 411,960 395,677 420,781 461,606 445,010 426,803
67,552
12,295
2006
7,209 92,460 162,829 155,335 143,360
2003
24,876 108,831 104,411 85,058 65,441 62,028 165,659 127,525 123,406 118,168 106,979 107,997
2002
N/A
2001
156,323 122,109 108,016
2000
N/A
1999
361
362 The Second Automobile Revolution
Markets and distribution While Volvo is most famous for its Uddevalla experiment, it did in fact make equally significant innovations in its distribution system. In 1992, Volvo launched the customer-ordered production (COP) model in an effort to increase customer responsiveness (Hertz et al., 2001; Holweg and Pil, 2004; Reichhart and Holweg, 2007). This was a key innovation in the European motor industry, which with the exception of some of the luxury producers, had made products based on a forecast model. Distribution challenges in terms of managing higher product variety led to the development of the build-to-order (BTO) model, which starts almost ‘accidentally’ in the Torslanda plant in the logistics department, and is rolled out in the entire Volvo organisation soon thereafter. This initiative was supported by a change in the company’s information systems and its order processing process. As one executive who was part of the initial project commented to us, the initial change had ‘brought down inventory levels across Europe almost overnight’. This change was driven by an initial prioritisation of orders coupled with a simple change in the order booking process. Volvo is one of the first vehicle manufacturers to implement such a system: Renault, BMW and Nissan following only half a decade later. The COP approach resulted in a drastic reduction of stock levels (see Figure 18.3 for the impact on stock levels in the European market), and further cemented Volvo’s position as a luxury maker as it was able to provide customised vehicles in short lead times. The COP model was rolled out to Europe, where more than 60 per cent of vehicles are currently built to order, while many of the overseas markets are still largely served from stock, in particular the US market. In terms of market structure, sales were distributed in 2005 as 57 per cent to Europe, 31 per cent to NAFTA, and 12 per cent to the rest of the world. Hence, the strong reliance on the US export market has not been solved for Volvo yet. With the initial conflicts between implementing such a lean distribution and production system, new conflicts have emerged, as shown in stock levels which have been increasing again after the Ford takeover, which is related to a decline in the BTO content. This trend continues despite the manufacturing arm’s capabilities to deliver customer-specific orders with acceptable leadtimes and high delivery precision. As one of Volvo’s vice presidents remarked to us in a recent interview: ‘We now have the worst of both worlds; a flexible production system [with the associated costs] and high inventory levels in the distribution system with customer discounts to get rid of the cars.’ The main reason for this increase has been a cultural clash between Ford’s wholesale stock-push mentality, and Volvo’s build-to-order or pull mentality. While Ford has been pushing Volvo hard to increase volumes, this has come at the price of compromising Volvo’s COP approach to sales and distribution. The Ford volume imperative was ‘lower cost due to higher volume’, with a
Volvo and its Malcontents 363
Days of demand (index: 1990 ⫽ 100)
100 90 80 70 60 50 40 30 20 10 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Year Figure 18.3
European stock levels in days of demand (index: 1990 = 100)
Source: Reichhart and Holweg (2007).
view towards raising overall output from the average 400,000–450,000 units to 600,000 units, and such wholesale operations clashed with Volvo’s BTO approach and its customer-driven planning and distribution system. In terms of development of markets, there is also the increasing importance of the US market, which also features in the product range (SUVs etc.) – a strategic issue that Volvo has been struggling with in the past, and will have to deal with in the future.
Concluding thoughts Volvo is celebrating its eightieth anniversary as we write this chapter. Over this period of time the first observation is its remarkable resilience: Volvo still portrays well with respect to safety, but it has changed fundamentally on a number of fronts. Its manufacturing strategy has evolved dramatically beyond the reflective production for which it was once known. However, Ford’s influence over the past eight years is also having an impact: from the abandonment of dock-build system at Uddevalla, to increased efforts to drive the Ford production system into all Ford-owned facilities, Volvo’s manufacturing character has changed dramatically. However, there are legacy impacts on how work is organised, and in particular, Volvo has followed efforts we also observe at Toyota, to create coherent group-level activities and to strengthen group identity and responsibility – these mirror efforts that were once centred on the individual (Pil and Fujimoto, 2007).
364 The Second Automobile Revolution
Beyond the workforce strategies, perhaps the biggest identity crisis we see at Volvo is the strategic clash between Ford’s wholesale mentality and Volvo’s BTO model, which has seen production volumes rise – yet stock levels rose at the same time. Overall it is clear though that Ford’s push to increase volume through increasing the model range and sales targets has only led to a minor increase in overall production volume, yet has resulted in a drastic fragmentation of the production volume. In comparison to the development of global vehicle production Volvo is clearly falling behind; it has missed out on the emerging markets, and – like many other luxury makers such as Jaguar or BMW – is heavily reliant on the US market, while facing strong financial pressure when expanding the model range into the lower segments of the market. Volvo has a number of strengths, including a very strong brand identity, the remnants of a unique ability to build to customer order, and strong employee engagement and good employee relations. Being part of a volume producer targeting greater efficiency and scale has put pressure on these strengths, but has also brought discipline on some fronts. This has at times created a schizophrenic environment, with competing directives. As Ford has been sending conflicting messages about divesting Volvo as early as 2007, we hope that the strengths that are Volvo’s own, and the financial discipline introduced by Ford, will be the dominant legacies. While tension between competing imperatives may be healthy in a munificent environment, these will need to be resolved in ways that leverage those strengths that have given Volvo its unique identity and historical endurance if it is to successfully navigate its next phase of ownership.
Bibliography Adler, P. S. and Cole, R. (1993) ‘Designed for Learning: a Tale of Two Auto Plants’, Sloan Management Review, 34(3): 85–94. Adler, P. S. and Cole, R. (1994) ‘Rejoinder to Berggren’s Critique’, Sloan Management Review, 35(2): 45–9. Argote, L. (1999) Organizational Learning: Creating, Retaining & Transferring Knowledge. Berlin: Springer. Berggren, C. (1993) ‘The Volvo Uddevalla Plant: Why the Decision to Close it is Mistaken’, Journal of Industry Studies, 1(1): 75–87. Berggren, C. (1994a) ‘NUMMI vs. Uddevalla: a Rejoinder’, Sloan Management Review, 35(2): 37–49. Berggren, C. (1994b) ‘Point/Counterpoint: NUMMI vs. Uddevalla’, Sloan Management Review, 35(2): 37–49. Berggren, C. (1994c) The Volvo Experience: Alternatives to Lean Production in the Swedish Auto Industry. Basingstoke: Palgrave Macmillan. Berggren, C. (1998) ‘A Second Comeback or a Final Farewell? The Volvo Trajectory, 1973–1994’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, pp. 418–40.
Volvo and its Malcontents 365 Cutcher-Gershenfeld, J. et al. (1998) Knowledge-driven Work: Unexpected Lessons from Japanese and United States Work Practices. Oxford: Oxford University Press. Ellegard, K., Jonsson, D., Engstrom, T., Johansson, M. I., Medbo, L. and Johansson, B. (1992) ‘Reflective Production in the Final Assembly of Motor Vehicles: an Emerging Swedish Challenge’, International Journal of Operations and Production Management, 12(7/8): 117–33. Hertz, S., Johansson, J. K. and de Jager, F. (2001) ‘Customer-oriented Cost Cutting: Process Management at Volvo’, Supply Chain Management: An International Journal, 6(3): 128–41. Holweg, M. and Pil, F. K. (2004) The Second Century: Reconnecting Customer and Value Chain through Build-to-Order. Cambridge, MA: MIT Press. Nelson, R. and Winter, S. (1982) An Evolutionary Theory of Economic Change. Cambridge, MA: Belknap Press. Pil, F. K. and Fujimoto, T. (2007) ‘Lean and Reflective Production: the Dynamic Nature of Production Models’, International Journal of Production Research, 45(16): 3741–61. Polanyi, M. (1966) ‘The Logic of Tacit Inference’, Philosophy, 41(155): 1–8. Prokesch, S. (1991) ‘Edges Fray on Volvo’s Brave New Humanistic World’, New York Times, 7 July. Reichhart, A. and Holweg, M. (2007) ‘Lean Distribution: Concepts, Contributions, Conflicts’, International Journal of Production Research, 45(16): 3699–3722. Sandberg, A. (ed.) (1995) Enriching Production: Perspectives on Volvo’s Uddevalla Plant as an Alternative to Lean Production. Aldershot: Avebury. Szulanski, G. (1996) ‘Exploring Internal Stickiness: Impediments to the Transfer of Best Practice within the Firm’, Strategic Management Journal, 17 (Winter Special Issue): 27–43. Volvo AB, Volvo Car Corporation and Ford Motor Company. Annual Reports, various years. Williams, K., Haslam, C. and Johal, S. (1995) ‘Fait accompli? A Machiavellian Interpretation of the Renault–Volvo Merger’, in A. Sandberg (ed.), Enriching Production: Perspectives on Volvo’s Uddevalla Plant as an Alternative to Lean Production. Aldershot: Avebury, pp. 309–26.
19
Production Counterfeits and Policy Collisions: the Rover Trajectory – a Salutary Tale Dan Coffey
Introduction Rover Group’s acquisition by BMW and its dissolution little more than five years later marks the end of an era for the British car industry. This chapter builds on earlier work by the author and others to provide an overview of Rover Group’s formation, the controversy over the break-up of its partnership with Honda and acquisition by BMW, and the subsequent trajectories of each of its principal production sites. Neglected contradictions between the market-oriented approach of BMW and Japanese-inspired manufacturing practices are noted – a striking example, played out at Rover in the 1990s, of a strategic mismatch in firm policy. More generally, the history of a Britishowned car industry abounds in instances where strategies not only ‘meet’, but ‘collide’ (Boyer and Freyssenet, 2002). A persistent theme to emerge throughout is that ‘British’ decline is not readily disentangled from the effects of internationalisation on the sector.
Contextualising the Rover trajectory The Rover story is complex and at times contradictory. The main essentials are perhaps most easily introduced by first considering Rover in relation to British Leyland (BL). From BL to Rover Group Ltd. The 1950s and 1960s saw a number of smaller British car and vehicle assemblers and parts suppliers draw together in umbrella organisations. The original Rover was one firm subsumed in this process, becoming part of Leyland Motors in the 1960s after a career of more than six decades’ standing as a carmaker, and boasting commercial successes which included the famous Land Rover, launched in 1948. Another major domestic car group, British Motor Holdings (BMH) had been formed (in 1966) by merger between Jaguar and the British Motor Company, itself the product of a somewhat earlier 366
The Rover Trajectory: a Salutary Tale 367
merger (in 1952) between Austin and Morris; and in 1968 BMH merged with Leyland – incorporating Rover – to form the British Leyland Motor Corporation (BLMC). In this way, a diverse array of independent British car and commercial vehicle producers were drawn into a single manufacturing organisation, with the ‘Rover’ business just one amongst many. The impetus for this movement was in part competitive pressure on smaller British firms and in part government policy direction, with hopes that a ‘national champion’ would emerge, to compete on a par (say) with VW in Germany. Such hopes were quickly dashed: the progressive freeing up of trade in Europe and British entry to the (then) EEC in 1973 saw car imports surge (from less than 10 per cent of the UK market in 1968 to almost 60 per cent by 1980: Wilks, 1984: 70–1), and the overall balance of payments (in cars) steadily worsened, entering deficit in 1975. Market share was now under heavy pressure, and in the midst of a recession. Nationalisation, prompted by BLMC’s mounting financing difficulties, saw the British-owned car industry brought largely into public ownership, first as British Leyland Ltd. in 1975, and then, in 1978, as BL Ltd. The ensuing period of public ownership, however, after the abandonment of an initially more positive supporting plan, proved to be one of retrenchment, closures and fragmentation, and the 1980s saw dismemberment of the group: whole branches, companies like Jaguar (acquired by Ford) and the old Leyland truck and van business (sold to DAF), were hived off and purchased by foreignowned transnationals intent on expanding their base of operations. The remaining rump of businesses, substantially reduced, became Rover Group plc in 1986, although trading at this juncture continued under the name Austin Rover. A single trading company was then created to form Rover Group Ltd., in 1989. At the same time the 1980s saw a wider policy shift towards privatisation. Rover was made a wholly owned subsidiary of British Aerospace plc (BAe) in 1988, when the government of the day sold its majority shareholding in the car business, making it part of what was then Britain’s largest manufacturing and engineering organisation. Car assembly operations were now concentrated in just three spots: small and lower-medium sized cars at Longbridge (Birmingham), upper-medium and large cars at Cowley (Oxford), and Land Rover products at Solihull. Each was a long-established production site with emblematic roots in the ‘British’ industry, being associated historically (and respectively) with Morris, Austin and Rover. The Rover brand itself was restored to a place of prominence, but in the greatly reduced circumstances of the remnants of a once substantial British-owned industry. Rover and Honda It is impossible to comprehend either this or later periods without noting the prominence of foreign players in the industry, and the successful
368 The Second Automobile Revolution
development of UK-based car (and commercial vehicle) production as one branch in the regional operations of transnationally mounted firms. In the period stretching through to the formation of British Leyland Ltd. in 1975, the main foreign players with UK manufacturing operations were American. The impact of US giants Ford and GM (also Chrysler) on the trajectory of the British industry in these years has long been recognised as important (Church, 1994; Dunnett, 1980; Wilks, 1984). By the 1970s, the (then) EEC was the relevant sphere of operations for these firms – with Ford both the UK’s largest car importer and its dominant foreign producer (Cowling 1982: 143; Wilks, 1984: 70–5), a situation contrasting with the growing relative weakness of British car firms in Europe. In these years, criticism of the conduct of the American transnationals was often sharp. The 1980s then saw the emergence of a second wave of foreign direct investment in UK-based car manufacture, led by Japanese firms. In this period the fortunes of the Rover Group overlapped significantly with the aspirations of Honda to develop and maintain a presence in the UK. The ‘partnership’ with Honda dated from the end of the 1970s. The 1981 launch of the Triumph Acclaim (a repackaged Honda Ballade) saw the fruition of what was in effect an arrangement to sell Honda vehicles under licence; this was followed, in 1984, by a successor range, the 200 series saloon (a Honda Civic). But the relationship soon progressed to encompass more substantial British ventures, commencing with the Rover 800, developed jointly with the Honda Legende, and launched in 1986, followed by the new Rover 200 and 400 series launched in 1989 – a complex project overlapping in design with, but by no means reducible to, the new Honda Concerto. In the same year, Honda acquired a 20 per cent stake in the British company’s equity. Collaboration reflected industry politics of the day. As Mair (1998) notes, BL was initially interested in Honda because the British firm was seeking means to develop new products and a ‘small’ Japanese partner appeared not to threaten ‘sovereignty’; for its part, it enabled Honda to bypass import restrictions. But as the decade progressed, aspirations for a significant British-owned car industry diminished, and government policy increasingly emphasised measures to encourage further foreign direct investment – bearing fruit in the decisions of Japanese car assemblers to establish plants in the UK, led by Nissan at Sunderland (in 1986) and followed by Toyota as well as Honda. The Rover–Honda break The years following Honda’s acquisition of a stake in Rover were rather mixed ones from the viewpoint of Rover Group’s main plants. Land and Range Rover products, manufactured at Solihull, and operating in this as in earlier and later periods as something of a stand-alone business venture, fared reasonably well. The new 200 and 400 series models, built at Longbridge, were well received, and at least ostensibly profitable – although as we see (in
The Rover Trajectory: a Salutary Tale 369
the next section) there were problems. But the decision to build these cars at Longbridge proved a serious blow for Rover’s Cowley plant (Greenhalgh and Kilmister, 1993: 45), which retained only an ageing set of models, with no major replacements in sight. The onset of the 1990s saw the prospect of closure, and a campaign was mounted to save the Oxford site, supported by academic activism (Hayter and Harvey, 1993). Nonetheless, so far as Rover and Honda are concerned public perceptions were almost entirely positive. Expectations were dashed when the Rover Group was abruptly sold by BAe in 1994 not to Honda, but to the German producer BMW.
The BMW controversy Commentators differ in assigning positive and negative motives to BMW’s decision to buy Rover. Eckardt and Klemm (2003) locate it in a longer-run trajectory towards internationalisation of operations dating as far back as the 1970s, the era in which BMW ‘established and sharpened its image as a producer of sporty, innovative and high-tech automobiles, with a presence in the premium market segments’. Overseas sales representation was established in the US (the largest export market) in 1975; the first international office for global sourcing followed less than ten years later. The 1990s saw this process enter a new phase, with plans to establish production facilities around the world – and the Rover acquisition in 1994 one step in this process. More negative and tactical reasons (although not wholly inconsistent) have also been suggested: fearing the disadvantages of small size and under pressure in the quality cars market, BMW aimed to eliminate a potential rival (Rover) and expand its portfolio without risking its own brand strength and exclusivity (Brady and Lorenz, 2001: 6–7, 14–15). It is broadly agreed, however, that BMW adopted a respectful attitude to its British acquisition, a ‘sister company’, informed by commitment to a distinctive Rover image, and by socio-cultural sensitivities (Eckardt and Klemm, 2003). One of the German firm’s first acts was to invest heavily in Cowley: a new model (the 75 series saloon) was put into development for the site, breaking the existing Rover product cycle. At the time there seemed much to commend the purchase. The 1989 rebranding of Rover Group – labelled ‘Roverisation’ – sought to re-establish the reputation which had been Rover’s before its absorption into BL, of advanced product engineering for prestige markets. Competition with the likes of Ford for volume sales gave way to a new strategy, embodied in the new Rover 200 (hatchbacks) and 400 (saloon) series launched the same year, aimed at repositioning products at the top end of the lower-medium size segment. It was envisaged that younger car drivers with sufficient income, car enthusiasts, and second car purchasers – as well as women drivers – would join the older male drivers hitherto buying into the marque; it was hoped
370 The Second Automobile Revolution
that some sales would even compete for custom with the bottom end of the BMW 3 series. The range would be expansive, with plans for an increasingly rich menu of factory-fit options. Coupé and cabriolet derivatives were added a few years after launch. Considerable hard investments had been made in the Longbridge site housing production, vis-à-vis body and paint shops, with some technologies directly purchased from Honda (Mair, 1998). Rover’s own design input for part of the range was also considerable, including a new aluminium K-series engine, developed to add to Honda-supplied engines. There was talk of a gradual reassertion of business independence, and a more confident future. Judged from the outside, BMW’s 1994 acquisition seemed propitious. Land and Range Rover products, performing well in sales, were known to be profitable. With its prompt investment in a new Cowley model, BMW had saved the site. And the new 200 and 400 series proved to be the Rover Group’s biggest single revenue generator. The question as to why the acquisition then ‘failed’ is difficult, and likely to generate lasting controversy. But little more than five years later news broke – on 14 March 2000 – that BMW intended to dissolve the Rover Group. BMW chose to maintain production at Cowley, and to retain the ‘Mini’ brand; Ford acquired the Solihull site and the Land Rover business (and the main Rover R&D facility at Gaydon); but in a dramatic turnaround of fortunes no buyer was found for Longbridge – a crisis loomed. Why did the marriage fail? One possible explanation is that BMW underestimated the amount of investment financing the Group needed, and then lost patience. By investing immediately in Cowley BMW had certainly saved the Oxford site from closing. But a breakdown of financial returns across the Rover divisions prior to the BMW purchase also indicates that Longbridge was struggling – albeit not in quite the same way. Large developmental overheads for the new 200/400 series (as well as the terms upon which Rover did business with Honda) meant that this product line was always close to falling below break-even despite good sales, with pressures to economise scaling back ambitions for the next generation of the line-up (relaunched in 1992) vis-à-vis quality and complexity. Only Land Rover was unambiguously self-sustaining. Problematically, Longbridge had also been experiencing severe problems in meshing marketing ambitions for the new 200/400 line-up with a simultaneous experiment in just-in-time (JIT) manufacture, also launched in 1989, following a decision to adopt what was perceived to be a ‘Japanese’ product supply structure. Executive policy decided that the new series would take a lead in repositioning Rover both as a quality marque commanding premium prices and as an ‘aggressive’ JIT producer. A new logistics unit was created at Longbridge: extra purpose-built loading decks were constructed around the site’s trim and assembly sections, and civil works undertaken; old routes to
The Rover Trajectory: a Salutary Tale 371
plant-side handling for parts (bulk delivery, decanting, storage, withdrawal, trackside delivery and assembly) were abandoned for as many parts as deemed practicable. But the scheme proved deleterious. A complex network of warehouses and distribution centres had been employed to provide a sort of ‘halfway house’ for suppliers whose own manufacturing facilities were dispersed not only across the British mainland but also Europe (and Japan). This was intended to allow parts to be shipped from suppliers in greater bulk than that reserved for ultimate ‘small-lot’ delivery to the assembly plant, while adding some scope for final collation and checking. But the absence of local stock-cover, combined with handling and transit lags for parts (and time needed by suppliers to organise their own production), severely compromised manufacturing flexibility: after only two years (by 1991) the ability of Longbridge to accommodate 200/400 series factory-fit orders deteriorated to the point where the lead time from order-submission to production was more than half as long again as for the whole Group average, and getting worse (Coffey, 2005, 2006, 2007). The difficulty became acute, exacerbated by mounting scheme costs: overheads for monitoring increased, to which was added an increased materials handling bill. These problems grew worse as more model derivatives were added to the line. Not surprisingly, this was not reflected in the readily managed and consistently positive media coverage of the day. It was generally supposed at the time that the just-in-time experiment was successful, improving plant ‘flexibility’ (Foreman-Peck et al., 1995: 234; Whisler, 1999: 395). One view even holds that BMW acquired Rover having confused the improving effects of Japanese manufacturing practices with the longer-term business viability of the Rover brand (Brady and Lorenz, 2001). The somewhat bleaker reality added to pressure at the Longbridge site for reduced model complexity – at the further expense of the product development strategy envisaged for 200/400 series cars. The ‘build to order’ plan, modelled on BMW, but prior to the BMW purchase, was quietly abandoned, a decision bolstered by the observation that Japanese carmakers typically eschewed customisation of this kind, a view supported by a Rover management team tour of Japan in the early 1990s (Coffey, 2005, 2006). Honda’s own practice of the day was ‘few choices within a model type: few colours even’ (Mair, 1994: 184, 237). The attitudes this engendered were inherited by BMW, in 1994. Then BMW chairman Bernd Pischetsrieder, who played a major part in the Rover purchase, is reported as later explaining that: Rover could never achieve truly premium brand quality and prices unless it got away from what BMW saw as the volume oriented Honda approach . . . The contrast was embodied in the different production systems . . . making cars in batches, which Rover was still compelled to use because of its dependence on the Japanese company, also conflicted with BMW’s
372 The Second Automobile Revolution
marketing objectives for Rover. BMW builds cars to individual customer orders and wanted Rover to do the same. (Brady and Lorenz, 2001: 43; emphasis added) But whereas Longbridge was aggressive in moving towards a Japan-inspired just-in-time production regime prior to the BMW acquisition, Cowley had been tentative. Each of these considerations, taken together, helps explain the subsequent shift in sentiment away from the Longbridge site, and its largest car line. A post-1998 slump in sales also hit the site badly: 200/400 series cars performed particularly poorly, especially in the home market. Eckardt and Klemm (2003) describe the situation as a ‘latent crisis’ becoming publicly manifest, as the profitability of the German parent and its British ‘sister’ diverged: BMW’s Western European sales were up in the first half of 1999 on the previous year, but Rover’s had fallen sharply. Underlying political tensions at BMW were laid bare; Chairman Pischetsrieder was ousted.
Fragmented trajectories: MG Rover and beyond A sense of the fragmented trajectories of Rover Group’s disassembled parts, and the emergence of new transnational interests in UK-based car manufacture, is best obtained by briefly considering subsequent events at each of the major sites in turn. Longbridge: MG Rover and beyond A proposal by venture capitalists (Alchemy Partners) to organise sports-cars production at the unwanted Longbridge site, on a much reduced scale of activity, was rejected despite some government support, following intense trade union lobbying. Instead, production recommenced under the auspices of the Phoenix consortium, fronted by an ex-Rover CEO (John Towers) whose history with Longbridge pre-dated BMW. The business was re-established as MG Rover; finances included a BMW loan (the consortium obtained the Longbridge site for a peppercorn fee of £10). Pledges made included maintaining the workforce, and a quick return to profitability. But while licensed to produce under the Rover brand, assets (which still included 25, 45 and 75 series cars) did not include Land Rover, while BMW kept the ‘Mini’ brand; moreover, little developmental capability was left for new products (Brady and Lorenz, 2001; Holweg and Oliver, 2005). The venture’s shortlived history was one of almost unyielding uncertainty: MG Rover passed into administration on 8 April 2005. It is hard to disagree with the judgement that MG Rover was too small to survive long as a stand-alone producer, given its aspiration to maintain a sizeable assembly unit at the Longbridge site. It lacked overall business scale, with an output in its final year of little more than 100,000 units (and in a context
The Rover Trajectory: a Salutary Tale 373
where a far larger ‘small’ firm like BMW still produced more than a million), and market share at less than 3 per cent of the UK home market (Holweg and Oliver, 2005: 8, 27); its product portfolio was limited and ageing – demanding resources it could not provide. Attempts to find partners proved nugatory or inadequate: avenues explored included acquiring Qvale, an Italian sports car producer, and developing a new niche product, the MG SV. Product quality was seemingly sustained, judged by customer feedback (Holweg and Oliver, 2005: 18). But size matters. A partnership deal with Shanghai Automotive Industry Corporation (SAIC) failed, and the Longbridge site was mothballed, with allegations of mismanagement. A struggle ensued over MG Rover assets. Another Chinese firm, Nanjing Automobile Corporation (part of the NAC Group), won the bid to acquire the Longbridge plant, reportedly paying between £50 and £60 million. Site machinery, whole production lines included, were ‘lifted and shifted’ to China, where a number of new factory sites had been founded. Prospects of resumed operations at Longbridge were initially stymied by a property rights struggle over MG Rover models, with Nanjing and SAIC contending to build versions of the 75 series saloon (property rights for which were originally acquired by SAIC along with rights to the smaller 25 series car). However, production of the MG TF sports car – with first models displayed in 2007 – was scheduled to commence at Longbridge early in 2008. Hopes exist of further production, and some longer-term restoration of operations. Nanjing’s parent company (NAC) announced cooperation with SAIC – ‘to jointly build an automotive group that will become the largest world-class automotive group in China’ (Milner, 2007) – and a planned merger is confirmed. The early emergence of new players on the world stage, pursuing their own strategies of internationalisation, is evident. Acquiring equipment and designs explains only part of the impetus vis-à-vis MG Rover. Europe-based selling is also part of the strategy, with affluent Chinese reportedly susceptible to European fashions: developing an active base in the West is therefore fully consistent with an immediate focus on growing domestic sales in China. Longbridge is now the European headquarters of Nanjing – aspiring to a presence in sports and small and medium cars.
Land Rover: Ford and beyond Losses were not the reason for BMW’s Land Rover sale to Ford, although the development of another four-wheel drive vehicle (the X5) under the German firm’s own brand was a possible consideration (Brady and Lorenz, 2001: 135). Disposal represented capitalisation of an asset, and product portfolio restructuring. Production continued at Solihull, but with some Land Rover operations too at Ford’s Halewood site. Currently, Ford is itself in the process of seeking to sell Land Rover, along with Jaguar. Viewed over the past twenty-five years, the production trajectory has been positive: volumes
374 The Second Automobile Revolution
almost trebled in the second half of the 1980s, and increased more than five-fold over the next fifteen years (Holweg and Oliver, 2005: 30). Despite suggestions that Ford’s strategy for Land Rover (and Jaguar) faced stronger than expected German competition (Froud et al., 2006: 262–3), this latest fund-raising sale (subject to concerns about a pension deficit which could swallow much of the sales price) is also taking place in the context of Ford’s global difficulties, with record losses in the US. Ford wishes to contract to supply engines, and is thought to be seeking a remaining stake. Again the changing contours of the world auto industry are evident. The Indian car maker Tata is now confirmed as Ford’s preferred bidder, with union support in the belief that production is likely to stay in the UK, including the new ‘baby’ Land Rover. Cowley: BMW and beyond For its part Cowley under BMW is now almost synonymous with the ‘Mini’, the German firm’s decision to retain the brand proving auspicious. The model was relaunched from Cowley in 2001, and again five years later, in 2006: at this juncture engine production for the car was switched from Brazil to the Hams Hall factory near Birmingham, reportedly raising the car’s British content towards 60 per cent. Plans to further expand production at Cowley were announced in early 2007 – by 2005 already in the region of 200,000 units a year, reflecting dramatic cumulative year-on-year growth for the model. In recent years Mini sales have fluctuated between one-sixth and one-seventh of all BMW sales worldwide: a growing US market is predicted. A new ‘Clubman’ variant has resuscitated another old British brand. What is perhaps most striking is the UK rehabilitation of BMW’s reputation as a foreign investor – with the ‘Britishness’ of assembly plant (Cowley), engine plant (Hams Hall) shop and press shops (Swindon) repeatedly emphasised in press coverage.
Lessons and prospects What can be made of this complex history? An obvious point of note is that decline of a British-owned car industry has proceeded alongside the growing importance of foreign transnationals and internationalisation of markets. Government In the run-up to the formation of BLMC in 1968, and entry to the EEC in 1973, government policy towards the British-owned car industry was motivated largely by concerns over business size and the growth of a national champion. But the state was also responsive to internal lobbying from significant tranches of industry for a freeing up of markets, quarters which certainly included important car industry players like Ford and GM, already operating as transnationals in the UK (Cowling, 1982). In an industry characterised by
The Rover Trajectory: a Salutary Tale 375
what Rhys (2005) describes as highly contested oligopoly, the ability to attack rivals’ markets without fear of effective retaliation is a major competitive advantage. UK government policy helped achieve an asymmetry of this kind – but what still impresses is the speed with which ‘British’ fortunes plummeted in the 1970s. In a period of home market fragmentation – as imports from Europe flooded in – the policy goal of achieving greater business scale and consolidated production at the expense of the individual identities of long-established British firms was badly misguided, even if all immediate organisational obstacles had been overcome (Mair, 1998). In the later period hopes for a national British champion slowly dissipated. Instead, institutional biases within government, defending a policy of ‘openness’ to inward foreign direct investment, come to the fore: by the late 1980s not only Japanese transplants to the car industry, but ‘Japanisation’, was a policy panacea. When public disquiet was caused by the fact that BMW dissolved the Rover Group without prior notice to or consultation with its workforce (mandatory in Germany), this was brushed aside as irrelevant by the British government (see HC, 2000–2001: xxxii–xxxiii): a strong policy commitment to free capital movement was justified specifically by reference to its previous success in attracting Japanese transplants to the UK (ibid.: 95). In the final period, with the short life and not unexpected death of MG Rover, the diminution in any significance attached to a British car firm is palpable. The failure of the company was announced during an election campaign: it had little impact. Chinese involvement in the Longbridge site is welcomed: but ‘ownership’ is not an issue. And gradually, BMW’s success at Cowley has leavened the embarrassment of Rover. Unions A parallel decline can be noted, with qualifications, in the relative strengths and bargaining positions of the trade unions representing workers. Industrial relations were not in a healthy state in the Britishp-owned car industry through the period now associated with BLMC and its nationalised successors, popularly attributed to the effects of a highly fragmented collective bargaining structure combined with a powerful shop steward movement (union representatives elected on the shop floor): complaints that management reforms were consistently ‘blocked’ were common, if exaggerated. But the structure of bargaining reflected the original structure of the industry, a potential strength had a different developmental path been taken in strategic planning for the sector (the historical discussion and comments of Freyssenet and Mair, 2000 are of interest here). In retrospect, there is much to be said for the ‘weak union’ hypothesis: union organisation was too fragmented to provide a significant countervailing force – resulting in a series of essentially reactive disputes in the 1970s.
376 The Second Automobile Revolution
But a more quiescent workforce in later decades did not resolve Rover’s later difficulties, with workers first at Cowley and then (ultimately) at Longbridge contending with plant closure. The Rover ‘New Deal’ of the early 1990s, the later three-year deal with BMW on pay and jobs, and a generally supportive union stance, did not save the Group. Production counterfeits and policy collisions Several types of production counterfeit are evident in Rover’s long history. The dream of a large-scale British-owned car manufacturer, able to compete in freer European and world markets as a national champion, proved a conceit: the policy aspiration collided badly with the reality of the underlying industrial structure. A foolhardy attempt by the Rover Group in 1989 to marry a BMW-style marketing aspiration with a product supply regime modelled on firms with a quite different approach proved to be another production ‘counterfeit’, which again ended badly for the firm. It is possible, although not certain, that more intelligible policies would have fared better. But if a single consistent theme runs through the history of the last major British car firm, it is the progressively dissipating effects on its fortunes of the internationalisation of the sector.
Appendix Table 19.1 Year
1968 1969 1970 1971 1972 1973 1974 1975 1976* 1977 1978 1979 1980 1981 1982 1983
Rover, 1968–2003
Total car production (Units)
Total sales outside UK (Units)
Total sales (£m)
Operating profit (£m)
Total employees
818,289 830,874 788,737 886,721 916,218 875,839 738,503 605,141 687,875 651,069 611,625 503,767 395,820 413,440 405,116 445,364
397,501 408,465 368,380 385,837 347,325 347,998 322,523 256,672 320,836 293,316 247,936 200,215 157,829 126,249 133,862 96,897
907 970 1,021 1,177 1,281 1,564 1,595 1,868 2,892 2,602 3,073 2,990 2,877 2,869 3,072 3,421
45.5 45.7 13.4 46.7 41.0 58.2 19.3 −38.1 116.1 56.1 71.3 −46.2 −293.9 −244.6 −125.8 4.1
188,247 196,390 199,524 193,703 190,841 204,149 207,770 191,467 183,384 194,610 191,853 176,790 157,460 126,267 107,763 103,216 (Continued)
The Rover Trajectory: a Salutary Tale 377 Appendix Table 19.1
Continued
Year
Total car production (Units)
Total sales outside UK (Units)
Total sales (£m)
Operating profit (£m)
Total employees
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995
383,324 465,104 404,454 471,504 474,687 466,619 464,612 395,624 378,797 406,804 e460,000 e475,000
78,664 112,783 123,558 165,708 137,048 138,540 168,928 187,873 159,107 161,932 NA e228,000
3,402 3,415 3,412 3,096 3,224 3,430 3,785 3,744 3,684 4,288 4,900 5,600
−11.7 −34.6 −246.4 16.8 71.0 73.0 65.0 −52.0 −49.0 56.0 82.0 91.0
96,001 77,849 73,396 49,207 42,300 39,900 41,400 38,300 34,600 33,100 36,238 40,100
504,125 522,466 486,105 429,157 174,885 163,144 142,928 136,065
272,228 292,581 (exLR) 168,000 (exLR) 141,000 70,000 63,000 43,000 36,000
6,251 6,324 5,482 4,801 2,310 1,321 1,493 1,307
−13.1 135.2 −447.8 −2,119.60 549.6 −182.5 −69.6 −88.2
39,335 38,763 38,688 31,393 14,409 5,035 5,062 4,615
1996 1997 1998 1999 2000 ** 2001 2002 2003
Sources and notes: 1968–1995 The entries follow Statistical Appendix 15 in Mair (1998), with data compiled from the Society of Motor Manufacturers and Traders (SMMT); Annual Reports for BLMC, BL, British Aerospace, BMW. * 1976 = fifteenmonth accounting period. e = author estimate. Original notes are as follows: ‘Total production is for passenger cars (includes Land Rover’s Range Rover and Discovery but not Land Rover utility vehicles, includes 20,000–30,000 Jaguar cars per year until Jaguar was sold in August 1984). Total sales outside UK includes direct exports and CKD production abroad of cars (CKD car production took place until 1983). Operating profit is before interest, tax, extraordinary expenditures, and transfers from reserves (from 1988–1993 rounded figures for operating profit are given by owner British Aerospace). Total employees is average employees during year (except 1994 and 1995: year end): overseas proportion 12–20% during 1970s, varies 9–28% during 1980s, not available after 1987 but limited to sales companies. Total sales, profits and employment refer to the whole BLMC/BL company including Jaguar and divisions making buses and trucks, components and industrial vehicles from 1968 to mid-1980s. Production and sales abroad refer to cars only. It is difficult to establish the proportion of turnover due to car sales; 65–75% during the 1970s and early 1980s, rising to 90% in 1987 with divestments and 95% in the early 1990s (the remainder being Land Rover utility vehicles).’ 1996–2003 The series for estimates of total production follows Holweg and Oliver (2005). Sales, operating profit and employee totals are taken from ICC Plum (Business Report). Entries have been cross-checked against each of the relevant annual accounts returned for Rover. **The operating profit shown for 2000 is calculated in the ICC Plum series as the sum of operating losses for that year plus profits on the disposal of Group assets (a breakdown is given on page 12 of the 2001 Annual Accounts). A change in accounting policy regarding provisions against investments and group debtors in the Annual Accounts revised estimates of profit and loss for 2001, and thereafter: the ICC Plum entries for operating profit incorporate the resulting restatement of profit and loss accounts for the Group. Operating profits originally reported in Rover accounts prior to this policy change deviate marginally from figures in the ICC Plum series (the original returns made for operating profits in the annual accounts for each year, sometimes with following year revisions, were: −£16.2m in 1996; +£130.5m in 1997; −£437.9m in 1998; −£2201.1m in 1999; −£605.2m in 2000, before adding profits on disposal of Group Assets; −£230.4m in 2001). Entries for employment levels are the average number recorded in annual accounts for each year. Estimates for exports outside of the UK were made as follows: 1996 and 1997 – annual reports indicate circa 54% and 56% of year-long production sold outside UK, respectively; 1998–1999 exclude Land Rover products, and are approximate estimates using UK registration data for Rover cars; 2000–onwards are similarly approximate estimates for MG Rover.
378 The Second Automobile Revolution
Acknowledgements Thanks are due Michel Freyssenet and Carole Thornley for suggestions on this chapter. Since this chapter was written Tata has acquired Jaguar. The British government is coordinating with the EU on bail-out measures for the industry in the face of economic crisis. The future is uncertain – not least for US-owned sites. Bibliography Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan Brady, C. and Lorenz, A. (2001) The End of the Road: BMW and Rover – A Brand Too Far. London: Financial Times/Prentice Hall. Church, R. (1994) The Rise and Decline of the British Motor Industry, Basingstoke: Macmillan. Coffey, D. (2005) ‘The BMW–Rover–Toyota Complex: Interpreting Strategic Matching in Car Assembly’, International Journal of Automotive Technology and Management, 5(3): 320–35. Coffey, D. (2006) The Myth of Japanese Efficiency: the World Car Industry in a Globalizing Age. Cheltenham and New York: Edward Elgar. Coffey, D. (2007) ‘Automotive Efficiency: Lean Beginnings’, Manufacturing, June/July: 18–19. Cowling, K. (1982) Monopoly Capitalism. Basingstoke: Macmillan. Dunnett, P. (1980) The Decline of the British Motor Industry. London: Croom Helm. Eckardt, A. and Klemm, M. (2003) ‘The Internationalization of a Premium Car Producer: the BMW Group and the Rover Case’, in M. Freyssenet, K. Shimuzu and G. Volpato (eds), Globalization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan, pp. 170–97. Foreman-Peck, J., Bowden, S. and McKinlay, A. (1995) The British Motor Industry. Manchester and New York: Manchester University Press. Freyssenet, M. and Mair, A. (2000) ‘De British Leyland à Rover: la recherche d’une stratégie de profit pertinente’, in M. Freyssenet, A. Mair, K. Shimuzu and G. Vopato (eds), Quel modèle productif? Trajectoires et modèles industriels des constructeurs mondiaux. Paris: La Decouverte, pp. 441–55. Froud, J., Sukhdev, J., Leaver, A. and Williams, K. (2006) Financialization and Strategy: Narrative and Numbers. London and New York: Routledge. Greenhalgh, C. and Kilmister, A. (1993) ‘The British Economy, the State and the Motor Industry’, in T. Hayter and D. Harvey (eds), The Factory and the City: the Story of the Cowley Automobile Workers in Oxford. London: Mansell, pp. 26–46. Hayter, T. and Harvey, D. (1993) The Factory and the City: the Story of the Cowley Automobile Workers in Oxford. London: Mansell. HC (2000–2001) Vehicle Manufacturing in the UK: Report, Together with the Proceedings of the Committee, Minutes of Evidence, and Appendices. House of Commons, Session 2000–2001, Trade and Industry Committee, Third Report (HC 128). Holweg, M. and Oliver, N. (2005) Who Killed MG Rover? A Special Report from the Cambridge-MIT Institute’s Centre for Competitiveness and Innovation (CCI). Centre for Competitiveness and Innovation, University of Cambridge, April.
The Rover Trajectory: a Salutary Tale 379 Mair, A. (1994) Honda’s Global Local Corporation, Basingstoke: Macmillan and New York: St. Martin’s Press. Mair, A. (1998) ‘From British Leyland Motor Corporation to Rover Group: the Search for a Viable British Model’, in M. Freyssenet, A. Mair, K. Shimizu and G. Volpato (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press. Milner, M. (2007) ‘Chinese Car Rivals Join Forces in Global Group’, The Guardian, 28 July and 2 December (www.guardian.co.uk./business/2007/dec/02, accessed 3 December 2007). Rhys, D. G. (2005) ‘Competition in the Auto Sector: the Impact of the Interface between Supply and Demand’, International Journal of Automotive Technology and Management, 5(3). Whisler, T. R. (1999) The British Motor Industry 1945–1994: a Case Study in Industrial Decline. Oxford: Oxford University Press. Wilks, S. (1984) Industrial Policy and the Motor Industry. Manchester: Manchester University Press.
This page intentionally left blank
Part IV Newcomers and Global Suppliers
This page intentionally left blank
20
Made in China: Joint Ventures and Domestic Newcomers Hua Wang
Introduction China is currently the second automobile producing nation in the world, after Japan. It overtook the United States in 2008 as total car production reached 9.3 million units. Between 2000 and 2008, the passenger car sector in China expanded at a striking annual growth rate of 21.1 per cent. This phenomenal growth was undoubtedly linked to the constant economic development in China since the ‘opening-up’ policy implemented in the late 1970s. China’s average annual GDP growth rate between 1978 and 2008 was 15.3 per cent. Passenger cars in China are produced by both Sino–foreign joint ventures and independent local carmakers. Leading original equipment manufacturers (OEMs) in Triads (USA, Europe and Japan) have all established equity joint ventures with large state-owned enterprises (SOEs). They produce global models and have taken a lion’s share of the Chinese market. In parallel, domestic companies are offering local own-brand customised cars. In 2006, thirty-five carmakers out of a total of fifty-one were independent local firms. The former produced 1.14 million units claiming a market share of 39 per cent. Leading Chinese OEMs were the Chery Group, the Geely Group and the First Auto Works Xiali affiliate, whose annual sales exceeded 200,000 units. Interestingly, carmakers that are now producing Chinese cars are dynamic small and medium enterprises (SMEs). Despite considerable entry barriers, these firms have successfully penetrated the market and are booming. They are competing directly with the global giants and though not all of them will survive, a study of their development provides valuable information concerning the fast-changing automobile industry in China. The emergence of Chinese carmakers may well pose challenges for global OEMs. The Chinese low-end segments, together with several niche markets, are ignored by foreign carmakers. This is due either to low profit margins or small production volumes. In the absence of competition, their Chinese counterparts have seized the opportunity to explore their comparative 383
384 The Second Automobile Revolution
advantage of low cost and low price. What is more, their designs are more appealing to local markets since they are a better match with local consumer preferences. From niche players to full-line carmakers, fledgling producers have taken to the wing, targeting both domestic and overseas markets. Mirroring the success of Japanese and Korean carmakers, Chinese OEMs are now establishing a firm foothold in the world automobile industry. Rather than analysing the competition between joint ventures and domestic OEMs, this chapter will focus on Chinese carmakers. It is important to understand the favourable social and economic conditions that have encouraged the emergence of homegrown Chinese automobile producers. Presenting a brief overview of several Chinese OEMs, this chapter highlights the case of the Geely Group, China’s first private carmaker. A summary of Geely’s productive model is included. The research framework of Boyer and Freyssenet (2002) facilitates future comparative work between the Geely model, the Taylor, Woollard, Ford and Sloan models employed in Western countries, and the Toyota Honda model of Japan.
Emergence of the Chinese automobile industry Creation of the automobile industry in the planned economy (1949–1978) When the People’s Republic of China was founded in 1949, the automobile industry did not exist. During the period of the planned economy, central government mobilised resources at national level to build several automobile factories for the production of trucks. The role of the government in the creation of the automobile industry was decisive. The first manufacturing site, First Auto Works (FAW), was established in 1953, located in Changchun City in Jilin Province, northeast China. This turn-key project was the fruit of cooperation with the former Soviet Union. China’s communist ally offered financial and technological support, manufacturing equipment, a system of management and a vehicle model. Production of the ‘Liberation’, a copy of the Soviet truck Zis 150, was started in 1956. An annual volume of 30,000 units was planned, equivalent to Toyota production levels at the time. However, 1962 marked the end of the Sino–Soviet collaboration due to bilateral political conflicts. FAW began manufacturing the vehicles independently. The firm also produced marginal quantities of the limousine ‘Red Flag’. Dongfeng Motors (DFM), located in Shiyan City, Hubei Province, was the second major national project launched during this period. Central government deliberately chose a remote and mountainous region in preparation for the Cold War. The Chinese economy was closed to the outside world and entered the 1960s period of autarky. For the production of a 2.5-ton military truck, DFM benefited from technology transfer from FAW. The new model ‘East Wind’ was in fact an improved version of the ‘Liberation’. It is worth
Made in China 385
mentioning that 95 per cent of the manufacturing equipment at DFM was produced by Chinese firms, in marked contrast to the massive equipment importation effected at FAW (HOCAI, 1996). This illustrates the astonishing progress of the Chinese automobile industry. Shanghai Automobile Factory, the predecessor of Shanghai Automotive Industrial Corporation (SAIC), initiated passenger car production in 1956. Though a relatively advanced industrial complex existed in Shanghai, the factory spent ten years bolstering production volumes due to the complexity of passenger cars. Based on the Mercedes 220s, the ‘Shanghai’ model was launched in 1966. Before 1970, average annual production was 1,000 units. Production expanded to more than 6,000 units in 1983, the peak period in its history. The ‘Shanghai’ monopolised the passenger car sector, maintaining 94 per cent market share throughout the 1970s. When the joint venture with Volkswagen started in 1985, production of the ‘Shanghai’ ceased. Between 1969 and 1970 the Chinese automotive industry experienced its first wave of expansion with a proliferation of vehicle manufacturing sites. In preparation for the Cold War several other national projects were established in backward provinces. In addition, local governments were offered incentives to develop industries in the regions under their jurisdiction. Since all factories were state-owned, technology was transferred at no cost from several ‘leading’ factories to the new assembling sites. Consequently the number of carmakers increased dramatically from twenty-two in 1966 to fifty-six in 1980. Automobile industry development and foreign direct investment in the transition period (1979–2001) During this period, global OEMs began entering the Chinese market through Sino–foreign joint ventures. This form of market entry was imposed by the Chinese government under their policy of ‘trading market for technology’. Due to the limited number of large Chinese carmakers, several global OEMs had to build alliances with the same Chinese firm. Examples include FAW’s partnership with Volkswagen, Toyota and Mazda, DFM’s alliance with PSA, Nissan, Honda and Hyundai-Kia, and SAIC’s cooperation with Volkswagen and General Motors. The Chang’an group established partnerships with Ford and Suzuki. The above list is by no means exhaustive. DaimlerChrysler, BMW, Fiat and Honda also established joint ventures with smaller players in Beijing, Shenyang, Nanjing and Guangzhou City respectively. New local carmakers emerged producing various models. During the previous period, the main vehicle product was the 4-ton truck. There was an enormous shortage of passenger cars, and light and heavy commercial vehicles. In 1980, the market share of passenger cars was a mere 2 per cent, or 5,418 units. With comfortable profit margins, large SOEs had less incentive to innovate. They were also subject to stifling bureaucracy and central government control. SMEs were more flexible, responding to consumer needs for
386 The Second Automobile Revolution
new types of vehicles and a burgeoning market. However, given that their technology was limited, each SME specialised in the production of one type of vehicle – a niche market strategy. Between 1980 and 1985 the number of carmakers doubled – from 56 to 114. After reaching a historical peak of 124 carmakers in 1992, the number of firms fell to 117 in 2001. With the exception of Tibet and Ningxia, automotive firms (including component suppliers) were located in all of China’s provinces (twenty-nine out of a total of thirty-one provinces and municipal centres) (CAIY, 2006: 495). By the mid-1990s, seventeen provinces had passenger car plants of which more than ten had established the automobile industry as the ‘pillar industry’ (CATARC, 1997: 166). The increasing number of domestic carmakers is not only explained by the profit-seeking opportunism of Chinese firms – the industry was also stimulated by the singular institutional environment in which these firms operated, notably the organisational complexity of the government and the federalism which characterised central/regional government relationships (Wang, 2004).
The rise of Chinese carmakers following China’s accession to the World Trade Organisation (2002 onwards) China has undergone profound change since its accession to the World Trade Organisation (WTO) in December 2001. At the macro level, China’s economic, legal and political institutions had to be brought in line with WTO rules. At the meso and micro levels, the new measures applied to trade, investment and tertiary sectors have reshaped the industry. For example, import tariffs for vehicles and components were slashed from 80–100 per cent and 15–50 per cent to 25 per cent and 10 per cent respectively. Importation quotas were phased out by 2006. In terms of trade-related investment measures (TRIMs), the entry barriers to foreign direct investment were lowered. With the exception of car assembly, the creation of joint ventures is no longer obligatory. Deregulation has been extended to automobile-linked financing services, international trade services, distribution and after-sales services (KPMG, 2004). The application of WTO rules has stimulated competition and increased industrial efficiency. As cited by IBM Consulting (2005: 7), prior to 2001, ‘Car models are mostly dominated by manufacturers, and the market is a sellers’ market. Few efforts have been done in studying local consumer habits.’ Between 2002 and 2003, car prices were cut by 20–30 per cent and thirty new models were introduced. Competition drove profit margins down at the industrial level from 20 per cent in 2003 to 10 per cent in 2004. Even so, the margins of foreign OEMs in China were still above the global industry level in 2004 (Accenture, 2005). FDI in China accelerated in this period. At industry level, FDI represented 37.7 per cent of the industrial output value and 65.1 per cent of gross profit.
Made in China 387 Table 20.1 Contribution of Chinese firms and FDI to the Chinese automotive industry by type of products, 2005
Engines Vehicles Components Motorcycles Refitted vehicles Whole industry
Industrial output value
Gross profit
Chinese firms
Chinese firms
52.0 62.3 50.8 82.1 93.6 62.3
FDI 48.0 37.7 49.2 17.9 6.4 37.7
19.2 24.7 41.7 73.9 N.A. 34.9
FDI 80.8 75.3 58.3 26.1 −11.1 65.1
Note: Investment from HK is integrated into FDI statistics. Source: CAIY, 2006, calculated by the author.
The contrast between these two figures confirms the positioning of FDI in the upper-range of each segment. For example, in the vehicle sector (including passenger cars and commercial vehicles), FDI realised 75 per cent of gross profit despite its industrial output of only 38 per cent. The domination of FDI in the engine segment was overwhelming, accounting for 81 per cent of industrial gross profit. Refitted vehicles and motorcycles were the least represented segments due to small volumes and low added value (see Table 20.1). The presence of Chinese firms was not negligible. Thousands of firms (including the components segment) contributed 62.3 per cent of the industrial output value and 34.9 per cent of gross profit for the whole industry. Due to the low technology level, all sectors exhibited lower profit margins than FDI. On the other hand, the performance of Chinese firms was impressive in terms of the industrial output value of the refitted vehicle and motorcycle segments, standing at 94 per cent and 82 per cent respectively. In the vehicle segment, local firms focused on low value-added cars and vehicles. We shall now take a closer look at the passenger car segment. Joint ventures clearly had the lion’s share. Among the top ten OEMs, eight were joint ventures with 56 per cent market share in 2006 (see Table 20.2). Latecomers such as BMW, DaimlerChrysler, Fiat and Ford, tried to expand market share by introducing their latest models and/or attractive prices. Competition was extended to include vehicle financing services. The strong presence of global OEMs was not only due to their competitive advantages of product, technology and management. Thanks to alliances with large Chinese automobile groups, they also benefited from considerable government support. The presence of Chinese OEMs is no longer negligible in the passenger car segment. Among fifty-one OEMs, thirty-five are Chinese carmakers with 38.5
1,224,000 1,165,700 932,300 708,700 685,100 352,300 302,500 266,800 210,200 204,400 6,052 000 1,164,000 7,216,000
SAIC FAW DFM Chang’an Beijing Auto Guangzhou Auto Chery Hafei Brilliance Geely
Subtotal Rest Total
83.9 16.1 100.0
17.0 16.2 12.9 9.8 9.5 4.9 4.2 3.7 2.9 2.8 Subtotal Rest Total
Shanghai GM* FAW VW* Shanghai VW* Chery+ Beijing Hyundai* Guangzhou Honda* FAW Toyota* Geely+ DFM Citroën* DFM Nissan*
By carmaker
Market share (%)
By automobile group
Volume of sales (units)
Passenger cars
All types of vehicles
Ranking of automobile sales in China by automobile firm and type of vehicle, 2006
Notes: * Sino–foreign joint venture. + Independent Chinese carmaker without foreign capital participation. Source: China Association of Automobile Manufacturers.
1 2 3 4 5 6 7 8 9 10
Ranking
Table 20.2
2,590,600 1,238,285 3,828,885
356,400 341,200 340,600 272,400 261,800 224,300 210,400 204,300 201,300 198,900
Volume of sales (units)
67.7 32.3 100.0
9.3 8.9 8.9 7.1 6.8 5.9 5.5 5.3 5.3 5.2
Market share (%)
388
Made in China 389 Table 20.3 Volume of production of major Chinese OEMs producing passenger cars, 2005 Volume of production Small economic car 1 Chery 2 Geely 3 FAW Tianjing 4 Chang’an 5 Changhe 6 BYD
185,588 148,182 192,964 168,242 45,039 11,236
Nature of ownership SOE Private, listed company SOE SOE SOE Private, listed company
Sedan 7 FAW Haima 8 FAW group 9 Brilliance
73,086 58,817 22,560
SOE SOE Listed company
SUV & MPV & minivan 10 Hafei 11 Dongnan 12 Nanjing Yuejin 13 Jianghuai 14 Changfeng 15 Jiangling 16 Ji’ao (Gonow)
49,893 45,167 35,832 31,397 24,763 17,610 10,332
SOE SOE SOE SOE Private SOE Private
Subtotal of 16 OEMs Total 35 Chinese OEMs Total 51 OEMs in China
1,120,780 1,138,381 2,958,400
Market share: 37.9% Market share: 38.5% Market share: 100%
Note: This table only integrates OEMs which produced more than 10,000 cars in 2005. Source: China Automotive Industry Yearbook (2006: 541–4).
per cent of market share (see Table 20.3). Most of them started with limited numbers of car models, specialising in economic cars, sedans or recreation vehicles. These cars provide a better match with the expectations of local consumers in terms of price, styling, interior design and equipment. With their ambition of building own-brand cars, Chinese OEMs are mostly independently owned. Through collaboration and technical contracts with overseas engineering and/or design companies, together with the strong support of regional governments, Chinese carmakers now produce for upgraded car categories. From the production of foreign brands to the creation of own-brand cars, the expansion of three major SOEs is largely thanks to central government support. The top three groups have taken different development paths. FAW’s own-brand cars are based on the imitation of, and technology transfer from, foreign models. For example, the ‘Red Flag’ was first an imitation of the Audi 100 and then a copy of Toyota’s Majesta. All models of FAW’s affiliate Haima
390 The Second Automobile Revolution
are copies of Mazda cars. SAIC is producing the Roewe model having acquired the intellectual property rights of the Rover 75 in 2006 at a cost of 3.68 billion yuan ($460 million). DFM is expected to finalise its own-brand car through a joint design contract drawn up between its own technology centre and an Italian company. Market launch of the new car is planned for late 2008. All three automobile groups are still at an early stage of commercial success. In contrast to the strong support enjoyed by large firms, central government has imposed huge institutional entry barriers on SMEs with the aim of rationalising a highly fragmented industry. They hoped to achieve this through the selection of champions among existing firms and by limiting the creation of new firms. The receipt of a licence to produce passenger cars became a rare occurrence. Geely and Lifan, two motorcycle giants, were exceptional cases. What is more, they were compelled to resort to unorthodox practices in order to overcome the barriers. The president of each group had to approach various authorities and lobby for licences. Other Chinese companies entering the industry have employed alternative methods. One approach involves the acquisition of existing small carmakers with a poor performance record. BYD Autos was a Hong Kong listed company producing batteries for mobile phones. In 2003, the company invested 270 million yuan to acquire 77 per cent equity of Qinchuan, a carmaker located in Xi’an Province. A second approach was to focus on less regulated segments such as recreation vehicles. The relatively easy access of Chinese firms to the automobile industry is also explained by Mitsubishi’s business activity in China. For various reasons, Mitsubishi began selling locally produced engines. The acquisition of key components through the market mechanism considerably decreases technology barriers for Chinese carmakers. Currently, more than twenty Chinese firms are buying engines from Mitsubishi. It is astonishing to note that small and medium OEMs are now taking a leading role in the production of Chinese cars, despite their disadvantages of limited technology, management and financial power (Wang, 2005). Geely, the first private OEM in China, has been chosen as the subject of an indepth investigation in order to explain the dynamics of Chinese SMEs in the industry sector.
Geely: from imitation to innovation Against all odds, Geely surpassed the barriers and started car production in 1998. In less than ten years, Geely has become the second largest carmaker in the economic segment (below 1.5L), the tenth passenger car manufacturer in China and number thirty-two in the world, with production totalling 205,000 units in 2006. At the end of 2007, Geely’s total annual production capacity reached 200,000 passenger cars, 200,000 engines and 200,000 transmissions based on single shift production.
Made in China 391
Roadster Coupê (60 K–80 K RMB, or 7.5–10K USD)
Beauty Leopard (Mybo) (1.0, 1.3)
B pricing range (70K–90 K RMB, or 8.7–11.3 K USD)
Vision (FC-1) (1.8, CVVT) King Kong (LG)
A pricing range (50K–70 K RMB, or 6.3–8.8K USD) A0 pricing range (30K–50 K RMB, or 3.8–6.3K USD)
Maple (1.3, 1.5, 1.8)
Merrie (MR) Ulion Merrie (MR 203) (1.0, 1.3) (1.0, 1.3) (1.0, 1.3) Haoqing (HQ) Haoqing (HQ Facelift) (1.0) (1.0) 1998
Figure 20.1
Free Cruiser (CK) (1.5, 1.6, 1.8) (1.3, 1.5)
2000
2002
2003
2005
2006
Product line-up of Geely, 1998–2006
Source: Geely website: www.geely.com
The Geely Group has set up four manufacturing sites: three in Zhejiang Province (Linhai, Luqiao and Ningbo City) and one in Shanghai. A further four new assembling plants in China will be operational by 2010 and the group envisages the creation of six overseas assembling sites in the near future. The company aims to reach a total manufacturing capacity of one million cars in 2010 and two million in 2015. In terms of internationalisation, the cumulated exportation before 2006 was 30,000 units. The strategic plan designates a large proportion of sales to overseas markets by 2015. Both Europe and the USA will be targeted in addition to current markets in developing countries such as the Middle East and North Africa. The Group has created and built eight models: Haoqing, Meiri, Ulio, Beauty Leopard, Maple, Free Cruiser, King Kong and Vision. These products extend from low-end economic cars (1.1L) to the sedan (1.8L) and include a roadster (see Figure 20.1). By 2010, the Group plans to offer fifteen models, eight series of engines, six series of manual transmissions, three series of automatic transmissions, one hybrid fuel project and one racing car project. From now on Geely aims to have the capacity to build three to four new models a year, including SUV, MPV and hybrid power cars. How has Geely managed to orchestrate such a phenomenal growth in just ten years? How is the Chinese car produced? The Geely case serves to illustrate the development path of a Chinese firm from the imitation of locally produced foreign models to innovation through international cooperation. The imitation stage Initially, Li Shufu, the founder of Geely, believed a car was simply a combination of capital, technology and human resources. The technology was sufficiently mature to be acquired either via market transactions or through the recruitment of engineers. With optimism running high, Geely’s car project was launched in the Linghai county of Zhejiang Province in
392 The Second Automobile Revolution
1997. The first prototype was based on the combination of two vehicles: a Mercedes-Benz and a Red Flag. The prototype body was constructed of glass fibre-reinforced plastic. Quality and security tests proved that the combination of several models was not viable for mass production. After the development of two additional prototypes (Model Anchi and 6410), Li reformulated his strategy and focused on the imitation of a single popular vehicle. This approach reduced the technology barrier for Geely. The first vehicle destined for volume production was the Haoqing, the fourth prototype car based on the Charade model of FAW’s Xiali (Model TJ7300). Xiali was the result of technology transfer from the Daihatsu Charade of the Toyota Group. Original components, copied Charade components and components from other vehicles were combined to produce the Haoqing. The body design was very similar to that of the Charade. Many suppliers of the Charade were contacted for the purchase of components. At the beginning, around 60 per cent of the components (including engine and transmission) were purchased directly from the supplier of FAW Xiali. Copied components of the Charade represented 10 per cent of Haoqing’s components. In total, around 70 per cent of the components were interchangeable with that of the Charade model. This type of product architecture can be described as quasiopen architecture because ‘imitation-turned-versatile parts are being gathered and assembled by numerous companies and this is different from a fullyfledged open architecture based on a carefully worked-out plan as seen in various digital products made by American companies’ (Fujimoto, 2002: 35). Other Geely models could also be characterised by their quasi-open modular architecture. The Maple 203 and 303 produced in Geely’s Shanghai plant were based on the imitation of Citroën ZX, the French car assembled in the DFM Citroën joint venture. Most of the imitation and remodelling was performed on the body and chassis. The majority of the components came from DFM’s suppliers. Unlike the Haoqing model, Maple cars were equipped with Geely engines and chassis. Geely’s engine was derived from a Toyota model and the architectural structure of the Maple was a combination of two foreign models: the Charade and the Citroën ZX. In this way Maple cars are more open-modular oriented than the Haoqing. Combining components from different integral models was a challenge for Geely. ‘During the design period, enough space has been reserved for an engine. We’ve tried to assemble different engines on the same car, and all of them work without the need to change the rest of the car’s architecture’, said vice general manager during a face-to-face interview conducted by the author in March 2006 at Geely’s Ningbo plant. Initially, the car bodies of Geely’s models Haoqing, Meiri and Huapu were produced by platework labour. Only at a later stage were blueprints completed and mathematical models built. Geely managed to produce imitated products at a lower price. For example, the 1.0 Haoqing hatchback sold at 37,999 yuan (US$4,588), whereas
Made in China 393
the Charade of FAW Xiali, its direct competitor, was priced at 40,000 yuan (US$4,830). Foreign brands of the same category sold at 100,000 yuan (US$12,075). The pricing strategy corresponded to the low-end segment where the importance of price outweighs that of quality or brand image. Innovation through international cooperation Between 2002 and 2007, Geely’s production underwent a rapid transformation as it evolved from craftsmanship to mainstream mass production. A second transformation is now underway as Geely moves from imitation towards innovation. Geely has invested heavily to ensure this transition and the successful implementation of innovation. Overall, R&D investment represents 6 per cent of annual sales. The Geely automobile research institute was established at a cost of 350 million yuan (US$44 million). To become a fully-fledged carmaker in the domestic market, three key measures were implemented: the construction of technology infrastructure, the internalisation of key components and international cooperation for the production of new cars. The construction of technology infrastructure was initiated in 2002. It consisted of a product development system, a technology management system (TMS), and the establishment of a product test and approval system. This ensured platform-oriented (NPD) and technology life-cycle management. Following the implementation of these three systems, Geely entered a learning phase in order to master the art of platform management. A platform strategy was developed both for existing cars and new models. Car models launched during the imitation stage were modified and upgraded. For example, new versions of the Huapu 303 and the Beauty Leopard were developed in 2003 having gone through the entire product development process. Geely gradually internalised the production of key components such as the engine and transmission system. For historical reasons, Tianjin Toyota Automotive Engine Co. Ltd. (TTAE), a 50–50 joint venture between Toyota and Tianjin Automotive Industrial (Group) Co. Ltd., had been the supplier of Geely’s engines but the procurement contract was terminated in 2003 when Toyota sensed fierce competition from Geely. The company had anticipated this type of hold-up situation due to the asset specificity of the engine. From 1998 onwards, Geely started designing its own engine through reverse engineering of the Toyota model. Key parts, such as the cylinder block and cylinder heads, were produced in-house and the rest (around 70 per cent) were purchased. Some of the suppliers of engine parts were also Toyota’s suppliers in China (Lee et al., 2002). One of the principal benefits of internalisation is a significant reduction in cost. Geely’s model (MR4790Q) can now be produced at one-third of the cost of the Toyota engine. International cooperation boosted Geely’s performance and the company entered a new stage of development. Early on, Geely simply purchased
394 The Second Automobile Revolution
foreign turn-key projects. The design contract with Maggiora SPA Auto Projects Group of Italy in 2002 marked its first international technology collaboration. The objective was to develop a complete car design of a new mid-sized model to be produced in 2005 at its Shanghai assembling plant. Geely had a dual objective: to acquire valuable knowledge concerning the entire process of car development and to become familiar with the European automobile industrial system. Geely engineers were sent to Italy to participate in a joint development project. When necessary, Italian experts visited China. A similar cooperative agreement was signed with German-based Ruecker AG in 2003 to develop an economic car. Geely rapidly became more involved in design and manufacturing through a learning by doing process. On 19 December 2002, a series of contracts were signed with Korea’s four main automotive designing and manufacturing companies led by Daewoo International Co. Ltd. The parties agreed to collaborate in an R&D and manufacturing project. Geely’s Korean partners were responsible for the following: (1) the pressing moulds and testing tools; (2) the welding line and related components design; and (3) simultaneous engineering. Geely aimed to learn and master the design process, and gain greater understanding of the detailed design. Each Korean expert worked with two Chinese engineers or technicians at Geely’s Ningbo assembling plant. In September 2004, the Free Cruiser car (internal code: CK-1) was on the market. Since 2006, several international cooperation agreements have helped Geely expand its market from China to other countries. A joint venture with Manganese Bronze Holdings plc of the United Kingdom, and a licence agreement with Antonov of Holland represent a big step forward as Geely prepares the production of upper-range models tailored to Western markets where quality issues pose a real challenge for Chinese manufacturers. The construction of new manufacturing plants dedicated to models for export is underway. In parallel, the group has begun to set up offshore SKD/CKD assembling plants with foreign partners. Product certifications have been requested in the United States and Europe. In 2006, Geely was one of the first companies to be included in the ‘State Designated Automobile and Auto Parts Export Base Enterprises’ listing drawn up by China’s Ministry of Commerce – a tribute to the company’s performance and leading role in exportation. There are at least two significant advantages of international cooperation: a shorter learning process and the speeding up of NPD. Geely identified its own needs and then worked with different foreign partners with whom it established specific and limited contractual relationships. Overall, new car design came from Europe, and moulds and manufacturing line upgrades were supported by Asian (mainly Taiwanese and Korean) companies. Geely’s research institute was actively involved in a learning by doing process throughout the period of international cooperation (see Table 20.4).
New car design (CI-1 model)
Construction of assembling line (including moulds) for CK-1 (Free Cruiser)
New car design (economic car)
Production of new taxi & highend limousine (3.0 and 3.5L)
Six-speed automatic transmission (TX6) for own use, but also for other OEMs around the world
Turn-key project, with low level participation by Geely
Turn-key project, with greater participation by Geely
Turn-key project, with greater participation by Geely
Equity Joint Venture & Technology Transfer agreement
Production licensing agreement
Source: Geely Annual Reports 2003, 2004, 2005 and 2006.
Objective of cooperation
Antonov of Holland
Manganese Bronze Holdings PLC of the United Kingdom
Ruecker AG of Germany
Daewoo International Co. Ltd and its Korean consortium
Maggiora S.P.A of Italy
International partner
A summary of Geely’s international cooperation
Nature of cooperation
Table 20.4
December 2006
October 2006
June 2003
December 2002
December 2002
Date of signing contract
N.A.
Registered capital: 440 million yuan (55 m US$)
N.A.
30 million US$
30–40 million US$
Value of contract
Payment of licence fees and production royalties
Geely group and its affiliate Huapu controls 52% of equity share, with 48% for Manganese Bronze Holdings plc
N.A.
One Korean expert guides two Chinese engineers at Geely’s plant
Development by Geely On-site consulting by Italian experts at Geely
Significant details
395
396 The Second Automobile Revolution
The Free Cruiser, King Kong and Vision are three cars representative of collaborative innovation. Their development was based at different assembling plants. Members of the Geely automobile research institute worked with engineers at assembling plants throughout the matrix organisation. The Free Cruiser was the first project to be carried through the whole NPD process. It took Geely three years to complete the project, from design to assembly line rollout. Simultaneous engineering, CAD and technology management systems were implemented. The entire car was digitalised. At the time, long-term strategic partnerships at business unit level between Geely and its foreign partners were new to the firm. Geely deliberately avoided major equity control by global OEMs in order to preserve the company’s independence. The dramatic changes seen at Geely were also closely linked to the acquisition of high-quality human capital. Chinese and foreign senior engineers with overseas work experience at DaimlerChrysler and Daewoo were appointed as top managers. Domestic specialists of engine and transmission systems were recruited to develop in-house models. For example, Xu Bingkuan became top manager at Geely in late 2002. He was formerly chief engineer of a national automatic transmission project which did not reach the final stage despite a huge investment of 800 million yuan (US$100 million). Geely invested 50 million yuan (US$6.25 million) to finalise the project. The first prototype was produced and tested in April 2004. The transition from prototype to volume production was a long process of trial and error over a three-year period (2003–6). About 110 cars were tested with an experimental transmission system of which more than 50 cars were driven more than 200,000 kilometres. Three hundred transmissions were damaged during testing. Due to the high risk involved, none of the state-owned suppliers cooperated with Geely. More than 130 transmission components were produced by over forty small and medium-sized suppliers. Volume production of the automatic transmission system was eventually launched with success. Geely achieved an annual production capacity of 20,000 units. The manufacturing line was designed in-house with a total investment of only 80 million yuan (US$10 million) – much less than that required for an imported manufacturing line estimated at 200 million yuan (US$25 million). The localisation of the production line considerably reduced costs. The unit price was about 6,000 yuan (US$750) compared to 10,000 yuan (US$1,250) for an imported model. Geely thus became the first Chinese carmaker to produce its own automatic transmission system.
Geely’s productive model Geely is now emerging from its ten-year childhood period (1998–2007) as a maturing company and established car manufacturer. However, it is perhaps
Made in China 397
too early to characterise Geely’s productive model which is still at an early stage of development. Nonetheless, we shall attempt to apply GERPISA’s analytical framework (Boyer and Freyssenet, 2002; Boyer et al., 1998; Freyssenet et al., 1998) to highlight key features of the productive model developed by Geely. The model covers aspects including product strategy, production method, labour conditions and buyer–supplier relationships. Definition of the Geely model is based on a literature review, an in-depth, longitudinal case history of Geely. Geely’s productive model is based on a low-cost and low-price strategy, with every possible effort made to keep costs down. The objective is not to increase the unit profit margin, but to become competitive enough in terms of price vis-à-vis foreign brand cars. Generally speaking, Chinese cars are at least 30 per cent cheaper than foreign brands of same category. The application of a series of measures explains the successful implementation of this corporate strategy. Firstly, Chinese carmakers adopt various approaches to achieve low cost during NPD. At the imitation stage, a radical measure – the development of quasi-open product architecture – significantly reduces costs. When moving away from imitation and towards innovation, Geely establishes limited contractual relationships in order to acquire the technologies and know-how required, such as car design, mould production and engineering consulting, rather than requesting the integral solution from a single foreign OEM partner. This approach not only assures Geely’s independence but is also cost-saving. Secondly, Geely’s production organisation is competitive and low cost. Salaries in the Chinese (and Indian) automobile sector are estimated to be the lowest in the industry worldwide (Perkowski, 2004). Geely does not intend to squeeze labour costs any further. Chinese OEM believes that cost reduction is mainly derived from localising manufacturing equipment. Thanks to the existence of a comprehensive industrial infrastructure in China, a higher proportion of equipment can be designed and produced by Chinese firms. This avoids heavy sunk costs. It is reported that Geely first became profitable when its annual production exceeded 20,000 units at plant level (Lu and Feng, 2004; 53). Thirdly, the nature of buyer–supplier relationships also favours tight cost control. Suppliers have played an active role in repetitive imitation and remodelling of components. These companies are run by grassroots entrepreneurs working diligently to satisfy the demanding needs of OEMs. They are prepared to provide low-cost components to the detriment of their own profit margins, labour working conditions and environmental protection. They also accept double sourcing, one-year open contracts, bear mould costs, and produce large quantities in short lead-times. The competence of Chinese private-owned suppliers has even become a source of inspiration for world tier-one suppliers. The latter are learning from their example as they
398 The Second Automobile Revolution
gain insight into how Chinese suppliers are able to achieve unbeatable prices whilst maintaining quality standards. In the years to come, more and more Chinese tier-two and tier-three suppliers will be integrated into the global supply chain. Further details concerning Geely’s productive model are outlined in Table 20.5 which summarises their ‘low-cost, low-price for the masses’ strategy. It is important to highlight Geely’s governance compromise which has guaranteed the successful implementation of this profit strategy. As founder of a private company, Li Shufu had a long-term vision from the start of turning Geely into a global player in the automobile industry. Profits from other business units were injected into car production to compensate for short-term losses. The management team has changed. At the early stage of Geely’s development managers were largely made up of employees nearing retirement age, a state of affairs explained by the perception of high risk and uncertainty pertaining to the fledgling company. Today, executive management is dominated by younger, upwardly mobile professionals. Newly recruited managers found it particularly rewarding to be involved in the production of own-brand cars; they were recognised and valued for their work. This contributed to the low staff turnover rate. Strong employee relations have also favoured fast growth of the company. Salary levels are above average for inland China. Cost reduction and innovation by workers have been encouraged through personal acknowledgement and financial rewards. Various paternalistic measures such as low-cost housing and child-care facilities have further reinforced staff commitment. For the past two years, on-site training has been bolstered in the aim of improving product quality and retaining qualified workers. The above measures have counterbalanced (and compensated for) imposed heavy work loads and kept the staff turnover rate at an acceptable level. Early on, Geely’s supplier network was an intricate and extended family of friends and relations. Suppliers played a major part in helping Geely through its infant years. They often made great sacrifices including agreeing on extremely low prices or delayed payment. Thus Geely was able to evolve, slowly but surely, adopting a long-term approach to its inter-firm cooperation where quality issues became more important. Last but not least, the strong support of local government during the early stage of development followed by preferential policies from central government later on have contributed to Geely’s success.
Conclusion China has become the most competitive automobile market in the world thanks to a large number of global and local OEMs. Global OEMs are still playing a dominant role. At the same time, the dynamics of emerging Chinese OEMs are clearly manifest. Geely and Chery, together with other local OEMs,
Price competitive low-end cars for domestic market
Only low-end cars Low price but inferior quality Key components are purchased
Car design: imitation of best selling foreign cars in China Car body: craftsmanship, made by platework labourers Key components: imitation through reverse engineering Manufacturing line: not built
Labour-intensive work, significant overtime compensated by higher salary, and better welfare Single-skill workers
Only people at retirement age can be recruited
Product strategy
Production method
Employment relations
Human resource management
Stage 1 (1998–2001)
Recruitment of professionals. Standardisation of HR management process In-house training for managers
Reinforcement of worker training Key positions at assembling plants operated by workers with a 3-year college degree Career development programme for workers
Car design: by European companies Car body: mould designed by Taiwanese company Key components: designed by its own R&D centre Manufacturing line: in-house design, local manufacturing equipment, key machines imported
Wider product range (from 1.0–1.8L) Low price, better quality Own engine and transmission system New cars targeting overseas markets
Price competitive full-range car for both domestic and overseas markets
Stage 2 (2002–2007)
Geely’s productive model: ‘The low-cost, low-price car’
Corporate strategy
Table 20.5
(Continued)
+: Standardisation methods −: How to increase overall quality of the management team
+: Greater flexibility than foreign OEMs −: Conflict between use of single-skill workers and production of sophisticated cars
+: Effective control of costs through ‘quasi-open modular product architecture’ −: Low cost to the detriment of car performance
+: Higher profit margin in upper range cars −: Expanding (too) fast, but limited capability and resources
+: Possible to become a global competitor −: Challenge the status quo and the synonym ‘made in China’ = ‘inferior quality’
Future perspective
399
Profile of suppliers: most of them are SMEs in close proximity, producing low-cost copied components Supplier selection: double sourcing, priority on low price Length and nature of contract: one year open contract, price is fixed but not quantity Purchasing system: decentralised, based on car models
Not yet implemented
Chinese market: Low-end segment: neglected by foreign OEMs. Consumers are price sensitive, display a lack of car culture and no brand loyalty Sedan and upper-range segment: concentration on foreign brand cars. Consumers have greater brand awareness.
Overseas market strategy
Market conditions
Stage 1 (1998–2001)
Continued
Buyer– supplier relationships
Table 20.5
Chinese market: Low-end segment: increasing presence of foreign cars, less price competitiveness. Consumers become more conscious of car appearance and quality Sedan and upper-range segment: greater competition with the arrival of Chinese OEMs. Emergence of price sensitive Chinese SMEs
Export to developing countries Assembly CKD in Indonesia and Ukraine On-going NPD for Western markets
Profile of suppliers: suppliers for large Chinese OEMs account for 50%, increasing proportion of foreign suppliers. Better quality Supplier selection: the same as before, increasing concern for quality Length and nature of contract: the same as before Purchasing system: to be centralised from late 2007. Plan to cut number of suppliers by half.
Stage 2 (2002–2007)
Chinese market: +: Greater understanding of Chinese consumer needs Advantages to operate in domestic market −: The most competitive market, huge number of competitors
+: Higher profit margin in foreign countries −: Great shortage of talent for overseas operations
+: One of the main suppliers of low-cost components −: Rationalisation of supplier networks to meet quality standards and rapid expansion of car model range for both domestic and overseas markets
Future perspective
400
Foreign market: In developed countries: The same as previous stage. The substitution of second-hand cars with new cheap cars (such as Renault’s Logan). In developing countries: The same as previous stage. With the exception of the marginal wealthy classes, mass consumers buy motorcycles and three-wheelers but also desire cheap four-wheel cars
Source: Company visit in March 2006, March and May 2007. Company website: www.geely.com
Foreign market: In developed countries: Domination of OEMs from Triads (Europe, USA, and Japan) but some leading OEMs suffer huge operational losses. Mature consumers. In developing countries: Domination of transplant of OEMs from triads. Lack of local players. Consumers lack purchasing power
Foreign market: In developed countries: +: Price advantage −: Severe regulations and standards. Consumers with rich car culture. In foreign developing markets: +: Price advantage, no local competitors, sizeable market potential −: Governmental regulations for market entry. Competitors also come from China. Risk of new wave of price competition
401
402 The Second Automobile Revolution
are expected to have 40 per cent of the domestic market by 2010, equivalent to around 2.5 million units. When foreign OEMs serve manufacturing sites in China for local markets, Chinese carmakers utilise their production plants for both domestic sales and exportation. Low-price cars have huge market potential due to increasing income disparity both in developing and developed countries. The main objective of this chapter is to draw attention to the emerging local OEMs and their mission to produce own-brand cars. Geely is a case in point. During the imitation stage, bestselling foreign models are copied and remodelled through reverse engineering. Previous closed integral cars are transformed into quasi-open modular types. Innovation in product architecture contributes to the realisation of a low-cost, low-price strategy as Geely’s key success factors illustrate. During this stage, the quality and performance of low-end cars are mediocre. Not content to remain a niche player, Geely has a driving ambition to become a full-range carmaker at international level. In many ways, its productive model is representative of the quintessential Chinese productive model. However, Chinese OEMs still have a long way to go. Aggressive pricing alone is not enough to compete successfully on the market. Manufacturers also need to build a brand image and distribution channels, offer high value-added services such as automobile financing, respect rigorous regulations and meet the quality standards of both national and international markets. It took Japanese and Korean OEMs several decades to establish a foothold in the global marketplace. Even today, the market share of Korean carmakers in European and North American markets is low. There is no sign that Chinese carmakers can further reduce the time required to catch up with their competitors in the development race. Geographic expansion will bring new challenges in manufacturing organisation, supply chain management and human resource management. Chinese OEMs will have to be responsive to change, with flexible management structures designed to adapt to constantly evolving internal and external factors. The highly responsive and adaptable productive model of Japanese and Korean OEMs (Christina and James, 2001; Pardi, 2003; Whitley et al., 2003; Winfield and Kerrin, 1996) is of high value for the Chinese and a benchmark for the future development of the Chinese automobile industry. Bibliography Accenture (2005) China: an Automotive Industry on the Verge. http://www.accenture. com/, accessed 20 July 2007. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Boyer, R., Charron, E., Jürgens, U. and Tolliday, S. (eds) (1998) Between Imitation and Innovation: the Transfer and Hybridization of Productive Models in the International Automobile Industry. Oxford: Oxford University Press.
Made in China 403 CAIY, China Automotive Industry Yearbook. Tianjin: China Automotive Technology and Research Centre (CATARC), various years. CATARC (1997) Qiche Gongye Guihua Cankao Ziliao [Automotive Industry Planning References 1997]. Tianjin: China Automotive Technology Research Centre. Christina, L. A. and James, R. L. (2001) ‘Keiretsu, Governance, and Learning: Case Studies in Change from the Japanese Automotive Industry’, Organisation Science, 12(6): 683–701. Freyssenet, M., Mair, A., Shimizu, K. and Volpato, G. (eds) (1998) One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford: Oxford University Press. Fujimoto, T. (2002) ‘Thinking about China’s Manufacturing Industry from the Perspective of Product Architecture’, Keizai Sangyo Janaru [Journal of Economies and Industries], June: 34–7 (in Japanese). English version at http://www. rieti.go.jp/en/papers/research-review/002.html, accessed 20 Feburary 2007. HOCAI (1996) Zhongguo Qieche Gongyehi [History of China’s Automotive Industry 1901–1990]. Beijing: Renmin Jiaotong Chubanshe (People’s Transportation Press). IBM Consulting Services (2005) ‘Inside China: the Chinese View Their Automotive Future’, 30 November, www.ibm.com/industries/automotive/doc/content/ bin/ge510-6229-inside-china.pdf, accessed 30 June 2007. KPMG (2004) ‘China’s Automotive and Components Market’, KPMG, http://www. kpmg.com.cn/redirect.asp?id=6051, accessed 10 December 2006. Lee, C. L., Fujimoto, T. and Chen, J. (2002) ‘Trends of Open Product Architecture and Internationalization of Private Companies in the Chinese Automobile Industry’, Discussion Paper for International Motor Vehicle Programme (IMVP), MIT, USA, August. Lu, F. and Feng, K. D. (2004) ‘Fazhan woguo ziwhu zhishi chanquan qiche gongye de zhengce xuanze’ [The Policy Choice for Developing the Chinese Automobile Industry with Own Property Rights], Research Institute of Firm and Government, Government Management College, Pekin University. Pardi, T. (2003) ‘Marges d’action et contraintes: Génèse et dynamique d’un compromis de gouvernement d’entreprise dans un transplant Toyota en Grande-Bretagne’, 11th GERPISA International Colloquium, ‘Company Actors on the Lookout for New Compromises: Developing GERPISA’s New Analytical Schema’, 11–13 June. Paris: Ministère de la Recherche. Perkowski, J. (2004) ‘China’s Auto Industry: Localisation, Globalisation and Management’, Deutsche Bank China Expert Series, July, www.asimco.com/news/jul.15.2004. ppt, accessed 20 August 2007. Wang, H. (2004) ‘Why Are There Hundreds of Carmakers in China? Essay on the Analysis of Entry Barriers’, International Conference on Industrial Organisation, Law and Economics, Porto Carras, Chalkidiki, Greece, 17–20 June. Wang, H. (2005) ‘When Will the Chinese Brand Car be Made? The Comparison of Development Strategy between Large and Small Carmakers’, 13th GERPISA International Colloquium, ‘Productive Organisations, Employment Relationships, Financialisation: Specificities of the Automotive Industry’, 16–17 June, Paris: Ministère de la Recherche. Whitley, R., Morgan, G., Kelly, W. and Sharpe, D. (2003) ‘The Changing Japanese Multinational: Application, Adaptation and Learning in Car Manufacturing and Financial Services’, Journal of Management Studies, 40(3): 643–72. Winfield, I. and Kerrin, M. (1996) ‘Toyota Motor Manufacturing in Europe: Lessons for Management Development’, Journal of Management Development, 15(4): 49–56.
21
Maruti-Suzuki’s Trajectory: From a National Champion to a Japanese-owned Subsidiary Florian Becker-Ritterspach
Maruti-Suzuki is the market leader in India’s passenger car market and has remained in this position despite losing ground to the growing competition. Understanding Maruti-Suzuki’s trajectory is interesting not only because it is one of the most successful players in one of the fastest growing emerging markets, but also because the case epitomises the transformation the Indian economy and society has undergone in the past decades. While MarutiSuzuki’s trajectory is strongly shaped by the deregulation and liberalisation of the Indian socio-economic context, this influence is not unidirectional. Being a national champion in the first decades of its existence, Maruti-Suzuki has become a driver of liberalisation and reform as a range of government polices are tailored to the company’s needs. Clearly, like no other automobile company, Maruti-Suzuki has changed and shaped the development of the Indian automobile industry. Drawing broadly on Boyer and Freyssenet’s (2003) work and definition of production models and their contextual embeddedness, this chapter focuses on the transformation of Maruti-Suzuki’s production model between 1992 and 2007. This discussion highlights the close conjunction between changes in the company’s production model and its context. The chapter is structured as follows. In the next section we briefly discuss Maruti-Suzuki’s trajectory until the early 1990s. The following sections discuss Maruti-Suzuki’s trajectory from the early to mid-1990s until 2001, and the period from 2002 until 2007. The chapter ends with a conclusion on Maruti’s-Suzuki’s past trajectory and presents a brief outlook on key contextual conditions structuring the company’s future trajectory.
Maruti-Suzuki’s origin and emerging production model In the early 1970s Sanjay Gandhi (the son of Prime Minister Indira Gandhi) founded the two private limited companies Maruti Technical Services Private Limited and Maruti Limited in Gurgaon (State of Haryana). The stated mission of the enterprises was the development of a ‘people’s car’ – an affordable, cost-effective, low maintenance and fuel-efficient car – designed and 404
Maruti-Suzuki’s Trajectory 405
produced for India’s middle class. However, despite government backing and support, the young company proved incapable of producing a single marketable car. Unable to establish a viable production model, Maruti Limited faced liquidation in 1977. In October 1980 the Government of India took over Maruti Limited and incorporated it in February 1981 by an act of parliament as a public limited company. Recognising the need for foreign collaboration to turn rechristened Maruti Udyog Limited into a success story, Maruti’s new management initiated the search for a suitable joint venture (JV) partner. After a difficult search period, the Indian government finally won Suzuki as a partner. On 2 October 1982, a JV and a ten-year licensing agreement were signed between Maruti and Suzuki. The JV agreement obliged Suzuki to take a 26 per cent stake in the company leaving 74 per cent to the Indian government (Mohanty et al., 1994; Shirali, 1984). Suzuki’s involvement in Maruti in 1982 marked the beginning of a tremendous market success for the JV. In less than a decade Maruti-Suzuki out-competed its two domestic rivals Hindustan Motors and Premier Automobiles. By 1990–1 Maruti-Suzuki held about 62 per cent market share, leaving 13.9 per cent to Hindustan Motors and 23.7 per cent to Premier Automobiles (Mohnot, 2001). As a government company, Maruti-Suzuki benefited in a particular way from a declining state-led growth regime and the first departures from it. On the demand side, the state-led regime created substantial social mobility (D’Costa, 2005) translating into rising demand for low-cost cars. At the same time, it was this demand that India’s planned economy left largely unsatisfied. On the supply side, the company benefited from preferential treatment (e.g. lower excise and customs duties for small cars) and the restricted competition ensured by a licensing system and protectionism. Yet, Maruti-Suzuki’s success cannot be understood without the additional condition of departures from the previous economic regime. On the demand side, this departure was signalled by the Indian government’s willingness to respond to middle-class consumer demands by no longer classifying passenger cars as luxury items. On the supply side, Maruti benefited from the relaxation of the licensing system and protectionism, which granted companies more flexibility in deciding their production capacity and product mix and selectively allowed international cooperation (D’Costa, 2005). These careful changes marked the Indian government’s acknowledgement that without a minimum of entrepreneurial freedom and international involvement neither Maruti-Suzuki nor the automobile industry as a whole could thrive. Moreover, there was recognition that the corporate governance conditions prevalent in Indian companies were unfit to modernise the industry and improve productivity (Venkataramani, 1990). Against this background, we can understand Maruti-Suzuki’s rise. While the company was shielded from international competition, it could outcompete its domestic rivals through preferential treatment and an international partner whose business strategy was specifically geared towards
406 The Second Automobile Revolution
low-cost small cars for developing economies. This strategy was also reflected in Maruti-Suzuki’s profit strategy and product policy. In the 1980s, its profit strategy rested on a volume strategy that mainly sought to keep cost under control through fast localisation of the production of three basic models. These were the Maruti 800 in the mini-segment (A), the Omni, a small multi-purpose vehicle (MPV), and the Gypsy, a sports utility vehicle (SUV).1 Targeting the lowest market segments, they served basic transportation needs and provided value for money. However, while the fit between the demand context, profit strategy and policy formed one side of the market success, the other side was based on matching forms of production organisation and labour relations. It was the Indian government’s hands-off approach, diverging from earlier modes of the government–public sector company nexus (D’Costa, 2005), which left operative decisions to experienced Indian managers and their Japanese counterparts and created the socio-political basis for a sustainable governance compromise in the first decade: specifically, by giving management the leeway to experiment with new forms of production organisations and employee relations. While the production organisation rested externally on tremendous efforts to develop suppliers and cooperative relations with them, it concentrated internally on the introduction of high involvement work practices unheard of in Indian industry. The latter built on a motivated workforce sustained by a mutually enforcing complex of company growth, the selection of a young inexperienced workforce, public sector job security, excellent pay for workers, training and new modes of employee relations. Above all these found expression in the adoption of an unaffiliated company union and culture-related efforts to keep hierarchical demarcations low in the company (Becker-Ritterspach, 2006).
Market expansion, new competition and the battle for control in the 1990s In the 1990s, Maruti-Suzuki’s market share and profitability saw a sharp rise and decline. Until 1997, Maruti-Suzuki strongly benefited from its first mover advantage. In fiscal year 1997–8 Maruti-Suzuki’s market share peaked with 83 per cent and so did its profits. However, the situation changed dramatically in the next years. Between 1999 and 2001 Maruti-Suzuki’s market share and profits slid. In 2001, Maruti-Suzuki posted its first loss in the company’s history. In 2000–1 the company’s market share was down to 57.6 per cent (see Appendix Table 21.1). The transformation of India’s state-led economic regime Maruti’s rise and decline in terms of market share and profitability mirrored a period of fundamental transformation, shifting the Indian economy from a state-led protectionist regime towards a more market-led and open trade
Maruti-Suzuki’s Trajectory 407
growth regime. Following the balance of payment crisis in the early 1990s, the Indian government launched stabilising measures and a long-term reform programme. In the passenger car industry the reforms arrived in 1993 with the abolishment of licences. The measure essentially removed major restrictions on market entry and capacity expansion. At the same time, import tariffs were reduced and foreign direct investments (FDI) in the automobile industry invited. However, like in other sectors, quantitative restrictions on imports stayed in place and FDI was kept under tight regulation in the automobile industry (Mohnot, 2001). Overall, the reforms contributed to the recovery and growth of the entire Indian economy between 1992 and 1997. The abolishment of the licensing system, the liberalisation of the trade and FDI regime, and the fiscal and monetary reforms (e.g. reductions of excise duty for cars) created the supply and demand side conditions for the Indian passenger car market to grow. Particularly in the first half of the 1990s Maruti-Suzuki expanded as it benefited from the reforms and the positive economic development. As an established player, the company was in an advantageous position with its high levels of local content and catering to entry level market segments. However, as the newcomers settled in and new market segments developed, the company felt the effects of a changing competitive environment. Competition particularly mounted as foreign players like Hyundai, Ford, Fiat, Honda and General Motors entered the market and domestic players like Hindustan Motors and Telco caught up or moved into the passenger car market. These effects built up in the second half of the 1990s and were aggravated by delayed model introductions. For, as market competition increased, MarutiSuzuki was distracted by an internal conflict between its JV partners. The conflict built up as Suzuki raised its equity in the company (to 40 per cent in 1989 and 50 per cent in 1992). In the mid-1990s, Suzuki made bids to take over the company. The Indian government, however, declined the offers and increased its direct influence in the company instead. While on the surface issues such as technology transfer and overstaffing, ways of financing a modernisation/expansion plan and the appointment of top company executives were the contentious issues (Radhika, 2002), the key underlying conflict was a ‘battle for control’ between the JV partners. As a successful government company marked by an Indian government foreign–private ownership structure, Maruti-Suzuki was at the epicentre of a socio-political conflict in India revolving around the privatisation of public sector enterprises and their sale to foreign investors. The close conjunction between India’s socio-political developments and the conflict between the company’s JV partners was also manifested in the sudden end to the conflict. Shortly after a new governing coalition came to power in 1998, reviving the stalled reforms and the privatisation policy, an out of court settlement was found for the contentious issue of managing director appointment. In the following year, the new government announced its decision to divest its stake
408 The Second Automobile Revolution
in the company, thereby opening the way for its privatisation and takeover by Suzuki. Volume growth, modest diversification and delayed model introductions Triggered by domestic market growth, Maruti-Suzuki continued to unfold its volume expansion strategy throughout most of the 1990s. However, in contrast to the 1980s the company started to diversify into higher passenger car segments. In the 1990 Maruti-Suzuki ventured, with the Maruti 1000, into the mid-size segment (C) (replaced in 1994 by the Esteem). Three years later, in 1993, the company offered with the Zen its first model in the compact segment (B). However, while most of Maruti-Suzuki’s models saw some facelifts and a few variant introductions, there were basically no new model introductions between 1993 and 1999. Many industry observers attribute this delay to the simmering JV conflict in this period (Muthukumar, 2004). Despite falling profits and market share, it was not until 1999, with the resolution of the JV conflict, that Maruti-Suzuki finally introduced a new range of models. Catering to the fast-growing compact segment (B), the Wagon R and the Alto were introduced in 1999 and 2000. Moreover, targeting the emerging mid-size segment (C) the Baleno was launched in 1999. To stem its sliding profits and market position, Maruti-Suzuki also started to cut prices. At the same time, the company tried to widen its customer base by introducing a wide range of services. In the early 2000s, Maruti-Suzuki entered into a number of automobile-related businesses to offer what it called ‘360 degree customer service’ (Mohapatra, 2003). Som (2004) described the shift as a change from ‘sell what we produce’ towards ‘marketing and customer focus’. With the change of the competitive environment at the end of the 1990s, Maruti-Suzuki’s profit strategy increasingly emphasised the reduction of costs. It was a profit strategy, however, that still built on expansion in the Indian market with a negligible role for exports (see Appendix Table 21.1). Capacity expansion and quality initiatives in the production organisation With its product offers in the lowest market segments, its established sales and service network and its high levels of production localisation Maruti was in an excellent position to capitalise on its first mover advantage and to further enhance its market share in the first half of the 1990s. This market success translated into a massive expansion of production capacity. With the opening of its second plant, Maruti-Suzuki reached an installed capacity of 200,000 units in 1995. In 1999, Maruti-Suzuki’s third plant became operational, enhancing the installed capacity to 350,000 vehicles per annum. With regard to the developments in the production and work organisation in the first half of the 1990s, Maruti-Suzuki placed a strong emphasis on training initiatives and quality improvements. Towards the end of the 1980s,
Maruti-Suzuki’s Trajectory 409
Maruti-Suzuki had identified a range of training gaps in its entire workforce. In response the company launched in the early 1990s an in-house training initiative that focused, besides cost reduction and leadership issues, massively on quality improvement (Chatterjee, 1990). This training was flanked by the introduction of new quality improvement practices such as total quality management and 5S (Okada, 1998). Moreover, in the 1990s, annually as many as 120 workers underwent training at Suzuki in Japan (Okada, 1998). As a result of training efforts and other measures, Maruti-Suzuki was able to almost quadruple its value-added per employee from Rs. 0.57 million in 1990–1 to Rs. 1.9 million in 1995–6 (Okada, 1998). However, while these improvements in quality and productivity terms, together with those initiated with suppliers were substantial, they proved insufficient to avert the company crisis. Falling market share and profitability at the end of the 1990s led to an ever growing emphasis on the introduction of cost reduction and productivity enhancing measures in the production and work organisation. Deteriorating employment relations and governance compromise Maruti-Suzuki’s increasing performance problems and looming takeover challenged the company’s established record of peaceful employment relations at the end of the 1990s. The first signs of deteriorating employment relations surfaced in 1998 with the Indian government’s privatisation policy and its divestment plan for Maruti-Suzuki. Opposing the takeover, MarutiSuzuki’s company union threatened a strike. It also threatened to call off the long-standing non-affiliation agreement. The takeover stirred fears about employment security and conditions that were seen to be safeguarded only by continued government involvement. These fears were compounded by the deteriorating company performance. In the difficult economic situation a management–union disagreement surfaced over the pensions and promotion policy and above all over proposed changes in the incentive scheme by the management. As both sides were unwilling to give in, Maruti-Suzuki saw its first strike in October 2000. However, the strike and the associated union demands proved unsuccessful. In January 2001 the strike came to an end with Maruti-Suzuki’s management essentially pushing through its position. The management not only unilaterally put through the new incentive scheme, but it also dismissed some workers and trainees and asked its workmen to sign ‘a good conduct undertaking’. Among other things the workers had to agree that they would not go on strike in future. What began in Maruti-Suzuki as a showcase of industrial peace reverted to a typical Indian industrial conflict. One sign of this shift was the formation of a second union during the labour conflict, challenging the company’s long-standing one-union policy. The period of labour unrest marked a rupture with the company’s previous governance compromise based on strong company growth, steady pay rises and secure employment conditions under the protective umbrella of Indian government ownership (Becker-Ritterspach, 2006).
410 The Second Automobile Revolution
Adapting to the new competitive scenario in the 2000s After a slump in 2000–1, Maruti-Suzuki’s profits have seen a recovery and the market share stabilised around 50 per cent in the early 2000s. In 2004–5 Maruti-Suzuki’s overall sales saw a rise and so did the company’s profits (see Appendix Table 21.1). The company also started to develop a stronger export orientation as part of Suzuki’s internationalisation strategy. Despite a shrinking market share, Maruti-Suzuki remains the leader in the Indian passenger car market in the 2000s, followed by Tata and Hyundai as second and third. The stabilisation of a market-led growth regime Maruti-Suzuki’s recovery and ongoing leadership in the 2000s reflect the company’s successful adaptation to a new socio-economic context in India. It is a context that is characterised by a stabilising shift from a state-led to a marketled growth regime, albeit with intermediate slowdowns. The stabilisation of the new economic regime is manifested in and secured by the Indian economy’s strong performance. After some fluctuation in the late 1990s and early 2000s, the Indian economy saw a steady and strong growth around 8 per cent from 2003 onward. On the demand side, the high growth rates and growing disposable incomes translated into a substantial sales growth. The Indian government continued to play a pivotal role in directly stimulating the automobile demand. Between 2001 and 2007 it further reduced income tax as well as customs and excise duties on automobiles. Low interest rates and easier financing additionally support consumer demand and growth. In the 2000s, India is one of the fastest-growing markets for passenger cars in the world. It is a market, however, in which the mini- and compact car segments combined account for more than 70 per cent of new car sales. Catering to highly price-sensitive customer segments in a highly price-sensitive market, Maruti-Suzuki benefits from this situation (ACMA, 2006; Nair, 2006). On the supply side, the stabilisation of the new growth regime has found expression in new government initiatives for the privatisation of public sector enterprises and the ongoing deregulation of the trade and FDI regime. For Maruti-Suzuki, these policies finally opened the way for the takeover by Suzuki. After a rights issue and an initial public offering in 2002 and 2003, Maruti-Suzuki finally became a subsidiary of Suzuki. In 2007, the Suzuki Motor Corporation had a 54.21 per cent stake in the company followed by institutional and other shareholders. However, the ongoing liberalisation is not only the precondition for Maruti-Suzuki’s changed ownership mode. At the same time, the ongoing deregulation of India’s trade and FDI regime in the 2000s renders the entry of new international competitors ever easier. With the entry of new players, particularly of Renault and VW, to the Indian market, Maruti-Suzuki has to prepare for a continuing intensification of competition.
Maruti-Suzuki’s Trajectory 411
Cost leadership, continued volume growth and diversification In the 2000s, Maruti-Suzuki’s profit strategy is still focusing on volume growth, but simultaneously emphasises massive cost reduction. While Maruti-Suzuki expects strong company growth, as part of the overall market growth, it is not counting on past levels of market share. The company re-emphasises ‘value for money’ as its ultimate selling point and assumes that the Indian passenger car market will remain an entry level market in the coming years. More than ever before, Maruti-Suzuki competes on price, which is also reflected in the product policy. On the one hand, the product policy strongly builds on cost leadership, expressed in series of price cuts for existing models or alternatively more features for the same price. On the other hand, four new base models have been introduced that show the company’s effort to upgrade into the executive (SX 4 in 2007) segment (D) and to widen its product range in the compact segment (B) (Swift in 2005), the MPV (Versa in 2001) and the UV (Grand Vitara in 2003) segment. These model and variant introductions are reflective of the company’s ‘focus on the small car segment, while offering products in most segments of the Indian passenger car market’ (Red Herring Prospectus, 2003) and at the same time ‘a policy of offering the cheapest cars in each vehicle segment in which it operates’ (The Economist Intelligence Unit Limited, 2006). At the end of 2007, out of its eleven base models, five cater to the small and compact car market. The most important development in terms of product policy, however, is the introduction of the new Swift, the Swift diesel (the company’s first diesel in the portfolio) and of other ‘World Strategic Models’ in the late 2000s. The introduction of the Grand Vitara, Swift, SX 4 and the planned launch of the Splash (mini-MPV) and A-Star (A segment) mark, on the one hand, an increasing shift towards adding more contemporary (world strategic) models to Maruti-Suzuki’s Indian product portfolio. On the other hand, it reflects Suzuki’s strategic reorientation, aiming at a global redistribution of R&D and production for different car segments. In particular, the involvement of Indian engineers in the Swift’s development and the allocation of its export production to India reflect Suzuki’s plan to develop India into the R&D and export hub for small cars (Mankad, 2006; Nair, 2006). Finally, in an effort to widen its customer base, Maruti-Suzuki continues to improve its service and expand its offerings in related businesses (Nair, 2006). Productivity enhancement and cost reduction in the production organisation Starting with Suzuki’s new directorship in August 1999 and more importantly Maruti-Suzuki’s takeover by Suzuki in 2002, the company has introduced a series of changes in the organisation of production. Suzuki proposed a road map and asked Maruti-Suzuki to present an action plan for enhancing productivity and reducing costs. In response and focusing ever more on cost
412 The Second Automobile Revolution
leadership, Maruti-Suzuki launched the ‘Challenge 50’ programme. The goals of the programme were improvements in safety, productivity, quality, cost and material management (Maruti Udyog Ltd., 2005). The defined goal was to improve productivity by 50 per cent (bringing down man hours per vehicle from twenty-four to twelve) and reduce cost by 30 per cent within three years (Muthukumar, 2004). The programme not only focused on Maruti-Suzuki’s internal organisation but also involved its key suppliers. In 2005, Maruti-Suzuki launched a follow-up programme called ‘Next Leap’. Like its predecessor, the programme mainly targets cost reduction and productivity increases. However, in contrast to its predecessor, the programme focuses more on the suppliers. Internally, Maruti-Suzuki’s drive for improved productivity and cost reduction in the 2000s relies only partly on higher levels of automation. The main thrust of productivity increases rests on the introduction of new work practices and training initiatives to bring Maruti-Suzuki’s operations further in line with the parent operations. As part of that, the Maruti Production System has been introduced. Derived from Suzuki, its major emphasis is ‘to identify and eliminate waste in operations, such as unnecessary movement of men and material, in process waiting and so on’ (Maruti Udyog Ltd., 2005). Apart from the new measures to boost the productivity of existing facilities, Maruti-Suzuki and Suzuki set up new production facilities in 2007. These facilities are not only envisioned to supply the growing domestic market, but to facilitate cost reduction through further localisation of critical components and to build production capacity for global exports. In 2007, a fourth assembly plant was inaugurated at Manesar, the state of Haryana. The initial capacity was set at 100,000 cars per year, with the goal to scale up production to about 300,000 cars by 2010. On the same compound, Suzuki inaugurated in 2007 Suzuki Powertrain India Limited, a diesel engine and transmission plant. Under a licence agreement with Fiat Auto, the company will produce diesel engines (Suzuki, 2007). With three fully integrated plants at Gurgaon near New Delhi and a new plant at Manesar, Maruti-Suzuki will be capable of producing close to a million cars per year by 2010 (Maruti-Suzuki, 2007). The introduction of the Swift also sets new production standards. In contrast to earlier model introductions, the Swift has had from the start an 85 per cent level of localisation. About 70 per cent of the car’s welding is automated with the use of 89 robots on the line. This contrasts with the Alto, which has automation levels in welding of about 50 per cent. Maruti-Suzuki is also able to use the Swift’s new welding line more flexibly than those of older models. With regard to its goal of reducing vehicle platforms, however, Maruti-Suzuki has only made modest progress. In 2007, the company was still working on producing its model portfolio (from six to five) on just three platforms. Overall, however, Maruti-Suzuki’s productivity enhancement efforts seem to be paying off. In 2004–5, Maruti-Suzuki reported that ‘Challenge 50’ improved productivity by 46 per cent (Maruti Udyog Ltd., 2005). Yet, some
Maruti-Suzuki’s Trajectory 413
caution is warranted, as productivity calculations may have been based on permanent employees only. Improving employment relations and a new governance compromise Following the takeover by Suzuki, the new cost cutting programmes and the strike experience in 2000–1, Maruti-Suzuki’s management pushed for a new governance compromise, mainly by altering the employee level, the employee composition as well as the compensation and incentive systems. The new governance compromise is a result of and facilitated by the new competitive situation for Maruti-Suzuki in the Indian automobile market and by the new ownership situation of the company. It is essentially imposed by Maruti-Suzuki’s management whose bargaining power increased with the company’s difficulties in the early 2000s and the dwindling influence of the Indian government. Professionalisation of the human resourse management, following the labour unrest, also played a pivotal role in implementing the new compromise (Som, 2004). Reflecting Suzuki’s drive to reduce what it sees as rampant overstaffing and the drive to bring down cost, MarutiSuzuki sought to reduce its level of permanent employees through voluntary retirement schemes (VRS). The first round of employee reduction started in 2001–2. As 19 per cent of Maruti-Suzuki’s employees accepted the scheme, the employment level fell by more than 1,000 employees, from 5,770 in 2000–1 to 4,590 in 2002–3. In 2003–4 Maruti-Suzuki initiated a second round of VRS, bringing the employee level further down to 3,334. With the opening of the fourth plant in 2007, the company saw for the first time in the decade a rise in the employee level (see Appendix Table 21.1). In addition, shifts in the workforce composition have taken place. There are strong indications that Maruti-Suzuki is increasingly turning ‘low-value jobs’ into contract labour (Red Herring Prospectus, 2003). While exact figures are not available, industry observers estimate that Maruti-Suzuki currently employs, in addition to regular workers, twice their amount as contract labour. This contract labour earns (Rs. 5,000) about one-fifth of their colleagues (Rs. 30,000–35,000) in regular employment relations. In addition, Maruti-Suzuki’s employees also face a ‘streamlining’ of their salary structure in the 2000s. This mainly involves a stronger alignment of cost to the company, company performance and employee performance with compensation, promotion and incentive packages for all employees, including managers (Maruti Udyog Ltd., 2001, 2005). While there is a stronger performance orientation, Maruti-Suzuki feels more and more the need to reduce the high turnover rate of managers, through a host of retention measures (Maruti Udyog Ltd., 2006, 2007). Maruti-Suzuki has also improved social security packages for the remaining core of employees (Red Herring Prospectus, 2003). By the same token, training focuses more on core employees. Apart from training initiatives targeting a stronger customer orientation within the
414 The Second Automobile Revolution
workforce, a crucial goal is to restore the work atmosphere by means of training programmes for a new corporate culture. Maruti-Suzuki has also sought to improve its damaged union relations. In 2002, Maruti-Suzuki’s management found an opportunity to derecognise the old company union that sparked the company’s first strike. Simultaneously, the management recognised the new union that formed during the labour conflict. In 2003 around 3,000 employees were represented by the new union, which appears to be more ‘cooperative’. By the mid-2000s, Maruti-Suzuki’s employment relations have returned to peace, signalling the settlement of a new governance compromise.
Conclusion and outlook Maruti-Suzuki’s trajectory from a national champion to a Japanese-owned subsidiary shows exemplarily how the firm’s production model and governance compromise was embedded in the changes to India’s socio-economic context. In the 1990s these contextual changes were substantial as India embarked on a far-reaching transformation and liberalisation of its economy. While Maruti-Suzuki initially, as a government-owned company, benefited from the socio-economic transformation, the ownership situation turned into a liability when wider societal conflicts were reflected in the company. For Maruti-Suzuki this development involved a crisis of its governance compromise and difficulties in maintaining its production model. Ultimately, the declining performance of the company in a new competitive environment, and a new reform push by the Indian government at the end of the 1990s, prepared the ground for the company’s privatisation and a new governance compromise in the following decade. By the mid-2000s, Maruti-Suzuki has become a private and foreign-owned company whose production model and governance compromise seem to have achieved again a good fit with conditions in the Indian automobile market. While it is an open question to what extent the global financial crisis will affect automobile demand in India, it will probably remain dominated by small and medium car segments. However, despite substantial changes, particularly in the company’s governance compromise, we also need to stress continuity in Maruti-Suzuki’s context and trajectory. While it is true that the company started to diversify into new segments from the 1990s onward, volume expansion, offering ‘value for money’ and catering to low market segments were strategic fix points throughout. On the one hand, this was related to Suzuki’s overall strategy. After all, Suzuki is an international player with a strong position in the small and sub-compact car segments. On the other hand, India has been and remains a developing country. Indian automobile demand is very price sensitive and volumes are generated in lower segments. Moreover, India is heavily dependent on oil imports and keeps a watchful eye on its balance of payments. Therefore, the conservation of energy and export growth have been
Maruti-Suzuki’s Trajectory 415
and will continue to be paramount objectives in India’s automobile policies. In addition, battling pollution and infrastructure problems are increasingly feeding into the country’s automobile policies. As a consequence, creating incentives for producers of small cars and their export is a contextual constant and is also reflected in the Indian government’s new automobile mission plan for 2006–16. The plan envisions developing India into a hub for the production and export of small cars. The mission plan reflects India’s past emphasis and experience in small car production and informs the country’s future path as government policies create new incentives for profit strategies and product policies that focus on small, clean and fuel-efficient cars. Clearly, having focused throughout much of its existence on such cars, Maruti-Suzuki will continue to benefit from these government policies. However, other players catering to the same segments will benefit as well. There are two particular competitors in the Indian market that have eaten away big chunks of Maruti-Suzuki’s market share and are also eyeing India’s low and entry segments. These are Hyundai and Tata Motors. Tata presented its super cheap Nano car early in 2008, to cost around 1,700 euros. If massproduced (350,000 vehicles per year targeted for first few years) at the price, it will be a vehicle that not only breaks into Maruti-Suzuki’s A-segment monopoly, but costs about half the price of Maruti-Suzuki’s cheapest offer in the segment (Matuti 800). While it remains to be seen whether Tata will be able to live up to its goal, growing price competition at the entry level is bound to further challenge Maruti’s bottom line, particularly under the conditions of the financial and economic crisis at the end of the 2000s.
416 The Second Automobile Revolution Appendix 21.1 Fiscal year 1984–85 1985–86 1986–87 1987–88 1988–89 1989–90 1990–91 1991–92 1992–93 1993–94 1994–95 1995–96 1996–97 1997–98 1998–99 1999–00 2000–01 2001–02 2002–03 2003–04 2004–05 2005–06 2006–07
Maruti-Suzuki, 1984–2007
Turnover Profit (million Rs.) Production Market of vehicles share (million Rs.) Before tax After tax
14,882 19,129 21,488 27,929 41,494 64,130 76,803 82,066 77,814 93,151 89,287 90,809 90,636 114,250 137,271 151,823 178,043
481 385 366 1,367 2,632 6,524 8,076 9,773 7,842 3,851 −2,692 1,183 2,821 7,698 13,049 17,500 22,798
9 20 102 223 265 419 483 291 378 861 2,476 4,276 5,101 6,519 5,230 3,301 −2,694 1,045 1,464 5,421 8,536 11,891 15,620
22,372 47,694 82,308 94,033 105,592 117,255 120,308 121,167 128,138 158,109 206,330 277,776 336,811 354,336 333,198 407,589 350,374 358,108 359,960 472,908 540,409 572,097 630,0001
Export
480
74.8 73.0
83.1 78.8 61.5 57.6 58.6 54.6 51.4 54.5 55.1 54.6
26,103 35,031 25,994 24,100 21,450 15,300 12,233 32,240 51,175 48,899 34,784 39,295
Employees
2,815 3,497 3,526 3,629 3,721 3,945 3,993 4,042 4,141 4,840 4,968 5,324 5,642 5,719 5,848 5,770 4,627 4,590 3,334 3,453 4,791 4,9932
Notes: 1. Planned. 2. Industry observers expect that, in addition to regular workers represented in this figure, twice their amount works currently as contract labour. Sources: Maruti Udyog Ltd., 2000–7; Okada, 1998; Red Herring Prospectus, 2003; Suzuki, 2007.
Maruti-Suzuki’s Trajectory 417
Note 1. Based on vehicle length and price India’s automobile market is commonly segmented as follows: A1/A mini segment (up to 3,400 mm; 30 und
146 (70 Japan, 76 overseas) 142 103 >350 180 159 121 300 197 232 398 222 191
33 und 19 26 29 29 25
265 291 80 211 206 206 158
Manufacturing sites World
Domestic
und
und
121 80 270 est. 165 137 110 159 143 171 214 132 170 215 229 69 147 129 87 95
RD, design, technical centres Foreign World
Domestic
Foreign
58
und
und
6
46 (NA) 6 60 est. 58 48 65 (NA) und 48 37 30 9 48
und 74 und 107 89 und und 95 134 184 123 122
21 20 und 15 22 11 und 26 25 und 18 14
6 (NA) 2 und 4 6 10 (NA) und 19 12 und 6 10
und 18 und 11 16 und und 7 13 und 12 4
und 62 15 41 (NA) 56 (EO) 15 28
und 167 54 und und 72 67
46 62 5 19 68 und 6
und 8 1 7 (NA) 42 (EO) und und
und 54 4 und und und und
Notes: EO: Western Europe; NA: North America; und: undetermined. The table is built along the following lines (sources: website/10-k form/Annual Report): – Columns 2 & 3: Company statements. – Column 4: Manufacturing: survey of sites by author, except where the firm has provided a summarised table. – Column 5 (R&D, design, technical centres): same compilation principles as manufacturing uses. The data here have been aggregated since firms have various ways of accounting for these three types of sites. When a production and an R&D site are associated, they are only accounted for once, as a production site.
jobs a year to LCCs; Lear has stated that, in addition to its internal components, 30 per cent of its other components will come from these sources (the target being 45 per cent by 2010). These strategies are not necessarily new if we remember the delocalisations that began in the late 1970s when US suppliers were first moving into Mexico (Carillo, 2004; Lara Rivero and Carillo, 2003). However, the trend has accelerated and it now affects all continents (Domanski et al., 2006; Frigant and Layan, 2009; Humphrey et al., 2000). It remains that productive constraints limit the distance at which a delocalisation can occur, meaning that this tends to materialise within the framework of a regional unit. Central zones all seem to have developed peripheral spaces
Winners and Losers in the Auto Parts Industry
435
Table 22.4 Distribution of staff and plants of FTSs between domestic and nearby peripheral zones (year end 2006; %) Company
Autoliv Bosch Bridgestone Continental Delphi Denso Faurecia Freudenberg GKN Johnson Controls Lear Magna Michelin Valeo Visteon ZF Fried.
Employees
Plants
Domestic (%)
Peripheral LCCs (%)
Domestic (%)
Peripheral LCCs (%)
67.8 84.6
32.2 15.4
63.3
36.7
58.6 76.7 64.4 52.7 82.4 86.1 94.8 83.3 86.6 89.2 84.2 80.0 72.7 86.2
41.4 23.3 35.6 47.3 17.6 13.9 5.2 16.7 13.4 10.8 15.8 20.0 27.3 13.8
82.5 97.2
17.5 2.8
77.9
22.1
75.4
24.6
Notes: Domestic: USA/Canada or Western Europe or Japan. LCCs: Mexico or Eastern Europe or Other Asia/Pacific. Example: Magna (Canada, North America): 22.1% of its North American employees are located in Mexico; 10.8% of its North American production plants are in Mexico.
where most basic components are manufactured, usually alongside complex components and even R&D activities. At the same time, not all suppliers have adopted similar spatial schemes and the variations can be significant (Table 22.4). One noteworthy point is that the most profitable suppliers are not necessarily the ones with the greatest number of units or staff members working in LCCs. One example is GKN, the most profitable firm but also one that has the fewest units in Eastern Europe. Inversely, Delphi’s strong presence in Mexico has not helped its earnings – a sign that the search for low production costs is not the only solution to the supply industry’s profitability problems.
Conclusion The last ten years have witnessed major organisational restructurings that have had a direct effect on automobile suppliers. Despite forecasts that greater carmaker-driven outsourcing would help suppliers to achieve their desired aims, recent analyses are more circumspect. It is true that a global elite has been born but this has occurred to the detriment of other suppliers who
436 The Second Automobile Revolution
have shown a lesser ability to master their trajectory. Despite rapid growth in the global market, certain FTSs have been unable to seize their chance or have made wrong strategic choices, meaning that they have suffered whilst rivals forged ahead. In this reshuffling of the deck, supplier subsidiaries have performed relatively well. But no one has really found an economic equilibrium – actors’ global profitability is low and many suppliers, mostly in the US, are in difficulty. Yet nothing is set in stone, at least not in terms of potential strategies. Companies can choose to pursue a proactive modular policy or else bide their time until the market matures; they can try to develop key technological and organisational competences or else focus on reorganising their productive configuration (number of units, locations, etc.). Given that targeting Tier 1 suppliers is not necessarily the only path possible (Herrigel, 2004) and since the companies in question are not starting from scratch, surely it will suffice if they adapt their current trajectories to market dictates. The notion of trajectory also helps us to understand that there is no such thing as a unique supplier profile. Our analysis emphasises the diversity of individual configurations, even if actors do have some problems in common. De facto, suppliers devise singular compromises in their ways of apprehending and adapting to restructured vertical relationships. We are willing to bet that the future will hold further surprises in terms of how different suppliers perform in meeting such challenges.
23,604 31,963 20,670 188,000 97,767 62,357 86,000 197,568 72,359 29,154 29,289 52,300 89,000 65,316 N/A 127,241 N/A 50,400 82,000 33,264
1998
32,860 36,500 22,580 194,000 101,489 62,155 84,200 214,200 80,795 31,808 29,667 N/A 95,000 121,102 59,000 130,434 90,100 51,700 77,000 29,858
1999 36,343 36,000 28,000 197,000 102,165 63,832 79,300 216,000 85,371 36,390 30,118 N/A 105,000 121,636 62,000 128,122 72,800 75,200 82,000 31,915
2000
Employees of FTSs, 1998–2006
Note: N/A = not available. Source: Annual Reports enterprise.
Aisin Seiki ArvinMeritor Autoliv Bosch Bridgestone Continental Dana Delphi Denso Faurecia Freudenberg GKN Johnson Controls Lear Magna Michelin TRW Valeo Visteon ZF Friedrichshafen
FTS
Appendix Table 22.1
40,234 33,000 28,300 218,000 104,700 65,293 70,000 206,000 86,639 55,089 27,578 49,800 112,000 113,577 67,000 127,467 N/A 70,000 79,000 33,959
2001 42,500 33,000 30,100 224,941 106,846 64,379 63,100 192,000 89,380 58,878 27,693 48,600 102,000 114,694 73,000 126,285 63,600 69,100 77,000 53,281
2002 44,132 32,000 32,100 231,600 108,741 68,829 59,000 190,000 95,461 59,578 28,479 47,900 108,000 111,022 75,000 127,210 60,800 68,200 72,000 53,487
2003 47,616 31,000 34,500 N/A 113,699 80,586 45,900 185,200 104,183 62,507 32,004 40,500 113,000 110,083 81,000 126,474 59,900 67,276 70,200 54,542
2004 53,237 29,000 34,100 250,975 123,727 79,849 44,000 184,200 105,723 61,722 33,385 39,900 114,000 115,113 82,800 127,319 63,100 70,304 49,575 53,940
2005
59,587 27,500 41,800 261,292 126,326 85,224 45,000 171,400 112,262 65,682 33,526 40,500 136,000 104,276 83,300 126,673 63,800 69,663 45,000 55,050
2006
437
100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
Aisin Seiki ArvinMeritor Autoliv Bosch Bridgestone Continental Dana Delphi Denso Faurecia Freudenberg GKN Johnson Controls Lear Magna Michelin TRW Valeo Visteon ZF Friedrichshafen
Source: Thomson Financial.
1998
97.1 116.0 109.3 108.4 93.2 135.4 105.0 102.5 107.1 177.4 106.3 124.8 128.2 137.2 155.8 110.2 88.7 128.2 105.1 102.4
1999 107.0 134.3 118.0 122.6 89.7 150.0 98.6 102.3 114.6 243.0 122.0 138.8 136.3 155.3 176.6 123.3 92.5 151.5 109.6 121.0
2000 120.7 177.4 114.4 132.2 95.4 166.6 82.2 91.6 136.5 400.0 117.7 115.5 146.4 150.4 183.6 126.3 85.5 170.0 100.5 133.8
2001 130.7 179.4 127.4 135.9 100.5 169.2 75.2 96.3 132.6 410.6 115.0 111.2 159.7 159.2 216.0 125.3 88.0 162.9 103.6 181.5
2002
Sales of FTSs (group level), 1998–2006 (100 = 1998)
FTS
Appendix Table 22.2
150.6 203.0 151.9 141.3 103.0 171.1 63.0 98.7 145.7 421.3 113.5 112.2 179.9 173.8 255.5 123.1 96.2 153.4 99.4 176.7
2003 171.7 209.4 176.1 155.5 108.0 186.8 71.7 100.5 159.2 446.1 129.7 117.2 211.0 187.2 343.9 125.7 101.8 156.8 105.0 195.9
2004 195.7 232.1 177.9 161.1 120.3 205.2 68.1 94.6 181.3 456.9 142.0 122.7 218.3 188.6 379.8 124.9 107.1 165.0 95.6 214.4
2005 226.9 239.7 177.4 169.7 133.7 220.8 67.3 92.7 205.2 484.8 148.4 122.3 256.1 196.9 402.6 131.2 111.4 165.6 64.3 230.8
2006
126.9 139.7 77.4 69.7 33.7 120.8 −32.7 −7.3 105.2 384.8 48.4 22.3 156.1 96.9 302.6 31.2 11.4 65.6 −35.7 130.8
Growth rate 1998–2006 (%)
Appendix Table 22.3 Year
Net incomes/sales (%) of FTSs, 1997–2006 2006
2005
2004 2003 2002 2001 2000 1999 1998 1997
Aisin Seiki 2.9 2.6 2.2 3.4 2.1 ArvinMeritor −1.9 0.4 −0.5 1.8 2.2 Autoliv 6.5 4.7 5.3 5.1 4.1 Bosch 4.7 5.4 4.5 3.4 2.0 Bridgestone 2.8 6.7 4.7 3.9 2.0 Continental 6.6 6.7 5.3 2.7 2.0 Dana −8.7 −18.6 1.0 2.8 0.4 Delphi −20.7 −8.7 −16.6 −0.2 1.3 Denso 5.7 5.3 4.7 4.3 4.8 Faurecia −3.8 −1.7 0.8 0.1 −0.6 Freudenberg 4.0 3.9 3.9 2.4 3.5 GKN 4.9 1.5 16.6 3.0 3.0 Johnson Controls 3.2 3.3 3.1 3.0 3.0 Lear −4.0 −8.1 2.5 2.4 0.1 Magna 2.2 2.8 3.3 3.3 4.0 Michelin 3.5 5.7 3.3 2.1 3.7 TRW 1.3 1.6 0.2 −0.6 2.1 Valeo 1.3 1.4 1.6 2.0 1.4 Visteon −1.5 −1.6 −8.0 −6.9 −0.5 ZF Friedrichshafen 2.5 2.3 1.7 −1.9 0.1
−0.6 1.8 0.5 4.2 1.2 4.1 1.8 4.2 0.8 0.9 −2.2 2.0 −2.9 2.7 −1.4 3.6 3.0 3.0 −0.6 −0.7 3.3 2.2 1.0 7.7 2.6 2.7 0.2 2.0 4.8 5.3 1.9 2.6 −0.2 0.9 −5.8 4.0 −0.7 1.4 2.2 2.4
0.9 4.4 5.2 1.4 4.3 2.6 3.9 3.7 3.3 1.3 2.2 9.7 2.6 2.1 4.6 1.1 1.1 7.3 1.5 –
1.5 3.8 5.4 1.6 4.7 2.9 4.2 −0.3 3.4 3.2 2.4 19.1 2.6 1.3 4.5 4.3 – 4.3 4.0 –
1.5 3.3 – 3.3 1.8 2.5 4.3 0.7 4.3 2.7 3.7 9.7 1.9 2.8 7.6 4.9 – 4.4 3.0 –
Sources: Thomson Financial and enterprises (Annual Reports).
Appendix Table 22.4 Year
EBIT/sales (%) of FTSs, 1997–2006 2006 2005
Aisin Seiki 6.0 ArvinMeritor −1.4 Autoliv 8.5 Bosch – Bridgestone 5.5 Continental 10.8 Dana – Delphi – Denso – Faurecia −2.7 Freudenberg – GKN 6.1 Johnson Controls 4.0 Lear – Magna 3.2 Michelin 7.3 TRW 4.4 Valeo 3.3 Visteon 0.4 ZF Friedrichshafen 5.7
5.2 1.5 8.4 8.2 9.7 11.2 −1.4 −7.9 8.7 −0.4 6.9 3.4 3.8 −5.5 4.4 10.0 4.3 2.8 −0.2 4.8
2004 2003 2002 2001 2000 1999 5.0 7.7 3.3 4.0 8.4 8.2 7.0 5.6 8.0 7.6 8.9 7.5 0.9 3.8 −1.7 0.1 8.1 7.4 2.3 2.1 6.9 5.7 17.2 3.5 4.7 4.9 4.3 4.6 5.7 6.8 6.6 5.3 3.4 3.5 2.3 3.1 −2.4 −6.3 4.5 −0.1
5.1 5.0 7.6 4.9 5.7 6.5 2.6 2.4 10.2 1.1 7.0 4.0 5.4 4.8 7.2 7.8 6.4 3.3 −0.3 2.2
Sources: Thomson Financial and enterprises (Annual Reports).
−0.6 2.9 4.4 4.8 −4.8 0.2 −1.7 −1.4 6.0 1.5 6.8 1.5 5.2 2.7 8.6 5.9 3.2 −4.6 −0.3 4.4
5.0 8.3 8.5 5.1 3.4 5.0 6.3 6.1 5.6 1.1 5.5 9.7 5.7 5.7 9.7 6.8 5.5 6.1 2.8 5.1
3.2 8.2 10.0 3.6 9.6 6.0 7.7 6.2 6.1 3.7 6.2 11.4 5.7 5.5 7.7 7.1 5.8 10.9 3.1 –
1998 1997 3.7 7.1 10.4 4.1 9.0 6.4 8.7 −0.2 6.4 5.3 6.2 20.7 6.0 3.6 7.7 10.4 – 5.7 6.6 –
4.0 5.8 – 6.6 4.0 5.9 10.0 1.7 7.2 5.0 8.1 12.5 4.9 6.1 11.3 10.6 – 5.5 5.0 –
440 The Second Automobile Revolution
Notes 1. In a study of thirty suppliers, we calculated that they had engaged in 953 takeover operations between January 1989 and July 2003, mostly in their country of origin and in the automobile supply sector (Frigant, 2004). 2. This analysis covered a sample of twenty FTSs, chosen from the main suppliers as ranked in 1998 (and not in 2006 to take into account the ‘failures’). Groups were excluded when the automobile amounted to less than 60 per cent of total sales (ThyssenKrupp, Siemens, Eaton), as was Yasaki due to a lack of data. We did add Continental, the first 2006 top thirty firm, which had not been selected using the aforementioned criteria. 3. An estimate is that in recent years, thirty-five US suppliers with revenues of over $1 billion sought protection under Chapter 11. 4. Particularly active private equity included The Carlyle Group, which acquired Beru in 2000 (and resold it in 2005 to BorgWarner), United Components in 2003, AxleTech Int. and Diversified Machine in 2005 and Allisson Transmission in 2006. Blackstone was also very active with the purchase of American Axle & Manufacturing in 1997, Collins & Aikman in 1998 and TRW in 2003. 5. Hedge funds control 55.6 per cent of Dana, 37.6 per cent of Delphi and 24.4 per cent of Visteon (data Thomson Financial, March 2007). 6. In 2007, for example, a fund called Pardus acquired a further 6 per cent stake in Valeo between January and March, with a view towards ultimately engineering a merger between this company and Visteon. 7. We consider gross spending (including customer invoices under long-term contracts) since this is the most frequently published form of data. The sample includes nineteen companies since Magna International does not provide any information.
Bibliography Aoki, M. (1988) Information, Incentives and Bargaining in the Japanese Economy. New York: Cambridge University Press. Asanuma, B. (1989) ‘Manufacturer–Supplier Relationships in Japan and the Concept of Relation-Specific Skill’, Journal of the Japanese and International Economies, 3: 1–30. Autobusiness (2004) Modules & Sytems. Stamford: AutoBusiness Ltd. Batchelor, J. (2006) ‘Modularisation and the Changing Nature of Automotive Design Capabilities’, International Journal of Automotive Technology and Management, 6(3): 276–97. Becker, M. and Zirpoli, F. (eds) (2005) ‘Knowledge and Task Partitioning in the Auto Industry: Coordination, Governance and Learning in New Product Development’, Special issue of International Journal of Automotive Technology and Management, 5(2). Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Brocard, P. and Donada, C. (2003) La chaîne de l’équipement automobile. Paris: SESSI. Carrillo, J. (2004) ‘Transnational Strategies and Regional Development: the Case of GM and Delphi in Mexico’, Industry and Innovation, 11(1/2): 127–53. Chanaron, J.-J. (2001) ‘Implementing Technological and Organizational Innovations and Management of Core Competences: Lessons from the Automotive Industry’, International Journal of Automotive Technology and Management, 1(1): 128–44. Clark, K. and Fujimoto, T. (1991) Product Development Performance: Strategy, Organization and Management in the World Auto Industry. Boston: Harvard Business School Press.
Winners and Losers in the Auto Parts Industry
441
Domanski, B., Guzik, R. and Gwosdz, K. (2006) ‘The New Spatial Organisation of the Automobile Industry in Poland’, Proceedings of the 14th GERPISA International Colloquium, 12–13 June, Paris. Fourcade, F. and Midler, C. (2004) ‘Modularisation in the Auto Industry: Can Manufacturer’s Architectural Strategies Meet Supplier’s Sustainable Profit Trajectories?’, International Journal of Automotive Technology and Management, 4(2/3): 240–60. Fourcade, F. and Midler, C. (2005) ‘The Role of 1st Tier Supplier in Automobile Product Modularisation: the Search for a Coherent Strategy’, International Journal of Automotive Technology and Management, 5(2): 146–65. Freyssenet, M. (2005) ‘Les “parcs de fournisseurs” à la lumière des nouveaux compromis de gouvernement’, in H. Dumez (ed.), Gouverner les organisations. Paris: L’Harmattan, pp. 217–25. Frigant, V. (2004) ‘L’internationalisation marchande et productive des équipementiers automobiles: une évaluation empirique’, Working Paper of GRES, No. 2004-16, http://ideas.repec.org/p/grs/wpegrs/2004-16.html, July. Frigant, V. (2007) ‘Between Internationalisation and Proximity: the Internationalisation Process of Automotive First Tier Suppliers’, Working Paper of GREThA, No. 2007-13, http://ideas.repec.org/p/grt/wpegrt/2007-13.html, July. Frigant, V. and Layan, J.-B. (2009) ‘Modular Production and the New Division of Labour within Europe: the Perspective of French Automotive Parts Suppliers’, European Urban and Regional Studies, 16(1): 11–25 Froud, J., Johal, S. and Williams, K. (eds) (2002) The Tyranny of Finance? New Agendas for Auto Research, Special issue of Competition & Change, 6(1). Garud, R. and Kumaraswamy, A. (1995) ‘Technological and Organizational Designs to Achieve Economies of Substitution’, Strategic Management Journal, 16: 93–109. GERPISA (1999) ‘International Division of Labor and Assembler-Supplier Relations’, Actes du GERPISA, Université d’Evry-val-d’Esssonne, No. 25, February. GERPISA (2002) ‘Vertical Relations and Modularisation in the Automotive Industry’, Actes du GERPISA, Université d’Evry-val-d’Esssonne, No. 33, March. Helper, S. (1991) ‘How Much Has Really Changed between US Automakers and Their Suppliers?’, Sloan Management Review, 32(4): 15–28. Helper, S. and Sako, M. (1995) ‘Supplier Relations in Japan and the United States: Are They Converging?’, Sloan Management Review, 36(3): 77–84. Herrigel, G. (2004) ‘Emerging Strategies and Forms of Governance in High-wage Component Manufacturing Regions’, Industry and Innovations, 11(1/2): 45–79. Humphrey, J., Salerno, M. and Lecler, Y. (eds) (2000) Global Strategies and Local Realities: the Auto Industry in Emerging Markets. New York: St Martin’s Press. Laigle, L. (2003) ‘The Internationalization of the French Automobile Component Industry and the Case of Valeo’, in M. Freyssenet, K. Shimizu and G. Volpato (eds), Globalization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan. Lara Rivero, A. A. and Carillo, J. (2003) ‘Technological Globalization and Intra-company Coordination in the Automotive Sector: the Case of DelphiMexico’, International Journal of Automotive Technology and Management, 3(1/2): 101–21. Larsson, A. (2002) ‘Learning or Logistics? The Development and Regional Significance of Automotive Supplier-Parks in Western Europe’, International Journal of Urban and Regional Research, 26(4): 767–84. Lecler, Y. (1993) La référence japonaise. Limonest: L’Interdisciplinaire.
442 The Second Automobile Revolution Lung, Y. (2001) ‘Coordinating Competencies and Knowledge: a Critical Issue for Regional Automotive Systems’, International Journal of Automotive Technology and Management, 1(1): 108–27. Sako, M. (1992) Prices, Quality and Trust: Inter-firm Relations in Britain and Japan. Cambridge: Cambridge University Press. Sako, M. (2003) ‘Modularity and Outsourcing: the Nature of Co-evolution of Product Architecture and Organisation Architecture in the Global Automotive Industry’, in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford: Oxford University Press, pp. 229–53. Sako, M. and Murray, F. (2000) ‘Modules in Design, Production and Use: Implications for the Global Automotive Industry’, Paper presented at the 8th GERPISA International Colloquium, 8–10 June, Paris. Salerno, M. (2001) ‘The Characteristics and the Role of Modularity in the Automotive Business’, International Journal of Automotive Technology and Management, 1(1): 92–107. Takeishi, A. (2005) ‘Structural Change in the Japanese Supplier System’, Communication at the IMVP-MMRC Asia Pacific Forum, 13 October, Tokyo. Takeishi, A. and Fujimoto, T. (2003) ‘Modularization in the Car Industry: Interlinked Multiple Hierarchies of Product, Production, and Suppliers Systems’, in A. Prencipe, A. Davies and M. Hobday (eds), The Business of Systems Integration. Oxford: Oxford University Press, pp. 254–78. Volpato, G. (2004) ‘The OEM-FTS Relationship in the Automotive Industry’, International Journal of Automotive Technology and Management, 4(2/3): 166–97.
23
Conclusion: the Second Automobile Revolution – Promises and Uncertainties Michel Freyssenet
The first automobile revolution was characterised by the adoption of a global standard – the internal combustion engine, sustained by a liquid fossil fuel (oil) and by the diffusion of automobiles, mainly throughout the industrialised world. After a brief era of confrontation between different energy producers, the tandem ‘internal combustion engine/petrol’ became the global standard and still dominates today (Bardoux et al., 1982). Aside from a dramatic shift in production volumes between the first and the second half of the twentieth century – as shown in Figure 23.1 – the latter half of the 80 70
Vehicles (millions)
60 50 40 30 20 10
18 9 19 8 0 19 1 0 19 4 0 19 7 1 19 0 1 19 3 1 19 6 1 19 9 2 19 2 2 19 5 2 19 8 3 19 1 3 19 4 3 19 7 4 19 0 4 19 3 4 19 6 4 19 9 5 19 2 5 19 5 5 19 8 6 19 1 6 19 4 6 19 7 7 19 0 7 19 3 7 19 6 7 19 9 8 19 2 8 19 5 8 19 8 9 19 1 9 19 4 9 20 7 0 20 0 0 20 3 06
0
Year Figure 23.1
Worldwide automobile production, 1898–2007
Notes: All types of automobile vehicles; numbers in millions. Sources: (main) WMVD, SMMT, JAMA, IRF, CCFA, OICA. Elaboration: Freyssenet (2004), up-to-date 2008. Website: freyssenet.com
443
444 The Second Automobile Revolution
century fulfilled the earlier promise of what can be called the automobilisation of industrialised societies. Achieving this promise was predicated on two political choices: the creation of an adapted and dense road network within a framework determined by state authorities; and the post-Second World War implementation in most industrialised nations of a ‘coordinated and moderately hierarchised’ mode of national income distribution. This mode triggered a self-sustaining process until 1974, one that ‘school of regulation’ economists have called Fordism, in homage to Antonio Gramsci, who was the first to use this term to refer to the variant of capitalism in which increases in real wages are viewed as a precondition for mass consumption and industrial development.1 The new forms of work organisation and their extension to ever more activities, along with the general development of social protection, education and public services, expanded the middle classes and more broadly the population of individuals who could afford to acquire a new vehicle. The end of this era (the 1970s, a time of monetary and oil shocks) led to a major slowdown in the growth of global automobile production. Carmakers long hoped that so-called ‘developing’ countries (some of which came to be known as ‘emerging’ markets) would become the new growth catalysts. First comprising oil and raw materials producers, the list would later include Latin America, followed by Southeast Asia and the former Soviet bloc. Things did not turn out as expected, however, with carmakers being forced to compete ferociously with one another in the world’s mature markets. This turn of events mainly benefited Japanese, Korean and German manufacturers, that is, until the liberalisation of global capital flows and the ensuing speculative bubbles revealed their financial fragility. Fears of a rapid depletion of oil and of skyrocketing fuel prices fell by the wayside by the mid-1980s, a period marked by a counter-oil shock and by variations in the exchange value of the US dollar, the world’s transaction currency. This turnaround led to an explosion in demand for a new type of vehicle, light trucks, which were both socially ostentatious and gas guzzlers. The decline in ‘coordinated and moderately hierarchised’ modes of national income distribution, replaced by ‘competitive’ modes based on ‘merit’, a local and categorial balance of power and financial opportunism, led to the emergence of new segments of society that wanted to show off their good fortune, and to do this as soon as possible, given the potentially ephemeral nature of their newfound wealth (Boyer and Freyssenet, 2002). American carmakers rode the wave of light trucks to return to profitability. Despite the efforts of the automobile industry to minimise or relativise criticism, consumer associations followed by political parties and governments were increasingly vociferous in their denunciation of the ‘externalised’ costs of the automobile society (road accidents, pollution, network maintenance, police, justice, relations with oil-producing countries, etc.). State authorities and carmakers long blamed each other for the time it was taking to come up with alternative driving systems. Then, at the start of the twenty-first century, two apparently unrelated
Conclusion: the Second Automobile Revolution
445
major events caused a dramatic change leading to the second automobile revolution.
Start of the race towards alternative driving systems and resurgence of the global automobile market thanks to the BRICs The continuous increase in the price of oil since 2005, shooting beyond $140 per barrell in 2008, along with rising extraction costs and difficulties, global warming, a customer base willing to pay a surcharge for vehicles emitting less CO2 (as shown by the success of hybrid vehicles since Toyota’s 1997 launch of the Prius in Japan), increasingly stringent environmental standards and restrictions on vehicle use in certain urban spaces – all of these factors have finally convinced carmakers and major suppliers that it is time to seek more sustainable solutions. Each of these issues will be resolved in its own time depending on the pressures exerted. One major contributory factor today is carmakers’ fear of being technically and commercially wiped out by the competition, all the more so given the general uncertainty about future solutions (better performance for existing engines, petrol-electricity hybrids or engines that run on gas, biofuels, stored electricity and/or fuel cells). Above all, there are still many questions about which category of actors will set the standards of the future – existing carmakers, battery makers, energy producers, tyre manufacturers, major suppliers, etc. In early 2008, each of the leading carmakers signed an agreement with one or another battery maker, announcing a large-scale production of vehicles featuring alternative driving systems for the not so distant future. So far, different paths have been envisaged for achieving this goal. Moreover, none has been ruled out. At the same time, the automobile markets have exploded in the politically ambitious continent-sized countries (Brazil, Russia, India and China, the socalled BRICs). Nothing on this scale has ever happened before in the history of the automobile, neither in the United States in the 1950s, nor in Europe or in Japan in the 1960s. The phenomenon is unprecedented in both absolute value and growth terms, as demonstrated in Figure 23.2. The condition of its long-term survival – to wit, broadly shared national income – has not been fulfilled, and disparities have in actual fact increased in the BRICs. Given the size of these countries’ populations, however, they should experience at least a few more years’ growth. Moreover, through a political management of their economies, they may very well be able to contain whatever imported or internal crises might arise. The effects of today’s growth will simply be the same, albeit accentuated, as previously observed growth in modern industrialised countries: industrial and economic boom, general mobility, massive mobilisation of the rural workforce, social problems, urban congestion, pollution, fatal road accidents, rising regulation and exports.
446 The Second Automobile Revolution
35 30
Vehicles (millions)
Asia-Oceania 25 20
Europe
15 10 5
BRIC America Africa
China
18 9 19 8 0 19 2 0 19 6 1 19 0 1 19 4 1 19 8 2 19 2 26 19 3 19 0 3 19 4 3 19 8 4 19 2 4 19 6 5 19 0 5 19 4 58 19 6 19 2 6 19 6 7 19 0 7 19 4 7 19 8 8 19 2 8 19 6 90 19 9 19 4 9 20 8 0 20 2 06
0 Year
Figure 23.2 Worldwide automobile production by continent and production of certain countries, 1898–2007 Notes: all types of automobile vehicles; numbers in millions; BRIC = Brazil, Rusia, India, China. Sources: (main) WMVD, SMMT, JAMA, IRF, CCFA, OICA. Elaboration: Freyssenet (2004), up-to-date 2008. Website: freyssenet.com
One aspect that will clearly be different is the price of oil. Instead of volume effects contributing to price reductions in dollar terms – as was the case in the 1950s and 1960s – demand from the BRICs markets is helping to push prices higher. The needs of these countries are so massive that they create extra tensions in an oil market that is already stretched because of increasing extraction costs and given the need to maintain fossil fuel reserves for future use (due to the lack of any known substitutes). It is true that the BRICs are currently refusing any restrictions on their consumption without the older industrialised countries quantifying their own priority efforts in this area. The new giants (especially India and China) are fully aware of their lack of oil resources and how rising prices will constrain their economic development. Thus, they too have a strong interest in alternative driving modes and energy sources. The Chinese already have millions of bicycles and more and more electricity-run scooters. There are probably more surprises to come along these lines. Brazil has long had a fully functioning biofuel sector and promoted the driving systems that correspond to this particular energy source. It intends to export ethanol and the technologies associated with this. As for Russia, it has a strong interest in building up the market for natural gas.
Conclusion: the Second Automobile Revolution
447
Two of the main events serving as a prelude to the second automobile revolution are therefore interdependent and mutually amplifying. Furthermore, and as mentioned above, the BRICs are politically ambitious countries that plan on becoming independent, or at least on gaining a foothold in many different areas of economic activity, notably the automobile industry. Local carmakers in India, China and Russia (and why not Brazil one day?) are working all-out to rise up the global automobile production league tables, at a time when carmakers from the countries that were such an integral part of the first automobile revolution are finding it increasingly difficult to keep up.
Why a revolution? The promises of a clean and safe car and the upheavals that will potentially ensue The transition to safer non-polluting automobiles will modify mobility conditions and life in general, enabling the automobilisation of new populations and shaking up the automobile industry’s current structure, geography and economy. The benefits from modern automobiles – time-saving, greater access to many locations, superior travel conditions, available and multi-usage vehicles – have been completely exhausted in the industrialised world and partially exhausted elsewhere. As a result, the shortcomings of this mode of transport and the problems caused by its excessive use have become patently obvious to everyone and difficult to accept at a human, social, political or economic level. The end result is a whole host of serious problems, which can be more or less divided into four different categories: pollution; energy security; physical and material accidents; and the consumption and destructuring of spaces. Pollution affects the air, water, land, climate, ecological balance and public health. Air pollution derives from gas and toxic particle emissions that directly and indirectly cause many illnesses and premature deaths. They make a substantial contribution to climate change, which is affecting all human activity worldwide and will have considerable geo-political consequences. Then there is land and water pollution caused by liquid substances (oil, fuel, detergents, solvents, etc.) that are spilled accidentally or voluntarily by many different sources, and by uncollected, non-recycled or non-recyclable waste coming from materials like steel, cast-iron, aluminium, plastic and rare metals. Lastly, there is sound pollution made by internal combustion engines or tyres as they roll on different road surfaces (asphalt, concrete, rocks, earth, etc.), noises that can exceed levels and durations considered medically acceptable. The source of most transportation energy today is a fossil resource – oil – that is neither renewable nor distributed evenly across the world. It is therefore destined to become increasingly rare, thus expensive, and will be harder
448 The Second Automobile Revolution
and harder to extract from new fields. Given its strategic importance for the world economy and for the leading consumer countries’ defence needs, oil has always demanded – now more than ever – political and/or military control over the main producer countries. Because it is so widespread, the tandem ‘internal combustion engine/petrol’ is a main determinant of modern international relations. Automobiles launched at very high speeds and driven by hordes of largely uncontrolled individuals in variable and unsafe circumstances can be just as destructive as military weaponry. Road accidents can be humanly unsupportable, socially destructive and economically ruinous. Indeed, automobiles are one of the main causes of death and illness, notably amongst young persons. This concerns not only drivers and passengers but also other road users and more generally anyone who happens to be in the general vicinity. Not only are lives destroyed but there can also be significant indirect damage. Preventing physical and material accidents and remedying them ex post facto has become the basis for a whole set of economic activities (insurance, expertise, police, judiciary, road safety, licensing, medical treatment, surgeries, regulations, automobile repairs, spare parts manufacturing, rental or replacement vehicles, safety equipment engineering and production, public projects, etc.). In accounting terms, these activities contribute significantly to national economic wealth. At the same time, they are also a distraction from other jobs that many working people would find socially more productive and useful. Automobile mishaps are very costly to individuals and communities. Because the automobile is a means of transportation that is more or less capable of moving people from a point of departure to a final destination, it also has the potential to destructure and destroy historical, natural, social and individual spaces that were not intended to accommodate it or have been unable to do so. After years of efforts to adapt cities and indeed all spaces to the automobile, there is finally a consensus to exclude it from certain zones. The purpose of having a car is to save time and money, increase access to relatively inaccessible locations and enable people to travel around comfortably (protected from bad weather, excessive fatigue, etc.). However, because of its ubiquity, concentration and bulk, the automobile is now being associated with some negative effects in certain places: slower speeds, time lost, high costs for individuals (fuel consumption, vehicle maintenance, need to replace old vehicles, road costs) and for communities (infrastructure maintenance, security and control problems) – not to mention problems relating to inaccessibility, stress or physical inactivity. It has become obvious that the automobile cannot become a universal means of transportation, one that will relegate all other alternatives to the odd, occasional use. In general, cars consume a great deal of space: because they are parked in lots or haphazardly across the landscape; because more people are affected by the congestion they create than actually use them; because their network is extremely dense
Conclusion: the Second Automobile Revolution
449
and requires an enormous amount of service and safety space, etc. Few doubt the need to limit automobile use today. The problems associated with this particular means of transportation can be more or less remedied depending on which alternative solutions prevail in the future. Some of its inconveniences are permanent, however, including physical and material accidents or the destruction of spaces that were not originally meant for automobile use or were incompatible with this. Using fuel sources other than oil to run internal combustion engines (natural gas, biofuels) will reduce some kinds of pollution (CO2 , greenhouse gas, dust, etc.) without eliminating them entirely. Above all, it will leave other variants unscathed (noise, odours, land pollution) and create several new problems. Securing oil supplies will become geo-politically less crucial if oil is replaced by natural gas, a resource distributed more widely across the world. If oil is totally or partially replaced by biofuels, a multitude of production branches will take shape, reflecting different regions’ agricultural specialisations (sugar cane, beetroot, corn, wood, etc.). Of course, it is also true that some of these branches offer poorer energy performance and compete with food production. Up until now, only cars with 100 per cent electrical engines were considered capable of eliminating all pollution (aside from any unrecycled materials and liquids). Moreover, where these cars use electricity produced from renewable energy sources or nuclear power, they can alleviate many of today’s supplyrelated political tensions. Of course, reliance on electrical sources is associated with serious and as yet unresolved problems of public safety and the storage of radioactive waste. A safe electrical car, i.e. one equipped with collision avoidance systems, might reduce the third category of problems that automobiles cause – physical and material accidents. Last but not least, automobiles consume and destroy spaces. The consumption of space dedicated to automobile services and safety should lessen slightly if future cars are powered by battery-stored electricity (although not if their electricity is produced by hydrogen fuel cells). Silent clean cars would create less havoc in spaces that were not originally intended for automobile use. Even so, more and more space will have to be used for traffic and parking purposes. It is not clear how the cars of the second automobile revolution will escape usage limitations compared to more collective or individual modes of transportation. Cars will especially struggle as their diffusion is facilitated by an alleviation of the nuisances they generate, and by a semi-unlimited availability of new energy sources. Indeed, cheap cars with accessible energy sources are a precondition for the automobilisation of the BRICs plus the other countries that will develop economically in their wake. The explosive rise in these regions’ automobile stocks cannot continue, however, if oil prices keep rising, as seems probable today, despite smaller or larger decline due to varying political
450 The Second Automobile Revolution
circumstances, financial manipulations and recessions. Limitations on the automobile’s diffusion – a probable scenario due to everything happening in the oil market – is the most certain catalyst for tomorrow’s clean(er) cars. The upheavals that the automobile industry will face due to these changes in energy and driving systems depend on which standards prevail in the end. If the changes only involve fuel types or hybrid petrol/electricity engines – technologies with which we are already familiar – they will not be too drastic. The changes will be more radical, however, if electricity is the only future source of fuel. In any event, new actors are bound to arise, ones who will try to appropriate a significant proportion of the value produced in this sector and maybe even control its distribution: producers of new fuels, electricity producers, battery makers, fuel cell manufacturers, manufacturers of new driving system or companies that design and produce the new electronic systems’ power, control and regulation functions. There is another possible consequence if electric driving systems ultimately prevail – a new profit strategy might emerge, combining commonalisation and innovation. As discussed in Chapter 2, this would be based on a real and not just metaphorical modularisation of vehicles. Such a strategy, which is technically possible today, would be appropriate for regions characterised by the cohabitation of a ‘coordinated and moderately hierarchised’ with a ‘competitive’ mode of national income distribution.
The uncertainties of the second automobile revolution This second industrial revolution is marked by considerable uncertainty relating to factors such as its speed of implementation, the stages it will go through, the future technical standards, actors, industrial structures and geography, the geo-politics of energy and its impact on mobility and on living conditions in general. The speed with which the revolution will unfold (clean cars and automobilisation of populations in continent-sized countries) clearly depends on how quickly the technical problems that continue to affect the different possible solutions can be resolved. The sooner people sense how urgent it is to overcome these problems, the quicker things will change, notably as real oil prices rise and supply problems worsen. Since 2006, a sense of urgency has mobilised carmakers, although no one knows how long this new frame of mind will dominate, notably amongst the legacy manufacturers, if the oil prices remain low because of the crisis. The deep problems currently besetting the American Big Three, all of whom were guilty of placing too much faith in light trucks, show that the warning signs are finally being taken seriously. Design teams have been formed and will probably no longer be dissolved, as they were in the past, even if oil prices were to fall back. Above all, new actors have appeared, ones determined to push their advantage (as energy producers or as suppliers) or overcome the inevitable constraints they will face (the Chinese and Indian carmakers).
Conclusion: the Second Automobile Revolution
451
Will progress occur step-by-step? On a quantitative level, will the first stage be an improvement in the environmental performance of today’s engines, leading to petrol/electricity hybrids to be followed by biofuels or battery-run electrical vehicles and culminating in fuel-cell electrical vehicles? Will each of these stages lend itself to the technical and economic preparation of its successor? There are no certainties at any of these levels. For the moment, all of these solutions are being presented as if they were contradictory. Partisans of improving current engines’ environmental performance affirm that hybrid engines are barely better than the status quo, not conducive to any great improvements and ultimately too expensive to be widely applied. Fans of hybrid engines doubt that it is possible, within the foreseeable future, to develop battery-run electrical vehicles with sufficient range to be used for anything other than local travel. Promoters of electrical cars like the fact that they are based on batteries being placed on existing models, meaning that there is no need for a completely redesigned automobile architecture. They also highlight the potential for considerable progress in terms of the batteries’ bulk, weight, power, safety and charging speed. Different driving systems evoke different problems, and this impedes their mutual resolution. Until now, the preference for one path as opposed to another seems to have been dictated more by the particular profit strategy that a carmaker was pursuing than by the search for an optimal technical solution. In 1998, for example, trying to leverage a ‘volume and diversity’ strategy to become a big player, PSA opted to improve the environmental performance of its existing diesel and petrol engines and decided to abandon further research into electrical cars. The risk it took was that it would no longer be viewed as an environmentally friendly carmaker, despite its product range’s excellent performance in this area. Toyota, on the other hand, chose a hybrid driving system for just one model while continuing to produce several higher emission vehicles. The reason for this decision was that its ‘cost reduction with constant volumes’ strategy involved minimising risks while progressively reducing the costs associated with the particular solution it had chosen, one that would make the company appear both innovative and environmentally friendly, helping it to benefit from government assistance and receive free publicity because of its new emblematic status. As usual, Honda tried to be really innovative by moving full scale into the fuel cell market from the very outset. As for Nissan and Renault, they decided to support the new generation of batteries and push cars with a medium-distance driving range, to be sold in less densely populated areas. What we now have is a fierce battle about standards. Countries and carmakers are starting to promote whichever solution suits their own purposes. The question is whether this will culminate in the coexistence of solutions reflecting the many different ways in which automobiles are used; the preference of each geographic region; or the degree of influence exercised by one or the other major automobile companies. Another possibility is the formation
452 The Second Automobile Revolution
of a new coalition, like the one between carmakers and oil companies that led to the triumph of the internal combustion engine/petrol standard at the start of the twentieth century, an outcome that turned the automobile into a product that could be exported, diffused and used across the world. The future remains uncertain and we wait to see whether the producers of biofuels, electricity or hydrogen will one day play the role that oil companies play today. The actors who will be well placed in the future will be those able to combine in a coherent architecture the automobile’s various (new and old) characteristics with a universally accessible and relatively inexpensive source of energy, and who can sharply reduce retail prices in a homogeneous and relatively extensive market characterised by solvent customers, abundant governmental assistance and very low production costs. These conditions might exist for new Chinese or Indian entrants; energy producers; major suppliers; or even the legacy carmakers. China and India feature a number of major industrial groups and possess financial resources, transferred technology, competent engineers, a trained labour force, competitive wages, subcontractor networks, vast consumer markets and a desire for autonomy and political revenge – all factors necessary to build up their own automobile industries. The transition to new driving systems might be the opportunity and the means for this to happen. Some electricity producers have been collaborating with automakers for years now. Solidly established and possessing significant financial reserves, they too could go on the attack. Michelin’s supply units have been testing automobile prototypes featuring tyres equipped with electrical engines that can be managed electronically. By supplying the tyres, they are supplying a driving system and therefore, conceivably one day, an entire vehicle. Today’s automobile electronic system designers might be tomorrow’s integrators. There are even some aircraft manufacturers (Dassault) and battery producers (Bolloré) who have launched their own electrical cars. Of course, the legacy manufacturers still have certain cards to play but their current profit strategies could be a handicap. The financial risk that is part of any major innovation is not really a feature of the ‘volume and diversity’ strategies that General Motors, Ford, Fiat, Volkswagen and PSA are pursuing, or of Toyota’s ‘cost reduction with constant volume’ strategy. These players may be inclined to wait until someone else assumes the risk for them and then copy whatever solution works out. Carmakers who owe their profits to conceptual innovation and productive flexibility, like Honda and Renault (or Chrysler in the past), might become increasingly entrepreneurial as long as they do not forget that they will need to simultaneously design a radically new automobile and find a customer base that is ready to adopt it right away and make it a commercial success. Specialist carmakers might also fulfil their traditional technical innovators’ role although it is not at all certain that their top-of-the-range customers will find the social signals they desire in these companies’ new driving systems.
Conclusion: the Second Automobile Revolution
453
And as regards environmental attitudes, it is not at all certain that social climbers focused on ostentatious material status are particularly concerned with fuel savings and reduced pollution. Carmakers that tend to pursue hesitant strategies (like Nissan and Mitsubishi) or else whose strategies need to be redefined (like Chrysler) might take advantage of this situation to develop a new strategy, although not all of them possess the financial and human wherewithal to do so. The history of the automobile has taught us that the optimal solution is not necessarily the one that dominates; that the leading inventors are not necessarily those people who have invested the most in R&D or those producers who know how to make profits from their inventions; that these producers can also be new entrants; and lastly, that the country (countries) of origin of those producers who are able to impose their particular standards is (are) not necessarily the market leader(s) in production and diffusion terms. The second main uncertainty relates specifically to the extension of automobile demand in continent-sized countries or in others that will follow in their wake. These markets will only continue to develop if their growth sources continue to perform and if their mode of national income distribution becomes less unequal (unless the mode they adopt is ‘coordinated and moderately hierarchised’). Foreign direct investment has played an important role up until now but it is not at all certain that this will continue at such a high level, as the 2008 crisis seems to confirm. Similarly, there is nothing spontaneous about the transition to a less unequal national income distribution, when this happens. History has taught us that the social segments that appear during a population’s development phase are generally not happy to relinquish any of their acquired privileges to help those social segments that have been unable to enjoy increased purchasing power. Offering low-cost vehicles will not help the working classes of developing countries to access new vehicles. The low-cost models currently being designed will target the new countries’ middle classes, as Henry Ford’s historical experience reminds us. What this means is that the choices of the future will be political in nature and involve building social compromises. Translation by Alan Sitkin. Acknowledgement to Richard Senter for his useful comments. Note 1. Historical truth and conceptual rigour require us to call this process not Fordism but Sloanism. This is because it was Alfred Sloan and not Henry Ford who understood that an increase in real wages could only be achieved by recognising employee unions and contractualising wage increases in exchange for getting workers to accept the kinds of work organisation that could enable a continuous growth in productivity. Ford may have been the first to pay workers high salaries (initially to retain those who refused assembly line work) – and had long theorised that this policy is an integral part of the mass production system – but he naively
454 The Second Automobile Revolution believed that he could convince his fellow industrialists of the soundness of his thinking. Ford was never going to recognise unions or state interventionism and would not countenance any weakening of employers’ power to do as they see fit. He was incapable of thinking in terms of the social compromise that is necessary to launch the kind of self-sustaining process in which mass consumption leads to mass production.
Bibliography Bardou, J.-P., Chanaron, J.-J., Fridenson, P. and Laux, J. (1982) The Automobile Revolution: the Impact of an Industry. Chapel Hill: University of North Carolina Press. Boyer, R. and Freyssenet, M. (2002) The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave Macmillan. Freyssenet, M. (2004, up to date 2008) ‘Worldwide Automobile Production, 1898– 2007’. Website: freyssenet.com.
Appendix: the GERPISA International Network The GERPISA (the Permanent Group for the Study of and Research into the Automobile Industry and its Employees) started out as a network of French economics, management, history and sociology researchers who were interested in the automobile industry. Founded by Michel Freyssenet (CNRS sociologist) and Patrick Fridenson (EHESS historian), it was transformed into an international network in 1992 in order to carry out a research programme on the ‘Emergence of New Industrial Models’. With Robert Boyer (CEPREMAP, CNRS, EHESS economist) and Michel Freyssenet supervising its scientific orientations and under the management of an international committee, the programme (which lasted from 1993 to 1996) made it possible, thanks to its study of the automobile firms’ (and their transplants’) trajectories, productive organisation and employment relationships, to demonstrate that lean production, which according to the authors of The Machine that Changed the World (Womack, Jones and Roos) was supposed to become the industrial model of the twenty-first century, was in fact an inaccurate amalgamation of two completely different productive models, the ‘Toyotian’ and the ‘Hondian’. Moreover, it showed that there are, have always been, and probably always will be several productive models that are capable of performing well at any one time. Shareholders, executives and employees are not only not obliged to adopt a one best way, they have to devise a ‘company governance compromise’ covering the means that will allow them to implement one of the several profit strategies that are relevant to the economic and social environment in which they find themselves. A second programme (running from 1997 to1999) entitled ‘The Automobile Industry, between Globalization and Regionalization’ and coordinated by Michel Freyssenet and by Yannick Lung (Bordeaux IV, economist), tested the analytical framework that had been developed during the first programme in an attempt to better understand the new wave of car manufacturer and component-maker internationalisation that had been observed over the previous decade. The outcome was that the viability of the choices being made depends primarily on the chosen profit strategies’ compatibility with the growth modes in the areas in which the investments are being made. The third programme (2000–2002) was developed under Yannick Lung’s coordination. It focuses on the issues at stake in the ‘Coordination of Knowledge and Competences in the Regional Automotive Systems’. Supplementing existing studies of forms of regionalisation in the automobile industry, the programme analysed the sector’s new contours as well as the development of new relational and cooperative modes amongst its actors. The fourth programme (2003–2006), entitled ‘Variety of Capitalisms and Diversity of Productive Models’ and coordinated by Bruno Amable (CNRS, Paris) and Yannick Lung, discussed the thesis of the national productive model. It showed the diversity of productive models in each type of capitalism, exploring the interaction between the micro, meso and macro levels. It emphasised the importance of the political compromises at meso and macro levels to manage the competition conditions and the employment conditions. The fifth programme (2007–2010) is developed under the direction of Bernard Jullien (Bordeaux IV). Its topic is ‘Sustainable Development and the Automobile Industry’. 455
456 Appendix It aims to analyse the emergence of this theme and to compare discourses and practices of the automobile enterprises. In 2008, GERPISA counted 350 members from 27 different countries. Affiliated with the Centre de Recherches Historiques (CRH) of the Ecole des Hautes Etudes en Sciences Sociales (EHESS) and acknowledged as a host structure by the French Ministry of National Education, its administrative offices are located in the Université d’Evry. It receives additional financial and material support from the French car companies, from their professional association (the CCFA), and from the European Union. The international steering committee comprises the following members: Robert Boyer (CNRS-EHESS, Paris), Jorge Carrillo (Colegio de la Frontera Norte, Mexico), Jean-Jacques Chanaron (CNRS, Lyon), Elsie Charron (CNRS, Paris), Michel Freyssenet (CNRS, Paris), Patrick Fridenson (EHESS, Paris), Takahiro Fujimoto (University of Tokyo), Olivier Hirt (École des Mines, Paris), Bernard Jullien (Université de Bordeaux IV), Bruno Jetin (Université Paris XIII), Ulrich Jürgens (WZB, Berlin), Yveline Lecler (MRASH/IAO, Lyon), Yannick Lung (Université de Bordeaux IV), Tommaso Pardi (EHESS, Paris), Sigfrido Ramirez (Institut Européen, Florence), Mario Sergio Salerno (University of São Paolo), Koichi Shimizu (University of Okayama), Koichi Shimokawa (Hosei University, Tokyo), Paul Stewart (Cardiff University), Giuseppe Volpato (Ca’Foscari University of Venise), and Karel Williams (Victoria University, Manchester).
GERPISA’s publications GERPISA edits in English and in French a quarterly review entitled Actes du GERPISA and a monthly newsletter called La Lettre du GERPISA. The review combines the writings that the network’s members have presented on a specific topic in various work meetings. The newsletter comments upon news from the automotive world and provides up-to-date information on what is happening in the network. Findings from the first and second programmes have been published in a series of books:
Programme: ‘Emergence of New Industrial Models’ Boyer, R., Charron, E., Jürgens, U. and Tolliday, S. (eds), Between Imitation and Innovation: the Transfer and Hybridization of Productive Models in the International Automobile Industry. Oxford and New York: Oxford University Press, 1998. Freyssenet, M., Mair, A., Shimizu, K. and Volpato, G. (eds), One Best Way? Trajectories and Industrial Models of the World’s Automobile Producers. Oxford and New York: Oxford University Press, 1998. French translation: Quel modèle productif? Trajectoires et modèles industriels des constructeurs automobiles mondiaux. Paris: La Découverte, 2000. Durand, J. P., Stewart, P. and Castillo, J. J. (eds), Teamwork in the Automobile Industry: Radical Change or Passing Fashion. Basingstoke: Macmillan, 1999. First edition in French: L’avenir du travail à la chaîne. Paris: La Découverte, 1998. Lung, Y., Chanaron, J. J., Fujimoto, T. and Raff, D. (eds), Coping with Variety: Product Variety and Production Organization in the World Automobile Industry. Aldershot: Ashgate, 1999. Shimizu, K., Le Toyotisme. Paris: La Découverte, 1999. Boyer, R. and Freyssenet, M., The Productive Models: the Conditions of Profitability. Basingstoke and New York: Palgrave 2002. First edition in French: Les modèles productifs. Paris: La Découverte, 2000. Other editions: Los modelos productivos. Buenos Aires,
Appendix
457
Mexico: Lumen Humanitas, 2001. Produktionmodelle, Ein e Typologie am Beispiel der Automobilindustrie. Berlin: Edition Sigma, 2003. Los modelos productivos. Madrid: Editorial Fundamentos, 2003. Oltre Toyota. I nuovi modelli produttivi. EGEA, Università Bocconi Editore, Milano, 2005. Boyer, R. and Freyssenet, M., Le Monde qui a changé la machine, un siècle d’histoire automobile, 1999, 2006, http://freyssenet.com
Programme: ‘The Automobile Industry between Globalization and Regionalization’ Humphrey, J., Lecler, Y. and Salerno, M. (eds), Global Strategies and Local Realities: the Auto Industry in Emerging Markets. Basingstoke and New York: Palgrave Macmillan and St. Martin’s Press, 2000. Freyssenet, M., Shimizu, K. and Volpato, G. (eds), Globalization or Regionalization of the American and Asian Car Industry? Basingstoke and New York: Palgrave Macmillan, 2003. Freyssenet, M., Shimizu, K. and Volpato, G. (eds), Globalization or Regionalization of the European Car Industry? Basingstoke and New York: Palgrave Macmillan, 2003. Carillo, J., Lung, Y. and van Tulder, R. (eds) Cars . . . Carriers of Regionalism? Basingstoke and New York: Palgrave Macmillan, 2004. Charron, E. and Stewart, P. (eds), Work and Employment Relations in the Automobile Industry. Basingstoke and New York: Palgrave Macmillan, 2004.
Programme: ‘Coordination of Knowledge and Competences in the Regional Automotive Systems’ Froud, F., Johal, S. and Williams, K. (eds), ‘The Tyranny of Finance? New Agendas for Auto Research’, Competition and Change, 6, double issue 1/2 (2002). Lung, Y. (ed.), ‘The Changing Geography of the Automobile Industry’, Symposium. International Journal of Urban and Regional Research, 4 (2002). Lung, Y. and Volpato, G. (eds), ‘Reconfuring the Auto Industry’, International Journal of Automotive Technology and Management, 2(1) (2002). Calabrese, G. and Lung, Y. (eds), ‘Designing Organisations to Manage Knowledge Creation and Coordination’, Special issue of International Journal of Automotive Technology and Management, 1/2 (2003).
Programme: ‘Variety of Capitalisms and Diversity of Productive Models’ Hirt, O. (ed.), ‘Variety of Capitalisms and Diversity of Productive Models’, Actes du GERPISA, 38 (2005). Pardi, T. (ed.), ‘State and Politics in the Automobile Industry’, Actes du GERPISA, 40 (2006). Freyssenet, M., ‘Stratégies et modèles nationaux de croissance. Proposition d’une démarche et esquisse d’iun schéma d’analyse’, Revue de la régulation, 3 (2008). Jullien, B. and Smith, A., Industries and Globalization: the Political Causality of Differences. Basingstoke and New York: Palgrave Macmillan, 2008. Lung, Y., ‘Modèles de firmes et formes de capitalismes’, Revue de la régulation, 2 (2008).
458 Appendix
Programme: ‘Sustainable Development and the Automobile Industry’ Jullien, B., ‘A Framework of Sustainable Development Issues for the Automobile Industry’, International Journal of Automotive Industry and Management, 1(1) (2007): 1–19. Freyssenet, M. (ed.), The Second Automobile Revolution: Trajectories of the World Carmakers at the Beginning of the 21st Century. Basingstoke and New York: Palgrave Macmillan, 2009.
Information on GERPISA’s activities can be obtained by contacting GERPISA réseau international. Université d’Evry-Val d’Essonne, rue du facteur Cheval, 91025, Evry cedex, France. Telephone: 33 (1) 69 47 70 23 - Fax 33 (1) 69.47.80.35. E-mail: [email protected]. Website: http//www.gerpisa.univ-evry.fr
Index Agnelli, G., 290, 291, 292, 293 Aisin Seiki, 425, 427, 429, 431, 434, 437, 438, 439 Alfa Romeo, 289, 290, 291, 295, 298, 301, 302, 304, 306 alliance–merger–takeover, 1, 3, 4, 9, 17, 19, 20, 21, 25, 27, 39, 40, 85, 86, 91, 95, 96, 98, 100, 101, 103–9, 123, 139, 142, 143, 144, 167, 168, 174, 176, 179, 186, 187, 190, 199, 200, 204, 205, 207, 208, 209, 210, 211, 212, 213, 216, 217, 218, 228, 229, 231, 237, 238, 239, 241, 246, 247, 249, 254, 255, 256, 258, 263, 267, 268, 269, 270, 273–7, 279, 280, 282, 283, 290, 303, 309–14, 316, 317, 321–7, 330, 333, 336, 337, 339, 353, 354, 362, 366, 367, 368, 369, 370, 372, 373, 385, 387, 390, 396, 408, 409, 410, 411, 413, 424, 425, 433, 440 Aoki, M., 419, 422, 440 Aston Martin, 4, 16, 21, 44, 187, 192, 195, 198, 204, 354 Audi, 18, 227, 229, 231, 232, 233, 234, 241, 243, 302, 332, 336, 349, 389 Avtoframos, 275, 281 Avtovaz, 41, 282, 283
Bosch, 155, 425, 426, 429, 430, 432, 434, 435, 437, 438, 439 Boyer, R., 5, 35, 65, 94, 183, 244, 265, 286, 307, 352, 378, 402, 417, 440, 454, 456, 457 BRIC (Brazil, Russia, India, China), 19, 30, 446 British Leyland, 204, 366, 367, 368, 378 British Motor Company, 366 Broustail, J., 252, 265 Bugatti, 16, 231, 238 Buick, 175 Cadillac, 175, 177, 200 Calvet, J., 248, 249, 250, 251, 252, 253, 254, 255, 257, 265 Camuffo, A., 306, 307 Cantarella, P., 290, 291 Carillo, J., 2, 5, 35, 36, 37, 65, 434, 441, 457 Castillo, J. J., 5, 265, 456 Castronovo, V, 306, 307 Cerberus Capital Management, 168, 170, 216–19, 221, 222, 309, 323, 324, 325, 326 Chana Automobile, 41 Chanaron, J.-J., 6, 36, 250, 251, 420, 440, 454, 456 Charron, E., 2, 5, 22, 35, 139, 271, 286, 352, 402, 456, 457 Chery, 19, 118, 218, 294, 325, 383, 388, 389, 397, 398 Chevrolet, 175 Chrysler, 15, 17, 18, 25, 26, 35, 40, 44, 45, 49, 85, 100, 167, 169, 170, 171, 174, 177, 183, 185, 186, 188, 194, 200, 201, 203, 206, 222, 246, 249, 262, 309–31, 368, 432, 452, 453 Chung, M.-K., 161 Church, R., 368, 378 Citroën, 246, 272, 306, 388, 392 Coffey, D., 378 Cray, E., 166, 183
Bajaj, 20, 281 Balcet, G., 306, 307 Becker-Ritterspach, F., 417 Beijing Automotive Industry Corporation (BAIC), 316 Belzowski, B., 15, 25, 35, 167, 183, 206, 311, 331 Bentley, 16, 231 Berggren, C., 16, 35, 354, 357, 364, 365 Bernhard, W., 212, 213, 240, 241, 313, 324, 337 BMW, 16, 17, 35, 40, 41, 42, 44, 46, 48, 49, 117, 127, 181, 200, 258, 263, 325, 327, 330, 332, 356, 362, 364, 366, 369–78, 385, 387 459
460 Index crisis financial, 95, 100, 141, 142, 151, 159, 214, 269, 311, 333 firm, 13, 17, 25, 27, 39, 50–5, 57–9, 95, 98, 100, 101, 108, 109, 117, 121, 125, 142, 165, 167, 168, 171, 172, 176, 188, 192, 193, 195, 198, 204, 215, 233, 237, 246, 248, 250, 261, 269, 272, 275, 278, 287, 291, 293, 300, 310, 311, 313, 314, 319, 320, 321, 323, 330, 333, 337, 343, 374, 377, 398, 401, 429, 430, 440 regional, 12, 21, 51, 52, 53, 100, 108, 151, 167, 186, 188, 201, 206, 229, 256, 262, 271, 272, 273, 275, 338, 354, 367 world, 7, 26, 98, 115, 122, 167, 186, 227, 228, 246, 291, 310, 311, 444 Cusumano, M. A., 96, 97, 98, 99, 111, 132, 140 Cutcher-Gershenfeld, J., 358, 365 Dacia, 4, 15, 40, 101, 111, 267, 272, 273, 275, 280, 281, 283, 285, 286 Daewoo, 19, 40, 176, 276, 288, 394, 395, 396 DAF, 43, 367 Daihatsu, 40, 84, 85, 86, 91, 93, 392 Daimler, 4, 13, 17, 25, 40, 44, 45, 46, 49, 100, 181, 186, 200, 206–13, 216–18, 309, 333 DaimlerChrysler, 4, 21, 42, 44, 45, 46, 48, 49, 64, 169, 174, 207–19, 220, 240, 309, 385, 387, 396, 426, 432 Dana, 425, 426, 429, 430, 431, 432, 434, 438, 439, 440 Delphi, 25, 63, 172, 173, 191, 203, 204, 324, 425, 426, 427, 429, 430, 431, 432, 434, 435, 437, 438, 439, 440 Denso, 27, 84, 425, 426, 427, 429, 430, 432, 434, 435, 437, 438, 439 diversification of activities, 2, 10, 14, 15, 38, 45, 58, 75, 123, 143, 200, 208, 310, 311, 317, 318, 337, 408, 411, 421, 422, 425 Dodge, 206, 215, 216, 217, 221, 319 Dongfeng, 254, 255, 262, 384, 385, 388, 390, 392 Durand, J. P., 1, 5, 265, 456
Eaton, R., 207, 212, 221, 310, 312, 313, 424, 425, 440 Eckardt, A., 17, 35, 319, 331, 335, 337, 350, 352, 369, 372, 378 Ellegard, K., 357, 365 employment relationship career, 9, 10, 14, 97, 136, 150, 191, 257, 332, 366 general, 13, 22, 23, 29, 31, 277, 455 job cuts, 146, 151, 152, 161, 320, 321, 322, 324, 325 job security, 9, 148, 152, 160, 198, 406 overtime, 12, 72, 78, 79, 80, 81, 235, 399 purchasing power, 25, 280, 324, 376, 401, 453 recruitment, 236, 257, 261, 332, 340, 342, 347, 391, 422 engineers, 71, 74, 86, 87, 98, 113, 117, 124, 125, 133, 209, 210, 213, 219, 252, 261, 275, 279, 301, 317, 332, 335, 347, 349, 391, 394, 395, 396, 411, 452 engines clean, 33, 325 combustion, 5, 27, 29, 85, 108, 114, 121, 443, 447, 448, 449, 452 diesel, 86, 121, 123, 144, 217, 249, 257, 258, 259, 260, 274, 303, 314, 320, 336, 411, 412, 451 electrical, 27, 38, 84, 449, 451, 452 Enrietti, A., 306, 307 environmental policies, 123, 173, 257, 260, 304, 349, 445, 449 exportations, 8, 18, 38, 70, 82, 91, 97, 100, 109, 114, 117, 119, 132, 142, 147, 156, 158, 159, 160, 203, 253, 275, 284, 285, 310, 311, 319, 336, 338, 346, 354, 362, 369, 377, 394, 400, 408, 410, 411, 412, 414, 415, 416, 433, 445, 446 Faurecia, 27, 258, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 437, 438, 439 Fiat, 4, 10, 11, 18, 25, 37, 41, 42, 44, 45, 46, 48, 49, 50, 64, 119, 176, 204, 241, 246, 253, 254, 259, 263, 281, 287–307, 385, 387, 407, 412, 426, 427, 452
Index finance activities, 41, 187, 199, 200, 209, 238, 310, 311, 320, 336 bubble, 10, 12, 13, 18, 25, 30, 246, 269, 270, 271, 444, 450 crisis, 95, 100, 141, 142, 151, 159, 214, 311, 333 financialisation, 2, 14, 25, 173, 180, 317, 321, 326, 348, 430 insurance, 24, 25, 170, 172, 183, 290, 422, 448 structure, 161, 226, 326, 407 firms’ players banks, 18, 26, 31, 39, 183, 195, 287, 326, 330, 333, 348, 403 corporate governance, 178, 180, 181, 242, 257, 309, 320, 348, 405 employees, 25, 31, 39, 72, 75, 80, 168, 172, 183, 247, 257, 282, 292, 322, 413 executives, 13, 24, 25, 31, 34, 39, 139, 166, 168, 169, 172, 173, 174, 179, 188, 195, 197, 200, 204, 206, 207, 209, 211, 212, 213, 216, 217, 218, 249, 269, 282, 407, 455 general, 16, 20, 22, 25, 39, 324, 326, 367, 368, 373, 374, 384, 385, 401, 404, 407, 410, 415, 424, 452 owners, 26, 40, 117, 139, 167, 169, 187, 193, 216, 219, 225, 226, 238, 242, 324, 326, 327, 364, 367, 375, 389, 407, 409, 410, 413, 414 shareholders, 9, 13, 18, 24, 25, 31, 33, 39, 57, 167, 183, 196, 216, 217, 219, 239, 269, 272, 289, 314, 316, 326, 327, 332, 333, 348, 407, 410, 429, 430 stakeholders, 312, 348 state, 226, 227, 228, 229, 240, 255, 269, 270, 367, 407, 408, 409, 410, 414 UAW (United Automobile Workers), 170, 171, 172, 194, 196, 320, 321, 322, 324, 340 unions, 31, 39, 74, 75, 77, 79, 80, 81, 87, 95, 96, 99, 100, 107, 109, 141, 146, 151, 152, 153, 160, 170, 171, 172, 194, 196, 213, 226, 227, 228, 229, 233, 235, 237, 242, 255, 257, 268, 313, 320, 321, 322, 323, 324,
461
325, 326, 330, 340, 342, 343, 348, 372, 374, 375, 376, 406, 409, 414, 453, 454 firms’ internationalisation foreign investments, 22, 53, 54, 55, 57, 119 joint ventures, 19, 82, 84, 132, 139, 158, 159, 176, 214, 233, 254, 264, 316, 319, 346, 383, 384, 385, 386, 388, 392, 393, 394, 405 licences, 119, 390, 407 local integration rate, 26, 254, 269, 407 transplants, 8, 12, 152, 156, 158, 174, 339, 340, 375, 401, 403, 455 First Auto Works, 41, 384, 385, 388, 389, 392, 393 Fligstein, N., 180, 183 Folz, J. M., 255, 256, 257, 260, 261, 262 Ford, 4, 10, 13, 15, 16, 17, 20, 21, 25, 26, 35, 38, 40, 42, 43, 44, 48, 49, 50, 53, 54, 55, 63, 64, 83, 96, 97, 100, 101, 119, 129, 130–40, 169, 170, 171, 174, 177, 183, 185–205, 207, 214, 215, 218, 225, 226, 228, 238, 258, 263, 266, 281, 294, 297, 303, 306, 315, 319, 325, 330, 333, 353, 354, 355, 356, 357, 359, 360, 362, 363, 364, 365, 367, 368, 369, 370, 373, 374, 384, 385, 387, 407, 426, 427, 428, 432, 452, 453, 454 Ford, H., 453 Ford, W. C., 193 Fordism, 9, 141, 148, 165, 166, 175, 185, 227, 306, 444, 453 Freeland, R. F., 179, 184 Freightliner, 309, 316 Freyssenet, M., 1, 2, 4, 5, 7, 10, 11, 13, 15, 16, 20, 21, 23, 31, 32, 35, 36, 37, 38, 39, 65, 70, 75, 94, 111, 127, 128, 161, 166, 183, 184, 185, 205, 225, 228, 244, 246, 249, 252, 260, 265, 266, 267, 268, 269, 270, 286, 306, 307, 308, 331, 350, 352, 365, 366, 375, 378, 384, 397, 402, 403, 404, 417, 432, 433, 440, 441, 443, 444, 446, 454, 455, 456, 457, 458 Fridenson, P., 454, 455, 456 Froud, J., 374, 378, 426, 441, 457 Fucci, J. J., 139, 140
462 Index Fujimoto, T., 6, 13, 36, 78, 95, 96, 97, 98, 101, 102, 103, 106, 111, 131, 135, 140, 281, 358, 364, 365, 392, 403, 420, 421, 423, 440, 442, 456 Fujisawa, T., 113, 122 Fukui T., 112, 113, 121, 126 Garuzzo, G., 306, 307 Geely, 4, 19, 118, 383, General Motors, 4, 15, 16, 17, 18, 20, 21, 25, 26, 28, 34, 35, 42, 44, 69, 96, 165–83, 185, 193, 199, 207, 215, 218, 219, 221, 229, 249, 263, 281, 282, 293, 322, 385, 407, 426, 452 GERPISA, 1, 2, 7, 10, 31, 32, 36, 94, 162, 225, 286, 397, 403, 441, 442, 455, 456, 457, 458 Gettelfinger, R., 321, 324 Ghidella, V., 287, 292 Ghosn, C., 23, 27, 101, 102, 106, 111, 136, 270, 273, 278, 279, 282, 286, 302 GMAC, 168, 169, 219, 324 Greggio, R., 252, 265 Gyllenhammar, P., 354, 357 Hanada, M., 10, 36, 95, 96, 97, 99, 111 Hartz, P., 234, 240, 244 Heller, D., 111, 140 Helper, S., 419, 441 Hertz, 187, 191, 194, 198, 200, 362, 365 Hindustan Motors, 405 Holweg, M., 365, 378 Honda, 1, 10, 12, 13, 15, 16, 19, 21, 25, 26, 36, 37, 40, 42, 45, 47, 48, 49, 50, 56, 59, 60, 62, 64, 112–26, 128, 132, 200, 203, 207, 330, 336, 340, 350, 366, 367, 368, 369, 370, 371, 378, 384, 385, 388, 407, 427, 451, 452 Honda, S., 112, 113, 116, 121, 125, 127 Humphrey, J., 2, 6, 22, 36, 184, 434, 441, 457 Hyundai, 40, 41, 42, 48, 49, 60, 61, 118, 141–61, 162, 208, 213, 214, 288, 309, 313, 314, 316, 324, 325, 327, 388, 407, 410, 415, 425, 427 Iacocca, L., 311 Imaki, H., 131, 136, 137 international public policies
exchange rate, 8, 11, 12, 26, 58, 64, 132, 156, 262, 337, 354, 444 protectionism, 33, 47, 103, 146, 156, 368, 383, 386, 390, 405, 406 Iran Khodro, 262 Isuzu, 40, 42, 45, 47, 48, 49, 57, 58, 64, 86, 102, 176, 178 Iveco, 296 Jaguar, 4, 16, 17, 20, 21, 40, 44, 139, 187, 190, 192, 195, 196, 198, 204, 205, 354, 359, 364, 366, 367, 373, 374, 377 Jeep, 188, 203, 206, 215, 216, 217, 221, 311, 315 Jetin, B., 19, 21, 36, 38, 43, 48, 64, 65, 101, 248, 253, 261, 265, 266, 456 Johnson Controls, 425, 426, 429, 432, 435, 437, 438, 439 Jullien, B., 14, 36, 280, 286, 455, 456, 457, 458 Jun, H., 92, 275, 286 Jürgens, U., 5, 16, 21, 25, 35, 36, 225, 233, 244, 402, 456 Kawamata, K., 99 Kawamoto, N., 112, 113, 115, 116 Kerkorian, K., 167, 168, 169, 180, 282 Kia, 40, 142, 143, 144, 147, 288, 427 Klemm, M., 17, 35, 321, 337, 350, 352, 369, 372, 378 Köhler, H.-D., 17, 25, 309, 322, 326, 330, 331, 352 Kuhlmann, U.-W., 244, 352 Kumon, H., 13, 36, 100, 111 labour bargaining, 24, 75, 151, 152, 156, 270, 271, 272, 274, 321, 322, 340, 344, 375, 413 division, 74, 86, 149, 420, 433 general, 10, 11, 12, 13, 19, 70, 71, 72, 74, 75, 76, 80, 87, 88, 95, 96, 99, 107, 114, 117, 126, 139, 141, 146, 148, 149, 151, 152, 153, 159, 160, 165, 170, 171, 182, 194, 196, 201, 220, 225, 226, 227, 229, 234, 236, 237, 251, 255, 257, 277, 320, 321, 322, 327, 330, 341, 342, 343, 344,
Index 345, 347, 348, 349, 357, 358, 359, 397, 406, 409, 413, 414, 416, 452 market, 10, 13, 75, 146, 151, 345 relations, 12, 74, 88, 139, 142, 146, 152, 153, 170, 225, 229, 236, 320, 321, 322, 330, 342, 347, 406 strike, 152, 153, 171, 321, 324, 338, 409, 413, 414 Lamborghini, 231, 311 Lancia, 254, 289, 290, 291, 296, 300, 301, 302, 303, 304, 306 Land Rover, 4, 16, 17, 20, 21, 40, 44, 192, 195, 196, 198, 203, 204, 337, 338, 350, 354, 366, 367, 370, 372, 373, 374, 377 LaSorda, T., 216, 217 Layan, J. F., 423, 433, 434, 441 Lear, 425, 429, 430, 432, 434, 435, 437, 438, 439 Lecler, Y., 6, 36, 419, 441, 456, 457 Lee, Y. H., 144, 162, 311, 393, 403 Lewandowski, J., 334, 352 Lexus, 86, 332, 336 Lincoln, 186, 187, 190, 192, 200, 202 Loubet, J. L., 251, 257, 266, 286 Lung, Y., 1, 5, 6, 16, 20, 35, 36, 37, 38, 39, 65, 177, 183, 266, 421, 441, 455, 456, 457 Lutz, R., 312, 331 Mack, 273 Magna, 169, 323, 425, 426, 429, 430, 431, 432, 434, 435, 438, 439, 440 Mahindra & Mahindra, 281 Mair, A., 5, 10, 13, 35, 36, 37, 94, 111, 114, 127, 161, 183, 184, 205, 244, 265, 266, 286, 307, 331, 352, 365, 368, 370, 371, 375, 377, 378, 403, 456 Marchionne, S., 287, 292, 293, 294, 295, 296, 297, 301, 302, 304, 306 market American, 43, 112, 119, 121, 122, 125, 167, 174, 196, 207, 214, 215, 216, 218, 289, 310, 314, 315, 319, 330, 362, 364, 374 Asian, 119 Brazilian, 301 British, 367
463
Chinese, 19, 50, 51, 157, 201, 255, 356, 383, 385, 400 emerging, 25, 118, 119, 120, 123, 156, 159, 207, 259, 261, 262, 310, 311, 312, 316, 323, 325, 364 European, 11, 19, 116, 122, 213, 253, 271, 302, 312, 336, 362 export, 159, 203, 275, 319, 362, 369 heteregeneous, 7, 14, 15, 17, 20, 156, 259, 423 industrialised countries, 133, 134, 187, 228, 288, 294 Korean, 267, 276, 281 Latin American, 117, 118 mass, 208, 210, 217, 228, 339 national, 33, 264, 433 Northern European, 267 South American, 158, 280, 319 United States, 11, 43, 58, 112, 114, 119, 121, 122, 125, 158, 166, 167, 174, 186, 193, 196, 201, 203, 207, 214, 215, 216, 218, 273, 289, 310, 314, 315, 319, 330, 362, 364, 374 Maserati, 301, 306, 311 Matra, 251, 252 Maybach, 311, 318, 319 Mazda, 4, 13, 20, 42, 45, 47, 48, 49, 50, 51, 55, 57, 58, 64, 75, 85, 101, 129–39, 186, 187, 200, 204, 205, 356, 359, 385, 390 Mercedes, 12, 16, 17, 200, 208, 211, 212, 213, 214, 216, 259, 263, 310, 332, 356, 385, 392 Mercury, 16, 186, 187, 190, 202 MG, 337, 372, 373, 375, 377, 378 Midler, C., 423, 441 Milberg, J., 337, 341 Mitsubishi, 4, 13, 17, 25, 37, 42, 45, 47, 48, 49, 57, 58, 64, 114, 186, 208, 213, 214, 259, 263, 309, 311, 313, 314, 316, 319, 324, 325, 327, 330, 390, 425, 453 model Anglo-Saxon, 24, 25 German, 227, 330 Japanese, 9, 13, 16, 203, 419, Monden, Y., 74, 94 Moulonguet, T., 106, 111 Mulally, A., 195, 201
464 Index Nasser, J., 191, 192, 198, 200, 204, 354 national income distribution, 8, 11, 13, 14, 15, 17, 28, 29, 33, 444, 445, 450, 453 Navistar, 42, 48, 49 Nissan, 4, 10, 13, 15, 20, 21, 23, 27, 36, 40, 42, 45, 47, 48, 49, 50, 56, 57, 59, 60, 64, 75, 84, 95–111, 114, 119, 121, 136, 218, 267, 272, 273, 274, 276, 277, 278, 279, 281, 282, 283, 286, 306, 330, 362, 368, 385, 388, 426, 427, 451, 453 Nobeoka, K., 132, 133, 139, 140 Ohno, T, 69, 72, 74, 80, 81, 94 Opel, 176, 253, 272, 295, 306 Paccar, 42, 43, 48, 49 PAG (Premier Automobile Group), 4, 16, 17, 44, 54, 192, 197, 198, 204, 353, 354, 360 Pardi, T., 12, 36, 402, 403, 456, 457 Pascale, R. T., 125, 127, 129, 130, 131, 140 Pelata, P., 286 period 1920s, 47, 96, 97, 184, 333 1930s, 96, 97 1950s, 1, 63, 81, 85, 116, 226, 227, 333, 336, 366, 445, 446 1960s, 8, 95, 98, 116, 123, 129, 185, 188, 216, 227, 333, 366, 384, 445, 446 1970s, 1, 7, 9, 14, 15, 17, 18, 23, 53, 54, 98, 125, 131, 165, 185, 197, 206, 207, 216, 226, 228, 334, 335, 357, 368, 369, 375, 377, 383, 385, 404, 434, 444 1980s, 10, 11, 16, 26, 51, 52, 54, 70, 71, 72, 75, 82, 87, 98, 139, 174, 183, 187, 189, 197, 203, 206, 208, 228, 229, 237, 246, 247, 249, 250, 267, 287, 288, 289, 310, 311, 327, 333, 334, 335, 339, 342, 367, 368, 374, 375, 377, 406, 408, 419, 421 Peugeot, 82, 246, 295 Piëch, F., 229, 240, 241 Pil, F., 365 Pininfarina, 358 Pischetsrieder, B., 240, 241, 337, 371, 372
Pontiac, 175 Porsche, 25, 225, 234, 241, 242, 327, 332, 348, 349 Powell, W. W., 179, 184 Pries, L., 17, 331, 332, 335, 344, 348, 352 product policy electronisation, 28, 29 commonalisation, 2, 15, 16, 17, 19, 20, 21, 28, 29, 32, 84, 86, 102, 103, 120, 121, 122, 132, 142, 143, 144, 156, 175, 177, 205, 206, 207, 209, 210, 213, 231, 232, 241, 246, 247, 252, 253, 254, 255, 258, 259, 260, 261, 262, 263, 264, 274, 276, 277, 279, 280, 282, 283, 291, 294, 297, 298, 306, 314, 316, 326, 341, 354, 355, 356, 359, 393, 412 general, 2, 15, 22, 26, 31, 95, 107, 125, 225, 226, 227, 228, 240, 241, 251, 263, 267, 279, 406, 411 innovative product, 10, 11, 12, 15, 16, 27, 28, 29, 206, 246, 247, 249, 251, 252, 260, 263, 264, 267, 269, 270, 271, 272, 276, 277, 278, 279, 280, 281, 283, 303, 315, 334, 337, 338, 339, 452 modularisation, 22, 24, 27, 28, 29, 107, 142, 144, 145, 148, 149, 154, 155, 159, 230, 232, 233, 263, 298, 311, 318, 319, 335, 392, 399, 402, 421, 422, 423, 428, 430, 436 production plan, 120, 261, 336, 337, 338, 339, 341, 343, 362, 363, 402, 435 quality control, 74, 88, 147, 148, 150, 357, 359 R&D, 83, 84, 86, 87, 102, 108, 116, 118, 120, 125, 132, 142, 143, 144, 146, 147, 155, 160, 172, 203, 261, 275, 292, 301, 302, 303, 305, 313, 320, 332, 336, 345, 346, 347, 348, 370, 393, 394, 399, 411, 420, 422, 423, 431, 432, 434, 435, 453 services, 12, 15, 24, 25, 114, 129, 138, 151, 167, 168, 170, 171, 172, 183, 187, 191, 198, 199, 214, 254, 275, 279, 290, 296, 297, 310, 311, 320, 386, 422, 448 teamwork, 10, 72, 74, 76, 77, 78, 209, 234, 299, 321, 335, 358
Index production type lean, 2, 7, 9, 10, 12, 30, 146, 150, 153, 186, 230, 232, 358, 455 mass, 71, 79, 141, 148, 185, 392, 393, 453, 454 modular, 24, 145, 148, 149, 153, 159, 421 reflexive, 77, 358, 363 productive model emergence, 2, 7, 9, 17, 30, 38, 40, 269, 288, 368, 372, 373, 383, 384, 444, 456 general, 1, 3, 4, 5, 7, 9, 13, 23, 24, 34, 39, 62, 70, 225, 227, 228, 231, 240, 242, 384, 396, 397, 398, 399, 402, 432, 455 Hondian, 122, 384, 455 Sloanian, 20, 246, 270, 384 Toyotian, 120, 330, 392, 393, 455 productive organisation buffer, 12, 76, 78, 340, 358 externalisation, 7, 24, 26, 27, 32, 277, 421 design, 15, 16, 19, 20, 24, 27, 29, 70, 71, 79, 87, 88, 102, 103, 113, 121, 124, 125, 132, 145, 155, 156, 158, 159, 160, 167, 186, 193, 197, 206, 215, 218, 235, 240, 242, 246, 252, 258, 259, 260, 264, 266, 270, 274, 275, 278, 279, 280, 283, 286, 290, 291, 292, 294, 296, 303, 312, 332, 340, 347, 353, 355, 356, 357, 358, 359, 368, 370, 389, 390, 392, 394, 395, 396, 397, 399, 420, 422, 434, 450, 452 general, 1, 22, 71, 75, 78, 88, 143, 149, 150, 154, 226, 227, 230, 234, 235, 241, 271, 311, 358, 421, 422 integration rate, 26, 254, 269 just-in-time, 69, 75, 76, 78, 98, 233, 261, 280, 335, 370, 371, 372, 422, 423 kaizen, 70, 71, 72, 73, 74, 75, 77, 79, 80, 81, 160 line, 71, 75, 76, 77, 78, 98, 145, 146, 149, 154, 155, 248, 258, 277, 338, 340, 357, 358, 396, 423, 453 logistics, 22, 76, 105, 274, 334, 342, 345, 346, 357, 362, 370, 422
465
machines and tools, 1, 8, 69, 74, 75, 77, 78, 80, 148, 149, 227, 230, 232, 250, 277, 283, 326, 340, 357, 399, 412, 422 maintenance, 10, 24, 117, 149, 150, 268, 314, 335, 404, 444, 448 overtime, 12, 72, 78, 79, 80, 81, 235, 399 productivity, 8, 9, 69, 72, 73, 74, 75, 76, 79, 80, 81, 88, 100, 117, 143, 145, 146, 148, 150, 151, 155, 186, 221, 229, 232, 233, 235, 236, 240, 241, 242, 274, 299, 310, 311, 341, 357, 358, 359, 360, 405, 409, 411, 412, 413, 419, 430, 453 profit foreign countries’ contribution, 51, 52, 53, 57, 58 operating, 55, 56, 57, 58, 59, 60, 61, 63, 64, 65, 101, 135, 216, 323, 350, 351, 376, 377 world, 51, 53, 56, 57, 58, 59 profit sources diversity, 9, 141, 341, 360, 362 flexibility, 13, 15, 29, 31, 32, 88, 120, 122, 141, 146, 148, 151, 159, 209, 219, 257, 258, 261, 264, 277, 321, 334, 335, 339, 340, 341, 342, 345, 347, 349, 356, 371, 399, 405, 452 innovation, 2, 7, 10, 15, 16, 28, 29, 31, 88, 116, 120, 121, 122, 123, 125, 141, 147, 161, 236, 246, 247, 249, 251, 252, 257, 258, 259, 260, 269, 270, 277, 279, 281, 303, 304, 312, 323, 334, 336, 340, 345, 347, 353, 357, 362, 390, 391, 393, 396, 397, 398, 422, 450, 452 quality, 8, 9, 10, 19, 23, 27, 31, 32, 45, 69, 71, 72, 74, 75, 78, 87, 88, 98, 105, 106, 116, 117, 118, 119, 120, 122, 124, 131, 141, 145, 146, 147, 148, 150, 152, 155, 158, 159, 160, 161, 186, 193, 203, 207, 210, 211, 215, 216, 232, 248, 267, 268, 269, 272, 274, 275, 276, 278, 279, 289, 290, 295, 297, 298, 299, 303, 312, 317, 318, 320, 335, 336, 338, 339, 340, 341, 342, 346, 357, 359, 360, 369, 370, 371, 373, 393, 394, 398,
466 Index profit sources – Continued 399, 400, 402, 408, 409, 412, 420, 422, 430 reduction in costs, 9, 23, 27, 71, 79, 87, 88, 104, 105, 106, 133, 136, 211, 213, 303, 316, 397, 409, 411, 412, 452 volume, 11, 16, 23, 28, 29, 31, 40, 41, 58, 104, 122, 141, 145, 155, 159, 165, 166, 175, 177, 226, 231, 247, 249, 252, 258, 259, 263, 264, 277, 282, 353, 420, 422, 423 profit strategy general, 1, 3, 4, 10, 11, 13, 23, 27, 31, 32, 33, 34, 35, 39, 62, 79, 246, 247, 269, 312, 326, 327, 348, 398, 406, 408, 411, 415, 450, 451, 452, 455 innovation and flexibility, 15, 16, 29, 249, 270, 312 volume, 406 volume and diversity, 10, 15, 16, 21, 28, 29, 228, 231, 246, 247, 252, 254, 255, 260, 263, 270, 283, 451, 452 PSA, 11, 19, 21, 25, 26, 27, 40, 42, 44, 46, 48, 49, 50, 51, 64, 204, 246–65, 292, 302, 303, 304, 307, 359, 360, 385, 426, 427, 428, 431, 432, 451, 452 Quandt, H., 332, 333, 334, 348, 350 Raff, D. M. C., 6, 36, 456 regional agreement and institution Common European Market, 288 European Union, 51, 189, 190, 226, 238, 239, 317, 322, 341, 350 Mercosur, 22, 246, 247, 256, 262, 263 NAFTA (North American Free Trade Agreement), 22, 171, 176, 182, 220, 316, 362 Renault, 4, 10, 11, 13, 14, 15, 18, 20, 21, 26, 36, 40, 42, 46, 48, 49, 51, 64, 85, 95, 98, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 111, 168, 203, 204, 246, 251, 252, 257, 266, 267–85, 292, 297, 298, 302, 306, 307, 353, 354, 359, 360, 362, 401, 410, 426, 451, 452 Renault–Nissan, 25, 168 Renault Véhicules Industriels, 273
Rolls-Royce, 238 Romiti, C., 287, 290 Rootes, 249, 311 Rosellen, H.-P., 333, 352 Rover, 4, 16, 17, 35, 40, 117, 122, 127, 204, 327, 337, 339, 343, 350, 351, 352, 366–77, 390 Saab, 16, 17, 178 SAIC (Shanghai Automotive Industry Corporation), 218, 373, 385, 388, 390 Sako, M., 419, 420, 421, 441, 442 Salerno, M. S., 6, 22, 36, 422, 441, 442, 456, 457 Samsung, 4, 15, 40, 267, 272, 273, 274, 275, 283, 285 Sandberg, A., 94, 357, 365 Sato, M., 116, 127 Scania, 42, 44, 45, 46, 48, 49, 64, 238 Schrempp, J., 209, 213, 214, 216, 310, 313, 318, 323, 330 Schumann, M., 234, 244, 344, 352 Schweitzer, L., 251, 257, 269, 270, 273, 278, 280, 286 Seat, 228, 231, 233, 237, 243 Segrestin, B., 103, 111, 274, 286 Senter, R., 35, 184, 453 Shimizu, K., 5, 36, 37, 65, 94, 205, 244, 403, 456, 457 Shimokawa, K., 13, 37, 123, 124, 128, 129, 140, 456 Shook, R. L., 125, 128 Siemens, 424, 425, 426, 440 Simca, 249, 251, 311 Škoda, 11, 228, 229, 231, 233, 237 Sloan, A. P., 165, 166, 179, 180, 184, 249, 266, 364, 384, 417, 441, 453 Smart, 17, 309, 310, 313, 314, 318, 319, 323, 324, 328, 329, 330 sourcing, 2, 7, 14, 19, 20, 22, 23, 25, 82, 83, 102, 103, 105, 107, 108, 115, 117, 124, 144, 145, 151, 152, 154, 155, 159, 168, 169, 170, 171, 176, 179, 186, 188, 190, 191, 192, 203, 208, 209, 210, 211, 212, 214, 216, 218, 219, 230, 231, 238, 255, 274, 275, 276, 277, 278, 280, 282, 288, 289, 292, 293, 294, 303, 306, 311, 317, 333, 334, 335, 338, 346, 353,
Index 360, 369, 370, 371, 392, 393, 397, 400, 401, 419, 421, 422, 426, 430, 431, 433, 435, 440, 453 Stallkamp, T., 211, 212, 221 Stevens, M., 106, 111 Stewart, P., 2, 5, 22, 35, 265, 286, 352, 456, 457 Streiff, C., 262, 303 Subaru, 40, 41, 42, 47, 48, 49, 57, 58, 84, 86 suppliers, 1, 2, 4, 9, 22, 23, 24, 26, 27, 28, 31, 39, 75, 83, 87, 102, 104, 105, 106, 107, 109, 120, 131, 144, 145, 147, 149, 154, 155, 156, 159, 166, 168, 170, 171, 172, 173, 194, 195, 206, 211, 213, 214, 221, 230, 233, 236, 248, 258, 269, 274, 275, 278, 282, 292, 295, 318, 320, 322, 326, 332, 335, 338, 340, 345, 347, 348, 349, 366, 371, 386, 392, 393, 396, 397, 398, 400, 406, 409, 412, 419–40, 420, 421, 422, 423, 424, 425, 426, 427, 428, 429, 430, 431, 432, 433, 434, 435, 436, 440, 445, 450, 452 supply chain, 22, 28, 104, 105, 106, 144, 153, 154, 155, 173, 211, 213, 345, 356, 398, 402, 420, 421, 422 Suzuki, 40, 42, 45, 47, 48, 49, 50, 57, 58, 59, 60, 61, 64, 65, 176, 178, 297, 385, 404–16
Takeishi, A., 102, 104, 111, 420, 423, 427, 442 Talbot, 246, 249, 250 Tanaka, T., 94 Taniguchi, M., 133, 139, 140 Tata, 20, 22, 41, 60, 118, 196, 281, 294, 303, 374, 410, 415 Taylor, E. W., 116, 128, 384 Thun, E., 19, 37 Tolliday, S., 5, 35, 139, 402, 456 Toyoda, K., 424, 426, 427 Toyota, 69–93, 99 Toyotaism, 37 Triumph, 368 Trotman, A. J., 187, 188, 189, 190, 197, 198
467
Valeo, 27, 423, 424, 425, 429, 430, 431, 432, 433, 434, 435, 437, 438, 439, 440, 441 value chain, 24, 238, 345, 428 van Tulder R., 5, 35, 36, 37, 38, 39, 65, 457 vehicles commercial, 14, 85, 86, 110, 143, 208, 253, 254, 259, 273, 282, 284, 305, 306, 310, 316, 329, 377, 385, 387 family cars, 280, 354 fleets, 14, 193 front-wheel drive, 186, 189, 249 fuel cell, 29, 63, 108, 121, 123, 124, 320, 445, 449, 450, 451 industrial vehicles, 85, 86, 93, 313, 317, 324, 327, 330, 377 light trucks, 15, 16, 26, 63, 101, 121, 132, 173, 182, 189, 196, 203, 206, 207, 208, 214, 218, 260, 280, 311, 315, 316, 444, 450 low emission, 120, 123, 325, 445, 449, 451, 452 luxury, 14, 16, 101, 115, 177, 200, 204, 208, 231, 232, 259, 312, 313, 319, 325, 337, 345, 353, 354, 355, 356, 362, 364, 369, 405, 417 medium-sized cars, 16, 145, 155, 232, 333 minivans, 14, 15, 16, 19, 91, 132, 189, 203, 206, 207, 208, 212, 218, 254, 259, 260, 269, 272, 279, 311, 314, 315, 316, 389 pickups, 15, 63, 203, 207 niche cars, 16, 20, 58, 119, 203, 225, 232, 256, 259, 260, 261, 268, 279, 290, 334, 339, 341, 346, 370, 373, 383, 384, 386, 402 small cars, 14, 60, 85, 101, 115, 117, 118, 122, 137, 158, 178, 214, 218, 234, 252, 289, 294, 306, 314, 333, 354, 390, 396, 405, 406, 411, 415 SUVs, 14, 15, 16, 63, 85, 86, 101, 121, 136, 139, 143, 144, 186, 188, 189, 191, 192, 193, 197, 198, 200, 201, 203, 205, 206, 207, 212, 252, 259, 260, 304, 316, 345, 347, 354, 355, 356, 360, 363, 389, 391, 406
468 Index vehicles – Continued top-of-the-range, 11, 14, 16, 17, 20, 253, 260, 263, 277, 279, 282, 452 used cars, 280, 291, 303, 320, 330, 401 world car, 20, 158, 187, 189 Verma, A., 162 Visteon, 25, 63, 156, 191, 198, 203, 204, 425, 426, 427, 428, 429, 430, 431, 432, 434, 435, 437, 438, 439, 440 Volkswagen, 1, 11, 16, 18, 19, 21, 25, 26, 36, 40, 44, 46, 49, 50, 119, 181, 225–43, 253, 256, 281, 295, 297, 298, 301, 302, 306, 336, 337, 344, 345, 350, 385, 452 Volpato, G., 5, 18, 19, 25, 35, 36, 37, 65, 94, 111, 127, 128, 161, 183, 184, 205, 244, 265, 266, 286, 287, 306, 307, 308, 331, 352, 365, 378, 403, 420, 421, 423, 441, 442, 456, 457 Volvo, 4, 12, 16, 17, 40, 44, 48, 64, 77, 139, 188, 191, 195, 202, 263, 267, 268, 270, 353–64 Volvo Trucks, 45, 273, 278 wage, 12, 14, 15, 23, 24, 72, 73, 79, 80, 81, 88, 100, 108, 113, 141, 149, 150, 151, 153, 172, 194, 211, 215, 234, 235, 236, 237, 247, 257, 262, 263, 321, 344, 385, 409, 413, 453 Wagoner, R., 168, 282 Wang, H., 403 Warburton, M., 241, 244 Wellhöner, V., 227, 245 Williams, K., 365, 378, 441, 457 Womack, J., 1, 6, 9, 30, 37, 178, 184, 230, 245, 312, 331
work crisis, 75 parcellised, 235, 397 skilled, 76, 77, 103, 106, 108, 109, 199, 251, 347, 349 workers, 12, 14, 69, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 87, 88, 107, 145, 146, 148, 149, 150, 151, 152, 155, 161, 166, 168, 170, 171, 172, 180, 194, 196, 198, 201, 213, 214, 236, 251, 262, 268, 277, 285, 311, 322, 324, 325, 326, 333, 337, 340, 342, 343, 344, 345, 347, 348, 349, 357, 358, 375, 376, 398, 399, 406, 409, 413, 416, 453 workforce, 13, 42, 43, 63, 89, 90, 91, 95, 99, 107, 148, 170, 172, 201, 204, 229, 236, 241, 264, 265, 284, 305, 314, 321, 344, 349, 351, 359, 364, 372, 375, 376, 406, 409, 413, 414, 445 working conditions, 13, 146, 150, 151, 226, 235, 237, 261, 397 working hours, 72, 76, 78, 79, 146, 150, 151, 229, 230, 234, 248, 257, 261, 277, 279, 342, 343 working time, 72, 76, 78, 79, 146, 150, 151, 229, 230, 234, 248, 257, 261, 277, 279, 342, 343 Zetsche, D., 169, 212, 213, 214, 215, 216, 221, 309, 314, 321, 325, 327 Zirpoli, F., 306, 308, 420, 440