Hybridization of MNE Subsidiaries: The Automotive Sector in India

Hybridization of MNE Subsidiaries The Automotive Sector in India

Florian A.A. Becker-Ritterspach

Hybridization of MNE Subsidiaries

This page intentionally left blank

Hybridization of MNE Subsidiaries The Automotive Sector in India Florian A. A. Becker-Ritterspach University of Groningen, Department of International Business and Management, The Netherlands

© Florian August Arthur Becker-Ritterspach 2009 Foreword © Arndt Sorge 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 author has asserted his right to be identified as the author 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: 978–0–230–20669–4 ISBN-10: 0–230–20669–7

hardback 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. Library of Congress Cataloging-in-Publication Data Becker-Ritterspach, Florian A. A. 1972Hybridization of MNE subsidiaries: the automotive sector in India / Florian A. A. Becker-Ritterspach. p. cm. Includes bibliographical references and index. ISBN-13: 978–0–230–20669–4 ISBN-10: 0–230–20669–7 1. Automobile industry and trade–India–Case studies. 2. International business enterprises–Case studies. I. Title. HD9710.I42B43 2009 338.8
87292220954–dc22 2008030139 10 18

9 17

8 16

7 15

6 14

5 13

4 12

3 11

2 10

Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne

1 09

To my wife Jutta

This page intentionally left blank

Contents List of Tables and Figures

ix

Foreword by Arndt Sorge

xi

Preface and Acknowledgements

xiii

List of abbreviations

xv

Part I Theoretical Foundations

1

1

Introduction

3

Background Research gaps and contributions The study’s theoretical contribution The study’s empirical contribution Research objective The structure of the study

3 3 4 5 5 7

2 Transfer and Hybridization in MNEs

8

3

Japanization or Transplant perspectives Institutionalist perspectives International Business perspectives Approaches compared

8 16 28 36

Subsidiary Hybridization – Conceptual and Methodological Tools

43

Introduction Production systems and hybridization outcomes defined Strategic and institutional distance, strategic choices and hybridization outcomes Research design and methodology

Part II 4

Empirical Results

43 44 47 63

75

India’s Business Context and the Automotive Sector

77

India’s institutional context India’s strategic context in the automobile sector

77 87

vii

viii Contents

5 The Case of Maruti–Suzuki India Introduction Analysis of the hybridization profile

6 The Case of Fiat India Introduction Analysis of the hybridization profile

7 The Case of Mercedes-Benz India Introduction Analysis of the hybridization profile

8 The Case of Skoda Auto India Introduction Analysis of the hybridization profile

9 Discussion Core findings – similarities and differences mapped and explained Conclusion Notes Bibliography Index

99 99 102

136 136 140

173 173 176

214 214 219

250 251 270 274 276 296

List of Tables and Figures Tables 2.1 2.2 2.3 2.4 2.5 3.1 3.2 3.3 3.4 3.5 3.6 3.7 4.1 4.2 4.3 5.1 5.2 6.1 6.2 7.1 7.2 8.1 9.1 9.2 9.3 9.4 9.5

Boyer’s four main types of hybridization Nature and likelihood of hybridization in different industrial models and national institutions Distinct environmental forces and their sources A comparison of four learning cycle models Strengths and weaknesses compared across different approaches Different hybridization outcomes Porter’s generic strategies Entry mode combinations Selection criteria and variance of cases based on theoretical sampling Summary of research activities at different research sites Summary of data collection regarding India’s institutional and strategic context Interviewees’ functions and positions across firms Institutional foundations of work dispositions in the Indian work context Development of India’s FDI regime in the automobile industry Major players in the Indian automobile industry Overseas training programmes in MSI Summary of MSI’s hybridization profile Fiat’s suppliers in India Summary of FI’s hybridization profile MBI’s hierarchical differentiation Summary of MBI’s hybridization profile Summary of SAI’s hybridization profile Hybridization outcomes of all four cases compared Overview of strategic choices of the four MNEs Generic strategy, internationalization experience and template availability for transfer Entry mode and local template availability Equity mode and local template availability ix

25 27 30 35 41 46 58 60 68 70 71 73 83 90 92 122 134 168 171 184 211 247 252 258 261 264 266

x List of Tables and Figures

Figures 3.1 3.2 3.3 3.4 3.5 4.1 4.2 4.3 4.4 4.5 6.1 6.2 7.1 7.2

Dimensions of a production system Different kinds of (mis)fits Different transfer scenarios, misfits, recontextualization modes and outcomes The fourfold embeddedness of subsidiary production systems Foreign ownership development of MSI, FI, MBI and SAI Market share in the passenger car sector by manufacturer in India, 1990–91 Market share in the passenger car sector by manufacturer in India, 1999–2000 Market share in the passenger car sector by manufacturer in India, 2004–2005 Development of the Indian passenger and commercial vehicle market (sales), 1995–96 to 2004–2005 Development of the Indian passenger car market by segment, 2001–2002 to 2005–2006 The UTE concept Organization of an UTE MBI’s ‘Clearly defined functions for our supervisors’ MBI’s ‘Family organization principles’

46 48 52 57 69 94 94 95 96 96 153 154 194 198

Foreword

When international business and management (IB&M) researchers think of India, they are more likely to think of software than motor cars. However, as this book shows, India is more interesting as an ideal setting in which to look at subsidiaries of motor-car manufacturers from all over the world, in all kinds of market segments, with all kinds of national or international production systems and local domestic content. Probably nowhere else in the world is the organizational mix of the strategy, organization, industrial relations and human resource arrangements in the subsidiaries of multinational enterprises as diverse and complete as in motor industry multinationals in India. It is amazing that no one has looked at it systematically before. Moreover, IB&M researchers, over-specialized and over-generalizing in the best traditions of the social sciences, including economics, have been adept at identifying the role of particular factors such as cultural or institutional distance, transaction costs, learning, micropolitics, and other factors in the internationalization of enterprises. But they have been less successful at unravelling the interaction of such factors. In that way, they have not adequately explained what happens in concrete and complex instances of real organizational life in multinationals. The real world was crying out for someone to go there and see what it is like, in theoretically sampled subsidiaries, cleverly combined to represent combinations of factors and developmental histories. Florian Becker-Ritterspach has done that, and in this way he has provided a valuable service, cutting through the vague and unhelpful discussions about ‘hybridization’. In a world in which races, institutions, cultures, strategies and markets evolve and intermingle; we are all hybrids, of course, including the cars that we drive and the organizations in which we work. Yet, the recombinations that occur have locally specific forms that are interesting beyond the banal fact of hybridity, and they can be explained quite well. This is what this book demonstrates in exemplary form. This what makes it valuable for everyone in or looking at multinationals, also those that do not know India or do not much care for the automobile industry in particular, like myself. Florian was ideally prepared for this task, having grown up for some years in Nepal, working in his mother’s business and having spent time in India, before launching himself into this study. I am very happy and xi

xii Foreword

proud to have served him, first, as a university teacher at Humboldt University, Berlin, then as a career counsellor, doctoral supervisor and his immediate boss in Groningen. The results of this book could not have been properly presented in an article, not even a series of articles. Even where messages are basically simple they may have complex qualitative ramifications; books are still indispensable to make these messages credible. I am also happy that Florian has demonstrated this to the number-crunching research-journal impact-score fetishists in our research evaluation establishment. Arndt Sorge University of Groningen and Wissenschaftszentrum für Sozialforschung, Berlin

Preface and Acknowledgements

Writing this book was a painstaking process in a number of ways: getting access to companies during my stay in India, living in a developing country on a tight budget, and filling threatening white pages with meaningful thought, appeared at times a never-ending, insurmountable task. Or using the analogy of flying an aircraft: I started with a plane loaded with fuel, all systems and engines running smoothly. However, once in India two of my engines were suffocated by the thick dust hanging over New Delhi, when two firms decided they no longer wanted to participate. After some intense efforts and a mobilization of all possible social capital, I managed to re-ignite the engines by replacing the lost firms with two alternatives. However, this was only the beginning; a long flight back was still lying ahead. On my way back to the Netherlands, loaded with voluminous empirical cargo, it felt too heavy, as if too much fuel was being burned without covering enough distance. And as if this was not enough, it felt as if all navigational systems were being lost during night-flight, when I started to explore and make sense of the empirical quagmire. Strangely, hard work, instinct and maybe a hand of God guided me and kept me on route. Approaching home, circling around the airport several times was unavoidable due to bad visibility. Running low on fuel, the clouds suddenly broke and the runway came into sight. I finally managed to touch down on the last drop of kerosene, leaving everything miraculously without major structural damage. It should not be forgotten here that this flight was only possible with the support of air traffic controllers, a highly supportive ground crew and infrastructure. Therefore, I would like to thank very much a number of people and institutions, without whom this project never could have got off the ground and brought back home safely. First of all, I would like to thank my parents and my wife Jutta. I would like to thank my parents, who lay the foundations of what I was able to achieve and never lost faith in my work when I was uncertain. I would like to thank my wife, who helped me to structure my thoughts in numerous discussions, who comforted me when I faltered, and who took on the arduous task of proofreading some of my work. In this context, I also would like to thank Christoph Dörrenbächer who gave me invaluable comments to revise my book for Palgrave Macmillan. xiii

xiv Preface and Acknowledgements

Moreover, I would like to thank a number of people and friends, who supported me during my research stay in India and helped with the difficult task of gaining access to companies. In this context, I would like to particularly thank a number of Indian women including Vijay and Pushpa Bhatia, Kavita Datta, Anandi Iyer, Sonia Prashar and Priya Jaikishan, who proved much more dependable than their male counterparts. I am also very grateful to Augustus Mallier, whom I got to know years ago, during my first research stay in India and who has become a close friend. Moreover, I would like to extend my thanks to a number of New Delhi–based organizations, which were very supportive, including the German Embassy, the Indo-German Chamber of Commerce, and the GTZ (Gesellschaft für Technische Zusammenarbeit). And I would like to thank of course the companies and organizations who granted me interviews enabling me to realize this project. Finally, I would like to thank my supervisors Arndt Sorge and Jacob Wijngaard. Here, I would like to thank especially Arndt Sorge, who has been a very dependable air-traffic controller, who monitored my flight and provided me with invaluable support and direction to master air turbulence and other navigational challenges. Arndt Sorge has been in the best sense of the German word a Doktorvater to me. Last, but not least, I gratefully acknowledge the grants received for this study from the German Academic Exchange Service (DAAD) and the Research School Systems, Organization and Management (SOM). Florian A. A. Becker-Ritterspach University of Groningen and Wissenschaftszentrum für Sozialforschung, Berlin

List of Abbreviations

ACMA AITUC BEMs BMS CITU CKD CPI DB DC EWA FDI FERA FIAL FI GoI GM HCV HML HMS HR HRM IB&M IAL IB IDA IEA IIT INTUC ITI JIS JIT JV KEPS

Automotive Component Manufacturers’ Association All India Trade Union Congress Big Emerging Markets Bharatiya Majdor Sangh Centre of Indian Trade Unions Completely Knocked Down Conduttori di Processi Integrate Daimler Benz DaimlerChrysler Engineering Workers’ Association Foreign Direct Investment Foreign Exchange Regulation Act Fiat India Auto Ltd Fiat India Private Ltd Government of India General Motors Corporation Heavy Commercial Vehicle Hindustan Motors Ltd The Hindu Majdoor Sabha Human Resources Human Resource Management International Business and Management India Auto Ltd International Business Industrial Disputes Act Industrial Employment Standing Orders Act Indian Institute of Technology Indian National Trade Union Congress Industrial Training Institute Just In Sequence Just In Time Joint Venture Kamgar Ekta Premier Sanghatana

xv

xvi List of Abbreviations

LCV MB MBI MBO MBPS MD MNE MoU MPS MRTP MSI MUEU MUKU MUL MUV PAL PLEMAs PSUs Rs SAI SIAM SKD SMC SMPIL SPS TELCO TUA UNIDO UTE VAT VRS VW WTO

Light Commercial Vehicle Mercedes-Benz Mercedes-Benz India Ltd Management by Objectives Mercedes-Benz Production System Managing Director Multinational Enterprise Memorandum of Understanding Maruti Production System Monopolies and Restrictive Trade Practice Act Maruti–Suzuki India Maruti Udyog Employees’ Union Maruti Udyog Kamgar Union Maruti Udyog Ltd Multi Untility Vehicle Premier Auto Ltd Periphery of Large Existing Market Areas Public Sector Undertakings Rupees Skoda Auto India Private Ltd Society of Indian Automobile Manufacturers Semi Knocked Down Suzuki Motor Corporation Standard Motors Products of India Ltd Skoda Production System Tata Engineering and Locomotive Company Ltd Trade Union Act United Nations Industrial Development Organization Unità Tecnologica Elementare Value Added Tax Voluntary Retirement Scheme Volkswagen AG World Trade Organization

Part 1 Theoretical Foundations When complex organizational systems developed in one social context (in this case Japan) are transferred to a different setting (the United States), they change, and the resulting system is neither a copy of the original model nor a replica of existing local patterns but something different. (Eleanor Westney in Liker et al. (1999), Remade in America: Transplanting and Transforming Japanese Management Systems, p. 385. By permission of Oxford University Press.)

This page intentionally left blank

1 Introduction

Background Multinational enterprises (MNEs) setting up production sites abroad face the ongoing challenge of how to configure their subsidiaries’ production systems. This organizational challenge involves the questions: Whether MNEs should define and transfer whole production templates or only apply a limited range of policies? What kinds of adaptations are possible and required to match the transferred templates with local or host context conditions? And which templates and practices may not be transferable at all but either have to be drawn from or customized to the local or host contexts? Based on these organizational challenges for MNEs, this study tries to understand how and why subsidiary production systems differ with regard to their contextual origin or constitution, that is, their hybridization profile. While answering such questions are important for any internationalizing firm, they are even more salient if provided for firms setting up production sites in large and newly emerging economies. By providing a comprehensive understanding about the conditions of India’s business context for foreign direct investment (FDI) with a particular focus on the automobile industry this work serves such a purpose.

Research gaps and contributions The aim of this study is to make a major theoretical and empirical contribution to hybridization research in organization studies and international business. Hybridization research refers here to different bodies of literature that are concerned with questions of transfer and adaptation of organizational forms and practices (including production models or systems) in MNEs or in cross-national settings. 3

4 Hybridization of MNE Subsidiaries

The study’s theoretical contribution While there is no organizational hybridization theory as such, there are a number of studies from different research traditions that have either contributed to or adopted a hybridization perspective (e.g. Abo et al., 1994; Boyer, 1998; Brannen, 1999). Many of these contributions have focused on MNEs because it is particularly challenging to understand the contextual constitution of organizations that are embedded in different national contexts. In this study, three main bodies of research are identified that have made seminal contributions to the question how we can capture hybridization outcomes and why these outcomes occur when organizational forms and practices are transferred across borders and units in MNEs. These bodies include: the Japanization or transplant literature (e.g. Turnbull, 1986; Ackroyd et al., 1988; Elger and Smith, 1994; Oliver and Wilkinson, 1988; Florida and Kenney, 1991; Pil and McDuffie, 1999), Institutionalist approaches (e.g. Rosenzweig and Singh, 1991; Westney, 1993; Kostova, 1999; Kostova and Roth, 2002; Whitley, 2001; Lane, 2000; Morgan, 2001; Sharpe, 1997; Lorenz, 2000; Saka, 2003) and contributions from the field of International Business (e.g. Bartlett and Ghoshal, 1998; Prahalad and Doz, 1987; Birkinshaw and Hood, 1998; Beechler et al., 1998; Gupta and Govindarajan, 2000; Chini, 2004). However, this work shows that all three bodies of literature remain unsatisfactory for our understanding of hybridization in MNEs, if they are left unconnected. Specifically, the discussion of the different approaches illustrates: (1) that hardly any body of literature systematically addresses the impact of both strategic and institutional contextual difference on production system hybridization; and (2) that most studies on production system hybridization fail to thoroughly theorize and empirically research how strategic choices at the corporate and subsidiary level influence hybridization outcomes in MNEs. Addressing these gaps, this work makes two main theoretical contributions to the emerging body of hybridization research in Organization Studies and International Business. The first contribution is to address the question how the complex contextual embeddedness of subsidiaries impact transfers and adaptations and thereby production system hybridization. In contrast to much of earlier hybridization research (see Chapter 2), it is particularly stressed that production systems emerge out of both specific strategic and institutional contexts. It is proposed that strategic and institutional differences constitute two distinct sources of contextual distance that impact production system hybridization. The second contribution is to explore the association between production

Theoretical Foundations 5

system hybridization and different strategic choices at the corporate and subsidiary level. In contrast to the neglect of this relationship in much of earlier hybridization research (see Chapter 2), it is proposed that generic product strategies of MNEs and entry modes of subsidiaries have a strong influence on hybridization outcomes.

The study’s empirical contribution Extant hybridization research, particularly Japanization research, has made important contributions to understanding the conditions of the transferability of production systems. However, the literature has largely neglected non-Japanese transfer origins as well as non-Western transfer destinations. This study addresses the empirical bias by focusing on the constitution of subsidiary production systems in India, involving not only Japanese companies, but also a German, an Italian and a Czech/German MNE. Moreover, although researchers from Organization, Management and International Business studies have paid increasing attention to Asian business contexts, most notably to the ‘Asian Tigers’ and China, India, one of the most important emerging economies in Asia, has been largely neglected. By focusing on India, the study provides insight into one of the most important emerging markets and business environments of the 21st century. Finally, with its focus on the emerging automobile industry in India, this study also makes a contribution to understanding the processes of internationalization in the automobile industry, one of the world’s most important and international industries.

Research objective The research objective of this study is to explore how and why the hybridization profiles of four automobile-subsidiary production systems in India differ. The how-question in this work involves identifying the hybridization profiles of the four subsidiaries’ production system. The term ‘hybrid’, which means in its Latin root ‘of two origins’, refers in this context to the emergence of organizational forms that are constituted by different contextual origins. In this research context a hybridization profile essentially describes how the different dimensions, defined to constitute the core of a production system, reflect different contextual origins. Drawing on and extending seminal contributions of hybridization research (e.g. Boyer, 1998) the different contextual origins of a production system can

6 Hybridization of MNE Subsidiaries

be captured by four ideal typical (hybridization) outcomes, comprising: imitation, localization, hybridization and customization/novelty. Thus, a hybridization profiles describes in this study the contextual origin of organizational configurations along a number of predefined organizational dimensions. It is important to note, that the approach in this work differs from much of the earlier work in this field in that it does not predefine a specific production system or model to be transferred and asking in a second step what is implemented and what is not (e.g. Abo et al., 1994). Instead, it takes a more open approach by defining core dimensions of a production system and by identifying then what contextual origin the respective dimensions reflect. Such an approach is based on the assumption that transfers in MNEs understood as ‘transnationals’ (Bartlett and Ghoshal, 1998) may have many sources, drivers, and may be, rather than being based on clearly intended and comprehensively defined transfer templates, based on transfer restraint, selective transfer intentions or even inarticulate, fuzzy transfer efforts. The why-question asks why similar or different hybridization profiles have come about in the production set up of the four automobilesubsidiaries in India. Drawing on and refining earlier research (Westney, 1993; Beechler, 1998; Brannen, 1999), hybridization outcomes in MNEs are seen to be constituted by the interaction of three varying factors comprising different: transfer scenarios, contextual misfit/recontextualization pressures and recontextualization modes (see Chapter 3). The transfer scenario involves the question, whether or not an MNE transfers a template or poses demands vis-à-vis a local production system. There are three ideal typical starting points for transfer scenarios: A foreign/parent template transfer, a host/local template use or, neither a foreign/parent template transfer nor a local/host template use. Based on the different transfer scenarios there are different kinds of fits or misfits these starting scenarios can face. For example, a foreign/parent context template can (mis)fit the local/host context just as a local/host context template may (mis)fit the foreign/parent context. Misfits of whatever kind tend to induce pressures for adaptation, that is, recontextualization pressures. The final factor involves the question how misfit induced recontextualization pressures are resolved. It is argued that two principle recontextualization modes exist that can be simultaneously at work. The first mode involves the adaptation of foreign/parent context (templates, demands or conditions) to the local/host context. The second mode involves the opposite, the adaptation of the local/host context to

Theoretical Foundations 7

the foreign/parent context (templates, demands or conditions). Depending on the interaction of these three variables we can expect different hybridization outcomes. It is argued in this study that the variance on these three constituent factors of production system hybridization can only be properly understood, if we consider: (1) the role of both strategic and institutional distance, (2) the impact of strategic choices at the corporate and subsidiary level, and (3) the possible interplay of strategic choices and contextual distance.

The structure of the study This study is organized as follows: In the first part of the book, comprising Chapters 1–3, develop the theoretical foundations of the study. Following the introductory chapter, Chapter 2 comprises the literature review and discussion. While Chapter 2 serves to identify research gaps with regard to subsidiary production system hybridization, it simultaneously provides building blocks for the analytical framework in Chapter 3. In Chapter 3, the analytical framework and the research methodology are presented. In this chapter the unit of analysis and different hybridization outcomes are defined. Moreover, propositions are developed, delineating the expected association between hybridization outcomes and institutional/strategic distance as well as between hybridization outcomes and strategic choices. Following the presentation of the analytical framework the study’s research design is presented at the end of the chapter. The book’s empirical part starts with Chapter 4. The chapter provides an introduction to institutional and strategic context conditions in India that are particularly relevant for production system hybridization. The chapter focuses particularly on the conditions for the automobile industry in the Indian business context. Presenting and analyzing the hybridization profiles of four automobile subsidiaries in India, Chapters 5–8 present the main empirical part of the study. Specifically, Chapter 5 covers Maruti-Suzuki India Limited, Chapter 6 Fiat India Limited, Chapter 7 Mercedes-Benz India Limited and Chapter 8 Skoda India Limited. Finally, Chapter 9 comprises a discussion and conclusion of the empirical study, including a critical reflection of the study as well as a brief outlook for further research.

2 Transfer and Hybridization in MNEs

Most studies that have made contributions to the questions of hybridization within the context of cross-border transfers in MNEs fall into at least one of the following three larger bodies of literature, namely: Japanization or Transplant, Institutionalist and International Business literature. While not all contributions can be exclusively classified as one or the other, most contributions can be subsumed under either of these bodies of literature (particularly their sub-streams) based on their main theoretical and empirical orientation. The following literature review introduces the Japanization, Institutionalist and International Business literature and their sub-steams and discusses their specific contribution to the question how- and why hybridization of organizational forms and practices – including production systems – occurs in MNEs and their subsidiaries. The chapter closes with a comparison of the strengths and weaknesses of the different bodies of literature. It is shown that no one approach makes an effort to relate systematically the hybridization of subsidiary production systems to the impact of both the strategic and institutional distance and to strategic choices at both the corporate and the subsidiary level.

Japanization or Transplant perspectives Hybridization perspectives have strong roots in the Japanization research of the 1980s and 1990s. Triggered by a stark rise in Japanese FDI to the US and UK, combined with the influential work of management gurus and a perceived superiority of Japanese business practices an immense body of literature evolved around the transferability of Japanese production systems and practices. Westney argues that Transplant research – as a part of the wider Japanization research – not only emerged as a subfield 8

Theoretical Foundations 9

in MNE-research but was ‘arguably the most widely studied aspect of the organization of multinational enterprise in the 1980s and 1990s’ (Westney, 2001, p. 640). Commonly, two major strands have been identified in the Japanization literature: the Labour Process and the Lean Production perspective (Saka, 2003; see Stewart, 1998 for a similar mapping of the debate). This distinction is particularly helpful with regard to mapping early contributions. For the two streams not only vary in focus – ‘managerial-user vs labour control’ (Saka, 2003) – but also offer strikingly different answers to the questions whether the cross-national transfer of Japanese production systems is possible and whether contextual differences play a constraining or modifying role for such transfers. Interestingly, the two bodies of the Japanization debate not only originate mainly from the US and the UK – reflecting Japanese FDI patterns in the 1980s and early 1990s – but they also reflect different research traditions on the two sides of the Atlantic.

The Lean Production perspective The North American side of the Japanization literature is firmly embedded in the Lean Production debate (Saka, 2003). Contributions located within this perspective are overall optimistic about the possibilities of successful Japanese production system transfer (Krafcik, 1986; Adler, 1993). They also tend to emphasize the progressive nature of Japanese production systems for the employment relations (e.g. Adler and Cole, 1993). With the managerial-user in mind, the Lean Production perspective’s main level of analysis is the firm. Although, the perspective substantially develops over time, its original proposition is the universal applicability and competitive superiority of Japanese production systems. The most prominent representations of this view are Womack et al.’s (1990) work and in a more elaborate expression the work by Florida and Kenney’s (1991a, 1991b; Kenney and Florida, 1993; see also Pil and MacDuffie, 1999). Based on both survey and case study research, these authors investigate to what extent predefined ideals of the Japanese production systems can be transferred to the United States. Excepting the work of Abo et al. (1994) (see discussion below), there is initially little elaboration and conceptualization of outcomes other than successful transfer or imitation of production systems. Although early contributions in this body of literature admit that adaptations are unavoidable, these are rated as ‘transfer-with-secondary-adaptations’ or as ‘functional equivalents’, which are not compromising the transfer of the core of the production system or its performance (Kenney and

10 Hybridization of MNE Subsidiaries

Florida, 1993; see also Oliver and Wilkinson, 1988; Mishina, 1998; Adler et al., 1998; Pil and MacDuffie, 1999). Nevertheless, despite overall optimism about transfer success, these studies already indicate that transfer is not equally successful across all dimensions of a production system and varies by transfer content. Typically, adaptations are found in the companies’ industrial relations and in specific aspects of human resource management. For example, aspects such as wage determination and labour relations are adapted to fit the US context (e.g. Adler et al., 1998; Pil and MacDuffie, 1999). Notwithstanding that adaptations to local or host contextual conditions are found, especially early contributions stress that contextual differences do not impede transfers in a meaningful way (Florida and Kenney, 1991b, p. 193). Japanese production systems are hailed as culture-free and independent of specific contextual conditions. In areas where contextual misfit and production system adaptation cannot be ignored entirely, the ability of resource- and powerful firms to select, change and/or to create the required internal and external context is emphasized (Florida and Kenney, 1991a). To be fair, while early contributions downplay transfer outcomes other than straight imitation and contextual differences as causes for transfer difficulties or restraint, more recent contributions in this stream take a different view. Scholars in this body of literature have come to embrace the notion of ‘hybridization’ or of similar concepts such as ‘third culture’, ‘transformation’ or ‘recontextualization’ to capture transfer outcomes in a more differentiated and complex way (e.g. Wilms et al., 1994; Liker et al., 1999; Adler et al., 1998; Adler, 1999; Babson, 1998; Brannen et al., 1999). Also, with respect to the question why transfer outcomes differ, there is a growing attention to how contextual differences and the context-boundedness of certain practices impact transfer propensities and adaptation pressures. In this context a number of contributions even identify institutional or societal difference as important factors influencing transfer and adaptation dynamics (e.g. Adler et al., 1998; Babson, 1998; Adler, 1999; Brannen et al., 1999; Liker et al., 1999; Pil and MacDuffie, 1999). The main reasoning is that as some practices (or dimensions of a production system) are more embedded in institutional environments than others. As institutional environments differ from country to country, the ease of transfer varies with the kind of practice (e.g. level of tacitness) and the level of institutional difference. Notwithstanding the increasing concern with institutional contexts and their impact on the transferability in this body of research, the conceptualization of institutional framework remains overall ad hoc and under theorized (c.f. Elger and Smith, 2006).

Theoretical Foundations 11

In contrast to the weak concern with institutional context, scholars from the Lean Production wing have stressed for some time the impact of task environmental and business contextual difference on transfer propensity and adaptation. For example, Abdullah and Keenoy (1995) show, in the case of Japanese electronics firms in Malaysia how low labour costs and low profit margins in the host context lead to transfer restraints on the part of the Japanese parent (see also Dedousis, 1995). In a similar vein, Kenney and Florida (1995) find substantial difference in transfer propensity depending on the sector or industry. Although Kenney and Florida (1995) do not develop a theoretical framework that systematically relates strategic choices to transfer propensity, adaptation requirements, different adaptation modes and hybridization outcomes, their findings suggest a close connection between transfer propensity and strategic factors, such as time of market entry and establishment modes. They also find associations between the nature of the production process (labour intensive task environment or not), the international division of labour in the MNE and the transfer propensity (Kenney and Florida, 1995; see also Abdullah and Keenoy, 1995; MacDuffie, 1995; Dedousis, 1995; Itagaki, 1997; Adler, 1999; Liker et al., 1999; Wilkinson et al., 2001).

Seminal contribution: Abo and colleagues The work of Abo et al. (1994, see also Itagaki, 1997; Kumon and Abo, 2004) is discussed here in more detail because it belongs to the first contributions to conceptualize and empirically research different hybridization outcomes within the context of cross-border transfers of production systems in MNEs. While Abo et al. (1994) share the conviction that Japanese production systems are ‘best practices’ and that anything but their full application is a trade off, they concede that some adaptation may be required as local contexts differ. It is a situation that they also define as the ‘Application–Adaptation Dilemma’ for Japanese firms. In a study of 34 Japanese auto assembly, auto parts, consumer electronics and semiconductor firms in North America, Abo and colleagues investigate how the outcomes of production system transfer differ. They distinguish four types of outcomes, including: outright application, revised application, active adaptation and passive adaptation. ‘Outright application’ involves that a predefined element or elements of the Japanese production system are applied without alternation. While outright application is the ideal scenario, the authors concede that adaptation may be a deliberate choice for certain elements to allow the application of other – potentially more important – elements. This is

12 Hybridization of MNE Subsidiaries

what they call a ‘revised application’. ‘Active adaptation’ implies that a firm deliberately draws on a wide range of typical elements of the local production system, actively seeking their duplication. Finally, a ‘passive adaptation’ means that an application has failed and that instead of the Japanese production system, the local production system is used or prevails. In a further step, Abo and his colleagues (1994) determine the degree of hybridization by the extent to which typical Japanese production system elements are found in Japanese owned production plants in the United States. Using an ‘application-adaptation-evaluation form’, the core question is to what extent the American subsidiaries apply predefined elements of the Japanese production system. By placing the Japanese and American systems at opposite ends of a continuum, the authors are able to determine the degree of application or adaptation. Based on this framework, Abo and colleagues identify in their survey different hybridization patterns across industries, plants, and production system elements. The authors show that the ideal Japanese production system, as initially defined, cannot be found in American plants. Their hybridization ratios show that transfer success is on average little more than 50 per cent. In other words, the authors find a dominance of hybrid systems, reflecting neither the Japanese nor the American production system in purity. Yet, while Abo et al. (1994) admit that ‘certain practices take on a new form, quite apart from anything that already exists in America or in Japan’ (Abo et al. 1994, p. 36), they do not pay attention to novel outcomes with regard to specific production system elements. In their survey analysis, Abo et al. (1994) identify hybridization profiles on an aggregate level by calculating to what extent the Japanese or the American production system has been ‘applied’ or ‘adapted’. This is essentially a dichotomous view of hybridization, based on either Japanese/imitation or American/localization outcomes across the elements of a production system. While the work of Abo et al. (1994) presents an elaborate framework how transfer outcomes may differ, there is only a weak and largely ex post theorization as to why such differences occur (c.f. Adler, 1999). These ex post theorizations include cultural as well as task environmental factors. Abo and colleagues reason, for example, that difficulties in application and requirements for adaptation are connected with the interrelatedness of different production system elements and their embeddedness in the Japanese culture and society. Moreover, as the Japanese production systems are strongly human centred, they require specific contextual conditions that are not readily available in historically and culturally

Theoretical Foundations 13

different environments (Abo et al., 1994). Finally, while Abo et al. (1994) refer to factors such as the nature of production process, operation size and operational characteristics to explain different application levels across industries, they leave the connection between strategic choices, transfer propensity and different hybridization outcomes largely unexplored. It also remains unclear whether the Japanese production system’s superiority is seen as universal or as contextually founded. This also leads to some ambivalence with regard to the question whether a full transfer is always the best option.

The Labour Process strand With strong roots in the Labour Process tradition (Braverman, 1974) the British side of the Japanization debate tends to be very skeptical about both the prospects of widespread transfer of Japanese production systems and its emancipatory value (e.g. Turnbull, 1986; Ackroyd et al., 1988; Dickens and Savage, 1988; McKenna, 1988; Marchington and Parker, 1988; Briggs, 1988; Morris, 1988; Crowther and Garrahan, 1988; Bratton, 1990; Garrahan and Stewart, 1992; Delbridge et al., 1992; Sewell and Wilkinson, 1992; Delbridge, 1995; Danford, 1997; Procter and Ackroyd, 1998).1 As far as the level of analysis is concerned, the Labour Process literature varies between the industry and the firm level. At the firm level, the question is not only whether Japanese MNEs succeed in transferring their production systems but also whether British firms are able to emulate such production systems. Turnbull (1986) is generally seen to have initiated the Japanization debate with regard to the question of the transferability of Japanese production systems to the UK. In a case study he asks whether British firms successfully emulate Japanese management practices. While Turnbull identifies industry-driven and government supported changes towards ‘manufacturing “a la Japanese style” ’, he is very skeptical about the possibility of transferring ‘ “high-trust” management techniques into an essentially “low trust” environment’ and stresses ‘the problems of “grafting on” Japanese production methods without some wider social or institutional parameters found in Japan’ (Turnbull, 1986, p. 204; see also Ackroyd et al., 1988; Procter and Ackroyd, 1998; Stewart, 1998). What does the Labour Process wing of the Japanization debate suggest with regard to the question how and why the outcomes of production system transfer differ? Essentially transfer outcomes are captured in this perspective as a dichotomy between, either successful transfer of Japanese elements – which is generally held unlikely or very difficult – or a continuation of local patterns. While early contributions from the

14 Hybridization of MNE Subsidiaries

Labour Process stream are little concerned with the emergence of hybrid or novel organizational outcomes, research findings suggest that the degree of Japanization varies across different dimensions of a production system due to the selective nature of transfer or practice adoption (e.g. Turnbull, 1986; Taylor et al., 1994; Procter and Ackroyd, 1998). With regard to the question why transfer outcomes differ, the Labour Process representatives stress the differences between home and host contexts. Production systems are seen to develop out of specific capitalist social relations. If such relations are different in the host country, transfer becomes very difficult, if not impossible. Scholars in the Labour Process wing tend to implicitly or explicitly stress society as well as region specific expressions of the universal tensions in capital–labour relations (e.g. Elger and Smith, 1994). With a strong focus on social relations in general and labour relations in particular, contributions in the British strand are sensitive to the institutional origins of transfer contents and different kinds of local institutional barriers to transfer or conflicts related to it (e.g. Turnbull, 1986; Ackroyd et al., 1988; Dickens and Savage, 1988; Graham, 1988; Procter and Ackroyd, 1998; Morris et al., 1998). The British Japanization debate also makes first steps towards hybridization research by pointing to processes of conflict and institutional constraints within the context of Japanese production system transfers. What is more, recent contributions in this research tradition have also moved beyond dichotomous views of transfer outcomes and recognize or adopt the notion of hybridization (Wilkinson and Ackers 1995; Mair, 1998; Scarbrough and Terry, 1998; Smith and Elger, 2000; Elger and Smith, 2006). However, with the notable exceptions of Elger and Smith’s work (introduced below) few contributions move beyond the home- and hostcontext difference to explain different transfer outcomes. The impact of strategic factors on transfer outcomes at the corporate and subsidiary level generally remain under explored. While few systematic comparisons exist, exploring the impact of strategic choices at the corporate level, some authors in this stream deal with the impact of subsidiary-level strategic choices. Some authors ask here whether Greenfield operations facilitate transfers and mediate institutional distance, coming to contrasting findings, however (c.f. Oliver and Wilkinson, 1988; Garrahan and Stewart, 1992; Smith and Elger, 1998).

Seminal contribution: Elger and Smith The work of Elger and Smith is discussed here because it is most elaborately concerned with questions of transfer and transferability of Japanese production systems within the Labour Process Perspective.

Theoretical Foundations 15

Although Elger and Smith are critical about the ease to transfer Japanese production systems and their emancipatory value, they do not deny that transfers may be successful to some extent (e.g. Smith and Elger, 2000). In fact, being critical about both an overestimation and an underestimation of institutional pressures, they see possibilities for hybrid outcomes to emerge from transfers. With regard to the question how transfer outcomes differ, they take a middle ground between seeing transfers as always reverting to local patterns in the face of nationally specific institutional systems and seeing transfers as remaining unaltered in the face of foreign parent pressure and demand (Elger and Smith, 1994). With regard to the contextual impact on transfer outcomes, Elger and Smith (1994) argue against ‘needlessly dualistic’ positions between context bound and context free positions. Instead they suggest that organizational forms and practices are ‘simultaneously the embodiment of general economic efficiencies, culturally specific institutional supports and dominant best practices of a powerful economy’ (Elger and Smith, 1994, p. 31). Accordingly, the authors see the transfer of organizational forms and practices and their adaptation constituted by different contextual pressures. In more recent contributions, they further specify these different factors or contextual effects and come to call them ‘system imperatives’ of capitalism, ‘societal effects’ exerted by the institutional environment of nation states and ‘dominance effects’ exerted by ‘specific hierarchies of economic, military/political and cultural dominance and subordination among national states’ (Smith and Elger, 2000, p. 227; see also Elger and Smith, 2005, 2006). Based on this broader framework of contextual effects the authors explore empirically the transfer of Japanese systems and practices to Japanese subsidiaries in Britain. They argue that the transfer of organizational practices primarily rests on three aspects, including: the capability of transfer, the propensity to transfer and the negotiated appropriateness. The ‘capability of transfer’ refers to those factors that condition whether a company can actually transfer. These are different human and financial capabilities of firms, depending on ‘standard contingencies of size, power resources, technology and international experience’ (Smith and Elger, 2000, pp. 229–230). The ‘propensity to transfer’ is related to the question whether firms want to transfer their systems and practices and is seen to depend on ‘strategic intentions of the MNC in building home-standard or more differentiated manufacturing facilities and methods of work organization in overseas subsidiaries’ (Smith and Elger, 2000, p. 232). In this context, Elger and Smith also suggest that more often than not transfers are piecemeal, rather then comprehensive and

16 Hybridization of MNE Subsidiaries

complete. Finally, the ‘negotiated appropriateness’ refers to the adaptation requirements transfers face, when they arrive in a new local context. Specifically those related to a particular workplace and locality. Interestingly, however, Smith and Elger see this adaptation process not as unidirectional. Instead, adaptation dynamics are found to be a ‘working out of transfer or a “fit”’ (2000, p. 232), involving adaptation pressures from both, the local/host and the foreign/parent context. In this process, the Japanese practices as well as the local context transform to some extent. In their empirical study on Japanese transplants in the UK, Smith and Elger find no evidence of a ‘full and coherent package of management techniques suggested by ideal type portrayals’ (2000, p. 236). While they find in some respects implemented features of an ‘ideal’ production system, the overall picture is one of diversity and variation.

Institutionalist perspectives Introduction Institutional perspectives form the second body of research that has increasingly become concerned with the transfer of organizational forms and practices in MNEs. In contrast to the Japanization literature discussed, Institutional perspectives elaborately theorize the social constitution of organizations. Despite all kinds of differences, a common denominator of these approaches is seeing organizational behaviour as socially embedded. Institutionalist approaches generally reject the notion that organizations are rational actors operating in response to singular and universal logics of economic efficiency (Saka, 2003). Although institutional approaches have only recently focused on questions of transfers and adaptations in MNEs and rarely focus on the transfer of entire production systems, they fill important gaps in hybridization research in general and in the research of production system hybridization in particular. First, Institutionalist approaches – even those that do not consider cross-border transfers – provide elaborate concepts for institutional contexts and differences that potentially impact production system hybridization. Secondly, they increasingly apply Institutionalist thought to MNEs and their subsidiaries, addressing the question how their contextual constitution and how the transferability of organizational forms and practices are impacted by their operation in different institutional contexts. While Institutionalist approaches share at a general level a common understanding of organizations as social contextually constituted, the approaches differ markedly in their understandings of institutions.

Theoretical Foundations 17

Moreover, different Institutional approaches have also provided – at least initially – markedly different answers with regard to the transferability question of organizational forms and practices. Based on these two major differences – concepts of institutions and the transferability question – two bodies of approaches can be distinguished. The first body of approaches is commonly described as the Varieties of Capitalism literature (e.g. Whitley, 1992; Lane, 1994; Hall and Soskice, 2001). The second body comprises approaches that build on New Institutionalist thought (e.g. Meyer and Rowan, 1977; Zucker, 1977; DiMaggio and Powell, 1983). Interestingly, both strands dominate again in the US and Europe respectively and are, therefore, also labeled American Institutionalism and European Institutionalism (Tempel and Walgenbach, 2003; Geppert et al., 2004). Although Institutional perspectives with their strong focus on the social constitution of organizational forms suggest themselves for analysing questions of organizational hybridization, early Institutional contributions were reluctant to do so. In fact, both the American and the European Institutionalism neglected the conceptualization of hybrid organizational forms and practices. This initial neglect has basically two reasons: first, a relatively modest concern with MNEs and their subsidiaries as special kinds of organizations facing a particular institutional contextual complexity, and second, different contextual frames of reference, with organizational fields being the main frame of reference on the American side and the nation on the European side. However, more recently contributions from both research traditions started to focus on the MNE and asked what happens when organizations face ‘institutional duality’ (Kostova and Roth, 2002) or are ‘organized across institutional divides’ (Morgan, 2001a).

American Institutionalism American Institutionalism or New Institutionalism is crucially defined by the seminal contributions of Meyer and Rowan (1977), DiMaggio and Powell (1983), Zucker (1991) and Scott (1995). One basic idea is that processes of isomorphism in organizational fields lead to an increasing homogenization of organizational structures, forms and even practices (DiMaggio and Powell, 1983). Regarding the question how transfer outcomes differ, early contributions in the American Institutionalism suggest that the diffusion, i.e. transfer and imitation of organizational forms and practices within structured organizational fields is not only possible but widespread. In this perspective there is only little room for concepts about alternations of what is being transferred. With respect to

18 Hybridization of MNE Subsidiaries

the question why such outcomes occur, the focus is on the unifying forces of the field and field embeddedness of the organization. Essentially, the shared institutional context and pressures facing organizations in the same field explain the diffusion of organizational forms and practices across organizations. In this context, the work of DiMaggio and Powell (1983) is most detailed in mapping these unifying institutional pressures, comprising of: coercive, mimetic and normative forces. (DiMaggio and Powell, 1983). The lacking consideration of hybrid or novel outcomes in the early American Institutionalism can probably be explained by the neglect of MNEs that straddle by their very nature substantially different (national) institutional contexts or fields (see also Temple and Walgenbach, 2003). To be fair, even early approaches from American Institutionalism cast doubts over the question whether a neat diffusion and imitation of organizational forms can be expected under all circumstances. They recognize that organizations may face different or even contradictory institutional pressures, which may be resolved by decoupling (Meyer and Rowan, 1977) or ceremonial adoption (DiMaggio and Powell, 1983) (c.f. Oliver, 1991). By the same token, DiMaggio (1988) admits that initial contributions from New Institutionalism did not pay attention to results short of perfect diffusion or institutionalization. More recently, New Institutional thought was taken up and applied to the MNE (e.g. Westney, 1993; Rosenzweig and Singh, 1991; Kostova, 1999; Kostova and Roth, 2002). In these contributions the main level of analysis shifts from the organizational field to the organization. In contrast to earlier contributions, this work explicitly addresses the questions: (1) how the subsidiaries of MNEs are contextually constituted given their embeddedness in different institutional contexts, conceptualized as ‘dual pressure’ (Rosenzweig and Singh, 1991) or ‘institutional duality’ (Kostova and Roth, 2002); and (2) how ‘institutional distance’ (Kostova, 1999) between transfer origins and destinations impact the adoption of such transfers in MNEs. With regard to the question how- and why transfer outcomes may not reflect perfect imitation and full diffusion these contributions also come to more differentiated answers. However, with the exception of Westney’s work (to be presented below) most contributions in this body of literature adopt a dualistic perspective to capture the contextual constitution of subsidiaries or transfer outcomes. Specifically, it is mainly seen to vary between different degrees of parent/home company patterns or local/host context patterns with little in-between. Regarding the question why such patterns beyond simple diffusion and imitation occur, it is argued that MNEs and their subsidiaries face different – oftentimes contradicting isomorphisms – that pull in different

Theoretical Foundations 19

directions. Which side prevails is then linked to set of local/host contextual variables and organizational variables – including strategic choices at the corporate and subsidiary level. The work by Rosenzweig and Singh (1991) and Kostova and Roth (2002) are exemplary of this body of research (see also Rosenzweig and Nohria, 1994; Tempel et al., 2006; Björkman et al., 2007). Kostova and Roth (2002) examine, for example, the adoption of organizational practices by subsidiaries of MNEs under conditions of ‘institutional duality’. They specify this condition as one where ‘each foreign subsidiary is confronted with two distinct sets of isomorphic pressures and a need to maintain legitimacy within both the host country and the MNC (Kostova and Roth, 2002, p. 216)’. Kostova and Roth (2002) investigate to which extent organizational practices mandated for implementation by the parent are actually implemented and internalized by MNEs’ subsidiaries and their employees (c.f. Kostova, 1999). The authors differentiate implementation and internalization levels between high or low and come to identify in their empirical study four different patterns of adoption: ‘active’ (implementation: high/internalization: high), ‘minimal’ (implementation: low/internalization: low), ‘assent’ (implementation: low/ internalization: high) and ‘ceremonial’ (implementation: high, internalization: low). While this approach shows that organizational practices transferred can be adopted by subsidiaries to different degrees, it tells us again little about qualitative changes in what is transferred. Such a research design does not allow discovering hybrid or novel organizational solutions because outcomes are again dichotomously pre-defined. Concerning the question why different degrees of implementation and internalization occur, Kostova and Roth (2002) reason that the successful practice adoption by subsidiaries is crucially related to two factors: (1) the ‘institutional profile’ of the host country and (2) the ‘relational context’ in the MNE. The first aspect suggests that the transfer success depends on the ‘favourability’ of the host countries’ regulatory as well as the cognitive and normative institutional profile. The concept of ‘institutional favourability’ builds on the concept of ‘institutional distance’, which Kostova (1999) developed earlier. Kostova (1999) defines ‘institutional distance’ as ‘difference between the institutional profiles of the two countries – the home country of the practice and the country of the recipient organizational units’ (Kostova, 1999: 316).The second aspect, the ‘relational context’ refers to the relation between the subsidiary and the parent. In this respect Kostova and Roth (2002) refer to organization or firm level factors (c.f. Kostova, 1999) and hypothesize that the level

20 Hybridization of MNE Subsidiaries

of dependence, trust, and the identification of the subsidiary with the parent organization crucially impacts the transfer success. In their empirical study of 104 subsidiary locations in ten countries, Kostova and Roth (2002) test their explanatory model and confirm that transfer success – practice implementation and internalization – varies across foreign subsidiaries depending on the institutional context in the host country and the relational context in the MNE.

Seminal contribution: Westney Westney (1987, 1993) is discussed here because her work deeply theorizes different transfer outcomes and the contextual conditions that give rise to such outcomes. On the one hand, her work (Westney, 1993) goes beyond other New Institutionalism inspired work in that it stresses innovations as likely transfer outcomes. On the other hand, particularly Westney’s early work (1987) goes beyond much work on hybridization in that it offers a comprehensive and systematic account of the different adaptation modes organization can engage in if a receiving context does not fit the requirements of organizational forms and practices to be transferred. We shall look briefly at these two contributions respectively. Westney (1993) suggests that Institutionalist thought and the idea that MNEs straddle different organizational fields can make an important contribution to explain the emergence of innovations. She identifies two reasons why innovations emerge. First, innovations emerge, ‘when an organizational pattern institutionalized in one field is introduced into another’ (Westney, 1993, p. 64). This outcome occurs, when organizational practices originating in one institutional context, are transferred to another and modified to fit the isomorphic pulls of a new context. As we will see below, there are unintended as well as intended factors that can cause transferred systems to change in a new context. Second, innovations emerge, ‘when conflicting isomorphic pulls produce new structures or processes’ (Westney, 1993, p. 65). In this scenario, there is not necessarily a transfer effort. A new structure or process simply emerges because an organization responds to different institutional demands. While Westney (1993) underlines that not all changes of organizational forms justify labeling them an innovation, pure imitations are the least likely outcomes to occur in processes of transfer. Now, in an earlier historical analysis, Westney (1987) explores in more detail why such innovations or departures occur within the context of system or model transfer. Specifically, in her work Imitation and Innovation: The Transfer of Western Organizational Patterns to Meiji Japan she asks what happened to Western organizational forms, when

Theoretical Foundations 21

they were transferred to the Japanese environment in Meiji Japan. Her major finding is that the transfer of organizational forms across societal contexts always brings about departures from the original model, for deliberate or unintended reasons. Unintended departures result from ‘imperfect information’ about the original model and ‘implicit alternative concepts’ of the actors in the receiving context (Westney, 1987). Thus, even if a perfect imitation is sought, lacking information and different local readings and interpretations cause changes in the implementation of the model. With respect to deliberate departures, Westney (1987) identifies three kinds, including: ‘selective emulation’, ‘adapting the patterns to different societal scales’ and ‘adapting the new organization to an environment that lacks some of the organizations that support it in the original setting’ (Westney, 1987, p. 25). A first reason for deliberate model departure lies in its ‘selective emulation’ due to possible conflicts with valued local patterns or a preference for only some aspects in the original model. Whichever way, it means that the transferred model is adapted to the local context, as certain elements are deselected or eliminated. As a result the transferred model undergoes change, when implemented in the new context. A second reason for model departure is rooted in different geographic and demographic scales of the receiving environment. In this respect, Westney (1987) essentially introduces contingency variables to explain model adaptation. However, the third and main reason for deliberate model departures is seen in different organizational environments in the receiving context. Westney (1987) argues that organizational forms are embedded in specific ‘organization-sets’ in their home environment. If transferred, the new host environment may either lack or differ in terms of organization-sets required. In the face of this environment misfit organizations have four principle options. They include: ‘elimination’, ‘internalization’, ‘functional equivalents’, and ‘organization creation’. In all four scenarios there is initially a different or lacking organizational context standing in the way of a full replication of the original organizational model. The elimination solution is about doing without or not transferring a certain element because the outside organizational context lacks the conditions needed for a transfer. As a result of elimination there is a departure from the original model. Internalization is about replicating a model by internalizing parts of the organizational context, which are in the original model placed in the organizational environment. In this scenario, the original model is also changed given the adaptations required for the internalization of the lacking external context. The functional equivalent solution implies less change

22 Hybridization of MNE Subsidiaries

to the original organizational model. Although the host context does not provide the exact organizational environment required, functional equivalents are available in the host context. Finally, in the organizationcreating solution, the external organizational context is again unfit for a replication of the original model. However, in this scenario an effort is made to create the needed context in the host environment. In this case, there is the least amount of departure from the original model. This implies, however, some change in the host organizational context. Westney’s approach essentially suggests different adaptation modes ranging from an adaptation of the transferred model – to fit the local/host context – to an adaptation of the local/host context – to fit the transferred model. However, as Westney’s (1987) historical analysis does not deal with model transfer in MNEs, she is not concerned with how strategic choices at different levels impact transfer outcomes. However, the strength of her approach lies in conceptualizing transfer outcomes other than imitation or localization and above all contrasting modes of adaptation under conditions of contextual misfit.

European Institutionalism European Institutionalism tends to be equated with the Varieties of Capitalism literature (Whitley, 1999; Hall and Soskice, 2001). However, not all European Institutionalists can be placed or like to be placed under the label of Varieties of Capitalism (e.g. Sorge, 2004). In contrast to American Institutionalism, European Institutionalism is more diverse. Tempel and Walgenbach (2003) state, for example, that European Institutionalists differ widely in focus and terminology. Major approaches to be subsumed under the broad label of European Institutionalism include the ‘societal effect approach’ (e.g. Maurice et al., 1980; Sorge, 1991) the ‘national business system’ approach (e.g. Whitley, 1992), the ‘industrial order’ approach (Lane, 1994) and the ‘social systems of production’ approach (Hollingsworth and Boyer, 1997). However, despite all diversity, European Institutionalists generally focus on the importance of national institutional settings and logics and posit these contexts as crucial for the contextual constitution of organizations. In line with this focus, the main analytical level tends to be the nation. European Institutionalists also share defiance against propositions about global convergence or statements denying the ongoing relevance of national institutional systems in how business is conducted (Tempel and Walgenbach, 2003). In fact, some European Institutionalists even argue that national distinctiveness not only persists despite of globalization but that globalization itself increases and generates it (e.g. Sorge, 2005).

Theoretical Foundations 23

Like American Institutionalist, European Institutionalists are initially not particularly concerned with organizational hybridization. European Institutionalists neglect organizational hybridization because they focus on the national distinctiveness of institutional systems and their organizations (mainly in comparative studies). This implies little concern for organizations straddling different institutional contexts, such as MNEs. Moreover, European Institutionalists are initially only little concerned with cross-border transfers of organizational practices. Even where such transfers are considered, national institutional forces are seen as overriding the foreignness of imported organizational practices (e.g. Sorge, 1995a). In other words, these early approaches suggest that transfers mainly revert to local/host context solutions, due to the overriding power of national institutional settings. However, since the late 1990s European Institutionalists have started to discover the MNE. They pay increasing attention to MNEs and ask: How do national institutional contexts shape the strategies and structures of firms that are organized ‘across institutional and national divides’? (Morgan, 2001a: 1). Like their American counterparts, European Institutionalists focus on the question (1) how MNEs and their subsidiaries are contextually constituted, given their embeddedness in different institutional contexts, and (2) how different institutional contexts between transfer origins and destinations impact the transfers of organizational forms and practices in MNEs. What is more, not only has it been realized that MNEs and their subsidiaries are impacted by institutional systems in complex ways, but it has also been questioned whether national institutional systems suffice to account for the contextual constitution of MNEs (e.g. Théret, 1997; Maurice, 2000; Morgan, 2001c). Approaches in this body of research have either focused on MNEs as a whole or subsidiaries as their main level of analysis. While at the MNE level the contributions mainly ask how the entire MNE is constituted, given their embeddedness in different institutional contexts, the subsidiary level analyses are more intimately concerned with the question how practice transfers across borders are impacted by institutional differences. Regarding the former there is a substantial disagreement whether hybrid solutions can be expected to emerge at the level of the MNE (e.g. Lane, 2001; Morgan, 2001a,b,c; Whitley, 2001). Whitley, for example, holds the development of hybrid MNEs unlikely due to the pervasive power of the MNEs home countries’ business systems. This contrasts with Morgan (2001a,b; see also Morgan and Whitley, 2003) and Lane (2001) who see much more scope for MNEs to develop into hybrids or into entities detached from any one particular business system. They see

24 Hybridization of MNE Subsidiaries

cross-national transfers and interaction across sites as potential drivers for the emergence of organizational forms that reflect either different institutional contexts or even constitute a context or institutional level in its own right. It should also be noted that strategic choices at the corporate level are considered to some extent in these contributions? These choices, however, are either seen to be determined by home-country institutional effects (Whitley, 2003) or not systematically conceptualized and empirically tested (Lane 2001; Morgan 2001a,b). Regarding the subsidiary level studies, there is little disagreement about the possible emergence of hybrid organizational forms. In this stream transfer outcomes typically vary between different levels of implementation. Although transfer outcomes are captured in different ways, a common strength of these subsidiary level approaches is their fine grained description of how these hybrid forms look like. Contributions typically examine on the micro-level how actors and their interactions (including conflicts) constitute the transfer outcomes and how different transfer outcomes, in turn, can be explained by the actor’s different institutional embeddedness (e.g. Sharpe, 1997; Lorenz, 2000; Ferner et al., 2001; Saka, 2003; Becker-Ritterspach, 2005). Over all, with the exception of Boyer’s contribution (see discussion below), European Institutionalists are weak in defining and typifying systematically a full range of different hybridization outcomes. What is more, few studies (a notable exception of Boyer) look at the transfer of entire production systems. Concerning the question why different transfer or hybridization outcomes occur, European Institutionalist focus on the condition that MNEs are embedded in different institutional environments and that these institutional environments strongly differ from one another (e.g. in strength, coherence, etc.). As practices and organizational forms grow out of specific institutional contexts and as institutional context differ substantially from one another, transfer outcomes are bound to differ. However, it is noteworthy that the approaches vary with regard to the level at which the relevant institutional context is located. Clearly, the national level has become but one level being looked at. Increasingly supra-national and regional contexts come into the focus. Also the level of the organization or firm and the nature of the transfer content itself receive growing attention (e.g. Sharpe, 1997; Lorenz, 2000; Saka, 2003; Edwards et al., 2005; Ferner et al., 2005). Nevertheless, different institutional contexts remain the core focus to explain why organizational forms and practices change when transferred in MNEs. In contrast, European Institutionalist pay little attention to the question how different task environments or supply and demand market

Theoretical Foundations 25

conditions in different markets impact transfer propensities, adaptation requirements and thereby hybridization outcomes. With the notable exception of a few, mostly recent contributions (Colling and Clark, 2002; Quintanilla and Ferner, 2003; Edwards and Kuruvilla, 2005; Almond et al., 2005),2 the embeddedness of the subsidiary in the organizational context of the MNE and the impact of strategic choices at the MNE or subsidiary level are not addressed systematically and empirically researched by European Institutionalists.

Seminal contribution: Boyer Boyer’s hybridization approach (1998) is selected here because it is probably one of the most systematic and comprehensive approaches with regard to capturing the how and explaining the why of production system hybridization. Boyer’s approach takes its starting point in a critique of ‘One-Best-Way’ claims of Lean-Production models. Although Boyer admits that certain production models (e.g. Fordist, Sloanist, Toyotist) achieved epochal dominance, he underlines that different production models have always coexisted in any given period of time. This coexistence has been and will remain possible because models of production are not universally superior but contextually optimized. In this view, a concept of global model-diffusion is misguided because transferring a model from one institutional context to another requires adaptations, calling into question its universal applicability. Boyer focuses on different transfer outcomes when MNEs transfer their production systems or rather productive models across borders. His approach not only allows for contextually mixed organizational forms – hybrids – but for a whole range of outcomes including: imitation and different modes of hybridization (see Table 2.1). He also

Table 2.1 Boyer’s four main types of hybridization Nature of process Breadth of process

Imitation

The search for a functional equivalent (FE)

Novelty (N)/ innovation

Partial: some components

Imitation 1 Partial imitation

Hybridization 1 Partial FE

Hybridization 3 Partial N

Complete: all components

Imitation 2 Complete imitation

Hybridization 2 Complete FE

Hybridization 4 Complete N

Source: Boyer, R., ‘Hybridization and Models of Production, Geography, History and Theory’ (1998) fig. 2.6, p. 35. (By permission of Oxford University Press).

26 Hybridization of MNE Subsidiaries

considers the possibility of complete transfer failure. By developing a taxonomy along the dimensions ‘nature of process’ and ‘breadth of process’ Boyer refines the concept of hybridization. In a first step, he distinguishes ‘straightforward imitation’ from hybridization and further distinguishes hybridization depending on whether it rests on functional equivalents or the emergence of novelty. ‘Functional equivalents’ refer to the replacement of specific practices of a model with host context practices, while the general principle of the model remains untouched. ‘Novelty’ or ‘innovation’, refer to the situation when a transplant comes up with principles and practices that neither show resemblance to host context practices nor to the principles and practices of the original production model (Boyer, 1998). Boyer, finally, distinguishes hybridization on yet another dimension which is the ‘breadth’ of the hybridization process. This indicates that imitations, functional equivalents as well as novelties can comprise of all, or only some of the components, of a given production model. In a second step, Boyer links the different hybridization outcomes to two core variables. He argues that different transfer outcomes are the result of the compatibility or incompatibility between the ‘requirements of a model of production’ of a firm and the ‘constraints and opportunities of the local institutions’ of the host space (Boyer, 1998, p. 34). Boyer indicates that the factors which influence the likelihood of particular outcomes are, on the one hand, related to the firm’s production model, i.e. how clearly defined and complimentary its principles and practices/routines are (i.e. dependence of transfer components on one another and context) and, on other hand, related to the institutions of the host space, i.e. how strong, coherent, and homogeneous institutional configurations are. By correlating the variable institutional context conditions with the variable production model, Boyer constructs a framework of likely trajectories of hybridization (see Table 2.2). It is important to note that Boyer understands hybridization as process of permanent adaptation and learning. Specifically, a ‘process of hybridization follows, at first a static order to get the factory up and running, and then [a] dynamic order to maintain and improve its initial performance and to respond to economic and social changes peculiar to a new space’ (Boyer, 1998, p. 38). Boyer’s approach is particularly helpful for this research context because it offers a comprehensive and systematic framework to capture production system hybridization. However, the framework suffers from a substantial weakness. We can criticize Boyer’s approach, albeit on a lower-level, on the same grounds he criticizes the Lean Production

Table 2.2

Nature and likelihood of hybridization in different industrial models and national institutions

National institutions of host space Firm’s model of production

Weak and heterogeneous

Rather strong, compatible with some diversity

Strongly coherent and homogeneous

Precisely defined principles and routines

Transplantation possible with some minimal congruence

Uncertainty because of the discrepancy between the profit strategy and national institutions

Conflict between the profit strategy and the national institutions of the host country

Clear principles, but some flexibility of routines

Partial hybridization is likely (same principles but different routines)

Hybridization as a functional equivalent is possible owing to some degree of freedom in the host institutions

Pressure towards hybridization as innovation: restructuring of the objectives and routines of the firm induced by the institutional environment

Neither principles, nor routines are strongly implemented

Likely incoherence and non-viable productive model

A new trajectory is possible in response to the institutions of the host country

Similarity of industrial model to that already implemented by firms of the host country

Source: Boyer, R., ‘Hybridization and Models of Production, Geography, History and Theory’ (1998) fig 2.8, p. 38. (By permission of Oxford University Press).

27

28 Hybridization of MNE Subsidiaries

approaches. For Boyer’s work rest on the assumption, that automobile MNEs have something like a defined a ‘One-Best-Productive Model’ or template which they always seek to transfer. While it cannot be denied that different MNEs follow different generic product strategies, it is doubtful if all automobile MNEs come to develop a specific productive model and seek its transfer to every institutional and strategic context in which they operate. Such an assumption disregards that MNEs, related to their divers strategic and institutional embeddedness, and related their different brands, strategic choices and international division of labour, may not have a transfer intent in the first place (see also Pries, 2003).

International Business perspectives The International Business (IB) perspective has like no other perspective furthered our understanding of MNEs, why they come into being, and how they are organizationally configured. While the explanation of why MNEs come into existence was strongly associated with economic paradigms (e.g. Hymer, 1976; Dunning, 2000), the study of organizational design and their relations to strategic contexts tented to be developed in corporate strategy and business policy literature (Ghoshal and Westney, 1993). The latter body of research has also come to be known as the Environment Structure Strategy Paradigm. This section will focus exclusively on this literature because it systematically relates the organizational configuration of MNEs to strategy and strategic contextual conditions. At the heart of the perspectives lies the assumption that there has to be ‘a good fit between strategy and environmental demands, and between organizational structure/processes and strategy’ (Harzing, 1999, p. 31). From relatively early on, this body of literature recognizes that MNEs and their subunits need to organizationally respond to internal and external contextual complexity (Westney and Zaheer, 2001). Thus, with regard to the contextual constitution of MNEs and their subsidiaries or different transfer outcomes this body of research considers mainly internal (mainly task environmental aspects) and external contingencies (mainly market related aspects). In the following paragraph a selective review of a variety of perspectives from IB is presented. Moving from the corporate to a subsidiary level of analysis, the main perspectives discussed here include: the Integration– Responsiveness framework, the related MNE Process Strand, the Strategic Role and the Knowledge Flow strand.

Theoretical Foundations 29

Integration–Responsiveness framework The Integration–Responsiveness framework can be seen as the mother of the Environment Structure Strategy Paradigm in the IB literature. The framework is introduced by Prahalad (1975) and subsequently taken up by a number of scholars. It rests on the contingency theory as presented by Lawrence and Lorsch (1967), that is, on the idea that firms face two fundamental environmental forces, i.e. pressures for differentiation and pressures for integration. Translated to the MNE, these pressures are labeled by Bartlett and Ghoshal (1998) ‘forces for global integration’ and ‘forces for national differentiation’ and by Prahalad and Doz (1987) ‘pressure for global integration’ and ‘pressure for local responsiveness’. Based on the integration-responsiveness framework (high/low global integration and high/low responsiveness) Prahalad and Doz (1987) distinguish multidomestic (low/high), international (medium/medium), global (high/low) and transnational (high/high) industries, strategies and structures. While the level of global integration represents the need for central or global coordination, the level of local responsiveness expresses the need for local approaches. However, the Integration– Responsiveness framework is not only applied at the level of industries or firms. Bartlett (1985) show that, even within the same company, some functions and tasks are more subject to global integration while others are more subject to local responsiveness. With regard to the question how and why MNEs are differently contextually constituted or how and why transfer outcomes differ, this body of research suggests, similar to the New Institutionalist contributions, dichotomous outcomes. Specifically, MNEs, subsidiaries and even different tasks with an MNE are likely to reflect either more local patterns or more global patterns. How a global pattern is origin-wise constituted remains open, however. Nevertheless, what kind of pattern prevails, is basically dependent on the relative importance of the different environmental pressures for different Industries, MNEs, subsidiaries and tasks or functions.

Different internationalization strategies in the MNE Strategy Structure stream Different authors in the IB literature come to relatively similar descriptions of what alternative internationalization strategies firms can choose (see Harzing, 1999 for a good overview). Bartlett and Ghoshal’s (1998) work can be viewed as typical for this body of literature. The main level of analysis is the MNE as a whole. Building on the Integration– Responsiveness framework, Bartlett and Ghoshal (1998) posit that MNEs

30 Hybridization of MNE Subsidiaries Table 2.3

Distinct environmental forces and their sources

Forces for global integration

Forces for local differentiation

Forces for worldwide diffusion

• Unified world market place • Convergence of consumer preferences • Market liberalization • Economies of scale • Reduced transportation and communication cost

• Nationally differentiated market structures • Difference in consumer preferences • Host government policies • Liability of geographic distance for transport and coordination • Flexible production technology

• Increasing parity among players • Rising R&D costs • Shortening product life-cycles • A shifts from freestanding products to integrated systems • Global standards and specifications

Source: Compiled and adapted from Bartlett and Ghoshal (1998).

need to respond to three distinct environmental forces to different degrees: ‘forces for global integration’, ‘forces for local differentiation’ and ‘forces for worldwide diffusion of knowledge’ (see Table 2.3). These environmental forces translate into three distinct and at times contradictory strategic needs for global efficiency, local responsiveness and worldwide diffusion of innovations. Based on MNEs responses to these different strategic needs, Bartlett and Ghoshal (1998) distinguish four distinct internationalization strategies including: the multinational, the global, the international and the transnational solution. While the multinational, the international and the global MNEs develop strategies that focus respectively on national responsiveness, global integration and transfer of knowledge; the transnational solution is able to respond to all three strategic needs at the same time. Although Bartlett and Ghoshal (1998) do not directly address the question how and why MNEs or subsidiaries are differently contextually constituted, we can infer from their work that depending on the relative strength of different environmental forces and the corresponding strategies, MNEs and their subsidiaries reflect either more local patterns or more global patterns. In the case of MNEs following a transnational strategy the contextual constitution transfer outcomes may be more complex, however, as the company responds simultaneously to different environmental forces. As different subsidiaries have different assets and capabilities, they assume different strategic roles and receive and distribute knowledge to different degrees. Therefore, some subsidiaries and functions will be tightly centralized and globally integrated reflecting global patterns. Others,

Theoretical Foundations 31

in turn, remain decentralized reflecting local context solutions. In this scenario different subsidiaries and different functions within subsidiaries can be very differently contextually constituted. Dichotomies between local vs global/parent, dependent vs independent give way to a universe of national or even transnational contextual influences (e.g. home, host and third countries context influences flowing together) and a whole spectrum of relations, ranging from dependence to interdependence (Bartlett and Ghoshal, 1998; see also Perlmutter’s concept of geocentric MNE). In fact, in the transnational the possibilities of contextual constitutions become so complex that the focus has to shift to subsidiaries’ particular roles.

Different strategic roles of subsidiaries To understand MNEs as transnational types or differentiated networks implies that subsidiaries take on different strategic roles. The Strategic Role stream in the IB literature, sets out to understand such differences in subsidiary roles (e.g. White and Poynter, 1984; Bartlett and Ghoshal, 1986; Ghoshal and Nohria, 1989; Jarillo and Martinez, 1990; Birkinshaw and Morrison, 1996; Taggart, 1997; Beechler et al., 1998) and their evolution (e.g. Birkinshaw and Hood, 1998; Chang and Rosenzweig, 1998; Delany, 1998; Peters, 1999; see Birkinshaw, 2001 for an overview). Particularly in the latter contributions, the subsidiary is increasingly seen as a unit that is ‘not just an instrument of the parent, but has certain degrees of freedom in shaping its own destiny’ (Birkinshaw, 2001, p. 383). Accordingly, the subsidiary is the main unit of analysis. Regarding subsidiary role definitions, many typologies are based on at least two of the following three aspects: the nature of the subsidiary’s local/host context, the nature of the subsidiary’s wider corporate context and the nature of the subsidiary’s resources and capabilities (e.g. Ghoshal and Nohria, 1989; Jarillo and Martinez, 1990; Dörrenbaecher and Gammelgaard, 2006). Birkinshaw and Hood (1998) and Birkinshaw (2000) probably most extensively explore subsidiary roles and their evolution. They adopt a network conceptualization of the MNE and model the subsidiary ‘as a semiautonomous entity, capable of making its own decisions but constrained in its action by the demands of head-office managers and by the opportunities in the local environment’ (1998, p. 780). A cornerstone of their concept is seeing subsidiary roles as based on two dynamic aspects, comprising their stocks of resources/capabilities and their charter. The charter is seen to be the visible manifestation of a subsidiary’s role and defined ‘in terms of the markets served, products manufactured,

32 Hybridization of MNE Subsidiaries

technologies held, functional areas covered or any combination thereof’ (Birkinshaw, 2000, p. 86). The Charter is typically a shared understanding between the subsidiary and the HQ regarding the subsidiary’s responsibilities. (Birkinshaw, 2000, p. 86) The relationship between the charters and resources/capabilities is a problematic one. Equilibrium between the two aspects is the exception, rather than the rule. More often than not, there will be mismatches between capabilities/resources and charters. Birkinshaw and Hood (1998) see the development of capabilities/resources and charters, in other words, the development of subsidiary roles, as driven by three contextual factors. These include: parent company factors, subsidiary factors, and host country factors. Parent company factors comprise the nature of the competitive internal resource allocation, the decentralization of decision making, and the ethnocentrism of parent management. They also refer to this factor as ‘head-office assignment’ which essentially refers to the task or charter assigned to subsidiaries. Subsidiary factors include the track record of the subsidiary, the credibility of subsidiary management, the entrepreneurial orientation of subsidiary employees, and the contestability of subsidiary’s existing charter. This dimension refers to abilities of and initiatives taken by the subsidiary to extend the charter. Such initiatives can be the development new capabilities/resources or ceasing new market opportunities. Host country factors, as the third aspect, include the strategic importance of the country, the relative cost of factor inputs and the dynamism of local environment. These factors are also summarized as ‘local environment determinism’ and refer to local market conditions (including customers, competitors, suppliers as well as government agencies) posing opportunities and constrains for the subsidiary’s role development. Now coming back to the importance of this stream for subsidiary hybridization, one can easily imagine that the contextual constitution of a subsidiary is impacted by role defining factors at the parent company, subsidiary and host context level. For example, the resources and capabilities held by the subsidiary may play an important role for the transfer propensity or adaptation requirement. By the same token, the division of labour in an MNE and the task assigned by the parent may play an important role whether or not subsidiaries are similar enough to transfer production system across units. Similarly, differences in the strategic conditions of host context may make transfers difficult and may call

Theoretical Foundations 33

for adaptations. Thus, although this body of literature does not explicitly deal with the question how and why the hybridization profiles of subsidiaries differ (with the notable exception of Beechler et al. (1998) to be discussed below), strategic role defining factors can be assumed to influence transfer and adaptation requirements. Without being able to explore this idea further at this point, the basic rationale is that if we wish to understand how subsidiaries are contextually constituted, we need to consider their different strategic roles in MNEs. For it can be assumed that the strategic distance between subsidiaries in terms of strategic roles and role defining contexts crucially impact transfer propensities and adaptation requirements.

Knowledge flows in the transnational corporation The last body of approaches from IB literature reviewed is the Knowledge Flow strand. In contrast to the other IB strands, this body of research comes closest to dealing directly with the question how and why transfer outcomes differ in subsidiaries. Starting from the assumption that the creation, diffusion and adoption of innovations is the single most important strategic challenge to MNEs (e.g. Bartlett and Ghoshal, 1998), this body of research focuses on the enabling and constraining conditions of knowledge flows in MNEs (e.g. Gupta and Govindarajan 2000). The main level of analysis in this body of literature is the subsidiary, which is embedded in an MNE conceptualized as a ‘differentiated network’ (Doz and Prahalad, 1991; Nohria and Ghoshal, 1997). Now, with regard to the question how transfer outcomes differ, the knowledge flow studies typically measure knowledge flows or successful knowledge adoption in terms of patent citations (Almeida and Phene, 2004; Singh, 2004; Yamin and Otto, 2004), product introductions (Tsai, 2001) or alternatively by the presence of predefined kinds of knowledge (Gupta and Govindarajan, 2000; Hansen, 2002; Schulz, 2003). The main interest is the successful diffusion, adoption or application of pre-defined knowledge in subsidiaries. In contrast, there is hardly any concern for potential transformations or alternations resulting from knowledge flows. Over all, there is little concern for hybrid organizational outcomes as the transfer outcome perspectives are largely dichotomous, i.e. between successful and failed transfer. Regarding the question why transfer outcomes differ, the Knowledge Flow literature focuses mainly on how complex organizational characteristics of MNEs, the characteristics of the knowledge transferred, and the knowledge-related characteristics of sending and receiving subunits, impact knowledge flows in MNEs. Specifically, frequently cited factors

34 Hybridization of MNE Subsidiaries

include: (1) The characteristics of the sender unit (Szulanski, 1996; Foss and Pedersen, 2002; Gupta and Govindarajan, 2000); (2) the characteristics of the receiving unit (Szulanski, 1996; Foss and Pedersen, 2002; Gupta and Govindarajan, 2000; Minbaeva et al., 2002; Tsai, 2001); (3) the characteristics of intra-organizational or relations context in the MNE (Szulanski, 1996; Almeida and Phene, 2004; Björkman et al., 2004; Foss and Pedersen 2002; Gupta and Govindarajan, 2000; Hansen, 1999, 2002; Hansen and Lovas, 2004; Teigland et al., 2001; Tsai, 2001); (4) the characteristics of the transferred knowledge (Szulanski, 1996; Foss and Pedersen 2002; Hakanson and Nobel, 2000; Hansen, 1999, 2002; Kotabe et al., 2003; Schulz, 2003); and (5) the characteristics of a unit’s business and technological environment as well as external/local network relations (Almeida and Phene, 2004; Forsgren et al., 1999; Foss and Pedersen, 2002; Frost, 2001; Mudambi, 2002; Pearce and Papanastassiou, 1999; Kotabe et al., 2003; Yamin and Otto, 2004). While the Knowledge Flow literature mainly focuses on the impact of organizational factors on knowledge processes, it also considers host context factors to some extent. However, like in the Subsidiary Role literature, these are largely confined to strategic context conditions (e.g. technological richness). In contrast, there is only scant attention to the question how institutional contexts3 in home, host, third countries or beyond impact knowledge flows and their outcomes.

Seminal contributions: Beechler, Bird and Taylor Beechler, Bird and Taylor’s (1998) work ‘Organizational Learning in Japanese MNCs: Four Affiliate Archetypes’ is discussed here because it can be grouped into the IB literature – with its focus on different affiliation types – but stands out in that it explicitly focuses on hybridization and its causes. Beechler et al. (1998) build on the notion that subsidiaries face dual contextual pressures from the parent and the local/host context. Their main interest is the ability of MNEs to learn in and from their overseas affiliates. Their basic argument is that: As organizations expand overseas, they must, by necessity, establish new systems for managing that operation. These events present an occasion for organizational learning. (Beechler et al., 1998, p. 333) Beechler et al. (1998) link such learning opportunities to different affiliate types or models. However, while Beechler et al. (1998) are mainly interested in the learning link, their work has important implications for the

Theoretical Foundations 35

question how and why these different affiliate types – or concretely their management systems – are differently contextually constituted. Thus, given that this study is interested in how and why different hybridization profiles in subsidiaries occur, the following review focuses on those aspects only. Concerning the question how transfer outcomes or better the contextual constitution of subsidiaries can differ, Beechler et al. (1998) distinguish four types of affiliates. These include: the exportive, the closed hybrid, the adaptive and the open hybrid. The authors argue that the exportive model or affiliate essentially mirrors a parent template (the foreign parent context) and that the adaptive model mirrors a local template (i.e. the local/host context). In contrast, the closed and open hybrids involve major adaptations. While the closed hybrid is likely to reflect the parent context, the open hybrid is most likely to reflect a mix (local as well as parent organizational) of different contexts. Why do these different outcomes occur? Beechler et al.’s (1998) work implies that different transfer scenarios, different misfit scenarios, and different adaptation modes explain the different outcomes. First, the authors suggest that foreign (Japanese) affiliates either draw on a local or a parent template. Secondly, parent templates can be more or less fitting with the local environment. And thirdly, misfits can be overcome by either adapting the template or by selecting or buffering the template from the local context. Based on these different situations and options, Beechler et al. (1998) construct different MNE affiliate types (see Table 2.4). The exportive model describes a successful transfer of a parent template to a permissive local context where no adaptations are required. Table 2.4 A comparison of four learning cycle models Exportive

Closed hybrid

Adaptive

Open hybrid

Enactment

Parent template

Parent template

Local template

Customized temporary template

Selection

Minor adjustments

Major adjustments

Minor adjustments

Major adjustments

Retention

Verification of fit; fine-tuning

Identification of success consistent with self-image

Local success

Cautions success

Source: Beechler et al. in Birkinshaw and Hood, Multinational Corporate Evolution and Subsidiary Development (1998), table 13.1, p. 355.

36 Hybridization of MNE Subsidiaries

The closed hybrid model describes the transfer of a parent template that does not fit the host context. However, instead of changing the template the template is shielded from the host context. The authors call this buffering and give the example of a careful workforce selection. The adaptive model, in contrast, is not based on a parent template transfer. Instead, the subsidiary set-up is either based on a local template (with minor adjustments to parent requirements) or customized to fit the local environment. The open hybrid is, finally, based on either a parent or a local template. Difficulties or misfits between the subsidiary set-up and the local context are the starting point in this scenario. As problems of misfit are potentially attributed to both the existing set-up or the template used and the local context, adaptations may involve the adaptation of the local context as well as adaptations of the existing template of the site. It involves ‘[u]sing a customised template, modified from the parent and local models, and open to change once they encounter difficulties’ (Beechler et al., 1998, p. 352). In this scenario, customized templates emerge that may differ from typical local templates and parent templates. The model of Beechler et al. (1998) is interesting because it offers a full range of hybridization outcomes and relates these outcomes systematically to different transfer scenarios, contextual (mis)fits, and different recontextualization modes. We draw on this conceptualization in a refined version in the analytical framework below. However, while Beechler et al. (1998) consider strategic or firm factors impacting the transfer propensity or template selection, the contribution is very vague in terms of explicating the cause and nature of contextual misfits. Here the authors only loosely refer to ‘local environment’, ‘culture’ or ‘local needs’.

Approaches compared The discussion above has shown that each body of literature has different contributions to make to the question how and why transfer outcomes or the contextual constitution of subsidiary production systems differ. Before constructing the analytical framework, we briefly present a summary and comparison of the respective approaches’ strengths and weaknesses.

The Japanization stream The Lean Production perspective initially ignores the possibility of transfer outcomes other than successful transfer and imitation. Conversely, the Labour Process perspective is initially pessimistic about transfer

Theoretical Foundations 37

success and stresses the continuation of local patterns. However, even early contributions from both strands admit that transfer success or failure varies by different dimensions of production systems. More recently both strands adopt or embrace notions of hybrid or mixed outcomes as a result of transfer processes. With regard to the question why different transfer outcomes occur, there are more marked differences between the two bodies of literature. While the Lean Production strand initially plays down that contextual differences impact transfer propensities and outcomes, the Labour Process strand emphasizes all along that societal differences form barriers to transfer. Lean Production contributions, in contrast, only slowly and selectively refer to the relevance of institutional contexts for transfer outcomes. While adaptations to certain institutional conditions are not denied, their relevance as transfer constraints tends to be applied in an ad hoc fashion. Institutional conditions rarely form an integral part of an overall explanatory framework. In short, there is no coherent or elaborate concept for institutional contexts in this stream. This has led some to suggest that Lean Production perspectives take on a universalistic approach that ‘ignores the contextualities of the diffusion process’ (Saka, 2003, p. 37). However, this is not entirely true, particularly not with regard to more recent studies. The Lean Production strand pays increasing attention to the impact of those contextual differences that are related to sectoral or firm specific task environments and strategic choices. Excepting the work of Smith and Elger (2000), this attention is largely absent in the Labour Process strand. Overall, there is little concern with the question how different task environments and strategic choices impact hybridization outcomes. Moreover, the Labour Process strand does not pay much attention to the question whether and how firms can shape their environment, which is emphasized in Lean Production contributions. The latter see internationalizing firms not merely as reactive to new environments but rather as strategic actors able to select, change and create their environments. Let us finally look at some common strength and weaknesses of both strands. The Japanization literature has not least been discussed here because it focuses on the hybridization of whole production systems. This is of particular relevance for this research context. Both strands increasingly highlight that firms may differ with regard to their transfer propensity depending on different kinds of contextual conditions. However, both strands of the Japanization literature – probably somewhat more the Labour Process strand – share the weakness that they fail to look more closely how transfers, adaptations, and hybrid outcomes are shaped by particular conditions of the subsidiary embeddedness in the

38 Hybridization of MNE Subsidiaries

MNE and more generally by MNEs as special kinds of organizations. Both strands draw only little on the Environment Strategy Structure paradigm of the IB literature or on elaborate institutional concepts from Institutionalist approaches. The disregard for the IB literature implies that much of this work suggests that transfers are based on one transfer source (parent company), one transfer template (parent company) and one transfer driver (parent company), rather than seeing subsidiaries embedded in corporate networks, where transfer sources, templates and drivers can be diverse and dispersed.

Institutionalist approaches Institutionalist approaches are initially little concerned with hybrid organizational constitutions or hybrid outcomes as a result of transfers in MNEs. This neglect changes, however, over time. American Institutionalist increasingly look at different degrees of organizational practices transfer or adoption and are concerned with the question to what extent the organizational elements of subsidiaries reflect the foreign parent or local/host context. However, while American Institutionalist come to admit that simple diffusion and imitation is unlikely if organizational practices are transferred from one institutional context to another, there still is little effort to identify new or hybrid outcomes. Much of the work offers hardly more than dichotomies typologies for transfer outcomes. European Institutionalists are also critical about propositions of an easy transfer of organizational forms across borders. While it is difficult to make generalizing statements about this diverse body of literature, it is fair to say, that most proponents of this strand either acknowledge the possibility of the emergence of hybrid forms or describe such outcomes in detailed case studies. However, although European Institutionalists start to capture hybrid solutions, there is overall little effort to typify them systematically and comprehensively. Although European and American Institutionalist differ markedly in their definitions of institutions, both approaches show that institutional contexts and differences between them have a strong impact on transfer outcomes and thereby on the contextual constitution of subsidiaries. However, there are some important differences between the approaches. European Institutionalist are strong in understanding the interplay of different institutional levels – ranging from the actors’ level, the regional level, the national level to the supranational or transnational level – on transfer success. Such a multi-level analysis finds only little attention in the work of American Institutionalist. American Institutionalists, for their part, emphasize other levels of institutional influence which

Theoretical Foundations 39

have not found much attention in the work of European Institutionalists. The strength is here the focus on the dual embeddedness of subsidiaries in a parent institutional context and a local/host institutional context. This focus reflects American Institutionalism’s proximity to the IB’s Integration-Responsiveness framework. By the same token, American Institutionalism tends to refer more to the IB literature and generally adopts a stronger focus on the relevance of MNEs’ internal organizational context for transfer outcomes. This also implies that American Institutionalism develops a more complex understandings how strategic choices and relational contexts in MNEs impact transfer outcomes. While some European Institutionalists suggest the relevance of strategic choices for the contextual constitution of MNEs, the approaches are generally weak in systematically addressing relationships between strategic choices and relational conditions in MNEs, on the one hand, and transfers in MNEs and the contextual constitution of subsidiaries, on the other. All in all, there remains little attention to the enormous body of IB literature which suggests that different roles of subsidiaries in MNEs impact transfer propensities and outcomes. Thus, neither the European nor the American Institutionalism explores systematically how different task environments or roles of subsidiaries, as a result of transnational strategies and the global divisions of labour, impact hybridization profiles. In both Institutionalist strands, there is also little systematic reasoning how different transfer propensities/scenarios and different adaptation modes affect hybridization outcomes in production systems. This neglect is probably related to the fact that Institutionalist are generally not concerned with the transfer of entire production systems and show little concern for the ability of firms to shape their environments.

International Business Compared to the other bodies of literature presented, the International Business perspectives is least concerned with identifying different hybridization outcomes as a result of cross-contextual transfer. However, while the Integration–Responsiveness framework, the Strategy–Structure strand as well as the Subsidiary Role strand do not address transfer processes and outcomes in any detail, these approaches suggest different outcomes with regard to the contextual constitution of subsidiaries. These outcomes can be seen to vary between global/parent and local/host country solutions. Specifically, the Integration–Responsiveness framework can be read to suggest that subsidiaries contextual constitution differs between global and local across its different functions and tasks.

40 Hybridization of MNE Subsidiaries

The Strategy-Structure strand, in turn, suggests that the overall contextual constitution of subsidiaries can vary between parent/global or host/local context. While, the Subsidiary Role and particularly the Knowledge Flow strand suggest themselves to go beyond dichotomous outcome perspectives, there are again few if any fine-grained descriptions of hybrid (or even novel) transfer outcomes or forms of contextual constitutions. The potential strength of the IB literature for this research context is that it allows conceptualizing strong affiliations between different business contexts, strategies, task environmental conditions, on the one hand, and the contextual constitution/hybridization profile of subsidiaries, on the other. The Integration–Responsiveness framework sensitizes us for the condition that transfer attempts and hybridization profiles of subsidiaries depend on nature of the business environment, function or even task within a function. The MNE Strategy–Structure strand suggests that transfer outcomes and hybridization profiles depend on the internationalization strategy of MNEs. What is more, the concept of MNEs as ‘transnationals’ or ‘differentiated networks’ suggests that transfers flow in many directions and that subsidiaries receive (send) transfers from (to) many poles. Moreover, the Subsidiary Role strand suggests that different organizational roles and the increasing division of labour in MNEs render it difficult to develop and transfer the same template to many sites. For the approach implies that what is transferred to a subsidiary is strongly dependent on the subsidiary’s local task and business environment as well as its local resource and capability endowment. In contrast to the other strands in the IB literature, the Knowledge Flow strand looks closest at conditions that impact transfer processes and outcomes in MNEs. Regarding these conditions, the focus is mainly on the organizational characteristics and the nature of the knowledge transferred. Curiously, this strand of literature not only neglects the impact of different institutional or cultural environments but also fails to focus on the effect of different internationalization strategies on transfer outcomes. The main reason for this neglect lies in the assumption that most MNEs have become ‘differentiated networks’ following a ‘transnational strategy’. To summarize: The IB literature suggests that contextual constitution of MNEs and its subsidiaries is related to external and internal contingencies of the MNE. Overall, the strength of the IB literature rests with its elaborate inward focus on MNEs, which comes at the costs of not comprehensively theorizing environment relations. We learn little about how environmental pressures are translated into strategies and strategies

Table 2.5

Strengths and weaknesses compared across different approaches

Approach

How

Why

Dichotomous vs Complex

Contextual distance

Strategic choice

Institutional distance

Strategic distance

Corporate or subsidiary level

From dichotomous to increasingly complex

Weakly considered

Somewhat considered

Somewhat considered

From dichotomous to increasingly complex

Somewhat considered

Weakly considered

Weakly considered

Largely dichotomous

Strongly considered

Weakly considered

Somewhat considered

Largely complex descriptive

Strongly considered

Weakly considered

Somewhat considered

International Business Integration– Not explicitly considered but Responsiveness suggests dichotomous outcomes

Weakly considered

Strongly considered

Somewhat considered

Strategy Structure

Not explicitly considered but allows for complex outcomes

Weakly considered

Strongly considered

Strongly considered

Subsidiary Role

Not explicitly considered but allows for complex outcomes Mainly dichotomous outcomes

Weakly considered

Strongly considered

Strongly considered

Weakly considered

Strongly considered

Strongly considered

Japanization Lean Production Labour Process Institutionalist American Institutionalism European Institutionalism

Knowledge Flow

41

42 Hybridization of MNE Subsidiaries

into structures or processes. For the most part, the focus is on internal contingencies involving a fairly mechanistic approach to the relations between different organizational characteristics. While different market conditions, task environmental conditions, strategies and strategic roles lie at the centre of why MNEs and their subsidiaries are differently contextually constituted, there is practically no attention to their institutional constitution.4 Finally, IB literature is generally not concerned with production systems, their hybridization profiles and how such profiles are related to different generic product strategies and entry modes. There is also no systematic attention to possible transfer misfits and different adaptation modes when and where such misfit occurs.

Conclusion The foregoing discussion of different approaches shows (see Table 2.5) that no single body of literature discussed, systematically addresses the impact of both strategic and institutional contextual difference on production system hybridization. More importantly, while the importance of strategic choices is not entirely ignored, there is hardly any work that thoroughly theorizes and empirically researches the question how foreign/parent strategic choices on the corporate and subsidiary level interact with transfer scenarios, contextual (mis)fits/recontextualization pressures and recontextualization modes to influence hybridization outcomes/profiles of production systems in MNEs.

3 Subsidiary Hybridization – Conceptual and Methodological Tools

Introduction The unit of analysis of this research project is the production systems of four automobile subsidiaries in India. The core goal is to identify, compare and explain the hybridization profiles of their production systems. Instead of identifying the hybridization profiles by asking if predefined aspects of a production template or model (e.g. the Japanese) were successfully applied or adapted (e.g. Abo et al., 1994), this project turns the identification of hybridization profiles upside down by asking about the contextual origins of the organizational configuration on functionally predefined dimensions of a production system. In such an approach transfers and adaptations are still relevant for the identification of the profile. However, this approach allows seeing hybridization profiles also as a result of more or less deliberate non-transfer. If, for example, a certain dimension of a production system is not targeted by transfer, it may either be locally constituted or neither locally nor foreign constituted. This perspective stresses the possibility that firms do not always undertake the transfer of clearly defined production templates. Instead, they may be very selective with regard to what they transfer, depending on particular strategic choices and strategic and institutional contexts of their subsidiaries. The advantage of this approach is that instead of defining some abstract transfer model in advance that may or may not exist, a more open approach is taken. This involves defining functionally the core dimensions of a production system and identifying what contextual origin the different dimensions reflect.

43

44 Hybridization of MNE Subsidiaries

Production systems and hybridization outcomes defined Dimensions of a production system Defining a production system is no easy task as the question arises how comprehensively or detailed it should be defined. Trist stresses that researching a production system as a functioning whole always calls for a consideration of both the social and the technical components of a system (Trist, 1975). In the early works of the socio-technical school the term ‘primary work systems’ takes the centre stage. A primary work system is understood as an operative subsystem of an organization. It centres on a functional task that assures system reproduction trough input-output transformations. Most importantly, a primary work system consists of a technical and a social subsystem and the relation between the two is a major field of inquiry (Trist, 1981). By the same token, ‘work organizations’ are not just social systems but social and technical systems ‘where the substantive factors – the people and the equipment’ interact (Trist, 1981, p. 10). Moreover, the use of the system term is a deliberate choice to stress the various interdependencies inside the system as well as between the system and its environment (Trist, 1981). The latter mainly concerns input and output relations between the system and its environment. In its original focus the socio-technical system approach has only organizational subsystems – primary work systems – in mind. Subsequently, however, the idea of socio-technical systems is broadened and applied to organizations as a whole (Emery, 1959; c.f. Sydow, 1985). As Sorge (1993) has adopted more recently such a broader understanding of a ‘socio-technical work systems’ and redefined it in a comprehensive manner, we draw in the following definition on his work. A production system is defined in this study as a system that manufactures a specific product or a range of products, which constitutes its primary task (e.g. parts, components or finished products such as automobiles as in this research case). The production system comprises four main components: an organization structure, a process organization, a technical configuration and a human resource profile (c.f. Sorge, 1993). The organization structure indicates how comprehensive systems are differentiated by function, geographic location, products and markets into subsystems. The process organization indicates how inputs of various kind move through the system and are transformed into outputs. The process organization is constituted by the interplay between a production system’s technical configuration (physical infrastructure of plant, factory layout, aggregates, machines, tools, levels of mechanization,

Theoretical Foundations 45

automation and flexibility, temporal-spatial organization of material flows) and human resource profile (skill levels, qualification, attitudes). These together define the work organization. Moreover, in line with Sorge (1993), two crucial relational aspects are looked at that closely interpenetrate and partly constitute a production system. These are: work and labour relations and inter-organizational relations (mainly supplier relations). Work and labour relations appear important to include as part of the production system because constructive work and labour relations are seen as crucial for the functioning of modern production systems, particularly for the implementation of high workforce involvement practices. The level of inter-organizational relations looked at in this work are pragmatically derived. Given the research focus, the analysis is restricted to those relations that revolve around the primary task, that is, the core material inputs to the production process. Therefore, the concentration lies on major aspects of supplier relations of the automobile manufacturing sites. It involves both relations that may be under one corporate legal structure and relations that are constituted by legally distinct entities. The centre of attention is, therefore, on the character of internal and external supplier relations. Such a comprehensive consideration of supplier relations is particularly justified in the view that the success of automobile production systems is increasingly attributed to the organization of the entire supply chain (Taylor and Brunt, 2001). It should also be noted that the reshaping of the division of labour between final assemblers and suppliers, to the extent of integrating suppliers on site for final assembly, additionally blurs the line between an ‘external’ supplier and the core production system itself (if defined by spatial boundaries) (c.f. Camuffo, 2002) (Figure 3.1 summarizes the constitutive dimensions of a production system).

Different hybridization outcomes Asking how the contextual constitution of subsidiary production systems differs is about identifying different contextual origins in their configuration. The literature review above offers different concepts, how we can capture a subsidiary’s contextual constitution. Approaches from IB are utterly weak in typifying or describing to what extent organizational forms and practices reflect different contextual origins. For the most part, such typifications are dichotomous and vary between global and local outcomes with little explicit consideration about their origin. The, IB literature offers practically no conceptualization to capture hybrid organizational forms. The same holds true for American Institutionalism, with the exception of Westney’s (1993) work. European

46 Hybridization of MNE Subsidiaries Organization structure: Functional differentiation Hierarchical differentiation Process organization: Technical configuration Human Resource profile Organizational relations: Work and labour relations Inter-organizational relations Figure 3.1 Dimensions of a production system

Institutionalism in contrast, details specific hybrid outcomes but fails, with exception of Boyer (1998), to develop a systematic typology to account for different kinds of contextual constitution. While the Japanization literature, most notably the work of Abo et al. (1994), started to develop such typifications, it is probably Boyer (1998) who developed the most systematic and comprehensive classification to capture Table 3.1

Different hybridization outcomes

Contextual origin

Foreign/parent

Local/host

Yes No

Yes

No

Hybridization Localization

Imitation Customization/novelty

the different contextual origins of production systems. We will, therefore, draw on Boyer’s framework with some minor adaptation. In line with Boyer (1998), we can imagine four ideal–typical outcomes how subsidiary production systems and/or their dimensions can be contextually constituted (see Table 3.1). These ideal types include: imitation, hybridization, customizations/novelties and localization. Specifically, a subsidiary or the different dimensions of its production system can: resemble a foreign/parent context organizational-template (imitation); resemble a local/host context organizational-template (localization);

Theoretical Foundations 47

resemble both, a foreign/parent context organizational-template and a local/host context organizational-template (hybridization); resemble neither a foreign/parent context organizational-template nor a local/host context organizational-template (novelty/customization). It should be noted that the category foreign/parent can include foreign context patterns that do not originate from any site of the parent company. By the same token, the category local/host can include local patterns that are peculiar to a local site, but not typical for the host context.

Strategic and institutional distance, strategic choices and hybridization outcomes Transfer, (mis)fit and recontextualization Drawing on Beechler et al. (1998) and Westney’s (1987) work we argue that an understanding of different hybridization outcomes rests first of all on the dynamic relation between three crucial factors. These include: The kind of transfer scenario, the kind of (mis)fit/recontextualization pressure and the mode of recontextualization. To explain the different hybridization outcomes it is crucial to understand that there are different starting points. Three ideal typical transfer scenarios are thinkable. The first principle starting point is the transfer of a foreign/parent template. A second principle starting point is the use of a local/host template (e.g. from a local Joint Venture partner). The third principle starting point is defined by the absence of both, a foreign/ parent template transfer and a local/host template use. Such an absence of a defined foreign/parent and local/host template does not mean, however, that the local production system does not have to respond to the foreign/parent and/or local/host contextual requirements. Such contextual requirements involve demands and condition the production system to be built has to respond to. Often such demands and conditions may just involve pressures to achieve a certain result (e.g. local content rates) rather than defining a coherent organizational pattern or structure to achieve such a result. A template, in contrast, is understood here as a coherent organizational pattern that either exists or is defined as an ideal to configure some or all dimensions of a production system. The different transfer scenarios can face different kinds of fits or misfits which influence the hybridization outcome. In the first transfer scenario, the foreign/parent template may either fit or misfit the local/host context. Depending on the misfit, the transferred template will be subject to

48 Hybridization of MNE Subsidiaries

MNE Host country Fit/Misfit Home/3rd country

MNE Host country Fit/Misfit Home/3rd country

MNE Host country Fit/Misfit Home/3rd country

Demands/conditions from foreign/parent or local/host context Foreign/parent context template transfer or local/host-context template use Figure 3.2 Different kinds of (mis)fits

pressures for recontextualization. Conversely, in the second transfer scenario, the local/host template can fit or misfit the foreign/parent context. Again, depending on the misfit, the local/host template will be subject to pressures for recontextualization. In the third transfer (or rather nontransfer) scenario, there is neither a foreign/parent template transfer nor a local/host template use. However, even in this scenario, we can picture fits and misfit (see Figure 3.2) between the local/host context and the foreign/parent contextual requirements. In this scenario, the local production system to be built faces (more or less contradictory) contextualization pressures from the foreign/parent and/or the local/host context in the form of demands and conditions that need to be accommodated. The final step in understanding different hybridization outcomes involves asking how different misfits and corresponding

Theoretical Foundations 49

(re)contextualization pressures are resolved. While different recontextualization modes are available depending on whether contexts or templates are adapted, they share the principle of adaptation. Drawing on Westney (1987) adaptation is seen to rest on three mechanisms of adaptation, including: selection/de-selection, change and creation. So, let us return to different transfer scenarios, (mis)fits and their resolution. If the foreign/parent template fits the local/host context well, there is basically no recontextualization pressure. Therefore, we can expect a smooth transfer and imitation as the outcome. However, if there is a misfit and a corresponding recontextualization pressure, three modes of recontextualization are imaginable. The first mode of recontextualization involves overcoming the foreign/parent template’s misfit by mainly adapting the local/host context to the foreign/parent template. This involves only little or no adaptation of the foreign/parent template to the local/host context. Depending on the extent to which the local/host context is adapted, we can imagine outcomes ranging from hybridization to imitation. The second mode involves an adaptation of the foreign/parent template to the local/host context. In this scenario there is only little or no adaptation of the local/host context to the foreign/parent template. Under this condition we can imagine outcomes ranging from localization to hybridization, depending on how comprehensive the foreign/parent template is adapted to the local/host context. The third form involves both an adaptation of the foreign/parent template as well as of the local/host context with hybridization as the most likely outcome. Starting from the second transfer scenario a similar range of modes of recontextualization are thinkable. On the one hand, there may be a good fit between the local/host template and the foreign/parent context. If there is such a fit, the local/host template is not likely to face much recontextualization pressure and we will probably see localization as the outcome. Things are different, on the other hand, if there is a misfit and a recontextualization pressure. The first mode of recontextualization involves an adaptation of the local/host template to the foreign/parent context, with little or no adaptation of the foreign/parent context to the local/host template. Here we can imagine results ranging from imitation to hybridization, depending on how comprehensive the local/host template is adapted to the foreign/parent context. The second mode involves an adaptation of foreign/parent context to the local/host template. In this case there is little or no adaptation of the local/host template to the foreign/parent context. Accordingly, we can imagine results ranging from localization to hybridization, depending on how much the

50 Hybridization of MNE Subsidiaries

foreign/parent context is adapted. The third form of recontextualization is again a combination of the first two, with hybridization as the most likely outcome. Finally, we can imagine a third (non-)transfer scenario. In this scenario, neither a foreign/parent nor a local/host template is used. There may, however, be different contextual requirements to be met in the site set up. To the extent that foreign/parent context requirements fit local/host context requirements and vice versa, no recontextualization of either is required. The local production systems may be simply set up to meet the different contextual requirements. Nevertheless, it is not easy to determine ex ante, if these organizational solutions come to resemble a typical foreign/parent template (close to imitation), a local/host template (close localization), both or none. Some aspects may be created in line with foreign/parent context in the absence of contradictory local/host context, which can – but does not have to – mean that they resemble imitated solutions. However, they may as well just be customized to meet foreign/parent context demands and conditions not reflecting any particular foreign/parent template or pattern. In turn, some aspects may be created in line with local/host context in the absence of a contradictory foreign/parent context, which can – but does not have to – mean that they resemble local/host templates. They may as well be customized to local/host context demands and conditions not necessarily reflecting typical local/host templates. Things become even more unpredictable under the scenario where foreign/parent context requirements misfit or contradict local/host context requirements. Again, two principle modes of recontextualization, including their combination, are possible in the site set up. On the one hand, the site’s set up may mainly respond to the foreign/parent context. At the same time, the contradictory local/host context is adapted to meet the foreign/parent context in the set up (with little or no adaptation of foreign/parent context to the local/host context). On the other hand, the site set up may mainly respond to the local/host context. In this scenario, a contradictory foreign/parent context is adapted to the local/host context (with little or no adaptation of the local/host context to the foreign/parent context). Again in a third scenario the site set up may reflect both processes. Now, while we are more likely to see possible variations in organizational set-ups between imitation and hybridization in the first case, we are probably more likely to see outcomes between localization and hybridization in the second. In the third scenario, in turn, we are probably most likely to see outcomes between hybridization and customization/novelty.

Theoretical Foundations 51

It is important to note that customized or novel solutions are possible whenever a production system responds to contextual requirements that do not specify organizational templates to achieve them. A response to foreign/parent or local/host context’s demands and conditions may often involve organizational solutions that resemble templates of the respective contexts. However, there is also always the possibility that such a response takes on a customized or novel form. While customizations or novelties can, therefore, occur in almost any transfer scenario, they are most likely to occur when organizational set-ups respond to contradictory contextual requirements that define organizational goals rather than organizational designs or patterns (see Figure 3.3 for a summary).

Two kinds of contextual distance To take the analytical framework presented above a step further, we need to ask about the grounds for different (mis)fits as well as transfer scenarios and recontextualization mode choices. We argue that contextual distance plays a key role in this regard and propose that two kinds of contextual distance are relevant: institutional or strategic distance. Whether the institutional or strategic distance is more relevant for certain dimensions of a production system is left open to exploration.

Institutional contexts The Labour Process wing of the Japanization debate and Institutionalist approaches have largely looked at the transferability of organizational forms and practices from one institutional context to another. The major focus has been on the role of national institutional contexts. The basic reasoning is that firms or their organizational forms and practices are embedded and grow out of distinct institutional contexts. As such, their functioning is inextricably linked to a specific institutional context (Kostova, 1999). In other words, organizations and their elements are institutionally constituted. While European Institutionalists have focused for the most part on institutional systems at the national level, more recent studies in this research tradition have extended the focus. In these studies the relevance of institutional conditions at the organizational level has also been recognized. Now, when foreign/parent templates are transferred to a different local/host institutional context, chances are that misfits emerge between what the transferred templates institutionally require and what the local/host context of the subsidiary institutionally offers or demands. This is mainly because production system templates rely, albeit to different degrees, on specific institutional

52 Hybridization of MNE Subsidiaries Transfer Scenario

Fit/Misfit

Fit

Foreign/parent template

Misfit

Fit

Local/host template

Misfit

Fit

No template, but different contextual demands

Misfit

Recontextualization Mode

Outcome

No Recontextualization

Imitation

Local/host context adaptation

Imitation to Hybridization

Foreign/parent template adaptation

Hybridization to Localization (Customization)

Foreign/parent template and local/host context adaptation

Hybridization

No Recontextualization

Localization

Local/host template adaptation

Hybridization to Imitation (Customization)

Foreign/parent context adaptation

Localization to Hybridization

Foreign/parent context and local/host template adaptation

Hybridization

No Recontextualization

Hybridization, Imitation, Localization or Customization

Local/host context adaptation

Hybridization to Imitation or Customization

Foreign/parent context adaptation

Foreign/parent and local/host context adaptation

Hybridization to Localization or Customization

Hybridization or Customization

Figure 3.3 Different transfer scenarios, misfits, recontextualization modes and outcomes

conditions, which may or may not be available in a local/host context of a subsidiary. Put differently, the misfit depends on the degree to which transfer templates depend on specific institutional contexts and the degree to which the receiving context provides such institutional conditions and favourable demands.

Theoretical Foundations 53

What is more, the institutional distance may not only induce the misfit and potential recontextualization pressure after a transfer took place, but may impact the transfer scenario in the first place. If, for example, institutional conditions in the local/host contexts are very distant but rated advantageous, the foreign parent may refrain from template transfer and draw on a local/host template. Similarly, if institutional conditions in the local/host contexts are very distant but rated not very conducive with regards to the institutional requirements of a foreign/parent template, the foreign parent may also refrain from its transfer. Thus, institutional distance is likely to impact the transfer scenario. An institutional distance is likely to cause transfer restraints and induce recontextualization pressures on transferred templates. In summary, it is proposed: Transfer scenarios and misfits vary because there can be more or less of an institutional distance between the origins and destinations of a transfer template. It should be added that the institutional distance is not only relevant for the fit/misfit between a foreign/parent template and local/host context conditions and demands. There may as well be an institutional distance between a local/host template and foreign/parent demands and conditions, causing misfits and recontextualization pressures. Finally, it is assumed that the recontextualization mode, the question whether an institutional distance induced misfit leads to an adaptation of the transfer template or to an adaptation of the local/host context, is linked to the interplay between the willingness and ability of a foreign/parent to invest resources such an adaptation requires and the nature and resilience of the receiving (local/host) institutional context constraining or enabling the adaptation.

Definitions Institutional distance is defined here as the difference between a foreign parent’s local/home or local/3rd country site institutional context and the local/host country institutional context of the subsidiary under investigation. An institutional misfit is defined here as the mismatch/incompatibility between what a transferred template institutionally requires or is designed for, and what the local/host context institutionally offers or demands; or alternatively as the mismatch/incompatibility between what a local/host template institutionally requires or is designed for and what the foreign/parent context institutionally offers or demands. Or in more simple terms, an institutional misfit is defined here as the mismatch/incompatibility

54 Hybridization of MNE Subsidiaries

between foreign/parent templates or demands and the local/host institutional context of a subsidiary or alternatively as the mismatch between local/host templates or demands and the institutional context of the foreign/parent. Drawing on Sorge (2004), institutions are understood here as habitual regularities. At the level of a country or the societal level such habitual regularities find expression in differentiated yet ‘reciprocally interdependent’ institutional domains (Sorge, 2004). Firms and organizations are institutional domains themselves and are in an interdependent relationship with other institutional domains such as education systems, industrial relation systems. The relevant institutional context for an organization in this approach is pragmatically derived, depending on ‘reciprocal interdependencies’ between the focal organization and other institutional domains. Such a pragmatic approach is also applied in this research context. Adopting and adapting earlier Societal Effect research in the manufacturing sector (Maurice et al., 1980), it is suggested that family structures, social stratification, education systems, industrial as well as supplier relations systems form core institutional domains that are in an interdependent relationship with a productions system’s configuration.

Strategic contexts The IB literature and also the Lean Production stream of the Japanization literature has been utterly weak in theorizing the institutional embeddedness of production system templates as well as the fact that institutional distance influences transfer scenarios, misfits and recontextualization pressures. There has also been little systematic attention to different recontextualization modes. In areas where institutional contextual misfits could not be entirely denied, it has been suggested that MNEs can adapt – select, change and create – contexts to their liking. This generally implies that the only mode of recontextualization considered is the adaptation of local/host institutional contexts to the foreign/parent template. Put simply, transfer contents do not have to be changed in the face of institutional contextual distance and misfit. Global or local solutions seem to be decided on purely rational strategic grounds with little concern for institutional mismatches upon arrival. However, while there seems to be some underlying conviction about the global transferability, if strategically required, the IB literature and the Lean Production body of the Japanization literature suggests that there are strategic contextual misfits that affect transfer scenarios and misfits/recontextualization pressures. These impacts on transfer scenarios and needs for recontextualization mainly derive from different

Theoretical Foundations 55

strategic roles, one could also say from the different strategic contexts of subsidiaries in MNEs. Adopting the IB literature’s differentiated network perspective of the MNE, in which subsidiaries are seen to take on different and evolving roles (Birkinshaw, 2000) has three important implications for our understanding of transfer scenarios and recontextualization dynamics in subsidiaries. First, the differentiated-network perspective allows us to envision many more transfer origins than have commonly been considered in Japanization and Institutionalist approaches. In such a perspective, the parent is but one transfer source: other subsidiaries and even other organizations, as part of local or international networks, come into the picture here. Secondly, by adopting a subsidiary perspective, we can see subsidiaries not only as instruments of the corporate parent, but as actors in their own right, more or less resourceful and capable, needy and greedy for transfers. This also allows considering local transfer initiatives as much as changing transfer propensities and pressures for recontextualization as subsidiaries and their environments evolve over time. Thirdly, and probably most importantly, seeing subsidiaries with Birkinshaw (2000) as differentiated by their role, understood as differences in resource/capability endowments and differences in charters, suggests different transfer scenarios and misfits/recontextualization pressures on strategic grounds. If a foreign/parent template is transferred from one subsidiary to another, chances are that a strategic misfit emerges between what the transferred production system template is strategic contextually optimized for – with regard to task profile (including the required capabilities and resources to master the task profile) and supply/demand market conditions – and what the receiving subsidiary’s strategic context offers or demands. This is the case because production systems rely, albeit to different degrees, on specific strategic contexts, which may or may not be present in the receiving local/host context. We can, therefore, imagine different degrees of strategic contextual misfit between what a foreign/parent production system template is strategicsally optimized for and the strategic context of a subsidiary. For example, the more distant the task profile and market conditions (e.g. quality, quantity and price of input and output factors) of a subsidiary from other operations in the MNE, the less likely a transfer of a foreign/parent template without a misfit and a recontextualization pressure. Now, it is important to note that the strategic distance may not only cause a misfit, a recontextualization pressure on what has been already transferred, but may influence the transfer scenario itself. If, for example, strategic conditions in the local/local contexts are very distant but rated advantageous

56 Hybridization of MNE Subsidiaries

(e.g. low labour costs) the foreign parent may refrain from template transfers and draw on a local/host template or a locally customized solution instead. Similarly, if strategic conditions in the local/host contexts are very distant but rated not conducive, with regard to the strategic contextual requirements of a foreign/parent template, the foreign parent may refrain entirely from its transfer. Thus, the strategic distance is likely to impact the transfer propensity and induce recontextualization pressures on transferred templates. Parallel to the reasoning on the institutional dimension, we can imagine that transfer templates can be more or less dependent on the strategic context of subsidiaries, defined by task profiles, capabilities/resources and supply/demand market conditions. In summary, it is proposed: Transfer scenarios and misfits vary because there can be more or less of a strategic distance between potential origins and destinations of a transfers template. It should be added, that the strategic distance is not only relevant for the (mis)fits between the foreign/parent template and the local/host strategic context. There may as well be a strategic distance between the local/host template and the foreign/parent strategic context causing a misfit and a recontextualization pressure. Finally, the question whether a strategic distance leads to an adaptation of the transfer template or to an adaptation of the local/host context is linked to the interplay between the willingness and ability of a foreign/parent to invest resources such an adaptation requires and the nature and resilience of the local/host strategic context constraining or enabling the adaptation.

Definitions Strategic distance is defined here as the difference between a foreign parent’s local/home or local/3rd country site strategic context and the local/host country strategic context of the subsidiary under investigation. A strategic misfit is defined here as the mismatch/incompatibility between what a transferred template strategically requires or is designed for, and what the local/host context strategically offers or demands; or alternatively as the mismatch/incompatibility between what a local/host template strategically requires or is designed for and what the foreign/parent context strategically offers or demands. Or more simply, a strategic misfit is defined here as the mismatch/incompatibility between foreign/parent templates or demands and the local/host strategic context of a subsidiary; or alternatively as the mismatch between local/host

Theoretical Foundations 57

Strategic context Home/3rd country MNE

Local/subsidiary context

Home/3rd country site context

Institutional context

Host country

Institutional or strategic demands from parent or local/host context Template transfer/use

Figure 3.4

The fourfold embeddedness of subsidiary production systems

templates or demands with strategic context of the foreign/parent. Drawing loosely on Birkenshaw (2000), strategic contexts are defined as the task profiles of firms – constituted by markets served, quality and quantity of products manufactured, as well as functional areas covered – and their supply and demand market conditions: quality, quantity and price of supplies/inputs and quality, quantity and price demands for outputs. Clearly, task profiles of firms and market conditions mutually constitute each other. However, in the case of foreign subsidiaries, the local task profile and the host market conditions can be decoupled from one another as the local site can serve global markets (see Figure 3.4 summarizing core ideas of the analytical framework).

Strategic choices: global product strategy and entry mode The final step in this analytical framework is to link the triad of transfer scenario, institutional and strategic (mis)fit/recontextualization pressure, recontextualization mode and consequently the hybridization profiles of production systems to strategic choices at the corporate and the subsidiary level. A core argument of this study is that this complex has been only weakly explored so far. It is asserted that specific strategic choices – the generic strategy of the MNE as well as entry modes – strongly interact with transfer scenarios, institutional and strategic (mis)fits and modes of recontextualization. Furthermore, since we are dealing here

58 Hybridization of MNE Subsidiaries Table 3.2

Porter’s generic strategies

Strategic strength Strategic scope

Low cost

Uniqueness

Wide range of market segments

Cost leadership strategy

Differentiation strategy

(Hybridization outcome)

(Between Localization, Hybridization, Customization & Imitation)

(Localization, Hybridization & Customization)

Narrow range of market segments

Cost focus strategy

Differentiation focus strategy

(Hybridization outcome)

(Imitation)

(Between Localization, Hybridization, Customization & Imitation)

Source: Adapted from Porter (1980).

with production systems, it is assumed that global product strategies will be of particular relevance. While leaving the specific patterns of interaction open to empirical exploration, a number of potential interactions that may impact hybridization outcomes are proposed.

Corporate level: generic strategies Porter (1980) distinguishes three or rather four generic strategies: The cost leadership, the differentiation, the cost focus, and the differentiation focus strategy. These strategies rest on a demand side dimension – the choice between a broad or a narrow market scope – and a supply side dimension – the strength or core competency of the firm – mainly varying between product uniqueness and low cost (see Table 3.2). With regard to the relationship between generic strategies and transfer scenarios, contextual misfits and recontextualization modes, the following associations can be asserted. Firms whose competitive strategy mainly rest on low costs and who have a narrow range of market segments, have the highest propensity to develop similar production sites as they internationalize. When the product portfolio does not vary much and cost control takes on a prime importance, there is a high incentive to develop standard templates for global operations. This is particularly the case when entry barriers to different markets disallow an exportled internationalization or servicing all markets from global/central hubs. In other words, firms with cost focus strategies are very likely to define and transfer templates. The moderate strategic distance between sites (based on narrow market scope implying similar task profiles and

Theoretical Foundations 59

market environments across sites) allows these firms to develop/define standard templates. At the same time, they also have strong incentives to control cost by applying standard templates in different locations. What is more, since contextual distance is moderate or low, these firms will face less recontextualization pressure when they engage in template transfer. Finally, as cost control through standard template application forms a central part of their competitive strategy, they will have a preference for an adaptation of the local/host context to the foreign/parent template, rather than the other way round. Thus, compared to MNEs with other product strategies: MNEs with a cost focus strategy are more likely to develop/define and transfer a foreign/parent template, less likely to face a high strategicdistance-related recontextualization pressure, more determined to adapt a local/host context to the foreign/parent template and to feature higher degrees of imitation as an outcome. This is very much in contrast to what we can expect when firms with a differentiation strategy internationalize. These firms’ competitive advantage rests on a wide range of market segments and product uniqueness. Since their sites are likely to be strategically distant from each other – based on more differentiated task profiles and market conditions – and since cost advantage plays less an important role for their competitive strength, they will also have a lower propensity to define/develop and transfer standard templates. However, when foreign/parent transfers occur or demands are voiced, there is a high chance of strategic misfit and recontextualization pressure because the sites tend to be strategically distant. At the same time, these MNEs may be more flexible and willing to adapt their templates or demands to the local/host context because catering to diverse demands in a unique way is a crucial element of their competitive advantage. Thus, compared to MNEs with other product strategies: MNEs with a differentiation strategy are less likely to develop/define and transfer a foreign/parent template, more likely to face a high strategic-distance-related recontextualization pressure if they transfer, more flexible to adapt their templates or demands to the local/host context and to feature as a result higher degrees of local, hybrid and customized solutions as outcomes. Finally, firms with cost leadership strategy or a differentiation focus strategy are likely to be between the poles of high and low imitation.

60 Hybridization of MNE Subsidiaries Table 3.3 Entry mode combinations Equity mode Establishment mode

Wholly owned

Joint venture

Greenfield Brownfield

Wholly owned Greenfield Acquisition

Greenfield JV Brownfield JV

Subsidiary level: entry modes The second aspect of strategic choice that has received remarkably little attention is the influence of entry modes – defined as the combination of establishment (Greenfield site vs Brownfield site) and equity modes (wholly owned vs JV) (Pan and Tse, 2000; Harzing, 2002) – on transfer scenarios, institutional/strategic misfits and recontextualization modes (see Table 3.3 for different entry mode combinations). There is evidence to suggest that Greenfield sites either reduce or mediate institutional distance which reduces, in turn, institutional misfit and recontextualization pressure (e.g. Sharpe, 1997; Saka, 2003). We argue that the establishment mode impacts all components constituting hybridization outcomes including: transfer scenarios, misfits/recontextualization pressures and recontextualization modes. In contrast to Brownfield sites, Greenfield sites lack an existing production system. Their local (meaning on site) institutional contexts and strategic contexts (task profile and corresponding resources and capabilities) have to be created afresh. This implies that they are in need of some configuration which increases the likelihood of a template transfer. Moreover, their low level of institutionalization implies a lower contextual distance compared to Brownfield sites, which poses less constraints for the transfer of a foreign/parent template or the realization of foreign/ parent demands. By the same token, Greenfield sites are likely to show lower levels of misfit and recontextualization pressures when transfer occurs. Again, lower levels of local institutionalization and preexisting local strategic context (task profile and corresponding resources and capabilities) imply that there is less of an exiting context that can be incompatible with what is being transferred. Finally, even if some misfit occurs, it is comparatively easier to adapt the local context to foreign/parent demands and templates because there is a lower level of institutionalization and strategic set up. It is therefore proposed: In contrast to Brownfield sites, Greenfield sites are expected to have a higher propensity for a foreign/parent template transfer, to have

Theoretical Foundations 61

lower levels of misfit when transfer occurs, find it easier to adapt a local context to foreign/parent templates or demands and feature, as a result, higher degrees of imitation. Conversely, Brownfield sites may require or seek less foreign/parent template transfer because of an existing local production system. And even if there is no intention to draw on or use the local institutional/strategic contexts of an existing template (e.g. because of institutional or strategic misfit with foreign/parent templates or demands), these very same existing conditions make a transfer of foreign/parent templates more difficult and misfit related recontextualization pressures more likely. Finally, an existing local/host template makes an adaptation of the local context to foreign/parent demands and templates more difficult than in Greenfield sites. It is therefore proposed: In contrast to Greenfield sites, Brownfield sites are less likely to see a transfer of a foreign/parent template, will face higher levels of misfit when transfer takes place, will find it more difficult to adapt a local context to the foreign/parent template and will feature higher degrees of localization as an outcome. However, the establishment mode is but one crucial aspect of entry mode choice to affect the hybridization outcomes. We propose that equity modes are a second crucial aspect impacting all components constituting hybridization outcomes. The core assumption is that the equity mode has a substantial influence on whose management controls the company and whose management decides: What is or can be transferred or demanded? What is rated as a misfit and requires recontextualization? And to what extent the foreign/parent templates/demands or local/host templates/demands have to be adapted, rejected or responded to? Now, in the case of a wholly owned subsidiary the situation appears to be rather clear. As the foreign parent is in full control, we can expect a high propensity of foreign/parent template transfer and/or demands. At the same time, given the high foreign/parent transfer and demand propensity, there is also more scope for misfit and recontextualization pressure. However, where misfits and recontextualization pressures occur, parent companies enjoying full ownership will find it easier to put through their templates and demands without facing the intervening power of a local partner. In addition, they will also be more willing to invest resources. As a consequence, adaptation of the local context to foreign/parent demands and templates may be a more prevalent recontextualization mode than in Joint Ventures ( JVs).

62 Hybridization of MNE Subsidiaries

In JVs, certainly depending on whether there is a minority or majority equity of the foreign partner, the situation is slightly different. Here shared ownership increases the number of template sources and demands (local partner vs foreign partner) that can potentially play into the configuration of a production system. Shared ownership can imply divided responsibilities and the simultaneous impact of different demands and templates. This situation probably increases the chance of hybrid and/or customized/novel outcomes because templates and demands from markedly different contexts may be integrated or followed at the same time. In JVs the equity share of the parties involved is assumed to play an important role. For the equity share may be decisive for the question whose management controls the site. For example, if the foreign parent takes a minority position, the foreign/parent may be less likely to transfer templates because of fears of knowledge loss or a lacking permission to do so. In this scenario the chance of misfit and recontextualization pressure is also assumed less because there is less foreign/parent influence or presence that can potentially cause a misfit and induce a recontextualization pressure. Moreover, even if a foreign/parent transfer takes place and a misfit occurs, the adaptation of foreign/parent demands and templates to the local/host context is more likely because the local parent is in control giving preference to local/host solutions. As a result, JVs in which the foreign parent takes a minority position will have a high propensity to feature outcomes between hybridization and localization. It is therefore proposed: The higher the local parent equity, the higher the chance that the site uses local/host templates and responds to local/host context demands, the lower the chance of misfits/recontextualization pressures, the higher the likelihood that foreign/parent templates and demands are adapted to the local context in the case of misfit and as a consequence, the more likely localization as an outcome. The opposite is asserted when the foreign parent holds the majority in the JV. If the foreign parent takes a majority position, the foreign parent may be more likely to transfer templates or pose demands because it controls the site. At the same time, the chance of misfit and recontextualization pressure is higher because there is more foreign/parent influence or presence that can potentially cause a misfit and induce recontextualization pressure. When misfit occurs, the adaptation of the local/host context to foreign/parent demands and templates is more likely because the foreign/parent is in power and will give preference to foreign/parent

Theoretical Foundations 63

solutions. As a result, JVs where the foreign parent takes a majority position will have a higher propensity to feature outcomes between hybrid and imitated solutions. Thus: The higher the foreign parent equity, the more likely are we to see the transfer of foreign/parent templates and/or the influence of foreign/parent demands, the higher the chance of misfit and recontextualization pressure, but also the more likely an adaptation of the local context to foreign/parent templates and demands as the dominant recontextualization mode and as a consequence a higher the likelihood of imitation as an outcome. Referring to entry modes as a whole, we can expect – based on the foregoing reasoning – wholly owned Greenfield sites to feature the highest degree of imitation and Brownfield minority JVs the highest degree of localization.

Research design and methodology A qualitative comparative case study with a strong explorative element In the view of the research objective a qualitative comparative case study with a strong exploratory element is adopted in this study. Understanding how hybridization outcomes differ and why such outcomes have come about is essentially about understanding qualitative changes in an organizational phenomenon in relation to its complex contextual embeddedness. Such a research focus requires qualitative methods that are defined by Marschan-Piekkari and Welch (2004) as ‘procedures for “coming to terms with the meaning not the frequency” of a phenomenon by studying it in its social context’ (Marschan-Piekkari and Welch, 2004, p. 6). Case studies are suggested to be particularly suited in IB research when cross-border or cross-cultural issues are involved (Ghauri, 2004). Moreover, Yin suggest that ‘case studies are the preferred strategy when “how” or “why” questions are being posed, when the investigator has little control over events, and when the focus is on a contemporary phenomenon within some real life context’ (Yin, 2003, p. 1). Both conditions apply when we are asking how and why hybridization profiles of production systems in MNEs differ. In addition, case studies employed in this research context should ideally adopt a multiple-case design (Yin, 2003). Pauwels and Matthyssens (2004, p.129) argue that the ‘only argument to switch from single to multiple case study research (at the risk of

64 Hybridization of MNE Subsidiaries

loosing depth) is to create more theory-driven variance and divergence in the data, not to create more of the same’. Based on this reasoning, a multiple-case design suggests itself when we are exploring associations between variations in strategic choices and hybridization profiles (Yin, 2003, Pauwels and Matthyssens, 2004). In multiple case studies, the case selection follows what Yin (2003) calls a ‘replication-logic’ or others the logic of ‘theoretical sampling’ (Glaser and Strauss, 1967; Eisenhardt, 1989; Pauwels and Matthyssens, 2004). The basic rational is that ‘each case must be carefully selected so that it either (a) predicts similar results (a literal replication) or (b) predicts contrasting results but for predictable reasons (a theoretical replication)’ (Yin, 2003: 47). Similarly, Eisenhardt (1989, p. 537) reasons that in case study research the ‘cases may be chosen to replicate previous cases or extend emergent theory, or they may be chosen to fill theoretical categories and provide examples of polar types’. In line with these suggestions and given the analytical framework presented, this research will be based on ‘theoretical sampling’ involving the selection of four ‘categorial types’ that differ systematically with regard to major variation causing factors theorized. More specifically, while the first research goal, the question whether both strategic and institutional distance impact hybridization outcomes could be investigated in a single case study (i.e. mainly through variation across different production system dimensions within a single case), the second research goal of this project, involving the question how different strategic choices impact hybridization outcomes, relies on ‘theoretical replication’ through a multiple case study. Although the selection of cases is driven by broad theoretical propositions, the study’s analytical framework is formulated in an open and tentative way (Eisenhardt, 1989).

Case selection: four automobile subsidiaries in India The comparative study involves four cases for two reasons. On the one hand, four cases allow a relatively complete variation of the major variance causing factors posited in this study (i.e. variation in generic strategy and entry mode). On the other hand, given the comprehensiveness of a production system as defined in this work and given the time-intensity of qualitative case studies as well as the limited financial resources of the project, four cases were considered the maximum that could be handled. The emergence of hybrid organizational forms and practices can principally occur in any kind of organization and production system. However, it can be expected to be particularly prevalent and observable in those organizations or production models that are embedded or at the

Theoretical Foundations 65

interface of markedly different strategic and institutional contexts. It is, therefore, the MNE that is in the focus of this study. At the same time, this work chooses the subsidiary as the main level of analysis because it is expected that different contextual pressures are more tangible and their empirical investigation more manageable at subsidiary level, compared to the corporate level. Production models of automobile subsidiaries in India are selected as the unit of analysis for three reasons. First, apart from the fact that the research of production system hybridization is justified in its own right, given its crucial importance for the internationalization of production in MNEs, production systems as defined here are multi-dimensional constructs whose different dimensions may be more or less impacted by different kinds of context. Thus, unlike a research that focuses solely on the transfer and hybridization of specific practices – such as human resource management (HRM) practices – a focus on whole production systems provides us with more within case variation that renders it easier to observe whether and how different organizational dimensions are impacted by different contexts. Second, the focus is here on automobile production because extant research shows that sectoral difference has a strong impact on hybridization outcomes (e.g. Florida and Kenney, 1995; Kenney, 1999; Abo et al., 1994; Smith and Elger, 2000). To rule out such a sectoral impact and to reduce research complexity, the study is confined to the automobile industry. Moreover, a focus on the automobile industry allows a selective comparison of the findings of this study to those of others. After all, most of the hybridization research on production systems has been conducted in the automobile industry. Third, past research has shown that host country variation has a strong impact on outcomes of transfers in MNEs (e.g. Kostova and Roth, 2002). Therefore, to reduce contextual complexity, subsidiaries were chosen, which not only operate in the same industry but also in the same host country. Finally, the choice of India as the focal host context is motivated by the theoretical sampling logic of this study and informed by the state of hybridization research. On the one hand, the Indian context offered a range of cases where both parent generic strategies as well as entry modes differed systematically in line with broad propositions posited in this study. Additionally, the within host context variation for the different cases was relatively low due to a modest variation of entry times. On the other hand, India was chosen because hybridization research has not paid much attention to emerging market contexts, let alone to India, which tends to be in the shadow of China or East Asia in general.

66 Hybridization of MNE Subsidiaries

Introducing the four cases The four cases selected for this research project include the subsidiaries Maruti-Suzuki India (MSI), Fiat India (FI), Mercedes-Benz India (MBI), Skoda Auto India (SAI). Although a thorough introduction of the four cases will follow in the individual case discussions in Chapter 6, a brief introduction of the cases is made here to render the sampling-logic of this project more comprehensible. The roots of Maruti Suzuki India date back to 1970 and 1971 when Sanjay Gandhi, the son of Prime Minister Indira Gandhi, establishes two companies (Maruti Technical Services Private Limited and Maruti Limited) for the development and production of a ‘people’s car’. However, it is not until 1983, when Suzuki enters the company that the company produces its first cars. As no meaningful production infrastructure and labour force exist, the establishment of the site is basically a Greenfield project largely driven by Suzuki’s entry in the early 1980s. In terms of equity mode, Suzuki is initially only involved as a minority JV partner. In 1992, when the Suzuki Motor Company (SMC) acquires a 50 per cent share in the JV, Maruti seizes to be a public sector company. In 2002, SMC finally acquires a majority in the JV. While SMC’s overall strategy is between cost leadership and cost focus, its emerging market strategy rests more on cost focus. Fiat enters the Indian market in 1996 in the format of a majority JV with Premier Automobiles Ltd (PAL). Although the company initially plans to set up a wholly owned Greenfield operation, market conditions force the company to freeze its plans. Instead, Fiat engages in a JV with the former technical-collaboration partner PAL and takes over its ‘Brownfield’ site. Practically from the beginning, Fiat has a majority (51 per cent) in the JV with PAL and raises its stake continuously. With the shift in equity the Indo-Italian JV, India Auto Ltd (IAL), is renamed to Fiat India Private Ltd (FI). In 2006, Tata Motors and Fiat establish in Ranjangaon, near Pune, a new 50:50 per cent JV to manufacture passenger cars, engines and transmissions for the Indian and overseas market. Fiat’s generic strategy can be described as combining a differentiation and cost leadership strategy. However, as far as Fiat’s generic strategy for emerging markets is concerned, it is more appropriate to label it cost focus. Mercedes-Benz India (in the meantime called DaimlerChrysler India) is founded in 1994 as an Indo-German JV between Daimler-Benz and the Indian truck producer Tata Engineering and Locomotive Company Limited (TELCO). From the start, the German partner holds a majority in the JV and continuously increases its equity in the operation. The establishment mode of MBI is basically a Greenfield format. Although

Theoretical Foundations 67

some employees are delegated from TELCO, the labour force is for the most part young and newly recruited. The production facilities are also newly installed in a new factory hall. Mercedes-Benz’s overall generic strategy can be best categorized as differentiation. However, in emerging markets the strategy rests more on differentiation focus. Skoda Auto India Private Ltd (SAI) is founded in 2000 as a wholly owned Greenfield site. The generic strategy of Skoda’s parent Volkswagen AG (VW) can be best described as combining a differentiation and cost leadership strategy. VW’s emerging market strategy with Skoda rests more on cost leadership, however. Table 3.4 outlines the major difference between the cases with regard to their strategic choices. Figure 3.5 shows the equity development of the four cases between 1982 and 2005.

Data collection method The study’s data collection rests on the triangulation method (Miles and Huberman, 1994). The study makes use of multiple sources of data and collection methods including: observation, documentation and guided open-ended interviews (Eisenhardt, 1989; Patton, 1990; Yin, 2003). The main benefit of triangulation and integration of multiple data sources is an increased internal validity of the study (Pauwels and Matthyssens, 2004). In this study triangulation is of paramount importance because there is an imbalance with regard to different types of data available for the respective cases. For example, while there is a high saturation with regard to interview data, there is only a moderate amount of external (press/media) documentation in the MBI case. This is just the other way round in the case of MSI, where the interview saturation-level is moderate and the availability of external documentation is very high. Thus, the combination of these different sources of data not only helps to crosscheck the validity of findings but is also a prerequisite to establish a sound comparison across different dimensions under conditions of financial resource and information/data access constraints.

Data collected The data collection method of this work rests on three pillars. The first pillar comprises the collection of documents available within and outside the companies. Data collection for this kind of data lasted practically throughout the whole period of the research project starting in February 2002 and ending in December 2005. This implies that data collection overlapped with data analysis, which is a key element and advantage of case study research (Eisenhardt, 1989). The second pillar involves observations made during factory and site visits and the third pillar

68

Table 3.4

Selection criteria and variance of cases based on theoretical sampling MSI

Strategic choices at corporate level Generic strategy SMC is between cost focus & cost leadership SMC’s emerging market strategy rests more on cost focus Strategic choices at subsidiary level Establ. mode Greenfield Equity mode

FI

MBI

SAI

Fiat combines cost leadership & differentiation

MB features differentiation

VW combines differentiation & cost leadership

Fiat’s emerging market strategy rests more on cost focus (world car project)

MB’s emerging market strategy rests more on differentiation focus

VW’s emerging market strategy with Skoda rests more on cost leadership

Brownfield

Greenfield

Greenfield

Joint Venture

Joint Venture > wholly owned

Joint Venture > wholly owned

Wholly owned

(From minority to majority JV)

(From majority JV to wholly owned)

(From majority JV to wholly owned)

(Wholly owned from the outset)

Theoretical Foundations 69 120.00 100.00

(%)

80.00 60.00 40.00 20.00 0.00

82

19

84 986 988 990 992 994 996 998 000 002 004 1 2 1 1 1 1 2 1 1 2 19 Maruti-Suzuki India

Mercedes-Benz India

Fiat India

Skoda Auto India Source: Compiled from company information.

Figure 3.5

Foreign ownership development of MSI, FI, MBI and SAI

consists of guided/semi-structured interviews. The factory visits and company interviews were mainly conducted during the research stay in India between 15 September 2002 and 29 February 2003. Only in the case of MBI there was an earlier research stay in India between 6 July and 27 July 1998 during which a total of 24 interviews were conducted. During the second and main research stay for this comparative study a total of 38 interviews were completed in the four subsidiaries as well as with different organizational and institutional representatives/experts outside these firms. Over all, the interviews served two broad goals. Interviews in the automobile subsidiaries were conducted to establish a solid understanding about their production system hybridization profiles and the contextual conditions that brought these about. Interviews with outside agencies were conducted to acquire a more solid understanding of the strategic and institutional context in India as well as to establish an external perspective on the companies. Table 3.5 gives an overview of the data collected, ordered by research case, by data-collection activity, and by date. Table 3.6 gives an overview about the interviews conducted with representatives of company external agencies and experts. It should be added that the geographical spread of the four sites, initial access difficulties and financial constraints regarding trips to and stays

70

Table 3.5

Summary of research activities at different research sites

Activity case Research period

Document collection

Factory visits

Interviews

MSI

Mainly between February 2002 Company website, company Tour of production site in and February 2003 documents and reports, IPO Gurgaon, the State of red herring prospectus, Haryana on 18.01.2003 academic books and articles, newspaper and website articles, press interviews

6 interviews conducted between 13.12.2002– 22.01.2003

FI

Mainly between February 2002 Company website, company and February 2003 documents and reports, academic books and articles, newspaper and website articles, press interviews

4 interviews conducted on 07.01.2003

MBI

Mainly between February 2002 Company website, company Tour of production site in and February 2003 documents and reports, Pimpri/Pune, the State of academic books and articles, Maharashtra on 12.11.2002 newspaper and website articles, press interviews

SAI

Mainly between February 2002 Company website, company and February 2003 documents and reports, academic books and articles, newspaper and website articles, press interviews

Total

Tour of production site in Kurla/Mumbai, the State of Maharashtra on 07.01.2003

10 interviews conducted between 10.11.02– 18.11.02 (24 interviews conducted during first research stay between 15.07.1998–29.07.1998)

Tour of production site in 3 interviews conducted Walju Industrial Estate/ on 25.01.2003 Aurangabad, the State of Maharashtra on 25.01.2003 23

71 Table 3.6 context

Summary of data collection regarding India’s institutional and strategic

Expert interviews

Interview date

Number of interviews

Factory visit where applicable

General Motors India Vice President Corporate Affairs

Gurgaon, 27.01.2003

1

–

Technical Vocational Educational and Training Consultant

New Delhi, 25.10.2002

1

–

Free Journalist – Motoring columnist

New Delhi, 25.10.2002

1

–

Eicher Motors Limited Chief Strategic Planning, General Manager Manufacturing and Director Supply Chain

Indore, 03.01.2003

3

Tour of production site in Pithampur/ Indore, in the State of Madhya Pradesh on 03.01.2003

Gesellschaft für technische Zusammenarbeit (GTZ) – Indo-German Tool Room Programme

New Delhi, 28.10.2002

3

–

Automotive Component Manufacturers Association (ACMA) of India – Executive Director

New Delhi, 06.11.2002

1

–

Society of Indian Automobile Manufacturers (SIAM) – Assistant Director

New Delhi, 29.11.2002

1

–

Indo-German Chamber of Commerce (IGCC) – Director General

Mumbai, 09.01.2003

1

–

Friedrich Ebert Stiftung – Expert on Industrial Relations in India

New Delhi, 12.12.2002

1

–

Business Standard Motoring – Assistant Editor – Features

Mumbai, 09.01.2003

1

–

Engineering und Design AG (EDAG) – Managing Director

New Delhi, 10.09.2002

1

–

Total

15

72 Hybridization of MNE Subsidiaries

at the different site locations did not allow prolonged research stays in some of the locations. Therefore, the number of interviews remained below the level originally targeted. Although case study protocols were carefully developed (c.f. Yin, 2003) including a detailed scheduling for the different kinds of research activities at different research sites, realities on the ground made a frequent rescheduling necessary. Nevertheless, the case study protocol served to systematically guide the research process and to track the achieved research goals.

Interview guide and interviewees The main goal of the company interviews and the document collection was gathering information on the how and why the local subsidiaries’ production systems differed with regard to their hybridization profiles. In addition, expert interviews that were conducted with representatives of institutions or organizations followed a different format. In this respect, interview guides were specifically tailored to the host institutional or strategic context to be inquired, i.e. host country conditions focusing on the educational system, the industrial relations system and the automotive industry in India. As far as the selection of interviewees was concerned, there was only a limited influence on the part of the researcher. However, the interviewrequest to companies involved asking for interviewees who were ideally: in a top to middle management position, preferably in a production or human resource management function; interviewees who knew well other sites of the foreign parent company or had even stayed there for a prolonged period; and interviewees who worked at the local subsidiary for an extended period of time. The reasoning behind this ideal interviewee profile was: that the interviewees needed to have a functional competence to answer production system and human resource related questions; that they needed to have a sufficient overview to answer questions concerning the whole production system; that the interviewees featured a comparative competence to answer questions about differences between the local site and other corporate sites; and that they should have some historical knowledge about the local site to be able to answer questions about developments and changes over time. Table 3.7 presents an overview of the functional and hierarchical positions covered by the interviewees in the respective subsidiaries.

Data analysis The data analysis of this project mainly follows Eisenhardt’s (1989) suggestion of starting with a within-case analysis and then moving on to a

Theoretical Foundations 73 Table 3.7 Interviewees’ function and position across firms Function/Position/Interview date

Code

MSI Assistant General Manager Production on 13.12.02 Assistant Manager Marketing on 18.01.03 Assistant Manager Production on 18.01.2003 Former Manager R&D on 20.01.03 General Manager HRM on 20.01.2003 Assistant General Manager Maruti Supplier on 21.01.2003

MSI 1 MSI 2 MSI 3 MSI 4 MSI 5 MSI 6

FI Senior Advisor to the Managing Director on 07.01.2003 Italian Head of Production on 07.01.2003 Manager Production on 07.01.2003 Manager Organization and Development on 07.01.2003

FI 1 FI 2 FI 3 FI 4

MBI German Vice President and Head of Production on 18.11.2002 Head of Human Resources on 12.11.2002 German General Manager Production on 18.11.2002 General Manager Human Resources on 13.11.2002 General Manager Materials on 13.11.2002 General Manager Quality on 14.11.02 Divisional Manager Human Resources on 12.11.2002 Divisional Manager Planning on 14.11.02 Divisional Manager Production on 14.11.02 German Meister on 10.11.2002

MBI 1 MBI 2 MBI 3 MBI 4 MBI 5 MBI 6 MBI 7 MBI 8 MBI 9 MBI 10

SAI Czech General Manager Technical/Quality on 25.01.2003 Head of Production on 25.01.2003 Production Worker on 25.01.2003

SAI 1 SAI 2 SAI 3

cross-case analysis. Moreover, the study follows Miles and Huberman’s (1994) suggestions concerning the main components of qualitative data analysis, based on the overlapping and intertwined activities of: data reduction, data display, conclusion drawing and verification. Following the ‘pattern matching logic’ (Yin, 2003; Pauwels and Matthyssens, 2004), the final analytical step of this study involved investigating and exploring whether or not the similarities and differences found across the cases could be attributed to predicted associations posited in the theoretical framework. This process was supported by extensive efforts of data display as suggested by Miles and Huberman (1994) (see also within-case and cross-case comparing tables). In addition to detecting

74 Hybridization of MNE Subsidiaries

matches/contradictions based on comparisons between theory and data, an effort was made to specify associations that were deliberately not explicated in the theoretical framework.

Validity of research Pauwels and Matthyssens’ (2004) suggest that the quality of qualitative research and its validation cannot be assured by output control but has to be built into the research design and process. They stress that a careful crafting of the research design, matching the research method, the data collection method, the sampling strategy and the data analysis with the research goal, are most important building blocs to assure research validity. This was done in this study, as we outlined in this chapter. While case study research cannot provide ‘statistical generalization’, it can provide ‘analytical generalization’ (Pauwels and Matthyssens, 2004). Analytical generalization involves ‘to generalize a particular set of results to some broader theory’ (Yin, 2003, p. 37). In the research context of this study, theories about subsidiary hybridization are the domain to which the results can be generalized.

Part II Empirical Results

This page intentionally left blank

4 India’s Business Context and the Automotive Sector

This chapter provides an introduction to India’s institutional and strategic context conditions that proved particularly relevant for understanding the production system hybridization. It should be noted, that the institutional patterns discussed here, describe tendencies at a rather general level.

India’s institutional context Hierarchy and demarcation in family, education, social stratification and beyond Family Traditionally the Indian family is an extended or joint family in which up to four generations live under one roof. The decision-making power rests mainly with the father of the family who involves the eldest son. It is also the eldest son to whom responsibility is conferred after the father’s death. The typical Hindu-family is patriarchal and hierarchical (Kakar, 1997), with the family bonding fundamentally based on the father-son and brother-brother relationship (Schmitt, 1984; Tayeb, 1988). While it is true that the Indian family is shifting towards a nuclear family – for example not living in a one household any more – it is also true that traditional family ties remain functional. If, for instance, a younger brother passes away, the eldest son still takes the responsibility for the family left behind. Regarding the upbringing of children, with all due caution, it may be stated that discipline, subordination, and deference towards the father, the elders or other authorities are among the cornerstones of socialization (Kanungo, 1994). Tayeb (1988) underlines the high degree of economic and emotional dependence of children on their family until 77

78 Hybridization of MNE Subsidiaries

late in life. Decisions related to educational, professional choices and even choices of marriage partner are largely taken by parents. The socialization goals and structures of families also interpenetrate into schools and ultimately into the work context. The authoritarian relationships in the family extend to authoritarian teacher-pupil relations in schools, with teaching methods resting on compliance and passive reproduction of knowledge (Schmitt, 1984). Kanungo (1994) and Sinha and Sinha (1994) show that patriarchal and hierarchical structures are also reproduced in the work context. Like the father in the family, the superior is consulted on all major issues. There is little scope for argument and open confrontation with superiors is disapproved of (Sinha, 1990a; Virmani, 2000). “‘Check with the boss” is the crux of the majority of decision making which naturally shifts the locus of control into the highest position in the organization’ (Sinha and Sinha, 1994, p. 167). The authors also state that clear superior-subordinate work roles are generally preferred over equality-based ones in the Indian work context (Sinha, 1980; Sinha and Sinha, 1994; Sinha, 1999). Sinha (1990b) also raises doubts about the ability to implement team work in the Indian context and suggest that the ‘only kind of teams which can function effectively is the vertical ones where the status differential is readily accepted and vertical solidarity develops’ (Sinha, 1990b: 482). Kanugo (1995) sees authoritarian and patriarchal institutional patterns as the root cause for work-dispositions marked by strong feelings of dependency, a tendency towards conformity and blind obedience. The overall implication of the socialization patterns cited is a general difficulty to implement equality based work concepts in the Indian work context. At the same time, hierarchical and patriarchic relations place a great emphasis on certain authority figures which renders it difficult to achieve ‘taking of responsibility’ at the lower levels in the organization. As a result it is difficult to implement concepts such as management-by-objectives (MBO) or delegation in general. Those work relations that have been described as effective in the Indian context, such as ‘nuturant task leadership’ (Sinha, 1980; Sinha, 1999), are based on strong personal relations where personal affection between the superior and subordinate takes on an important role for the task achievement. To be fair, India is undergoing tremendous change. In the urban centres traditional patterns of socialization described are seeing a transformation and an increasing heterogeneity (Saraswathi and Pai, 1997; Becker-Ritterspach, 2000). On the other hand, the patterns still form basic tendencies and certainly hold true for the generation of employees researched in this case study.

Empirical Results 79

Higher technical education The Indian higher education system is compartmentalized with different educational paths posing alternatives rather than building on one another (Heitmann, 1995). The technical education and vocational training in India broadly rest on a ‘three tier system’ (Palit, 1998; Thete, 1999). At the lowest level we find trade-courses for skilled workers and craftsmen, which are either taught at Industrial Training Institutes (ITI) or Higher Secondary Schools. At the middle level we find technical training programmes of Polytechnics that produce technician engineers or so called diploma engineers. The highest level is constituted by ‘first degree and post-graduate courses in engineering and technology’ (Thete, 1999: 25). These courses are imparted at Engineering Colleges or the highly prestigious Indian Institutes of Technology (IIT) which are producing graduate engineer degrees. It is important to understand that the structural conditions of the three tier system and the financial endowment (Schmitt, 1984; Heitmann, 1995) required for education, practically rule out vertical mobility across the three levels. Moreover, the educational content, in terms of the share of practical experience and theoretical training, varies starkly with the three levels. Two interviews with representatives of a developmentproject for vocational training (interview Indo-German Tool Room Project, 2002; Heitmann, 1995) explained that while ITI graduates have some degree of practical training and conclude their education with an apprenticeship, diploma engineers and even more so graduate engineers have hardly any practical hands-on training as a part of their education. There is a striking resemblance between the Indian technical education – which was founded to a large extent during British colonialism (Thete, 1999) – and the British system in this respect. For India, like in the British system (Sorge, 1995b), it holds true that educational programmes with strong links between academic knowledge and practical experience are associated with lower prestige. Similar to Britain, the link is stronger in polytechnics than at universities, and stronger at regular engineering colleges than at the prestigious IITs. Like their British counterparts (Sorge, 1995b), Indian graduates from the prestigious institutions have little practical experience. As a result, diploma engineers and engineering graduates feature a managerial work-identity that clearly does not have a physical involvement in the manufacturing process within its ambit. In India, different educational paths create a strong professional distance. The overall effect of such demarcations is that different employee categories tend to distance themselves from one another, creating communication and interaction boundaries in the organizations.

80 Hybridization of MNE Subsidiaries

Social stratification The professional distance between different employee categories generated by India’s educational system is complemented or rather in a ‘reciprocal interdependence’ (Sorge, 2004) with a social distance produced by traditional caste-related principles of social-stratification. Bouglé (1997) identifies three constituting elements of the Indian caste system. These are ‘repulsion, hierarchy and hereditary specialization’ (Bouglé, 1997: 65). Like Bouglé, Dumont sees these principles as constitutive but argues that they can be reduced to the one core principle of the opposition of the pure and the impure (Dumont, 1997). For: This opposition underlies hierarchy, which is the superiority of the pure to the impure, underlies separation because the pure and impure must be kept separate, and underlies the division of labour because the pure and impure occupations must likewise be kept separate. (Dumont, 1997, p.477) Interestingly, the structuring principles of the caste-system are ‘reproduced non-identically’ (Sorge, 2004) in India’s modern industrial workcontext in two core respects. Firstly, the principle of ‘hereditary specialization’ is socially functional in that there is a strong correlation between certain caste-backgrounds, educational-choices and professions (Dupont, 1992; Bronger, 1996; Panini, 1996; Sharma, 1997). Secondly, and more significantly, the principle of the ‘opposition of the pure and impure’ remains strong. It finds expression in a marked social distance between different employee categories (c.f. D’Costa, 2003) and in a strong preference of mental over physical work (Gosalia, 1992; Ramaswamy, 1996; Panini, 1996). The ‘reciprocal interdependence’ between the education system and traditional stratification principles of the caste-system constitute strong socio-professional demarcations between different employee-categories as well as a low prestige for physical work in general. The low prestige for physical work finds different expressions. Ramaswamy states, for example that ‘[b]land labels such as ‘operators’, ‘technician’ and ‘craftsman’ are increasingly the mean or minimum expectation, especially in high technology industry where the more conventional ‘worker’ is itself an opprobrium’ (Ramaswamy, 1996, p. 35). Industry is replete with evidence of the search for designations, especially expressive ones which stress rank and progression from physical

Empirical Results 81

to mental work, and the creation of new layers of the hierarchy to accommodate new designations. (Ramaswamy, 1996, p. 36) Generally, manual activities provide much lower prestige compared to clerical work no matter how simple it may be (Ramaswamy, 1996; interview Indo-German Chamber of Commerce, 2003). Needless to say that cleaning activities are amongst those with the lowest prestige. Perceived as ‘polluting’, cleaning activities on the shopfloor are still largely reserved for lower castes.

Social stratification and organizational hierarchy Indian companies, particularly Public Sector Undertakings (PSUs), feature extensive hierarchies. The proliferation of hierarchical levels and corresponding designations in PSUs can be partly explained by public sector policies that leveled differences in pay between workers and managerial levels (c.f. Okada, 1998). As compensation based differentiation proved difficult to realize, PSUs offered status differentiation by promoting particularly managerial employees to ever higher designations. However, it is not only the PSUs that feature an extensive differentiation of hierarchical levels and designations. Apart from the educational segmentation, India’s social stratification and family patterns play a crucial role in explaining the extensive organizational hierarchies. India’s contemporary social stratification can hardly be captured by referring only to the caste-system. Instead, ascriptive and meritocratic status-criteria exist side by side. Particularly in the urban and industrialized regions, with the emergence of new modern professions, traditional caste-related professions have partly lost their meaning. While meritocratic status-criteria are becoming increasingly important, it is also true that ascriptive (i.e. Jati and Varna) and meritocratic statuscriteria (income, educational degree, and designation held) are not readily separable (Bronger, 1996; Sharma, 1997). After all, the social status of an individual or rather of the whole Joint-Family is crucially defined by a combination of class- and caste-specific attributes. There is a coexistence of modernism and traditionalism where formal educational degree, profession and designations held in a company, combine with traditional caste-backgrounds. The most shining examples of this coexistence are weekly marriage market advertisements. Under the category

82 Hybridization of MNE Subsidiaries

‘Brides Wanted for Brahmin’ we can read in the Hindustan Times, for example: Wanted: Really Beautiful Professional girl below 25 yrs for Delhi based Status Gaur Brahmin boy B.E., MBA, 30/174/ 12 Lakhs [income] p.a.. Manager in Top MNC. Reply with Horoscope & Photo to. (Hindustan Times, 2002) Professional advancement in an organization and the status of the family are intimately connected. Tayeb (1988) stresses that professional success of an individual, translates into a success of the whole family. As hierarchical designations are one of the most visible signs of success, there is a permanent family-induced pressure for advancement in the organizational status hierarchies (Ramaswamy, 1996). Although, job contents often do not even change, companies have increasingly given in to demands for ever better designations to avoid loosing their best employees (Ramaswamy, 1996). The overall result of this development is that many Indian organizations have seen a tremendous inflation of hierarchical designations and reporting/responsibility levels (Bhadury, 2000). In India, ‘[t]he typical organizational chart would show structures which are neither flat and modern, nor even traditionally pyramidal, but shaped like a barrel, with bulging midriff and (counting out the chief executive) a flat top not very much smaller than the bottom’ (Ramaswamy, 1996: 39). Ramaswamy (1996) underlines that job-rigidity and poor labour-utilization in the Indian work context is strongly related to the steep hierarchical differentiation in organizations. Table 4.1 summarizes associations between institutional conditions and behavioural dispositions in the Indian work context.

Industrial relations in India The Industrial Relations System in India has been commonly described by three characteristics: First, a state domination of the industrial relationship, including a strong affiliation between unions and political parties; second, a legislative framework that creates employment inflexibility, particularly with regard to recruiting, deploying, disciplining, and dismissing labour; and third, a legacy of highly confrontational and adversarial employer-union and employment relations.

The basic legal framework Different scholars have pointed out that the prominent position of the state in India’s Industrial relations is rooted in India’s labour laws,

Empirical Results 83 Table 4.1 Institutional foundations of work dispositions in the Indian work context Institutional foundations

Expectations & behavioural dispositions

Compartmentalized vocational training & technical education enforced by caste-system related principles of social-stratification

Indian employees (IE) perceive high vertical professional and social distance between white and blue collar occupations close to the shopfloor

Practical and theoretical training content varies strongly with different educational programmes

IE have generally a low appreciation for physical/manual work

Polytechnics trained supervisors, forming the first level of management, generally lack practical know how

Supervisor levels dislike physical involvement on the shopfloor, feel distant from workers and play a weak roles as practical trainers of worker

Industrial Training Institute trained workers generally have superior practical manufacturing know how compared to supervisors or other managerial levels

Workers find it hard to accept supervisors as practical trainers & to form a close ties with them

Patriarchal/hierarchal-orientated relationships in family, school and beyond based on socialization patterns serving to create deferential and collectivist personalities

IE find it hard to take individual responsibility and tend to associate taking of responsibility & decisionmaking with specific social/ organizational authority roles or figures.

Strong socio-religious and professional hierarchies

Demand for ever higher, status-giving organizational designations

which were largely drafted before independence. Interestingly, they strongly reflect the British colonial effort to keep labour conflict down and production up, during the Second World War (e.g. Gupta and Sett, 2000; Ramaswamy, 2000). According to Gupta and Sett (2000) and Ramaswamy (2000) three laws have played a pivotal role in shaping Indian industrial relations as a distinct system (Ramaswamy, 2000). These are the Trade Union Act (TUA), 1926, The Industrial Employment Standing Orders Act (IEA), 1946, and the Industrial Disputes Act (IDA), 1947. The TUA creates the legal foundation for unions to exist. It ‘confers on unions the basic minimum legality which protects them from civil and criminal liability arising out of industrial disputes and enables to

84 Hybridization of MNE Subsidiaries

function as lawful associations’ (Ramaswamy, 2000: 113). Under the law any seven workers can form a union. For companies this means that even outsiders, non-company members, could form a union and pose demands on behalf of company employees. For some external union leaders this has even been a lucrative business, as one of my interviewees stressed (MBI 2). However, as Gupta and Sett (2000, p. 145) point out a ‘[m]ere registration of a union under the TUA does not entitle it to recognition by the employer as a legitimate representative of his workers, or to the bargaining relationship that would arise from such recognition’. Thus, one crucial reason for the prominent position of the state in Indian industrial relations is related to the condition that the TUA includes no provision that recognizes trade unions as collective bargaining agents. This condition is further strengthened by the IDA. Under the IDA, the government enjoys full discretionary power whether or not, when and how to intervene in an industrial dispute actual or threatened. It may or may not decide to conciliate, it may refuse to send a dispute for adjudication or it may decide not to implement the award of a Labour Court or Industrial Tribunal. While it may prohibit strikes or lock-outs, or it may refuse permission for a lay-off, retrenchment or closure, no court can take cognizance of an offence under the IDA, by way of an illegal strike/lock-out/layoff/retrenchment/closure, unless the government formally notifies such an action as illegal and makes a complaint to the court. (Gupta and Sett, 2000, p. 145) The IDA cements the vital role of the government in Indian industrial relations and forms the basis for the strong political involvement in Indian unions. Ramaswamy (2000) even suggests that the system creates a ‘fertile ground for political meddling’ and underlines that ‘[t]he history of industrial disputes is replete with instances of the abuse of political power and the subversion of conciliation to promote political interest’ (Ramaswamy, 2000, p. 121). However, the IDA has not only been problematic for unions and their recognition. For employers, the IDA and its subsequent amendments (in 1976) imply substantial employment and labour market inflexibilities. After all, lay-offs, retrenchments or closures are under the ultimate guard of political leaders as they require government permission. Moreover, the system implies that small unions whether or not they have a constituency in companies, can push through their demands as long as they have political patronage (Gupta and Sett, 2000; Ramaswamy, 2000).

Empirical Results 85

The last piece of legislation that crucially impacts Indian industrial relations is the Industrial Employment (Standing Orders) Act (IEA). The IEA ‘requires every establishment employing a hundred or more workers to have a set of certified standing orders defining conditions of employment such as hours of work, attendance, proper conduct, punishment for misconduct, and means available for redressal against unfair treatment’ (Ramaswamy, 2000, p. 114). The law’s basic purpose is to guarantee a minimum protection of the workforce, where no union is established.

The basic union structure and levels of collective bargaining The legal framework and India’s struggle for independence implied that there has been traditionally a strong nexus between political parties/movements and unions (Mamkoottam, 2000). In 2003, there are about five central trade union federations that are more or less affiliated with a political party. These include as the most important ones: The All India Trade Union Congress (AITUC), the Bharatiya Majdor Sangh (BMS), the Indian National Trade Union Congress (INTUC), the Centre of Indian Trade Unions (CITU) and the Hind Majdoor Sabha (HMS). According to Bhattacherjee (2001), India’s planned economy and growing public sector after independence ‘provided a terrain for large scale unionization’ and ‘quite naturally led to the formation of public sector unionism’ (Bhattacherjee, 2001, p. 248). Over all, the strong connectedness between political parties and the proliferation of unions between the 1960s and 1970s implied that: (1) political party or movement rivalry was carried all the way down into rivalling unions in companies, (2) political movements or politicians instrumentalized labour conflict for their personal or partisan interest, but also that (3) local labour disputes led to a wider politization by calling in politicians and political parties (Mamkoottam, 2000). However, from the 1980s onward and accelerated by India’s liberalization in the 1990s, India’s union landscape sees a shift. Already in the 19080s, failed industrial action and disillusionment with politically affiliated unions (Sinha, 2002; Budhwar, 2003) lead to a ‘rise and proliferation of “independent” unions operating in the major industrial centres and competing with the traditional party affiliated unions’ (Bhattacherjee, 2001, p. 254). On average this development deradicalizes industrial action and leads to less labour conflict, particularly in private sector companies. Moreover, leftist unions loose increasingly support which is rooted in their opposition to decentralized bargaining (Bhattacherjee, 2001). It is also suggested that the increased attractiveness of unaffiliated unions is related to their greater success in achieving wage increases (Bhattacherjee, 2001). All in all, these developments lead

86 Hybridization of MNE Subsidiaries

to a growing diversity in labour-management/union-employer relations between industries, states, and regions (Ratnam, 2001; Bhattacherjee, 2001). Presently, different bargaining levels and union structures co-exist in the Indian economy. In the private corporate sector, plant-level bargaining takes place with enterprise based unions that may or may not be affiliated with political parties. In the public sector enterprises, centralized union federation that are affiliated with to political parties bargain with the states (as employer) at the industry- and/or national-level. Central state government employees in the service sector (transportation, postal services, banking and insurance, police and firefighters, etc.) have their (typically) politically-affiliated unions bargaining at the national and/or regional levels. (Bhattacherjee, 2001, p. 247) Although politically affiliated unions have lost power, they still play a role depending on region, sector, and industry. And even if a company union has no political affiliation such a possibility always looms as a potential threat for employers and has created willingness on their part to foster good relationships with internal unions. While unaffiliated company unions have claimed industrial relations in private sector enterprises in general and in MNEs in particular, an unaffiliated company union is still no guarantee for absent union militancy or industrial peace. Although union power and membership has steadily declined since the late 1970s (Gupta and Sett, 2000; Sinha, 2002; Budhwar, 2003), India loses in the late 1990s still more days annually to strikes and lock-outs than any other country (Bhattacherjee, 2001)

Company industrial relations Company industrial relations in pre-liberalized India have been frequently described as adversarial in a great number of ways. At its worst, different company unions would fight each other, employer-union relations would be highly confrontational, and so would be labourmanagement relations. Even relations between different employee categories would be antagonistic. Lack of labour commitment, discipline as well as a culture of militant trade unionism, traditionally characterized core challenges to managing the Indian workforce (Bhadury, 2000; see also Venkataramani, 1990). Industrial conflicts, once they broke out, could be long-lasting, frequently involving go-slows, strikes, hunger-strikes, political agitation

Empirical Results 87

and even physical threat and violence. The employers’ side would most commonly react to such industrial action with lock-outs (Gupta and Sett, 2000). The state conciliation machinery, in turn, which has a prominent role for settling labour disputes, could not be counted on, as it reacted arbitrarily with regard to referring disputes for adjudication (Gupta and Sett, 2000). Furthermore, judicial processes would be slow and long-lasting with court cases pending over years. However, while this antagonistic mode of company industrial relations has still not perished entirely from the Indian industrial relations landscape, new companies, and especially MNEs, are increasingly able to establish an alternative scenario. This scenario generally implies the presence of just one unaffiliated company union and involves cordial and cooperative employee relations. Even the much bemoaned employment inflexibility, has become less of a problem in the wake of the Indian government’s initiation of the new economic policy in 1991 (Gupta and Sett, 2000). In fact, India’s post-liberalization governments resort themselves to public sector retrenchment through so-called ‘golden handshakes’ or Voluntary Retirement Schemes (VRS). In doing so, a government sanctioned mode for retrenchment is set that is widely adopted by private sector firms (c.f. Gupta and Sett, 2000). In addition, employment flexibility increases through all kinds of new contractual provisions, involving ‘bans on recruitment, job transfer to non-bargainable categories, introduction of parallel production, mergers, suspension of industrial action for a period of five years, concession bargaining’ (Bhattacherjee, 2001: 259). A case in point is Hyundai Motor India, seeking to avoid labour inflexibility and unionization by employing major parts of its labour as ‘trainees’ (Lansbury et al., 2006).

India’s strategic context in the automobile sector India’s economic reforms In 1991 the Congress-led government launches a comprehensive reform programme that marks a transition of the Indian economy. Prior to this turning point, India follows a mixed economic model that is marked by protectionism and state-led economic development. While investment in key sectors – such as heavy, basic and capital goods industries – is reserved for the public sector, the private sector is allowed to invest in some consumer goods industries that are defined by the five year plans. However, even in industries where the private sector is allotted a role,

88 Hybridization of MNE Subsidiaries

no full market competition unfolds. Instead, a comprehensive licensingsystem, strict investment control and protectionism shield many consumer industries from domestic and international competition. Despite some successes, India’s economic model becomes unsustainable by the early 1990s (Krueger and Chinoy, 2002). Following a severe balance of payment crisis in the early 1990s, the Indian Government launches stabilizing measures and initiates a new economic policy. First stabilizing measures include the reduction of the fiscal deficit and the devaluation of the Indian rupee. While the stabilization measures aim at short term alleviation of the economic crisis, the reform programme addresses structural problems in the Indian economy with a more long term approach. Internally, the reforms focus on shifting the economy from a state-led to a market-led growth regime (D’Costa, 2005). This implies the massive de-regulation of private sector controls and a step-wise privatization of public sectors and their enterprises. Externally, the reforms aim at opening up the Indian economy by liberalizing the trade and foreign direct investment regime (Krueger and Chinoy, 2002).

The emergence of a FDI regime in the Indian automobile industry India’s trade and FDI regime for the automobile industry can be broadly divided into four phases. These include: the licence-phase, from the 1950s until the mid-1980s; the phase of deregulation, from the mid1980s until the early 1990s; the phase of emerging liberalization, from the early 1990s until the early 2000s and the phase of full liberalization from 2000 onward (cf. Becker-Ritterspach, 2008). The first, the licence-phase, is characterized by stringent restrictions on investments and imports in the industry. Technology imports require government permission and investments in the sector are subject to government licences. Even capacity enhancement of existing operations requires permission (WTO, 1998; Degnbol-Martinussen, 2001). In the 1970s, additional legislation is put in place that substantially restricts trade and FDI. In 1974 the Foreign Exchange Regulation Act (FERA) is put into force, restricting foreign equity in Indian companies to a maximum of 40 per cent (Degnbol-Martinussen, 2001). The phase of deregulation starts with Rajiv Gandhi in 1985. In this phase 32 industries are freed from the requirement of obtaining a licence for new investments. On a case-to-case basis the Indian Government invites a few foreign companies to collaborate with local players in the automobile industry. At the same time, the Indian Government introduces the system of ‘broad-banding’ (Rieger, 1989; Mohnot, 2001). Although the new system is applied to a range of industries from 1983

Empirical Results 89

onward, it isn’t until 1985 that it is also applies to the automobile industry. The ‘broad-banding’ system essentially allows automobile and automotive parts manufacturers with an existing licence to produce a different product range with their existing facilities (Rieger, 1989). The deregulation phase also sees a range of other developments such as a loosening up of the Monopolies and Restrictive Trade Practice Act (MRTP) and reductions on import duties (Mohnot, 2001). Clearly, Suzuki’s involvement in government-owned Maruti and the company’s requirements for component imports in the beginning are a crucial reason for first liberalizations of the trade regime. At the same time, FDI is still required to follow a ‘Phased Manufacturing Programmes’ obliging companies to achieve a local content of 95 per cent. The third phase begins with the launch of India’s new economic policy in 1991. As the increase in market driven productivity, product quality and demand satisfaction becomes high on the political agenda, FDI is for the first time actively invited on a larger scale by the Indian government. The main goal is to create a competitive industrial base in India by gradually lifting investment-, production- and sales-restrictions. With the abolishment of the licensing system for the automobile industry in 1993 and the possibility of foreign equity of 51 per cent and above, India sees the entry of major international players in the 1990s (see table 4.3). Yet, a host of restrictions on FDI remain in place until the late 1990s. In the early 2000s, most of these remaining restrictions are lifted, marking the phase of full liberalization for the Indian automobile industry. Table 4.2 summarizes India’s emerging FDI regime in the automobile industry.

Demand market conditions in India’s automobile industry With the launch of the new economic policy in 1991, the Indian government not only reforms the trade and investment regime but also reforms the financial sector and introduces a range of fiscal and monetary reforms. Apart from an overall stimulation of economic growth, these reforms also have a direct impact on automobile demand. In this respect, the steady reduction of the excise duty on automobiles has played a particularly important role to stimulate automobile demand. As a result of economic reform and growth, the Indian automobile market sees a sea change in qualitative and quantitative terms. The market develops from a sellers market, involving only a few domestic players, protected from internal and external competition, into a highly competitive buyers market, involving the presence of almost all global players in the automobile industry. In the early 1980s, India’s annual automobile production is around 40.000 vehicles (Mohanty et al., 1994; Becker-Ritterspach, 2008). At that

90 Hybridization of MNE Subsidiaries Table 4.2

Development of India’s FDI regime in the automobile industry

Time frame

Policy measures

1947–1984 Protectionism and state-led economic development

• Cars considered luxury products subject to government price controls • Introduction of comprehensive licensing system in 1951, restricting market entry and capacity expansion • GM and Ford leave India, following introduction of ‘Phased Manufacturing Programme’ (involving stiff local content requirements) in the early 1950s • Imposition of stringent import and foreign exchange controls • 1970 Monopolies and Restrictive Trade Practice Act (MRTP) regulates activities of large business houses, including automobile industry • 1973 Foreign Exchange Regulation Act (FERA) introduces heavy restrictions on FDI in India (40% equity restriction for foreign investors) • Quantitative restrictions on imports in place • Tariff structure designed to restrict the market

1985–1992 First deregulation measures introduced

• Foreign direct investment in the auto industry is limited to a few selected Japanese players (Suzuki for passenger cars, Honda for two wheelers, Mazda and Mitsubishi for light commercial vehicles) • 40% equity restriction for foreign investors still in place • Local market liberalization through introduction of ‘broad-banding’ system in 1985 • Licence requirements and qualitative restrictions on imports stay in place • Import tariffs are lowered

1993–2000 Shift towards internal and external market liberalization

• 1993 abolishment of licensing system for passenger car manufacturers • FDI up to 51% (fast track approval) and more allowed • Reformulation of ‘Phased Manufacturing Programme’ • Pre-entry security for investment decisions becomes obsolete • Several international players enter the market • FDI still subject to a host of restrictions involving the signing of Memorandum of Understanding, stipulating: minimum investment levels, foreign exchange neutrality, export requirements and localization/local content schedule (50% in 3rd and 70% in 5th year) • Import tariffs further reduced • Reduction of excise duty for passenger cars

Empirical Results 91 Table 4.2 (Continued) Time frame

Policy measures

Since 2000 New liberalization push

• • • •

100% FDI through automatic route is introduced Further reduction of import tariffs Removal of quantitative resections on imports Abolishment of Memorandum of Understanding including local content requirements and dropping of minimum levels for investment • New range of international automobile companies enter or are about to enter the Indian market, including Renault, BMW, VW and Audi • Further reduction of excise duty

Source: Compiled from Mohnot (2001), Becker-Ritterspach and Becker-Ritterspach (2008), ACMA, (2006), Ministry of Heavy Industries and Public Enterprises Government of India (2006).

point in time, there are only three players – Premier Automobiles Limited (PAL), Hindustan Motors Limited (HML), and Standard Motors Products of India Limited (SMPIL) – producing passenger cars for the Indian automobile market (Mohanty et al., 1994; Becker-Ritterspach, 2008). This scenery changes dramatically with Suzuki’s market entry in the 1980s and the entry of new players in the wake of the 1990s market liberalization. Table 4.3 provides an overview of the major players entering the Indian automobile market after the country’s deregulation and liberalization. For Indian consumers these developments imply a shift from being able to choose among two to three models that are marked by outdated technology to being able to choose from an ever growing range of stateof-the-art car models. By the 2000s, the Indian automobile market has developed into a differentiated automobile market. Based on vehicle length and price, India’s automobile market is commonly segmented as follows: A1/A mini segment (up to 3400 mm;
Table 4.3

Major players in the Indian automobile industry Partner (production site)

Share of the foreign partner

Car or MUV models

Production (2006–2007), in units

BMW India (2007)

(Chennai, Tamil Nadu)

100%

BMW 3 and 5 series

(Goal: 1700 p.a.)

Daewoo (1994)

Formerly DCM (Surajpur, Uttar Pradesh,)

74% → 91% → 100% → 0%

Matiz, Cielo, Nexia

Production stopped

Fiat India Pvt. Ltd (1996/97)

Premier Automobiles (PAL) (Mumbai, Maharashtra)

51% → 76% → 93% → 95% → (0%)

Uno, Siena, Palio, Petra, Adventure

Production shifted (see below)

Fiat India Automobiles Private Ltd (2007)

Tata Motors (Pune, Maharashtra)

50%

Palio, Grande Punto, Linea and TATA cars

1715

Ford India Pvt. Ltd (1995/2005)

Mahindra & Mahindra (Chennai, Tamil Nadu)

50% → 85% → 90% → 100%

Ford Ikon, Mondeo, Endeavour, Fusion, Fiesta

39423

General Motors India Pvt. Ltd (1994)

Hindustan Motors (Halol, Gujarat & Talegaon, Maharashtra)

50% → 85% → 100%

Opel Corsa & Sail, Chevrolet Optra, Tavera, Spark, Aveo, Captiva

36879

Hindustan Motors Ltd (1942)

(Uttarpara, West Bengal; Chennai, Tamil Nadu; Hosur, Karnataka; Vadodara, Gujarat; Indore, Madhya Pradesh)

Ambassador

13481 (including Mitsubishi models below)

Mitsubishi

Hindustan Motors (Chennai, Tamil Nadu)

10%

Lancer, Pajero & Montero

Honda Siel Cars India Ltd (1995)

SIEL Limited (Noida, Uttar Pradesh)

90%

Honda City, Civic, Accord, CR-V

59152

Hyundai Motor India Ltd (1996)

(Chennai, Tamil Nadu)

100%

Santro, Getz, Accent, Elantra, Verna, Sonata,

314604

92

OEM (foundation date)

Armada, Bolero, Scorpio, MaXX, Savari, Mahindra Major, MM550, Commander

91216

49%

Logan

614

Government of India (Gurgaon & Mansear Haryana)

1982: 26% 1989: 40% 1992: 50% 2002: 54,2%

Maruti 800, Zen, Alto, Swift, Omni, Versa, WagonR, Baleno, Esteem, Grand Vitara, Gypsy, SX 4

666935

Mercedes-Benz India (1994)

Tata Motors (Pune, Maharashtra)

51% → 76% → 86% → 100%

C-class, E-class, S-class

2053

Peugeot (1995–97)

Premier Automobiles (PAL)

50% → 33,96% → 0%

Peugeot 309

Production stopped

Premier Automobilies (1944)

Formerly Peugeot and Fiat (Mumbai, Maharashtra)

Padmini

Production stopped

Skoda Auto India Pvt. Ltd (2000)

(Aurangabad, Maharashtra)

Octavia, Superb, Laura, Fabia & Audi models

12748

Tata Motors Ltd. (Telco) (1945)

(Pune, Maharashtra)

Indica, Indigo, Marina (Nano) & Sumo, Safari, Spacio

245556

Toyota Kirloskar Cars India Ltd (1997)

Kirloskar Group (Bangalore, Karnataka)

70% → 88,86%

Corolla & Innova

50210

Volkswagen (2006)

(Pune, Maharashtra)

100%

Goal is production of Passat, Jetta, Tuareg

–

Mahindra & Mahindra Ltd (1945)

(Mumbai, Nashik, Igatpuri, Maharashtra; Zaheerabad, Andhra Pradesh)

Mahindra Renault Pvt Ltd (2005)

(Nashik, Maharashtra)

Maruti-Suzuki India Ltd (formerly Maruti Udyog Ltd) (1981)

100%

93

Source: Adopted from Becker-Ritterspach (2008) and SIAM (2008).

94

Hindustan Motors 14%

Premier Automobiles 23%

Maruti 63%

Source: Compiled from Mohnot (2001), p. 61; D’Costa 2005, p. 86.

Figure 4.1 Market share in the passenger car sector by manufacturer in India, 1990–91

Ford 6.4%

Fiat 3.9%

Tata Motors Ltd. 12.3%

Hindustan Motors 2.7%

Honda Siel 2.0%

Andere 1.7%

Maruti 54.1%

Hyundai 16.9%

Source: Compiled from Centre for Industrial & Economic Research (2002), p. 9.

Figure 4.2 Market share in the passenger car sector by manufacturer in India, 1999–2000

Empirical Results 95 Honda Siel 4.35% Hindustan Motors 1.78%

General Motors 1.91% Daimler Skoda Chrysler 0.89% 0.22%

Fiat 0.66%

Toyota 1.33%

Ford 3.03%

Tata Motors Ltd. 17.66%

Maruti 50.87%

Hyundai 17.30%

Source: Compiled from SIAM (2006), p. 107.

Figure 4.3 Market share in the passenger car sector by manufacturer in India, 2004–2005

While India’s pre-liberalization production capacities are not able to satisfy market demand, India’s post-liberalization production capacities soon surpass actual market demand. In 2003, the Indian automobile industry is suffering from a severe production overcapacity. Although market demand grows steadily (see Figure 4.5), it remains below expectation for many international automobile entrants. In the 2000s, it is mainly the mini, the compact and the mid-size segment that provide decent production volumes for the automobile manufacturers in India. Interestingly, it is not the mini, but the compact segment, that grows fastest in recent years (see Figure 4.6).

Supply market conditions in India’s automobile industry Suppliers As the Indian automobile industry is for decades very small, isolated, and monopolized sellers market, the less than a handful of auto manufacturers in the market provide little volume for the development of a

96 1400000 1200000

Numbers

1000000 800000 600000 400000 200000

5 00

4 20

04

–2

00

3 20

03

–2

00

2 20

02

–2

00

1 20

01

–2

00

0 20

00

–2

00

9 19

99

–2

99

8 –1

99 98 19

–1 97 19

–1 96 19

19

95

–1

99

99

7

6

0

Year Passenger cars

MUVs

LCVs

M&HCVs

Source: SIAM (2002); SIAM (2006).

Figure 4.4 Development of the Indian passenger and commercial vehicle market (sales), 1995–1996 to 2004–2005

800000 700000 600000

Mini (bis 3400 mm/⬍5000 €) Compact (3401–4000 mm/5000–8000 €) Mid-size (4001–4500 mm/8000–13,000 €) Executive (4501–4700 mm/13,000–22,000 €)

500000

Premium/Luxury (4701–5000 mm/22,000 €⫹)

400000 300000 200000 100000 0 2001–02

2002–03

2003–04

2004–05

2005–06

Source: Compiled from SIAM (2006) p. 17 (data for 2001–2002 to 2004–2005); ACMA (2006) (data for 2005–2006).

Figure 4.5 Development of the Indian passenger car market by segment, 2001–2002 to 2005–2006

Empirical Results 97

strong supplier industry. In addition, the licensing system, low volumes, supply uncertainties and the governing tax regime provide little incentive for car manufactures to outsource parts or components (D’Costa, 2003). In the absence of market growth and competition, existing suppliers have little capital and incentive to upgrade their manufacturing technology and facilities (Bhargava, 2002). Moreover, the dispersal of the suppliers over different states in India and the absence of the Value Added Tax discourage concentration processes (ACMA and SIAM Interviews). As a result, the Indian automobile suppliers remain small players, feature low production capacities, low technological know how and low quality levels (D’Costa, 2003; Bhargava, 2002). The contractual relations in the industry are based on low trust and arms length relationships. The supply logistics are shaped by supply uncertainties, the geographical dispersion of suppliers and the poor transportation infrastructure. As a result, high inventories are for many manufacturers a necessity. Before the beginning of deregulation in 1980s, Indian suppliers are unable to live up to international standards, especially not in terms of quality, quantity and price. This scenario changes, however, with the market entry of Suzuki in the early 1980s and the entry of other international automobile suppliers asking their suppliers to follow and forge local linkages (Bhaktavatsala, 1993; D’Costa, 2003; Sutton, 2004; Becker-Ritterspach, 2008). The ‘follow sourcing strategy’ implies that big international suppliers, such as Bosch, Cummins, Delphi, Denso, Valeo, or Visteon, follow their buyers and either establish wholly owned subsidiaries or engage in local tie-ups and JV (Humphrey, 1999; Humphrey, 2000; Bhavani, 2002; Humphrey, 2003). For local Indian suppliers the entry of the international suppliers translates, in turn, into a huge challenge. The pressure to catch up with international cost, quality and technology standards is in many cases only possible by means of international cooperation. Humphrey also stressed in this context, the entry of international suppliers has led to a ‘clear marginalization of locally owned companies’ (Humphrey, 2003, p. 136). Nevertheless, in the 2000s the automobile manufacturers in India can draw on a local supplier base that is increasingly internationally competitive in terms of manufacturing cost, capacity, technology and quality (ACMA, 2006).

Human resources India’s education efforts have a legacy of focusing on higher education (Becker-Ritterspach, 2000). While the number of university graduates is among the highest in the world, India as a whole faces still high levels

98 Hybridization of MNE Subsidiaries

of illiteracy. This imbalance is partially mirrored in the availability of human resources for the automobile industry. While the Indian labour market offers a decent supply of highly qualified managers and a large low-cost labour pool, concerns have been raised about a shortage of skilled labour and engineers (Belzowski et al., 2007). Apart from the shortage of human resources in quantity terms, there is also a concern about the quality of graduates for the automobile industry (Belzowski et al., 2007). A common complaint about India’s human resources is a lack of practical skills. It is a problem that is rooted in the segmentation of India’s higher technical education system, a socio-religiously founded prejudice towards physical/manual work, a poor funding of educational facilities and outdated curricula (Heitmann, 1995). Finally, although the Indian automobile industry has seen a double digit wage growth in recent years, the country remains a low-cost labour country.

5 The Case of Maruti–Suzuki India

Introduction In the 1960s and 1970s the Suzuki Motor Company (SMC) starts with the internationalization of motorbike production. However, it isn’t until the 1980s that the company develops a strong internationalization of automobile production. The company’s first automobile assembly site opens in 1976 in Indonesia. In 1981, SMC signs its first business tieups with the General Motors Corporation (GM). In 1982, SMC starts an assembly operation in Pakistan and in the same year a JV agreement is signed with Maruti Udyog Limited (MUL) for the set-up of an integrated production site in India. SMC internationalization efforts in the 1980s also include a JV with Santana Motors in Spain (1985), an agreement with General Motors Corporation of Canada to establish a production JV – better know as CAMI (1986) – and the establishment of assembly sites in New Zealand (1984), Columbia (1987) and Egypt (1989). In the 1990s this pattern of internationalization continues. In Asia SMC establishes wholly owned operations, ties ups and JVs for the production of passenger cars in Korea, China, Vietnam and Myanmar. In Europe, SMC builds an integrated production plant in Hungary (Suzuki, 2004). In addition, SMC enters into a number of collaborations and strategic alliances, the most important of which with General Motors in 1998. The strong international expansion in the 1980s and 1990s is followed by a phase of consolidation and strategic reorientation in the 2000s. Maruti–Suzuki India (MSI), for example, is no longer mandated to mainly serve the domestic market, but receives the mandate to produce for the European market (The Financial Express, 2002). In the 2000s, SMC is in the process of reorganizing its international division of labour. SMC envisions MSI to develop into a global R&D and production hub for small cars. 99

100 Hybridization of MNE Subsidiaries

Global product strategy SMC’s overall generic product strategy in the passenger car segment is between a cost focus and a cost leadership strategy (cf. Porter, 1980). SMC focuses mainly on light, sub-compact and small car market segments (Kasahara,1994). Along with this product focus, the company targets the less wealthy but more populated areas of the world, including India, China and Eastern Europe (Reference for Business, 2003). Although Suzuki has widened its range of segments, its product strategy for emerging economies can be best described as a cost focus orientation. As a logical concomitant of this product strategy, SMC relies on high volumes and stresses constant cost reduction in its production organization. While SMC has in 2003 an overall product range of 14 models, MSI offers in India only nine – most of which targeting the lowest market segments (MSI 4). All of these models are essentially developed in Japan and the level of customization to the Indian environment is rated low (MSI 2). In the past, SMC has been hesitant to offer the latest models in its developing country markets. This is also reflected in MSI’s product portfolio showing a time-lag and a technology gap compared to model introductions in more developed markets (MSI 1, 2, 3). However, with the intensification of competition in the Indian automobile market and MSI’s increasing international production mandate, the time-lag/gap narrows in the 2000s.

Establishment mode and equity development SMC enters the Indian passenger car market in the early 1980s in the format of a JV with Maruti Udyog limited, a PSU at the time. Although the company is established before SMC’s involvement, the JV is basically a Greenfield start-up. Founded by Sanjay Gandhi in 1970 and taken over after its failure by the Indian Government in 1981, Maruti Udyog offers little more than a plot of land for production (MSI 4). For SMC, despite the risks involved, a market entry into India provides an opportunity to advance its lagging internationalization (Venkataramani, 1990). The entry allows getting a foothold in a market, which remains closed for other international competitors in the passenger car sector, which lacks serious domestic competition and which offers a huge unsatisfied demand in those lower segments, SMC specializes in. The Indian Government, in turn, looks for a foreign collaborator because it lacks the capabilities to build a modern, large scale passenger car production facility (Mohanty et al., 1994). After a prolonged search period, the Indian Government finally selects SMC as its partner. SMC convinces with its small car product portfolio. Moreover, SMC promises to

Empirical Results 101

transfer the much sought after Japanese manufacturing culture. Most importantly, SMC’s equity participation offer is higher than that of all other contenders (Venkataramani, 1990). On the 2 October 1982 a Joint Venture and a 10 year licensing agreement (for the transfer of technology) are signed between Maruti and Suzuki. The equity participation between the Indian Government and SMC is set at 74 per cent and 26 per cent. The JV agreement includes a provision allowing SMC to raise its equity to 40 per cent. In 1989, SMC exercises this right and raises its stake to 40 per cent. When the ten-year agreement between the JV partners ends in 1992, a new JV agreement is signed, allowing SMC to raise its share to 50 per cent. Around the same time, India embarks on its new economic policy, which fundamentally changes the country’s FDI conditions. In 1993, the licensing system is abolished for the automobile Industry, which essentially invites potential competitors for MSI into the country. Feeling the heat of the emerging competition, SMC asks the Indian Government for a majority stake in MSI in 1997. The Indian Government declines the request, however, and period of conflict between the JV partners follows. It is not until 2002, following a revived disinvestment policy by a newly elected government, that SMC is finally able to increase its stake. After a rights issue and an initial public offering in 2002 and 2003, MSI becomes a subsidiary of Suzuki. While the Indian Government continuously reduces its stake in the company in the 2000s, SMC’s stake remains stable. In 2007, the SMC has a 54.21 per cent stake in the company followed by institutional and other shareholders.

Location, production programme and market share SMC’s dominant mode of establishing foreign production sites has been setting up satellite assembly plants (Reference for Business, 2003). Only a few of SMC’s sites are fully integrated production sites. MSI is one of them and is in 2003, with an installed capacity of 350,000 units and around 4600 employees the biggest and most important site among SMC’s foreign operations. MSI’s three main production plants are located in Gourgaon, the State of Haryana. They are all integrated plants with flexible assembly lines. Most of the lines are able to handle different variants in changing sequences (Muthukumar, 2004; Mohapatra and Dittakavi, 2003). The plants perform all major automobile production steps, including: engine manufacturing in the three machine and engine assembly shops; blanking and forming in the balking line and press shop; welding and painting in the three weld and paint shops; and assembly in trim, chassis and final line, followed by final vehicle inspection (MSI 3). In 2007, a forth plant is opened in Manesar at new site in Haryana. The

102 Hybridization of MNE Subsidiaries

new site raises the installed capacity to 450,000 (and the actual capacity close to 700,000) units. In line with SMC’s core competence in the small car segment and the Indian Government’s wish to mass-produce a ‘people’s car’ (a small, energy efficient and affordable car) production commences in 1983 with the A-segment Maruti 800. As an affordable car for India’s growing middle class, the Maruti 800 proves to be a run-away success. Although the company introduces new and also higher segment models over time, the models catering to the A and B market segments remain MSI’s bread and butter. In 2003, for example, the A segmented Maruti 800 still accounts for 43.4 per cent of the company’s domestic sales (Chavan, 2003). At the time of research, the site produces nine base models (Maruti 800 (A), Alto (B), Wagon R (B), Zen (B), Baleno/Altura (C), Esteem (C), Gypsy (UV), Omni (MPV), Versa (MPV)) in over 50 variants. Enjoying preferential treatment by the government in the first years (Chatterjee, 1990, p. 39) and a first mover advantage, thanks to a restrictive licensing policy until 1993, MSI becomes the undisputed market leader. In 1998, MSI’s market share peaks at around 83 per cent. Although foreign and domestic competitions cut into MSI’s market share, the company remains the market leader. In the 2000s, MSI still has an overall market share of around 50 per cent. Until 2008, MSI dominates the small car market with a share of 100 per cent in the A segment. While it remains to be seen what effect the introduction of Tata’s low-priced Nano-model will have on MSI’s market share, it is certain that no other company has dominated and institutionalized the Indian automobile market and industry as MSI has in the past.

Analysis of the hybridization profile The general transfer scenario The set-up of the MSI’s production system follows from the beginning a comprehensively defined template. This template is SMC’s main plant in Kosai, Japan. The template transfer becomes possible because, despite some differences in production volumes and product variety, the overall task profile is sufficiently similar to follow a transplant approach. Both, the available literature (e.g. Mohanty et al., 1994) and the interviewees (MSI 1, 2, 3, 4, 5) underline that SMC’s Kosai plant is from the start until the present day the main point of reference. This is also reflected in the following comment: No as far as I can see from where I am, it is definitely a one-way flow and Kosai is the benchmark and everything that we do tries to

Empirical Results 103

emulate what happens in Kosai. Yea, there is no benchmarking with other plants in the rest of the world or anything like that either. It is all with respect to Japan and with respect to Kosai. (MSI 2) In no other research case, do interviewees suggest with a similarly clear language – using the words ‘replica’ or ‘duplicate’, etc. – the copycat character of the production system set up (MSI 1, 2, 3, 4, 5). Even more than SMC’s management, it is the Indian top management that actively supports the comprehensive replication of SMC’s production system (Chatterjee, 1990; Venkataramani, 1990; Mohanty et al., 1994). According to Mohanty et al. (1994), this comprehensive transfer intent is already fixed in the initial technology transfer agreement between the JV partners. The transfer essentially targets all major dimensions of the production system ranging from structural attributes to core aspects of the process organization (Mohanty et al., 1994, see also Venkataramani, 1990). The organizational structure of MUL is devised on exactly the similar pattern of its Japanese partner i.e. Suzuki Motor Company. The staffing of people, factory layout, installation of plant and machinery, etc. are inherited in toto form Suzuki Motor Company, Japan. The basic reasoning of such a structure is to realise the industrial success based on Japanese Business Management Systems in India. (Mohanty et al., 1994, p. 133) While the transfer targets all dimension of MSI’s production system the main thrust of the transfer effort is on human resources and aims at replicating behavioural patterns and attitudes of SMC’s Japanese work culture (Mohanty et al., 1994; Kasahara, 1994, p. 78). The transfer effort strongly aims at the internalization of soft practices and attitudes within the Indian workforce (MSI 1, 2, 5). It is stressed that such a transfer goal cannot be achieved in one step and requires intensive human interaction, instead of document transfer, as a transfer mechanism. (Shirali, 1984; Venkataramani, 1990; Kasahara, 1994).

The functional differentiation of the organization structure Transfer scenario MSI’s functional differentiation is designed to follow the example of the Kosai plant in Japan (MSI 1, 4, 5). Despite being the junior partner in the JV the entire production set-up, including the functional differentiation, is SMC’s responsibility. After all, SMC introduces the products

104 Hybridization of MNE Subsidiaries

and is chosen as a JV partner because the Indian partner lacks the capability to design and produce automobiles on its own (Venkataramani, 1990). As a Greenfield site, MSI also lacks a pre-existing functional set-up (MSI 4). At the same time, there are no indications that any local/host strategic or institutional distance prevents the transfer of SMC’s home plant functional set-up. Specifically, as the MSI operation aims at the establishment of an integrated plant with a similar task profile (high volume production of low segment cars) and level of vertical integration, the functional differentiation required, is about the same. The Greenfield nature of the site also implies that there is no existing local institutional context that discourages a transfer ex ante. In fact, local management very much invites the comprehensive transfer of SMC’s production organization. Thus, even though SMC is the minority partner, the capability gap between SMC and its local partner, the Greenfield nature of the project and the absence of major strategic and institutional distance suggest the transfer of SMC’s home plant’s functional differentiation.

Misfit The interviews suggest that the implementation of the functional layout of the Kosai template faces no major institutional or strategic misfits (MSI 1, 2). Given that MSI is a Greenfield site, there is no pre-existing task profile and corresponding functional configuration standing against the requirements of the foreign/parent template. There is also no indication that the local institutional context demands any changes in the transferred template. Again, the Greenfield nature of the site may be important explanation here as there are no pre-existing local institutional patterns that contrast with the template’s requirements. While there is no on-site strategic or institutional misfit and demand market conditions also suggest a transfer, there is some misfit between the foreign/parent template and the supply market conditions in the wider host context. The Kosai functional differentiation is based on an extremely low vertical integration of no more than 26 per cent in-house value addition (Bhargava, 2002; UNIDO, 2003). This requires, however, a developed supplier base (including reliable supply of energy), which the Indian strategic context does not provide in the early 1980s.

Mode of recontextualization Some adaptation of the foreign/parent template to the host context Once transferred, there is only little contextual (mis)fit/recontextualization pressure working against the implementation of Kosai’s functional differentiation. Yet, there is some misfit that results in an adaptation of

Empirical Results 105

the foreign/parent template. It is the absence of a developed supplier base and other infrastructural shortcomings that lead to some functional extensions, rather than fundamental changes, in the site’s functional differentiation. Due to the strategic misfit in supply market conditions, MSI creates additional functions that are not part of the functional differentiation in the Japanese operation. A case in point is the ‘Vendor Upgradation Division’, which has the task to develop a supplier base (Mohanty et al., 1994; UNIDO, 2003). MSI even establishes an on-site electricity unit, to secure its energy supply (Gulyani, 1999). Thus, there are some minor adaptations of the foreign/parent template through extension or internalization (Westney, 1993). However, overall, there is not much willingness on the part of the Japanese parent to change the template by, for example, internalizing the absent supplier base into the organization.

Major adaptation of local/host context to the foreign/parent template While there are no adverse local strategic or institutional demands or conditions calling for a substantial template change, there is also no locally existing structure to draw on. This implies that the organization, and the functional differentiation as a part of that, has to be newly created. This organization building is realized through a substantial resource mobilization by SMC. Specifically, in the form of Japanese expatriates who are transferred and provided with a mandate to establish MSI’s functional differentiation. The establishment is facilitated by a resilient local context, which is constituted by a top management that welcomes the transfer and young newly recruited workforce with few preconceived notions (MSI 1, 2). The strategic misfit with regard to host context supply conditions, in turn, leads to a massive host context adaptation. We shall see in more detail later, how SMC and MSI engage in a substantial effort to build a host context supplier base. This context creation allows MSI to replicate a crucial aspect of Kosai’s functional set up, namely its low level of vertical integration (UNIDO, 2003).

Outcome The functional differentiation of MSI’s production system can be best understood as an imitation of SMC’s home plant Kosai. Without providing detailed information, interviewees underline that MSI’s functional differentiation largely follows the mother plant Kosai in terms of different divisions, departments, sections and work groups (MSI 1, 2; cf. Mohanty et al., 1994). Moreover, available literature underlines the ambitious replication of the home plants levels of vertical integration equally

106 Hybridization of MNE Subsidiaries

suggesting a similar functional set up (UNIDO, 2003). This outcome rests on a foreign/parent template transfer that meets only a weak local strategic and institutional misfit. While there is some strategic misfit, mainly with regard to an absent supply market in the host context, this misfit is overcome by a recontextualization mode that involves a minor adaptation of the foreign/parent template and a major adaptation of the host context. Specifically, while the functional template is largely imitated, some functions are added. As these additional functions neither mirror typical foreign/parent nor local/host context patterns (D’Costa, 2003), they can be best understood as customized/novel solutions. Nevertheless, the dominant recontextualization mode comprises a creation of the local/host context in line the foreign template’s requirements. We can, therefore, describe the contextual constitution of MSI’s functional differentiation as dominantly imitation involving some customizations.

The hierarchical differentiation of the organization structure Transfer scenario Despite the comprehensive transplant approach, there are no indications that SMC also transfers its home plant’s hierarchical differentiation. Instead, the hierarchical differentiation of MSI shows clear signs of using a host context template similar to other PSUs in India (MSI 4, 5). The fact that this aspect is left out can be attributed to MSI’s equity mode. SMC enters a JV with a Government of India company as a minority partner. While SMC is mandated to shape the production organization, the shaping of certain structural aspects follows PSU rules and regulation. This is particularly the case with regard to policies of recruitment, staffing, remuneration and promotion and has strong implications for MSI’s hierarchical differentiation. MSI’s former chairman and managing director Bhargava (quoted by Chatterjee, 1990) underlines the point as follows: Yes, we have made some ‘adaptations’ of Japanese practices. We could not blindly copy everything. I do not know whether you should call it dilution or not there have been modifications in some of the management practices we follow. For example, in our manpower planning and in our manpower policies we have to take note of the fact that we work in a system where the kind of freedom which is available to the Japanese in rewarding employees, and granting pay scales, these things are not available (to us); so we have to follow a different thing in the system of assessment of people, the promotion policies and in all of these we have to follow something which is different from the

Empirical Results 107

Japanese. In our system, on the financial side we have to take note of the fact that we are a Government company and there are certain rules and regulations prescribed by the Government to be followed, whereas in Japan there is much less of that kind of thing. (Chatterjee, 1990, p. 134) Being essentially a PSU, it is the local partner’s mandate to take care of human resource planning and major human resource policies. This, in turn, impacts the transfer scenario, namely the use of a local/host template for the hierarchical differentiation of the company. Apart from the Government of India ownership or majority in early years, there are also indications that a transfer of SMC’s flat hierarchical differentiation would have been difficult given the local/host institutional demands for extensive organizational hierarchies.

Misfit Findings are inconclusive whether the local/host context template faces a misfit with regard to SMC’s foreign/parent context. Looking at the situation from SMC’s side, there is a substantial institutional distance between the host context template and what SMC probably would have preferred with regard to MSI’s hierarchical differentiation. A number of Japanese researchers who had the opportunity to research MSI underline the institutional distance between typical hierarchical differentiation in Japanese automobile companies and MSI’s 19 levels of hierarchy (Okada 1998). Kasahara (1994) finds that MSI, in comparison to SMC, features a much higher level of positional demarcation and rigidity. For example, in contrast to the Japanese context, it is impossible for operators to cross the line into supervisory levels in the first years of operation. Kasahara (1994) argues that in SMC, or Japan more generally, ‘professional duties’ tend to be only few and vaguely defined. Moreover, changes to ‘professional duties’ are readily accepted by employees based on management needs, personal training requirement and a strong in-house qualification system (Kasahara, 1994).1 The opposite is the case in MSI, which Kasahara (1994) attributes to the fact that staffing of different organizational levels is tightly coupled with specific educational qualifications and with the wider condition that hierarchical demarcations play a central role for an Indian employees’ work identity (Kasahara, 1994). Yet, while the institutional distance appears substantial, there is little evidence that the institutional distance translates into a recontextualization pressure on the local/host template from the foreign/parent. After all, SMC is for a long time the junior partner in the JV and any serious effort to reduce

108 Hybridization of MNE Subsidiaries

hierarchical designations or levels would be up against a strong resistance from the local/host institutional context where steep hierarchical demarcation are deeply rooted. Finally, there is little evidence to suggest that the local/host template of hierarchical differentiation causes a misfit with foreign/parent strategic demands or conditions.

Mode of recontextualization Adaptation of foreign/parent demands to the local/host template

Based on MSI’s equity mode, SMC is not able to transfer the hierarchical differentiation of the Kosai template. At the same time, interviews remain inconclusive with regard to the question whether SMC pushes for a change of the local/host context template. Even if SMC pushes for such a change, these demands are not easy to realize. For SMC is until 2003 in no majority position in the JV. While SMC delegates have the mandate to shape crucial aspects of MSI’s production system, the ultimate control remains until 2003 in the hands of Indian managers. Based on the equity mode, SMC’s ability to shape the organization is limited. Furthermore, in the Indian institutional context, hierarchical differentiation and regular promotions to ever better designations are expected and play a very important motivating role. Against this background, it is doubtful whether SMC seeks a reduction of hierarchical designations and levels after the takeover. Both the equity mode and entrenched local/host institutional conditions render it likely that SMC either refrains from demands for a flatter hierarchical set-up or that demands from SMC are adapted (or flatly rejected) to maintain the local/host context template.

Minor (ceremonial) adaptation of the local/host template to foreign/parent context There are strong indications that the hierarchical differentiation of MSI is created in line with PSU rules and regulations. As such it reflects a typical host context template. Even when MSI seizes to be a PSU and when its majority is acquired by SMC, no major reductions of hierarchical designations are reported. However, at the time of research it is too early to tell if the shifts in equity translate into foreign/parent demands for a changed hierarchical differentiation. The only adaptation of the local/host template observed, appears to be of ceremonial nature. A human resource manager interviewed suggested that MSI has a flat hierarchy in terms of ‘responsibility levels’ and ‘salary levels’ comparable to those of SMC (MSI 5). According to the manager, MSI has only three ‘responsibility levels’ – as publicly purported by MSI’s website2 . At the same time, the human resource manager (MSI 5) and other interviewees (MSI 1, 2) admited that there are many more hierarchical levels

Empirical Results 109

than reporting levels. It, therefore, remains unclear whether there is a meaningful difference between responsibility levels and hierarchical designations. While MSI’s possible decoupling of hierarchical designations and reporting levels may mirror a response to demands from the Japanese parent, such ‘decoupling’ is also not uncommon in typical Indian companies. However, there is little indication that MSI’s publicly purported flat structure is more than ‘ceremonial adoption’. We can conclude that the host context template of hierarchical differentiation is, if at all, only marginally adapted to foreign/parent context demands. This is either the case because the foreign/parent does not pose demands for change or because the host template is not very resilient due to the equity mode and the deeply ingrained institutional patterns in the local/host context.

Outcome Despite its overall transplant approach, MSI’s organizational hierarchy shows clear signs of being based on a host context template. MSI’s hierarchical differentiation is created in line with PSU rules and regulations. Referring to MSI Kasahara (1994) concludes along similar lines: ‘[t]he fact that along with the occupational categories, a hierarchy is also clearly defined, is a characteristic of firms run by the Indian government’ (Kasahara, 1994, p. 76). The structure that MSI features is a steep structure of 19 hierarchical designations (MSI 5). Chatterjee (1990) explicates these levels as follows: Employees are divided into levels beginning at one going up to 19. Initial recruitment has been generally made from among candidates with an ITI qualification for levels 3–7. Level 8 employees are normally diploma holders and the group 8–10 work as Supervisors. Level 11 recruits are drawn from Engineering graduates on the technical side. Levels 11–13 are in the grade of Section Managers. Above this layer is the level 13–17 group having Department Managers. Finally there are Divisional Managers who vary between levels 15–19. (Chatterjee, 1990, p. 47) These 19 designations include: several Operator designations, Supervisor, Executive, Manager, Senior manager, Deputy General Manager, General Manager, Deputy Departmental Manager, Departmental Manager, Divisional Manager, Joint Managing Director, Managing Director (MD), Chairmen (MSI 5). Thus, MSI features a hierarchical differentiation that is very similar to other companies in India, particularly in the public sector.

110 Hybridization of MNE Subsidiaries

In summary, a host context template is used that is institutionally distant from SMC’s set-up. However, this distance does not amount to an effective foreign/parent recontextualization pressure on the host template. The only sign for such a recontextualization is the possible decoupling of responsibility levels and hierarchical designations. Yet, such a decoupling is not uncommon in Indian companies. Overall, we can conclude that there is little or no adaptation of local/host template to foreign/parent demands and conditions. Therefore, the hierarchical differentiation can be best described as localization.

Technical configuration of the production process Transfer scenario As part of the overall transfer approach the Kosai plant is used as a template for the set-up of the Indian site’s production processes. Again, the local JV partner’s lacking capability for developing suitable cars and corresponding production processes are a key reason why the Indian Government and its management invite a template transfer by SMC. Irrespective of being the junior JV partner, it is SMC’s task to establish the site’s production process. This comprises the set up of the plant’s basic factory layout, process design and technical hardware. The Greenfield nature of the operation, combined with the lacking capabilities of the local JV partner and the fact that the cars to be produced are SMC models, explain why no local/host template is used. However, while SMC delegates are in charge of establishing the technical configuration of the MSI site, there is also no all out transfer effort of the Kosai template. With regard to adopting Kosai’s template for MSI’s basic factory layout and process design, there are no local/host strategic or institutional misfits. Transfer is pursued in this respect because the Indian demand conditions – mass demand for a small low-segment car – imply a task profile for MSI similar to that of SMC’s Kosai plant. In addition, there are no local/host institutional obstacles working against the transfer of Kosai’s basic production layout and process design (MSI 4). This is different, however, with respect to the transfer of Kosai’s technical hardware configuration. On the one hand, there is a substantial strategic misfit between what the foreign/parent template is strategically designed for and what the host strategic context offers. Operating in a developing country implies not only a highly price sensitive demandside for low-segment cars, but also a supply side with labour costs that are much lower than in high-wage economies like Japan. This, in turn, rules out a comprehensive transfer of Kosai’s capital intensive technical

Empirical Results 111

hardware configuration, marked by high levels of mechanization and automation. In addition to this strategic misfit, there is also an institutional misfit working against a full transfer of Kosai’s technical hardware configuration. Especially, in the early years, the Indian JV partner pushes hard for a labour-intensive production set-up. As MSI is a PSU, the Indian Government and union representatives demand employment-generating production concepts. Although MSI sees neither a foreign/parent template transfer, nor a local/host template use, there is a foreign/parent and local/host context the site has to respond to with regard to its technical hardware configuration.

Misfit There are few indications that the foreign/parent template of production layout and process design face any substantial strategic or institutional misfit once transferred. There are only some limited constraints with respect to the site’s geographical properties and inherited buildings. With regard to the company’s technical hardware configuration there are, however, signs of substantial misfit. This misfit primarily revolves around contradictory demands between the foreign/parent strategic and the local/host institutional context regarding the appropriate technology/labour mix. Specifically, while the local JV partner demands labour intensive production, SMC’s increasingly demands a reduction of what it perceives as ‘rampant overstaffing’. Looking at it from SMC’s side we can call this as a strategic misfit because SMC finds the institutional demands of the local partner problematic in terms of production cost and quality.

Mode of recontextualization Adaptation of the foreign/parent template and demands to the local/host context As far as the basic factory layout and process design of the Kosai template is concerned, there is little adaptation of the transferred template to the local/host context. After all, there is little recontextualization pressure in this respect due to a strategic fit and an absent institutional misfit. This is different with regard to the configuration of the technical hardware. In this respect, there is only a selective transfer of the Kosai template. While the technical hardware of up-stream process steps – such as pressing and stamping – sees a transfer, downstream process steps – such as body welding and final assembly – see much less of a transfer (MSI 3). In the latter areas, the technical hardware set-up is not built on the Kosai template but rather created or tailored

112 Hybridization of MNE Subsidiaries

to respond to the foreign/parent as well as to local/host context. Interestingly, there is some changing response to these partly contradictory demands over time. While in the first years, the technology/labour mix reflects more an adaptation of foreign/parent demands to local/host context demands, the situation turns the other way around as the equity shifts in favour of SMC. Thus, under the condition of minority ownership the local/host context is not very resilient to meet foreign/parent demands for more productivity and cost reduction through lower staffing levels and increased automation.

Adaptation of local/host context to the foreign/parent template and demands To implement the factory layout and process design in line with the Kosai template, a local context has to be created at the MSI site. The strong presence of Japanese expatriates, massive transfers of personnel across all levels, from India to Japan and vice versa, the Greenfield nature of the site as well as the local management support, allows a replication of the Kosai set-up (MSI 1, 2, 5). What is more, this process of transfer effort continues over the years as both MSI and Kosai grow and develop. In the 2000s, for example, SMC initiates yet another initiatives called ‘Challenge 50’ and ‘Next leap’, to bring MSI’s processes and productivity performance even further in line with those of Kosai (Business Standard, 2005). In the area of the technical hardware configuration there is much less of local context creation in line with a foreign/parent template. Apart from some transfer of special machinery and equipment, the technical hardware configuration is based on the simultaneous response to the foreign/parent and local/host context. On the one hand, the technical hardware configuration (like the entire process) has to satisfy SMC’s strategic demands of manufacturing of a specific range of models at certain volumes, quality and cost (adaptation of the local site set up to foreign/parent strategic context). On the other hand, the set-up has to respond to a not very resilient host strategic and institutional context, marked by low labour cost, highly price sensitive customers and demands for labour intensive manufacturing (adaptation of the local site set up to local/host strategic and institutional context). As a result of different (partly contradictory) demands, a shifting equity situation between the JV partners, and a growing local capability, a customized technical hardware’s configuration emerges. Particularly in the 2000s, when MSI faces increasing competition and SMC acquires a majority stake in the company, the local technology/labour mix is increasingly adapted to foreign/parent demands for cost reduction and productivity

Empirical Results 113

increases. It involves repeated rounds of layoffs, a re-composition of the workforce, combined with an increased application of in-house developed automation technology (Muthukumar, 2004; c.f. Maruti Udyog, 2004). Reflecting Suzuki’s aim to reduce what it sees as rampant overstaffing and to bring down cost, Maruti–Suzuki reduces its level of permanent employees through VRSs. The first round of employee reduction starts in 2001–2002. As 19 per cent of Maruti–Suzuki’s employees accept the scheme, the employment level comes down by more than 1050 employees, reducing it employment levels from 5770 in 2000–2001 to 4590 in 2002–2003. In 2003–2004 Maruti–Suzuki initiates a second round of VRS, sacking another 1250 workers. In addition, shifts in the workforce composition take place. MSI increasingly sources out ‘lowvalue jobs’ to contract labour, many of these contract workers are former permanents (Red Herring Prospectus, 2003). In 2006, out of around 5000 workers only 1300 are permanent workers. Compared to the wage of the shrinking core of permanent workers (earning between Rs25000 and Rs35000, including incentives), the contract workers earns Rs4500–5500 (Gurgaon Worker News, 2007).

Outcome The hybridization outcome of MSI’s technical configuration has to be described in a differentiated manner. MSI’s basic factory layout and process design can be best captured as imitation, based on foreign/parent template transfer, low misfit/recontextualization pressure and a creation of the site in line with the template. An Indian engineer familiar with both sites indicated that MSI factory layout and process design has even grown closer to the Kosai plant over the years (MSI 1). Similarly, MSI’s material flow system is based on SMC’s Kanban concept and is said to have moved very close to Kosai’s practices (MSI 4). Mohanty et al. (1994) and Gulyani (2001) also indicate that MSI’s in-house logistics increasingly approach Kosai’s levels of work-in-progress-inventories. In addition, the MSI site’s Maruti Production System (MPS), which ‘focuses on elimination of wasteful activities’, is said to mirror SMC’s production system (Red Herring Prospectus, 2003). This imitation outcome contrast with the outcome of MSI’s technical hardware configuration, which can be better described as being between imitation and customization. A part of Kosai’s technical hardware configuration is also transferred. This is mainly the case for upstream process steps such as blanking and pressing as well as with respect to specialized equipment such as CNC-machines, special tooling, jigs and a limited number of robots (MSI 1, Red Herring Prospectus, 2003). An imitation

114 Hybridization of MNE Subsidiaries

of Kosai manufacturing equipment can be found where it is either a process-related requirement (e.g. for quality reasons) or where the automation of highly standardized production steps offers cost savings despite low labour costs. In other areas, however, SMC refrains from such a transfer and creates local hardware solutions that are specifically customized to the foreign/parent and the local/host context. Areas where such customizations can be found are the weld shops and the final assembly lines. Here, the Indian site contrasts sharply with the mother plant Kosai. An Indian production manager commented in this context: Factory layout would be pretty much similar, yes. Certain portions, because I have visited certain plants with SMC, are of course heavily robotized and the weld shop here is very man-intensive, [there are] a lot of people here. (MSI 1) This contrast becomes more explicit when comparing productivity ratios. In 2001, MSI and SMC (Kosai) produce an annual per employee output of 100 cars and 140 cars respectively. While MSI achieves this output with around 100 robots, SMC’s Kosai operation relies on some 1400 robots (Mohan, 2002). In contrast to MSI, Kosai’s welding and machine shop is practically completely automated. Production in India, particularly the welding activities, have been far more labour intensive throughout MSI’s existence. For as long as comparatively low labour cost and high price sensitivity stays in place, a boost in MSI’s productivity levels is unlikely to be driven by imports of expensive automation technology (MSI 4). While process-related necessities and the absence of local/host context capabilities in pre-liberalized India of the early 1980s render the selected technical hardware transfers necessary, this necessity changes as the MSI accumulates know how. With the rise in capabilities, MSI develops in a number of areas home grown customized solutions. This is even more so the case as equity shifts in favour of SMC. Responding to foreign/parent strategic demands for higher productivity and cost reduction and, at the same time, host strategic context conditions of price sensitive demand and low labour cost supply conditions, MSI develops its own low cost equipment and automation technology (MSI 1). Basically, the McKinsey Global Institute’s study argues that, rather than just replicating jobs/processes from high-wage countries to lowwage ones, it makes more sense to actually use local talent to develop new processes. The study cites the example of MUL [later renamed to

Empirical Results 115

MSI] developing its own robots for its assembly lines at a fraction of the cost of the ones used by SMC in Japan. (Jain, 2003) In 2003, MSI has 122 six-axis and 25 in-house produced two-to-four axis robots (Red Herring Prospectus, 2003). Over time, the ratio of locally built robots increases further. In 2008, MSI has about 150 robots, out of which 71 have been developed in-house (Maruti–Suzuki, 2008). While SMC probably never entirely stops the transfer of certain technical hardware, MSI increases the development of customized solutions for its specific strategic context. These include: dies, two-to-four axis robots, multispot welders, welding jigs, automated trolleys, machine shop equipment (Red Herring Prospectus, 2003). The crucial incentive for their local development is their low cost compared to imports from the parent. Even though these customized solutions are locally developed, they are no typical local solutions. MSI technical configuration has always been more mechanized and automated than comparable sites in India. At the same time, it features much less automation than the foreign/parent’s home site. In summary, the overall outcome of MSI’s technical configuration of the production process can be best described as combining imitation and customization.

Work organization and human resource profile Transfer scenario From the beginning the transfer or a ‘Japanese work culture’ is a defined goal of the Indo-Japanese JV. This transfer goal even guides the search for a foreign JV partner and influences the choice of a Japanese partner (Venkataramani, 1990; Parekh, 1984). The transfer of SMC’s work concepts, corresponding skills and more general work dispositions, also referred to as ‘attitudinal know how and work-ethics’, are at the heart of the whole template transfer (Chatterjee, 1990; Parekh, 1984; Shirali, 1984). According to Mohanty et al. (1994) the transfer of such ‘soft’ aspects was part of the JV’s technology transfer agreement: The technology transfer agreement lays considerable emphasis on human resource i.e. the people who are ultimately responsible for the fulfilment of organizational goals. Their attitudes and approach to work need proper orientation. Methods for creating awareness about quality, productivity, and cost control are also ingrained in the collaboration agreement. Moreover, the agreement underlines the need

116 Hybridization of MNE Subsidiaries

to inculcate new work culture and work ethos in the organization for greater performance by undermining hierarchical rigidity. (Mohanty et al., 1994, p. 135) Although being the minority partner in the JV, SMC has the mandate to set up the work organization and develop a human resource profile in line with its home operation (MSI 1). SMC’s mandate suggests itself because the company introduces the product and has tested work concepts readily available. However, the main reason for SMC’s transfer mandate lies in the wish of the local JV partner to avoid the adverse institutional conditions in India at the time of market entry. While strategic context conditions – demand market conditions – are not in the way for such a transfer, the institutional conditions in the Indian context, are about the opposite, of what SMC’s work concepts require. Put differently, the perceived Japanese superiority with regard to work concepts and related human resource practices are seen as a solution to overcome, at least for MSI, the problematic employment relations that paralyze many industrial organizations in India at the time (Venkataramani, 1990; Mohanty et al., 1994; Okada, 1998; D’Costa, 2003). These conditions are characterized by mutually enforcing complex of hostile labour-management relations, extreme hierarchical-demarcations, labour-inefficiency, low labour involvement, frequent labour-unrest and low identification with the company (Venkataramani, 1990). Apart from the Indian side’s interest in overcoming adverse intuitional conditions by making an ‘experiment’ (MSI 1; Venkataramani, 1990: 124) and a fresh new start, it is also the Japanese JV partner that supports the emphasis on such a human resource centred transfer approach (Kasahara, 1994). Given the absent local manufacturing capabilities at the Greenfield site and given that SMC has a brand image to lose, SMC has to engage in a substantial transfers of work concepts, skills and work practices. Moreover, as the Indian site operates with much less automation technology than the home plant, the role of human resources becomes even more important. Thus, SMC takes the transfer mandate seriously and engages in a comprehensive transfer of its home plant’s work organization and human resource profile. As far as the work concepts are concerned the transfer not only comprises concepts related to direct manufacturing. In addition, many concepts are introduced over time that target indirect activities and practices that are geared at continuous improvement in products and manufacturing processes. These activities focus on permanent cost reduction, productivity increases as well as accident and waste reductions

Empirical Results 117

(MSI 1, 2). MSI’s human resources, in turn, are developed to match the basic skills and work dispositions of their Japanese colleagues. This is seen as a prerequisite to allow the implementation of high involvement work concepts (Shirali, 1984).

Misfit The fact that SMC’s work concepts and matching human resource profile are transferred because and despite of the institutional distance implies that they are either not readily compatible with extant institutional conditions or meet adverse local institutional demands. Clearly, the work concepts transferred neither meet a warm welcome nor are the skills and basic work dispositions required for their implementation readily available in the local/host context. In order to understand the initial misfits, we need to look at what is being transferred. Regarding the direct manufacturing organization on the shopfloor, much stress is put on transferring SMC’s teamwork concepts. SMC work concepts underline the crucial role of supervisors. Moreover, it calls upon the entire production management to take a hands-on-approach and to be close to the shopfloor (Venkataramani, 1990). The teamwork on the shopfloor involves a wide scope of job enrichment and job-rotation. A substantial range of responsibilities are integrated into the production line, such as quality assurance (zero-defect production). Like in Japan, workers can autonomously initiate line-stoppages. The job rotation not only involves aims at transferring worker across different work stations, but also across different departments. In addition, many concepts revolve around the maintenance, the continuous improvement of products and manufacturing processes and are based on individual as well as small group activities. The most prominent concepts followed in this regard, include: 5S, Kaizen, Quality Circles and Suggestion Schemes. To realize these concepts the Indian workforce has to acquire skills and work dispositions similar to those of their Japanese counterparts. Interviews (MSI 1, 2) and a literature study reveals that the transfer of work dispositions involves, in turn, three focus-areas: Work commitment and discipline (stressing the taking of responsibility, punctuality and attendance), cleanliness and quality awareness, cooperation and communication (D’Costa, 2003, p. 76).3 Now, the implementation of these work concepts, matching skills and work dispositions is not without difficulty. Particularly in the beginning the transfer of a teamwork concept, the central role of the supervisor, the taking of responsibility and the practice of job-rotation faces institutional misfits. For example, the implementation of teamwork faces

118 Hybridization of MNE Subsidiaries

substantial difficulties as team members find it hard to cooperate (e.g. Khare, 1999; Pitroda, 2001). These difficulties reflect socio-professional demarcations produced by India’s social stratification and education system (see Chapter 4). Similarly, there are indications that the central role of the supervisor is not readily accepted in the Indian work context. The following paragraph from a MSI’s manual allows such an interpretation: SUPERVISORS ARE MANAGEMENT MEN WHO MAKE THEIR WORKSHOPS SUPERIOR TO THOSE OF OTHER COMPANIES. Supervisors are in reality “front line managers” and should therefore, be “viewed with importance.” The Supervisor is “the pivot in the workshop operation” and holds the “important responsibility” for improving the workshop’s performance in safety, orderliness, productivity, loss avoidance and cost reduction. On him too rests responsibility for ensuring high morale of workers. (Venkataramani, 1990, p. 111) Apart from the supervisors, production managers in general, do not identify themselves with manufacturing operations on the shopfloor. There is also reluctance on the part of production management and engineers to be close to or get involved on the shopfloor. A senior SMC deputationist said in an interview that some young Indian engineers coming to work straight from a university and with no hands-on experience, tended to be impatient listeners. While all around them in the plant there were problems to be tackled and improvements to be made, they were ready to engage in discussions on other matters, especially on far out technologies calling for materials and equipment that could possibly be made available. Many of them had a theoretical bias rather than a practical orientation. They loved to sit at their desks and discharge their duties from the desk. “I try to push them on the shop-floor. ‘This is a car company’, I tell them. ‘Go out on the floor and at least see what a car looks like’.” (Venkataramani, 1990, p. 171) Equally related to socio-professional demarcations, the implementation of job-rotation also proves difficult. Kasahara (1994) reports a stark contrast between the ease to rotate personnel in Japanese companies and in MSI. He attributes the difficulties to ‘a strong prejudice based on the type of occupation’ in the Indian society (Kasahara, 1994). There are indications that the implementation of indirect high involvement work concepts is not passionately embraced either in the beginning (c.f. Khare,

Empirical Results 119

1999). In a case study on MSI, Som (2004) reports that it is difficult to implement quality circles, as Indian employees disperse immediately after work. At the same time, there are also complaints from Japanese expatriates about lacking practical skills and basic work dispositions (Venkataramani, 1990). Apart from skill related training needs, the larger number of training workshops on issues such as ‘quality and discipline’ (level 0 to 7), ‘quality and leadership’ (level 8 to 10), and ‘quality and cost reduction’ (level 11 and above) suggest that initial work dispositions with regard to discipline and quality awareness are far from satisfactory (Chatterjee, 1990, p. 103). Work commitment and the taking of responsibility is another aspect that is reported by Japanese expatriates to have caused problems in the beginning (Venkataramani, 1990:131). Sinohara, a Japanese expatriate who serves as the Director Production and representative of SMC at MSI, bemoans lacking responsibility, discipline and perseverance among Indian employees: Shinohara had empathic views on both “responsibility” and “keeping promises”. “Responsibility means that all the consequences arising out of an activity are to be borne by the person concerned.” Nobody claimed responsibility when a job was not performed properly whereas alibis were promptly offered. Putting the finger on one all-too-familiar Indian practice, he said. (Venkataramani, 1990, p. 131) Yet again, the biggest implementation difficulties are related to those work concepts and work dispositions that clash with the existing socioprofessional demarcations. The communication and cooperation across socio-professional divides is a big issue at MSI and so are a number of egalitarian concepts that are meant to bridge those gaps. Most of the concepts that promote openly visible egalitarianism, such as a shared canteen, company uniforms, a shared toilets and open offices, meet stiff resistance from many managerial employees in the beginning (c.f. Som, 2004). My own interviews indicate, for instance, that the concept of common uniforms meets strong resistance (MSI 3). These finding find further confirmation in comments made by MUL/MSI’s former Managing Director Bhargava: Behind MUL’s success is another significant aspect of Japanese work culture – parity among the employees. In Indian conditions, we initially did face problems introducing the Japanese management ethos in MUL. There was, especially among managers, a certain amount of reluctance and hesitation about wearing a uniform and eating in the

120 Hybridization of MNE Subsidiaries

same canteen, sitting in the open offices, and all that. It required the top management to spend time with these managers and convince them that this was the only way to do things. It worked after some time. And as we went along, our managers realized that the concept not only produced results but also that it was not a bad thing to wear a uniform and eat in a common canteen with the workers. Similarly, when we introduced the open office concept, initially there were some partitions. But gradually, those partitions were also removed and the office became a totally open hall. People got used to it and realised that it did not adversely affect them. (Bhargava and Behl, 1997) Similarly, while all employees are asked to eat in the same canteen, there is, at least in the beginning, not ‘as much fraternization in MUL’s dining room among executives, supervisors, and workers as in Japan’ (Venkataramani, 1990, p. 225; c.f. Som, 2004). Indications are that middle to lower management shows the strongest signs of resistance. Kasahara (1994) underlines that white-collar workers, engineers and junior management levels have the biggest difficulties in accepting equality-based practices and also stressed their lacking willingness to share knowledge. The initial management resistance towards egalitarian practices is probably also compounded by the fact that management, compared to workers, is relatively poorly paid by Indian industry standards at the time. This leads, in turn, to turnover problems among these employees (Okada, 1998). Thus, middle management is the group whose identity is most strongly encroached by measures aiming at leveling hierarchy (Kasahara, 1994). Nevertheless, while the transfer of institutionally distant concepts and practices is clearly not received without resistance, the fact that the MSI is basically a Greenfield operation with a young workforce implies that local institutional conditions are probably as conducive as they could possibly be at that time (MSI 1, 2, 5).

Mode of recontextualization Some adaptation of the foreign/parent template to the local/host context From the very beginning it is the goal to transfer and imitate SMC’s home plant work concepts and corresponding human resource profile as far reaching as possible. While the host strategic context conditions are no major obstacle to such a transfer, the adverse local/host institutional conditions are both a major reason and obstacle for the transfer. Especially in the beginning, institutional misfit manifests in implementation difficulties (limited local resilience). For example, MSI’s work practices

Empirical Results 121

and their employees’ work dispositions are marked by much stronger social-professional demarcations than those envisioned by SMC’s work concepts. Due to India’s pronounced hierarchical distance some practices are not transferred at all. A case in point is the Ringi-System, which is based on consensual decision making (Chatterjee, 1990; Kasahara, 1994). Interviews suggested that decision-making practices in MSI and particularly in the factory remain very hierarchical and top-down rather than consensual or participative (MSI 2; c.f. Som, 2004). However, although there is some adaptation of the transfer template and not all work practices are implemented from the start (Shirali, 1984), no other company in the researched sample features a similar degree of effort to avoid a deflection of the foreign/parent’s template in the area of work organization and human resource profile.

Substantial adaptation of local/host context to the foreign/parent template MSI and SMC invest a tremendous amount of resources to bring work concepts and basic work dispositions in line of those of the foreign/parent template. The only way to transfer SMC’s work concepts, skills and basic work disposition with as little departures as possible is to select, change and create a different local/host institutional context for MSI. And in fact, in no other company researched in this study, is there a comparable systematic and comprehensive effort to adapt the local/host context to the requirements of the foreign/parent template of work organization. The effort to rests on the interplay of the following factors: Top management commitment, massive training, personnel transfer and socialization efforts, a young workforce, monetary and normative incentives and the cultivation of a culture of equality. From the start MSI’s Indian top management is not only committed but actively seeks the transfer (Venkataramani, 1990; Kasahara, 1994). Indian top managers are repeatedly depicted as religious followers of the Japanese work concepts and practices. Interviewees stress that the Indian top management sets examples and is committed to SMC’s work concepts and practices beyond lip service (MSI 1, 2). For example, top management is also not exempt from practices such as using a common toilet or working in an open office. The second aspect that is said to have enabled a far-reaching change of the local context is the comprehensive personnel transfer. SMC engages in massive personnel transfer from Japan to India (Okada, 1998). All major production departments are either headed or co-headed in a tandem-like fashion by Japanese expatriates. At the same time, there is a continuous training (on the job and in workshops) on skill and

122 Hybridization of MNE Subsidiaries Table 5.1

Overseas training programmes in MSI

Overseas training: Training held in co-ordination with SMC, Japan and AOTS (Association for Overseas Technological Scholarship) (covered 1600 employees under the various schemes) • 6 months SMC Training for Technicians – OJT in SMC, Japan (2 batches/yr of 50 each) • 9 months Javada Training for Press, Tool & Die Specialists – Design and Maintenance • AOTS Managerial Training (4–10 weeks for Manager & above – Managerial Best Practices) • AOTS Technical Training (3.5–6 months) for Supervisors and above – Technological Knowhow • R&D Training (2 yes) – Research on new Technologies Source: Adapted from MSI’s HR department (2003).

attitudinal development (Okada, 1998; Chatterjee, 1990). The transfer of Japanese expatriates to India as well as elaborate on-site and off-site training programmes in India, are only one part of the training effort. On top of that, MSI has a comprehensive overseas training programme which involves all employee categories (see Table 5.1). For example, as of March 2003, 1900 employees out of 4590, have been trained one way or another in Japan. In 2008, after repeated rounds of voluntary retirements, the ratio of the workforce that has been trained in Japan is about half of the total employee population (Maruti–Suzuki, 2008). A third aspect that is seen as crucial for the transfer success is the condition that most workers have high formal industrial qualifications, but no prior work experience (MSI 2). They come straight from industrial training institutes (ITIs) (c.f. Chatterjee, 1990, p. 41). Okada states that MSI prefers ‘fresh ITI graduates to workers who have work experience in other industries’ (Okada, 1998, p. 30). One of the interviewees points out that MSI is among the first companies to start the industry wide trend of hiring ITI graduates for worker positions (MSI 4). In any event, interviewees at MSI stress that the young workforce is a major factor that enable a far-reaching transfer of Japanese work concepts and work dispositions (MSI 1, 4). I think what also happened was that in the initial stages, when those concepts were brought in, the work force was very young. It was a very young workforce, you know, and that helps because, I suppose,

Empirical Results 123

it is more adaptable, more open. But I think as far as the production set-up is concerned, the factory set-up is concerned, I don’t think that there were too many issues. Because, I think, the young workforce thing really played a big role. So you don’t inherit this workforce from somewhere else. So they don’t have any preconceived notions about what should be right and what is not right, I don’t know, to that extent. Of course people who come in new, at middle management level, they have a bit of a problem. (MSI 2) In contrast to the workers, the managers recruited do have a prior work experience. However, an advantage is that even they have no prior experience of working together (Venkataramaini, 1990, p. 134). Two of the interviewees independently underlined that a great deal of transfer success – particularly with regard to different high involvement or continuous improvement concepts and practices – is related to MSI’s monetary and normative incentives structures (MSI 1). MSI has come to put great stress on productivity-linked payment schemes (Red Herring Prospectus, 2003). For example, the worker level incentivebased pay makes up a much higher proportion than the base salary. Moreover, quite deliberately no difference is made between blue- and white-collar workers to enhance cooperation between these traditionally distant groups (Bhargava and Behl, 1997; D’Costa, 2003). However, pay is not only linked to overall productivity and performance of the company. In addition to overall productivity-linked pay, compensation is also linked to attendance levels, group and individual suggestions as well as their implementation and related cost savings (Chatterjee, 1990, p. 124). Replying to the question how successful the transfer of quality activities and suggestion schemes really is, an interviewee explains: It is also some amount of incentive system and rewards, I think. The suggestion schemes, if you actually put your mind into it, you can make a fair amount of money on it, you know. That kind of thing, it all helps. (MSI 2) The payment of high salaries to workers and salaries directly linked to company, group and individual performance are seen to have a positive effect on the implementation of the Japanese high involvement practices. High salary levels also contribute a low level of industrial conflicts and a highly motivated labour force (Okada, 1998). Overall, the intricate system of incentives has direct effects on MSI’s work

124 Hybridization of MNE Subsidiaries

discipline, attendance-levels and on the successful implementation Kaizen activities, quality circles and suggestion schemes which are also confirmed as exemplary by a number or independent sources (c.f. MSI’s industry awards). In short, while failure to perform in line with Japanese work practices and concepts lead to quick financial punishment, complicance and above average performance not only pays off financially, but also offer ‘normative recognition’ in the company (MSI 1). The final aspect that MSI engages in is a cultivation of a culture of equality by introducing a number of highly visible practices that aim at reducing hierarchical demarcations. Among these are a common canteen, uniform, toilets and transportation for everybody, open offices (Afsar, 2003). The introduction of these visual manifestations of equality provoke as much initial resistance as they are said to have helped to bridging professional and social demarcation in the company (D’Costa, 2003). In summary, it is fair to say that while there is some adaptation of the foreign/parent template to the local/host institutional context, the dominant recontextualization mode in the set up of MSI work organization and human resource profile is an adaptation of the local/host context to the foreign/parent template. Although MSI is a Greenfield project with a young workforce (offering a much more resilient local context than a comparable Brownfield PSU), the local context’s resilience has limitations at the outset. These limitations are largely overcome by a long term approach and massive resource mobilization.

Outcome A substantial local context adaptation to the foreign/parent template suggests that imitation is a dominant outcome with regard MSI’s work concepts and human resource profiles. At the time of research, there are a number of observers from inside and outside the company who unanimously suggest that SMC has been relatively successful in transferring work concepts and human resource profiles in line with its home plant standards (MSI 1, 2; D’Costa, 2003; Okada, 1998). To what extent Japanese attitudes have been internalized cannot be answered. Interviewees stress, however, that MSI employees generally find it hard to work in a ‘normal Indian work environment’ after having worked for the company (MSI 3, 5). Undoubtedly, enormous efforts have been made to institutionalize Japanese work practices in MSI. Comprehensive training and socialization efforts bring MSI’s work concepts and the required human resource profile relatively close to those of the Japanese parent. Such a finding does not to deny that there are aspects of

Empirical Results 125

work organization where transfer and imitation proves difficult or unsustainable. These areas are predominantly those related to institutionally founded socio-professional demarcations between different employee categories. Many concepts and practices promoting egalitarianism are at least initially difficult to replicate. However, even if hierarchical and occupational demarcations remain more pronounced in MSI than in SMC’s home plant, we can expect more co-operation across socioprofessional divides in MSI than in typical Indian companies. This is not least the case because it literally pays off to change behaviour. In summary: Based on the some adaptation of the template to the local/host context and simultaneously a substantial adaptation of the local context to the foreign/parent template we see MSI’s work organization and human resource profile to vary between imitation and hybridization.

Industrial relations Transfer scenario From the beginning a key goal is to replicate a Japanese/SMC company union concept (MSI 4, 5). In contrast to all other companies researched, MSI is the only company with a clear intent to transfer a template of foreign/parent industrial relations. In fact, transferring Japanese labour relations is a crucial cornerstone of the whole transfer effort (Chatterjee, 1990). The goal is to transfer the Japanese company union concept with its emphasis on harmonious labour relations. To ensure harmonious relations, different discussion forums-concepts are established to solve problems at an early stage (MSI 1). Problems should not even have the chance to come up, but should be tackled as they arise. Although SMC is in the minority position the adoption of SMC’s company industrial relations is an expressed transfer goal. It is particularly the Indian top management that strongly supports the transfer of the industrial relations template (Chatterjee, 1990; Venkataramani, 1990; D’Costa, 2003). The institutional distance between Indian conditions and the idealized company industrial relations in Japan are the main driver of the local transfer request. Put differently, the transfer of SMC labour relations are deliberately sought as a means to break away from obstructive labour relations common in Indian enterprises at the time. For Indian and Japanese managers alike, it is clear that only if typical local labour relations are avoided, can a non-antagonistic work climate be achieved that is required for the implementation of SMC’s high involvement work concepts. Thus, a crucial part of the transfer effort is to

126 Hybridization of MNE Subsidiaries

establish a new mode of cooperative industrial relations that radically depart from what is dominant in Indian industry (D’Costa, 2003). The main point of departure is having only one company union, as opposed to many, and having the union not externally-affiliated as opposed to being established by external unions or affiliated with political parties.

Misfit Given that the foreign/parent template transfer is sought because of institutional distance, there is a potentially adverse local/host institutional context standing in the way of a smooth transfer. The establishment and maintenance of a single company union that is independent of outside unions and political parties clearly turns out to be a challenge for MSI. Repeatedly outside union activists affiliated to INTUC, Hind Mazdoor Sangh, Lok Dal and Lok Mazdoor Sangh seek to establish competing unions in MSI (Chatterjee, 1990; see also Gurgaon Worker News, 2007). At the same time, the internal union repeatedly threatens to align externally. Over the years, MSI’s foreign/parent template faces increasing recontextualization pressure, pulling MSI’s company industrial relations towards local/host context solutions. These pulls rest on an institutional complex of stringent labour laws, strongly protecting labour in formal sectors, long standing radical-unions movements, many of which an arm of the political parties and multiple unions competing and fighting each other in companies (see Chapter 4).

Mode of recontextualization Adaptation of foreign/parent template to the local/host context

From the inception until my research in 2003, MSI and its top management make an enormous effort not to adapt the company’s Japanese/SMC template to the local/host institutional demands and conditions. However, the foreign/parent template cannot withstand the local/host context institutional pressures. Apart from external unions seeking to get a foothold in the company, MSI’s company union itself, the Maruti Udyog Employees Union (MUEU), repeatedly threatens to align with external politicians and departs from MSI’s consensual approach. In 1998, amid the Government of India’s new openness for SMC’s takeover bid, MSI’s company union threatens to call off its non-affiliation agreement (Indian Express, 1998). As the company faces a tougher competitive environment in the 2000s and announced a new incentive scheme a strike breaks out. The labour conflict displays many features of a typical Indian labour conflict

Empirical Results 127

including: tool-downs, indefinite strikes, lockouts, political affiliation and a dharna (a hunger strike). What is more, during the labour conflict a second union, the Maruti Udyog Kamgar Union (MUKU) is formed, which stalls the company’s one union policy. Although MSI is able to revert formally to its one-company-union policy in the 2000s, the union behaviour and conflict pattern during the strike are highly reflective of local/host institutional patterns, departing radically from those in SMC’s home plants.

Adaptation of local/host context to foreign/parent template

To transfer the institutionally distant industrial relations concept, MSI engages in an adaptation of its local institutional context. To establish the foreign/parent template, SMC and MSI management take a number of measures and invest resources to create an enabling local institutional context. Among the most important measures is the commitment of the Indian Government and the ruling party at the time of establishment to discourage any party involvement in MSI’s labour relations (MSI 4; Indian Interviewee Friedrich Ebert Stiftung, 2003). The selection of a young work-force, personnel transfers to Japan and excellent pay for permanent workers, well above Indian industry standards, are all investments that contribute MSI’s overall peaceful labour relations (MSI 4). Q: So do they have a strong history of industrial conflict or do they rather? A: No, they don’t. I mean Maruti is one of the best paying companies today in this region. And I guess that has done a lot to keep them away from . . . I mean they had some industrial conflict but I don’t think it has been anywhere near as bad as in many other organizations. (MSI 4) Yet, following the first strike in the 2000s, MSI management reverts to more draconic measures to assure industrial peace and to re-install the company’s one-union and a non-affiliation policy. In reaction to the strike, MSI’s management not only pushes through a new incentive scheme which is the major bone of contention, but also asks its workmen to sign ‘a good conduct undertaking’. MSI also dismisses a number workers and trainees due to their strike-related activities. The signing of the ‘good conduct undertaking’ is made a precondition for workers to reenter the plant. Among other things, the workers have to sign that they will not go on strike in future. In 2002, MSI is finally able to revert to a one union policy. Following financial irregularities, MSI de-recognized the

128 Hybridization of MNE Subsidiaries

MUEU – MSI’s first union – and recognized the MUKU instead. Also, the massive re-composition of the workforce to a small core of well paid permanent workers and a huge pool of contract labour and apprentices that lack union representation play a role in controlling industrial action in the 2000s (Gurgaon Worker News, 2007). While MSI’s industrial relations are very peaceful by Indian standards and have probably set an example for a new mode of industrial relations in modern Indian enterprises, the Japanese template of harmonious industrial relations with one company union gets corrupted over time. While it remains to be seen how stable the company’s recent return to a single union policy proves in future, it is fair to say that India’s local/host context is not overly resilient with respect to industrial relations. As the institutional pattern of industrial relations is deeply entrenched in the Indian society, politics and legislation, a departure from them requires a substantial and continuous mobilization of resources. We can conclude that the recontextualization mode includes both: A substantial adaptation of the local/host context to the foreign/template (particularly in the beginning) and over time an increasing adaptation of the foreign/parent template to the local/host institutional context. Clearly, the adaptation of the local context to the foreign/parent template strongly relied on a massive mobilization of resources and commitment given the not very resilient local/host industrial relations context.

Outcome MSI receives a foreign/parent template, which marks a deliberate departure from local/host institutional conditions. However, because of such an institutional distance, it faces a misfit with the local/host institutional context. Being aware of that, efforts are made to adapt the local institutional context by taking a number of measures to ensure the survival of the contrasting concept. As a result, MSI’s company industrial relations depart particularly in the first years from traditional Indian industrial relations. This concerns the level of industrial peace, the provision of forums and processes to address grievances as well as the level of cooperation between labour and management. However, what begins as a showcase of industrial peace modeled on foreign/parent industrial relation practices, increasingly moves towards local/host industrial relations patterns (cf. Sachitanand, 2000). The involvement of multiple, politically and externally affiliated unions as well as a tradition of radical industrial action pose a constant threat. In MSI this local/host mode of industrial relations surfaces as the company’s

Empirical Results 129

profitability comes under pressure and a less favourable incentive scheme is introduced for workers. Thus, MSI’s industrial relations start off as an imitated solution. Over time, this solution turns into a hybrid (reflecting Japanese origins as well as host context elements), if not increasingly into a localization, in the face of local/host context institutional pressure (MSI 1). At the time of research, the company industrial relations could be best described as somewhere between hybridization and localization. It remains an open question if the establishment of a new company union in the 2000s and the employee re-composition towards a small core of permanent employees heralds a shift back towards an imitation foreign/parent industrial relations.

Supplier relations Transfer scenario From the start the set-up of MSI’s supplier relations is strongly guided by SMC who has the configuration mandate despite being the junior JV partner. SMC takes a step-by-step approach to replicate the foreign/parent template. The transfer effort comprises the contractual, the structural and the process-related dimension of the supplier relations. In contractual terms, the goal is to establish close ties between MSI and its suppliers. In structural terms, SMC seeks to increasingly implement a tierisation of MSI’s supplier relations. And even in supply flow terms, SMC pushed for just-in-time (JIT) supply logistics. From the very beginning, SMC management insists that MSI follows a comprehensive outsourcing policy excepting only the most critical components. As part of this effort, in house value addition is to be no more that 26 per cent (UNIDO, 2003). Based on this foreign/parent demand, a substantial proportion of the value addition, has to come from outside MSI. In theory, MSI could import the outsourced parts and components from Japan but there is a host institutional and strategic context that rules out such an option. First, high import tariffs set by the host institutional context make a largely imported vehicle too expensive for price-sensitive entry-level customers. Secondly, MSI has to observe local content requirements that form part of the JV agreement and the Indian Government FDI policy. Specifically, it is agreed that out of the 70 per cent of the non-company value addition, at least 60 per cent have to be locally procured (Mohanty et al., 1994). Thirdly, even without high local content requirements and high import tariffs the combination of a high Yen and a low market segment focus, rule out the import of

130 Hybridization of MNE Subsidiaries

costly components in the long-run. In short, SMC’s demands, the Indian Government demands, a high Yen and a prices sensitive market demand suggest a central role to be played by suppliers located in India (MSI 1; Okada, 1998). However, this supplier infrastructure hardly exists and where it exists host context conditions are institutionally too distant for MSI’s requirements in terms of contractual relations, structure and the process and consequently too strategically distant in terms of technology, quality, cost and volume capability. In the early 1980s, the Indian automobile supplier Industry is small, weak and fragmented. It is dominated by small and technologically weak players (Khare, 1997; Bhargava, 2002). Tierization and JIT supply logistics are unheard of and contractual relations are arms-length and low trust in nature (D’Costa, 2003). At this point in time, Indian suppliers are not capable to supply MSI with parts and components in the right quality and quantity (Mohanty et al., 1994; Joseph, 1990). Given that SMC introduces the product and given the absence of local/host capabilities SMC assumes the mandate to establish MSI’s supplier relations. While the host context market demand conditions allow a far reaching replication effort, it is the absent supply conditions make such a massive transfer effort a requirement.

Misfit The transfer intent of foreign/parent supplier relations implies that what is going to be transferred faces either an absent or a partly adverse institutional and strategic context. The strategic context reflects institutional conditions in the supplier industry that are still shaped by India’s state-led growth regime in the 1980s. For example, as the Indian supplier market is not concentrated and as the financial and technological capabilities are low, the cost reduction by means of a tiered supplier structure is impossible in the beginning (Humphrey and Salerno, 1999). The lacking reliability of subcontractors also puts a break on tierisation efforts. Moreover, the absence of VAT means that a tierisation of suppliers increases the cost for supplies (ACMA Interview, 2002; SIAM Interview, 2002). Another reason why tierization remains slow is related to the low production volumes in the Indian automobile industry (Okada, 1998). A somewhat similar situation can be found with regard to JIT supplies. Here, the high geographical spread of MSI suppliers, poor road conditions, unreliable suppliers, frequent strikes and seasonal rainfalls pose difficulties to its fast implementation. Finally, getting the Indian suppliers to invest in and cooperate with MSI is no easy task to begin with (Bargava, 2002). A heritage of arms

Empirical Results 131

length and low trust patterns in the Indian supplier industry, make it difficult to win over potential suppliers for long term cooperation (Bhargava, 2002; D’Costa, 2003). Suppliers also fear that their investments for a lofty Indian Government project may not pay off. One of the first challenges is, therefore, to overcome mistrust of suppliers and convince them that state-owned MSI is determined to realize the targeted volumes.

Mode of recontextualization Adaptation of foreign/parent template to the local/host context

As there is no existing supplier infrastructure, SMC has the crucial role to orchestrate its set up. While this set up is largely based on the foreign/parent template of supplier relations, not all aspects of the template can be implemented over night. There are strong indications that substantial adaptations of the foreign/parent template to the host (institutional and geographical) context are required because the host context conditions are not very resilient (e.g. wide geographic spread of suppliers, poor roads, taxation policies). This particularly involves the de-selection of tierisation and JIT in the beginning. Clearly, a part of the recontextualization effort rests on the adaptation of the foreign/parent template to the local/host strategic and institutional context.

Increasing adaptation of local/host context to the foreign/parent template over time The dominant recontextualization mode with regard to the supplier relations is, however, an adaptation of the local/host institutional/strategic context to the foreign/parent template. In order to tackle MSI’s supply problem, SMC and MSI embark on a massive ‘supplier development programme’ (UNIDO, 2003). The programme essentially centres on an immense institutionalization effort. Specifically, the selection, change and creation of the local/host supplier context in line with SMC’s set up of home supplier relations. In the initial phase, the institutionalization and capability building effort focuses on realizing SMC’s outsourcing policy through the development of suitable suppliers in India (UNIDO, 2003). To achieve the goal, a multi-pronged approach is taken. In a first step, MSI tries to build up trustful relations with existing suppliers which are selected as suitable partners. It does so by giving all kinds of support, ranging from financial to technical assistance to the Indian suppliers (Khare, 1997; Bhargava, 2002; UNIDO, 2003). Even some personnel is sent from MSI to suppliers for longer periods of time. Additionally, the mistrust of suppliers is overcome by doing away with the annual tender system and replacing

132 Hybridization of MNE Subsidiaries

it with long term contracts. The new system gives assurances to the suppliers with regard to production volumes, prices and dates of payment (Bhargava, 2002; UNIDO, 1999). However, simply supporting existing local suppliers proves insufficient. In a second step, MSI gets involved in suppliers and founds JVs with local suppliers (Okada, 1998). MSI not only helps in bringing together SMC and Indian suppliers for JVs, but also enters into 15 JVs with suppliers (Maruti–Suzuki, 2008). Although MSI holds only minority equity positions in these companies, it is strongly involved in their quality and productivity assurance. Yet, the transfer effort does not stop at the contractual level. In the late 1990s, MSI starts to reduce its suppliers. This development gains new momentum when SMC takes over the company in the 2000s. While MSI has about 406 suppliers in 1998, it has almost halved the number to 220 in 2004 (Maruti–Suzuki, 2004). Also with regard to JIT, MSI is able to change the situation through a host of measures (Shelley, 2002). From the beginning, there are a number of key suppliers that build their production facilities directly on the MSI compound. Next to this, MSI’s JVs, producing the most important components, are located near its plant (Halasyam, 2001, former Director Finance in MSI). What is more, the majority of MSI’s parts and components are provided by suppliers that are within a radius of 100 km (Red Herring Prospectus, 2003). Getting more and more supplier to locate near its plant also reflects MSI’s effort to lure in far away suppliers by means of incentives. Such incentives comprise: offering subsidised, well located and industrially developed land, sales tax concessions and reliable power, which supply MSI generates itself (Gulyani, 2001; D’Costa, 2003). In addition, MSI asks suppliers unwilling to relocate, to build or at least make use of warehouses near its plant. In these cases, MSI is willing to bear the expenses incurred to maintain the warehouse (Venkatachari, 2000). To better co-ordinate and facilitate JIT, MSI also develops a low budget IT-solutions for its suppliers helping to schedule JIT deliveries (Kulkarni, 2002). While these efforts are far reaching efforts to bring MSI ever closer to the foreign/parent template in terms of the supply flow, there are still factors, such as the transportation systems, that work against a full the realisation of the JIT concept (c.f. Gulyani, 2001). To conclude: MSI has taken a long term approach to implement major features of the foreign/parent template in terms of supplier relations and there are still some adaptations to the local/host context. Nevertheless, the dominant recontextualization mode has become an adaptation of the local/host context to foreign/parent template. While the certain conditions in the host supplier context remained non-resilient, SMC’s and

Empirical Results 133

the local JV partner’s commitment of substantial resources created an increasing host context resilience for an adaptation to the foreign/parent template.

Outcome SMC embarks on a huge transfer effort with regard to its supplier relations. This is possible because of a strategic fit with regard to the demand market conditions. At the same time, it is necessary because of the combined effect of foreign/parent demands and host supply market conditions. Put differently, SMC has to undertake a huge transfer effort because it wants to observe its outsourcing policy, but faces an institutional/strategic environment that does not allow the fulfilment of such a policy without major transfers of home practices. In the face of a weakly and adversely institutionalized market for parts and components, SMC develops suppliers by founding JVs and assisting suppliers. In order to transfer its supplier relations SMC embarks on a massive institutionalization effort that involves a selection, change and creation of local/host supplier relations. The overall result is that MSI’s supplier relations move closer to a pattern of imitation over time. For example, in contractual terms MSI develops relations with suppliers that come very close to those of SMC, departing radically form those typical in the host context (UNIDO, 2003). Like in MSI’s supplier development programme, SMC in Japan had been from the 1950s onward very actively involved in the establishment of its supplier network through special developmental activities and direct support (e.g. SMC Supplier Cooperative Union). Like in MSI, SMC in Japan provided financial support and training to its suppliers. Similar, to the MSI set-up, SMC supported its suppliers in Japan in the early 1960s to move to a nearby supplier park and also directly influenced the management of these enterprises (c.f. Price, 1997). And, like in the MSI set-up, SMC had the policy at home to have at least two suppliers for all major parts. Additional support for a replication strategy of supplier relations comes from an UNIDO (2003) report. The report suggests that MSI’s supplier relations are strongly modelled on Japanese examples. The overall result is that, in contrast to old Indian practices, substantial outsourcing, high trust and cooperative supplier relations develop. While MSI’s initial transfer focus is more on establishing suppliers and having close relations with them to assure the quality and quantity of supplies, the focus widens over time. More recently, efforts are made to replicate the structural and process-related features of the Japanese supplier set-up. At the time of research, the tierisation and supply logistics

Table 5.2

Summary of MSI’s hybridization profile MSI

Functional differentiation

Transfer scenario: Transfer of foreign/parent template Misfit: Minor strategic misfit between template and local/host context Mode of Recontextualization: Low template adaptation to local/host context; high adaptation of local/host context to foreign/parent template Outcome: Mainly imitation

Hierarchical differentiation

Transfer: Use of host context template Fit/Misfit: Little (known) institutional misfit between host context template and foreign/parent demands Mode of Recontextualization: Likely adaptation of foreign/parent demands to host context template; creation of site in line with host template with minor host template adaptation to foreign/parent context Outcome: Between localization (and hybrid)

Technical configuration

Transfer: Transfer of foreign template with regard to factory layout and process design; only selective and increasingly less transfer with regard to technical hardware Fit/Misfit: Hardly any strategic misfit between selected aspects of foreign/parent template and local/host context; some misfit between foreign/parent strategic and local/host institutional demands Mode of Recontextualization: Little adaptation of those foreign/parent template aspects transferred; mainly creation in line with foreign/parent template or demands and host context Outcome: Between imitation and customization

Work organization/ HR profile

Transfer: Transfer of foreign/parent template Fit/Misfit: Institutional misfit between foreign/parent template and local/host context Mode of Recontextualization: Some adaptation of the template (de-selection) to local/host context; mainly adaptation of local context to foreign/parent template Outcome: Between imitation and hybridization

Industrial relations

Transfer: Transfer of foreign/parent template Fit/Misfit: Misfit between foreign/parent template and local/host institutional context; Mode of Recontextualization: Adaptation of local context to foreign/parent template as well as adaptation of the foreign/parent template to local/host context Outcome: Between hybridization and localization

Supplier relations

Transfer: Transfer of foreign/parent template Fit/Misfit: Foreign/parent template misfits local/host institutional/strategic context Mode of Recontextualization: Initially adaptation of foreign/parent template to host context; over time major adaptation of local/host context to foreign/parent template Outcome: Between hybridization/customization and increasingly imitation

134

Firm > PS dimension

Empirical Results 135

of MSI are not exactly those of SMC. The number of direct suppliers is still more compared to the Japanese parent and the supplier hierarchy is less tiered than in SMC. Similarly, the level of JIT is still behind that of the Japanese parent (Okada, 1998). However, MSI is narrowing the gap. Although MSI is in 2003 still away from full JIT-deliveries, it has come closer to the ideal over time. An indication is MSI’s falling inventory times. While in 1992 MSI’s total inventories are 57 days of stock, they have come down to 19 days of stock in 2003-2004 (c.f. Gulyani, 2001; Maruti Udyog, 2004). In summary: MSI has come closer and closer to an imitation of its home supplier relations. While in the beginning, MSI’s supplier relations reflect more a hybrid solution-based on a combination of imitation and localization (possibly also customized solutions reflecting foreign/parent demands and local/host context conditions) – they develop towards imitation (see Table 5.2 for a summary of MSI’s hybridization profile).

6 The Case of Fiat India

Introduction Fiat’s internationalization starts already with car export before the First World War and is triggered by the company’s small domestic market (Volpato, 2000/2001). After the Second World War Fiat’s internationalization strategy increasingly focuses on setting up production sites for small cars in emerging markets. Volpato (2000/2001) calls Fiat’s path also a ‘forced specialization’. He sees it as the result of tax regimes (i.e. heavy tax progression with engine size) and gasoline prices in Italy, dating from the fascist period and continued after the Second World War. Volpato argues that the business conditions in Italy ‘progressively reduced the possibilities for Fiat to compete in foreign markets on higher segments of automobile production’ (Volpato 2000/2001, p. 2). Fiat’s ‘forced specialization’ explains why Fiat shifts its internationalization focus comparatively early to emerging markets, which eventually culminates in an emerging market World Car strategy in the 1990s and beyond. Apart from some licence-production agreements with Premier Automobiles Limited/India in 1951, with Zavodi Crvena Zastava/Serbia in 1954 and the establishment of engines and transmissions plants in Brazil in 1958, Fiat’s internationalization does not take off before the 1970s. In 1971, Fiat starts the Tofas-Bursa plant in Turkey (JV with the Koc Group). In 1973, the Betim plant is founded in Brazil and a year later the Tychy plant in Poland. While Fiat’s internationalization sees some disinvestment and slow down in the 1980s, caused by the oil crisis and domestic conflicts (Balcet and Enrietti, 2002a,b,c), it receives a new boost in the 1990s. This boost is driven by Fiat’s World Car project. For the production of its World Car range, Fiat 136

Empirical Results 137

transforms its existing plants (Brazil/Betim 1996, Poland/Bielsko-Biala, Turkey/Bursa 1998), enters into new JVs (Morocco/Casblanca 1997, Egypt/Cairo 2002, India/Kurla 1997, China/Nanjing 1999) or licensing agreements (South Africa/Johannesburg 1998 and Vietnam/HoChiMinh City 2000) and establishes new operations (Argentina/Cordoba 1996, Venezuela/LaGuaira 1997). However, facing a crisis in the 2000s and only a modest success of its World Car range in major Asian economies, including India and China, Fiat’s World Car driven internationalization spree comes to a halt and a period of consolidation follows. Yet, despite disappointing sales, the strategic importance of the Indian operation is held on to in the 2000s (Fiat Group, 2005a). Fiat’s Indian operation is still considered an important strategic pole for Fiat’s World Car production in Asia. This is not least related to the fact that Fiat enters into a strategic partnership with Tata Motors in 2006. Apart from the joint distribution and production of Fiat models, the partnership also envisions a joint production in Fiat’s Latin American operations (Fiat India, 2006, 2007). In contrast to this ‘focussed globalization’ oriented towards emerging economies (Balcet and Enrietti 2002b), Fiat’s production internationalization in the developed triad markets – even in Europe – is comparatively weak. This also has do with the failed involvement in Citroen in the early 1970s, the failed merger with Ford Europe in mid 1980s and the failed expansion into the USSR in the early 1990s (Volpato, 2000/2001). Fiat’s European production internationalization is, therefore, largely confined to the company’s cooperation with GM. With the Fiat Group’s crises in the 2000s, Fiat’s internationalization strategy receives a shift in orientation. Apart from announcements to close down some of its international sites, the internationalization strategy shifts more towards forging new international alliances with limited or ‘no equity’ commitment (Fiat Group, 2005b). Interestingly, a good amount of these alliances are forged with strong players in emerging markets (e.g. Tata Motors, Maruti-Suzuki).

Global product strategy Based on Porter’s (1980) classification of generic strategies, Fiat’s global product strategy can be best described as a combination of cost leadership and differentiation. In a similar vein, Boyer and Freyssenet label the company’s profit strategy a ‘diversity and volume’ strategy (Freyssenet, 1998; Boyer and Freyssenet, 2003). Already in the 1960s Fiat starts to broaden its brand portfolio. The major road post of becoming a multibrand company are the takeover of Autobianchi in 1958, Lancia in 1969, Ferrari in 1966, Alfa Romeo in 1986 and Maserati in 1993.

138 Hybridization of MNE Subsidiaries

Fiat’s global product strategy has been regionally differentiated for some time. This involves offering state-of-the-art products in developed markets (production of which mainly located in Italy) and ‘mature’ or outdated products in second and third world economies (production of which mainly located in emerging markets). In the 1990s, this pattern changes and continues at the same time. With the World Car project Fiat introduces a range of new models that are based on one platform and specifically developed for emerging markets. This implies a continuation of a strong regional differentiation, however, with models that are newly designed for emerging markets. While Fiat’s overall global product strategy can be labelled a combination of cost leadership and differentiation, its product strategy vis-à-vis emerging markets can be more appropriately described as cost focus. Despite developments in the 2000s towards less regional differentiation, Fiat’s strategic orientation of cost focus is still very present in Fiat India (FI) in 2003.

Establishment mode and equity development In 1951, Fiat enters into a license agreement with the India automobile company PAL. The agreement involves the license production of the Fiat 500 and later the Fiat 1100. In 1972 this agreement expires. However, the model continues to be built in India under the name Premier Padmini (Mohanty et al., 1994). In 1981, a second technology agreement is signed with PAL (Mohanty et al., 1994). In the 1990s with India’s market liberalization, Fiat’s involvement in India sees a revival. Like other international auto manufacturers, Fiat identifies a huge potential in the Indian car market. What is more, in the 1990s Fiat focuses on developing countries as the major future markets for its products. In line with this strategic focus, Fiat announces in 1995 its World Car project which is specifically designed for emerging markets and envisions India as a production pole (Alzona and Prakash, 2001). Against the background of its longstanding connection with PAL and its World Car project, Fiat opts for a two-pronged entry into the Indian market (Goldstein, 2002). In 1996 Fiat enters into a technical collaboration with PAL for the import and assembly of the Fiat Uno from CKD-kits. The Uno assembly is conferred to PAL and established at PAL’s Brownfield site in Kurla in the outskirts of Mumbai (Subhadra, 2002). Simultaneously, Fiat plans to set-up a wholly owned Greenfield plant for the production its World Car model range. To realize the project, Fiat founds a new subsidiary, Fiat India Auto Ltd. (FIAL) and buys a factory site in Ranjangaon near Pune. In 1997 Fiat even signs a MoU for the Greenfield project with the State Government of Maharashtra for the annual production of 100000 cars. The project’s

Empirical Results 139

plan is to start production by the end of 1999. However, things don’t work out as planned. PAL which takes care of the Fiat-Uno production proves to be a problematic partner. Fiat becomes very unhappy with the way PAL management handles the Fiat-Uno project and PAL’s unwillingness to step up investment in the JV (Goldstein, 2002; Subhadra, 2002; FI 2; FI 3). In addition, a strike brakes out followed by a lockout, which leads to delivery delays and vast cancellations of car bookings (Subhadra, 2002). Overall, the PAL-led Fiat-Uno project contributes to a huge image loss for Fiat in India. In reaction to these problems, Fiat takes over the Uno project in 1997. To realize the take-over, Fiat acquires a majority in the newly founded JV India Auto Ltd (IAL) with PAL and raises its stake continuously (Subhadra, 2002; FI 4). Parallel to these developments, the Indian automobile market moves into recession and sees increasing competition at the end of the 1990s. As a consequence, initial sales projections of 100,000 vehicles per year are scaled down to a mere 25,000 vehicles per year. The changed demand scenario leads Fiat to reconsider its Greenfield project. The problems with PAL and doubts whether the Greenfield investment pays off, lead Fiat to put its Greenfield project on hold and turn to the jointly owned Brownfield facility as an alternative for its World Car production. Instead of building a new production site, Fiat increases its stake in the JV IAL and concentrates on refurbishing the Kurla Brownfield facility. At the end of 1998, Fiat increases its stakes in IAL and announces the freeze of Rangjangaon Greenfield project. In 2000, Fiat increases its share in IAL again from 76 per cent to 95 per cent and in 2004 to 99 per cent (Alzona and Prakash, 2001). At the time of research, IAL is renamed to Fiat India Private Ltd. (FI) and has practically become a wholly owned subsidiary of Fiat Auto Italy. In 2006, as part of a new strategic partnership with Tata Motors, Fiat establishes a new 50:50 per cent production JV. The new JV is named Fiat India Automobiles Private Limited (FIA).

Location, production programme and market share In 2003, Fiat’s automobile factory in India is located in Kurla in the agglomerations of Mumbai (State of Maharashtra). The factory site and the factory halls are inherited from former JV partner PAL. Fiat invests substantially to upgrade the Brownfield facilities which are in a poor condition (Atzeni, 2001; Subhadra, 2002). At the time of research, Fiat manufactures three models: the Uno, the Sienna/Sienna Weekend and the Palio in about twelve variants. In terms of market-segments, the models range from the lower B to the lower C segment. Having difficulties to sell the C-segment Sienna, the company

140 Hybridization of MNE Subsidiaries

plans to focus more on the B segment to realize higher sales and production volumes. While all models share the same platform, it is specifically the Sienna, Sienna Weekend, Palio that belong to the World Car range (FI 1). In 2003 FI has an installed capacity of about 60,000 vehicles per annum and runs its operation with around 2000 employees. Daily output was at 150–200 vehicles manufactured in two shifts. Until 2002, the site produced the Uno and the World Cars in two separate lines. From 2003 onwards, they are all integrated into one flexible manufacturing line. FI’s on site production steps comprise: a body shop, a paint shop, an assembly line and a finishing line. In addition, there is an on-site engine assembly facility. The bottleneck of the FI’s manufacturing set-up is its paint shop. As the paint-shop is only capable of painting one car at a time, which lasts about 4 minutes, the whole process organization is adapted to the paint shop’s cycle time (FI 2). In 2006, Fiat shifts its entire passenger car production from Kurla to its new production JV with Tata in Ranjangaon near Pune (State of Maharashtra). The new plant – which is not established at the time of research – is the long awaited state-of-the-art Greenfield plant. It will have an installed capacity to produce 100,000 cars and 200,000 engines and plans to operate with about 4000 employees (Fiat Group, 2007). FI’s market share in the 2000s reflects the company’s modest success with the introduction of its World Car models to the Indian market. While especially the Palio (launched in 2001) gives Fiat’s sales in India a boost, the new introductions do not live up to the expectations. In terms of sales volume, FI sells in the B segment 17,067 cars in 2001– 2002 (out of a total of 293,131 sold in the B segment) and 16,366 cars in 2002–2003 (out of a total of 158,291 sold in the B segment). These are market shares of 5.8 per cent and 10.3 per cent in the B segment in the respective years. In the C segment the tendency is similar. In 2001–2002 Fiat sells 898 vehicles (out of a total of 72,610 sold in the C segment) and in 2002–2003, 1850 vehicles (out of a total 37,957 sold in the C segment). These are market shares of 1.7 per cent and 4.9 per cent in the two years (data provided by Red Herring Prospectus, 2003). Despite efforts to boost its market share with new model introductions, Fiat’s overall market share in the passenger car segment in India remains weak in the 2000s.

Analysis of the hybridization profile The general transfer scenario The transfer intent and effort of Fiat in India is closely linked to Fiat’s World Car project code-named ‘178’. Apart from a standard range of cars

Empirical Results 141

based on a single platform, the project comprises of a production-system template. Camuffo and Volpato (2000/2001) describe the main goals of the World Car project also code-named ‘178’ as follows: 1. Define a family of new models, based on the same platform, to be produced and sold with basically no significant change, in a number of countries. 2. Create a worldwide supply chain to manufacture, in different places of the world, a family of new models suited to the motorization needs of BEMs [big emerging market]. 3. Take advantage of cost differentials, namely labour cost differentials, available in PLEMAs [peripheral of large existing market areas]. 4. Guarantee absolute standardization of each version of the models produced, even if they are targeted to different markets. 5. Establish an organizational learning process capable of conciliating the design of a centralized product with the needs of various final markets. 6. Diffuse, by replication and adaptation to foreign plants, the lesson learned in the setting up and operation of the highly successful green-field plant in Melfi (Italy). 7. Develop a global supply chain, flexibly and efficiently using the production capacity and the supplier base available in different countries. (Camuffo and Volpato, 2000/2001, p. 4) The World Car’s production template is based on the ‘Fabbrica Integrata’1 concept, which Fiat has implemented at its Italian Greenfield site in Melfi (Camuffo and Volpato, 2000/2001). According to Volpato, the Melfi plant serves as a ‘paradigm’ to be applied ‘to the various new plants made under “Project 178” in a flexible and adaptive way’ (2000/2001, p. 9). While the ‘Fabbrica Integrata’ experience is the paradigmatic starting point for the World Car production-systems, changes are made to adapt it to developing country conditions. These comprise mainly lower wages, lower volumes and local content requirements. In contrast to Melfi, the standard production volumes for the World Car production template are lower and as is the targeted level of automation. In addition, a human resources focus is defined as a corner stone of the template. Concerning supplier relations, also some key concepts are drawn from Melfi. These include a nearby supplier park and substantial levels of outsourcing (c.f. Camuffo, 2000/2001). To develop the World Car and its production template, Fiat even establishes a pilot plant in Turin and transfers technicians, mainly from South American, for a better understanding of developing country requirements. The pilot plant in Turin becomes a permanent testing

142 Hybridization of MNE Subsidiaries

and developing ground for the product and the production template of the World Car project (FI 2). According to Volpato, the pilot plant also serves as a ‘conservatory of weights and lengths’ which ‘dictates the product and manufacturing specifications to which all Fiat and supplier plants must conform to’ (2000/2001, p. 14). In 1996, Fiat starts with the implementation of its World Car project in Brazil. In the same year, Fiat builds a whole new Greenfield site in Cordoba, Argentina, where the World Car production template is most faithfully realized (Volpato, 2000/2001). It is also this plant that initially serves as a model for the planned Greenfield site in India (Autojunction, 2003). Fiat’s World Car production template is very comprehensive (Camuffo, 2000/2001, Cardoso, 2004). The template defines major organizational structural parameters such as hierarchical levels, functional differentiation and a standard concept of work organization. It includes a standard production layout for 400 vehicles per day (which can be downgraded to 200) and specifies the technical configuration (emphasizing low levels of automation) of the production process. It even defines a range of concepts and policies with regard to supplier relations and employment relations. Apart from the World Car related template and policies, the Fiat-group has global standards for industrial relations and supplier relations. In this respect, however, subsidiaries have a substantial freedom to respond to specific host context conditions (FI 4). While Fiat seeks to transfer World Car its template to its Indian operation, the decision to start the production at PAL’s Brownfield site implies from the start that certain aspects of the template cannot be implemented. This mainly concerns the transfer of more advanced manufacturing technology and the nearby supplier park (FI 1, 2). However, despite these adaptations, the interviewed managers insist that core aspects of the World Car template have been transferred to India.

The functional differentiation of the organization structure Transfer scenario Fiat transfers to its Indian operation the functional configuration of the World Car production template. This comprises two crucial concepts, the ‘Management by Process’ and the ‘Unità Tecnologica Elementare’ (UTE), that are based on integrating different functions in organizational units for the production of a specific product family. ‘Management by process’ involves the entire production process. Here, different functional departments are allocated to a certain product family (e.g. Palio or Uno).

Empirical Results 143

The product family is headed by a ‘process manager’ who oversees the whole process. As part of this concept, the process manager is also responsible for human resource management including issues such as recruitment and promotions (Cardoso, 2004). The same idea of functional integration can be found in the organization of different process steps (e.g. final assembly) in operational units called UTEs. One UTE generally integrates the functions production, maintenance, product and process quality assurance, material handling (FI 2; cf. Cardoso, 2004). The Italian interviewee pictured them as a ‘little factory’ comprising about 50 people, each (FI 2). In a common approach to all its world car sites (Cardoso, 1998; Cardoso, 2004), Fiat transfers the template’s concepts of functional differentiation or rather integration also to its Indian site Kurla (FI 1, 2). Based on its entry mode, Fiat could have drawn on the existing functional template of the Brownfield site. However, given Fiat’s control over the site, its product ownership, its strategic intent to implement the World Car template and the capability gap between Fiat and its local partner PAL, a continued use of the local template is out of the question. The fact that Fiat has practically from the beginning a majority in the JV, plus the fact that Fiat introduces the vehicle and has essentially the mandate to establish the production of its World Car, allows the foreign/parent to change the local template. At the same time, there are no reports that host context institutional or strategic demands and conditions triggered any changes in transfer intent of the World Car template’s functional differentiation.

Misfit While there is no strategic or institutional distance that deters the template transfer up front, it is also clear that the World Car template faces an existing local template of functional differentiation. However, there is no indication that the implementation of the new functional differentiation, which requires a change of the existing template, causes adverse reaction among the local employees (FI 2, 3, 4). While some human resource and shopfloor related practices of the UTE concept prove hard to be implemented, due to the sites institutional heritage, the formal structural aspects and functional differentiation of the UTE systems meet little institutional misfit. What is more, the condition that the host strategic context features a lower market demand than the ideal World Car template requires in volume terms, does not cause a strategic misfit/recontextualization pressure for the ‘management by process’ model or UTE concept (FI 2). It appears that the inbuilt possibility to

144 Hybridization of MNE Subsidiaries

scale down the template (Camuffo, 2000/2001), gives the template some robustness vis-à-vis varying volumes.

Mode of recontextualization Minor adaptation of foreign/parent templates to the local/host context There are no indications that Fiat’s World Car template undergoes major adaptations with regard to its functional differentiation. Neither is there any report that the existing local template obstructs the implementation of the World Car template’s functional differentiation, nor that the local/host context strategic and institutional conditions and demands induce an adaptation of the foreign/parent template (FI 1). Although there are no reports in this regard, some adaptations of the template to local/host context demands and conditions are likely.

Major adaptation of local/host template to the foreign/parent template The template transfer and the absence of an adverse local/host strategic and institutional context with regard to the World Car template’s functional differentiation, allow an adaptation of the existing template of PAL in line with the foreign/parent template. This replication success is mainly achieved through the transfer of Italian expatriates to the Indian site and the transfer of Indian employees to the Italian pilot plant in Turin. The Italian side is willing and able to restructure the functional set-up along in own template because the steady increase in equity of the JV and the product ownership provide the mandate to restructure the site. On the one hand, the adaptation of the extant local template is based on Fiat’s firm willingness to invest resources in the form of substantial personnel transfer (FI 1). On the other hand, there is a high resilience of the local context with regard to a change in the functional differentiation. This is not least the case because the change of the local site’s functional differentiation does not fundamentally challenge existing status order at the site.

Outcome Despite some strategic distance between the Indian and other World Car sites, the Indian site is strategically close enough to implement the basic functional configuration. There is also no indication that strategic or institutional context, be it the existing local template or the wider host context, call for a major modification of the functional differentiation transferred. In strategic terms this is probably related to the fact that the template’s functional differentiation is already an adaptation of the

Empirical Results 145

‘Fabbrica Integrata’ to emerging market conditions. While the ‘management by process’ and the UTE concept seem to have been formally implemented, there is little information if the behavioural underpinnings, such as ‘teamwork free from hierarchies’ and ‘participated management’ have taken root (cf. Cardoso, 2004). Nevertheless, there are clear indications that the formal structure of the foreign/parent template has been transferred with regard to the site’s functional differentiation (FI 2). Thus, a clear transfer-intent, low recontextualization pressure, due to a moderate strategic and institutional misfit, and an adaptation of the local context as the dominant recontextualization mode, lead to an imitation of the World Car template’s functional differentiation.

The hierarchical differentiation of the organization structure Transfer scenario From the outset Fiat plans to transfer the hierarchical differentiation as an integral part of its World Car template. Derived from its ‘Fabbrica Integrata’ concept, Fiat’s World Car production template includes a standard hierarchy of five hierarchical levels (c.f. Camuffo, 2000/2001; Camuffo and Volpato, 1994). However, as market demand falls below expectations, Fiat drops its Greenfield project and decides to move its World Car project to the existing Brownfield site Kurla, where it commenced earlier the production of the Fiat Uno. While this shift to the Brownfield site does not cause Fiat to change its plan to transfer the World Car template’s hierarchical differentiation, the shift does imply that the template’s ideal hierarchical structure faces the much steeper hierarchical differentiation of an existing site. That is, of the production site Fiat has taken over from its Indian JV partner PAL. Yet, neither this potential institutional misfit, nor a local/host strategic distance cause Fiat to revise its intent to transfer the World Car template’s hierarchical differentiation. Fiat’s control over the site, its product ownership, its strategic intent to implement the World Car template and the capability gap between Fiat and its local partner PAL suggest and allow Fiat to pursue a transfer of the world car template’s hierarchical differentiation.

Misfit While there is no strategic misfit putting a recontextualization pressure on the foreign/parent template once the transfer commences, there is an institutional misfit between the foreign/parent template’s hierarchical

146 Hybridization of MNE Subsidiaries

differentiation and the existing local template institutionalized at the Brownfield site. After all, the existing hierarchical differentiation of the Kurla site mirrors the wider host institutional context conditions in India, marked by extensive socio-professional demarcation and hierarchical differentiation. As a result, there is a substantial resistance to the changes that go along with the implementation of the World Car template’s hierarchical differentiation (FI 4). Specifically, the new hierarchical differentiation implies a reduction from about twelve to five hierarchical levels (FI 4). This de-hierarchization causes resentment among those who feel that their seniority- or experience-based status is disrespected. Depending on how one looks at it, there is, a strong recontextualization pressure on either the existing local/host template or on the foreign/parent template to dissolve the misfit.

Mode of recontextualization Minor adaptation of foreign/parent template to the local/host context

FI shows hardly any signs of having adapted the hierarchical differentiation of the World Car template to the existing local template, local demands or the wider host institutional context. With one more hierarchical level than the ideal foreign/parent template envisions, Fiat India has come close to fully implementing the template’s hierarchical differentiation. The only aspects remaining from the local/host template are the titles of the former hierarchical designations (FI 4).

Major adaptation of local/host templates to the foreign/parent template Fiat carries through the reduction of hierarchical levels against local resistance. In line with its World Car template, the hierarchical levels of the existing Brownfield template are reduced, from previously twelve (and these are probably only counting managerial levels) to six levels and there is still one more level to be cut: When we were having in Fiat earlier Premier Automobile, there were many hierarchical levels. I don’t remember exactly how many but it was more than ten or twelve hierarchical levels. Now presently we are working, comparatively, with a very flat structure. We are having six layers in officers. That is: Vice President, General Manager, Assistant General Manager, Manager, Assistant Manager and Executive, so these are the levels. […] That is six grades. As per Italian standard we need to have as far as officer grades, white collar grades, are concerned, we need to have five levels. For that we have a tool, which has been developed by Italy. (FI 4)

Empirical Results 147

Fiat is in the position to change the local context or rather the existing local template’s hierarchical differentiation – despite its Brownfield background – because the company is quick to acquire a majority in the JV with PAL. As the Italian side is in control and the site’s top management staffed with Italian expatriates (with a clear mission for template implementation), there is little local top management left that can successfully oppose the changes in the organizational hierarchy. The dominant recontextualization mode with regard to the site’s hierarchical differentiation is, therefore, an adaptation of the local/host template to the foreign/parent template.

Outcome With around 2,000 employees and six levels of hierarchy, Fiat India has the flattest organizational hierarchy in the sample of the four research cases. Unlike the other cases, Fiat India does not feature a decoupling between formal hierarchical levels and a great number of hierarchical designations. Being a Brownfield site, FI faces an existing local organizational template. At the same time, Fiat has a clearly defined template for the site’s hierarchical differentiation. Even though the institutional distance between the local template and Fiat’s World Car template is substantial and causes resistance, the company pushes through the hierarchical differentiation of its World Car template. The fast equity shift allows Fiat to replace the old hierarchy with Fiat’s template without facing the intervening power of the local JV partner. Despite employee resistance, FI quite rigorously implements its foreign/parent template. As far as the formal hierarchical structure of FI is concerned, there is a clear shift from a local/host context solution, as is prevalent in PAL, to a foreign/ parent standard. The result can be labeled an imitation of the hierarchical differentiation of Fiat’s World Car template (cf. Cardoso, 2004).

Technical configuration of the production process Transfer scenario In line with the World Car project the structuring of FI’s basic factory layout, process design and technical hardware configuration is envisioned to follow the World Car production template. Specifically, the Indian World Car site is planned as a Greenfield project, based on the Argentinean site in Cordoba. As an integral part of its World Car strategy, Fiat aims at transferring similar technical configurations to all its World Car sites, including India. Coming to modularity in organization, with regard to intra-firm organizational design, the roll out of Palio in foreign plants followed

148 Hybridization of MNE Subsidiaries

a common technological and organizational template, exported and replicated in the different countries. As already mentioned before, each technological/organizational “module” was characterized by certain equipment and degree of automation (with possible variants and potential adaptation), and was organized to meet a given production capacity, which is 400 vehicles per two-shift day. (Camuffo 2000/2001, p. 22) However, this comprehensive transfer intent of a standard technological/ organizational module’ is partly dropped in the Indian case, as the strategic context proves too distant to allow the set-up of a completely new Greenfield site. The World Car template is optimized for 400 vehicles per day (Camuffo, 2000/2001). The Indian market demand allows, however, a production of hardly more than 200 vehicles a day. This strategic misfit makes it impossible for Fiat to set up an all-new World Car template as it has originally planned (The Tribune, 1999). The shift of operations to a Brownfield site instead, implies a task profile and spatial/physical conditions that differ from the strategic context the template is optimized for. Not only does the site have to cope with production volumes lower than the World Car template is designed for but there are also spatial restrictions related to the factory’s location in the agglomerations of Mumbai (FI 1, 2, 3). For example, the existing layout of the paint shop poses throughput limitations. This, in turn, poses limitations for the realization of the World Car template’s ideal cycle times. The lower volumes and spatial limitations also rule out the implementation of the World Car template’s nearby Supplier Park for JIT supplies, which translates into an obstacle to fully transfer the template’s Kanban system. While the strategic misfit between the local/host context and the template is poses already some barrier to transferring the template’s factory layout and process design, the situation is even more pronounced with regard to the technical hardware configuration. Despite the fact that the template is already optimized for low volumes and cost sensitive customers and labour intensive production in low labour cost countries, there is too much of a strategic misfit to transfer the template’s ideal technical hardware configuration. Thus, highly price sensitive customers – probably even more than in other emerging economies – and modest demand conditions are responsible for the misfit and transfer restraint. Although the market demand in India and the ensuing shift of the World Car production to a Brownfield site have some effect on Fiat’s transfer intent, a use of the local template in an unchanged manner is

Empirical Results 149

not an option. For one thing, because Fiat introduces its own product and finds the existing manufacturing infrastructure as strategically too distant to produce the World Car range of vehicles, at the quality, quantity and cost required. For another, because Fiat is quick to control the site, has the undisputed configuration mandate and thereby the ability to change the extant site as much as possible in line with its template.

Misfit While there is some transfer restraint of the World Car template, due to a host strategic misfit (low demand), there is also a misfit due to the existing local template of the Kurla site. The site’s existing technical configuration and infrastructure is, strategically distant from the World Car task profile in volume, quality and product specification terms. It does not fit the requirements of the World Car models and is, therefore, under recontextualization pressure to fit the foreign/parent template requirements or, where transfer is refrained from, to fit foreign/parent demands for producing the World Car under specific strategic host context conditions. It should be noted that this strategic misfit is not accompanied by an institutional misfit. There are no indications that the technical configuration of the foreign/parent template is perceived as a threat to or by the local/host institutional context. In fact, as far as host institutional demands are concerned, the World Car template is designed to respond to common emerging economy demands, such as high local content requirements and high import tariffs.

Mode of recontextualization Adaptation of the foreign/parent template to the local/host context The World Car template itself is already an adapted template, because it mirrors the adaptation of Fiat’s ‘Fabbrica Integrata’ to developing country conditions. However, as far as the application of the World Car template’s technical configuration is concerned this adaptation is not sufficient for the Indian strategic context. On the one hand, local supply and demand conditions, call for an adaptation or a de-selection of aspects of the foreign/parent template. On the other one hand, the extant physical infrastructure of the Brownfield and its location in Mumbai limit the resilience of the site for change. As a combined result of strategic and spatial misfit, Fiat is only selectively able to transfer its ideal World Car factory layout and process design. With regard to its technical hardware configuration the strategic misfit rules out the template transfer entirely. Those aspects that are transferred, however, do not meet further

150 Hybridization of MNE Subsidiaries

recontextualization pressure and are implemented in line with the template. Despite strategic and physical/spatial misfits, FI’s managers insist that it has been possible to reproduce major aspects of the factory layout and process design of the World Car template, such as a U-shaped flow layout (FI 2). Moreover, in line with the Cordoba plant and other World Car plants, the Indian plant achieves similar levels of vertical integration and outsources core process steps including welding (although performed on site) and forging. While some adaptation of the foreign template is reported with regard to the template’s factory layout and process design, such as a selective use of the Kanban system, there is a more substantial adaptation of the template’s technical hardware configuration to the host context (FI 2, 3). In fact, the strategic misfit makes Fiat refrain almost completely from transferring the technical hardware configuration. This, in turn, doesn’t mean the full absence of foreign/parent demands vis-à-vis the sites technical hardware configuration. After all, the site has to be able to produce a certain volume at a defined quality and cost level.

Adaptation of local context (existing template) to the foreign/parent template or demands To achieve a factory layout and process design in line with its World Car template, Fiat initiates a renovation and restructuring of the Kurla site. According to Atzeni (2001; see also Subhadra, 2002), Fiat upgrades manufacturing technology, machinery and renews buildings and other facilities (Atzeni, 2001). Regarding the site’s factory layout and process design, the upgrading largely follows the foreign/parent template implying an adaptation of the existing local template (FI 1, 2). As far as the set-up of the technical hardware configuration of the site is concerned, the local context is not adapted to the foreign/parent template (FI 1, 2). Instead, the local context is adapted to meet foreign/parent demands for the production of the World Car under specific (not too resilient) local/host demand and supply conditions. The adaptation of the local Brownfield context is, therefore, a customized solution and does not aim at replicating the World Car’s typical technology/labour mix. The Italian interviewee spoke in this context even of the use of old technology: A: So this factory is old. But the project is for a new car, the Palio. If you compare, this project is realized with old technology . . . The project is realized with old technology. Q: This is possible? A: Yes it is possible because the robot activity is here a manual activity. Here all is manual and it is old. I think Fiat didn’t want to invest in

Empirical Results 151

Kurla because it is a little factory, there is not enough space to put the cars, and the warehouse park is little. (FI 2) While there are few automated manufacturing aggregates, the factory tour revealed, a substantial level of mechanization. This mainly involves numerous welding stations as well as a few semiautomatic welding units for bigger body panels. Other upstream activities, such as blanking and pressing of body panels, are not performed on site, but supplied from outside. However, except for a few narrowly circumscribed manufacturing activities, the technical hardware configuration is characterized by very low levels or almost absent automation. It appears that transfer of advanced automation technology is tightly restricted to those areas in which it is a process-related requirement. This is the case in the paint-shop. The Indian market is very price sensitive, so if I am bringing technology then it is adding to the cost. As and when there is a requirement we put in technology. We have put in a robot in our paint-shop now because it was a requirement. If I am saving some money, maybe on labour or any other parameter maybe my delivery time or total time is decreasing, yes, then we are putting that money into the system. (FI 4) Overall, FI’s technical hardware configuration is a customized response to low and price sensitive demand and low labour costs in India. As the Kurla site is a run down, it requires some upgrading of the existing technical hardware configuration. Yet, this upgrading leads nowhere near other World Car site’s automation levels, let alone, Fiat’s state-of-theart technical hardware configurations in Italy. The fact that FI not even tries to replicate the technical hardware configuration of other World Car sites can be readily understood when comparing volumes. In Brazil, for example, Fiat has an annual output of about 500,000 cars. In India, the total annual production output of the entire Indian passanger car sector in 2002/2003 is around 600,000 cars. While the Brazilian output is at roughly 1500 cars per day, FI barely realizes 200 (FI 2). To conclude, the dominant model of recontextualization is an adaptation of the local site in line with the foreign/parent template as well as in line with foreign/parent demands and host context conditions.

Outcome According to the managers interviewed, the Indian site has successfully imitated the basic factory layout and process design of the World Car

152 Hybridization of MNE Subsidiaries

template. Here, some customizations are required which are related to the lower volumes and the physical/spatial restrictions of the Brownfield site in Mumbai. FI’s outcome with respect to its technical hardware configuration is for the most part a customized solution. This finding is also in line with Camuffo’s findings (2000/2001) showing that the Indian site’s automation level is among the lowest across the World Car sites and does not reflect the World Car template’s typical technical standard: An optimal “module” of four hundred vehicles was replicated for every Operating Unit in each of the Palio project foreign plants, with the exception of India (Brownfield plant in Kurla, Mumbai, in the Maharashtra state of Southwest India). Here, demand expectations and labour costs suggested a module adaptation and downsizing (characterized by partly different machinery and a lower level of automation) leading to a “degraded” module with a capacity of two hundred vehicles. (Camuffo, 2000/2001, p. 23) While the technical hardware configuration of the site relies on some transfers of technical hardware from Italy, this is not a replication effort of any template or existing site. Although the technology of Fiat’s technical hardware configuration e.g. welding guns, semiautomatic weldingunits, testing equipment – are transferred from Italy – the site reflects neither Fiat’s state-of-the-art technology configuration nor any of the other World Car sites. For the most part the technical configuration rests on technical solutions, customized to meet foreign/parent demands under specific host strategic context conditions that differ markedly from the template’s ideal context conditions. FI’s technical hardware configuration is mainly customized to low volumes and comparatively low labour costs in the host context – a finding also confirmed by Goldstein (2001). We can, therefore, best describe the technical hardware configuration as customized solution and the overall technical/physical organization of the production process as being between imitation and customization.

Work organization and human resource profile Transfer scenario The UTE concept is the corner stone of the World Car template’s work organization and human resource profile. Similar to other World Car sites (Cardoso, 1998; Cardoso, 2004), the transfer of the UTE concept to the Indian production site is an important if not the most important transfer goal (FI 2, 4).

Empirical Results 153

Figure 6.1

UTE 1

UTE 2

UTE 3

Body shop

Paint shop

Assembly

The UTE concept

The UTE concept is first introduced in Italy and becomes subsequently an integral part of the World Car production template. UTE’s are the central operational units that integrate all tasks and jobs to ensure the continuous production of a certain product at a defined volume, cost and quality. Different UTEs are organized along the production process and are linked in supplier-customer relationship to one-another (see Figure 6.1). The idea of the UTE is to integrate vertically all responsibilities to the lowest possible level and horizontally all activities and functions required for a particular process segment. The UTE is designed as a highly autonomous and integrated unit that takes care of: operational planning, smooth production/variance absorption, quality assurance, problem solving and even continuous process improvement, learning and training of its human resources (Camuffo and Volpato, 1994, 1998). Structurally a UTE chief heads a UTE or ‘a team’ comprising on average of about 50 people. As part of the UTE, technologists (product and process technologist, line technologist, quality technologist, specialist technologist) and technicians take on different functional tasks to support the UTE chief (cf. Camuffo and Volpato, 1998). Cardoso (2004) explicates the role of the technologist as follows: The target of the technologist of product and process is to support the integration between the technical team of the product project and production control. The line technologist is interested in each unit in terms of the technical management of work resources, as well as the objective of continuous improvement in cycle times, notably in the context of technical cost. The role of the quality technologist, simply put, is to ensure high product quality. Finally, the specialist technologist is responsible for technical support at project level, which includes providing solutions to technical anomalies. (Cardoso, 2004, pp. 65–7) As far as the direct work organization on the shop floor is concerned, a group leader – a Conduttori di Processi Integrati (CPI) – reports to the UTE-chief and heads a group of workers. Thus, the CPI is positioned

154 Hybridization of MNE Subsidiaries UTE Chief Four Technologists

Technicians

Integrated process conductor (CPI) Production operators Source: Adapted from Cardoso (2004).

Figure 6.2 Organization of an UTE

between the UTE chief and the operators/workers who are performing the direct production operations. One UTE can consist of varying number of work groups – depending on the kind of production line segment (see Figure 6.2). As a deliberate alternative to Taylorist work concepts, the UTE concept builds on delegating responsibilities to the lowest possible levels and emphasizes the important role of workers. Cardoso (1998, 2004) points out that generally the UTE chief, the CPI, as well as the supporting technicians and the technologist are recruited from the ranks of the workers. They are usually highly skilled and experienced workers. Moreover, the role of workers on the line is emphasized. They are asked to perform a wide range of activities which are not directly related to production such as self-certification and quality control, continuous improvement and housekeeping etc. (FI 2, 4). The UTE concept builds on a human resource profile that stresses skill-polyvalence and high-involvement of the worker level. Now, it is this concept Fiat also tries to transfer to its Indian operation. Fiat is able to do this because the company is quick to take control of the JV with PAL, introduces the vehicles and has the configuration mandate. Moreover, after quality problems with the local partner in the joint UNO production, Fiat finds the introduction of a different form of work organization as crucial to meet the volume, quality and cost requirements of the World Car range. While Fiat rules out to continue with existing patterns of the work organization at the Kurla site, it does work with the extant human resources. As far as the transfer of the basic formal concept of the UTE system is concerned, institutional or strategic distance pose no transfer obstacle ex ante. The strategic context regarding the demand and supply conditions as well as the corresponding task profile are similar enough to allow a transfer of the World Car template’s work organization

Empirical Results 155

and human resource profile. After all, the UTE system is designed for the strategic conditions of low labour costs and low volumes. However, the distant local/host institutional context, the entrenched behavioural patterns of the Brownfield site, lead to some transfer constraint with regard to high-involvement practices that form part of the work concepts of the World Car template (Atzeni, 2001; Goldstein, 2001).

Misfit When the transfer effort begins, it becomes clear that institutional conditions at the Fiat Brownfield site are not conducive to apply all aspects of the work organization and human resource profile of the foreign/parent template. However, some differentiation in terms of institutional fit and misfit is again required here. The local institutional conditions do not stand in the way of transferring and implementing the formal structure of the UTE system, which forms a central part of the World Car work organization. For example, there are no reported demarcation problems between CPI’s (supervisors) and operators. Part of the institutional fit in supervisor-operator relations is probably related to the fact that Fiat’s UTE concept is not interfering with (or circumventing) managerial work identities. As part of the UTE concept, supervisors – the CPI – are recruited from the ranks of the workers. This implies that Fiat does not ask employees with a managerial work identity to work as supervisors close to the shopfloor. Simply speaking, the UTE concept does not seem to call into question local/host socio-professional demarcations. At the same time, even though UTEs are integrated units, they are hierarchically and functionally differentiated (c.f. Camuffo and Micelli, 1997). This, in turn, fits well with the Indian work context, where social stratification and educational systems produce strong socio-professional demarcations. While the structural aspects of the UTE work organization do not seem to cause a major local/host institutional and strategic misfit, this is different with regard to work dispositions and practices envisioned by the template. The local human resource profile, the old workforce of the Brownfield site, proves to be institutionally about the opposite of what is required for practices such as continuous improvement, allocation of responsibilities down in the line, quality circles, and self-certification. Particularly in the beginning, Fiat faces an obstructive workforce, which it inherits from the JV partner PAL. The workforce reflects traditional Indian patterns of low efficiency, low involvement and contentious industrial relations. In fact, Fiat’s Kurla facility is renowned for its hostile labour relations. Against this background, Atzeni (2001) who researches

156 Hybridization of MNE Subsidiaries

the Indian site in late 1990s, reports a lot of frustration on the part of Italian expatriates. And these frustrations with the local workforce have not entirely faded when the site is visited in 2003 for this study (FI 2). The biggest misfits between the foreign/parent template and/or demands revolve around the issues of quality awareness and cleanliness, the taking of responsibility and most of all work commitment and involvement (Atzeni, 2001). Apart from this quality-related work disposition, the Italian expatriate interviewed complains about the initial reluctance of the Indian employees to take responsibility and to work in a self-directed manner (FI 2). Similarly, Atzeni (2001) and also Goldstein (2001) report that it is very difficult to transfer a range of high-involvement activities due to the local institutional conditions of a Brownfield site. Goldstein – who also has the opportunity to research FI – finds in this context: Despite introducing in Kurla one of the basic organisational features of the integrated factory model – the so-called elementary technological units (UTE) – fully implementing lean manufacturing has been very difficult. Because of labour market regulations, Fiat has been unable to substitute PAL workers with a younger workforce and found it hard if not impossible to adopt concepts such as continuous improvement, total employment involvement, and quality control circles and change the value system. (Goldstein 2001, p. 15) Moreover, Fiat faces resistance when it begins to introduce housekeeping activities to the work profile of operators (FI 2). This resistance is rooted in the pronounced socio-professional demarcations in India’s institutional context. Finally, Italian expatriates and Indian managers complain about unsatisfactory skill levels of newly recruited workers. This misfit has its institutional roots in the weak practical training component of the Indian industrial training system (see Chapter 4).

Mode of recontextualization Substantial adaptation of foreign/parent templates to the local/host context Interviews underline that there is no adaptation of the formal structure of the UTE concept to the local/host context. While the formal aspects of the World Car template’s work organization are transferred, notably the UTE concept, there are indications that corresponding human resource profile deviates from the ideal staffing pattern. In contrast to other World Car sites, for example in Brazil (Cardoso, 1998, 2004), the UTE chiefs in India are not recruited from the ranks

Empirical Results 157

of workers. In India, this position is staffed with polytechnic graduates (officers) and not with workers. It may well be that Fiat, in order to offer at least some hierarchical differentiation to its managers, pulls down the managerial hierarchy into the UTE concept. Thus, placing non-workers into the UTE’s usual profile can be seen as an adaptation of the concept to local/host qualification-related staffing patterns. At the same time, such a higher qualification approach to the staffing of the UTE does not cause a strategic misfit due to the low labour cost in the host context. A more important adaptation of the World Car template’s work organization to the local/host institutional context is that Fiat drops or is very slow to implement many of the UTE template’s underlying practices. Cases in point are the high-involvement activities, such as continuous improvement, total employment involvement, and quality control circles, self-inspection. For example, under the UTE concept workers are ideally responsible for quality assurance based on self-certification. In FI, this concept is not implemented and relies, instead, on a line-external quality control (FI 2). Thus, indications are that FI does not fully succeed in implementing the high involvement concepts of World Car template’s work organization. In this respect, Fiat either drops or adapts them to the work dispositions of the Brownfield site’s human resource profile that proves not very resilient.

Substantial adaptation of local/host template to foreign/parent template Having said that Fiat does not succeed in transferring certain aspect of its UTE concept is not to say that there are no efforts to adapt the local institutional context or the existing local template. In fact, FI takes a whole range of measures to change or re-institutionalize the local human resource profile. These measures include: a re-composition of the work-force, socialization efforts through personnel transfers, and on site training and schooling. One measure to change the local human resource profile is based on filtering and laying-off a substantial amount of the workforce and replacing it with a new one: See, when Fiat came into this site, instead of a Greenfield, which was having old technology due to a lack of investment and the manpower with an old culture, you know, to overcome this, we were thinking to invest and gradually to joining together there was a scanning and people were filtered. A good amount of those filtered received all sorts of training and were given to the human resources. (FI 3)

158 Hybridization of MNE Subsidiaries

Based on overstaffing, a problematic old work culture and a lacking economic success, FI also sees the need to reduce its old workforce through VRSs. In 1998 FI lays off about 1500 workers, getting rid of about half of its old workforce. In the early 2000s, when the introduction of the Palio leads to a rise in demand, Fiat takes the opportunity to change the composition of its work force by hiring 600 new ITI graduates (FI 3). An important part in adapting the local context to the template, are also investments in socialization through personnel transfers and in on-site training and schooling. At the time of research, eleven Italian expatriates are heading various top management positions in FI. With FI’s decision to take over the majority in PAL and to set up the World Car production in Kurla, Italian expatriates take over the control of the site’s development. Italian expatriates have the task and the positional power to socialize, train, or enforce a work organization and corresponding human resource profile in line with the World Car template. That the latter is only gradually possible has been discussed earlier. However, despite adverse local conditions, the commitment to implement UTE’s high involvement practices remains an ongoing goal of the Italian expatriates (FI 2). Open information policies, reducing hierarchical distance and behavioural consistency are also mentioned as actively pursued practices by Italian management to come closer to the desired work dispositions (FI 2). With regard to creating quality awareness within the Indian workforce, personnel transfers also play a crucial role. Expatriates serve as trainers to institutionalize higher quality awareness. Additionally, special teams are sent for shorter periods from the pilot plant in Italy to train the required quality standards in India (FI 2). At the same time, Indian quality specialists are regularly transferred to Italy for training. Fiat even transfers around 100 workers to the World Car pilot plant in Turin for training. Than there were one, two, three, four batches of workmen were sent to Italy. Around 50–100 were sent to Italy, many sent there for two months, three months, six months to understand the work culture of the European Auto Fiat. How they maintained the shopfloor, how they work, what is the dress code there, how they behave, how the testing is done, technically and also all other things. Three months, four months the people were sent there. So the people are sent there and then they are brought here. . . . The line was set up by the Italians along with the Indians, so it was making a Fiat Auto in India in this factory. Everybody was, the local man, were very proud that I am

Empirical Results 159

part of the team which is making Fiat Auto. So mentally, psychologically everything was starting to change. It was not only the technical, the instruments which was changing. Also the minds of people were changing. Also the technique was made available, it was seeing and believing. So people were sent, this is not a magic what Europeans are doing, but you really go and see. It is not a magic, it is a practical problem, so hundreds of workmen were sent to Fiat factory to see physically and believe and start changing their own. Those 100 people have already changed the other people. (FI 2) Upon return, the workers transferred to Italy assume positions as CPI in the UTEs and are asked to transfer their know how into their work groups (FI 2, 3). While there are no indications that FI uses incentives or compensation systems to institutionalize different work dispositions, its effort to promote industrial peace through pay rises probably also plays a role. Moreover, by appointing an Italian manager experienced with radical union action as the head of human resources and by changing the human resource approach to more transparency, the company succeeds in bringing about industrial peace. Such an industrial peace, in what uses to be a very conflict-prone company, also suggests that work dispositional changes are achieved (c.f. Atzeni, 2001). Nevertheless, an Italian expatriate interviewed in 2003 still displayed a great deal of frustration despite the various measures mentioned to bring the local workforce in line with the human resource profile required by the Word Car template work concepts. To conclude: The recontextualization mode shaping FI’s work organization includes both a substantial adaptation of the foreign/parent template to the local context/template and vice versa. FI’s Brownfield human resource profile prove to be not overly resilient for a change towards high involvement work disposition. At the same time, FI’s mobilization of resources for change is also not as comprehensive as in MSI’s case.

Outcome Indications are that Fiat succeeds in transferring and imitating the formal or structural aspect of its UTE concept of World Car template’s work organization. What we are having, we are having an Italian system implemented more or less implemented here in India. If you look at the typical

160 Hybridization of MNE Subsidiaries

production area, where we are having a, that is, each assembly line we have divided into a system called UTE. So UTE is basically a common group of working, where one particular similar kind of activity is being manufactured. Than total, or let’s say for example, 178-Siena, which has been broadly divided into broadly three categories: so UTE one, UTE two and UTE three. Each UTE is headed by an UTE-chief. He is basically the officer responsible for the workmen. Between workmen and that officer we have a group leader. He is called Capo UTE [also called CPI]. (FI 4) However, Fiat faces severe problems in transferring concepts of high involvement activities that are an integral part of the ideal UTE concept. Especially in the beginning, the site’s human resource profile reflects work dispositions that are traditional Indian industrial work relations and not conducive to implementing post-Taylorist work concepts. Due to the site’s Brownfield heritage it has been very difficult, if not impossible, to transfer work dispositions to fully implement the World Car template’s work organization and shopfloor practices. Only slowly, through the gradual adaptation of the local institutional context, a selection and change of the local human resource profile has FI been able to move closer to the implementing of its World Car work concepts and the work dispositions these concepts require. In summary: Fiat’s work organization and human resource profile features elements of typical local institutional patterns and elements mirroring the foreign/parent template. Particularly in the beginning, there is a strong decoupling between the imitation of formal, structural elements of the foreign/parent template, and work disposition reflecting local/host patterns. Over time, Fiat seems to have achieved changes in line with its template. Over all, adaptations of the transferred work organization – by dropping or changing certain practices – as well as adaptations of the local human resource profile – through selection, socialization and training – suggest that the resulting work organization and corresponding human resource profile could be best captured as a hybrid solution, reflecting both foreign/parent template and old local/host template origins.

Industrial relations Transfer scenario Concerning FI’s industrial relations there is no conclusive evidence whether Fiat seeks the transfer of a specific industrial relations pattern.

Empirical Results 161

While it is part of the Fiat group’s defined policy to handle industrial relations ‘in compliance with the law and practice of the individual countries’ (Cantarella, 1997), there is, at the same time, a broader formulated demand for developing cordial and participative labour relations. Recognizing that industrial relations tend to be different from country to country, the Fiat group formulates only broad guidelines, rather than defining specific modes of company industrial relations to be transferred as an integral part of the World Car template. This approach to industrial relations is also reflected in comments by an Indian interviewee: Globally, you can run globally in terms of systems maybe human resource developmental systems. Like what I have just shared with you competence assessment system, than performance management system, than people satisfaction survey. All this can run globally. But when we talk about local environment workers settlement, agreements with the workmen, wage negotiations, that has to be localized, what we are having. Ok, we take a guideline from Italy, but it has to be handled, it has to be taken care in the local way. (FI 4) Fiat’s guidelines are a reflection of the company’s overall reorientation towards emphasizing the importance of human relations in the 1990s that goes hand in hand with the introduction of the ‘Fabbrica Integrata’. Within this new production paradigm, human resources are given a central role. It is stressed that such a constructive role can only be realized if hostile industrial relations are overcome. Cardoso (2004) even argues that the ‘Fabbrica Integrata’, involves a clear anti-union stance, because unions stand in the way of integrating workers deeply into the company and creating a work identity based on the idea of the ‘Fiat family’. As Fiat’s World Car production template is an offspring of the ‘Fabbrica Integrata’, it too rests on a concept of cordial labour relations and probably an anti-union stance. Clearly, the human-factor, specifically a high involvement and commitment of the workforce are strongly emphasized in the World Car template. So, while there is no comprehensively articulated template to structure industrial relations, there are broader policies from the foreign/parent calling for constructive, peaceful and ideally non-unionized labour relations. Now, Fiat’s Brownfield site Kurla has an existing template of typical local/host industrial relations that contrast markedly with the foreign/ parent’s industrial relations policy. In fact, FI’s company industrial relations profile has a record of hefty labour unrest and strong external union involvement. As the existing template is not conducive for the

162 Hybridization of MNE Subsidiaries

implementation of the World Car template’s ideal work organization and stands against Fiat’s corporation wide policies, FI’s management is not willing to accept the local situation. The fact that Fiat is in control of the JV and Italian managers assume top management positions, including in human resources, gives Italian managers the mandate to change the local situation of industrial relations.

Misfit Recognition of the fact that industrial relations vary by country does not mean that Italian managers are satisfied with the company industrial relations they find in FI. In fact, when Fiat takes over the Kurla site from PAL, industrial relations conditions are about the opposite of what Fiat policies ideally envision and require for its high involvement work concepts. The local situation is a far cry from the broadly demanded cordial industrial relations let alone a non-union site. This is not surprising given PAL’s history of hostile company industrial relations. The company has different rivaling unions. At least one of them is externally linked and, in addition, the company sees prolonged periods of strike and labour unrest. When Fiat takes over the Brownfield operation, FI’s industrial relations have all the characteristics of traditional Indian adversarial labour relations (Atzeni, 2001). Atzeni (2001) who researches the site in the late 1990s reports that company industrial relations are so adverse – and compounded by cultural differences between Italian and Indians – that Italian managers see little scope for the implementation of constructive and participatory labour relations. Questioned if they had promoted or had planned to introduce a different model of IR, the Italian managers answered that there was no room for any type of discussion around the argument. The Indian reality was so different and contrasting to that of Italy that any attempts or ideas to move forward were considered risky and fruitless from the start. They also argued that a union such as the one existing, with their members and their “original” way to behave was a sufficient reason to keep labour relations as they were conflictive, chaotic and not stable, but at least they knew with whom they were dealing. . . . Later on, further attempts to see labour relations in the plant from a different perspective were specifically rejected. “We are in India, a place where industrial relations are like the times of Valletta. There is no room for dialogue. Concertation, participation are very far from this reality”. (Atzeni, 2001, p. 9)

Empirical Results 163

Put differently, there is a huge distance between the existing local template and the foreign/parent policies. However, while the institutional distance seems to rule out any change of the local/host template in the beginning, this situation changes over time. Italian managers do not accept the extant industrial relations scenario at their plant and put a pressure on the local/host template to change in line with foreign/parent policy demands.

Mode of recontextualization Major adaptation of foreign/parent demands to the local/host context In the first years of operation, the local institutional conditions of the Brownfield site prove so un-resilient that Italian management drops the intention to bring industrial relations in line with Fiat’s industrial relations policies. It essentially implies an adaptation (surrender) of foreign/parent demands to the local/host context. However, at the time of research in 2003, indications are that the company has made a big leap towards to fulfiling Fiat’s industrial relations policy demands. After all, these demands are not overly comprehensive and mainly ask for cordial industrial relations (ideally without unions) in line with local/host context conditions.

Increasing adaptation of local/host template to foreign/parent demands While initially cooperation and participatory relations are out of reach, the Italian and local management makes an effort to turn the company industrial relations into a peaceful local/host mode. A part of this effort expresses itself in the preferential treatment of an internal union that emerges during a prolonged strike. After a long strike and lockout in 1996 the external union, the Engineering Workers’ Association (EWA) which leads the industrial action, loses support. In the wake of EWA’s weakening, a new internal union, the Kamgar Ekta Premier Sanghatana (KEPS), is formed. This union emerges from a fusion of the existent internal union, the PAEU (abbreviation origin unknown) and some members of the EWA. In this scenario of a weakening external union, Italian management ceases the opportunity to support the internal union (FI 2, 3). This is seen as the best option, given the industrial relations conditions in India in general and at the FI’s site in particular. Atzeni states in this context that the ‘internal union exists just because management has preferred to deal with it rather than run the risk to bargaining with external powerful unions’ (2001, p. 7). Although the Kurla site has a wild industrial relations record, with one of its peaks in 1996, culminating in a six months strike and lockout,

164 Hybridization of MNE Subsidiaries

things calm down after Italian management takes root in the company. Indications are that this change is achieved through a mix of confidence building measures. Apart from a different communication approach, change also results from the re-composition of the local workforce as well as personnel transfers and training of Indians in Italy. The appeasement is probably also related to the fact that Fiat transfers human resource managers from Italy. According to Atzeni (2001), these are highly experienced and tested in dealing with industrial conflicts. What is more, although the internal union has initially little backing in the workforce (Atzeni, 2001), pay rises and improvements in working conditions help the internal union to gain increasing acceptance or toleration within the local workforce. Finally, a new generation of Italian managers comes to the Indian site. They express the commitment to more cooperative and participatory modes of labour relations (FI 2). Thus, while Fiat has to initially adapt its demands to a non-resilient local institutional context, it is over time able to adapt the local context to foreign/parent demands. It is able to do so by mobilizing resources to change the work force composition and dispositions, helping to bring about a more resilient local context.

Outcome Being aware of differences in industrial relations across countries, Fiat refrains from making specific modes of industrial relation part of its World Car template. Instead, Fiat promotes an industrial relations policy that broadly aims at achieving cordial labour relations in harmony with local/host rules and regulations. However, even in this broad sense, Italian managers see initially little scope for a policy implementation. The existing local/host industrial relations template of the Brownfield site proves so adverse that the alternative approach demanded by the foreign/parent policy is put aside. There are profound differences between Italian managers, unionists and workers. Different cultures, different languages and mutual mistrust are factors that do not contribute to change the climate of labour relations from conflict to cooperation. To this we should add the heritage of anarchy that dominated the plant during the last decades. The research has shown, at least for what concern Fiat’s plant in India, that lean production is not always coupled with a more co-operative labour/management approach. (Atzeni, 2001, p. 15)

Empirical Results 165

Over time the site’s industrial relations turn more peaceful. FI’s last strike is in 1998 and there is no new one since (FI 3). While FI’s industrial relations have probably not changed into a showcase of cooperative industrial relations, they have become very peaceful from the Indian and Italian management’s point of view. The Italian manager interviewed even describes FI’s company industrial relations as ‘paradise’, compared to the situation in Italy (FI 2). In contrast to what Atzeni (2001) found earlier, my findings suggest very smooth labour relations in 2003 (FI 2, 3). This change reflects efforts of changing the site’s industrial relations into a more cordial and less conflict-prone mode. This adaptation of the existing template does not imply a shift toward a typical foreign/parent pattern. In rather means a shift to a different local/host context pattern, more in line with Fiat’s policy. That is, a mode of industrial relations, characterized by one, rather weak, unaffiliated internal union and low levels of industrial action. As such, FI’s industrial relations shift from a traditional to a more modern local/host context pattern. We can, therefore, label FI’s industrial relations as localization.

Supplier relations Transfer scenario In line World Car template Fiat plans the transfer of a number of supplier relations concepts to its Indian operation. In can do so because the company is quick to take over the site, introduces the vehicle and has the configuration mandate of supplier relations. The original transfer intent envisions the set-up of a new site in India. This Greenfield project is based on extensive outsourcing and includes the establishment of an adjacent supplier park. The suppliers, in turn, are expected to supply just-in-time allowing low levels of inventory. With regard to the supplier structure the World Car plants typically rely on global suppliers that follow Fiat worldwide (Camuffo and Volpato, 2000/2001). Yet, Fiat is not following a rigid template transfer strategy with regard to supplier relations. The main goal is rather to ‘[d]evelop a global supply chain, flexibly and efficiently using the production capacity and the supplier base available in different countries’ (Camuffo and Volpato, 2000/2001, p. 4). Thus, more important than strictly following an ideal template is that supplier relations fit into Fiat’s ‘world material flow’ and respond to purchasing policies that are part and parcel of the World Car programme. The ‘world material flow’ can be understood as a ‘double network’: comprising an internal exchange of ‘makes’ between different World Car sites and an external supply of ‘buys’ (Camuffo and Volpato,

166 Hybridization of MNE Subsidiaries

2000/2001). The internal network is composed of global poles of Fiat group companies, producing parts and components for the World Car. The external supply of ‘buys’, in turn, is composed of a network of first tier suppliers that are asked to follow or at least supply the World Car sites, wherever Fiat decides to establish them. The kind of suppliers and their supply location is not a coincidence, either, but is defined by Fiat’s World Car purchasing policy. The choice of a particular supply mode depends on two variables: The first variable is the ‘technological characteristics of the part/component, i.e. the complexity of know how required to produce it’ and the second variable is the ‘incidence of logistic cost’ (Camuffo and Volpato, 2000/2001, p. 11). Based on these two variables, know how complexity (high/low) and transportation costs (high/low), Fiat draws on different supply modes (Camuffo and Volpato, 2000/2001). The above discussion of Fiat’s world material flow and material purchasing policies highlights that there are specific concepts and policy demands, the Indian site has to respond to. However, from the outset a part of the transfer intent cannot be realized. As the demand scenario in India is much below initial projections, Fiat puts its Greenfield project on hold. This, in turn, has strong repercussions for FI’s supplier set-up. Given the overcapacities of the Indian supplier industry and given FI’s low volumes, the company has difficulties to convince many of its global suppliers to open new ventures in India. FI has already problems to give attractive orders to those suppliers who follow to India (Alzona and Prakash, 2001). Moreover, even though a few suppliers follow to India, FI relies for the most part on local suppliers or other international suppliers who have come to India earlier. The disadvantage of this is, however, that many of these suppliers are dispersed all over the country. This, together with FI’s decision to use the Brownfield site Kurla for its World Car production, makes the realization of JIT-supplies from a nearby supplier park impracticable. The spatial constraints of the Kurla operation essentially renders the concept of a nearby supplier park with JIT deliveries impossible (Goldstein, 2002). While Fiat gives up certain elements of its ideal World Car production template, many of the World Car material flow policies apply and have a shaping impact on FI’s unfolding supplier relations. At the same time, FI’s supplier relations are also built in response to host institutional and strategic context. Specifically, observing institutional demands such as local content requirements or the supply market conditions for parts and components in India. Thus, the site supplier relations are mainly built in response to foreign/parent and host context demands and conditions.

Empirical Results 167

Misfit While some of the core World Car template’s supplier relations concepts cannot be transferred because of strategic misfit, there is also a strategic misfit between the local template of the Kurla operation and the strategic context of foreign/parent. The existing supplier relations at PAL neither fit with the structural and contractual concepts of the World Car’s supplier relations template nor with Fiat’s supplier policies (above all in quality terms). Therefore, the existing supplier relations scenario comes under substantial recontextualization pressure. Overall, however, Fiat’s supplier policies and the host institutional demands fit each other quite well. After all, these policies are preconceived to meet host institutional demands of developing countries and are on top of that flexibly defined. Host institutional conditions such as the availability of a local supplier base, taking advantage of local labour costs differentials, observing the incidence of logistics costs, local content requirements, are all factored into the Fiat’s World Car purchasing policy and suggest situation specific responses. For example, in a scenario where a site is facing limited volumes, high local content requirements in a low wage environment, where other major World Car production poles (the Latin American ones) are far away and where many other international auto supplier are already in the market, the policy suggests a strong localization of production, taking advantage of the existing host context suppliers infrastructure. Moreover, given that Fiat produces a low to medium segment vehicle for price sensitive customers, production and transportation costs have to be kept at a minimum. It is, therefore, in Fiat’s own interest to localize substantial parts of its production in India. Thus, while Fiat meets some strategic misfits that rule out to transfer of some of its preferred supplier concepts, host strategic and institutional conditions and demands do not call into question the application of Fiat’s core supplier policies.

Mode of recontextualization Some adaptation of the foreign/ parent template/demands to the local/host context At the outset, Fiat plans to transfer a number of supplier concepts of its World Car template. However, in the face of strategic misfit, FI de-selects the transfer of major structural and logistical concepts. As a consequence, Fiat’s main first tier suppliers are not supplying FI locally. Instead, FI draws much more on locally existing suppliers than other World Car sites. Fiat’s core World Car suppliers are either not able

168 Hybridization of MNE Subsidiaries Table 6.1

Fiat’s suppliers in India

Cluster

Delhi Chennai Bangalore Nagpur Indore Baroda Aurangabad Pune Mumbai

Distance from Kurla (kms)

Transit time (days)

No. of suppliers

No. of Items

% of total value

1408 1367 1033 861 600 500 400 200 100

5 5 4 3 3 2 2 1 1

29 11 7 2 1 1 1 25 43

124 53 22 6 1 9 1 146 386

34 19 3 0 3 9 3 20 10

Source: Goldstein, A., Fiat Group’s Projects in E-Business’, in ‘Fiat’, reproduced in ‘Electronic Commerce for Development’, OECD (2002), table 5.4, P. 108. http://www.oecd.org/bookshop? 9264099549. (By permission of OECD.)

or willing to set up their own ventures in India under such low demand conditions. Comau and Teksid, for example, two of Fiat’s main suppliers are not supplying or servicing the company. Fiat also drops the nearby supplier park concept and with it the concept of JIT deliveries (FI 2). Facing lower volumes than expected, implies that FI has to work with local suppliers or international suppliers already in India. These, in turn, are located all over India (see Table 6.1). As a result, FI has to take into account longer transit times, transport related risks of long distance deliveries under poor road conditions, which translates into higher inventory levels. The fact that Fiat has to drop core aspects of its World Car supplier concept does not mean that FI’s supplier relations do not comply with Fiat’s core supplier policies any more. In this respect, there is little adaptation of foreign/parent demands to the host context as the policies fit quite well with the host strategic and institutional context. In line with its policies, FI integrates into the ‘world material flow’. FI receives from other World Car production poles (the internal network) a number of crucial components (about 10 per cent of the car’s value). Body-panels come, for example, from Brazil and engines from Italy (FI 4). FI’s external network of ‘buys’, is also in line with the companies purchasing policies. FI succeeds, for instance, in a substantial localization of production, claiming to have achieved a real local value addition between 80 per cent and 90 per cent (FI 2, 3). These efforts are highly reflective of Fiat’s policy of controlling logistics costs by means of reducing transportation costs, costs related to import tariffs or local content

Empirical Results 169

requirements. That is, ultimately to check the cost for a product that caters to price sensitive customers in developing countries (FI 2). In line with Fiat’s strong outsourcing policy FI even sources out major on-site production steps, specifically the body welding, similar to what is being done at other world car sites (Camuffo, 2000/2001). Overall, the FI’s supplier relations comply with foreign/parent demands and partly resemble those in Argentina and Brazil – in structural and contractual terms. In structural terms, like in Argentina and Brazil, FI forms partnerships and cooperates with suppliers that are either international suppliers or have JVs with international suppliers. In accordance with its policy, FI takes advantage of existing local capabilities and capacities. Cases in point are supplier relations with Mico-Bosch, Tata-Johnson Controls, Automotive Stampings and Assemblies LTD, etc. Indications are that some of Fiat’s core suppliers have contracted out major components to other suppliers in India. Here also there was a provision when we started in this period there were three phases. Phase A was when the contract was done with the internal supplier and with the Indian supplier. Phase A was the Indian supplier will take the total component ready from the Italian supplier and supply under his name to this company and till that time he develops more and more. Phase B was instead of assembly he will get some parts, which are very critical and some he has already developed. He assembles them here and gives them to us. And phase C, the supply from Italy stops completely and he makes the whole on his own. (FI 3) To ensure the required standard quality for the World Car, Italian expatriates strongly cooperate and get involve with the Indian suppliers. Like in the other cases, the cooperation is not only a policy requirement but a necessity to assure quality. In this context, the Italian expatriate underlines his strong personal involvement to achieve the desired qualities for the World Cars produced in India (FI 2). In India, Fiat relies probably less on its global group of core suppliers than elsewhere. Yet, it is still following, in contractual terms, its template’s approach of selecting new local suppliers as global suppliers to be tied into long-term relationships. In summary, Fiat has to deselect/adapt core aspects of its ideal World Car supplier set-up to local/host context conditions. At the same time, it does not have to adapt its global supplier policies to the local/host context because they are flexibly defined and fit well into a developing country context.

170 Hybridization of MNE Subsidiaries

Some adaptation of local/host templates to foreign/parent demands Fiat clearly finds the existing local/host supplier relations unfit to meet its requirements. Given that FI is a World Car site, given that Fiat is quick to control the site and given the technological and task environmental distance between the PAL’s operations and the World Car requirements, it is out of the question to use the Brownfield site’s existing supplier relations as a template. Instead, FI’s supplier relations have to be tied into the world material flow, respond to Fiat’s World Car supplier policies and respond to the specific host context strategic conditions and institutional demands (such as local content requirements). Fiat, therefore, engages in an effort to adapt the local/host supplier relations to its needs, observing at the same time local/host context demands and conditions. Regarding the latter, Fiat has no problem to fulfil the host institutional demands for increasing local value addition, because this is in its own interest. How these local content requirements are realized, mainly follows from Fiat’s global purchasing policy. The dominant mode of recontextualization with regard to FI’s supplier relations is selecting, changing and partly creating supplier relations (host context resilience), in line with both foreign/parent policy demands and the local/host institutional and strategic demands and conditions. The existing suppliers in India offer FI a developed pool of capable suppliers from which it can chose. Still some involvement of Fiat personnel (resource mobilization) is required to bring these suppliers in line with Fiat’s quality standards. However, although some adaptation of these local suppliers is required, the set-up of Fiat’s supplier relations in nowhere near the institutionalization or (re-) institutionalization effort first movers, such as Maruti-Suzuki, had to undertake. For when Fiat enters India, the supplier context is comparatively developed. Outcome Fiat is not able to transfer a number of supplier relations concepts of its World Car template. This transfer restraint is mainly caused by a strategic misfit between the foreign/template requirements and the local/host strategic context. At the same time, there is also a strategic misfit between foreign/parent demands and the existing local/host supplier context of the Brownfield site. It is important to note that Fiat never rigidly applies and imitates its ideal supplier relations template to all World Car sites. After all, different World Car sites have different roles in the internal supply network of the world material flow. This rules out a similarity of supplier relations across all World Car sites in the first place. As Fiat is not able to

Table 6.2 Summary of FI’s hybridization profile FI

Functional differentiation

Transfer: Foreign/parent template transfer (meets local existing template) Fit/Misfit: Minor strategic misfit between foreign/parent template and local/host template Mode of Recontextualization: Mainly adaptation of local template – change and creation – to foreign/parent template Outcome: Imitation

Hierarchical differentiation

Transfer: Foreign/parent template transfer (meets local existing template) Fit/Misfit: Institutional misfit between foreign/parent template and locally existing template Mode of Recontextualization: Hardly any foreign/parent template adaptation; adaptation of local template to foreign/ parent template requirements Outcome: Imitation

Technical configuration

Transfer: Selective foreign/parent template transfer Fit/Misfit: Some strategic misfit between existing local template/manufacturing infrastructure of Brownfield site and foreign/ parent template and demands Mode of Recontextualization: Mainly adaptation of local template (i.e. change of Brownfield context) to foreign/parent template, plus creation of site’s technical hardware configuration in line with foreign/parent strategic demands as well as specific local/ host strategic conditions Outcome: Between imitation and customization

Work organization/ HR profile

Transfer: Transfer of foreign/parent template Fit/Misfit: Institutional misfit between foreign/parent template’s work concept and local HR profile of existing Brownfield template Mode of Recontextualization: Adaptation of foreign/parent template to local/host context; increasing adaptation of local site/template to foreign/parent template Outcome: Hybridization

Industrial relations

Transfer: Foreign/parent demands pitted against locally existing template Fit/Misfit: Institutional misfit between foreign/parent demands and local/host industrial relations template Mode of Recontextualization: Adaptation of foreign/parent demands (i.e. withdrawal of foreign parent demands for alternative); over time more adaptation of local context to foreign/parent demands in line with host content Outcome: From traditional to modern localization

Supplier relations

Transfer: Foreign/parent and local/host context demands and conditions to be met Fit/Misfit: No misfit between foreign/parent strategic and host institutional demands; strategic misfit between the foreign/ template requirements and the local/host strategic context Mode of Recontextualization: Little adaptation of foreign/parent demands, little adaptation of host context demands; adaptation of local/host supplier relations to meet foreign/parent and local/host context demands and conditions Outcome: Between customization and imitation

171

Firm > PS dimension

172 Hybridization of MNE Subsidiaries

transfer its foreign/parent supplier template and is not willing to use the local/host template of its Brownfield operation, it customizes its supplier relations. It customizes them by building supplier relations that respond simultaneously to the foreign/parent and local/host demands and conditions. Specifically, the supplier relations respond to low demand market conditions in India, they observe host institutional demands for the localization of production and they observe the foreign/parent demands comprising quality assurance, taking advantage of international cost differentials, transportation cost reduction and drawing on existing supplier industries in emerging economies. To build its supplier relations, FI largely draws on the locally existing supplier infrastructure. However, this does not mean that Fiat simply uses its local JV partner’s supplier relations as a template. Among other things, this is ruled out by Fiat’s the equity mode, product ownership of the World Car and the related strategic distance with regard to the task profile and technology level between the World Car and PAL’s main model, the old Padmini. Instead, Fiat selects some suppliers and develops intense contractual relations with them. Clearly, this requires some adaptation of the local/host supplier relations to meet Fiat’s demands. As a result of selective template transfer and responses to Fiat’s supplier policy, FI comes in certain respects to structural and contractual solutions that resemble the supplier relations of other World Car sites. In that sense, there is some degree of imitation. However, for the most part FI’s supplier relations are customized solutions, responding to both foreign/parent and local/host demands and conditions. To conclude, Fiat’s supplier relations can probably be best described as somewhere between customization and imitation (see Table 6.2 for a summary of FI’s hybridization profile).

7 The Case of Mercedes-Benz India

Introduction While the internationalization of Daimler’s commercial vehicle production starts already in the 1900s, the internationalization of passenger car production, starts late compared to other auto manufacturers. At the beginning of the 1990s, Daimler-Benz (DB) has practically no internationalization experience in passenger car production (GrunowOsswald, 2006; Pries, 2000). This changes, however, in the 1990s. The most noteworthy steps of internationalization are the mergers/takeovers of Chrysler in 1998 and Mitsubishi in 2001. Proving unsuccessful, the two mergers/takeovers are dissolved in 2007 and 2004 respectively. Apart from its international mergers, DB initiates organic internationalization efforts with its Mercedes-Benz (MB) brand. In the United States DB sets up an integrated plant for Sports Utility Vehicles (M-class) in Tuscaloosa/USA in 1997. At the same time, the company targets newly emerging markets and establishes a number of Semi Knocked Down (SKD)/Completely Knocked Down (CKD) assembly sites throughout the 1990s. In 1994 Daimler’s Indian site is established and belongs, together with the Indonesian and the South African operation, to DB’s first passenger car production sites aboard. In 2003, the India site (assembling the S, E and C-class) shares its basic task profile and strategic role as a SKD/CKD assembly site with similar sites in Indonesia (assembling the S, E and C-class), Malaysia (assembling the S, E and C-class), Thailand (assembling the E and C-class), Vietnam (assembling the E-class), the Philippians (assembling the E-class). Daimler, moreover, establishes a new SKD/CKD operation in China in the 2000s for the assembly of the S-, E- and C-class (Petri, 2004). While Mercedes-Benz India (MBI) is, in terms of production volume and variety, strategically very distant from 173

174 Hybridization of MNE Subsidiaries

the main plant Sindelfingen in Germany (where mainly the E and S class are built), there is much less of a strategic distance to other SKD/CKD operations worldwide.

Global product strategy Drawing on Porter’s (1980) concept of generic strategies, DaimlerChrysler’s (DC) global product strategy in the early 2000s can be best described a differentiation strategy. Daimler and particularly its brand Mercedes-Benz are focusing on the premium market segments with an increasing product variation in these premium segments. Especially in the 1990s, Daimler-Benz’s (DB) and then DC’s product strategy shifts from a narrow premium-niche market strategy to a more differentiated premium-niche market strategy. Now, as far as Daimler’s emerging market SKD/CKD sites for the Mercedes-Benz brand are concerned, the company allocates only a limited range of its full model portfolio to the respective assembly sites. The models produced include a narrower up-market segment range and reflect, at the same time, the state-of-the-art of the company’s product offerings. We can, therefore, also call the product strategy applied to MB’s emerging market sites a differentiation focus.

Establishment mode and equity development In 1994, Daimler Benz (DB) establishes the Greenfield JV Mercedes-Benz India Limited (MBI) with Tata Engineering and Locomotive Company Limited (TELCO) for the CKD production of the Mercedes-Benz E-class in Pune. DB’s selection of a local JV partner is based on the following grounds. First, DB has a long standing relationship with TELCO. Already in 1954, DB signs a technical collaboration for the production of trucks with TELCO. While this collaboration ends in 1969, DB continues to be shareholder (10 per cent) in TELCO. When DB sees new opportunities in India’s passenger car market, following the 1991 market reforms, a JV with the old partner is an obvious starting point to enter the Indian market. Moreover, choosing a market entry within a JV framework is, at the time, not only an option but linked to incentives, namely a fast track approval process for all JVs in which foreign equity does not exceed 51 per cent. Against this background, MBI features an initial equity situation in which a 51 per cent share is held by DB and the remaining 49 per cent by TELCO. Along with the sharing of equity, there is also a task sharing between the JV partners. The German partner is responsible for getting production up and running and the Indian partner has to handle

Empirical Results 175

administrative matters, including personnel administration. TELCO is asked to bring in its market knowledge and distribution network to sell the vehicle. In line with high market expectations – 10,000 vehicles domestic sales and 10,000 for export – a production capacity is installed for 20,000 vehicles. DB’s market entry mainly aims at servicing the Indian market with the Mercedes brand and getting a foothold in one of Asia’s largest countries. In this context India is considered as a strategic market for the future (MD, 1998). India’s potential for global sourcing of labour intensive material-groups is also part of DB’s entry consideration. However, things don’t work out as expected for a number of reasons. The Indian market potential for luxury vehicles is overestimated. The company also makes the mistake to enter the market with an outdated model which causes resentment among potential Indian buyers who are well aware of MB’s latest models. Moreover, TELCO with its sellersand truck-market background proves too inexperienced in marketing a high-segment luxury-brand. TELCO is also unwilling to keep pace with Daimler-Benz’s investments into the site (MD, 1998). The combination of these problems triggers not only a downscaling of the initial production set-up but also an equity and control shift in favour of the German partner. In 1997, equity shifts from 51:49 per cent to 76:24 per cent. In 2001, Mercedes-Benz India becomes finally a 100 per cent subsidiary of DC. In the same year, the company changes its name to DaimlerChrysler India Private Ltd. However, with the breakup of the merger between Daimler and Chrysler the Indian’s operations name changes once again. In 2007, the company is renamed to Mercedes-Benz India.

Location, production programme and market share MBI’s factory is located in Pune (State of Maharashtra) one of the core industrial locations for automobile production in India. MBI’s production starts initially in a factory hall on the huge factory compound of the JV partner TELCO in Pune. Along with the final equity shift, MBI also shifts its operations outside the TELCO -premises into a new building. In the new location the manufacturing activity takes place in a newly built factory hall, which is built by and leased from former JV partner TELCO. The new factory is still very close to the TELCO factory compound and remains a CKD operation. The new factory features, however, a lower installed capacity than the initial one. MBI starts production in March 1995 with the W124, an E-class model, which is at the time not the latest model in its class. When MBI is visited in 2002, it has advanced to producing a selected range of the MB’s latest models (C-class, E-class and S-class) in a limited number of variants

176 Hybridization of MNE Subsidiaries

(diesel/petrol, automatic/manual transmission, classic/ elegance interior trim). The output of the E- and C-class adds up to three to four cars per day, the S-class output is no more than three to six vehicles per month. While MBI’s possible plant capacity is at 9000 vehicles per annum, its actual production capacity is only 1500 cars per annum, which is realized with around 320 employees in 2002. Segment-wise, all MBI-models are categorized as belonging to the D- and E-segment (i.e. premium- or luxury segment) in the Indian market. Throughout the 1990s and the 2000s MBI sales in India are modest (e.g. 810 vehicles in 2000 and 1,335 in 2001) (DaimlerChrysler 2004b) remaining well below initial expectations. From the inception in 1994 until the time of research sales figures grow only slowly to around 2000 vehicles per annum. While MBI’s total market share in the Indian automobile market is negligible (0.22 per cent in 2004–2005), it has in 2003 a market shares of 14 per cent and 100 per cent in the premium and luxury segments respectively (SIAM, 2006).

Analysis of the hybridization profile The global transfer scenario MBI’s production system is neither modeled on a specific home-plant blueprint nor on any other defined template. Although DC is in the process of developing such a template for SKD/CKD sites and also draws on MBI’s experience to develop it, no such template is available at the time of MBI’s set up (MBI 3). Besides, the transfer of an all-encompassing template could have been hampered by the fact that MBI starts its Indian operation within the format of a JV involving divided responsibilities between the JV partners. As part of the initial JV deal, a number of human resource related responsibilities, including initial recruitment, staffing, and basic organization building, are given to the Indian partner. Indian managers, who are transferred from TELCO to the JV, draw selectively TELCO’s organizational set-up. Similarly, German managers, who are mainly responsible to establish the production process, draw on their home plant experiences in the set up of MBI. After all, the products manufactured at the Indian site are the same and have to meet the same specifications than those produced at the home operations. For German managers the site set-up mainly involves the planning and the supply of documentation of technical aspects of the production-system. This includes documentation on vehicle assembly, production layout, the transfer of technical equipment and quality systems and procedures. The German head of the factory states in this

Empirical Results 177

context that technical equipment and the technical process of building a car are clearly derived from home operations in Germany (MBI 1). Interestingly, the set-up of MBI’s work organization and human resource profile is not defined or planned in any way up front (MBI 1). Instead, their realization is left to the experience of expatriate managers in cooperation with their Indian partners. However, to state that German managers have no defined template for the transfer of ‘soft’ contents, is not to say that they have no demands with vis-à-vis MBI’s work organization and human resource profile. After all, MBI has a slight majority in the JV, provides the product and has a strong expatriate presence particularly in the early years of the operation. These expatriates have more or less explicit concepts in their mind of how things ought to work. Specifically, demands for change and transfer intentions emerge and become visible as German managers see behavioural patterns of Indian employees as inappropriate or problematic, based on their own performance standards and goals. Finally, in 2002 MBI takes the initiative to introduce as the first CKD plant the Mercedes-Benz Production System (MBPS). However, as MBPS is designed for the production systems of integrated plants, it is in many respects completely unsuitable for MBI’s task environment. Being aware of these limitations, the MBI’s management is free to implement the system, where it is deemed possible, and to alter it, where an adaptation to the local/host task environment of a CKD plant is required. In this process, MBPS does not change the local production systems as much as the local production systems changes MBPS. After all, this is not an unintended consequence, but part of the whole exercise as the reformulated MBPS is fed into a blueprint for other CKD plants. However, just as in the case of the initial transfer of technical documentation, the head of production (MBI 1) stresses that MBPS is a system defining technical solutions and benchmarks (technical set-ups, technical equipment, tools and standards) rather than soft management practices.

The functional differentiation of the organization structure Transfer scenario MBI’s production system’s functional differentiation is not targeted by the transfer of a foreign/parent template. At the same time, there is also no mentioning of a local/host template being used to configure the Greenfield site’s functional differentiation. The strategic distance, in the task profile and demand market conditions between the foreign/parent strategic context and the local/host strategic context, explain why there

178 Hybridization of MNE Subsidiaries

is no effort to replicate one of MB’s home plant functional differentiation at MBI. Moreover, at the time of MBI’s establishment no foreign/parent template is available for the set-up of SKD/CKD sites. Like the nontransfer of a home plant template, the non-use of a local template from the Indian JV partner is also explained in part by the strategic distance between the task profile and capabilities of an Indian large volume truck manufacturer and the task profile of MBI involving the production of a premium product in small numbers. The use of a local/host template is ruled also out by the fact that the German side introduces the vehicle and by the fact that the German side has the mandate to establish the functional structure of production system. However, while the MBI’s functional set-up is not reflective of a particular foreign/parent or local/host template, the site has to respond to foreign/parent and host context demands and conditions.

Misfit As neither a foreign/parent nor local/host templates is transferred or used for the Greenfield site, there is also no misfit between a template and a local/host or foreign/parent context. There is also no indication that local/host demands and conditions and foreign/parent demands are contradictory. Yet, there are different strategic and institutional demands and conditions that call for integration in the newly established functional differentiation. The high presence of German expatriates in the first years, their configuration mandate and their conviction that certain functions are a must, no matter if 300 or 300,000 vehicles are built, suggests the presence of a range of strategic foreign/parent demands to create functions in line with foreign/parent company standards (MBI 1, 3). After all, in terms of quality-requirements MBI’s products are not strategically distant from the vehicles built at the German home sites. There are, therefore, clear demands to set up functions, such as quality audits and vehicle testing. At the same time, the site has to respond to a host strategic context; characterized by extremely low demand for up-market vehicles and a corresponding low volume task profile (MBI 3). And finally, the site has to respond to host institutional demands. The most important of which are the India’s local content requirements.

Mode of recontextualization Some adaptation of foreign/parent demands to the local/host context Since there is no transfer of a foreign/parent template there is also no adaptation of it. There are, however, foreign/parent demands

Empirical Results 179

to which the site responds. These demands vis-à-vis the site’s functional differentiation are adapted in the sense that they are aligned with the sites specific task environment of low volume production involving, for example, substantially less vertical integration than foreign/parent home plants. MBI is only an assembly site, which implies that certain functions/departments (machine shop departments) and certain sizes of functions/departments are not required.

Some adaptation of local/host context to foreign/parent demands There are few indications that the host context is substantially adapted to meet foreign/parent demands. Nevertheless, the site is a Greenfield context and has to be functionally configured or created. This creation or adaptation of the local context happens in line with different contextual demands. Specifically, the site’s functional set-up is customized to meet simultaneously foreign/parent demands, host strategic conditions and host institutional demands. For example, the quality function comprises the departments ‘Vehicle in-house’, ‘Supplier parts & Quality audit’. This set-up reflects both foreign/parent demands for comprehensive quality audits and the site specific situation involving system suppliers that needed to be closely monitored in quality matters. Moreover, MBI’s functional differentiation is tailored to the specific host country strategic context conditions in volume and product variety terms. In this respect, the operation’s production system is customized to a task environment that differs very much from parent operations with respect to vertical integration (low), supply logistics (long distance) as well as in terms of product variants (much less) and volumes (much lower) (MBI 3). The main production departments are tailored to the specific assembly activities performed by the Indian operation and comprise: ‘Body shop & Equipment maintenance’, ‘Trim & Mechanical Line’, ‘Finish Line & Paint Shop’. In addition, the small operation has to take care of a range of functions that are not directly linked to production (e.g. ‘Marketing Services, Marketing & Sales’, and ‘After Sales’). In a home manufacturing operation these would not be part of the overall task profile. However, MBI’s functional differentiation not only reflects foreign/ parent demands and the market demand conditions in the host strategic context (MBI 5). A look at the organization chart unveiled the presence of functions that are the direct result of host institutional demands (i.e. for local value addition). Cases in point are the ‘Customs affairs’, the ‘Warehouse/Container station’, and the ‘Local content/Direct purchases’ departments, which are a reflection of the specific

180 Hybridization of MNE Subsidiaries

institutional demands and low volume strategic conditions in the host context. We can, therefore, conclude the dominant (re)contextualization mode regarding MBI’s functional differentiation is a creation in line with different demands and conditions from the host and foreign/parent context. Low levels of misfit and a high resilience of the site’s Greenfield context provide an enabling context for such a (re)contextualization mode.

Outcome While German expatriates have home operations as their functional frame of reference and seek to implement company policies which require the establishment of certain functions, there is no indication that any sort of full or deliberate template transfer is sought. The host strategic conditions are simply too different from those at home. Moreover, there is no foreign/parent template defined for MBI’s specific strategic and institutional conditions. At the same time, the increasing control of the JV by the German side, the vehicle introduction by the German side, the Greenfield nature of the operation and lastly the strategic distance between the local JV partner’s task profile and MBI’s task profile explain, why no local template is used. As no foreign/parent or local/host template is used, the site’s functional differentiation is tailored to foreign/parent demands under specific host strategic conditions and institutional demands. Thus, the overall outcome is the result of an absent local and parent template transfer/use, the specific foreign/parent context demands in the form of policies and expatriate translations of home functional concepts as well as host strategic context condition and institutional demands that together translate into a functional differentiation that can be best described as customized or tailored.

The hierarchical differentiation of the organization structure Transfer scenario When MBI is founded in 1994, there is no template transfer effort from the foreign/parent side targeting the hierarchical differentiation of MBI. This has to do with two factors. First, MB has no defined template for SKD/CKD sites. Second, resulting from the equity mode and the related division of responsibilities in the JV, the German side sees its mandate largely restricted to the establishment of the technical and functional dimensions of the production system, that is, those aspects that are directly related to the manufacturing of the product. Other matters, such as personnel administration and human resource management, are left

Empirical Results 181

in the hands of the Indian JV partner. As many of the Indian managers have worked for TELCO before, they draw on TELCO’s existing hierarchy as a template and derive a downscaled version for the new and smaller MBI site (MBI 2). TELCO’s template counts 27 levels of hierarchy, including 18 managerial designations and nine levels of blue and white collar worker designations. The initial hierarchical set-up of MBI is based on a local template which is related to the company’s entry mode: This has to do with history. We started with a Joint Venture with TELCO and at that time there were 18 levels [managerial] and the whole system was taken and transferred to MB India because TELCO had the responsibility in administration and HR. And having put people into 18 different levels, we had to collapse it into less, without taking away the perceived work concept, if I could call it like that. So somehow, we arrived at ten. And even within the ten, the reporting of all employees from ten to five is to level four. So reporting levels, there are only five, not even five, level one we don’t have any more. XY was the last level one. Now it is level two, Executive/MD; level three, General Manager; level four, Divisional Manager; and level five is everybody else. Four levels only. (MBI 2) The absent transfer of a foreign/parent template to MBI is not connected to any particular strategic or institutional misfit. Rather it is caused by the equity situation that implies that the set-up of the hierarchical differentiation is organized by the local parent and its delegates. Instead of a foreign/parent template transfer, a local parent template is used. While MBI’s hierarchical differentiation is derived from TELCO, it is from the outset clear that, no one-to-one transfer of the TELCO template is appropriate. Given MBI’s size of operation, an unaltered transfer of the TELCO’s hierarchical differentiation is not considered (MBI 2).

Misfit The downscaled version of TELCO’s template poses some institutional misfit with foreign/parent demands (MD interviewed 1998; MBI 3). German managers still find the hierarchical differentiation much too steep for an operation of MBI’s size. However, it should also be mentioned that not all German managers find MBI’s hierarchical differentiation, as inherited from TELCO, problematic. Germany managers do not unanimously share the opinion that designations should be radically reduced. While some see excessive designations rendering the

182 Hybridization of MNE Subsidiaries

organization inefficient (MBI 3), others, particularly the head of production, sees no problem in having more hierarchical designations than actual reporting levels (MBI 1). Nevertheless, as the equity shifts, demands by German expatriates to bring the hierarchy more in line with foreign/parent company standards increase and receive more weight. The foreign/parent pressure for recontextualization of the local template, in turn, meets resistance from host country managers.

Mode of recontextualization Minor adaptation of foreign/parent demands to the local/host context While foreign/parent demands for a reduced hierarchy increases over time and receives more weight with the shift in equity, the local template never fully develops towards foreign/parent or home country configurations. The reason for this is that there is no clearly defined foreign/parent template or transfer intent in the first place. At the same time, there is resistance from the local institutional context (MSI 5). While the reporting levels ultimately come close to foreign/parent demands and standards, the many remaining hierarchical designations are still reminiscent of the original TELCO template and a tribute to the local institutional pressures for hierarchical differentiation. Particularly, in those areas where some clubbing of designations takes place the company witnesses opposition from Indian employees. They see these measures as a breach with the seniority principle and as a disrespect of their experience. There is strong evidence to suggest that the company cannot and does not want to ignore the local institutional demands for status-giving designations entirely. After all, there are still many more hierarchical designations than reporting levels in MBI. An additional reason why these designations are held on to, is related to the company’s lacking growth. As no new job profiles can be offered to the largely young and ambitious employees, the very least the company can do, is to compensate for lacking career opportunities by offering symbolic promotions in the form of better designations (MBI 2). Finally, the human resource function is, at all times, in the hands of host country nationals. The Indian human resource managers have a strong say in the company’s hierarchical differentiation and cushion some foreign/parent demands despite the changing equity situation.

Adaptation of the local/host context to foreign/parent demands

MBI’s hierarchical differentiation sees a substantial change over time. While the site reflects in the beginning a scaled down version of a local/host template this template is adapted to foreign/parent demands over time.

Empirical Results 183

Among the most important changes are a reduction of the hierarchical designations and a decoupling of the remaining hierarchical designation with reporting levels. The adaptations of the local template are made in response to foreign/parent demands. They are made possible by the shift in equity and the increased control of MBI by the German side. However, while MBI succeeds in reducing the number of hierarchical designations, the company still has more hierarchical designations than reporting levels. It appears that no further effort is made to bring the Indian hierarchical differentiation fully in line with foreign/parent institutional context. Given the continued top management position of a host country national in human resource management and given the local opposition to a complete change as well as the economic situation of the site, the local resilience for change is limited.

Outcome MBI’s hierarchical differentiation starts with a derivative of a local/host template. As equity shifts, the institutional misfit between the local template and foreign/parent demands leads to a reduction of hierarchical designations and a decoupling of reporting levels from hierarchical designations. When MBI is researched in 2002, it has, in the hierarchical order, the following designations: Managing Director, General Manager, Divisional Managers, Senior Managers, Managers, Supervisors/ Executives, Assistants and Operators. These hierarchical designations are not hierarchical levels in the sense of reporting levels. Reporting levels, there are only four. The first reporting level is the Executive Director (Vorstand/E2), the second General Managers (Abteilungsleiter/E3), the third Divisional Manager (Leiter/E4), the forth level reporting level includes everybody from Supervisor/Executive to Senior Manager (see Table 7.1 showing the change to local template). Although the German side achieves some reduction of the hierarchy by reducing reporting levels to four, pushing for an imitation of German standards, there remain much more hierarchical designations than reporting levels – reflecting the local template and local/host institutional demands. Similar to the MSI case, it remains unclear, whether the hierarchical designations have an effect on the company’s hierarchical relations. At the very least, these decoupled structures coexist side by side. While the hierarchical designations are reflective of the Indian institutional scenario, the number of reporting levels corresponded with MB’s categorization of hierarchical levels. The German head of production underlines that while the reporting levels are deviating from

Table 7.1

MBI’s hierarchical differentiation MBI India 1994/5

Oct 98

2002

Executive Director EG1a Senior General Manager EG1b General Manager EG1c Deputy General Manager EG2a Assistant General Manager EG2b Senior Divisional Manager EG3 Divisional Manager EG4 Senior Manager TM6 Manager TM5 Deputy Manager TM4 Assistant Manager TM3 Senior Officer TM2 Officer TM1 Junior Officer Supervisor 4 Supervisor 3 Supervisor 2 Supervisor 1 White collar worker A White collar worker B White collar worker C Operator 1 Operator 2 Operator 3 Operator 4 Operator 5 Operator 6

Managing Director Vice President

L1 Managing Director L2 Vice President

– L2 Executive Directors & Managing Director

Executive Manager

L3 General Manager

L3 General Manager

Divisional Manager

L4 Divisional Manager

L4 Divisional Manager

– Team Manager

– L5 Senior Manager

– L5, Senior Managers L6, L7, Managers L8, L9, Executives/Supervisor L10 Assistants

L6, L7 Managers

Supervisor 1-4 L8, L9 Executives/Supervisor L10 Assistants White collar worker

Operators

Operators

Operators

184

Telco designations 1994

Empirical Results 185

the foreign/parent company, they are very well in line with a mid-sized company in the home country context (MBI 1). In summary: A local template is used that does not fit with the foreign/ parent (strategic and institutional) demands for a flatter hierarchy. The misfit is overcome by adapting the local template to the foreign parent/ demands. However, the foreign/parent demands remain half hearted due to local counter pressures and an absence of a clearly defined template for the Indian site. The result is a hybrid solution, wherein we find a side-by-side of foreign/parent elements (reporting structure) and local partner/host elements (designations).

Technical configuration of the production process Transfer scenario When MBI’s production site is established in India there is no template serving as a model to structure the factory layout, process design and the technical hardware configuration of the site. The use of home plants as a template is ruled out by the strategic distance between the foreign/parent and local/host strategic context. Specifically, the strategic distance in supply and demand market conditions rules out the transfer of the factory layout, process design and technical hardware configuration of any existing integrated home plant in Germany. This becomes evident if we consider that the demand scenario in the Indian market allows little more than a production of 1500 vehicles per annum. Such a production volume is about the daily output of MB’s home plant in Sindelfingen. Apart from the strategic distance in volumes, MBI’s task profile also contrasts with the home plant’s vehicle variety. With its choice of producing the C, E, and S class, the Indian site’s task profile features a much higher base-model variety in a much lower number of variants. Next to the strategic misfit in the demand conditions, a supply-side strategic misfit rules out the transfer of the home plant technical hardware configurations. For MBI’s technical hardware configuration also has to take advantage of low labour costs in India. We don’t use robots here because of the low number of units. The mechanization of a work place costs us including a robot roughly 200,000 German Marks [100,000 Euro]. Given our labour cost here – an employee earns here about 6500 Rupees per Month – that just doesn’t pay off. (General Manager Production, 1998)

186 Hybridization of MNE Subsidiaries

While the transfer of any home plant’s technical configuration of the production process is ruled out by strategic distance, there is also no transfer of other SKD/CKD sites’ templates. Interestingly, it is not before the 2000s that efforts are made to develop such a template for CKD sites (MBI 3). In this context, MBPS is redefined for a possible use at CKD sites. Yet, when MBI is established neither its factory layout, process design nor its technical hardware configuration is based on such a template (MBI 3). Conversely, despite having entered the Indian market within the format of a JV, no effort is made to draw on a template from the local partner. Such a local/host template use is ruled out by the fact that the vehicles are introduced by the German partner who holds the majority in the JV and has the mandate to establish the site’s production process. Also the strategic distance of the local partner’s task profile as a truck manufacturer stands against such a local template use. While there is no specific template to be transferred, MBI’s technical configuration of the production process has to meet foreign/parent and local/host demands and conditions.

Misfit The set up of MBI’s process design faces a substantial misfit between local/host institutional demands and foreign/parent strategic demands. On the one hand, MBI’s factory layout, process design and technical hardware configuration has to respond to foreign/parent strategic demands with regard to a specific product mix, cost and quality. At the same time, the site has to respond to the specific local/host strategic conditions, including low volumes based on host demand conditions and low labour cost based on host supply conditions. On the other hand, MBI’s factory layout, process design and technical hardware configuration has to respond to host institutional demands, notably India’s FDI policies. Like other foreign company’s, MBI has to observe foreign exchange neutrality, import tariff structures and local content requirements. However, in contrast to other companies (e.g. Fiat or MarutiSuzuki) that have higher volumes and operate in price sensitive market segments, MBI has no intrinsic motivation to localize production. As its customers are not as price sensitive and demand is very low, there is little incentive for MBI to shift major process steps to India.1 For low volumes render the set-up of capital intensive up-stream manufacturing steps (to achieve the required local content rate) very uneconomical. Thus, there is a misfit between the Indian institutional context’s demand for local content and the foreign parent’s strategic demand for profitable manufacturing under low volume/low demand strategic context conditions.

Empirical Results 187

Mode of recontextualization Adaptation of foreign/parent demands to the local/host context

Given the strategic distance between home and host country conditions and given that there is no defined template for CKD sites, there is also no foreign transfer template to be adapted to the local/host context. Instead, there are foreign/parent and local host demands. These foreign/parent demands vis-à-vis the site’s technical configuration are partly in contradiction with local/host demands. Yet, indications are that the technical configuration of the production process largely complies with the foreign/parent demands involving an adaptation of host context demands (MBI 1).

Substantial adaptation of local/host demands to the foreign/parent context Given that the product is introduced by the German partner, given that the German side holds a majority in the JV and given the Greenfield nature of the set-up, there is no local/host template use and adaptation to foreign/parent demands. Instead, MBI’s factory layout, process design and technical hardware configuration is created to meet foreign/ parent and local/host demands and conditions. The partial misfit that exists between foreign/parent and local/host demands, is mainly dissolved by adapting the host institutional demands to the foreign/parent demands. The factory layout and process design is built to meet the quality and assembly requirements of a selected range of models which are similar to those manufactured in the German parent plants. While basic quality processes and assembly steps are in principle the same, they have to be customized to the low volume production (MBI 3). This customization involves different aspects. First, MBI’s production process features a comparatively low vertical integration because capital-intensive process steps (e.g. stamping, pressing, painting) are performed in Germany or off-site. Low volumes, in a low labour cost context, suggest low levels of vertical integration and a localization of labour-intensive assembly activities. Second, although the remaining assembly activities are in principle the same as in the home plants, low volumes call for a revised clubbing of the assembly tasks. For example, unlike in Germany, gluing activities are performed at one station only (MBI 10). Similarly, while the testing and programming equipment is the same as in Germany, it diverges in how it is applied. In Sindelfingen, testing is done separately in various line-sections. In India by contrast, it is done only once at the end. In contrast to the German home plants, the factory layout and process design is characterized by much fewer stations performing much more

188 Hybridization of MNE Subsidiaries

operations which translate into much higher cycle times of around 60 to 70 minutes. These contrasts sharply with German operations that feature cycle times of little more than a minute. Third, internal logistics are customized to low volumes and long distance supplies. For extensive storage requirements for bulk shipments of CKD-kits render the realization of home-plant Kanban supply-logistics meaningless. Fourth, MBI designs its operation to meet local content requirements. Apart from local sourcing of labour-intensive material groups (e.g. horns) for worldwide production, MBI sources out the assembly of knocked-down components to local module suppliers. While these components are also delivered in kits from Germany, they are not assembled on site. Instead, they are assembled by local suppliers, thereby counting as local content. This is in essence an adaptation or rejection of host institutional demands. It implies an even lower vertical integration in MBI, compared to other CKD sites. As a result, the host institutional demand for more local content does not lead to more local value addition, but rather to some tactical outsourcing, which increases the formal, but not the actual rate of local content. Also with regard to the site’s technical hardware configuration there are customized responses to the foreign/parent demands and local/host context conditions. Low volumes and low labour costs, rule out the transfer of capital-intensive process steps and technology such as: transfer presses, paint shop technology or automated welding technology. Though the technical hardware configuration receives particularly initially technical equipment from Germany, such as welding guns, jigs, fixtures and testing equipments, there is no replication of home plant mechanization and automation levels. What is more, the technical equipment used, is increasingly developed on site (MBI 8, 9). MBI’s technical hardware configuration is based on a ‘low capital intensity strategy’ translating into low mechanization and automation levels. The most sophisticated equipment is located in the areas of final testing and software-upload. The highest level of mechanization is in the weld-shop, consisting of a few stations equipped with manual welding-guns. This contrasts with high automation levels in the home plants’ body shops. Likewise, final assembly in Germany has higher levels of mechanization and automation (e.g. windscreen insertion) (MBI 10). Even vehicle testing is more automated in Germany and more manually based in the Indian set-up. To conclude, the main (re-)contextualization mode found in this area is a configuration in line with foreign/parent and local/host demands and conditions. While the local content requirements (as MBI can

Empirical Results 189

circumvent them) posed by the host institutional context as well as the local strategic context of the Greenfield site prove resilient in the set up of the technical configuration, the local demand conditions prove to be a not very resilient host strategic context.

Outcome The outcome of MBI’s technical configuration of the production process can be best described as a customization. This customization is constituted by an absence of a template transfer/use and specific responses to foreign/parent and local/host strategic and institutional demands and conditions, instead. The factory layout and process organization respond to the strategic context of a high base-model mix at low volumes. It translates into a low vertically integrated production process. Specifically, the factory layout consists of two parallel lines: one short line for the S-class and a longer mixed line for the C- and E-class. While the S-class has a very short line, purely consisting of assembly activities, the C- and E-class line consists of more process steps and stations. Although the home plants serve as a broad frame of reference to sequence the production process, the low volumes in India imply different and higher work contents at much fewer manufacturing stations. What is more, the low vertical integration of the production process, which is even lower than at other CKD sites, is connected to local content requirements in the Indian context. For example, to boost its local content rate and to keep investments into volumesensitive production steps down, MBI sources out painting to TELCO. As a result, the production line for the E and C class is divided into two subsections with a directly opposite product flow. The production starts with the welding of most body parts from CKD-kits. It continues with the transfer of the welded car body to TELCO for painting. When the painted bodies return from TELCO they are handed over to final assembly and checking activities. (For the S-class, the process is different as no local welding and painting takes place. High profit margin in the S-class and the low volumes make such low local value addition despite punishing import tariffs a viable option). Similarly, the absence of automated technology and the increasing use of local technical solutions are a response to low labour costs and rising capabilities in the local/host strategic context. In summary, the sites technical configuration of the production process is a customized solution to integrate foreign/parent demands related to the production of particular range of models within a particular host strategic and institutional context, which is defined by sluggish demand

190 Hybridization of MNE Subsidiaries

for luxury vehicles, low labour costs and the Indian Government’s indigenization requirements.

Work organization and human resource profile Transfer scenario According to the German production management there is no defined template to structure the work organization or the human-resource profile of the subsidiary’s production system. The transfer of the home plant work organization and human resource profile is ruled out by strategic and institutional distance between the foreign/parent and the local/host context. On the one hand, there is a misfit between what home plant work concepts are optimized for in terms of supply and demand market conditions and what the India’s strategic context has to offer in this respect. Specifically, one of the most important reasons for the transfer restraint is that the group-work as practised at home is optimized for a task profile marked by short cycle times in mass production and a minimization of labour costs in a high wage strategic context. The work profile of groups in Germany involves highly repetitive work content, marginally ‘enriched’ by job-rotation and self-certification (MBI 3, 10). The groups are conceptualized as semi-autonomous, self-directed working units. They featured, self-training responsibilities, self-directed rotation and self-directed group meetings, chaired by group-speakers. The human resource profile of these groups is increasingly based on a semi-skilled workforce. In this work context, the traditional role of the ‘Meister’, a highly skilled craftsman, is changing to a more distanced monitoring and coaching role, involving less direct involvement. Now, German management underlines that it has no intention to transfer this group concept (German General Manager Production 1998; MBI 3). It is stresses that the ‘the American model’ – which they see now applied in Germany – is not applied to the Indian site because of the nature of the production process (MBI 1, 3). In their view, MBI’s task profile is more akin to workshop production and requires a more qualified and skilled workforce compared to a strategic context marked by high volumes, extreme division of labour and short cycled times. At the same time, drawing on high qualifications does not cause a cost problem in India, given the labour supply conditions in the host strategic context. After all, the Indian strategic context provides high industrial qualifications at comparatively low costs. The following citation underlines that the German expatriates do not seek the transfer of home group work concepts and corresponding human

Empirical Results 191

resource profiles because of the strategic distance between home and subsidiary site operations: A: In Germany the main goal of group work was, let’s say, to increase the average utilization of the workers on the line. That is, to reduce the cycle segment compensation (Taktausgleich) and to thereby reduce the costs for the company. You only have a very small work content, two to one and a half minutes, and if you have one more worker rotating the cycle segment compensation (Taktausgleich) is less. Q: Then the work also becomes more interesting? A: That is how we have made it palatable to the employee. But in reality we have increased his average utilization through this enrichment and rotation from 94 per cent to 98 per cent. The cycle segment compensation could thereby be dropped. We could tact (eintakten) the employees differently and sold it as enrichment, sure. And this system doesn’t work here. I have here a work content of an hour and I cannot do group work here. I would have to put four men in a group and they would have four hours of work content and which would cause quality problems, in turn. No, the group is here rather a unit under the supervisor. (MBI 3) Another manager interviewed in 1998 (German General Manager Production, 1998) argues along similar lines, stating that it makes absolutely no sense to rotate workers at the Indian site. With a cycle time of more than an hour, operators have already a more complex work profiles than most of their counterparts in German plants. Job rotation under these circumstances would seriously compromises product quality. For a similar reason, German quality assurance and self certification concepts are not transferred. In MBI, it relies heavily on line-external quality inspectors. Although operators do have to certify their work, the company relies on external inspectors and a high frequency of product audits. Again, the strategic distance, lower volumes and lower labour cost in India, rule out the transfer of German quality practices: Self-certification is generally being practised in Germany but not in India to the same extent. It is important to see why a practice has been introduced in the first place. The first intention was to change the employees’ thinking and the second issue was cost reduction. More important, however, was the cost issue. But in this respect we haven’t got such a great pressure. Cost doesn’t play a role here because if you employ one or two workers it doesn’t play a role due to low

192 Hybridization of MNE Subsidiaries

labour cost. So self-certification here is more a question of awareness creation. (MBI 3) While the strategic distance is a main reason for not transferring the German group concept, the institutional distance also plays a critical role. The pronounced patriarchic-hierarchical orientations in India render the implementation of egalitarian team-work concepts as practised in MB’s home plants a highly difficult, if not impossible, endeavor (MBI 3; German General Manager Production 1998). What is more, not only do German managers refrain from transferring a home plant template, but they generally have little guidance how to configure the work organization and human resource profile in overseas operations. Instead, the establishment of work organization is largely left to the experience of expatriate managers. The German head of the factory states in this context: Processes involving humans, their culture and behaviour are generally not a part of the overall planning. I have seen that here and in Yugoslavia. Those back home are not capable of planning work organization and its prerequisites. This part is left to the expatriates who are sent abroad. (MBI 1) Overall, German managers interviewed appear to have little sense of being on cultural mission and stress the need for local learning and observing host context customs (MBI 1). However, to state that German managers lack comprehensively defined package of work concepts, skills and basic work dispositions for transfer is not to say that they have no demands vis-à-vis the work organization and human resource profile of the site. While German managers deliberately refrain from transferring a number of shopfloor related work concepts from the integrated home-plants and while they have at the time of initial establishment no defined CKD-template to draw on, it would be wrong to suggest that their home-plant background does not serve as a strong conceptual frame of expectation and reference with regard to the developing MBI’s work organization and human resources profile. Based on their background expatriates have more or less explicit concepts in mind how things ought to work. It should be added, that using a local template from the JV partner TELCO is also not an option. The German partner has the majority in the JV, introduces the vehicles and has the mandate to set up the work organization as part of the whole production process. Above all, German

Empirical Results 193

managers have to assure that the final product lives up to the foreign parent’s global quality standards of a luxury brand. Instead of having a grand transfer plan, foreign/parent demands surfaced in the day-today Indo-German interaction. Specifically, apart from a few quality and continuous improvement related formal concepts, German transfer intentions mainly emerge when and where German managers view local behavioural patterns as inappropriate.

Misfit Although hardly any formal concepts of work organization and human resource profiles are transferred from home plants and although there is no defined transfer template for CKD sites, there are still substantial foreign/parent demands. These demands for a recontextualization of the local context emerge because the strategic context of the Indian site is still similar enough not to let go quality standards and because there are institutional misfits between the German expatriate expectations and the local behavioural patterns at the site. Most of the surfacing foreign/parent demands aim at overcoming local institutional conditions that are perceived as problematic in the view of producing a world-class luxury vehicle. It emerges from interviews that the biggest misfit are shopfloor-related and concern basic work dispositions. Specific areas of concern are: the work roles/identities of supervisors and workers, including their work relations; the taking of responsibility and communication practices; and the cleanliness and quality awareness. In the first years of MBI’s establishment, work relations between Indian supervisors and workers are seen by the German production management as one of the most crucial aspects of misfit. For German production managers, supervisor–operator relations and supervisor work roles are of utmost importance in organizing production and ultimately producing a quality product. Supervisors are expected to work cooperatively in a team with operators, to train and coach them hands-on. It also requires them to be always close to the production line and able to step in and personally demonstrate whenever manufacturing problems emerge. Additionally, supervisors are seen as having a bridging function between management and workers. However, there is a substantial discrepancy between German expectations and Indian behavioural patterns. These are related to the strong socio-professional demarcations produced by India’s social stratification and higher education/vocational training system (Becker-Ritterspach, 2005). In contrast to German work concepts and expectations, supervisors see themselves as managerial superiors to operators, defining the work relation as a clear-cut

194 Hybridization of MNE Subsidiaries

superior-subordinate relationship, rather than cooperative teamwork. Supervisors have a ‘white-collar’ work identity and show reluctance to being close to the manufacturing process, let alone to be physically involved in it. Operators, in turn, exhibit low acceptance levels of their supervisors, as these are seen as lacking practical manufacturing competence and skills. As a result, the potential bridging function of supervisors is undermined from both the supervisors’ and the operators’ side. In contrast to the working relationship envisioned by German managers, there is a substantial hierarchical demarcation between supervisors and operators. The following shopfloor display (Figure 7.1) – to be signed by supervisors – is an expression of both German expectations and their precarious fulfillment. However, it is not only the supervisors who reject certain task profiles envisioned by German production management. For German managers, building a world class quality product begins with a clean workplace. The Indian operators, for their part, similarly based on strong socioprofessional demarcation consciousness, reject to get involved in cleaning activities of the work place. For cleaning of factory floors is perceived by operators as unacceptable, clearly not within the purview of their task profile. This leads us to second complex of misfit. Especially in the beginning, lacking cleanliness and quality awareness of the Indian workforce is a big issue. German expatriates see a need for change in this respect. Thus, there is a great mismatch between expatriate expectations and local behavioural realities. Here, supervisor and worker behaviours, which are seen as determining the product quality, are also a matter of concern. Supervisor • QUALITY is my first Priority. • I stand on my line for 95% of the time. • I give help and feedback to my operators. • I take care of quality by coaching my operators. • I coach my operators by showing them work with my own hands. Supervisor Source: copy from a shopfloor billboard (Mercedes-Benz India Private Ltd).

Figure 7.1 MBI’s ‘Clearly defined functions for our supervisors’

Empirical Results 195

A third complex of misfit concerns expectations about taking of responsibility on the shopfloor and open feedback of the Indian employees. A central goal of German management is to define clear areas of responsibility, related work objectives and decision-making rights. A cornerstone of the German leadership- and management-concept is a comprehensive delegation of production-related decisions to where they emerge. On the shopfloor this implies that a team, with a supervisor and a maximum of 15 operators, is responsible for a certain line-section and has to take all decisions directly related to the achievement of daily quality and output targets. However, these seemingly simple modes of organization and management are difficult to implement with Indian managers in general and supervisors as well as operators in particular. Instead of seeing a taking of responsibility for the assigned tasks, German management complain about delays in task accomplishment, widespread responsibility-diffusion and an inability to work in a self-directed manner. It is also bemoaned that Indian employees are not willing or able to speak up openly when and where problems occur. Decisions tend to be permanently re-delegated upward with the effect that the simplest objects-of-decision end up on the German production manager’s desk. Unlike the German concept of work relations between supervisors and operators, Indian employees are not opposed to taking more responsibility. In fact, those interviewed embrace it as an improvement to what they had experienced in previous work relations (MBI 8, 9). At the same time, a number of Indian employees stress that it is difficult to get used to taking of responsibility, as this style is in stark contrast to what they are used to. However, it should also be noted that there are many areas where German expectation and local institutional conditions fit each other well. For example, expatriates have no complaints about lacking skilland discipline-levels of Indian workers. This also has to do with the very site-specific fact that Daimler has a long-standing relationship with TELCO. DB first involvement with TELCO dates back to the 1950s (Grunow-Osswald, 2006). Along with licenses for manufacturing of Daimler-Benz trucks, the German partner also introduced an apprenticeship system at the time. Curiously, more that 30 year later, MBI is able to draw on this heritage. According to the German head of production, MBI is too small to establish its own apprenticeship and training system. Instead, TELCO provides MBI with workers whose skill profile is broadly in line with MBI’s expatriate expectation. In contrast to workers, German managers are rather unhappy about the skill profile of production managers and engineers. In their view, this reflects the theory bias

196 Hybridization of MNE Subsidiaries

of Indian higher education. Professional training of managerial ranks is seen as too remote from the practical realities of production (MBI 1; German General Manager Production, 1998).

Mode of recontextualization Major adaptation of foreign/parent demands to the local/host context Expatriates dispatched from Germany’s home plants lack a comprehensively and formally defined template to shape MBI’s work organization and human resource profile. Related to the strategic distance between the Indian and the home operations, German expatriates either refrain from transferring core home-plant work concepts or adapt them to the local/host strategic conditions. Cases of the former are the non-replication of home plant group-work concepts and qualification-levels. Examples of the latter, are adapted work concepts in the areas of quality assurance and product auditing. Quality inspection in Germany is largely integrated into the production line through self-inspection and random-based external audits. This contrasts, with MBI, where almost every vehicle is externally inspected and goes through a product audit. Thus, to ensure home-country quality-standards, MBI adapts its quality assurance concepts to local/host strategic conditions of low volumes combined with low labour costs. It should also be mentioned that some work concepts are transferred that are not directly concerning manufacturing. These involve mostly small group or continuous improvement activities, such as quality circles. In this regard, there is no indication that these concepts are or have to be adapted to the local/host context. Finally, while German expatriates are, overall, not very eager to transfer formal home plant concepts, they have a number of demands vis-à-vis the basic work dispositions of MBI employees. This is particularly concerns work dispositions that are seen to have a strong impact on product quality. Namely: the supervisor and operator work roles and their work relations, the taking of responsibility as well as the cleanliness and quality awareness. While foreign/parent demands are uncompromising (not very resilient) with regard to the behavioural outcomes, they are utterly adaptive with respect to the work concepts to achieve them. In this latter respect foreign/parent demands are adapted to the local/host institutional context as we will discuss below.

Major adaptation of the local/host context to foreign/parent demands The establishment of MBI’s work organization and human resource profile does not simply reflect a transfer restraint and an adaptation of

Empirical Results 197

foreign/parent concepts and demands to the local/host context. Instead, there is also an adaptation of the local site context to foreign/parent demands. The implementation of these demands involves a substantial adaptation effort of MBI’s work organization and human resource profile. There are two core mechanisms for the realization of German demands. One the one hand, changes of the local institutional conditions and demands are achieved through training and socialization efforts by expatriate transfer and on site training. On the other hand, adaptation of local institutional context to parent demands is achieved through host context selection, that is, by selecting a highly qualified and young workforce and by selectively drawing on host institutional patterns. The corner stone of the establishment of MBI’s work organization and the human resource profile is the transfer of German expatriates in the beginning. When MBI is set up, 20 expatriates are delegated to the Indian site, taking all major organizational positions in production. By the time of the second visit in 2002, this number has come down to three expatriates. However, the managing director and the heads of production are still German. MBI does not engage in a large-scale transfer of host context personnel to the German parent plants. There is just some limited transfer of a selected few Indian managers – high potentials – transferred to adopt the foreign/parent or rather home country company standards. These Indian managers are hand picked and systematically prepared to take over key tasks as German expatriates withdraw. They are expected to apply German standards with the same rigor. Massive expatriate transfer in the first years and their continued presence in core positions, functions as a major vehicle to bring basic work dispositions in line with foreign/parent demands. After all, it is the German side that provides the product, is responsible for getting production up and running and has from the beginning the majority in the JV. Thus, German production management has the mandate, control and takes the initiative to shape things. Expatriates clearly initiate training and socializations measures to create and change basic work dispositions of the Indian work-force and to bring it in line with their own concepts and expectations. For example, together with the human resource department a range of workshops on cleanliness and quality are organized. Similarly, young Indian engineers are sent to work on the production line, to acquire practical experience (German General Manager, 1998). While some of these efforts clearly involve changes to bring basic work dispositions in line with the foreign/parent standards, they do not aim at replicating formal work concepts from the home plants. Instead, German managers seek an

198 Hybridization of MNE Subsidiaries

implementation of basic work dispositions and product-related standards, irrespective of the institutional origin of the conceptual means to achieve them. This implies flexibly in how to achieve the behavioural expectations and goals. There is a repeating pattern of achieving the desired work dispositions by observing local institutional conditions through selectively drawing on alternative host institutional templates. Cases in point are the family model and a number of measures that are linked to raising levels of cleanliness and quality awareness. The ‘family model’ is, for example, a team-concept for the shopfloor. Every production line comprises a number of teams called families. Like in a typical Indian family, there is an ‘eldest son’ (selected by the ‘father’) who acts as the ‘deputy’ in the ‘father’s’ absence. The supervisor or the ‘father’ has far reaching administrative and operational responsibilities for his ‘children’. The core principles of the ‘family model’ are put up on shopfloor billboards (see Figure 7.2). There is evidence to suggest that that the ‘family concept’ mainly serves to satisfy two basic work dispositional demands of the German expatriates. First, it aims at ensuring high quality by instilling a sense of responsibility with those who are directly involved in production, most of all the supervisors. The following citations from an interview with two different German production managers suggest such an interpretation: The family concept comes from the Indian culture, no doubt about that, that’s from here. You know, the philosophy that someone has

Top 10 family organisation Principles 1. P ersonal cleanliness is as important as a clean house and a car body. 2. R emember to inform your father about the discussion you had with any of the superior. 3. I n the ‘Family’ the target of ‘Faults per car’ must be known to all. 4. N umber of ‘Family members’ in a ‘Family’ should not exceed 15. 5. C oaching by the ‘Family father’ to the ‘Family members’ is a must. 6. I n the ‘Family’ the ‘Father’ is always there. In his absence the ‘Eldest son’ takes care. 7. P lace of the meeting of the ‘Family’ is the ‘Information board’. 8. L et us put the stickers on the cars to mark the defects. 9. E very line fault is to be discussed with the concerned operator by the supervisor personally. Special care should be taken to avoid repeated mistakes. 10. S upervisor works like a ‘Father’ of a ‘Family’. Source: copy from a shopfloor billboard (Mercedes-Benz India Private Ltd).

Figure 7.2 MBI’s ‘Family organization principles’

Empirical Results 199

to be responsible for an area, you know. I mean a ‘self’-responsibility you know. I don’t see otherwise any other option to build such a high quality. (German General Manager Production, 1998) We also have our family principles. There we tried to take the Indian mentality into account. I think that it is very important for us to say, that is, the father figure, especially the supervisor on the line; well, it was very important, that the supervisor really is the father of the family and that, when he is not there that there must be someone – the eldest son for example – who looks after things. Well, that is something they understand. (German General Manager Quality Assurance, 1998) The link between the ‘family concept’ and the taking of responsibility becomes even clearer if we bear in mind that the role of the father and the eldest son are those positions in the Indian society that are most intimately associated with the taking of responsibility (see Chapter 4). Secondly, it emerges from interviews that the family concept serves yet another core demand. The interview with the German head of production indicates that the family concept is equally a tool to achieve closer team-like cooperation and bridge the socio-professional distance between the supervisors and operators: There is a management level in India which I see very critical and that is for me the Meister-level which doesn’t exist here. Here we have the supervisors and there is a clear divide between the ones beneath and the ones above. And back home in Germany we have the Meister who has a very, very high importance in managing things in that he is a link between workers and management. And he is a man who understands both. And the interface here is a very dangerous one. That is, those below know how it works, they are just meant to work, and the supervisor is already a studied man and walks around with a tie and a shirt, is another breed of human and also doesn’t want to dirty his hands. This is culturally a big topic and management has to pay a lot of attention here. Here is my closest attention. Therefore, I highlight especially the supervisors. This is also why he is the family father on the line and to lift him out and to call upon him to serve as a link and to make him personally demonstrate the work. In other words, that is very crucial and there is a huge gap in Indian management and some haven’t even recognized it but that’s the ways it is. (MBI 1/German head of production, 1998)

200 Hybridization of MNE Subsidiaries

In this context, the ‘family model’ – by means of allocating the ‘father role’ to the supervisor – aims at creating the link that is commonly formed in Germany by the ‘Meister’. Thus, an effort is made to bridge the professional and social distance between supervisor and operators by drawing on an Indian institutional template, the father-son/-child relationship. Interestingly, the ‘family concept’ is not the only example where the fulfillment of foreign/parent demands is achieved by drawing on local institutional patterns. In the case of housekeeping a similar solution is sought. To circumvent operator-resistance, particularly vis-à-vis cleaning the shopfloor, German management imports cleaning machines. Using cleaning machines is in line with the institutionally founded workidentities of operators. The following citation of the German General Manager Production underlines this as follows: Half a year later I said: Well, now everyone cleans his work place! And I was well aware from colleagues that this was going to be a problem because it is really the lowest to clean the floor. And in order to forestall any discussion, I organized beautiful Kärcher cleaning-machines and said: You are not cleaning, you are operating a machine! It was a big fight but now we do it and it works. (MBI 3) Although the ‘family model’ is the most prominent solution, there are other examples where German demands are realized by drawing on locally available templates. For example, a host of measures that aimed at raising cleanliness and quality standards depart from home company practices. MBI uses, for example, a Q-graph to visualize the quality performance of each operator. This Q-graph has been adopted from a local Japanese supplier. It should be noted, again, that none of these measures replicate home country patterns. In fact, the use of Q-graphs would be in breach of data privacy legislation in Germany and would trigger serious work council opposition. Finally, a crucial enabling factor in creating MBI’s work organization and human resource profile in line with parent demands is that the company is a Greenfield operation with a young work-force. Besides, MBI’s local JV partner has prior experience in automobile production. The old relationship with TELCO and the market liberalization of the 1990s allows MB to choose a JV partner that is not entirely inexperienced in automobile manufacturing. At the same time, MBI managers do not have to fight an old conflict-ridden Brownfield culture. MBI’s establishment

Empirical Results 201

is set up afresh, in a new factory hall, with a mainly young workers who have been either newly recruited or hand picked and trained in the JV partner’s apprenticeship programme (MBI 1). This young workforce is quite open to demands posed by German expatriates. As many of the young operators are ITI-graduates who have received apprenticeship training at TELCO, MBI’s workforce not only features high formal qualification-levels but also decent levels of skill. German expatriates, therefore, do not have to start from scratch as far as the human resource profile and matching skill levels are concerned. In fact, German managers praise the skill levels and discipline of the workers inherited from TELCO. A concern with inappropriate skill levels is mainly identified in the case of young engineers and production managers who are newly recruited. Here German managers try to institutionalize a different behaviour by training on-the-job and by acting as role models (German General Manager Production, 1998). In summary: The dominant recontextualization mode of MBI’s work organization and human resource profile can be described as a setup to meet foreign/parent demands, while respecting and drawing on local institutional demands and conditions. The recontextualization involves both the adaptation of some foreign/parent concepts (resilient foreign/parent context) to local/host strategic and institutional context and an adaptation of local institutional conditions and demands (resilient local context) to foreign/parent demands. The adaptation of the local context to foreign/parent demands is enabled by the Greenfield site’s contextual resilience and the resource mobilization of the foreign/parent expressed in a strong expatriate presence.

Outcome The hybridization outcome of MBI’s work organization and human resource profile is the combined result of on an absent transfer of formally defined templates, institutional and strategic misfits between foreign/parent demands and local institutional conditions and a recontextualization mode that mainly aims at meeting foreign/parent demands by observing and drawing on institutional templates available in the local/host institutional context. While crucial aspects of MBI’s work organization and human resource profile are shaped to meet foreign/parent demands, this does not imply an imitation of German home plant work concepts and human resource profiles. After all, this is also ruled out by a substantial strategic distance. Yet, we cannot talk of localization, either, because the organizational solutions are no clear reflections of typical local work arrangements and

202 Hybridization of MNE Subsidiaries

human resource profiles. The best way to describe MBI’s work organization and corresponding work profile is probably between hybrid and customized/novel. The emerging family model is indicative of this. The family concept aims at changing certain local work dispositions which the German management deems problematic with regard to the Indian workforce. It is most of all a lack of co-operation between operators and supervisors, the initial unwillingness of supervisors to engage in ‘bluecollar-activities’ and the general problem of allocating responsibility to lower organizational-levels. For the German management the familyconcept is a very powerful metaphor to get certain things done their way, while using a well-known host institutional template. The idea of family and children allows bridging the professional and social gap that exists between supervisors and operators and allows fostering the idea of a team, a family with common goals. The father, like in a family is asked to be a role model. He has to be able to show the work with ‘his own hands’ which is, if achieved, also a departure from a typical workidentity of an Indian supervisor. Moreover, the family metaphor helps to allocate responsibility down onto the shopfloor by using the father figure – a role most strongly associated with responsibility taking in Indian society. However, the hierarchical emphasis of the family-model also implies a compromise from the German side. A more equality based group concept as promoted and introduced in MB’s home plants is not even considered for MBI. German managers admit that the Indian teamwork concept is a departure from the group-work concept as followed by MB in Germany. The Indian family-team is very much a hierarchical configuration, with the supervisor being the undisputed leader. Thus, MBI’s work organization is hybrid in the sense that it draws on a typical Indian patriarchal-hierarchical institutional pattern to further the implementation of German demands.2 The family model is, at the same time, customized or novel in that it responds to different contextual demands and is neither a clear reflection of the shopfloor practice in the German parent company nor a reflection of what is typically found in India’s industrial work contexts. For on the German shopfloors no ‘family concept’ exists and the work organization has increasingly shifted towards autonomous group-work where hierarchies are less pronounced and ‘Meisters’ take on more distanced advisory roles. In a typical Indian manufacturing company, in turn, the demarcation between supervisors and operators would be taken-for-granted (c.f. Ramaswamy, 1996) and hardly an object of change through a de-contextualized ‘family model’. Not only is MBI’s central work concept a hybrid, but the corresponding human resources profile is as well. Both skill-levels and more so basic

Empirical Results 203

work dispositions, departed from a typical Indian and foreign/parent home plant set-up. According to German management, quality awareness and responsibility taking have substantially improved and depart from typical local conditions. Factory discipline, in turn, is seen to even surpass German standards. Moreover, the average Indian worker working on MBI’s lines has a higher qualification than his counterpart on the home production lines. On the other hand, while attitudes of hierarchical demarcation are seen to have been reduced and quality awareness enhanced, German management is doubtful if the changes are so deeply internalized that they would stay without ongoing German presence. Taken together the human resource profile of MBI is a hybrid in the sense that some work dispositions and formal qualifications (related to organizational position) remained institutionally Indian, while others move closer to the German ones. Finally, while formal concepts are one-thing actual behavioural patterns are quite another. Formal concepts may be more often than not superimposed and actual informal practices may diverge from them. In MBI it appears that the family concept is the result of local on-site dynamics, rather than an abstract concept designed and superimposed from anywhere else. This is not to suggest that there is always congruence between the concept’s behavioural underpinnings and actual behaviour. However, according to German and Indian managers, the model achieved its core goals in work dispositional terms (MBI 1, 3, 5, 6). Supervisors are said to visit their operators privately and their resistance to co-operate with operators and their unwillingness to go down to the ‘dirty line’ is said to have faded over time. On the other hand, the lingering presence of documents presented above can also be seen as an indication that certain behavioural issues require ongoing attention. All in all, the work organization and human resource profile is hybrid. They are hybrid in the sense that they reflected foreign/parent and host context institutional elements. These elements from different contextual origins are pieced together in a highly idiosyncratic way. Some of the work organization configuration simply reflected a specific response to a complex set of different demands. As such the configuration also features elements that are customized or novel.

Industrial relations Transfer scenario As far as company industrial relations are concerned, there is neither a template nor are there specific demands German expatriates want to see

204 Hybridization of MNE Subsidiaries

implemented. There is, for example, no intention to transfer German style works councils. One of the main reasons for the transfer absence is the fact that company industrial relations are under the purview of the human resource department which is staffed throughout MBI’s history by experienced Indian human resource managers. In the first years, Indian management handles the human resource department because it is the Indian JV partner’s responsibility. However, even after DB/DC fully acquires MBI, the human resource department remains in the hands of Indian managers. In fact, the human resource department is the only function that remains exclusively in Indian hands throughout the company’s existence. This is not least the case because of the jungle and pitfalls of India’s labour legislation and industrial relations system. The institutional distance between the German and Indian industrial relations scenario and the lacking familiarity of German managers with host context conditions in this regard make them to rely on their Indian counterparts. So, next to the fact that there is no defined template for transfer, the institutional distance between the home and the host context deters German management to seek a transfer. Besides, there are no indications that German managers find home modes of labour relations so advantageous as to seek their replication in India. This, in turn, allows Indian managers to build MBI’s industrial relations by drawing on a local/host template.

Misfit There are no indications that German managers have specific demands or problems with the way the Indian human resource managers handles company industrial relations. At least, there are no signs that German managers finds MBI’s labour relations implemented problematic. In fact, the opposite is the case; German managers praised the local discipline of labour. Thus, while the institutional distance may have indirectly influenced German managers not to transfer their home company industrial relations, there is no misfit between local/host template and foreign/parent demands. German expatriates either have no demands or their expectations vis-à-vis MBI industrial relations are not contradicted by local conditions. There are, however, changing local/host institutional demands at MBI to which the Indian human resource managers respond to. In the early years, when MBI operates on TELCO’s compound, the company is shielded from the direct impact of host context industrial relations. This has to do with the circumstance that MBI is from an outsides perspective not perceived as an independent unit and from inside not perceived as a unit in its own right. Specifically, MBI

Empirical Results 205

adopts, in the first years, the wage settlements which are negotiated by TELCO’s union. In this scenario, there is not much threat of the formation of a much feared external union. However, when MBI is taken over by DB/DC and moves to a new location outside the premises of TELCO, union activist approach MBI employees and probe for possibilities of creating an external union. The change in equity and the following shift to an outside location, change local/host institutional demands and conditions for MBI with respect to its company industrial relations.

Mode of recontextualization No adaptation of foreign/parent context to the local/host context In the absence of a foreign/parent template for MBI’s industrial relations, no such template can be adapted. Similarly, there are either no foreign/parent demands or no contradictions between foreign/parent demands the local/host set up of industrial relations. Either way, there is no requirement for an adaptation of foreign/parent demands to the local/host context.

No adaptation of local/host template to the foreign/parent demands As the local/host context template faces no foreign/parent demands, there are also no adaptation requirements. Instead, the host country managers who respond to local/host institutional conditions and demands create MBI’s industrial relations in line with a modern host context template. Specifically, in the first years, they install a ‘coordination committee’ which has the function of an ‘early warning system’ and addresses worker grievances (MBI 2). These efforts are mainly geared at assuring industrial peace at the site. As MBI is shielded in the first years from institutional pressures of the Indian industrial relation context, Indian managers are happy to keep out any sort of union activity. Up until 2000, the site’s industrial relations are mainly regulated through the ‘coordination committee’ which is constituted by labour representatives and chaired by the head of human resources (MBI 2). However, when MBI moves to its new location outside the premises of TELCO and outside activists approach the MBI employees, the Indian human resource management takes a proactive step and founds an internal company union. According to the Indian head of human resources, any organization in India would sooner or later have a union. But, to prevent the union from being externally linked and to prevent union radicalism from entering the company, MBI’s human resource managers pursue the foundation of an internal union. In doing so, they draw on local/host context template

206 Hybridization of MNE Subsidiaries

for industrial relations that is increasingly followed by modern companies in India. This option gives the company’s labour a stronger voice, a forum to raise grievances and wage demands. At the same time, it serves to keep Indian union radicalism out of the company. The union’s foundation by Indian management is a deliberate, pro-active step to select the most preferred institutional mode for company industrial relations available within the framework of the Indian industrial relations system. It is probably the resilience of the local institutional context of the Greenfield site with a young workforce that allows the establishment of such a non-contentious and modern mode of industrial relations.

Outcome MBI’s industrial relations are neither impacted by a foreign/parent template nor by foreign/parent demands. Instead, industrial relations are created by host country human resource managers in response and in tune with local/host institutional conditions and demands. They respond to these conditions by establishing a typical host template, an internal unaffiliated union. This structure reflects a mode of industrial relations followed in many modern companies in India. The interaction with this new union is rated good although not as smooth as in the early years. Although there is some minor tension with the union, at the time of the second visit, due to wage negotiations, there is no threat of strike or any radical industrial action. Rather it is perceived as normal to have a bit of tension during wage negotiations. The head of human resource, however, also stresses that he has a watchful eye on the situation as the formation of a second union can never be ruled out (MBI 4). Nevertheless, employee relations are by and large described as smooth and peaceful. Another indication of this is that the site has low levels of absenteeism and high levels of labour discipline. It is fair to say that the industrial relations of MBI reflect a typical local mode of industrial relations.

Supplier relations Transfer scenario Regarding MBI’s supplier relations there is neither a comprehensive transfer of home plant template nor a comprehensive replication effort of any existing CKD plant’s supplier relations. Yet, there is a loose reference to the just-in-sequence (JIS) and supplier park concept practised at some sites in Germany. The main factor ruling out a transfer of home plant supplier relations – most notably the tiered supplier structure and JIT supply logistics – is the

Empirical Results 207

strategic distance. Given the host strategic situation of low demand, it makes no sense to build an integrated manufacturing unit. Instead of an integrated plant, a low volume CKD set-up is chosen. This, in turn, means that it is for suppliers very unattractive to set up or invests in capital-intensive manufacturing facilities for MBI. As it is in the very nature of CKD sites to receive most parts and components from home operations, JIT supply logistics are rendered meaningless. In this scenario, inventory costs and times become less important cost variables. Cost minimization is mainly achieved through bulk container shipments, which comes at the cost of higher inventories (MBI 3). Thus, given that MBI mainly performs assembly activities and relies on the long distance supply of CKD-kits, a number of supplier-related concepts that apply for integrated plants are entirely irrelevant for MBI. While it is evident why there is no intention to replicate home-plant supplier structures and supply logistics one could imagine that MBI would adopt supplier relations similar to other CKD sites. This, however, is only partly possible because MB has no defined template for CKD supplier relations and because institutional demands in India, namely the local content requirements, force MBI to adopt supplier relations that differ from other CKD sites. At the same time, there is also no intention to use the local JV partner’s supplier relations as a template. This option is ruled out by DB/DC’s control of the site, its global supplier policies and its introduction of a premium product that has different quality and technology requirements compared to TELCO’s products. Having said that hardly any templates are used is not to say that there are no local/host demands and foreign/parent with respect to MBI’s supplier relations. Important demands from the local/host institutional are the local content and other foreign direct investment requirements. Moreover, the supply and demand conditions define a strategic context the site’s supplier relations have to respond to. Additionally, there are also foreign/parent demands affecting MBI’s supplier relations. Like other international automobile companies, DB/DC formulates global policies and concepts for its supplier chain. The most important ones are the ‘tandem’ concept and the ‘follow sourcing’ policy as well as global quality systems and processes (MBI 5). While the ‘tandem concept’ aims at establishing cooperative relations with key/first tier suppliers, the ‘follow sourcing’ concept envisions that key global suppliers provide their products and services wherever DB/DC manufactures globally. Apart from these policies that are mainly concerned the contractual relations with suppliers, there are worldwide quality polices DB/DC suppliers have to adhere to (MBI 6).

208 Hybridization of MNE Subsidiaries

Misfit Although there is practically no local/host or foreign/parent template serving as a blueprint, there are different demands and conditions impacting the supplier relations. As far as the host context is concerned, there are institutional demands with the goal to develop India into a supplier base. Foreign exchange neutrality/export requirements, high import tariffs and local content requirements are the principle tool to force foreign companies to increase local value addition. For companies with high volume demand scenarios and price sensitive customers, demands for high local value addition pose no problem. However, for manufacturers like MBI, facing a low demand scenario for high-segment vehicles, local value addition, beyond a localization of labour intensive production, steps is not economically viable. Given the low demand, these companies cannot invest in capital-intensive production steps which require high production volumes to amortize. By the same token, it is very difficult to convince international and local suppliers to set up manufacturing units and accept low volume orders. There is, thus, a substantial misfit between the local content demands, posed by the host institutional context and the strategic demand of profitable production on the part of the foreign/parent and its suppliers. It is this principle dilemma MBI faces in the set up of its supplier relations. Without the pressure of local content requirements, MBI would probably not even have considered to work with suppliers in India. However, given local content requirements, MBI chooses to work with local suppliers. Once the decision is made to work with suppliers in India, certain supplier related policies apply. These policies are mainly the ‘tandem’, the ‘follow sourcing’ and quality policies. Given the strategic and institutional conditions in India – low volumes and lower quality standards of suppliers – the application of the concepts is not an easy task and translates into change pressures on local suppliers.

Mode of recontextualization Little adaptation of foreign/parent templates and demands to the local/host context As there is hardly any template transfer, there are also few possible adaptations of such a template to the local/host context. Those concepts mentioned are the JIS and supplier park concept. These home plant concepts are not or only marginally applied due to the strategic distance between the home plant and local/host context. For example, interviewees suggested that, instead of JIT, just-in-sequence (JIS) was transferred to shape supply logistics. However, there are strong indications that

Empirical Results 209

even JIS is not implemented as in the integrated plants at home. With component deliveries three days in advance – though in the right vehicle sequence – India’s supply flow is a far cry from time-critical pull-systems which both JIS and JIT essentially are. Naturally, the ‘real’ transfer of these practises is no option as supply logistics in CKD set-ups follow a different logic compared to integrated plants. Similarly, there is also no implementation of the supplier park concept as practised in the A-class production in Raststatt, Germany. While MBI’s systems are located near the plant, there is no supplier park as such where all the suppliers are located. While there is hardly any supplier template transfer, there are foreign/parent demands that can be potentially adapted to the local/host context. One indication of such an adaptation of the foreign/parent demands to the local/host context is the condition that MBI works with local suppliers despite being a CKD site. To satisfy local/host institutional requirements for local content, MBI establishes supplier relations with six ‘module’ or ‘system suppliers’ (MBI 5). Specifically, some of the CKD components are separated out and shipped directly to ‘module or system suppliers’ in India. These suppliers assemble the components (such as seats, axles and cockpits etc.) at nearby sites in Pune and supply them to MBI. As the systems and components supplied by these suppliers count as local content, no matter how much of their value is created abroad, MBI is able to achieve a ‘deemed local content’ of about 60 per cent (MD, 1998). Thus, in supplier structural terms, to fulfil India’s FDI policy, MBI develops a small supplier base in India. While, MBI formally fulfils local/host demands, the ‘real local content’ remains under 20 per cent. In this perspective, we could also view MBI’s supplier set up a partial adaptation of local/ host institutional demands (host context resilience) to foreign/parent strategic demands for profitable operation (non-resilience of foreign/parent). Regarding DB/DC’s supplier related policies there is no sign of adaptation to the local/host context. Among the most important are the ‘follow sourcing’, the ‘tandem’ and quality policies. As we shall see next, this requires some adaptation – creation and change – of the local/host supplier context.

Some adaptation of local/host context to foreign/parent demands Once the decision is made to work with suppliers in India, DB/DC supplier policies apply. This, in turn, requires some adaptation of the local/host context to satisfy MBI’s requirements in the establishment of supplier relations. Specifically, out of MBI’s six system suppliers in India

210 Hybridization of MNE Subsidiaries

(SM Auto/Eberpaecher, TELCO, Dana Spicer India, Tata Johnson, Bajaj Tempo, Draexlmaier-Motherson Sumi), four are long-standing DB/DC suppliers serving the company already in other parts of the world. In India these suppliers form local JVs and take charge of MBI’s the subcomponent assembly (MBI 5). Similarly, in line with the tandem policy suppliers are treated as closely integrated business partners, ‘extended Arms’ of the corporation (MBI 5). Particularly, in the case of the newly developed local suppliers the company provides support. In some cases, MBI delegates employees and even expatriates to the sites of system suppliers. MBI also actively assists suppliers to adopt quality standards and polices. Without such close cooperation and the involvement of global suppliers, MBI could not achieve the required quality-levels. Such assistance is not simply a formal requirement in line with the ‘tandem policy’ but a necessity in the face of adverse institutional/strategic conditions in an emerging market automobile sector. Thus, in contractual terms, the establishment of supplier relations in India is based on an adaptation of the local/host supplier context to meet foreign/parent policy demands. To conclude: MBI’s supplier relations are mainly created to integrate and respond simultaneously to foreign/parent as well as host context strategic and institutional demands and conditions . In structural terms they reflect host institutional demands; in contractual terms the site largely reflects foreign/parent policy demands; in terms of supply logistic the site reflects the local/host strategic context of low demand and the corresponding task profile of a low volume site (low volumes and a strong reliance on long distance supplies for CKD kits).

Outcome MBI’s supplier relations are not based on any template transfer. The strategic distance, between the local/host strategic context and the home strategic context, rules out the transfer of home plant supplier relations. In addition, no supplier relations template is defined for CKD sites, at least not at the time when MBI’s supplier relations are established. Simply using the local JV partner’s existing supplier relations as a template is also ruled out by a number of factors including: strategic distance between the JV partner’s task profile, a technology gap, DB/DC’s product introduction and majority in the JV as well as its global supplier policies. While no template is used or transferred, there are foreign/parent and local/host institutional and strategic conditions that have to be factored into MBI’s supplier relations. These different conditions and demands are partly contradictory and are resolved with a pro-forma compliance

Table 7.2

Summary of MBI’s hybridization profile

Firm > PS Dimension

MBI

Functional differentiation

Transfer: Foreign/parent strategic demands and local/host institutional demands and strategic conditions to be met Fit/Misfit: No misfit between different demands and conditions, need for their accommodation Mode of Recontextualization: Adaptation (i.e. creation of functional set-up) of local context to meet different demands and conditions Outcome: Customization

Hierarchical differentiation

Transfer: Use of local template Fit/Misfit: Increasing institutional misfit between local template and parent/foreign demands Mode of Recontextualization: Some local template adaptation; some foreign/parent demand adaptation (rejection/limitation) to local template Outcome: Hybridization (coexistence)

Technical configuration

Transfer: Only selective and increasingly less transfer of technical configuration; mainly foreign/ parent strategic demands plus the pressure to meet host strategic conditions and institutional demands Fit/Misfit: Host context institutional and foreign/parent strategic demands partly contradictory Mode of Recontextualization: Adaptation (i.e. creation) of local context to meet foreign/parent demands and host demands Outcome: Mainly customization (Continued)

211

212

Table 7.2

Continued

Firm > PS Dimension

MBI

Work organization/ HR profile

Transfer: Foreign/ parent demands and local/host conditions and demands to be met Fit/Misfit: Misfit between foreign/parent strategic/institutional demands and local/host institutional demands and conditions Mode of Recontextualization: Adaptation (i.e. creation) of local site to integrate foreign/parent demands and local/host demands; little adaptation of foreign/parent demands; adaptation of local context by rejecting local demands; yet, drawing on a de-contextualized host-context template Outcome: Between hybridization and customization

Industrial relations

Transfer: Local/host template in use Fit/Misfit: No misfit between local/host template and foreign/parent demands Mode of Recontextualization: Creation of company industrial relations in response to local/host institutional demands and conditions Outcome: Localization

Supplier relations

Transfer: Foreign/parent strategic/institutional demands and host institutional/strategic demands and conditions to be met Fit/Misfit: Some misfits between the local/host and foreign/parent strategic/institutional demands and conditions Mode of Recontextualization: Only little adaptation of foreign/parent demands; some rejection of host context demands; for the most part adaptation creation/customization of supplier relations to integrate foreign/parent and local/host demands and conditions Outcome: Customization

Empirical Results 213

of MBI supplier relations with local/host institutional demands and an actual compliance with foreign/parent demands and policies. All in all, the site’s supplier relations are creation or customized solutions to meet foreign/parent policies under the specific strategic conditions of a low volumes, long distance CKD supplies, and FDI regulations in India’s institutional context. The result can be best captured as a customization (see Table 7.2 for a summary of MBI’s hybridization profile).

8 The Case of Skoda Auto India

Introduction VW’s internationalization in the passenger car segment starts early compared to other German auto manufacturers. In the 1950s VW opens its first CKD-operations in Brazil (1953), Mexico (1954) and South Africa (1956). Responding to stringent protectionism in the respective countries, these plants soon develop into fully integrated production sites (Eckardt et al., 2000). The internationalization pattern of setting up production sites for ‘mature products’ in peripheral markets continues throughout the 1970s and 1980s and involves the establishment of new sites in Nigeria (1975), Argentina (1980), Egypt (1981) and China (starting in 1983). This development is paralleled by VW’s Europe-centred internationalization (Eckardt et al., 2000). In 1982, VW starts cooperation with SEAT, which leads to SEAT’s takeover in 1986. According to Eckardt et al. (2000), the 1980s herald a qualitative change in VW’s internationalization strategy towards ‘transnationalization’. As a result, the international division of labour becomes more integrated, involving more exchange of parts and components between different international production sites. In addition, the centre– periphery imbalance with regard to strategic roles and product mandates begins to crumble. A number of foreign plants are no longer mandated to exclusively build outdated models for peripheral markets but are assuming product mandates for new, core market-models. A case in point is the New Beetle built at VW’s Mexican Puebla plant. In addition to this qualitative shift, the 1990s bring a quantitative boost in VW’s internationalization. Among the most important developments are probably: the establishment of the JV with China’s First Automobile Works, the 214

Empirical Results 215

takeover of Skoda and the foundation/renewal of a number of production site in Eastern Europe and Latin America. The establishment of SAI has to be seen within the context of VW’s global product strategy: that is, using the Skoda brand to cater to Central/Eastern European and Asian markets. It remains unclear, however, whether the establishment of Skoda India is a strategy-driven and coordinated effort by the Volkswagen group, or rather a subsidiary initiative that grows out of Skoda itself. In any case, while the involvement of Skoda in India may have been the result of a central decision by VW, SAI’s establishment, at least at the time of research, is very much left to Skoda. It appears (SAI 1) to be an internationalization effort driven by the Skoda subsidiary, rather than by the VW group. Nevertheless, the site serves as bridgehead for the VW-group to venture into India. As such the site has an important strategic role for the future. Moreover, Skoda’s cars meet a comparatively good initial reception in the Indian market. With regard to the strategic distance, the Indian site is, in production volume and product variety terms, strategically very distant from Skoda’s home plants in the Czech Republic, Mlada Boleslaw (2001/350,000 cars), Vrchlabí (2001/70,000 cars) and Kvasiny (2001/12,000 cars). At the same time, the strategic distance to home operations is in the Skoda India case is less dramatic than in the MBI case. After all, Skoda’s home operations are also based on a ‘high-labour and low-investment approach’ and rely heavily on suppliers (Kochan, 2001). Strategically, the Indian operation is probably closest to Skoda Auto’s Eastern European CKD-assembly operations, namely in Sarajevo (Bosnia & Herzegovina), Poznan, (Poland), Solomonovo (the Ukraine) as well as in Kaluga (Russia) and in Ust’ Kamenogorsk (Kazakhstan), where the Skoda vehicles are assembled (Skoda, 2008).

Global product strategy Drawing on Porter’s (1980) generic strategies, VW’s overall global product strategy can be best described as combining a differentiation and cost leadership strategy. Along similar lines, Boyer labels VW’s profit strategy a ‘diversity and volume’ strategy (Boyer and Freyssenet, 2003; Freyssenet, 2000). In the 1990s, the differentiation aspect of the product strategy sees a strengthening and is expressed by an ‘upward opening’ of VW’s product portfolio to new luxury brands and the upward extension to luxury segments within the VW brand itself (Pries, 1999). The Volkswagen brand group – comprising the Volkswagen Passenger Cars, Skoda, Bentley and Bugatti brands – offers products ranging from

216 Hybridization of MNE Subsidiaries

the small sub-compact cars to luxury saloons. The aim of the brand is to set class-beating standards in all its products from the sub-compact Lupo through to the Phaeton luxury saloon. Skoda, with its Fabia, Octavia and Superb models, offers a range of products extending up to the mid-class segment which above all deliver outstanding value for money. (Volkswagen AG, 2005) VW follows from relatively early on different product strategies for different regional markets. This mainly involves introducing older, specifically customized or even outdated models to peripheral markets in Latin America, Africa or China and state-of-the-art models to the developed triad markets. While this disparity narrows in the 1990s, VW still differentiates its product offerings according to different markets or world regions. For example, in contrast to VW’s overall product strategy, the product strategy of the Skoda brand rests more on the cost-leadership orientation, targeting mainly Central Eastern Europe and Asia. In 2005, Winfried Vahland, Vice Chairman of Board of Management Commercial Affairs of Skoda Auto, defines ‘[e]stablishing of a CKD-Strategy to conquer other markets in Eastern Europe and Asia, incl. China’ as a corner stone of Skoda’s future strategy (Vahland, 2005). In fact, Skoda’s market success in Western Europe is originally not intended and has cannibalizing effects for the VW brand. Despite different regional product strategies, VW increasingly integrates its product strategy by developing shared platforms for its four main passenger car brands: VW, Audi, Seat and Skoda. This even allows establishing production sites that are capable of producing a mix of the different car brands (Eckardt et al., 2000).

Establishment mode and equity development SAI is founded as a wholly owned Greenfield subsidiary in the year 2000. A press interview with Skoda India’s managing director suggests that Skoda/VW chooses a wholly owned Greenfield as the entry mode because it wants to build up the site in line with VW/Skoda operational standards. Q: Coming to your manufacturing plans, it seems strange that in this market, which has excess capacity, you are adding to it. Wouldn’t it have made sense to use some of the existing capacity, especially since you will be starting with an assembly operation? A: That’s right. But existing capacity doesn’t mean that it is really suitable capacity for us. New capacity at the same time does not attribute to over-capacity. It’s the question whether the capacity that has been

Empirical Results 217

invested for is being accepted finally by the customer. What we want to do is have our own little thing which we can build up from scratch with our way of doing things of manufacturing and of quality management. That’s the most important part for us. We have decided to locate in Maharashtra because the quality labour and the infrastructure is there. Also, Mumbai port is on the right side of India for our shipments from Europe. (Autocar India, 2001) In contrast to other international automobile companies, the VW-group is a late comer in the Indian market. Unlike in China, the VW-group has been undecided for some time about whether, when and how to enter the Indian market. Already in 1996, the VW-group is considering to enter the Indian market. In 1997, Posth, the President of VW Asia-Pacific Limited is asked in a press interview about the size of investment in India. He states that VW is considering a US$130–190 million operation. Production volumes are not decided, but are intended to be at least 20,000 units per year. Concerning the brand-choice, Skoda is mentioned among others brands (Mukherjee, 1997). The principle dilemma that the VW-group faces in India is that the volume segments A and B are in the firm grip of first movers, mainly MSI and Hyundai. At the same time, VW has seen MBI’s difficulties in offering premium-segment cars in India. Therefore, VW is careful about introducing a high-segment brand such as Audi. However, offering a low-segment car is equally risky. Like many other foreign car manufactures that have entered earlier, VW/Skoda comes to realize that the Indian market demand is weaker than originally projected. Instead of the bigger investment plans, the company has earlier, a careful look-and-see approach is adopted (SAI, 2; Mukerji, 2002). Thus, as a latecomer in a slow market growth scenario, VW/Skoda scales down its initial project. Ultimately, the choice is made to enter the Indian market within the format of a small SKD-operation and with the Skoda brand. The brand is seen to cater to a specific market-niche between the C and D segment which is identified as vacant. In 2000, Skoda Auto signs the MoU and founds SAI. SAI is set up as a wholly owned subsidiary. This equity situation is still unchanged, in 2003. The choice of entry is structured by the following conditions: First, Skoda wants to build its ‘own little thing’. The easiest way to do this is a wholly owned Greenfield site. Probably Skoda has also observed that most auto JVs in India are beset with JV partner problems which generally triggered a continuous equity shifts in favour of the foreign partner; secondly, neither VW nor Skoda has a long-standing history in the Indian

218 Hybridization of MNE Subsidiaries

market. Therefore, no specific Indian partner suggests himself; thirdly, when Skoda enters in 2000, India’s FDI regime has substantially changed, rendering a market entry in the format of a wholly owned subsidiary much easier.

Location, production programme and market share Skoda’s first assembly facility is located in an industrial area near the city of Aurangabad (State of Maharashtra). The facility commences as a low-investment project in the format of a SKD-facility. Skoda’s assembly activity takes place in a leased building on the compound of a Siemens Switchgear factory. In early 2003, SAI assembles only one model, the Octavia in two basic variants (1.9 litre turbo diesel/2.0 litre petrol available in six colours) (SAI 2). The site has an installed capacity of 10,000 SKD units per year. When the site is visited, the daily output is at 35–40 vehicles on a double shift basis. Manufacturing activities are characterized by pure assembly of SKD-kits. The company has already moved on from an initial SKD 00 status to an SKD 0 status, implying slightly more local assembly activity. Moreover, the site is preparing the commencement of CKD production at another location. This means that the car will arrive in a more disintegrated manner requiring more local assembly. In addition to its CKD site in Aurangabad, VW is planning a new production facility in India by 2010. The goal is to have an integrated production unit in Pune. In terms of segmentation Skoda enters the Indian market with the lower-D-segment Octavia which is produced in two basic variants (1.9 litre turbo diesel/2.0 litre petrol available in six colours). Given the Octavia’s positioning in the Indian market, between the upper C and lower D segment, the car is well received and is considered locally a luxury vehicle at a low price. Following the positive market response and a new import duty structure SAI plans in 2003 to move to a new site, shifting operations from SKD to CKD assembly. However, when SAI is researched, its operation is still a SKD facility, assembling one car model only. The site has an installed capacity of 10,000 units/SKD-kits p.a. and has about 230 employees. Skoda realizes in 2002, in its first year of full operation, sales of around 4400 units. In 2003, Skoda has a sales target of 8500 cars but has to lower it to 4500 units. The downscaling of the sales targets is not caused by sluggish demand but by a slowdown in production, related to shifting operations from a SKD- to a CKD-assembly set-up. In 2006, the Indian plant manufactures a total of 12,599 vehicles and sells 12,105 vehicles. This is 35.2 per cent more than in 2005 (SIAM, 2006). In 2006, the plant employed 378 people and has production capacity of around 28,000

Empirical Results 219

vehicles per annum. Whether Skoda can be called a big success is too early to say. However, by Indian standards, the market entry is a success. Compared to MBI, Skoda’s market entry is a run away success. After a little more than one year of full operation, Skoda achieves in 2003 a sales volume of 4656 units (The Hindu, 2005) and in 2004–2005 a market share of close to 1 per cent (SIAM, 2006). This market share is four times bigger than MBI’s, which operates already ten years in the Indian market.

Analysis of the hybridization profile The global transfer scenario Interviews at SAI reveal that the company does not draw on any comprehensively defined production template (SAI 1, 2). While the parent plants, at which the Octavia is manufactured, do serve as a competence pool, particularly in matters of product specifications and assembly expertise, there is neither a specific home country plant nor any other SKD/CKD site that serves as a defined blueprint for the Indian site’s set-up. Very similar to MBI, the officially planned transfer effort is restricted to documentation and planning of the technical side of the organization of the production process: documentation of assembly sequences, systems and procedures for manufacturing and quality assurance. While there is no comprehensively defined template for the Indian operation, there is a selective transfer-intent. This mainly involves the implementation of VW’s global quality and auditing system as well as the Skoda Production System (SPS). The latter defines circumscribed aspects of shopfloor work organization, including a number of tools and parameters to control quality and work-force performance. The Czech interviewee summarizes the aspects to be implemented in the Indian subsidiary as follows: The product, the systems, the parameters for regulation have to be the same. The methods have to be the same. We have to audit with the same scores which is direct pass ratios. (SAI 1) The transfer intent centres on the implementation of two parent company systems defining a limited range of aspects for the shopfloor work organization. These are, for the most part, outcome monitoring systems, tracking performance by means of various numerical indicators. How these systems are implemented and embedded in day-to-day interaction is largely left to Indian managers who receive a brief training in the parent plants. Apart from the core-system implementation, the Indian and

220 Hybridization of MNE Subsidiaries

the Czech interviewee state that there are no intentions to imbibe any sort of foreign work culture (SAI 1, 2). To the contrary, the Czech interviewee stresses that he has to learn from his Indian colleagues how things go and emphasizes that he has no ambitions to change many things as they belong to ‘their own habits’ (SAI 1). However, he repeats that the core-systems have to be implemented. This he considers his main task as an expatriate. Along similar lines, the Indian interviewee stresses that while certain technical systems are transferred, various cultural aspects have to be Indian (SAI 2). Interestingly, the limited or selected transfer intent is also reflected in an utterly polycentric staffing approach. There are only two expatriates at the site, a German managing director (a Pakistan-born Non-Resident Indian), with a marketing background, and a Czech expatriate, with a quality assurance background responsible for production. In fact, the Czech expatriate complains that more expatriates and more personnel exchange would be required. Given the tight budget of the start-up this is not possible, however. He is sure that had the start-up been a VW project, it would have received more resources, including a higher rate of expatriate presence.

The functional differentiation of the organization structure Transfer scenario SAI functional differentiation is not based on defined foreign/parent template or a local/host template. The fact that no local/host template is being used is related to the site’s entry mode. As the site is a wholly owned Greenfield subsidiary, the use of existing template or using a template from a local JV partner is already ruled out. Apart from the fact that the foreign/parent provides no template for SAI’s functional differentiation, the strategic distance in the task profile and demand market conditions between the home strategic context and the local/host strategic context explain why there is no effort to replicate Skoda’s home functional differentiation. Yet, while the transfer of Skoda’s home site’s functional set-up is ruled out by the strategic distance, the replication of other – strategically closer – CKD/SKD site’s functional set-ups would have been an option. However, at the time of SAI’s establishment, no such template is available or defined for the setup of SKD/CKD sites, at least not for Skoda (SAI 1). Although SAI has no defined template to draw on for its functional differentiation, the site has to respond to some degree to its foreign/parent and local/host conditions and demands.

Empirical Results 221

Misfit As neither a foreign nor local/host parent template is transferred or used, there is no recontextualization pressure on any template. There is also no indication that local/host demands and conditions and foreign/parent demands are outright contradictory. Yet, there are different contextual demands and conditions that call for an accommodation in the newly established functional differentiation of SAI. The foreign/parent demands involve the implementation of global policies. For example, in line with VW’s global quality policy, SAI is obliged to have an independent quality auditing department. What is more, as the task environment is with regard to the product and its quality not much different from the vehicles built at the Czech home sites, there are strategic demands to set up specific functions and processes in a similar way. However, given the low presence of expatriates at the site, there are only few specific functional structural demands other than those global standards or policies related to assembly process and product quality (SAI 1). Conversely, those few foreign/parent demands that are defined with regard to the functional differentiation meet only little if any local resistance. This is related to the condition that: the site is a Greenfield site, the workforce is young and most of the Indian managers have worked for other international automobile companies in India before. There are few functional demands in the first place and those foreign demands present, are not in conflict with local/host strategic or institutional conditions and demands. SAI’s the functional differentiation has to respond to the strategic requirements of producing a specific foreign/parent car model under specific local/host strategic and institutional conditions and demands. The latter are characterized by moderate demand levels for medium to high segment cars and local content and import tariff demands from the host institutional context.

Mode of recontextualization No adaptation of foreign/parent templates or demands to the local/host context? Since no use is made of any template, there is also no recontextualization pressure on any template. However, there are some foreign/parent demands with regard to the functional differentiation. These demands are met without much deviation.

Adaptation of local/host template, demands or conditions to the foreign/ parent context The dominant (re-)contextualization mode shaping SAI functional differentiation is a creation to meet different demands and

222 Hybridization of MNE Subsidiaries

conditions from the foreign/parent and local/host strategic and institutional context. With regard to foreign/parent context demands, the site establishes a quality department in line with global quality policies. VW’s global quality system calls for the function of a quality auditor who has to report to a global quality department of VW. By the same token, the product requires some manufacturing operations and functions similar to those of the home plants. For the most part, however, the functional set-up is customized to the specific local/host strategic context of the site. In other words, the operation’s functional set-up is mainly customized to a specific task environment that differs from the parent operations in terms of vertical integration (much less), supply logistics (long distance) and with respect to product variants (much less) and volumes (much lower). Moreover, the small operation takes care of a whole range of functions that are not directly related to production (e.g. sales and marketing, service and spares) and are not part of the task profile of a home operation. Finally, the site responds functionally to demands from India’s institutional context. These demands are connected to SAI’s comprehensive import of parts and components (SKD/CKD kits). We can, therefore, conclude the dominant (re)contextualization mode regarding SAI’s functional differentiation is a creation in line with different demands and conditions from the host and foreign/parent context. Low levels of misfit and a high resilience of the site’s Greenfield context provide an enabling context for such a (re)contextualization mode.

Outcome SAI’s functional differentiation is a customized solution responding to different contextual demands and conditions. The functional differentiation is not based on any foreign/parent template. The reason for this is that the parent company has not defined a template matching the strategic context (and task profile) of SKD/CKD sites. As far as foreign/parent operations are concerned, a transfer is not an option due to the substantial strategic distance to the local/host strategic context. The site is also not based on a local/host template which is mainly due to the fact that the site is a wholly owned Greenfield operation. The sites functional set up rather reflects foreign/parent demands about the manufacturing of a vehicle that needs to meet technical and quality specification with regard to the product and production process and at the same time producing such a product profitably under local/host conditions. The latter are defined by a low volume demand scenario related to the model choice and time of operation in the local market as well as requirements for

Empirical Results 223

certain levels of local value addition. SAI’s functional differentiation can, therefore, be best described as customization.

The hierarchical differentiation of the organization structure Transfer scenario SAI’s hierarchical differentiation does not draw on a foreign/parent template. This is on the one hand related to the absence of a suitable template and on the other to the site’s staffing pattern. While there is no foreign/ parent SKD/CKD template available for transfer, the transfer of Skoda’s home plant’s hierarchical differentiation is not considered an option, either. Both the Czech expatriate and his Indian counterpart, share the conviction that such a transfer would be incompatible with the Indian institutional context. The Indian head of production comments in this context: Q: The six (seven) levels you mentioned you have oriented yourself on the Czech hierarchical levels? A: No it was not. Not Czech hierarchical levels. It is Indian levels. In CR we don’t have these levels. They are totally different and neither have we tried to adopt that. The supervisor is at the top level there. In India supervisor is the lowest level of management. So there is a big difference and we have never tried to adopt this. Q: So they have only two levels. Operators and supervisors? A: No they have different levels of course. But when I say supervisor he is of a managerial level of the whole set-up and he is a very powerful person. Whereas in the Indian culture, the supervisor belongs to the lowest level of management. So it is a totally different scenario. In between they have, between operators and the supervisor levels, they have again levels, which are different from what we have in India. In India again the qualification, his experience, all comes into the picture. We have diplomas, we have engineering graduates, and they begin with different levels, so it is a different kind of structure here and there. (SAI 2) From the Indian manager’s perspective the transfer is ruled out by the different professional-education structure and qualification-related entry-points to specific organizational levels. He also stresses the role of status differentials of managerial positions in the home and the host context.

224 Hybridization of MNE Subsidiaries

However, while the site doesn’t follow a foreign/parent template, there is also no existing template or a locally defined template. Given that the site is a wholly owned Greenfield start-up, there is neither an existing hierarchical differentiation nor a local JV partner pressing for the implementation of a particular local template. Yet, Skoda’s polycentric staffing approach implies that the development of SAI’s hierarchical differentiation is in the hands of Indian managers. Moreover, the expatriates on site display a hands-off attitude in all areas but those that are related to their immediate task or responsibility. The Czech expatriate propagates, for example, a non-intervention policy and explains this approach with his minority status and his conviction that there is no reason to change ‘local habits’ as long as basic performance parameters are met (SAI 1). Thus, as there is no foreign/parent template transfer, little foreign/parent demands and no locally existing template, the shaping of the site’s hierarchical differentiation mainly faces demands voiced by the local/host management.

Misfit SAI’s hierarchical differentiation is not based on a foreign/parent template or demand and faces, therefore, no local/host misfit and recontextualization pattern. Being mainly set up by host context managers and responding to local/host institutional demands, the site rather comes to reflect a typical local/host pattern of hierarchical differentiation. It is a pattern that Indian management, considers lean and flat, very much in line with other modern enterprises in India (SAI 2). While the foreign/parent management does not quite understand and approve of this local/host pattern, it also poses no demands for change or modification. For example, the Czech expatriate is not happy with and does not understand the need for up to eight hierarchical designations for an operation of SAI’s size. Yet, he induces no recontextualization pressure (although he does question, whether this tendency of excessive levels of designations would be financially sustainable in the long run). The absent recontextualization pressure, despite a perceived institutional and strategic misfit by foreign/parent management, is due to the foreign/parent’s polycentric staffing approach as well as the utter handsoff attitude of the Czech expatriate. The fact that the Czech expatriate and his Indian managerial counterpart are hierarchically more or less at the same level, also add to the difficulty demanding a change in the hierarchical set up.

Empirical Results 225

Mode of recontextualization No adaptation of foreign/parent templates or to the local/host context

As no foreign/parent template is transferred and hardly any foreign/parent demands vis-à-vis SAI’s hierarchical differentiation are voiced, there is also no adaptations to the local/host context.

No adaptation of local/host templates or demands to the foreign/ parent context Conversely, as there are no meaningful foreign/parent demands with regard to SAI’s hierarchical differentiation, it does not have to be adapted to the foreign/parent context. Instead, SAI’s hierarchical differentiation is created in line with local/host institutional demands of the Indian management on site. Thus, given the absence of foreign/parent template transfers or demands, and given the polycentric staffing approach at SAI, the dominant (re-)contextualization mode is a creation of the local context, in line with local/host institutional demands and conditions.

Outcome The hierarchical differentiation of the site is newly created in line with local/host institutional demands, that is, in line with the institutional concepts of Indian managers who have the task to set up the organization’s hierarchical differentiation. Specifically, there are between seven to eight hierarchical designations from top to bottom excluding operators. In the hierarchical order the designations at SAI are: Managing Director, Director, General Manager, Senior Managers, Managers, Assistant Managers, Senior Officers, Officers, (probably also Junior Officers) and Operators. Given these many hierarchical designations for so small a company (130 employees out of which 90 are operators), the question arises whether real task profiles are linked to the many designations. While the functional heads (i.e. general managers) are responsible for the core functions of SAI: technical, sales & marketing, production, service & spares and finance, it is less clear to what extent the other designations denote real task profiles or levels of responsibility. The link between hierarchical designations and task profiles or levels of responsibility remains unclear (SAI 1, 2). It appears that SAI decouples, just as other companies in India, responsibility levels and designations. In summary: The absence of a foreign/parent template or demands and a polycentric staffing approach leave the development of SAI’s hierarchical set-up to host context managers. Moreover, a perceived institutional distance between the parent company hierarchical structure and the local/host institutional conditions as well as the hands-off approach

226 Hybridization of MNE Subsidiaries

of foreign/parent management in many areas, all favour SAI’s strong response to local/host institutional demands and, thereby, the adoption of a typical local/host context solution. We can, therefore, describe SAI’s hierarchical differentiation as localization.

Technical configuration of the production process Transfer scenario Neither a home plant nor a CKD/SKD site or model served as a template for SAI’s factory layout, process design and technical hardware configuration. This is remarkable because VW has a long experience in setting up international operations, including CKD/SKD sites. The strategic distance between the host and the home country context, is the main factor ruling out the transfer of a home plant set-up. While SAI, with its medium to high segment vehicle, has a better market response than MBI, it faces a similar situation of contrasting demand conditions and task environments between the home and host context environment. The Indian task profile, involving the assembly of around 40 vehicles per day, contrasts with Skoda’s home plants Mlada Boleslav and Vrchlabi where the same model is built at the volume of around 500 and 150 units per day (Kimberley, 2001). In fact, fearing a low demand scenario in India is a main reason why the VW group has been hesitant to enter the market. However, SAI’s task profile not only contrasts with home operations in terms of production volumes but also in terms of model variety. The SAI operation produces only one base model in a very limited number of variants. Thus, in comparison to Skoda’s home operation its small facility in India features a low base model variety in very limited variants. The difference in model variety adds to the strategic distance that results from different production volumes. In addition to the strategic distance regarding market demand, there is a supply-side strategic distance that rules out the transfer of the home plant’s technical hardware configurations. Difference in labour costs also have to be factored in to the set-up. This equally rules out the replication of home plant technical hardware configurations, which are already less automated than comparable VW plants in Germany. While there is no foreign/parent template transfer, there is also no local/host template use. After all VW/Skoda wants to build its own thing from scratch, which is facilitated by the condition that SAI is a wholly owned Greenfield subsidiary. However, although no specific template is transferred, the main plant Mlada Boslav in the Czech Republic, serves as a limited point of reference for the Czech-Indian team responsible for the initial establishment. Apart from foreign/parent demands regarding the

Empirical Results 227

production requirements of a specific model, the factory layout and process design and technical hardware configuration has to respond to specific strategic conditions and institutional demands in the host context.

Misfit While the strategic distance rules out a home plant transfer, the site has to respond to foreign/parent and local/host strategic and institutional demands and conditions. First, SAI’s vertical integration, factory layout, process design and technical hardware configuration has to respond to parent strategic demands. Despite the strategic distance in production volume and variety, the model to be produced is still the same as the one manufactured in the home plants. It requires similar final assembly activities and the same quality levels. Secondly, SAI’s vertical integration, factory layout, process design and technical hardware configuration have to respond to the specific host strategic conditions of low volumes on the demand side and low labour costs on the supply side. Thirdly, SAI’s factory layout, process design and technical hardware configuration would have to respond to host institutional demands. Like all other foreign companies, SAI signs a MoU in 1999 and is required to fulfil an indigenization schedule involving increasing local content rates. SAI also has to observe specific import tariffs for imported parts and components. It is particularly the institutional demand for increasing local content rates that prove very problematic for SAI, given the modest market demand and the corresponding production volumes. SAI’s low volumes render the set-up of capital intensive up-stream manufacturing steps, especially in the initial market entry phase, uneconomical. There is, therefore, a misfit between host institutional demands for increasing local content and the foreign/parent demands for profitable production under low demand conditions.

Mode of recontextualization No adaptation of the foreign/parent demands to the local/host context Given the early market entry stage, given the strategic distance and given that there is no defined template for SKD/CKD sites, there is also no adaptation of a foreign/parent template to the local/host context. Similarly, there is also no indication that Skoda’s foreign/parent demands vis-à-vis the technical configuration of the production process sees any substantial adaptation to the local/host context.

Adaptation of local/host demands and conditions to foreign/parent context SAI is based on neither a foreign/parent nor on local/host template. This implies that there is also no adaptation of local/template to foreign/

228 Hybridization of MNE Subsidiaries

parent demands and conditions. Instead, SAI’s factory layout, process design and technical hardware configuration is created to meet strategic demands of the foreign/parent and the strategic conditions and institutional demands in the local/host context in India. The site’s establishment mode implies that the factory layout, process design and the technical hardware configuration have to be newly established. A technical support team of expatriates from a foreign parent plant realizes the initial site establishment. Specifically, the site is set up by a group of six expatriates who are transferred at beginning for a short period of time: We had expatriates and, as I said, the basic focus has been on setting up the shop, the production line. This is basically a technical support function. How to do it? How to produce this? How to incorporate systems and procedures? This is where their support came in. And once this is taken up and learnt by our people, than this team went away. We had a good number of people here, I think six people, it was in the initial phase, once it was stabilized and everything is OK, than they were off. (SAI 2) Lacking a transfer template this creation expresses a customized response to foreign/parent and local/host context demands and conditions. The factory layout, process design is designed to meet the quality and assembly requirements of the selected model that is similar to those in the parent plants in the Czech Republic. While basic quality processes and final assembly steps are principally the same, the set-up team has to customize the site to substantially lower volumes. Furthermore, with regard to the site’s technical hardware configuration, there are also customized responses to foreign/parent demands and host context conditions. Low volumes and low labour costs rule out the transfer of capital-intensive process steps and technology such as transfer presses, paint shop technology or welding technology. It should be emphasized, however, that SAI is at an early stage of entry. SAI’s factory layout, process design and technical hardware configuration are first and foremost specific responses to host institutional demands. That is, responses to qualify as a SKD venture. Only if the site can prove a minimal amount of local value addition, is Skoda permitted to import parts and components at substantially lower rates than for full vehicles imports. The process design essentially reflects the effort of importing a completed vehicle at the lowest possible import tariff. Moreover, given the early stage of entry, SAI has not yet shifted its site to CKD operations

Empirical Results 229

and does not procure parts from India. Against this background, internal logistics processes are entirely customized to long distance deliveries of SKD-kits from parent plants in the Czech Republic. Over all, the dominant (re-)contextualizatuion mode of SAI’s technical configuration is a creation to meet foreign/parent and host context demands and conditions. The Greenfield nature of the site implies a resilient local context for set up such a (re-)contextualizatuion mode.

Outcome SAI’s basic factory layout, process design and technical hardware configuration can be best described as customized solutions. These customized solutions are the result of an absent template transfer – neither foreign nor local – yet customized responses to foreign/parent and local/host strategic as well as institutional demands and conditions. The set-up of the site is institutionally triggered by India’s FDI policy. For without the policy, the low volumes in the Indian market would probably have implied a preference for an import strategy. Skoda’s lacking transfer intent has largely to do with its careful market entry approach, with the fact that the Indian site is strategically very distant from the Skoda’s home sites (at least at this early stage) and the condition that no template has been defined for SKD operations. However, while there is no direct transfer, there are foreign/parent demands as well as local/host strategic and institutional demands/conditions to which the factory layout, process design and technical hardware responds. Specifically, to qualify as a SKD site, SAI receives SKD-kits which consist of completed vehicles, apart from a few components separated out for local assembly (SAI 2). Accordingly, the site’s process has a very low vertical integration, not comparable to any of Skoda’s home sites. SAI’s process design is such that complete car-bodies (including interiors) are manually pushed into a garage-like hall. In the hall the transmission, the engine, the front- and rear-suspension are mounted onto the vehicle on a central lever (SAI 3). Apart from this main station, there are only two or three additional stations. These comprise of minor final assembly activities (e.g. door-sealing), paint check and repair in light-tunnels. The car finally comes to life with a software upload, testing and fuelling. The product flow is discontinuous and has a workshop character. SAI’s low volumes and the high level of vehicle completion upon arrival imply a low vertical integration. Low volumes and a small number of remaining assembly tasks mean that SAI features a fraction of the home plants assembly stations. The few assembly tasks are clubbed in a small number of stations translating into high cycle times of about 40 minutes.

230 Hybridization of MNE Subsidiaries

Internal logistics as practices at home sites have no relevance for SAI, either, as the cars arrived almost completed. Similarly, SAI’s technical hardware configuration is also a customized solution. It responds to foreign parent context manufacturing requirements and the local/host institutional demands and strategic context conditions, manifested in local content requirements, low demand and low labour cost conditions. The resulting technical hardware configuration comprises of little more than simple hand tools and some final testing equipment. Compared to the home plants, SAI’s technical hardware profile shows no automation and hardly any mechanization. Apart from one lever for under-body assembly, simple hand tools, some fixtures and testing-equipment, there is no technical equipment on site. What is more, the technical hardware used is mainly sourced from vendors in India. There is only some critical equipment, such as fixtures, which are directly transferred from the parent. In summary: The production process is a customization to integrate foreign/parent strategic demands related to the production of a particular model within a particular host country strategic and institutional context, marked by low volumes, low labour costs and a specific FDI policy, pushing for local production.

Work organization and human resource profile Transfer scenario SAI work organization and human resource profile sees some selective foreign/parent transfer (SAI 1, 2). Referred to by the interviewees as ‘quality’ and ‘technical systems’, the Skoda Production System (SPS) and VW’s global quality systems are two defined templates that aim at structuring circumscribed areas of work organization and human resources. Originally derived from a VW-group production system, SPS involves a work concept for the shopfloor level. It defines a shopfloor team concept including an average team-size, team-responsibilities, training and development targets and a number of visual tools to monitor worker performance. These include charts for monitoring attendance, production, quality, cost and safety targets as well as 5S (Mudd, 2000). The Indian head of production describes SPS as follows: [SPS] talks mainly about, team working, more responsibilities for the operators, better organizing his work and his team members, training and developing his team members, ensuring better productivity and better quality, implement security. So it is more focused on first line operation. . . . So that is what it focuses on. Another major difference

Empirical Results 231

was a focus on absenteeism. This seems to be very good in this production system. Than control on quality this monitoring, tracing out the problems and getting direct reactions is very good. (SAI 2) SPS covers formal structural aspects of the work organization as well as tools to monitor performance outcomes on the shopfloor. With respect to the human resource profile, SPS rests on polyvalent workers, skilled enough to work at different stations. It relies to a certain degree on high involvement work dispositions. Workers are responsible for quality based on integrated self-certification and also some degree of maintenance based on housekeeping activities. In a way, it implies a non-Taylorist concept of work organization combined with tight work-force control. The second template transferred is VW’s global quality system. Unlike SPS, this system is a standard template implemented worldwide in the VW group. The system only marginally touches the work organization and is essentially a separate monitoring unit that reports directly to VW. Its main task is to randomly check the product quality. As far as VW’s global quality system and SPS are concerned, there is no reported strategic or institutional misfit that leads to an up-front transfer restraint. However, SPS only covers a limited range of the work organization directly related to the shopfloor. Above this level, Skoda’s home plant’s work organization is not considered for transfer because of the combined effect of strategic and institutional distance. Apart from the fact that the polycentric staffing policy and low resource endowment make a comprehensive transfer home plant work organization and human resource profile difficult, the institutional and strategic distance seem to rule out a comprehensive transfer effort. Both the Indian production manager and the Czech expatriate share the conviction that above all institutional incompatibilities rule out a replication of the work roles as found on the home plant’s shopfloors. The Czech expatriate sees a big gap or mismatch with regard to the supervisor work roles. In the home plants, supervisors govern high respect and have about 50 to 60 people reporting to them. In India, a supervisor is generally a young man, ‘a beginner who knows almost nothing’ (SAI 1). Apparently, key work roles of the home plant’s work organization such as the supervisor role required a specific human resource profile in terms of qualifications and status expectations, which are either not available or incompatible with the Indian institutional context. However, the transfer restraint not only follows from the sheer fact of institutional incompatibility between home and host context. In addition to the institutional distance, the strategic context of the Indian

232 Hybridization of MNE Subsidiaries

site also rule out investments to replicate the elaborate home-plant schooling and training standards Skoda is renowned for in the Czech Republic (Lyle, 2002). The modest volumes and the size of the operation, at least at the initial stage, do not allow the establishment of comprehensive three-year schooling and vocational training programmes as Skoda practices at home (SAI 1). Thus, the absence of a comprehensive transfer effort also has to do with a combination of both institutional and strategic distance ruling out a comprehensive replication of home country the work organization and the corresponding human resource profiles. Thus, while VW’s global quality system and the SPS aims at structuring certain aspects of the work organization with implications for the human resource profile and while SAI’s mother plant in Mlada Bosleslaw serves as a limited point of reference – especially for personnel transfers from India – there is no holistic or comprehensive replication effort of any foreign/parent plant’s work organization and human resource profile. Instead, the main focus of the transfer is to set up the two core work concepts, that is, putting their formal structure and monitoring tools in place to be able to track performance levels. The Czech expatriate frankly admits that his transfer ambitions and ability doesn’t go any further and repeatedly states that the human resources profiles and ‘habits’ cannot be changed to those in the Czech Republic or in Germany (SAI 1). Comments by his Indian counterpart also confirm this set up scenario: We have one production General Manager here, so basically he is here to support guide and see everything is happening well. But ultimately the interaction on the shopfloor, the culture, it has to be set by the Indians. (SAI 2) An important explanation of the transfer restraint is Skoda/VW’s polycentric staffing approach. Given the low presence of expatriates, the set-up of the SAI’s work organization and human resource profile is largely left to Indian management. Although SAI officially declares that it has chosen a Greenfield site to build up its ‘own little thing’ and although SAI is from the outset a wholly owned subsidiary, SAI relies most extensively on host country nationals for the establishment of the site’s work organization and human resource profile.

Misfit The set up of SAI’s work organization and human resource profile involves two sources of misfit. The first potential source is the

Empirical Results 233

institutional misfit between the limited foreign/parent template and the local/host context. The second source is the institutional misfit between the local/host work organization and human resource profile established by Indian managers and the foreign/parent context. Regarding the latter, it should be noted that Czech expatriate is highly critical about a number of local/Indian work dispositions. This, however, does not trigger any transfer ambitions or demands for change. For example, the enormous hierarchical distance between managerial levels, particularly between supervisors and workers/operators is critically reflected and attributed it to the caste system (SAI 1). In this context, the Czech expatriate mentions that in the beginning when everything had to get started quickly, he personally helped building the cars. However, this physical involvement met the disapproval of his fellow Indian engineers. They were appalled by it and found it unsuitable. Moreover, the Czech expatriate complains about low productivity levels among the least educated, a missing identification with the company among employees and the tendency of his Indian counterparts to overstaff the operation. He also stresses the narrow professional focuses and demarcation consciousness among different staff which make personnel transfers across departments difficult (SAI 1). However, although he finds a number of ‘local habits’ problematic, he has no intention to change them for two reasons. First, he accepts them as ‘Indian habits’, habits that are rooted in the host context and not for him to interfere with. Secondly, he has to run the whole production set-up on a very tight budget (SAI 1). This implies limited resources for personnel transfers for further training or expatriate involvement. After all, he is the only expatriate on site responsible for production and is busy with getting basic systems implemented and monitored. As a sheer matter of necessity, he has to leave substantial aspects of organization building in the hands of his Indian counterparts. While the Czech expatriate makes no effort to transfer a comprehensive home plant work organization and human resource profiles, he does have the mission to implement a basic shopfloor work template and related performance monitoring mechanisms. Although there is no reported strategic distance hindering transfer and implementation of SPS, there are some perceived institutional misfits. First, the skill levels of the newly recruited ITI graduates are rated as insufficient. This is a deficiency shared by the Indian General Manager (SAI 2) and attributed to the weak components of practical training in the Indian vocational or industrial training system. Moreover, as the operators are newly recruited, they have not worked for another company and are

234 Hybridization of MNE Subsidiaries

not familiar with the operation of systems such as SPS. At the same time, Skoda’s home vocational training programme is also not available for training. However, apart from these skill deficiencies, the operators’ work dispositions, mainly ‘attitudes and discipline’ are rated good (SAI 1). Second, in contrast to workers who mainly lack practical training and skills, the managerial ranks and particularly the lower managerial levels are seen to lack motivation and proper adherence to the SPS. From the Czech expatriate’s point of view, ‘work attitudes’ of lower managerial levels, most notably ‘officers’ are poor. He sees officers more concerned with salary and status matters than with the work. He is also highly critical about their employment history marked by frequent ‘job-hopping’. Apart from motivational problems there is another problem related to the fact that many of the officers have worked for other auto companies before. According to the Czech expatriate the implementation of SPS is particularly difficult because officers are used to other systems which were ‘deeper under their skin’ (SAI 1). However, these reported misfits can also be read in a different light. No one of the newly recruited employees has come from a traditional obstructive local/host working culture. The recruited workers are newly recruited and the managers are familiar with systems typically practised in international automobile companies. Working with experienced Indian managers implies a comparatively lower institutional distance with regard to work dispositions such as the taking of responsibility or quality assurance issues. This may explain why the Czech expatriate complains comparatively little about such problems.

Mode of recontextualization No adaptation of foreign/parent template to the local/host context

SAI’s transfer effort is limited with regard to the work organization and the human resource profile. It is largely restricted to the implementation of the SPS and VW’s global quality systems. Apart from the implementation of these systems, that see little adaptation to the local/host context, Skoda’s polycentric staffing policy implies that there are not many foreign/parent demands that can be adapted because there are hardly any expatriates to pose such demands. Although the expatriate in the area of production is critical about certain ‘Indian habits’ or work dispositions, he sees this misfit as a matter of personal opinion, rather than a reason for change or intervention. In addition, he is hierarchically more or less on an equal footing with his Indian managerial colleagues. So, again, there are few foreign/parent demands that can be adapted to the local/host context. As far as the implementation of the SPS work

Empirical Results 235

concept and the VW quality system is concerned, the Czech expatriate is determined to implement it without deviation.

Adaptation of local/host context to the foreign/parent template

In those limited areas where a foreign/parent transfer is sought, the local context of the site is adapted to the template’s requirements. This adaptation is achieved through a selection and creation. The selection involves the recruitment of specific human resources in the host context. In addition, it is based on limited personnel transfer in focus areas and using experienced Indian managers as trainers. But let us take a closer look. As far as managerial levels are concerned, SAI makes an effort to headhunt and recruit Indian managers who have worked for other international auto manufacturers before. Although, the Czech expatriate complains about the recruited manager’s deep-seated familiarity with other international auto manufacturer’s systems, the Indian General Manager of production is convinced that the recruitment strategy has facilitated the transfer of SPS (SAI 2). Similarly, although both the Czech expatriate and Indian manager identify skill deficits and training requirements regarding the newly recruited ITI-qualified workers, their general attitude and discipline are rated conducive to establishing SPS. Ours being a new plant, we had no problems of adaptation because it was new. The operators were new, though we have the managing people, the engineers, managers, and assistant managers they also were exposed to such concepts. Like we have people from Daewoo, we have people from Ford, so they are also exposed to this type of production systems. Maybe there is some slight fine tuning required, which is not a major problem. So the operators being totally new, so we began with these systems, so we not had any resistance. In my past experience I have seen this resistance coming through because I was in my 15 years, my old company, a 30 to 35 years old plant, there was tremendous resistance for these changes. I would say, there was a different work culture. Totally, Indian work culture, where there were no controls, no systems, proper system set-ups. Now, when new concepts then come in there comes in a lot of resistance for change. . .. In our case nothing of this type, because it was all-fresh. We began with this thing so there was nothing. (SAI 2) A second corner stone of implementing transferred systems is a limited transfer of personnel. The personnel transfer scenario comprises of two building blocs. Before the actual set-up of SAI, an Indian start-up team is

236 Hybridization of MNE Subsidiaries

transferred to the home country for training at Skoda’s mother plant in Mlada Boleslav. This initial team comprises 20 Indians, including operators/workers and officers. For the site set up, the Indian team returns to India together with a Czech team. The two teams’ responsibility is to kick off the production in cooperation with their Indian counterparts. The Czech team’s involvement is restricted to a relatively short time-frame, however. After the Czech team leaves, the ongoing establishment of SPS is left to the Indian start-up team and the remaining Czech expatriate in production (SAI 2). Thus, after this initial transfer only one expatriate remains at the site. His restricted, but focused role, is to monitor the implementation of core systems most important of which the SPS. A similar focused transfer takes place with regard to VW’s global quality system. In this regard, an Indian quality specialist – whose job is to establish and conduct the global quality control – is transferred to Germany for training on VW’s global quality system. Yes we do have an audit which is like a global audit. This person is trained in Germany. He is tuned or calibrated, I should say, to the same level of knowledge and ability to identify defects in line with the VW auditors and this person picks up a vehicle at random every day and audits it. And then he gives the feedback and that gives us a feeling of what exactly we are producing. (SAI 2) While SAI sees regular personnel transfers to establish the home plant’s production system and to align the local human resource profile with its core requirements, there is some sense of frustration on the part of the Czech expatriate. He complains about too few personnel exchanges and blames it on the tight budget for SAI’s set-up. He even argues that had the site been a VW venture, there would have been more resources available, more technical personnel exchanges and more expatriates transferred. Referring to himself, he states, ‘one man here is not enough’ (SAI 1). In the absence of more intensive expatriate transfer, SAI strongly relies on host country nationals who are briefly trained at Skoda Mlada Boleslav and/or have worked before for another international auto company in India. Thus, in the absence of expatriates and home plant training facilities, operators are mainly trained on-the-job and familiarized with SPS through Indian officers and managers. In addition to training on-the-job, there is also some on-site-classroom training on ‘professional habits’ in the areas of safety, protection, technical aspects (SAI 1). While there is some limited foreign/parent transfer effort targeting certain aspects of SAI’s work organization and human resource profile, most

Empirical Results 237

aspects of the work organization and human resource profile are created in line with local/host context institutional demands and conditions. Clearly, as far as SPS’s formal structure and measurable aspects of the required work behaviour are concerned, the local site follows the foreign/parent template. At the same time, most behavioural patterns and work dispositions remain local/host context specific. Their structuration is entirely in the hands of Indian managers, involving little expatriate interference. Such an assessment is also suggested by the interview with the Indian General Manager of production: Rules, discipline or the shopfloor procedures etc. are all decided by Indians and the day-to-day interaction is among Indian officers or engineers. So I would say it is basically an Indian culture with the support of broad guidelines from Czech as to how certain things have to happen. We have the Skoda Production System, quality systems these are technical systems or the way how to produce which these guidelines tell. But incorporating and developing a culture, it has to be a local culture; it cannot be a foreign culture at al. But there are certain good things, which have to be incorporated and they have been taken care of, all the good aspects have been incorporated. (SAI 2) Initial personnel transfers, training and some limited change effort by the expatriate in day-to-day interaction apart; there is no massive effort to bring local skill levels and basic work dispositions of the workforce in line with parent company standards. On the one hand, this has to do with the fact that there are hardly any expatriates involved on a day-to-day basis, trying to reshape basic work dispositions of the Indian human resources. On the other hand, the selection of Indian managers who worked for other auto multinationals before, imply less of an institutional distance and socialization need as compared to what the early movers into the Indian market have felt. Thus, the possibility to recruit Indian managers with prior work experience in an auto multinational probably renders personnel transfers and large scale training and socialization efforts a less pressing requirement. We can therefore conclude that there are two modes of recontextualization: Adaptation (creation) of the work organization and human resource profile to meet the foreign/parent template plus an adaptation (creation) of the work organization and human resource in line with local/host demands and conditions. The adaptation of the local context to the foreign/parent template is facilitated by resilient conditions that are

238 Hybridization of MNE Subsidiaries

constituted by the Greenfield nature and the selective recruitment of experienced local human resources.

Outcome In the case of SAI’s work organization and hybridization profile it makes sense to differentiate the ‘hybridization outcome’. Indications are that SAI succeeds in transferring the SPS and its global quality system without much deviation. According to the Czech expatriates and Indian manager, SPS as well as the global quality system are implemented. Yes, our engineers have been trained there. They do understand the advantages of this and they like to adopt it without any alternation from there because this is the base system. And of course there is monitoring how it is going. (SAI 2) Although the Czech interviewee states that it is at times difficult to get the new systems established with managers and engineers who worked for other Auto MNE before, a factory-visit confirms that SPS’s main structural features and the global auditing systems are implemented (SAI 3). For example, in line with the SPS, SAI follows a team concept on the shopfloor with seven to eight operators making up team. One of the team members is a team leader taking care of ‘moment-to-moment problems’. As far as the integration of indirect tasks is concerned, housekeeping (5S), self-certification and quality audits are implemented in line with the SPS and VW quality system template. Worker flexibility charts indicate efforts to achieve a polyvalent skill profiles. Thus, the limited transfer intent comprising a few of Skoda/VW systems, their moderate local misfit in the local/host institutional environment and the adaptation of the local human resource profile to the foreign/parent template facilitate this imitation. Focusing mainly on quantitatively measurable practices and outcomes, their formal implementation is relatively easy to track and achieve. However, while the formal systemic features of SPS are transferred and imitated, other aspects of the work organization and human resource profile, follow typical local/host patterns. For example, there are no indications that Skoda transfers its home plant supervisor–worker relations. There are strong indications that basic behavioural patterns on the shopfloor mirror typical Indian patterns characterized by strong professional and social distance. The only mediating factor is the fact that many of SAI’s managers and officers worked for international companies before. It should also be emphasized that SPS covers only a limited range

Empirical Results 239

of the company’s work organization. The remaining aspects are defined and structured by the Indian managers who organize operation without major interference from the parent. For example, when the Indian General Manager Production is asked if he could imagine an operator to be promoted to being an officer he rules out such an option. Q: And would somebody who is an operator be able to graduate to being an officer at some point? A: No, I don’t think so. Generally operators would not be that equipped with to move into an officers’ category. The management requirements are different. Seldom would you find an operator who can take up these responsibilities. I have seen officers who were taken and promoted to officer levels from operators. But you would find a big difference in their approach. It is not so easy; I have seen this personally with these people managing the shopfloor. Their perspective of management is different. Q: So you rather would not do that? A: No, I don’t think so. (SAI 2) This occupational compartmentalization contrasts with the parent plants in the Czech Republic. According to the Czech interviewee, workers at home have the opportunity to graduate into supervisor levels and beyond (SAI 1). In other words, it is convincing when the Indian manager states that apart from broad structural-guidelines, the work organization and human resource profile is culturally Indian because day-to-day interactions take place among Indians with little foreign/parent interference. By the same token, the Czech expatriate’s observations suggest the prevalence of Indian work dispositions in many respects. These are work dispositions he does not want to interfere with, as long as the main systems are implemented and measurable results are satisfactory. The work organization and human resource profile at SAI reflects two origins: a foreign/parent (imitation) and a host contextual one (localization). Taken together, we may call the result hybridization. However, the two aspects don’t seem to be tightly integrated. Related to the transfer of SPS, some formal work organizational aspects on the shopfloor as well as some related work disposition reflect the narrowly circumscribed templates of the parent companies VW and Skoda without much deviation. Most other work organizational and human resource related aspects reflect local/host context patterns, however. To say that many work organizational and human resource related issues follow a local/host pattern, calls for some specification. On the one hand, the Indian managers worked for other auto MNE before. Their socialization

240 Hybridization of MNE Subsidiaries

is, therefore, not ‘typically Indian’. On the other hand, is not easy to define ‘typically Indian’ patterns of work organization and human resource profiles in the SAI case. For at the time of Skoda’s market entry, the Indian economy is marked, by substantial differences between ‘old’ and ‘modern’ companies (SAI 2).

Industrial relations Transfer scenario When SAI is visited in 2003, there is neither the intent to transfer a foreign/parent industrial relations template nor are there any foreign/parent demands in this regard. Given the Greenfield nature of the site, there is also no existing configuration of local industrial relations. The early stage and small size of the set-up mean that there is not even a human resource department in operation which handles such matters. At the same time, the Czech expatriate at the site is busy with getting basic manufacturing systems in place and is not concerned with the organization industrial relations of the young company. The Czech expatriate also stresses that he is not familiar with human resource matters in the Indian context (SAI 1). The polycentric staffing policy and the knowledge required about industrial relations implies that SAI’s human resource management and industrial relations matters are left to the Indian managers. They, in turn, do not feel the immediate requirement to take the initiative of establishing a formal industrial relations structure at SAI. At this early stage of the site, company industrial relations are not an issue, neither for the Czech expatriate nor for the Indian managers. However, the Czech expatriate is well aware of a substantial institutional distance between India and his home country in labour and human resource related matters.

Misfit Without a foreign/parent template for transfer and without foreign/parent demands, there is also no strategic or institutional misfit between a transferred template and the local/host institutional context. Conversely, without an existing local/host company industrial relations template and local/host demands, there is also no misfit between a local/host template and foreign/parent demands. Labour relations are calm and there is no (re)contextualization pressure from any side. Although there is some institutional distance between Skoda’s home industrial relations and SAI’s company industrial relations, there is no perceived misfit on the part of the Czech expatriate triggering demands for change. After all, the local site has no labour representation and does not experience any industrial conflict. As the site is small, new

Empirical Results 241

and located on the premises of a Siemens factory, there are also no external institutional demands or needs to respond to pressures from any particular industrial relations actor. Also, the location and the size of the operation do not make it particularly visible and attractive for external union activists to approach it. However, the Indian manager interviewed is well aware that eventually such local/host institutional demands can arise from within or outside the company. The Indian management expects that at some point there will be some institutional misfit as the company grows. SAI’s informal industrial relations will come under recontextualization pressure as the site grows in size, gets older and employees align with the local/host institutional conditions and demands (SAI 2).

Mode of recontextualization No adaptation of foreign/parent templates or demands to local/host context? As neither foreign/parent templates are transferred nor foreign/parent demands posed, there is also no contextual misfit, recontextualization pressure and adaptation to the local/host context.

No adaptation of local/host template or demands and conditions to foreign/parent context At the time of research, SAI has not formalized its industrial relations set up. Matters that fall into the area of industrial relations are dealt with in an informal way. This mode of industrial relations largely reflects the young age and small size of the operation. The industrial relation set up sees no mismatch with foreign/parent demands calling for an adaptation. There is, however, some awareness that the status quo of SAI’s company industrial relations is likely to face at some point local/host institutional demands for change. While no pro-active steps are taken to meet prospective local/host institutional demands – either by SAI’s own labour or outside union activist – there are indications that the company management will, in the long run, take an initiative to form an internal union. The Indian manager interviewed envisions such a path, common to modern Indian enterprises in the formal sectors. Unsurprisingly, the creation of an internal non-affiliated union is favoured. Such an internal union would not only help to reduce the risk of an intrusion of an external union and but would also serve as tool for internal communication and to deal with labour grievances in a formal way. Q: Do you have a union here? A: No we don’t have a union as of now.

242 Hybridization of MNE Subsidiaries

Q: This is a policy, you have a non-union policy? A: No, this is a new company; we would not like to have a nonunion policy because a union is necessary. A union is necessary from worker’s point of view because, because as a bargainable category of people they need to have some voice. But not only that, there would be certain things where operators would not feel free to come directly to management, so that’s where the union comes into the picture. We definitely, Q: Sooner or later A: Definitely, I think it will come into the picture. Q: The only thing is, whether it will be a non-affiliated union? A: Yes, yes, we hope so. A union has to be there and it has to be a good union. That’s what one would expect because management also prefers a union. Q: I mean some companies in India, they have taken the initiative as management and said we are starting a union for you and thereby they were able to control the whole thing? A: Yes, yes it is essential. If you have people from outside, they would not understand the full objective of business. And they would have their own personal objectives, which will clash with the business objective and there are all sorts of problems. And ultimately both sides end up suffering, it is not just one side, both will get affected. (SAI 2) Given the polycentric staffing policy at SAI, the Czech expatriate’s lacking familiarity with industrial relations management in India and the expected emergence of local/host institutional demands as the site grows bigger and older, it is very likely that industrial relations will eventually follow a typical local/host institutional pattern. The Indian manager indicates that the site will probably use a common Indian template of industrial relations, namely a non-affiliated internal union. Applying this template will be enabled by the Greenfield nature of the site.

Outcome When SAI is researched, the company does not pay much attention to the establishment of company industrial relations. There is neither a foreign/parent template to be transferred nor are there parent demands for change. SAI’s expatriate management also shows no ambition to change the local scenario or implement home company practices. However, a typical host context template is not established, either. The site basically lacks a formal industrial relations structure and labour representation.

Empirical Results 243

In addition, it has experienced no labour unrest so far. This status is neither a reflection of a non-union policy on the part of the parent Skoda nor a reflection of a local management strategy. It rather has to do with the fact that the site is at an infant stage, small in size and relatively hidden on the Siemens compound. Local/host institutional demands have not yet been voiced, demanding a particular solution or change to the status quo. However, indications are that the formation of formal industrial relations will be left to the Indian management and will sooner or later take on a typical host context form. Specifically, the form of a non-affiliated internal union, as it has become common in modern industrial enterprises in India. We can, therefore, call the outcome as customized to SAI’s infant condition with a high potential for localization, however.

Supplier relations Transfer scenario Apart from the supply of some consumables (i.e. fuel and other liquids), SAI has not established any supplier relations in India. The SKD site relies at the time of research entirely on the import of SKD-kits from Skoda’s parent plant, Mladá Boleslav. Because of its early stage, the site has not set up any local supplier relations (SAI 2). What is more, there is neither a local/host nor a foreign/parent template defined for the site’s prospective supplier relations. A transfer of Skoda’s home plant supplier relations (marked by modular manufacturing concepts, involving: integrated suppliers, a tiered supplier structures, and JIT supply logistics) is ruled out by the strategic distance between what home supplier relations require in terms of production volumes and what the host strategic context has to offer in the first years of operation. Against the background of modest demand in India, SAI decides to start carefully with a SKD set-up to be turned into a CKD facility. As volumes are low, most parts and components are received in a knocked down fashion from home. This means, in turn, that local/host suppliers play a marginal role. For as long as production volumes are low, the preference is on importing knocked-down-kits. From a pure economic point of view, it makes no sense to develop an elaborate local/host supplier base for SKD/CKD sites. Also for potential suppliers, unless they are already in India, it is unattractive to set up local manufacturing facilities for low volumes. And this is even more so the case for SAI. SAI is a late comer in the Indian market, many international auto MNEs have already brought their suppliers. Most of these international suppliers in India are eying orders from other auto MNEs, as they

244 Hybridization of MNE Subsidiaries

are suffering from low capacity utilization. Thus, as long as the strategic distance in market demand and production volume terms remains, it is very unlikely that SAI is to develop a tiered supplier structure in India. Similarly, SAI’s modest volumes and long distances supplies ruled out the implementation of home plant JIT supply logistics. SAI’s production volumes imply that the site is mainly performing assembly activities and relies on long distance supplies. This suggests that a number of supplier related concepts that apply for integrated plants are irrelevant for the Indian site. While the strategic distance rules out the transfer of home supplier relations, the lacking definition of a template for SKD/CKD sites also means that SAI has no defined template for implementation. Yet, while there is no template transfer, there are host context and foreign/parent demands and conditions that structure the set up of a SAI supplier relations. Among the most important demands are those posed by the host institutional context. Although local content requirements are lifted in 2002, SAI still faces import tariffs that create an incentive to localize production. Besides, when SAI is established in 2000 local content requirements are still in place and put the company under pressure to establish a local supplier base. At the same time, Skoda/VW also has demands vis-à-vis prospective supplier relations in India. Particularly, the parent VW formulates global policies with regard supplier relations. These global policies involve among other things selection criteria for suppliers and the definition of contractual modes with suppliers. They also included quality polices and standards, prospective VW/Skodasuppliers have to adhere to. In this respect, VW suppliers have to live up to specific quality standards as well as to specific audit and certification systems. Next to host institutional and foreign/parent demands, SAI’s supplier relations have to respond to host context strategic conditions. Although the company meets a good initial reception in the Indian market and considers localizing some production to bring down cost, it will have to tailor its supplier relations to a low-scale production scenario.

Misfit Although there is no defined transfer template for the site’s supplier relations, there are different supplier related demands from the local/host environment and the foreign parent. The host institutional context poses demands in the form of local content requirements and high import tariffs, putting the company under pressure to explore potential suppliers in India. Yet, the set up of local supplier relations is difficult under host strategic demand conditions. For under conditions of modest demand for middle to high-segment vehicles, an increased local value addition,

Empirical Results 245

beyond the localization of labour intensive assembly activities, is uneconomical. As long as SAI produces exclusively for the host market, it will be very difficult to convince international and local suppliers to set up manufacturing units or accept orders for low volume jobs. It is this principle misfit between local institutional context demand and foreign/parent strategic demand for profitable production that impact the development of SAI’s supplier relations. At the same time, any localization effort means that VW’s global supplier policies apply. Among the most important policy requirements is the demand that suppliers in India have to be either existing ‘global suppliers’ or have to qualify to become such suppliers. Both options are not easy to realize for SAI. Drawing VW’s global suppliers to India is not easy, unless these suppliers are already in India. For the low volumes offered by SAI make ‘follow sourcing’ very unattractive. Qualifying as a new VW global supplier is not easy, either, because of VW’s ‘one source policy’. This meant that any prospective global supplier in India would have to outperform existing global suppliers in price and quality. The realization of these VW policies and demands do not fit the host context demand conditions. However, compared to other manufacturers, SAI has the advantage of an established local ancillary sector when it entered India. Many international suppliers who followed other auto MNE are in principle capable of living up VW’s global quality standards and systems.

Mode of recontextualization No adaptation of foreign/parent templates and/or demands to the local/host context As no supplier relations template is defined for transfer, there is also no adaptation of such a template. However, when SAI is established in 2000, the development of local/host supplier relations is still a must, due to host institutional demands. At the time of research, VW/Skoda responds to these demands and explores possibilities for local production. The potential local suppliers, in turn, have to meet VW’s stringent global supplier policies. Despite difficulties in establishing a local/host supplier structure, SAI showed little signs to move away from these stringent policies (low foreign/parent resilience). Simply spoken, SAI has two options: Either it asks global supplies to establish operations in India to supply locally or it takes of other suppliers already in India – Indian or international suppliers. The first option is hardly realistic, unless these suppliers are already in India. This leaves the company to look more closely into the second option of developing some suppliers in India to become global suppliers.

246 Hybridization of MNE Subsidiaries

However, as far as this option is concerned, the company seems to have faced difficulties due to VW’s one source policy and its stringent quality and price demands. MD Hassen states, for example, in 2001: Skoda India intends to have a “real localization” level of about 18 per cent in the Octavia by end-2002. Mr Hassen maintained that indigenization continued to be a challenge, as for some reason Indian vendors even as they succeed in meeting quality norms lose out on the pricing front. (The Hindu Business Line, 2001) When SAI is researched, the company has not achieved its localization goal because it is not willing to adapt its foreign/parent demands to the local/host context. Although VW is, as a latecomer, in the comfortable position of coming into a market where many auto multinationals have either brought their suppliers or have local JVs, VW/Skoda appears slow in taking advantage of them. However, of late, VW engages in talks with potential suppliers in India, including international suppliers such as: Goodyear, Johnson Controls and MICO. As the following abstract suggests, that there are efforts to work with suppliers in India who could qualify as global suppliers: While the list of vendors is not yet finalised, Skoda is in talks with Goodyear for supply of tyres, and with Johnson Controls for sourcing seating systems. Further, cockpit instrument panels and wheels are other components that are under consideration, company executives said. Sourcing of components from India will also ensure localisation of SkodaAuto’s offerings in India. Currently, all the parts of the Octavia that is sold in India are imported. Imran Hassen, managing director, SkodaAuto India said, “We are looking to localise components used in our cars. However, it does not make sense for us to source components just for the Indian operations in isolation. SkodaAuto is interested in sourcing components from vendors who meet its stringent quality and cost requirements. Indian vendors will have to compete with those in China and South America.” Globally, Skoda sources all components for a particular car from the same vendor as it creates economies of scale for its vendors, which in turn gives them a cost advantage. Further, it also ensures the same quality of parts anywhere in the world, as the source is common. (Business Standard, 2003) The press report may be seen as an indication that VW/Soda gives up its ‘one source policy’ at least in India and considered working with

Table 8.1

Summary of SAI’s hybridization profile

Firm PS dimension

SAI

Functional differentiation

Transfer: Foreign/parent strategic demands and local/host institutional demands and strategic conditions to be met Fit/Misfit: No misfit between different strategic demands and conditions, need for their accommodation Mode of Recontextualization: Adaptation of local context (i.e. creation of functional set-up) to meet different demands and conditions Outcome: Customization

Hierarchical differentiation

Transfer: No foreign/parent template transfer, local/host institutional demands posed Fit/Misfit: Minor institutional misfit between foreign/parent demands and local/host context demands Mode of Recontextualization: Adaptation/rejection of weak foreign parent demands; set-up of local site in line with local/host institutional demands Outcome: Localization

Technical configuration

Transfer: Foreign/parent strategic demands and conditions as well as local/host strategic conditions and institutional demands (local content and import tariffs) to be met Fit/Misfit: Minor misfit between foreign/parent strategic and local/host institutional demands Mode of Recontextualization: Adaptation of local context – creation – to meet foreign/parent demands and the host strategic conditions and institutional demands Outcome: Mainly customization (Continued)

247

248

Table 8.1

Continued

Firm PS dimension

SAI

Work organization/ HR profile

Transfer: Selective transfer of foreign template; little foreign/parent demands otherwise Fit/Misfit: Some misfit between foreign/parent template and local/host institutional demands Mode of Recontextualization: No adaptation foreign parent template transferred; some adaptation – selection and creation – of local context in line with template; for the rest set-up responds to local/host institutional context and demands Outcome: Between localization and hybridization (imitation coexists with local solutions)

Industrial relations

Transfer: No local/host or foreign/parent demands or templates Fit/Misfit: No misfit Mode of Recontextualization: Absent company industrial relations reflect early stage, size and the Greenfield site. Yet, likely development is adaptation – change/creation – of local site company industrial relations to meet local/host institutional demands as they arise Outcome: Customization most likely to develop into localization

Supplier relations

Transfer: Foreign/parent and host context demands and conditions Fit/Misfit: Misfits between the local/host strategic context, foreign/parent strategic demands and local/host institutional demands Mode of Recontextualization: Not entirely clear; possible adaptation of foreign/parent policy demands; possibly also some adaptation/rejection of local/host institutional demands; most likely adaptation (creation) of local/host supplier relations in line with different foreign /parent strategic/institutional demands as well as local/host institutional demands and local/host context strategic conditions. Outcome: Most likely customization

Empirical Results 249

suppliers which do not qualify as global suppliers. Overall, there are few signs, however, that VW/Skoda is willing to adapt its global supplier policy demands to host country demands and conditions.

Little adaptation of local/host context to the foreign/parent context There is no indication that SAI can reject host context institutional demands with respect to local content requirements or import tariffs. While local content requirements are dropped in 2002, SAI responds to the host institutional demands and starts to establish supplier relations in India. In this respect the local context is not very resilient for change. As regards the cooperation with local/host suppliers or suppliers in India, SAI will find it much less necessary than the early movers to adapt the local supplier relations to its standards and policies. By the time SAI enters India, the Indian supplier base is developed and well aware of global cost, technology and quality standards.

Outcome The organization of supplier relations is a reflection of the early entry stage of the site. No local suppliers are established and the supply logistics are largely customized to respond to host strategic conditions of low demand with a corresponding low volume task profile. Consequently, there is a strong reliance on long distance supplies of SKD/CKD kits from the Czech home plant. As long as SAI remains a SKD/CKD site, the transfer of home plant supply-logistic concepts, such as JIT, remains meaningless. Rather than being based on a defined transfer template for SKD/CKD sites, SAI’s supplier relations are most likely to develop into a customized response to specific host and foreign/parent context demands and conditions. Although supplier relations are not fully established, indications are that they will be a customized response to the interplay of VW/Skoda’s global supplier policies, the demand market conditions for Skoda’s cars in India, the supply market conditions in India’s supplier sector and the demands of the liberalizing institutional context. The result of such complex responses to different contextual demands and conditions will most likely be customized or novel solutions rather than imitations. At the time of research, however, Skoda’s supplier relations were most of all a reflection of its start-up phase and its task profile as a SKD site (see Table 8.1 for a summary of SAI hybridization profile).

9 Discussion

The goal of this project is to identify the hybridization profile of the production systems of four automobile MNEs’ subsidiaries in India and to explore whether and how possible variations in these profiles can be related to both institutional and strategic distance as well as to the systematic variation in strategic choices on the corporate and subsidiary level. Derived from earlier research different hybridization outcomes are conceptualized as the combined effect of different transfer scenarios, (mis)fits/recontextualization pressures and recontextualization modes. The first crucial addition to hybridization research is to consider the effect of both strategic and institutional distance on transfer scenarios and (mis)fits/recontextualization requirements and thereby on hybridization outcomes. Key questions for exploration are therefore: Are both strategic and institutional distance relevant for hybridization outcomes? Are institutional and strategic distances, respectively, more relevant for the hybridization outcomes on certain dimensions of a production system? The second crucial addition to hybridization research is related to the question how different strategic choices – at the corporate and the subsidiary level – impact hybridization outcomes. Specifically, what is the link between strategic choices and the triad of transfer scenario, (mis)fit/recontextualization pressure and recontextualization mode and consequently the hybridization outcomes? And, whether and how specific strategic choices – the generic strategy of the MNE as well as establishment and equity modes – interact with transfer scenarios, institutional and strategic misfits induced recontextualization pressures as well as recontextualization modes to constitute hybridization outcomes? 250

Empirical Results 251

Core findings – similarities and differences mapped and explained Hybridization outcomes compared across dimension and cases The comparison of the hybridization profile across different production system dimensions reveals the following pattern. The production system dimensions: industrial relations and hierarchical differentiation have the strongest tendency towards localization outcomes. In contrast, the dimensions of functional differentiation, technical configuration and supplier relations feature the least amount of localization outcomes. Specifically, while the dimensions of technical configuration and supplier relations have a high incidence of customization solutions, the dimension of functional differentiation have a more moderate incidence of customization. Compared to all other dimensions the dimension of work organization and HR-profile features the highest occurrence of hybridization outcomes (see Table 9.1). Looking at the outcome patterns across cases reveals the following differences. While MSI and FI show a high incidence of imitation, MBI and SAI feature much less imitation outcomes and instead much more customization solutions. The biggest difference can be identified between MSI and SAI. The MSI case has come closest to imitation. This contrasts with SAI which has come closest to localization solutions as an overall outcome.

Institutional and strategic distance and different hybridization outcomes The analysis of the four cases’ hybridization profile (Chapters 5–8) shows that both institutional and strategic distance have a distinct explanatory value for understanding hybridization outcomes of the international automobile companies’ production subsidiaries in India. The study suggests that strategic distance – different task profiles and different supply and demand market conditions in different contexts – as well as institutional distance – different institutional conditions and demands have a substantial impact on the hybridization outcomes. The case studies suggest that the strategic distance or proximity between different MNEs’ sites is associated with the definition or development of templates, their transfer and the recontextualization pressure they face. For example, both SMC and Fiat develop and define transfer templates as the modest strategic distance between different sites creates a strong incentive to do so. At the same time, in both cases the strategic distance in local/host supply market conditions is too pronounced to

252

Table 9.1

Hybridization outcomes of all four cases compared

Dimension/case

MSI

FI

MBI

SAI

Functional differentiation

Imitation

Imitation

Customization

Customization

Hierarchical differentiation

Localization

Imitation

Hybridization

Localization

Technical configuration

Between imitation and customization

Between customization and imitation

Customization

Customization

Work organization & HR-profile

Between imitation and hybridization

Hybridization

Between hybridization and customization

Between localization and hybridization

Industrial relations

Between hybridization and localization

Localization

Localization

Localization

Supplier relations

Between hybridization/ customization to increasingly imitation

Between customization and imitation

Customization

Customization

Empirical Results 253

transfer all aspects of the foreign/parent templates. Low labour costs in India – and in the FI case additionally lower market demands – prohibit a full transfer of templates and suggest, instead, a transfer restraint, an adaptation of the original template or customized solutions (e.g. with regard to the sites’ technical configuration). Likewise, institutional distance proves to have an important explanatory value for understanding the hybridization outcomes. Institutional distance impacts transfer scenarios and (mis)fits/recontextualization pressures. In the MSI case, for example, the institutional distance is a core reason why SMC is invited and even pressed to transfer its template. For host institutional conditions in India are rated as so obstructive that a foreign/parent template transfer is sought as a deliberate alternative to host context conditions. Cases in point are transfers of the work organization and labour relations. However, in the very same company and on the same dimension of the production system, some institutional patterns between the home and the host context are so distant that the transfer of certain concepts fails, or is not even pursued. An example is SMC’s consensual approach to decision making, which is ruled out by extreme hierarchical socio-professional demarcations in India. Conversely, in the MBI case, institutional distance induces a recontextualization pressure from the foreign/parent side on the local/host template, thereby impacting the hybridization outcome. In contrast to the assumption posited, the research shows that contextual distance not necessarily implies a transfer restraint with regard to the transfer scenario. The opposite may also be the case. A foreign/parent company may decide to transfer a template because the local/host strategic conditions are so adverse that there is no way to work with them as they are. Contextual distance can, therefore, lead to both, a triggering of foreign/parent transfer or demands, where no transfers or demands are initially intended and a transfer restraint, where a transfer is initially intended. An example for the former is the MBI case, where German expatriates find local institutional patterns in the work organization unacceptable and pose demands for change. The latter we see in FI, where adverse local institutional conditions prevent Fiat from transferring high involvement elements of the work organization. In line with the proposition posited, the cases presented show that both institutional and strategic distances impacts transfer scenarios or lead to (mis)fits/recontextualization pressures thereby impacting hybridization outcomes. The study also shows that recontextualization pressures induced by contextual distance can operate in different directions. For instance,

254 Hybridization of MNE Subsidiaries

just as foreign/parent templates and demands in subsidiaries can come under recontextualization pressures, if distant and incompatible with local/host context (as we see in FI and even MSI), so can local/host templates initially used in subsidiaries (as we see in MBI) come under recontextualization pressure, if too distant from or incompatible with foreign/parent context demands and conditions. This finding is very much in line with the work of New Institutionalists (e.g. Kostova and Roth, 2002) who have stressed the double embeddedness of subsidiaries in foreign/parent and local/host institutional contexts. The findings add, however, to insights from New Institutionalism that in addition to the impact of institutional distance we need to look at the impact of strategic distance on transfer processes in MNEs. Broadly speaking this suggests adopting a more complex perspective of subsidiary embeddedness, placing them in institutional and strategic contexts as well as in parent company and host country contexts. The case analysis also suggests that some dimensions of a production system are more severely affected by institutional distance and others more by strategic distance. For instance, the high incidence of localization outcomes on the dimensions hierarchical differentiation and industrial relations can be related to entrenched institutional patterns in the host context that penetrate deeply into the companies, irrespective of generic or internationalization strategy and experience related template availability and transfer. Similarly, the high incidence of hybridization outcomes on the dimension of work organization and HR-profile is strongly related to the interplay of whatever is being transferred from or demanded by the foreign/parent and local/host institutional context. Regarding the latter, it is most notably the strong socio-professional demarcation produced by the Indian social stratification and education system that pulls transfers towards hybridization solutions. In fact, in no case in the research sample – no matter how strong the foreign/parent’s transfer intent or demand for an alternative work pattern – is the foreign parent able to overcome this institutional or societal effect. Regarding the functional differentiation, the technical configuration and supplier relations host/local institutional demands or misfit are comparatively less relevant than local/host strategic conditions for the outcome. On these latter production system dimensions, the host country demand and supply market conditions and their strategic distance to home or third country operations, are causing a high incidence of customization outcomes. Thus, while the outcomes of functional differentiation and technical configuration appear to be more affected by strategic distance;

Empirical Results 255

hierarchical differentiation, labour/industrial relations and certain aspects of work organization appear to be more affected by institutional distance. It is important to note, however, that although some dimensions (and even some aspects within certain dimension) appear to be more ‘institutional-context-sensitive’ and others more ‘strategic-contextsensitive’, it would be wrong to suggest that strategic and institutional distance can be neatly separated in their effect on specific production system dimension as has been suggested by some (e.g. Pil and McDuffie, 1999). Indications are that both of them almost always play a role or play together in affecting hybridization outcomes. A case in point is the establishment of MBI’s work organization. In this case, it is essentially the strategic distance – lower volumes and labour costs – that deter the transfer of a home plant group-work concept because its application only financially paid off in a high volume and a high wage context. Simultaneously, the much more pronounced socioprofessional demarcation in India renders the transfer of the very same group concept – implying substantial equality among group members – not a viable option. Similarly, in the SAI case the Czech expatriate perceives a problematic institutional distance between the qualifications and skills of human resources available in India and those he is used to from the home context. Yet, given the low volumes of the Indian site, the strategic distance to the home plant context, there is no possibility to set up an alternative training programme and to adapt the local human resources to foreign/parent institutional standards. Thus, no matter whether there is an institutional fit or misfit, if there is a strategic misfit; it may be simply not possible to transfer a certain template. Conversely, there may be some strategic fit to transfer a template, but institutional conditions may be too adverse to transfer it. The research evidence suggests that both contexts need to be considered in combination. While one may take the prominence over the other and may be more relevant on certain dimensions the other context in the background always matters. It should be noted that institutional and strategic contexts, while analytically distinct, may be to some extent two sides of the same coin. For example, lacking or weak supplier relations can be seen as a distant or an un-conducive institutional context (if we look, for example at typical patterns in structural or contractual terms). At the same time, they can be seen as a distant or un-conducive strategic context in terms of supply market conditions (supply market conditions that are not able to provide the right quality, quantity and price of input factors). The same holds true for other input factors, such as human resources. Based on

256 Hybridization of MNE Subsidiaries

institutional conditions, such as education and industrial training systems, human resources are equipped with specific qualifications, skills and work dispositions. At the same time, these institutional conditions play a crucial role in constituting markets for human resources, i.e. the quality, quantity, and price at which human resources are available. Or looked at the other way around, market conditions, for example demand for specific human resources, shape to some degree specific institutional profiles of the education systems. Finally, what contextual distance (be it institutional or strategic) cannot explain, or at least only partly predict, is in what direction it will impact transfer scenarios, misfit/recontextualization pressures and why a certain recontextualization mode kicks in. Clearly, when the contextual distance is low or absent there is little misfit and recontextualization pressure and no recontextualization is required in the first place. However, strategic and institutional distance in isolation, without considering the constraining and enabling conditions of specific firms’ contexts and strategic orientations, on the one hand, and the constraining and enabling conditions of a specific local/host context conditions, on the other, cannot explain: When contextual distance leads to a transfer restraint? Under which conditions contextual distance translates into misfit? When a perceived misfit translates into recontextualization pressure? And finally, by what mode of recontextualization a misfit/recontextualization pressure is resolved. As posited these questions can only be answered by looking at willingness and ability of the foreign/parent to invest resources and the nature and resilience of the receiving local/host context. While the former aspect is crucially related to strategic choices at the corporation and subsidiary level (see next paragraph), the latter appears to be related to the level of embeddedness of local/host demands and conditions. It appears that the more adaptations threaten fundamental local/host identities, norms and resources the less resilient the local/host context and the more resource intensive the adaptation. A case in point is the socio-professional demarcation in India proving not very adaptable to foreign/parent demands or templates. Here, the most comprehensive adaptations where achieved by MSI which massively invested resources to adapt local/host patterns to foreign/parent templates and demands.

Strategic choices and different hybridization outcomes The second mayor research objective of this study involves the question: Whether and how strategic choices – the generic strategy of the MNE as well as entry modes – are related to transfer scenarios, contextual

Empirical Results 257

(mis)fits/recontextualization pressures, modes of recontextualization and consequently hybridization outcomes? See Table 9.2 for an overview of the strategic choices of the four MNEs.

Generic strategies With regard to generic or global product strategies it is expected that particularly the cases featuring cost focus strategies – such as SMC and Fiat in emerging markets – are more likely to define and transfer a foreign/parent templates, to face comparatively less strategic misfit and recontextualization pressure, to be more determined to adapt the local/host context to the foreign template and to feature, as a result, higher degrees of imitation. Conversely, it is expected that cases with a differentiation strategy or strategies that contain uniqueness or wide segment scope elements – such as MB and Skoda – are more likely to face strategic distance between different sites, less likely to define and transfer a foreign/parent template, more likely to experience strategic misfit induced recontextualization pressure and, should a transfer occur, more flexible with regard to adapting their templates or demands to the local/host context. In short, these firms were expected to feature higher degrees of localization, hybridization or customization outcomes. A first analysis of the empirical material confirms this proposition. MSI with its cost focus strategy has the highest degree of imitation. SMC has a foreign/parent template and seeks to transfer it. SMC’s template also meets low levels of strategic misfit and strategic context related recontextualization pressure. Moreover, the company invests more than the other companies resources to adapt the local/host context to the foreign/parent template. Similarly, Fiat has a foreign/parent template and transfers it to the Indian site. However, despite having designed a template for similar – developing country environments – Fiat experiences a substantial strategic misfit as the company underestimates the strategic distance between its Indian and other World Car operations. Given this strategic distance, Fiat decides to adapt – deselect – some aspects of its template to make it fit with the site’s extant Brownfield set up and the Indian strategic context of low demand and low labour cost market conditions. In contrast to the proposition posited, Fiat faces a substantial degree of strategic misfit and is less willing and able than SMC to invest and adapt the local/host context to its foreign/parent template. Here, the generic strategy only partly predicts the transfer scenario, misfit and recontextualization mode because other factors come into force such as the Brownfield entry mode and Fiat’s modest market success in India.

258

Table 9.2

Overview of strategic choices of the four MNEs MSI

Strategic choices at corporate level Generic SMC is between cost strategy focus & cost leadership SMC’s emerging market strategy rests more on cost focus Strategic choices at subsidiary level Entry time 1982

FI

MBI

SAI

Fiat combines cost leadership & differentiation

MB features differentiation

VW combines differentiation & cost leadership

Fiat’s emerging market strategy rests more on cost focus (world car project)

MB’s emerging market strategy rests more on differentiation focus

VW’s emerging market strategy with Skoda rests more on cost leadership

1997

1994

2000

Establishment mode

Greenfield

Brownfield

Greenfield

Greenfield

Equity mode

Joint Venture

Joint Venture > wholly owned

Joint Venture > wholly owned

Wholly owned

(From minority to majority JV)

(From majority JV to wholly owned)

(From majority JV to wholly owned)

(Wholly owned from the outset)

Empirical Results 259

Thus, while there is evidence to suggest that SMC’s and Fiat’s availability and transfer of a foreign/parent template is related to these firms’ cost focus strategies and the expected low strategic distance between their different global operations, the generic strategy seems to be a weak predictor for the actual strategic fit/recontextualization pressures and modes. The FI case shows that even templates that are devised specifically for developing countries may not strategically fit all developing countries’ conditions. The FI case disconfirms the Japanization literature’s Lean Production wing in a particular convincing way because it shows that ‘best practises’, even those customized for specific business environments, run the risk of facing a high recontextualization pressures. By the same token, the cases of MBI and SAI show that the generic strategy alone, without an additional attention to MNE internationalization experiences and internationalization strategies do not suffice to explain foreign/ parent template availability and transfer. Mercedes-Benz and VW have a differentiation and a differentiation and cost leadership strategy, respectively. At first glance, these firms’ overall product strategies, with their element of differentiation, seem to predict their sites’ hybridization outcomes. Both companies neither define/develop nor transfer foreign/parent templates to their Indian sites. They also abstain from transferring their home site set-ups due to a vast strategic distance. They also appear to be flexible to adapt their production systems to the demands and conditions of the local/host strategic context, featuring as a result low degrees of imitations as outcomes. However, a closer look at their generic strategies for emerging markets suggests that it is not selfexplanatory why these two cases come to so few imitation outcomes. After all, both companies feature no pure differentiation strategy and also have – with their focus (MBI in emerging markets) and cost (Skoda) element in their strategies – some incentive to develop templates. Particularly, Mercedes-Benz’s focus element in the strategy and its recent strategic reorientation towards Asian markets involves the set-up of a number of Asian CKD sites, with similar task profiles, suggesting strategic proximity and some incentive to develop and transfer a foreign/parent template. Interestingly, such a template is, at the time of research, even in the making. However, as MBI is established as one of the first of the international CKD sites, Mercedes-Benz has not yet defined such a template and probably lacks the experience to do so. In other words, when MBI is established the strategic shift is at an early stage and no such template exists. Instead, there is a vast strategic distance to home operations, no other foreign/parent template available or transferred and consequently little strategic (mis)fit/recontextualization pressure resulting from such

260 Hybridization of MNE Subsidiaries

a transfer. Similarly, in the Skoda case one could have expected a template transfer to SAI. Such an expectation is even more justified given VW’s substantial internationalization experience and a portfolio of lower cost vehicles for emerging economies. Specifically, Skoda, with its cost-leadership orientation has the strategic task of exploring emerging markets in the format of CKD operations. However, Skoda is acting rather independently in setting up the Indian site and doesn’t have the experience or corporate pressure to develop and transfer a comprehensive template to its infant SKD site. Skoda, like Mercedes-Benz, has no specific template to transfer, has no defined transfer intent and faces a vast strategic distance between its home and host operation, ruling out the use of home plants as templates. Instead of using a foreign/parent template, those dimensions strongly dependent on strategic contexts conditions are adapted to the specific strategic demands of the foreign/parent and local/host context resulting in high degrees of customizations. The cases only partly confirm that generic strategies are relevant for the question as to whether templates are defined/developed and transferred or not (see Table 9.3). The findings support that there is an association between companies with a cost focus strategy and the development and transfer of organizational templates. Clearly, the development and transfer of templates increases the likelihood of coming to imitation outcomes. For, without transfers or demands from the foreign/parent context, there is also no chance of coming to imitations as outcomes. Generic strategies of MNEs play an important explanatory building bloc in understanding hybridization outcomes. However, even the isolated aspect of whether templates are defined/developed or transferred depends on the interplay with other crucial factors such as: internationalization experience, whether developed or developing markets have been mainly served in the past (i.e. are there import restrictions), unexpected strategic misfits and institutional misfits as well as the relevance of other strategic choices at the subsidiary level. Moreover, generic strategies do not suffice to explain the difference across different dimensions of a production system within a particular case. The case discussions above show that a cost focus strategy do not automatically mean that a template covers all dimensions of a production system, that there is the intent or possibility to transfer all aspects of a template, that there are no strategic (or other) recontextualization pressures and that the adaptation of the local/host context will always be the dominant recontextualization mode. First, even if there is a template for transfer, actual transfer may be very selective, given other intervening

Table 9.3

Generic strategy, internationalization experience and template availability for transfer MSI

FI

MBI

SAI

Generic strategy in emerging markets

Cost focus

Cost focus

Differentiation focus

Cost leadership

Strategic distance between units

Moderate between home and Indian operation and also a number of other integrated foreign operations

High between home and Indian operation; low between World Car operations

High between home and Indian operation; low between emerging market operations

High between home and Indian operation; low between emerging market operations

Internationalization experience

High

High

Moderate

Moderate

Comprehensive template availability and transfer

Yes

Yes

No (but in the making)

No

261

262 Hybridization of MNE Subsidiaries

and interacting factors such as entry mode choices and/or institutional constraints and opportunities in the local/host context. There may be, for instance as in the case of FI, templates from the local/host context or from a local JV partner competing with the foreign/parent template. Second, even if there is a transfer of templates, they still may not perfectly fit with the local/host strategic and institutional context conditions. These misfits, in turn, call for recontextualization that can involve as much adaptations of the local/host context or templates as adaptations of the foreign/parent context or templates. Third, although there are indications that companies in which the foreign/parent company defines and seeks to transfer a template are also more willing to adapt the local/host context, there are other crucial intervening factors. Thus, while a cost focus strategy does not automatically imply template availability and comprehensive transfer, low strategic distance and fit across all dimensions and local/host context adaptation as the exclusive recontextualization mode; a differentiation strategy, especially when it is not a pure one (as we find in MB), does not automatically mean an absent foreign/parent transfer intent or demands, large strategic distance and misfit across all dimensions and the adaptation of foreign/ parent templates or demands as the exclusive recontextualization mode. For example, even if a foreign/parent lacks a comprehensive template or decides not to transfer certain aspects of a template, there still may be organizational demands vis-à-vis the local site’s production system. Such demands can be part of globally defined corporate policies or direct foreign/parent reactions to local/host institutional or strategic conditions channeled through expatriates. Strategic and institutional distance may trigger foreign/parent transfers or demands like in the case MBI, although there was initially no defined transfer intent. A locally existing template may also not be in line with foreign/parent context expectations and standards. To what extent such foreign/parent demands are raised, strongly depends on the formulation of global policies and on the presence of expatriates on site, which is related to the overall staffing policy. Moreover, like in the case of a foreign/parent template transfer, foreign/parent demands may be more or less distant or contradictory to the local/host strategic and institutional context. In the empirical study we see in MBI and SAI different recontextualization modes in response to such demands. One possibility is the adaptation of the local/host conditions or the rejection of local/host context demands. The other possibility is an adaptation of foreign/parent context or a rejection of foreign/parent demands. Whether one, the other, or both recontextualization modes become active, depends, as we shall see next, not only on

Empirical Results 263

generic strategies but among other factors on the establishment mode related resilience of a local context, wider host context resilience as well as the equity mode related ability to induce changes and the willingness to mobilize and invest resources on the part of the foreign/parent company. Finally, the findings confirm that looking at an MNE’s overall generic strategy is misleading in understanding hybridization outcomes. Instead we need to look at the strategy of specific brands for specific world regions.

Entry modes – establishment and equity modes This study posits that establishment modes impact all three components constituting hybridization outcome, that is, transfer scenarios, contextual (mis)fits/recontextualization pressures and recontextualization modes. First, it is assumed that Greenfield sites are much more likely to receive foreign/parent templates than Brownfield sites because they lack an existing configuration and because contextual distance is lower in comparison. Second, it is proposed that Greenfield sites either reduce or mediate contextual distance, which reduces the misfit and the corresponding recontextualization pressure in the case of a foreign/ parent transfer. Third, related to the second point, it is assumed that Greenfield sites render an adaptation of the local context to distant foreign/parent templates and demands easier because of lower levels of institutionalization. As a result, it is expected that Greenfield sites feature higher levels of imitation than Brownfield sites. With respect to equity modes it is also suggested that they impact all components constituting hybridization outcomes. The argument is here that the equity mode has an influence on whose management controls the company and whose management can decide: What is or can be transferred or demanded? What is rated as an unacceptable contextual distance or misfit? And to what extent the foreign/parent template and contexts or local/host templates and contexts have to be responded to or can be rejected? It is proposed that the higher the foreign/parent equity, the more likely are we to see the transfer of foreign/parent templates or the influence of foreign/parent demands, the higher the level of misfit/ recontextualization pressures and the more likely an adaptation of the local/host context to foreign/parent templates, demands and conditions will be the dominant recontextualization mode. Combining the two entry mode related propositions, it is suggested that we can expect wholly owned Greenfield sites to feature the highest and Brownfield JVs, the lowest degree of imitation. Applied to the four cases of this study these propositions suggest SAI and MBI to have the

264 Hybridization of MNE Subsidiaries

highest incidence of imitation; and FI and MSI to have the lowest incidence of imitation. Surprisingly the opposite is empirically found. How can we explain this? One important explanation is the generic strategy related difference in the availability and transfer of a foreign/parent template in the cases MSI and FI. But there are other factors that play a crucial role as we shall discuss below.

Establishment modes With regard to establishment modes the assumption is that the Brownfield site FI – in comparison to SAI, MBI, MSI – sees less of a template transfer, faces higher levels of misfit in the case of foreign/parent template transfer and finds it more difficult to adapt the local context to foreign/parent templates and shows, as a result, higher degrees of localization outcomes. Interestingly, this proposition is only partly confirmed by the empirical results. Largely related to its generic strategy FI – together with MSI – belongs to the only two companies in the sample that transfers a comprehensive foreign/parent template and has a high incidence of imitation. At the same time, the comparison of MSI and FI confirms that FI faces more local contextual misfit and more difficulties to adapt the local context to the foreign/parent template, due to its establishment mode. But let us take a closer look if and how the establishment mode impacts the hybridization outcomes. The case comparison shows that the establishment mode has an impact on the transfer scenario because it implies, first of all, the presence or absence of an existing local template (see Table 9.4). We see this in the FI case, the only Brownfield site in the sample, with an extant local template. At the same time, the case comparison also shows that the establishment mode, viewed in isolation, is only a week predictor for the transfer scenario. MSI, MBI and SAI are all Greenfield sites, yet, out of these, it is only SMC that transfers a comprehensive foreign/parent template. FI in contrast, the only Brownfield site, does see a foreign/parent template transfer. The findings indicate that although a Greenfield site implies a need for establishment, this configuration must not necessarily be based on a foreign/parent template transfer or be subject to strong foreign/parent demands. Instead, indications are Table 9.4

Entry mode and local template availability

Establishment mode Local template at site

MSI

FI

MBI

SAI

Green No

Brown Yes

Green No

Green No

Empirical Results 265

that whether a Greenfield site is shaped by a foreign/parent template or subject to strong foreign/parent demands crucially depends on: (1) Whether a foreign/parent template is available and which production system dimensions it covers?; (2) Whether the foreign/parent template fits the local/host strategic and institutional context?; (3) Who (foreign partner vs. local partner; expatriates vs. host context nationals) has the configuration mandate on a specific dimension of a production system? For example, Skoda has practically no transfer template for its Indian Greenfield operation. Skoda has not defined a template for SKD sites and the home plants are strategically too distant to be used as templates. If at all, the Skoda Production System is a very narrowly circumscribed template, which covers only a very limited range of the production system. Most importantly, due to its polycentric staffing approach, the configuration of the site is in the hands of host country nationals who shape many production system dimensions in line with local/host conditions and demands. However, the establishment mode is not only proposed to impact the transfer scenario but also to mediate contextual distance and to have an influence on the recontextualization mode. In contrast to the establishment mode’s proposed impact on transfer scenarios, there is more empirical evidence to support the latter two assumptions. Indications are that the transfer of foreign/parent templates or the realization of foreign/parent demands are substantially easier at the Greenfield sites SAI, MBI, MSI as compared to the Brownfield site FI. Employees at the Greenfield sites have less preconceived notions working against foreign/parent template elements or demands, which contribute to lower levels of misfit. And even where host context notions are in contradiction with foreign/parent demands or templates, the local context proves to be more adaptable to them. This condition can be nicely seen in the comparison of MSI and FI. Both subsidiaries receive a foreign/parent template, however, the institutional misfit is more pronounced in FI than in MSI, which is crucially related to FI’s Brownfield heritage. More importantly, although institutional misfit occurs in both cases, it proves easier to adapt the local context to the foreign/parent template in MSI as the greefield site is largely constituted by a young workforce. Thus, there is evidence to suggest that Greenfield operations render it substantially easier to adapt a local context to foreign templates/ demands as compared to Brownfield sites. These findings confirm earlier work that has shown the relation between establishment modes and transfer successes (e.g. Sharpe, 1997). It should be noted, however, that while the establishment mode has an impact on a local site’s contextual

266 Hybridization of MNE Subsidiaries

resilience, there are other factors that play into the question, which recontextualization mode unfolds. For instance, whether a contextual distance is rated positively or negatively, the foreign/parent’s ability and willingness to invest resources and put through its demands and templates as well as the wider host context constraints penetrating into the local context are crucial. The establishment mode alone is hardly sufficient to explain recontextualization modes, as the local context of a site is embedded in the wider host contexts. In such a perspective even Greenfield contexts are only partly resilient as we see in the case of MSI, where the foreign/parent does not fully succeed to adapt the local context to the foreign/parent template. Despite substantial resource mobilizations, the local adaptability is constrained by the limited resilience of host institutional context shading into the local context of the site.

Equity modes Following our proposition about the impact of equity modes on hybridization outcomes, the expectation is that MSI features the highest incidence of localization and SAI the lowest. The basic rationale behind this is that the equity mode directly impacts whose management is in control and thereby the transfer scenario, the (mis)fit/ recontextualization pressure and mode. However, the empirical analysis of the cases shows that the opposite is the case as SAI has a higher incidence of localization outcomes than MSI. While the configuration mandate plays an important role in explaining this outcome, the proposed connection between configuration mandate and equity mode is only a weak one. The case comparison shows that the equity mode influences the transfer scenario by having first of all an impact on the potential availability of a local template. In the sample, MSI, MBI and FI have a local JV partner and thereby the opportunity or constraint to draw on their local partner’s template or aspects thereof (see Table 9.5). Table 9.5

Equity mode and local template availability MUL

Equity mode

FI

DCI

SAI

Minority JV > Majority JV > Majority JV > Wholly owned Majority JV Wholly owned Wholly owned

Local template Yes, partially available from local partner

Yes, partially

Yes, partially

No

Empirical Results 267

In contrast, SAI, the only wholly owned operation from the start, has no such opportunity. The case comparison shows that the equity mode not only impacts the availability of local/host templates, but more generally, the use/transfer of available local/host or foreign/parent templates. The equity mode has an impact on the transfer scenario because it has an influence on the question who has the configuration mandate for the production system set up (local parent vs foreign parent; host country nationals vs expatriates)? For example, Fiat is very quick to turn its JV into a wholly owned subsidiary. With this move the company is free to unilaterally decide to transfer crucial aspects of its World Car template. For MBI and MSI this is not possible to the same extent. SMC, which is for a long time a minority partner in the JV, faces some transfer limitations on certain dimensions of the production system. These limitations are rooted in the equity mode and the corresponding division of configuration mandates among the JV partners across different dimensions of the production system. This is also the case in MBI, which has put the configuration of a number of organizational aspects in the hands of its Indian JV partner (and later host country nationals) who uses a host/local parent template for the hierarchical differentiation of the site. Thus, there is empirical evidence to suggest that the equity mode has an impact on the question whose management is responsible and has the mandate to configure a certain dimension of a production system and, as a result, whose templates or demands are more likely to be used/transferred or have a weight. However, the empirical material equally suggests that the equity mode is neither sufficient to explain whose templates or demands are more likely to be used/transferred, nor is the equity mode as such sufficient to fully explain who has the configuration mandate (on certain dimensions of a production system). Even if a foreign/parent company holds a site’s entire equity and even if it has a defined foreign/parent template, it may still decide not to transfer such an available template or all of its parts because of strategic or institutional distance in the local/host context. What is more, the cases show that even if a host or foreign company holds the majority of a site’s equity; this does not necessarily imply that the majority partner assumes the configuration mandate. The equity mode is simply not enough to determine who has the configuration mandate for setting up a production system’s different dimensions. While the equity mode influences the configuration mandate, there are other influencing factors. These factors include: the product ownership, the local/host context capabilities (and their development over time), the capabilities of the respective partners on certain production system dimensions as well as the staffing policy.

268 Hybridization of MNE Subsidiaries

In this respect, MSI and SAI are two telling cases. The equity mode of these two cases suggest for SAI the most comprehensive configuration mandate and in MSI the least comprehensive configuration mandate for the foreign partner and its expatriates. However, the opposite is the case. In SAI this is mostly related to Skoda’s polycentric staffing policy, which implies that most configuration work is delegated to experienced host country management. It is equally linked to the host context capability – i.e. capable host country managers – which is available by the time Skoda enters the Indian market. Interestingly, such an availability of local/host capability is connected to Skoda’s late market entry, allowing the company to head-hunt experienced Indian managers from other international Automobile companies, an option the early movers do not have. In MSI, in turn, local/host context capability – or rather its absence – plays a key role in explaining SMC’s comprehensive configuration mandate, despite being the junior partner in the JV. Given the lacking local competence – particularly in ailing Maruti – as well as in other Indian automobile companies in the 1980s, the Japanese partner is asked or invited to assume the configuration mandate on most dimensions of the production system. In other words, despite being in the minority equity position, SMC’s capability and product ownership earns it the comprehensive configuration mandate for most production related dimension of the production system. This allows SMC to transfer its foreign/parent template rather comprehensively. Similarly, in MBI the local capability plays a crucial role for staffing its HR-function (throughout the company existence) with Indian managers, giving them the configuration mandate on core HR-issues. While with the equity shift all top management positions are staffed with Germans, HR is the only function where the host country management remains in charge. This is related to the Indian managers’ intimate knowledge with regard to recruitment, retention, redundancy procedures and industrial relations in India – capability German managers do not have or cannot acquire at short notice. Now, the equity mode is not only proposed to impact the transfer scenario, but also the occurrence of misfits/recontextualization pressures and to influence recontextualization modes. Clearly, to the extent that an equity shift in favour of the foreign/parent leads to increased presence of foreign/parent management at the foreign site the chances for misfit/recontextualization pressure increases. A case in point is MBI. As equity shifts, German managers replace Indian managers in a number of areas and find certain local practices inappropriate and to be changed to foreign/parent demands. Yet, there is again only limited evidence to suggest that the equity mode alone impacts contextual

Empirical Results 269

misfit/recontextualization pressure between the foreign/parent templates or demands and local/host contexts and vice versa. Rather than equity modes, the configuration mandates have more explanatory power. Despite a foreign/parent majority, the host country management may have the configuration mandate on certain dimensions of a production system. Under such a condition the misfit and recontextualization pressure is lower because of a lower incidence of foreign/parent template transfer or demands. Moreover, even if there is a foreign/parent template transfer from a distant context, the distance will be mediated by local/host country management, which may also defuse the misfit. A case in point is SAI. Despite being a wholly owned subsidiary from the start, the local set-up of the site does not involve much recontextualization pressure from the foreign/parent side. There is only a very limited foreign/parent template transfer and the largely indigenously designed set up of the site – although it is institutionally very distant from the home configuration – does not trigger much recontextualization pressure from the foreign/parent. In this case the absence of foreign/parent template transfer, misfit perception and recontextualization pressure have to do with the staffing policy, translating into a dominant configuration mandate for host country managers. Thus, although the equity mode is not without influence on the configuration mandate – i.e. what kind of management is in control of or influences a site – and, thereby, on misfit perceptions and the inducement recontextualization pressures, the equity mode does not determine it. By the same token, the equity mode has some influence on the recontextualization mode because it influences the configuration mandate and the question which side has the motivation, power and legal right to push for adaptations. For instance, SMC management in MSI is probably not very happy about the company’s excessive hierarchical differentiation, which is very different from its own in Japan. This aspect of the production system is, however, difficult to change as long as SMC is the junior partner. Thus, the equity mode plays an important part in enabling and constraining the involvement and ability of different parties to adapt foreign/parent and/or local/host templates, demands or conditions. Yet, the equity mode is only one factor among many others to impact the mode of adaptation. Although a foreign/parent company in a majority position may wish to adapt the local/host context to its template, it may lack the resources to do so or may face a local/host context that is very distant and not very resilient for change. Moreover, despite a foreign/parent majority ownership, some organizational aspects may remain in the hands of a local/host partner, depriving

270 Hybridization of MNE Subsidiaries

the foreign/parent side of the ability to induce a recontextualization pressure and an adaptation of the local/host context. We can summarize that equity modes impact transfer scenarios, the occurrence of contextual misfits/recontextualization pressures and recontextualization modes. However, we have to look at how they interact with other factors discussed earlier to explain hybridization outcomes. Above all, we need to look at the distribution of configuration mandates across the different dimensions of production systems that are only partly linked to equity modes.

Conclusion Summary of findings This work, in line with the basic proposition posited, shows that hybridization outcomes are impacted by strategic and institutional distance. Their primary influence on hybridization outcomes rest with their effect on transfer scenarios and inducing contextual misfit related recontextualization pressure. Moreover, it is found that while strategic and institutional contexts form distinct influences and take on different prominence in affecting certain dimensions of a production system, they should always be considered as interacting influences on any production system dimension. It would be erroneous to state that the hybridization of some dimensions of a production system only depends on strategic or institutional context and distance. Both contexts generally matter, albeit to different degrees. Secondly, this work shows that strategic choices at the corporate and subsidiary level have a substantial impact on hybridization outcomes. Generic product strategies, for example, play an important role for the availability and transfer propensity of foreign/parent templates, for the strategic misfit/recontextualization pressure and even affect the recontextualization mode, although this last aspect requires further exploration. Similarly, this work shows in the four cases researched that strategic choices at the level of the subsidiary affect transfer scenarios, misfits/recontextualization pressures, recontextualization modes and thereby hybridization outcomes. The work also offers some refinement with regard to the impact of strategic choices on hybridization outcomes. While generic strategies appear to have an important impact on the availability and transfer of foreign/parent templates, other factors are required to explain such template development and transfer. The findings suggest that internationalization experience and internationalization strategies of MNEs

Empirical Results 271

may be of crucial relevance. Similarly, while establishment and equity modes seem to impact transfer scenarios, misfits/recontextualization pressures and recontextualization modes, they need to be considered in interaction with other factors. Both establishment mode and equity mode do not suffice to explain transfer scenarios, misfits and recontextualization modes. It emerges in this context that configuration mandates – involving the question whether host country nationals or expatriates; local parent or foreign parent management is in control – plays a paramount role to understand hybridization outcomes. Although equity modes have an important impact on such configuration mandates, they do not determine them. Thirdly, the study shows that there are important interactions between strategic choices and contextual distance that impact hybridization outcomes. For example, companies with a cost focus strategy face a lower degree of strategic distance between their operations which creates incentives to develop and transfer templates. Another example are the establishment modes that mediate the contextual distance between potential transfer origins and destinations. Apart from its specific contribution for our understanding of hybridization processes in MNE, this study shows that a quasi-experimental research design involving rich array of cases, with great variation of combinations, allow us, if studied interactively in a procedural perspective, to establish refined patterns. The strength of such an approach is that it sensitizes us for complex interactions of factors that constitute processes and outcomes in MNE

Contribution to current debates This study shows that hybridization research has much scope for the cross fertilization of different fields. In contrast, to the normative and stylized models of production systems in the Japanization literature (see also Elger and Smith, 2005, 2006), this study offers an alternative perspective on production system hybridization by functionally defining their core dimensions and asking how they are contextually constituted. Such an approach avoids the assumption that MNEs generally define and seek the transfer of comprehensive templates to their subsidiaries. It stresses that firms do not always undertake the transfer of clearly defined production templates and are very selective (see also Meardi and Toth, 2006) with regard to what they transfer, depending on particular strategic choices and strategic and institutional contexts of their subsidiaries. Similarly, foreign/parents are not the only transfer drivers in MNEs. Partners in local/host context,

272 Hybridization of MNE Subsidiaries

particularly in emerging market JVs, may ask for transfers due to resource and capability gaps or poor institutional conditions (see also Meardi and Toth, 2006). While this study, adds to the Japanization literature, it equally contributes to hybridization and cross-border transfer research in the field of IB and Institutionalism, by synergizing the strength of different approaches. The study adds to hybridization research from European Institutionalism a strategic or contingency perspective. It does so by drawing on IB and conceptualizing the MNE as a special kind or organization with an increasing strategic role differentiation and an international division of labour (see also Edwards and Kuruvilla, 2005). Conversely, the study adds to the IB research on transfers in MNEs, by introducing a hybridization perspective and how institutional contexts and their difference across countries and sites impact transfer intents and transferability of production systems. Finally, in contrast to most studies contributing to the wider field of hybridization research, the study is not only open to different starting points of production system hybridization and different kinds of contextual misfits, but also to the condition that hybridization outcomes can be affected by different modes of recontextualization. Specifically, firms can react to or act on local/host misfits in different ways. Depending on the local/host context resilience and their ability and willingness to mobilize resources, they can react by adapting their systems and demands to the local/host context or be pro-active by adapting the local/host context to their systems and demands.

Limitations and outlook While this work makes a contribution to hybridization research by bringing together the insight from three crucial research streams, some points still remain unclear and some limitations of this work need to be addressed. First, the biggest remaining question of this work is what factors interact and determine a specific recontextualization mode in the case of a contextual misfit. While certainly the ability and willingness of the foreign/parent as well as the resilience of the local/host context provide a first answer, we need to develop a better understanding what factors determine foreign/parent ability and willingness, what factors determine the local/host context resilience as well as how the former and the latter interact. Some of the relevant factors have been discussed here such as strategic choices and contextual distance but their interplay remains still elusive.

Empirical Results 273

The study also invites to explore in more detail the relationship between configuration mandates and hybridization profiles. Specifically, the question how and why local/host or foreign/parent management is selected to set up different dimensions of a production system. While equity modes, product ownership, know-how and capabilities offer an important starting point, there needs to be a better understanding how staffing policies impact hybridization outcomes. The cases discussed also show that contextual constitution of a production system, its hybridization profile, is a highly dynamic phenomenon. As time unfolds the cases feature different outcomes on the different dimensions of their production systems. This profile change goes hand in hand with firm-specific changes at the corporate and subsidiary level (e.g. equity or ownership change, build up of local know how) and the development of local/host context (build up capability, changes in legislation). Thus, a more fine grained longitudinal perspective on the hybridization dynamic of individual cases would be an important further contribution, which is beyond the scope of this study. A final challenge this work is not able to address is the question how hybridization processes unfold at the micro-level between specific actors that are embedded in or originate form different strategic and institutional context. It involves asking: How specific hybridization outcomes have come about by taking a closer look at how actors are differently affected and able to shape such processes based on their systemic positioning in corporate and societal contexts? How hybridization processes are socio-politically constituted? What kinds of decision making processes constitute hybridization processes? Whether we should consider transfer processes as ‘translation processes’ or ‘dialectical transformation’ (Becker-Ritterspach, 2006)? As these questions require an embedded – ideally an ethnographic – micro-perspective for their exploration, they are beyond the theoretical scope and empirical access conditions of this work.

Notes 2 Transfer and Hybridization in MNEs 1. Some authors take a middle ground. Wilkinson et al. (1995), for example, are critical about the emancipatory potential of Japanese production system but see their transfer as generally feasible. 2. It is also interesting to note that a number of authors in this body of research have increasingly combined American and European Institutionalism to understand the constitution of HRM practices in MNEs (e.g. Ferner and Quintanilla, 1998; Tempel et al., 2006). 3. While institutional context have found only little attention, there are a few contributions that focus on the impact of cultural context (e.g. Bahgat et al., 2002; Subramaniam and Venkatraman, 2001; Chini, 2004; Cui et al., 2006). 4. It should be noted, however, that an increasing number of approaches in the area of Strategic International Human Resource Management – which could be viewed as a subfield of IB – have integrated New Institutionalist or Cultural perspectives into their models (e.g. Schuler et al., 1993; Taylor et al., 1996; Schuler et al., 2002).

5 The Case of Maruti–Suzuki India 1. Price who researched SMC’s production system in Japan points in the same direction and states ‘the absence of job descriptions for workers meant that SMC employees had no specific job assignment or classification and could be rotated or transferred relatively easily’ (Price, 1997, p. 165). 2. MSI prides itself to have implemented ‘a flat organisational structure’ with ‘only three levels of responsibilities ranging from the Board of Directors, Division Heads to Department Heads’ (Maruti, 2003). 3. These three work dispositional transfer goals can also be read as problem areas that require particular focus in the beginning.

6 The Case of Fiat India 1. Camuffo describes the corner stones of the Fabbrica Integrata concept as follows: ‘This model, named Fabbrica Integrata, is an Italian, adapted version of lean manufacturing developed in the early ’90s and fully implemented in the highly successful greenfield plant in Melfi, South Italy. Following ‘lean’ principles, the main features (Camuffo and Volpato, 1995) of Fabbrica Integrata are: advanced and flexible production technology, the adoption of lean manufacturing concepts (just in time, synchronous kanban, kaizen, job rotation, management by sight, quality tracking, etc.), key partner suppliers located

274

Notes 275 close to the assembly plant, a ‘flat’ organizational structure, organizational units based on process logic and linked to client-supplier logic, decentralisation of responsibilities and functions, focus on skills and human resources as performance drivers.’ (Camuffo, 2000/2001, p. 42).

7 The Case of Mercedes-Benz India 1. The higher the segment of vehicles, the bigger the profit margins and the lower the price sensitivity of customers and the lower the volumes – making scale economies difficult – the more attractive it becomes, to import non-labour intensive parts and components despite penalizing import tariffs. 2. See the parallels between MBI’s family model and J.B.P Sinha’s (1990a) concept of a ‘nurturant task leader’, which D. Sinha (1999) has called a ‘blend of exogenous and indogenous’, i.e. a blend between foreign and Indian management practices.

Bibliography Abdullah, S. and T. Keenoy (1995) ‘Japanese Management Practices in the Malaysian Electronics Industry’, Journal of Management Studies, 32, 747–65. Abo, T. (1998) ‘Hybridization of the Japanese Production System in North America, Newly Industrialising Economies, South-East Asia, and Europe: Contrasted Configurations’ 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). Abo, T. (ed.) (1994) The Hybrid Factory: The Japanese Production System in the United States (Oxford: Oxford University Press). Ackroyd, S., G. Burrell, M. Hughes and A. Whitaker (1988) ‘The Japanisation of British industry?’ Industrial Relations Journal, 19, 11–23. ACMA (2006), ‘Indian Automotive Component Industry. Engine of Growth Driving the Indian Manufacturing Sector’, http://acmainfo.com/docmgr/Status_of_ Auto_Industry/Status_Indian_Auto_Industry.pdf, date accessed 15 December 2006. ACMA (2004) ‘State of Industry’, http://acmainfo.com/docmgr/Presentations_ 7th_EC_Meeting/Industry_Review_August_2003.pdf, date accessed 13 May 2004. Adler, P. S. (1999) ‘Hybridization: Human Resource Management at Two Toyota Transplants’ in J. K. Liker, W. M. Fruin and P. S. Adler (eds) Remade in America: Transplanting and Transforming Japanese Management Systems (New York: Oxford University Press). Adler, P. S. (1993) ‘The Learning Bureaucracy: New United Motors Manufacturing, Inc.’ in B. M. Staw, L. L. Cummings (eds) Research in Organizational Behavior, Vol. 15 (Greenwich, CT: JAI Press). Adler, P. S. and R. E. Cole (1993) ‘Designed for Learning: A Tale of Two Auto Plants’, Sloan Management Review, Spring, 85–94. Adler, P. S., B. Goldoftas and D. I. Levine (1998) ‘Stability and Change at NUMMI’ in R. Boyer, E. Charron, U. Jürgens and S. Tolliday, S. (eds) Between Imitation and Innovation: The Transfer and Hybridization of Productive Models in the International Automobile Industry (Oxford: Oxford University Press). Afsar, A. (2003) ‘Maruti Parts and Service 2003 – about Maruti’, http://www. fansamotors.com/marutiparts/AboutMaruti.htm, date accessed 26 May 2003. Almeida, P. and A. Phene (2004) ‘Subsidiaries and Knowledge Creation: The Influence of the MNC and Host Country on Innovation’, Strategic Management Journal, 25, 847–64. Almond, P., T. Edwards, P. Gunnigle, T. Colling, A. Ferner, M. Müller-Camen, J. Quintanilla and H. Wächter (2005) ‘Unraveling Home and Host Country Effects: An Investigation of the HR Policies of an American Multinational in Four European Countries’, Industrial Relations 44, 276–306. Alzona, G. and V. Prakash (2001) ‘Italian Companies Entering, Penetrating, and Escalating into the Indian markets. Europe-India Business Linkages Proceedings

276

Bibliography 277 of the EU-India Project’, http://www. saa.unito.it/pdf/euindia/1annualconf.pdf, date accessed 1 December 2005. Atzeni, M. (2001) ‘Technological Innovations and Traditional Labour Relations: A Case Study of the Industrial Relations from Fiat’s Plant in Kurla, India’, unpublished MA Research Paper, Institute of Social Studies, The Hague. Autocar India (2001) ‘One-to-one. Interview with Skoda’s MD Imran Hassen’, http://www.autocar india.com/features/onetoone/onenone.asp?f_id=10, date accessed 10 November 2005. Autojunction (2003) ‘Major Players. Ind Auto Ltd.’, http://auto.indiamart.com/indauto-ltd/, date accessed 3 March 2006. 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). Balcet, G. and A. Enrietti (2002a) ‘Fiat’s Focused Globalisation and its Impact. World Markets Research Centre’, Global Automotive Manufacturing & Technology, 1–4. Balcet, G. and A. Enrietti (2002b) ‘The Impact of Focused Globalisation in the Italian Automobile Industry’, Journal of Interdisciplinary Economics, 97–133. Balcet, G. and A. Enrietti (2002c) ‘Partnership and Global Production: Fiat’s Strategy in Turkey’, Actes du Gerpisa, 34, http://www.univ-evry.fr/PagesHtml/ laboratoires/gerpisa/actes/34/34-5.pdf, date accessed 29 November 2005. Barney, J. B. (1991) ‘Firm Resources and Sustained Competitive Advantage’, Journal of Management, 17, 99–120. Bartlett, C. A. (1985) ‘Global Competition and MNC Managers’, ICCH Note no. 0-385-287 (Boston: Harvard Business School. Bartlett, C. A. and S. Ghoshal (1998) Managing Across Borders: The Transnational Solution, 2nd edn (Boston, MA: Harvard Business School Press). Bartlett, C. A. and S. Ghoshal (1986) ‘Tap Your Subsidiaries for Global Reach’, Harvard Business Review, 64, 87–94. Becker-Ritterspach, F. A. A. (2006) ‘The Social Constitution of Knowledge Integration in MNEs: A Theoretical Framework’, Journal of International Management. Becker-Ritterspach, F. A. A. (2005) ‘Transfer, Intercultural Friction and Hybridization: Empirical Evidence from a German Automobile Subsidiary in India’, Asian Business and Management, 4, 365–87. Becker-Ritterspach, F. A. A. (2000) India 2010 – Where is the Elephant Going?: Development Prospects for the Indian Subcontinent (Berlin: DaimlerChrysler Research Publication). Becker-Ritterspach, J. C. E. (2008) Chinas und Indiens Automobilindustrien im Vergleich. Die Rolle staatlicher Industriepolitik (Saarbrücken: VDM Verlag Dr. Müller). Becker-Ritterspach, F. A .A . and J. C. E. Becker-Ritterspach (2008) ‘Die Entwicklung des Automobilismus in Indien’, in Canzler, W. and G. Schmidt (eds) Zukünfte des Automobils. Aussichten und Grenzen der autotechnischen Globalisierung (Berlin: Sigma), 257–86. Beechler, S., A. Bird and S. Taylor (1998) ‘Organisational Learning in Japanese MNCs: Four Affiliate Archetypes’ in Birkinshaw, J. and N. Hood (eds) Multinational Corporate Evolution and Subsidiary Development (Basingstoke: Macmillan Press).

278 Bibliography Belzowski, B. M., A. Henderson and P. Koppinger (2007) ‘Inside India. Indians view their automotive future’. IBM Global Business Services, http://www. gerpisa.univ-evry.fr/rencontre/15.rencontre/GERPISAJune2007/Colloquium/ Papers/UM.IBMInsideIndiaFinalReport.pdf, date accessed 13 February 2008. Berger, P. and T. Luckmann (1967) Social Construction of Reality (New York: Anchor Books). Bhadury, B. (2000) Managing the Indian Workforce. Challenges for the Manufacturing Industry (New Delhi: Sage). Bhagat, R. S., B. L. Kedia, P. D. Harveston and H. C. Triandis (2002) ‘Cultural Variations in the Cross-border Transfer of Organizational Knowledge: An Integrative Framework’, Academy of Management Review, 27, 204–21. Bhaktavatsala, R. C. (1993) ‘Technology and Industry Structure: A Strategic Management Perspective’, ASCI Journal of Management, 22, http://www.asci.org.in/ publications/ascijl/v22/ v22_2_3_bha.htm, date accessed 6 July 2004. Bhargava, R. C. (2002) ‘Strengthening SMEc Through Strategic and Marketing Linkages’, http://www.ficci.com/media-room/speeches-presentations/2002/dec/ dec-sme- bhargava.ppt, date accessed 27 March 2006. Bhargava, R. C. and V. Behl (1997) ‘Maruti Faced Problems Introducing the Japanese Management Culture’, Interview with former Maruti MD Bhargava, Rediff On The Net Business News, http://www.rediff.com/business/sep/03mul. htm, date accessed 26 May 2003. Bhattacherjee, D. (2001) ‘The Evolution of Indian Industrial Relations: A Comparative Perspective’, Industrial Relations Journal, 32, 244–63. Bhavani, T. A. (2002) Impact of Technology on the Competitiveness of the Indian Small Manufacturing Sector. A Case Study of the Automotive Component Industry, Discussion Paper no. 2002/76, http://www.wider.unu.edu/publications/dps/dps2002/ dp2002-76.pdf, date accessed 6 July 2004. Birkinshaw, J. (2001) ‘Strategy and Management in MNE Subsidiaries’ in A. M. Rugman and T. Brewer (eds) The Oxford Handbook of International Business (Oxford: Oxford University Press). Birkinshaw, J. (2000) Entrepreneurship in the Global Firm (London and New Delhi: Sage and Thousand Oaks). Birkinshaw J. and N. Hood (1998) ‘Multinational Subsidiary Evolution: Capability and Charter Change in Foreign-Owned Subsidiary Companies’, Academy of Management Review, 23, 773–95. Birkinshaw, J. M. and A. J. Morrison (1996) ‘Configurations of Strategy and Structure in Multinational Subsidiaries’, Journal of International Business Studies, 26, 729–94. Björkman, I., W. Barner-Rasmussen and L. Li (2004) ‘Managing Knowledge Transfer in MNCs: The Impact of Headquarters Control Mechanisms’, Journal of International Business Studies, 35, 443–55. Björkman, I., C. F. Fey and H. J. Park (2007) ‘Institutional theory and MNC subsidiary HRM practices: evidence from a three-country study’, Journal of International Business Studies, 38, 430–46. Bouglé, C. (1997) ‘The Essence and Reality of the Caste System’ in D. Gupta (ed.) Social Stratification (New Delhi: Oxford University Press). Boyer, R. (1998) ‘Hybridization and Models of Production. Geography, History, and Theory’ 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).

Bibliography 279 Boyer, R. and M. Freyssenet (2003) Produktionsmodelle. Eine Typologe am Beispiel der Automobilindustrie (Berlin: edition sigma). Brannen M. Y., J. K. Liker and W. M. Fruin (1999) ‘Recontextualization and Factoryto-Factory Knowledge Transfer’ in J. K. Liker, W. M. Fruin and P. S. Adler (eds) Remade in America. Transplanting and Transforming Japanese Management Systems (Oxford: Oxford University Press). Bratton, J. (1990) Japanization at Work. Managerial Studies for the 1990s (Houndsmills, Basingstoke, Macmillan Press). Braverman, H. (1974) Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century (New York: Monthly Review Press). Briggs, P. (1988) ‘The Japanese at Work: Illusions of the Ideal’, Industrial Relations Journal, 19, 24–30. Bronger, D. (1996) Indien. Größte Demokratie der Welt zwischen Kastenwesen und Armut (Gotha: Klett-Perthes). Budhwar, P. (2003) ‘Employment Relations in India’, Employee Relations, 25, 132–48. Business Standard (2005) http://www.businessstandard.com/strategist/rbi/story. asp, date accessed 3 June 2005. Business Standard (2003) ‘Skoda to Source Spares Locally, 30.04.2003’, http:// www.businessstandard.com, date accessed 3 June 2006. Camuffo, A. (2002) ‘Globalization, Outsourcing and Modularity in the Auto Industry’, http://www.sses. com/public/events/euram/complete_tracks/ modularity_and_division /camuffo.pdf, date accessed 28 November 2005. Camuffo, A. (2000/2001) Rolling Out a ‘World Car’: Globalization, Outsourcing and Modularity in the Auto Industry, IMVP Working Papers, http://imvp.mit.edu/ papers/0001/camuffo1.pdf, date accessed 28 November 2005. Camuffo, A. and S. Micelli (1997) ‘Mediterranean Lean Production. Supervisors, Teamwork and New Forms of Work Organization in Three European Car Makers’, The Journal of Management and Governance, 1, 103–22. Camuffo, A. and G. Volpato (2002) ‘Partnering in the Global Auto Industry: The Fiat-GM Strategic Alliance’, International Journal Automotive Technology and Management, 2, 335–52. Camuffo, A. and G. Volpato (2000/2001) Global Sourcing in the Automotive Supply Chain: The Case of Fiat Auto ‘Project 178’ World Car, IMVP Working Papers, http:// imvp.mit.edu/papers/ 0001/cam-volp1.pdf, date accessed 28 November 2005. Camuffo, A. and G. Volpato (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: Oxford University Press). Camuffo, A. and G. Volpato (1994) ‘Making Manufacturing Lean in the Italian Automobile Industry: The Trajectory of Fiat’, Actes du Gerpisa, no. 10, http://www.univ-evry.fr/labos/gerpisa/actes/10/10-2.pdf, date accessed 29 November 2005. Cantarella, P. (1997) Fiat Group Values and Policies (Fiat company source). Cardoso, L. A. (2004) ‘The Rationalized Factory and Changes in Work Management: The Case of the FIAT Post-Fordist Auto Plant in Brasil’ in Charron, E. and P. Stewart (eds) Work and Employment Relations in the Automobile Industry (Houndsmills, Basingstoke: Palgrave Macmillan), 58–75. Cardoso, L. A. (1998) ‘Globalization and the Changes in the Labor Relations in an After-Fordist Era: The Case of the Automobile Industry in Brazil’ in XIV

280 Bibliography Sociology World Congress. Abstracts of the XIV Sociology World Congress. (Montreal: ISA – International Sociology Association). Centre for Industrial & Economic Research (2002) Auto Industry Faces Roadblocks. Update 2002–03 to ‘Automobile Industry: 2001 and Beyond’ (New Delhi: Cier intecos). Chang, S.-J. and P. Rosenzweig (1998) ‘Functional and Line of Business Evolution Processes in MNC Subsidiaries: Sony in the USA, 1972–1995’ in J. M. Birkinshaw and N. Hood (eds) Multinational Corporate Evolution and Subsidiary Development (London: Macmillan). Chatterjee, B. (1990) Japanese Management. Maruti in the Indian Experience. (New Delhi: Sterling Publishers Private Limited). Chavan, S. (2003) ‘IPO Profile Maruti Udyog’, http://content.icicidirect.com/ ULFiles/UploadFile_ 20038113039.asp, date accessed 30 March 2006. Chini, T. C. (2004) Effective Knowledge Transfer in Multinational Corporations. (Houndsmills, Basingstoke: Palgrave Macmillan). Cohen, W. M. and D. A. Levinthal (1990) ‘Absorptive Capacity: A New Perspective on Learning and Innovation’, Administrative Science Quarterly, 35, 128–52. Colling, T. and I. Clark (2002) ‘Looking for ‘Americanness’: Home Country, Sector and Firm Effects on Employment Systems in an Engineering Services Company’, European Journal of Industrial Relations, 8, 301–24. Crowther, S. and P. Garrahan (1988) ‘Corporate Power and the Local Economy’, Industrial Relations Journal, 19, 51–9. D’Costa, A. P. (2005) The Long March to Capitalism – Embourgeoisment, Internationalisation and Industrial Transformation in India (Houndsmills, Basingstoke: Palgrave). D’Costa, A. P. (2003) ‘Institutions and Industrial Governance in India: Learning to Cooperate the Japanese Way’, Asian Business and Management, 2, 63–89. D’Costa, A. P. (1998) ‘An Alternative Model of Development? Co-operation and Flexible Industrial Practices in India’, Journal of International Development, 10, 301–21. Cui, A. S., D. A. Griffith, S. T. Cavusgil and M. Dabic (2006) ‘The influence of market and cultural environmental factors on technology transfer between foreign MNCs and local subsidiaries: A Croatian illustration’, Journal of World Business, 41, 100–11. DaimlerChrysler (2004a) ‘Company at a Glance’, http://www.daimlerchrysler. com/dccom/0„0-5-7155-1-12898-1-0-0-0-0-0-8-7155-0-0-0-0-0-0-0,00.html, date accessed 8 July 2004. DaimlerChrysler (2004b) ‘DaimlerChrysler India Private Limited’, http://www. daimlerchrysler.com/dccom/0„0-5-8790-1-36884-1-0-0-0-0-0-91-7155-0-0-0-00-0-0,00.html, date accessed 3 April 2004. Danford, A. (1997) Labour Control and Intensification Through Teamworking and Kaizen, Labour Studies Working Papers, no. 14, Centre for Comparative Labour Studies, Coventry: University of Warwick. Das, N. (2004) ‘DaimlerChrysler Turns the Corner’, http://www.domainb.com/people/interviews/ 20040525_hans_michael_huber.htm, date accessed 10 December 2005. Datamonitor (2003) ‘Suzuki Motor Corporation, Company Profile’, Reference Code 1622, www.datamonitor.com, date accessed 18 October 2005.

Bibliography 281 Davis, L. E. (1971) ‘The Coming Crisis for Production Management: Technology and Organization’, International Journal of Production Research, 9, 65–82. Dedoussis, V. (1995) ‘Simply a Question of Cultural Barriers? The Search for New Perspectives in the Transfer of Japanese Management Practices’, Journal of Management Studies, 32, 731–45. Degnbol-Martinussen, J. (2001) Policies, Institutions and Industrial Development Coping with Liberalisation and International Competition in India (New Delhi: Sage). Delany, E. (1998) ‘Strategic Development of Multinational Subsidiaries in Ireland’ in J. M. Birkinshaw and N. Hood (eds) Multinational Corporate Evolution and Subsidiary Development (London: Macmillan). Delbridge, R. (1995) ‘Surviving JIT: Control and Resistance in a Japanese Transplant’, Journal of Management Studies, 32, 803–17. Delbridge, R., P. Turnbull and B. Wilkinson (1992) ‘Pushing Back the Frontiers: Management Control and Work Intensification under JIT/TQM Factory Regimes’, New Technology, Work and Employment, 7, 97–106. Dickens, P. and M. Savage (1988) ‘The Japanisation of British industry? Instances from a High Growth Area’, Industrial Relations Journal, 19, 60–8. DiMaggio, P. J. (1988) ‘Interest and Agency in Institutional Theory’ in G. L. Zucker (ed.) Institutional Patterns and Organizations: Culture and Environment (Cambridge, MA: Ballinger). DiMaggio, P. J. and W. W. Powell (1983) ‘The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields’, American Sociological Review, 48, 147–60. Dörrenbächer, C. and J. Gammelgaard (2006) ‘Subsidiary role development: The effect of micro-political headquarters–subsidiary negotiations on the product, market and value-added scope of foreign-owned subsidiaries’, Journal of International Management, 12, 266–83. Doz, Y. and C. K. Prahalad (1991) ‘Managing DMNCs: A Search for a New Paradigm’, Strategic Management Journal, 12, 145–64. Dumont, L. (1997) ‘Hierarchy, Status and Power: The Caste System and Its Implications’ in D. Gupta (ed.) Social Stratification (New Delhi: Oxford University Press). Dunning, J. H. (2000) ‘The Eclectic Paradigm as an Envelope for Economic and Business Theories of MNE Activity’, International Business Review, 9, 163–90. Dupont, V. (1992) ‘The Significance of Caste in the Segmentation of the Urban Labour Market: Study of a Middle-sized Industrial Town in Western India’, presented at the 12th European Conference on Modern South Asian Studies, Berlin, Sept. 23–26. Eckardt, A., H.-D. Köhler and L. Pries (2000) Auf dem Weg zu global operierenden Konzernen? Fallstudien zu den Internationalisierungsverläufen deutscher Automobilkonzerne in den 90er Jahren. (Erlangen: Institut für praxisorientierte Sozialforschung und Beratung e.V.). Edwards, T., P. Almond, I. Clark, T. Colling, and A. Ferner (2005) ‘Reverse Diffusion in US Multinationals: Barriers from the American Business System’, Journal of Management Studies, 42, 1261–1286. Edwards, T. and S. Kuruvilla (2005) ‘International HRM: National Business Systems, Organisational Politics and the International Division of Labour in Global Value Chains’, International Journal of Human Resource Management 16, 1–21.

282 Bibliography Eisenhardt, K. M. (1989) ‘Building Theories from Case Study Research’, Academy of Management Review, 14, 532–50. Eisenstodt, G. (1993) ‘A 4 Billion People Market’, Forbes, 152, 27.09.1993, 48–50. Elger T. and C. Smith (2006) ‘Theorizing the Role of the International Subsidiary: Transplants, Hybrids and Branch-Plants Reviested’, in Ferner, A., J. Quintanilla and C. Sanchez-Runde (eds) Multinationals, Institutions and the Construction of Transnational Practices: Convergence and Diversity in the Global Economy (London: Palgrave), 53–85. Elger T. and C. Smith (2005) Assembling Work: Remaking Work Regimes in Japanese Multinationals in Britain (Oxford: Oxford University Press). Elger, T. and C. Smith (1994) Global Japanisation?: The Transnational Transformation of the Labour Process (London: Routledge). Emery, F. E. (1959) ‘Characteristics of Socio-technical Systems’ in L. E. Davis and J. C. Taylor (eds) Design of Jobs (Harmondsworth: Penguin). Ferner A., P. Almond and T. Colling ‘Institutional theory and the cross-national transfer of employment policy: the case of ‘workforce diversity’ in US multinationals’ Journal of International Business Studies, 36, 304–21. Ferner A., J. Quintanilla and M.Z. Varul (2001) ‘Country-of-origin effects, hostcountry effects, and the management of HR in multinationals: German companies in Britain and Spain’ Journal of World Business, 36, 107–27. Fiat Group ( 2005a) ‘Main Fiat Auto Plants in the World (at 12.31.2005)’ http://www.fiatgroup.com, 29 November 2007 date accessed. Fiat Group (2005b) ‘Fiat Group in Competitive Mood, US Roadshow’, http://www.fiatgroup.com/en- us/shai/iinfo/presentations/Documents/Fiat%20 US%20December%202005.pdf, 10 May 2008 date accessed. Fiat India (2006) ‘Tata Motors and Fiat Group announce three additional cooperation agreements’, http://www.fiat-india.com/FiatAdmin/MediaCoverage/Doc Pdf/MediaDoc_4. PDF, 10 May 2008 date accessed. Fiat India (2007) ‘Fiat Group and Tata Motors announce establishment of Joint Venture in India’ http://www.fiat-india.com/Fiat_Press/PressDetails.aspx? PressId=26, 10 May 2008 date accessed. Fiat, n.d. ‘Main Fiat Auto Plants in the World’, http://www.fiatgroup.com/comuni/ php/file_get.php? w=G6ADS03XQL02TX5FZ6VT, date accessed 6 June 2004. Florida, R. and M. Kenney (1991a) ‘Transplanted Organizations: The Transfer of Japanese Industrial Organization to the U.S.’, American Sociological Review, 56, 381–98. Florida, R. and M. Kenney (1991b) ‘Organisation vs Culture: Japanese Automotive Transplants in the U.S.’, Industrial Relations Journal, 22, 181–96. Forsgren, M., T. Pedersen and N. Foss (1999) ‘Accounting for the Strength of MNC Subsidiaries: The Case of Foreign-owned Firms in Denmark’, International Business Review, 2, 181–96. Foss, N. J. and T. Pedersen (2002) ‘Transferring Knowledge in MNCs: The Role of Sources of Subsidiary Knowledge and Organisational Context’, Journal of International Management, 8, 49–67. 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: Oxford University Press).

Bibliography 283 Frost, T. (2001) ‘The Geographic Sources of Foreign Subsidiaries’ Innovations’, Strategic Management Journal, 22, 101–23. Fucini, J. J. and S. Fucini (1990) Working for the Japanese: Inside Mazda’s American Auto Plant (Basingstoke Macmillan). Garrahan, P. and P. Stewart (1992) The Nissan Enigma: Flexibility at Work in the Local Economy. (London: Mansell). Geppert, M., D. Matten and P. Schmidt (2004) ‘Die Bedeutung institutionalistischer Ansätze für das Verständnis von Organisations- und Managementprozessen in multinationalen Unternehmen’, Berliner Journal für Soziologie, 14, 379–97. Ghauri, P. (2004) ‘Designing and Conducting Case Studies in International Business Research’ in R. Marschan-Piekkari and C. Welch (eds) Handbook of Qualitative Research Methods for International Business (Cheltenham, UK/Northampton, MA: Edward Elgar Publishing Ltd.). Ghoshal, S. and N. Nohria (1989) ‘Internal Differentiation within MNCs’, Strategic Management Journal, 10, 323–37. Ghoshal, S. and D. E. Westney (1993) Organization Theory and the Multinational Corporation (New York: St. Martin’s Press). Glaser, B. G. and A. L. Strauss (1967) The Discovery of Grounded Theory (Chicago: Aldine). Goldstein, A. (2002) ‘Local Entrepreneurship in the Era of E-business: Early Evidence from the Indian Automobile Industry’ in A. Goldstein and D. O’Connor (eds) International Development, OECD http://www.instinformatica.pt/v20/cid/biblioteca_digital/comercioelectronico/ ElectronCom merceforDeveolpm.pdf, date accessed 1 January 2005. Goldstein, A. (2001) ‘Local Entrepreneurship in the Era of E-business: Early Evidence from the Indian Automobile Industry,’ Paper presented at the International Conference on “E-commerce for Development: Reviewing Early Experiences, Comparing New Ideas”, co-organised by the OECD Development Centre and the School of Development, Innovation and Change, University of Bologna, 4–5 May 2001. Gosalia, S. (1992) Indien im südasiatischen Wirtschaftsraum. Chancen der Entwicklung zu einem regionalen Gravitationszentrum (Hamburg: Deutsches Übersee-Institut). Graham, I. (1988) ‘Japanisation as Mythology’, Industrial Relations Journal, 19, 69–75. Grunow-Osswald, E. (2006) Die Internationalisierung eines Konzerns. Daimler-Benz 1890–1997. Vaihingen: DaimlerChrysler Konzernarchiv. Gulyani, S. (2001) ‘Effects of Poor Transportation on Lean Production and Industrial Clustering: Evidence from the Indian Auto Industry’, World Development, 29, 1157–77. Gulyani, S. (1999) ‘Innovating with Infrastructure: How India’s Largest Carmaker Copes with Poor Electricity Supply’, World Development, 27, 1749–68. Gupta, A. K. and V. Govindarajan (2000) ‘Knowledge Flows within Multinational Corporations’, Strategic Management Journal, 21, 473–96. Gupta, A. K. S. and P. K. Sett (2000) ‘Industrial Relations Law, Employment Security and Collective Bargaining in India: Myths, Realities and Hopes’, Industrial Relation Journal, 31, 144–53. Gurgaon Worker News (2007) Welcome to the Machine -Summary on re-structuring at Maruti Suzuki, Automobile Industry Part Nine, http://gurgaon

284 Bibliography workersnews.wordpress.com/gurgaonworkersnews-no8/, date accessed 10 May 2008. Hakanson, L. and R. Nobel (2000) ‘Technology Characteristics and Reverse Technology Transfer’, Management International Review, 40, 29–47. Halasyam, A. R. (2001) ‘Session VII: Operational Excellence: A Product of Multiple Lever at Work’, Conference of IMA India, A. R. Hayasyam, Director Finance, Conference Speaker Maruti Udyog Limited, http://www.imaindia.com/autoroundtable/session7.htm, date accessed 26 May 2003. Hall, P. A. and D. Soskice (eds) (2001) Varieties of Capitalism. The Institutional Foundations of Comparative Advantage (Oxford: Oxford University Press). Hansen, M. T. (2002) ‘Knowledge Networks: Explaining Effective Knowledge Sharing in Multiunit Companies’, Organization Science, 3, 232–48. Hansen, M. T. (1999) ‘The Search-Transfer Problem: The Role of Weak Ties in Sharing Knowledge across Organization Subunits’, Administrative Science Quarterly, 82–111. Hansen, M. T. and B. Lovas (2004) ‘How Do Multinational Companies Leverage Technological Competencies?: Moving from Single to Interdependent Explanations’, Strategic Management Journal, 25, 801–22. Harzing, A.-W. K. (2002) ‘Acquisitions versus Greenfield Investments: International Strategy and Management of Entry Modes’, Strategic Management Journal, 23, 211–27. Harzing, A.-W. K. (1999) Managing the Multinationals. An International Study of Control Mechanisms (Cheltenham, UK/Northampton, MA: Edward Elgar). Heitmann, J. (1995) ‘Berufsbildung in Indien’, unpublished desk-study compiled by the GTZ (Gesellschaft für Technische Zusammenarbeit). Hitagaki, H. (ed.) (1997) The Japanese Production System: Hybrid Factories in East Asia. (Houndsmills, Basingstoke: Macmillan). Hollingsworth, J. R. and R. Boyer (1997) ‘Coordination of Economic Actors and Social Systems of Production’ in J. R. Hollingsworth and R. Boyer (eds) Contemporary Capitalism. The Embeddedness of Institutions (Cambridge: Cambridge University Press). Humphrey, J. (2003) ‘Globalization and Supply Chain Networks: The Auto Industry in Brazil and India’, Global Networks, 3, 121–41. Humphrey, J. (2000) ‘Assembler–Supplier Relations in the Auto Industry: Globalisation and National Development’, Competition & Change, 4, 245–71. Humphrey, J. (1999) ‘Globalisierung und nationale Entwicklung. Die Transformation der Hersteller-Zulieferer-Beziehungen in der brasilianischen und indischen Automobilindustrie’ in H. Kilper and L. Pries (eds) Die Globalisierungsspirale in der deutschen Automobilindustrie (München, Mering: Rainer Hampp). Humphrey, J. and M. Salerno (1999) ‘Globalization and Assembler-Supplier Relations: Brazil and India’, Actes du Gerpisa, no. 25. Hymer, S. (1976) ‘The International Operations of National Firms: A Study of Direct Investment,’ PhD Thesis (Boston: MIT Press). Indian Express (1998) ‘Maruti Union Goes on the Warpath’, Indian Express Newspapers (Bombay) Ltd, http://www.expressindia.com/fe/daily/19980610/ 16155074.html, date accessed 3 June 2003. Jain, S. (2003) ‘Maruti’s Second Skin, 24.11.2003, http://www.rediff.com/money/ 2003/nov/ 24guest.htm, date accessed 3 March 2006.

Bibliography 285 Jarillo, J. C. and J. I. Martinez (1990) ‘Different Roles for Subsidiaries: The Case of Multinational Corporations in Spain’, Strategic Management Journal, 11, 501–12. Joseph, A. (1990) ‘Japan’s FDI in India’s Industrial Development: A Case Study of Maruti Udyog Ltd.’, The CTC Reporter, no. 30, http://www.law.harvard. edu/Administrative_Services/Registrar/ exams_97-98/html/smit02.html, date accessed 26 May 2003. Kakar, S. (1997) The Inner World. A Psycho-analytic Study of Childhood and Society in India. (New Delhi: Oxford University Press). Kanungo, R. N. (1994) ‘Culture and Work Alienation. Western Models and Eastern Realities’ in H. S. R. Kao, D. Sinha and N. Sek-Hong (eds) Effective Organisations and Social Values (New Delhi: Sage). Kasahara, K. (1994) ‘Japanese Automobile Firms in India – Maruti Suzuki and the transfer of technology’, Japan Academy of International Business Studies, Rikko University, 73–84. Kenney, M. (1999) ‘Transplantation? A Comparison of Japanese Television Assembly Plants in Japan and the United States’ in J. K. Liker, W. M. Fruin and P. S. Adler (eds) Remade in America: Transplanting and Transforming Japanese Management Systems (New York: Oxford University Press). Kenney, M. and R. Florida (1995) ‘The Transfer of Japanese Management Styles in Two US Transplant Industries: Autos and Electronics’, Journal of Management Studies, 32, 789–802. Kenney, M. and R. Florida (1993) Beyond Mass Production: The Japanese System and Its Transfer to the US (New York: Oxford University Press). Khare, A. (1997) ‘Strategic Advantages of Good Supplier Relations in the Indian Automobile Industry’, Technovation, 17, 557–68. Khare, A. (1999) ‘Japanese and Indian Work Patterns: A Study of Contrasts’ in S. R. Kao, D. Sinha and B. Wilpert (eds) Management and Cultural Values. The Indigenisation of Organisations in Asia (New Delhi: Sage). Kimberley, W. (2001) ‘Skoda. An Eastern European Sucsess’, http://www.auto fieldguide.com/ columns/will/0601euro.html, date accessed 5 November 2005. Kochan, A. (2001) ‘Skoda Takes a Low-cost Approach to Production of Its New Luxury Car’, Automotive News Europe, 6. Kostova, T. (1999) ‘Transnational Transfer of Strategic Organisational Practices: A Contextual Perspective’, Academy of Management Review, 24, 308–24. Kostova, T. and K. Roth (2002) ‘Adoption of an Organizational Practice By Subsidiaries of Multinational Corporations: Institutional and Relational Effects’, Academy of Management Journal, 45, 215–33. Kotabe, M., X. Martin and H. Domoto (2003) ‘Gaining from Vertical Relationships: Knowledge Transfer, Relationship Duration, and Supplier Performance Improvement in the U.S. and Japanese Automobile Industries’, Strategic Management Journal, 24, 293–316. Krafcik, J. (1986) Learning from NUMMI. International Motor Vehicle Program (Cambridge, MA: MIT). Krueger, A. O. and S. Chinoy (2002) ‘The Indian Economy in Global Context’, in: Krueger, A.O and S. Chinoy (eds) Economic Policy Reforms and the Indian Economy (New Delhi: Oxford University Press), 9–46. Kulkarni, V. (2002) ‘On a Smooth Road, Financial Daily from THE HINDU Group of Publications’, http://www.blonnet.com/ew/2002/01/30/stories/2002013000 230300.htm, date accessed 26 May 2003.

286 Bibliography Kumon, H. and Abo, T. (2004) The Hybrid Factory in Europe: The Japanese Management and Production System Transferred (Houndsmills, Basingstoke, Palgrave Macmillan). Lansbury, R., S-H. Kwon, and C-S. Suh (2006) ‘Globalisation and Employment Relationsin the Korean Auto Industry: The Case of the Hyundai Motor Company in Korea, Canada and India’, Asia Pacific Business Review 12, 131–47. Lane, C. (2001) ‘The Emergence of German Transnational Companies: A Theoretical Analysis and Empirical Study of the Globalization Process’ in G. Morgan, P. H. Kristensen and R. Whitley (eds) The Multinational Firm. Organizing Across Institutional and National Divides (Oxford: Oxford University Press). Lane, C. (2000) ‘Understanding the Globalization Strategies of German and British Multinational Companies: Is a ‘Societal Effects’ Approach Still Usefull?’ in M. Maurice and A. Sorge (eds) Embedding Organizations. Societal Analysis of Actors, Organizations and Socio-economic Context (Amsterdam/Philadelphia John Benjamins Publishing Company). Lane, C. (1994) ‘Industrial Order and the Transformation of Industrial Relations: Britain, Germany and France’ in R. Hyman and A. Ferner (eds) New Frontiers in European Industrial Relations (Oxford: Blackwell). Lawrence, P. and J. Lorsch (1967) ‘Differentiation and Integration’, Administrative Science Quarterly, 12, 1–30. Liker, J. K., W. M. Fruin and P. S. Adler (1999) ‘Bringing Japanese Management Systems to the United States’ in J K. Liker, W. M. Fruin and P. S. Adler (eds) Remade in America: Transplanting and Transforming Japanese Management Systems (New York: Oxford University Press). Lorenz, E. (2000) ‘The Transfer of Business Practices to Britain and France’ in M. Maurice, M. and A. Sorge (eds) Embedding Organizations: Societal Analysis of Actors, Organisations and Socio-Economic Context (Amsterdam/Philadelphia John Benjamins Publishing Company). Lyle, F. (2002) ‘Fresh Talent from Skoda’s Own College’, Automotive News Europe, 7, 5. MacDuffie, J. P. (1995) ‘International Trends in Work Organization in the Auto Industry: National-Level versus Company-Level Perspectives’ in K. S. Wever and L. Turner (eds) The Comparative Political Economy of Industrial Relations. (Madison, WI: Industrial Relations Research Association). Mair, A. (1998) ‘Internationalization at Honda: Transfer and Adaptation of Management Systems’, Employee Relations, 20, 285–302. Mamkoottam, K. (2000) ‘Economic Reforms and Labour Relations in India – An Academic Perspective’ in V. C. S. Ratnam and P. Sinha (eds) Trade Union Challenges at the Beginning of the 21st Century, India Industrial Relations Association (New Delhi: Execel Books). Marchington, M. and P. Parker (1988) ‘Japanization: A Lack or Chemical Reaction’, Industrial Relations Journal, 88, 272–85. Marschan-Piekkari, R. and C. Welch (2004), Qualitative Research Methods in International Business: The State of the Art’ in R. Marschan-Piekkari and C. Welch (eds) Handbook of Qualitative Research Methods for International Business (Cheltenham: Edward Elgar Publishing Ltd.). Marshall, C. and G. B. Rossman (1995) Designing Qualitative Research (London: Sage).

Bibliography 287 Maruti Udyog Ltd. (2000–2008) Annual report, 1999/00–2007/08 (New Delhi). Maruti (2003) ‘Maruti Culture, http://www.marutiudyog.com/ab/careers.asp? ch=1&ct=9&sc=1, date accessed 3 March 2006. Maurice, M. (2000) ‘The Paradoxes of Societal Analysis. A Review of the Past and Prospects for the Future in M. Maurice and A. Sorge (eds) Embedding Organizations. Societal Analysis of Actors, Organizations and Socio-economic Context (Amsterdam/Philadelphia: John Benjamins Publishing Company). Maurice, M., A. Sorge and M. Warner (1980) ‘Societal Differences in Organizing Manufacturing Units: A Comparison of France, West Germany, and Great Britain’, Organization Studies, 1, 59–80. McKenna, S. (1988) ‘ “Japanisation” and Recent Developments in Britain’, Employee Relation, 10, 6–12. Meardi, G. and A. Tóth (2006) ‘Who is Hybridizing What?: Insights on MNCs’ Employment Practices in Central Europe’, in Ferner, A., J. Quintanilla and C. Sanchez-Runde (eds) Multinationals, Institutions and the Construction of Transnational Practices: Convergence and Diversity in the Global Economy (London: Palgrave), 155–83. Meyer, J. W. and B. Rowan (1977) ‘Institutionalized Organizations: Formal Structure as Myth and Ceremony’, American Journal of Sociology, 83, 340–63. Miles, M. B. and A. M. Huberman (1994) Qualitative Data Analysis, 2nd edn (Thousand Oaks, CA: Sage). Milkman R. (1991) Japan’s California Factories: Labor Relations and Economic Globalization (Los Angeles: Institute of Labor Relations, University of California). Minbaeva, D., T. Pedersen, I. Björkman, C. F. Fey and H. J. Park (2002) ‘MNC Knowledge Transfer, Subsidiary Absorptive Capacity and HRM’, Academy of Management Proceedings. Ministry of Heavy Industries & Public Enterprises (2006) ‘Automotive Mission Plan 2006–2016. A Mission for Development of Indian Automotive Industry’ Government of India. http://www.siamindia.com/Upload/AMP.pdf, date accessed 13 February 2008. Ministry of Heavy Industries & Public Enterprises (2002) ‘Auto Policy. Vision: To Establish a Globally Competitive Automotive Industry in India and to Double Its Contribution to the Economy by 2010, http://www.dhi.nic.in/autopolicy.htm, date accessed 7 May 2004. Mishina, K. (1998) ‘Making Toyota in America: Evidence from the Kentucky Transplant’ 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). Mohan, N. C. (2002) ‘Reliance Leads the Charge, but Others too are on the Way to Becoming World Class Companies. Seeking a Place in the Global Sun’, The Financial Express, Columnists, 07.03.2002, http://fecolumnists.expressindia. com/print.php?content_id=3974, date accessed 3 March 2006. Mohanty, A. K., P. K. Sahu and S. C. Pati (1994) Technology Transfer in Indian Automobile Industry (New Delhi: Ashish Publishing House). Mohapatra, H. and P. Dittakavi (2003) Maruti Udyog in 2003. ICFAI Knowledge Centre, Hyderabad, India. Case study distributed by European Case Clearing House, England and USA. Mohnot, S. R. (2001) Automobile Industry 2001 and Beyond (New Delhi: Cier Intecos).

288 Bibliography Morgan, G. (2001a) ‘The Multinational Firm: Organizing Across Institutional and National Divides’ in G. Morgan, P. H. Kristensen and R. Whitley (eds) The Multinational Firm. Organizing Across Institutional and National Divides. (Oxford: Oxford University Press). Morgan, G. (2001b) ‘Transnational Communities and Business Systems’, Global Networks, 1, 113–30. Morgan, G. (2001c) ‘The Development of Transnational Standards and Regulations and their Impacts on Firms’ in G. Morgan, P. H. Kristensen and R. Whitley (eds) The Multinational Firm. Organizing Across Institutional and National Divides (Oxford: Oxford University Press). Morgan, G., D. R. Sharpe, W. Kelly and R. Whitley (2002) ‘The Future of Japanese Manufacturing in the UK’, Journal of Management Studies, 39, 1023–44. Morgan, G. and R. Whitley (2003) ‘Introduction’, Journal of Management Studies, 40, 609–16. Morris, J. (1988) ‘The Who, Why and Where of Japanese Manufacturing Investment in the UK’, Industrial Relations Journal, 19, 31–40. Morris, J., J. Lowe and B. Wilkinson (1998) ‘ “Front-end Reflections”: Supervisory Systems in the UK’s Japanese Transplants and in “Japanized” Companies’, Employee Relations, 20, 261–70. Mudambi, R. (2002) ‘Knowledge Management in Multinational Firms’, Journal of International Management, 8, 1–9. Mudd, T. (2000), ‘The Last Laugh: Czech Automaker Skoda Turns the Tables on Jokesters with Highly Efficient Operations Turning out Hot-selling Sedans’, http://www.industryweek.com/CurrentArticles/asp/articles.asp?ArticleID=898/, date accessed 10 June 2003. Mukerji, R. (2002) ‘Czeching Out, Skoda at Its Plant in the Czech Republic’, Auto, Motor & Sport India, October. Mukherjee, S. (1997) ‘VW Nears Decision on Plans for India; Skoda Model Likely’. Interview with Volkswagen AG’s Asian operations head Martin Posth, Automotive News, 71, 29. Muthukumar, R. (2004) Operations Management at Maruti. ICFAI Knowledge Centre, Hyderabad, India. Case study distributed by European Case Clearing House, England and USA. Nohria, N. and S. Ghoshal (1997) The Differentiated Network: Organizing Multinational Corporations for Value Creation (San Francisco: Jossey-Bass). Noronha, C. (2002) ‘Culture-specific TQM in China: Case Studies for Theoretical Consideration’, Asian Business and Management, 1, 125–45. Okada, A. (1998) ‘Does Globalization Improve the Employment and the Quality of Jobs in India? A Case from the Automobile Industry’, Alfred P. Sloan Foundation Project on Globalization and Jobs, Research Note No. 3, http://ipclis.mit.edu/globalization/Okada%20India.pdf, date accessed 18 June 2004. Oliver, C. (1991) ‘Strategic Responses to Institutional Processes’, The Academy of Management Review 16(1), 145–79. Oliver, N. and B. Wilkinson (1988) The Japanisation of British Industry (Oxford: Blackwell). Palit, S. K. (1998) ‘The Development of Engineering and Technical Education in India’, Global Journal of Engineering Education, 2, 317–26. Pan, Y. and D. K. Tse (2000) ‘The Hierarchical Model of Market Entry Modes’, Journal of International Business Studies, 31, 535–54.

Bibliography 289 Panini, M. N. (1996) ‘World-Views and Life Worlds: Case Studies of Industrial Entrepreneurs in Faridabad’ in A. M. Shah, B. S. Baviskar and E. A. Ramaswamy (eds) Social Structure and Change: Complex Organizations and Urban Communities, Vol. 3 (New Delhi: Sage). Parekh, H. C. (1984) ‘An Experiment in Trusteeship’, http://www.3pjobs.com/ 3PNTSite/Cfp/ M16.htm, date accessed 26 May 2003. Patton, M. Q. (1990) Qualitative Evaluation and Research Methods (Newbury Park: Sage). Pauwels, P. F. J. and P. Matthyssens (2004) ‘The Architecture of Multiple Case Study Research in International Business’ in R. Marschan-Piekkari and C. Welch (eds) Handbook of Qualitative Research Methods for International Business (Cheltenham, UK: Edward Elgar Publishing). Pearce, R. and M. Papanastassiou (1999) ‘Overseas R&D and the Strategic Evolution of MNEs: Evidence from Laboratories in the UK’, Research Policy, 28, 23–41. Penrose, E. (1959) The Theory of the Growth of the Firm (New York: Oxford University Press). Peters, E. (1999) ‘Plant Subsidiary Upgrading: Some Evidence from the Electronics Industry’ in N. Hood and S. Young (eds) The Globalization of Multinational Enterprise Activity and Economic Development (London: Macmillan). Petri, H. (2004) ‘Automobilstandort Deutschland – heute und morgen’, http://www.presse.uni-karlsruhe.de/img/presse/Gastprof_Vorlesung1.pdf, date accessed 5 December 2005. Pil, F. and J. MacDuffie (1999) ‘What Makes Transplants Thrive: Managing the Transfer of ‘Best Practice’ at Japanese Auto Plants in North America’, Journal of World Business, 34, 372–91. Pitroda, S. (2001) ‘The Magic of Team Work’, http://www.nilacharal.com/stage/ katturai/team.html, date accessed 23 March 2006. Polanyi, M. (1962) Personal Knowledge (London: Routledge and Kegan Paul). Porter, M. (1980) Competitive Strategy (New York: Free Press). Prahalad, C. K. (1975) The Strategic Process in a Multinational Corporation, School of Business Administration, Harvard University, unpublished doctoral dissertation. Prahalad, C. K. and Y. L. Doz (1987) The Multinational Mission (New York: The Free Press). Price, J. (1997) Japan Works. Power and Paradox in Postwar Industrial Relations (Ithaca/London: Cornell University Press). Pries, L. (2003) ‘Emerging Production Systems in the Transnationalisation of German Carmakers: Adaptation, Application or Innovation? New Technology, Work and Employment, 18, http://www.ruhr -uni-bochum.de/soaps/download/publ2003_lp_emergingproductionsystems.pdf, date accessed 19 December 2005. Pries, L. (2000) ‘Globalisierung und Wandel internationaler Unternehmen. Konzeptionelle Überlegungen am Beispiel der deutschen Automobilkonzerne’, Kölner Zeitschrift für Soziologie und Sozialpsychologie, 52, 670–95. Pries, L. (1999) ‘The Dialectics of Automobile Assemblers and Suppliers Restructuring and Globalization of the German “Big Three” ’ in Actes du Gerpisa, No. 25, 77–91. Procter, S. and G. Ackroyd (1998) ‘Against Japanization: Understanding the Reorganization of British Manufacturing, Employee Relations, 20, 261–70.

290 Bibliography Quintanilla, J. and A. Ferner (2003) ‘Multinationals and human resource management: between global convergence and national identity’, International Journal of Human Resource Management, 14, 363–86. Ramaswamy, E. A. (2000) Managing Human Resources: A Contemporary Text (New Delhi: Oxford University Press). Ramaswamy, E. A. (1996) ‘Wealth and Power Convert into Status: The Impact of Society on Industry’ in A. M. Shah, B. S. Baviskar and E. A. Ramaswamy (eds) Social Structure and Change. Complex Organizations and Urban Communities, Vol. 3 (New Delhi: Sage). Ratnam, C. S. V. (2001) Globalization and Labour–Management Relations – Dynamics of Change (New Delhi: Response Books). Red Herring Prospectus (2003) Maruti Udyog Ltd. 2003: IPO, http://www.myiris.com/ shares/ipo/ draft/MARUDYOG/MARUDYOG.pdf, date accessed 19 December 2005. Reference for Business (2003) Osamu Suzuki. http://www.referenceforbusiness.com/ biography/S-Z/SMC-Osamu-1930.html, date accessed 24 April 2004. Rieger, H. C. (1989) ‘Aktuelle Trends in der indischen Wirtschaftspolitik und Wirtschaftsentwicklung’ in W. Draguhn (ed.) Indien in den 90er Jahren. Politisch-soziale und wirtschaftliche Rahmenbedingungen (Hamburg: Institut für Asienkunde). Rinehart, J., C. Huxley and D. Robertson (1997) Just Another Car Factory? Lean Production and Its Discontents (Ithaca: Cornell University Press). Rogers, E. M. (2003) Diffusion of Innovations (New York/London/Toronto/Sydney: Free Press). Rosenzweig, P. M. and N. Nohria (1994) ‘Influences on human resource management practices in multinational corporations’, Journal of International Business Studies, 25, 229–51. Rosenzweig, P. and J. Singh (1991) ‘Organizational Environment and the Multinational Enterprises’, Academy of Management Review, 16, 340–61. Sachitanand, N. N. (2000) ‘Where Does Maruti Go from Here? The Hindu, http://www.hindu.com/ 2000/11/13/stories/0613000b.htm, date accessed 3 June 2003. Saka, A. (2003) Cross-national Appropriation of Work Systems, Japanese Firms in the UK (Cheltenham, UK/Northampton, MA: Edward Elgar). Saraswathi, T. S. and S. Pai (1997) ‘Socialisation in the Indian Context’ in H. S. R. Kao and D. Sinha (eds) Asian Perspective on Psychology (New Delhi/ Thousand Oaks/London: Sage). Scarbrough, H. and M. Terry (1998) ‘Forget Japan: The Very British Response to Lean Production’, Employee Relations, 20, 224–36. Schmitt, E. (1984) Indien (Berlin: Express-Edition). Schuler, R.S., P.S. Budhwar and G.W. Florkowski (2002) ‘International human resource management: review and critique’, International Journal of Management Reviews, 4, 41–70. Schuler, R.S., P. J. Dowling, and H. De Cieri (1993) ‘An Integrative Framework of Strategic International Human Resource Management’, Journal of Management, 19, 419–59. Schulz, M. (2003) ‘Pathways of Relevance: Exploring Inflows of Knowledge into Subunits of Multinational Corporations’, Organization Science, 4, 440–59. Scott, W. R. (1995) Institutions and Organizations (Thousand Oaks: Sage).

Bibliography 291 Sewell, G. and B. Wilkinson (1992) ‘Someone to Watch Over Me: Surveillance, Discipline, and the Just-in-time Labour Process’, Sociology, 26, 271–91. Sharma, K. L. (1997) Social Stratification in India, Issues and Themes (New Delhi: Sage). Sharpe, D. R. (1997) ‘Compromise Solutions: A Japanese Multinational Comes to the UK’ in R. Whitley and P. H. Kristensen (eds) Governance at Work: The Social Regulation of Economic Relations (Oxford: Oxford University Press). Shelley, V. (2002) ‘The Race Begins’, corporate report, Business India, November 11/24. SIAM (2008) Profile of the Indian Automobile Industry 2006–07 (New Delhi). SIAM (2006) The Indian Automobile Industry. Statistical Profile 2005–06 (New Delhi). SIAM (2002) Profile of the Indian Automobile Industry (New Delhi). Singh, J. (2004) ‘Multinational Firms and Knowledge Diffusion: Evidence Using Patent Citation Data’, Academy of Management Best Conference Paper. Sinha, D. (1999) ‘Approaches to Indigious Mangement’ in H. S. R. Kao, D. Sinha and B. Wilpert (eds) Management and Cultural Values. The Indigenization of Organizations in Asia (New Delhi: Sage). Sinha, J. B. P. (1990a) Work Culture in the Indian Context (New Delhi: Sage). Sinha, J. B. P. (1990b) ‘The Salient Indian Values and Their Socio-ecological Roots’, The Indian Journal of Social Science, 3, 477–488. Sinha, J. B. P. (1980) The Nurturant-task Leader: A Model of the Effective Executive (New Delhi: Concept Publishing Co.). Sinha, J. B. P. and D. Sinha (1994) ‘Role of Social Values in Indian Organizations’, in H. S. R. Kao, D. Sinha and N. Sek-Hong (eds) Effective Organisations and Social Values (New Delhi: Sage). Sinha, P. (2002) ‘Trade Unions in India – Spreading Wings’ in W. Dragun (ed.) Indien 2002 (Hamburg: Institut für Asienkunde). Skoda (2005) ‘Skoda World Wide’, http://www.skoda-auto.com/global/company/ worldwide/, date accessed 11 November 2005. Skoda (2006) ‘Production abroad’, http://new.skoda-auto. com/COM/worldwide/ production/Pages/ProductionAbroad.aspx, date accessed 10 May 2008. Smith, C. and T. Elger (2000) ‘The Societal Effects School and Transnational Transfer: The Case of Japanese Investment in Britain’ in M. Maurice and A. Sorge (eds) Embedding Organizations: Societal Analysis of Actors, Organisations and Socio-Economic Context (Amsterdam/Philadelphia: John Benjamins Publishing Company). Smith, C. and T. Elger (1998) ‘Greenfields and ‘Wildebeests’: Management Strategies and Labour Turnover in Japanese Firms in Telford’, Employee Relations, 20, 271–84. Som, A. (2004) Masuki Limited: Challenges of Redesign of a Japanese JV in India. (Paris: ESSEC Business School). Sorge, A. (2005) The Global and the Local: Understanding the Dialectics of Business Sytems (Oxford: Oxford University Press). Sorge, A. (2004) ‘Cross-national Differences in Human Resource and Organization’ in A.-W. Harzing and J. van Ruysseveldt (eds) International Human Resource Management (London: Sage). Sorge, A. (1995a) ‘Cross-national Differences in Personnel and Organization’ in A.-W. Harzing and J. van Ruysseveldt (eds) International Human Resource Management: An Integrated Approach (London: Sage).

292 Bibliography Sorge, A. (1995b) ‘Labour Relations, Organization and Qualifications’ in A.-W. Harzing, J. van Ruysseveldt and J. van Hoof, J. (eds) Comparative Industrial & Employment Relations (London: Sage). Sorge, A. (1993) ‘Arbeit, Organisation und Arbeitsbeziehungen in Ostdeutschland’, Berliner Journal für Soziologie, 4, 549–67. Sorge, A. (1991) ‘Strategic Fit and the Societal Effect: Interpreting Cross-National Comparisons of Technology, Organization and Human Resources’, Organization Studies 12, 161–90. Sorge, A. and M. Warner (1986) Comparative Factory Organization – An Anglo German Comparison of Management and Manpower in Manufacturing (Aldershot: Gower). Stewart, P. (1998) ‘Out of Chaos Comes Order: From Japanization to Lean Production: A Critical Commentary’, Employee Relations, 20, 213–23. Strang, D. and J. W. Meyer (1993) ‘Institutional Conditions for Diffusion’, Theory and Society, 22, 487–511. Subhadra, K. (2002) Fiat’s Indian Journey (Hyderabad: ICFAI Center for Management Research). Subramaniam, M. and N. Venkataraman, (2001) ‘Determinants of transnational new product development capability: testing the influence of transferring and deploying tacit overseas knowledge’, Strategic Management Journal 22, 359–78. Susman, G. I. (1976) Autonomy at Work: A Sociotechnical Analysis of Participative Management (New York: Praeger). Sutton, J. (2004) ‘The Auto-component Supply Chain in China and India: A Benchmarking Study’, http://personal.lse.ac.uk/sutton/auto_component_sutton. pdf, date accessed 17 April 2004. Suzuki (2004) ‘Suzuki Outline History’, http://www.globalSuzuki.com/corp_info/, date accessed 3 April 2006. Sydow, J. (1985) Der soziotechnische Ansatz der Arbeits- und Organisationsgestaltung. Darstellung, Kritik. Weiterentwicklung (Frankfurt/M.: Campus). Szulanski, G. (1996) ‘Exploring Internal Stickiness: Impediments to the Transfer of Best Practice within the Firm’, Strategic Management Journal, 17, 27–43. Taggart, J. H. (1997) ‘Autonomy and Procedural Justice: A Framework for Evaluating Subsidiary Strategy’, Journal of International Business Studies, 5, 447–68. Tayeb, M. H. (1988) Organisations and National Culture. A Comparative Analysis New Delhi: Sage). Taylor, D. and D. Brunt (2001) Manufacturing Operations and Supply Chain Management – The Lean Approach (London: Thomson). Taylor, B. (1999) ‘Patterns of Control within Japanese Manufacturing Plants in China: Doubts about Japanization in Asia’, Journal of Management Studies, 36, 853–73. Taylor S., S. Beechler, and N. Napier (1996) ‘Toward an Integrative Model of Strategic International Human Resource Management’. Academy of Management Review 21, 959–85. Teigland, R., C. F. Fey and J. Birkinshaw (2001) ‘Knowledge Dissemination in Global R&D Operations: An Empirical Study of Multinationals in the High Technology Electronics Industry’, Management International Review Special Issue, 1, 49–77. Tempel, A., T. Edwards, A. Ferner, M. Muller-Camen and H. Wächter (2006) ‘Subsidiary responses to institutional duality: Collective representation practices of US multinationals in Britain and Germany’, Human Relations, 59, 1543–70.

Bibliography 293 Tempel, A. and P. Walgenbach (2003) ‘Global Standardization of Organisational Forms and Management Practices?, Combining American and European Institutionalism’, paper presented at the 3rd Conference of the European Academy of Management, Milan. The Financial Express (2002) ‘After Alto, Maruti Plans New Model for Overseas Markets’, http://www.financialexpress.com/fe_full_story.php?content_id= 14339, date accessed 3 March 2006. The Hindu (2005) ‘Skoda Car Sales up’, 05.01.2005, http://www.hinduonnet.com/ 2005/01/05/stories/ 2005010502421600.htm, date accessed 3 March 2006. The Hindu Business Line (2001) ‘Skoda India May Be Allowed to Import CBUs for Octavia’, 21.11.2001, http://www.blonnet.com/2001/11/22/stories/0222079q. htm, date accessed 3 March 2001. Théret, B. (1997) ‘Méthodologie des Comparaisons Internationales, Approches de l’Effet Sociétal et de la Régulation: Fondements pour une Lecture Structuraliste des Systèmes Nationaux de Protection Sociale’, L’Année de la Régulation, 1, 163–228. Thete, A. R. (1999) A Comparative Study of the System of Technical Education in Germany and India with Special Reference to Maharashtra State, PhD. dissertation (Aurangabad: Marathwada College of Education). The Tribune (1999) ‘Fiat Investment’, 06.01.1999, http://www.tribuneindia.com/ 1999/99jan06/ biz.htm, date accessed 3 March 2006. Trist, E. (1981) The Evolution of Soci-technical Systems. A Conceptual Framework and an Action Research Program, occasional paper no. 2 (Toronto: Ontario Quality of Working Life Centre). Trist, E. L. (1975) ‘Sozio-technische Systeme’ in W. G. Bennis, K. D. Benne and R. Chin (eds) Änderung des Sozialverhaltens (Stuttgart: Klett-Cotta). Tsai, W. (2001) ‘Knowledge Transfers in Intraorganizational Networks: Effects of Network Position and Absorptive Capacity on Business Unit Innovation and Performance’, Academy of Management Review, 44, 996–1004. Turnbull, P. J. (1986) ‘The Japanisation of Production and Industrial Relations at Lucas Electrical’, Industrial Relations Journal, 17, 193–206. UNIDO (2003) ‘Guide to Supplier Development – For Programmes to Be Implemented by Industrial Subcontracting and Partnership Exchanges (SPXs)’, Vienna, http://www.unido.org/file-storage/download/?file_id=9607, date accessed 13 May 2005. UNIDO (1999) ‘In-depth Evaluation of Selected UNIDO Activities on Development and Transfer of Technology’, Component 4: Transfer of Technology in Subcontracting, http://www.unido.org/ userfiles/PuffK/dtt-com499.pdf, date accessed 23 March 2006. Vahland, W. (2005) ‘Volkswagen AG. Central European Field Trip: Skoda Auto a.s.’, http://www2.volkswagen-ir.de/fileadmin/vw- ir2/dokumente/inv_ presentations/2005/Q2/20050523_ Mr_Vahland.pdf, date accessed 12 December 2005. van Maanen, J. (1983) ‘Reclaiming Qualitative Methods for Organizational Research: A Preface’ in J. van Maanen (ed.) Qualitative Methodology (Beverly Hills, CA.: Sage). Venkatachari, J. (2000) ‘New Maruti Initiatives’, Business Domain.com., http:// www.domain-b.com/ companies/companies_m/maruti/20000309newmaruti_ initiatives.html, date accessed 9 March 2000. Venkataramani, R. (1990) Japan Enters Indian Industy. The Maruti–Suzuki Joint Venture (New Delhi: Radiant Publishers).

294 Bibliography Virmani, B. R. (2000) Managing People in Organisations: The Challenge of Change (New Delhi: Sage). Volkswagen, A. G. (2005) ‘Investor Relations – Group Structure’, http://www. volkswagen-ir.de/Group_ Structure.473.0.html, date accessed 11 November 2005. Volpato, G. (2000/2001) Fiat Auto: Toward Globalization, IMVP Working Papers, http://imvp.mit. edu/papers/0001/volpato1.pdf, date accessed 28 November 2005. Westney, D. E. (2001) ‘Japan’ in A. M. Rugman and T. L. Brewer (eds) Oxford Handbook of International Business (Oxford: Oxford University Press). Westney, D. E. (1999) ‘Organisation Theory Perspectives on the Cross-border Transfer of Organisational Patterns’ in J. K. Liker, W. M. Fruin and P. S. Adler (eds) Remade in America: Transplanting and Transforming Japanese Management Systems (New York: Oxford University Press). Westney, D. E. (1993) ‘Institutionalization Theory and the Multinational Corporation’ in S. Ghoshal and D. E. Westney (eds) Organization Theory and the Multinational Corporation (New York: St. Martins Press). Westney, D. E. (1987) Imitation and Innovation: The Transfer of Western Organizational Patterns to Meiki Japan (Cambridge, Mass: Harvard University Press). Westney, D. E. and S. Zaheer (2001) ‘The Multinational Enterprise as an Organization’ A. M. Rugman and T. L. Brewer (eds) Oxford Handbook of International Business (Oxford: Oxford University Press). White, R. E.and T. A. Poynter (1984) ‘Strategies for Foreign-owned Subsidiaries in Canada’, Business Quarterly, 59–69. White, M. R. and M. Trevor (1983) Under Japanese Management (London: Heinemann). Whitley, R. (2001) ‘How and Why are International Firms Different? The Consequences of Cross-Border Managerial Coordination for Firm Characteristics and Behaviour’ in G. Morgan, P. H. Kristensen and R. Whitley (eds) The Multinational Firm. Organizing Across Institutional and National Divides (Oxford: Oxford University Press). Whitley, R. (1999) Divergent Capitalisms. The Social Structuring and Change of Business Systems (Oxford: Oxford University Press). Whitley, R. (1998) ‘Internationalization and Varieties of Capitalism: The Limited Effects of Cross-national Coordination of Economic Activities on the Nature of Business Systems’, Review of International Political Economy, 5, 445–81. Whitley, R. (ed.) (1992) European Business Systems: Firms and Markets in Their National Contexts (London: Sage). Wilkinson, A. and P. Ackers (1995) ‘When Two Cultures Meet: New Industrial Relations at Japanco’, International Journal of Human Resource Management, 6, 849–71. Wilkinson, B., J. Gamble, J. Humphrey, J. Morris and D. Anthony (2001) ‘The New International Division of Labour in Asian Electronics: Work Organization and Human Resources in Japan and Malaysia’, Journal of Management Studies, 38, 675–95. Wilkinson, B., J. Morris and M. Munday (1995) ‘The Iron Fist in the Velvet Glove: Management and Organization in Japanese Manufacturing Transplants in Wales’, Journal of Management Studies, 32, 819–30.

Bibliography 295 Wilms, W. W., A. J. Hardcastle and D. M. Zell (1994) ‘Cultural Transformation at NUMMI’, New United Motor Manufacturing Inc., Sloan Management Review, 36, 99–114. Womack, J. P., D. T. Jones and D. Roos (1990) The Machine that Changed the World (New York: Macmillan). World Trade Organisation (1998) World Trade Policy Review India (Geneva: WTO). Yamin, M. and J. Otto (2004) ‘Patterns of Knowledge Flows and MNE Innovative Performance’, Journal of International Management, 10, 239–58. Yin, R. K. (2003) Case Study Research. Design and Methods (Thousand Oaks/London/New Delhi: Sage). Zucker, L. G. (1977) ‘The Role of Institutionalization in Cultural Persistence’, American Sociological Review, 42, 726–43. Zucker, L. G. (1991) ‘Role of Institutionalization in Cultural Persistence’ in W. Powell and P. J. DiMaggio (eds) The New Institutionalism in Organizational Analysis (Chicago/London: The University of Chicago Press).

Index Ceremonial adoption 108 Challenge 3, 33, 86, 97, 126, 131, 144, 246, 273 ‘Challenge 50’ 112 of transfer 3 Charter definition of 31–2 of subsidiary 31–2 Chennai (former Madras) 92, 93, 168 China 5, 65, 99, 100, 137, 173, 214, 216, 217, 246 Citroën 137 Company union 86–7, 125–9, 205 Completely knocked down (CKD) kits 259, 260 Fiat 138 Mercedes-Benz India 173–8, 180, 186–9, 192–3, 206–7, 209–10, 213 Skoda 214–16, 218–20, 222–3, 226–8, 243–4, 249 see also Semi knocked down (SKD) kits Components 25–6, 44, 60–1, 73, 97, 263, 275 Fiat 166, 168–9, Maruti–Suzuki 129–33 Mercedes-Benz India 188, 207, 209, Skoda 214, 222, 227–9, 233, 243, 246 Conditions institutional context conditions 26, 146, 262 strategic context conditions 7, 34, 77, 114, 116, 120, 152, 179, 186, 230 Configuration mandate 129, 149, 154, 165, 178, 265–9 Consumers, Indian 91 consumer electronics 11 consumer goods 87 consumer industries 88 consumer preferences 30

Active adaptation 11, 12 Adaptation modes 11, 20, 22, 35, 39, 42 see also recontextualization modes Adaptation requirements 11, 16, 25, 33, 205 Adoption 14, 18, 19, 33, 38, 109, 125, 226 patterns of 19 American Institutionalism 17–18, 22, 39, 45 see also Neo-Institutionalism Application–Adaptation Dilemma 11 Audi 91, 92, 216, 217 Aurangabad 70, 93, 168, 218 Automotive Components Manufactures Association (ACMA) xv, 71, 91, 96, 97, 130 Balance of payment (BOP) crisis 88 Bangalore, Karnataka 93, 168 Best practice 11, 15, 122 BMW 91–2 Brazil 136–7, 142, 151, 156, 168–9, 214 Broad-banding 88–90 Business system 22, 23 national business system approach 22 Capability 104, 110, 112, 130, 131, 143, 145 gap 104, 143, 145, 272 of local/host context 114, 267–8 Capacity 88, 90, 97, 101, 140, 141, 148, 152, 165, 175, 176, 216, 218, 244 Capitalism 15, 17, 22 Capital–labour relations 14 Capitalist social relations 14 Case study 9, 13, 63–7, 72–4, 78, 119 Caste 80–1, 83, 233

296

Index 297 Context business context 3, 5, 7, 11, 40, 77, see institutional context; strategic context work context 78, 82–3, 118, 155, 190, 202 Context free 15 Contextual constitution 4, 16, 18, 22, 23, 28, 30–2, 35, 36, 38–40, 45–6, 106, 273 Contextual distance 4, 7, 41, 51, 54, 58, 60 Contextual misfits 36, 54, 58, 270, 270 see also misfits Contingencies 15, 21, 28–9, 40, 42, 272 see also strategic context Continuous improvement 116, 117, 123, 153–7, 193, 196 see also Kaizen Control cost control 58, 115 quality control 154, 156, 157, 236 Convergence convergence of consumer preference 30 global convergence 22 Coordination 29, 30, 205 Culture xi, 10, 12, 36, 86, 100, 103, 115, 116, 119, 121, 124 Customization 6, 46–7, 50–2, 100, 106, 113–15, 134, 152, 171, 172, 187, 189, 211–13, 223, 230, 247–8, core findings 251–2, 254, 257, 259–60 Czech Republic 215, 226, 228–9, 232, 239 Daewoo 92, 235 Daimler-Benz (DB) 66, 173–8, 195, 205–10 DaimlerChrysler (DC) 66, 174–6, 204, 205, 207–10 Data analysis 72–4 Data collection 67–72, 74 method of 67, 74

DCM 92 Demand market conditions 55–7, 89, 104, 133, 172, 177, 185, 190, 221, 249, 251 compare supply market conditions Differentiated network 31, 33, 40, 55 see also transnational Dominance effects 15 Economic reform 87–8 Economies of scale 30, 246 Education 77, 78–83 educational systems 54, 72, 223, 254, 256 caste, stratification principles 80, 81, 107, 118, 155, 193 higher education 79, 83, 97–8, 196, 223 Ethnocentrism 32 Eicher 71 Embeddedness 63 social 12, 18 of subsidiary 4, 18, 23–5, 28, 37–9, 54, 57, 254, 256 Emulation 14 Entry mode 57, 60, 61, 64, 143, 181, 216, 220, 259, 262–4 definition of 59–60 of Fiat India 138–9 of Maruti–Suzuki 100–1 of Mercedes-Benz India 174–5 of Skoda Auto India 216–18 see also establishment and equity mode Environment selection, creation, change 49 Environment Strategy Structure perspective/paradigm 38 Environmental forces 29–31 Environmental pressures 29–31 Equity mode 59–61, 66, 68, 100, 101, 106, 108, 109, 138, 139, 172, 174, 175, 180, 216, 217, 250 see also entry mode Establishment mode 11, 59–61, 66, 68, 85, 100, 104, 138, 174, 216, 228, 250, 258, 263–5, 270–1 see also entry mode

298 Index European Institutionalism 17, 22, 272, 274 Export export requirements 90, 208 production for 58, 136, 175,

Fabbrica Integrata 141, 145, 149, 161, 274 Family 77, 81, 198 Fiat 136–72 Fiat India (FI) 92, 136–72, 250–73 India Auto Limited (IAL) 66, 139 Ford 90, 94, 95, 137, 235 Ford India 92, 94, 95 Foreign Direct Investment (FDI) 3, 8, 9, 88, 90, 101, 129, 186, 207, 209, 213, 218, 229, 230 Foreign Exchange Regulation Act (FERA) 88, 90 Foreign ownership development 69

Gandhi, Indira 65 Gandhi, Rajiv 88 Gandhi, Sanjay 66, 100 General Motors (GM) 92 Generic product strategies 5, 28, 42, 270 see also generic strategies and global product strategies Generic Strategies 5, 28, 42, 65, 57–8, 100, 137–8, 174, 215–16, 250, 254, 257–63, 270 definition of 57–8 see also global product strategy Germany 174, 177, 181, 185, 187–91, 196, 199–202, 206, 209, 226, 232, 236 Global efficiency 30 Global integration 29–30 Global product strategy 57–8, 100, 137–8, 174, 215–16 see also generic strategies Gujarat 92 Gurgaon, Haryana 70, 71, 93, 113, 126, 128

Halol, Gujarat 92 Haryana 70, 93, 101 Hindustan Motors (HM) 92 Heavy Commercial Vehicle (HCV) 96 Honda Motors 92–3 Honda-Siel 92–3 Hosur, Karnataka 92 Human resources (Indian automobile industry) 97, 98, 103, 116–17, 141, 153–62, 192, 202, 205, 230–8, 255–6 Hybridization breadth of process 25–6 definition of 5, 45–7 modes of 25 nature of process 25–6 of production system 44–7 ratios of 12 trajectories of 26–7 Hybridization outcomes definition of 45–7 Hybridization profile compared 251–2 definition of 5 explained 251–70 Fiat India’s 140–72 Maruti–Suzuki’s 102–35 Mercedes-Benz India’s 176–213 Skoda Auto India’s 219–49 Hybridization research 3–5, 14–16, 65, 250, 271–2 definition of 3 Hybridization theory 4 Hyundai Motors 87, 93–5 Igatpuri, Maharashtra 92 Import tariffs 90–1, 129, 149, 168, 189, 208, 227, 244, 247, 249, 275 Incentive system 123, 126–8 India automobile industry 87–98 business context 77–98 FDI regime 88–95 Indian Automobile industry major players 92–3 suppliers 95, 97 Indian Institute of Technology (IIT) 97

Index 299 Indian passenger car market development of 91, 94, 95 by manufacturer 92–3 market share 94, 95 by segment 96 Indore, Madhya Pradesh 92 Industrial Disputes Act 83 Industrial Relations (IR) Fiat India’s 160–5 India’s 82–7 Maruti–Suzuki’s 125–9 Mercedes-Benz India’s 203–6 Skoda Auto India’s 240–3 Industrial Training Institute (ITI) 79, 83, 109, 122, 158, 201, 233, 235 Institutional context 51–4, 99–249, 250–73 constraints and opportunities 26–7 definition of 54 India’s 77–87 Institutional distance 7, 8, 14, 18, 19, 41, 47–54, 104, 117, 125–8, 143, 147, 163, 190, 192, 204, 225, 231, 234, 237, 240, 251–6 Institutional duality 17–19 Institutionalist approaches 16–28 Integration-responsiveness framework 28–9, 39–40 Internationalization experience 173, 260–1, 270 Internationalization strategies 29–31, 40, 259, 270 Inventories 97, 113, 135, 165, 168, 207 Italy 136, 138–41, 146, 151–3, 158–69, 274 Isomorphism 17–20 Japan 12, 13 20, 21, 99–135, 269 Japanese production system 9–14 Japanization literature 8–16 Jati 81 Just in sequence (JIS) 206, 208 Just in time (JIT) 129, 165, 274 Kaizen 117, 124, 274, Kirloskar 93 Knowledge flow perspective 33–6, 41

Kosai plant 102–14 Labour relations see industrial relations Labour Process perspective 13–16 Labour unrest 116, 161–2 Lean Production perspective 9–13 Learning cycle model 35 Level of analysis 9, 13, 18, 23, 28, 29, 33, 64 Light commercial vehicles (LCV) 96 Local content 47, 89–92, 129, 149, 166–70, 178–9, 186–9, 207–9, 221, 227, 230, 244–9 Local differentiation 30 see also environmental forces Local responsiveness 29–30 see also environmental forces Logistics 97, 113, 129, 130, 133, 167–8, 179, 188, 206–9, 222, 229–30, 243–4, 249 Madhya Pradesh 71 Maharashtra 70, 92–3, 138–40, 152 Mahindra & Mahindra 93 Mahindra Renault 93 Management by objectives 78 Manufacturers in India 92–3 Manesar, Haryana 93, 101 Market reform 174 see also economic reform Market segmentation 91 Market share Fiat’s 139–40 By manufacturer 94–5 Maruti–Suzuki’s 102 Mercedes-Benz India’s 175–6 By segment 96 Skoda Auto India’s 218–19 Maruti Udyog Ltd. 93, 99–100, 113, 126–7, 132, 135, Maruti Production System 113 Maruti–Suzuki India (MSI) 7, 65–73, 93, 99–135, 137, 159, 170, 182–3, 186, 217 Mass production 190 of peoples’ car 102

300 Index Mazda 90 Memorandum of Understanding (MoU) 90–1 Mercedes-Benz (MB) 66–7, 68, 173–6 Mercedes-Benz India (MBI) 7, 65–73, 69, 84, 93, 173–213, 215, 219, 226, 251–68 Mercedes-Benz Production System 177 Middle class 102 Misfits 6, 10, 21–2, 35–6, 42, 47–63, 104–34, 143–71, 178–212, 221–48, 254–72 kinds of 6, 47–57 see also recontextualization pressure Mitsubishi 90, 93, 173 Model diffusion 25 Monopolies and Restrictive Trade Practices Act (MRTP) 89–90 Multinational enterprises Strategies of 29–31 Structures of 29–31 Multi Utility Vehicles (MUVs) 92, 96 Mumbai (formerly Bombay) 92–4, 138–9, 148, 152, 168

Nashik, Maharashtra 92 New Delhi 71, 82, 168 New Institutionalism 17–18, 20, 254, see also American Institutionalism Noida, Uttar Pradesh 92 North America 9, 11

Opel 92 Organization structure 44, 46, 103, 106, 142, 145, 177, 180, 220 Organizational relations 45–6 inter-organizational relations 45–6 Outright application 11 Outsourcing 129, 131, 133, 141, 165, 169, 188 Overcapacity 95, 216 Ownership development 69

Passive adaptation 11–12 Peugeot 93

Phased Manufacturing Program (PMP) 89–90 see also Local content Pithampur, Madhya Pradesh 71, 92 Polytechnics 79, 83 Public Sector Undertakings (PSUs) 100, 106–11, 124 Premier Automobiles Ltd. (PAL) 66, 91–4, 136, 138, 146, 163 Process organization 44, 46, 103, 140, 189 Profit strategy 27, 137, 215 see also generic strategy and global product strategy Production location 92–3 Fiat’s 139–40 Maruti–Suzuki’s 101–2 Mercedes-Benz India’s 175–6 Skoda Auto India’s 218–19 Product ownership 143–5, 172, 267–8, 273 Production model 26 Production system definition of 44–5 Functional differentiation 44, 103–6, 142–5, 177–80, 220–3, 251–4 Hierarchical differentiation 44, 106–10, 145–7, 180–5, 223–6, 251–4 Industrial relations 45, 125–9, 160–5, 203–6, 240–3, 251–4 Supplier relations 45, 129–35, 165–72, 206–13, 243–9, 251–4 Technical configuration 44–5, 110–15, 147–52, 185–90, 226–30, 251–4 Work organization and human resource profile 45, 115–25, 152–60, 190–203, 230–40, 251–4 Production template 43, 47, 102–3, 140–2 definition of 47 Productive model 25 requirements 26 Productivity 89, 112, 114–16, 118, 123, 132, 230, 233

Index 301 Quality Quality circle 117, 119, 124, 155, 196 Quality system 222, 231–2, 235–6, 238 Recontextualization kinds of 36, 47–51 see also recontextualization pressure compare transfer scenarios; recontextualization modes Recontextualization mode 6, 36, 42, 49, 51–4, 57, 58, 60, 61, 63, 104–34, 143–71, 178–212, 221–48, 254–72 Recontextualization pressure 6, 35, 47–51, 104–34, 143–71, 178–212, 221–48, 254–72 see also misfit Relational context 19–20, 39 Research gap 3, 7 Research traditions 4, 9, 14, 17, 51 Resources and capabilities 31–2, 60 Responsibility taking of 77–8, 82–3, 117, 119, 156, 193, 195–204, 234 Revised application 11–12 Russia 215 SEAT 214, 216 Semi knocked down (SKD) kits 260, 265 Mercedes-Benz India 173–4, 176, 178, 180, 186 Skoda 217–20, 222–3, 226–9, 243–4, 249 Skoda 215–16 Skoda Auto India (SAI) 65, 67, 69–73, 93, 214–49, 251–69 Skoda Production System 219, 230–7, 265 Socialization 77–8, 83, 121, 124, 157–60, 197, 237, 239 Social constitution of organizations 16–17 Societal effect (approach) 22, 54, 254 Society for Indian Automobile Manufactures (SIAM) 71, 95–7, 130, 176, 218, 219

Social stratification 80–2 Socio-professional distance/demarcation Staffing policy 231, 234, 240, 242, 262, 267–9 Standard Motors Products of India Ltd. (SMPIL) 91 Strategic choices 4–7, 57–63 corporate level 58–9 Fiat’s 137–9 Maruti–Suzuki’s 66, 68, 100–1 Mercedes-Benz India’s 66, 68, 174–5 Skoda Auto India’s 66, 68, 215–18 subsidiary level 59–63 Strategic context 7, 28, 34, 54–7, 77, 87, 99–249, 255–9 definition of 57 Strategic distance 7, 33, 41, 51–9, 144–5, 154, 172, 174–80, 185–7, 191–2, 196, 201, 207–10, 174, 177–80, 185–7, 190–2, 196, 201, 207–10, 215, 220, 222, 226–7, 231–3, 243–4, 250–62 Strategic role perspective 28, 31–3, 41, 55, 173, 215, 272 see also subsidiary role Strikes 84, 126–7, 139, 162–5, 206 Subsidiary evolution 31 Definition of 31 see also strategic role charter 31–2, 57 Subsidiary role 31, 34, 39–41, 55 Subsidiary production system 7 see also production system Supply market conditions 95–8, 104–5, 116, 133, 166, 249, 251, 254–5 Surajpur, Uttar Pradesh 92 Suzuki Motors Corp. (SMC) 66, 99, 103 System imperatives of capitalism 55 Talegaon, Maharashtra 92 Task environment 11, 37, 177, 179, 221, 222

302 Index Tata Motors 66, 92–5, 137, 139 see also TELCO Tata Nano 93, 102 Technical configuration 46, 110, 134, 147, 171, 185, 226, 252 TELCO 66, 174–6, 181–2, 189, 192, 195, 200–1, 204–5, 207, 209 see also Tata Template availability 254, 259–66 see production template Theoretical foundations 1, 5, 7, 8–42 Theoretical sampling 64–5, 68 Tierization of suppliers 130, 166–7, 207 Toyota 95, 93 Toyota Kirloskar 93 Trade unions see unions Training Vocational and technical in India 79, 83, 107, 119–22, 124, 133, 153–64, 190–7, 201, 219, 230–7, 255–6 Transfer capability of 15 content 10, 24 cross-national 9, failure 26 intent 28, 103, 130, 140, 143, 145, 147, 165–6, 182, 219–20, 229, 238, 254, 260–2 of knowledge 30 of organizational forms and practices 15–16, 21, 38, 260 of production systems 9, 11, 13 propensity of 11, 13, 15, 32, 36–7, 65–2, 270 of templates 6, 36, 47–8, 53, 57–61,102, 106, 110–15, 126, 143–4, 149, 165, 171–2, 180–1, 189, 208–10, 224, 226, 229, 244, 247, 253, 260–4, 269

Transferability question 17 Transfer outcomes complex view 41 definition of 45–7 dichotomous views 41 see also Hybridization outcomes Transfer scenarios 53, 56 compare misfits; recontextualization pressure; recontextualization modes Transnational (corporation and strategy) 29–33, 38–40 Transportation 30, 86, 97, 124, 134, 166–8, 172

Unions 82–7, 111, 125–33, 159–65, 205–6, 241–3 United States 9, 12, 173 Universities 79 UTE 142, 143, 145, 152–60 Utility Vehicles 102 Uttar Pradesh 92, 93 Uttarpara, West Bengal 92

Vadodara, Gujarat 92 Varieties of Capitalism (VoC) 17, 22 Varna 81 Vertical Integration 104–5, 150, 179, 187–9, 222, 227, 229 Volkswagen (VW) 67, 68, 93, 215–16,

Work discipline 77, 86, 117–19, 124, 195, 201, 203–6, 234–5, 237 World car 68, 136–8, 140–2, 258, 267 Fiat’s world car project 136, 138, 140–2, 145 Worldwide diffusion 30

Zaheerabad, Andhra Pradesh 93