Corporate Research and Development in a Late Industrializing
Transcrição
Corporate Research and Development in a Late Industrializing
Corporate Research and Development in a Late Industrializing Context The Case of Singapore DISSERTATION der Universität St. Gallen, Hochschule für Wirtschafts-, Rechts- und Sozialwissenschaften (HSG) zur Erlangung der Würde einer Doktorin der Wirtschaftswissenschaften vorgelegt von Yvonne Elise Helble aus Deutschland Genehmigt auf Antrag der Herren Prof. Dr. Li Choy Chong und Prof. Dr. Martin Hilb Dissertation Nr. 2848 Difo-Druck GmbH, Bamberg 2004 Die Universität St. Gallen, Hochschule für Wirtschafts-, Rechts- und Sozialwissenschaften (HSG), gestattet hiermit die Drucklegung der vorliegenden Dissertation, ohne damit zu den darin ausgesprochenen Anschauungen Stellung zu nehmen. St. Gallen, den 4. November 2003 Der Rektor: Prof. Dr. Peter Gomez II Acknowledgements This dissertation could only be achieved with the support and collaboration of many individuals and organizations. First, I would like to sincerely thank Prof. Dr. Li Choy Chong, my doctoral advisor, and Prof. Dr. Martin Hilb, my co-supervisor. I am deeply indebted to Prof. Dr. Li Choy Chong, who far exceeded his duties as doctoral advisor in terms of academic guidance and support. Whether in Switzerland or Singapore, he would always take time to discuss my research topic and he continuously encouraged me to progress further. I would also like to express my sincere gratitude to Prof. Dr. Martin Hilb for his enthusiasm, positive motivation and academic guidance. I would also like to extend my gratitude to Prof. Dr. Arnoud De Meyer, INSEAD Deputy Dean, Dean of Administration, and Prof. Dr. Hellmut Schütte, Dean of Asia Campus, for accepting me as Visiting Research Associate at INSEAD, while carrying out field research in Singapore and during the writing stage in Fontainebleau, France. I would especially like to thank Prof. Dr. Arnoud De Meyer for his academic expertise. As a pioneer in the field of international R&D management, he gave me very valuable academic guidance. I considered it a great privilege to be able to discuss my research with him. I am also indebted to Prof. Dr. Roberto Mariano and Prof. Dr. Augustine Tan for accepting me as Visiting Research Associate at the Wharton-SMU (Singapore Management University) Research Center in Singapore. In addition, I greatly appreciate the support from Dr. Jasbir Singh from the Agency of Science, Technology and Research, Singapore. This dissertation would not have been accomplished without interviews with many executives who took time off from their hectic schedules to discuss their R&D activities and strategies with me. I would particularly like to thank Prof. Dr. Paul Herrling, Head of Corporate Research at Novartis, Dr. Thomas Keller, Head of Chemistry at the Novartis Institute for Tropical Diseases, and Dr. Richard Harrison, Head of Staff of Novartis Pharma Research. Special thanks go to Prof. Dr. Paul Herrling for his exceptional support. I would also like to thank Mr. Ah Bee Goh, III Managing Director of Leica Instruments Singapore (LIS), and Ms. Germaine Tan, Senior R&D Manager at LIS, for their extraordinary help and kind support. I am also greatly indebted to Dr. Santosh Mishra, Managing Director of Lilly Systems Biology Singapore, Mr. Thomas Frischmuth, Managing Director of Siemens Singapore, and Mr. Michael Tiefenbacher, Vice President, Development Center Singapore, Infineon Technologies. I am truly thankful to the Schweizer Nationalfonds for granting me a scholarship, which enabled me to be a Visiting Research Associate at INSEAD in Singapore and France and at the Wharton-SMU Research Center in Singapore. I am equally indebted to the Cusanuswerk e.V. for granting me a PhD scholarship. I would like to thank in particular Dr. Ingrid Reul of the Cusanuswerk e.V. In addition to my supervisors, the interview partners and the scholarship institutions, a number of friends have been important during my doctoral studies. My two friends, Dr. Kyung-won Kim and Marla Kameny, were critical for my social life during my doctoral studies. Special thanks go to Dr. Kyung-won Kim for proofreading and encouragements in the final phase of my dissertation. I would also like to thank Nicole Köhmstedt and Andreas Kühn for their wonderful friendship and for being the best hosts in St. Gallen! Furthermore, I would like to thank Verena Hässig for her friendship and support and for being my Swiss guide. Special thanks also go to Dirk Böhe, who gave me excellent research suggestions and comments, and to Dr. HansPeter Hentze, who provided me with the inspiration for my dissertation topic. I would also like to thank Linda, Wilson and Ethan Fong for being my family ‘away from home’ during my stay in Singapore. Their loving support, friendship and kindness made my stay in Singapore unforgettable. I would also like to sincerely thank Doris Lee, Wang Liang Toon and Alin Nainggolan for their extraordinary friendship during my stay in Singapore. The same is true for my friend Cheihwee Chua and Mr. and Mrs. Delaunay, my hosts in Fontainebleau. IV My brother Matthias Helble and his girl-friend Nathalie Benoît are highly regarded for their inexhaustible patience in editing this thesis and for giving me continuous moral support in the final writing stage. I would especially like to thank Matthias for always being a great brother. Finally, I would like to express the most sincere appreciation to my parents, Johanna and Carl Helble, who always supported me in their best way during my academic education. Without their love and unconditional support, I would not be where I am today. I would especially like to thank my mother, the best mum in the world, for always being there for me. Her love and encouragement make it all worthwhile. This thesis is dedicated to my loved ones, my parents, my brother and my grandmother. St. Gallen, November 2003 Yvonne Elise Helble V This dissertation is dedicated to my parents Johanna und Carl Helble, to my brother Matthias Helble and to my grandmother Maria Helble. VI Wer glaubt, etwas zu sein, hat aufgehört, etwas zu werden. (Sokrates) VII Table of Contents EXECUTIVE SUMMARY ___________________________________________ IX 1 INTRODUCTION ________________________________________________1 2 LITERATURE REVIEW __________________________________________6 2.1 Definition of International Corporate R&D Activities _________________6 2.2 Development of the Literature on International R&D__________________7 2.2.1 Determinants of R&D Internationalization ________________________7 2.2.2 International R&D Organization and Management ________________13 2.2.3 R&D Internationalization Process______________________________21 2.3 3 Research Gaps in the International R&D Literature __________________26 RESEARCH METHODOLOGY___________________________________31 3.1 Fundamental Approaches to Research Methodology _________________31 3.2 Research Methods Used in this Dissertation ________________________32 3.2.1 Archival/Theoretical Analysis _________________________________32 3.2.2 Case Study ________________________________________________33 3.2.3 Survey____________________________________________________35 3.2.3.1 Large Scale Survey_______________________________________36 3.2.3.2 In-depth Survey _________________________________________37 3.3 4 Overview of Research Methods Applied___________________________39 LATE INDUSTRIALIZING CONTEXT: SINGAPORE AS A NONTRADITIONAL R&D LOCATION ________________________________41 4.1 Singapore’s Science and Technology Policy________________________41 4.2 Challenges for Singapore as a Non-Traditional R&D Location _________47 4.2.1 Structural Factors __________________________________________48 4.2.1.1 Insufficient Local Human Resources _________________________48 4.2.1.2 Overdependence on MNEs for Innovation _____________________48 4.2.2 Social and Cultural Factors___________________________________49 4.2.2.1 Lack of Entrepreneurship __________________________________49 4.2.2.2 Lack of Creativity________________________________________50 4.3 Concluding Remarks __________________________________________52 i 5 EMPIRICAL EVIDENCE ________________________________________54 5.1 Quantitative Empirical Findings: R&D Internationalization Determinants and International R&D Organizations ____________________________54 5.1.1 Conceptual Framework of a Metanational R&D Organization _______54 5.1.1.1 The Metanational Organization _____________________________54 5.1.1.2 The Metanational R&D Organization ________________________55 5.1.2 Operationalization of Major Variables __________________________63 5.1.2.1 Leveraging of Technological Hierarchy_______________________63 5.1.2.2 Number of Knowledge Bases _______________________________65 5.1.2.3 R&D performance _______________________________________67 5.1.3 Discussion of Results ________________________________________70 5.1.3.1 Different Types of R&D Organizations _______________________70 5.1.3.2 Exploratory Performance Implications________________________73 5.2 Quantitative Empirical Findings: R&D Internationalization Process _____81 5.2.1 Conceptual Framework ______________________________________81 5.2.1.1 Technological Capability __________________________________81 5.2.1.2 Technological Capability Upgrading _________________________82 5.2.2 Levels of Technological Capabilities of R&D Subsidiaries___________88 5.2.3 Typology of Technological Paths of R&D Subsidiaries______________92 5.2.3.1 Derivation of Technological Paths ___________________________92 5.2.3.1.1 Technological Path I _________________________________93 5.2.3.1.2 Technological Path II and Technological Path III ___________94 5.2.3.2 Performance Implications of Technological Paths _______________95 5.2.3.2.1 Technological Path I _________________________________96 5.2.3.2.2 Technological Path II and Technological Path III ___________97 5.2.3.2.3 Temporal Sequence of Technological Stages ______________99 5.2.3.2.4 Discussion of Results_________________________________99 5.2.3.3 Impact of Key Factors on Technological Capability Upgrading ___106 5.2.3.3.1 Role of Internal R&D Network Linkage _________________106 5.2.3.3.2 Role of External R&D Network Linkage_________________108 5.2.3.3.3 Discussion of Results________________________________110 ii 5.2.4 Managerial Implications ____________________________________115 5.2.4.1 Internal and External R&D Management Needs in a Late Industrializing Context ___________________________________115 5.2.4.2 Discussion of Results ____________________________________120 5.3 Qualitative Findings _________________________________________123 5.3.1 A Metanational R&D Organization in the Making: Novartis Institute for Tropical Diseases (NITD) in Novartis’ Research Organization ______123 5.3.1.1 Introduction ___________________________________________123 5.3.1.2 The NITD in Novartis’ Research Organization ________________123 5.3.1.3 Novartis’ R&D Organization as Metanational R&D Organization _125 5.3.1.4 Leveraging of the Technological Hierarchy ___________________125 5.3.1.4.1 Sensing of the Knowledge Base in the Periphery __________125 5.3.1.4.2 Mobilizing the Knowledge Base in the Periphery __________126 5.3.1.4.3 Integrating the Knowledge Base into the R&D Organization _127 5.3.1.5 Knowledge Base in the Periphery __________________________130 5.3.1.6 Conclusion ____________________________________________131 5.3.2 Case Study: R&D Activities at Leica Instruments Singapore ________132 5.3.2.1 Introduction ___________________________________________132 5.3.2.2 Underlying Rationale for Initiation of R&D Activities __________132 5.3.2.3 General Technological Path at LIS__________________________133 5.3.2.4 Methods of Technological Capability Upgrading ______________136 5.3.2.5 Leica Microsystems: Process of Technological Capability Upgrading _____________________________________________________137 5.3.2.6 Leica Geosystems: Process of Technological Capability Upgrading 139 5.3.2.7 Management Capabilities During the Process of Technological Capability Upgrading ____________________________________142 5.3.2.8 Impact of Technological Capability Upgrading on LIS’ Performance _____________________________________________________143 5.3.2.9 Main Challenges Ahead __________________________________145 iii 5.3.3 Internal and External R&D Network Linkage: The Establishment of Lilly Systems Biology in Singapore ________________________________146 5.3.3.1 Introduction ___________________________________________146 5.3.3.2 Motivation for Establishing Lilly Systems Biology in Singapore __147 5.3.3.3 Internal R&D Network Linkage ____________________________147 5.3.3.4 External R&D Network Linkage ___________________________148 5.3.3.5 Interaction between Internal and External R&D Network Linkage _149 5.3.3.6 Management Capabilities and Challenges Ahead ______________150 5.4 6 Summary of Findings ________________________________________150 IMPLICATIONS FOR THEORY, PRACTICE AND POLICY ________153 6.1 Implications for Theory _______________________________________153 6.2 Implications for Practice ______________________________________155 6.3 Implications for Policy _______________________________________158 7 CONCLUSION ________________________________________________161 8 REFERENCES ________________________________________________165 9 APPENDIX ___________________________________________________177 9.1 Questionnaire as a Basis for the In-Depth Interviews ________________177 9.2 Open-ended questions asked during the in-depth interviews __________183 9.3 Definition of Technological Stages provided for Section C of the Questionnaire ______________________________________________184 9.4 Letter Asking for Interview Participation _________________________185 9.5 Interview Partners ___________________________________________187 iv List of Exhibits Exhibit 1: Structure of Dissertation 5 Exhibit 2: Overview of Literature on R&D Internationalization Determinants 8 Exhibit 3: Major Determinants of R&D Internationalization 12 Exhibit 4: Overview of Literature on International R&D Management 17 Exhibit 5: Different International R&D Organizational Models 19 Exhibit 6: Literature on the R&D Internationalization Process from a Corporate Perspective Exhibit 7: 22 Literature on the R&D Internationalization Process on a Subsidiary Level 24 Exhibit 8: Research Gaps in the International R&D Literature 29 Exhibit 9: Research Questions 30 Exhibit 10: Selected Macroeconomic Indicators for Singapore 1965-2000 43 Exhibit 11: Determinants of R&D Internationalization 57 Exhibit 12: Leveraging of Technological Hierarchy in the Metanational R&D organization 59 Exhibit 13: Proposed Model for the Metanational R&D Organization 60 Exhibit 14: Comparison of Traditional R&D Models versus the Metanational R&D 62 Organization Exhibit 15: Classification Scheme for Different International R&D Organizations 66 Exhibit 16: Classification of R&D Organizations 71 Exhibit 17: Regression Results on Number of New Product Developments versus Ethnocentric and Meta/Hub/Integrated R&D Organizations (Electronics Industry only) 76 Exhibit 18: Regression Results on Number of New Product Developments versus Ethnocentric and Meta/Hub/Integrated R&D Organizations (Other Industries) 78 v Exhibit 19: Framework for a Metanational R&D Organization: Implications for the 80 Periphery Exhibit 20: Framework for Technological Capability Upgrading at R&D Subsidiary Level 84 Exhibit 21: Level of Technological Stages 86 Exhibit 22: Industry Breakdown of Different Technological Stages Followed by R&D Subsidiaries of MNEs (1995-2002) 89 Exhibit 23: Comparison of Technological Levels of R&D Subsidiaries of MNEs versus Local R&D Subsidiaries (1995-2002) 91 Exhibit 24: Typology of Technology Paths of R&D Subsidiaries Exhibit 25: R&D Performance Behavior of Technological Paths I, II and III 92 101 Exhibit 26: Performance Behavior of R&D Subsidiaries (Fast and Slow Path Sequence) 105 Exhibit 27: Internal R&D Network Linkage Level 107 Exhibit 28: External R&D Network Linkage Level 109 Exhibit 29: Technological Sophistication and Internal as well as External R&D Network Linkage (All Industries) 111 Exhibit 30: Technological Sophistication and Internal as well as External R&D Network Linkage (Electronics Industry only) 112 Exhibit 31: Technological Sophistication and Internal as well as External R&D Network Linkage (Biomedical Sciences Industry only) 113 Exhibit 32: Classification of R&D Subsidiaries according to their Internal and External R&D Network Linkages 117 Exhibit 33: Classification of R&D Subsidiaries in the Sample 121 Exhibit 34: LIS’ General Technological Path (Technological Path I) 135 Exhibit 35: Process of Technological Capability Upgrading at BU-SM 139 Exhibit 36: Process of Technological Capability Upgrading at Leica Geosystems 141 Exhibit 37: LIS’ Financial Ratios 1992-2003 vi 144 Abbreviations A*Star Agency for Science, Technology and Research AG Aktiengesellschaft BMRC Biomedical Research Council BU-SM Business Unit Stereomicroscopy CEO Chief Executive Officer COE Create, Own and Exploit CPAD Corporate Planning and Administration Division CRO Clinical Research Organization DF Dengue Fever EDB Economic Development Board et al. and others (in Latin: et alii) ETPL Exploit Technologies Private Limited FDI Foreign Direct Investment FMI Friedrich Miescher Institute FY Fiscal Year GDP Gross Domestic Product GNF Genomics Institute of the Novartis Research Foundation GNP Gross National Product vii HQ Headquarters IP Intellectual Property IT Information Technology LIS Leica Instruments Singapore MNE Multinational Enterprise NITD Novartis Institute for Tropical Diseases NUS National University of Singapore PA Patent Applications PD Product Developments R&D Research and Development RISC Research Initiative Scheme SERC Science and Engineering Council SIMTech Singapore Institute for Manufacturing Technology TB Tuberculosis TP Technological Path UK United Kingdom US United States WHO World Health Organization viii Executive Summary This dissertation examines international R&D (research and development) organizations in Singapore. After introducing the topic in chapter 1 and reviewing the literature pertinent to international R&D in chapter 2, the research methodology is discussed in chapter 3. Based on the literature review, research gaps have been identified. In essence, R&D internationalization has so far been confined to the triad nations, so that implications beyond this geographical area have been neglected. The research questions in this dissertation therefore address these implications by analyzing what type of R&D model enables an R&D organization to tap into knowledge existing in non-traditional R&D locations; this dissertation also demonstrates why R&D subsidiaries in non-traditional R&D locations are still at the periphery and how they can increase their level of technological sophistication; it also analyzes performance and managerial implications in the periphery context. The research thus aims to provide a first study on international R&D organizations, present also in late industrializing countries, a major, but neglected research area. The periphery context is analyzed in chapter 4. Singapore is a late industrializing country, which has been able to attract high quality foreign direct investment, involving activities of higher value added and more complex technology. However, it has yet to develop full-fledged R&D activities. So far only few R&D organizations conduct research in Singapore and most R&D subsidiaries’ focus is on development. Lessons learned from the Singapore experience include the need to develop sufficient local R&D expertise as well as to change the general mindset to focus on creativity rather than on precise execution. The empirical study in this dissertation is based on 85 in-depth interviews with 61 R&D subsidiaries of MNEs (multinational enterprises) in Singapore, 10 Singaporebased R&D subsidiaries, two public research institutes (Singapore Institute of Manufacturing Technology and Institute of Bioengineering) and the two main government bodies (Agency for Science, Technology and Research and the Economic Development Board). In the in-depth interviews, responses to a standard questionnaire ix were sought, but open-ended questions were also asked and thus important firmspecific contexts could be discussed. Based on these in-depth interviews, the fifth chapter presents and discusses both the quantitative and qualitative empirical findings of this dissertation. The first part of chapter 5 proposes a framework for a metanational R&D organization. Metanational R&D organizations are characterized as R&D organizations that are capable of optimally leveraging the technological hierarchy internationally and which use a large number of knowledge bases to their advantage. This means that metanational R&D organizations leverage different technological stages within their R&D organization based in many critical knowledge clusters, including non-traditional R&D locations. From the data analysis in the first part of chapter 5, it is found that (1) the metanational organization applies to only a limited extent to R&D organizations, that (2) there is a relationship between the type of international R&D organization and R&D performance at subsidiary level and (3) that the notion of the metanational R&D organization is certainly not mere imagination, even if it is reality for only a few R&D organizations today. It could, however, possibly be the new organizational model of international R&D organizations in the future. In such a transition to a metanational R&D organization, R&D subsidiaries’ managers face the challenges of non-existent or marginal perceptions of the periphery by headquarters and a lack of full-fledged R&D activities in their context. Given this situation, therefore, a change in the perception of the periphery is necessary. Such a change can be achieved by increasing the level of technological sophistication of R&D subsidiaries in the periphery, thus creating a critical knowledge cluster, a necessary condition for R&D organizations to tap into the periphery and thus to become metanational. A second part of chapter 5, therefore, concentrates on how R&D subsidiaries can achieve such an increase in their technological sophistication, which will allow R&D subsidiaries to manage increasingly complex knowledge and thus to remain competitive. Both the level and the evolution of technological capabilities are investigated. While the level of technological capabilities gives a static picture, the evolution of technological capabilities shows the dynamics behind technological x upscaling. Key factors which influence technological sophistication are analyzed, namely internal and external R&D network linkage. Based on this analysis, this dissertation draws some implications for management. Analysis of the data in this second part of chapter 5 shows that (1) the level of technological capability is mostly at the technological stages of development, that (2) interfirm differences in technology paths may result in differential R&D performance, that (3) technological capability upgrading within the same type of technological capabilities occurs at a faster pace than between different types of technological capabilities, that (4) internal R&D network linkage has a greater impact on technological sophistication than external R&D network linkage and that (5) the interaction of internal and external R&D network linkage is critical for an R&D subsidiary in the periphery. The last two findings imply that R&D subsidiary managers need to increase the strategic importance of their R&D site within the internal corporate R&D organization and need to create an efficient local network of external players. If these internal and external R&D management issues are properly addressed, the R&D subsidiary can contribute effectively to the internal corporate R&D organization and be a crucially important partner in the local external research network. Through such an interaction of internal and external R&D network linkage an R&D subsidiary in a non-traditional R&D location may reach the same status as R&D sites in the triad nations. A third part of chapter 5 discusses qualitative findings in the form of case studies. The first case uses Novartis’ R&D organization as an example of a metanational R&D organization in the making. The Novartis Institute for Tropical Diseases (NITD) is currently being built up as an R&D site in Singapore for conducting research into the tropical diseases of tuberculosis and dengue fever. Consequently, Novartis’ R&D organization has tapped into knowledge residing in the periphery and has mobilized and integrated this knowledge in the overall R&D organization, creating a metanational advantage. The second case study analyzes the technological capability upgrading of Leica Instruments Singapore (LIS). LIS started its R&D activities as a manufacturing support unit and increased its level of technological sophistication to that of an exploratory development unit, engaging in several external research xi collaborations. Management capabilities and challenges during this process are discussed. The third case study investigates the internal and external R&D network linkage of Lilly Systems Biology. This R&D site has been created recently in Singapore. The building of an internal and external R&D network linkage is examined and how its interaction is managed is shown. A final part in chapter 5 summarizes the empirical findings of this dissertation. Chapter 6 points out implications for theory, practice and policy. In theoretical terms, this dissertation hopes to provide a small step towards a more advanced understanding of international R&D management by analyzing the implications of R&D internationalization beyond the triad nations. It thus develops a framework for a metanational R&D organization and investigates the process of technological upgrading of R&D subsidiaries in the periphery. Certainly, more theoretical development is warranted in future studies. Practical implications refer mostly to the perception gap between headquarters and R&D subsidiaries. It is frequently found that the R&D subsidiary’s level of technological sophistication is overestimated by the R&D subsidiary itself and is underestimated by headquarters. Intense communication with headquarters and the creation of an awareness of the local context are important managerial implications. With respect to policy implications, the role of the Singapore government is decisive in fostering more and higher level R&D activities. Singapore’s science and technology policy may indeed serve as a role model for other late industrializing countries. Chapter 7 presents conclusions, indicates limitations of the dissertation and suggests areas for future research. There is a need for studies examining, for instance, more late industrializing countries and their differences with regard to R&D. These and other issues provide ample scope for future studies in the field of international R&D management. xii 1 INTRODUCTION Different corporate functions show a different development regarding their internationalization. Multinational enterprises (MNEs) first largely internationalized their sales and production functions. The R&D (research and development) function, however, was typically concentrated in the home country. Concerned that the potential benefits of overseas research would be outweighed by the costs incurred from duplicated efforts, scale diseconomies and knowledge leaks, few organizations were willing to internationalize their R&D activities. During the past fifteen years, however, a large number of MNEs have located considerable portions of their R&D activities abroad (Boutellier, Gassmann and von Zedtwitz, 2000: 3-8; Florida, 1997: 85-86; Kuemmerle, 1997: 61-62). The emergence of more knowledge centers worldwide, for instance, has provided a strong incentive to internationalize corporate R&D activities in order to tap into and leverage on new knowledge. A study by Kuemmerle (1999) of 32 multinational firms in 5 countries shows that overseas R&D efforts by these firms increased from 6,2% in 1965 to 25,8% in 19951. At the beginning of corporate R&D internationalization, knowledge centers were limited to the triad nations (US, Europe and Japan). More recently, this development has extended beyond the triad nations and also includes the ‘periphery’, non-traditional R&D locations, which are emerging as critical knowledge bases (Boutellier, Gassmann and von Zedtwitz, 2000: 38). Given the increasing propensity to internationalize the corporate R&D function, there is a correspondingly strong interest in firms’ international R&D activities from an academic perspective (Penner-Hahn, 1998: 149). Research in the field of international 1 European firms are at the forefront of this internationalization process, but US firms and Japanese firms seem to have less internationalized their R&D activities compared to their European counterparts. European MNEs operating in small countries have particular high shares of foreign corporate R&D. Firms based in Belgium and the Netherlands, for instance, perform more of their R&D activities outside the home country than inside it (Granstrand, Hakanson and Sjölander, 1993: 414). Swiss firms spend more than 50% of their R&D expenditure abroad (Boutellier, Kloth and Bodmer, 1996: 282). Foreign shares of R&D for firms in larger European countries such as Sweden and the UK vary between 23 and 42% (Granstrand, Hakanson and Sjölander, 1993: 414; Granstrand, 1999: 279). These figures contrast with foreign R&D activities of 10%-12% by US firms (v. Zedtwitz, 1999: 31). Japanese firms are late, but fast internationalizers regarding their R&D internationalization (Grandstrand, 1999: 278). 1 R&D management has examined various issues such as the determinants, the management and the process of this R&D internationalization. First, the determinants of R&D internationalization have been the subject of various studies (De Meyer, 1993; Florida, 1997; Kuemmerle, 1999; Le Bas and Sierra, 2002; Patel and Vega, 1999; Ronstadt, 1977, 1978; Westney, 1993). A second research stream focuses on the management of the internationally dispersed R&D facilities and their respective roles because the ongoing process of the internationalization of R&D raises the issue of its effective management (Boutellier, Kloth and Bodmer, 1996; Gassmann and von Zedtwitz, 1998; Gerpott, 1990; Gerybadze and Reger, 1999; Medcof, 1997; Medcof, 2001; Nobel and Birkinshaw, 1998; von Zedtwitz, 1999; von Zedtwitz and Gassmann, 2002). Third, researchers have investigated the internationalization process of corporate R&D (Belderbos, 2003; Cantwell, 1989; Cantwell, 1992; Grandstrand, 1999; Kuemmerle, 1999; Pearce, 1989; Pearce and Pooni, 1996; Pearce and Singh, 1992) As can be seen, most studies of overseas R&D activities address three research questions: 1) What are the determinants to conduct R&D outside the home country? 2) How should MNEs manage a globally dispersed portfolio of R&D sites? and 3) What is the nature of the R&D internationalization process? Unfortunately, these research questions have only been explored for R&D sites in the triad nations and neglect further implications beyond that geographical area, ignoring non-traditional R&D locations, namely late industrializing countries, which are increasingly gaining in importance as critical knowledge bases. This observation is confirmed by Mahmood and Singh (2003: 1053) who state that so far research has focused on developed countries, leaving room for further research on innovation in late industrializing countries. Little is known about technological capability development in late industrializing countries and its impact on the R&D organization (Figueiredo, 2002: 73). Furthermore, the current literature focuses mostly on the phenomenon of R&D internationalization from a corporate perspective. The general subsidiary literature (for 2 instance Birkinshaw and Hood, 1998 and Andersson, Forsgren and Holm, 2002) addresses subsidiary specific issues, but not related to the R&D function. Almost no attention has been paid to R&D subsidiaries as such. In an increasingly competitive environment, however, the capability of deploying and leveraging technological competencies at R&D subsidiary level is a major source of gaining and sustaining competitive advantage for MNEs (Birkinshaw and Hood, 1998: 773). In essence, there has been no examination of how R&D internationalization affects R&D subsidiaries in the periphery. Previous studies examined R&D internationalization from a headquarters perspective and focused on the triad nations. The present dissertation represents a modest attempt to illuminate this blind spot in the literature. The research gap identified leaves several important questions unanswered: What type of international R&D model enables R&D organizations to tap into knowledge in non-traditional R&D locations? Why are non-traditional R&D locations still at the periphery and how can R&D subsidiaries in non-traditional R&D locations increase the importance of the periphery? What are management and performance implications in non-traditional R&D locations? Hence, this dissertation attempts to address these important, unanswered questions of R&D internationalization for the periphery. More specifically, it examines and proposes a metanational R&D organization, investigates why R&D subsidiaries in non-traditional R&D locations are still at the periphery and how R&D subsidiaries in the periphery can increase their level of technological sophistication, discusses internal and external R&D management needs and draws performance implications. It is one of only a few empirical research projects which examine R&D subsidiaries in the periphery. It attempts to advance our understanding of international R&D organizations and of technological capability development in non-traditional R&D locations. The study’s principal contribution to the field include the creation of a new primary data set, the formulation of simple theories and associated propositions that address the neglected research questions. This dissertation also develops measures that operationalize the leveraging of technological hierarchy, technological paths, internal and external R&D network linkage and innovative performance. 3 This dissertation is organized as follows: chapter 2 reviews the existing literature on international R&D management. Chapter 3 discusses the research methodology. Chapter 4 analyzes the context of the periphery, namely Singapore, where the R&D investment under investigation takes place. This chapter is devoted to Singapore’s science and technology policy and its recent efforts to create a biomedical sciences hub. It considers in particular Singapore’s transition towards more research activities. Chapter 5 presents and discusses both the quantitative and qualitative findings of this dissertation. Chapter 6 draws implications for theory, practice and policy. Chapter 7 identifies limitations of the study, suggests directions for future research and presents conclusions. The structure of the dissertation is shown in exhibit 1: 4 Exhibit 1: Structure of Dissertation Introduction Literature Review Research Methodology Late Industrializing Context: Singapore as a Non-Traditional R&D Location Empirical Evidence Quantitative Findings Qualitative Findings Implications for Theory, Practice and Policy Conclusion Source: Author 5 2 LITERATURE REVIEW 2.1 Definition of International Corporate R&D Activities Before reviewing the literature on the internationalization of corporate R&D, the central term must be defined. R&D refers to the systematic approach of gaining new scientific knowledge through the combination of production factors (Gauglitz-Lüter, 1998: 6). Research, the ‘R’ of R&D, denotes the process of discovering this knowledge, thus providing a platform for product and process development for specific markets, which is the major task of the development unit (v. Zedtwitz, 1999: 16-17). In general, R&D activities show several singular characteristics: First, they exhibit a high degree of innovation. This degree varies according to the development or research function. Second, the tasks within R&D are of a unique and non-repetitive nature or in other words R&D activities are often unstructured and intangible. Third, the R&D process reveals a high degree of complexity. This being the case, R&D activities are also characterized by a high degree of uncertainty and risk. Fourth, strategic control of R&D activities is considered crucial since R&D knowledge is an important invisible asset of the firm (De Meyer, 1993: 110; for technology as a source of competitive advantage also see Barney, 1991; Mahoney and Pandian, 1992; Wernerfelt, 1984). Internationalization of corporate R&D refers to the fact that a significant portion of a firm’s R&D activities are conducted in an international setting. Global R&D management is defined as the management of corporate R&D efforts in different countries (De Meyer and Mizushima, 1989: 135). The R&D definition used in this dissertation relies on the R&D classification developed by Amsden and Tschang (2003) and Medcof (1997), who distinguish between different technological stages within R&D. This classification has been adopted because it is a comprehensive definition of R&D and is suitable for the late industrializing context (for more details see chapter 5). 6 2.2 Development of the Literature on International R&D The internationalization of corporate R&D was a marginal subject for research up to the 1980s (e. g. Behrman and Fischer, 1980; Ronstadt, 1977, 1978). Academic interest in this topic started in the early 1990s (e. g. De Meyer and Mizushima, 1989; De Meyer, 1993). Up to that point in time, the internationalization of R&D attracted only peripheral and passing attention in the literature (Granstrand, Hakanson and Sjölander, 1993: 414; Granstrand, 1999: 276; Gerybadze and Reger, 1999: 251). One of the reasons for the relatively late development in theory and practice of R&D internationalization was a widely held assumption that corporate R&D activities needed to be centralized. It was argued that the protection of firm-specific technology was only possible by centralizing corporate R&D activities. Furthermore, it was believed that only through such centralization could economies of scale, and as such the necessary efficiency in corporate R&D, be reached. Another important reason for centralized corporate R&D is historical: R&D facilities were traditionally established close to headquarters (Gerpott, 1990: 230). However, with the internationalization of other corporate activities such as sales and production, the internationalization of corporate R&D activities followed suit, overthrowing the long-held belief in the centralization of corporate R&D activities (Gerybadze and Reger, 1999: 251). A literature review in the field of international R&D management revealed that previous studies in international R&D management include R&D internationalization determinants, international R&D organization and management and the R&D internationalization process. These different literature streams are discussed in more detail in the following sections. 2.2.1 Determinants of R&D Internationalization A substantial body of literature examined the determinants for R&D internationalization (for instance De Meyer, 1993; Florida, 1997; Kuemmerle, 1999; Le Bas and Sierra, 2002; Patel and Vega, 1999; Ronstadt, 1977, 1978). The different 7 determinants indicate why R&D organizations conduct R&D abroad. The following exhibit gives on overview of this literature stream: Exhibit 2: Overview of Literature on R&D Internationalization Determinants Authors Purpose of Study Main Findings of R&D Internationalization Determinants Creamer, 1976 Differences between homecountry R&D and overseas R&D Market-driven and labor cost determinants Ronstadt, 1977; 1978 In-depth case studies of overseas R&D investment of seven US MNEs Four types of R&D units: Transfer technology units, indigenous technology units, global technology units and corporate technology units ⇒ Market-driven determinants and technology driven determinants Hewitt, 1980 Foreign R&D activities by US Market-driven determinants MNEs based on 1966 data (technical support) and technology-driven determinants Pearce, 1989 R&D activities by US MNEs Market-driven and technologydriven determinants Pearce and Singh, 1992 Survey of the internationalization process of corporate R&D by the world’s leading enterprises Various determinants (characteristics of parent and subsidiary laboratories important) Westney, 1993 Cross-Pacific Technology driven determinants internationalization of R&D by for Japanese firms versus market US and Japanese firms driven determinants for American firms Serapio and Dalton, 1993 Survey of 255 US research facilities of foreign firms 8 Japanese R&D facilities more oriented towards commercial markets than European R&D facilities Exhibit 2 continued: Authors Purpose of Study Main Findings of R&D Internationalization Determinants De Meyer, 1993 16 clinical case studies of Technical learning European, North American and Japanese firms Kuemmerle, 1996; 1999 Investigation into the international allocation of R&D activities by MNEs Home-base augmenting (technology driven) versus home-base exploiting (marketdriven) determinants Florida, 1997 Foreign direct investment into R&D in the US by foreignaffiliated R&D laboratories Technology-driven determinants Serapio, 1999 Foreign direct investment in R&D in the US Different determinants depending on R&D subsidiary’s nationality Patel and Vega 1999 US patenting activities of 220 of the most internationalised firms in terms of their technology in the 1990s. Market driven determinants still important Le Bas and Sierra, 2002 345 MNEs with greatest patent Confirmation of findings by Patel activities (replication of Patel and Vega (1999) and four types and Vega’s study for Europe) of strategy according to a firm’s technological advantage Gassmann and von Zedtwitz, 2002 290 research interviews and Technology and market-driven database research in 81 MNEs determinants; research is concentrated in only five regions worldwide; development is more globally dispersed. Source: Author The various studies on R&D internationalization determinants identified two main categories of determinants: technical and market driven determinants. 9 The first determinant, namely, the need to provide technical support to the operations of foreign affiliates, is important, as overseas subsidiaries evolve and become differentiated from their parent companies. This type of determinant implies that different consumer preferences and/or government regulations need to be considered. Foreign R&D subsidiaries are responsible for adapting extant product ranges or, in some cases, for creating new products to satisfy local needs. Creamer (1976) examines the difference between home country and host country R&D and found that market-driven as well as cost-related criteria are important determinants for R&D internationalization. In more recent studies, Patel and Vega (1999) and Le Bas and Sierra (2002) examine patterns of technological activities outside their home countries. The study of Patel and Vega (1999) is based on data of US firms, while Le Bas and Sierra (2002) based their analysis of R&D internationalization determinants on data of European firms. Both studies indicate that adapting products and processes to suit foreign markets and hence to provide technical support are still major factors for R&D internationalization, which means that R&D subsidiaries exist to extend abroad the firm-specific advantage of the parent firm (Meyer-Krahmer and Reger, 1998: 2). This implies that R&D knowledge created in the home base is transferred and applied to the host country’s market or in other words the home base R&D knowledge is exploited by applying it to the host market. In contrast to these market driven determinants, a second stream of literature emphasizes technology driven determinants for R&D internationalization, which imply that R&D knowledge created in the home base is augmented by R&D knowledge created in overseas R&D subsidiaries. Florida (1997) analyzes foreign direct investment (FDI) into R&D in the US by foreign-affiliated R&D laboratories, and as such European MNEs’ R&D internationalization into the US. Kuemmerle (1999) focuses on two industrial sectors, namely electronics and pharmaceutical, and examines laboratory sites of 32 multinational firms domiciled in five countries. Both authors agree that the major determinant for R&D internationalization is more technology driven than market driven; that is MNEs seek to harness external new scientific and technological 10 capabilities abroad. Other authors also suggest that firms in many industries decreasingly undertake overseas R&D activities as a mere support function for existing sales or production operations (Belderbos and Iwasa, 1999: 3). The competitiveness of firms is becoming increasingly dependent on their ability to establish a presence at an increasing number of locations in order to access new scientific capabilities (Edler, Meyer-Krahmer and Reger 2002: 149-150, Kuemmerle, 1996: 69-71 and Shan and Song, 1997: 268-269). This development is particularly important in the context of increasing R&D costs, fiercer foreign and domestic competition, intensifying product differentiation, and market globalization (Mutinelli and Piscitello, 1998: 491). Besides these two literature streams, which identified either market or technology determinants, several studies have identified both market and technology driven determinants for R&D internationalization, for instance Ronstadt (1977, 1978), Pearce (1989), Gassmann and von Zedtwitz (2002). Ronstadt (1977; 1978) argues that an overseas R&D unit can evolve from being a technology transfer unit to being an indigenous technology unit. Such a transition implies that the original market-driven determinant (to transfer technology and adapt it to local conditions) can shift to a technology-driven determinant (indigenous technology is developed) (Ronstadt, 1978: 15-16). The following exhibit summarizes the main determinants of R&D internationalization. 11 Exhibit 3: Major Determinants of R&D internationalization Technological Advantage of Home Country Strong Weak Home-Base Exploiting Home-Base Augmenting Market Seeking Technology Seeking Weak Strong Technological Advantage of Host Country Source: Kuemmerle (1996); LeBas and Sierra (2002) Analysis of the literature on the determinants of R&D internationalization shows most notably the distinction between home and host country base. In effect, the literature is characterized by a dichotomous nature. Depending on the home country’s strengths or weaknesses either the home base is exploited or augmented. A home country’s strengths or weaknesses refer to the technological advantage or disadvantage of the home base. It is argued here that the R&D internationalization determinant home base exploiting is logically also a market seeking determinant, since the home base is exploited by adapting products and processes to a new market and hence the market is sought after. This market-seeking determinant applies even if the home country is not strong in this particular technical field. Existing products of the parent firm are adapted to a local 12 market even though these products may not be of a high level of technological sophistication. The same holds true for the technology-seeking determinant, which at the same time is also a home base augmenting determinant. The home base is augmented by seeking new technologies in the host country, meaning investment in cross-border R&D takes place in order to acquire knowledge from overseas locations with a specific scientific or technological competence. The review of the literature shows that the determinants of R&D internationalization have been the subject of a variety of studies. Without more formal development in theory, however, little progress will be made towards explaining and predicting how different processes influence overseas R&D subsidiaries. The literature suggests that the determinants of R&D internationalization are characterized by a dichotomous nature (home base exploiting versus home base augmenting). What is missing in the literature, however, is a more holistic view of the determinants for R&D internationalization. Consequently, this dissertation attempts to overcome the dichotomy of home versus host country base in the literature on the determinants of R&D internationalization and develops a more holistic view of the determinants of R&D internationalization (see chapter 5). 2.2.2 International R&D Organization and Management As has been stated in the introduction, the internationalization process of R&D activities has intensified over the years. As a result of this development, it is important to manage the benefits and costs of this increasing R&D internationalization optimally. The literature on international R&D organization and management elucidates the interplay of these benefits and costs: There are numerous benefits resulting from the internationalization of corporate R&D activities and an effective R&D management would attempt to take advantage of these benefits, which include resource access, product market access and infrastructure access (Brockhoff, 1998: 28-29; Boutellier, Kloth and Bodmer, 1996: 282-283). Resource access usually refers to know-how as well as personnel access, the latter 13 being for instance the availability of qualified R&D personnel. The first refers to local scientific communities as centers of excellence. A MNE can thereby profit from the latest technology trends and from innovation impulses created by these foreign R&D subsidiaries. This is confirmed in a study by Gerybadze and Reger (1999), who emphasize the importance of the knowledge generating capacities of foreign R&D locations and the dynamic interactions between different R&D locations. The importance of foreign R&D locations as knowledge generating units was also pointed out in a study by Birkinshaw and Hood (2001) who found that crucial innovations often emerge from foreign R&D units due to their local proximity and their low attachment to headquarters’ procedures. Product market access is a pre-condition for local adaptation of manufacturing conditions, product development and processes (Belderbos, 2001: 314). Hence, local R&D allows proximity to local customers and hence to suit local markets (v. Zedtwitz, 1999: 43; Gerybadze and Reger, 1999: 261). Such a product market access is important if a company plans to enter new markets. Infrastructure access implies, for instance, a favorable political infrastructure, highly supportive of the respective R&D unit or a positive cultural and educational structure, fostering technology awareness and acceptance (Gassmann and v. Zedtwitz, 1998: 150-152). The MNE can benefit from such a favorable infrastructure, for example in the form of different national governments’ support in establishing foreign R&D units through subsidies, as is the case in Singapore. A further benefit of internationalizing corporate R&D activities is the possibility of making use of different time zones. R&D units in Asia, for instance, may start working on R&D projects, transferring their data to the R&D unit in Europe and then to the R&D unit in the US. This assures a 24-hour research effort. Furthermore, firms can gain comparative advantage by having R&D facilities in several countries since customer demands, knowledge bases, and regulatory requirements differ from country to country. This means that it might be comparatively more advantageous to conduct corporate R&D activities in one country 14 as compared to another, even though one particular country might have an absolute advantage in all R&D fields. Besides the benefits accruing from internationalizing corporate R&D activities, there are also costs involved in the internationalization of corporate R&D activities. Effective international R&D management would attempt to neutralize these costs. High coordination costs, for instance, are the consequence of intense communication between the international R&D units and between the international R&D units and headquarters in order to provide a well-functioning global R&D organization. Furthermore, information and transfer costs arise due to knowledge exchanges between the different locations (Granstrand, Hakanson and Sjölander, 1993: 415). At the same time, global R&D activities can result in headquarters partially losing control of research results. This is because R&D activities are conducted in a more international and hence more decentralized setting. Since R&D knowledge is a firm’s invisible asset, then the partial loss of strategic control of such an asset can result in major costs (De Meyer, 1993: 110). An international R&D setting with various R&D locations might also induce parallel R&D efforts and thus create redundancies (Reger, 1999: 72). This problem might especially occur if coordination mechanisms between the various R&D locations are weak. Another cost arising from the internationalization of corporate R&D comes from the fact that an international R&D organization might not be capable of fully realizing economies of scale and synergy effects in contrast to a centralized R&D organization (Brockhoff, 1998: 31). The underlying reasoning behind this argument is that a critical number of researchers is needed in order to be able to realize economies of scale and scope (De Meyer, 1993: 110). 15 Based on studying the diverse benefits and costs of R&D internationalization, the literature on managing overseas R&D units is concerned with structural arrangements for the coordination and control of worldwide R&D activities. These studies are devoted to managing international R&D activities so as to carefully balance benefits and costs resulting from R&D internationalization. In the ideal case, benefits are maximized and costs minimized. The following exhibit gives an overview of the relevant literature. 16 Exhibit 4: Overview of Literature on International R&D Management Authors Purpose of Study Main Findings of International R&D Management Behrman and Fischer, 1980 Structured interviews with 56 R&D subsidiaries of US, European and Japanese MNEs Managerial styles range from absolute centralization to total freedom for R&D units; majority of R&D units either managed by participative centralization or supervised freedom De Meyer and Mizushima, 1989 In-depth case studies of global R&D activities by 7 European and 15 Japanese firms Need for a new model for the management of internationally dispersed R&D laboratories pointed out Gerybadze and Reger, 1999 In-depth analysis of R&D internationalization in 21 large corporations in Europe, Japan and the US Streamlining of R&D activities, resulting dominant form of R&D organization: multiple R&D centers with one dominant center of coordination Gassmann and von Analysis of R&D activities Zedtwitz, 1999 by 33 MNEs Identification of five different types of R&D organizations Boutellier, Gassmann and v. Zedtwitz, 2000 International R&D Major International R&D Management (18 case studies Management Practices in the pharmaceutical, electronics/software and electrical/machinery industry) Medcof, 2001 Analysis of internal R&D networks and their strategic importance Overseas technology units heterogeneous with regard to their resources, therefore different management styles are required Source: Author In an early study, Behrman and Fischer (1980) conducted structured interviews with R&D managers at 56 U.S., European, and Japanese MNEs and found that in a continuum from absolute centralization to total freedom the managerial styles of 17 supervised freedom or participative centralization are the most frequent managerial modes for R&D subsidiaries. De Meyer and Mizushima (1989) examined global R&D management and suggest that new types of organizational structures are necessary to address increasing R&D internationalization. They make several assertions as a basis for such a new framework of international R&D organization. In essence, these assertions state that R&D internationalization has become a key component of the MNE, driven by both unrelated and related technology determinants. Due to this development global R&D management requires new organizational types. The authors suggest a network organization of peer laboratories, but also point out that the precise mechanisms of managing such a network are not clearly specified (De Meyer and Mizushima, 1989: 144-145). A later study, which assesses the management of international corporate R&D in 21 MNEs, has been published by Gerybadze and Reger (1999). Their analysis focuses on recent changes in the global management of corporate R&D activities. Their findings suggest that due to the strong internationalization of corporate R&D between 1985 and 1995, R&D organizations of international firms proved to be extremely complex and hence difficult to manage. As a consequence, these transnational firms streamlined their R&D activities. The dominant form of organization resulting from this consolidation is multiple centers with one dominant coordinating center. Five different types of R&D organizations are identified by Gassmann and von Zedtwitz (1999). These types are illustrated below based on Gassmann (1997) and Gassmann and v. Zedtwitz (1999). This classification is important since it provides an overview of different international R&D organizations. This classification is extended in this dissertation (see chapter 5). The exhibit below visualizes these different international R&D organizations: 18 Exhibit 5: Different International R&D Organizational Models Decentralized R&D Polycentric, decentralized R&D Distribution of Internal Competences and Knowledge Bases Centralized R&D Integrated R&D Network R&D Hubmodel Ethnocentric, centralized R&D Competition Geocentric, centralized R&D Cooperation Degree of Cooperation between Locations Source: Gassmann, 1997a: 49 and 1997 b: 49; Gassmann and von Zedtwitz, 1999: 245 In the ethnocentric, the most traditional model, all corporate R&D activities are centrally located at headquarters. As a result, no transnational R&D processes take place. Such a centralized approach is based on the assumption that corporate R&D must be centralized and implies that only such R&D activities allow economies of scale and synergies to be realized and unintended technology transfer to be avoided. Given the primacy of the home base in this model, this organizational form is insensitive to local market demands. Moreover, an ethnocentric R&D organization cannot benefit from external technologies since knowledge is created only in the home base. 19 As opposed to the first concept, the second model is international in nature and adopts a geocentric, centralized approach to R&D. In this model international R&D knowledge is acquired through intense collaboration with local manufacturing, suppliers and lead customers. The physical R&D location, however, is located in the home base. As a result of this, this R&D structure is sensitive to local markets and may benefit from foreign technological trends. The international awareness of the R&D personnel is fostered as well, since researchers are exposed to international markets. This model might be inadequate if an R&D organization becomes increasingly international for its centralized character might then hinder achieving a critical presence in diverse R&D locations. The third model is the polycentric approach, where various R&D locations in different foreign markets exist without centralized control. These R&D units have mostly been established as a result of the presence of local distribution facilities and manufacturing plants. Consequently, the decrease in the primacy of the home base assures a high degree of local sensitivity and the utilization of local resources. Due to the strongly decentralized character of this R&D organization and its lack of coordination mechanisms, parallel R&D efforts might result in duplication and thus in major costs for a MNE. Opportunities for innovation through combination are foregone in this model. The fourth model, the Hub-Model, is basically an ‘in-between’ model, which has on the one hand a central R&D location similar to the ethnocentric model, but on the other hand has dispersed R&D locations. The central R&D location leads in most technological fields. On the dimension of the degree of cooperation, it is situated between cooperation and competition. Due to coordination of R&D efforts, a high degree of efficiency is achieved and a suboptimal resource allocation avoided. Potential high coordination costs might, however, be a drawback of this model. The last model, and similar to the hub model, is that of the integrated network. Within this concept, the central R&D location loses its overall dominance. All R&D locations are equally important, play a strategic role and interact through multiple and complex coordination mechanisms. Each R&D location focuses on one specific product, 20 component or technology area. Consequently, the home versus host country dichotomy loses its significance. The obvious benefits of this model are the realization of specialization, learning and synergy effects, and the utilization of local competencies. On the other hand, high coordination costs occur because of its complexity. Furthermore, the decision making process is more difficult due to the equal status of the different international R&D locations. While this division into five different international R&D models provides an overview of international R&D models, it is not an all-encompassing classification. Not all R&D activities of international companies can be clearly attributed to one of these models. Furthermore, the differentiation between the different models is not always completely clear-cut; the hub-model in particular as an ‘in-between’ model does not seem to be fully defined. However, this classification of international R&D models is a contribution to visualizing different international R&D concepts and serves in enhancing our understanding of international R&D activities. This dissertation attempts to expand this framework in chapter 5. While the review of this literature stream revealed how different R&D organizations can be managed, it does not provide any insights on the evolution of overseas R&D subsidiaries. The third literature stream, R&D internationalization process, is reviewed in the next section. 2.2.3 R&D Internationalization Process The increasing internationalization of corporate R&D activities raises two important research questions. First, previous literature has examined this development from a firm level perspective. Or in other words, corporate internationalization patterns have been studied. Second, later studies have examined the role and charter change on a subsidiary level, resulting from the intensified R&D internationalization. The following exhibit shows the literature on the R&D internationalization process from a corporate perspective: 21 Exhibit 6: Literature on the R&D Internationalization Process from a Corporate Perspective Authors Purpose of Study Main Findings of International R&D Management Cantwell, 1992 Analysis of corporate R&D internationalization patterns Implications for competitiveness on a firm as well as a country level Kuemmerle, 1999 Analysis of corporate R&D internationalization patterns and entry modes Greenfield investments prevailing entry mode; incremental R&D internationalization process Grandstrand, 1999 Comparison of corporate R&D internationalization patterns in Swedish and Japanese firms Swedish firms early R&D internationalizers in contrast to Japanese firms; psychic distance more relevant for research than for development Source: Author Cantwell (1992), for instance, analyzes corporate R&D internationalization patterns of corporate R&D activities of MNEs and their implications for competitiveness on a firm as well as a country level. Kuemmerle (1999) investigates this internationalization process by analyzing the entry modes for foreign R&D. His findings suggest that greenfield investments are the prevailing mode of entry, firms establishing R&D sites first at home before establishing such sites abroad. This suggests that R&D internationalization takes place in incremental steps, an observation in line with the thinking of Johanson and Vahlne (1977: 25), who found that firms first establish foreign sites in nearby markets and then further expand into more distant markets and incrementally increase their foreign activities, an incremental internationalization process. A further study by Granstrand (1999) examines the internationalization process of corporate R&D in two different countries, namely Sweden and Japan. The starting point of corporate R&D internationalization and its speed seem to be important criteria 22 which determine how this internationalization process unfolds over time. While Swedish firms were early internationalizers in their R&D activities, Japanese firms started late in this process, but later were rapid in their R&D internationalizing process. Granstrand (1999) also applies the concept of psychic distance, first introduced by Johanson and Vahlne (1977), to the internationalization process of corporate R&D. Psychic distance denotes the distance the parent firm has from certain foreign markets, for instance in terms of differences in economic development, language and culture. Psychic distance seems to have less explanatory power for the internationalization process of corporate R&D, having more influence on the internationalization process of research rather than on the internationalization process of development and limited influence on the internationalization process of R&D in Japanese firms (Granstrand, 1999: 293). While these studies examine the R&D internationalization process on the level of a firm, there is not a corresponding research stream, which analyzes the process from a subsidiary perspective. Characteristically, these studies have examined the nature of the subsidiary as a result of firm internationalization. The exhibit below gives on overview of this literature: 23 Exhibit 7: Literature of the R&D Internationalization Process on a Subsidiary Level Authors Purpose of Study Main Findings of International R&D Management Lall, 1992 Implications of industrial strategy on corporate and national technological capabilities Careful and selective government intervention is critical for industrial success in late industrializing countries Birkinshaw, 1998 Analysis of Subsidiary Initiatives Global, local, internal, and global-internal hybrid initiatives; entrepreneurship at subsidiary level is critical Birkinshaw and Hood, 1998 Subsidiary Evolution and Initiatives Subsidiary evolution is a function of capability and charter change, five generic subsidiary evolution processes are identified influenced by parent, subsidiary and host country factors Frost, 2001 Investigation into the geographical sources of knowledge sources utilized by foreign subsidiaries Illustration of the conditions under which foreign subsidiaries tap into knowledge in the home country or host country Birkinshaw, Hood Study of 24 MNEs with and Young, 2002 regard to their competitive dynamics Three types of subsidiaries, internally focused, externally focused and dual focused arena, with different performance implications Costa and De Queiroz, 2002 Analysis of technological learning in MNEs versus local firms Higher level of technological sophistication reached by subsidiaries of MNEs compared to local firms Figueiredo, 2001 and 2002 Two case studies on technological capability accumulation paths in two Brazilian steel firms Operational performance improvements can be accelerated if technological capabilities are deliberately enhanced, possibly resulting in financial benefits Source: Author 24 The general literature on subsidiaries discusses foreign subsidiary roles and their dynamics. One study within this literature focuses on the geographical sources of foreign subsidiaries’ innovations by analyzing the conditions under which innovating subsidiaries are likely to draw upon sources of knowledge located in the home base of the firm and/or the subsidiaries’ host country environment (Frost, 2001: 101-123). While Frost (2001) emphasizes the dichotomy between home and host country and its impact on the origin of knowledge sources for the subsidiaries, Birkinshaw, Hood and Young (2002) distinguish between internal and external competitive forces on MNEs’ subsidiaries and their impact on subsidiary performance. More specifically, their study distinguishes between an internally focused competitive arena, an externally focused competitive arena, and a dual-focused arena and analyzes the implications for subsidiaries’ performance. An earlier study by Birkinshaw and Hood (1998) examines multinational subsidiary evolution with regard to capability and charter change in foreign-owned subsidiary firms. Based on the identification of five subsidiary evolution processes, their findings suggest that charters are mobile for subsidiaries and that subsidiary capabilities are critical for MNEs (Birkinshaw and Hood, 1998: 783792). Yet another study by Birkinshaw (1998) characterizes subsidiary initiatives in multinational corporations by looking at 39 such initiatives, classifying these initiatives in global, local, internal, and global-internal hybrid initiatives and concluding that entrepreneurship at the subsidiary level has the potential to enhance local responsiveness, worldwide learning and global integration. The subsidiary literature has also examined technological capabilities in foreign affiliates. Costa and de Queiroz (2002), for instance, analyze the deepening of technological capabilities of foreign affiliates. Their results indicate that foreign affiliates are engaged in more complex technologies than their local counterparts (Costa and de Queiroz, 2002: 1432-1433). Figueiredo (2002) found that technological capability-accumulation paths account for interfirm differences in operational performance improvement in his study of two Brazilian steel firms. His findings show a strong association between rates of operational performance improvement and the rate of accumulation and the consistency over time of the technological capability accumulation paths. Furthermore, Figueiredo (2002) points out that it has not been 25 discussed how firms’ efforts on learning and technological capability accumulation can be a competitive advantage, especially in the late industrializing context. This dissertation attempts to fill in this void in the R&D subsidiary literature. Previous research has investigated general subsidiary evolution in terms of their roles and technological capability upgrading. However, what has been neglected is what consequences R&D internationalization entails for overseas R&D subsidiaries in a late industrializing context. This dissertation makes an attempt to analyze this issue by examining the technological capability upgrading of R&D subsidiaries in a late industrializing country. This literature review has discussed previous research on R&D internationalization determinants, international R&D management and R&D internationalization process. Research gaps are derived in the final section of this chapter. 2.3 Research Gaps in the International R&D Literature While the literature on the determinants of R&D internationalization has focused on the dichotomous nature of these determinants (home-base augmenting versus homebased exploiting), a more comprehensive understanding of these determinants is lacking. Consequently, this dissertation attempts to overcome this dichotomy by proposing a holistic determinant. Closely linked to the determinants of R&D internationalization are different types of international R&D organizations. While the ethnocentric R&D organization centralizes all international R&D activities in one location, an international R&D network manages various strategic R&D locations of equal status. The basic tradeoff in international R&D management is to find an optimal balance between the benefits and costs of R&D internationalization. This also implies that an optimal balance between internal (inside the corporate R&D organization) and external (in the local research environment) needs to be achieved. Based on a more comprehensive understanding of the determinants of R&D internationalizations, this dissertation 26 attempts to provide a new framework for an international R&D organization and aims thus to enhance the literature stream on international R&D management. The literature review of the R&D internationalization process showed that previous studies have examined the process of R&D internationalization mainly from a corporate perspective and to a very limited extent from a subsidiary perspective. While technical learning is an important determinant for foreign R&D activities (De Meyer, 1993: 111-112), there is no clear understanding of technological capability accumulation in an overseas R&D subsidiary (Figueiredo, 2002: 73). It has only been pointed out that an increase in credibility in the subsidiary R&D site is important for the evolution of the R&D subsidiary (De Meyer, 1993: 112-113). Such an increase in credibility is only possible if the competence of the laboratory is acknowledged (De Meyer, 1993: 112). Such competence acquisition can only be achieved through technological capability upgrading, especially in the context of a late industrializing country, which then increases the importance of the periphery. This deepening of technological capabilities is crucial for R&D subsidiaries in late industrializing economies if they are to reach the economic status of the most advanced nations and increase their strategic importance in the internal R&D organization. Such a technological capability upgrading in late industrializing countries has frequently been associated with local affiliates of foreign MNEs (Lall, 1992: 166-169). However, these MNEs retain the more complex technologies (such as R&D) in their home countries. This, on the other hand, means that MNEs transfer technological knowledge, but not the process of generating new knowledge (Costa and De Queiroz, 2002: 1432). Therefore, a clear framework analyzing different technological capabilities and key influencing factors on technological capability upgrading is lacking. Moreover, previous literature has not investigated the relationship between technological capability upgrading and its performance implications at a subsidiary level. Therefore, this research will be advanced by examining the evolving process of technological capability upgrading at overseas R&D subsidiaries in a late industrializing country and its implications for R&D performance. 27 Managerial implications are drawn as well: more specifically, how R&D subsidiary managers need to address both internal as well as external R&D management needs in a latecomer country such as Singapore. These needs are examined from a network perspective in this dissertation, with the goal of contributing to a more differentiated view on these internal and external R&D management needs than exists at present. The following exhibit summarizes the research gaps in the international R&D literature and shows how this doctoral dissertation attempts to fill these gaps. 28 Exhibit 8: Research Gaps in the International R&D Literature Current Findings of Current Findings of Current Findings of R&D Inter- International R&D R&D Inter- nationalization Management nationalization Determinants Different types of Process Home-base international R&D Sequential R&D augmenting and organizations and home-base different managerial exploiting styles Internationalization determinants Research Gaps Research Gaps Research Gaps More comprehensive International R&D Implications of R&D approach lacking Management: Need Internationalization for new framework on R&D subsidiaries Contribution of the Contribution of the Contribution of the Dissertation Dissertation Dissertation Development of a Development of a Analysis of holistic determinant framework of a technological metanational R&D capability upgrading organization of R&D subsidiaries Source: Author The following exhibit presents the corresponding research questions: 29 Exhibit 9: Research Questions 1. What type of international R&D model enables an R&D organization to tap into knowledge existing in non-traditional R&D locations? 2. What characteristics would such an international R&D organization show? 3. What is the relationship between the type of R&D organization and R&D performance at subsidiary level? 4. Why are R&D subsidiaries in non-traditional R&D locations at the periphery? 5. How can R&D subsidiaries in non-traditional R&D locations create a critical knowledge base? 6. What is the nature of the process of upgrading the level of technological sophistication of R&D subsidiaries in non-traditional R&D locations? 7. What are managerial implications for R&D managers in a late industrializing context? Source: Author 30 3 Research Methodology 3.1 Fundamental Approaches to Research Methodology The researcher with a positivist view of the world regards reality as objective. Positivists assume that the researcher is independent of and is not biased by his/her research matter and that independent causes lead to observed effects (Bentz and Shapiro, 1998: 125). From this positivist point of view, quantitative analyses of data obtained are emphasized in an attempt to identify common patterns or processes, with the objective of generalizability (Bryman, 1998: 139). Phenomenologists, on the other hand, assume that the researcher is not independent of the phenomenon under investigation and that reality is not objective. Characteristically, phenomenologists focus on qualitative methods, their models not necessarily being mathematical, but rather verbal, diagrammatic, or descriptive (Remenyi at al., 1998: 32-37). Many scholars suggest that to overcome this distinction and hence overcome the drawbacks of single research methods using a variety of research methods in the form of triangulation is recommendable (Jick, 1979: 608-609). Research methods inform us only of narrow realities and are only capable of providing descriptions that reflect incomplete images of organizations, which are complex. Therefore, several paths of inquiry are recommendable and their individual quality is to be distinguished in order to increase the overall validity of findings (Daft, 1980: 633). This dissertation adopts empirical understanding as a basis for the research, while a systematic approach is undertaken, both with regard to gathering data as well as to the testing of propositions (Black, 1999: 3). Most noticeably, this dissertation adopts the point of view that combining both qualitative and quantitative research methods leads to superior results than those achieved when only one approach is applied. The aim thereby is to combine the advantages of both methods and to eliminate the disadvantages of a single method research design (Jick, 1979: 608-609; Daft, 1980: 31 633). Hence, a multi-method approach in the form of triangulation is used in order to gain an improved understanding of the research subject. How this dissertation attempts to achieve such an improved understanding is explained in the next sections, where three research methods, namely archival analysis, case study and survey, are evaluated in more detail. 3.2 Research Methods Used in this Dissertation 3.2.1 Archival/Theoretical Analysis In an archival analysis, the sources of data are various types of documentation, public records, or other units of analysis. Dane (1990) defines archival research as any research which deals with public records as the unit of analysis. Content analysis as one form of archival analysis, for instance, proceeds systematically and makes inferences from theory (Dane, 1990: 170). What distinguishes archival analysis from other research methods is that information is available through archival analysis before one’s own research has begun (May, 1997: 160-161). Disadvantages of archival analysis include the potentially considerable age of data and differences in the unit of analysis used in previous studies and that used in one’s own research. Dependence on the quality of data from previous research is a further problem, the reliability and validity of data collected by others being difficult to determine (Dane, 1990: 187). Through such an archival analysis in the form of a formal theoretical inquiry, however, new knowledge based on extant knowledge can be created by means of combining, extending, analyzing, and integrating existing research areas. This means making use of cross-fertilization (Bentz and Shapiro, 1998: 141). Consequently, one advantage of archival analysis is that such an analysis of different disciplines and theories, namely an interdisciplinary approach, allows the researcher to gain new insights. For example, the international R&D literature stands at the intersection of at least five intellectual domains – organizational theory, international business, technology and innovation management, strategic management, and economics. Therefore, no single perspective 32 is able to provide a complete understanding of such a multifaceted subject. As a result, archival analysis may help one to gain new insights. A further advantage is that this research method is economical regarding research resources (such as financial, temporal, and human constraints). This is especially true where access to the research object is costly due to geographical or social barriers or non-reactive research. Archival analysis/theoretical inquiry is used in this dissertation to derive the research questions from the identified research gaps. Equally importantly, archival analysis/theoretical inquiry into different research areas, for instance on international business and technology and innovation management, can create new knowledge by means of cross-fertilization of the individual research areas. Moreover, constant archival/theoretical analysis is necessary to ensure that the findings of the latest publications on international R&D management are included in the dissertation. And finally, archival research (for instance in the form of analysis of corporate information) is a valuable complement to other research methods, for instance, in the form of data made available through in-depth interviews. Hence, archival/theoretical analysis is an essential element in this dissertation, not sufficient on its own, but used to complement other research methods. Such analysis focuses on data collected in the past, the case study approach, however, focuses on contemporary events. This research method is explained below. 3.2.2 Case Study According to Eisenhardt (1989: 534), a case study is a “research strategy, which focuses on understanding the dynamics present within single settings”. The case study can be of a qualitative as well as a quantitative nature, depending on the data collection methods (Eisenhardt, 1989: 534-535). Case studies are well suited to answer how/why research questions. Furthermore, case studies can be used when there is little or no control over behavioral events, when the research questions examine a contemporary 33 event, and when the boundaries between phenomenon and context are not clear (Yin, 1981: 59; Yin, 1989: 17). Case studies can be used for providing description, testing theory, and building theory (Eisenhardt, 1989: 535). Eisenhardt (1989) focuses on the last aspect and proposes an eight-step approach for generating theory from case study research. First, the research questions need to be defined. Second, the researcher is confronted with the selection of cases (for a discussion of single versus multiple case study research see Dyer and Wilkins, 1991 and Eisenhardt, 1991) based on theoretical sampling. Furthermore, multiple data collection methods are applied. In a next step, the data is analyzed, dealing primarily with within-case issues as well as with cross-case patterns. The shaping of hypotheses constitutes the sixth step before the generated theory is compared both with similar as well as conflicting literature and closure is reached. Weaknesses of this research method are that it can result in overly complex theory due to the large amount of data or in excessively narrow theory because the case study research is very specific (Eisenhardt, 1989: 547). Moreover, case study research only allows theoretical generalization, but no empirical generalization. Another problem which is highlighted by Gable (1994: 113) refers to the difficulty experienced in case study research when seeking to manipulate independent variables. This in turn implies that it is not a single task to establish causality in case study research. Moreover, there is the risk of improper interpretation and the lack of ability to randomize cases (Gable, 1994: 113). Consequently, certain limitations seem apparent in case study research. On the other hand, the case study approach also offers unique strengths. This research method is especially appropriate for new topic areas (Gable, 1994: 113), very likely to generate new theory, to be testable and to be highly empirically valid (Eisenhardt, 1989: 546-547). Moreover, the case study method allows the researcher to comprehend the nature and complexity of the research matter under investigation (Yin, 1989: 14). Evidently, case studies are suitable for examining highly complex, recent phenomena (for instance R&D internationalization beyond the triad nations) and for obtaining in34 depth insights into the research topic. Such research brings the researcher detailed insights and contributes to a richer understanding of international R&D management. For these reasons, case study research is important in this dissertation. The first case which is analyzed in this dissertation is Novartis’ Research Organization as an example of a metanational R&D organization in the making. The second case analyzes the technological capability upgrading of Leica Instruments in Singapore. Analysis of the newly established R&D site of Lilly Systems Biology and its internal and external network linkage is part of a third case study. Managerial implications are also drawn from these cases. Consequently, the case study approach is very appropriate as a way of obtaining in-depth insights into the research questions this dissertation addresses. Based on the recognition that different research methods lead to superior results, better than those provided by one research method, this case study approach is complemented by an in-depth survey of 61 R&D units in Singapore. 3.2.3 Survey A survey allows information to be obtained from participants directly or indirectly, either orally or in a written form. It allows “questioning persons and record their responses for analysis” (Emory and Cooper, 1991: 318). The survey can be directed at single respondents in a firm, multiple respondents, single or multiple expert panels, and to both the corporate and subsidiary offices of a firm (Snow and Thomas, 1994: 462). Surveys can be conducted in various ways: face-to-face interviews, telephone interviews, mail and electronic surveys (Emory and Cooper, 1991: 320-343; for electronic surveys see Zhang, 2000). Three different types of information can be collected: facts, opinions, and behaviors (Dane, 1990: 119-123). The data analyses depend on the particular study and type of data that needs to be collected. A survey is considered an appropriate research method when the research question is of a “who/what/where/how many/how much” nature (Yin, 1989: 17). Furthermore, a survey is applicable when no control over behavioral events is required and when contemporary events are examined (Yin, 1989: 17). A survey is also suitable when data across several time periods is to be collected (Snow and Thomas, 1994: 462). 35 Remenyi at al. (1998) distinguish between two forms of a survey, in-depth surveys and large scale surveys. Whereas the in-depth survey attempts to obtain detailed and rich evidence from a relatively small number of informants, the large scale survey, on the other hand, attempts to collect large quantities of data or evidence (Remenyi at al., 1998: 56), its sample ideally that which is the most representative and most consistent with the hypotheses (Black, 1999: 28). 3.2.3.1 Large Scale Survey A large scale survey is addressed to an important number of informants, the objective being to discover relationships based on a quantitative analysis, relationships that are common across organizations. It aims to provide generalizable statements about the phenomenon under investigation (Gable, 1994: 114). Evidently, a major strength of a large-scale survey is its good geographical coverage since respondents in various geographical areas can be researched. Equally importantly, by conducting a large-scale survey a large sample can be covered in a cost efficient and convenient manner (Snow and Thomas, 1994: 462; Remenyi et al., 1998: 56). This is especially valuable when statistical relationships are examined and are to be generalized (Bechhofer and Paterson, 2000: 75). A further strength of a large scale survey is that the respondent can take more time to answer questions, rethink them, and can reply more carefully than is the case in an in-depth survey, which usually involves face to face interviews (Cooper and Schindler, 1998: 304). In addition since the large-scale survey is more anonymous, the respondent may be more willing to reveal more information than in a direct interview (Emory and Cooper, 1991: 333). Since a large-scale survey addresses an important number of informants, the nature of evidence may, however, be rather superficial (Remenyi at al., 1998: 57). A large-scale survey encompassing objectivity and testability might be carried out at the cost of a richer understanding of the phenomenon under investigation (Gable, 1994: 114). Moreover, the informant might interpret a question or concept very differently from what the researcher’s intends and, hence, may answer a different question (Cooper and Schindler, 1998: 304). Furthermore, the informant might deliberately give a false answer or an answer without knowing, because he/she feels obliged to respond (Emory 36 and Cooper, 1991: 319). A further disadvantage of a large scale survey is its low response rate, which reduces the confidence in generalizability; or in other words, a low response rate questions the extent to which results from the survey can be generalized to the whole sample (Snow and Thomas, 1994: 462). Given these weaknesses, an in-depth survey seems to be more suitable for this dissertation, as can be seen in the discussion of the following section. 3.2.3.2 In-depth Survey Since an in-depth survey involves personal or telephone interviews, this method is very time-consuming and requires good cooperation from respondents. The researcher, however, can obtain more detailed evidence in comparison to a large-scale survey (Cooper and Schindler, 1998: 291). Due to the direct interaction between researcher and research subject, the researcher can immediately respond to the information given, can ask additional questions, can clarify doubts, and can gather supplemental information through observation (Emory and Cooper, 1991: 320). This research method also allows the researcher to constantly improve the in-depth interview, for instance questions may be phrased more clearly and the interaction between respondent and researcher may be optimized. Given the advantages of such a research method, it seems that an in-depth survey is important for the research questions of this dissertation, in order to complement the case study research and hence to obtain more generalizable results. At the same time, by using an in-depth survey, the researcher overcomes a potentially low response rate, a major problem of large-scale surveys, especially in an Asian context. The Asian context is rather intransparent, with respondents reluctant to volunteer information. That a large-scale survey results in a very low response rate in Asia has been confirmed by my conversations with several academic scholars and R&D managers in Singapore2. 2 One academic scholar, for instance, sent out a large-scale survey to more than 4,000 firms and the response rate was so low (despite measures to increase the response rate) that it was impossible to work with the sparse data received. 37 In an in-depth survey involving personal interviews, non-response as well as response error can occur. In the first case, the researcher is not able to locate the interviewee or to encourage the person to participate. In the second case, the response error, the data reported differs from the actual data (Cooper and Schindler, 1998: 297-298). These two types of error, however, also apply to a large-scale survey. Various measures have been taken in order to avoid non-response errors as well as response errors. The identified R&D managers received a letter asking them to participate in the research project (see appendix). Those R&D managers who did not reply automatically were sent a second or third letter and/or contacted by phone in order to persuade them to agree to an appointment for an interview. Of the 100 firms identified, 61 participated in this research study, resulting in a response rate of 61%. The sample of 100 firms contains the R&D subsidiaries with the highest technological sophistication based on information of A*Star (Agency for Science, Technology and Research). 39 R&D subsidiaries declined to cooperate for various reasons, for instance due to corporate restructuring, time constraints or discontinuation of their operations in Singapore. In order to avoid response error, the researcher explained the purpose of the research study, attempted to phrase questions clearly and to clarify doubts or misunderstandings immediately during the interviews. Through the in-depth interviews of 61 R&D subsidiaries in Singapore, the researcher was able to obtain a rich primary data set on R&D organizations in the periphery, namely Singapore. The three research methods, archival analysis, case study and in-depth survey involving personal interviews, allow a detailed investigation in the form of case studies and a general overview in the form of an in-depth survey on R&D organizations in Singapore, complemented by archival research. Hence, this work bridges the different research traditions of quantitative versus qualitative research and takes advantage from such a triangulated approach. 38 3.3 Overview of Research Methods Applied A preliminary exploratory field study was conducted in April 2002 (in-depth interviews with seven R&D subsidiaries of MNEs in Singapore were conducted), which allowed the researcher to identify important variables and problems for further investigation and to discuss major issues pertinent to international R&D management. This exploratory phase, which allowed getting close to the phenomenon under investigation, led to the development of a survey (see appendix) whose objective was to examine the relationships among the key variables identified. This survey is of an explanatory and statistical nature. The exploratory phase was also used for pre-testing the survey instruments and as a crosscheck against questionnaire responses, which improved internal validity and the interpretation of quantitative findings. The subsequent fieldwork was carried out in personal, face-to-face, in-depth interviews, with the survey as a basis, and complemented by archival research. This approach allows standardized data to be obtained by addressing a specific set of questions for each company (in the form of a questionnaire). Besides this, the researcher can investigate important firm-specific issues in the personal interview which enhances an understanding of the firm-specific context. For instance, the researcher gains knowledge of the firm’s technology strategy, its geographical R&D dispersion, and its R&D management practices. The firm-specific context is further explored by case studies focusing on the issues under investigation. The unit of analysis is the firm specific R&D organization (R&D site or R&D department). Interview partners are R&D directors, senior R&D managers of the respective R&D units/R&D departments at subsidiary level and managing directors of the subsidiary in Singapore. As unit of analysis an R&D site/R&D department was chosen because it is relatively easy to identify, its size can be determined and involves a longer-term commitment than for instance a single research agreement (Kuemmerle, 1996: 50). The study examines mostly R&D subsidiaries of MNEs, since they are particularly active in establishing R&D sites abroad (Boutellier, Gassmann and von Zedtwitz, 2000: 8-10); local R&D subsidiaries are also examined in order to establish interesting comparisons. 39 External validity refers to the generalizability and representativeness of the sample (Black, 1999: 49). This type of validity shows the extent to which the inference drawn from a research method can be generalized to or across time, settings, and persons (Mitchell, 1985). The sample was chosen carefully. Overall, the researcher conducted 85 interviews with 51 R&D subsidiaries of MNEs, 10 Singapore based R&D organizations, two research institutions (Singapore Institute of Manufacturing Technology and Institute of Bioengineering) and the two main government bodies (Economic Development Board and A*Star) in this cross-sectional study. A specific set of R&D subsidiaries was revisited several times to gain more in-depth insights into the type of R&D organization they presented. Both the managing directors as well as the R&D managers of the different R&D departments were interviewed. Out of the 51 R&D subsidiaries of MNEs, 20 are American, 19 European, 10 Japanese and 2 R&D subsidiaries with other parent firm nationalities. Out of the 51 R&D subsidiaries of MNEs, 14 R&D subsidiaries belong to the biomedical sector, 13 to the electronics sector, 7 to the chemical sector, 9 to the information technology and communication sector, 4 to the engineering sector and 4 to other sectors (food and aviation sectors). The industry classification is based on the National Survey of R&D in Singapore 2001, which is conducted annually by A*Star. According to this survey, there are 206 private firms (either wholly foreign owned or with less than 30% local ownership) conducting R&D in Singapore in 2001 in the above-mentioned industries. The sample of 51 R&D subsidiaries therefore translates into a response rate of 25% for R&D subsidiaries of MNEs only. Besides these 51 R&D subsidiaries of multinational firms, 10 Singaporebased firms were interviewed. Hence, 61 R&D organizations have been interviewed overall. Based on the 100 most technologically sophisticated R&D organizations according to information provided by the A*Star, the response rate is 61%. As has been discussed, the research methods applied in this dissertation bridge the different research traditions of quantitative (questionnaire) versus qualitative research (firm-specific context) and take advantage of such triangulation. After reviewing the literature and evaluating the research methodology, the following chapter analyzes the context of the R&D subsidiaries under investigation, namely a non-traditional R&D location, Singapore. 40 4 Late Industrializing Context: Singapore as a NonTraditional R&D Location 4.1 Singapore’s Science and Technology Policy The following outline of Singapore’s science and technology policy serves as a basis for understanding the periphery context in terms of a non-traditional R&D location. The internationalization of R&D has up to today mainly been confined to the triad nations (Europe, Japan and the US). Singapore is a non-traditional R&D location, with R&D investment into Singapore a relatively recent phenomenon. Only under the recent science and technology policy adopted by the Singapore government3 has the focus on R&D activities been intensified. The following discussion attempts to illustrate this recent science and technology policy, based on a short analysis of Singapore’s historical economic development. Singapore’s general industrial policy focused on manufacturing at the start of its industrialization process after gaining national independence in 1965. Singapore adopted a technology leverage strategy from MNEs due to its limitations of natural resources and the size of its domestic market. FDI by MNEs has been encouraged through highly favorable conditions. This strategy, which focused mainly on FDI in manufacturing, allowed the country to gain access to new technologies and to create a significant number of jobs. The favorable incentive schemes offered by the Singapore government have also generated a self-sustaining positive feedback loop (Song, 2002: 192; 196). This is the case because domestic firms’ skills and expertise have been enhanced through the supply of goods and services to MNEs (also see Feinberg and Majumdar, 2001). Furthermore, these incentive schemes have created a first mover advantage by attracting high value added production. They also provide a motivation for MNEs to keep on upgrading local operations. The overall goal of this industrial 3 The Agency for Science, Technology and Research (A*Star) was established in 1991 in order to implement Singapore’s science and technology policy. Three five-year national technology plans (1991-1996, 1996-2001, 2001-2006) have been adopted since then. 41 policy is to raise the level of technological sophistication. This outward oriented policy led to accelerated industrialization from 1967-79 (Hobday, 1995). The electronics industry started as the first industry with manufacturing (assembly) of simple consumer goods thanks to cheap labor costs and stable working conditions (Hobday, 1995: 1173; Matthews, 1999: 56). Due to Singapore’s technology leverage strategy from MNEs, Singapore’s electronics industry relied almost exclusively on foreign companies. For instance, Philips began its operations in 1951 with a trading office of four employees and expanded into the production of transistor radios in the early 1960s (Hobday, 1995: 1174). MNE investment in the electronics sector helped to start up and further train local firms. In the 1970s, simple manufactured goods were produced, before the next phase of professional electronics was entered into in the 1980s. From 1982-85, a cluster of hard disk drive producers was created and by 1991 Singapore was a major producer of such components (for a discussion on the hard disk drive industry see McKendrick, Doner and Haggard, 2000). This example shows how the technology leverage strategy adopted led to the building up of expertise in the electronics industry. By the 1980s and 1990s, Singapore had developed a high-tech industry and a regional services center, shifting low wage production into the surrounding region (Song, 2002: 192). Overall, this technology leverage strategy, which was mostly based on manufacturing, has resulted in sustainable industrial development (Matthews, 1999: 56). The annualized real GDP growth rate, for instance, was 9.6% for the time period 19701974, 8.5% for 1975-1979, 6.3% for 1980-1984, 8,5% for 1985-1989, 9.2% for 19901994 and 4.3% for the years 1995-1999. The following exhibit gives a more detailed overview of major macroeconomic indicators: 42 Exhibit 10: Selected Macroeconomic Indicators for Singapore 1965-2000 Years Real GDP Consumer Price growth rate % Index Inflation Unemployment Indigenous GNP Rate % Rate % GNP per per capita in capita in Singapore dollars Singapore per year dollars per year 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 6.6 10.6 13.0 14.3 13.4 13.4 12.5 13.3 11.3 6.8 4.0 7.2 7.8 8.6 9.3 9.7 9.6 6.9 8.2 8.3 -1.6 2.3 9.7 11.6 9.6 9.0 7.1 6.5 12.7 11.4 8.0 7.6 8.5 0.1 5.9 9.9 0.3 2.0 3.3 0.7 -0.3 0.4 1.8 2.2 19.6 22.3 2.6 -1.9 3.2 4.8 4.0 8.5 8.2 3.9 1.2 2.6 0.5 -1.4 0.5 1.5 2.4 3.4 3.4 2.3 2.3 3.1 1.7 1.4 2.0 -0.3 0.0 1.3 n.a. 8.9 n.a. 7.3 n.a. 8.2 (census) n.a. n.a. 4.4 3.9 4.5 4.4 3.9 3.6 3.3 3.5 (census) 2.9 2.6 3.2 2.7 4.1 6.5 4.7 3.3 2.2 1.7 (census) 1.9 2.7 2.7 2.6 2.7 3.0 2.4 3.2 4.6 4.4 (census) n.a. n.a. n.a. n.a. n.a. 2478 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 8343 9798 11027 12474 13515 12973 12842 13814 15692 17788 20075 21380 23887 258893 29500 30998 32489 35098 34423 35928 38445 Source: Adapted from Peebles and Wilson, 2002: 273 43 1618 1773 1945 2188 2499 2825 3233 3798 4575 5498 5996 6353 6817 7558 8577 9962 11067 11942 13574 14853 14666 14576 15515 18093 20381 22411 24021 25510 28200 31872 34420 35482 39494 37226 38832 42212 The macroeconomic indicators presented in the exhibit show that Singapore has experienced sustainable economic growth since its independence in 1965. GNP per capita, for instance, increased from 1618 Singapore dollars in 1965 to 42,212 Singapore dollars in 2000. Indigenous GNP per capita refers to GDP earned by Singaporeans only. The difference between GNP per capita and indigenous GNP per capita is thus the amount of GDP which is produced by foreigners. These two measurements have been adopted since a large number of foreigners works in Singapore (Peebles and Wilson, 2002: 135-136). The consumer price index inflation has been low with the exception of the years 1973, 1974 and 1980, 1981. The real GDP growth rate is 8.7% on average for the years 1965-2000. There are two views on this continual development of growth in the literature. The accumulation view of growth sees it as result of high savings and investments that made it possible for late industrializing countries to use technologies inherited from the world’s technological leaders and to use them better (proponents of this view are, for instance, Young, 1995; Collins and Bosworth, 1996). This view assumes that the output growth can be explained by the increase in the quantities of inputs of capital and labor (Peebles and Wilson, 2002). On the other hand, the assimilation view states that the source of growth is an outcome of productivity growth resulting from increases in the learning, entrepreneurship and innovation that these economies have experienced. Thus, not only the adoption of foreign technologies but also development of indigenous technologies have been possible (proponents of the assimilation view are, for instance, Hobday, 1995 and Kim, 1998). This dissertation assumes that the accumulation view probably applies to the beginning of Singapore’s economic development. With Singapore’s transition towards more R&D activities, however, the latter view (assimilation view) should gain in importance as a way of explaining the country’s economic growth. More R&D activities may result in indigenous innovation, entrepreneurship and learning. Therefore, the assimilation view may be a way of explaining Singapore’s present and future economic growth. The reasoning for a shift towards more R&D activities is based on recent economic developments. Emerging economies, such as China and Malaysia, have increasingly gained importance in manufacturing due to their lower manufacturing costs compared 44 to those in Singapore. China in particular has attracted a large number of foreign producers. Therefore, it follows that Singapore’s economy can no longer rely only on manufacturing for its present and future economic development. Singapore’s most recent economic policy thus aims at the growth of economic activities beyond manufacturing. This in turn implies that a strong emphasis is placed on fostering R&D activities. Singapore’s recent science and technology policy is therefore based on building up R&D activities as they are conducted in the triad nations (US, Europe, and Japan; the internationalization of R&D is still confined to these triad nations, see Edler, Meyer-Krahmer and Reger 2002: 159-160). The objective is for new and innovative products and services to contribute at least 15% to the revenue of manufacturing companies by 2008, to reach the Swiss standard (1993) of 237 patents per 100,000 inhabitants and to have at least 50 successful new seed venture capitaltype start-ups per year (Haley and Low, 1998: 545-546). R&D activities are already to some extent conducted in the electronics industry, but Singapore’s science and technology policy attempts to promote such R&D activities in the biomedical sciences field also. The R&D capabilities in this sector comprise the pharmaceuticals, medical devices, biotechnology, and healthcare services sector according to the definition of the Agency for Science, Technology and Research (A*Star). As part of its strategy, the Singapore government has decided to build the biomedical sciences industry as the economy's ‘fourth pillar’, the other three ‘pillars’ being electronics, chemicals, and engineering (Wess, 2002: 1). The two main government institutions, the Economic Development Board (EDB) and the Agency for Science, Technology and Research (A*Star), are responsible for Singapore’s technology policy in all of these four fields. The major government institution in charge of the national science and technology policy is A*Star, which has established two research councils – the Biomedical Research Council (BMRC) and the Science and Engineering Council (SERC) to support and advance research. The BMRC is responsible for strengthening R&D capabilities in genomics, molecular biology, bioinformatics, bioengineering, and bioprocessing and oversees five public research institutes (Agency for Science, Technology and Research, 2001: 1). Furthermore, an intellectual property framework 45 and a research infrastructure, have been created, for instance Singapore Science Park I, II, and III, and the Tuas Biomedical Park (Wess 2002: 2-3). In addition, the education system has undergone changes in order to create local expertise in the biomedical sciences. BMRC, for instance, initiated a new graduate scholarship scheme in which the hundred top graduates will be given scholarships to pursue their PhDs at the National University of Singapore and to take up a two-year post-doctoral fellowship at a foreign university. Besides this new scheme, BMRC also initiated the National Science Scholarships Program in 2001 to provide scholarships from undergraduate to PhD level (Yong, 2003: 4). An example of one of the first firms investing in the biomedical sciences is the Swiss pharmaceutical firm Novartis, which has established a Novartis Institute for Tropical Diseases to conduct research on the diseases of dengue fever and tuberculosis (also see case study). A further example is the investment by the American pharmaceutical firm Eli Lilly in the form of Lilly Systems Biology, which was as one of the first firms to receive funding from the Singaporean government’s US$600 million biomedical sciences fund (also see case study). The growth target for the biomedical sciences sector is a 10% contribution to Singapore’s total manufacturing output by 2010 (Yong, 2003: 4). The SERC is responsible for the enhancement of R&D capabilities in areas such as microelectronics, material science, electronics, engineering, information communication and chemistry and oversees eight public research institutions (Agency for Science, Technology and Research, 2001). In addition, Exploit Technologies Pte Ltd (ETPL) was formed to commercialize the intellectual property created by A*Star’s Research Institutes and Centers. The Corporate Planning and Administration Division (CPAD) supports the two research councils and ETPL. A*Star has identified three key thrusts in its science and technology policy: to strategize public research to integrate with industry clusters; to train human capital for research and industry; and to create, own and exploit (COE) intellectual capital, for instance by offering incentives schemes such as the ‘R&D Assistance Scheme’ or the ‘Cooperative Research Program’. The stimulation and entrenchment of R&D in 46 Singapore-based private sector firms are key goals (Agency for Science, Technology and Research, 2001). Besides A*Star, the EDB also plans and executes strategies to sustain Singapore’s competitiveness. It enables multinational and Singapore-based companies to start operations, to enhance them and to upgrade them to higher value-creating operations. Furthermore, the EDB offers various incentives for fostering R&D activities. Examples include the ‘Research Incentive Scheme for Companies’ (grants may be offered to support the development of in-house R&D capabilities among Singaporebased companies for a maximum of 50% of the total research efforts for a period of five years) (Swinbanks, 1997: 2). This overview of Singapore’s science and technology policy shows how a nontraditional R&D location can foster R&D activities and thus increase its level of technological sophistication and create a critical knowledge cluster. The barriers Singapore as a non-traditional R&D location is facing in this process are discussed in the next section. 4.2 Challenges for Singapore as a Non-Traditional R&D Location Despite the efforts of the recent science and technology policy in Singapore, challenges remain on the way to more R&D activities. Since Singapore has a relatively short industrial history, it faces the challenges of a smaller R&D scale compared to the triad nations. It also has a weaker technical capability since the Singapore economy was mostly based on manufacturing, as has been illustrated above. It is therefore important to increase technological sophistication. This may help to change the perception of the periphery and create a critical knowledge cluster. Three major challenges to Singapore’s transition towards more R&D activities have been identified, namely insufficient local human resources, overdependence on MNEs for innovation, and lack of entrepreneurship and creativity. Many interview partners referred to these challenges during the personal interviews conducted for this 47 dissertation. The first challenge to be tackled lies in overcoming the current deficiency in local human resources. 4.2.1 Structural Factors 4.2.1.1 Insufficient Local Human Resources A barrier in the transition towards more R&D activities, which has been reiterated in many of the interviews the researcher has conducted, is a significant deficiency in the availability of scientific and technical manpower. When the R&D managers were asked how they acquired the first human resources to build up the R&D subsidiary, most of them indicated that they had to rely on foreign talent rather than local manpower. The majority of the interview partners of MNEs indicated that either foreign talent from within the R&D organization had been transferred to Singapore or foreign talent was newly hired either on a global or regional basis in order to build up R&D subsidiaries. This foreign personnel help to train and upgrade local manpower, either at headquarters or at the respective R&D subsidiary and the Singapore government often supports such training. The Singapore government has implemented various measures to meet the lack of local human resources. Various scholarship programs and reforms of the university system have been undertaken (Peebles and Wilson, 2002: 264-266). It will, however, take time to fill the void. Besides this challenge to develop critical human resources, Singapore’s future economic growth depends on the building up of indigenous innovative capabilities, which would mean that the relative overdependence on MNEs for innovation would be reduced. 4.2.1.2 Overdependence on MNEs for Innovation According to Mahmood and Singh (2003), who study technological dynamism in various Asian countries, the relative contribution of innovation by multinational subsidiaries is highest in Singapore and India and minimal in Taiwan and South Korea. Peebles and Wilson (2002) confirm that a high share of innovation results from MNEs 48 and state that much of Singapore’s technological base has been imported with MNEs, but not through indigenous innovation and enterprise. Although Singapore appears to have developed relative specialization in electronics and other high technology areas, a large part of Singapore’s patenting activity continues to come from MNEs rather than from domestic entities: 46% of the patenting activity arising from Singapore in the years 1990-1999 is the result of MNEs’ R&D efforts (Mahmood and Singh, 2003: 1044). Furthermore, the innovative activity is highly concentrated; the fraction of Singapore’s patents held by its top 50 assignees is 70% (Mahmood and Singh, 2003: 1052). The relative role of domestic entities in patenting activity, however, is increasing: 59 out of 148 patents were granted to domestic entities for 1990-1994 and 287 out of 499 total for 1995-1999 (Mahmood and Singh, 2003: 1044). It seems that the recent adoption of a more R&D oriented policy by the government is helping Singapore to begin developing indigenous innovative capabilities. While MNEs are important in Singapore’s transition towards more R&D activities, the creation of indigenous innovative capabilities is critical in sustaining the transition towards more R&D activities. Besides these structural challenges, social and cultural factors play an important role in Singapore’s transition towards more R&D activities. 4.2.2 Social and Cultural Factors 4.2.2.1 Lack of Entrepreneurship An important factor in Singapore’s transition towards more R&D activities is the creation of an entrepreneurial climate. Different perspectives on entrepreneurship exist in the literature. Authors such as Cooper and Dunkelberg (1986) and Schumpeter (1934) view entrepreneurship as the identification and pursuit of market opportunities by recombining and allocating diverse resources. Other authors such as Usbasaran, Wright and Westhead (2003) emphasize heterogeneity in entrepreneurship. 49 The literature has also emphasized the economic importance of entrepreneurship. Especially in times of strong international competition, entrepreneurial activities are regarded as a driving force for innovation and as such sustained competitive advantage (Luethje and Franke, 2003: 135). The literature has analyzed the proclivity for entrepreneurship as a function of various factors such as personal and cultural factors (e. g. Huefner, Hunt and Robinson, 1996 and Smith, 2003). An adverse social attitude to failure, both on an individual as well as societal level, may lead to an unfavorable environment for entrepreneurship (Peebles and Wilson, 2002: 254-255). Since entrepreneurship may involve high risk, failure might occur. Singapore has an adverse social attitude to failure, an attitude which may be detrimental for the creation of an entrepreneurial environment. In Singapore, more secure career options, as offered by the public sector or by MNEs, are preferred. The public sector may skim off outstanding entrepreneurial talent because it offers high salaries, prestige and security (Haley and Low, 1998: 540). The various MNEs also offer highly attractive career paths. These options discourage entrepreneurialism. In Singapore’s transition to more R&D activities then, the Singapore government is attempting to stress the need for more entrepreneurship and risk-taking in order to create a more favorable environment for entrepreneurialism. 4.2.2.2 Lack of Creativity Lack of creativity is a major hindrance to indigenous innovation since all innovation begins with creative ideas (Amabile et al., 1996: 1154). Three dimensions of creativity are distinguished, namely individual creativity, group creativity and organizational creativity (Amabile, 1997: 39; Oldham and Cummings, 1996: 607). On the individual level, creativity depends on creative thinking skills, task motivation and expertise (Amabile, 1998: 76). Individual creativity affects group creativity, which is in turn characterized by group composition (e.g. individuals of diverse backgrounds), group characteristics (e.g. group size, group cohesiveness) and group processes (e.g. communication patterns, problem solving strategies) (Woodman, Sawyer and Griffin, 1993: 301). Organizational creativity is seen as the creation of valuable, relevant output for the organization by a complex social system, be it in the form of a new product, new service or process (Woodman, Sawyer and Griffin, 1993: 303). Hence, 50 organizational creativity is viewed as subset of innovation, which can also encompass the adaptation of preexisting products or services, or those created outside the organization (Woodman, Sawyer and Griffin, 1993: 293). Other authors agree with this definition of creativity as the production of new and useful outcome and emphasize the implementation aspect of innovation, which is viewed as the successful functioning of the creative output (Oldham and Cummings, 1996: 607; Amabile, 1996: 1155). Thus, organizational creativity is by nature idiosyncratic, rare, hard to imitate and not substitutable. It is thus a source of competitive advantage. Previous studies explored the relationship between organizational properties and organizational creativity in order to determine what the key enablers of organizational creativity are. Woodman, Sawyer and Griffin (1993), for instance, found out that the availability of resources, free internal and external communication and information flows, as well as organizational structures, such as R&D networks, foster organizational creativity. These results are confirmed by Amabile (1997), who views the properties of organizational encouragement (for instance in the form of orientation towards risk), the availability of resources, and management practices as important for the prospering of organizational creativity. Similar findings are provided by Shalley, Gibson, Blum (2000), who regard organizational complexity, autonomy, and low organizational controls as crucial for organizational creativity. Oldham and Cummings (1996) looked both at individual and organizational creativity and found out that R&D scientists’ creative output was significantly related to the extent to which supervisors developed an understanding for the employees’ feelings, one of their key results being that the contextual measure, that is the organizational properties are independently and positively related to organizational creativity, hence underlying the importance of organizational creativity (Oldham and Cummings, 1996: 616-617). All of the above authors consider the following organizational properties, namely organizational encouragement (including free information and communication flows and organizational autonomy), organizational complexity, and the availability of resources as enablers of organizational creativity. It is a critical question whether Singapore can create creativity potential. While Singaporean culture – well crafted by the government – transformed Singapore into an 51 attractive investment location for MNEs, providing a good infrastructure and a graftfree business environment, it also appears hierarchical and highly disciplined (Haley and Low, 1998: 532-534). These characteristics seem to diminish creativity and the ability to innovate since they are in contrast to the properties of autonomy and complexity, which are enablers of organizational creativity. Haley and Low (1998: 550-551), for instance, point out that that the technocratic and authoritarian Singapore government system has a negative impact on creativity and entrepreneurship. A study by Shane (1992) also indicates that the extent to which a society stresses social hierarchy is negatively related to inventiveness. Hence, it is essential for Singapore to create an environment conducive to R&D creativity, where regulations are only minimal in order to reach continued competitiveness (Man, 2001: 230-233). The government-led policy more recently attempts to retain and foster creativity among Singaporeans, for instance in the education sector. Yet, despite these governmental efforts, it remains to be seen if such further engineering of the Singaporean culture may result in the creative entrepreneurial labor force that is vital for R&D activities. Whether Singapore can strike this balance of government intervention and the increasing need for creativity and entrepreneurship has not yet been demonstrated (Haley and Low, 1998: 551). 4.3 Concluding Remarks While Singapore faces challenges in its transition towards more and higher level R&D activities, it seems well on the way in this transition. The Singapore government has adopted a large number of measures to create a local pool of highly skilled labor and to create indigenous innovation (Yoshida, 2001: 6). For example, several spin-offs of research institutions resulted in new local enterprises. Creative Technology, a leading sound card maker, for instance, was founded by entrepreneur Sim Wong Hoo (Haley and Low, 1998: 547). The researcher conducted interviews with such spin-offs, which have been acquired by MNEs. Given this development towards more R&D activities, social and cultural factors may change over time as well. An entrepreneurial and more creative environment may 52 result from a general change towards more R&D activities in the economic structure. As can be seen, the Singapore economy is already in the process of overcoming some of the barriers which have been discussed. It will be interesting to see how Singapore’s future economic development will evolve. 53 5 Empirical Evidence 5.1 Quantitative Empirical Findings: R&D Internationalization Determinants and International R&D Organizations 5.1.1 Conceptual Framework of a Metanational R&D Organization 5.1.1.1 The Metanational Organization According to Doz et al. (1997) and Doz, Santos and Williamson (2001), MNEs face two major challenges today in maintaining their competitiveness. First, transferring knowledge created in the home base to the host country is insufficient for creating sustained competitive advantage. In today’s competitive environment, the international exploitation of strategic advantages coming from the home base is no longer a valuable strategy. This is the case because new important locations of technology and sophisticated customer demand emerge in non-traditional geographical areas. For example, initially manufacturing sites were established in low cost locations. Over time, such manufacturing sites can, however, develop their own abilities to create new technologies. Semiconductor manufacturing plants in South East Asia, for instance, have made innovations in the technologies for packaging semiconductors (Doz et al., 1997: 5). The challenge lies therefore in overcoming the dichotomy of home versus host country base (see chapter 2). This is called the location challenge. The second challenge comes from the fact that MNEs increasingly face more and more complex knowledge. Knowledge originates from more dispersed parts of the world; products and customers require more complex knowledge combinations (Doz, et al., 1997: 6-10). Firms so far have been able to manage simple knowledge, but are now required to manage increasingly complex knowledge. This challenge is especially important in an R&D context where knowledge is of a highly complex nature, especially in research. It is assumed that the higher the degree of knowledge complexity, the higher the resulting sustainable competitive advantage (Doz et al., 1997: 6). 54 According to the metanational framework, a metanational advantage is necessary when these challenges become relevant for MNEs. Three distinct capabilities are required to build such a metanational advantage, namely sensing, mobilizing ever more complex knowledge and putting it into operations. In other words, MNEs need to create sensing nodes to detect critical knowledge, mobilize this specific knowledge and then connect as well as apply this knowledge (Doz, Santos and Williamson, 2001: 5-9). That means that MNEs need to project, integrate and orchestrate knowledge, while at the same time improving their capability to address ever more complex types of knowledge, following a metanational strategy. This in turn implies that MNEs need to augment their capabilities to leverage highly complex and dispersed knowledge to create competitive advantage (for more details see Doz, Santos and Williamson, 2001). The term ‘meta’ stands for ‘beyond’. According to Doz, Santos and Williamson (2001), this term has been chosen because metanational firms do not draw their competitive advantage from their home country or from various host countries, a dichotomy which is overcome in the metanational context. Metanational organizations view the world as an entirety with pockets of different, specialist knowledge. 5.1.1.2 The Metanational R&D Organization The following section analyzes how an R&D organization could respond to the two challenges identified by proposing a metanational R&D organization: Location challenge: Large number of knowledge bases In accordance with the metanational framework, the international R&D organization is viewed in its entirety. Thus, overall R&D allocation is optimized and integrated accordingly. R&D organizations adopt a comprehensive strategy, allocating their R&D activities anywhere in the world where specific knowledge is available. This knowledge does not only refer to internal, but also to external knowledge bases, in contrast to the traditional R&D models, where only internal knowledge bases are considered (see chapter 2). Internal knowledge bases are knowledge bases within the R&D organization, whereas external knowledge bases are knowledge bases residing outside the organizational boundaries, namely in the external research environment. 55 The following example attempts to illustrate this comprehensive approach: While Bangalore (in India) has specific knowledge in software development, Ang Mo Kio (in Singapore) has specific expertise in the development of electronic related products and Shanghai (in China) has a strong manufacturing base. Hence, an R&D organization would conduct its software development activities in Bangalore, its electronic development activities in Ang Mo Kio, and an R&D unit with manufacturing support in Shanghai and thus would take advantage of the technological hierarchy within R&D. Following such a metanational strategy, the R&D organization is no longer concerned with exploiting or augmenting their knowledge base from a home country perspective. The corporate R&D organization is more concerned with tapping into specialized knowledge centers anywhere in the world and integrating this technological expertise to the fullest for the R&D organization, resulting in a highly specialized R&D organization. This means that the international R&D organization is not determined from a home versus host country perspective, but is determined by a holistic perspective (where it is best to allocate R&D activities based on the global landscape of the R&D organization). The international R&D organization acts as a ‘global scanner’ picking up and exploiting new technology wherever it evolves (Zander, 1998: 19). Hence, previous literature is extended by including this determinant, which will be referred to in the following as a non-dichotomous determinant of R&D. It is depicted below. 56 Exhibit 11: Determinants of R&D Internationalization Technological Advantage of Home Country Strong Home-Base Exploiting Home-Base Augmenting Non-dichotomous determinant (globally optimizing) Weak Market Seeking Technology Seeking Weak Strong Technological Advantage of Host Country Source: Author extending Kuemmerle (1996) and Le Bas and Sierra (2002) This non-dichotomous determinant is presumably a metanational R&D organization’s starting point for tapping into different knowledge bases worldwide. The idea of such a non-dichotomous determinant can be found in the notion of an international network of R&D laboratories (De Meyer and Mizushima, 1989: 145; De Meyer, 1993: 112; 115-119) which is considered crucial for the creation and diffusion of both internal as well as external know-how (De Meyer, 1993: 117-119). This network idea is further developed in the concept of the metanational organization. A metanational R&D organization needs to be able to sense, mobilize and integrate knowledge bases world wide, regardless of home or host country base. It is assumed here that sensing different knowledge bases worldwide would require MNEs to increase their degree of R&D internationalization. Otherwise, if R&D activities are 57 restricted to a low number of knowledge bases, it would be difficult to sense dispersed knowledge. This difficulty applies to a great extent to R&D since R&D is mostly focused on the triad nations (namely US, Europe and to some extent Japan). The periphery (non-traditional R&D locations) has so far been neglected, especially because it is considered that due to lower value added activities there, the R&D organization does not need to consider the periphery. But examples prove that through developments in their science and technology policy (see chapter 4) and constant technological capability upgrading (see second part of this chapter), the periphery can contribute to the whole R&D organization and should not be neglected. The negative perception of the periphery also renders the mobilizing of different knowledge bases difficult. Integration of knowledge refers again mostly to the integration of knowledge residing in the triad nations. A metanational R&D organization, however, would take advantage of a large number of knowledge bases (internal and external knowledge bases) including the periphery (see dimension of knowledge bases on the x-axis in exhibit 13). Knowledge Complexity Challenge: Leveraging of the technological hierarchy The second dimension of the proposed metanational R&D organization (see exhibit 13) refers to leveraging the technological hierarchy, which refers to all the technological levels depicted in the technology ladder in exhibit 12. The hierarchy can provide R&D knowledge with regard to market support, manufacturing support, development or research. As stated in the above-mentioned example, this R&D knowledge can reside in development expertise in South East Asia, in R&D knowledge related to manufacturing in East Asia and research expertise in Europe. It is essential to sense this knowledge within the technological hierarchy. The next step is to mobilize it. It is important that the periphery is not ignored during this step and that all relevant R&D knowledge regardless of its technological level is considered, especially because different R&D functions require different technological capabilities (also see exhibit 12). To integrate this knowledge is the third component in the metanational advantage and requires the capability to not only integrate the different knowledge bases, but also to integrate operation, improvement and generation capabilities in a fruitful way to create competitive advantage, since different technological stages require different management skills (for a more detailed 58 explanation of the technological hierarchy also see the second part of this chapter). This second dimension is shown in exhibit 12 below and is depicted on the y-axis in exhibit 13: Exhibit 12: Leveraging of Technological Hierarchy in the Metanational R&D Organization R&D subsidiary R&D in host country Other R&D subsidiaries headquarters Technological Hierarchy: Sense and mobilize different knowledge bases in the technological hierarchy Integration Capabilities Pure Science Unit (S) Basic Research Unit (R2) Applied Research Unit (R1) Generation Capabilities Exploratory Development Unit (D2) Advanced Development Unit (D1) Improvement Capabilities Manufacturing Support Unit (M2) Market Support Unit (M1) Operation Capabilities Source: Author’s extension of Medcof, 1997; Costa and de Queiroz, 2002; Amsden and Tschang, 2003. After describing the different dimensions of the metantional R&D organization, the proposed framework for the metanational R&D organization is visualized below. Furthermore, the positioning of the metanational R&D organization is compared to the traditional R&D models (for an illustration of the international R&D models see chapter 2). 59 Exhibit 13: Proposed Model for the Metanational R&D Organization Metanational Geocentric high R&D organization Leveraging of Technological Hierarchy R&D organization Integrated R&D network R&D Hub Model low Ethnocentric R&D Polycentric R&D organization organization few many Number of Knowledge Bases Source: Author Traditional international R&D models have been discussed in chapter 2. The elaboration of a metanational R&D organization has been presented in this chapter. How the different types of R&D organizations can address the location and knowledge complexity challenges is now to be analyzed. As can be seen from exhibit 13 and based on the discussion of the traditional international R&D models in chapter 2, the metanational R&D organization is an extension of the integrated R&D network. Within the model of the integrated R&D network, the home base loses its overall dominance, all R&D locations being equally important, play a strategic role and interact through multiple and complex coordination mechanisms. As can be seen, the three capabilities for a metanational advantage, namely sensing, mobilizing and integrating different knowledge bases, are partly 60 apparent in this international R&D model. Hence, an integrated R&D network may have to reinforce these capabilities to reach the state of being a metanational R&D organization. The same holds true, but to a lesser degree for the R&D Hub Model, an in-between model as described above. The home base is more important than in the integrated R&D network. The central R&D location leads in most technological fields. On the dimension of the degree of cooperation, it is situated between cooperation and competition. The importance of the home base should be decreased and knowledge management skills increased. The ethnocentric R&D organization is furthest away from the metanational R&D organization, the primacy of the home base being critical. As opposed to this concept, the geocentric, centralized approach of R&D is more international in nature. The same applies to the polycentric R&D organization. Both the geocentric as well as the polycentric R&D organization are strong in one dimension of the metanational R&D organization (namely leveraging of the technological hierarchy for the geocentric model, and a large number of knowledge bases for the polycentric model), but lack another dimension. The geocentric R&D model can leverage the technological hierarchy because international R&D activities are coordinated. In contrast, the polycentric R&D organization can tap into different knowledge clusters, but is highly decentralized. Finally, the metanational R&D organization acts as ‘global scanner’, picking up and integrating technologies in all relevant and emerging critical knowledge clusters. Thus, this proposed type of R&D organization takes advantage of a large number of knowledge bases and at the same time is highly effective in leveraging the technological hierarchy. The following exhibit gives an overview of the differences of all traditional R&D model versus the metanational organization. 61 Exhibit 14: Comparison of Traditional R&D Models versus the Metanational R&D Organization Dimensions of Conceptual Framework Traditional R&D Metanational models R&D organization Location Challenge: Number of Knowledge Bases Determinants of R&D internationalization Dichotomous in nature Comprehensive in nature Extension of R&D internationalization Only in the triad nations Triad nations and periphery Knowledge Complexity Challenge: Leveraging of Technological Hierarchy Degree of specialization Low to medium High Knowledge development Critical R&D personnel at headquarters or at key R&D subsidiaries Development of critical R&D personnel also in the periphery Locus of critical knowledge (innovation) At headquarters Anywhere in the and/or at key R&D R&D organization subsidiaries Source of knowledge Mostly internal Internal and external Source: Author As illustrated in exhibit 14, the metanational R&D organization is based on a comprehensive strategy of dropping the home base versus host country perspective, is present also in the periphery and is highly specialized (the different stages in the technological hierarchy are optimized). In contrast, the traditional R&D models emphasize the home versus host base dichotomy (to a lower or higher degree depending on the R&D model). Characteristically, the traditional R&D models are also restricted to the triad nations and are characterized by a lower degree of specialization than the metanational R&D organization. These differences apply to all traditional R&D models in comparison to the metanational R&D organization. 62 The metanational R&D organization also differs in its leveraging of the technological hierarchy. Critical knowledge development takes place in the whole R&D organization, the locus of knowledge is anywhere in the R&D organization and the source of knowledge is internal and external in nature. In contrast, traditional R&D models develop critical knowledge mostly at headquarters and at key R&D subsidiaries. As a result, the locus of knowledge is residing at headquarters and at key R&D subsidiaries and its source is mostly internal. After proposing this framework for the metanational R&D organization and discussing its differences in comparison to the traditional R&D models, we shall empirically examine the extent to which existing R&D organizations present in the periphery have evolved into metanational R&D organizations. The periphery is a good location for examining metanational R&D organizations because in contrast to traditional R&D models, a metanational R&D organization is also present in the periphery and taps into knowledge residing in the periphery. Singapore, as the context of periphery, is well suited because it has a long history of MNE investment and has evolved from being a manufacturing base into a base where high value added activities are carried out, e.g. R&D activities. Equally importantly, Singapore is still at the periphery, i.e. not part of the triad nations, and has little tradition in carrying out R&D activities. The following section elaborates on the operationalization of the major variables for the classification of the R&D subsidiaries under investigation. 5.1.2 Operationalization of Major Variables 5.1.2.1 Leveraging of Technological Hierarchy The R&D subsidiaries under investigation are classified according to the conceptual framework. This is done by analyzing the illustrated dimensions, namely leveraging of the technological hierarchy and number of knowledge bases. The leveraging of the technological hierarchy is measured according to the R&D subsidiaries’ relationships with other R&D subsidiaries and R&D headquarters. In the R&D organization, three types of relationships (tie modalities) have been distinguished (Vereecke, Van Dierdonck and De Meyer, 2002: 9-12). First, relationships of human resources refer to 63 the human resources flow between the different R&D sites and with headquarters. More specifically, the human resources flow refers to the development of critical R&D personnel in the internal R&D organization. It shows if such R&D personnel is also developed in the periphery. In the in-depth interviews, the R&D managers were asked about the composition of their R&D personnel (number of personnel at technician level versus research scientist/engineer level). Technician level refers to R&D personnel holding a bachelor’s degree, whereas research scientist/engineer level refers to R&D personnel holding a master’s degree or a PhD. If more than half of the R&D personnel belonged to the second category, it was assumed that critical R&D personnel were developed in the periphery. This assumption is similar to the reasoning of Deeds, DeCarolis and Coombs (2000) who state in another context that the number of R&D personnel holding a PhD (or master’s degree) as a percentage of the management team can be an important performance indicator. Here it is not seen as a performance indicator, but as an indicator for critical human resources development at subsidiary level. Furthermore, the R&D managers were asked to what extent their R&D subsidiary has influence on the acquisition and development of human resources and to what extent they conduct training for their R&D personnel (see survey in appendix). Overall, these items attempt to reflect to what extent development of critical R&D personnel occurs in R&D subsidiaries in the periphery. Second, the innovation configuration is examined to investigate how far the locus of innovation is at the R&D subsidiary, at headquarters and/or at other R&D sites, namely whether R&D sources are at the subsidiary R&D site or whether core technologies are transferred for further development to the respective R&D subsidiary. This will give an indication of the primacy of the home base. In the in-depth interviews, respondents (R&D managers/directors or managing directors) were asked to indicate to what extent they are able to participate actively in the global R&D program, to what extent they are recipients of core technologies from the home base and to what extent they can initiate own R&D projects. Moreover, the respondents indicated to what extent they conduct R&D activities in a field where headquarters or other R&D sites have no expertise and to what extent the innovation locus in their R&D organization is equally balanced (also see survey in appendix). Overall, these items attempt to reflect the innovation configuration. 64 Third, the degree of freedom (autonomy) the R&D subsidiary enjoys is examined. Respondents were asked to indicate to what extent their R&D site has to follow rules and regulations by headquarters, to what extent their R&D site can engage freely in external research collaborations and to what extent their R&D site can interact freely with other R&D sites (also see survey in appendix). Overall, these three measures attempt to give an indication of leveraging the technological hierarchy. 5.1.2.2 Number of Knowledge Bases The number of knowledge bases is determined partly by the interviews and partly by corporate archival research. The answers obtained from R&D managers and directors at subsidiary level were complemented by corporate archival research (annual reports, corporate web pages) documenting their type of R&D organization and the number of different knowledge bases of the corporate R&D organizations. In general, the interview partners indicated the number of knowledge bases. Data on the number of knowledge bases was also found through corporate archival research. The number of knowledge bases is considered small when the knowledge base is essentially located at headquarters. The number is medium when the knowledge base is located at headquarters and a few key R&D subsidiaries. An R&D organization with a large number of knowledge bases has R&D locations in all critical knowledge clusters, also in non-traditional R&D locations. The following exhibit summarizes the classification scheme for the various types of R&D organizations: 65 Exhibit 15: Classification Scheme for Different International R&D Organizations Type of R&D organization Leveraging of Technological Hierarchy Critical human Innovation Degree of resources freedom Locus also at Number of Knowledge Bases development at R&D subsidiary (autonomy) of R&D R&D subsidiary subsidiary Ethnocentric No No Low Small Geocentric No Yes4 Low Small R&D Hub Yes Yes Medium Medium Yes Yes High Medium Polycentric Maybe Maybe High Medium Metanational Yes Yes High Large Model Integrated R&D network R&D organization Source: Author An R&D organization would be classified as an ethnocentric R&D organization if the number of knowledge bases is small, in this case only at headquarters. Logically, the leveraging of the technological hierarchy is low. Critical human resources development does not take place at R&D subsidiary level. The innovation locus is at the home base and hence not at the R&D subsidiary. Due to the primacy of the home base, the degree of freedom of the R&D subsidiary is low. The geocentric R&D organization is also present at headquarters only. But in this model, different international R&D collaborations are entered into. Therefore, the innovation locus is not entirely at headquarters. Thus, leveraging of technological hierarchy is higher than under the ethnocentric model. 4 Even though the geocentric R&D organization does not have international R&D subsidiaries, it does engage in international R&D collaborations. Therefore, the innovation locus is not only at headquarters. 66 The R&D Hub model, the integrated R&D network as well as the polycentric R&D organization are all present in the triad nations. They differ with regard to their ability to leverage the technological hierarchy. By contrast, a metanational R&D organization is present in a large number of knowledge clusters and is very good at leveraging the technological hierarchy. 5.1.2.3 R&D performance In order to analyze the performance implications of the different international R&D organizations, which have been discussed, different R&D performance measures are discussed in the following: From an organizational perspective a measure of R&D output refers to criticality, substitutability, interaction, and immediacy (Brockhoff, 1998: 76-77). In this context, criticality reflects the degree to which the overall success of the organization depends on the work of the individual laboratory. The term criticality shows the possibility of substitution of work done in an R&D unit. The higher this variable, the lower the positive output of this R&D unit. Intense interaction of one R&D unit with other R&D units might increase the innovation output of the respective R&D unit. Finally, immediacy refers to the time lag between the stoppage of work in a laboratory and the cessation of work in the whole organization, that is, to what extent a stoppage by the R&D unit constitutes a block for the rest of the organization. These concepts are relatively hard to measure due to their abstract nature and are more useful in theoretical discussions than in an empirical study. Other R&D measures include its efficiency (to do things right) and its efficacy (to do the right things). Efficiency refers to the use of R&D resources in a favorable relation between costs and usefulness, whereby efficacy refers to R&D projects which contribute to overall firm goals. In other words, low efficacy means a high share of money being wasted for R&D (von Boehmer, 1995: 106-107). The major disadvantage of these types of measurements is that these measures are exposed to considerable subjective evaluation by the R&D manager. This might lead to biased results. 67 An alternative measure of R&D output is the number of drugs in development or in the ‘pipeline’. In general, one distinguishes between the number of products at stage I, II and III and the number of products approved for sale (Coombs and Deeds, 2000: 238). The strength of a product pipeline ensures future cash flows and increases the likelihood of firm survival. Hence, this measure of R&D output indicates the future potential value of a firm. Obviously, a product pipeline does not necessarily translate one for one into innovative products. Moderator variables, such as the firm’s ability to commercialize these products effectively, can play a significant role. One possible solution to this problem would be to consider only the products in stage III of the pipeline since they have the highest probability of being translated into innovations. This measure, however, applies solely to certain industries such as the pharmaceutical industry. A further R&D performance measure refers to intellectual human capital, which can be measured as the number of Ph.D.s and/or master’s degrees in sciences as a percentage of the management team (Deeds, DeCarolis and Coombs, 2000: 221). This performance measure is difficult to apply in this dissertation since the management of the R&D subsidiaries under investigation usually consists of only person who in nearly all cases holds a Ph.D. or a master’s degree. Other alternatives which reflect R&D performance can be found in the quality of the R&D personnel, for example in the number of star scientists. While the number of star scientists may reflect the quality of the R&D personnel, it may not translate into high R&D performance. This measure is not applicable in this dissertation since the R&D team at the R&D sites under investigation is usually too small to have a sizable number of star scientists. A further, frequently used measure is R&D expenditure, which can be found in annual reports and 10-K reports and is hence available for all public firms. However, R&D expenditure is a very crude measure since R&D expenditure does not necessarily translate into research performance. Knowing that the R&D expenditure is high does not necessarily mean that the R&D outcome is high. Hence, it is not clear what R&D 68 expenditure really measures. Another disadvantage of this measure is that it is almost impossible to obtain figures for R&D expenditure for private firms. Another performance measure refers to the number of innovative strategic actions undertaken by the R&D subsidiary, which could be found in public sources. However, there might be a reporting bias to present the R&D organization positively, so that size and impact of innovative strategic actions (quality vs. quantity) might be unclear. These actions might also apply for a limited time horizon only. The number of publications and citations is one more measure of R&D performance (Boutellier, Gassmann and v. Zedtwitz, 2000: 46). The number of publications is important since it reflects the degree of research output by the R&D site, but it does not necessarily reflect the quality of these publications. A further measure of R&D output can be seen in the introduction of new products and services. The definition of such new product or service developments, however, differs. Besides the definition problem, the link between new product developments and market performance might not be evident. Furthermore, not all new product developments have the same impact; some new products might have a strong market impact (i.e. blockbuster), while other new products might result in a minor impact. Hence, the market performance of the new product is not known or will only be known in lag time. Despite the weaknesses of this measure, however, it is undeniable that this measure is easily quantifiable: some authors consider the number of new product developments as a more appropriate measure than patents (Westhead, 1997: 47). The weaknesses and strengths of this measure are analyzed below. R&D output can also be measured by the number of patents applied for by a firm (Westhead, 1997: 47). Even though patent applications seem to be an obvious measure for innovation, such a criterion is not without drawbacks. First, patent applications leading to patented inventions do not automatically result in innovations (Coombs and Deeds, 2000: 238). Second, patent applications cannot measure all the innovative activities of a firm since not all intangible assets are patentable (Shan and Song, 1997). Third, the economic significance of inventions can vary greatly (Ernst, 2001: 145). 69 Fourth, not all patentable innovations lead necessarily to a patent application since there may be other possibilities of appropriating R&D earnings (Ernst, 2001: 145). Finally, the quality of patent applications is heterogeneous. However, patent applications constitute crucial indicators of innovative activity and as such of important technology positions. One study by Ernst (2001) examines the relationship between a firm’s patent applications and subsequent changes in its performance, concluding that patent application data is a major output indicator of R&D activities. The number of patent applications is important in the context of this dissertation because it reflects to what extent the R&D subsidiary is research based versus development based. The latter orientation would be reflected in the number of new product developments. This first part of the empirical findings examines the relationship between the type of organizational form and its performance implications, using the data of the 51 R&D subsidiaries under investigation, according to the R&D performance measures applicable, namely the number of patent applications and the number of new product developments. 5.1.3 Discussion of Results 5.1.3.1 Different Types of R&D Organizations Based on the dimensions of leveraging of the technological hierarchy and number of knowledge bases, the R&D subsidiaries under investigation have been classified (exhibit 16). The average number of patent applications per researcher and the average number of product developments per researcher as of year 2001 have been calculated for each type of R&D organization. The number of R&D organizations belonging to each category has been listed as well. The sample is 42 since the remaining R&D subsidiaries, namely 9, started their R&D operations only in 2002. 70 Exhibit 16: Classification of R&D Organizations Leveraging of Technological Hierarchy high Integrated R&D network (4) Metanational R&D Organization (1) Not applicable Ø no. of PA/R: 0.02 Ø no. of PD/R: 0.17 Geocentric R&D Hub (1) R&D organization (11) Ø no. of PA/R : 0.03 Ø no. of PD/R: 0.24 Pre-metanational R&D Organization (0) medium Ø no. of PA/R : 0.2 Ø no. of PD/R: 0.05 Ø no. of PA/R : 0 Ø no. of PD/R: 0.15 Ethnocentric Polycentric R&D Organization (23) R&D Organization (6) low Ø no. of PA/R : 0.08 Ø no. of PD/R: 0.19 Ø no. of PA/R : 0.02 Ø no. of PD/R: 0.08 small medium Highly dispersed R&D Organization (0) many Number of Knowledge Bases Ø no. of PA/R = Average Number of Patent Applications per Researcher on Subsidiary Level in Year 2001 Ø no. of PD/R = Average Number of Product Developments per Researcher on Subsidiary Level in Year 2001 () Number of R&D organizations of this type Source: Author 71 Exhibit 16 shows the different types of R&D organizations according to the classification scheme. Besides the six types of R&D organizations, three more R&D organizations have been identified. In the first case, the number of knowledge bases is small, but leveraging of technological hierarchy is high. This type of R&D organization is practically not possible. If present at headquarters only, an R&D organization can logically not be strong at leveraging the technological hierarchy internationally. While the geocentric R&D organization can do this to some extent due to its various R&D collaborations, this type does not seem to be practically possible. A highly dispersed R&D organization is logically conceivable. It would imply that the R&D organization is present both in the triad as well as the non-triad nations. Due to its low leveraging of technological hierarchy, any integration of R&D efforts is foregone. Even though theoretically conceivable, however, an R&D organization would forego even more than in the polycentric model any benefits resulting from a large number of knowledge bases and would probably be highly inefficient. One can imagine its temporary existence after, for example, a corporate merger. But it does not seem to be sustainable. The third ‘new’ type of R&D organization can be seen as a precursor to a metanational R&D organization. This model is present in all critical knowledge clusters. In contrast to the highly dispersed R&D organization, this type of R&D organization has learnt to leverage the technological hierarchy to a medium extent. Therefore, it is considered as a pre-form of a metanational R&D organization. The metanational form of R&D activities occurs only rarely in our sample. There was in fact only one metanational R&D organization with corresponding performance data; the other two metanational R&D organizations did not have corresponding performance data. The rare occurrence of this type of R&D organization is the case despite the factors speaking in favor of a metanational R&D organization, factors which are the erosion of competitive advantage from simplification through imitation and expansion of the opportunities for profitable knowledge orchestration. Such a profitable knowledge orchestration is made possible by the emergence of new knowledge clusters around the world combined with improvements in the range and 72 efficiency of connectivity mechanisms available (Doz et al., 1997: 20-21). That only few R&D organizations show the characteristics of a metanational R&D organization may partly be so because the R&D function differs considerably from other corporate functions. Due to the sensitive function of R&D (fear of imitation or diffusion of knowledge), high value added activities such as R&D are still mostly performed in the triad nations, not in the periphery. This dominance of the home base, plus the perceived unimportance of R&D subsidiaries in the periphery and the notion that local adaptations can only be applied locally, are major barriers to creating a metanational R&D organization (Doz, Santos and Williamson, 2001: 49-51). The primacy of the home base is reflected by the fact that still most high value added activities are performed in the home base. The periphery is still perceived more as a location where corporate functions such as production or marketing are performed, but is not seen as an R&D hub. And finally, the importance of local applications beyond the local or regional market may not be taken into account by R&D organizations. The barriers described apply more notably to research than to development. In development, they are less pronounced. This is also reflected in a higher degree of internationalization for development than for research activities of a firm (von Zedtwitz and Gassmann, 2002: 573-575). How these barriers can be overcome and a metanational advantage be obtained in an R&D organization are explained in more detail in the case study on Novartis’ Institute for Tropical Diseases. 5.1.3.2 Exploratory Performance Implications As can be seen from exhibit 16, the number of patent applications per researcher is highest for the geocentric R&D organization, followed by the ethnocentric R&D organization, by the metanational and then by both the integrated R&D network and the polycentric R&D organization. In terms of the number of new product developments, the metanational R&D organization scores highest, followed by the ethnocentric R&D organization, the integrated R&D network, the R&D hub, the polycentric and geocentric R&D organization. Overall, the number of new product developments per researcher is higher than the number of patent applications per researcher. These differences might 73 indicate that technological capabilities at subsidiary level are more focused on product development rather than on the number of patent applications. This in turn might also imply that non-traditional R&D locations need to reinforce their research efforts to move from being a development location to becoming a research hub. These differences may also be due to various other factors, such as the patenting strategy of the respective R&D organization or type of industry. The analysis of this study shows that it is important to distinguish between research and development in terms of performance. While the geocentric R&D organization seems to be strong in patenting activity, the ethnocentric and the integrated R&D network, the R&D hub and the metanational R&D organization show an important number of new product developments. Within these four R&D organizations, which are strong in the number of new product developments, it is important to examine whether there is a difference from one industry to another. This reasoning comes from the fact that roughly one third of the R&D subsidiaries in the sample, namely 12, belong to the electronics industry. The rest belong to other industry sectors such as the biomedical sciences, chemical and food industry. It is argued that due to the nature of the electronics industry, the R&D organizations of metanational, hub and integrated network would be more suited to developing new products than an ethnocentric R&D organization for R&D organizations in the electronics sector. This might be the case because the knowledge in the electronics industry, in particular knowledge about product developments and applications, is more dispersed than it is in the pharmaceutical sector for instance. Doz, Santos and Williamson (2001) emphasize this dispersion of knowledge in the electronics sector by presenting the case of ST Microelectronics, a firm which was only able to sustain its competitiveness by combining dispersed knowledge in order to create a new semiconductor chip. Based on this reasoning, the following regressions were run (also see Emory and Cooper, 1991: 629). First, the number of new product developments versus the type of 74 R&D organization (ethnocentric organization versus the metanational, hub, and integrated R&D network taken together), for the electronics industry only. Second, the same regression was run for all R&D subsidiaries not pertaining to the electronics industry. Results are presented below: 75 Exhibit 17: Regression Results on Number of New Product Developments versus Ethnocentric and Meta/Hub/Integrated R&D Organizations (Electronics Industry only) Variables Entered/Removed(b) Variables Model 1 Variables Entered Removed Method . Enter SIZE, METHUBIN, ETHNO(a) a All requested variables entered. b Dependent Variable: PRODEV Model Summary Adjusted R Std. Error of Model R R Square Square the Estimate 1 .962(a) .926 .898 22.32723 a Predictors: (Constant), SIZE, METHUBIN, ETHNO ANOVA(b) Sum of Squares df Mean Square F Sig. Regression 49782.376 3 16594.125 33.288 .000(a) Residual 3988.041 8 498.505 Total 53770.417 11 Model 1 a Predictors: (Constant), SIZE, METHUBIN, ETHNO b Dependent Variable: PRODEV Coefficients(a) Standardized Unstandardized Coefficients Model Coefficients t Sig. 2.358 .046 -.171 -1.518 .168 25.247 .951 9.125 .000 -.153 -1.323 .222 B Std. Error (Constant) 32.328 13.710 ETHNO -22.959 15.127 METHUBIN 230.385 -20.713 15.658 Beta 1 SIZE a Dependent Variable: PRODEV Source: Author 76 As can be seen from exhibit 17, R squared in our model is 0.926, that is 92.6% of the variation of the dependent variable, that is the number of new product developments, is explained by the regressor, namely the type of R&D organization. The regression results are significant at p < 0.01. The variable ethnocentric R&D organization shows no statistical significance, however, the variable of meta, hub and integrated R&D network (‘methubin’) showing statistical significance at p < 0.01. The F value of our model exceeds the critical F value at p < 0.01. These findings are a first step towards suggesting that it is important for R&D organizations in the electronics industry to be organized either as a metanational, hub or integrated R&D network since these organizational forms have a positive impact on the number of new product developments. The findings imply as well that an ethnocentric R&D organization is ill suited for delivering a large number of new product developments. In the regression, the variable ethnocentric R&D organization is statistically insignificant (p at 0.168). The same regression was run for the R&D organizations in the other industries. Results of this second regression are presented below: 77 Exhibit 18: Regression Results on Number of New Product Developments versus Ethnocentric and Meta/Hub/Integrated R&D Organizations (Other Industries) Variables Entered/Removed(b) Variables Model 1 Variables Entered Removed Method . Enter SIZE, ETHNO, METHUBIN(a) a All requested variables entered. b Dependent Variable: PRODEV Model Summary Model 1 Adjusted R Std. Error of R R Square Square the Estimate .748(a) .559 .508 5.80350 a Predictors: (Constant), SIZE, ETHNO, METHUBIN ANOVA(b) Sum of Model 1 Squares df Mean Square F Sig. Regression 1110.545 3 370.182 10.991 .000(a) Residual 875.697 26 33.681 Total 1986.242 29 a Predictors: (Constant), SIZE, ETHNO, METHUBIN b Dependent Variable: PRODEV Coefficients(a) Standardized Unstandardized Coefficients Model Coefficients t Sig. -.133 .895 .451 3.186 .004 3.429 .078 .545 .591 .588 4.444 .000 B Std. Error (Constant) -.243 1.827 ETHNO 7.342 2.304 METHUBIN 1.868 12.838 2.889 Beta 1 SIZE a Dependent Variable: PRODEV Source: Author 78 As can be seen from exhibit 18, R squared in our model is 0.559, that is 55.9% of the variation of the dependent variable, that is the number of new product developments, is explained by the regressor, namely the type of R&D organization. The regression results are significant at p < 0.01 The variable ethnocentric R&D organization shows statistical significance at p < 0.05. The variable of meta, hub and integrated R&D network (‘methubin’) show no statistical significance. The F value of our model exceeds the critical F value at p < 0.01 These findings are a first step towards suggesting that for the R&D organizations in the other industries, it is more favorable to be organized in the ethnocentric form. The types of metantional, hub and integrated R&D network do not contribute to the R&D performance in terms of the number of new product developments. Thus, our observation that it is important to be organized as either the metanational, hub or the integrated R&D network for the electronics industry can be confirmed. These results refer to the number of new product developments. These performance implications constitute exploratory performance implications. Further studies are needed to confirm these findings, in particular studies with a larger sample. Given the barriers described for a metanational R&D organization, it is not surprising that only few R&D organizations reach the state of a metanational R&D organization. The findings therefore suggest that few firms utilize and leverage knowledge residing in the periphery, especially with regard to R&D. The periphery can raise its importance through increasing its level of technological sophistication. How such an evolution in the form of technological capability upgrading can take place is analyzed in the second part of this chapter. This will give a more precise indication of the leveraging of the technological hierarchy from the perspective of the periphery. The following exhibit shows the framework of a metanational R&D organization and how the subsequent findings of this chapter contribute to this framework. 79 Exhibit 19: Framework for a Metanational R&D Organization: Implications for the Periphery Metanational R&D Organization R&D Sub. R&D Organization Level R&D Sub. R&D HQ Internal R&D network linkage Technological capability upgrading of the R&D subsidiary R&D Sub. R&D Subsidiary Level External R&D network linkage Source: Author While the first part of the empirical findings analyzed different international R&D organizations and proposed a framework for a metantional R&D organization, the second part examines technological capability upgrading at the R&D subsidiary level. This logic of structure follows because only through technological capability upgrading are R&D subsidiaries in the periphery capable of playing an important role in the overall corporate R&D organization. Therefore, the process of this technological capability upgrading is analyzed in more detail. Such an investigation is important in 80 order to understand why non-traditional R&D locations are still at the periphery, which means they are not yet part of the group of more advanced R&D locations, and how they can progress further in their technological capability upgrading. In this process, the impact of both variables of internal and external R&D network linkage is examined (see exhibit 19). 5.2 Quantitative Empirical Findings: R&D Internationalization Process 5.2.1 Conceptual Framework 5.2.1.1 Technological Capability This section attempts to provide a definition for the key terms of this second part of the quantitative empirical findings, namely technological capability and technological capability upgrading. There is no standard definition of these terms and they are not used only in an R&D domain (see for instance Lall, 1992 and Coriat and Dosi, 2002). Lall (1992) has developed an illustrative matrix of technological capabilities. He distinguishes between routine, adaptive and innovative technological capabilities (Lall, 1992: 167). Dahlman and Westphal (1982) use the term “technological mastery”. Technological capability is also defined as “skill, knowledge, and experience required for a firm to achieve technological change at different levels” (Costa and de Queiroz, 2002: 1433). It is thereby assumed that technological capabilities are accumulated over time as the firm seeks to undertake technological tasks. Following the definition of Costa and De Queiroz (2002), two types of technological capabilities are distinguished: functional and meta-technological capabilities. Within functional technological capabilities, these authors further distinguish between operational, improvement, and generation capabilities. While operational capabilities refer to an efficient functioning of productive activities, improvement capabilities refer to the improvement of technologies from external sources. Finally, generation capabilities allow the firm to achieve original results (Costa and de Queiroz, 2002: 1434-1435). 81 Based on Amsden and Tschang’s (2003) R&D classification, it is argued here that the different stages of an R&D site incorporate these different types of capabilities. Since an R&D unit with manufacturing support has to ensure an efficient production process as its main function, the R&D site’s operational technological capability is strong. A similar logic applies to an advanced development unit and an exploratory development unit, where the focus is on improvement capabilities. Extant technology is further developed and improved. Finally, generation capabilities apply to an R&D site with a strong research focus (applied research, basic research or pure science). Generation of new technologies is important. It has become obvious that most R&D sites under investigation in this study attempt to obtain generation capabilities in tandem with meta-technological capabilities, that is, the capability to manage this technological capability upgrading process (Costa and de Queiroz, 2002: 1434-1435). It is assumed that a metanational R&D organization is capable of applying these capabilities to the different stages of the technological hierarchy in tandem with the meta-technological capabilities. 5.2.1.2 Technological Capability Upgrading Closely linked to the definition of technological capabilities is the definition of technological capability upgrading. Again, there is no standard definition of this term and different authors have a different understanding of the concept. Capability upgrading in general can be seen as entry into progressively more complex new activities. Technological capability upgrading can be regarded as increasing competence in more complex technologies (Costa and de Queiroz, 2002: 1433; Intarakumnerd, Chairatana and Tangchitpiboon, 2002: 1445). It can also be viewed as technical learning, as the “creation of a technical know how base, the accumulation of technical knowledge” (De Meyer, 1993: 112). This dissertation combines the views of Medcof (1997), Costa and de Queiroz (2002) and Amsden and Tschang (2003). Technological capability upgrading is defined as moving along the technological ladder indicated in exhibit 20. A higher level of technological capability is defined when a company has achieved the ability to perform a technological activity that it had not been able to perform before 82 (Figueiredo, 2002: 78), for instance when an R&D unit moves from simply supporting manufacturing to conducting development activities. While supporting manufacturing involves mostly operation capabilities, improvement capabilities are required for development functions. The framework of Amsden and Tschang (2003) has been chosen because they based their research on fieldwork in a late industrializing country, namely Singapore. Therefore, their R&D classification seems to be suitable to fit the late industrializing context. The work of Amsden and Tschang (2003) is complemented by that of Medcof (1997). While the former focuses on the differences within R&D, it does not consider the impact of R&D activities in a geographical perspective. So, in order to provide a more comprehensive framework of technological capability upgrading, it is important to include the work of Medcof (1997), who provides a systematic classification of R&D sites not only according to their technological level, but also according to their geographical output. Hence, exhibit 20 is derived from these two works of Medcof (1997) and Amsden and Tschang (2003). 83 Exhibit 20: Framework for Technological Capability Upgrading at R&D Subsidiary Level5 Technology Ladder (local, regional and global scope) Pure Science (S) Basic Research Unit (R2) Applied Research Unit (R1) Generation Capabilities Exploratory Development (D2) Advanced Development (D1) Improvement Capabilities Manufacturing Support (M2) Market Support (M1) Meta-Learning Capabilities Operation Capabilities Source: Author expanding Medcof (1997); Costa and De Queiroz (2002) and Amsden and Tschang (2003). The first two levels in the technology ladder apply to market and manufacturing support. According to Medcof (1997), market support is defined as customer support and/or the adaptation of already established product technology to particular customer requirements. This function (market support) is carried out by the R&D unit in collaboration with the marketing unit, without significant collaboration from manufacturing. In contrast to market support, manufacturing support refers to the adaptation of an already established process technology to some particular condition, usually to improve the manufacturing process. Manufacturing support is carried out in tandem by the R&D unit and manufacturing, but without significant cooperation from marketing (Medcof, 1997: 306). In late industrializing economies these two functions are crucial if the R&D site is to acquire more technologically complex capabilities (Amsden and 5 This chapter examines technological capability upgrading only. However, the author is aware of the fact that technological capability downgrading occurs as well. The objective of a late industrializing country, however, is to upgrade rather than downgrade the technological capabilities in order to increase its economic competitiveness. This is why technological capability upgrading is discussed in this dissertation. 84 Tschang, 2003: 566-569). These basic technological capabilities allow R&D subsidiaries to achieve higher levels of technological development (Figueiredo, 2002: 74-79). The next two levels in the technological ladder point to development and can be defined as the creation of new products and processes with commercial value by means of applying current scientific knowledge (Medcof, 1997: 306). Development is carried out by R&D units in collaboration with marketing and manufacturing units (Medcof, 1997: 306-308). Within development, two categories are distinguished (Amsden and Tschang, 2003: 555; Medcof, 1997: 306). Product development or advanced development refers to the development of manufacturable and commercially viable new products (Medcof, 1997: 306; Amsden and Tschang, 2003: 555). An advanced development unit aims to deliver immediate market results and is related to production. Techniques used include engineering design tools including simulation and testing. The R&D personnel has an engineer level (bachelor’s and/or master’s degree). A higher level of development is termed as process development or exploratory development, whereby the R&D unit aims to find a detailed product design or prototype in a system. Thus, the search is not focused on a new product, but on a new and commercially viable process. The research objective is to implement this process as engineered system. An exploratory development delivers short-term market results and requires R&D personnel on an engineer level (bachelor’s and/or master’s degree). Techniques employed comprise engineering design tools including simulation, but not testing. Given these characteristics, exploratory development is therefore technologically more advanced than an advanced development unit (Amsden and Tschang, 2003: 555, 560). The next more sophisticated stage in the technological ladder is applied and basic research, carried out by R&D units without significant consultation with marketing or manufacturing. Applied research refers to the application of scientific techniques in order to find a differentiated product for a specific market. The objective is to transform and reapply a known concept for a new application. The output is a 85 differentiated product for a specific market with intellectual property, which is created in the medium and short-term. The R&D personnel is well trained, experienced and has a bachelor’s, master’s degree or a PhD and applies scientific techniques (Amsden and Tschang, 2003: 555). Basic research refers to the discovery of new knowledge for new marketable products. In contrast to applied research, it seeks applications that are unknown or diffuse. R&D units conducting basic research adopt a long-term horizon. The output is product-based research for transfer to applied research or exploratory development. Scientific techniques are applied by highly qualified R&D personnel. Finally, pure science refers to the search for intrinsic knowledge in order to uncover new scientific principles (Medcof, 1997: 306-307; Amsden and Tschang, 2003: 555). This type of R&D is, however, performed by universities or other public sector institutions, and is usually not performed by the private corporate sector. The different levels of technological sophistication are presented below: Exhibit 21: Level of Technological Stages Technological Stage Market Support Manufacturing Support Advanced Development Exploratory Development Applied Research Basic Research Pure Science Level 1 2 3 4 5 6 7 Source: Derived from Amsden and Tschang (2003: 555), Medcof (1997: 306) Besides this technological ladder, which shows the technological complexity an R&D site faces, it is also important to know the geographical area which the R&D site targets through its research activities in order to evaluate its significance from this 86 perspective as well. Thus, it is attempted to provide a more comprehensive understanding of technological capability upgrading. In addition to the different technological functions, their geographical scope, namely local, regional, and global scope, is distinguished. Local scope refers to R&D output, which is targeted specifically for Singapore. Regional scope refers to R&D output for the Asia-Pacific region. Logically, a global scope implies an R&D output applicable to the global market. Since the Singapore market is very small, the local scope is obviously not of much significance. An example for regional innovations (regional R&D scope) refers to medicines against diseases pertinent to the Asia-Pacific region only. An example for a global output would be a product development used globally, for instance the hardware for cell phones. This scope dimension is included in the framework in exhibit 20. Even though the suggested framework presents different technological capabilities according to different levels or stages, it does not presume that all firms will necessarily build up capabilities in a linearly sequenced process, or start and end at the same stages. Most often, R&D subsidiaries are at different technological stages at the same time, and some of them being closely interconnected. Different R&D sites with different technologies would adopt different sequences, depending on various factors such as industry participation. Obviously, the suggested framework is not without drawbacks; it does, however, attempt to provide a first step in analyzing technological capability upgrading of the R&D function6. The next section presents empirical evidence on the levels of technological capabilities of the R&D subsidiaries under investigation. 6 Previous studies focused mainly on technological capability upgrading in general, but not specifically on the R&D function. 87 5.2.2 Levels of Technological Capabilities of R&D Subsidiaries The following exhibit shows the number of R&D subsidiaries at different technological stages according to the conceptual framework. It also presents the average year of establishment of the specific technological stage. Since the different R&D subsidiaries experience or are going through different technological stages during their evolution, the number of R&D subsidiaries naturally does not add up to the overall number of R&D subsidiaries, namely 61. For instance, one R&D subsidiary experiences the technological stages of manufacturing support and advanced development. Therefore, the number of technological stages is two for one R&D subsidiary. Consequently, the number of technological stages does not correspond to the number of R&D subsidiaries because each R&D subsidiary usually goes through more than one technological stage (also see section C in the survey in the appendix). The following exhibit shows the industry breakdown of the different technological stages of the R&D subsidiaries and the average year of establishment of the technological stage. 88 Biomedical Sciences - - 0 (0%) Pure Science 1997 1994 1993 0 (0%) 4 (11%) 4 (11%) 12 (33%) 1988 Basic Research Research Applied Development Exploratory Development Advanced Support Manufacturing 11 (31%) 1991 5 (14%) Market Support Year of total) Stages 8 7 0 (0%) 0 (0%) 1 (5%) 2 (10%) 9 (45%) 2 (10%) 6 (30%) of total) - - 1999 2000 1997 1997 1994 Year 0 - 2002 2000 1998 1998 1991 1995 (% of total) Source: Author 1 (3%) 4 (14%) 5 (18%) 10 (36%) 3 (11%) 5 (18%) of total) 0 (0%) 1 (5%) 2 (13%) 2 (13%) 5 (31%) 2 (13%) 4 (25%) of total) Number (% Other Industries - 1998 1998 1997 1999 1996 1995 year 0 (0%) 0 (0%) 1 (5%) 2 (10%) 7 (33%) 8 (38%) 3 (14%) of total) - - 1990 1991 1996 1996 1998 year Average Number (% Average Chemical Industry 0 (0%) 2 (2%) 12 (10%) 15 (12%) 43 (36%) 26 (21%) 23 (19%) of total) Number (% - 2000 1997 1996 1997 1994 1995 year Average MNEs overall 89 Number refers to the number of technological stages; % of total refers to the percentage of technological stages at a particular stage of the total number of technological stages. 8 Average year refers to the average year of establishment of a particular technological stage. 7 IT Industry Technological Number (% Average Number (% Average Number (% Average year Electronics Industry Exhibit 22: Industry Breakdown of Different Technological Stages Followed by R&D Subsidiaries of MNEs (1995-2002) As can be seen from exhibit 22, the biomedical sciences, chemical and electronics industry have a relatively high share in applied research (14%, 13% and 11% respectively). About one third of the R&D subsidiaries are at the stage of advanced development (36%). Overall, 76% of them are at the technological stages of advanced development, manufacturing support and market support. The findings suggest that the first type of technological stage entered into is manufacturing support. This type usually resulted from the manufacturing base the MNE had in Singapore. The average year of establishment of this technological stage is 1994. The other technological stages were established slightly later: 1995 for market support, 1997 for advanced development. Only a few R&D subsidiaries conduct applied and most notably basic research. Therefore, the ‘R’ of R&D is not conducted by a large number of R&D subsidiaries of MNEs. It seems, however, that the research function is gaining in importance. For example, two technological stages indicated by the R&D subsidiaries in the sample refer to basic research with the average year of establishment in 2000 and 12 technological stages are at the level of applied research with the average year of establishment in 1997. The breakdown by nationality of the MNEs shows that R&D subsidiaries of European MNEs have a higher level of technological sophistication compared to other MNEs as of the year 2002. 40% of R&D subsidiaries of European MNEs are at the technological stages of applied and basic research, whereas 22% of Japanese R&D subsidiaries and 12% of American R&D subsidiaries conduct applied research. Correspondingly, 76% of US R&D subsidiaries and 66% of Japanese R&D subsidiaries conduct advanced development compared to 45% of European R&D subsidiaries in 2002. The breakdown by local versus expatriate management shows that it is usually expatriate management which is entrusted with technologically more advanced activities. Under such expatriate management 30% of these R&D subsidiaries conduct applied or basic research. This figure compares to 22% for local management. 74% of R&D subsidiaries under local management conduct advanced development as opposed to 46% doing so in R&D subsidiaries under expatriate management. 90 While this analysis shows different factors impacting the level of technological capabilities in R&D subsidiaries of MNEs, it is also interesting to compare these R&D subsidiaries with local R&D subsidiaries. The next exhibit compares the level of technological capabilities of R&D subsidiaries of MNEs with local R&D sites. Exhibit 23: Comparison of Technological Levels of R&D Subsidiaries of MNEs versus Local R&D Subsidiaries (1995-2002) MNEs Overall Local R&D Subsidiaries Technological Number of Average year of Number of Average year of Stages technological stages establishment of technological stages establishment of (% of total) technological stage (% of total) technological stage 23 (19%) 1995 7 (23%) 1997 26 (21%) 1994 8 (27%) 1993 43 (36%) 1997 11 (37%) 1997 15 (12%) 1996 3 (10%) 1998 12 (10%) 1997 1 (3%) 1996 Basic Research 2 (2%) 2000 0 (0%) - Pure Science 0 (0%) - 0 (0%) - Market Support Manufacturing Support Advanced Development Exploratory Development Applied Research Source: Author The findings summarized in this exhibit show that the local R&D sites have slightly weaker technical capabilities than the R&D subsidiaries of MNEs. Their technological stages are usually situated at advanced development, manufacturing and market support (overall 87%). None of them conducts basic research and a very small number do applied research (3%). With regard to the average year of establishment of the technological stage, there is not much difference between R&D subsidiaries of MNEs and local R&D sites. After illustrating different levels of technological capabilities of the R&D sites under investigation, the next section focuses on the technological paths these R&D 91 subsidiaries have followed. While the technological levels give a comprehensive, but rather static, picture of technological capabilities in the R&D subsidiaries under investigation, technological paths show the technological evolution these R&D subsidiaries have made and thus provide a more dynamic picture. 5.2.3 Typology of Technological Paths of R&D Subsidiaries 5.2.3.1 Derivation of Technological Paths Based on the in-depth interviews with the R&D mangers in this study and based also on the conceptual framework, different technological paths have been identified in order to understand R&D subsidiary evolution. The following exhibit shows the technological paths identified: Exhibit 24: Typology of Technological Paths of R&D Subsidiaries Characteristics Technological Path I Technological Path II Technological Path III (TP 1) (TP2) (TP3) Geographical Scope at path beginning Global Local or regional Local Geographical Scope at further stages Global Global Regional Pace Sequential or exponential Sequential or exponential Sequential or exponential Possible Technological Sequences M2 M2, D1 M2, D1, D2 M2, D1, D2, R1 M2, D1, D2, R1, R2 or any subsequence within these technological stages M1, M2 M1, M2, D1 M1, M2, D1, D2 M1, M2, D1, D2, R1 M1, M2, D1, D2, R1, R2 or any subsequence within these technological stages M1, M2 M1, M2, D1 M1, M2, D1, D2 M1, M2, D1, D2, R1 M1, M2, D1, D2, R1,R2 or any subsequence within these technological stages Source: Author As can be seen from the exhibit, three different technological paths have been identified. They are described in more detail in the following: 92 5.2.3.1.1 Technological Path I Technological path I denotes a technological path where the R&D subsidiary’s output is targeted for the global market from the beginning of the R&D subsidiary’s evolution and in subsequent technological stages of the R&D subsidiary. This means that the R&D subsidiary is not responsible for local or regional functions, but delivers to the global market during its entire evolution. The time sequence refers to the time differentials between the different stages in the technology ladder, which are directed towards the global market. While a sequential global path is characterized by equal time differentials between the different technological stages, an exponential path refers to progressively shorter time intervals in relation to the degree of technological sophistication reached. This technological path can entail several different technological stages: M2, D1, D2, R1, R2. M1, market support, is not a technological stage for this type of technological path. This is the case because market support is defined as customer support and/or the adaptation of already established product technology to particular customer requirements. These specific customer demands usually refer to a specific market on a local or regional level. Such a market function performed by R&D subsidiaries was a major market determinant in the earlier R&D internationalization (see chapter 2). Since this technological path has a global scope, the technological stage M1 does not apply. In general, the different technological functions could have been followed in complete sequence (M2, D1, D2, R1 and R2) or in partial sequence (subsequence), for instance D1, D2. This also implies that some technological functions can be left out. For example, a sequence of D1 and R1 is possible as well. Such a sequence can be the case when the R&D site did not evolve from a manufacturing base, but was established as a greenfield R&D site from its beginning. How technological path II differs from technological path I is explained below: 93 5.2.3.1.2 Technological Path II and Technological Path III In contrast to the technological path I, this type of R&D subsidiary is mostly concerned with supporting manufacturing and adapting products and processes to a local or regional market (technological stages of M1 and M2) in the beginning of its evolution. Therefore, this technological path starts with M1 or M2. The R&D activities result either from market support or a manufacturing base or both. With an increase in the technological level (technological stages of D1, D2, R1, R2), the output of the R&D subsidiary reaches beyond the local or regional scope for a global market. Within the different technological stages (M1, M2, D1, D2, R1, R2), any subsequence can be followed. It is also possible that not all technological stages are undergone by the R&D subsidiary. For instance, the R&D subsidiary can follow D1, D2, R1. The time differentials between the different stages can be sequential or exponential as in technological path I. Technological path III is equal to technological path II, but with a difference in the geographical scope of the R&D output. While the R&D subsidiary’s output is mostly local at the beginning, it becomes regional in the evolution of the R&D subsidiary. However, this R&D subsidiary does not reach the global level. The other characteristics apply as for technological path II. The conceptual framework has been defined and the different technological paths have been derived. The next section develops propositions. Singapore’s science and technology policy was discussed in chapter 4 and serves as a basis for understanding the late industrializing context in which the technological capability upgrading at the R&D subsidiary level under discussion takes place. 94 5.2.3.2 Performance Implications of Technological Paths While the section on the conceptual framework has derived different technological paths, it is not clear what performance implications these technological paths entail. Misaligning technological capability upgrading with the global R&D organization in terms of time sequence and output level may reduce the profitability to be gained from these technological competencies abroad and may increase operational instability (Luo, 2000: 365). For instance, upgraded technological capabilities of foreign subsidiaries need to be in accordance with well-developed management systems. Only if upgraded technological capabilities can be integrated in the internal R&D organization are they of real value to the internal R&D organization. Such a technological capability upgrading cannot be an isolated process at R&D subsidiary level if it is to be effective. For an R&D subsidiary manager it is crucial to know how to impact positively the evolution of his/her R&D subsidiary. Questions that arise in this context are: Is it better to start on different technological stages at the same time or is it more advantageous to develop different technological stages sequentially? What geographical scope is optimal for the R&D subsidiary? It is also important to note that different technological stages imply different performance measures. While manufacturing support is targeted at the efficiency of the production process (in terms of cost reduction, quality of the product), the development function focuses on the number of new product developments. The higher end research activities, namely applied and basic research, focus on the number of patent applications and the number of publications. Given these qualifications, it seems very difficult to measure R&D performance. Besides the general disadvantages of the different R&D performance measures, R&D performance itself is also contingent on the technological level. If, for example, the number of patent applications is taken as a yardstick, but the R&D subsidiary’s main focus is the number of new product developments, the number of patent applications does not adequately reflect the R&D subsidiary’s performance. 95 Given the analysis of the technological capabilities in section 5.2.2, it is important to realize that the two R&D performance measures of the number of new product developments and the number of patent applications probably differ for the R&D subsidiaries under investigation. Since more R&D subsidiaries are at the technological stages of development than at research (see exhibit 22), one may suggest that the number of new product developments is higher than the number of patent applications for the R&D subsidiaries. The technological stages of research would rather emphasize the number of patent applications as an R&D output measurement. The following section attempts to develop propositions for the different technological paths and their performance implications. 5.2.3.2.1 Technological Path I If this technological path is followed, it implies that from the start of the R&D subsidiary one or usually more technological stages are entered into simultaneously or sequentially with the objective to serve the global market. These technological stages can be followed in sequence or subsequence as described above. According to the ‘organizational learning’ perspective in late industrializing countries, latecomer R&D organizations need to build up their learning system first (Figueiredo, 2002: 74). If the R&D site aims at a global scope from the beginning of its technological path, organizational learning requires both external as well as internal knowledge acquisition (Figueiredo, 2002: 73-74; Luo, 2000: 369). External knowledge acquisition would imply that the R&D subsidiary acquires knowledge about the global market in terms of R&D. Internal knowledge acquisition would involve continuous integration and sharing of knowledge within the R&D organization. Obvious advantages of this technological path for the R&D subsidiary are that the R&D subsidiary can establish a high strategic importance within the R&D organization from the beginning of its existence since it is not only considered as an R&D site with local scope, but as an R&D site with global relevance. The first stage, market support, is not 96 entered into because of the global scope of this R&D site. From the start, higher technologically functions are performed by this type of R&D site. This technological path, on the other hand, is very demanding for the respective R&D subsidiary. It requires strong R&D management skills and systems since the R&D output is measured according to global standards (see Tallmann, 1992). The site cannot evolve from performing local or regional functions to performing global functions. It is therefore assumed that the R&D site will need a period of adjustment to its global scope. Given this situation, it is to be expected that the R&D subsidiary will show low R&D performance initially. Performance will be stronger once it has overcome the initial hurdles of global adjustment at the beginning of its technological path. As noted previously, the number of new product developments is likely to be higher than the number of patent applications because more R&D subsidiaries in our sample are at the technological stages of development than research. Proposition 1 follows from this reasoning: Proposition 1: The relationship between technological path I and subsidiary R&D performance is of an increasing linear nature, R&D performance being higher for the number of new product developments than for the number of patent applications. 5.2.3.2.2 Technological Path II and Technological Path III This type of R&D subsidiary faces fewer challenges in the beginning of its technological path than those R&D subsidiaries, which follow technological path I, since at the beginning it serves a local or regional market. This technological path usually starts with market support tasks, adjusting existing technologies to specific customer requirements. Therefore, the R&D subsidiary is excluded from the main international sources of R&D at the beginning of its path (Hobday, 1995: 1172). In its subsequent evolution, however, this type of R&D subsidiary targets its R&D activities at a global scope. In order to reach this stage, both internal as well as external 97 knowledge needs to be acquired. Externally, knowledge about the mainstream international market the R&D subsidiary wishes to supply is critical. Internally, the increase in importance of the R&D subsidiary needs to be managed (also see De Meyer, 1993). Overall, these R&D subsidiaries adjust more sequentially to a global standard of their R&D output than in technological path I. At the beginning of their technological path, they perform M1 and M2 for a local or regional market and then follow a global technological path (with the functions of D1, D2, R1, R2 in sequence or subsequence). During this transition period, it may be difficult for the R&D subsidiary to show a strong performance. This is especially the case because the R&D subsidiary has to upgrade its technological capabilities from operation to improvement and/or generation capabilities. While operation capabilities focus on an efficient functioning of technological stages M1 and M2, improvement capabilities require development capacities in the R&D subsidiary. And generation capabilities require an even higher technological sophistication focusing on achieving original results (Costa and De Queiroz, 2002: 1434). Management issues during this transition period include achieving better quality standards and acquiring knowledge of more complex technologies. It is therefore assumed that during this transition period performance will decrease or stagnate before it increases again. The performance implications of technological path III are similar to those of technological path II. Again, the R&D subsidiary has to undergo a transition period to provide results mostly for a local market and then for a regional market. The same logic applies, with the difference that this transition is less difficult than the one following technological path II. This is the case because technological path III involves a transition from a local to a regional scope and not to a global scope. Proposition 2 follows: Proposition 2: The relationship between technological path II or technological path III and R&D performance is of a step-wise increasing nature, R&D performance being higher for the number of new products developments than for the number of patent applications. 98 5.2.3.2.3 Temporal Sequence of Technological Stages While different technological paths result in different performance implications, the technological paths also occur at a different pace. It is assumed that once the R&D subsidiary has gained experience in technological capability upgrading, the time differentials between the different evolutionary stages will be shorter (Grandstrand, 1999: 278; for the argument of experience also see Johanson and Vahlne, 1977: 25). More specifically, it is argued that the time differentials within the stages pertaining to one type of technological capabilities, be it operation, improvement or generation capabilities, are shorter than the time differentials between these stages. This reasoning is based on the argument that to achieve different types of technological capabilities is more difficult than to stay within one type of technological capabilities (Costa and de Queiroz, 2002: 1434, Lall, 1992: 167). The same type of technological capabilities requires similar competences. For instance, both the technological stages of advanced and exploratory development require development skills, at a product or process level. These development skills involve knowledge associated with creative development of technologies adopted (Costa and de Queiroz, 2002: 1434). The next technological stage, applied research, however, requires the application of scientific techniques in order to find a differentiated product for a specific market. Accordingly, an R&D subsidiary, for example, needs more time to move from exploratory development to applied research than from advanced development to exploratory development. Proposition 3 follows from this: Proposition 3: R&D subsidiaries which follow technological paths I, II or III are characterized by shorter time differentials within the same type of technological capabilities than between different types of technological capabilities. 5.2.3.2.4 Discussion of Results The analysis of different measures of R&D output in the first part of this chapter showed that no R&D performance measure is without drawbacks. Therefore, the performance implications of the different technological paths followed by 51 R&D subsidiaries should be measured according to several R&D performance indicators, in order to be able to draw more accurate performance implications. These measures 99 include the number of new product or service developments and the R&D subsidiaries’ patenting behavior. The propositions have all been tested against each of these performance measures, which means that for each technology path each performance measure was analyzed. The underlying assumption is that different performance measures apply to different technological stages in the evolution of an R&D subsidiary. The 51 R&D subsidiaries were classified according to technological path I, II and III. Based on this classification, the performance behavior of each technological path was examined along the dimensions of the average number of new product developments per firm and per year, the average number of patent applications per firm and per year for the years 1995-2002. This time frame was chosen since most R&D subsidiaries in Singapore have recently been established. The average year of establishment of the R&D subsidiary under investigation is 1994. Therefore, the time frame of 1995-2002 represents a comprehensive time frame. Exhibit 25 shows the performance behavior of technological paths I, II and III along the performance measures of the number of new product developments and patent applications. 100 Exhibit 25: R&D Performance Behavior of Technological Paths I, II and III 24 22 20 18 R&D Performance 16 14 12 Number of new product developments of TP II Number of new product developments of TP I 10 8 6 Number of patent applications of TP I 4 Number of new product developments of TP III Number of patent applications of TP II 2 0 1995 Number of patent applications of TP III 1996 1997 1998 1999 2000 2001 2002 Years Source: Author As can be seen in exhibit 25, the performance behavior of technological path I follows a more or less linear structure. The average number of new product developments per firm increases in general with a slight decline in the years 1997 and 1998 and increases in the years afterwards. The same is true for the performance measure of the average 101 number of patent applications per firm with a decline in 1997. As predicted, the number of new product developments is higher than the number of patent applications for technological path I. The decline in the performance measures in the years 1997-1999 may be explained by the Asian financial crisis (for a strategy framework during turbulence see Chakravarthy, 1997). During this period of instability, the R&D subsidiaries might have been less capable of continuing their linear performance due to number of possible reasons: First, their R&D budget might have been reduced or delayed due to the Asian financial crisis, which in turn restricted certain R&D activities and as such limited R&D output in terms of performance behavior. A second reason could be that due to the Asian financial crisis the technology strategy was revised and readjusted. This readjustment might have partly reduced the focus on performance and increased the focus on adjustment to the new environment. This period of instability may also have led to a stagnation of technological capability upgrading or technological downgrading may have occurred as well. The findings suggest that the linear performance continued with a period of stagnation and slight decline in R&D performance during the Asian crisis. Therefore, the empirical findings are consistent with proposition 1, stating that the R&D performance of technological path I is of a linear nature, while the number of new product developments is higher than the number of patent applications. The performance behavior of technological path II is characterized by a step-wise structure, especially regarding the number of new product developments; the years 1997-1999 reflect a certain stagnation in the performance behavior of the R&D subsidiaries before their performance increases again. Again, this stagnation may be explained by the Asian financial crisis. Several reasons could explain this stagnation. Like any crisis, the Asian crisis allowed readjusting and revising of the existing technology structure. This allows the R&D subsidiaries to adjust to new and competing technologies and to integrate new ideas in their processes. Since technological path II involves a transition from local/regional to global scope, the Asian crisis could have enabled some R&D subsidiaries to achieve a global scope from previously mainly 102 serving the local/regional market. After the Asian crisis and a possible readjustment to a global market, the performance behavior increases again. Therefore, proposition two can be confirmed. The findings further suggest that the performance behavior of technological path III is also characterized by a step-wise structure, whereby the years 1997-1998 seem to be a phase of stagnation. This period can again be explained by the Asian crisis. It is suggested that since this technological path includes an evolution from local to regional scope, these R&D subsidiaries are more affected by the Asian crisis than the R&D subsidiaries, which follow technological paths I or II. The absolute numbers of new product developments and patent applications are the lowest compared to technological path I and II. On a more positive note, however, one may put forward the notion that the Asian financial crisis enabled the R&D subsidiaries to reach beyond the local market to the regional market. This reassignment of the technology strategy to a regional market allows the R&D subsidiaries of technological path III to reach a stronger performance again. Therefore, proposition three can be confirmed. An important observation which can also be drawn from exhibit 25 is that the R&D subsidiaries in the sample were all able to increase their number of patent applications and their number of new product developments in the years 1995-2002. This is particularly so from the year 1999 onwards. This development may indicate that R&D activities intensified in Singapore and are moving in the direction of R&D activities as they are conducted in the triad nations. In terms of the number of new product developments, technological path II seems to be the best evolution for an R&D subsidiary. The number of new product developments for technological path I and III is lower than that of technological path II and similar in nature. Overall, R&D performance is lowest for technological path III. Technological path I is strongest in the number of patent applications. The findings suggest that a sequential R&D evolution (technological path II) is optimal with regard to R&D performance in terms of the number of new product developments. 103 Regarding proposition three, the time differences between the technological stages have been measured (for the different technological stages see exhibit 20). The respondents were asked in what year they moved from one technological stage to the next and consequently how long they stayed at each technological stage. In our sample, the 51 R&D subsidiaries needed on average around six months to move from market support to manufacturing support. It took them longer to reach the next level in the technological ladder, namely advanced development from previously being manufacturing support (3½ years). Within this next phase of development capabilities, the move from advanced development to exploratory development was shorter than two years (1 2/3 years). Again the next step to generation capabilities was 2 years, while the move from applied to basic research took only 1½ years (only two R&D subsidiaries indicated that they conducted basic research and none indicated that they carried out pure science research). These results are consistent with our proposition four. The move within the same type of capabilities seems to be easier than the move from one type of capabilities to the next. The move from operation capabilities to improvement capabilities seems especially difficult. Once the R&D subsidiary has gained improvement capabilities, it seems to be easier to achieve generation capabilities also. In addition to the analysis of the time differentials, the performance implications have been examined. Two groups of R&D subsidiaries were considered: those which need on average more time to move within and between the different technological capabilities and stages and those which need on average less time to move within and between the different technological capabilities and stages. The results are shown in exhibit 26 below: 104 Exhibit 26: Performance Behavior of R&D Subsidiaries (Fast and Slow Path Sequence) 30 28 26 24 22 Average number of new product R&D performance 20 developments per firm (Slow Path Sequence) 18 16 Average number of new 14 product developments per 12 Average number of patent applications 10 firm (Fast Path Sequence) per firm (Slow Path Sequence) 8 6 4 2 Average number of patent applications per firm (Fast Path Sequence) 0 1995 1996 1997 1998 1999 2000 2001 2002 Years Source: Author As can be seen from exhibit 26, the performance behavior of those R&D subsidiaries which follow a slower technological path than the average is stronger than those whose technological path follows a relatively fast path sequence, most notably in the number of new product developments. While a firm achieves about 30 new product developments a year in 2002 in the slower than average technological path, this number amounts to only 8 for the faster than average technological path in 2002. This finding is consistent with our finding that technological path II is most favorable in terms of the number of new product developments since the R&D subsidiaries which follow a slower technological path than average follow mostly technological path II. As a result, a sequential development of technological capabilities seems most favorable in terms of R&D subsidiary performance and in particular in terms of the number of new product developments. 105 After this analysis of the level of technological capabilities and proposing a typology of different technological paths and discussing its performance implications, the next section investigates key influencing factors on technological capability upgrading. While the level of technological capabilities gave an overview of the technological sophistication of the R&D subsidiaries in the sample, the typology of technological paths showed their evolution. It has been pointed out that internal and external knowledge acquisition is important for technological capability upgrading. These issues are analyzed more closely in the next section. Overall, an attempt is made to provide a comprehensive analysis of technological capability upgrading at subsidiary level in a late industrializing context. 5.2.3.3 Impact of Key Factors on Technological Capability Upgrading In order to tap into knowledge on a subsidiary level in the periphery and then to integrate this knowledge in the overall R&D organization, the R&D subsidiary needs to be both integrated in the internal R&D organization as well as in the external research environment. Both the internal R&D organization and the external research environment are important for an R&D subsidiary to upgrade its technological capabilities. In the following both the terms internal R&D network linkage and external R&D network linkage refer to the ties of the R&D subsidiary with the internal R&D organization and the external R&D environment. This dissertation does not intend to provide a general network definition. The terms used are specifically for an R&D context. 5.2.3.3.1 Role of Internal R&D Network Linkage Through the internal network linkage, the R&D subsidiary can gain critical knowledge for upgrading its technological sophistication. Internal R&D network linkage refer to the R&D site’s relationships with other internal R&D sites as well as headquarters. In the internal R&D network three types of relationships (tie modality) are distinguished (Vereecke, van Dierdonck and De Meyer, 2002: 9-12): critical human resources development, innovation configuration and degree of freedom (also see first part of this chapter). They reflect the degree to which the R&D subsidiary is embedded in the internal R&D organization. Or in other words, they reflect how far technological 106 leveraging takes place in the internal R&D organization. This internal network linkage and its level are discussed in more detail below: Exhibit 27: Internal R&D Network Linkage Level Internal R&D Network Linkage Linkage Level No impact by the R&D subsidiary on human resources development, innovation 1 and no autonomy Limited role of the R&D subsidiary in human resources development, innovation 2 and autonomy; Linkage with HQ only Basic role of the R&D subsidiary in human resources development, innovation 3 and autonomy; Linkage with HQ and other R&D subsidiaries Intermediate role of the R&D subsidiary in human resources development, 4 innovation and autonomy; Linkage with HQ only Active role of the R&D subsidiary in human resources development, innovation 5 and autonomy; Linkage with HQ and other R&D subsidiaries Advanced active role of the R&D subsidiary in human resources development, 6 innovation and autonomy; HQ only Highly advanced role of the R&D subsidiary in human resources development, 7 innovation and autonomy; Linkage with HQ and other R&D subsidiaries Source: Author extending Ariffin and Figueiredo (2001; 2003) This classification attempts to reflect the internal network linkage of the R&D subsidiary, but is certainly not without drawbacks. Three dimensions such as human resources development, innovation locus and autonomy cannot fully characterize a multidimensional construct such as an internal R&D network linkage. Each of these dimensions is multifaceted. Therefore, this dissertation attempts to focus in particular on human resources development at subsidiary level, innovation locus at subsidiary level and degree of autonomy by the R&D subsidiary. If tie modalities occur with headquarters and other R&D subsidiaries, it is assumed that the R&D subsidiary is better integrated in the internal R&D organization than if tie modalities are limited to an exchange with headquarters. If the internal R&D network linkage is limited to headquarters, the R&D subsidiary is obviously only partially 107 integrated in the overall internal R&D organization. The linkage is limited to one internal party only. 5.2.3.3.2 Role of External R&D Network Linkage External R&D networks refer to the R&D site’s relationships with external parties in the local context, namely research institutions, other firms (local and multinational), and the government (for more definitions of networks see Blankenburg Holm, Eriksson and Johanson, 1999; Gulati, 1998; Gulati, Nohria and Zaheer, 2000; Hage and Hollingsworth, 2000; Manor and Tasi, 2001; Powell, Koput and Smith-Doerr, 1996; Stuart, 1998). Other firms, local or multinational firms, can refer to customers, suppliers and competitors. Through these external network linkages, R&D subsidiaries can gain new and complementary knowledge outside the boundaries of the R&D subsidiary (McEvily and Zaheer, 1999: 1134). Thus, these external parties are important resources which can help to increase the technological sophistication of the R&D subsidiary. Within the external R&D network, both the tie modality along the dimensions of human resources, innovation and information as well as the membership structure are examined. This is in accordance with the literature, which distinguishes between relational and structural external network linkage (Andersson, Forsgren and Holm, 2002: 982-988). While these concepts have been used in a general context, they have not been applied to R&D subsidiaries. By examining both the tie modality as well as the membership structure of the external network, an attempt to adequately reflect the external network linkage for R&D subsidiaries is made. 108 Exhibit 28: External R&D Network Linkage Level External Network Linkage Linkage Level No links 1 One to two external parties on a short-term basis 2 Two to three external parties on a short-term basis 3 One to two external parties on medium-term basis 4 Two to three external parties on a medium-term basis 5 One to two external parties on a long-term basis 6 Two to three external parties on a long-term basis 7 Source: Author This operationalization attempts to classify the external linkage level. In the in-depth interviews, the R&D managers indicated with which external parties they had R&D collaborations (local firm, local research institution, MNE). The R&D managers were explicitly asked about R&D collaborations, not about collaborations with other aims. R&D collaborations refer to collaborations when the R&D subsidiary and an external party or parties conduct common R&D projects. Furthermore, the respondents indicated the duration of these R&D collaborations. During the exploratory phase of the research, the R&D mangers were asked to indicate what a meaningful definition of short-, medium- and long-term duration would be in an R&D context. Short duration refers to a duration on a short-term basis, i.e. R&D collaborations are only based on short-term projects (less than six months). In contrast, long-term R&D collaborations denote R&D collaborations which are based on longterm interests (more than two years). Medium-term R&D collaborations are collaborations which are based on several projects, but which have are not entered into on a long-term basis (between six months and two years). Furthermore, the R&D managers were asked to what extent external parties contribute to the competitiveness of the R&D subsidiary, how these external parties are managed and why they are entered into. 109 Obviously, these measures are not without drawbacks. The number of external parties and their duration may not translate one for one into the external R&D network linkage level. Some R&D collaborations might be of medium duration, but are not highly important for the R&D subsidiary, i.e. the duration of R&D collaborations may not indicate their significance. Moreover, the number of external parties may not translate into a high external R&D network linkage since tie modalities with these external parties might only be superficial. On the other hand, it can be assumed that the number of external parties and the duration of R&D collaborations can give a certain, albeit imperfect proxy of the external R&D network linkage. These measures attempt to reflect in how far the R&D subsidiary is a critical external player in the late industrializing context. Based on the measures illustrated, a regression was run with technological sophistication as dependent variable (see exhibits 29, 30 and 31) and external as well as internal network linkage as independent variables. Results are presented and discussed in the next section. 5.2.3.3.3 Discussion of Results Exhibits 29, 30 and 31 present the empirical evidence. The first regression includes the entire sample, while exhibit 30 presents regression results for the electronics industry and exhibit 31 for the biomedical sciences industry. 13 R&D subsidiaries belong to the electronics industry and 14 R&D subsidiaries to the biomedical sciences industry, the largest groups in our sample. Therefore, two separate regressions were run for these industries. The following exhibit shows the regression for the entire sample: 110 Exhibit 29: Technological Sophistication and Internal as well as External R&D Network Linkage (All Industries) Variables Entered/Removed(b) Variables Variables Model Entered Removed 1 EXTERLIK, . INTERLIK(a) a All requested variables entered. b Dependent Variable: TECHSOPH Method Enter Model Summary Adjusted R Model R R Square Square 1 .376(a) .142 .104 a Predictors: (Constant), EXTERLIK, INTERLIK Std. Error of the Estimate .95996 ANOVA(b) Model 1 Sum of Squares df Regression 6.844 2 Residual 41.468 45 Total 48.313 47 a Predictors: (Constant), EXTERLIK, INTERLIK b Dependent Variable: TECHSOPH Mean Square 3.422 .922 F 3.713 Sig. .032(a) Coefficients(a) Model 1 (Constant) INTERLIK EXTERLIK Unstandardized Standardized Coefficients Coefficients Std. B Error Beta 2.499 .458 .194 .114 .252 .114 .084 .202 a Dependent Variable: TECHSOPH t Sig. 5.456 1.702 1.366 .000 .096 .179 Source: Author As can be seen from the regression results, R square is 0.142, that is 14.2% of the dependent variable, namely technological sophistication, is explained by the two independent variables, namely internal and external R&D network linkage. External R&D network linkage, however, is not statistically significant. Internal R&D network linkage is statistically significant at p < 0.1. 111 Exhibit 30: Technological Sophistication and Internal as well as External R&D Network Linkage (Electronics Industry only) Variables Entered/Removed(b) Variables Removed Model Variables Entered 1 NATIONAL, EXTERLIK, INTERLIK, YEAR(a) Method . Enter a All requested variables entered. b Dependent Variable: TECHSOPH Model Summary Adjusted Std. Error of Model R R Square R Square the Estimate 1 .923(a) .851 .766 .42926 a Predictors: (Constant), NATIONAL, EXTERLIK, INTERLIK, YEAR ANOVA(b) Sum of Squares df Mean Square F Regression 7.377 4 1.844 10.009 Residual 1.290 7 .184 Total 8.667 11 a Predictors: (Constant), NATIONAL, EXTERLIK, INTERLIK, YEAR b Dependent Variable: TECHSOPH Model 1 Sig. .005(a) Coefficients(a) Model Unstandardized Coefficients B Std. Error 1 (Constant) 1.612 .467 INTERLIK .262 .093 EXTERLIK .212 .070 YEAR -.039 .022 NATIONAL 1.223 .353 a Dependent Variable: TECHSOPH Standardized Coefficients t Sig. 3.450 2.829 3.017 -1.787 3.459 .011 .025 .019 .117 .011 Beta .452 .472 -.354 .678 Source: Author R square is 0.851, that is, 85.1% of the dependent variable, namely technological sophistication, is explained by the two independent variables, namely internal and external R&D network linkage (nationality and year of establishment of R&D subsidiaries were taken into account). Both internal and external R&D network linkages are significant at p < 0.05. The following regression shows the results for the biomedical sciences industry. 112 Exhibit 31: Technological Sophistication and Internal as well as External R&D Network Linkage (Biomedical Sciences Industry only) Variables Entered/Removed(b) Variables Variables Model Entered Removed 1 EXTERLIK, . INTERLIK(a) a All requested variables entered. b Dependent Variable: TECHSOPH Method Enter Model Summary Adjusted Std. Error of Model R R Square R Square the Estimate 1 .743(a) .553 .453 .77234 a Predictors: (Constant), EXTERLIK, INTERLIK ANOVA(b) Model 1 Sum of Squares df Mean Square Regressi 6.631 2 on Residual 5.369 9 Total 12.000 11 a Predictors: (Constant), EXTERLIK, INTERLIK b Dependent Variable: TECHSOPH 3.316 F Sig. 5.558 .027(a) t Sig. 3.866 3.293 -1.191 .004 .009 .264 .597 Coefficients(a) Model Unstandardized Coefficients B Std. Error (Constant) 3.019 .781 INTERLIK .642 .195 EXTERLIK -.214 .179 a Dependent Variable: TECHSOPH Standardized Coefficients Beta 1 .847 -.306 Source: Author R square is 0.553, that is, 55.3% of the dependent variable, namely technological sophistication, is explained by the two independent variables, namely internal and external R&D network linkage. Internal R&D network linkage is significant at p < 0.01, whereas external R&D network linkage is not significant. Overall, the regression results suggest that internal R&D network linkage is positively related to technological sophistication, i.e. an R&D organization’s ability to integrate 113 and diffuse knowledge to its R&D subsidiaries has a positive impact on the technological sophistication at R&D subsidiary level. For the electronics industry, both the internal and external R&D network linkage seem to be important for the level of technological sophistication at subsidiary level. For the biomedical sciences industry, internal R&D network linkage is positively related to the level of technological sophistication. The results are more robust for the electronics industry and the biomedical sciences industry than for the entire sample (R square is 14.2% for the entire sample, while it is 85.1% for the electronics industry and 55.3% for the biomedical sciences industry). This finding may suggest that internal and external R&D network linkage, and in particular internal R&D network linkage, are more important for the level of technological sophistication in the electronics and biomedical sciences industry than in other industries. Overall, internal R&D network linkage seems to be more important than external R&D network linkage for the level of technological sophistication at R&D subsidiary level. This finding is in contrast to Frost, Birkinshaw and Ensign (2002). They state that external parties such as customer, suppliers and competitors are more important as sources of competence development than internal actors (Frost, Birkinshaw and Ensign, 2002: 1012). Their findings are based on a Canadian sample. Possible reasons for the difference in results may be various. Apart from differences in the definition of external and internal parties, the R&D subsidiaries in our sample are based in a late industrializing country (in contrast to a highly advanced economy such as Canada) and are based more at the technological stages of development than research. This may suggest that they are not fully integrated in the internal R&D organization, which in turn may indicate that more interaction with the internal R&D organization may help the R&D subsidiary in the late industrializing context to improve its technological capabilities. Our findings also suggest that Singapore’s science and technology policy may be effective in connecting R&D subsidiaries with important external parties. Because of that, external R&D network linkage may have less influence on the technological sophistication of R&D subsidiaries. This discussion shows that certainly more studies, in particular with a large sample and possibly comparing R&D subsidiaries in advanced and late industrializing countries, are needed in order to 114 validate the findings of this dissertation and to further enhance insights into this important topic. This previous section has analyzed the impact of internal and external network linkage on the technological sophistication. The following section attempts to analyze managerial implications of the interaction of internal and external network linkage. The analysis is based on the in-depth interviews conducted for this research. 5.2.4 Managerial Implications 5.2.4.1 Internal and External R&D Management Needs in a Late Industrializing Context While internal and external R&D network linkages are important for the R&D subsidiary, its interaction is critical as well. Based on the internal and external R&D network linkage, the R&D subsidiaries under investigation are classified according to four different types of R&D subsidiaries (see exhibit 33), namely as loosely linked R&D subsidiary, semi-linked-externally oriented R&D subsidiary, semi-linkedinternally oriented R&D subsidiary and fully linked R&D subsidiary. The internal R&D network is important for the R&D subsidiary in three ways. It enables the R&D subsidiary to gain and develop critical human resources, to actively participate in the R&D program and to be internally connected information-wise. Being a critical external partner in the local R&D network allows the R&D subsidiary to gain new and/or complementary knowledge and to enhance and/or maintain its competitiveness. In general, it has been posited that there is a link between internal and external R&D network linkage (Asakawa, 2001: 1-14). Examples of such a simultaneous interaction include that adequate support by the internal R&D network will lead to important external R&D network linkage. On the other hand, well-managed external R&D network linkages will provide key R&D sources for the internal R&D network. The main caveat in this interaction is to maintain an optimal balance between the two network linkages in order to avoid dispersion and information leakage in the external 115 network linkage and a too strong control by headquarters and/or other R&D sites in the internal network linkage. This problem was also pointed out by De Meyer (1993), who states that the information and knowledge learned locally has to be diffused in the company, which means that a self-organizing local external R&D network can only be effective if it is linked to a strong internal R&D network. The core management problem resulting from this interaction of internal and external network linkage is “how to find the right balance between a central control of the activities to avoid inefficiencies and unintentional duplication and a level of autonomy which is high enough to allow for an optimal deployment of local entrepreneurship and technical competence” (De Meyer, Mizushima, 1989: 139). This part of the dissertation attempts to provide a more systematic approach to this interaction between internal and external R&D network linkages. Four scenarios in this interaction have been depicted in exhibit 32 and are illustrated in more detail. 116 Exhibit 32: Classification of R&D subsidiaries according to their Internal and External R&D Network Linkages high Semi-linked R&D subsidiary Fully linked R&D subsidiary Loosely linked R&D subsidiary Semi-linked R&D subsidiary External R&D network linkage low low Internal R&D high network linkage Source: Author Scenario 1: Loosely linked R&D subsidiary This type of R&D subsidiary is dislocated from the main R&D sources both in the internal R&D organization and in the external R&D network. As Hobday (1995) correctly states, this isolation makes it difficult for the R&D subsidiary to reach a higher technological level. The number of internal relationships the R&D subsidiary is engaged in is low. This refers to all types of tie modalities, namely human resources, innovation and information. No critical human resources development takes place. With regard to innovation, the R&D subsidiary is the recipient of core technology, shows no R&D initiatives of its own and cannot participate in the global R&D program. Within the information flow, rules and regulations are to be followed according to 117 headquarters. Externally, the nature of tie modalities is similar. No critical human resources development and acquisition from external resources takes place. There are no external innovation impulses and no information exchange with external parties. Overall, the R&D subsidiary is of low strategic importance. It neither contributes to the internal R&D organization in a significant way nor is it seen as a critical partner in the external R&D network. Therefore, the loose simultaneous interaction of the two linkages negatively affects the competitiveness of the R&D subsidiary. It is particularly difficult for this type of R&D subsidiary to leave this stage and to reach a level of higher strategic importance. Management needs for this type of R&D subsidiary include identifying critical partners both internally and externally. Communication is a key element within the internal R&D network to report progress on ongoing R&D projects in order to increase trust. This in turn allows the internal R&D organization to assign more technologically complex R&D projects to this type of R&D subsidiary and it thus contributes to the development of human resources in this R&D subsidiary. In the context of late industrializing economies, it is crucial to develop external relationships, especially with government bodies, since they can provide both information on the external R&D network as well as R&D grants. Scenario 2: Semi-linked-externally oriented R&D subsidiary This type of R&D subsidiary is regarded as a critical partner in the external R&D network. However, is considered of low importance by the internal R&D organization. Overall, the R&D subsidiary is of medium strategic importance. The critical human resource acquisition and development takes place from the external network only. With regard to innovation, the R&D subsidiary is isolated from core technologies from the corporate R&D organization, but receives innovation impulses from the external network. Information-wise, it is loosely connected with the internal R&D organization. The information flow with the internal R&D organization is infrequent and at a low level in contrast to the external information flow which is frequent and mutual. 118 It is of paramount importance to communicate to the internal R&D organization that the R&D subsidiary has managed to be a critical partner in the external network and that the knowledge of the external R&D network is important and should be effectively used and transmitted in the internal R&D organization. Scenario 3: Semi-linked-internally oriented R&D subsidiary This type of R&D subsidiary is the opposite of R&D subsidiaries belonging to group 2, but is also of medium strategic importance. It has none to few relationships externally, but is highly interlinked in the internal R&D organization. Critical human resource acquisition and development takes place internally only. Furthermore, the R&D subsidiary participates actively in the internal global R&D program, but receives no external innovation impulses. The information exchange takes place mostly internally. Obviously, the interaction between internal and external network linkages allows an efficient internal connectivity, but not sufficient local autonomy for the R&D subsidiary. The R&D subsidiary is not engaged in crucial external research collaborations. The R&D subsidiary should hence start and/or increase external collaborations to be perceived as a critical external party and to be able to tap into external knowledge and to receive new external innovation impulses. This will allow the R&D subsidiary to increase its external strategic importance in the late industrializing economy. Scenario 4: Fully linked R&D subsidiary The last cluster is so to speak the ideal case. The R&D subsidiary is perceived as critical partner both internally as well as externally. This allows the R&D subsidiary to have the “best of both worlds”: strong internal as well as external tie modalities. Internally, it is connected through critical human resources, core technologies and mutual information exchange. Externally, the R&D subsidiary can access critical human resources, complementary and new knowledge and is able to receive grants from the local government since it is considered critical for the economic development in the late industrializing country. 119 To reach this stage has been considered by various respondents as “walking a fine line.” The management of this type of R&D subsidiary requires a constant balance between internal and external R&D management needs. If this balance can be maintained, this type of R&D subsidiary may reach the same status as R&D subsidiaries in the triad nations. 5.2.4.2 Discussion of Results The 51 R&D subsidiaries examined have been classified according to the four groups, which were discussed in the previous section. The result of this classification is depicted in exhibit 33. 120 Exhibit 33: Classification of R&D Subsidiaries in our Sample Group four (mostly R&D subsidiaries from the electronics sector) Group two high External R&D network linkage low Group one low Close to group three high Internal R&D network linkage Electronics Industry Chemical Industry IT/Communications Industry Aviation Industry Engineering Industry Food Industry Biomedical Sciences Industry Source: Author As one can see from exhibit 33, few R&D subsidiaries belong to groups one or three, that is, their external network linkage is low, their internal network linkage low to medium. Most R&D subsidiaries seem to be semi-linked-externally oriented R&D subsidiaries, i.e. they manage to be a critical partner externally, but fail to be perceived as a critical R&D subsidiary internally. Hence, it is of paramount importance to communicate to the internal R&D organization that the knowledge of this external network is important and should be used effectively and transmitted in the internal R&D organization. This in turn would contribute to a stronger internal network linkage 121 and improve the interaction between internal and external R&D network linkage in order to ideally reach group four. Some R&D subsidiaries have indeed reached or are close to reaching the ideal stage of group four. Some of the respondents of these R&D subsidiaries indicated that the interaction of internal and external R&D network linkage is important for the competitiveness of their R&D organization. For instance, this interaction helps to identify important research fields and to reconfigure and refine the R&D organization. It also helps the R&D subsidiaries to be efficiently internally connected and to have sufficient local autonomy. The firms of group four belong mostly to the electronics sector. This result is not too surprising since Singapore’s economic policy has been effective in building up expertise in that sector. The question then arises if this development is replicable for the biomedical sciences sector, which is relatively new to Singapore’s economy. So far, most R&D subsidiaries of MNEs belonging to the biomedical sciences sector are part of group two, that is they are critical external partners, but are not perceived as such in the internal R&D organization. It remains to be seen whether they will reach group four. Related to these linkages is the question with whom these linkages arise (network membership). An interesting result of analysis is that the internal network linkage is dominated by relationships between R&D subsidiaries and headquarters. The interaction with other R&D subsidiaries is usually limited. Out of the 51 respondents, only about half, 24, R&D subsidiary managers indicated that they have linkages with other intra-firm R&D units besides headquarters. This part of quantitative empirical findings is substantiated by qualitative findings. The following section presents three case studies. The first discusses Novartis’ R&D organization as an example of a metanational R&D organization. The second analyzes technological capability upgrading at Leica Instruments Singapore. And the third shows how internal and external R&D network linkage is important in building up technological sophistication in an R&D unit of Lilly Systems Biology. 122 5.3 Qualitative Findings 5.3.1 A Metanational R&D Organization in the Making: Novartis Institute for Tropical Diseases (NITD) in Novartis’ Research Organization9 5.3.1.1 Introduction As example of a metanational R&D organization this case study illustrates the research organization of Novartis and more particularly the Novartis Institute for Tropical Diseases (NITD), established in Singapore in 2002. This research institute focuses its research efforts on infectious diseases, more specifically on tuberculosis (TB) and dengue fever (DF). These diseases account for about 10% of the global disease burden10. Despite the prevalence of these diseases, however, the pharmaceutical industry has not focused its research efforts on them due to the low return on investment in research and the high costs of drug discovery and development. As for TB, basic science is mostly done by the public sector and usually fails to enter the phases of predevelopment and development. As for DF, there are no drug discovery activities. As a result, there are no adequate remedies for these diseases. Given this situation, Novartis created the NITD in order to further the discovery of preventive and effective treatments for these neglected diseases. 5.3.1.2 The NITD in Novartis’ Research Organization The NITD is part of Novartis’ worldwide research organization, consisting of 9 research institutions (3 in the US, 4 in Europe and 2 in Asia, including the NITD in Singapore) with in all around 3,000 researchers. These 9 research institutions are divided into two major R&D focus groups. 9 The information of this case study is based on in all five interviews with Prof. Dr. Paul Herrling, Dr. Thomas Keller and Dr. Richard Harrison (see list of interview partners in appendix) and on the attendance of the NITD Inaugural Symposium in January 2003. 10 Nearly 1% of the world's population is newly infected with TB each year. One third of the world's population is currently infected with the TB bacillus and two million people die from this disease each year (NITD Inaugural Symposium, 2003). 123 The Novartis Institutes for Biomedical Research, headed by Mark Fishman, comprise the research institutes in Basel (Switzerland), Horsham/London (United Kingdom), Vienna (Austria), Tsukuba (Japan), Cambridge/Boston and East Hanover (US). The Novartis Corporate Research Institutes, headed by Paul Herrling, consist of the Friedrich Miescher Institute (FMI) in Basel (Switzerland), the Genomics Institute of the Novartis Research Foundation (GNF) in La Jolla (US), and the NITD in Singapore. The NITD’s main roles are to conduct basic research, target finding, assay development, screening and medical chemistry on the diseases of TB and DF. The same technologies that are used by Novartis Institutes for Biomedical Research are applied by the NITD in the research areas of TB and DF. The establishment of the NITD enables Novartis to develop new research capabilities in the area of tropical diseases. Furthermore, it is planned to leverage expertise from the other corporate research institutions and from the Novartis Institutes for Biomedical Research for the NITD (for instance through sabbaticals or post doctoral studies at the NITD and through regular scientific symposia). Overall, the NITD therefore acts as both a discovery institute and a training institute (in particular as an external training institute). Novartis’ long-term goal is to help reduce the global burden of infectious diseases and thus to improve the prosperity of developing countries by researching those diseases which are prevalent in the developing world. The mission of the NITD is stated in the following: “The NITD aims to discover novel treatments and prevention methods in respect of major tropical diseases. In those developing countries where these diseases are endemic, Novartis AG intends to make treatments readily available and without profit. The discovery technology is state-of-the-art and the scope of the activities range from target discovery through to screen development and compound optimization” (http://www.nitd.novartis.com/mission/index_mission.shtml as of August 30, 2003). 124 5.3.1.3 Novartis’ R&D Organization as Metanational R&D Organization As described in this chapter, a metanational R&D organization is able to take advantage of a large number of knowledge bases and talent, including non-traditional R&D locations, and is able to leverage the international technological hierarchy to the fullest for the R&D organization. In order to do this, R&D organizations need to sense, mobilize and integrate complex and dispersed knowledge. This metanational advantage is analyzed with the example of Novartis’ R&D organization. 5.3.1.4 Leveraging of the Technological Hierarchy 5.3.1.4.1 Sensing of the Knowledge Base in the Periphery In mid 2000, Philip Yeo, Chairman of A*Star and Co-Chairman of the EDB, introduced the RISC (Research Initiative Scheme) to Novartis in Basel. In January 2001, a delegation of Novartis visited Singapore at EDB’s invitation in order to examine the local research context and conditions. In a next step, in June 2001, the Novartis Executive Committee approved the TB and DF initiative. In other words, it decided to conduct research on these diseases. In August 2001, the negotiation of key terms took place. On November 8, 2001, Daniel Vasella, Chairman and CEO of Novartis AG, officially announced the planned establishment of the NITD. On May 13, 2002 the first employee began to work for the NITD. On Oct 15, 2002 temporary facilities at Capricorn, Singapore Science Park II, were occupied until the NITD can move to a newly established science park (called Biopolis@one-north) in 2004. The NITD plans to hire 73 scientists and technicians (in about equal proportion) working on TB and DF. By the end of 2003, 36 scientists will have been hired and by the end of 2004 another 37 scientists will have joined the NITD. Besides the 73 scientists, 27 scientists, either on a post-doc basis or as part of their doctorate, will work on a temporary basis for the NITD per year. Besides sensing this knowledge base in the research area of tropical diseases, it is also critical to mobilize this knowledge. 125 5.3.1.4.2 Mobilizing the Knowledge Base in the Periphery The following section shows how the knowledge base in the periphery is mobilized along the dimensions of human resources acquisition and development, innovation configuration and information exchange (for the dimensions also see section 5.1.2.1). This will give an indication of how leverage of the technological hierarchy is used in Novartis’ R&D organization. Referring to human resources, a team of about five to six scientists from Novartis’ R&D organization is responsible for the establishment of the NITD in its initial phase. Human resource acquisition and development takes place by means of global hiring of research scientists through various measures (recruitment advertisements in major scientific journals, organization of recruitment events and scientific symposia). In addition, research scientists are also trained by the NITD, if they, for instance, lack capabilities for research in a commercial setting in contrast to an academic setting. Scientific assistants are hired locally. The scientists recruited so far are highly motivated to work for the NITD because TB and DF may be diseases endemic in their home countries, so by working on these diseases, they feel that they can contribute and improve their home country’s situation. Furthermore, it is planned to have frequent human resources exchanges between the NITD and other Novartis’ research institutes. With regard to innovation, the research program is determined by research experts; in this respect, they examine how the NITD can best achieve its mission, namely to discover novel treatments and prevention methods in the research area of tropical diseases. Proposals for the research program are presented to the Scientific Advisory Board, which is comprised of Sydney Brenner, Duane Gubler, Barbara Imperiali, Stefan Kaufmann and Rolf Zinkernagel. The proposals will then be prioritized and, based on this prioritization, will be implemented into specific experiments. Both scientific information and information in terms of experience are readily exchanged in Novartis’ R&D organization. With regard to scientific information, all internal publications by Novartis are globally accessible through Novartis’ intranet and through internal workshops. In addition, a main research conference takes place once a year and various scientific symposia take place regularly. Furthermore, the researchers 126 at the NITD are creating a web page on TB and DF as an open information forum for researchers, scientists, patients and medical doctors. In the research areas of TB and DF, the number of researchers is limited, which makes the information exchange relatively straightforward. Information in terms of experience in building up research institutes is transmitted in the following way: Paul Herrling built up the Sandoz research institute in Bern, Novartis’ research institute in Japan, the GNF in La Jolla and did major pre-work for the research institute in Cambridge. Therefore, this experience could be transmitted for the establishment of the NITD. Furthermore, the Novartis respiratory center was built in Horsham (UK) in 1997 where Thomas Keller (currently interim director at the NITD) worked for five years. Hence, Thomas Keller can transfer his experience of building up this research site in the UK to the NITD, for instance in terms of laboratory design, laboratory changes and efficiency in the process of building up the new institute in Singapore. As an expert in medical chemistry, Thomas Keller can also transfer his scientific knowledge to the NITD. A screening expert, David Beer (from Novartis’ research institute in Horsham) contributes his knowledge in this area to the NITD. Sabine Daugelat (from the Max Planck Institute in Berlin) built up the Biosafety Level 3 facility there. She is now responsible for the same task at the NITD. Alex Matter who spearheaded the discovery of the breakthrough cancer medicine Glivec has now been named as the Inaugural Director of the NITD. Being a highly important drug discovery scientist, he can transfer his drug discovery experience to the NITD. 5.3.1.4.3 Integrating the Knowledge Base into the R&D Organization In the following it is analyzed to what extent the different research institutions complement each other and how their different knowledge bases are used for the R&D organization. This analysis shows how the NITD is integrated in the overall R&D organization of Novartis. As mentioned above, the Novartis Corporate Research Institutes, headed by Paul Herrling, consist of the Friedrich Miescher Institute (FMI) in Basel (Switzerland), the 127 Genomics Institute of the Novartis Research Foundation (GNF) in La Jolla (US), and the NITD in Singapore. The FMI conducts basic biomedical research. Its research results in the field of gene sequencing could possibly also be used for research on DF at the NITD. The GNF in La Jolla (US) conducts research in the fields of genomics and proteomics. Their R&D activities on technology screening and identification of targets could also involve research on infectious diseases. Therefore, research projects conducted at the GNF and at the NITD could complement each other. Consequently, research synergies are formed with the other corporate research institutes, namely the FMI in Switzerland and the GNF in the US. Besides the corporate research institutes, there may also be research synergies with the institutes for biomedical research. The Novartis Institutes for Biomedical Research, headed by Mark Fishman, comprise the research institutes in Basel (Switzerland), Horsham/London (United Kingdom), Tsukuba (Japan), Cambridge and East Hanover (US). These institutes work on infectious diseases in a commercial way. One example will illustrate potential research synergies between the NITD and Novartis’ institutes for biomedical research. The NITD conducts research on DF and the biomedical research institutes on Hepatitis C and potentially the West Nile Virus. Research synergies may result, since the Hepatitis C virus and the DF virus are similar in nature. That is why research techniques and knowledge resulting from research conducted on Hepatitis C can possibly be used in researching the DF virus. A preliminary scientific understanding is that these two viruses (Hepatitis C und DF virus) may distinguish themselves in their surface. A further synergy may result from the sharing of clinical research knowledge. The NITD will gain clinical expertise in the most affected countries of tropical diseases. This expertise may be useful to Novartis’ institutes for biomedical research. Another example of a research synergy between the NITD and Novartis’ institutes for biomedical research is compound synergy. A compound from therapeutic areas of Novartis’ institutes for biomedical research, for instance, may be useful for the research conducted at the NITD, which in turn means 128 this compound would go through a new channel, possibly creating new research knowledge. On the technology side, the NITD is working closely with lead finding (at headquarters in Basel) since such a test system for drug target finding is expensive and hence cannot be built up at each Novartis’ research site. After the lead finding, lead optimization takes place at the NITD to identify suitable new potential drugs. After the lead optimization, the toxicology will either be conducted at headquarters in Basel or will be carried out by clinical research organizations (CROs). Pre-clinical studies will take place at the NITD. On the policy side, the NITD will collaborate with Novartis generics (antibiotics) in Kundl (Austria) since this site is experienced in working with international organizations such as the World Health Organization (WHO) or the TB Alliance. In this cooperation it is discussed how the NITD can optimize its collaborations with these international organizations. Issues to be addressed are the collaboration in clinical research studies and the determination of a suitable drug distribution. In order to conduct clinical research in those countries where tropical diseases are endemic, collaboration with international organizations is important, because these organizations can serve as important negotiation partners with the respective governments in these countries. For the determination of a suitable drug distribution, the cooperation of these international organizations is sought due to their knowledge of country-specific economic and political conditions. Overall, these international organizations are essential to overcome public policy failures in countries where DF and TB are endemic. In these countries, the infrastructure is often insufficient or fails to deliver drugs to patients in need. Reasons for such an insufficient infrastructure can vary, for instance, ineffective government or inappropriate transportation and handling of medical supplies. The illustration of sensing, mobilizing and integration of the knowledge base of a nontraditional R&D location has shown the leveraging of the technological hierarchy beyond the triad nations. The next section is devoted to the knowledge base in the 129 periphery and shows that not only the traditional knowledge bases, but also knowledge bases in the periphery, can be an important source of competitive advantage. 5.3.1.5 Knowledge Base in the Periphery In the following, discussion focuses on the extent to which the knowledge base in the periphery is important for Novartis’ research organization. Due to Novartis’ strategic location in Singapore’s biomedical sciences hub, the knowledge base in the periphery offers several advantages. Novartis’ R&D organization is close to major regions, where TB and DF are endemic. Access to a good hospital network (an efficient infrastructure) is a further advantage. Furthermore, the Singapore government is supportive of the NITD and provides both substantial financial as well as network support (for example contacts to relevant research institutions). Novartis’ R&D organization, in particular the NITD, can enter into collaborations with academic institutions and local firms, which also focus their research efforts on biomedical sciences. The collaboration with important academic institutions, for instance, allows the NITD to gain new and complementary knowledge and to access the local infrastructure. The NITD plans to work with all relevant universities, for instance the National University of Singapore (NUS), and with all relevant government research institutions such as the Institute for Molecular Biology, the Institute for Bioinformatics and the Institute for Genomics. These collaborations will also involve joint appointments between universities and the NITD, postdoctoral fellowships at the NITD (for the development of local expertise in the biomedical sciences), sabbaticals or teaching requirements by conjunct professors at the NITD and at universities and possibly providing a Master’s Program in Tropical Diseases. With regard to local firms, the NITD plans to collaborate with local biotechnology firms. Such collaborations will apply to the drug discovery process. Furthermore, there will be collaborations with clinical researchers in the region. One external relations manager at the NITD is responsible for managing all these collaborations in order to avoid dilution. 130 5.3.1.6 Conclusion This case study shows that Novartis’ research organization incorporates the elements of a metanational R&D organization. Knowledge bases worldwide have been sensed, mobilized and integrated as can be seen from the example of the establishment of the NITD, namely, knowledge bases reach beyond the triad nations to a non-traditional R&D location such as Singapore. Important management capabilities in building up the NITD as a knowledge base in the periphery are considered the following: the acquisition and development of key researchers for the NITD, building of trustworthy relationships with critical partners and strong persistence in the establishment of the NITD. Key researchers with expertise in the diseases of TB and DF are critical for ensuring high quality research. Trustworthy relationships both with critical internal and external partners are important for internal integration into the R&D organization and for playing an important role in the external research environment. And finally, discipline and persistence help to ensure an efficient process of building up the NITD. Despite the NITD’s successful establishment, management challenges remain. It is essential to formulate and implement a successful drug discovery strategy for the NITD (including portfolio management) because such a drug discovery strategy will determine the research results for the coming years. To find highly qualified and suitable scientists with a drug discovery mindset and to manage these scientists from diverse backgrounds is critical to assure an efficient research process. It will be most interesting to see the results of the NITD in the years to come. 131 5.3.2 Case Study: R&D Activities at Leica Instruments Singapore11 5.3.2.1 Introduction Leica was formed by the joint venture of Wild-Leitz and Cambridge Instruments in 1990. There are two major business divisions: Leica Surveying Group and Leica Microscopy Group (Goh et al., 2000: 524). Leica Instruments Singapore (LIS) was established as manufacturing subsidiary of Leica in Singapore in 1971. Technological capability upgrading started in April 1993 with the initiation of R&D activities at LIS, with one R&D Manager and one R&D engineer. Over the last 10 years, the staff strength has increased to 22 in 2003 (out of whom 20 are engineers and 2 are technicians) headed by a senior R&D manager, Ms. Germaine Tan. From 1993 onwards, the R&D activities at LIS increased in scope and complexity from mainly supporting manufacturing to performing exploratory development in 2003. This increase in scope and complexity is also reflected in an increase in the diversity of the R&D personnel, who come from pure mechanical to electronics, optics and surveying application fields. The R&D department in Singapore now acts as an extended branch of Stereomicroscopy R&D at headquarters in Switzerland (Goh et al., 2000: 537). This upgrading of R&D capabilities has been critical to the development of LIS as it has enabled LIS to support the Leica R&D organization by being of high strategic importance in the manufacturing, design and development of stereomicroscopes and surveying instruments. The case study illustrates this process of technological capability upgrading. 5.3.2.2 Underlying Rationale for Initiation of R&D Activities Leica started operations in Singapore in 1971 in the form of a manufacturing site for basic optical and mechanical components. The main reason for the establishment of the manufacturing site was the availability of low cost labor. In the course of Singapore’s economic development, however, labor costs increased sharply. Given this development 11 If not indicated otherwise, the information of this case study is based on in all two interviews with Mr. Goh and Ms. Tan (see list of interview partners in appendix). 132 and the consequent competition from low cost labor countries, Mr. Ah Bee Goh, LIS’ first local managing director, decided to pursue a strategy of focusing on higher value added activities in order to increase LIS’ strategic importance and thus to ensure its sustained competitiveness. As a consequence of this strategy, an R&D department supporting manufacturing was established in 1993. Before the technological capability upgrading for each business unit is analyzed (business units of stereomicroscopy and geosystems), LIS’ general technological path is illustrated. 5.3.2.3 General Technological Path at LIS The newly created R&D team was initially mainly responsible for product support, product adaptation and sustained engineering. It was in charge of ensuring the production of a new generation of stereomicroscopes and accessories in 1993. These tasks required enhanced process engineering capacities. In order to build local R&D capabilities, several engineers were recruited and trained in order to reach the level of competence to adequately communicate with headquarters on the product transfer. The R&D engineers needed to acquire design and manufacturing knowledge in castings, plastic injection and metal forming techniques (Goh et al., 2000: 531). In 1994, the technological level of the R&D department was increased to a level where it could carry out product design and development activities, the R&D team being required to redesign existing products to increase cost competitiveness. Furthermore, basic concepts were delivered from headquarters to LIS for conversion into products (Goh et al. 2000: 532). As a result of this first step of technological capability upgrading, LIS was awarded the ISO 9001 certification in the design and development of optical instruments. Besides this vertical technological capability upgrading, R&D capabilities were also enhanced horizontally, that is, the R&D team has increasingly been able to support a wider range of products (the initial product range of CMO Stereomicroscope and Automatic Level was for example replaced with a wider range of technologically 133 superior products such as the Greenough Stereomicroscope). In 1995, the R&D scope was enlarged to include surveying projects. This led to capabilities in electronics, including hardware and software for surveying products (Goh et al., 2000: 532). Besides the horizontal as well as vertical expansion of the R&D activities, R&D activities were also enhanced laterally from purely mechanical design to optomechanical design, electronics firmware design support, tooling and product software programming, optics design support and opto-electronics product support. For instance, the initial 2D mechanical drafting tool (Medusa) was replaced by 3D modeling tools. As a result of this continuous technological capability upgrading, the current R&D activities at LIS have reached the technological stage of exploratory development as of 2003 (common R&D projects are currently undertaken with SIMTech (Singapore Institute of Manufacturing Technology) and Nanyang Polytechnic). The exhibit below shows LIS’ general technological path according to the typology of technological paths identified in this chapter. 134 Exhibit 34: LIS’ General Technological Path (Technological Path I) Characteristics Technological Path I (TP 1) LIS’ Technological Path 1993-2003 Global Global Global Global Pace Sequential or exponential Sequential Technological Sequences M2 M2 (Product Support and Design in followed M2, D1 the mechanical field) M2, D1, D2 D1 (Advanced Development in the M2, D1, D2, R1 opto-mechanical area) M2, D1, D2, R1, R2 D2 (Exploratory Development in the or any subsequence within opto-mechanical-electronics field) Geographical Scope at path beginning Geographical Scope at further stages these technological stages Source: Author Remembering the typology of technological paths elaborated in this chapter, it is clear that LIS’ general technological path started with manufacturing support (M2), continued with advanced development (D1) and is currently at the technological stage of exploratory development (D2). LIS’ general technological path has been classified as a global linear technological path I since the R&D site’s output has always targeted the global market. Furthermore, it follows a linear rather than an exponential technological path because there is no empirical confirmation that its technological capability upgrading occurred at a faster pace later in the evolution of the R&D site than it did earlier in the evolution process. In addition to LIS’ general technological path, the two main divisions, stereo microscopy and geosystems, follow different technological paths. Before these technological paths are analyzed, however, major methods of technological capability upgrading need to be examined for they indicate how an R&D subsidiary is able to increase its level of technological sophistication. How these methods have been applied 135 by LIS will be analyzed in the discussion of the different technological paths of the two main divisions. 5.3.2.4 Methods of Technological Capability Upgrading The R&D subsidiary can choose various methods of technological capability upgrading, namely (1) production transfer from headquarters to the R&D subsidiary, (2) development from scratch by the R&D subsidiary, (3) linking R&D subsidiary personnel to key R&D engineers at headquarters or at other R&D subsidiaries, (4) collaborating with public research institutions in the local research environment and (5) initiation of new R&D projects by the R&D subsidiary itself. These methods will be briefly illustrated in general. First, production transfer from headquarters to the R&D subsidiary requires product engineering by the R&D subsidiary in order to sustain the products which are transferred. This, in turn, means that through intimate product knowledge and corresponding product engineering, the R&D subsidiary is capable of increasing its technological capabilities. In this process, technologies and skills are identified that are needed to ensure a smooth production. Tools, devices, instructions and process plans accompany such a production transfer, which is complete upon acceptance of pilot-run units (Goh et al., 2000: 528). Such a product transfer allows to gain competitiveness in manufacturing costs and consistency in quality. This method of technological capability upgrading is usually applied at the beginning of an R&D subsidiary’s evolution. Second, another method of technological capability upgrading usually applied later in an R&D subsidiary’s evolution is development from scratch, where the R&D subsidiary fully develops a new product or process on its own or in partial collaboration with headquarters. Therefore, the R&D subsidiary reaches further capabilities than those needed for a manufacturing support unit and becomes an R&D unit with development capabilities. Third, newly hired R&D engineers are linked to key R&D personnel at headquarters for training. In turn, corporate R&D engineers visit and work at the respective R&D 136 subsidiary for important milestone activities of the respective R&D projects. They are there to ensure the upgrading of technological capabilities. Such a close intrafirm collaboration allows exchange and upgrading of technological competences. Fourth, R&D collaborations with local research institutions are also important to upgrade the level of technological sophistication since such collaborations allow the R&D subsidiary to tap into different fields of expertise and into foreign talent working for these research institutions. These external collaborations allow the R&D subsidiary to gain new and complementary knowledge and to take advantage of the local infrastructure (not available within the firm) and to show headquarters that a certain R&D content is reached. Finally, new R&D projects can be initiated by the R&D subsidiary during the process of technological capability upgrading. In other words, it brings in its own ideas, which materialize into product developments. Thus, headquarters’ expectations can be exceeded and the strategic importance of the R&D subsidiary will be increased. How these methods have been applied in the process of technological capability upgrading at LIS in the division of stereomicroscopy is analyzed below. 5.3.2.5 Leica Microsystems: Process of Technological Capability Upgrading The process of technological capability upgrading of this unit is outlined through the major projects the stereomicroscopy business unit (BU-SM) has undertaken based on the respective interviews and on Goh (2002). Correspondingly, it shows the various methods of technological capability upgrading. In the first project (the product transfer of the MZ8 optics carrier project) LIS SM R&D was responsible for coordinating LIS production, LIS vendors and SM R&D. The objective was to create an efficient and effective communication between these three parties regarding all technical issues and the packaging design of all new system articles within the MZ8 optics carrier project. After this first product transfer, LIS R&D 137 continued to be involved in further product transfers (transfer of other new optics carriers, namely MS5, MZ6, MZ12 and MZAPO). In 1994, SM-R&D capabilities at LIS were increased through the Photo-Video Tube project, the first co-design project between headquarters and LIS-SM. In contrast to the first project of product transfer, LIS R&D was involved in the design and development of about fifteen system articles in this project. Given the experience and knowledge gained from the two previous projects, LIS R&D started its own R&D initiatives, for instance, by reviewing the technological features of new products, by proposing optimized solutions for existing products and by modifying the design of existing system articles to respond to customer requirements. This allowed LIS to reach the technological stage of advanced development (D1). The next important stage in the technological capability upgrading of R&D activities at LIS was the CATS Project in 1998, where LIS R&D actively participated in the conception stage of the new Greenough Stereomicroscope line. In this project, LIS R&D designed the full range of eyepieces, supplementary lenses, focus drives and bases (overall 23 system articles), under the conditions of a strict cost target, better performance features and a short time-to-market schedule. This CATS project resulted in the successful launch of the new stereo zoom line in 2000, and the filing of a patent titled “Focus Drive for Optical Instruments” in Europe and the USA, the joint invention of Mr. Peter Soppelsa at headquarters and Mr. Chris Loh of LIS R&D. As a result of the continuous technological capability upgrading at BU-SM, the LIS BU-SM R&D team has reached the technological stage of exploratory development (D2) and is now fully integrated in the internal corporate R&D organization of Leica. The main tasks in future technological capability upgrading for BU-SM will involve the development of more technologically sophisticated products and core modules, for instance product developments of optics carriers, which require expertise in both the opto-mechanical as well as the electronics field. In other words, the capability of integrating software, electronics and optics elements is critical for future technological 138 capability upgrading. The following exhibit visualizes the process of technological capability upgrading for BU-SM. Exhibit 35: Process of Technological Capability Upgrading at BU-SM Manufacturing Advanced Exploratory Support Development Development Mechanical Field Opto-Mechanical- Opto-Mechanical Field Electronics Field Activities: Product Support Product Design (Own R&D initiatives) Product Development Products: MS5, MZ6, MZ8, MZ12, MZAPO Photo Video Tube Greenough Stereomicroscope Line Increase in the Level of Technological Sophistication Source: Author As can be seen from the exhibit, BU-SM upgraded its technological capabilities to the level of exploratory development. While at the beginning of this process product transfers helped to upgrade the technological level, collaboration with headquarters and its own R&D initiatives enhanced the level of technological sophistication in the later stages of the R&D subsidiary’s evolution. 5.3.2.6 Leica Geosystems: Process of Technological Capability Upgrading The R&D activities in this business unit started later than the R&D activities in the stereomicroscopy unit. The first project in which Leica Geosystems’ R&D participated was the NA800 automatic level, which aimed at reducing the bench mark cost of the N2000 Level through redesign and process optimization. During this cost reduction project, Leica Geosystems’ R&D served as a communication channel for all technical 139 issues between the corporate R&D engineers, the operational departments in LIS and the vendors in Singapore. Upon project completion, LIS R&D was responsible for sustaining the NA800 Automatic Level (Goh, 2002: 73). The Pipe Laser Project followed the NA 800 project, involving the production transfer of the Pipe Laser and Interior Laser from COS Laser Technology, a Swiss company which had been acquired by Leica, to LIS, where Leica Geosystems’ R&D tasks were similar to those in the NA800 project (Goh, 2002: 73). The next R&D project in 1996 had the objective of reducing the manufacturing cost of NA800 automatic levels. As a result, Leica Geosystems’ R&D initiated the Cookie 99 (NA700 Auto-Level) project, which was highly significant for Leica Geosystems’ R&D since for the first time a complete geodesy instrument was designed and developed in LIS with close technical collaboration from headquarters. As part of the project, Leica Geosystems’ R&D was also responsible for designing eleven modules of the automatic level, starting in 1997. As a result, the new NA700 Level series was launched in 1999 (Goh, 2002, 73). With this initiative the life span of this automatic level was extended significantly. To date, this product has been more than 15 years in the market. This shows that this product’s competitiveness could be improved through R&D capabilities. After the Cookie 99 project, several product transfers of opto-mechanical-electronics products took place. This enabled Leica Geosystems’ R&D to gain the capability to manufacture these products and then subsequently to increase its know-how about them. Hence, the opto-mechanical and electronics know-how could be enhanced during this time and the corresponding R&D capabilities intensified. Given the knowledge obtained through these projects, Leica Geosystems’ R&D became part of the development team of technologically advanced products, such as the electronic theodolite and the total positioning system, where LIS’ R&D, with close support from the corporate R&D team, was responsible (1) for optimizing the design of a standard telescope for use in the T100 theodolite and (2) for developing a shiftable tribrach. This project enabled Leica Geosystems’ R&D to pursue further collaboration efforts with the corporate R&D team, which covered, for instance, the optimization of a 140 diagonal eyepiece for use with the T100 telescope and re-engineering of the T100 theodolite (Goh, 2002: 74). The most recent developments of Leica Geosystems’ R&D were initiated in September 2001 at a meeting between Mr. Pius Vorburger, Surveying and Engineering Development Director, and Ms. Germaine Tan, LIS’ R&D Senior Manager, when collaboration strategies were discussed and mapped out between LIS R&D and corporate R&D. As a result of this meeting, the most recent project, on barcode technology, started in April 2002 in cooperation with five researchers from SIMTech and with the corporate R&D team at headquarters. This product not only involves optomechanical, but also electronics parts, i.e. the product complexity is high. The following exhibit visualizes the technological capability upgrading at Leica Geosystems. Exhibit 36: Process of Technological Capability Upgrading at Leica Geosystems Manufacturing Advanced Support Development Opto-Mechanical Electronics Field Activities: Manufacturing Cost Reduction and Product Support Product Design and Development Products: NA700 and NA800 Automatic Level, Pipe Laser Barcode Technology Increase in the Level of Technological Sophistication Source: Author As major methods of technological capability upgrading, product transfers and collaboration with headquarters enabled Leica Geosystems to reach a higher level of technological sophistication. In contrast to SM, LIS’ Geosystems has not yet reached 141 the technological stage of exploratory development (D2). LIS’ Geosystems went through the technological stages of manufacturing support and advanced development. Major management capabilities for this technological capability upgrading process are discussed in the following section. 5.3.2.7 Management Capabilities During the Process of Technological Capability Upgrading During technological capability upgrading, several management capabilities have been considered critical by the R&D managers at LIS. First, a smooth intrafirm collaboration is important to assure continuous technological capability upgrading. Such an intrafirm collaboration requires open communication, mutual understanding, awareness of sensitive issues (such as potential emotions associated with know-how transfer), the building of trust and an appreciation of different cultures. It was important, for example, to explain to headquarters why an R&D site in Singapore would be beneficial not only for LIS, but also for the overall R&D organization. In the subsequent process of technological capability upgrading, diplomatic handling of knowledge transfer was important. Second, the capability of developing a product in the shortest time possible, at the lowest cost possible and at the best quality is critical. This, in turn, requires a tight work schedule, which can be highly demanding for the R&D team, which has to learn and work simultaneously. In order to achieve this capability, the creation of commitment and the quick handling of potential conflicts are essential. Third, human resources management with regard to R&D was important in order to build a highly qualified and well-functioning R&D team. All R&D engineers, for instance, joined the team with a three-year working contract in order to avoid short-term interests so that an R&D team on a long-term basis could be created. It was further essential that all new R&D engineers had had several years of professional experience in design and development prior to joining LIS (Goh et al., 2000: 532-533). 142 Forth, LIS’ management team has incorporated a set of core values over the years under the leadership of Mr. Ah Bee Goh. These core values refer to operating principles, performance, business and reward ethics and agility (Goh, 2003: 1). Of high importance are the ROFO (Responsibility, Ownership, Focus and On-time corrective action) principles and the learning philosophy. These core values form a strong corporate culture and have helped LIS’ employees to pursue their objectives successfully and to continue technological capability upgrading. This learning philosophy, visible in every employee, was crucial for the different R&D projects. 5.3.2.8 Impact of Technological Capability Upgrading on LIS’ Performance The upgrading of R&D activities at LIS has been critical for LIS in order for it to remain competitive and to increase its strategic importance. Since 1995, 55% of the products at LIS have a product age of less than two years. In addition, all product launches by LIS have been on time. Close collaboration of the R&D department with manufacturing reduced time-to-market. Prior to the R&D efforts, it was difficult for manufacturing to generate a short time to market. The R&D efforts also enabled Leica to be close to markets. Competitive products can be created by fulfilling costumer needs in the growing markets of Asia with design for manufacturing (Goh et al., 2000: 539). As a result of the R&D activities being upgraded, the number of new product developments could be increased from 5 in 1993 to 18 in 2002. In addition, one patent was filed. The number of R&D projects also increased from 2 in 1993 to 7 in 2002. The contribution of LIS to the overall Leica R&D organization in terms of development content has become important (for BU-SM from 10% in 1993 to 80% in 2002). At the beginning of the R&D activities, light performance measurements were applied such as the number of projects LIS could work on or the reduction in manufacturing costs. In the course of technological capability upgrading, performance measurements were tightened, for instance, by considering the number of projects which are joint developments between headquarters and LIS or by including the number of new products. Furthermore, the extent to which research is conducted is increased with the close support of the corporate R&D team, especially in the fields of software and electronics imaging. 143 The following exhibit shows LIS’ financial ratios in terms of worldwide sales and profit before tax as a percentage of Leica’s profit before tax overall. Exhibit 37: LIS’ Financial Ratios 1992-2003 100,000 50% Start of LIS' R&D Activities % of Leica 40% overall in Singapore Dollars '000 90,000 80,000 70,000 30% 60,000 50,000 20% 40,000 30,000 10% 20,000 10,000 0% Sales FY03 FY02 FY01 FY00 FY99 FY98 FY97 FY96 FY95 FY94 FY93 (10,000) FY92 0 -10% Profit Before Tax % Source: From LIS Management 2003 with permission As can be seen from exhibit 37, sales have improved roughly threefold since the beginning of the R&D activities in the year 1993 (fiscal year (FY) 1994). Over the same period, the workforce decreased from 620 employees to 500 employees. Furthermore, profit before tax increased steadily reaching around 10% of Leica’s overall profit in the years 2002 and 2003. Despite the successful technological capability upgrading at LIS, main challenges remain for LIS R&D in the years to come, which are discussed below. 144 5.3.2.9 Main Challenges Ahead First, training programs with corporate R&D should be continued and intensified. Current programs include, for example, the training attachment program, whereby LIS R&D personnel spend 1½ years at Leica Geoystems and Microsystems at headquarters, supported by the EDB. This program is crucial since it allows faster technological capability upgrading at LIS and critical interaction with headquarters. Second, it is important to maintain the right balance between technology/innovation and time to market. Customer needs are critical, especially their price requirements. An important challenge remains in how to enable technology to serve customers. Therefore, it is critical to identify the core technologies to be used in products in order to shorten the product cycle. The last challenge refers to human resources in R&D. It is critical to employ and retain highly qualified R&D engineers with the correct expertise. Human resources management needs to provide high motivation on a long-term basis. It is also important that the R&D engineers possess the relevant knowledge for new technology waves. It will be most interesting to see LIS’ further technological capability upgrading in the years to come. 145 5.3.3 Internal and External R&D Network Linkage: The Establishment of Lilly Systems Biology in Singapore12 5.3.3.1 Introduction Due to various reasons pharmaceutical firms are increasingly under pressure to effectively use new knowledge. The emergence of several new technologies in the last five years created a large amount of data. Examples of such new technologies and the corresponding new data are, for instance, newly obtained information on the mapping of the human genome, gene expression profile data (DNA chip) or improvements in screening and bioinformatic capabilities. Given such developments in the scientific field, it is crucial to integrate this information in order to create new knowledge on, for example, gene(s) functions and their interactions under different biological conditions of disease as well as under the effects of existing and new chemical entities. In order to achieve this integration, it is necessary to combine the fields of biology, chemistry and pharmacology heavily leveraging on the advances made in the information technology area. This development, which leads to the creation of new knowledge, is paralleled by a decreasing productivity in R&D. R&D expenditure doubled in the last 20 years, but at the same time research productivity increased by only 10%. Reasons for the decline in new chemical entities are numerous. One possible explanation is that research is not conducted in the correct information space (also see Achilladelis and Antonakis, 2001). Based on the need for accelerating the drug discovery process, research has to be indeed conducted in the correct information space. Lilly Systems Biology is attempting to integrate the newly emerged technologies’ fields and hence to conduct R&D in the right information space. Eli Lilly and Company is the first pharmaceuticals company to initiate the Systems Biology efforts by establishing the Lilly Systems Biology center in Singapore. ‘Systems Biology’ refers to the examination of human biology in the context of pharmaceutical discovery and as such to a broader understanding of biology. In other 12 The information in this case study is based on in all two interviews with Dr. Santosh Mishra, Managing Director of Lilly Systems Biology. 146 words, systems biology is a creative approach to studying whole biological systems utilizing contemporary biotechnology strategies, including bioinformatics, in parallel with conventional biomedical research. Thus, Lilly Systems Biology conducts R&D at the intersection of biology, chemistry, and pharmacology with the objective of accelerating the drug discovery process and increasing R&D productivity. 5.3.3.2 Motivation for Establishing Lilly Systems Biology in Singapore The R&D site, Lilly Systems Biology, Lilly’s first R&D site in Asia, was established in 2002 with planned staffing of 50 scientists. Researchers for the R&D site were hired regionally and on a multi-disciplinary basis in order to create new research capabilities. The strong government commitment enforced the decision to establish Lilly Systems Biology in Singapore. The current science and technology policy with the objective of building a biomedical sciences hub enables Lilly Systems Biology to find potential synergies with other pharmaceutical firms, which are in close geographical proximity. The regulatory framework created by the Singapore government in terms of intellectual property is an important consideration as well. 5.3.3.3 Internal R&D Network Linkage With regard to human resources, critical human resources development takes place at Lilly Systems Biology. As has been stated, multidisciplinary scientists have been hired. Requirements for the recruitment are that the R&D personnel has an academic and/or professional background in one or more scientific discipline(s) and knows how to apply this knowledge to the biological field. For instance, a researcher may have expertise in statistics and knows how to apply this expertise to the field of biology. If there is a lack of expertise in one area, the R&D personnel will be trained within the internal R&D organization of Lilly. In 2002, the year of Lilly Systems Biology’s establishment, six managers were sent from Lilly’s headquarters in the US to Singapore to help build up the R&D site. Up to now there has not been a human resources flow from Lilly Systems Biology to headquarters or to other R&D sites but such an exchange between R&D sites is planned 147 for the future in the form of internships and post doctoral work at Lilly’s R&D centers in Europe and the USA. With regard to the innovation flow, research projects have been worked on simultaneously by headquarters and Lilly Systems Biology since the establishment of the subsidiary in 2002. Lilly Systems Biology is also free to conduct its own R&D projects. Most internal research projects, however, are in collaboration with headquarters, since the majority of researchers and scientists are located there. In 2003, Lilly Systems Biology will work on many more research projects, determined in close cooperation with headquarters, depending on resources and expertise at the different locations. Overall, the research projects are determined both by market needs as well as by internal needs (for instance in terms of computational infrastructure and various technologies), which are driven eventually by market needs, namely to meet unfulfilled medical needs, for example in the areas of cancer or other disease areas. With regard to the information flow, a computational database is available for all R&D personnel. Furthermore, senior managers at each R&D site know the major R&D projects at all other R&D locations. The information flow between Lilly Systems Biology, headquarters and other R&D sites takes place on a mutual and constant basis. As can be seen from the analysis of the internal network linkage, Lilly Systems Biology is an important R&D site in the internal R&D organization of Lilly. 5.3.3.4 External R&D Network Linkage Lilly Systems Biology is currently at just the beginning to discuss potential external research collaborations, of which headquarters is highly supportive, in order to find synergies and to create win-win situations. Mutual rewards, risk sharing and openness are the most important criteria for such collaborations. Furthermore, these collaborations help the company to acquire new and complementary knowledge and to tap into the local infrastructure. One research alliance management team at headquarters oversees all Lilly research collaborations in order to avoid potential management 148 problems and dilution. The same general guidelines apply for all R&D sites with regard to external research collaborations, but specific details apply for each region. Barriers to external research collaborations are various. The pharmaceutical or scientific value of the research collaboration needs to be identified, the budget needs to be determined and the intellectual property rights have to be clarified by all parties. Since the establishment of Lilly Systems Biology took place only in 2002, it is difficult to determine what characteristics the external network linkage shows. It can, however, be assumed from the information given that high external network linkage is important for Lilly Systems Biology. 5.3.3.5 Interaction between Internal and External R&D Network Linkage Given the previous analysis, it can be said that the internal network linkage is strong and the external network linkage is currently being built up. Internal network linkage helps to access important external network configurations. For instance, the manager responsible for building up the research clinic in Singapore and various R&D sites in Europe has also been in charge of building up Lilly Systems Biology in Singapore. Thus, Lilly Systems Singapore can profit from Lilly’s previous experience. External research collaborations are mostly determined by Lilly Systems Biology Singapore within the general guidelines: the internal R&D organization allows Lilly Systems Biology Singapore to engage freely in external research collaborations and headquarters might even suggest or make Lilly Systems Biology Singapore aware of potential external research collaborations. Lilly Systems Biology Singapore can be seen as an extension of research conducted at headquarters and thus increases Lilly’s strategic R&D presence with regard to Asia. 149 5.3.3.6 Management Capabilities and Challenges Ahead In order to develop the newly created R&D site further, Lilly Systems Biology applies various management capabilities. Creating an open, motivating and challenging environment, where Lilly’s corporate principles of respect for people, integrity and excellence apply is crucially important. In such a working environment, projects are given as the major guideline, but the micro management (how and when the R&D scientists work on the projects) depends on the scientists themselves. The emergence and conversion of many technologies has created a large amount of data and a new information space. The challenge is now to understand this information space and to find out how scientists can work on these technologies simultaneously. This also means to raise the awareness of the importance of doing multidisciplinary work. And finally, the challenge is to find well-trained scientific researchers, a challenge because there is an insufficient number of such people worldwide. 5.4 Summary of Findings The first part of the quantitative findings of this dissertation examined different international R&D organizational models and discussed exploratory performance implications. It has been found that a metanational R&D organization, that is, an R&D organization, which is both present in the triad nations and the periphery and which is capable of optimally leveraging the technological hierarchy, is rarely found in corporate practice. Potential barriers to this type of R&D organization are the dominance of the home base, the perceived unimportance of R&D subsidiaries in the periphery and the notion that local adaptations can only be applied locally. However, even though the metanational R&D organization is reality for only a few R&D organizations today, it could be the new model of international R&D organizations in the future. The results also suggest that different international R&D models entail different R&D performance implications. While international R&D models such as the metanational R&D organization, integrated R&D network and the R&D hub model, seem to be more suited to produce an important number of new product developments, the ethnocentric 150 R&D organization may be a good organizational form in order to achieve a significant number of patent applications. Non-traditional R&D locations can raise their importance by increasing their level of technological sophistication. Such an investigation is important in order to understand why non-traditional R&D locations are still at the periphery, which means they are not yet part of the group of more advanced R&D locations, and how they can progress further in their technological capability upgrading. In a second part of the quantitative findings, it has, therefore, been examined how the periphery can raise its strategic importance through such technological capability upgrading. The current level and types of technological capabilities of the R&D subsidiaries in the sample were discussed. We found that the R&D subsidiaries are currently more at the technological stage of development than of research. A typology of technological paths was also derived and performance implications analyzed. The findings indicate that a sequential technological path, first focusing on a local or regional scope and then on a global scope, seems most favorable in terms of the number of new product developments. In a next step, key factors, which influence technological capability upgrading, namely internal and external R&D network linkage, were analyzed. The analysis shows that internal R&D network linkage is more important for technological capability upgrading than external R&D network linkage (with the exception of the electronics industry). This observation is also confirmed by examining the interaction of internal and external R&D network linkage. Most R&D subsidiaries are semi-linked-externally oriented R&D subsidiaries, i.e. they are critical external parties, however lacking strategic importance in the internal R&D organization. Management implications were discussed accordingly, namely that it is critical to communicate intensely with headquarters on ongoing R&D projects and to further increase the level of technological sophistication. In this way, the R&D subsidiaries can also reach a high strategic importance internally. The qualitative findings attempted to substantiate the quantitative findings. The first case studied, that of Novartis, analyzed Novartis’ research organization as an example of a metanational R&D organization in the making. Novartis’ R&D organization is both present in the triad nations and the periphery. The metanational advantage in terms of sensing, mobilizing and integrating the knowledge base in the periphery has been 151 achieved. Novartis established the Novartis Institute for Tropical Diseases (NITD) in 2002 in Singapore, mobilized knowledge in the research areas of tropical diseases and integrated this knowledge accordingly in the internal R&D organization. Major management capabilities include the acquisition and development of key researchers, building of trustworthy relationships with critical partners and strong persistence. The second case study, Leica Instruments Singapore (LIS), showed how an R&D subsidiary can increase and maintain its strategic importance in the overall R&D organization by increasing its level of technological sophistication. LIS started its R&D activities in 1993 in form of a manufacturing support unit. In 2003, LIS has reached the technological stage of exploratory development. The different divisions of LIS have undergone different developments. As major management capabilities, the management of emotions associated with technology transfer, achieving a short time to market and a human resources management on a long-term basis are pointed out. The third case study, Lilly Systems Biology, showed the importance of internal and external R&D network linkage for the building up of a new R&D site. Lilly Systems Biology conducts research at the intersection of biology, chemistry, and pharmacology with the objective of accelerating the drug discovery process and increasing R&D productivity. Lilly Systems Biology Singapore is well integrated in the internal R&D organization and is currently building up a strong external R&D network linkage. Management capabilities comprise the creation of a challenging working environment and conducting research in the correct information space. The next chapter discusses implications for theory, practice and policy. 152 6 Implications for Theory, Practice and Policy 6.1 Implications for Theory This dissertation attempts to take a small step towards a more advanced understanding of the literature on international R&D. Three research streams were distinguished, namely R&D internationalization determinants, international R&D management and the R&D internationalization process. Regarding the first research stream, this dissertation developed a holistic determinant for R&D internationalization in contrast to previous studies which emphasize market and technology driven determinants. With respect to the second research strand, this dissertation developed a framework for a new R&D organization, namely the metanational R&D organization, and discussed the consequent implications for R&D management. In doing this, it provided a comprehensive illustration of international R&D organizations. Traditional international R&D organizations emphasize the home versus host country dichotomy, are present only in the triad nations and have a low to medium degree of specialization. Furthermore, traditional international R&D organizations develop key R&D personnel only at headquarters or at key R&D subsidiaries, where the locus of innovation is situated. Their source of knowledge is mostly internal. By contrast, a metanational R&D organization adopts a comprehensive approach, is present both in the triad nations and the periphery and has a high degree of specialization. Critical R&D personnel are developed also in the periphery, the innovation locus is anywhere in the R&D organization and the source of knowledge is both internal and external. Furthermore, exploratory performance implications of different international R&D organizations were drawn. The case study of Novartis’ R&D organization presents a metanational R&D organization in the making. 153 The analysis in this dissertation shows that the literature on international R&D organization and management is evolving in nature (also see Doz, 2003). For example, there is as yet no comprehensive theory of international R&D organizations and management. The third research stream, the R&D internationalization process, has examined this process from a corporate perspective and in advanced economies. Therefore, the analysis of technological capability upgrading in this dissertation is a modest contribution to the literature on R&D subsidiaries in late industrializing countries because the literature has so far largely ignored the process of R&D internationalization on R&D subsidiaries in the periphery. Instead, it has focused on R&D subsidiaries in the triad nations, where technological capabilities are substantially available. Our understanding of R&D subsidiaries in the periphery is still limited. The periphery can increase its strategic importance through technological capability upgrading, and thus create a critical knowledge base, a necessary condition for metanational R&D organizations to tap into the periphery. Therefore, a detailed analysis of technological capability upgrading was provided. The level of technological capabilities was examined, a typology of technological paths identified and key factors (internal and external R&D network linkage) on technological sophistication investigated. The study underscores the importance of technological capability research. Results showed that technological capability upgrading is at least partially a function of internal R&D network linkage at R&D subsidiary level in late industrializing countries. If further studies support this conclusion, this would provide important theoretical implications for R&D management in a late industrializing context. Overall, the dissertation attempts to provide some elements for a more advanced theory of international R&D management. The literature so far has focused on various issues in international R&D management, but has not yet developed a systematic and more concise approach to international R&D management theory. It would be important to integrate the various elements in international R&D management in order to possibly develop an international R&D theory. More theoretical development is certainly needed. 154 6.2 Implications for Practice Most of the research implications of this study are directly applicable to the practice of overseas R&D management. The interview partners were asked what lessons they had learnt in building up an R&D subsidiary in the periphery and/or what management challenges they had been facing, are facing and are likely to face. These important managerial implications are discussed in the following: An important managerial implication concerns the gap in perception between headquarters and R&D subsidiaries. The headquarters-subsidiary relationship is crucial as subsidiaries are organized through interdependent change (Birkinshaw et al., 2000: 322). Birkinshaw et al. (2000) have found that home-country managers and overseas site directors are not ’on the same page’ when it comes to the mission or role that individual subsidiaries are expected to play within the organization (Birkinshaw et. al., 2000: 326). This finding was confirmed for the R&D context by Grevesen, (2001). In his study, he showed that chief corporate R&D officers sometimes categorized a particular R&D subsidiary unequivocally as either a home base augmenting or home base exploiting unit while the R&D manger in the host country selected the opposite classification (Grevesen, 2001: 113-114). This observation was confirmed in many of the in-depth interviews conducted for this dissertation. While at headquarters the specific R&D subsidiary was classified as a technical support unit, the view at the R&D subsidiary itself was very different. The R&D subsidiary would usually see itself at a higher level of technological sophistication. Analyzing official corporate information led to the same result. While the R&D subsidiary in the periphery is not mentioned at all in official corporate sources, the R&D subsidiary is, however, highly important in the local context. In some cases, respondents at headquarters even indicated that they had no R&D activities in Singapore; the contrary truth emerged, however, during fieldwork in Singapore. Consequently, it is frequently found that the R&D subsidiary’s level of technological sophistication is overestimated by the subsidiary and underestimated by headquarters. In general, it seems that more R&D is conducted at the subsidiary level than 155 headquarters is aware of. In order to overcome this difference in perception, it is important for an R&D manager at subsidiary level to make headquarters aware of the R&D subsidiary’s existence and capabilities. The subsidiary R&D manager also needs to encourage awareness of the local context and thus create more credibility. This, in turn, requires intense communication with headquarters and constant technological capability upgrading. It is also critical to be an important external player, which in turn increases the credibility of the R&D site. If important external R&D collaborations are entered into, this will help the R&D subsidiary to gain new external R&D knowledge. It has been reiterated in many interviews that headquarters needs to be convinced that the R&D subsidiary has the capability and credibility to conduct certain R&D projects. Continued communication is necessary to convince headquarters that it is important to conduct certain R&D projects at a subsidiary level and not at headquarters, or in other words, to show that the R&D subsidiary in Singapore has the same or similar capability as headquarters to conduct R&D. Otherwise, the R&D subsidiary will be given only small and technologically less demanding R&D projects, which will make the technological progress of the R&D subsidiary slower and more difficult. If the R&D site’s credibility and self-sustainability is not constantly proven, the R&D subsidiary’s existence might even be thrown into jeopardy. Once the underestimation of the R&D subsidiary’s role is overcome, however, it can start or continue to play an important role in the overall R&D organization. Another managerial implication which has been pointed out by many of the interview partners refers to the acquisition of suitable R&D resources. Due to a local shortage of R&D manpower (see chapter 4), it is important to hire key R&D personnel on a global basis. Such a pool of key researchers is critical in building up an R&D site. Once this acquisition of human resources is managed, R&D subsidiary managers need to create a challenging and highly motivating work environment. The challenge lies in creating curiosity, passion and spirit of innovation. Equally importantly, human resources management of R&D personnel should attempt to retain the critical R&D personnel on a long-term basis in order to ensure that the building up and development of the R&D subsidiary continues. 156 Human resources management needs to be in accordance with project management. It is important for R&D subsidiary managers to create the correct technological roadmap and then to follow it consistently. Project management requires foresight to seize important technological trends, which has been pointed out as an important managerial challenge. In a pharmaceutical context, for instance, project management in the form of a drug discovery strategy determines the future pipeline and thus medicines for several years ahead. If such a technological roadmap or strategy has little foresight, it may be difficult to achieve good R&D results. In this respect, it is also important to follow the technological roadmap closely once it is determined in order to avoid dilution or a slowing down of the R&D process. In the interviews, R&D managers mentioned that it is critical to communicate this project schedule to R&D personnel because some of the researchers may have been former academic researchers, who now need to follow a commercial schedule in industrial based R&D. R&D managers were also asked what management challenges they consider important for the future of their R&D subsidiaries. They identified different challenges ahead. Accelerating knowledge, development towards increasing product complexity, increasingly fierce competition and maintaining high research productivity were identified as the major challenges. The challenge of accelerating knowledge refers to the fact that R&D managers face increasingly more, more complex and fast changing R&D knowledge. Closely linked to this challenge, product complexity is increasing as well. In spite of this increasing knowledge and product complexity challenge, high R&D productivity has to be maintained or increased in order to be able to cope with intense competition. It will be interesting to see how these challenges will be tackled in the future. 157 6.3 Implications for Policy Besides theoretical and managerial implications, this dissertation is also concerned with policy implications. Its analysis of technological capability upgrading of R&D subsidiaries in Singapore highlighted the divide between development and research. While most R&D subsidiaries are at the technological stages of advanced or exploratory development, only a few have reached the technological stages of applied or basic research. Nonetheless, it is apparent that this divide may be narrowed, a change in which Singapore’s science and technology policy is playing a critical role. The government is not only providing financial incentives but also IP protection and an efficient infrastructure in the form of government R&D laboratories. In Singapore, general expenditure on R&D as a share of gross domestic product increased from 1% in 1991 to 1.89% in 2000 (Amsden and Tschang, 2003: 565). Systematic government support is critical in fostering R&D activities (also see Bartholomew, 1997). In an effort to upgrade the technological activities and thus bridge the divide between development and research, Singapore faces challenges. These challenges for late industrializing economies in attracting more and high level R&D activities are threefold: First, the Singapore economy needs to develop sufficient local expertise, especially in the newly created biomedical sciences, as has been elaborated on in chapter 4. The question arises if the success which was achieved in building up an electronics cluster is replicable for the biomedical sciences sector. Currently, there is a shortage of high-level research personnel. For instance, only 8.2% of the 1,930 researchers in government research institutions are Singaporeans with PhDs, which amounts to only 160 PhD holders (Yong 2003: 4). Thus, for the development of critical human resources, an expansion and upgrading of the university system is necessary as well. The Singapore government has addressed these issues in the shortage of R&D labor by granting subsidies for personnel training. More than half of all research grants are devoted to such training (Amsden and Tschang: 565). It has also adopted liberal immigration laws in order to attract foreign professionals with high talents. In addition, various reforms 158 have been initiated in order to expand and upgrade the university system, with the objective of producing high-end researchers. Second, closely linked to this development of critical human resources is the challenge of changing the mindset of researchers from mere execution to more creativity. While lower-end R&D activities, which are closely linked to manufacturing, require more execution skills, higher-end research projects require more creativity (also see chapter 4). Several R&D managers pointed out that the local workforce is better suited to development than to research due to its lack of creativity and its lack of initiative. It is more difficult for the Singapore government to address these issues. However, with Singapore’s increasing transition towards more research based activities these issues may resolve themselves with more time. Third, the interviews provided some insights into local R&D subsidiaries, whose level of technological sophistication is slightly lower than that of R&D subsidiaries of MNEs. The ten Singapore-based companies indicated that it is very difficult to support R&D, especially as a small and medium sized company, constantly having to strike a balance between R&D and commercial value. This observation is also confirmed in a study by Kam et al. (2003: 17), who state that in Singapore much innovation arises from the application of technologies that were developed and are already in use elsewhere. The most common innovation activity in the Singapore manufacturing sector, for instance, is the acquisition of machinery, equipment and software linked to product and process innovation (80% of innovating manufacturers), followed by R&D (66%) (Kam et al. 2003: 17). Given the preceding context, it is important to foster the R&D activities by local companies and thus to create indigenous innovation. This, in turn, will reduce Singapore’s dependence on MNEs for innovation. In summary, it is apparent that the role of the Singapore government is decisive in fostering more and higher level R&D activities that otherwise would not have occurred. In view of this critical role, Singapore’s science and technology policy will remain important in the country’s transition towards more research activities. Based on the limited evidence of this dissertation, it seems that systematic policy formulation and implementation continues to be of vital importance if more research activities as 159 opposed to development activities are to be created. Pro-active government policies and the operation of public research laboratories are critical. Future science and technology policy would have to consider how to balance the focus on biomedical sciences with other industries. While the emphasis on biomedical sciences is important, science and technology policy should not neglect other industries. A balance would ensure the upscaling of technological sophistication in all industrial sectors. Some interview partners felt that their R&D efforts are not adequately recognized because they are not part of the biomedical sciences sector. The building up of a biomedical sciences cluster, however, seems to be important as it constitutes a second major cluster besides the electronics industry. This would also provide the Singapore economy with a broader economic diversity in the case of external shocks to the economy, as was the case during the Asian financial crisis. Within the biomedical sciences industry, which encompasses the pharmaceuticals, medical devices, biotechnology, and healthcare services sectors, it may be necessary to focus on one or two of these subsectors. Due to its small economy (with a population of about 4.1 million), it may not be feasible for Singapore to acquire expertise in all of these subindustries. It may be better to have one strong focus within the biomedical sciences in order to avoid the danger of dispersion of resources. Thus, important expertise could be built up in one subsector, instead of having lower expertise in several areas. In general, the role of a solid science and technology policy seems to be decisive in fostering research activities in late industrializing countries. Singapore’s national innovation system may serve as a role model for other late industrializing countries which are on the same route, but at a lower stage of development. 160 7 Conclusion Previous research on R&D internationalization has provided valuable insights into the nature of scientific and technological activity conducted by MNEs outside their home countries. More specifically, previous literature has helped considerably in explaining the motivation for conducting overseas R&D, has developed different R&D organizational models and has analyzed the R&D internationalization process from a corporate perspective. Until now, however, little consideration has been given to the implications of R&D internationalization beyond the triad nations and on an R&D subsidiary level. This research area, however, is an important one, since non-traditional R&D locations are emerging as critical knowledge clusters. Little is known about the nature of R&D subsidiaries in these late industrializing countries. Therefore, this dissertation intends to provide a first step in this major, but neglected research area and to further increase interest in this important topic. It analyzed different R&D organizational models and in particular developed a framework for a metanational R&D organization, that is, an R&D organization which is also present in the periphery and leverages the technological hierarchy internationally. In order for the periphery to increase its strategic importance, technological capability upgrading towards more research activities is critical. This process of an increase in an R&D subsidiary’s technological sophistication is examined in detail. The late industrializing context in this study is Singapore, which is an emerging R&D hub. Singapore’s science and technology policy plays a decisive role in fostering R&D activities. Caution should be applied in interpreting the results of any research study, particularly when the investigation is exploratory in nature. Given the almost complete absence of prior studies of R&D subsidiaries in a late industrializing context, the results of this dissertation should be regarded as tentative until subsequent studies based on more sophisticated models, larger samples and different analytical approaches can either confirm or refute them. 161 This study has several limitations. First, the modest sample size limited the types of statistical analyses possible and limited the number of independent and control variables that could be considered. Therefore, only exploratory performance implications could be drawn for different international R&D organizations and only tentative conclusions could be drawn on key factors which influence the level of technological sophistication. Even though some effects were found insignificant, this may not be so because they lacked explanatory power but because the sample size did not produce adequate statistical power to detect them. Second, measurement issues are another limitation of this dissertation. Even though we attempted to adequately reflect the measures in this study, none of the measures used (for instance R&D performance) is without drawbacks. Third, in most cases (with the exception of the case studies and several other R&D subsidiaries), the R&D manager or managing director of the subsidiary was a single respondent. It would have been more ideal if for all the R&D subsidiaries under investigation in this dissertation several respondents could have been interviewed. However, this was practically not possible because R&D subsidiaries were usually not willing to provide more than one in-depth interview (with the exceptions as mentioned). Given these limitations, this study arrives at several tentative conclusions. Preliminary evidence is provided, showing that the metanational R&D organization is rare in corporate practice. However, it is a critical organizational form which can tap into knowledge residing in emerging, but so far non-traditional, R&D locations. Furthermore, it is important for R&D subsidiaries in these non-traditional R&D locations to upgrade their technological level and capabilities. Different technological paths entail different performance implications. Internal and external R&D network linkage and more specifically its interaction have a critical impact on R&D subsidiaries’ technological sophistication and strategic importance. Internal and external R&D network linkages are important for effective internal and external knowledge acquisition. This dissertation encourages further research to advance the work begun here. Future studies should investigate the concept of the metanational R&D organization further and should enhance the theoretical development of this new organizational form. On the 162 empirical side, future studies should be devoted to performance implications of the implementation of the metanational R&D organization in a larger sample. It would also be of considerable interest to further examine the technological capability upgrading of R&D subsidiaries in late industrializing countries. More studies analyzing this phenomenon are necessary to confirm the findings of this study. There is a need for clinical studies of R&D subsidiary evolution and more detailed examination of various aspects of the phenomenon, such as the interplay between R&D headquarters and R&D subsidiary management and the impact of host country policies on R&D subsidiary evolution. Future studies should also include more late industrializing countries in order to examine if the results of this dissertation also apply to other such countries with different degrees of economic development and to identify and analyze potential differences between late industrializing countries. For example, it would be interesting to examine differences in late industrializing countries of Asia versus South America, for example, in their approach to foster R&D activities. An entire research agenda should be devoted to the complex phenomenon of R&D output measurement. In this respect, a fruitful extension of this study would be an examination of several measures of R&D output. In order to evaluate the different patent applications adequately, for instance, one may have to consider the respective relevance of each patent for not every patent application seems to be equally relevant to the corporate R&D organization in terms of innovative contribution. One solution to this problem may lie in assigning different weights to different international patent applications, based on their probable contribution to corporate R&D output. This, however, raises the problem of determining such a probable contribution. Furthermore, future studies should investigate which performance measures, both on a national and international level, are most appropriate in evaluating a government’s efforts to foster R&D activities. There may be important implications for the concepts developed here, both for the role of R&D subsidiaries of MNEs in late industrializing countries and for the theory of the 163 MNE itself. Although it is too soon to predict how such research extensions will transpire, the hope is that this dissertation provides a first framework of concepts and ideas around which subsequent studies can be built. Consequently, the issues raised here for subsequent research provide ample scope for future studies in the field of international R&D management. 164 8 References Achilladelis, B. and Antonakis, N. 2001. The Dynamics of Technological Innovation: The Case of the Pharmaceutical Industry. Research Policy, 30: 535-88. Agency for Science, Technology and Research. 2001. National Survey of R&D in Singapore 2001. Singapore: Agency for Science, Technology and Research. Amabile, T. M. 1998. How to kill creativity. Harvard Business Review, 76(5): 76-87. Amabile, T. M. 1997. Motivating Creativity in Organizations: On Doing What You Love and Loving What You Do. California Management Review, 40(1): 39-58. Amabile, T. M., Conti, R., Heather, C., Lazenby, J. and Herron, M. 1996. Assessing the Work Environment for Creativity, Academy of Management Journal, 39(5): 115484. Amsden, A. and Tschang, T. 2003. A new approach to assessing the technological complexity of different categories of R&D (with examples from Singapore). Research Policy, 32: 553-72. Andersson, U., Forsgren, M. and Holm, A. 2002. The strategic impact of external networks: subsidiary performance and competence development in the multinational corporation. Strategic Management Journal, 23(11): 979-96. Ariffin, N. and Figueiredo, P.N. 2003. Internationalisation of Innovative Capabilities: Counter-Evidence from the Electronics Industry in Malaysia and Brazil. Paper presented at DRUID's Summer 2003 Conference, Copenhagen. Ariffin, N. and Figueiredo, P.N. 2001. Technological Capability Building and Innovation in the Electronics Industry: Evidence from Manaus: Research Programme on Technological Learning and Industrial Innovation, EBAPE-FGV. Asakawa, K. 2001. Evolving headquarters-subsidiary dynamics in international R&D: the case of Japanese multinationals. R&D Management, 31(1): 1-14. Barney, J. 1991. Firm Resources and Sustained Competitive Advantage. Journal of Management, 17(1): 99-120. Bartholomew, S. 1997. National Systems of Biotechnology Innovation: Complex Interdependence in the Global System. Journal of International Business Studies, 28(Second Quarter): 241-65. Bechhofer, F. and Paterson, L. 2000. Principles Of Research Design In The Social Sciences. London: Routledge. 165 Behrman, J.N. and Fischer, W. A. 1980. Overseas R&D activities of transnational companies. Cambridge, MA: Oelgeschlager, Gunn and Hain. Belderbos, R. 2003. Entry Mode, Organizational Learning, and R&D in Foreign Affiliates: Evidence from Japanese Firms. Strategic Management Journal, 24(3): 235-59. Belderbos, R. 2001. Overseas Innovations by Japanese Firms: An Analysis of Patent & Subsidiary Data. Research Policy, 30(2): 313-32. Belderbos, R. and Iwasa, T. 1999. Research And Development in Foreign Affiliates: Evidence From Japanese Firms. Working Paper, Netherlands Institute of Business Organization and Strategy Research. Bentz, V. M. and Shapiro, J. J. 1998. Mindful Inquiry in Social Research. Thousand Oaks, London, New Delhi: Sage. Birkinshaw, J. 1998. Entrepreneurship in Multinational Corporations: The Characteristics of Subsidiary Initiatives. Strategic Management Journal, 18(3): 207-29. Birkinshaw, J. and Hood, N. 1998. Multinational Subsidiary Evolution: Capability and Charter Change in Foreign-Owned Subsidiary Companies. Academy of Management Review, 23(4): 773-95. Birkinshaw, J. and Hood, N. 2001. Unleash Innovation in Foreign Subsidiaries. Harvard Business Review, March 2001: 131-37. Birkinshaw, J., Holm, U., Thilenius, P. and Arvidsson, N. 2000. Consequences of Perception Gaps in the Headquarters-Subsidiary Relationship. International Business Review, 9: 321-44. Birkinshaw, J., Hood, N. and Young, S. 2002. The Impact of Internal and External Competitive Forces on MNC Subsidiary Performance. Academy of Management Conference 2002: 1-26. Black, T.R. 1999. Doing Quantitative Research in the Social Sciences: An Integrated Approach to Research Design, Measurement, and Statistics. London: Sage. Blankenburg Holm, D., Eriksson, K. and Johanson, J. 1999. Creating value through mutual commitment to business network relationships. Strategic Management Journal, 20(5): 467-86. 166 Boutellier, R., Gassmann, O. and von Zedtwitz, M. 2000. Managing Global Innovation: Uncovering the Secrets of Future Competitiveness. Second revised ed. Berlin et al.: Springer-Verlag. Boutellier, R., Kloth, B. and Bodmer, C. 1996. Neue Organisationsformen globaler Forschung und Entwicklung. Zeitschrift Führung und Organisation, 65(5): 282-87. Brockhoff, K. 1998. Internationalization of Research and Development. Berlin et al.: Springer. Bryman, A. 1998. Quantitative and Qualitative Research Strategies in Knowing the Social World. In May, T. and Williams, M., editor, Knowing the Social World. Buckingham, Bristol: Open University Press. Cantwell, J. 1992. The Internationalization of Technological Activity and its Implications for Competitiveness. In Grandstrand, O., Hakanson, L. and Sjölander, S., editors, Technology Management and International Business: Internationalization of R&D and Technology. Chichester: John Wiley Sons. Cantwell, J. 1989. Technological Innovation and Multinational Corporations. Oxford, Cambridge: Basil Blackwell. Chakravarthy, B. 1997. A New Strategy Framework for Coping with Turbulence. Sloan Management Review(Winter): 69-82. Collins, S.M. and Bosworth, B. 1996. Economic Growth in East Asia: Accumulation versus Assimilation. Brookings Papers on Economic Activity, 2: 135-204. Coombs, J.E. and Deeds, D.L. 2000. International Alliances as Sources of Capital: Evidence from the Biotechnology Industry. Journal of High Technology Management Research, 11(2): 235-53. Cooper, A.C. and Dunkelberg, W.C. 1986. Entrepreneurship and Paths to Business Ownership. Strategic Management Journal, 7: 53-68. Cooper, D.R. and Schindler, P.S. 1998. Business Research Methods. 6th ed. Boston et al.: Irwin/McGraw-Hill. Coriat, B. and Dosi, G. 2002. The Nature and Accumulation of Organizational Competences/Capabilities. Revista Brasileira de Inovacao, 1(Julho/Dezembro): 275-326. Costa, I. and De Queiroz, S.R.R. 2002. Foreign direct investment and technological capabilities in the Brazilian industry. Research Policy, 31: 1431-43. 167 Creamer, D. B. 1976. Overseas research and development by United States multinationals, 1966-1975: Estimates of expenditures and a statistical profile. New York: The Conference Board. Daft, R.L. 1980. The Evolution of Organization Analysis in ASQ, 1959-1979. Administrative Science Quarterly, 25: 623-36. Dahlman, C. J. and Westphal, L.E. 1982. Technological Effort in Industrial Development - An Interpretative Survey of Research. In Stewart, F. and James, J., editors, The Economics of New Technology in Developing Countries. London: Frances Pinter. Dane, F. C. 1990. Research Methods. Pacific Grove: Brooks/Cole. Deeds, D., DeCarolis, D. and Coombs, J. 2000. Dynamic Capabilities and New Product Development in High Technology Ventures: An Empirical Analysis of New Biotechnology Firms. Journal of Business Venturing, 15: 211-29. De Meyer, A. 1993. Management of an International Network of Industrial R&D Laboratories. R&D Management, 23(2): 109-20. De Meyer, A. and Mizushima, A. 1989. Global R&D Management. R&D Management, 19(2): 135-46. Doz, Y. 2003. The Evolving Theory of the Multinational Firm. Paper presented at the Academy of Management Conference, Seattle, USA. Doz, Y., Asakawa, K. Santos, J. and Williamson, P. 1997. The Metanational Corporation. INSEAD Working Paper, 97/60/SM. Doz, Y., Santos, J., and Williamson, P. 2001. From global to metanational: how companies win in the knowledge economy. Boston: Harvard Business School Press. Dyer, G.W. and Wilkins, A.L. (1991). 1991. Better Stories, Not Better Constructs, to Generate Better Theory. Academy of Management Journal, 16(3): 613-19. Edler, J., Meyer-Krahmer, F. and Reger, G. 2002. Changes in the strategic management of technology: results of a global benchmarking study. R&D Management, 32(2): 149-64. Eisenhardt, K.M. 1991. Better Stories and Better Constructs: The Case for Rigor and Comparative Logic. Academy of Management Review, 15(3): 620-27. Eisenhardt, K.M. 1989. Building Theories from Case Study Research. Academy of Management Review, 14(4): 532-50. 168 Emory, C.W. and Cooper, D.R. 1991. Business Research Methods. 4th ed. Boston: Homewood. Ernst, H. 2001. Patent Applications and Subsequent Changes of Performance: Evidence form Time-Series Cross-Section Analyses on the Firm Level. Research Policy, 30(1): 143-57. Feinberg, S.E. and Majumdar, S.K. 2001. Technology Spillovers from Foreign Direct Investment in the Indian Pharmaceutical Industry. Journal of International Business Studies, 32(2): 421-37. Figueiredo, P. 2002. Does technological learning pay off? Inter-firm differences in technological capability-accumulation paths and operational performance improvement. Research Policy, 31(1): 73-94. Figueiredo, P.N. 2001. Technological Learning and Competitive Performance. Cheltenham, Northhampton: Edward Elgar Publishing Limited. Florida, R. 1997. The Globalization of R&D: Results of a Survey of Foreign-affiliated R&D Laboratories in the USA. Research Policy, 26: 85-103. Frost, T. 2001. The Geographic Sources of Foreign Subsidiaries' Innovations. Strategic Management Journal, 22: 101-23. Frost, T., Birkinshaw, J. and Ensign, P.C. 2002. Centers of Excellence in Multinational Corporations. Strategic Management Journal, 23(11): 997-1018. Gable, G.G. 1994. Integrating Case Study and Survey Research Methods: An Example in Information Systems. European Journal of Information Systems, 3(2): 112-26. Gassmann. 1997a. Internationales F&E-Management. München, Wien: Oldenburg Verlag. Gassmann, O. 1997b Management transnationaler Forschungs- und Entwicklungsprojekte: Eine empirische Untersuchung von Potentialen und Gestaltungskonzepten transnationaler F&E-Projekte in industriellen Grossunternehmen. Doctoral Dissertation, University of St. Gallen, Bamberg. Gassmann, O. and v. Zedtwitz, M. 1999. New Concepts And Trends in International R&D Organization. Research Policy, 28: 231-50. Gassmann, O. and von Zedtwitz, M. 1998. Organization of Industrial R&D on a Global Scale. R&D Management, 28(3): 147-61. Gauglitz-Lüter, S. 1998 Effektivitäts- und effizienzorientiertes Forschungs- und Entwicklungsmanagement. Doctoral Dissertation, Universität Köln, Lohmar. 169 Gerpott, T. 1990. Globales F&E-Management: Bausteine eines Gesamtkonzeptes zur Gestaltung einer weltweiten F&E-Organisation. Die Unternehmung, 44(4): 22646. Gerybadze, A. and Reger, G. 1999. Globalization of R&D: Recent Changes in the Management of Innovation in Transnational Corporations. Research Policy, 28: 251-74. Goh, A.B. 2003. Core Values. Singapore: Leica Instruments (S) Pte Ltd. Goh, A.B. 2002. Marching On. Singapore: Leica Instruments (S) Pte Ltd. Goh, A.B., Rotterdamm, R., Schulz, L. and Woon, C. 2000. Leica Microsopy: International Transfer of R&D Activities. In Boutellier, R., Gassmann, O. and von Zedtwitz, M., editors, Managing Global Innovation: Uncovering the Secrets of Future Competitiveness. 2nd ed. Berlin et al.: Springer. Granstrand, O. 1999. Internationalization of Corporate R&D: A Study of Japanese and Swedish Corporations. Research Policy, 28: 275-302. Granstrand, O., Hakanson, L. and Sjölander, S. 1993. Internationalization of R&D - A survey of some recent research. Research Policy, 22: 413-30. Grevesen, C. 2001 The Internationalization of Technological Activity and R&D Performance in Multinational Enterprises. Doctoral Dissertation, The State University of New Jersey, Newark, NJ. Gulati, R. 1998. Alliances and networks. Strategic Management Journal, 19: 293-317. Gulati, R. 1999. Network location and learning: the influence of network resources and firm capabilities on alliance formation. Strategic Management Journal, 20(5): 397420. Gulati, R., Nohria, N. and Zaheer, A. 2000. Strategic networks. Strategic Management Journal, 21(3): 203-15. Hage, J. and Hollingsworth, J.R. 2000. A Strategy for the Analysis of Idea Innovation Networks and Institutions. Organization Science, 21(5). Haley, U. and Low, L. 1998. Crafted Culture: Government Sculpting of Moderns Singapore and Effects on Business Environments. Journal of Organizational Change Management, 11(6): 530-53. Hewitt, G. 1980. Research and Development Abroad by U.S. Manufacturing Multinationals. KYKLOS, 33: 308-27. 170 Hobday, M. 1995. East Asian latecomer firms: learning the technology of electronics. World Development, 23(7): 1171-93. Huefner, J.C., Hunt, H.K. and Robinson, P.B. 1996. A Comparison of Four Scales Predicting Entrepreneurship. Academy of Entrepreneurship Journal, 1(2): 56-80. Intarakumnerd, P., Chairatana, P. and Tangchitpiboon, T. 2002. National innovation system in less successful developing countries: the case of Thailand. Research Policy, 31: 1445-57. Jick, T.D. 1979. Mixing Qualitative and Quantitative Methods: Triangulation in Action. Administrative Science Quarterly, 24(2): 602-11. Johanson, J. and Vahlne, J.E. 1977. The internationalization process of the firm: A model of knowledge development and increasing foreign commitments. Journal of International Business Studies, 8: 23-32. Kam, W. P., Kiese, M., Singh, A. and Wong, F. 2003. The Pattern of Innovation in Singapore's Manufacturing Sector. Singapore Management Review, 25(1): 1-34. Kim, L. 1998. From Imitation to Innovation: Dynamics of Korea's Technological Learning. Boston: Harvard Business School Press. Kuemmerle, W. 1997. Building effective R&D capabilities abroad. Harvard Business Review, 75(2): 61-70. Kuemmerle, W. 1999. Foreign Direct Investment in Industrial Research in the Pharmaceutical And Electronics Industries: Results Form A Survey of Multinational Firms. Research Policy, 28: 179-83. Kuemmerle, W. 1996 Home base and foreign direct investment in research and development - An investigation into to the international allocation of research activity by multinational enterprises. Doctoral Dissertation, Graduate School of Business Administration, Harvard University, Boston. Lall, S. 1992. Technological capabilities and industrialisation. World Development, 20(2): 165-86. Le Bas, C. and Sierra, C. 2002. 'Location versus home country advantages' in R&D activities: some further results on multinationals' locational strategies. Research Policy, 31: 589-609. Luethje, C. and Franke, N. 2003. The 'Making' of an Entrepreneur: Testing a Model of Entrepreneurial Intent Among Engineering Students at MIT. R&D Management, 33(2): 135-47. 171 Luo, Y. 2000. Dynamic Capabilities in International Expansion. Journal of World Business, 35(4): 355-78. Mahmood, I. and Singh, J. 2003. Technological Dynamism in Asia. Research Policy, 32: 1031-54. Mahoney, J.T and Pandian, J.R. 1992. The Resource-Based View within the Conversation of Strategic Management. Strategic Management Journal, 13(5): 363-89. Man, J. 2001. Creating Innovation. Work Study, 50(6): 229-33. Manor, B. and Tasi, W. 2001. Knowledge Transfer in Intraorganizational Networks: Effects of Network Position and Absorptive Capacity on Business Unit Innovation and Performance. Academy of Management Journal, 44(5): 996-1004. Matthews, J.A. 1999. A Silicon Island of the East. Creating a semiconductor industry in Singapore. California Management Review, 41(2): 55-78. May, T. 1997. Social Research: Issues, Methods and Processes. 2nd ed. Buckingham, Bristol: Open University Press. McEvily, B. and Zaheer, B. 1999. Bridging Ties: A Source of Firm Heterogeneity in Competitive Capabilities. Strategic Management Journal, 20: 1133-56. McKendrick, D.G., Doner, R.F. and Haggard, S. 2000. From Silicon Valley to Singapore: location and competitive advantage in the hard disk drive industry. Stanford: Stanford University Press. Medcof, J. 2001. Resource-based strategy and managerial power in networks of internationally dispersed technology units. Strategic Management Journal, 22(11): 999-1012. Medcof, J. W. 1997. A taxonomy of internationally dispersed technology units and its application to management issues. R&D Management, 27(4): 301-18. Meyer-Krahmer, F. and Reger, G. 1998. European Technology Policy and Internationalization: An Analysis behind the Background of the Innovation Strategies of Multinational Enterprises. Paper presented at DRUID Summer Conference on "Competence, Governance, and Entrepreneurship", Bornholm. Mitchell, T.R. 1985. An Evaluation of the Validity of Correlational Research conducted in Organizations. Academy of Management Review, 10(2): 192-205. Mutinelli, M. and Piscitello, L. 1998. The Entry Mode Choice of MNEs: An Evolutionary Approach. Research Policy, 27: 491-506. 172 Nobel, R. and Birkinshaw, J. 1998. Innovation in Multinational Corporations: Control and Communication Patterns in International R&D Operations. Strategic Management Journal, 19: 479-96. Novartis. 2003. Novartis Institute for Tropical Diseases - Mission. Internet: http://www.nitd.novartis.com/mission/index_mission.shtml as of August 30, 2003. Oldham, G.R. and Cummings, A. 1996. Employee Creativity: Personal and Contextual Factors at Work. Academy of Management Journal, 39(3): 607-34. Patel, P. and Vega, M. 1999. Patterns of Internationalization of Corporate Technology: Location vs. Home Country Advantages. Research Policy, 28: 145-155. Pearce, R.D. and Singh, S. 1992. Globalizing research and development. London and Hong Kong: Macmillan in association with the Graduate School of European and International Studies, University of Reading. Pearce, R.D. 1989. The Internationalization of Research and Development by Multinational Enterprises. London: Macmillan Press in assocation with the Graduate School of European and International Studies of Reading. Pearce, R.D. and Pooni, G.S. 1996. The Globalisation of R&D in Pharmaceuticals, Chemicals and Biotechnology: Some New Evidence. Discussion Papers in International Investment and Business Studies, Series B, 8(213): 1-33. Peebles, G. and Wilson, P. 2002. Economic Growth and Development in Singapore: Past and Future. Cheltenham, Northhampton: Edward Elgar. Penner-Hahn, J.D. 1998. Firm And Environmental Influences on the Mode and Sequence of Foreign Research and Development Activities. Strategic Management Journal, 19(2): 149-68. Powell, W.W., Koput, K.W. and Smith-Doerr, L. 1996. Interorganizational Collaboration and the Locus of Innovation: Networks of Learning in Biotechnology. Administrative Science Quarterly, 41(March 1996): 116-45. Reger, G. 1999. How R&D is coordinated in Japanese and European Multinationals. R&D Management, 29(1): 71-88. Remenyi, D., Williams, B., Money, A. and Swartz, E. 1998. Doing Research in Business and Management: An Introduction to Process and Method. London, Thousand Oaks, New Delhi: Sage. 173 Ronstadt, R.C. 1978. International R&D: The Establishment and Evolution of Research and Development by Seven U.S. Multinationals. Journal of International Business Studies, 9(1): 7-24. Ronstadt, R.C. 1977. Research and Development Abroad by U.S. Multinationals. New York: Praeger Books. Schumpeter, J. 1934. The Theory of Economic Development. Cambridge: Harvard University Press. Serapio, M.G. and Dalton, D. 1993. Foreign R&D Affiliates in the United States. Research Technology Management, 36(6): 33-39. Serapio, M.G. and Dalton, D. 1999. Globalization of Industrial R&D: An Examination of Foreign Direct Investments in R&D in the United States. Research Policy, 28: 303-16. Shalley, C.E., Gilson, L.L. and Blum, T.C. 2000. Matching Creativitiy Requirements and the Work Environment: Effects on Satisfaction and Intentions to leave. Academy of Management Journal, 43(2): 215-23. Shan, W. and Song, J. 1997. Foreign Direct Investment and the Sourcing of Technological Advantage: Evidence from the Biotechnology Industry. Journal of International Business Studies, 28(2): 267-84. Shane, S.A. 1992. Why do some societies invent more than others? Journal of Business Venturing, 7: 29-46. Smith, J. 2003. Building an Entrepreneurial Knowledge Culture in a National Research Laboratory. R&D Management, 33(2): 231-37. Snow, C.C. and Thomas, J.B. 1994. Field Research Methods in Strategic Management: Contributions to Theory-Building and Testing. Journal of Management Studies, 31(4): 457-80. Song, J. 2002. Firm Capabilities and Technology Ladders: Sequential Foreign Direct Investments of Japanese Electronics Firms in East Asia. Strategic Management Journal, 23(3): 191-210. Stuart, T.E. 1998. Network Positions and Propensities to Collaborate: An Investigation of Strategic Alliance Formation in a High-Technology Industry. Administrative Science Quarterly, 43: 668-98. 174 Swinbanks, D. 1997. Singapore's Government will double its R&D investment over next five years: Aiding Private Firms/Universities. Research Technology Management, 40(1): 2-3. Tallmann, S. 1992. A Strategic Management Perspective on Host Country Structure of Multinational Enterprises. Journal of Management, 18(3): 455-71. Usbasaran, D., Wright, M. and Westhead, P. 2003. A Longitudinal Study of Habitual Entrepreneurs: Starters and Acquirers. Entrepreneurship & Regional Development, 15(July-September): 207-28. Vereecke, A., Van Dierdonck, R. and De Meyer, A. 2002. A typology of plants in global manufacturing networks. INSEAD Working Paper, 67. Von Boehmer, A. 1995. Internationalisierung industrieller Forschung und Entwicklung: Typen, Bestimmungsgründe und Erfolgsbeurteilung. Wiesbaden: Gabler. Von Zedtwitz, M. 1999 Managing Interfaces in International R&D. Doctoral Dissertation, University of St. Gallen, Bamberg. Von Zedtwitz, M. and Gassmann, O. 2002. Market versus technology drive in R&D internationalization: four different patterns of managing research and development. Research Policy, 31: 569-88. Wernerfelt, B. 1984. A Resource-Based View of the Firm. Strategic Management Journal, 5: 171-80. Wess, L. 2002. Regional Development: From Pill to Pillow. The Bernstein Report on BioBusiness: BioCentury: 1-8. Westhead, P. 1997. R&D 'inputs' and 'outputs' of technology-based firms located on and off Science Parks. R&D Management, 27(1): 45-62. Westney, D.E. 1993. Cross-Pacific Internationalization of R&D by US and Japanese Firms. R&D Management, 23(2): 171-82. Woodman, R.W., Sawyer, J.E. and Griffin, R.W. 1993. Toward a Theory of Organizational Creativity. Academy of Management Journal, 18(2): 293-321. Yin, R. 1981. The Case Study Crisis: Some Answers. Administrative Science Quarterly, 26(1): 58-65. Yin, R. 1989. Case Study Research: Design and Methods. Newbury Park: Sage. Yong, J.A. 2003. PhD pool to be expanded with new scholarships, The Straits Times. January 4, 2003, Singapore. 175 Yoshida, P.G. 2001. Asian Economies Striving to Enhance Innovation Capabilities. Research Technology Management, 44(1): 2-6. Young, A. 1995. The tyranny of numbers: confronting the statistical realities of the East Asian growth experience. Quarterly Journal of Economics, 110(3): 641-80. Zander, I. 1998. The Evolution of Technological Capabilities in the Multinational Corporation - Dispersion, Duplication and Potential Advantages from Multinationality. Research Policy, 27: 17-35. Zhang, Y. 2000. Using the Internet for Survey Research: A Case Study. Journal of the American Society, 51(1): 57-68. 176 9 Appendix 9.1 Questionnaire as a Basis for the In-Depth Interviews Survey: R&D Investment in Asia University of St. Gallen, Asia Research Center Prof. Dr. Li Choy Chong Email: [email protected] Yvonne Helble, Research Associate Email: [email protected] Handphone: 9729-6769 Name: Position: Date: Company: Location: Industry: Section A: Motivation for R&D activities in Singapore 1. What are the reasons for your firm to conduct R&D activities in Singapore? Please indicate the strength of your agreement or disagreement. Strongly disagree 1. To adapt products and services to local requirements and conditions Strongly agree 1 2 3 4 5 2. To access cutting-edge science & technology specific to the host country 1 2 3 4 5 3. To access top researchers and scientists 1 2 3 4 5 4. To access a creative workforce 1 2 3 4 5 5. To learn from foreign lead markets or lead customers 1 2 3 4 5 6. To take advantage of technology developed by a foreign firm 1 2 3 4 5 7. To keep abreast of foreign technologies 1 2 3 4 5 8. To comply with local market access regulations or pressures 1 2 3 4 5 1 2 3 4 5 (e.g. to support non-domestic production) 9. To be close to major markets 10. To take advantage of government support 1 2 3 4 5 11. To take advantage of favorable infrastructure and 1 2 3 4 5 12. To compensate for an inappropriate environment at home 1 2 3 4 5 13. To develop new research capabilities 1 2 3 4 5 14. To comply with M&A (R&D site in Singapore is outcome of M&A) 1 2 3 4 5 1 2 3 4 5 regulatory framework (e.g. public R&D programs and institutions) 15. Other: Please specify: 177 Section B: R&D Unit Profile 1. Please indicate in what year your R&D site in Singapore was founded: 2. Please state the year and the location of your firm’s first R&D site established outside the firm’s home country: Year Location: 3. Was there any other firm activity in Singapore before R&D activities? Yes (R&D site is outcome of previous activities in Singapore) No Yes (R&D site is independent of previous activities in Singapore) 4. If yes, what were your firm’s first activities in Singapore? Sales Manufacturing Logistics Other form of activity: 5. Total number of employees at R&D unit in Singapore: with the following breakdown: • Number of technical personnel at technician level: Number of personnel at research scientist/engineer level: • Number of non-technical personnel at management level: Number of non-technical personnel below management level: 6. Ratio of local R&D expenditure (in Singapore) as percentage of global corporate R&D expenditure: 7. Ratio of international R&D expenditure versus domestic R&D expenditure (at headquarters): 178 Section C: Evolution of the R&D Unit Please indicate the different steps taken by your firm in the evolution of the R&D site and specify the year for each stage of evolution. If a certain stage of evolution does not apply, please skip. If a certain stage of evolution is not included, please add. Main Task of R&D site Level Form of Evolution Market Support Local Support by the Singaporean government Regional Collaboration (with local firms or research institutions) Global Independent activities Other form: Please specify: Manufacturing Support Local Support by the Singaporean government Regional Collaboration (with local firms or research institutions) Global Independent activities Other form: Please specify: Advanced Development Local Support by the Singaporean government Regional Collaboration (with local firms or research institutions) Global Independent activities Other form: Please specify: Exploratory Development Local Support by the Singaporean government Regional Collaboration (with local firms or research institutions) Global Independent activities Other form: Please specify: Applied Research Unit Local Support by the Singaporean government Regional Collaboration (with local firms or research institutions) Global Independent activities Other form: Please specify: Basic Research Unit Local Support by the Singaporean government Regional Collaboration (with local firms or research institutions) Global Independent activities Other form: Please specify: Other stage of evolution: Local Support by the Singaporean government Please explain: Regional Collaboration (with local firms or research institutions) Global Independent activities Other form: Please specify: 179 Year Section D: R&D Unit’s Interaction with Internal and External Parties 1. How does your R&D site in Singapore interact with other R&D facilities in your firm? a) Human Resources (HR) Flows Please indicate the number of R&D personnel transfers for 1997 and 2001: 1997___ 2001__ Headquarters Your R&D 1997___ 2001 1997__ _2001 Other R&D subsidiary site(s) Strongly disagree 1. Strongly agree To what extent does your R&D site have influence on the acquisition of human resources? 1 2 3 2. To what extent does your R&D site have influence over the development of human resources? 3. To what extent do you conduct training for your R&D personnel? 4 5 1 2 3 4 5 1 2 3 4 5 b) Innovation Flows Please indicate the flow of the number of innovative projects and products for 1997 and 2001: 1997___ 2001__ Headquarters Your R&D 1997___ 2001 1997__ _2001 Other R&D subsidiary site(s) Strongly disagree 4. Strongly agree To what extent does your R&D site participate in the global R&D program of your R&D organization? 1 2 3 4 5 1 2 3 4 5 3 4 5 5. To what extent can you initiate own R&D projects? 6. To what extent are you the recipient of core technologies from HQ or other R&D sites? 1 2 7. To what extent do you conduct R&D activities in a field where HQ or other R&D sites have no expertise? 8. To what extent is the innovation locus in your R&D organization equally balanced? 1 1 180 2 2 3 4 5 3 4 5 c) Information Flows Please indicate the respective % for the direction of the information flows for 1997 and 2001. % of direction 1997___2001 % of direction 1997___ 2001__ Headquarters % of direction 1997___ 2001 Your R&D Other R&D subsidiary site(s) Strongly disagree 9. Strongly agree To what extent does your R&D site have to follow rules and regulations by HQ? 1 2 3 4 5 2 3 4 5 2 3 4 5 10. To what extent can your R&D site engage freely in external research collaborations? 1 11. To what extent can your R&D site interact freely with other R&D sites? 1 2. External to your R&D site with which parties do you collaborate? Time period of collaboration (in years, e.g. 1997-2002) Party 1: Local Research Institution: Party 2: Local Firm: Party 3: Multinational Firm: Party 4: Government: Party 5: Other external party: Please specify: 3. Referring to your R&D unit’s collaboration with external parties, please indicate the strength of your agreement or disagreement for the following statements. Strongly disagree 1. External parties generate new knowledge for our R&D unit. 2 3 4 5 2. External parties provide complementary knowledge for our R&D unit.1 2 3 4 5 3. External parties provide local infrastructure for our R&D unit. 1 2 3 4 5 4. Our R&D unit relies on innovations from these external parties. 1 2 3 4 5 5. These external projects are strictly controlled by our R&D unit. 1 2 3 4 5 6. A mutual information flow between R&D unit and external parties exists. 1 2 3 4 5 7. Internal parties’ interaction is linked to external parties’ interaction. 1 3 4 5 181 1 Strongly agree 2 8. Frequent internal parties’ interaction is conducive to initiating external parties’ interaction. 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 9. The centrality of the R&D unit within the internal parties is conducive to initiating external parties’ interaction. 10. Interaction with internal parties significant to the R&D unit inhibits external parties’ interaction. Additional Comments to Section D: Section E: R&D Unit Performance Please provide the following information of your firm’s R&D site in Singapore for as many years as possible: Year Number Number Number of Contributions of R&D site in Number of Ratio of of new of patent external Singapore to the overall corporate research output of product appli- publications R&D organization on a local, projects of R&D site in develop cations (if applicable regional, or global level (please R&D site in Singapore as according to give an estimate in %) Singapore for % of overall company the overall corporate policy) corporate R&D out put ments R&D organization 1995 Local: Regional: Global: 1996 Local: Regional: Global: 1997 Local: Regional: Global: 1998 Local: Regional: Global: 1999 Local: Regional: Global: 2000 Local: Regional: Global: 2001 Local: Regional: Global: 2002 Local: Regional: Global Does your company have other R&D facilities in Asia? If yes, where: India China Other locations, please specify: Please indicate if you would like to have a report of this study’s research findings: Yes No Thank you very much for your participation! 182 9.2 Open-ended questions asked during the in-depth interviews Section A, B, C: 1. How did you acquire/build up your first resources for the R&D subsidiary? 2. How did you move from one technological stage to the next, for instance from manufacturing support to advanced development? 3. How important was the role of the Singapore government in helping your R&D site to upgrade and build its level of technological sophistication? Concluding Questions: 1. What are your lessons learnt in building up/managing your R&D site? 2. What do you consider major future management challenges? 183 9.3 Definition of Technological Stages provided for Section C of the Questionnaire Market support: Customer support and/or the adaptation of already established product technology to particular customer requirements, carried out by the R&D unit in collaboration with the marketing unit, without significant collaboration from manufacturing. Manufacturing support: Adaptation of an already established process technology to some particular condition, usually to improve the manufacturing process, carried out in tandem by the R&D unit and manufacturing, but without significant cooperation from marketing. Product development/Advanced development: Development of manufacturable and commercially viable new products with the objective of immediate market results; techniques used include engineering design tools including simulation and testing. Process development/Exploratory development: Development of a new and commercially viable process with the research objective of implementing this process as engineered system and to deliver short-term market results; techniques employed comprise engineering design tools including simulation, but not testing. Applied research: Application of scientific techniques in order to find a differentiated product for a specific market with the objective of transforming and reapplying a known concept for a new application. The output is a differentiated product for a specific market with intellectual property, which is created in the medium and short-term. Basic research: Discovery of new knowledge for new marketable products on a longterm horizon; output is product-based research for transfer to applied research or exploratory development. Scientific techniques are applied by a highly qualified R&D personnel. (based on Amsden and Tschang, 2003; Medcof, 1997) 184 9.4 Letter Asking for Interview Participation Prof. Dr. Li Choy Chong Email: [email protected] Phone Number: 6822-0718 Yvonne Helble, Research Associate Email: [email protected] Handphone: 9729-6769 Company Address Date…….. Dear Ms./Mr. …….., the Asia Research Center at the University of St. Gallen, Switzerland, is currently conducting an academic study on R&D internationalization. The motivation behind this research stems from the fact that multinational firms have increasingly internationalized their R&D activities. This R&D internationalization process has been expanding to countries such as Singapore, China, and India. As a result, corporate R&D organizations are ever more international, exposing R&D managers to important management challenges. The objective of this research is to examine the underlying motivation, process, and performance implications of such an R&D investment in Singapore. It also attempts to collect information on how subsidiary R&D sites are embedded both in the overall corporate R&D organization and in the local context. On a managerial level, this research project aims to provide useful insights for R&D managers on how to manage subsidiary R&D facilities most effectively, and how to integrate them into both the corporate R&D organization and the local research environment. In investigating these issues, this research project examines R&D facilities in Singapore. 185 Since your company conducts R&D activities in Singapore, you are in a good position to contribute to this research project. Therefore, it would be great if you could let us know if a short meeting of about 30, maximum 45 minutes, were possible in the next two weeks. This research project is an academic study. The information you provide will be dealt with complete confidentiality. No information will be disclosed with a specific link to a specific firm. Any publication will refer to the whole sample, not to individual firms. We thank you very much in advance for your kind cooperation. In return for your efforts, we would be more than happy to provide you with a report of the study’s research findings free of charge once the study is completed. This report will give you good insights into the current R&D management practices of major multinational firms. We look forward to your reply and thank you very much in advance. If you have any further questions, please do not hesitate to contact us. Best regards, Prof. Dr. Li Choy Chong Yvonne Helble 186 9.5 Interview Partners No. Company 1 ABB 2 3 4 Name Mr. Schmaderer Title Date Location Vice President R&D 11/22/2002 Switzerland China and Singapore (Phone Interview) Addvalue Mr. Kay Wee Kiat Manager Business 10/22/2002 Singapore Communications Pte Ltd, Development former ATT Agilent Technologies Mr. Wong Chee R&D Manager 11/27/2002 Singapore (Singapore Vision Keong Operation) Pte Ltd ASTI Holdings Limited Mr. Woo Kwek Chief Financial 1/29/2003 Singapore Kiong Officer 5 Aventis Pharma International S.A. 6 Becton Dickinson Dr. David Capes Medical Products Pte Ltd Borland (Singapore) Pte Dr. George Yuan Ltd 7 Dr. Peter Hodsman Medical Director North & South Asia 10/18/2002 Singapore R&D Director 10/31/2002 Singapore Asia Pacific Region Vice President, R&D 11/11/2002 Singapore 8 Cerebos Pacific Ltd 9 Chartered Semiconductor Dr. Lap Chan Mfg Pte Ltd Director 10/24/2002 Singapore University/Research Institute 10 Chartered Silicon Mr. Tom Joy Partners Pte Ltd, Joint Venture with Agilent Technologies Ciba Specialty Chemicals Mr. Kam-ho Tan (Singapore) Pte Ltd. Deputy Director Yield Enhancement 1/30/2003 Singapore 4/17/2002 Singapore and 8/27/2002 1/27/2003 Singapore 11 Dr. Ono Hiroyuki Vice President R&D 10/29/2002 Singapore and Dr. Daniel Tsi and Manager Scientific Research 12 Ciba Specialty Chemicals Dr. Ekkundi (India) Pte Ltd. Regional Manager Asia Pacific Technical Centre, Segment Plastic Additives Head, Research Technology 13 Covance (Asia) Pte Ltd Dr. Philip J. Masters Laboratory Director 14 DaimlerChrysler South East Asia Pte Ltd Dr. Udo F. Loersch Vice President External Affairs 187 8/9/2002 India (Survey) 4/16/2002 Singapore 15 DuPont Singapore Pte Ltd 16 Technical Manager 10/31/2002 Singapore Ericsson Dr. Andreas Telecommunications Pte Fasbender Ltd Ericsson (China) Pte Ltd Dr. Wang Director Ericsson Cyberlab Singapore Director China Research Lab 9/26/2003 Singapore 18 ExxonMobil Chemical Operations Pte Ltd Mr. Dave Beattie 1/30/2003 Singapore 19 GE Aviation Service Operation Pte Ltd Mr. Chen Keng Nam 20 Gemplus Technologies Asia Pte Ltd Mr. Michel Escalant Planning and Technology Supervisor Singapore Olefins Business Leader Technology and Engineering Technical Director Singapore R&D 21 GES S'pore Pte Ltd R&D Manager 2/18/2003 Singapore 23 Glaxo Wellcome Manufacturing Pte Ltd Dr. Jim E. Plant 11/20/2002 Singapore Director Flavor Science and CAO/Regional Finance Director Director of Technical 11/15/2002 Singapore Development 24 GSL Group Sense Technology (Singapore Pte Ltd) 25 IBM Zürich Research Laboratory, Rüschlikon Mr. Stuart Tan Hua Business 11/21/2002 Singapore Koon and Mr. Eng Development & Chong Meng Technical Support Manager and Engineering Director Dr. Phillippe Vice President of the 2/11/2002 Switzerland Janson IBM Academy of Technology 26 IBM Singapore Pte Ltd Mr. Chin Yook Sing 27 Infineon Technologies (Asia Pacific) Pte Ltd Mr. Michael Tiefenbacher 28 Infineon Technologies (Asia Pacific) Pte Ltd Mr. Michael Tiefenbacher 29 Kenwood Electronics Mr. K. H. Tan Technologies (S) Pte Ltd Leica Instruments Mr. A. B. Goh (Singapore) Pte Ltd Leica Instruments Ms. Germaine Tan Senior Manager (Singapore) Pte Ltd R&D 17 22 30 31 Ms. Hoe Kum Yoke Mr. Yeow Kim Chai Givaudan Singapore Pte Mr. Willi Grab and Ltd Mr. Stefan Giezendanner Director IBM Emerging Technology Centre Vice President Development Center Singapore Vice President Development Center Singapore Manager (HR / Administration) Managing Director 188 9/30/2002 China (Survey) 10/29/2002 Singapore 11/5/2002 Singapore 11/1/2002 Singapore 4/18/2002 Singapore 9/26/2002 Singapore 10/5/2001 Singapore 11/22/2002 Singapore 2/11/2003 Singapore 32 Lilly Systems Biology Pte Ltd Dr. Santosh K. Mishra Managing Director 11/25/2002 Singapore 33 Lilly Systems Biology Pte Ltd Dr. Santosh K. Mishra Managing Director 2/18/2003 Singapore 34 3M Asia Pacific Pte Ltd Mr. Howard D. Tam Laboratory Manager 11/26/2002 Singapore 3M Innovation Center 35 Matsushita Refrigeration Mr. Khoo Chew Industries (S) Pte Ltd Thong Senior Manager R&D Center 11/14/2002 Singapore 36 Micron Semiconductor Asia Pte Ltd Mr. Jen Kwong Hwa Managing Director 2/20/2003 Singapore 37 Mitsubishi Chemical Infonics Pte Ltd Mr. Yoshiyuki Kisaka Managing Director 10/30/2002 Singapore 38 Molex Singapore Pte Ltd Mr. Yeo Khee Teck 39 National Starch and Mr. Joseph M. Chemical (Singapore) Pte Light Ltd 40 Natsteel Ltd Dr. Josephine Kwa 41 NEC Electronics Singapore Pte Ltd Ms. Helen Chua 42 43 Nestlé R&D Centre Pte Ltd Nestlé/Nestec Ltd 44 Nestlé/Nestec Ltd 45 Novartis Pharma AG Dr. Singh and Mr. Wissgott Mr. Helio Waszyk and Mr. Carl Branscom Mr. Helio Waszyk and Mr. Carl Branscom Prof. Dr. Paul Herrling 46 Novartis Pharma AG Prof. Dr. Paul Herrling Manager DIE Design 10/12/2002 & Development & Product Development Support Director Technical 11/20/2002 Services & Applications, Food Asia Pacific Executive Vice 12/9/2002 President Technology Senior Consulting 2/27/2003 Manager Strategic Planning and Marketing Division Managing Director 4/15/2002 and Group Manager Corporate Research 4/30/2002 and Development Management Corporate Research 9/11/2002 and Development Management 10/28/2002 Head of Corporate Research Professor for Drug Discovery Science University of Basel Head of Corporate 1/31/2003 Research Professor for Drug Discovery Science University of Basel 189 Singapore Singapore Singapore Singapore Singapore Switzerland Switzerland (Survey) Singapore Singapore 47 Novartis Institute for Tropical Diseases (NITD) Pte Ltd Novartis Institute for Tropical Diseases (NITD) Pte Ltd Novartis Pharma AG Dr. Thomas Keller Head of Chemistry 2/21/2003 Singapore Dr. Thomas Keller Head of Chemistry 2/24/2003 Singapore Dr. Richard Harrison 2/27/2002 Switzerland 50 Oki Techno Centre (Singapore) Pte Ltd Mr. Yutaka Kumagai 51 Olympus Technologies Singapore Pte Ltd 48 49 52 Head of Staff of Novartis Pharma Research Managing Director Mr. Goh Soh Lian Division Manager (Finance & Administration) Pharmacia Singapore Pte Dr. Melvyn Teillol- Regional Senior Ltd Foo Director - Asia Pacific 11/8/2002 Singapore 10/24/2002 Singapore 2/25/2003 Singapore 53 Philips Electronics Singapore Pte Ltd Mr. Ferdinand Coehoorn 54 Philips Consumer Electronics Mr. Piet Coelewij 55 Philips Software Centre Private Ltd Dr. Bob Hoekstra 56 Prima Ltd Mr. Lim Kay Kong Group R&D Manager 1/28/2003 Singapore 57 Quintiles East Asia Pte Ltd 2/20/2003 Singapore 58 Rhodia Asia Pacific Pte Ltd Dr. Brian O'Keeffe President Product Development Asia Pacific Dr. Ji Li Regional Technical Director 59 Roche Diagnostics Asia Pacific Pte Ltd Mr. Ralph E. Graichen Technology Officer 9/17/2002 Singapore 60 Schneider Electric Industrial Development Singapore Pte Ltd SembCorp Engineers and Constructors Pte Ltd Serial System Ltd Mr. Jean-Marie Periot Managing Director 11/26/2002 Singapore Dr. Lim Teck Yong, Daniel Mr. Chin Yeow Hon Vice President (Civil) 11/5/2002 Singapore 61 62 63 Senior Manager 10/17/2003 Singapore Personal Appliances & Personal Care R&D Garment Care Senior Vice President 1/20/2003 Singapore General Manager BCT MTV PS& P and Marketing BCUTV CEO Philips 10/21/2002 Singapore Innovation Campus, India 9/25/2002 Singapore 2/18/2003 Singapore Serono Singapore Pte Ltd Dr. Theodor Wee Tit Gin Regional Medical 4/12/2002 Singapore Manager Asia Pacific Region 190 64 Shell Global Solutions (Singapore) (Pte) Ltd Mr. Eric Holthusen Managing Director & 11/8/2002 Singapore Fuels Manager 65 Siemens Pte Ltd Mr. Thomas Frischmuth 66 Siemens VDO Automotive Pte Ltd Mr. Azmoon Ahmad 67 Siemens Pte Ltd 68 Siemens Medical Instruments Pte Ltd 69 Siemens China Pte Ltd, SSMC/ICM/MO Dr. Borger President 70 Singapore Research Laboratory (Sony) 71 ST Aeorospace Dr. Kanzo Okada and Mr. Tong Kok Leong Mr Lim Tai Fui Division Director and 11/11/2002 Singapore Assistant General Manager Senior Vice President 3/3/2003 Singapore 72 Stratech Systems Pte Ltd Ms. Evelyn Soh 73 Sumitomo Bakelite Singapore Pte Ltd SurroMed Pte Ltd now Institute of Bioengineering Teraoka Weigh-System Pte Ltd Mr. Tan Tat Hong R&D Manager 12/19/2002 Singapore Ms. Ting Dor Ngi 11/13/2002 Singapore 76 Thomson Multimedia Asia Pte Ltd Mr. Gerard Dongois 77 United Test & Assembly Mr. C. K. Tan Centre Singapore Pte Ltd 78 Volume Interactions Pte Ltd, acquired by Bracco Group Xerox Singapore Software Centre Fuji Xerox Asia Pacific Pte Ltd Y3 Technologies Vice President, Science & Technology R&D Division Manager, Techno Centre General Manager TV & Video Product Development Vice President Worldwide Business Strategy President & Chief Technology Officer Technical Program Manager 11/18/2002 Singapore 74 75 79 80 Managing Director 7/15/2002 Singapore Vice President, R&D 11/12/2002 Singapore Chief Operating Officer Dr. Chua Kee Head of Department 4/12/2002 Singapore Chaing ICM Mobile Core R&D Dr. Anthony Chay R&D Manager 9/11/2002 Singapore Mr. Liaw Fong Chong Dr. Luis Serra Mr. Lui Kok Kwang 10/15/2002 China (Survey) Senior Vice President 2/13/2003 Singapore (HR and Administration) 1/30/2003 Singapore 11/12/2002 Singapore 1/21/2002 Singapore 1/29/2003 Singapore Mr. Eng Chye Yeo Manager Regional IT 2/24/2003 Singapore and Operation 191 Interviews with Government Institutions No. Company 81 Agency for Science, Technology and Research 82 Agency for Science, Technology and Research Name Dr. Lim Khiang Wee Dr. Jasbir Singh 83 Singapore Institute of Dr. Christopher Manufacturing Technology John Holmes Title Director Science and Engineering Council Head Computational, Mathematical & Physical Sciences Section Research Fellow Date Location 4/18/2002 Singapore Several Singapore meetings in 2002 and 2003 10/8/2002 Singapore 11/13/2002 Singapore 84 Institute of Bioengineering Ms. Dor Ngi Ting Vice President, Science & Technology Nanotechnology Laboratory 85 Economic Development Dr. Swan Gin Beh 2nd Director 3/1/2003 Singapore Board Biomedical Sciences 192 CURRICULUM VITAE Yvonne Elise Helble, born September 23, 1975 in Balingen, Germany. EDUCATION 04/2001 – 09/2003: PhD Program in International Management with a Focus on Asia at the University of St. Gallen, Switzerland, as a PhD Scholar of the Cusanuswerk e.V. 08/2002 – 07/2003: Visiting Research Associate at INSEAD, Singapore and Fontainebleau, France and at the Wharton-SMU Research Center, Singapore, as a PhD Scholar of the Swiss National Fund. 10/1995 – 07/2001: Master's Program of International Business Administration at the Otto-Friedrich-University of Bamberg, Germany. 08/1998 – 05/2000: MBA/MA in Economics Dual Degree Program at the University of Delaware, Newark, USA as a Scholar of the Federation of German and American Clubs. PROFESSIONAL EXPERIENCE 04/2001 – 05/2002: Research Assistant at the Institute for International Management (FIM), Asia Research Center, University of St. Gallen, Switzerland. 05/2000 – 01/2001: ‘Werksstudentin’ in Group Development – Regional Strategies at Siemens AG Transportation Systems, Erlangen, Germany. 06/1999 – 08/1999: Loan Administrator in the Global Loan Support Service at Citibank, New Castle, USA. 08/1997 – 10/1997: Internship in the Materials Management Center of the Plastics Business Unit at BASF AG, Ludwigshafen, Germany. 03/1997 – 04/1997: Marketing Assistant Consultant at KPMG Fidorga, Caen, France. 08/1996 – 10/1996 : Internship in the Human Resources and Production Department at Salmson/WILO AG, Laval/Paris, France. 193