One Way or Two Way Globalization? The Double-network Theory and Chinese R&D Internationalization

Transcription

1 One Way or Two Way Globalization? The Double-network Theory and Chinese R&D Internationalization Jieyin Zhang La Scuola Superiore Sant' Anna di Pisa Xiaohong Quan San José State University Alberto Di Minin University of California, Berkeley BECOME A MEMBER OF THE INDUSTRY STUDIES ASSOCIATION BY VISITING

2 One Way or Two Way Globalization?-The Double-network Theory and Chinese R&D Internationalization ABSTRACT In this paper we claim that the phenomenon of R&D internationalization of Chinese firms is an emerging and varied phenomenon. We find evidence that both different modes (experiential and cooperative) and different goals (explorative and exploitative) coexist. Our qualitative study is designed to identify and describe this phenomenon, showing its richness and complexity. Our findings also set the stage for further studies in the causalities and managerial implications. 1. INTRODUCTION In a globalized economy, the knowledge creation processes of technology-based companies have become increasingly global (Gassmann and von Zedtwitz, 1999: p.231). According to WIR (2005) data, in the past few decades, FDI into developing and transition countries has increased, and China has become the most attractive host country for FDI. This is partly attributable to the face that, China has worked hard in recent years to attract foreign investment in R&D with the aim of enhancing the technology capabilities of Chinese firms (Wu and Callahan, 2005). In addition, Chinese MNCs (Multinational Corporations) have also begun to expand overseas (Tung, 2005). In the dynamic global competitive environment, this paper attends to understand how the emerging MNCs from developing countries can find innovation opportunities and conduct their international R&D activities globally to obtain key resources and knowledge through their network organization. This raises the question: do the emerging MNCs from developing countries adopt different learning strategies than the MNCs from developed countries in the R&D internationalization process? During the last decade various scholars have adopted a macroscopic perspective to investigate FDI from emerging countries (Hong and Sun, 2006; Taylor, 2002; Xie and White, 2006). They found that there are two waves of MNCs from the developing world, each with a different push. The first wave was driven outward by the difficulties and restrictions local firms faced in their home countries, and the other one is mainly the result of the rising pressures of the global economy (Mathews, 2006b). The conventional theories, which seek to explain the reasons behind the first wave of MNCs from developing countries, focus on the exploitation of domestic assets abroad with multinationals existing advantages (Mathews, 2006b). Newer theories explaining the second wave of MNCs investment assert that the latecomer MNCs are seeking strategic assets, resources, and knowledge through internationalization in order to obtain competitive advantages (Deng, 2007; 2008; Hong and Sun, 2006; Mathews, 2002; 2006b). As a latecomer in the global knowledge economy, Chinese firms have the chance to cultivate their innovation capabilities and move from the position of late-follower to the position of rapid-follower or even leader through different routes (Child and Rodrigues, 2005; Fan, 2006; Gao, et al., 2007; Liu and Li, 2002; Wong, 1999) 1. 1 Child and Rodrigues (2005) described three internationalization routes taken by Chinese firms. (1) OEM/JVs route: many Mainland Chinese firms choose to start JVs, cooperating with foreign MNCs through OEM or technologies licensing. Gradually, they get the technologies or capabilities they need and move up the value chain. (2) Acquisition route: acquisition can accelerate the process to accrue market strength, to gain access to

3 Imitation and cooperation are obvious tactics for latecomers to accumulate knowledge and accelerate their learning process. China has depended upon these tactics greatly in its efforts to catch up. This approach is limited, however; to achieve sustainable competitive advantage, Chinese firms should go beyond imitation and promote independent innovation. In-house R&D is no longer sufficient and external knowledge has become more and more relevant to the MNCs from developing countries (Gilsing and Nooteboom, 2006). There is some discussion on international learning strategies of latecomers. However, both theoretical and empirical studies offering insights into managing practices and development of emerging MNCs are still limited (Hobday, 1995; Mathews, 2006a; Wong, 1999). In this study, we try to fill in the gap in the existing literature with both theoretical building and empirical evidence from Chinese MNCs. A large number of studies from developed countries has placed emphasis upon international R&D learning strategies of MNCs in a double-network (Ambos, et al., 2006; Asakawa, 2001; Bas and Sierra, 2002; Niosi, 1999). We see from the literature on theoretical development of MNC s network structure that the MNCs from developed countries experienced an evolution of organizational learning and of R&D activities in particular (Ambos, et al., 2006; Asakawa, 2001; Bas and Sierra, 2002; Niosi, 1999). R&D seemed to follow two evolutionary trends. First, they underwent a shift from market-oriented motivations to technology-oriented (Almeida, 1996; Bas and Sierra, 2002; Florida, 1997; Ronstadt, 1978). Second, MNCs evolved from internal knowledge transfer and creation (HQ-subsidiary) to bi-directional vertical (HQ-subsidiary) and horizontal (subsidiary-subsidiary) knowledge transfer and creation (Ambos, et al., 2006; Kogut and Zander, 2003; Tsai, 2001). The research on developed country MNCs shows that traditional organizational learning happens inside the MNC organizations whose headquarter is the knowledge generator, embodying the firm s competitive advantages, and whose subsidiaries exploit the received knowledge from their HQ (Gupta and Govindarajan, 2000; Hymer, 1976; Kogut and Zander, 2003). Since the 1990s, subsidiaries have become more active in learning and creating new knowledge through interaction within internal networks, including both parallel units and HQ (Ambos, et al., 2006; Niosi, 1999; Tsai, 2001). More recent literature reveals an explorative learning role of subsidiaries involving collaboration with a variety of external networks to absorb and learn new knowledge (Andersson, 2002; Doz, et al., 2001; Helble and Chong, 2004; Lehrer and Asakawa, 2002). Whereas the literature on R&D internationalization of developed countries abounds, research on the internationalization of R&D among emerging countries is quite limited (Chen and Tong, 2003; Gao, et al., 2007; Mathews, 2002; von Zedtwitz, 2005). And most of this research does not show us how MNCs from emerging countries tap into advanced knowledge resources through dispersed organizational configurations. What is missing is a discussion on international organizational learning and the orchestration of a double-network of dispersed R&D units. In this study, we aim to explore R&D internationalization from a double-network R&D organization perspective among 9 Chinese R&D centers in Europe. This paper will therefore focus on the following research questions: technology, to secure research and development skills, and to acquire international brands. (3) Greenfield investment: a pure international expansion which aims not only at technology exploitation to satisfy the needs of the local market and global brand recognition, but also intends to enhance managerial control and global integration.

4 Is Chinese double-network R&D organization guided by home-based exploration or home-based exploitation? Are Chinese MNCs in Europe building up double-network R&D organizations? Are they becoming part of a local innovation system? What are the dynamics of R&D internationalization strategies? How do they develop? This paper is organized in six sections. The next section reviews the studies on Chinese R&D internationalization and looks at the literature on double-network R&D organization; the third section defines research questions and hypotheses; the fourth section introduces the research methodology and data collection; the fifth section provides an interpretation of case studies; and the sixth section concludes this study. 2. LITERATURE REVIEW Relevant Studies on Chinese MNC R&D Internationalization Technological innovation and R&D are becoming one of the main driving forces of the international operations of MNCs. According to Gassmann and von Zedtwitz International R&D has risen from a by-product of business internationalization to a quite important and far-reaching phenomenon (1998: p.152). MNCs play a key role in R&D internationalization and account for a major share of international R&D, which therefore increases their presence and relevance in the national development strategy. WIR (2005) revealed that MNCs account for almost half of the total global R&D expenditure. 10% of global 500 companies are headquartered in developing countries, and most of these companies are playing an important role in the internationalization of R&D 2. Without doubt, China has become a popular destination not just for low-cost manufacturing but for research and development as well (Tung, 2005). WIR (2006) reveals that in the FDI recipient countries, China s ranking rose from the fourth place in 2004 to third in Outward FDI from China has taken off since the mid 1990s (Tung, 2005) and have continued to grow steadily into the early 21 st century (see figure-1). Chinese MNCs can be separated into two generations (see WIR, 2006). According to the report, the first generation of Chinese MNCs was dominated by large state-owned enterprises that controlled monopolies such as financial services, shipping, and natural resources. Many of the Chinese MNCs took advantage of the open door policy implemented in China in the late 1970s and started to expand their operations abroad. The second generation of Chinese MNCs emerged in the early 1990s and has had diverse ownership structures 3 that differ from the first generation of Chinese MNCs. This generation of MNCs is focused in sectors where international competition is higher, such as electronics, ICT 4 and other high-tech manufacturing industries. Today, more and more Chinese firms in the high-tech industry are investing overseas. Some of the bolder moves by these companies have gained significant international attention, such as Lenovo s acquisition of IBM s personal computer (PC) business, and TCL s acquisition of Schneider Electronics. To a large extent, Chinese enterprises going abroad are encouraged to do so by the Chinese government. The Chinese government not only retains maintains great influence on the 2 Our elaboration of global fortune 500 ranking list, According to WIR (2006), the ownership structures include private ownership, local government ownership and foreign participation. 4 ICT: Information and communication technologies.

5 Chinese market (Wu and Callahan, 2005) but also has a great policy influence on the international orientation of Chinese firms. The going out (zouchuqu) strategy was introduced by the Chinese government in The amendments made to China's Foreign Economic Development Strategy sought to shift priorities from merely attracting foreign direct investment to also developing going out strategies. In the Third Plenary Session of the 16th Central Committee of the Communist Party of China in 2003, the Chinese government once more claimed that the government would encourage the development of Chinese MNCs 5. The first alteration to China s innovation system, initiated in 1985, was criticized for its inability to create interactive learning between local Chinese firms and foreign MNCs (Gu and Lundvall, 2006). In other words, the innovation policy was unable to support a coherent absorption of foreign technology and there was little assistance for domestic innovation. Some scholars have argued that Chinese R&D engineers are not creative because the Chinese education system and culture don t encourage individualistic expression and creativity (Von Zedtwitz, 2006). They argue that China should change the status quo from Made in China to Created in China. The latest policy priorities indicate that the Chinese government has realized the importance of developing core technologies and technological capabilities as well as national brands (Gu and Lundvall, 2006). The national strategy of endogenous innovation (independent innovation) and continuous reforms to build harmonious development, as well as the Guiding Vision for the 11 th National Economic and Social Development Program ( ) introduced the guidelines of the Chinese government and explained the government s attitude towards innovation. In 2006, the Chinese government included Strengthening Innovation Capacity and Building up Innovation-Typed State as an essential step of the National Medium and Long-Term Science and Technology Plan 6. Chinese firms have recently played a major role in innovation in China. While governmental funds for R&D are decreasing steadily, the Chinese government is still providing substantial financial support to R&D policies (see figure-2). Compared to the MNCs from developed countries, Chinese firms have made a small but steady step in R&D internationalization. Most are relatively young (and therefore small) and focused on the domestic market (von Zedtwitz, 2005: p. 5). Chinese firms often purchase core components and technologies from foreign MNCs, and then undertake system integration and develop features for the final product for the consumer market (Wu and Callahan, 2005: p.175). This traditional cooperation pattern causes domestic Chinese firms to depend heavily on the technological support of foreign MNCs. As Xie and White (2006) pointed out, Chinese firms are accessing advanced technology abroad by establishing technology listening posts or R&D labs and by forming alliances with multinationals (2006: p.235). This trend began in 1995 but became significant in 1999 when outward FDI from China received a boost (Tung, 2005). Moreover, Chinese firms are developing a clear technical competence that contributes both to the local community and to the parent s international R&D network (von Zedtwitz, 2004: p.443). If this is the road ahead, Chinese firms are in their initial stage of R&D internationalization, and it is likely that they 5 See the document a number of decisions of CPC Central Committee on improving the socialist market economic system, 2003, 6 Chinese President Hu Jintao speech at the national conference on science in 2006.

6 will face problems similar to those of MNCs from developed nations. In the next section we will review the main theories found in the literature regarding the evolution of foreign R&D direct investment. The evolution of network Knowledge Organization Today we see decentralized R&D units not only as knowledge transfer units, but also as knowledge creation centers in a fully integrated network (Lehrer and Asakawa, 2002). Many empirical studies suggest that MNCs have switched from a centralized hub organization to a more decentralized federation of units and to integrated networks (Hakanson, 1990) for the purpose of facilitating not only traditional forward knowledge transfer, but also reverse and lateral knowledge transfer. Indeed, ever since resource-based and knowledge-based views of the firm have emerged, it has been difficult to treat knowledge transfer only as a one-way movement of methods from headquarters to foreign subsidiaries (Li, 2004; p.8). Also, the presence of technological enablers and an ever-increasing pressure to adjust to the demands of the market have triggered such a transformation and are changing the roles of subsidiaries in the MNC organization (Zanfei, 2000). O Donnel emphasized that important resources and knowledge upon which the firm s competitive advantage hinges exist at the subsidiary level (2000: p. 530), and that effective resources are transferred from a subsidiary to other international locations. A lateral network of intra-subunit linkage with a high degree of interdependence is required. It is not, however, sufficient to focus on the intra-organization knowledge flows. In recent years, management theories have taken us beyond corporate borders to understand the role of subsidiaries in MNCs. The company s external network has been greatly emphasized in the metanational corporation (Doz, et al., 2001). According to these authors, companies use their international network to sense, mobilize, and operationalize resources and opportunities located outside the boundaries of MNCs. Scholars and managers cannot, therefore, neglect the interaction between the internal and the external networks. Independence from the headquarters is only one of the features to consider for the new model. Zanfei (2000) suggests that the MNC can be seen as a double network, which includes both an internal network (i.e. headquarter and subsidiaries) and an external network (i.e. local partners, suppliers, customers etc.). While the transfer of knowledge in a traditional network organization is vertical and unidirectional, and units can only absorb knowledge passively, in the double-network organization, units not only absorb knowledge but also generate and circulate new information and knowledge. Units increase the potential for usage and generation of knowledge with local firms and institutions in the same area. The new organizational structure is more cooperative and interdependent. When applying the double network theory to MNCs, subsidiaries play a crucial role in interacting with a variety of external actors such as universities, companies, research institutions etc. (Helble and Chong, 2004; Kuemmerle, 1999; Lehrer and Asakawa, 2002; Pagano, 2006; Piscitello and Rabbiosi, 2006), and each of them might detain resources and capabilities which are critical for MNCs strategy (Pagano, 2006: p.6). The integration between the external network and the internal network has been regarded as an important competitive advantage (Andersson, 2002; Bjorkman, et al., 2004; Helble and Chong, 2004). Setting up external linkages becomes an important mission of the overseas subsidiaries. Through such cooperation, overseas R&D subsidiaries can tap into local scientific and

7 technological knowledge pools. A Four-pattern Learning Model in a Double-knowledge-network In a decentralized and more open MNC structure, the organization has plenty of opportunities to learn and change through interactions with a variety of networks. This is increasingly seen as one of the main motivations to decentralize R&D investment. Chiesa (1996) theorizes that the increasing importance of international R&D is due to the need to access external sources of knowledge relevant to a firm s innovation process and the related need to shorten the time spent to acquire, internalize, and utilize this knowledge to perform innovations. There are two main motivations for R&D decentralization according to Belderbos, (2003) and Kuemmerle, (1999). They are: (1) home-based exploitation (HBE): exploitation of a firm s technologies overseas by adapting the technologies to local circumstances in order to get access to foreign markets (Wu and Callahan, 2005); and (2) home-based exploration (HBA): exploration of a firm s technologies through access to overseas technology and know-how. Exploration and exploitation can be seen as two different processes in organizational learning (Holmqvist, 2004) 7. Such learning can take place through various channels, and in particular Hitt et al. (2005) isolate two dimensions: cooperative learning and experiential learning. Cooperative learning means creating new knowledge largely or wholly unrelated to the current knowledge stock or enriching the current knowledge through knowledge transfer by cooperation with partners; experiential learning emphasizes learning by doing (Hitt, et al., 2005) and self-experience accumulation. If we combine the dimensions of motivations/learning processes (exploitation and exploration) and learning channels (cooperative and experimental), we obtain four different types of learning patterns (Hitt, et al., 2005) (see figure-2). In the context of double-network organization, dispersed R&D units play a learning role both with the external network and the internal network. In the external network, R&D units can learn through tapping into the local knowledge pool through cooperation, and self-experience. In the internal network, R&D units can fully utilize the knowledge stock within the firm to refine, implement or create knowledge. In the four different modes of international learning we can identify four different roles that R&D subsidiaries play in the double-network framework. R&D Subsidiaries as observation outposts (Cooperative-exploratory learning): The main role of the R&D subsidiary is to bring new sources of knowledge into the corporate network. This happens through the interaction with foreign external networks. R&D Subsidiaries as remote centers of excellence (Experiential-exploratory learning). In this case, subsidiaries have already learned or acquired the most relevant local resources and the new mission is to generate new technology-related knowledge. R&D Subsidiaries as market gatekeepers (Cooperative-exploitative learning). The foreign subsidiary has to adapt knowledge for a distant market. Close interaction and cooperative development with important customers is necessary. R&D Subsidiaries as market colonizers (Experiential-exploitative learning). The adaptation of production for remote markets remains the main mission of the subsidiary. 7 As Holmqvist (2004) concluded: exploitation can be viewed as the processes by which organizations create reliability in experience through refinement, production, and focused attention ; exploration can be viewed as the processes by which organizations create variety in experience through experimentation, trialing, and free association.

8 Rather than interaction with key customers, the headquarters are requesting the subsidiary to experience directly and learn from the new market, codifying and transferring new knowledge, which is necessary for market access. 3. THEORETICAL HYPOTHESES Based on the four learning-pattern models, we explore the coexistence of modes and goals of R&D internationalization of Chinese firms, and look at their evolutionary dynamics. Home-based Augmenting and Home-based Exploiting As mentioned above, emerging countries lack two main resources when competing with developed countries: lead user markets and technological innovation (von Zedtwitz and Gassmann, 2002; Wong, 1999). Hong and Sun (2006) supposed that external sources of knowledge are increasingly valued by Chinese firms today. Acquiring resources, natural resources in particular, has been one of the key strategic considerations for China s outward FDI since the beginning of China s development. Parallel to resource-seeking investments, Chinese firms have been urged to obtain access to advanced foreign technologies and managerial know-how in hopes of establishing themselves in the international markets. In this case, the main motivation of Chinese firms was technology-driven. The common belief of MNCs from advanced Western countries is that it is better to practice marketing in the host countries and technical innovation in the home country. This means international R&D activities from developed countries are mainly market or technology exploitation-oriented. Technical innovation could be a competitive advantage for the firms from developing countries, which suggests a technology exploration-oriented view of international R&D activities from developing countries. Lee and Lim (2001) concluded that most of the technology-oriented, newly industrialized economies remain in the imitation or listening stage, and that these economies always try to catch up with developed countries by assimilating and adapting the comparatively obsolete technologies of developed countries. Moreover, human resources-driven motivation has been emphasized in many studies that low-cost, and technologically well trained human resources are the main reason for MNCs from developed countries to delocalize R&D activities in China (Motohashi, 2006; Wu and Callahan, 2005). Similarly, the MNCs from China would also like to use high-quality knowledge-based human resources in the host country (Von Zedtwitz, 2006). Thus, we propose the following research hypotheses: Research hypothesis 1A: There is evidence in Chinese MNCs of experiential/collaboration learning focused on exploitation of technological advantages by adapting technologies to local circumstances. Research hypothesis 1B: There is evidence in Chinese MNCs of experiential/collaboration learning focused on exploration of technological advantages by accessing overseas technologies and know-how. Experiential learning and Cooperative learning Organizational learning, rather than knowledge transfer from the parent company to the host country, has been regarded as the core activity of international R&D subsidiaries (De Meyer, 1993; Lam, 2003). For the multinationals in the early stage of R&D internationalization, both cooperative learning and experiential learning are necessary organizational learning channels.

9 Self-accumulated experience is an important path of organizational learning. And in particular, international experiences have been regarded as the prime source of organizational learning for MNCs (Belderbos, 2003). The overseas R&D units can enhance their learning capability by obtaining the knowledge stock from the knowledge center (i.e. HQ) (Zhao, et al., 2005). Furthermore, they can both explore new codified & tacit knowledge and exploit their existing knowledge stock by accumulating self-experience in different geographic locations. Cooperative learning is another very effective organizational learning path for MNCs. By developing modern international market activities and increasing decentralized R&D operations, the dispersed R&D subsidiaries have more opportunities to interact with global knowledge pools. Self-accumulated experience is no longer the only learning mode for firms. Latecomer MNCs with a relatively low knowledge stock can tap into more advanced technologies and accelerate the learning process through cooperation. As latecomers, most Chinese firms have only existed for a few decades. Compared with the MNCs from developed countries, Chinese firms are still in their initial development stage. The shortage of both international experience and core technologies encourages Chinese firms to learn by doing as well as to learn by cooperating. According to the above discussion, we make the following research hypotheses: Research hypothesis2a (Experiential learning): There is evidence in Chinese MNCs of close coordination between HQ and subsidiaries and experiential learning. Research hypothesis 2B (Cooperative learning): There is evidence in Chinese MNCs of collaboration with local/external partners and cooperative learning. The Evolution of the Modes and Goals of R&D Internationalization of Chinese firms The data of R&D investment abroad from developed countries shows that international R&D subsidiaries do not always stay in a state of rest, but experience dynamic evolutionary processes from one pattern to another (Asakawa, 2001; Gassmann and von Zedtwitz, 1999; Lehrer and Asakawa, 2002; Ronstadt, 1978). MNCs from developing countries also go through an evolution of the modes and goals of R&D internationalization. As Mathews mentions, latecomer firms, like latecomer nations, are able to exploit their late arrival to tap into advanced technologies, rather than to replicate the entire previous technological trajectory (2002: p.470). Gao et al. (2007) indicate that from a technical resource-based perspective, using advanced countries as R&D bases is useful in order to (a) acquire local technology and science, and to (b) support local product development (von Zedtwitz, 2005). Chen and Tong (2003) show some evidence that Chinese firms focused on the domestic market and that their international R&D activities have non market-driven factors. Meanwhile, he also proposes a 3-stage evolution pattern that predicts dispersed Chinese R&D units finally developing to a maturity stage aimed at a global market. Some scholars assert that Chinese latecomers should expand into external resource networks to narrow the resource gaps (e.g. Chen and Tong, 2003; Gao et al., 2007; von Zedtwitz, 2005). Chinese R&D units were most probably established in the countries with an extensive knowledge pool and with high-quality human resources. However, the R&D units cannot get enough technological support from the HQ because most Chinese firms have neither the

10 strong technical competences nor the international experience to compete with large MNCs. Consequently, the inabilities and shortcomings of independent R&D in the start-up stage prevent totally independent experiential learning by the R&D subsidiaries themselves. To some degree, cooperation is the only solution to help Chinese R&D units to tap into the knowledge pool easily. Here, we propose the following research hypothesis: Research hypothesis 3: Chinese MNCs first start from explorative learning and cooperative learning, and then move into exploitative learning and experiential learning. 4. METHODOLOGY Multi-case Study Without doubt, the international R&D activity of Chinese MNCs is such a new phenomenon that it has not undergone much academic research. The methodology of this multi-case study was specifically chosen for the exploration of this uncharted theoretical ground (Ghauri, 2004). In an exploratory study, in-depth interviews with R&D managers and a supplementary questionnaire are the main methods used to collect necessary data. Daniels and Cannice (2004) explained the reason for the application of interview-based research methods in case studies, stating that interview-based research is suited for exploratory and theory building studies, and when there is a small population of possible respondents. Furthermore, interviews allow researchers to develop a deeper relationship with informants (Daniels and Cannice, 2004). Therefore, the multi-case study method is quite suitable to investigate our proposed research question of what purposes do Chinese dispersed R&D subsidiaries serve and how? Data Collection As Ambos said, learning in high-technology firms takes place preponderantly in R&D units (2002: p 3). China is one of biggest developing countries. In this paper, we take the Chinese R&D subsidiaries as our object of research. Moreover, this research focuses on the overseas Chinese R&D units in Europe. Europe is deemed as of the most popular ODI destinations for emerging countries, including China, as evidenced by the large number of Chinese MNCs here present. Also, Europe has a diversified technological base and is one of the leading consumer markets. In order to collect the data we first identified the location and established time of subsidiaries by combining a variety of sources 8. In this phase, we identified 26 R&D units established by Chinese firms in Europe. We selected 9 of these subsidiaries belonging to various industries for this study to obtain various perspectives (Creswell, 1998)(see Table-1). Face-to-face or in-depth telephone interviews were then conducted. The open-ended interview questions are listed in Table-2. Most of the interviewees were senior R&D managers of the R&D unit, as well as some engineers. Prior to the interview, we asked the respondents to fill out a questionnaire for the quantifiable questions, which were used for pre-testing the survey instruments (Helble and Chong, 2004) and to verify the validity of the interviews findings. Finally, we cross-checked and integrated the data from the questionnaires and interviews. 8 Secondary information available in newspapers, and the internet, personal interviews and the database compiled by fdi Markets,

11 Whenever the answers provided in the questionnaire or in the interviews were unclear, the respondents were contacted again and asked for clarification MULTI-CASE STUDY ANALYSIS Although only a small proportion of worldwide Chinese R&D subsidiaries, we believe the 9 cases from different industries examined in our study are a good sample and accurately reflect the R&D internationalization of Chinese firms. Chinese Double-network R&D Organization are mainly Guided by Home-based Exploration The evidence gathered in our research among Chinese MNCs in Europe suggests the co-existence of two goals of learning-both home-based augmenting and home-based exploiting. In our cases, some have dual equivalent motivations, while others are largely motivated by the advanced technologies (see figure-3). Cases of the former, including cases D, F, and I, have normally occupied a steady share of the domestic market as well as have attained a certain degree of independent R&D capabilities. Expansion into the European market has become an integral part of their internationalization strategy. Lacking in both international market experience and technological know-how, their competitive disadvantages force them to establish R&D branches in Europe. Take Case D as an example. As a domestic LCD TV giant, the parent company of case D owns TV production bases in Hungary, France, and South Africa, and sales offices in the U.S., Europe, Australia, and Japan. In the past few years, the parent company has focused on the European and North American market, which are viewed as the biggest LCD TV markets. The establishment of Case D has a strong market-driven motivation. If you hope to develop the products that can make the European customers satisfied, you have to find out the situation of the local market, the technological development, and the consumer consumption and habit (interview, case D). After evaluating the technical personnel sources, suppliers, costs of investment, traffic conditions, geographical location, and language environment, Eindhoven was chosen as the location of case D. Technology-motivated firms account for the majority of our cases; the Chinese R&D investments in Europe that we have interviewed are mainly driven by technology. In our cases, a straightforward technology-driven motivation factored into the decision of many firms to establish R&D centers in technology-intensive areas. This provided Chinese firms with the opportunity to have close interactions with the leading local technology providers. These R&D centers try to increase local embeddedness to absorb offshore knowledge and some actively engage in new knowledge creation with a high degree of autonomy. Chinese firms apparently prefer to catch up with the technological aspects by consolidating and enlarging the market share in China first, and then expanding into the international market. The dominant technology-driven motivation can be broken down into the following four sub-motivations. Proximity to the centers of technological excellence When mapping our cases according to their location selection in Europe, we discovered a 9 The information and data that could not be obtained directly from some of the companies were obtained from secondary sources, such as LexisNexis Academic ( Factiva ( and official websites.

12 clear strategy: settling down close to technological excellence centers. Two main benefits can be obtained by setting up shop near technological leaders. First, Chinese firms can receive overseas technological support and compete in advanced technology R&D. As the vice director of Case C said, Turin is a world class design center, where the surroundings, infrastructure, supporting facilities are satisfactory. There are a lot of local modeling companies in Turin that can be utilized to build car models. The same motivation is also expressed by the interviewees of Case A, B, D, and F. Second, being close to centers of excellence enables Chinese firms to track technology & product development. Different from the discussion of Chen and Tong (2003), technological scanning is no longer the main task but just one of the necessary components according to the interviewees. For case B, C, D, F, and I, an overall grasp of the technology & product developing trend in Europe is quite useful for future R&D project planning and operating. Case F, the most recently established R&D center of the 9, mainly undertakes the tasks of information collection and technology monitoring. It has built informal but close and long-standing audio technological connections with the local senior engineers, R&D and consulting companies, and universities. Strengthening and improving the cooperation and communication with famous international technological partners. Another strong technology-oriented motivation expressed by our cases was the desire to establish new external knowledge networks or to strengthen existing local cooperative relationships. Case B is viewed as an advance troop, not only for design & style, but also in their initiative in outsourcing knowledge, integrating resources and monitoring the developing trends in the automobile industry. We came here to have a good front-line control and localized management. Meanwhile, we can as well develop a close cooperation with the local partners (interview, case B). Case E is another representative case with an exclusive technology-driven motivation. Before the foundation of this R&D center, our company had established a long and close cooperative relationship with the European companies, especially the parts & components manufacturers. We always have 3-4 projects at the same time. Meanwhile, many of our R&D employees are working in Europe with local companies. In such case, we establish this European R&D center in order to run all the projects smoothly (interview, case E). Specific technology or strategic product development Cases A, E, G, and H also aim at the development of strategic products that are critical for them to have a leg up on the competition. For instance, case A focuses on the 3G technology; case G focuses on the R&D of large gantry machining; case E focuses on the R&D of the internal-combustion engine; and case H focuses on the R&D of electronic jacquard machinery. These cases had clear product develop plans before they expanded their R&D operations into Europe. R&D talents recruitment and cultivation Last but not least, high quality specialized human resources are among the most important technology-driven motivations for setting up overseas R&D units in Europe. Cooperation with local, high-level human resources has been viewed as one of the main reasons for the enhancement of R&D capabilities through the establishment of overseas R&D centers. Chinese firms desire to not only use high-skilled researchers and engineers from the local environment, but to also cultivate their own R&D professionals through interaction with

13 various external knowledge networks. Using the local human resources with advanced technological knowledge is the most effective way for the R&D units to tap into the local knowledge networks. Human resource localization helps to achieve local R&D embeddedness. In addition, plentiful, local, and high-quality human resources contribute to high-skilled R&D human cultivation of Chinese firms. This is shown by both of the automobile companies (case B and C) who employ designers and engineers from the plentiful and high-quality local human resources to accomplish the projects as well as to train their own designers and engineers. The high-skilled researchers, engineers and designers in the local environment are not only a key factor in finishing R&D projects, but also an effective way to train their own R&D human resources. As the interviewee said, the biggest achievement for us is the training of our own designers and engineers (interview, case B). Chinese MNCs in EU are Building up Double-network R&D Organizations The evidence from our cases also suggests the co-existence of two modes of learning. The overseas R&D units of Chinese firms we have interviewed engage in both cooperative learning and experiential learning to enhance their R&D capabilities. In our cases, all the R&D units have established some degree of cooperation with the external knowledge network. Through the process of cooperation, the overseas R&D units get access to local knowledge pools and take advantage of the advanced technologies they haven t mastered. At the same time, the interviewees also emphasized the importance of experiential learning and close coordination with HQ. As part of the global R&D system of their companies, these R&D units are supported by the HQ in terms of capital, human resources, etc. As geographically detached R&D subsidiaries, these R&D units can self-experience different knowledge contexts and generate innovative gems. Chinese MNCs in EU are building up external R&D network organization From figure-4 we can see that the overseas R&D units of the Chinese firms we interviewed tap into external knowledge networks through cooperative learning with local actors in Europe, especially suppliers, private research institutes, and customers. Joint work with highly skilled partners leads to significant HR upgrading, which is an important method of cooperative learning for the R&D units of these Chinese firms. The knowledge lag of Chinese firms led them to seek resources and knowledge from the external network. This process enhances their competitive capabilities in the Chinese market first and in the international market later. The linkage with the external network is cooperative. The R&D centers of Chinese firms turn to the local technological environment, especially the large local firms with advanced knowledge resources that haven t been mastered by Chinese firms. They take the local companies as knowledge sources and cooperate with the local companies to fully and effectively utilize the technological resources. They learn and accumulate knowledge and experiences, while also training their own R&D human resources. The most useful and common cooperation strategies are contracting-out & consigned processing, cooperative R&D projects, and human resources outsourcing. These methods of cooperation also provide several opportunities to learn from the external knowledge network. Here we give some examples of how Chinese dispersed R&D subsidiaries learn through each cooperative mode.

14 1. Contract out and consigned processing When case B first started in Europe, they could not build the necessary platform for their projects operation. They therefore contracted out 75%-80% of their design projects to the famous local automobile design companies. For example, Pininfarina 10 has a high cost but good automotive design strength. For case B, however, the cost of carrying out the project by themselves was still higher than that of contracting out to this company. Through contracting with Pininfarina, case B gradually developed their own design/r&d skills and succeeded in cooperating with some local SMEs, while also cutting costs by 40%. We tell the local SMEs what we want to do and we can control them (90% are done by ourselves, and 10% contract-out) (interview, case B). 2. R&D project-based cooperation When case C started their international R&D activities to get their own independent intellectual property rights, they chose an independent R&D strategy. Even at the beginning, case C s main objective was to participate in joint-r&d projects with the firms occupying more advanced technologies. In the process of CM8 design & development, the R&D employees of case C can only do some assistant work, maybe 20%. But in project CV6, the R&D employees of case C can assume 50%. In the following projects, the R&D employees of case C can assume 70%-80%. (Director of Case C, 2006) R&D human resources outsourcing Hiring advanced technical human resources from local specialized partners is considered an effective approach of cooperative learning by the interviewees. Signing contracts with the suitable R&D specialists from local companies to work for the corresponding projects with the R&D experts in the center effectively covers the high-quality skilled R&D human resource shortage of Chinese firms. Case C cooperates with local companies and employs some engineers and designers to work together on a project. In the cooperative project, for example: we take the designers from IDEA-a famous automotive design company as the main force, and our designers participate in the project. The best way is to pair up the foreign designers and our designers, the foreign designer grading our designer every month (Director of case C, 2006). Case C leveled up the skills of their own designers in the projects. Anyway, the Italian designers are more advanced. We also employ the local designers to work in our R&D center on a long-term and full-time basis (Director of Case C, 2006). 12 The manager of case B was an engineer who used his human resources network to set up an R&D platform in a very short time. To improve case B s human resources, he sent his employees to the cooperative companies to learn the design and development processes of an automobile. Chinese MNCs in EU are Building Internal R&D Network Organization From figure-5 we find that there is also close coordination between the overseas Chinese R&D units we interviewed and their HQs. Formal coordination mechanisms are commonly used by HQs to supervise R&D subsidiaries. Standardized and routinized procedure or planning and report systems submitted by R&D units to HQs are always done with the sole purpose of integrating HQ and the subsidiary. Besides formal managerial instruments that can 10 Pininfarina is a major supplier of automotive design, product & process engineering and manufacturing. 11 Report from 21st Century Business Herald, 2006 (Chinese) 12 Report from 21st Century Business Herald, 2006 (Chinese)

15 facilitate HQ control of R&D subsidiaries, informal coordination and communication mechanisms are commonly used between R&D subsidiaries and HQs to accelerate the internal experiential learning and creation process of new knowledge and bi-directional transfer of new knowledge and information. According to our cases, the most important informal mechanisms for knowledge learning and transferring are communication tools, professional transfer, HQ-subsidiary teamwork, and integrating person/department. 1. Communication tools To solve distance problems and conquer know-how exchanging barriers, informal contacts through internet, s, and telephone have become the main daily communication means between R&D subsidiaries and HQs. For example, video meetings become an effective tool to connect R&D subsidiaries and HQs. Equipped with a professional translation team, the Chinese engineers in the HQ of case I communicate smoothly and regularly with its Italian designers and engineers through video conference. 2. Professional transfers and HQ-subsidiary teamwork Professional transfers and HQ-subsidiary teamwork facilitate close interactions and mutual learning between R&D subsidiaries and HQ. Our cases can be grouped into two situations based on the technology level differences that exist between R&D subsidiaries and headquarters. The cases belonging to the first situation have a relatively high technology level compared to their headquarters, such as cases A, G, H and I. Case A stresses localized human resource recruitment and has achieved a high degree of foreign embeddedness. Case G, H and I all retained the preexisting R&D branches, regulars and patented technologies after acquisitions of local firms. For these cases, large technological gaps exist between R&D subsidiaries and HQs. Short-term professional transfers from HQ to R&D subsidiaries are motivated by technological learning. As the interviewee of Case H said, I can only say that the chief reason for periodical visiting of engineers from our HQ is studying. We learn their product concept development, but don t really participate in R&D activities. These subsidiaries mostly dispatch their selected R&D teams back to HQ according to the specific requirements of HQ-subsidiary joint project. These project-based R&D teams not only take charge of technical experts delivery from R&D subsidiary to HQ and skill training, but also assist in project planning and decision-making. In the second situation, Case B, C, D, E and F have relatively weak technological accumulation. A high percentage of Chinese R&D professionals are transferred from the headquarters to accomplish specialized R&D projects. Compared to the first situation, these cases in the second group tend to foster close and direct interaction between R&D expatriates and local technological experts and tend to participate in R&D activities in the frontier of MNCs (i.e. R&D subsidiaries). Due to high frequency of R&D personnel transferring, information/know-how can be brought back to HQ easily when joint-projects are finished. 3. Integrating person / department Furthermore integrating person / department is a common practice used to smooth communication and couple information between HQ and the R&D subsidiaries, according to our cases. Take Case B for example. It has a project department in HQ that is in charge of the communication between the department in HQ and the overseas partners. Case B tries to

16 protect the authenticity of information. It s a one to one information transfer path. For each project, HQ sends a specific project manager to take charge of all the communication between HQ and R&D for that particular project. Tripartite teamwork: Chinese MNCs are building up a double-network R&D organization Teamwork can be considered to be the cooperative networking and communication between or within subsidiaries or units of a firm (Mudambi, et al., 2007). In our study, a tripartite of project-based teamwork, comprised of R&D subsidiaries, HQ, and external networks, makes up the double R&D network organization of Chinese MNCs. The three entities utilize their respective knowledge advantages to accomplish R&D projects jointly (see figure-6). 1. HQ-R&D subsidiary teamwork For many of our cases (case B, C, E, G, and H), their primary mission in joint R&D projects is to undertake research & design activities, while their domestic R&D sectors assume technological application and development, product commercialization and production industrialization. To a large extent, this division of R&D tasks aims to compensate for the disadvantage of the Chinese firms low technological/design level, while taking full advantage of the low-cost of production. A typical sample is case H: German side engages in the R&D of the third generation of jacquard machine and mainly focuses on conceptual research and prototyping. Chinese side is majorly with responsibility for market-facing product development and production industrialization. Our Chinese R&D professionals first communicate with our German R&D professionals about the conceptions and principles of new jacquard machine, and then take charge of cost reduction, sinicization of parts and components, product experiments and approval these specific jobs. (Interview, Case H) For the other cases (case A, D, F, and I), satisfying local customer demands and developing new products aiming at local market are important tasks, while their domestic R&D counterparts are mainly focusing on China s market. Case I gives us some evidence here: Our European center is responsible for high end motorcycles research and development, and mainly cooperates and communicates with local suppliers; reversely, the research and development of scooters primarily depend on the cooperation with the domestic R&D sector (interview, case I). 2. R&D Subsidiary-external partners teamwork Our cases choose their local partners according to their various technological requirements, especially for those projects that are difficult for their parent companies to accomplish. For small-scale, dispersed Chinese R&D units, accomplishing an R&D project relies heavily on the external technological bases, especially the support from core components suppliers and private research institutes/companies providing R&D services. Case D is a representative case engaged in R&D subsidiary-supplier teamwork of joint R&D project: At the moment, we don t own the independent capabilities to develop chips applied for European market. We have to rely on NXP, ST these professional companies to provide us mature chip solutions. The foreign specialists sent by suppliers fully involved in the whole project that provide us technical support and assist us to complete relevant software development and debugging(interview, case D)

17 3. HQ-external partners teamwork For case B, C, D, E, R&D subsidiaries become the mediators between internal R&D networks and external R&D networks, where the R&D employees from HQ in home countries can work and interact with the external R&D networks directly along with joint-project technological professional transfer. Human resources cooperation does not only involve the local specialists and the employees in the R&D centers, but also involves the R&D human resources in the HQs. To sum up, in a tripartite teamwork framework, HQ-subsidiary professional transfers facilitate communication between Chinese R&D employees and local specialists, upgrade the skill level of Chinese R&D human resources, and accelerate the learning process within Chinese firms. Case E is a project-concentrated R&D center that can be viewed as an overseas department of HQ. The average R&D employees are all dispatched from the HQ and keep transferring under the necessity of project requirements. Although overly frequent personnel transferring induce the problem of unstable personnel management, the dynamic project-based cooperative method achieves the goal of training and cultivation of high quality R&D employees 13. At present, this R&D center takes R&D projects as the central task, and employees vary with changes of projects. This arrangement just realized our purpose of R&D professional cultivation. Originally, the parent company of case E established a strategic alliance with AVL, which is a world-class company famous for the development of power train systems with internal combustion engines. With the increase of cooperative partners and the improvement of technical capabilities, case E gradually became an independent R&D center. The dynamics/development of R&D internationalization strategies The evidence of this study suggests an evolution of modes and goals of learning. According to both the R&D motivations/processes and learning channels, our 9 cases can be classified into three learning patterns. A classification of cases Cases A, B, C, and E are observation outposts. Regarding learning motivations/processes, we observed that case B and case C are neither serving the local European markets countries nor exploiting the technologies transferred from the HQs. These two Chinese automotive companies established their first overseas automotive design centers abroad because Turin is one of the most famous design centers, with a number of qualified engineers and designers. Similarly, case A has been dedicated to the exploration of the most advanced wireless technologies in Sweden ever since it was established 6 years ago. Case E s pure motive for internationalizing was exploring technology for the diesel engine. Regarding learning channels, case A, B, C and E all have close interaction and cooperation with their local partners. Cooperative learning with external technology networks makes up a large proportion of technological learning compared to knowledge acquisition from HQs and self-experiencing. Again, the automotive companies serve as very illustrative cases. Car design is a very creative job, and the key factor of success is high-skilled designers and engineers. The two companies clearly recognize their disadvantages and set up their observation outposts in the most intensive knowledge pools to utilize abundant local human resources. They have established multiple cooperative linkages and the extent of their involvement in the cooperation has 13 The next step of case E is to implement the practice of managerial personnel immobilization.

18 increased, providing them with unique learning opportunities and allowing them to cultivate their own high-quality designers and engineers. Case D, F and I can be viewed as market gatekeepers. Their parent companies all possess a certain degree of R&D capacity and many have grasped a share of the European market. The parent company of case D has emerged as the market leader in air-conditioners, computers, mobile phones, LCD televisions and so on in China s electronics industry. At present, the Chinese company owns TV production bases in Hungary, France, and South Africa, and has sales offices in U.S., Europe, Australia, and Japan. In the past few years, the company has begun to eye the European and North American markets, which are viewed as the biggest LCD TV market. The parent company of case F is a major supplier to several international firms, such as Siemens and GNNetcom % of their handset receiver products are exported to Asia, Europe and America. Having close interaction and cooperation with key customers is important for indentifying the newest market demands and for seeking potential product development projects. However, due to the uncrossable technological gaps between their European customers, technological learning is still inevitable for Cases D and F. After the acquisition of case I, the Chinese parent company has not only reached a higher technological level, but also successfully implemented an international brand arrangement from low-end brands to mid-range and high-end brands in the mainstream European motorcycle market. This is exactly what they hoped to achieve when they acquired case I. Case G and H can be viewed as remote centers of excellence. Both case G and case H are set up through merger and acquisition (M&A) and focus on the R&D of specific products (case G focuses on large gantry machining center and case H focuses on the electronic jacquard machine). In the acquisition of European companies cases G and H seized key strategic assets, including technologies, patents, human resource, and managerial skills. This was a highly strategic and successful move, as the parent companies of cases G and H wanted and needed these assets in order to build up competitive advantages. Both retained and enhanced the previous R&D departments of their acquired companies. Cases G and H became centers of excellence whose R&D capabilities are leveraged by Chinese parent companies and disseminated to other parts of internal knowledge networks (Frost, et al., 2002). Observation outposts are evolving into market gatekeepers At present, case B and case C are still struggling to compete with the global automotive MNCs for a larger Chinese automotive market share. The two Chinese automotive companies are in their first stage of internationalization; neither of the companies has begun to sell their products in the European market. The European market is, however, an inevitable part of the future plans of both companies. During the interview, the general manager of case B expressed their hunger for the local market. The R&D internationalization process of case A can be explicit evidence of the evolution from HBA to HBE. The parent company of case A takes an international market strategy that first occupies the developing country s market, and then expands to developed countries 15. By contrast, its international R&D strategy is a reverse process. This Chinese company set up its R&D center in the countries with the most advanced telecommunication technologies. As 14 and , entered the market of South Asia and Africa; , marched into India, Russia, and Brazil these emerging markets. Source: official website of case A and

19 early as 1998, it opened its first R&D institute in the U.S. The decision to set up a European R&D center in Sweden shows a strong technology-driven motivation to get overseas technological support and compete in 3G technology R&D. It is not enough that we only rely on the R&D forces in China to catch up these competitors in a short time, unless we have a good technological support (interview, case A). Along with the technological development and the enlargement of its market share, this Chinese telecommunications company is also gradually shifting its attention to the European market. Especially since 2005, it is emerging as a 3G competitor in Europe, signing many cooperation agreements with important European telecommunication operators and companies such as France Telecom. In the meantime, case A is also adjusting its position and is becoming not only a simple technology explorer but also a technology exploiter serving the European market. In addition to technological considerations, case A also tracks the global telecom standards of major operators, analyzes telecom market trends, and assists in strategic formulation. Since we assist our company in formulating product planning, we are obligated to have a deepening communication. We have to understand the operators demands and support market operation (interview, case A). Obviously, technology exploration is the most important mission for the Chinese R&D subsidiaries in Europe according to our preliminary evidence. At the same time, these originally HBA-dominated Chinese R&D subsidiaries also tend to be involved in more HBE R&D activities. Observation outposts are evolving to remote centers of excellence Our cases also provide significant insight into the transformation of cooperative modes. Case B, C, E as well as Case A, depend on the local partners to a large extent, and in the beginning had to contract out their R&D projects. During the process of cooperative learning, these R&D subsidiaries upgraded their R&D capabilities and improved the quality of R&D human resources. Moreover, these R&D subsidiaries increased their involvement in the R&D projects by switching to the cooperation mode of the joint R&D project and enhanced their self-experience and self-discovery skills in the new technological environment. One example is case A. After several years of development, case A has become a relatively mature R&D center, and handles the R&D activities of core 3G technologies. The explanation in the interview is the following: we master the core technologies through the cooperation with the local technological experts, and sequentially we finish the R&D projects through the cooperation with the R&D team in the HQ (interview, case A). Another representative example is case E. Since its establishment at the very beginning, case E has never changed its technology exploration motivation or revealed its ambition for the European market. Consistently upgrading its own technological level by continuous project cooperation with local technological partners, case E has achieved a qualitative leap from cooperative R&D of Landking (Euro III) series Diesel Engines to independent R&D of Landking (Euro IV) series Diesel Engines. Originally, it was absolutely impossible for us to complete the R&D task of Euro III, and we had to turn to the outside brain, i.e. AVL, for help. During the project cooperation, our own R&D employees had the opportunity to participate in the whole R&D process. After the cooperation of Euro III project, we have not only developed several euro III series diesel engines but also independently designed and developed Euro IV and V series diesel engines except some key component support from our suppliers, such as Bosch

20 (interview, case E). Although case E still relies on the advanced component suppliers, it has internalized the most relevant aspects of R&D capabilities and is shifting to a remote center of excellence. The missing learning pattern of market colonizer Unfortunately, we haven t identified any market colonizers from our European cases. We d like to discuss the potential evolutionary trajectories of remote centers of excellence to market colonizers or from market gatekeepers to market colonizers, even though these patterns were now found in our research (see figure-7, dotted arrows). Case G and H are the remote centers of excellence that grasp the key technologies and products that haven t been mastered by their Chinese parent companies. The immediate concern of the HQ is to digest and disseminate the knowledge transferred from the centers of excellence within the MNC. As the interviewee of case H explained, Only a small portion of Chinese staff here participate in R&D projects. The R&D employees dispatched by HQ are always learning oriented, but not R&D project oriented (interview, case H). However, due to the rapid technological progress, market-driven motivation is also revealing. In addition to advanced and sophisticated technologies, M&A also assures the acquisition of brands, markets, and distribution channels. FORD, GE, Volkswagen, etc. are all customers of the company we acquired. We consolidate and further expand the existing global markets after M&A (interview, case G). A thorough assimilation of specialized technological capabilities in the internal R&D network is a basic premise for remote centers of excellence to change their technology exploration role and evolve to market colonizers. Case I can be seen as a potential sample evolving from market gatekeeper to market colonizer. Case I is not a simple R&D center with independent design capability for motorcycles and scooters, but owns its renowned brand, factory, distribution, and after-service. Let s take the scooter part as an example, our scooters are designed in Europe, and manufactured in China. Here is also a sales service center and our scooter products will be delivered to Europe, the U.S., and worldwide (interview, case I). Except the cooperation concerning high-end motorcycle parts with suppliers, case I has internalized the most relevant knowledge resources and is positioned as a springboard to launch into the European motorcycle market. CONCLUSIONS: THE NEW TRENDS OF CHINESE R&D INTERNATIONALIZATION The above discussions support the claim that different modes and goals coexist in Chinese dispersed R&D subsidiaries, and that some R&D units are evolving from one combination of mode/goal to another. The evidence in this study of Chinese subsidiaries in Europe reveals some new evolutionary characteristics of R&D internationalization from emerging countries, distinct from that of developed countries. Motivation: from HBA to HBE The traditional explanation of R&D internationalization motivations of MNCs is from home-base exploiting (HBE) to home-base augmenting (HBA) (Almeida, 1996; Asakawa, 2001; Bas and Sierra, 2002; Florida, 1997; Lehrer and Asakawa, 2002; Ronstadt, 1978). As Asakawa (2001), Lehrer and Asakawa, (2002) concluded, there is a classic evolution path for R&D internationalization. First, the overseas R&D units learn to adapt in the host country the firm-specific knowledge they ve gained in their home country. These R&D units

21 undertake more and more research tasks in the evolution process. The early international business literature explain that firm-specific monopolistic advantages are transferred across borders through subsidiaries and exploited in the local markets of host countries (Caves, 1971; Hymer, 1976; Rugman, 1981). More recent studies show that knowledge augmenting by establishing a new presence in advantageous locations is becoming the dominated motivation of global FDI (Bas and Sierra, 2002; Kuemmerle, 1997). However, Almeida (1996) found differences in the motivations of MNCs from different countries and argued that foreign firms should be viewed as technological knowledge contributors rather than technological knowledge absorbers. The European and Korean firms are motivated by offsetting home country technological and perform as U.S.-technology absorbers, while the Japanese firms perform more as U.S.-technology participants that contribute to inter-firm knowledge flow with domestic firms. Also, Patel and Vega (1999) reveal the evidence that product adaptation and technical support to local markets remain the major factors promoting R&D internationalization. The overseas R&D units of Chinese firms in Europe reveal a different R&D internationalization pattern than the traditional HBA to HBE (See figure-7, Arrow1). Different from the MNCs from developed countries, most Chinese firms possess comparatively weak innovative capabilities in headquarters and there are not enough internal competitive advantages to be exploited for R&D subsidiaries to compete in host countries. Tapping into external knowledge networks to explore new technological advantages and augment previous home-based knowledge turns into the initial motivation for Chinese firms to expand R&D activities internationally. As technological capabilities get continuously upgraded, the motive of technology exploitation and participation of host market competition will gradually emerge and coexist with the motive of technology exploration. From Cooperative Learning to Experiential Learning Another traditional evolution of organizational learning of MNCs is from experiential learning to cooperative learning (Ambos, et al., 2006; Kogut and Zander, 2003; Tsai, 2001). Lehrer and Asakawa (2002) viewed the R&D subsidiary evolution from a perspective of organizational configuration and concluded the classic evolution path of foreign R&D subsidiaries is from knowledge transferor to fully integrated knowledge creator. More explicitly, traditional organizational learning channels inside MNC organization emphasize internal knowledge learning, transferring, sharing, integrating and synthesizing (Gupta and Govindarajan, 1991; Gupta and Govindarajan, 2000; Hymer, 1976; Kogut and Zander, 2003; Tsai, 2001). Gradually, external knowledge networks are regarded as an important competitive advantage and serve as sources of learning (Andersson, 2002: p: 981). The ultimate state of integrated network configuration is one that facilitates bi-directional flow of both internal and external knowledge (Kurokawa, et al., 2007). The overseas R&D units of Chinese firms in Europe reveal a different R&D internationalization pattern from cooperative learning to experiential learning (See figure-7, Arrow2). Although Chinese R&D units seem more like the third path, i.e. from knowledge incubators to knowledge integrators(lehrer and Asakawa, 2002), some significant differences existing in Chinese R&D subsidiaries cannot be ignored. The key mission of knowledge incubators is to create new knowledge 16. Chinese R&D subsidiaries mainly perform as 16 The cases used in this study are research-based incubators.

22 knowledge learners that absorb and transfer knowledge from external knowledge stock. In this case, Chinese firms will firstly quest for external knowledge sources through cooperative learning, where reverse knowledge flow from external knowledge network to internal knowledge network mainly happens. Along with internal knowledge accumulation, Chinese firms are diffusing and integrating their learning within multinational organization. Internal reinforcement of technological strength also stimulates self-learning and self-experiencing of R&D subsidiaries in the host knowledge environment. Finally, both cooperative learning and experiential learning will be utilized simultaneously by Chinese firms in a double-knowledgenetwork. Next Steps & Limitations We find evidence of different modes and goals, but we cannot say anything about dominant types of such goals. Even if we spot trends, our analysis is only partial and we cannot generalize results claiming to have identified a clear evolution from one combination of mode/goal to another. We hope that our analysis will lead to more in depth analysis of management of subsidiaries at the organizational and strategic level. REFERENCES Almeida, P., (1996). 'Knowledge sourcing by foreign multinationals: patent citation analysis in the US semiconductor industry'. Strategic Management Journal: Ambos, T.C., Ambos, B. and Schlegelmilch, B.B., (2006). 'Learning from foreign subsidiaries: An empirical investigation of headquarters' benefits from reverse knowledge transfers'. International Business Review, 15 (3): Andersson, U., Forsgren, M. and Holm, U., (2002). 'The strategic impact of external networks Subsidiary performance and competence development in the multinational corporation'. Strategic Management Journal, 23:pp Asakawa, K., (2001). 'Evolving headquarters-subsidiary dynamics in international R&D: the case of Japanese multinationals'. R&D Management, 31 (1):1-14. Bas, C.L. and Sierra, C., (2002). ''Location versus home country advantages' in R&D activities: some further results on multinationals' locational strategies'. Research Policy, 31 (4): Belderbos, R., (2003). 'Entry mode, organizational learning, and R & D in foreign affiliates: evidence from Japanese firms'. Strategic Management Journal, 24 (3): Bjorkman, I., Barner-Rasmussen, W. and Li, L., (2004). 'Managing Knowledge Transfer in MNCs: The Impact of Headquarters Control Mechanisms'. Journal of International Business Studies, 35 (5): Caves, R., (1971). 'International corporations: the industrial economics of foreign investment'. Economica:1-27. Chen, J. and Tong, L., (2003). 'R&D Internationalization and the Reformation of Chinese S&T System'. George Mason Law School, Child, J. and Rodrigues, S.B., (2005). 'The Internationalization of Chinese Firms: A Case for Theoretical Extension?'. Management and Organization Review, 1 (3): Creswell, J., (1998). Qualitative inquiry and research design: Choosing among five traditions: Sage Publications. Daniels, J. and Cannice, M., (2004). 'Interview Studies in International Business Research'. In: Marschan-Piekkari, R. and Welch, C. (eds). Handbook of qualitative research methods for international

23 business: Edward Elgar Publishing. De Meyer, A., (1993). 'Management of an international network of industrial R&D laboratories'. R&D Management, 23 (2): Deng, P., (2007). 'Investing for strategic resources and its rationale: The case of outward FDI from Chinese companies'. Business Horizons, 50 (1): Deng, P., (2008). 'Why do Chinese firms tend to acquire strategic assets in international expansion?'. Journal of World Business, 44 (1 ):74-84 Doz, Y., Santos, J. and Williamson, P., (2001). From Global to Metanational: How Companies Win in the Knowledge Economy: Harvard Business School Press. Fan, P., (2006). 'Catching up through developing innovation capability: evidence from China's telecom-equipment industry'. Technovation, 26 (3): Florida, R., (1997). 'The globalization of R&D: Results of a survey of foreign-affiliated R&D laboratories in the USA'. Research Policy, 26 (1): Frost, T.S., Birkinshaw, J.M. and Ensign, P.C., (2002). 'Centers of excellence in multinational corporations'. Strategic Management Journal, 23: Gao, X., Zhang, P. and Liu, X., (2007). 'Competing with MNEs: developing manufacturing capabilities or innovation capabilities'. The Journal of Technology Transfer, 32 (1): Gassmann, O. and von Zedtwitz, M., (1999). 'New concepts and trends in international R&D organization'. Research Policy, 28 (2-3): Ghauri, P., (2004). 'Designing and Conducting Case Studies in International Business Research'. In: Marschan-Piekkari, R. and Welch, C. (eds). Handbook of Qualitative Research Methods for International Business: Edward Elgar Publishing. Gilsing, V. and Nooteboom, B., (2006). 'Exploration and exploitation in innovation systems: The case of pharmaceutical biotechnology'. Research Policy, 35 (1):1-23. Gu, S. and Lundvall, B., (2006). 'China's Innovation System and the Move Toward Harmonious Growth and Endogenous Innovation'. Innovation: Management, Policy & Practice, 8 (1):1-26. Gupta, A. and Govindarajan, V., (1991). 'Knowledge flows and the structure of control within multinational corporations'. ACADEMY OF MANAGEMENT REVIEW, 16 (4): Gupta, A.K. and Govindarajan, V., (2000). 'Knowledge Flows within Multinational Corporations'. Strategic Management Journal, 21 (4): Hakanson, L., (1990). ' International decentralisation of R&D the organisational challenges. In: Bartlett, C.A., Doz, Y., Hedlund, G. (Eds.), Managing the Global Firm. Routledge, London'. pp Helble, Y. and Chong, L.C., (2004). 'The importance of internal and external R&D network linkages for R&D organisations: evidence from Singapore'. R&D Management, 34 (5): Hitt, M.A., Li, H. and Worthington, W.J., (2005). 'Emerging Markets as Learning Laboratories: Learning Behaviors of Local Firms and Foreign Entrants in Different Institutional Contexts'. Management and Organization Review, 1 (3): Hobday, M., (1995). 'East Asian latecomer firms: Learning the technology of electronics'. World Development, 23 (7): Holmqvist, M., (2004). 'Experiential Learning Processes of Exploitation and Exploration Within and Between Organizations: An Empirical Study of Product Development'. Organization Science, 15 (1): Hong, E. and Sun, L., (2006). 'Dynamics of Internationalization and Outward Investment: Chinese

24 Corporations' Strategies'. The China Quarterly, 187: Hymer, S., (1976). The international operations of national firms: MIT press Cambridge, Mass. Kogut, B. and Zander, U., (2003). 'Knowledge of the firm and the evolutionary theory of the multinational corporation'. Journal of International Business Studies, 34 (6): Kuemmerle, W., (1997). 'Building effective R&D capabilities abroad'. Harvard Business Review (March April) ((March April)): Kuemmerle, W., (1999). 'The Drivers of Foreign Direct Investment into Research and Development: An Empirical Investigation'. Journal of International Business Studies, 30:1-24. Kurokawa, S., Iwata, S. and Roberts, E., (2007). 'Global R&D activities of Japanese MNCs in the US: A triangulation approach'. Research Policy, 36 (1):3-36. Lam, A., (2003). 'Organizational Learning in Multinationals: R&D Networks of Japanese and US MNEs in the UK*'. Journal of Management Studies, 40 (3): Lehrer, M. and Asakawa, K., (2002). 'Offshore knowledge incubation: the "Third path" for embedding R&D labs in foreign systems of innovation'. Journal of World Business, 37 (4): Liu, H. and Li, K., (2002). 'Strategic Implications of Emerging Chinese Multinationals: The Haier Case Study'. European Management Journal, 20 (6): Mathews, J., (2006a). 'Catch-up strategies and the latecomer effect in industrial development'. New Political Economy, 11 (3): Mathews, J.A., (2002). 'Competitive Advantages of the Latecomer Firm: A Resource-Based Account of Industrial Catch-Up Strategies'. Asia Pacific Journal of Management, 19 (4): Mathews, J.A., (2006b). 'Dragon multinationals: New players in 21st century globalization'. Asia Pacific Journal of Management, 23 (1):5-27. Motohashi, K., (2006). 'R&D of Multinationals in China: Structure, Motivations and Regional Difference'. RIETI discussion papers. Mudambi, R., Mudambi, S. and Navarra, P., (2007). 'Global innovation in MNCs: The effects of subsidiary self-determination and teamwork'. Journal of Product Innovation Management, 24 (5): Niosi, J., (1999). 'The Internationalization of Industrial R&D: From Technology Transfer to the Learning Organization'. Research Policy, 28 (2-3): Pagano, A., (2006). 'Multinational Companies and the Globalization of R&D in the ICT Sector: Current Patterns in Italy'. Draft. Piscitello, L. and Rabbiosi, L., (2006). 'How does Knowledge Transfer from Foreign Subsidiaries affect Parent Companies?Innovative Capacity?'. DRUID Working Paper. Ronstadt, R., (1978). 'International R&D: the establishment and evolution of research and development abroad by seven US multinationals'. Journal of International Business Studies:7-24. Rugman, A., (1981). Inside the multinationals: The economics of internal markets: Croom Helm, London. Taylor, R., (2002). 'Globalization Strategies of Chinese Companies: Current Developments and Future Prospects'. Asian Business & Management, 1 (2): Tsai, 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): Tung, R.L., (2005). 'Perspectives-New Era, New Realities: Musings on a New Research Agenda?from an Old Timer'. Asia Pacific Journal of Management, 22 (2):

25 von Zedtwitz, M., (2005). 'International R&D Strategies in Companies from Developing Countries-the Case of China'. UNCTAD, Januar:1-11. Von Zedtwitz, M., (2006). 'Internationalization of R&D. Perspectives from outside and inside China, presentation at MOST'. OECD Conference, Chongqing, October. 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 (4): Wong, P.K., (1999). 'National innovation systems for rapid technological catch-up: an analytical framework and a comparative analysis of Korea, Taiwan and Singapore'. DRUID's summer conference on National Innovation Systems. Rebild, Denmark. Wu, J. and Callahan, J., (2005). 'Motive, form and function of international R&D alliances: Evidence from the Chinese IT industry'. Journal of High Technology Management Research, 16 (2): Xie, W. and White, S., (2006). 'From imitation to creation: the critical yet uncertain transition for Chinese firms'. Journal of Technology Management in China, 1 (3): Zanfei, A., (2000). 'Transnational firms and the changing organisation of innovative activities'. Cambridge Journal of Economics, 24 (5): Zhao, Z., Anand, J. and Mitchell, W., (2005). 'A Dual Networks Perspective on Inter-Organizational Transfer of R&D Capabilities: International Joint Ventures in the Chinese Automotive Industry'. Journal of Management Studies, 42 (1):

26 Figure-1 Chinese FDI outflows and the number of Chinese MNCs in top 500 global companies Source: Our elaboration of UNCTAD, WIR 2008 ( and Global 500 reports ( Figure-2 four patterns of international organizational learning Motivations/Learning processes Technology Exploration Cooperative-exploratory learning Experiential-exploratory learning Technology Cooperative-exploitative Exploitation learning Cooperative Learning Source: Hitt et al., 2005 Experiential-exploitative learning Experiential Learning Learning channels Table-1 the basic information of the 9 R&D subsidiaries R&D sub. Set-up Time Sub. location HQ location In China Entry mode Industry Case A 2002 Sweden Guangdong Greenfield investment Telecommunication Case B 2005 Italy Anhui Greenfield investment Automotive Case C 2005 Italy Chongqing Greenfield investment Automotive Case D 2007 Holland Shandong Greenfield investment Consumer electronics Case E 2003 Austria Shandong Greenfield investment Diesel engine Case F 2008 Denmark Zhejiang Greenfield investment Acoustic equipment Case G 2006 Germany Shenyang M&A Machine tool Case H 2005 Germany Zhejiang M&A electromechanical products Case I 2005 Italy Zhejiang M&A Motorcycle

27 Table-2 Interview Protocol General 1 Some basic information of your R&D unit Motivation 2 What are the main motivations of your company to establish this R&D unit? External 3 Who are the local cooperation partners of your R&D unit? R&D network 4 How does your R&D unit cooperate with and learn from the local partners? 5 Can you discuss about the knowledge transfer between your subsidiary and local partners? Internal R&D 6 Can you please briefly discuss the level of autonomy of your R&D unit? network 7 Can you please briefly discuss the coordination and control mechanisms used between your R&D unit and HQ? 8 Can you please briefly discuss how does your R&D unit interact with the other R&D units of the MNC? 9 Can you discuss about the knowledge transfer within your company? FIGURE-3 International R&D motivations of Chinese firms Source: Our elaboration of interviews and questionnaires (Where 1 = totally disagree, and 4 = totally agree)

28 FIGURE-4 Collaboration with local/external partners and cooperative learning Source: Our elaboration of interviews and questionnaires (Where 1 = no interaction, and 5 = Very frequently (at least once a week)) FIGURE-5 Coordination between HQ and R&D subsidiaries and experiential learning Source: Our elaboration of interviews and questionnaires (Where 1 = never/few times (once or twice), and 5 = Very frequently (almost every day)) FIGURE-6 A Model of Tripartite Teamwork Project