THE LOCATION CHOICE OF
CHINESE MNEs
WHICH LOCATION SPECIFIC ADVANTAGES OF THE ADVANCED EUROPEAN COUNTRIES INFLUENCE THE LOCATION CHOICE OF CHINESE RESEARCH AND DEVELOPMENT
INVESTMENTS?
Myrthe Kolsteren 11207639
Date: 27 January 2017 Final Master Thesis
MSc in Business Administration: International Management University of Amsterdam
STATEMENT OF ORIGINALITY
This document is written by Myrthe Kolsteren who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
ABSTRACT
Literature suggested that Europe is losing its competitive advantage as location for foreign direct research & development (R&D) investment. Therefore, the question remains, when Chinese MNEs want to invest in R&D abroad, are they still attracted to Europe and why? This thesis examined which location specific advantages (LSAs) of the most advanced European countries (AECs) influence the location choice of Chinese multinational enterprises (MNEs) when investing in R&D. Existing literature did not address this combination of the location choice of Chinese MNEs when investing in R&D and the LSAs of the AECs. In total, four LSAs have been found, which included the technology infrastructure, the knowledge infrastructure, the quality of human resources in science and technology and the support provided by the government and have been examined via quantitative research. In the sequel, a Binary logistic regression model provided answer to the research question: “Which LSAs of the AECs influence the location choice of Chinese MNEs to these countries when they invest in R&D?” Findings revealed that when examining the cohesive effect of all four LSAs, the technology infrastructure and the quality of human resources in science and technology significantly influence the location choice of Chinese MNEs to the AECs when investing in R&D.
_____________________________________________________________________________ KEY WORDS: Research & Development; Europe; Location specific advantages; FDI; Chinese MNEs, Location choice;
TABLE OF CONTENTS
INTRODUCTION 3,4
CH1. LITERATURE REVIEW 7 - 15
2.1 R&D related firm specific advantages of Chinese MNEs 8
2.2 Country specific advantages of China 8 - 10
2.3 Location specific advantages of the AECs 11 - 14
2.4 Conclusion 14, 15
CHAPTER 3. HYPOTHESIS DEVELOPMENT 16 - 19
3.1. Technology infrastructure 16
3.2 Knowledge infrastructure 16 - 18
3.3 The quality of human resources in science and technology 18
3.4 Support provided by the government 19
CHAPTER 4. METHODOLOGY 20 - 26
4.1 Description of the sample 20, 21
4.2. Data collection 21, 22 4.3 Methods 22 4.4. Variables 22 - 26 4.4.1. Dependent variable 22 4.4.2. Independent variables 22 - 25 4.4.3. Control variables 25, 26 CHAPTER 5. RESULTS 27 - 32
5.1. Descriptive statistics and correlations 27, 30
5.2. Model specification 30 - 32
CHAPTER 6. DISCUSSION 29, 30
6.1 Academic contribution 34 – 36
6.2 Managerial implications 36
6.3 Policy implications 36, 37
6.4 Limitations & Future research 37, 38
CHAPTER 7. CONCLUSION 39, 40
ACKNOWLEDGEMENT 41
REFERENCES 42 - 44
LIST OF FIGURES
Fig 1. The conceptual model 15
LIST OF TABLES
Table 1: Descriptive statistics, means, standard deviations and correlations 28 Table 2: Results of Binary regression analysis for industry activity 29
CHAPTER 1. INTRODUCTION
Over the past decade, China has received enormous attention due to its tremendous growth. However, many questions faced by International Business scholars on world’s second largest economy remain unresolved. One of these popular questions includes foreign direct investment (FDI) by Chinese multinational enterprises (MNEs) and their location choices. In particular, recent evidence indicates that the process of locating Research & Development (R&D) units abroad has accelerated, and that the pattern and nature of internalization of R&D has also changed (Atkinson, 2007; Lewin et al., 2009; Manning et al., 2008). In an article in the Financial Times, Joe Jimenez, chief executive of Novartis, states: “China’s going to become very important in R&D in this industry, not in the short term but over the long term.” (Waldmeir, 2016, p1). Whereas previously, R&D activities were aimed mainly at adapting products to the local markets, in recent years, an increasing number of R&D activities have been performed abroad in order to tap into the pool of competences of the host locations (Cantwell, 1995; Castellani, Jimenez & Zanefi, 2013). In emerging countries as China, only few organizations possess well-established R&D operations. Therefore, tapping into already existing world knowledge stock seems to be a natural way of bridging the technology gap that they face, rather than trying to enhance the domestic technology frontier by themselves (Hu, Jefferson & Jinchang, 2005). Although the location choice of Chinese MNEs is one of the most crucial decisions they have to face, there still remains a lot to examine.
Previous literature revealed many research gaps left for further investigation. First, examination of the current literature found serious limitations on the amount of firm level control variables (Duanmu, 2012). These firm level control variables include characteristics such as firm size, experience and performance. It is likely that these variables can affect the location choices of MNEs. Additionally, previous research indicates that future research should dive deeper into the complexity of global, regional and local levels. Research acknowledges that MNE activities often reflect different layers of relationships, inter alia regions that deserve further attention. Furthermore, previous studies showed that it is of importance to examine the possible interaction effects between firm specific and location- specific factors on the location choice of FDI (Makino, Lau & Yeh, 2002). It is demonstrated that outward FDI is significantly influenced by the specific motivations and capabilities of MNEs, suggesting that the effects of location specific advantages (LSAs) and firm specific advantages (FSAs) might be better specified as an important interaction (Dunning, 2000; Makino, Lau & Yeh, 2002). Although literature has found that Chinese MNEs are not endowed
with such knowledge based FSAs, they do possess important country specific advantages (CSAs) on which they are building scale economies (Rugman & Li, 2007). Therefore, this thesis will mainly focus on China’s R&D related CSAs and the LSAs of the host location in order to examine the location choice considered by Chinese MNEs when investing in R&D.
The location choice of Chinese MNEs will be limited to the most advanced countries of Europe (AECs). Previous literature found that the emerging countries within Asia Pacific have become more popular with MNEs as offshore R&D locations, while Western Europe seems to be losing its competitive advantage (Atkinson, 2007; Huggins et al., 2007; Demirbag & Glaister, 2010). Although Europe does not seem to be the number one location for Chinese R&D MNEs in terms of stock, in terms of growth rates, it has outpaced other regions such as the United States (Nicolas, 2014). The question remains, are Chinese R&D MNEs still attracted to the AECs and if so, which factors drive them to establish their units in Europe? For the purpose of this analysis, the most advanced European countries will be measured on national level.
To identify the importance of various LSAs on location choice, a comparison will be made between the R&D related FSAs of Chinese MNEs, the CSAs of China and the LSAs of the AECs. An econometric analysis will further be conducted by using micro and macro data from Chinese R&D MNEs on outward FDI to the AECs as important player of the regional triad. The contribution of the analysis is that it enables to quantify the importance of FSAs, CSAs and LSAs and their influence on the location choice of Chinese MNEs.
The structure of this thesis is as follows. First, the next section will discus the existing literature on Chinese FSAs, China’s CSAs and the LSAs of the AECs. Consequently, in the third section, the hypotheses derived from the theoretical framework will be further elaborated. The fourth section describes the methodology part, where the data collection, the variables included in this thesis and the research method utilized in this thesis will be explained. In the sequal, the fifth section describes the results gained from the statistical analysis. Furthermore, the discussion part will further elaborate on the outcomes gathered in the result section and discusses the contribution of this thesis, its implications, limitations and possibilities for future research. Finally, this thesis will finish with its final conclusions in the last section.
CHAPTER 2. LITERATURE REVIEW
MNEs investing in R&D aim to develop new technology or information in order to improve the effectiveness of products or to enhance production processes more efficiently. For that reason, R&D demands for a high level of knowledge and expertise. In particular, this thesis will focus on knowledge-intensive R&D activities executed by Chinese MNEs. They gain competitive advantage through the use of social and human capital that make up their unique trading assets (Swart & Kinnie, 2003; Alvesson, 2001; Frenkel et al, 1999; Lei et al, 1999; Newell et al, 2001; Purvis et al, 2001; Starbuck, 1992). Additionally, most work is said to be of an intellectual nature and where well-educated, qualified employees form the major part of the workforce (Swart & Kinnie, 2003; Alvesson, 2000).
In this thesis, the FSAs of Chinese MNEs, the CSAs of China and the LSAs of the AECs will be used to explain the location choice of Chinese MNEs. The FSAs of an organization determine the organization’s competitive and sustainable advantage when executing foreign direct investment. They can be defined as knowledge bundles that take the form of intangible assets, learning capabilities, and even privileged relationships with outside actors (Rugman and Verbeke, 2003). Additionally, they might be achieved through technology, distributional skills or marketing. LSAs on the other hand are unique to the host country where the business activities are performed. Examples of LSAs include natural resources endowments, cultural factors and labor costs.
It can be stated that the growth of emerging economies as China, over the last decade, has particularly led to a considerable expansion of various kinds of strategic asset seeking FDI when considering the ownership, location and internalization advantages. Conventional wisdom also states that when MNEs go to developed countries, they tend to do that in terms of technology-seeking. They argue that Chinese MNEs investing in R&D localize their subsidiaries in countries where knowledge is advanced, particularly when they aim to acquire state-of-the-art knowledge that is not available at the home country (Asakawa & Som, 2008; Peng & Wang, 2000; Song & Shin, 2007). Gassmann & Han (2004) talk about supply-oriented factors that Chinese MNEs are seeking in advanced countries. These factors include sophisticated scientific infrastructure, knowledge inputs and access to cutting-edge technology. Moreover, they argue that the lacking availability of well-educated R&D specialists in China, combined with low personnel costs are further incentives for MNEs to establish FDI in advanced countries. This has been confirmed by Di Minin, Zhang & Gammeltoft (2012) who argue that technology exploration is currently the most important goal of Chinese R&D MNEs. However, they also
acknowledge that previously technology exploration-dominated Chinese MNEs also tend to combine technology-exploration more with technology-exploitation activities, due to maturation of technology. Similarly, Lu, Liu & Wang (2010) imply that emerging MNEs need to balance strategic asset exploitation and exploration in the process of FDI and use FDI as a springboard to acquire strategic resources.
2.1 R&D related firm specific advantages of Chinese multinational enterprises
As mentioned before, FSAs determine the organization’s competitive and sustainable advantage. Therefore, it is of importance to examine the FSAs of Chinese MNEs r e l a t e d t o R & D in order to define their needs for outward FDI. However, previous firm level studies found that Chinese R&D MNEs lag behind Western firms in the development of their FSAs, in particular technology. Specifically, they tend to be labor intensive, protected, inefficient firms with low technology skills. Consequently, it is a challenge for them to improve their FSAs (Rugman & Li, 2007). On the other hand, Chinese MNEs are in the advantage of certain Chinese CSAs as cheap, unskilled and skilled labor and the support they get from the government.
To conclude, it is not likely that Chinese MNEs will go abroad in any significant numbers over the next decade on the basis of their FSAs (Rugman & Li, 2007). In particular, Chinese R&D MNEs are likely to overcome their disadvantages by tapping into locational advantages abroad.
2.2 Country specific advantages of China
As mentioned before, Chinese MNEs are in the possession of certain CSAs rather than FSAs when it comes to the R&D sector. CSAs are typically Ricardian type resource endowments (Kedia and Mukharjee, 2009; Demirbag & Glaister, 2010), including infrastructure and country institutions as political risk and local policies. The next section will cover the CSAs that give Chinese R&D MNEs a significant advantage or disadvantage.
Country specific disadvantages
Managerial expertise
A frequent CSA mentioned often in literature, or more accurately, the lack of it, concerns managerial expertise. Von Zedtwitz (2005) states that Chinese firms do not have
much experience in running or participating in international firms, consequently only few of them are qualified to manage R&D activities. They lack managerial capabilities and suffer from a Penrose effect of top management talent (Rugman & Li, 2007). Similarly, Deng (2007) argues that Chinese MNEs need to expand to other countries in order to gain managerial know-how to offset their competitive disadvantage. Given this lack, the majority of upper R&D management are staffed by foreign expatriates (Gassmann & Han, 2004). In reaction to these problems, the Chinese government encourages PhD holders from universities in the United States to return to China with lucrative incentives (Saxenian, 2006; Asakawa & Som, 2008)
The quality of human resources in science and technology
According to the data in the Annual Census of Industrial Enterprises (2002; Rugman & Li, 2007), the average labor productivity of local Chinese firms has augmented by 45%. With that, the productivity gap from foreign companies in China, measured by the ratio of the average productivity of foreign organizations to that of local Chinese ones, decreased from 5.25 to 3.29. It therefore can be suggested that local Chinese organizations have improved their CSAs over time, but in terms of productivity they still lag behind the foreign companies in their own country. On the one hand, China offers one of the largest human resources pools when it comes to R&D (Li & Zhong, 2003). On the other hand, the quality lags far behind that of developed countries (OECD, 2008; Di Minin, Zhang & Gammeltoft, 2012). Chinese designers and engineers are criticized for their lack of originality or creativity because ‘‘The Chinese education system and culture don’t encourage individualistic expression and creativity’’ (von Zedtwitz, 2006; Di Minin, Zhang & Gammeltoft, 2012).
Technology infrastructure
One of the most striking CSAs that Chinese R&D MNEs do not seem to possess according to current studies is advanced technology. Noland (2004; Rugman & Li, 2007) found only little evidence that Chinese MNEs are able to develop knowledge of the systems integration skills that characterize successful Western MNEs. According to him, the Chinese organizations tend to be inefficient, labor intensive, with low technology skills. This is confirmed by Deng (2007; Nolan, 2001) who states that because Chinese MNEs suffered from a weak innovation system, they lack proprietary technology and innovative capabilities. This is further illustrated by their R&D business spending that was only a fraction of that of the NAFTA, Japan and some European countries. In the contrary, Motohashi (2015) argues that China has abundant science and technology human resources and that even foreign MNEs can
enjoy the advantages of China’s local science base. Furthermore, some of the research conducted by Chinese leading universities is regarded magnificent by international standards. However, most scholars still argue China to lack of advanced technology compared to many other countries in Europe and the NAFTA. They argue that due the lack of superior technology, which has been called the biggest disadvantage, Chinese MNEs could overcome this disadvantage by tapping into locational advantages abroad.
Country specific advantages
R&D labour costs
Sun, Zedtwitz & Simon (2013) emphasize that Chinese firms provide extensive supplies of relatively low-cost labor, which provides them a comparative advantage in low-cost labor. Moreover, interviews conducted by Rui & Yip (2008) confirmed that China is in advantage of low labor costs. Similarly, Lu, Liu & Wang (2010) mention that top researchers in countries as China only cost a fraction of their counterparts in the Unitated States (Camel, 2003).
Support provided by the state
The Chinese government has expressed its concerns together with Chinese academia about how R&D capabilities of Chinese MNEs can be cultivated in order to compete with global MNEs (Di Minin, Zhang & Gammeltoft, 2012). The Chinese government has initiated a R&D policy that emphasizes building up a country that is innovative with an enterprise-centered national technology innovation system. Due to this governmental support, the industrial and science parks are likely to become centers of excellence according to interview partners in research conducted by Gassmann & Han (2004). Also Li & Zhong (2003) mention China’s long-term development policy in science and technology and its forthcoming rapid growth. Currently, the country is investing in advanced materials as biotechnology, industrial automation, microelectronics and so forth.
In short terms, Chinese firms are believed to use FDI as a means to leverage their competitive advantages (Boisot, 2004; Zeng & Williamson, 2003; Rui & Yip, 2008). Literature revealed that their current competitive advantages mainly stem from low labor costs, a large pool of R&D human resources and governmental support. Through internalization they must overcome their disadvantages of the lack of managerial expertise, labour skills and efficiency and a poor technology infrastructure.
2.3 Location specific advantages of the AECs
The LSAs of the AECs can be customized in order to examine the location choice of Chinese R&D outward FDI into the AECs. This can be done by measuring the differences between China and the AECs in terms of factors that support or impede the creation of knowledge. In the study by Demirbag & Glaister (2010) and Kedia and Mukharjee (2009) these R&D related LSAs have been conceptualized as location level advantages, human capital related advantages, advantages of labor arbitrage, advantages of knowledge arbitrage, and advantages of time arbitrage. Some of these LSAs have also been acknowledged by Yizhi (2009), who states that important motives of Chinese MNEs are to obtain high quality local research and the advanced development of human resources. For that reason, he argues, MNEs will choose regions where high quality of science and technology is available and where one can find a pool of talents and universities. Furthermore, besides the previous factors mentioned, Dunning (2000) showed how management expertise within a country and reputation for being established in a prestigious market is of importance. Supporting this assertion, von Zedtwitz (2005) noted that it is of importance to have experience in participating in international companies in order to gain managerial R&D expertise.
Chinese MNEs tend to search for slightly different LSAs than other emerging or developed countries. Many MNEs typically invest in neighboring countries in such a way that the level of development of the host country is equal or lower than their own (Kumar, 1998; Peng 2007). Chinese MNEs on the other hand tend to invest more in developed countries due to their superior investment environment, advanced management methods and technology know-how. This is confirmed by Wang (2002) who reveals that more than 70 percent of Chinese overseas subsidiaries have been established in the industrial countries. Moreover, Ambos & Ambos (2011) also acknowledge scientific and technological capital as one prominent factor for Chinese MNEs to base their location choice on. By offshoring to leading technological nations, organizations aim to increase their technological base through capturing spillovers created by the institutions and organizations of the host firm (Feinberg and Gupta, 2004; Ambos & Ambos, 2011). Besides technological capital, Buckley, Devinney, and Louviere (2007), also demonstrate that human capital is of importance. Previous literature has learned that expansion to the regions of the NAFTA and Europe has allowed Chinese organizations to obtain such capital and with that enhance their competitive advantage. Furthermore, the institutional environment of the host country is also often mentioned as external factor that pushes Chinese MNEs into a certain region. Finally, literature found that managers are primarily attracted by a host country large market size, when they want to
exploit their knowledge abroad (Ambos & Ambos, 2011).
The LSAs of the AECs are typically Ricardian type resource endowments. They represent the benefits that an organization can obtain by internalizing its activities within the value chain within Europe. Examination of previous literature on the LSAs have found some general advantages of the AECs related to the location decisions of MNEs.
Technology infrastructure
In his study, Carlsson (2006) argues that Europe seems to be the only supranational technological and scientific region, compared to Japan and the interaction between Canada and the USA. In particular, they state that technological excellence is increasingly driving MNEs in their locational decisions. Also, Di Minin, Zhang, Gammeltoft (2012) denote that Europe is one of the most popular destinations for Chinese outward FDI, inter alia because of its diversified technological base. In particular, Chinese MNEs seek for technological support to compete. In one of their interviews, it is stated: “It is not enough to rely solely on the R&D forces in China to catch up with our competitors in a short time, unless we have good technological support’’ (Di Minin, Zhang, Gammeltoft, 2012). In other words, Chinese MNEs establish their units in Europe to extend their technology strategy towards Europe’s technology base so that they can get external aid concerning key technology issues. They act as knowledge-seekers for technology and emphasize their roles as learners/absorbers. Additionally, it is found that the exploration of technology is still more important for Chinese MNEs than exploiting. However, adjacent to the maturation of technology, these technology-exploration activities now tend to combine technology- exploration with technology-exploitation activities (Di Minin, Zhang, Gammeltoft, 2012). Furthermore, according to Erken & Kleijn (2010) the attractiveness of Europe is dependent on several requirements. First of all, sources of knowledge need to be present. Second, it is of importance that there are industry-specific and cluster-based spillovers. Finally, high technological specialization must be available to capture general-purpose spillovers.
Knowledge infrastructure
Falk (2012) finds that knowledge intensive FDI location decisions are likely to be driven by the availability of a knowledge base and qualified universities. This has been acknowledged by Siedschlag et al (2010; Daniels and Lever 1996; Florida 1997; Zedtwitz and Gassmann 2002; Dicken 2004; OECD 2008) who point to the access to a strong knowledge- base as important advantage of Europe that drives MNEs to establish their units within this
region. Moreover, Abramovsky et al (2007; Siedschlag et al., 2010) finds that foreign R&D units base their location in the proximity of university-centers in the United Kingdom, because they provide MNEs access to high quality researchers for basic scientific research. Furthermore, Europe is in possession of top ranked universities that enables high-quality standards of knowledge. Also Cantwell & Piscitello (2002) refers to the effort of MNEs to gain access to advanced technology via public research centers, adequate educational systems, universities and a high quality science-base (Kline and Rosenberg, 1986; Nelson, 1993; Nelson and Rosenberg, 1999; Rosenberg and Nelson, 1996; Breschi, 2000).
The quality of human resources in Science and Technology
Di Minin, Zhang, Gammeltoft (2012) consider human capital as one of the key factors for Chinese MNEs to establish their subsidiaries in Europe. In particular, they refer to the high-qualified European scientists and engineers. More specifically, they state that the use of European human resources and their advanced technological knowledge gives Chinese MNEs access to the local knowledge environment. Their study reveals that highly skilled engineers and designers are the key factors of success. Additionally, they mention that in order to improve the technical learning capabilities of Chinese human resources, Chinese designers and engineers should seek for opportunities to cooperate with European colleagues. In other words, specialized human resources in Europe is one of the drivers for Chinese MNEs to set up overseas units not only to get external technological assistance but also to cultivate the development of high-quality Chinese human resources (Di Minin, Zhang, Gammeltoft, 2012). Similarly, Falk (2012) states that Chinese R&D outward FDI is unlikely to be driven by low labor costs, but rather is dependent on inter alia skilled workers. In particular, he argues that a skilled labor force in European host countries increases the likelihood for Chinese MNEs to invest in this region. Moreover, Siedschlag et al. (2010) also emphasizes human capital as important advantage of European countries.
Support provided by the government
European countries actively try to take a position in the minds of investors as destination for R&D purposes and also invest in international advertisement campaigns for the same purpose. At the same time, they increase their efforts to encourage foreign investors to get involved in R&D and to participate in national funding programs (guimon, 2011). In particular, they provide a favorable tax treatment to R&D expenditure and may take the form of accelerated depreciation, tax credits or import tariff exemptions. Although literature suggests that incentives
do not drive the choice of location significantly, it is recognized that they are able to influence the final decision in terms of competing destinations based on other attraction factors (Cantwell and Mudambi, 2000; Guimon, 2011).
Market
Many studies also emphasized the important role of the European market to drive MNEs towards Europe. Chinese organizations base their location choice inter alia on Europe’s attractive consumer market according to Di Minin, Zhang, Gammeltoft (2012). Additionally, they state that Chinese MNEs also locate their units in Europe to satisfy the demands of European customers. Moreover, results of the study conducted by Siedschlag et al. (2010) suggests that European market potential increases the likelihood of R&D foreign units in the European region. Also Falk (2012) confirms the market size of Europe as an important determining variable that guides Chinese MNEs towards European countries.
To summarize, the AECs are endowned with several LSAs that attract MNEs to invest in R&D in these countries. These LSAs include technology infrastructure, knowledge infrastructure, the quality of human resources in science and technology, support provided by the government and the market.
2.4 Conclusion
This literature review has been written in order to conduct research on how Chinese MNEs are guided towards their location choice when performing outward FDI in R&D. In particular, it took a view upon the F S A s o f C h i n e s e M N E s , Chinese r e l a t e d R&D CSAs, and the LSAs of the AECs. It can be stated that Chinese MNEs are not significantly in the possession of important R&D related FSAs and they therefore are likely to overcome these disadvantages by tapping into the LSAs of host countries, more specifically, the AECs. By comparing the CSAs of China with the LSAs of the AECs, there has been a consensus on which variables seem to be involved for the purpose of this study. Therefore, several hypotheses can be formulated.
First of all, this literature review came up with the most important CSAs related to R&D in China. It seems that R&D labor costs and R&D support by the government are two important positive R&D related CSAs of China, while China’s technology infrastructure, managerial skills and the low quality of human resources in science and technology are regarded as negative CSAs of the country. On the contrary, the AECs do seem to possess the quality of
human resources in science and technology, technology infrastructure and knowledge infrastructure in terms of important LSAs. Therefore, Technology infrastructure, knowledge infrastructure and the quality of human resources in science and technology are included in the conceptual model. Furthermore, the AECs are likely to be in the possession of strong governmental R&D support. Consequently, this variable has also been included in the final conceptual model: figure 1: The conceptual model. This figure includes the four independent variables that will be examined in this thesis. Because examination of existing literature found that Chinese MNEs are more likely to invest their R&D activities abroad due to reasons of strategic asset seeking in stead of market-seeking, the European market has not been included.
All in all, further research needs to be conducted in order to test the previous mentioned relationships between the LSAs of the AECs and Chinese MNE location choice.
CHAPTER 3. HYPOTHESIS DEVELOPMENT 3.1. Technology infrastructure
Examination of the literature has found that the technological infrastructure of China lags behind the infrastructure of the AECs. Therefore, scholars argue that Chinese MNEs should overcome this location specific disadvantage by tapping into countries abroad that are in the possession of an appropriate technological infrastructure. European countries are often the most popular destinations because of their diversified technological base. Carlsson (2006) even states that Europe seems to be the only supranational scientific and technological region. This has partly been confirmed by Patel & Pavitt (1987), as they also acknowledge the strong technological position of the European countries. However, they differentiate between different industries. In particular, they argue that non-electrical machinery and automobiles are important sectors of Europe’s technological strength. Moreover, the global competitiveness report (2010-2011) states that it are especially the Nordic members of Europe that continue in holding a privileged position in the world-wide technological ranking. They furthermore show that the United Kingdom is harnessing the latest technologies available to improve their productivity rates. In order to accomplish this, they invest in sophisticated and innovative business, characteristics that enhance productivity. Furthermore, according to Erken & Kleijn (2010), the attractiveness of Europe is dependent on several requirements. First of all, sources of knowledge need to be present. Second, it is of importance that there are industry-specific and cluster-based spill overs. Finally, high technological specialization must be available to capture general-purpose spill overs. Based on the review of the literature, it can be concluded that the AECs are in the possession of an advanced technological infrastructure above the global threshold. Hence:
H1. The technological infrastructure of the advanced European countries influences the location choice of Chinese R&D investments to these countries.
3.2. Knowledge infrastructure
Several studies have emphasized that the location decisions of R&D investments are likely to be guided by the availability of a strong knowledge base and qualified universities within a country. Specifically, Demirbag & Glaister (2010) argue that the main motive for R&D investments in terms of location choice is the possibility to augment the knowledge base of the firm by tapping into knowledge advantageous locations. Additionally, they state that Europe is known to have a high level of education.
It is argued that the advanced European countries are in the possession of this strong knowledge-base that serves as an important location specific advantage fore these countries. Moreover, foreign R&D units are often based in the proximity of university-centres available in the advanced European countries. Also Thursby & Thursby (2006) showed that it is mainly the intellectual capital and the collaboration between universities that act as an important trigger for foreign MNEs to invest in countries with these characteristics. Keeble et al (1999) point out to the effect of collective learning on the growth of high technology SMEs in the Cambridge region and highlights the importance of spin-offs, local network and linkages; and labour market recruitment for innovative capacity. Another example they mention is the Sophia-Antipolis in the Cote d’Azur, which contributes to the development of new high technology capabilities due to collective learning. Also, the European Union announced that they have the goal of becoming the most competitive and dynamic knowledge-based society in the world. They will try to achieve this by production and diffusion of knowledge, which they believe is the engine of economic and social progress. Hamdouch & Moulaert (2006) arrived at nine key findings regarding knowledge creation and innovation. One of these findings argues that knowledge, innovation, learning and competences are key factors for the European countries, which determine their economic growth and competitiveness. Furthermore, in the case of the United Kingdom, Foray (2006) revealed evidence that countries that invested in the United Kingdom benifited from the country’s knowledge-base. However, it is important to note that these results showed strong variations amongst countries, with particular large spillover effects for Switzerland and Sweden. Also, a study by Antonelli & Calderini (2008) confirmed the strong and effective role of knowledge compositeness as an indicator of the actual technological and competitive performance of the European firms in their study. Finally, knowledge seems to be embedded in human capital and enhances distinctive competences to discover innovation opportunities (Cehn & Huang, 2009). In other words, they argue that human capital is necessary in order to produce creative ideas, innovate new approaches and exert new opportunities by their absorption of knowledge. Similarly, Hatch & Dyer (2004) showed that human capital on firm level is a resource that stems from learning. If employees acquire knowledge, they become capable of making contributions to the learning performance of the company, and thus human capital becomes a source of competitive advantage. Most knowledge remains tacit in the understanding and skills of the employees.
In short, review of the literature showed that the European countries possess an advanced knowledge infrastructure, which acts as an important motive for foreign MNEs to invest in these
countries. Moreover, review of the literature found several indirect effects between knowledge and the location choice for an AEC and R&D investments via enhancing the technological infrastructure and human resources in scientists and technicians. Hence:
H2. The knowledge infrastructure of the advanced European countries influences the location choice of Chinese R&D investments to these countries.
3.3. Quality of human resources in science and technology
China possesses one of the biggest R&D human resources pools. However, the efficiency of this human resources pool lags far behind that of the advanced European countries. In particular, Di Minin, Zhang, Gammeltoft (2012) argue that human capital in terms of R&D is one of the key factors of the advanced European countries that guide Chinese R&D MNEs to invest in these countries. Moreover, Di Minin, Zhang & Gammeltoft (2012) found that specialized European human resources guide Chinese R&D investments towards these countries to get technological assistance and additionally cultivate the development of high-quality human resources.
First, Gassman (1998) examined that there was a significant increase of foreign attracted R&D human resources in the European countries. Especially, Austria, Switzerland, France, the United Kingdom and Sweden ought to be European centres for R&D personnel. In particular, 5500 people have been employed for software purposes alone. Furthermore, Siedschlag et al. (2010) also emphasizes human capital as important advantage of European countries. Moreover, the OECD Science, Technology and Industry Scoreboard (2009) point out to the average of R&D intensity of research and expenditure for Finland, Sweden and Iceland, which was above the OECD average.
To conclude, the quality of human resources in scientists and technicians seems to influence the location choice for European countries positively when it concerns R&D investments. Especially, the AECs seem to be in the possession of this high quality. Furthermore, the quality of human resources in scientists and technicians also seems to have a positive influence on the technological infrastructure of a country. Hence:
H3. The quality of human resources in science and technology of the advanced European countries influences the location choice of Chinese R&D investments to these countries.
3.4. Support provided by the government
The advanced European countries actively try to take a position in investors’ minds when they seek for a destination for R&D purposes through the aid of international campaigns. Moreover, they encourage investors via favourable tax treatments. When reviewing existing literature it seems that governments of advanced European countries have a higher R&D expenditure and also provide more effective tax treatments for R&D MNEs, than many other countries.
Moreover, research of the European regions showed that funding has a significant higher influence on the productivity of researchers than collaboration within the network (Pavitt, 1998). In particular, when funding supports R&D in terms of technology, it can have major unintentional technological spill overs. Also Thursby & Thursby (2006) showed that European governments offered significant tax breaks and/or direct assistance. Furthermore, Carayannis, Alexander & Ioanndis (2000) revealed that company partnerships with inter alia government agencies contributes to the most effective outcome of knowledge integration. The moderate subsidy of the European Commission leverages further investments across national borders, which enhances collaborative research. Countries that excelled themselves include France, the United Kingdom, Sweden and Germany. Moreover, according to Allen et al (1978) industrial innovation does not occur in vacuum, but with aid from environmental factors as government policies. Highly skilled capital, in particular the migration of it, is concerned to be an inseparable segment of national technology and development policies as well, according to Mahroum (2000). In a later study, he confirms the importance of the availability of skilled human resources and its influence on the economic competitiveness. Hence:
H4. The R&D support by the government of the advanced European countries influences the location choice of Chinese R&D investments to these countries.
CHAPTER 4. METHODOLOGY
This thesis aims to research the location choice of Chinese Research & Development (R&D) investments made in the most advanced European countries (AECs). Existing literature has not or hardly examined the factors that have an influence on the location choice of R&D investments made by Chinese MNEs. Hence, this thesis aims to fill that gap by answering the question: Why do Chinese MNEs perform their R&D investments in AECs? In particular, this study will focus on the location specific advantages (LSAs) of the AECs and their influence on the location choice of Chinese R&D investments to the AECs. Consequently, this thesis provides answer to the question: Which LSAs of the AECs influence the location choice of Chinese R&D investments to the AECs?
4.1. Description of the sample
The dependent variable in this thesis is the Chinese investment made in an AEC and whether this investment is an R&D-investment or not. In particular, this dependent variable serves as a dummy variable that equals one if the foreign direct investment (FDI) concerns an R&D investment in an AEC and equals zero if not. It is therefore of importance to specify which countries can be counted for as an AEC and which investments include R&D investments.
First, literature has not or hardly compared European countries based upon their R&D components. Therefore, this study faced difficulties in determining the most advanced countries in Europe. Nevertheless, due to the use of the global competitiveness Reports for the data collection of the independent variables, these same reports could also be used to determine the most AECs. In particular, for the timespan of 2003-2011, the scores of all European countries on four competition pillars related to R&D were compared in order to reveal the AECs. These pillars included ‘higher education and training’, ‘technological readiness’, ‘Business Sophistication’ and ‘Innovation’. The comparison of the scores on these four dimensions revealed a total of nine European countries that scored very high on average and therefore were selected as an AEC. These nine countries included Austria, Denmark, France, Finland, Germany, the Netherlands, Sweden, Switzerland and the United Kingdom. This result has been confirmed by several other studies. First, the industrial research institute (2016) acknowledged almost all of the aforementioned nine countries, except for Switzerland, as the most atrractive countries in Europe for R&D activities. They did not include Switzerland, because they only focused on the countries that belong to the European Union in stead of Europe in general. Furthermore, the European Commission (2013) listed almost the same European countries as
most important countries as well. Again, Switzerland was ignored due to the reason that the country is not a member of the European Union.
Second, the investments that contained R&D activities had to be identified. These were mentioned in the dataset as ‘Research & Development’ and ‘Design, Development & Testing’, and included a total of 133 investments. These two activities were chosen for the sample, because they represent the goal of MNEs investing in R&D conform the definition in the literature review: “MNEs investing in R&D aim to develop new technology or information in order to improve the effectiveness of products or to enhance production processes more efficiently”.
4.2. Data collection
In order to provide answer to the research question, data on Chinese R&D investments in the AECs and the location specific advantages of the AECs had to be collected through data held by variegated databases.
This thesis used a dataset of 2093 Chinese Greenfield investments made from 2003 to 2011 provided by the Financial Times database (Financial Times, 2011). Within this dataset, the dependent variable, which is a dummy variable that equals 1 if the investment comprised an R&D investment and equals 0 if not, was easily identified. This dataset was highly appropriate due to its wide timespan and rich information, which could be used for most of the control variables in this study. Furthermore, the profiles of each company were also provided throughout this dataset.
In order to create the final dataset out of the Financial Times dataset of 2093 Greenfield investments, several steps had to be conducted. In the first place, the investments that contained R&D activities under the list of the industry activities needed to be identified and be provided with a dummy variable that equals one.
Second, the Chinese investments made in the AECs had to be identified. To smoothen this process, the filter button for each AEC, described in the previous section, eased the search for these nine countries.
Finally, data on the independent variables, which include the LSAs of the AECs, have been compiled by data held by variegated other databases. These LSAs include: ‘the technological infrastructure’, ‘the knowledge infrastructure’, ‘the quality of human resources in science and technology’, and ‘support provided by the government’. Mainly the OECD
database and the Global Competitiveness Reports of the World Economic Form have been used.
these reports provide data for the United States and all ten AECs included in this study. Moreover, these reports were found in the UvA library and the OECD databank by using various combinations and varieties of the keywords that are affiliated to the mediator variables.
4.3. Methods
As a result of the deductive research approach of this thesis, a quantitative research method will be used. This method is highly appropriate, as the purpose of this research is to seek for the explanation why Chinese MNEs perform their R&D investments in the AECs. Through quantitative research it can be hypothesized if the LSAs of the AECs drive R&D investments of Chinese MNEs to these AECs. As a result, relations between the LSAs of the AECs and the location choice of Chinese R&D investments in the AECs should be found through collecting and analysing numerical data in quantitative research. Because data has only been collected for the AECs, the location choice of an AEC could not be treated as outcome, as there was nothing to compare. Therefore, the Chinese investment made and whether this investment is an R&D investment or not became the dependent variable. The LSAs of the AECs are the independent variables. In particular, the four LSAs of the AECs comprise the four independent variables that explain the likelihood that the investment in the AEC is an R&D investment when seeking for the influence of the LSAs. Results will be obtained by measuring the effects of the independent variables on the dependent variable. The results should reveal if the relations between the dependent variable and the independent variables are significant in order to accept the hypotheses.
4.4. Variables
4.4.1. Dependent variable
R&D vs. Non-R&D investment
The dependent variable of this study is the investment made by a Chinese MNE, which is a dummy variable that equals one if the Chinese MNE made an R&D investment and equals a zero if they made a non-R&D investment. The dummy variable is useful to sort the data of the dependent variable into two mutually exclusive categories.
4.4.2. Independent variables
The independent variables have been found during a thorough review of the literature. These four variables include ‘the technology infrastructure’, ‘the knowledge infrastructure’, ‘the
quality of human resources in science and technology’and ‘support provided by the government. For each independent variable, statistics had to be gathered from 2003 to 2011 for the AECs. Hereafter, comparisons could be made between the averages of the AECs on each variable compared to the R&D investments and the non-R&D investments.
Technology infrastructure
This variable contains the technological infrastructure of the AECs. For China it can be stated that the technological infrastructure is a locational disadvantage, while the technological infrastructure serves as a LSA for the AECs. It can be argued that in order to measure the technological infrastructure, many constructs can be used. After review of the current literature, this thesis will focus on the construct of technological infrastructure by Archibugi & Coco (2005). They measured the technology infrastructure as construct of the three main categories of technology: 1) Innovative capacity, 2) ICT diffusion and 3) Technology transfer.
The statistics on the technology infrastructure conform the construct given by Archibugi & Coco (2005) were mainly found in the Global Competitiveness Reports for the years 2004-2005, 2006-2007, 2008-2009, 2009-2010 and 2010-2011. Results for the years 2003-2004, 2005-2006, and 2007-2008, were obtained via interpolation of the data for the other years. These reports are the property of the World Economic Form and assess the competitiveness landscape of 140 economies by focusing on 12 pillars. First of all, the innovative capacity as component of the technological infrastructure was given for all countries included in this study. However, ICT diffusion and technology transfer were not given precisely or were only given for a small part of the timespan. Therefore, this study replaced ICT diffusion for ICT relating laws and technology transfer for technological absorption. ICT diffusion was only given for 2003 and 2004 and did not continued. Therefore, this component could not be used in order to measure the technological infrastructure for all countries from 2003 to 2011. To resolve this problem, laws relating to ICT replaced this component as it served as an extension on ICT diffusion. Where ICT diffusion focuses more on ICT investments and the use by households (Pilat, Ahmad & Schreyer, 2004), laws relating to ICT builds upon this component by measuring the laws relating to the use of information technology and wether they are non-existent or well developed and enforced. Moreover, technology transfer was neither given for most years and has therefore been replaced by technological absorption. The reason behind this replacement is that technology transfer is strongly associated with technology absorptive capacity. In particular, technology absorptive capacity impacts the effectiveness of technology transfer performance (Lin, Tan & Chang, 2002).
The Global Competitiveness Reports provided the statistics on all three components of technology infrastructure discussed above for all given countries. They gave all countries an average between 1 and 7 to compare the countries. For each country, the three averages of the components were added together and divided by three to get the final average.
Knowledge infrastructure
Knowledge infrastructure concerns measuring the knowledge indices of the AECs. Statistics on the knowledge infrastructure are not a given. This variable is a construct of several components. For this thesis, a composite index of in total four knowledge indicators is created in order to measure the knowledge infrastructures of each country, which is a similar approach of that of Demirbag and Glaister (2010). These four indicators include ‘technological readiness’, ‘quality of scientific research institutions’, ‘firm level technology absorption’, and ‘quality of maths and science’ and can be found back in the Global Competitiveness Report as well for each country and almost all years given. However, results for the years 2005-2006 and 2007-2008, were obtained via interpolation of the data for the other years. For each country, the four averages of the components were added together and divided by three to get the final average for the infrastructure of knowledge.
Quality of human resources in scientists and engineers
The quality of human resources in science and technology refers to the high quality level of R&D human capital on average within a country. Literature revealed that this variable serves as a LSA for the advanced European countries, while it is a location specific disadvantage for China. There was only limited data available on the quality of R&D personnel in the AECs, due to the difficulty of defining and measuring quality. However, data on the quality of human resources in science and technology for the years 2004, 2006, 2008, and 2010 have eventually been found in the dataset provided by the OECD Science, Technology and Industry Scoreboard. Results for the years 2005, 2007 and 2009 were obtained via interpolation of the data for the other years. This scoreboard provides the latest data available and compares the OECD countries with each other in order to explore the growing interaction between knowledge and globalization at the heart of the ongoing transformation of OECD economies.
The dataset provided by the OECD Science, Technology and Industry Scoreboard was appropriate for the definition of the quality of human resources in science and technology argued in this thesis as it set some high conditions to meet the definition of high-qualified R&D human capital. First, data on high-qualified R& personnel in this dataset only encompasses
workers in professional, technical or managerial occupations. Second, qualified R&D personnel in this dataset have completed a tertiary-level education successfully or were employed in a Science & Technology occupation that requires a high qualification where the innovation potential was high.
The OECD Science, Technology and Industry Scoreboard provided the percentages for human resources in science and technology according to the conditions mentioned as a percentage of the total employment in the country for each year given. These percentages were later on converted into decimals.
Support provided by the government
This variable measures the support by the governments of the AECs, concerning the R&D investments of MNEs. In a similar way as with the quality of human resources in scientists and engineers, this variable is also a given for the timespan used in this thesis and can be found back for each country in the Knoema databank under ‘Government budget appropriations or outlays for RD’. The Knoema databank is an open platform for users with interests in statistics and data analysis of several topics, amongst them R&D.
Government budget appropriations or outlays for RD are gathered by using the NABS 2007 classification. The outcome is conveyed in PPP dollars- current prices, which refers to the current purchasing power parity in Dollars. Moreover, the objective of governmental support is specifically expressed as ‘General advancement of knowledge: R&D financed from General University Funds (GUF)’ as well as ‘General advancement of knowledge: R&D financed from other sources than GUF’.
4.4.3. Control variables
Capital Investment
Capital investment as control variable refers to the amount of capital that is invested in the R&D investment abroad. This amount is also provided by the dataset of 2093 Green Field investments made by Chinese MNEs (Financial Times, 2011). It is of importance to take this variable into account, as it is possible that the capital invested influences the location choice of the R&D investment. It might me that the higher the capital invested, the more a Chinese MNE searches for the LSAs mentioned in the literature because of the risks that are at higher stake.
Gross Domestic Product (GDP)
The GDP of a country is measured by the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products (Knoema, 2016). Statistics on this control variable have been found in the Knoema databank, an open platform for users with interests in statistics and data analysis of several topics. The GDP was measured in current US dollars. Moreover, this control variable is of importance as it serves as an accurate indication of the size of a country’s economy. Previous literature has stated that the attractiveness of the advanced European country’s economy is likely to play a role in the location determination of MNEs when investing abroad (Di Minin, Zhang, Gammeltoft, 2012).
Time Dummies
The final control variable refers to the year the Chinese R&D investment is made. This could range between 2003 and 2011. Each year will get its own dummy variable. The period is included as a dummy variable in order to take into account possible changes in years due to events that had influence on the variable.
5. RESULTS 5.1. Descriptive statistics and correlations
Table 1: Descriptive statistics, means, standard deviations and correlations presents the descriptive statistics of the dependent, independent, and control variables. In total, this thesis had a sample of 501 Chinese investments made in the AECs. From the 501 investments made by Chinese MNEs, only 8,5% were made due to R&D reasons. This small amount of R&D might create problems for the generalizability of the results. Furthermore, the average for the technological infrastructure for the AECs ranged between 4,5 and 6,8 with an average of 5,8, which comes close to 7, which is the highest score possible. Moreover, the mean for the knowledge infrastructure contained 5,3 with a range between 5.0 and 6.4. Additionally, the scores for the quality of human resources in Science and Technology ranged between 0.45 and 0.54, with an average of 0,5. Finally, governmental support ranged between 1007,34 and 15827.73.
Also in table 1: Descriptive statistics, means, standard deviations and correlations, the correlations between the dependent, independent and control variables can be found. Only one issue of multicollinearity seems to apply: the GDP of the AEC and support provided by the government showed a significant correlation coefficient of 0.853 (P= .000). This seems odd as both the GDP of the host country and support provided by the government included very different components in their constructs. The GDP of a country has been measured by the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products, while the governmental support is measured by summing up the R&D financed from General University Funds and R&D financed from other sources than General University Funds. Additionally, both variables were gathered from two different data sources. However, as a diagnostic for multicollineairty, VIF scores have been calculated. Each independent and control variable has been compared individually to the other variables. A VIF outcome above the threshold of 3 would bring concerns. Nevertheless, none of the results crossed the criterion of 3,0. The high correlation coefficient between the GDP of the host country and support provided by the government only showed VIF scores of 1.663 and 2.653. Therefore, multicollineairty between the variables in this thesis is rejected.
5.2. Model specification
To measure the probability that it are the four LSAs of the AECs mentioned above that trigger Chinese R&D investments to the AECs, this thesis utilizes the Binary Logistic Regression model. This model fits the examination of this study as it compromises a regression analysis where the dependent variable takes only two values, in this case a dummy variable. It allows the research to assess how well the independent variables predict or explain the categorical dependent variable. Furthermore, it gives an indication of the adequacy of the model through assessing how well the model fits. In other words, this model is able to confirm or reject that the independent variable, a LSA, is related to one of the specific categories of the dependent variable, in this case an R&D investment made opposed to a non-R&D investment made. It is important to note that this type of regression is highly sensitive to high correlations between the independent variables. Moreover, outliers also have an influence on the results of this model. It is therefore of importance to first examine multicollineraity and outliers. As mentioned above, multicolineairty is out of the question in this research. The model is specified as follows:
𝑳𝒏 𝑷
A Binary logistic regression analysis has been conducted to predict if the four LSA’s of the AECs would lead to the location choice of R&D investments made by Chinese MNEs to the AECs. First, the control variables have been included into the regression analysis to the dependent variable, to observe their effect on the dependent variable. Then, each independent variable has been included seperetaly with all control variables to examine their individual effect on the dependent variable. Finally, all independent variables were included into the regression analysis together with all control variables in order to analyse their cohesive influence on the dependent variable. Subsequently, six models have been carried out, which can be found back in table 2: Results of Binary regression analysis for industry activity.
Model 1a describes the model where only control variables are included. Model 2a has been included to test hypothesis 1, model 3a has been included to test hypothesis 2, model 4a examines hypothesis 3 and model 5a tests hypothesis 4. The last model, model 6a, examines the contribution of all four independent variables together on the dependent variable. The Beta standardized coefficients designate if the control and independent variable included in the regression analysis contribute to the prediction of the dependent variable. If the significance level of this Beta standardized coefficient is below 0.05, the variable in question makes a significant contribution to the prediction of the dependent variable. Finally, the adjusted R2
values of .051, .059, .063, .062, .052 and .104 indicate that the variables explain approximatelly 5,1% of the dependent variable in model 1a, 5.9% in model 2a, 6.3% in model 3a. 6.2% in model 4a, 5.2% in model 5a and 10.4% of the dependent variable in model 6a.
When comparing each independent variable separately to the dependent variable, two out of the four LSAs revealed significant results. The first hypothesis in this study states that the technology infrastructure of the AECs influences the location choice of Chinese R&D investments to these countries. However, the results from the regression analysis showed a non-significant coefficient of the technological infrastructure (b=.135, p=.560). Therefore, hypothesis 1 is rejected. Furthermore, hypothesis 2 argues that the knowledge infrastructure of the AECs influences the location choice of Chinese R&D investments to these countries. Although results of the binary regression analysis showed significant results, the beta coefficient was negative (b=-1.699, p=.023), indicating a negative influence of the knowledge infrastructure of the AECs on the location choice for R&D investments. Therefore, hypothesis 2 is rejected. Hypothesis 3 argues that the quality of human resources in science and technology of the advanced European countries influences the location choice of Chinese R&D investments to these countries. The coefficient of the quality of human resources in science and technology is
significant (b=24.996, p=0.048) and positively related to the industry activity. Hypothesis 3 is therefore supported. Finally, hypothesis 4 states that the R&D support by the government of the advanced European countries influences the location choice of Chinese R&D investments to these countries. However, the coefficient of this variable is non-sginifianct (b=.000, p=.295). Hence, hypothesis 4 can be rejected.
The control variables did not show influence on the dependent variable. The result for the relation between the GDP of the AECs and the industry activity was significant, though negative (b=-.001, p=.000), indicating that the GDP of the home country would negatively influence the location choice of Chinese MNEs when investing in R&D abroad. Moreover, the amount of the capital invested did not reveal any relationship and was not siginificant (b=.000, p=.810).
Finally, although most of the independent and control variables individually do not seem to influence the location choice of R&D investments made by Chinese MNEs, when including all variables together in the statistical model, model 6a, all independent variables contribute significantly to the dependent variables. In this model, the technology infrastructure and the quality of human resources in science and technology indeed explain why Chinese MNEs would invest in R&D activities in AECs (chi square = 42,468, p < .001 with df = 14). Nagelkerke’s R2 of .187 indicated a weakly relationship between independent variables and grouping. Prediction success overall was 91,3%. Finally, the Exp(B) values indicate that when technological infrastructure is raised by one unit (score) the odds ratio is 3.655 times as large and therefore it is 3.655 more likely that Chinese MNEs invest in AECs. Similarly, the Exp(B) scores showed scores of .112 for knowledge infrastructure, 2.568 for the quality of human resources in science and technique and 1.00 for support provided by the government.
