How institutions shape the international strategy of world’s biggest manufacturers.

(1)

How institutions shape the international strategy of world’s

biggest manufacturers

Master Thesis

MSc. Business Studies – International Management

Supervisor: Dr. J.P. Lindeque Second reader: Dr. A. Muller

Student: N.W. Baak

Student ID: 5980925

Date: 11 September 2014

(2)

How institutions shape the international strategy of world’s biggest manufacturers

N.W. Baak 1

Abstract

In light of a broader debate about the regional / global dimensions of MNEs, this study focuses on the specific industry dynamics of the power- generation / energy sector in order to explain to which degree International Manufacturers’ (IM’s) business strategies are shaped by formal institutions. In order to collect the data, a longitudinal multiple case study research design was conducted. At first, quantitative data was collected to describe the regional profiles of six MNEs from the triad-regions and their development over time. Second, Financial Times and Nikkei Asian Review articles were used for thematic coding in order to explain the antecedents of the quantitative findings. Within- and cross-case analyses of the data confirm home-region and bi-regional orientations of these IMs. These analyzes also show that IMs IB strategies are shaped by institutions within the regions and also on a regional level. The nuclear Fukushima disaster in 2011 in Japan was identified as an event that reversed the institutional context of IMs in Asia. Furthermore, evidence shows that IMs use inter- regional alliances to overcome inter-regional LoF and intra-regional alliances to strength international competitiveness.

Keywords

Institutions, international manufacturers, regionalization, regional dimensions, home country effects, alliances, energy, FSAs, inter-regional LoF, intra-regional LoF.

(3)

N.W. Baak 2

Acknowledgements

Writing up this master thesis had never took place without the invaluable support and enthusiasm of my thesis supervisor dr. Johan Lindeque. He motivated me to continue writing and finish the thesis. I will also thank him for his patience which was challenged during the process. I also want to thank my fellow students, which were also part of the research group, for the pleasant cooperation especially during the first part of the writing process in which several meetings took place. I also thank the second supervisor in advance for assessing this thesis. Last but not least I thank my wife Tessa for here support during the writing process and for all the days, weekends and evenings she was patient while I was working on this project.

(4)

N.W. Baak 3

Table of contents

1. Introduction ... 5

2. Literature review ... 7

2.1 Globalization and regionalization debate ... 7

2.2 MNE’s as the key drivers of globalization ... 8

2.3 Liability of (regional) foreignness ... 9

2.4 An institution- based view of business strategy ... 11

2.5 A resource- based perspective of business strategy... 12

2.6 Importance of the location ... 13

2.7 International manufacturers ... 15

3 Longitudinal multiple-case study research design... 19

3.1 Research philosophy ... 19

3.2 Research design ... 19

3.3 Theoretical sampling strategy... 20

3.4 Data collection ... 20

3.5 Analysing methods ... 22

4. Within-case analysis ... 26

5. Cross-case analysis ... 52

6. Discussion and conclusion ... 57

7. References ... 59

7.1 Literature ... 59

(5)

N.W. Baak 4

Tables

Table 2.1: Industry specificity in production in Japan. 18

Table 3.1: Data sources and articles used for coding 22

Table 3.2: Code categories and codes used for coding in Nvivo 22

Table 3.3: Background information MNE’s 24

Table 4.1: Regional profile General Electric 28

Table 4.2: Regional profile Alstom 33

Table 4.3: Regional profile Siemens 37

Table 4.4a: Regional profile Hitachi 41

Table 4.4b: Regional profile Toshiba 45

Table 4.4c: Regional profile MHI 48

Table 5.1: Cross- case analysis 53

Table 5.2: Regional trends in the distribution of sales 55

Table 5.3: Regional orientations of international manufacturers 57

Figures

(6)

N.W. Baak 5

1. Introduction

In the international business (IB) literature, a debate has sparked about the regional dimensions of Multinational Enterprises (MNEs) (Ghemawat, 2003; Rugman and Verbeke, 2004, 2007) on the question whether MNE’s are really global or not. A milestone in this debate were the influential papers of Rugman (2001, 2004) ‘the myth of global strategy’, indicating that global strategy came to its end. Researchers elaborated on this idea (Asmussen 2008, Filippaios and Rama, 2008, Hall 2001, Hejazi 2007) and advocated for more regional oriented strategies (Ghemawat 2005, Khan 2010). Recent developments highlight industry dynamics in the power-related industry. In June 2014, the US MNE General Electric (GE) tried to buy most of the energy-related activities of the Europese MNE Alstom. However, the French government intervened and demanded a 20% stake in these activities before they could be sold to GE. So it was. This event reflects the role of formal institutions (North, 1990) in the internationalization process of MNEs. In order to investigate in more depth the role of institutions on the international profile of MNEs, this research tries to identify the institutional mechanisms that shape and constrain international business activities of MNEs.

An interesting dimension of this research is shaped by the fact that this research is part of a broader investigation of the regional versus global dimension of world’s biggest MNEs. As this thesis thus is part of a team research, it was possible to collect data of the 500 biggest MNE’s and to measure the degree of internationalization in the same way as Rugman and Verbeke (2004), the research team collected data of the year 2012. Interestingly, these findings are somewhat different than the findings of Rugman and Verbeke in 2004. Overall, in 2012, MNE’s were less home- region oriented (36,8%) and more bi-regional (14,8%), host- region (2,6%) and global (5,6%) than in 2004. The big question is now, how can these differences be explained? And what is the role of specific firms or sectors? This research cannot answer all these questions but will instead focus on a specific industry in order to contribute to the overall knowledge or the internationalization patterns of MNEs.

Although a lot of research has been done regarding the internationalization strategies of MNEs, the stream of research is increasingly focusing on specific sectors to uncover the sector-specific dynamics. Although different sectors have been investigated, the internationalization patterns of international manufacturing (IM) MNEs remain largely unexplored, leaving a gap in the literature. More specifically, the subject of investigation are the formal institutions that shape the international strategy of world’s biggest manufacturers, specifically General Electric (GE), Siemens, Alstom, Hitachi, Toshiba and Mitsubishi Heavy Industries (MHI). To explore these influences, specific attention will be paid to energy-related activities of these firms, as they are supposed to explain most of the impact that formal institutions have on the manufacturing MNEs. Furthermore, alliances are expected to be an important strategy to acquire host country and host region capabilities and thus will

(7)

N.W. Baak 6

play an important role in the geographic diversification strategy of manufacturing MNEs. In line with the institution-based view, it is expected that both formal and informal institutions can facilitate and restrict the internationalization of MNE’s, both in the home- and in the host region. In order to explore the internationalization strategy of MNE Engineering firms, the following research question is formulated:

To what degree do institutions shape the geographic diversification of world’s biggest manufacturers?

Answering this research question adds value both academically and practically. At first, knowledge about the reasoning behind the internationalization strategy will add to the academic debate regarding internationalization/ regionalization patterns and orientations of MNE’s. Second, insight in effect of institutions on the international transferability of MNE FSAs will add to the academic knowledge regarding regionalization patterns.

In order to answer the research question, working propositions are developed in the literature review, followed by a visualization of these propositions in chapter three. The research design adopted is a multiple-case study approach, which is seen as a good way to do theory building (Yin 1981, Eisenhardt 1989). Overall, the research can be divided into two parts. The first part consists of determining the degree of internationalization. In order to determine the degree of internationalization, longitudinal quantitative data will be collected from annual accounts of the aforementioned firms over a period of 5 years. In order to understand the strategy behind this degree of internationalization data, an archival study will be conducted, which consists of gathering articles from the Financial Times of the aforementioned MNE’s, over the same period of 5 years. Regarding the analyses, in line with Eisenhardt (1989), within-case analysis will be conducted, followed by cross-case analysis. The report ends up with a conclusion, limitations and directions for further research.

(8)

N.W. Baak 7

2. Literature review

The literature review starts with an overview of the current global versus regional strategy debate concerning the 500 biggest companies in the world. Afterward, paragraph 2.2 pays attention to MNEs as the key drivers of globalization and paragraph 2.3 explains the concept of the liability of foreigness (LoF), as the costs resulting from the LoF are inevitable in doing business abroad and are thus an important theme in the IB (international business) literature (Zaheer, 1995). Using the concept of LoF, paragraph 2.4 and 2.5 discusses respectively the institution-based view (IBV) and the resource-based view (RBV) as possible explanations of the LoF MNEs experience. Furthermore, paragraph 2.6 elaborates on key themes in the regionalization literature and paragraph 2.7 pays attention to industry specificity as it discusses the globalization and regionalization dimensions of the home-region oriented industry of IM’s. In this last paragraph the characteristics of these MNEs are linked to the general regionalization literature and working propositions and a conceptual research model are developed.

2.1 Globalization and regionalization debate

Globalization has been defined as ”the international integration of markets in goods, services and capital” (Garrett, 2000, p. 942). The effects of globalization and the role of the government are discussed in the literature. Countries can raise or lower tariffs and non-tariffs barriers, influencing the attractiveness of a country in terms of international trade. This ‘openess’ should speed convergence in the distribution of wealth on a cross-country level (Brune and Garrett, 2005, p. 410).

Ghemawat (2003) concludes that, based on economic evidence, “the international integration of markets indicates that we fall in between these extremes, into a state of incomplete cross-border integration that I refer to as semi globalization” (Ghemawat, 2003, p. 138). Gemawat (2003) distinguishes between two sides of integration; product market integration and factor market integration. From the perspective of product market integration, trade flows, foreign direct investment (FDI) and price integration (law of one price), a cross-country analysis shows “an overall inference that product market integration has increased significantly in recent decades” (Ghemawat, 2003, p. 143). In terms of factor market integration, capital market integration shows a similar path, while cross-border integration of labour and knowledge seems far from integrated (Ghemawat, 2003, p. 145). Rugman and Hodgetts (2001) are more extreme and they proclaimed the end of global strategy (Rugman, 2001). Their reasoning is that there are common ‘global’ misunderstandings about globalization. In contrast to the actual view on globalization, which is simply ‘providing the same output everywhere’, undermined with a couple of cases, they explains that in most cases, the business strategy of the MNEs is more triad / regionally focused than global (Rugmand and Hodgetts, 2001, p. 333). MNEs have to adapt their products / services to local markets. By example, there’s no

(9)

N.W. Baak 8

worldwide, global car, cars are built in a region (Japan, the US) and sold in that region (Rugman and Hodgetts, 2001, p. 334).

2.2 MNE’s as the key drivers of globalization

MNE´s are the key drivers of globalization, as they account for most of the world trade and investment (Rugman and Verbeke, 2004). The literature shows evidence of a very regional world trade, indicating three strong regional blocks: The United Stated, the EU and Japan (Rugman and Hodgetts, 2001, p. 334). According to Ohmae (1985 in Rugman and Verbeke, 2004), these regions share a number of commonalities such as a low macro-economic growth, a similar technological infrastructure, the presence of both large, both capital and knowledge intensive firms in most industries, homogenization of demand and protection. In terms of corporate strategy, theoretically, the distribution of sales of the MNEs in these regions should be in line with respectively country / regional GDP (Rugman and Verbeke, 2004). However, it is not. In practice MNEs are not able to penetrate the two host markets, which is necessary to recover innovation costs (Ohmae, 1985 in Rugman and Verbeke, 2004). Rugman and Verbeke (2004) identify five causes of this phenomenon. The first one is that the MNEs products are not really equally accessible and / or attractive to consumers all over the world, in spite of many MNEs attempting to adapt their products to local demand. The second lack of global market success could be the reflection of the limits to the non-location-bound nature of the MNEs knowledge base (or FSAs). This means that the international transferability of FSAs such as brand names and technological knowledge is limited (Rugman and Verbeke, 1992). The third lack could be the inability to access and deploy location-bound FSAs in the host country. The fourth point is that, depending on the market situation, different strategies are required. This should be reflected in the deployment of specific combinations of non-location-bound (NLB FSAs) and location-bound FSAs (LB FSAs) in each region (Rugman and Verbeke, 2004, p. 5-6).

The main attribute of the paper of Rugman and Verbeke (2004) is that they test whether MNEs are regional or global. In order to do so, they investigated the 500 biggest MNEs from the Global Fortune 500. Because the fact that from 120 MNEs there’s no or not enough data available, they use a sample of 380 MNEs (Rugman and Verbeke, 2004, p. 6). In his book ‘The Regional Multinational’ (2005), Rugman explains a classification of these MNEs, bases on their sales. The main results are presented in table 2.1.

(10)

N.W. Baak 9

Strategic Orientation Decision Criteria

Home oriented: Of the top 500 firms, 320 have at least 50% of their sales in their own region of the triad. So 80,3% of the firms is home region oriented. Bi-regional: Only 25 of the 500 firms are bi-regional, with over 20% of their sales in at

least two parts of the triad plus less than 50% of sales in their home region. 42% Of the MNEs is bi-regional.

Host region oriented: 11 host- region oriented firms with over 50% of sales in a triad region other than their own. 30.9% of the MNEs is host region oriented.

Global: Only none of the top 500 MNEs are global, defined as having sales of 20% or more in each of the three regions of the triad, but less than 50% in any one region. 38.3% of the MNEs is global.

Table 2.1: Regional orientations of MNEs. Source: Rugman, 2005, p.4

Thus they found that most of the MNEs are mainly focused on their own home region, without having a substantive business in the other two regions (Rugman and Verbeke, 2004, Collinson and Rugman, 2007, 2008). Although there has been some critique on the classification of Rugman and Verbeke (2004) used (see Osegowitsch et al., 2008), the classification is still robust (Rugman and Verbeke 2008c). The main point is that there are very few really ‘global’ MNEs and that there’s strong evidence for regional / triad MNE activity. Only nine MNEs are really ‘global’ (Rugman, 2005, p. 5). The explanation for this pattern of internationalization within regions is argued to be the relative degree of intra- and inter-regional LoF (Rugman and Verbeke, 2004).

2.3 Liability of (regional) foreignness

On a regional level, LoF can exist within (intra-regional LoF) or between regions (inter-regional LoF) (Rugman and Verbeke, 2004). In terms of the internationalization of MNEs; if the LoF MNEs face in host regions is relatively higher than the LoF MNEs face within their home-region, inter-regional LoF is shaped, indicating that it is easier for MNEs to internationalize within the home-region instead of internationalize to other regions. Doing business abroad comes along with extra costs that result in a competitive disadvantage (Zaheer, 1995). There are several sources of LoF. At first, these costs can arise from spatial distance, such as the costs of travel, transportation and coordination over distance and across time zones (Zaheer, 1995). The second source of these costs are firm-specific, ‘based on a particular company’s unfamiliarity with and lack of roots in a local environment’(Zaheer, 1995). The third source of LoF are the costs resulting from the host country environment (such as lack of legitimacy) and the fourth source are costs from the home country (such as the restrictions on high-

(11)

N.W. Baak 10

technology sales) (Zaheer, 1995). In order to overcome the LoF, multinational´s subunits can import capabilities embodied in the organizational practices of their parent enterprises (Zaheer, 1995). A second option to compete with local firms in the host country, more on an institutional level, is to imitate practices of local firms, called isomorphism (Zaheer, 1995).

Because it could be argued that, ´regional strategies of MNEs are embedded in the broader competitive, organizational, and institutional contexts at the regional level’ (Rugman, 2005, p. 77), the concept of inter-regional LoF, describe the costs of inter- regional ‘distance’. These regional strategies are reflected in the preference of European firms to grow mainly in Europe, of the North American firms mainly focus on the NAFTA and in the focus of the Asian MNEs on Asia before expanding to other regions (Rugman and Verbeke, 2007, p. 201). Rugman and Verbeke (2007) argue that the main reason of this phenomenon is that ‘the liability of intra-regional expansion appears to be much lower than the liability of interregional expansion’ (Rugman and Verbeke, 2007, p. 201). This is because of two reasons. The first one is the concept of ‘distance’ as elaborated on by Zaheer (1995). The second one is that many FSAs are, although non-location-bound, appear to be home region bound rather than having global deployment and exploitation potential (Rugman and Verbeke, 2007, p. 201). The importance of the difference between the liability of country foreignness and the liability of regional foreignness is indicated in a recent article written by Qian, Li and Rugman (2013), who found different effects on firm performance. The next sections elaborate on the LoF from both a resource- based and an institutional-based perspective.

LoF in Upstream and downstream activities or asset specificity

The degree of LoF MNEs face in terms of intra of inter-regional activities also depends on the kind of activities in terms of upstream (production) and downstream (sales) and between assets (‘asset specificity’). Regarding asset specificity, in this thesis, energy-related business units (thus activities) of the manufacturing MNEs are investigated. Li and Li (2007 in Kolk, 2010, p. 57) point to the fact that ‘firms’ sales are less internationalized than operations’. The reason is that ‘production and sourcing involve two-sided commitments (also by other corporate partners in the value chain such as suppliers and logistics providers), which does not apply to sales, where there are no guarantees from the side of the customers (see e.g., Rugman and Verbeke 2008c). In that sense, risks are higher in the case of downstream activities in view of the one-sided commitments. If upstream partners are located in the host region, then this helps reduce the inter-regional LoF (Kolk, 2010, p. 57).

(12)

N.W. Baak 11

2.4 An institution-based view of business strategy

This section is about the ‘fit’ of MNEs international activities to host countries / regions. Furthermore, this section discusses how the LoF can be shaped by institutional constrains in host environments. As no firm can be immune for institutional frameworks in which it is embedded, Peng (2002) concludes that institutions matter for (international) business strategy. North (1992, p. 97) defines institutions as “the humanly devised constraints that structure political, economic and social interaction”. In addition to that, Scott (1995) defines institutions as “cognitive, normative and regulative structures and activities that provide stability and meaning to social behaviour”. In the literature, both perspectives are labelled as the ‘new institutionalism’. Although Peng (2002) chooses for an integrative approach using both the economic perspective of North (1990) and the sociological perspective of Scott (1995), this research will focus more on the economic approach as it suits better in terms of explaining international differences between countries and regions. In Peng (2002, p. 252), Davis and North (1971) define an institutional framework as ‘the set of fundamental political, social and legal ground rules that establishes the basis for production, exchange and distribution’. As a result, institutions can reduce uncertainty. Reduced uncertainty within the region will decrease intra-regional LoF and if the uncertainty in host-regions is relatively higher than in the home-region, this will increase inter-regional LoF. Institutions can be both formal and informal. Formal constraints are political rules, judicial decisions and economic contracts (Peng, 2002, p. 252). Examples of informal constraints are sanctions, taboos, customs, traditions and codes of conduct, and formal rules as constitutions, laws and property rights (North, 1991, p. 97). In fact, any strategic choice that firms make is inherently affected by the formal and informal constraints of a given institutional framework (North, 1990 in Peng, 2002, p. 252).

Institutions on a regional level

As indicated by Rugman and Verbeke (2004, p. 5), formal institutional frameworks in the regionalization literature exist on a global and on a regional level. In line with Ohmae’s (1985) theory of the triad regions, on a global level, the World Trade Organization (WTO) plays a role in integrating the triad regions from an economic perspective. As the WTO is the only example for Rugman and Verbeke (2004) of a global economic institution arrangement, there seems to be a very low degree of global (or inter-regional) economic integration. The phenomenon of institutions facilitating intra- regional economic integration seems to be more common. On a regional level, institutional arrangements are visible in the US as well in Europe and in Asia. The Free Trade agreements in the Americas, the NAFTA in 1994 and its expansion to the Free Trade Area ‘ represent the agglomeration of attractive, proximate foreign markets (from a geographical, cultural, economic and administrative perspective) into a ‘broad triad region’ (Rugman and Verbeke, 2004, p. 5). This results in even deeper

(13)

N.W. Baak 12

intra-regional market penetration (see also Hejazi, 2007, p. 24). The free trade agreement of the 10 members of the Association of South East Asian Nations (ASEA) indicates intra-regional arrangements within the Asian region. In Europe, the EU is the most powerful institutional arrangement of economic integration which is still expanding (Rugman and Verbeke, 2004).

More recent research of Dikova, Rao Sahib and van Witteloostuijn (2010, p. 223) shows that formal and informal institutions indeed effect the internationalization process of firms, as they moderate the likehood that cross-border acquisitions will be completed. Organization learning seems to be an interesting phenomenon as it moderates the effect of institutional distance; “Past experience with completed cross-border acquisition deals increases the likehood of a subsequent deal completion in institutionally closer environments, but shortens the deal duration in institutionally distant environments” (Dikova, Rao Sahib and van Witteloostuijn, 2010, p. 223).

2.5 A resource-based perspective of business strategy

This section discusses the internationalization of MNEs from the resource-based perspective, thus focuses on the internal resources and capabilities of these MNEs. From an IB perspective, the RBV ‘has helped to specify the nature of resources required to overcome the LoF and provided a bridge to investigate the resources that provide the foundation for product and international diversification’ (Barney, Wright and Ketchen, 2001, p. 629).

One of the founders of the RBV was Barney (1991), who argued that sustained competitive advantage derives from the resources and capabilities a firm controls that are valuable, rare, imperfectly imitable, and not substitutable. These assets can be both tangible and intangible. In the IB literature MNE resources and capabilities are collectively referred to as firm specific advantages (FSAs) (Rugman and Verbeke, 1992). Rugman and Verbeke (2004) use the concept of FSAs to explain the success and failure of MNEs in doing business in their home and host regions of the triad. They relate the success of a MNE of doing business in another region to the degree to which a firm is able to transfer NLB FSAs to the host region and to meet local requirements which specific LB FSAs (Rugman and Verbeke, 1992). As it is clear that most of the MNE’s are more regional than global oriented, Rugman and Verbeke (2004) introduce a new category of FSAs, that is; region-bound FSAs, typically related to the home region of a MNE. Due to the lack of success in the host market, most of the MNE’s seem to be not able to deploy specific region-bound FSAs in host regions, which can also be explained “as a reflection of the limited customer value attributed home triad FSAs” (Rugman and Verbeke, 2004, p. 12). Rugman and Verbeke (2004) distinguish between FSAs needed in upstream (global sourcing etc.) and downstream (selling) activities. A lack of knowledge bundles at the downstream end differs from the knowledge bundles which are required in the home-region, which can explain the lack of success in the host region of most MNEs (Rugman and Verbeke, 2004).

(14)

N.W. Baak 13

Rugman also indicates that, although there’s little evidence, it is possible that the ‘back end’ production of the value chain is more globalized than the ‘front end’ of sales (Rugman, 2005, p. 7), due to the fact that customer-end FSAs are hard to transfer (Rugman, 2005, p. 73). In terms of international strategy, recombination is about the ability of a MNE to deploy capabilities in foreign markets. As capabilities are necessary to sustain competitive advantage on a global scale, as explained in previous sections, there can be distinguished between capability exploitation and capability building, called dynamic capabilities (Luo, 2002, p. 48).

As capability exploitation refers to extent to which a firm exploits firm specific, rent- generating resources, capability building concerns the extent to which a firm commits to building new capabilities in new situations. This author found that capability exploitation and capability building are influenced by environmental hazards, market orientation and entry mode (Luo, 2002, p. 58). In between the two extremes of entry modes, that is entering an export market and setup an wholly owned subsidiary, a firm will use different strategies to deploy their capabilities. By example, a wholly owned entry mode is positively related to capability exploitation and a joint venture entry mode is positively related to capability building. In the case of targeting a host market, a firm will deploy higher levels of capability exploitation and building compared to entering an export market. Depending of the choice of entry mode, environmental hazard plays a less vs more important role (Luo, 2002, p. 58).

As the RBV/ RBT explains competitive advantage, transferring resources and capabilities to host countries can overcome the LoF, as described in previous sections. Another option for MNEs to get access to resources and capabilities in host countries is to form alliances with firms in host countries. From an RBV/ RBT point of view, the capabilities to learn from partners ‘may be a tacit resource underlying a firm’s competitive advantage’ (Hamel, 1991 in Peng, 2001, p. 812). Thus forming alliances can be interpreted as an option of an MNE to acquire intangible assets in order to overcome the LoF and to gain competitive advantage.

2.6 Importance of the location

As the home country can potentially explain the home-regional orientation of MNEs in several cases (Rugman and Verbeke, 2004), this section discusses literature regarding themes that matter for the home regional orientation of US, EU and Asian MNEs. As there’s evidence (Rugman and Verbeke, 2004) that most of the MNEs in the triad are home-region based, the antecedents of location specificity will be discussed in the first section. Second, theories which explain differences in internationalization patterns between industries will be discussed.

(15)

N.W. Baak 14

In the literature there is a lot of evidence that the country or the region a MNE is situated in, matters in terms of internationalization patterns of these MNEs. Past decades, all three triad-regions are investigated in order to gain knowledge about what factors shape the IB strategy of MNEs. One of the first authors highlighting the role of location specificity in the IB literature was Ghemawat (2003).

On a country level, Rugman and Verbeke (2004) highlighted the importance of country specific advantages (CSAs). In their theory, country factors can lead to competitive advantage. CSAs can be ‘environmental factors as ‘political, cultural, economic and financial factors’ (Collinson and Rugman, 2008, p. 219). In terms of these authors, the CSAs can be based on home-country natural resource endownments (minerals, energy, forests) or on the labour force, or on distinctive cultural factors. The CSA can also include demand condition, the political, cultural and regulatory systems and infrastructure’. Government regulation, such as tariff and non- tariff barriers, are also aspects of CSAs.

Later research on cross-border integration not only focused on integration on a country-level, but, in line of the regionalization theory of Rugman and Verbeke (2004), tried to explain differences between regions in the triad (Rugman and Oh, 2012). These authors also found that both regional and industry effects explain most of the geographic expansion of MNEs (Rugman and Oh, 2012, p. 1).

Regarding US MNEs, Hejazi (2007, p. 23) found that the activities of these firms are concentrated in the US and that these activities are driven by the national dimension. Asmussen (2008) found similar evidence. Oh (2009) found similar results for European MNEs, as he concludes that ‘no evidence exists that European MNEs focus on global markets or were becoming geographically diversified..’(Oh, 2009, p. 336). Regarding the Asian region by example, research data of Rugman and Verbeke (2004) shows that 57 out of the 64 Japanese countries in the triad region have over 80% of their assets and sales in their home country/ home region. Thus they are intra-regional MNEs in terms of production as well as of sales (Collinson and Rugman, 2008, p. 227). Furthermore, regional strategies of Japanese MNEs are investigated in order to get more insight into the specific characteristics of the Asia-Pacific region (Collinson and Rugman, 2007). The lattest authors found that Japanese firms ‘(1) are innately tied to the regional and country specific factors, the political, economic, social context and business infrastructures of Japan and (2) have evolved FSAs specifically to compete in this environment’(Collinson and Rugman, 2008, p. 227). Specific location-bound FSAs are HR practices, organizational characteristics, R&D processes and network- related FSAs such as in- depth local connections with suppliers and customers (Collinson and Rugman, 2008, pp.225 - 226). Lifetime-employment and culture aspects as consensus-based decision making, self-management, which emphasise commitment and obligation as part of a general view of employment as a social rather than a financial contract (Clegg&Kono, 1998 and others in Collinson and Rugman, 2008, p.

(16)

N.W. Baak 15

225). Results from research measuring the performance of international firms are mixed. Qian (1996, p.262) found that it depends on the industry the firm is operating in and Delios and Beamish (2005) found that in the Asian region, well-internationalized firms are more successful than home- or host oriented ones.

2.7 International manufacturers

As it comes to ‘industry specificity’, according to Kolk (2010, p. 56) there can be distinguished between broader categories such as manufacturing versus services but also more specific, detailed (sub)sectors such as automotive, retail, cosmetics, accounting, financial services. As the sector of IMs is not very scrutinized in terms of regional vs global orientations, as indicated in the introduction chapter, this research aims to explore the regional vs global characteristics of the MNEs in this sector in more depth, using the broad category of manufacturing firms. Evidence suggest that overall, the biggest manufacturers are highly home-region oriented (Rugman and Verbeke, 2004).

From a resource-based perspective and as concluded in previous sections, forming alliances can be interpreted as an option of an MNE to acquire intangible assets in order to overcome the LoF and to gain competitive advantage. The strong home-region orientation of (international) manufacturers reflects strong home- region FSAs of these MNEs (Rugman and Verbeke, 2004). As evidence suggests (Collinson and Rugman, 2008, pp.225 - 226) and as described in previous sections, these FSAs are mostly very location-location bound, which means that it is (almost) impossible to transfer these FSAs to host regions. In this context, forming alliances could be an attractive strategy for IMs to enter host- regions. This leads to the following proposition:

Working proposition 1: International manufacturing MNEs use strategic alliances as an internationalization strategy to overcome the LoF and to enter host regions

All MNEs are exposed to institutional frameworks (Peng, 2002), these are also important for IM MNEs. The effects of country specific aspects (CSAs) are discussed in the literature to explain home – region effects of MNEs (Rugman and Verbeke, 2004 and Collinson and Rugman, 2008). As CSAs include demand condition, the political, cultural and regulatory systems and infrastructure, the formal institutions (Peng, 2002, p. 252) of political conditions could be important to the manufacturing industry, in light of developments in supply and demand of energy activities of these MNEs in the

(17)

N.W. Baak 16

context of changes in supply and demand of energy (by example the shale-gas revolution). Therefore, the following proposition is developed:

Working proposition 2: Regulatory systems (formal institutions) in the home region of international manufacturers, to a high degree, explain the regional orientation of these MNEs

As MNEs face different degrees of LoF in terms of upstream and downstream activities, as explained in previous sections, this seems to have important consequences for the regional versus global strategy of these MNEs. Because there is a two-side commitment as it comes to upstream (production) activities of manufacturing MNEs and a one-side commitment as it comes to downstream activities, sales activities seem to be more vulnerable in terms of market risks. Another reason for the phenomenon of limited success of the ‘front-end’ of the value chain / activities in host countries is the limited transferability of the ‘front-end’ FSAs (Rugman, 2005), leading to the following propositions:

Working proposition 3a: Upstream activities of international manufacturing MNEs face a lower LoF because production activities’ face less market risks that downstream activities

Working proposition 3b: Downstream activities of international manufacturing face a higher LoF than upstream activities.

The theory of regional institutions of the triad region, such as the NAFTA, the European Union and ASEA, argues that these ‘regional institutions’ increase geographic diversification within a region, resulting in a lower intra-regional LoF and decrease inter-regional geographic diversification, resulting in a higher inter-regional LoF (Rugman and Verbeke 2004, 2005, 2007, Zaheer, 1995). The vulnerability of MNEs to host region institutional constraints is even more important for a knowledge- intensive industry like manufacturing, as they are expected to be more vulnerable to host- region institutional frameworks. Thus government regulation, such as tariff and non-tariff barriers, could also play an important role (Collinson and Rugman, 2008). This leads to the following propositions:

(18)

N.W. Baak 17

Working proposition 4b: Stronger intra- regional formal institutions decreases intra- regional LoF

Another aspect of industry specificity points to the fact that the significance of transportation costs differ per industry, thus leading to industries that are more vulnerable to this aspect of LoF than other industries (Collinson and Rugman, 2008, p. 218). This is in line with Zaheer (1995, p. 343) who indicated that ‘these (distance) costs can arise from spatial distance, such as the costs of travel, transportation and coordination over distance and across time zones’. Research data of Collinson and Rugman (2008, p. 218) shows that in terms of the distribution of production from Asian MNEs, there are differences between industries. Data of the Japan Bank for International Cooperation (JBID, 2004, in Collinson and Rugman, 2008, p. 218) shows that, as it comes to, in this case, Japanese multinational manufacturing companies, these firms ‘had an average of 26.1% of their production operations outside Japan’ and almost 74% of their production operations were in Japan and over 91% of their production operations were in the Asia region. More specific, data shows differences across industries as shown in the following table:

Type of industry % Of production overseas

Chemicals 16.9

General Machinery 18.4

Electrical Equipment and Electronics 38.8

Automobiles 26.8

Table 2.1: Industry specificity in production in Japan. Source: Data extracted from Collinson and Rugman, 2008, p. 218.

As Collinson and Rugman (2008, p. 227) point to transportation costs as the cause of less internationalised industries, this leads to the following proposition:

Working proposition 5: Inter-regional LoF of international manufacturers is increased by transportation costs.

(19)

N.W. Baak 18

In order to provide an overview of the working propositions, a conceptual research model is developed in figure 2.1. The regional orientation of international manufacturers will be investigated from both an resource- based and an institutional-based perspective. From the resource-based perspective, alliances are expected to explain the international diversification of international manufacturers (working proposition 1). From an institutional-based perspective, the home-country effects in the form of institutional regulatory aspects will explain the home-region orientation of these MNEs (working proposition 2). At the other / lower side of the figure, transportation costs are expected to constrain the inter-regional diversification (working proposition 5), while formal institutions in the host region are also expected to constrain inter-regional geographic diversification (working proposition 4a), intra- regional formal institutions are expected to explain growth within the region (working proposition 4b). As a review of the literature learned that upstream and downstream activities faces different forms of LoF (Collinson and Rugman, 2008), working propositions 3a and 3b pay attention to this phenomenon.

Figure 2.1: Conceptual research model. Source: Author Alliances (wp 1)

Shape geographic diversification Formal institutions in

the HC (wp 2)

Constrain geographic diversification Regional oriëntation international

manufacturers Regional formal institutions (wp4a) Distinquish between upstream and downstream activities (wp3a, b) Regional formal institutions (wp4a)

Intra- regional formal institutions (wp 4b)

Transportation costs (wp 5)

(20)

N.W. Baak 19

3 Longitudinal multiple-case study research design

In order to test the propositions that are developed, this chapter explains how to do so. It starts which explaining why a post-positivist research philosophy (Hyde, 2000; Gephart, 2004) is adopted to this research and develops the research design. Furthermore, the sampling strategy and the way in which the data will be collected will be discussed.

3.1 Research philosophy

As ontology describes how we see the world around us, epistemology describes how the world around us can be investigated. There are two broad ontological perspectives; the objectivist and the subjectivist perspective. As a research philosophy is important because ‘to know what you're doing, you need to know how your model of knowing affects what you are doing (Miles and Huberman, 1984, p. 20), a postpositivist (realist) research philosophy is adopted in this research. This is in line with Hyde (2000) and Gephart (2004), who use the post-positivist approach for qualitative research. According to Gephart (2004, p. 456), the positivist and the postpositivist stances are often used in qualitative research. They both rely on the assumption of an ‘objective world external to the mind that is mirrored by scientific data and theories’. Furthermore, postpositivism differs from positivism ‘in holding that reality can be known only probabilistically’ Gephart (2004, p. 456), and also differs in the sense that the starting point is falsification, not verification. This is exactly what this research aims to do as propositions will be tested using qualitative data.

3.2 Research design

In order to answer this research question, a longitudinal multiple-case study research design will be used. Using a case-study design actually becomes more and more popular in the scientific literature (Siggelkow, 2007). Gephart (2004) argues that case studies can provide insights that are difficult to produce with quantitative research, because of the cases are studied in a real-life context. Eisenhardt and Graebner (2007, p. 25) describe case-studies as ‘one of the best (if not the best) of the bridges from rich qualitative evidence to mainstream deductive research’. These authors further argue that, in line with Yin (1994), multiple-case studies typically provide a stronger basis for theory building, and enable broader explanation of research questions and theoretical elaboration (Eisenhardt et al., p. 27). This is exactly the case here. Because it is the aim to get insight in the regionalization between three regions, it is useful to investigate MNE’s from all three regions. Therefore, different MNE’s will be investigated. As Burgelman (2011, p. 599) indicated the importance of longitudinal research, especially for the field of IB, a longitudinal dimension is added to this research. In addition to that,

(21)

N.W. Baak 20

adopting a longitudinal approach, will ‘more readily be able to document the intersection of continuities and contingencies, and the effects of context on outcomes’ (Burgelman, 2011, p. 599).

3.3 Theoretical sampling strategy

Theoretical sampling can be seen as a suitable way to select cases (Yin, 2003; Eisenhardt, 1989). Furthermore, as it comes to select a sample strategy, Coyne (1997, p. 603) argues that ‘the researcher should find out what information is most needed and most useful in a given situation, and then employ the most suitable methods. As it is the aim of this research to explore the institutions that both shape and constrain the intra- and inter-regional diversification of IM’s, information-oriented selection is applicated to select a representative sample. This is in line with Eisenhardt and Graebner (2007), who argue that theoretical sampling can be used if it is suitable to the aim of the research. According to Flyvbjerg (2006, p. 229), information-oriented selection is a suitable way of sampling, because the strategic selection of cases can increase the generalizability of the results. Furthermore, as a second criteria, MNE’s are purposeful selected on the criteria of having substantial energy-related activities. This is because MNE’s with substantial energy-related activities are expected to reflect the influence of, at least formal, institutions more than other manufacturers. The reasoning is that because, from an institution-based perspective, it is expected that energy-related activities are more vulnerable to formal institutions as governmental policies and restrictions, as they are strategically important in terms of social relevance (every country / region needs the supply of energy).

To pay attention to the differences between the three regions in the Triad (Ohmae, 1985 in Rugman and Verbeke, 2005), MNEs from all three regions will be selected. Based on these criteria, the following six companies are selected (at least one in each region): Based in Europe: Siemens and Alstom, based in Asia (Japan): Hitachi, Mitsubishi Heavy Industries and Toshiba and based in the US: General Electric (GE). The tables 3.1 a and 3.1b on the next page gives an overview of these companies. It is not claimed that these MNEs are representative for al manufacturing MNEs in the Global Fortune 500. According to Thomas (2011, p. 514), this is not necessary because the significance of the analysis does not rest in the representativeness of the subject.

3.4 Data collection

The data will be collected in two ways, both quantitative and qualitative. In order to see the patterns of regionalization, these data will be extracted’ from 2007 till 2012, thus making it a longitudinal study. This timeframe has been choosen because it includes the financial crisis, and therefore making it possible to determine which influences this crisis had on the regionalization of the MNEs. The

(22)

N.W. Baak 21

longitudinal character of the study makes it possible to see trends and patterns within and between MNE’s.

Establishing the degree of regionalization

To determine the degree of regionalization of the sample, specific data will be extracted from the annual reports of the MNEs. In line with Rugman and Verbeke (2004, p. 5), the main proxy will be the sales data, as ‘these constitute the ultimate reflection of market success’. In their 2008 article, Collinson and Rugman (2008, p. 218) distinguish between proxies to measure upstream and downstream activities because ‘prior studies that rely solely on sales data have been open to criticism that these provide a ‘one-sided’ measure of internationalisation. For this reason, both measures for upstream and downstream activities will be used in this research. Collinson and Rugman (2008, p. 218) use sales data as a ‘downstream measure of firm outputs and indicate the geographic distribution of a firm’s customers and its home-region market dependence’. They use assets as the proxy for upstream activities and also as industry specific measures of activity / presence, as it is ‘the single most representative and uniformly reported measure for the input side’. The robustness of these measures is confirmed by Oh (2009, p. 336), who found that measures, which are based on sales and assets, are better than scope measures, which are based on country counts’. Furthermore, employee- data, in line with Rugman and Verbeke (2004), will be used as a third proxy to measure the geographic distribution of MNEs activities.

Explaining the degree of regionalization

The second way of data collection regards the ‘story’ behind these quantitative data. Within the timeframe as mentioned before, Financial Times (FT) articles which are related to these MNEs will be collected and coded, to find key-events which can explain the ‘story’ (Pratt, 2009, p. 860) behind the data. If necessary, thus if there are too little FT articles available, additional data sources such as the Nikkei Asian Review (NAR) are used. As the sample is limited to six companies, also the generalizability of the research will be limited.

The next table indicates which data sources are used per MNE. LexisNexix (a database) was used to select the articles. The first number indicates the total amount of articles when the ‘power search’ function was used in LexisNexix within the timeframe 2008-2012. The second number (in brackets) indicates the amount of articles that are used for coding. This number was found after filtering the first results adding the criteria of ‘Power’, ensuring that articles were selected which information about power-related activities. After that, the selection of articles was filtered on company name. In total, 831 articles were used for coding.

(23)

N.W. Baak 22

Company name Nikkei Asian Review Financial Times

Hitachi 782 (100) 593 (22)

Toshiba 738 (144) 769 (79)

Mitshubishi Heavy Industries 276 (39) 97 (93)

Alstom 413 (65)

Siemens 1893 (154)

General Electric 2710 (135)

Table 3.1: Data sources and articles used for coding

3.5 Analysing methods

Regarding the analyses, in line with Eisenhardt (1989) and Doz (2011), within-case analysis will be conducted, followed by cross-case analysis. The qualitative data will be analysed using ‘thematic coding’. In line with Ryan and Bernard (2003, p. 87), in the context of this research, the definition of a theme or themes is: ‘fundamental concepts we are tying to describe’. In this research, these themes are the IB and regionalization concepts that are discussed in the propositions. As these authors furthermore identify proofreading as a suitable way to start analysing because they make some as yet inchoate sense’ (Ryan and Bernard (2003, p. 88), proofreading is applied to get an idea of the most important themes in the articles. After that, codes are developed per working proposition, as is summarized table 3.2.

Code category Code Description

Working proposition

FSAs Alliances Activities in host countries related to forming alliances WP 1 Home country Governmental support Support of the government to firms in home countries _{WP 2} effects Non- governmental home

country effects _{Home country effects which are not governmental- related} _{WP 2} LoF

Formal constraints

upstream _{Upstream activities related to market risks} _{WP 3a} Formal constraints

downstream The LoF MNEs face from formal institutions in host _{countries related to downstream activities} _{WP 3b} Entry barriers _{Tariff and non tarrif barriers that constrain inter regional}

geographic diversification WP 4a, b

Transportation costs _{Costs related to tranportation activities} _{WP 5} Intra LoF _{The Lof that manufactoring MNEs face intra regional} _All Inter LoF _{The Lof that manufactoring MNEs face inter regional} _All

(24)

N.W. Baak 23

After developing codes, these codes are programmed in Nvivo. Furthermore, the relevant financial times, and from other sources articles, are uploaded in Nvivo and after that, the text is coded. During the coding process, is was possible to compare the data across different sources in order to uncover similarities and differences between units of text, also called the ‘constant comparison method’ (Ryan and Bernard (2003, p. 91). The most relevant-meaningful sentences are used in the within case analysis to illustrate and explain the quantitative data and regionalization patterns in the next chapter.

(25)

Name Hitachi, Ltd Toshiba Corporation Mitsubishi Heavy Industries

Home country Japan Japan Japan

Formation year 1920 1978 1885

Ownership Listed Listed Listed

Amount of employees in 2012 323540 203000 68887

Total sales in 2012 9665883 (m yen) 6100262 (m Yen) 2820932 (m Yen)

Total assets in 2012 9418526 (m Yen) 851365 (m Yen) 3963987 (m Yen)

Corporate Strategy in 2008 Diversified engineering and electronic conglomerate Diversified electric/ electronic manufacturer Diversified engineering and electronic conglomerate

Corporate Strategy in 2012 Diversified engineering and electronic conglomerate Diversified electric/ electronic manufacturer Diversified engineering and electronic conglomerate

Growth mode Alliances, M&A Alliances, M&A Alliances, M&A

Markets Information and Telecommunication, Power Systems, Social _{Infrastructure&Industrial} _Systems, _Electronic Systems&Equipment, Construction Machinery, High Functional Materials & Components, Automotive Systems, Components & Devices, Digital Media & Consumer Products, Financial Services

Digital Products, Electronic Devices, Social Infrastructure, Home Appliances Segment

Shipbuilding & Ocean Development, Power Systems, Machinery & Steel Infrastructure Systems, Aerospace Systems, General Machinery & Special Vehicles, Others

Power- related Fossil Thermal Power Generation Systems Thermal Power Generation Systems Thermal Power Generation Systems activities Renewable Nuclear, hydro, wind, solar Nuclear, hydro, wind, solar Nuclear, wind

MNE's reporting about high impact

environmental issues 2011: Great East Japan Earthquake, floods in Thailand.

2011: Great East Japan Earthquake, floods in

Thailand. 2011: Great East Japan Earthquake

(26)

N.W. Baak 25

Name Alstom Siemens GE

Home country France Germany North America

Formation year 1928 1847 1879

Ownership Listed Listed Listed

Amount of employees in 2012 93998 405000 305000

Total sales in 2012 19934 (mE) 78296 (mE) 147,4 bn dollar

Total assets in 2012 31047 (mE) 108282 (mE) 684,2 bn dollar

Corporate Strategy in 2008 Diversified engineering and electronic conglomerate Diversified engineering and electronic conglomerate Diversified engineering and electronic conglomerate

Corporate Strategy in 2012 Diversified engineering and electronic conglomerate Diversified engineering and electronic conglomerate Diversified engineering and electronic conglomerate

Growth mode Alliances, M&A Alliances, M&A Alliances, M&A

Markets Thermal Power Systems Energy Power & Water

Renewable Power Systems Healthcare Oil & Gas

Transport Industry Energy Management

Grid Infrastructure & cities Aviation

Healthcare Transportation

Home & Business Solutions Power- related Fossil Thermal Power Generation Systems Thermal Power Generation Systems Thermal Power Generation Systems activities Renewable

Nuclear, hydro, wind, solar, New Energies (geothermal/ solarthermal/ biomass/ tidal stream enery/ wave energy)

Nuclear, hydro, wind, solar Wind, solar

MNE's reporting about high impact environmental issues

No No No

(27)

4. Within-case analysis

This chapter contains the within-case analysis of the six MNEs that are identified using theoretical sampling. Each within-case analysis contains a table including quantitative data (based on annual reports) and qualitative data (based on the articles) which is the core of the within-case analysis. Consequently, these tables are discussed per case. In the case of the Asian MNEs, the tables of these MNEs are presented separately, however the rest of the within case-analysis of the Asian MNEs is performed altogether. Representing the data in this way is more suitable as the evidence in Asian articles is overlapping.

North American MNE General Electric

As the data confirms and not surprisingly (Rugman and Verbeke, 2004), the diversified conglomerate General Electric is a home-region (North America) oriented MNE. Not only sales data but even more the downstream measure of internationalization, assets, indicate that between 2008 and 2012 their sales and assets were not equally distributed about the triad (Ohmae, 1985), but instead show that most of the sales and assets are assigned to the US, followed by respectively Europe and Asia as the second and third regions in terms of presence. Former chairman Mr. Immelt is aware of the fact that GE’s activities are not equally distributed about the Triad as he says "We are an imperfect global company, but we're a better global company" (Crooks and Marsh, 2012). The energy sector is important for GE, as energy sales came to $31bn in 2011 (McLannahan, 2012). Regional data regarding the geographical distribution in terms of employees was not given in the annual reports.

Evidence in the FT articles shows that the positive investment climate in the US, because of subsidies for renewable energy related technologies, that is shaped by the US government, will expand the market for renewable power (Guerrera, 2008) and will thus facilitate expanding of GE’s activities in renewable energy in their home country (McNulty and Baer, 2009). This suggests that the institutional environment in the US seems to be attractive for GE, indicating that this specific home- country effect at least partly explains the home-regional orientation of GE in North America. On the other hand, companies like GE face pressures from the local government to be ‘in complicance’ with specific regulations such as the Clean Air Act (Baer, Guerrera and Chung, 2008), which suggests that international manufacturers in the US face isomorphic pressures from the government in the energy- market.

Obviously, the inter-regional Lof is increased by the protectionist activities of host governments in order to protect their markets and hold, by example in the case of France with respect to Alstom, specific companies in local hands (Hollinger, 2009). International competitiveness is

(28)

N.W. Baak 27

increased by overseas competitors who benefit from scale-advantages (Lemer, 2011). In the articles no evidence is found, from an institutional perspective, that upstream and downstream activities of GE face different forms of LoF. Although there’s evidence of protectionism, as the call of an GE executive to the Chinese government to open their markets , suggests (Crooks, 2012), no evidence of tariff- and non-tariff barriers is found in the articles with regard to the rest of the triad-regions.

(29)

Regional profile General Electric 2012 2011 2010 2009 2008 Region

NA EU AS ROW NA EU AS ROW NA EU AS ROW NA EU AS ROW NA EU AS ROW

Regional presence (% sales) 47,7 18,6 16,6 17,1 47,4 19,7 15,8 17,2 50,2 20,7 13,9 15,2 46,2 23,5 13,2 17,0 46,7 24,1 12,9 16,2 Regional presence (% assets) 50,7 28,2 8,2 12,9 47,0 29,7 8,7 14,7 51,9 26,7 8,2 13,2 49,9 28,1 8,4 13,6 49,7 28,6 9,4 12,2 Regional presence (% employees) NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA Orientation

Bi- regional Bi- regional Home Regional Bi- regional Bi- regional

Dimensions Evaluation Explanatory quotes

Alliances (inter and intra regional)

Alliances are used by GE to be more competitive, to increase parnterships with governments (public- private partnerships) and to generate synergies between business units

"A venture in nuclear may be more complicated, said analysts, as Hitachi has an alliance with GE, while MHI is close to Areva of France" (McLannahan, 2012).

"GE's nuclear operations are now part of a joint venture with Hitachi of Japan that has two businesses: Hitachi GE Nuclear Energy, which is owned roughly 80-20 by the Japanese and US groups, and is based in Japan, and GE Hitachi, which has 60-40 US-Japanese ownership and covers the rest of the world. Although the businesses are formally separate, they are closely linked" (Crooks, 2011).

"The deal with XD, based in the central city of Xi'an, reflects GE's strategy of seeking to use alliances to make progress in an important but challenging market" (Crooks, 2012).

"General Electric, the largest US industrial group, is signing a series of deals with China this week in energy, rail and aviation as it drives to improve its disappointing performance in the Chinese market. GE says the announcements reflect the promise by Jeff Immelt, chief executive, that its collaborations with Chinese companies would create global sales and safeguard US jobs" (Crooks, 2011).

"General Electric and Cisco Systemshailed yesterday a new "era of public-private partnerships" that will enable companies to win a large share of the billions of dollars spent by governments around the world to stimulate their economies" (Guerra, 2009).

"GE plans to run Converteam as a standalone business unit but still expects to generate operating synergies worth about $250m a year. GE also hopes to substantially grow the company's revenues by rebranding it and selling its products more globally" (Lemer, 2011).

(30)

N.W. Baak 29

"Google and General Electric on Wednesday announced an unusual alliance to promote greater use of renewable energy in the US" (Waters, 2008).

"GE's long-standing relationships with power companies, its international footprint and its ability to fund energy projects through its financial services arm all add to its potency (Lemer, 2011)".

Regulatory home- country effects

Growth in energy - related activities is enhanced by the policy of the US government to invest in renewable energy.

"Mr Obama's backing for alternative energy helped makers of wind turbines" (Wood, 2008).

"The statement by two of the largest US multinationals underlines corporate America's realisation that governments have become increasingly important customers at a time when companies and consumers have been hit by the slowdown (Guerrera, 2009)".

"Mr Urquhart said the Obama Administration's emphasis on energy will open up opportunities for further investments in the sector. Given difficulties obtaining credit in the US amid the crisis, he said this year a "big focus'' of its funds will be in lending, though other types of investment will be made, too (McNulty and Baer, 2009)".

"GE also said US environmental regulators warned that a Pennsylvania power plant owned by a GE Capital subsidiary was not in compliance with the Clean Air Act (Baer, Guerrera and Chung, 2008)".

"But the French government has asked the domestic consortium of Schneider Electric and Alstom, which initially pitched the lowest bid, to modify their offer to be more competitive with GE and its fellow foreign rival Toshiba of Japan, in what appears to be a sign of the desire to see the business remain in local hands (Hollinger, 2009)". "The market for renewable energy is worth an estimated $60bn a year and is expected to expand rapidly as governments and utilities strive to reduce their dependence on fossil fuels and turn to cleaner energy (Guerrera, 2008)". LoF differences upstream and downstream activities No evidence found - Inter- regional LoF of formal institutions Intra- regional LoF of formal institutions No evidence found -