To what extent do industrial clusters impact location decisions of Multinational Enterprises (MNEs) in relation to intangible assets?

(1)

Master thesis MSc. Business Administration –

International Management

Supervisor Erik Dirksen MSc. Second reader Dr. Ilir Haxhi

Student Frederick Martijn van Ginkel

Student nr. 10994815

Date Monday, February 29, 2016

Word count 23233

Institute University of Amsterdam ABS

“To what extent do industrial clusters impact

location decisions of Multinational Enterprises

(MNEs) in relation to intangible assets?”

(2)

Statement of originality

This document is written by Student F.M. van Ginkel 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.

(3)

Abstract

This research investigates the influence of industrial clusters on location decisions of Multinational Enterprises (MNEs) and how intangible assets, which become increasingly important in the contemporary world economy, influence this relationship. A multiple case study was conducted through interviews and examination of secondary data from four firms that already have value creating activities through subsidiaries abroad or are planning to do so. Evidence is found that industrial clusters provide appropriate grounds for the development of new Firm Specific Advantages (FSAs) especially in the form of intangible assets. The reason for this is that firm proximity in industrial clusters leads to firm specialisation that increases the need to develop new brands, relationships, loyalty programmes, alliances, access to local business networks and a positive balance on knowledge spill overs. Evidence is also found that FSA development can be accelerated by entering industrial clusters through Joint Ventures with strategic local firms leading to a more efficient and quick offset of Liabilities of Foreignness (LoF) in industrial clusters than in non-industrial clusters.

(4)

Acknowledgements

Firstly, I would like to thank my supervisor Erik Dirksen MSc. for his supervision and guidance through the process of creating this thesis and the second reader Dr. Ilir Haxhi for his examination of this research. Secondly, I would deeply thank my mother, Mieke van Ginkel MBA, for her guidance, help and encouraging words. Further I would like to thank my father Fred, brother Jordy. They really deserve credits for their unlimited support during this MSc. program. Without their support, help, patience and love I would not be able to finish this research.

(5)

List of tables

Table 1: The attributes of global cities 14

Table 2: Review on location theory 17

Table 3: The regional orientation of MNEs listed in the Global Fortune 500, 23 Table 4: Three ways for MNEs to engage with their institutional context 25

Table 5: The different sources of data for Nvivo 37

Table 6: The codebook used for Nvivo analysis 38

Table 7: The results on the five working propositions 50

List of figures

Figure 1: The conceptual model of The Uppsala Model 19

Figure 2: The hierarchical model of entry modes 20

Figure 3: Sources of competitive advantage: A transaction costs model 26

Figure 4: Interpretation of the four generic roles of MNEs 27

Figure 5: The conceptual model of The Uppsala Internationalization process model 28

Figure 6: The conceptual framework of the author 30

Figure 7: The conceptual framework including the five propositions and theoretical concepts 31

Figure 8: The four layer model 48

(9)

1 Introduction

Location theory is one of the oldest concepts in International Business (IB) literature and gives direction for MNEs where to invest. Location theory is also traditionally related to MNEs’ spatial behaviour and its need to achieve economies of scale and scope while minimizing costs of doing business abroad (CDBA) and maximizing firm performance (Mariotti, Piscitello, & Elia, 2010) (Chung & Kalnins, 2001). In contrast to the traditional concept of location the following trends play a role in the domain of International Business (IB) literature:

Increase in industrial clustering or spatial agglomeration

Besides the critical role of location specificity in location decisions, spatial agglomeration or industrial clustering is argued to play an increasingly important role too. It functions as a pull factor for MNEs’ location choices in terms of supporting industries and infrastructure or availability of proficiency in the English language (Laamanen, 2012 ). According to Mariotti (2010) agglomerative behaviour is influenced by information externalities and potential knowledge spill overs which may act as centrifugal or centripetal forces (Mariotti, 2010). A practical example is Silicon Valley in California where various successful computer related and other high tech firms are established. MNEs mainly establish there to grasp some of the present knowledge in that area. Mariotti (2010) also claims that MNEs do not tend to agglomerate with domestic local companies as perceived knowledge outflows are higher than knowledge inflows. Instead, MNEs prefer to agglomerate with other MNEs as they want to maximize the balance between knowledge inflows and outflows, unless domestic firms have significant competitive advantages that lower information costs (Mariotti, 2010).

A changing world economic scenario

The changes in location- and internationalization theory and the increasing importance of industrial clusters have resulted in serious implications for MNE’s entry modes, locations decisions and Foreign Direct Investment (FDI) decisions. According to Dunning (2009) during the last two decades three dominant changes occurred in the world of international business activity: The first is the emergence of intellectual capital and intangible assets as the key wealth creating assets in most industrial economies. Dunning’s (2009) empirical evidence claims that the market value of most industrial economies has risen from 1.5 times in 1982 to between 2.5 and 5 times their intangible assets in 1995. Stewart (1997) claimed that the annual expenditure on Information Technology already exceeds that on production technology and that the knowledge component in manufactured goods has risen from 20% in the 1950s to 70 percent in 1995. Another indicator of the significance of intangible assets as key wealth creating assets is the growth of services, in particular knowledge intensive services. In 1995 the services industry accounted for 63 percent of the world GDP compared to 45 percent in 1965 (Stewart, 1997).

Secondly and shortly mentioned in the first paragraph is the significant increase in global trade made possible by technological development and economic integration through reduction of trade barriers. These developments allow firms to pursue value-adding activities on a global scale (Dunning, 1998) and emphasize the increasing role of Location Bound Assets (LBA) as complements to downstream Firm Specific Advantages (FSAs).

The third characteristics of the current global economy stated by Dunning (2009) is the emergence of “alliance” capitalism (or collective-, stakeholder-, and collaborative capitalism). This form of capitalism is characterized by growing extent to which stakeholders are collaborating and working together to reach their respective goals. In this, such collaborations include closer intra firm relations

(10)

and cooperative agreements between suppliers, customers and even competitors in a specific geographic area, such as an industrial cluster. This results in implications by which intangible assets are transferred in the MNE network.

Clearly a paradox has emerged related to the emergence of industrial clustering precisely at a time when globalization seems to dominate economic activity. This paper tries to resolve this paradox by explaining why regional focus matters for location choices by focusing on the role of intangible assets, which play an increasingly dominant role in the changing economic world scenario.

Research problem and research question 1.1

As stated in the previous paragraph the changing world economic scenario is characterized by the increasing importance of intangible assets such as intellectual capital, knowledge- and learning capabilities, access to networks ,experience and brand image. Currently MNEs deliberately locate in industrial clusters to gain certain advantages as a result of firm proximity. For example Huawei that wants to take advantage from the knowledge and subsequent knowledge spillovers in Silicon Valley, where many R&D flagship centres are established. Another example is Bank of China that opens its subsidiary in London, which is the banking capital of the world, to upgrade its image and reputation to a world player in the financial market. This research aims to address this phenomenon by answering the following main research question: “To what extent do industrial clusters impact location decisions of MNEs in relation to intangible assets?”

Research objectives 1.2

This research paper aims to achieve the following goals:

1. Be able to explain the primary motives for MNEs when selecting for locations industrial clusters.

2. Be able to explain the difficulties MNEs might face in accessing and developing new FSAs when choosing for a location in an industrial cluster.

3. Be able to explain how location decisions in industrial clusters can result in performance implications for MNEs.

4. Be able to explain how location decisions in industrial clusters can result in competitive advantages for firms.

5. Be able to name paradoxes that play a role among the research constructs. 6. Be able to critically examine relevant literature among the research constructs. 7. Be able to develop new theory based on inductive- and explorative research.

Scientific relevance 1.3

This study adds new insights and dimensions in the ongoing debate of traditional location- , agglomeration- and internationalization theories and elaborates further on contributions by Rugman, Verbeke & Nguyen (2011). Secondly, this study evaluates traditional and contemporary literature of these theories by adding the role of intangible assets and by arguing that MNEs’ successfulness in internationalization is largely determined by the development of these intangible assets (Caves, 1997). Thirdly, this paper tries to solve a paradox that has emerged related to the emergence of industrial clustering precisely at a time when globalization seems to dominate economic activity.

Managerial relevance 1.4

This study will help managers to find appropriate locations for new value creating activities. The emphasis of this research lies on locations in industrial clusters. Furthermore, this research helps to

(11)

identify possible advantages as result of industrial clusters and helps to identify factors to be considered inherent to industrial clusters. Also, factors are identified helping MNEs’ subsidiaries to develop new Firms Specific Advantages (FSAs) more efficient and, by doing so, overcoming Liabilities of Foreignness (LoF) more efficiently and in a shorter timeframe.

Research outline 1.5

This research consists of eleven chapters. The first chapter gives a short introduction of the research problem and research objectives. The study then shifts the theoretical background incorporating concepts related to the three variables: (i) Industrial clustering (ii) Location choice and (iii) Resource Based View (RBV). During the examination of these concepts research gaps are identified and five subsequent research propositions are made. After the literature review the study proceeds into the methodology chapter. This chapter describes the research design for the examination of the central research question based on the conceptual framework. Then the methodology chapter describes the fundamental assumptions related to the research design and applied research methods. After this the results will be examined through within case- and cross case analysis. The emergent paradigms from the within case analysis will form the bases of new theory that is developed and reported in the recommendations chapter. After the recommendation chapter this study proceeds into the discussion chapter in which the result on the five propositions are discussed and scientific-and managerial relevance is reported. Lastly, the conclusion chapter will report the most important evidence and will answer the central research question.

(12)

2 Introduction of the literature review

This part of the study is aimed to shed light on the attractiveness of industrial clusters and the implications for MNEs’ location decisions. It does so by focusing on intangible assets that become increasingly important in the contemporary economic activity. In mainstream IB-literature scholars support the ideas of Country Specific Advantages (CSAs) and Location Specific Advantages (LSA’s). However, the role regional specific advantages is still considered as a grey area in the literature. Based on evidence from (Mariotti, Piscitello, & Elia, 2010) I expect that MNEs can benefit from their presence in industrial clusters and that this can lead to competitive advantages for MNEs and their network of subsidiaries. We aim to address this gap starting with a short examination on the last fifty years of IB literature. By doing so, the focus is brought on the subsidiary and its role of its location decision. Then we examine concepts that are related to the central research question: “To what extent do industrial clusters impact location decisions of MNEs in relation to intangible assets?”. To examine this relationship, I first critically examine related literature leading to the research objectives and– propositions. The subsequent conceptual framework can be found in the sixth paragraph of this research paper.

Short examination of the literature on International Business over the last fifty years 2.1

Over the past fifty years the field of International Business (IB) has matured and there have been shifts in the core unit of analysis. The work by Vernon (1966), Dunning J. (1958) and Rugman (1980) clearly used country factors as their core unit of analysis, using national statistics on international trade flows and investments. This stream builds on the neoclassical assumption that countries have differences in factor endowments and Country Specific Advantages (CSAs) that will lead to international transactions.

During the 1970s the focus shifted to MNEs and their Foreign Direct Investments (FDIs). In this second stage Hymer (1960) has made fruitful contributions by positioning MNEs and their FSAs at the core unit of analysis of his approach. Hymer’s (1960) insights have led to the recognition of firms having Firm Specific Advantages (FSAs), both stand-alone such as image and R&D and higher order capabilities such as routines and recombination capabilities, that are required to overcome the Liability of Foreignness (LoF) (Hymer,1960) (Zaheer, 1995). The concept of LoF will be examined in more detail later on in the literature review.

In the 1980s more attention was dedicated to MNEs and their differentiated network of subsidiaries and clusters. In addition, more research was dedicated on specific, cultural, economic and institutional context of subsidiaries and their implications in the MNE-network. Also, during this stage an increasing importance on subsidiary initiative is developed. By building further on this stage the focus of FSA generation is now shifted from the parent company to a collective responsibility for the whole MNE-network including subsidiaries. In this stage, the subsidiary’s internal resources, in terms of FSAs, combined with subsidiary initiative have a strong impact on the contributory role of new FSAs.

Cleary the field of IB has matured bringing the contemporary view on the subsidiary and its value creation with FSAs and CSAs in home and host country. This research builds further on this contemporary view by focusing on the subsidiary as its core unit of analysis and its value creation in a particular geographic region or industrial cluster.

(13)

3 Key Themes within Industrial Clustering literature

Industrial clustering in an era of increasing global competition involves a paradox. It is recognized by famous scholars such as Porter (2000) and Dunning (2000) that changes in technology and competition have decreased the traditional roles of location such as sources for capital or technology. During that same time industrial clusters are becoming increasingly important and more apparent in every national, region and metropolitan economy, especially in developed countries (Porter, 2000).

In IB literature there are various definitions of industrial clusters and so there is much ambiguity in the way this concept is used (McCann & Gordon, 2000). In this research paper the definition of Porter ( 1998, p. 78) is used stating that industrial clusters are “geographic concentrations of interconnected companies and institutions in a particular field or region which encompass an array of linked industries and other entities important to competition”. In addition to this definition of Porter, industrial cluster are seen as a regional concept in International Business. It is embedded beneath the level of the country, but above the local level (Leydesdorff & Cooke, 2006). In geographical terms this is how industrial clusters can be put into perspective in IB- theories and this research paper.

The passage mentioned below will discuss key themes related to industrial clustering: agglomeration economies, knowledge spill overs, information externalities and social- and knowledge networks. By doing so, the focus is on the first variable, the independent variable, and its associated concepts.

Agglomeration economies 3.1

Agglomeration economies are seen as positive externalities as a result of industrial clustering (Ellison & Glaeser, 1992). Nachum & Keeble (2003) argue that firms that collaborate jointly in a region through similar formal and informal practices can profit from advantages due to their proximity. Krugman (1991) claims that geographical proximity leads to industrial specialization and increased returns in industrial clusters. Furthermore, industrial clustering can have positive impact on firm performance through agglomeration economies as newly locating MNEs make use of existing facilities such as knowledge networks, infrastructure and decreased transaction costs (Folta, Cooper, & Bak, 2006).

Shaver & Flyer (2000) argue that if firms are heterogeneous they benefit differently from agglomeration economies. This insight implies that firms in an industrial cluster differ in the net benefits they receive from agglomeration. Based on their evidence on the location choice and survival of foreign greenfield investments in U.S. manufacturing industries, Shaver & Flyer (2000) argue that “firms with the best technologies, human capital, training programs, suppliers, or distributors will gain little, yet competitively suffer when their technologies, employees, and access to supporting industries spill over to competitors.” (Shaver & Flyer, 2000 p. 1175). This implies that these firms have little to benefit from agglomeration economies and geographical proximity. In contrast, firms that have a weak position on the aforementioned positions have little to lose and a lot to gain so they are motivated to locate in industrial clusters. Hence, the creation and transmission of knowledge in terms of knowledge spillovers can be seen as an important advantage or disadvantage of a location in an industrial cluster (Paci & Usai, 1991) (Mariotti, Piscitello, & Elia, 2010).

(14)

Knowledge spillovers 3.2

An important benefit of industrial clustering is the creation and transmission of knowledge (Paci and Usai, 1999) leading to innovation and learning. For example firms present in an industrial cluster can benefit from social learning, production efficiencies and higher demand for products (Chung & Kalnins, 2001). These are the result of information flows between firms and customers supported by their location proximity. Location proximity permits firms to access leading techniques and allow consumers to evaluate multiple options as firms are located near one another (Chung & Kalnins, 2001). Hence, industrial clustering increases the possibility for information to circulate and reach other firms or customers.

Mariotti, Piscitello, & Elia’s (2010) empirical evidence, based on the spatial distribution of 686 Italian subsidiaries, claims that firm’s location behaviour is influenced by information externalities, resulting in high information costs from uncertainty, and supposed knowledge spillovers acting as centripetal and centrifugal forces. Based on the traditional approach of Ellison & Glaeser (1992), where MNEs co-locate in industrial clusters due to a decrease in transport- and transaction costs, Mariotti, Piscitello, & Elia (2010) claim that industrial clustering is necessary for social learning processes but is not sufficient for interaction among MNEs. Mariotti, Piscitello, & Elia (2010) also claim that interaction not always leads to a positive balance between incoming and outgoing knowledge spillovers. In this, an example of incoming knowledge spillovers is the hiring of top managers from other firms in an industrial cluster. Examples of outgoing knowledge spillovers are technological leakages in terms of patent loss to other firms in the cluster.

The leakage of intellectual property, (McCan & Mudambi, 2004) can have serious impact on the competitive advantage of MNEs and is therefore seen as a negative effect of industrial clusters. In IB literature knowledge spillovers occur in two forms: intra-industry spillovers, which are spillovers between firms in a similar industry (Baptista & Swann, 1998) and inter-industry spillovers, which are spillovers between firms from different industries (Steurs, 1995) (Kugler, 2006). This means that when firms choose for a location in an industrial cluster they have to examine knowledge inflows and outflows inside industries and between industries and as they impact the firm’s competitive advantage. More specifically, MNEs prefer to agglomerate with other MNEs as they assume that knowledge inflows are higher than knowledge outflows, unless collaboration with domestic or local firms might result in a competitive advantage (Demirbag & Glaister, 2010). This makes industrial clustering a trade-off decision that depends on intangible assets (i.e. knowledge based view) of the MNE and its information position to local competitors and other MNEs (Mariotti, Piscitello, & Elia, 2010).

Information externalities 3.3

Information externalities refer to the information costs for MNEs as result of uncertainty on local institutions and a lack of experience in local markets (Mariotti, Piscitello, & Elia, 2010). MNEs that enter a host country will experience uncertainty due to their liability of foreignness (LOF) (Zaheer, 1995) resulting in information asymmetries and subsequent information costs. Engagement and organizational learning are ways to overcome LOF and eventually minimize information costs. To do so, and prevent these high costs of doing business abroad, MNEs learn from others MNEs through observation and adoption (Mariotti, Piscitello, & Elia, 2010). In this context mimetic isomorphism is seen as the adoption of successful practices of competitors. It implies that for MNEs entering a new host location, behaviour of its predecessors related to the composition of resources, knowledge spillovers and information externalities are copied and adopted. However, in IB literature there are other streams (Barney, 1986) (Bartlett & Ghoshal, 2013) that claim that MNEs have their own

(15)

culture, reputation, networks and behaviour as the important sources for their competitive advantages. We therefore made the following proposition:

Proposition 1: MNEs establishing industrial clusters are likely to adopt practices of successful

local firms based on an assessment of their intangible assets.

Industrial clusters as social- and knowledge networks 3.4

The role of social networks in the development of industrial clusters has received much attention among scholars (McCann & Gordon, 2000) (Sorensson, 2005). In their empirical evidence, the informal contacts and networks of employees are seen as main carriers of knowledge between firms in an industrial cluster). Through a survey among engineers in an industrial cluster in Denmark, Dahl & Pedersen (2004) claim that knowledge rated as of medium or high importance for their company is exchanged through informal contacts. Even confidential information regarding innovation in their firm’s product portfolio is exchanged with the possibility of technological leakages and loss of patents (Mariotti, Piscitello, & Elia, 2010) (Dahl & Pedersen, 2004).

Also, an important factor of industrial clustering according to Porter (1998, 2000) is the role of supporting institutions such as universities. The economic world scenario nowadays is characterized as a knowledge-based economy or knowledge economy in which knowledge is a vital asset and learning and innovations are its processes (Baptista & Swann, 1998) (Cooke & Leydesdorff, 2006). In these processes universities are supporting industries for firms to acquire knowledge through recruitment of students.

Metropolitan environments 3.5

Since the 1980s there is an increase in economic development of metropolitan cities, which are developed and densely populated city areas with less populated suburbs. This trend is supported by the concentration of economic activity and an increase in industrial clusters around these areas. Therefore these metropolitan environments are attractive for MNEs. Also the possibility for MNEs to tap from talent pools and the ability to build relationships with stakeholders are seen as key factors in attractiveness of these metropolitan environments. According to a study of Goerzen, Asmussen, & Nielsen (2013) metropolitan environments is often linked with the concept of global cities. According to them the concept of global cities consists of three attributes: (i) The abundance of advanced user services such as finance, law and other supporting services for MNEs (ii) Cosmopolitan environments referring to cities that developed their own culture, education and communication (iii) A high degree of interconnectedness, which is further explained in the following sub paragraph. These three attributes have implications for MNEs’ location strategies and competitive advantage. In table 1 below these three attributes are shown and for each attribute an example is given.

Table 1: The attributes of global cities

(16)

Empirical evidence of Diez & Berger (2005) claims that metropolitan environments have their own Metropolitan Innovation Systems (MIS) that attracts FDI. Based on evidence from survey data from European cities such as Barcelona, Stockholm and Asian cities such Singapore, Diez & Berger (2005 claim that these metropolitan have different R&D and innovation characteristics that attract MNEs from various industries. Further, Diez & Berger (2005 also claim that MNEs in Europe are more oriented to product innovation while their peers from Asia are more attracted to process innovation. To conclude, metropolitan environments are seen as highly developed regions containing possibilities to tap from Metropolitan Innovation Systems (MIS) and cosmopolitan environments. Also the abundance of advanced user services supplying MNEs and the high degree of interconnectedness as result of firm proximity are seen as important benefits of locations in metropolitan environments. Therefor to I want to examine the following proposition:

Proposition 2: The more a metropolitan environment in an industrial cluster, the higher the

likelihood for a location choice of MNEs.

Relationships and business networks 3.6

In the paragraph related to key themes within industrial clustering literature it is already mentioned that industrial clusters can be seen as social-and knowledge networks. Business networks can be seen as formal and informal networks between agents in an industrial cluster. The relationships that develop between agents have implications for firms in an industrial cluster. These relationships between agents do not develop without a reason, they provide firms local bargaining power, knowledge and innovation (Haucap, Heimeshoff, Klein, Rickert, & Wey, 2013) (Håkansson, 1987). The degree of interconnectedness refers to the effects of inter-organizational relationships on other inter-organizational relationships (Ritter, 2000). In this, “No business is an island” (Håkasson & Snehota, 1989) which means that firms are interconnected through relationships and business networks. Firms develop multiple relationships over time and each firm has a portfolio of relationships in which it is embedded. In this, firms not only access resources through suppliers and customers but also through financial institutions, shareholders, governments, universities etc. (Easton, 1992)

The degree of interconnectedness also impacts the complexity that MNEs face in host environments (Friedman, 1986). It adds the complexity of managing business networks and relationships as actors in these networks create situations that are beneficial for themselves and their firms (Håkansson, 1987). The approach of Ritter (2000) presents a view of interconnectedness from the perspective of a black box. In this black box the interconnectedness is not often overestimated but often underestimated (Ritter, 2000). industrial cluster contain business networks and relationships that make the flow of information more easy as a result of high degree of interconnectedness. This flow of information can be positive or negative. The author therefor wants to examine the following proposition:

Proposition 3: The greater the degree of perceived interconnectedness in an industrial

cluster, the higher the chance that MNEs choose for a location for their economic activities.

(17)

4 Key Themes within location choice literature

The concept of location is a key area of interest among international business scholars such as Alcacer & Chung (2007), Porter(2001) and Goerzen (2013). The concept of location gives answer on the questions where and why MNEs choose to locate specific value creating activities in particular locations. For firms, a thoroughly determined location in a host country, region or city enlarges its value creation opportunities in terms of its adaption to local markets and exploitation of economies of scale and- scope. It also gives the possibility to tap optimal locations for value adding activities and resources and enlarges knowledge transfers in the MNE network (Goldstein & Gronberg, 1984).

Location Specific Advantages (LSA’s) 4.1

Location choice has always been a major focus in international business and a central question in international business research. Literature on location choice is primarily based on the examination of value adding activities of MNEs in different location, specific advantages of locations (Dunning ,1998) and the allocation of resources to these locations. These Location Specific Advantages (LSAs) can, according to Dunning & Lundan (2008), be categorized in three broad categories: (i) Endowment effects, which explain why firms, business activities and location decisions are ‘naturally’ drawn to particular locations. For example the abundance of natural resources. (ii) Agglomeration effects which refer to low cost access to input factors such as labour, infrastructure or knowledge spillovers and (iii) Policy induced effects referring relocation of MNEs through governments subsidization. Location theory is often linked to the Dunning’s (2000) OLI-paradigm or Eclectic paradigm, which will be explained later on in this research. However there are more theories underlying MNEs’ location choices. In table 2 a summary is given on the most influential contributions last decades:

(18)

Table 2: Review on location theory

(19)

Liability of Foreignness (LoF) and Psychic Distance

4.2

For over decades IB scholars agree with the position that MNEs face costs when doing business in a host location (Hymer, 1960) (Zaheer, 1995) (Nachum, 2015). These costs arise from the Liability of Foreignness (LOF) (Zaheer, 1995) that refers to the unfamiliarity of the host environment that foreign firms face above local firms. This unfamiliarity is the result of cultural, administrative, geographic and economic differences between the host environment and home environment (Ghemawat, 2007). Hence the key driver of LoF can be seen as institutional distance between the home and host country (Eden & Miller, 2004). The LoF refers to the impact of these various forms of distance and explains why MNEs experience difficulties when operating in host environments, especially when local opponents are not hindered, giving these local firms a competitive edge. As this applies to every MNE, the LoF is seen as a fundamental concept of IB literature (Buckley & Casson, 1976).

Further it is argued that in order to overcome LoF and compete successfully against local rivals, MNEs need to provide their subsidiaries with Firms Specific Advantages (FSAs) such as managerial capabilities, which when recombined with host CSAs, can lead to the development of new local FSAs (Hymer, 1960). This Resource Based View (RBV) on international expansion suggests that subsidiaries in the MNE network will import FSAs of the parent firm in order to overcome LoF (Zaheer, 1995) and develop new Location Based (LB) and Non-Location Based (NLB) Firm Specific advantages (FSAs) in order to gain a sustainable competitive advantage in the host location.

Besides LoF there is another related concept that impacts location decision in MNEs’ international expansion, namely ‘psychic distance’ (Johanson & Wiedersheim, 1975). This concept states that internationalization starts in foreign countries that have similarities with the MNEs’ home country. More institutional similarities between home and host country make it easier to overcome LoF. After success, MNEs gradually enter countries that have less similarities with their home market and by doing so, enter countries that involve a bigger psychic distance. This means that the larger the psychic distance, the larger the LOF (Hymer, 1960).

Path dependency or Springboard Perspective 4.3

Path dependency is closely related to the previously mentioned concepts of LoF and Psychic Distance. It proposes that internationalization is a path dependent or cumulative process in which a cost benefit analysis is needed to weigh the creation of new FSAs against its costs and efforts (Rugman, 2011). Path dependency is seen as one of the fundamental assumption of the Uppsala Internationalization Process Model of (Johanson & Vahlne, 1977). This model suggest that firms start their international ventures by following an incremental pattern of expansion into countries that are often geographically close to their home market and by using low commitment entry modes such as exporting. As these firms offset LoF they expand to more distant countries and the process starts again, starting with recombining FSAs with host country CSAs and developing new FSAs and using low commitment entry modes.

The Uppsala Model

4.3.1

The foundation of the Uppsala Model (Johanson & Valne, 1977) is based on the pioneering work of Carlsson (1966), one of the first scholars that described the phenomenon of international value creation by firms that apply an incremental way of value creation in foreign markets. Based on the foundations of bounded rationality, uncertainty and behavioural theory of the firm, the Uppsala model proposes an explanation for managerial decision-making based on risk taking and resource commitment in host markets (Vahlne & Johanson, 2013). It also provides an explanation how

(20)

experience and learning processes affect managerial decision-making. In the conceptual model in figure 1 the four dimensions of the original Uppsala model (Johanson & Valne, 1977) are mentioned. It contains two state aspects: Market knowledge and Market commitment and two change aspects: Commitment decision and Current activities. The model proposes a positive relationship between Market knowledge and Commitment decisions. As MNEs’ foreign market knowledge increases, it will affect the Commitment decision in these markets. Current activities are also positively associated with Market commitment, leading ultimately to higher Market commitment.

Springboard Perspective

4.3.2

In contrast to the concept of path dependency in which firms create value in ‘psychological close’ markets and increase commitment through learning in evolutionary stages, there is the Springboard Perspective. The Springboard Perspective describes the internationalization process of emerging market multinational enterprises (EM MNEs). According to the scholars Luo & Tung (2007) EM MNEs use internationalization to acquire strategic resources and reduce home market institutional constraints. As these EM MNEs experience latecomer disadvantages they try to overcome these by series of aggressive and risk taking acquisitions of local firms, or by buying these critical assets from other MNEs. By doing so, EM MNEs try to compensate their latecomer disadvantages and improve on their competitive weaknesses. A second reason for EM MNEs to acquire other firms is the spatial pre-emption of scarce assets, in which a firm limits the room for other profitable firms through acquisitions of strategic partners and selection of profitable niches (Prescott & Visscher,1977). In contrast to the concept of path dependency, where firms start their expansion in geographically close markets, the Springboard Perspective proposes that EM MNEs expand to geographically far locations and by doing so, skipping stages of traditional expansion as described by the Uppsala Model of Johanson & Valne (1977).

As both the Uppsala Model and the Springboard Perspective, are plausible methods for internationalization dependent on the firm’s situation, resources strategy and goals. I want to examine whether firms are willing to locate further than their home region in terms in terms of geographic distance when choosing for a location in an industrial cluster. This brings us to the following proposition:

Proposition 4: MNEs locating in industrial clusters are willing to increase their resource

commitment regarding geographic distance.

Source: Forsgren & Hagestrom (2007) Figure 1: The original Uppsala Model

(21)

Entry modes 4.4

Whereas location decisions give answer on ‘where’ MNE should invest in value adding activities, entry mode decisions give answers on the question ‘how’ they should make this investment. The entry mode decision is a central aspect of the Uppsala model and the Springboard Perspective. In the case of the Uppsala model of internationalization, Johanson and Valne (1977) refer to entry mode decisions as the ‘establishment chain’. The Uppsala Model argues that entry mode is an incremental resource commitment process. In this, firms start their foreign value creation with low commitment entry modes such as exporting or contractual agreements. Once they become more familiar with their host location or environment, they consequentially will shift to more higher commitment modes (i.e. equity modes) as perceived LoF and subsequent risks or uncertainty decreases. In contrast to this reasoning, Pan & Tse (2000) argue that the decision for equity modes or non-equity modes of entry depends on industry and country specific factors.

Traditionally the choice of entry mode has been associated with the amount of commitment and knowledge about that particular host market (Pan & Tse, 2000). This proposition is also been argued in terms of uncertainty about the host location: the lower the LoF in a particular host location, the more resources will be committed in terms of high commitment entry modes such as equity entry modes. (Zaheer, 1995) (Pan & Tse, 2000) (Johanson & Valne, 1977) Furthermore, Johanson and Valne argue that firms that internationalize possess strong non Location Bound firm specific advantages (NLB FSAs) that are needed to overcome LoF.

Regarding entry modes, Pan & Tse (2000) make a distinction between non-equity modes and equity modes. Equity based entry modes relate to wholly owned subsidiaries and equity joint ventures whereas non-equity modes relate to export and contractual agreements. Pan & Tse’s (2000) hierarchical model categorizes entry modes from low resource commitment modes referring to non-equity modes) to high commitment entry modes referring to equity modes. According to Brouthers & Hennart (2007), the model of Pan & Tse (2000) can be interpreted in two ways. The first way sees non-equity and equity modes on an continuum of increasing control, commitment and risk. The second view sees two distinctive main categories: non-equity and equity. Figure 2 mentioned below provides an overview of the two distinctive main categories and their respective modes of entry:

(22)

Joint ventures as real options

International expansion implies risks and uncertainty regarding demand in a host location. This problem increases when the new business is not directly related to current activities (Kogut, 1985). A firm’s initial investment can be considered as buying the right to expand in the future (Kogut , 1991). According to Kogut (1991) MNE’s Equity Joint ventures (EJV) as stated in the model of Pan & Tse (2000), can be framed as real options. “The joint venture serves as a way to bridge these options through pooling of resources of two or more firms” (Kogut, 1985 p.240). These joint ventures give MNEs the opportunity to expand and grow in the future. As joint ventures involve two firms that bundle their FSAs into a third, new entity, exercising this option requires further commitment of capital and renegotiation among involved firms. Traditionally, joint ventures are related to risk sharing, however, they do not only share risks but also decrease total investment of both firms. Other benefits of JVs as real options are the possibility to share scale economies and the excess of capacity (Kogut, 1991). Due to these benefits, joint ventures are considered as an attractive mechanism for entry modes in host locations.

In order to benefit from these advantages Kogut (1991) considers two options: The first is that firms involved in joint ventures should wait to invest in full acquisitions as it pays off to wait before making a further commitment. The second is that it is necessary to have enough resources in order to expand in the future. Furthermore, the timing of the acquisition is critical. The acquisition is only beneficial for the buying firm when the perceived value is higher than the price of the acquisition. This especially applies to strategic assets that impact future competitiveness and competitive advantage.

Internalization (Foreign Direct Investment) 4.5

International business involves uncertainty in terms of imperfections in the markets for goods and services. “Internalization serves to determine the reason for the foreign production and sales of MNEs, namely that these take place in response to imperfections in the goods and factor markets” (Rugman, 1980 p. 365). Hence, MNEs want to bypass these imperfect markets through making use of internal markets in terms of internalization or foreign direct investment (FDI) instead of arm’s length transactions. The creation of the internal market by MNEs permit them to exploit its Firm Specific Advantages (FSAs) in host countries and will keep the use of information internal to the firm (Rugman, 1980). Foreign Direct Investment (FDI) as a general theory of the MNE (Rugman,1981) is based on the central ideas of the Coasian transaction theory (Coase, 1937), which states that firms bypass imperfect external markets by creating internal markets. This situation is typically for markets for intangible assets such as information, reputation and knowledge (Dunning, 2000) impacting the competitive advantage of the firm, and therefore strictly applies to main research question of this paper.

Through years scholars have developed various theoretical approaches to explain MNEs’ strategic investment motives. Dunning ‘s (2000) OLI tripod or Eclectic Paradigm is still considered to be the dominant paradigm that explains the pattern foreign value creation of MNEs. Dunning (2000) integrates specific motivations on firm level and country level to explain FDI in cross border activities. Dunning (2000) proposes that his OLI-paradigm consists out of three variables that act independently and determine the extent, the location and industrial composition of MNEs’ foreign value creation. Each variable of the OLI-tripod can be seen as a three legged stool that consists of its own underlying motives for a FDI in which all three variables have to be present for an FDI to exist

(23)

(Dunning, 2000): Firstly, ownership advantages, which can be divided into asset advantages referring to both tangible and intangible assets, and transactional variables referring to the advantages of a geographically widespread MNE network of subsidiaries. These ownership advantages are also referred to as competitive advantages of FDI-seeking MNEs (Dunning, 2000). According to Dunning (2000), the greater the ownership advantages in a particular host country, the more likely a MNE increases its FDI in that specific country.

Secondly, location advantages (referring to specific CSAs in terms of an abundance of production factors, special taxes, low wages and other factors that give insight in the attractiveness of host countries above the home country. In this sub paradigm ‘’the more the immobile, natural or created endowments, which firms need to use jointly with their own competitive advantages, the more firms will choose to augment or exploit their ownership advantages by engaging in Foreign Direct Investment” (Dunning, 2000 p.164).

Thirdly, internalization advantages referring to the benefits of the exploitation and creation of FSAs internally instead of through external contractors. According to Rugman & Verbeke (2004) the set of ownership-and location advantages are important determinants for MNEs’ internalization advantages.

Although the OLI-paradigm is still the dominant paradigm, it has been criticized as it does not integrate country and firm level interactions. In this context Itaki (1991) pointed out that an ownership advantage may be derived by a firm’s internalization advantages and so both cannot be seen as separate determinants. Despite the criticism regarding the redundancy in its determinants the eclectic paradigm is still a comprehensive framework for FDI.

In addition to his OLI-paradigm Dunning (1998,2000) developed four motives for FDI activity: (i) Resource seeking: The motivation for MNEs to gain access to natural resources such as commodities or unskilled labor. (ii) Market seeking: The motivation for MNEs to satisfy a particular foreign market in order to create value. (iii) Efficiency seeking: The motivation for MNEs to promote a more efficient division of labor or asset portfolio. (iV) Strategic asset seeking: The motivation for MNEs to protect its ownership advantages against its competitors. The motivation for strategic asset seeking, is becoming significantly important. This is mainly due to strategic considerations for firms to invest in intangible assets such as technical knowledge, learning experiences, managerial expertise (Dunning , 1998). This development is best demonstrated by the increasing role of mergers and acquisitions which take place through FDI. Evidence can be found in UNCTAD (1997) which states that between 55 percent and 60 percent of FDI flows over the period 1985-1995 was accounted for by these mergers and acquisitions. UNCTAD (1997) also states that most of these M&As were concentrated within North America, Europe and Japan, and in knowledge and information-intensive sectors. The rapid growth of strategic asset seeking FDI over the last decades, the increasing importance of intangible assets in the contemporary world economic scenario and growing interest for industrial clusters have their implications for MNEs’ entry modes. Therefor the following position is brought forward:

Proposition 5: MNEs locating industrial clusters are willing to choose for high commitment

entry modes.

(24)

The globalization or regionalization debate 4.6

Globalization of economic activity has been a trend for decades that, empowered by technological advancements in transport and communication technologies, has profound implications on international business. Globalization in terms of integration of markets for goods, services and capital (Garret, 2000)l can have a positive impact on firms that search for competitive advantages or new FSAs (Garret, 2000) as it makes the exchange of goods and services easier and increases the possibility of international trade (Brune & Garret, 2005). Despite these advantages integration measures do not achieve their goal of global market integration. If this was the case then IB-strategy would not differ from mainstream strategy (Ghemawat, 2003). As a result Ghemawat (2003) claims that MNEs operate in the middle of these extremes and highlights the critical role of location specificity in IB strategy and location decisions.

In globalization, global presence is argued to be highest order of internationalization giving MNEs possible competitive advantages derived from economies of scale and scope or growth opportunities (Prahalad & Doz, 1987). However recent streams in IB literature are debating whether globalization really exists by arguing for a regional approach of international expansion (Rugman & Verbeke, 2004).

Building further on the debate whether globalization or regionalization is really the contemporary IB- paradigm. Rugman & Verbeke’s (2004) empirical evidence shows that most MNEs are not global companies as they do not penetrate widespread foreign markets. Table 3 shows the results on Rugman & Verbeke’s (2004) empirical evidence. The results imply that global firms are rare and that most of them are in fact bound to their home region. The research of Rugman & Verbeke’s (2004) is based on the dispersion of sales of world’s largest MNEs across categorized geographic regions consisting of European Union, North America and Asia. The basis of this broader categorization draws on the original work of Ohmae (1986), who initially categorized the geographic regions as North America, Japan and Europe.

Source: Rugman and Verbeke (2004)

This new strand contributes to an emerging paradigm where lower levels of globalization are the new standard (Ghemawat, 2003) (Rugman & Verbeke, 2004). In this so-called ‘semi-globalization’ (Ghemawat, 2003) most MNEs’ economic activity takes place in regional dimensions instead of global dimensions in terms of geographic distribution of value creation. Rugman announced the end of the global strategy of the MNE by calling it a myth. Especially the phenomenon of ‘global impasse’ points out that many MNEs are unable to be present in all three legs at the same time (Ohmae, 1986) placing a large emphasis on ‘region’ as unit of analysis in IB literature.

(25)

Industry specificity 4.7

It is argued in IB literature that industry specificity is an another important dimension within the regionalization literature (Rugman & Verbeke, 2004). It implies that some industries are more global or regional oriented than other industries (Rugman & Oh, , 2006). For example, Li & Guisinger (1992) examined the level of regionalization and globalization in the service- and manufacturing sector. Their empirical evidence claims that factors such as cultural distance or market size impact the level of internationalization of MNEs in different ways. , Li & Guisinger (1992) found that the greater the interaction with customers, the greater the barriers for foreign expansion. A second study that illustrates the importance of sector specificity is conducted by Filipaios and Rama (2008). Their study was on the degree of internationalization within the food and beverage sector. Their empirical evidence claims that most firms within this industry operate on regional level despite the fact that these firms have enough resources to pursue a global strategy. This implies that there are greater barriers to internationalization outside the home region, than inside the home region (Filipaios & Rama, 2008) (Rugman & Verbeke, 2004).

There are also various studies that examined which industries are more industrial cluster oriented. According to Asheim & Coenen (2005), the traditional industrial cluster is nearly always to be found in the context of industries related to synthetic knowledge (i.e. engineering based industries) and analytical knowledge based industries such as (i.e. IT or R&D based industries). A practical example can be seen in the industrial cluster of Silicon Valley in California where all leading edge tech companies are situated.

Institutions 4.8

As last concept related to location, the role of institutions is taken in to consideration as it plays a dominant role in location decisions of MNEs. Institutional theory, often referred to as the institution based view of MNEs, plays an important role in cross national business contexts. Firms are embedded into institutions as they form “the set of fundamental, political, social, and legal ground rules that establishes the basis for production, exchange, and distribution” (Davis, North, & Smorodin, 1971 p. 135). However there are multiple definitions for institutions and outcomes. Firstly, the definition of Douglas North: “Institutions are the rules of the game in a society or, more formally, the humanly devised constraints that shape human interaction”. Secondly, the definition of Scott (1995) arguing that “institution are cognitive, normative, and regulative structures and activities that provide stability and meaning to social behavior” (Scott, 1995p. 124). Besides the numerous definitions of the concept of institutionalism, the concept itself is also subject to scientific contributions. In 1990 the concept of “New Institutionalism” is introduced by DiMaggio & Powell in 1991 entailing a distinctive focus on both an economic and a sociological version. The former, economic focus, has its focus on efficiency. The latter, sociological focus, has its focus on legitimacy. In a recent study of Dikova and van Witteloostuijn (2007) on new institutionalism, the foundations of the work of Williamson (1985) are used to investigate the institutional effects in on MNEs’ dual investment choices (i.e. both the mode of entry and the mode of establishment). They conclude that the degree of a host country’s institutional development in terms of progressive changes in its institutional environment moderates the effects the technological intensity in relation to these investment choices.

According to one of the founders of the institution based view, Peng (2000), all firms are exposed to institutional powers when creating value in host locations. In the field of IB literature it is widely recognized that institutions matter and that international strategy cannot simply focus on industry conditions and CSAs as the only differentiating drivers between home and host countries (Peng,

(26)

2000). Peng argues that institutions can help to reduce uncertainty for organizations as institutional frameworks give MNEs insight which ones are tolerable and viable for their activities. Cantwell, Dunning, & Lundan (2010) argue three ways to for MNEs to engage with their institutional environment. In table 4 these three types of engagement and their main features are mentioned.

Concerning the influence of industrial clusters on institutional development, Newlands (2003) argues that benefits of clusters result from cooperation between firms in the cluster. Also, institutional development will inevitably benefit all firms in the industrial cluster, placing an emphasis on collaborative processes. However empirical evidence claims of Newlands(2003) that this relationship varies across regions and countries.

Table 4: Three ways for MNEs to engage with their institutional context

(27)

5 Key themes within Resource Based View (RBV) literature

The Resource Based View (RBV) has always been an important theoretical concept impacting a firm’s strategy. The original version of Barney (1991) considers the sustainable competitive advantage as a result of a firm’s heterogeneous resources. According to the RBV, a firm has a number of resources that can be seen as “anything that could be thought of as a strength or weakness of a given firm” (Wernerfelt, 1984). These resources can be divided into tangible- and intangible assets, which, when bundled, form unique characteristics for every firm. In turn, these specific bundles of tangible- and intangible assets determine whether a firms owns a competitive advantage. In line with these findings Rugman (1981) claims that each MNE consists of an idiosyncratic set of FSAs which gives it a competitive advantage or competitive edge in relation to other firms. These FSAs are developed when a firms create special knowhow or other capabilities that cannot be replicated by other firms, except in the long run, at high costs, or both. This especially applies to a firm’s intangible assets. Ultimately, these intangible assets can reflect a firm’s capability of decreasing on transaction costs and improving on coordination and control of assets.

In this paragraph the transferability of FSAs, the knowledge based view, the role of learning capabilities and relationships are examined as they are important concepts related to RBV-literature in the light of our main research question.

The transferability of FSAs through the MNE network 5.1

By drawing further on the third stage of IB, the subsidiary as core unit of analysis, Birkinshaw (1996) has introduced the concept of subsidiary initiatives. In his research the focus is on the MNEs’ recombination capabilities of CSAs in both home and host country and the development of new FSAs by subsidiaries in these host countries. Birkinshaw’s (1996) approach builds further on Hymer’s (1960) empirical analysis which claims that firms have both location based firms specific advantages (LB FSAs), which are bounded to a location and non-location based firm specific advantages (NLB FSAs), which are not bound to a location and transferable throughout the MNE network at low marginal costs. Both Birkinshaw (1996) and Hymer (1960) recognized the importance of home country CSAs and host country CSAs. In line with these empirical findings, yet approached form a transaction cost economics perspective, Rugman & Verbeke (2009) suggest that MNEs make dual use of these CSAs in a leveraged way through recombination FSAs. Their critical view leads to two important claims. Firstly, NLB-FSAs are not necessarily created in the parent firm but may also be created by the subsidiary in a host country. Secondly, many FSAs may be perceived to be NLB-FSAs whereas they are in fact LB-FSAs that cannot be easily transferred across borders and therefore need to be recreated by the subsidiary in the host country. Rugman & Verbeke (2009) analysis on the four generic organizational types of the MNE contributed by Bartlett and Ghoshal (1989) is represented in figure 3 and is called the transnational solution.

(28)

The findings mentioned in figure 3 have still unanswered one main question: ”Where is the location of the core sources of the MNEs’ future competitive advantage?” Firstly, According to Bartlett and Ghoshal (1989) all MNEs’ operations must take a double test namely perceived contribution of FSA development for MNEs’ competitiveness and the perceived potential of CSAs for MNEs’ competitiveness. Both tests are mentioned on the axes in figure 4.

A strategic leader from quadrant three scores high on both axes whereas a contributor has high potential on FSA-development but low potential for CSAs for MNEs’ competitiveness. The opposite is true for respectively implementor-and black hole subsidiaries. Clearly international operations marked as strategic leader or contributor should be selected for the creation FSAs, both location bound and non-location bound, that can be leveraged throughout the MNE network (Rugman & Verbeke, 1992).

Referring back to the increasing importance of intangible assets, Kaplan & Norton (2004) propose that intangible assets constitute a dominant part in the contribution of the development of new FSAs in firms. This means that these intangible assets constitute a great part of the competitive advantage and distinctive capabilities of firms in relation to its rivals in an industrial cluster.

Knowledge based view of the firm

5.1.1

As proposed by Kaplan & Norton (2004) and Dunning (2009), the contemporary world economy is increasingly driven by intangible assets such as knowledge, learning and experience. This change emphasizes the importance of the knowledge based view (Grant, 1996) or intellectual capital view (de-Castro et al.,2011) of the firm. The knowledge based view in this research involves an remarkable paradox: How can internationalization of products, goods and services, particularly knowledge intensive activities be concentrated in an industrial cluster, concerning the possibility of negative knowledge spillovers? This paradox will be solved in the following three subparagraphs.

Learning and experiential knowledge

5.1.2

Experiential knowledge and its process of learning through experience are dominant factors in the Uppsala Model of internationalization (Johanson & Valne,, 1977). It not only reduces risks but also provides opportunities in doing business abroad and a vehicle for knowledge acquisition both internal and external to the MNE (Eriksson, Johanson, Majkhard, & Sharma, 1997). In this, the more experiential knowledge a MNE has gained through its operations in host environments, the lower the perceived risks of operating in other host environments and the easier it is to grasp opportunities in these foreign markets (Johanson & Valne, 1977).

Figure 4: Interpretation of the four generic roles of MNEs