Master Thesis
Business Studies – International Management (MSc.) Faculty of Economics and Business
Graduate School of Business University of Amsterdam
“Upstream and Downstream Firm-Specific Advantages (FSAs) of Multinational Enterprises (MNEs): The Relationship Between their Scope and Scale of
Internationalization and the Moderating Role of International Strategy”
30 June, 2014
Wessel Smit – 10447245 1st Supervisor: Dr. Niccolò Pisani 2nd Supervisor: Prof. Dr. Suzana B. Rodrigues
Abstract
Scholars have often argued that an asymmetry may exist between internationalization patterns of upstream and downstream FSAs (e.g. Rugman, Verbeke & Nguyen, 2011). However, research on the relationship between the internationalization of FSAs along different parts of the value chain remains
conspicuously absent. In an effort to fill this academic gap, this research has initiated the endeavor to investigate the relationship between the internationalization of scope and scale of upstream and downstream FSAs among Global Fortune 500
companies. Additionally, this research investigates the extend to which this relationship is moderated by international strategy (i.e. strategic focus on local responsiveness). The findings of this research demonstrate that upstream FSAs are less location bound than downstream FSAs. It is found that the relationship between the scale of internationalization of FSAs on opposite ends of the value chain is positive and nonlinear, with the slope greater at low levels of scale of
internationalization for upstream FSAs. On the contrary, we found that the
relationship between the scope of internationalization of upstream and downstream FSAs is linear, suggesting that the internationalization of FSAs shows different patterns according to whether internationalization takes place inside or outside the home-region of the firm. The present research has proven to be of academic relevance by shedding light on the asymmetry of, and relationship between, the internationalization patterns of upstream and downstream FSAs.
Keywords:
internationalization, globalization, regionalization, home-region orientation, semi-globalization, firm-specific advantages (FSAs), multinational enterprises (MNEs), value chain, upstream, downstream, international strategy, local responsiveness
Acknowledgements
First and foremost, I would like to express my sincere gratitude to my thesis supervisor, Dr. Niccolò Pisani, Assistant Professor of International Management at the University of Amsterdam, for offering guidance and excellent feedback
Table of Contents
Introduction 7
Literature Review 10
Internationalization 10
Internationalization Defined 10
Scope and Scale of Internationalization 10
Scope of Internationalization 10
Scale of Internationalization 12
Upstream versus Downstream Firm-Specific Advantages (FSAs) 14 Scope and Scale of Firm-Specific Advantages (FSAs) 14 Internationalization of Firm-Specific Advantages (FSAs) 16
International Strategy 17
Research Gap 19
Conceptual Framework 20
Internationalization of Upstream and Downstream FSAs 21
Strategic Focus 22
Methodology 24
Sample and Data Collection 24
Sample 24 Data Collection 24 Measures 25 Dependent Variables 25 Independent Variables 25 Moderator Variables 26 Control Variables 26
Statistical Analysis and Results 27
Statistical Analysis 27
Results 29
Discussion 36
Academic Relevance 37
Limitations and Suggestions for Future Research 38
Conclusion 39
List of Tables and Figures
Figure 1: Conceptual Framework 20
Figure 2: Scatterplot - Scope of Downstream FSAs vs. Scope of Upstream FSAs 32 Figure 3: Scatterplot - Scale of Downstream FSAs vs. Scale of Upstream FSAs 32
Table 1: Means, Standard Deviations and Correlations 33 Table 2: Regression Results - Hypothesis 1 and Hypothesis 3 34 Table 3: Regression Results - Hypothesis 2 and Hypothesis 4 35
1. Introduction
In the past two decades some scholars have postulated that we are moving towards a state of globalization - a fully integrated market in which barriers are
becoming less significant and economies are increasingly interdependent (e.g. Levitt, 1983; Ohmae, 1989; Kobrin, 1991; Birkinshaw, Morrison & Hulland, 1995). Although scholars have come to different conclusions relative to the state and rate of
globalization (e.g. Ghemawat, 2003), it is commonly argued that the ongoing (trade) liberalization and technological advancements (i.e. communication, transport) are among the main external contributors to the emergent globalization process
(Ghoshal, 1987; Yip, 1992; Garrett, 2000). Therefore, in an effort to exploit synergies and arbitrage across boundaries, firms have often utilized global strategies (i.e. standardized products and processes) (Levitt, 1983; Ghemawat, 2005).
More recently, however, a new stream of research has emerged in which scholars put into perspective the globalization construct and claim that markets are in fact internationalizing and strategizing regionally rather than globally (Rugman, 2000; Rugman & Verbeke, 2004). Referred to as regionalization literature, this stream of research has empirically demonstrated that firms are generally home-region oriented as opposed to globally oriented (Rugman & Verbeke, 2004; Rugman, 2005; Rugman & Girod, 2003).
The recent emergence of the regionalization construct has led to a current theoretical state that is rather inconsistent and in which many new concepts have emerged over the past decade, with some scholars claiming that regionalization is a stepping stone towards globalization and others claiming regionalization is a post-globalization phenomenon (Stevens & Bird, 2004; Rugman & Verbeke, 2004, 2008; Benito, Grøgaard & Narula, 2003). However, despite the contradictory scholarly
literature, scholars generally agree that for multinational enterprises (MNEs) to operate across borders they are ultimately dependent on the utilization of their competitive advantages and their understanding of environmental factors impacting strategy and structure (i.e. local, regional, global) (Rugman, 1981; Rugman & Verbeke, 1992).
Research on internationalization has traditionally emphasized the importance of firm-specific advantages (FSAs), which can be distinguished at different levels along the value chain, as those factors that determine a MNEs’ competitive advantage (Vernon, 1979; Rugman, 2005; Cerrato, 2009). Customer-end, or downstream FSAs occur in activities with a direct interface with the customers and include, for example, product quality, product scope, reputation and brand name. Back-end, or upstream, FSAs are found in the activities associated with a firm’s key activities, or as Hu (1995: 75) states, “with what the firm is, what it does, and what it has - i.e. its superior attributes, activities, assets, skills, and internal and external relationships”. Furthermore, upstream FSAs are found in the ability and willingness to continuously learn from, and improve, internal/external sources and processes (Hu, 1995).
In terms of geographic dispersion, it is argued that an asymmetry may exist between upstream and downstream FSAs (Rugman & Verbeke, 2004). Some scholars have argued that achieving global distribution of upstream FSAs may
potentially be easier than achieving global distribution of downstream FSAs (Rugman & Verbeke, 2004; Rugman, Verbeke & Nguyen, 2011). However, results remain largely inconclusive with other scholars claiming the existence of different internationalization patterns (Muller & Van Tulder, 2005; Oh, 2009). Collinson & Rugman (2008), for example, illustrate that both upstream and downstream FSAs
have a limited geographic scope, consistent with theories suggesting that firms inevitably face a liability of foreignness (LOF) when expanding abroad (e.g. Zaheer, 1995).
Therefore, in an effort to contribute to internationalization research and to increase our understanding of the rationale that has made way for emergent
regionalization patterns, this thesis extends the work of Rugman & Verbeke (2004), Rugman & Sukpanich (2006) and Collinson & Rugman (2008) by theoretically framing and empirically testing the relationship between the scope and scale of upstream and downstream FSAs among Fortune Global 500 companies.
In addition, as it is traditionally argued that firms often face conflicting
environmental and strategic pressures (i.e. standardization, customization) (Prahalad & Doz, 1987; Bartlett & Ghoshal, 1989), we hypothesize that international strategy (i.e. strategic focus on local responsiveness) moderates the relationship between the internationalization of upstream and downstream FSAs.
The remainder of the paper is structured as follows. First, extant literature on internationalization, and more specifically, regionalization/home-region orientation is reviewed. Second, a conceptual framework explaining the relationship between the scope and scale of internationalization for upstream and downstream FSAs is presented, followed by a description of the research methods and procedures used to test the hypothesized relationships. Next, the findings of the study are presented. Finally, the paper concludes with a discussion and limitations section.
2. Literature Review
2.1 Internationalization
2.1.1 Internationalization Defined
Internationalization is generally described as the outward (international) movement of (part of) a firm’s operations. Nevertheless, scholars have not yet unanimously agreed on the definition of the internationalization concept (Andersen, 1997). Some scholars define internationalization as an incremental and orderly
process of increased international involvement (e.g. Johanson & Vahlne, 1977, 1990; Reid, 1981). Calof & Beamish (1995: 16) define internationalization as “the process of adapting firms’ operations to international environments”. Others, have broadened this definition by including both the inward and outward side of the process (e.g. Welch & Luostarinen, 1993). Lastly, Ahokangas (1998: 63) adds to the debate by defining internationalization as “the process of mobilizing, accumulating and developing resource stocks for international activities”.
2.2 Scope and Scale of Internationalization 2.2.1 Scope of Internationalization
In an effort to explain the internationalization process, several theories have been developed over the years, most of which can be traced back to industrial
organization and economics (Axinn & Matthyssens, 2002). In spite of much scholarly attention, the internationalization process remains a phenomenon that still calls for additional research (Kuivalainen & Sundqvist, 2006; Rugman & Verbeke, 2005). Rugman & Verbeke (2005), for example, argue that many scholars have made a quantum leap, equating internationalization with globalization. It is argued, therefore, that existing theories on the internationalization process need to be
Consequently, there has been a hefty scholarly debate with regard to
internationalization, with an increasing number of scholars questioning the traditional globalization claims made in earlier research (e.g. Levitt, 1983; Baden-Fuller & Stopford, 1991). Scholars in this field of research argue that globalization is a myth and that we are instead in a state in between localization and globalization referred to as semi-globalization, regionalization or home-region orientation (HRO), a state in which “we observe neither extreme geographical fragmentation of the world in national markets nor complete integration” (Rugman & Verbeke, 2004: 6, 2008; Ghemawat, 2003; Rugman, 2000). Additionally, Rugman & Verbeke (2008: 398) claim “the home-region orientation of most firms implies that the reality of
globalization has been vastly exaggerated”.
Propelled by these provocative insights, recent studies have started to explore the nuances of these claims. Studies in this field of research are mainly concerned with the scope of MNEs’ key activities, which generally denotes the international geographic reach of a MNEs’ business (George, Wiklund & Zahra, 2002), and therefore primarily focuses on region, country and firm-level effects (e.g. Arregle, Miller, Hitt & Beamish, 2013; Banalieva & Dhanaraj, 2013; Rugman & Sukpanich, 2006; Dunning, Fujita & Yakova, 2007). Nevertheless, a systematic theory explaining the observed regionalization patterns among MNEs remains conspicuously absent (Banalieva & Dhanaraj, 2013). As is noted by Osegowitsch & Sammartino (2007: 46): “scholars need to urgently confront the question why, in an age of purported
globalization, many of the world’s largest firms have barely ventured beyond the confines of their home region”.
Historically, Rugman & Verbeke (2004) were among the first scholars to empirically demonstrate the regional nature of most of the Fortune Global 500
companies across the ‘triad’ regions of Europe, North America and Asia-Pacific. In their iconoclastic study, Rugman & Verbeke (2004) showed that an average of 80.3 percent of MNEs’ sales are made in their home-region. Subsequently, Ghemawat (2007) found (longitudinal) support for these findings by demonstrating that intra-regional trade significantly increased over a period of over 40 years. In a similar vein, Ghemawat (2005) argued that the exorbitant costs of operating across cultural, administrative, geographic and economic distances are among the main causes of the current state of ‘semi-globalization’.
Thus far, scholarly literature on regional strategies has generally been treated as being halfway up the continuum towards globalization (Rugman & Verbeke, 2008). Also, explanations of the regionalization phenomenon have generally been based on normative observations and thus regionalization was often considered a natural phenomenon with scholars generally arguing that the occurrence of regionalization patterns was due to the emergence of regional trading blocs (Ghemawat, 2005). Furthermore, globalization was argued to be restricted by governmental institutions not being properly equipped to handle global trade (Emmerij, 1992). However, after Rugman (2000) and Rugman & Verbeke (2004) empirically demonstrated that most economic interactions take place within rather than between the triad regions of Europe, North America and Asia-Pacific, regionalization theory gained momentum and became recognized as an interesting and legitimate field of research among IB scholars (Oh, 2009).
2.2.2 Scale of Internationalization
Apart from the prevalent home-region orientation relative to the geographic scope of international activities, firms are often found to have a limited scale of internationalization, which is referred to as the extent to which firms rely on foreign
markets in their operations. Past research on internationalization theory has
generally focused on the concept of liability of foreignness (LOF), a context-specific variable suggesting that in order for firms to successfully internationalize they need to overcome the inherited disadvantage of being unfamiliar host-country specific factors (Hymer, 1960; Zaheer, 1995).
In an effort to offset such liability of foreignness firms are ultimately dependent on their possession of FSAs, those factors that determine MNEs’ competitive
advantage (e.g. process technology, managerial capabilities, reputation) (Cerrato, 2009; Hymer, 1960; Zaheer, 1995). However, (structural) market imperfections as a result of government intervention may exist, making it difficult to capitalize on these advantages (e.g. monopolies, entry barriers) (Dunning & Rugman, 1985). The existence of various forms of distance (cultural, administrative, geographic, economic) arguably explains why many firms have difficulties operating across borders and thus have a relatively limited geographic scale, especially when facing competition not hindered by such distance (i.e. local firms) (Ghemawat, 2001).
Johanson & Vahlne (1977) claim that there are stages of internationalization. The Uppsala model suggests that the benefits of operating across boundaries need to be weighted against the potential risks. Consequently, firms internationalize incrementally and initially venture to countries in nearby geographic regions with similar country-specific advantages (CSAs), and thus relatively low cultural,
administrative, geographic and economic distances. As a firm learns to overcome the liability of foreignness, it incrementally (step-by-step) expands to more distant
countries. It is argued, therefore, that in achieving international success and thus an increased geographic scale, firms are dependent on their ability to combine FSAs with CSAs. However, as CSAs are not always freely accessible, Hennart (2009)
claims that in case resources are controlled by host-country actors (e.g.
governments), the MNE is dependent on the development of relationships with powerful local actors to open up access to these resources.
2.3 Upstream versus Downstream Firm-Specific Advantages (FSAs) 2.3.1 Scope and Scale of Firm-Specific Advantages (FSAs)
In terms of FSAs’ geographic scope, scholars have argued that it is important to make a distinction between upstream and downstream activities along the value chain instead of merely looking at the organization as such (Rugman, 2005). Rugman (2005) states that it is possible for activities at the back-end of the value chain to be more globalized than activities that occur at the front-end. More
specifically, Rugman (2005) claims that managers need to urgently recognize that some parts of the value chain (customer-end) may function regionally, whereas other parts (back-end) may function globally. It is important, therefore, to recognize that activities along the value chain can have different levels of local responsiveness and/or global integration (Rugman & Verbeke, 2008).
Over the past few years several theories have been pushed forward in trying to explain the asymmetry between the scope of upstream and downstream FSAs. First, Rugman & Verbeke (2004) argued that activities that occur on the back-end of the value chain (upstream) are less location bound than those that occur on the customer-end (downstream). Besides, Rugman & Verbeke (2004) have argued that costs incurred in the development of downstream FSAs are one-sided, meaning that there is a large degree of uncertainty as to whether these investments lead to
commitment from the customer, whereas the costs incurred at the back-end (upstream) are relatively many-sided (i.e. spread out over multiple actors). Both these arguments suggest that upstream activities are generally more geographically
dispersed than their downstream counterparts. More recently, however, Rugman & Verbeke (2005) stressed the importance of home-region production networks on the upstream side of the value chain. Similarly, Muller & Van Tulder (2005) argue that the region-boundedness of upstream activities centers around the importance of those activities to home-region stakeholders, also referred to as ‘stickiness’. Also, Muller & Van Tulder (2005) claim that the many-sidedness of upstream activities constrains firms in their ability to relocate those activities beyond the confinements of their home-region.
Thus far, one of the key arguments for the overwhelming representation of home-region oriented firms is the inability to transfer FSAs beyond the home-region, also referred to as the home-region boundedness of FSAs (Rugman & Verbeke, 2007). In internationalization literature, it is generally argued that a firm’s degree of multinationality is dependent on its ability to integrate FSAs with CSAs (Rugman, 1981; Rugman & Verbeke, 1992). More specifically, it is suggested that
internationalization is dependent on a firms’ non-location bound FSAs (NLB FSAs), those advantages that can be exploited globally (Rugman & Verbeke, 1992).
Scholars generally distinguish between two types of FSAs, non-location bound FSAs (NLB FSAs) and location-bound FSAs (LB FSAs). NLB FSAs can be globally
exploited at relatively low marginal costs without substantial adaptation and lead to benefits of scale and scope. In contrast, LB FSAs are context dependent and only benefit the firm in a particular location and thus lead to benefits of national
responsiveness (Rugman & Verbeke, 1992). Scholars have generally argued that the international scale of FSAs is to a large extent dependent on the interaction and relative strengths and weaknesses of these two factors (Rugman & Oh, 2008).
Based on the idea that regions are the critical unit of analysis in the formation of international strategies, the notion of liability of foreignness (LOF), which has traditionally been conceptualized in the context of the entry into a specific country, has recently been rethought at a regional level, suggesting that the liability of
foreignness firms face outside their home-region is significantly higher than the one they face inside their home-region (e.g. Rugman & Brain, 2003; Cerrato, 2009;
Rugman & Verbeke, 2004). Rugman & Verbeke (2004) also claim that, in transferring FSAs across national boundaries, barriers are greater for downstream FSAs than for upstream activities, as these are argued to be less location-bound. Additionally, it is stated that “…there are insurmountable difficulties in moving and replicating a highly embedded capability without moving critical parts of the context in which the
capability has been developed” (Forsgren, 2008: 58). Therefore, in order for us to better understand the internationalization of FSAs, Rugman & Verbeke (2004) suggest that FSAs need to be broken down into upstream and downstream components.
2.3.3 Internationalization of Firm-Specific Advantages (FSAs)
Previous research on the internationalization of FSAs has shown that there exists an asymmetry between both the scope and scale of upstream and
downstream FSAs (e.g. Oh, 2009). It is argued that in large companies, upstream FSAs are likely to develop faster than their downstream counterparts (Rugman & Oh, 2007). The asymmetry between upstream and downstream FSAs arguably explains why many firms have dispersed assets, but have simultaneously been incapable of capitalizing on these FSAs at the customer-end (Rugman & Verbeke, 2004, 2008). However, evidence remains largely inconclusive. Oh (2009), for example,
upstream and downstream FSAs. Additionally, in their study on the
internationalization of Japanese MNEs, Collinson & Rugman (2008) find that almost none of the 64 companies investigated have upstream FSAs that extend beyond the home region. Rugman & Verbeke (2004) argue that international integration is dependent on internalization arbitrage, which is referred to as a firm’s ability to combine upstream FSAs with host countries’ locational advantages. Scholars have argued that in order for an MNE to integrate its operations on a regional or global basis, it should balance between sales and assets (Oh & Rugman, 2012). However, MNEs are believed to differ in their ability to decouple upstream and downstream FSAs (Rugman & Verbeke, 2008). Verbeke & Kano (2012), therefore, argue that it is the ability of a firm to decouple downstream and upstream FSAs that makes that service organizations are generally more home-region oriented than manufacturing organizations.
2.4 International Strategy
Scholars have generally agreed that both scope and scale of
internationalization are impacted by international strategy and related strategic decisions (e.g. Harzing, 2002; Cantwell, Dunning & Lundan, 2010). Traditionally, most scholarly work on international strategy underpins the conceptualization of the integration-responsiveness dichotomy (trade-off), suggesting that firms encounter (often conflicting) environmental pressures towards either local responsiveness or global integration (Prahalad & Doz, 1987; Bartlett & Ghoshal, 1989). Pressures for global integration may prompt responses such as global distribution of standardized products or mandating a uniform global marketing strategy. On the contrary,
pressures for local responsiveness might lead to a firm adapting its marketing strategy to local market conditions (Venaik, Midgley & Devinney, 2004). But, in
focusing on the extreme ends of the prevalent dichotomy, firms face the danger of either over-standardization (products/services) or over-centralization
(processes/organization) (Schlie & Yip, 2000). The sharp tradeoff nature of this model and the desire to achieve both global integration and local responsiveness resulted in the emergence of a multifocal strategy in which firms make a compromise and attempt to find a balance between the conflicting demands of the two
approaches (Prahalad & Doz, 1987). Accordingly, firms may benefit from developing regional strategies in addition to an overall global strategy (Schlie & Yip, 2000). In trying to establish a system that will effectively accommodate both customization and standardization - a daunting challenge - regional strategies can potentially pose a viable balanced compromise vis-à-vis global integration or local integration (Morrison & Roth, 1992). Regional strategies involve clustering countries (regional systems) that exhibit similar characteristics (market conditions) and addressing homogenized regional demand and thereby minimizing customization costs (Schlie & Yip, 2000). Bartlett & Ghoshal (1989) extended the integration-responsiveness framework by adding a third option, arguing that firms could transcend the tradeoff and become so-called transnationals. A transnational strategy seeks to simultaneously achieve high, as opposed to moderate, levels of both customization and standardization and by doing so attempts to overcome the dichotomy. In a highly competitive environment where firms face pressures to both standardize and customize a transnational
strategy is arguably most effective (Bartlett & Ghoshal, 1989). Yet, according to Ghemawat (2008), a transnational merely adds organizational complexity. Also, Ghemawat (2008) claims that from a single-country perspective, a firm is better of focusing on either global integration or local responsiveness, than on a hybrid strategy. The importance of international knowledge transfers, frequently cited in
proposed ‘third options’ (i.e. transnational, multifocal, metanational, differentiated network) is never really justified and looks questionable at best (Ghemawat, 2008). The lack of mutual exclusivity in a transnational strategy leads managers to believe that both ends of the framework are equally important, resulting in limited strategic impact (Ghemawat, 2008).
2.5 Research Gap
As a result of the foregoing literature review, the relationship between the scope and scale of upstream and downstream FSAs still calls for additional research. Moreover, no research has been conducted on the possible moderating effect of international strategy on the association between the scope and scale of
internationalization of upstream and downstream FSAs. Scholars have argued that the recently discovered home-region orientation among the world’s largest firms calls for further research on the internationalization of upstream and downstream FSAs (Rugman & Verbeke, 2007; Qian, Li, Li & Qian, 2008). In an effort to address such questions and contribute to the internationalization debate, this paper attempts to explore the following questions:
“What is the Relationship Between the Scope and Scale of Internationalization for Upstream and Downstream Firm-Specific Advantages (FSAs)? How is this
H1 + H2 +
H3+ H4 +
3. Conceptual Framework
Our theory builds on two core constructs: internationalization of upstream and downstream FSAs and strategic focus. We theorize a positive (nonlinear) relationship between the scope and scale of internationalization for upstream and downstream FSAs, with the slope greater at low levels of scope and scale of internationalization for upstream FSAs. This implies that MNEs, at the start of the internationalization process, are highly dependent on the utilization of their less location-bound upstream FSAs (assets), which will ultimately be caught up by, and result in, the creation of downstream FSAs (sales). In addition, we theorize that a strategy based on
achieving local responsiveness positively moderates the relationship between both the scope and scale of internationalization for upstream and downstream FSAs. Figure 1 synthesizes our conceptual model:
Figure 1: Conceptual Framework Independent:
Scope (H1) and Scale (H2) of Internationalization
of Upstream FSAs
Dependent: Scope (H1) and Scale (H2) of Internationalization
of Downstream FSAs
Moderator: Strategic Focus on Local
Responsiveness (H3 & H4)
3.1 Internationalization of Upstream and Downstream FSAs
According to Rugman & Verbeke (2008) firms can only penetrate foreign markets if they can overcome the disadvantage of being unfamiliar with local
conditions and successfully utilize their non-location bound firm-specific advantages (NLB FSAs). NLB FSAs are generally referred to as those advantages that can be exploited globally and either leads to benefits of scope/scale or exploitation of locational differences (Rugman & Verbeke, 1992). It is argued that activities that occur on the back-end (upstream) of the value chain are less location bound than those that occur on the customer-end (downstream) (Rugman & Verbeke, 2004). As a result, Rugman (2005) states that it is possible for upstream activities to be more globalized than downstream activities.
However, redeployability of such NLB FSAs seems limited given the regional nature of most MNEs (Rugman & Verbeke, 2004; Ghemawat, 2003). As mentioned before, one of the key arguments for the overwhelming representation of home-region oriented firms is the inability to transfer FSAs beyond the home-home-region, also referred to as the home-region boundedness of FSAs (Rugman & Verbeke, 2007). Instead of developing NLB FSAs, it seems firms are increasingly developing region-bound FSAs (RFSAs) (Arregle et al., 2013). Also, MNEs face different levels of liability of foreignness in different activities along the value chain. For example, Rugman (2005) claims that downstream FSAs face a higher liability of foreignness than upstream FSAs as a result of the learning pace associated with the utilization of such FSAs.
In this study, therefore, we claim that upstream FSAs initially internationalize faster than downstream FSAs, as upstream FSAs are relatively less context
time passes, we expect that through a process of organizational adaptation, upstream FSAs will facilitate the creation of downstream FSAs. However, as
organizational adaptation presumably occurs at a decreasing rate and the marginal contribution of upstream FSAs has a lower impact over time, we expect the
relationship to be non-linear with the slope greater at low levels of scope and scale for upstream FSAs. Thus, we hypothesize:
Hypothesis 1: The relationship between the scope of internationalization for upstream and downstream FSAs is positive and nonlinear, with the slope greater at low levels
of scope of internationalization for upstream FSAs.
Hypothesis 2: The relationship between the scale of internationalization for upstream and downstream FSAs is positive and nonlinear, with the slope greater at low levels
of scale of internationalization for upstream FSAs.
3.2 Strategic Focus
Traditionally, most research in the field of international business (IB) emphasizes the importance of two building blocks of international strategy, firm-specific advantages (FSAs) and country-firm-specific advantages (CSAs). A firm’s degree of multinationality is ultimately dependent on its ability to integrate FSAs with CSAs (Rugman, 1981; Rugman & Verbeke, 1992). Consequently, geographic expansion generally requires location-specific linking investments determined by, for example, local institutional environment (Kolk, 2010). The most widely accepted approaches in dealing with integration, or lack thereof, are global integration and local
responsiveness strategies (Prahalad & Doz, 1987; Bartlett & Ghoshal, 1989). A global integration strategy attempts to build efficient operations networks by taking maximum advantage of similarities across locations. In contrast, a local
responsiveness strategy is concerned with the attempt to respond to different local needs across locations (Luo, 2001). The essence of international strategy is the response to or management of pressures towards these two imperatives (Roth & Morrison, 1990).
Since a local responsiveness strategy is built around flexibility and adjustment to changing (local) market conditions (e.g. consumer needs, government policies, competitive landscape) we expect that a strategy focused on local responsiveness allows for faster and more efficient utilization of upstream FSAs. Thus, we expect that in firms focused on a local responsiveness strategy, upstream FSAs supposedly lead to downstream FSAs more rapidly than in firms focused on a global integration
strategy. In this study, therefore, we theorize that an international strategy focused on a local responsiveness positively moderates the relationship between the scope and scale of internationalization for both upstream and downstream FSAs. Therefore, we hypothesize:
Hypothesis 3: A focus on a strategy based on achieving local responsiveness positively moderates the relationship between the scope of internationalization for
upstream and downstream FSAs.
Hypothesis 4: A focus on a strategy based on achieving local responsiveness positively moderates the relationship between the scale of internationalization for
4. Methodology
4.1 Sample and Data Collection 4.1.1 Sample
The hypotheses developed in the conceptual framework section will be tested using a sample of 500 companies listed in the 2013 Fortune Global 500. The Fortune Global 500 list is ranked according to firms’ revenues and thus includes the world’s largest firms. Prior research has shown that firm size is an important determinant with regard to a firm’s ability to internationalize and overcome the costs of crossing
borders (i.e. liability of foreignness) (Buckley & Casson, 1976; Kumar, 1984). Therefore, firms listed in the Fortune Global 500 are most likely to have (partly) internationalized both upstream and downstream activities (Rugman, 2005).
In total, the sample includes companies from 38 countries distributed over 6 continents (Europe, 30%; North America, 29.2%; Asia, 37.2%; South America, 0.018%; Oceania, 0.016%; Africa, 0.002%). The largest number of Fortune Global 500 companies is found in the United States (132 companies; 26.4%), followed by China (84 companies; 16.8%), Japan (61 companies; 12.2%), France (30
companies; 6%) and Germany (28 companies; 5.6%).
4.1.2 Data Collection
Data on the dependent, independent, moderator and control variables will be (hand) drawn from company annual reports (fiscal year 2012) and supplemented with data from Bureau van Dijk’s (BvD) ORBIS database.
4.2 Measures
4.2.1 Dependent Variables
Following prior research, a firm’s scale of downstream FSAs will be measured with the ratio of foreign sales to total sales (FS/TS). In addition, based on
regionalization literature, the scope of downstream FSAs is measured with the ratio of regional (excl. domestic) sales to total sales (RS/TS) (e.g. Delios & Beamish, 2005; Rugman & Verbeke, 2004; 2008). In this study we will deliberately exclude domestic sales (and assets) from our scope measure(s), as Asmussen (2009) argues that Rugman & Verbeke (2004), by aggregating home-country and home-region sales, may be confusing between a home-country and home-region focus among MNEs.
4.2.2 Independent Variables
In accordance with Rugman, Oh & Lim (2012), Collinson & Rugman (2008), Oh (2009) and Oh & Rugman (2012) we will use the single most representative and uniformly reported measure of upstream FSAs (assets) and calculate the geographic scale of firm competitiveness (outcome measure of upstream FSAs) by dividing foreign assets by total assets (FA/TA). In addition, we will calculate the degree to which assets are present in the MNEs’ home region, or the geographic scope of upstream FSAs, by dividing home-region (excl. domestic) assets by total assets (RA/TA). In this study we will use data on asset distribution instead of corporate-level resources (e.g. marketing intensity, R&D intensity, managerial capability) as these measures cannot be assigned to a specific geographic region. In addition, it is argued that these input measures (resources) are generally developed at corporate headquarters (e.g. Hennart, 2007). Also, it is impossible to assign assets to a
particular internationalization motive, making it is the single most representative and uniformly reported measure of upstream FSAs (Sammartino & Osegowitsch, 2013).
4.2.3 Moderator Variables
As it is traditionally argued that firms often face conflicting environmental pressures toward either a local responsiveness strategy or a global integration strategy (Prahalad & Doz, 1987; Bartlett & Ghoshal, 1989), we have included a moderator variable to test if, and to what degree, a strategy focused on local
responsiveness moderates the hypothesized relationships between the scale/scope of upstream and downstream FSAs.
Following prior scholarly research, the moderating effect of international strategy (local responsiveness) will be proxied by advertising intensity
(advertising/sales) (e.g. Franko, 1989). Theoretically, the use of this specific local responsiveness proxy makes sense as it is generally argued that a local
responsiveness strategy involves relatively high marketing expenses, since local preferences, beliefs and cognitive structures differ from one location to another (i.e. there is no ‘one-size fits all’) (Takeuchi & Porter, 1986).
4.2.4 Control Variables
In this study we will control for a wide range of alternative explanations to the observed phenomena, including industry type (0 = primary sector, 1 = secondary sector, 3 = tertiary sector), firm size (number of employees), firm age (year of
analysis minus year of establishment), company ownership (0 = privately-owned, 1 = government-owned), continent of origin (0 = Europe, 1 = North America, 2 = Asia, 3 = South America, 4 = Oceania, 5 = Africa) and profitability (profitability/total sales).
In order to increase the reliability of the regression analysis, all non-binary variables (industry type and continent of origin) have been recoded into mutually exclusive categories (dummy variables) to indicate the absence or presence of a categorical effect.
4.3 Statistical Analysis and Results 4.3.1 Statistical Analysis
Table 1 presents the descriptive statistics and correlations of all the control, independent and moderator variables used in this research. Correlations between non-dependent variables upwards of 0.7 are generally considered to be too strong and therefore problematic for the validity of the research (e.g. Field, 2009). It is found that our sample shows some minor issues in terms of multicollinearity as one of the control variables has a correlation coefficient of 0.727, indicating a relatively high shared variability between the secondary and tertiary industry control variables. Therefore, in order to retain the validity of the research, the control variable ‘secondary sector’ has been excluded from the regression analysis. None of the other (non-quadratic) variables used in this research show any issues in terms of multicollinearity as all the bivariate correlations are below 0.7,which means that the amount of variability that is shared between any two (non-dependent) variables in this research is admissible, with the highest correlation coefficient being 0.543 (Scale of Upstream FSAs versus Scope of Upstream FSAs).
Additionally, Table 1 illustrates that the average company age highly deviates around the mean score of 61.22 years. Much like the average company age, the average company size (measured in number of employees) has a high standard deviation, with the average being 120.221 employees per company. Also, it is shown that average profit to sales ratio is 0.061, the vast majority (82 percent) of Global
Fortune 500 companies are privately owned and that most of the companies are operating in either secondary or tertiary sectors of the economy. Of those companies for which it was possible to operationalize a local responsiveness variable (N=152), the average marketing to sales ratio is 0.038. Furthermore, an average of 44.6 percent of Global Fortune 500 assets are situated outside the home country, 22.7 percent of which are found within the home-region of the company.
In a preliminary effort to explore the nature of the relationship that exists between the dependent and independent variables (hypothesis 1 and hypothesis 2) we have constructed two scatterplots (not controlling for alternative explanations). Figure 1, which shows the relationship between the scope of downstream FSAs and the scope of upstream FSAs (hypothesis 1), has a R2 of 0.457 and the shape of the regression line follows the equation: y = 0,07+1,13*x+-0,77*x^2. In addition, Figure 2, which illustrates the relationship between scale of downstream FSAs and scale of upstream FSAs (hypothesis 2), has a R2 of 0.621 and the regression line is shaped according to the following equation: y = 0,13+1*x+-0,24*x^2.
Afterwards, all continuous variables used in this research were standardized to equalize their range and data variability. This process has been undertaken to prevent variables measured at different scales from unequally contributing to the analysis. After the data has been standardized, all continuous variables have a mean value of 0 and a standard deviation of 1.
The hypothesized quadratic (non-linear) relationships were subsequently tested by performing a curvilinear regression analysis, as this test, contrary to the often-used curve estimator, allows for an assessment of the incremental predictive capacity by including a non-linear term (quadratic) in the regression equation.
First a base model (Model 1a,b) was developed, which only includes the
dependent variable and the control variables. The second model (Model 2a,b) tests for a linear relationship (straight line) between the dependent variable and the
independent variable. Model 3a,b, subsequently, tests whether Model 2a,b significantly improves by adding a quadratic (independent) variable that represents the quadratic function (Y' = a + b1X1 + b2X12), or a single bend in the regression line. In Model 4a,b we have added the moderator variable to see whether or not Model 3a,b significantly improves, and thus, whether or not a strategic focus on local responsiveness
provides any additional explanatory power for the relationship between the
dependent variable and the independent variable. Finally, Model 5a,b represents the interaction term (moderating variable x independent variable) and tests whether the slope of the regression line significantly changes at different values of moderation.
The regression results, presented in Table 2 and Table 3, are divided over two tables, since hypothesis 1 and 3, and hypothesis 2 and 4, have a different dependent variable. As can be noted, some dummy variables have been excluded from the regression analysis as they were constants or had missing correlations, resulting in a final sample size of N=46 for hypothesis 1 and 3, and N=89 for hypothesis 2 and 4.
4.3.2 Results
The regression results presented in Table 2 and Table 3 provide us with two values that are of critical importance to the research. First, the significance (*,**,***,†) shows whether the findings are reliable and thus whether or not we can accept the hypothesized relationship and if we can do so, at what specific significance or alpha (α) level. Second, the standardized beta (β) coefficient shows the direction of the relation, with a positive beta signaling that an increase in the independent variable will result in an increased probability of a 1-unit change in the dependent variable, or,
in other words, it refers to the rate of change of the dependent variable with respect to a change in the independent variable.
Model 2a shows that adding the scope of upstream FSAs to the control variables adds significant explanatory power and is significant at an α-level of 0.05. This implies that a positive linear relationship significantly explains the relationship between scope of downstream FSAs and scope of upstream FSAs (controlling for alternative explanations). In Model 3a, we test whether adding a quadratic variable, representing a single bend in the regression line, significantly improves Model 2a, and thus, whether hypothesis 1 is supported. It is found that Model 3a (non-linear) does not provide significantly more explanatory power to Model 2a (linear). The hypothesized quadratic effect (hypothesis 1) is therefore not supported. Model 4a tests whether a strategic focus on local responsiveness has a significant impact on the hypothesized relationship between scope of downstream FSAs and upstream FSAs. It is found that strategic focus on local responsiveness only provides little (non-significant) additional explanatory power and thus hypothesis 4 is also not supported. Like Model 4a, Model 5a (interaction term) is non-significant and provides little additional explanatory power over Model 4a.
Model 1b illustrates that much of the variability in scale of downstream FSAs is explained by the control variables identified in this research (e.g. ownership, origin and industry). Model 2b shows that a linear relationship between scale of
downstream FSAs and scale of upstream FSAs is significant at an α-level of 0.01. Model 3b, like Model 3a, tests whether adding a quadratic (non-linear) variable significantly improves Model 2b. In Table 3, it is shown that Model 3b does improve Model 2b but only at an α-level of 0.10, hypothesis 3, therefore, is (partially)
coefficient (β) for the squared term (-0.115) shown in Table 3 signals that (in line with our expectations) the slope is greater at low levels of scale of internationalization for upstream FSAs (decreasing rate). Subsequently, Model 4b, testing for the moderating effect of local responsiveness, is found to be significant only at an α-level of 0.10. However, as Model 5b, which illustrates the interaction between the independent and moderator variable, does not lead to a substantial F change the moderating effect of local responsiveness is non-significant and thus hypothesis 4 is not supported.
Figure 2: Scatterplot (H1) - Scope of Downstream FSAs Figure 3: Scatterplot (H2) - Scale of Downstream FSAs versus Scope of Upstream FSAs versus Scale of Upstream FSAs
Table 1: Means, Standard Deviations and Correlations Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1. Company Age 61,224 51,550 1,000 2. Company Size 120221 152198 -0,004 1,000 3. Profitability 0,061 0,010 -0,040 -0,003 1,000 4. Company Ownership 0,182 0,386 -0,239** 0,111* -0,035 1,000 5. Europe 0,300 0,459 0,197** 0,034 -0,079 -0,197** 1,000 6. North America 0,292 0,445 0,093* 0,017 0,105* -0,259** -0,420** 1,000 7. Asia 0,372 0,484 -0,297** -0,032 -0,062 0,436** -0,504** 0,494** 1,000 8. South America 0,018 0,133 -0,030 -0,023 0,036 0,014 -0,089* -0,087 -0,104* 1,000 9. Oceania 0,016 0,126 0,123** -0,033 0,088 -0,060 -0,083 -0,082 -0,098* -0,017 1,000 10. Africa 0,002 0,045 -0,022 -0,024 0,032 0,095* -0,029 -0,029 -0,034 -0,006 -0,006 1,000 11. Primary Sector 0,154 0,361 -0,095* -0,129** 0,029 0,151** 0,023 -0,079 0,050 0,026 -0,010 -0,019 1,000 12. Secondary Sector 0,344 0,476 0,013 0,049 -0,074 0,001 -0,079 -0,085 0,183** -0,066 -0,059 0,062 -0,309** 1,000 13. Tertiary Sector 0,502 0,501 0,056 0,047 0,049 -0,110* 0,058 0,138** -0,210** 0,045 0,063 -0,045 -0,428** 0,727** 1,000 14. Local Responsiveness 0,038 0,549 -0,050 -0,071 0,307** -0,108 0,194* 0,032 -0,199* -0,031 -0,051 0,000 -0,089 0,296** -0,237** 1,000
15. Scale of Upstream FSAs 0,446 0,294 0,108 0,040 -0,046 -0,203** 0,396** -0,062 -0,303** -0,083 -0,149* 0,000 -0,005 0,129* -0,122 0,302** 1,000 16. Scope of Upstream FSAs 0,227 0,228 -0,049 0,174* -0,040 -0,049 0,253** -0,152* -0,152* 0,138 -0,086 0,000 -0,015 -0,077 0,086 -0,143 0,543** **. Correlation is significant at the 0,01 level (2-tailed).
Table 2: Regression Results - Hypothesis 1 and Hypothesis 3 Dependent Variable: Scope of Downstream FSAs
n N=46 Model 1a or Base Model Model 2a or Linear Relationship Model 3a or Non-Linear Relationship (H1) Model 4a or Strategic Focus on Local Responsiveness (H3) Model 5a or Interaction Term (Loc. Resp. vs. Scope of Upstream FSAs) Control Variables Company Age -0,316* -0,267* -0,286* -0,284* -0,299* Company Size -0,080 -0,087 -0,092 -0,093 -0,104 Profitability -0,219† -0,171 -0,143 -0,149 -0,182 Company Ownership -0,042 -0,026 -0,036 -0,032 -0,004 Continent of Origin North America -0,288* -0,138 -0,091 -0,086 -0,070 Asia -0,296† -0,176 -0,104 -0,093 -0,067 Industry Primary Sector -0,035 -0,022 -0,015 -0,011 0,014 Tertiary Sector -0,138 -0,165 -0,137 -0,129 -0,060 Independent Variable
Scope of Upstream FSAs 0,439** 0,676** 0,681** 0,672†
Scope of Upstream FSAs - Squared Term -0,275 -0,277 -0,194
Moderator Variable
Local Responsiveness 0,016 0,166
Interaction Term
Local Responsiveness x Scope of Upstream FSAs 0,182
Adjusted R2 0,070 0,246 0,256 0,235 0,229
F Change 1,433 9,854** 1,518 0,008 0,713
Table 3: Regression Results - Hypothesis 2 and Hypothesis 4 Dependent Variable: Scale of Downstream FSAs
n N=89 Model 1b or Base Model Model 2b or Linear Relationship Model 3b or Non-Linear Relationship (H2) Model 4b or Strategic Focus on Local Responsiveness (H4) Model 5b or Interaction Term (Loc. Resp. vs. Scale of Upstream FSAs) Control Variables Company Age 0,031 -0,004 -0,027 -0,014 -0,008 Company Size 0,098 0,055 0,047 0,051 0,052 Profitability 0,195* 0,090 0,108† 0,058 0,062 Company Ownership -0,297** -0,204** -0,196** -0,183* -0,180* Continent of Origin Europe 0,340** 0,228** 0,264** 0,235** 0,231** Asia -0,094 0,020 0,060 0,091 0,081 Oceania -0,030 0,031 0,055 0,063 0,062 Industry Primary Sector -0,175 -0,097 -0,095 -0,074 -0,075 Tertiary Sector -0,532*** -0,374*** -0,367*** -0,322*** -0,337*** Independent Variable
Scale of Upstream FSAs 0,479*** 0,471*** 0,468*** 0,470***
Scale of Upstream FSAs - Squared Term -0,115† -0,110† -0,093
Moderator Variable
Local Responsiveness 0,138† 0,138†
Interaction Term
Local Responsiveness x Scale of Upstream FSAs -0,039
Adjusted R2 0,354 0,523 0,529 0,538 0,533
F Change 6,415*** 29,475*** 1,941† 2,505† 0,195
5. Discussion
The findings of this research demonstrate that upstream FSAs are less location bound than downstream FSAs, consistent with theories suggesting that upstream and downstream FSAs face a different liability of foreignness (e.g. Rugman, 2005). It is found that the relationship between the scale of
internationalization of FSAs on opposite ends of the value chain is positive and nonlinear, with the slope greater at low levels of scale of internationalization for upstream FSAs. These results suggest that upstream FSAs initially internationalize faster than downstream FSAs and that investments in upstream FSAs will ultimately result in the creation of downstream FSAs. In addition, these results suggest that the marginal contribution of upstream FSAs decreases over time and that coordination costs increase over time, resulting in a single bend in the regression line (decreasing rate). Finally, despite international strategy did not significantly moderate the
relationship between the internationalization of upstream and downstream FSAs, we did find international strategy to have a direct effect on the scale of upstream and downstream FSAs.
This research failed to find a quadratic effect in the relationship between the scope of internationalization of upstream and downstream FSAs. In addition, we failed to find a positive moderating effect of a strategic focus on local responsiveness on the relationship between the scope and scale of internationalization of upstream and downstream FSAs. Instead, we found that the relationship between the scope of internationalization of upstream and downstream FSAs appears to be linear. This implies that firms can internationalize within their home region more easily than outside their home region. This is consistent with theories suggesting that the intra-regional liability of foreignness is lower than the inter-intra-regional liability of foreignness
(e.g. Asmussen 2009; Brain & Rugman, 2003). The limited intra-regional liability of foreignness arguably allows firms to almost simultaneously (linearly) benefit from their investments in upstream FSAs, possibly explaining the lack of a direct effect and non-significant contribution of a strategic focus on local responsiveness on the scope of internationalization of upstream and downstream FSAs.
5.1 Academic Relevance
This study adds to internationalization research by shedding light on the
asymmetry of, and relationship between, the internationalization patterns of upstream and downstream FSAs. Thus far, scholarly research had been largely incapable of addressing the relationship between the internationalization of FSAs along different parts of the value chain. Therefore, by investigating the degree to which the
internationalization of upstream and downstream FSAs are related, this research has both important academic and managerial relevance. In addition, this study adds to internationalization research by investigation the moderating effect of international strategy.
Based on the foregoing theoretical and empirical analysis, the following three insights emerged. First, evidence from Fortune Global 500 companies supports the notion that upstream FSAs are generally less-location bound than downstream FSAs and thus upstream FSAs are found to initially internationalize faster. Second, it is found that upstream and downstream FSAs show different internationalization patterns according to the location of the upstream FSAs (i.e. inter-regional/intra-regional). Third, it is found that international strategy (i.e. strategic focus on local responsiveness) does not play a significant role in the scope and scale of
5.2 Limitations and Suggestions for Future Research
It is important to note that this study has its limitations and that future research can extend this study in a number of useful ways. First, this research has only
focused on Global Fortune 500 companies. It is possible that the relationships identified in this research are not generalizable across contexts. Therefore, future research could test for the generalizability of the findings of this study in a different context (i.e. small and medium enterprises). In addition, this research has focused on relationships in a single (fiscal) year. However, it is possible for the observed
relationships to change over time, especially since the observed fiscal year (2012) can be considered to be relatively short after the global financial crisis. Therefore, it is important for future research to (longitudinally) examine the hypothesized
relationships over a long(er) period of time. Finally, the sample size is relatively small due to the lack of consistency in annual reports published by Global Fortune 500 companies. It is important, therefore, to repeat the study with a larger sample size.
Apart from the limitations and extensions to the current research, this research has opened up several other opportunities for future research. Future research could, for example, attempt to investigate the rate at which upstream and downstream FSAs internationalize in more detail. Also, the rate at which coordination costs occur calls for additional research. Finally, more insights into the observed internationalization phenomena can be obtained through a survey-based approach.
6. Conclusion
The primary purpose of this research was to contribute to internationalization research by exploring the relationship between the scope and scale of
internationalization for upstream and downstream FSAs. Besides, the research was undertaken in an effort to increase our understanding of the rationale that has made way for the emergent regionalization patterns (Rugman & Verbeke, 2004).
Scholars have previously argued that an asymmetry may exist between the internationalization of upstream and downstream FSAs (e.g. Rugman, Verbeke & Nguyen, 2011). Nonetheless, research on the relationship between the
internationalization of FSAs along different parts of the value chain remains
conspicuously absent. Therefore, in an effort to fill this academic gap, this research has initiated the endeavor to investigate the relationship between the scope and scale of internationalization of upstream and downstream FSAs. Additionally, this research investigates the extend to which this relationship is moderated by
international strategy (i.e. strategic focus on local responsiveness).
By performing a curvilinear regression analysis, it is found that the relationship between the scale of FSAs on opposite ends of the value chain is positive and
nonlinear, with the slope greater at low levels of scale of internationalization for upstream FSAs. The hypothesized positive and non-linear relationship between the scope of upstream and downstream FSAs is found to be non-significant. Instead, we found that the relationship between these two variables is positive and linear. In addition, it is found that neither scope nor scale of internationalization is significantly moderated by a strategic focus on local responsiveness. However, we did find a strategy focused on local responsiveness to have a direct effect on the scale of internationalization of upstream and downstream FSAs.
Thus, in line with our expectations, in the early phases of the
internationalization process, MNEs are highly dependent on the utilization of their less location-bound upstream FSAs, which will ultimately be caught up by, and result in, the creation of downstream FSAs. It is shown, however, that the relationship between and internationalization upstream and downstream FSAs shows different patterns according to the location of the upstream FSAs. It appears that investments in upstream FSAs result in downstream FSAs in a linear fashion inside the home-region and at a decreasing rate outside the home-home-region.
The findings of this research have both academic and managerial implications. First, this research has filled the identified gap in scholarly literature and presented venues for future research. Thus far, some scholars have argued that upstream FSAs are less location bound than their downstream counterparts (e.g. Rugman & Verbeke, 2004). This research has demonstrated that this assumption holds true in the context of Global Fortune 500 companies. Second, the illustration of the
existence of an asymmetry between upstream and downstream FSAs and the empirical findings of this research will potentially allow MNEs to more effectively capitalize on their upstream FSAs in foreign markets.
7. References
Ahokangas, P. 1998. Internationalization and resources: an analysis of
processes in Nordic SMEs. University of Vaasa.
Andersen, O. 1997. Internationalization and market entry mode: A review of theories and conceptual frameworks. MIR: Management International
Review, 27-42.
Arregle, J. L., Miller, T. L., Hitt, M. A., & Beamish, P. W. 2013. Do regions matter? An integrated institutional and semiglobalization perspective on the
internationalization of MNEs. Strategic Management Journal, doi: 10.1002/smj.2051.
Axinn, C. N., & Matthyssens, P. 2002. Limits of internationalization theories in an unlimited world. International Marketing Review, 19(5): 436-449. Baden‐Fuller, C. W., & Stopford, J. M. 1991. Globalization frustrated: the case of
white goods. Strategic Management Journal, 12(7): 493-507.
Banalieva, E. R., & Dhanaraj, C. 2013. Home-region orientation in international expansion strategies. Journal of International Business Studies, 44(2): 80-116.
Bartlett, C.A. and Ghoshal, S. 1989. Managing Across Borders: The
Transnational Solution. Harvard Business School Press.
Bell, J., McNaughton, R., Young, S., & Crick, D. 2003. Towards an integrative model of small firm internationalization. Journal of International
Entrepreneurship, 1(4): 339-362.
Benito, G. R., Grøgaard, B., & Narula, R. 2003. Environmental influences on MNE subsidiary roles: economic integration and the Nordic countries. Journal of
Birkinshaw, J., Morrison, A., & Hulland, J. 1995. Structural and competitive determinants of a global integration strategy. Strategic Management
Journal, 16(8): 637-655.
Buckley, P. J., & Casson, M. 1976. The future of the multinational enterprise. Macmillan.
Calof, J. L., & Beamish, P. W. 1995. Adapting to foreign markets: explaining internationalization. International Business Review, 4(2): 115-131. Cantwell, J., Dunning, J. H., & Lundan, S. M. 2010. An evolutionary approach to
understanding international business activity: The co-evolution of MNEs and the institutional environment. Journal of International Business Studies, 41(4), 567-586.
Cerrato, D. 2009. Does innovation lead to global orientation? Empirical evidence from a sample of Italian firms. European Management Journal, 27(5): 305-315.
Collinson, S., & Rugman, A. M. 2008. The regional nature of Japanese
multinational business. Journal of International Business Studies, 39(2): 215-230.
Delios, A., & Beamish, P. W. 2005. Regional and global strategies of Japanese firms. MIR: Management International Review, 19-36.
Dunning, J. H., & Rugman, A. M. 1985. The influence of Hymer's dissertation on the theory of foreign direct investment. The American Economic Review, 228-232.
Dunning, J. H., Fujita, M., & Yakova, N. 2007. Some macro-data on the
regionalization/globalization debate: a comment on the Rugman/Verbeke analysis. Journal of International Business Studies, 38(1): 177-199.
Emmerij, L. 1992. Globalization, regionalization and world-trade. Columbia Journal
of World Business, 27(2): 6-13.
Field, A. 2009. Discovering statistics using SPSS. Sage publications.
Forsgren, M. 2008. Theories of the Multinational Firm: A Multidimensional
Creature in the Global Economy. Cheltenham: Edward Elgar Publishing.
Franko, L. G. 1989. Global corporate competition: Who's winning, who's losing, and the R&D factor as one reason why. Strategic Management Journal, 10(5): 449-474.
Garrett, G. 2000. The causes of globalization. Comparative Political Studies, 33(67): 941-991.
George, G., Wiklund, J., & Zahra, S. A. 2005. Ownership and the
internationalization of small firms. Journal of Management, 31(2): 210-233.
Ghemawat, P. 2001. Distance still matters. Harvard business review, 79(8), 137-147.
Ghemawat, P. 2003. Semiglobalization and international business strategy. Journal
of International Business Studies, 34(2): 138-152.
Ghemawat, P. 2005. Regional strategies for global leadership. Harvard Business
Review, 83(12), 98.
Ghemawat, P. 2007. Redefining global strategy: Crossing borders in a world
where differences still matter. Harvard Business School Press.
Ghemawat, P. 2008. Reconceptualizing international strategy and organization.
Strategic Organization, 6(2): 195-206.
Ghoshal, S. 1987. Global strategy: An organizing framework. Strategic
Harzing, A. W. 2002. Acquisitions versus greenfield investments: International strategy and management of entry modes. Strategic Management
Journal, 23(3), 211-227.
Hennart, J. F. 2007. The theoretical rationale for a multinationality-performance relationship. Management International Review, 47(3): 423-452.
Hennart, J. F. 2009. Down with MNE-centric theories! Market entry and expansion as the bundling of MNE and local assets. Journal of International Business Studies, 40(9), 1432-1454.
Hu, Y. S. 1995. The international transferability of the firm's advantages. California
Management Review, 37(4): 73-88.
Hymer, S. A. 1960. The international operations of national firms: a study of
direct foreign investment. Massachusetts Institute of Technology.
Johanson, J., & Vahlne, J. E. 1977. The internationalization process of the firm- a model of knowledge development and increasing foreign market commitments. Journal of International Business Studies, 23-32. Johanson, J., & Vahlne, J. E. 1990. The mechanism of internationalization.
International Marketing Review, 7(4).
Kobrin, S. J. 1991. An empirical analysis of the determinants of global integration.
Strategic Management Journal, 12(S1): 17-31.
Kolk, A. 2010. Social and sustainability dimensions of regionalization and (semi) globalization. Multinational Business Review, 18(1), 51-72.
Kuivalainen, O., & Sundqvist, S. 2006. Profitability of rapid internationalization: the relationship between internationalization intensity and firms' export
Kumar, M. S. 1984. Growth, acquisition and investment: An analysis of the
growth of industrial firms and their overseas activities. Cambridge:
Cambridge University Press.
Levitt, T. 1983. The Globalization of Markets. Harvard Business Review, 61(3): 92-102.
Luo, Y. 2001. Determinants of local responsiveness: Perspectives from foreign subsidiaries in an emerging market. Journal of Management, 27(4): 451-477.
Muller, A., & Van Tulder, R. 2005. Exploring Patterns of Upstream
Internationalization: The Role of Home-region ‘Stickiness’.
ERS-2005-084-ORG.
Oh, C. H. 2009. The international scale and scope of European multinationals.
European Management Journal, 27(5): 336-343.
Oh, C. H., & Rugman, A. M. 2012. Regional integration and the international strategies of large European firms. International Business Review, 21(3): 493-507.
Ohmae, K. 1989. Managing in a borderless world. Harvard Business Review, 67(3): 152-161.
Osegowitsch, T., & Sammartino, A. 2007. Exploring trends in regionalization.
Research in Global Strategic Management, 13: 45-64.
Osegowitsch, T., & Sammartino, A. 2008. Reassessing (home-) regionalization.
Journal of International Business Studies, 39(2): 184-196.
Prahalad, C. K., & Doz, Y. 1987. The multinational mission: balancing local
Qian, G., Li, L., Li, J. and Qian, Z. 2008. Regional diversification and firm
performance. Journal of International Business Studies, 39: 197–214. Reid, S. D. 1981. The decision-maker and export entry and expansion. Journal of
International Business Studies, 101-112.
Roth, K., & Morrison, A. J. 1992. Implementing global strategy: characteristics of global subsidiary mandates. Journal of International Business Studies, 715-735.
Rugman, A. M, & Girod, S. 2003. Retail multinationals and globalization: the evidence is regional. European Management Journal, 21(1): 24-37. Rugman, A. M, & Sukpanich, N. 2006. Firm-specific advantages intra-regional
sales and performance of multinational enterprises. The International Trade
Journal, 20(3): 355-382.
Rugman, A. M. 1981. Inside the multinationals. Croom Helm. Rugman, A. M. 2000. The end of globalization. Random House.
Rugman, A. M. 2005. The regional multinationals: MNEs and 'global' strategic
management. Cambridge University Press.
Rugman, A. M., & Brain, C. 2003. Multinational enterprises are regional, not global. Multinational Business Review, 11(1): 3-12.
Rugman, A. M., & Oh, C. H. 2007. Multinationality and regional performance, 2001–2005. Research in Global Strategic Management, 13: 31-43. Rugman, A. M., & Verbeke, A. 1992. A note on the transnational solution and the
transaction cost theory of multinational strategic management. Journal of
