SEARCH

SEARCH BY CITATION

Keywords:

  • liability of foreignness;
  • global strategy;
  • regional strategy;
  • cultural distance;
  • institutional distance;
  • emerging markets;
  • resource-based view

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

While recent research has pointed to the importance of regional strategy and the ‘interregional liability of foreignness,’ critics have pointed out that this argument obscures important differences within regions as well as the similarities across them. Bridging these diverging viewpoints, our research is designed to unpack this debate into cultural, institutional, and regional components. Using a large data set, we find that firms are significantly more dispersed across cultural and, in particular, institutional boundaries, than they are across geographically defined regional boundaries. Further, our results indicate that certain firm-specific resources influence firms' global dispersion; in particular, we find that a nuanced interplay of proprietary capabilities such as technology, marketing, and partnering capabilities has an impact on the location of firm activities.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

In a challenge to the literature on ‘global strategy’ (see, e.g., Bartlett and Ghoshal, 1989; Govindarajan and Gupta, 2001; Yip, 2002), a number of scholars have argued that global firms are rare and that globalization, as most commonly conceptualized, is exaggerated (Rugman and Verbeke, 2004). By examining firm-level data on worldwide sales patterns, these researchers show that world trade occurs predominantly within, rather than between, each of the Asia-Pacific, European, and North American geographic clusters of countries (hereafter, ‘regions’) and that most so-called ‘global’ multinational enterprises (MNEs) actually derive their revenues primarily from countries within their own home region. This evidence has led to the suggestion that most MNEs take a regiocentric orientation when designing their location strategy. Subsequent studies have made progress by exploring the conceptualization and measurement (Asmussen, 2009), dynamics (Osegowitsch and Sammartino, 2008), and performance implications (Goerzen and Asmussen, 2007) of regional and global orientation.

There are two important gaps in this literature, however, that impede our understanding of the process of firm globalization. First, critics have challenged a key assumption in the theory of regional MNEs, which is that the triad regions exhibit relatively high degrees of intraregional homogeneity and interregional heterogeneity. This assertion, they claim, ignores important types of differences within regions as well as links between them (Galan and Gonzalez-Benito, 2006), thereby underestimating the importance of the cultural and institutional dimensions (Dunning, Fujita, and Yakova, 2007). It is an empirical question, therefore, as to whether a regional delineation is sufficient or whether cultural and institutional elements also need to be incorporated within the theoretical model of MNE organization. Second, since firms make a wide variety of location choices resulting in different degrees of dispersion in regional, cultural, and institutional space, it is important to improve our understanding of the conditions under which MNEs pursue these varying strategies. Such variation is significant to both scholars and practitioners because MNEs that vary in their geographic pattern of FDI are quite different in terms of the challenges they face and the managerial responses required. To improve our knowledge of MNE location strategy, therefore, we need to understand not only the macroeconomic, political, and technological foundations of globalization, but also the drivers and inhibitors of this process for individual firms.

To address these questions, our article develops a framework that simultaneously considers the interregional, intercultural, and interinstitutional dispersion of MNEs. Drawing on internalization theory, we analyze the contingencies that underpin the degree to which the foreign direct investment (FDI) of MNEs is globally oriented along these three dimensions. We begin by developing our theoretical perspective on the spatial location patterns of MNEs and the fungibility of intangible assets and partnering capabilities. We analyze firm-level data on 247 MNEs to test our hypotheses and then discuss the implications of our findings with some suggestions for future research.

Conceptualizing and Operationalizing Firm Globalization

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

While the concept of economic globalization is invoked in both international management research and the popular press, it is often conflated with internationalization, i.e., the process of firm dispersion beyond home countries into foreign markets. Hence, the questions traditionally asked by researchers interested in the MNE have focused on the drivers of the degree of firm internationalization, ignoring the emerging idiosyncratic spatial distribution of these international activities based, perhaps, on the underlying assumption that a global presence is a logical endpoint of an internationalization trajectory.

More recently, however, scholars have begun to reexamine the concept of internationalization by differentiating explicitly between the degree and pattern of foreign involvement in recognition that these are distinct notions with different antecedents and consequences (Goerzen and Beamish, 2003; Hutzschenreuter, Voll, and Verbeke, 2011; Vermeulen and Barkema, 2002). For example, two firms can have an equal degree of internationalization by having 50 percent of their assets in foreign countries yet, at the same time, these firms could have completely different internationalization patterns in which one firm had all of its FDI in a single foreign market proximate to its home country, and another's operations could be spread across all corners of the world. Thus, while equally ‘internationalized,’ these two firms would still be completely different in terms of spatial organization; one firm need only organize collaboration between its closely located units while the other must control and coordinate its activities across great divides. This suggests that the dispersion of a firm's foreign operations in terms of its remoteness from the home base is an important component of MNE scope and needs to be considered along with its degree of internationalization.

Dimensions of dispersion

In the literature dealing with MNE dispersion, international business (IB) scholars have promoted three broad theoretical perspectives, each of which ascribes particular importance to a certain type of boundary and conceptualizes its impact on the strategies and performance of MNEs. In most early work, the focal boundary has been a cultural one, as examined by Hofstede (1980), Ronen and Shenkar (1985), and Gupta, Hanges, and Dorfman (2002). More recently, scholars have also become preoccupied with institutional boundaries—inspired by the rise of emerging markets and new MNEs originating from these markets (Luo and Tung, 2007) 1—as well as with regional boundaries based on the observed dispersion of large MNEs (Rugman and Verbeke, 2005). With very few exceptions (see, e.g., Berry, Guillén, and Zhou, 2010), however, there is little integrative theoretical and empirical modeling that considers these dimensions jointly as distinct but related components of MNE location strategy. This extension is important since these elements, as we will argue, are individually incomplete but mutually complementary.

To illustrate this point, the ‘foreign dispersion cube’ in Figure 1 demonstrates the dimensions of dispersion for firms of two possible home country origins: Japan and the United States. The y axis follows Rugman and Verbeke's (2008) decomposition of a firm's foreign operations into two components: those located in the ‘rest of the region’ (ROR) and those located outside the home region, in the ‘rest of the world’ (ROW), the former being relatively proximate geographically and the latter distant. For the x axis, we use the GLOBE methodology (Gupta et al., 2002) to subdivide the ROR and ROW locations into those that are within the home country's own cultural cluster (OC) versus those in a foreign cultural cluster (FC). Finally, the z axis of Figure 1 uses World Bank Governance Indicators (Kaufmann, Kraay, and Mastruzzi, 2004) to distinguish between developed and emerging markets, where U.S. and Japanese firms (both originating from locations of high institutional development) face ‘similar institutions’ (SI) and ‘different institutions’ (DI).

figure

Figure 1. Foreign dispersion cube for Japan and the U.S.

Download figure to PowerPoint

Figure 1 reveals a complex relationship between the three dimensions, which are configured in a way that depends on the firm's country of origin. Sometimes cultural and regional boundaries coincide—for example, Japan is culturally similar and geographically proximate to China, while Japanese firms will encounter dissimilar cultures as soon as they venture outside their home region. In other cases, there may be a weak and even inverse relationship; Japan is institutionally less similar to China than to a large number of other culturally and geographically distant countries in North America and Europe. Furthermore, as seen by the differences between the two home country examples, the relationships among the three dimensions vary with the focal country of origin. Whatever the direction and strength, the possibility of correlations between regional, cultural, and institutional boundaries implies that there is a risk of conflating their effects if treated together. It is important, therefore, to unpack these dimensions in a way that disentangles their effects—a challenge that we address in this article.

Internationalization, Proprietary Assets, and the Liability of Foreignness

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

Why are some firms willing and able to expand more than others across cultural, institutional, and regional boundaries? The basic premise of the theory of the MNE (Buckley and Casson, 1976; Hennart, 1982) is that the growth of MNEs, both in number and size, can be attributed to their abilities to deal with inefficient markets—largely for knowledge-based assets—through internal organization in foreign markets. A related school of thought views the MNE as a vehicle for knowledge creation and replication across borders (Ghoshal, 1987; Kogut and Zander, 1993). The underlying requirement of both of these perspectives is that MNEs' proprietary assets or core firm-specific advantages (FSAs) are non-location bound (Rugman and Verbeke, 1992). The benefits of these FSAs, however, must be weighed against the costs of internationalization, which are escalated by host market conditions including local stakeholder discrimination against outsiders, foreign firms' lack of important local knowledge (Kostova and Zaheer, 1999), and undeveloped network relations with local organizations (Johanson and Vahlne, 2009). The notion that foreign firms incur costs over and above those of local incumbents was pioneered by Hymer (1976) and subsequently examined by Zaheer (1995) and Mezias (2002), who provided evidence of the existence of the liability of foreignness (LoF) as an inhibitor of MNE performance.

This view of FSAs as drivers of firm internationalization that are constrained by the LoF is an argument that, in itself, suggests no inherent limitation to firm internationalization. As long as the MNE's proprietary assets retain their value, rarity, inimitability, and non-substitutability (Barney, 1991) (and are, at the same time, strong enough to compensate for the LoF), they can be applied anywhere in the world. It is tempting to conclude, therefore, that the natural destination of an MNE is to develop a worldwide presence through a global strategy (Ghoshal, 1987; Govindarajan and Gupta, 2001; Yip, 2002), created by ‘global strategic capability’ (Hamel and Prahalad, 1985), yielding a worldwide corporate identity (Bartlett and Ghoshal, 1989).

The viability of global strategy has been challenged, however, as international management scholars have begun to explore the prevalence and performance implications of different geographic internationalization patterns. Some studies have indicated that the complexity and costs of pursuing global scope may outweigh the benefits in certain circumstances. For example, Goerzen and Beamish (2003) show that MNE cultural and geographic diversity inhibits performance, Vermeulen and Barkema (2002) that geographic dispersion lowers the performance benefits of internationalization, and Hutzschenreuter et al. (2011) that the extant cultural diversity of an MNE network inhibits further international diversification efforts. These findings highlight the challenge of appropriating value from proprietary assets abroad; thus, instead of pursuing a global strategy, MNEs arguably should grow within a proximate set of similar countries to maximize their returns on foreign operations. This implies that most FSAs may be more location bound than previously assumed and, therefore, that MNEs may reach the limits of organization before exhausting the value of their proprietary assets. Indeed, a highly constrained geographic scope may be the natural outcome if, as asserted by Eden and Miller (2004), the LoF is largely determined by the diversity and distance between home and host markets.

Unpacking outsidership

It is still not well understood how the diversity and distance argument should be operationalized since there have been few attempts to decompose the LoF. Traditionally, the MNE has been conceptualized as a national outsider in its host market, suffering additional costs as a result. However, given the aforementioned research on international diversity, we need to complement the idea of national outsidership with other potentially more important aspects of outsidership. We will argue for the merits of also considering cultural, institutional, and regional outsidership as distinct constructs.

Cultural outsidership

Scholars have long emphasized the various negative consequences of national cultural distance between MNE home and host country. Kogut and Singh (1988) suggested that MNEs would find it more difficult to manage culturally distant foreign operations and Gómez-Mejia and Palich (1997) argued that cultural distance would lead to higher agency costs. Most of this literature is motivated by an expectation that ‘firms operating in countries with a culture distant from that of their home country would experience a liability of foreignness’ (Mezias et al., 2002: 408), suggesting the existence of an intercultural LoF.

Institutional outsidership

A complementary argument can be made for the effect of institutional boundaries. Scholars are increasingly cognizant of the fact that countries differ in terms of the quality and effectiveness of their institutions and of the implications of this fact for the operations of MNEs (Henisz, 2000). Most often, this view of institutions is used to draw a distinction between developed markets with highly developed institutions, and emerging markets characterized by institutional voids. Hence, when MNEs internationalize across developed and emerging markets, they effectively span institutional boundaries. This is important inasmuch as the LoF faced by a firm in a foreign market depends on the institutional similarity of the home versus host market (Cuervo-Cazurra and Genc, 2008). Supporting this idea, Khanna, Palepu, and Sinha (2005) ascribed the failure of U.S. firms to internationalize into emerging markets to their inability to handle institutional voids in these markets, whereas Cuervo-Cazurra and Genc (2008) found that emerging market MNEs were successful in internationalizing into other emerging markets where they could employ their ability to manage such voids. This suggests that developed market MNEs face an interinstitutional LoF when they internationalize into emerging markets. In a parallel argument, when emerging market MNEs enter developed markets, MNEs will be at a disadvantage relative to entrants from other developed markets that know the ‘rules of the game’ (North, 1990) in such institutional settings. Irrespective of their origins, MNEs spanning institutional boundaries face the added complexity of managing business models and resource bundles in a way that is applicable to locations of varying institutional strength.

Regional outsidership

Finally, MNEs may encounter a liability of interregional foreignness as they venture into foreign markets outside their home region, leading them to incur relatively higher costs than what they would face in foreign markets within their home region. The limited interregional scope of most firms (Rugman and Verbeke, 2004) appears to be an indication that managers avoid these costs (Asmussen, 2009). Thus, a Japanese MNE may find it more difficult to adapt to the U.S. market than would a Canadian MNE, whereas both would be at a disadvantage compared to a U.S. firm.

In most literature on regional MNEs, a region is operationalized as a group of geographically proximate countries. Conceptually, however, geographic proximity is accompanied by—and indeed is a driver of—other types of proximity as well (Rugman and Verbeke, 2004, 2005) because it facilitates the exchange of goods, people, ideas, and the construction of supranational institutions. Rugman and Verbeke (2005) explain how multilateral regional frameworks such as NAFTA, the EU, and ASEAN have arisen among geographically proximate countries as transaction cost efficient vehicles for cross-border integration. In turn, as elaborated by Rugman, Verbeke, and Nguyen (2011: 770), ‘regional economic integration fosters institutional coordination, and may contribute to lowering actual cultural distance by increasing mobility of labor and managerial best practices. At the same time, improving the common transport infrastructure, adding transport connections (e.g., in air travel) in terms of frequency and quality may reduce the impact of geographical distance.’ Hence, institutions (such as economic freedom) may also be geographically contagious and spill over between neighboring countries (Leeson, Sobel, and Dean, 2006). The implication of all this is that, when MNEs expand across regions, they may face a compounded distance, i.e., a simultaneous and superadditive distance in multiple dimensions including geography, culture, and institutions (Rugman et al., 2011).

While this insight helps explain the regional nature of most MNEs' dispersion, there is not a perfect correspondence between geography, cultures, and institutions. Regions are, in fact, relatively heterogeneous culturally and, conversely, there are important cultural links across regions (Galan and Gonzalez-Benito, 2006). Leeson et al. (2006: 26) noted that their economic freedom spillover effect was only 20 percent and concluded that ‘while there is strong support that the domino effect exists, its impact on overall regional freedom is relatively modest.’ This suggests that we need to cleanse the interregional LoF of its cultural and institutional components to distill the costs that arise exclusively as a consequence of regional outsidership.

Even without those components, though, some very important costs remain. These are the ones, related mostly to geographic distance, that Eden and Miller (2004: 9) call ‘activity-based’ economic costs that relate primarily to the costs of transportation and communication, as well as to barriers to trade and FDI which tend to correlate with regional boundaries. Transportation of goods across geographic distance is associated both with direct costs (Hummels, 2007) and with indirect costs from delays and increased response time (Levy, 1997). Costs of communication and knowledge flows are also known to increase with distance (Jaffe, Trajtenberg, and Henderson, 1993), and regions are separated by time zones, which is a particularly strong barrier to coordination in MNEs (Asmussen, 2012; Stein and Daude, 2007). Finally, trade and investment agreements are generally negotiated among a group of proximate countries, as reflected in gravity models of economic exchange (see, e.g., Frankel and Rose, 2002). For the remainder of this article, therefore, we define the interregional LoF narrowly as the sum of these geographically derived effects (as opposed to the extant broader definition that includes institutional and cultural components).

The differential impact of cultural, institutional, and regional boundaries

Our analysis suggests that we need to reconceptualize the factors that influence MNE internationalization; rather than collapsing elements into the simple assumption that the LoF inhibits MNEs' global dispersion, we propose the finer-grained notion that: (1) intercultural LoF inhibits MNEs' intercultural dispersion; (2) interinstitutional LoF inhibits MNEs' interinstitutional dispersion; and (3) interregional LoF inhibits MNEs' interregional dispersion.

While the LoF has traditionally been defined very broadly, Zaheer (1995: 343) identified several sources as being of prime importance, including ‘costs directly associated with spatial distance, such as the costs of travel, transportation, and coordination over distance and across time zones; firm-specific costs based on a particular company's unfamiliarity with and lack of roots in a local environment; and costs resulting from the host country environment, such as the lack of legitimacy of foreign firms and economic nationalism.’ We suggest that each of these sources, summarized as uncertainty, discrimination, and complexity, have cultural, institutional, and regional components as illustrated in Table 1. While intercultural, interinstitutional, and interregional LoF are, thus, underlying dimensions of the broader phenomenon of LoF, their effects are not the same as we will demonstrate.

Table 1. Cultural, regional, and institutional components of LoF
Source of LoF (Zaheer, 1995)Intercultural LoFInterinstitutional LoFInterregional LoF
UncertaintyManagerial inability to acquire mind-sets of culturally distant customers, partners, and suppliersManagerial inability to understand ‘rules of the game’ in an institutionally different environmentCommunication costs, information distortion, and other difficulties related to acquiring information over geographic distances
DiscriminationConsumer discrimination against firms and products from distant culturesConsumer backlash against cultural and political imperialism or negative country of origin brand spilloverNational and regional policies that favor firms from geographically proximate countries, such as free trade agreements (FTAs)
ComplexityOrganizational costs related to employees' cultural differences (e.g., lack of trust, agency costs)Costs of variety and of integration while maintaining legitimacy in diverse institutional environmentsTravel costs and coordination costs inflated by geographic distance, lack of face-to-face contact, operating in different time zones
Uncertainty

The LoF is built, in part, on the idea that local firms have a better understanding than do foreign ones of cultural-based characteristics such as customer needs and local value systems. However, foreign firms appear to be able to learn about such characteristics (Petersen and Pedersen, 2002) before entry (Johanson and Vahlne, 1977) as well as afterward (Zaheer and Mosakowski, 1997), and such learning may be enhanced by locally ‘embedding’ the subsidiary through, for example, the hiring of host country nationals (Mezias, 2002). This also applies in the institutional dimension, where managers can acquire an increased understanding of the local ‘rules of the game,’ including government regulation and business laws. As noted by Cuervo-Cazurra and Genc (2008), developed country MNEs may, thus, learn to operate in emerging markets over time, thereby reducing their relative LoF.

While the MNE can, thus, achieve a high degree of cultural and institutional understanding, it may still face a residual degree of uncertainty stemming from its geographic dispersion. The MNE's local presence requires its FSAs (often home-based) to be combined with the host country's location-specific advantages (Rugman and Verbeke, 1993), suggesting that changes either in markets or technology create a need to transfer knowledge across geographic space. It has been shown that geographic distance—even in the absence of cultural distance—is associated with information flow barriers (Jaffe et al., 1993). One reason may be the difficulty of transferring tacit knowledge when the costs of face-to-face interaction are high. Conversely, national cultural distance in itself has been found not to significantly inhibit knowledge spillovers (Hussler, 2004), particularly for MNEs using expatriates (Saxenian and Hsu, 2001). Hence, regardless of whether the MNE comes to the host market to exploit its preexisting knowledge or explore new knowledge, it will face less intercultural and interinstitutional than interregional uncertainty, ceteris paribus.

Discrimination

Since humans respond favorably to stimuli that are recognized as being close to their own reality (so-called similarity-attraction theory, see e.g., Williams and O'Reilly, 1998), local consumers and workers may contribute to the LoF by deliberately, or even unconsciously, stereotyping cultural outsiders (Tajfel, 1982) and discriminating against foreign products and firms. In response, however, the MNE may try to educate nationalistic consumers about its home country culture or use its presence in foreign markets to break down stereotypes of foreigners. It has been suggested that foreign firms may, thus, become ‘insiders’ over time (Zaheer and Mosakowski, 1997). Similarly, anecdotal evidence has informed us on the challenges to legitimacy that MNEs experience when they cross institutional boundaries; for example, developed market MNEs being perceived as agents of neocolonization or emerging market MNEs being associated with lower brand value due to their origins (Kostova and Zaheer, 1999)—a reflection of the idea that nations, like firms, have brand equity (Kim and Chung, 1997). Like cultural sources of discrimination, however, institutional discrimination is also about perceptions and can, therefore, be managed or changed over time.

In contrast, it may be more difficult to overcome discrimination enforced by government policies. Such discrimination is often highly correlated with regional boundaries such as the EU, NAFTA, and ASEAN (Crawford and Laird, 2001), since they levy tariffs or other frictions on regional outsiders. 2 These trade restraints often reinforce and coincide with direct barriers to FDI—regional trade blocs, for example, tend to enforce most-favored nation privileges for regional investors to support their internal markets. Furthermore, they tend to lead to increased intraregional rivalry and consolidation processes in which strong firms expand within the region whereas weaker firms are forced to exit or try to improve their competitive position though efficiency-seeking FDI (Rugman, Verbeke, and Luxmore, 1990). Cuevas et al. (2005), thus, estimated that trade agreements could enhance FDI flows by up to 60 percent. Taken together, these arguments suggest that regional boundaries are a significant influence on FDI location.

Complexity

Finally, information asymmetries combined with self-interest seeking may lead to transaction costs in dispersed organizations, such as the costs of shirking, cheating, and monitoring (Hennart, 1991). At face value, these costs would be inflated by cultural differences and the associated behavioral uncertainties (Brouthers and Brouthers, 2003). However, the underlying assumption behind this argument—that economic imperatives are sidelined by cultural biases—may not hold true in large, highly internationalized, and professionally run MNEs. Executives in such organizations are often experienced in managing foreign nationals and coordinating culturally diverse teams. Accordingly, studies of international joint ventures point out that cultural differences between team members become less important over time (Stahl et al., 2009).

Managing across institutional boundaries also implies a certain complexity due to the difficulty of designing compensation systems and business models that work well and are perceived as equally legitimate in developed and emerging markets where the ‘rules of the game’ are different. Nevertheless, according to Khanna et al. (2005), it is possible for developed country MNEs to adapt their business models to emerging market contexts without sacrificing their core competences. Like cultural complexity, institutional complexity is also about managing diverging stakeholder expectations and can, therefore, be overcome (although perhaps more slowly than cultural complexity). In contrast, the geographical sources of coordination and monitoring costs are more difficult to mitigate; for example, one cannot simply learn how to avoid the challenges of operating across time zones or the costs of international travel. Together, these ideas imply that cultural differences and, to a lesser extent, institutional differences can be mitigated, whereas the difficulties associated with geographic distance cannot.

Implications for MNE dispersion

Our analysis suggests that cultural distance is a potentially manageable barrier because cultural boundaries are cognitive constructs that exist mainly in the minds of managers and consumers, whereas geographical distance is a phenomenon that exists in the physical world. This view has received some support from extant studies; Dow and Karunaratna (2006) estimate that the effect of geographic distance on trade flows is four times that of cultural distance. At the firm level, Benito and Gripsrud (1992) do not find support for a hypothesis of increasing cultural distance in MNEs' internationalization trajectories and, in their meta-analysis, Tihanyi, Griffith, and Russell (2005) find no universally significant impact of cultural distance on entry mode, diversification, and MNE performance. While similar research on institutional diversity is scarcer, we suggest that institutional costs share an important characteristic with cultural costs—they can be diminished through training, experience, and cognitive effort unlike the costs that result from geographic distance. This idea can also be derived from Table 1 where the relatively most important sources of intercultural and interinstitutional LoF arguably appear in the uncertainty and discrimination categories which, at the same time, are of a more transitory nature than is complexity. Taken together, this discussion suggests that, all else being equal, firms should find it easier to overcome cultural and institutional barriers than regional barriers and, therefore, they should be more internationalized, on average, in the cultural and institutional dimensions than in the regional dimension. This leads us to hypothesize:

  • Hypothesis 1: MNEs are less interregionally dispersed than they are interculturally or interinstitutionally dispersed.

The effect of intangible assets on interregional and intercultural dispersion

Intangible assets are often considered one of the major enablers of global dispersion and the strength of these intangible assets has been found to correlate with the global orientation of MNEs' foreign activity (Goerzen and Beamish, 2003; Morck and Yeung, 1991). In a complementary argument, we propose that firms with strong intangible assets are likely to be more globally dispersed because they are able to overcome the LoF (Nachum, 2003) and, therefore, can tolerate better the complexity of global organization. This is consistent with evidence by Hennart and Park (1994) and Chang (1996), who found that Japanese firms with high R&D intensity were more likely to internationalize into the U.S. market, and Morck and Yeung (1991, 1992), who found that firms' R&D to sales ratios had a positive moderating effect on the returns to internationalization. The underlying argument in this literature stream is that these assets enable MNEs to overcome the LoF, which would otherwise inhibit their expansion into global markets. Nachum (2003) suggested that the proprietary assets of MNEs may sometimes more than compensate for the LoF leading foreign firms to outperform local incumbents.

While this is essentially a market-seeking argument, more recent research suggests that it may also hold for strategic asset seeking. Nachum (2011) estimates a hierarchy of location-specific resources that differ in the extent to which they are accessible to foreign firms who suffer discrimination and uncertainty in their international sourcing efforts. For example, being an outsider to the local knowledge environment, the MNE have to incur costs over and above those of local insiders in order to generate organizational learning in a foreign country (Narula, 2012). This may include the transfer of internal knowledge to their subsidiaries in order to enable these subsidiaries to act as recipients of external knowledge (Asmussen, Foss, and Pedersen, forthcoming), a type of ‘metanational absorptive capacity’ argument (Doz, Santos, and Williamson, 2001). Consistent with this idea, Penner-Hahn and Shaver (2005) showed that international R&D activities led to an increase in innovation output, but only for firms with preexisting technological capabilities. The implication of all this is that firms with strong intangible assets will tend to be better at strategic asset seeking and, therefore, have an incentive to pursue a high degree of global dispersion in order to become exposed to high knowledge diversity (Cantwell and Janne, 1999). This suggests the following hypothesis:

  • Hypothesis 2: MNEs that have stronger intangible assets are more globally (interculturally, interinstitutionally, and interregionally) dispersed.

Previous research, however, does not distinguish between the different dimensions of FDI dispersion, and we suggest that intangible assets are not equally important in all of these dimensions. Indeed, as a corollary to our first hypothesis, intangible assets may be particularly critical when firms aim for a high degree of interregional dispersion which, as argued earlier, is associated with stronger LoF. This is because these assets provide the firm with a competitive advantage and the increasing returns to such an advantage may compensate for even particularly strong disadvantages of outsidership (Banalieva and Dhanaraj, forthcoming). Conversely, if the intercultural and interinstitutional LoF are more easily overcome than the interregional LoF to begin with, the need for intangible assets may be weaker when firms expand across cultures and institutions as compared to expansion across regions. This leads us to hypothesize:

  • Hypothesis 3: Intangible assets are more strongly and positively related to interregional dispersion than they are to intercultural or interinstitutional dispersion.

Partnering capabilities and globalization

The intangible assets of the firm are not the only proprietary resources that are likely to matter for its global dispersion. As suggested by Harbison and Pekar (1998), as markets globalize and competition for markets intensifies, managers step up their search for new resources through interfirm alliances. Theorists have suggested that there are three particularly important potential economic benefits that accrue to an organization with effective interfirm relations (Burt, 1992). The first benefit is that of access to information, given that alliances can provide information well beyond what an organization could possess alone. The second is timing, where information provided early is advantageous to the recipient. For these reasons, an organization with an effective set of interfirm ties will have established contacts wherever useful information is likely to surface, thereby mitigating the uncertainty aspect of the LoF. The third benefit is that of referrals, where the focal firm's interests are represented positively to third parties (Burt, 1992), suggesting that alliance benefits also include mitigating the LoF by increasing legitimacy. The knowledge, resources, stability, and associative legitimacy that partners confer tend to compensate for the disadvantages of organizational experience (Baum, Calabrese, and Silverman, 2000) and can provide the external endorsement of its operations (Baum and Oliver, 1991) as well as improve the perceived quality of its products and services among potential customers (Stuart, Hoang, and Hybels, 1999). This increased legitimacy may enable the MNE to deploy its own capabilities more effectively in foreign markets by recombining these capabilities with complementary location-specific resources (Rugman et al., 2011) such as the local reputation of partner firms.

Not all firms may find it equally easy, however, to initiate and manage international alliances (Goerzen, 2005; Goerzen and Beamish, 2005). As suggested by Schilke and Goerzen (2010), organizations with strong partnering capabilities possess routines that allow for efficient and effective alliance management, and such organizational alliance management routines enable firms to grow. When these partnering capabilities are extended in international markets they provide the MNE with an effective way to bring its own proprietary assets to global markets. This suggests the following hypothesis:

  • Hypothesis 4: MNEs that have stronger partnering capabilities are more globally (interculturally, interinstitutionally, and interregionally) dispersed.

As was the case for intangible assets, we expect partnering capabilities to have an asymmetric impact on the three types of dispersion, but in a different way due to the fundamentally different nature of these two types of resources. Whereas intangible assets are intrinsic to the firm and provide it with a competitive advantage that may compensate for its outsidership, partnering capabilities are, by definition, relational and as such may have a direct influence on some elements of this outsidership. In particular, partnering capabilities may accelerate the learning and legitimization processes by which MNEs overcome uncertainty and discrimination (as argued earlier) by tapping into partner firms' knowledge and reputation. This suggests that partnering capabilities should be more strongly related to intercultural and interinstitutional dispersion because learning and legitimization barriers play a more important role in these dimensions than they do in interregional dispersion. These arguments lead to our final hypothesis:

  • Hypothesis 5: Partnering capabilities are more strongly positively related to intercultural and interinstitutional dispersion than they are to interregional dispersion.

Methodology

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

Data description

The central source of data for this study was a survey of the subsidiaries of Japanese corporations listed on the Tokyo stock exchange. The 1999 survey results were published by Toyo Keizai Shinposha (Toyo Keizai, 1999). The surveys were completed with a response rate of 60 percent by the subsidiary general managers and included subsidiary location, industry, employees, annual revenue, capital investment, and equity partner identities. These subsidiary-level data were then aggregated into a database of the geographic distribution of FDI.

Using Stopford and Wells' (1972) early definition that MNEs have operations in six or more countries, we tested our theoretical framework with a data set of 247 Japanese MNEs that have a total of 13,529 subsidiaries. The core corporate-level data were augmented with relevant data from Compustat and the Analysts' Guide (Daiwa Institute of Research, 1999).

Dependent variable measures

We operationalize intercultural, interinstitutional, and interregional dispersion by segmenting the firm's foreign operations according to the distinctions in Figure 1a. For each dimension of dispersion, we code each host country as being ‘similar to’ or ‘different from’ Japan. We believe this is a parsimonious and theoretically meaningful approach that is consistent with prior research (Hitt et al., 2000), as well as with our theoretical framework, which builds on a widely accepted distinction between insiders and outsiders and on the idea that the latter suffer a particular type of LoF (Hymer, 1976; Johanson and Vahlne, 2009; Zaheer, 1995). The alternative—using continuous distance measures—raises a number of questions of linearity and functional form by implicitly assuming away discontinuities and threshold effects (Shenkar, 2001). Gupta et al. (2002: 11) note that country clusters ‘are a useful way to summarize intercultural similarities as well as intercultural differences’ and we contend that the same is valid for interregional and interinstitutional similarities and differences. We operationalize MNE activities with three items—foreign capital invested, foreign revenue generated by the subsidiaries, and foreign employees in these subsidiaries—and average the dispersion of these into a composite measure for each dimension.

Interregional dispersion

We define geographic regions using the triad definition consistent with Rugman and Verbeke (2004) in which Japanese firms belong to the Asia-Pacific region. We do this in order to establish comparability with previous research and because our theoretical arguments require this type of aggregation (e.g., pertaining to time zones and FTAs). Our composite measure of interregional dispersion has a Cronbach's alpha of 0.89.

Intercultural dispersion

We used the GLOBE study (Gupta et al., 2002) to delineate different national cultures because it is one of the most recent efforts, based on extensive data gathering and analysis, and more comprehensive in scope than the alternatives (see, e.g., Hutzschenreuter and Voll, 2007, for a detailed discussion of these attributes). Earlier cultural clustering efforts have key limitations that make them problematic, particularly in the context of our sample of Japanese MNEs. According to Ronen and Shenkar's (1985) review of nine studies that propose different cultural clustering approaches (including Hofstede's [1976] and [1980] efforts), three studies did not include Japan, and another five classified Japan as an ‘independent’ nation with no culturally similar countries. Only one approach, by Redding (1976), included Japan in a ‘Far East’ cluster (which, as the name indicates, seems more a geographic than a cultural distinction). Ronen and Shenkar (1985: 452) attributed the difficulty of clustering Japan to incomplete methodologies that exclude most Asian countries and note that ‘if researchers had included more countries, especially those with cultural dimensions in common with Japan, a cultural cluster including Japan might have emerged.’ We argue that the GLOBE method is thereby more complete, which is why it is extensively used by international business researchers (Hutzschenreuter and Voll, 2007; Hutzschenreuter et al., 2011). GLOBE places Japan in the ‘Confucian cluster’ of countries, which also includes China, Hong Kong, Singapore, South Korea, and Taiwan. These cultures are characterized by high future orientation, high performance orientation (valuing excellence and encouraging training and development), and high degrees of in-group and institutional collectivism wherein group membership is important and group performance and awards are emphasized (Gupta and House, 2004; Javidan et al., 2006). Accordingly, we classify operations in these countries as being within our firms' cultural cluster and all other locations as being outside it. The composite measure has a Cronbach's alpha of 0.84.

Interinstitutional dispersion

While there is broad agreement on the characteristics on emerging markets and developed markets, there is no consensus about how to classify all the countries of the world into these two groups and, thus, we had to construct our own method. As a starting point, we obtained measures of the countries' development level in form of the World Bank Governance Indicators (Kaufmann et al., 2004), which are widely accepted measures of institutions (see, e.g., Cuervo-Cazurra and Genc, 2008; Dikova and Van Witteloostuijn, 2007). The six items in this scale—voice and accountability, political stability, government effectiveness, regulatory quality, rule of law, and control of corruption—are highly correlated, with a Cronbach's alpha of 0.96. Following Dikova and Van Witteloostuijn (2007), we calculated a standardized average of these six items to form an ‘institutional development score.’ 3 Countries that had institutional development above the median were then defined as institutionally similar to Japan (developed markets) and the rest as institutionally different countries (emerging markets). The results of this procedure largely reflect other definitions; for example, the ‘advanced economies’ as defined by IMF appear exclusively in our developed markets category. 4 The resulting (composite) measure of interinstitutional dispersion has a Cronbach's alpha of 0.80 in our sample. As a robustness test, we also divided the countries around the 40 and 60 percentiles and used these divisions for alternative definitions of the variable.

Independent variables

Intangible assets

Following prior research (Delios and Beamish, 1999), proprietary technological assets were measured by R&D intensity and proprietary marketing assets by advertising intensity, i.e., the firm's R&D or advertising expenditures, respectively, as a share of its total sales. The main theoretical rationale is not only that these expenditures in themselves produce intangible benefits, but that firms possessing tacit and immeasurable capabilities may reap a higher return from (and, thus, incur more of) such expenditures. We note that our intensity measures are not direct observations of tacit capabilities since such capabilities are, by definition, difficult to observe and alternative measures of knowledge assets have their own limitations. For example, patents are also a noisy indicator of firm-specific technological assets because patenting propensity is industry specific and driven by other concerns besides the quality and quantity of the firm's technological innovations. Interestingly, however, it has been shown that there is a strong correlation between R&D expenditures and patent output (Scherer, 1983), suggesting that they are both useful measures of technological capabilities.

Partnering capabilities

The firms in our sample extensively engage in international joint ventures (IJVs) in international markets. Yet, the total number of IJVs may not be a meaningful measure of partnering capabilities, as it is conflated with firm size and says little about the added value of each alliance. In fact, a large number of the international alliance partners of our firms are either repeat partners or Japanese partners which, all else being equal, would be both less demanding of the firm's partnering capabilities and less useful as opportunities to tap into diverse foreign market knowledge (Goerzen, 2007). Thus, we measure the firm's partnering capabilities with two items: the ratio of unique to total JV partners and the ratio of non-Japanese to total partners in foreign markets (Cronbach's alpha = 0.75), thereby capturing the firm's ability to manage a diverse portfolio of foreign partners.

Control variables

We also control for a number of variables that could influence our dependent variables, including firm size, international scale, international experience, knowledge-seeking orientation, and industry.

Firm size

As large firms may be more internationalized, we controlled for firm size operationalized with the natural log of the firm's total revenue, assets, and employees. The composite score of these items had a Cronbach's alpha of 0.95.

International scale

The more extensive the foreign operations of the MNE, the more likely that it will also be globally dispersed. For example, highly internationalized firms may saturate proximate markets and, therefore, need to seek further growth in more distant markets. To capture this, we include the absolute size of international operations as a control variable influencing the degree of global dispersion. We obtained three globalization indicators, i.e., the subsidiaries' total capital invested, number of employees, and revenues to create a composite score (Cronbach's alpha = 0.81).

International experience

The longer the firm's international assets have been deployed, the more experiential knowledge acquired by its managers. This, in turn, increases managerial capacity for globalization and enables the penetration of markets with high psychic distance (Johanson and Vahlne, 1977), as is necessary to succeed with a globally oriented strategy. Therefore, we also control for international experience measured as the average age of the MNE's subsidiaries.

Knowledge-seeking FDI motive

Since diverse knowledge is believed to be valuable (Page, 2007), a knowledge-seeking FDI motive might give the MNE an incentive to pursue a larger degree of global dispersion. As part of the original survey, each subsidiary manager was also asked about the original purpose(s) of establishing the subsidiary. Among the purposes in the list offered to the respondents were three which at face value relate to knowledge seeking: ‘information collection,’ ‘product development and planning,’ and ‘entry into a new business.’ Using this data, we constructed a measure of knowledge seeking by counting the affirmative answers within the subsidiary network of each MNE.

Industry

Since global dispersion patterns may be industry specific, we used the main SIC code reported by each firm to include two industry dummies we would expect to be related to the firm's ability to internationalize. The first dummy captures high-tech industries, since these are often believed to be more global than other industries (Roth and Morrison, 1992). Based on the definition by the American Electronics Association, 5 we defined high-tech firms as being in the computer, electronics, communications, and software industries. We also include a dummy for retail and wholesale industries, since retail business models often appear to be quite location bound (as illustrated, for example, by Wal-Mart's early experiences in Europe). Descriptive statistics and correlations for all items are given in Table 2.

Table 2. Correlations* and descriptive statistics
VariableMeanSD1234567891011
  1. *Correlations at or above 0.10 significant at the p < 0.05 level. **Global dispersion scores are normalized factors, raw FDI scores reported in Table 3.

 1. Technological assets0.020.03          
 2. Marketing assets0.010.020.07         
 3. Partnering capabilites0.001.00−0.040.03        
 4. Firm size0.001.00−0.060.01−0.04       
 5. International experience10.80.170.02−0.09−0.010.07      
 6. International scale0.001.000.22−0.06−0.11−0.030.08     
 7. Knowledge seeking4.190.220.08−0.13−0.19−0.030.020.17    
 8. High-tech industry0.330.020.13−0.150.18−0.080.070.19−0.20   
 9. Retail and wholesale industry0.070.01−0.13−0.16−0.06−0.01−0.00−0.02−0.01−0.16  
10. Interregional dispersion**0.001.000.300.020.100.05−0.010.030.080.02−0.21 
11. Intercultural dispersion**0.001.000.210.140.080.03−0.020.040.200.06−0.270.76
12. Interinstitutional dispersion**0.001.00−0.170.10−0.08−0.01−0.100.020.04−0.130.01−0.42−0.28

Empirical methods

As is evident from Table 2, our three dependent variables are correlated; interregional and intercultural dispersion are strongly positively correlated, while both are moderately negatively correlated with interinstitutional dispersion. As noted earlier, this is an expected outcome given the way these boundaries are drawn from the perspective of Japan (see Figure 1). As a result, we perform two different types of econometric analysis to uncover the key relations.

First, to test our hypotheses we apply a seemingly unrelated regression (SUR) model to the three raw global dispersion scores. 6 Second, we supplement this analysis with a variance decomposition exercise in an effort to disentangle the effects on our dependent variables' shared and unique variances, enabling us to draw out more nuanced findings from our data. This is because correlated dependent variables may lead to an empirical conflation issue, similar to multicollinearity, which manifests itself among our dependent variables. For example, as a Japanese firm increases its interregional dispersion this, in itself, leads to higher levels of intercultural dispersion because there are no Confucian countries outside the Asia-Pacific region (which explains why the two dependent variables are correlated in the first place). We posit that this issue of dependent variable conflation, while rarely identified explicitly, has been a significant barrier for IB researchers in their endeavors to disentangle the different dimensions of internationalization.

Addressing this problem requires a transformation of the variables so as to make them orthogonal (Stephan and Uhlaner, 2010). A direct way to accomplish this is to isolate the shared variance from the unique variance of each dependent variable. Hence, as a supplementary analysis, we perform a variance decomposition test on each dependent variable in which we regress that variable on the two other dependent variables, storing the predicted and residual values. The predicted values then contain that variance which is shared and can be fully explained by the two other dependent variables. We will refer accordingly to this variable as the ‘shared component’ of the dependent variable. The residuals, however, will contain that dependent variable's unique variance, henceforth labeled its ‘orthogonal component,’ which is uncorrelated with the other dependent variables and, thus, maps cleanly onto the corresponding dimension of LoF. Finally, for each dependent variable, we run regressions with all our independent variables on each of these two components. This enables us to decompose the effect of a given independent variable into (1) the effect on the shared component of the dependent variable and (2) the effect of the orthogonal component of that variable, with 1 + 2 being the total effect that we have already estimated using the raw global dispersion scores in the SUR. So, for example, when using the orthogonal component of intercultural dispersion as a dependent variable, a significant effect means that an increase in the independent variable leads to an increase in intercultural dispersion while holding all else constant, because, by construction, the orthogonal component of intercultural dispersion varies independently of (is uncorrelated with) interregional or intercultural dispersion.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

Hypothesis 1 predicted greater intercultural and interinstitutional dispersion as compared to interregional dispersion and this needs to be tested separately from other hypotheses. A challenge in testing this hypothesis is that the three dimensions are not directly comparable because of the different economic sizes of the segments defined by each dichotomy. Therefore, following earlier studies (Asmussen, 2009; Fisch and Oesterle, 2003), we normalize the dispersion measures by the underlying GDP distribution. For example, since 92 percent of World GDP (excluding Japan) is located outside the Confucian cluster, a ‘fully interculturally dispersed’ Japanese firm should also have 92 percent of its FDI in countries outside this cluster and, therefore, we divide the intercultural FDI distribution of each firm by this benchmark of 92 percent. Normalized dispersion scores are calculated in a similar fashion for the two other dimensions and are presented in Table 3.

Table 3. Tests of Hypothesis 1
FDI dispersionRawGDPNormalizedt (vs. 2)t (vs. 3)
  1. *p < 0.10; **p < 0.05; ***p < 0.01.

1. Interregional56.5%88.2%64.1%−15.5***−8.3***
2. Intercultural72.6%91.6%79.3%−6.4***
3. Interinstitutional20.7%17.0%121.8%6.4***

As can be seen from the first row of the table, the average firm in our sample has 56.5 percent of its FDI outside the Asia-Pacific region. However, the GDP distribution is such that 88.2 percent of non-Japanese economic activity is outside this region. We, therefore, conclude that our firms have progressed only 56.5/88.2 = 64.1 percent toward a ‘fully global’ level of interregional dispersion. Applying similar computations to the two other dimensions, we see that firms are indeed more dispersed in the cultural (79 percent) and institutional dimensions (122 percent) versus the regional dimension (64 percent). We performed t-tests on these differences and found that they were significant at the p < 0.001 level, thereby supporting Hypothesis 1. As an aside, it is interesting that firms are more interinstitutionally dispersed than is the GDP distribution, implying either that interinstitutional LoF does not exist or that firms are often able to develop the skills to overcome it.

Our remaining hypotheses are tested with SUR and regression models, 7 which are reported in Tables 4, 5, and 6. Intangible assets emerge as important determinants of international dispersion as previous literature has demonstrated (Morck and Yeung, 1991). We find partial support for Hypothesis 2, as technological assets and marketing assets have a positive impact on the firm's dispersion across regional (see Table 4) and cultural boundaries (see Table 6). However, only technological assets have an impact on interinstitutional dispersion (see Table 5), and this impact is negative. Hence, Japanese firms with strong technological assets are likely to avoid emerging markets, perhaps reflecting a greater fungibility of these assets across cultures and regions than across institutional settings. We also find partial support for Hypothesis 3 since we can see that intangible assets have a particularly strong effect on interregional expansion. In Table 4, both types of intangible assets have a positive coefficient on the orthogonal component of interregional expansion, suggesting that these assets explain some of the variance that is unique to interregional expansion (although this is significant only for technological assets).

Table 4. Drivers of interregional dispersion

Dependent variable:

Interregional dispersion

Total (shared + orthogonal)Shared componentOrthogonal component
EstimateStd errorBeta
  1. *p < 0.10; **p < 0.05; ***p < 0.01.

Technological assets8.71***2.000.275.00***3.71***
Marketing assets5.19**2.560.133.37*1.82
Partnering capabilites0.10**0.050.120.08**0.02
Firm size0.350.330.060.170.19
International experience−0.000.02−0.000.00−0.00
International scale−0.030.10−0.01−0.030.00
Knowledge seeking0.020.010.100.03***−0.01
High-tech industry−0.040.12−0.02−0.050.01
Retail and wholesale industry−0.95***0.33−0.17−0.86***−0.09
Percent of variance in DV100%55%45%
Pseudo R-square / R-square0.160.150.05
SUR chi-square / F46.0***4.65***1.47
Breusch-Pagan chi-square192.3***
Table 5. Drivers of interinstitutional dispersion

Dependent variable:

Interinstitutional dispersion

Total (shared + orthogonal)

Shared

component

Orthogonal component
EstimateStd errorBeta
  1. *p < 0.10; **p < 0.05; ***p < 0.01.

Technological assets−4.99***1.93−0.17−3.20***−1.79
Marketing assets−2.572.47−0.07−1.83*−0.74
Partnering capabilites−0.040.05−0.05−0.03*−0.01
Firm size−0.110.32−0.02−0.130.03
International experience−0.020.01−0.10−0.00−0.02*
International scale0.110.100.08−0.010.11
Knowledge seeking−0.000.01−0.01−0.000.00
High-tech industry−0.22*0.11−0.13−0.00−0.22**
Retail and wholesale industry−0.170.32−0.030.27**−0.44
Percent of variance in DV100%16%84%
Pseudo R-square / R-square0.060.140.05
SUR chi-square / F17.0**4.17***1.25
Breusch-Pagan chi-square192.3***
Table 6. Drivers of intercultural dispersion

Dependent variable:

Intercultural dispersion

Total (shared + orthogonal)Shared componentOrthogonal component
EstimateStd errorBeta
  1. *p < 0.10; **p < 0.05; ***p < 0.01.

Technological assets5.42***1.860.185.78***−0.35
Marketing assets3.98*2.380.103.47**0.50
Partnering capabilites0.10**0.050.130.07**0.03
Firm size0.200.310.040.24−0.04
International experience−0.000.01−0.01−0.00−0.00
International scale0.000.100.00−0.010.01
Knowledge seeking0.04***0.010.210.010.02***
High-tech industry−0.150.11−0.09−0.04−0.11
Retail and wholesale industry−1.34***0.31−0.25−0.68***−0.66***
Percent of variance in DV100%50%50%
Pseudo R-square/R-square0.060.160.09
SUR -square/F17.0**4.93***2.51***
Breusch-Pagan chi-square192.3***

Similarly, Hypothesis 4 is supported for interregional and intercultural dispersion but not for interinstitutional dispersion. Thus, we find mixed support for Hypothesis 5 as, on the one hand, partnering capabilities seem more strongly related to interregional than to interinstitutional dispersion, perhaps reflecting the relative importance of the regional over the institutional LoF. On the other hand, partnering capabilities also have a strong effect on intercultural dispersion, which is consistent with our theoretical arguments. This might suggest that partnering capabilities are particularly valuable as a means of addressing cultural barriers given that a local partner may provide an understanding of host country culture.

Several interesting findings emerge among the control variables as well. For example, the dummy for retail and wholesale industry has a negative coefficient on interregional (see Table 4) and intercultural expansion (see Table 6). This seems to confirm our initial idea of retail business models being more location bound than business models in other industries. Interestingly, there does not seem to be an effect on the ability to disperse across institutional boundaries (see Table 5), suggesting that Japanese retail MNEs are as successful in emerging markets as are other Japanese companies. High-tech industry, on the other hand, has a negative effect on interinstitutional dispersion, but not on the other two dimensions, perhaps reflecting the lack of technological infrastructure in many emerging markets. Finally, firms of a knowledge-seeking orientation are more likely to be interculturally dispersed, perhaps reflecting a search for diverse knowledge.

Taken together, the shared components regressions are suggestive of something we call a dominant internationalization pattern, in which some Japanese firms ‘leapfrog’ across emerging markets within their physical proximity to venture outside their home region and culture. The firms that pursue this pattern have strong technological and marketing assets and strong partnering capabilities and are in industries outside retail or wholesale. Apparently, such firms tend to pursue a certain type of international profile, emphasizing markets like North America and Western Europe.

Since the shared components collapse our three dimensions of global dispersion, we cannot say whether regional, cultural, or institutional factors are the dominant explanations for this pattern, yet the orthogonal components do shed light on that question. Technological assets is a driver specifically of regional dispersion, all other types of expansion held constant, suggesting that these assets make Japanese firms more likely to locate, for example, in the U.S. than in Australia. Similarly, being in a high-tech industry and having extensive international experience makes firms less interinstitutionally dispersed, all other types of expansion held constant, suggesting that such firms tend to locate in Korea rather than in China, for example. Finally, knowledge seeking and being outside the retail business has a unique effect on intercultural expansion, all other types of expansion held constant. Firms with those characteristics will, therefore, tend to expand into culturally distant markets (e.g., into Australia rather than into Korea).

Robustness tests

As mentioned earlier, we replicated our models with different cutoff points for emerging and developed markets, using 40 and 60 percentile classifications rather than the median. Institutional dispersion had a Cronbach's alpha of 0.80 and 0.86, respectively, based on these two alternative definitions. The test of Hypothesis 1 led to the same conclusions, but in a stronger form; with both alternative cutoffs, the GDP-corrected level of interinstitutional dispersion was more than 140 percent (compared to 122 percent for the median definition). Hence, our use of the median is probably conservative, as it underestimates the extent to which Japanese firms go to emerging markets. The predictors of interinstitutional dispersion were also similar to the ones we found based on the median with a few changes in the significance levels. With the 40 percent cutoff (a more demanding definition of developed markets), the negative effect of marketing capabilities and international experience on interinstitutional dispersion became significant at the p < 0.05 level. With the 60 percent cutoff (a more liberal definition of developed markets), experience also became significant (although only at the p < 0.10 level), whereas the effect of high-tech industry became insignificant.

As a comparison with our variance decomposition analysis, we also ran a principal components analysis to derive three orthogonal (i.e., uncorrelated) factors from our three dependent variables. The first factor (capturing 64 percent of the variance) reflected the ‘dominant internationalization paradigm’ identified earlier, with positive loadings for regional and cultural dispersion and negative loadings for institutional. The regression on this factor confirmed the drivers we identified for this pattern, with all three capability variables, as well as knowledge seeking, having a positive impact and retail and wholesale industry having a negative one. The second factor (with 27 percent of the variance) may be considered a ‘soft’ global dispersion factor, with a high loading of the institutional dimension and a moderate loading of the cultural dimension. This factor was negatively influenced by international experience and positively influenced by knowledge seeking, and it was also lowered by being in a high-tech or retail/wholesale industry. The third factor (with 9 percent of the variance) distinguished between the regional and cultural dimensions, with positive loadings for the former and negative on the latter. Only knowledge seeking had a significant (negative) impact on this factor, driving firms to become less interregionally and more interculturally dispersed. Overall, these findings seem to mirror our earlier analyses.

Finally, we ran our models on a number of subsamples to test for the universality of our estimates. However, in most cases this reduced our sample size to the extent that most parameters became insignificant. The only clear effect emerging from these analyses was that, among the drivers of interregional and intercultural dispersion, marketing assets seemed to be more important in high-tech industries, and both marketing assets and partnering capabilities had a larger effect on firms with strong technological assets, perhaps reflecting a complementarity between technologies and other types of assets that can be used to bring these technologies to foreign markets. Conversely, international experience had a stronger effect in low-tech industries, perhaps because these industries require more local adaptation and, hence, a better understanding of foreign stakeholders.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

A common observation is that economic globalization is becoming ever more pervasive, driven by technological progress and political liberalization. Initially, this trend led to an expectation that where firms located their operations would become less relevant and that economic activity would become more evenly dispersed around the globe—firms would act as centrifuges, offshoring their operations and expanding their worldwide networks (Bhagwati, Panagariya, and Srinivastan, 2004). Yet, scholars have pointed out that the globalization of economic activity has not been evenly distributed at all (Dunning, 2000; Scott, 2001), an observation that can be traced as far back as Hymer (1972). Thus, even though the concept of location is a central component of international business, it has become clear that our understanding of MNE location strategy remains underdeveloped (Beugelsdijk, McCann, and Mudambi, 2010).

To address this shortcoming, researchers have begun to analyze the nature of location more carefully to address the seeming paradox of lower barriers to international activity that, nonetheless, are associated also with uneven economic development. What we have learned in this literature stream is that regional boundaries appear to matter a great deal since firms are able to transfer and assimilate FSAs more easily within their home regions as compared to the rest of the world where the managerial challenge becomes substantially greater (Rugman and Verbeke, 2004). Where productive assets are located has a significant role in the factors that influence the transfer and combination of firm-specific assets and, ultimately, firm performance. As discussed by Rugman et al. (2011), these challenges become greater as distance increases, whether economic, cultural, institutional, or geographic. As internationalizing firms must cross multiple boundaries that must be managed simultaneously, it is critical to reflect on the different components as they combine and interact. Our research, therefore, was designed to unpack these components to ascertain their relative impacts on MNE organization.

We find with our sample that firms have significant home region bias in their global dispersion patterns, a result that reinforces earlier work (e.g., Asmussen, 2009; Rugman and Verbeke, 2004). Further, our results indicate that certain firm-specific resources influence the global dispersion of the firm's foreign operations; in particular, we find that strong proprietary capabilities in technology propel firms beyond their home regions. Thus, we find that regional borders are an inflection point for the internationalization process, leading to the existence of an interregional LoF. However, our contribution goes beyond those earlier studies by unpacking the concept of regions by simultaneously testing the intercultural and interinstitutional expansion of our firms and finding that our MNEs are relatively better at overcoming distinct cultural and, especially, institutional biases.

Our findings contribute to our understanding of outsidership in several ways. First, the moderately high level of intercultural dispersion supports the view that cultural barriers are more or less fluid social constructions that can be partly (but apparently not fully) overcome or that the costs of intercultural dispersion are mostly counterbalanced by the benefits of locating in different cultures. This is consistent with Ottaviano and Peri (2006), for example, who found that cultural diversity is conducive to innovation as it brings together diverse viewpoints and complementary approaches. This search for knowledge diversity may stimulate firms to expand across cultural boundaries (Cantwell and Janne, 1999) while staying within their home region so as to minimize their exposure to LoF. The fact that knowledge seeking has an effect specifically on intercultural dispersion supports this view. An important qualification based on our results, however, is that the MNE's industry is relevant since retail and wholesale firms were found to be less interculturally dispersed, all else being equal, suggesting that the ability to satisfy shopping preferences is a location-bound FSA that is less fungible across cultural space.

Another contribution to our understanding of outsidership is derived from our finding of a very high level of institutional dispersion, suggesting that our firms expand in a way that does not reflect an interinstitutional LoF. While this is consistent with our theoretical predictions, the fact that firms in our sample are more than 100 percent dispersed in that dimension may seem to need more explanation. However, considering the negative correlation between institutional dispersion on the one hand and cultural and regional dispersion on the other (suggestive of the two distinct internationalization paths available to Japanese firms), this finding may not be puzzling. In particular, in expectation of a relatively high interregional and intercultural LoF, Japanese MNEs may wish to avoid culturally and geographically distant markets. This would leave them to pursue growth opportunities in neighboring countries—which also happen to be emerging markets—not because they do not face a LoF in these markets, but because that LoF is relatively low. Hence, our findings provide insight into the relative LoF in the three key dimensions of distance.

Our study also informs the growing literature addressing the antecedents and consequences of LoF. Very few studies have provided a systematic attempt at unbundling the dimensions of the LoF as we do here (see, e.g., Eden and Miller, 2004 for an exception). On a conceptual level, we find that the LoF can be decomposed into regional, cultural, and institutional components, a simple procedure which nevertheless has profound implications both for the theoretical underpinnings of the LoF and for the empirical implications for MNEs' geographic scope. Yet, our article provides a counterpoint to Eden and Miller's (2004: 9) argument that the ‘hard costs’ of doing business abroad are less important than the ‘soft costs’ because the former ‘can be anticipated and measured, and may well be finite.’ On the contrary, we argue that—precisely because hard costs can be measured—managers anticipate hard costs and choose to stay out of markets where they are seen as too high. Hence, while soft sources of LoF may be significant for firms that are already internationalized into distant cultures and institutional environments, the hard sources provided by physical distance appear to be more important as predictors of MNE internationalization patterns, which is our preoccupation in this article.

Another contribution of our article is that we examine the concept of partnering capabilities in the context of MNE location strategy. According to Johanson and Vahlne (2009), it is not the LoF per se that matters, but rather being an outsider to relevant business networks in new local contexts. As argued by Rugman et al. (2011), the presence and the strength of the relationships between the MNE and relevant actors operating in the host environment has been one specific parameter related to the distance dimension that has been neglected systematically—relational capabilities may well be a key missing factor to overcome compounded distance and to allow access to coveted resources in host regions. Thus, we contribute to this literature by providing a theoretical elaboration and empirical test, finding that MNE skill in working within alliances is indeed related to interregional dispersion. Further, partnering capabilities have an equally strong effect on intercultural dispersion, suggesting that MNEs more accomplished in working within alliances are more capable of addressing intercultural LoF given that a local partner may provide an understanding of host country culture. This finding reinforces Hutzschenreuter et al.'s (2011) discussion that the key challenge in international expansion is that becoming an insider in local networks is a complex, timing consuming, and difficult task.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

Overall, our study indicates that it is useful, and indeed necessary, to break down MNE geographic scope into multiple dimensions of location. One might expect a positive relationship between all these dimensions since large international operations enable the MNE to build managerial capacity for handling and reducing the LoF and, therefore, may be a precursor to a global strategy. Yet, our findings suggest that they are also distinct properties of geographic scope and we even find intercultural and interregional dispersion to be inversely relative to interinstitutional dispersion, suggesting that firms need to make hard choices of how to allocate scarce managerial resources to their expansion in these three dimensions. Hence, whereas prior literature has implicitly treated geographic scope as a one dimensional variable ranging from local to global (perhaps with regional as intermediate point), we show that firms appear to internationalize in highly idiosyncratic ways. Furthermore, since firms' internationalization patterns differ in a systematic fashion, it would seem that we need such a multidimensional approach in our studies of internationalization—for example in the internationalization-performance literature 8—which has been criticized for being more preoccupied with ‘how much’ than with ‘how,’ the MNEs expand across borders (Hennart, 2011).

While our intent was to advance the current debate about MNE geographic configuration, our study has limitations that warrant further research into this topic. For example, we are using a cross-sectional sample to draw inferences about relationships between our variables. Our model shows that the data is, in fact, largely consistent with the hypothesized relationships between the variables in our framework; nevertheless, our empirical design did not enable us to test issues of directionality or time lags in these relationships. Hence, it is important to consider potentially important alternative explanations for our results, including ones in which the causality runs in the other direction. For example, while our theoretical argument is that firms with partnering capabilities are less susceptible to the LoF, we also know that firms enter into partnerships with foreign firms in an effort to overcome the LoF they experience as a result of their international expansion. An alternative or perhaps complementary explanation, therefore, is that firms have diverse partners because they have expanded to distant places. A similar interpretation is possible with respect to intangible assets that also might be derived from diverse locations; for example, do MNEs learn from research institutions and demanding customers in the foreign markets they enter?

While these ideas are not incompatible with our theory, future research should build on panel data in order to corroborate or challenge the results we have presented here and to provide more definitive tests of causality. Also, it would be useful to update our findings with more recent data (to assess, for example, whether the recent downturn in global economic growth has an impact on MNE internationalization patterns). Finally, it is not clear whether country of origin plays a significant role in the propensity to engage in globally dispersed location strategies. This is particularly important given the different manifestations of our two-by-two-by-two dispersion cube in Figure 1 and given that distance is inherently asymmetric (Shenkar, 2001; Tung and Verbeke, 2010). For example, even though Japanese firms do not seem to suffer high interinstitutional LoF in emerging markets and do not seem to need proprietary assets to enter these markets, it is possible that emerging market firms would find it more difficult to enter Japan or that they would need different sets of capabilities to do so. More cross-regional, cross-cultural, and cross-institutional research should be designed to address such issues.

As our study indicates, feasible international strategic options may be broader than previously believed. MNEs are apparently characterized by a wide range of different internationalization patterns. The question raised in recent studies is whether globally oriented firms are pursuing flawed strategies or whether they are merely different from regional firms in ways which enable them to exploit the potential of globalization. Our findings point to the latter explanation; we found that the possession of intangible assets enable knowledge-exploiting firms to globalize across geographic and cultural space rather than across institutional space. This suggests that the idea of the interconnected world is a seductive but perhaps dangerously misleading notion for MNEs and that managers as well as researchers should be increasingly concerned with the fit and the interactions between proprietary resources, firm strategy, and geographic scope.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References

We would like to thank Professors Doug Baer, Charles Dhanaraj, Nicolai Foss, Niron Hashai, Jasper Hotho, Takehiro Isobe, Will Mitchell, Torben Pedersen, Bent Petersen, and Alan Rugman for helpful comments on earlier drafts of this article. The article has also benefited from interaction with seminar audiences at Duke University, Reading University, and the University of Victoria. Finally, we gratefully acknowledge the comments of the anonymous reviewers and the guidance of Associate Editor J. Myles Shaver, as well as the research assistance of Zijun Zhang. Any remaining errors and omissions are our own.

Footnotes
  1. 1

    There is a debate in the IB literature as to whether cultural and institutional distances are identical, overlapping, or distinct constructs (Peng and Pleggenkuhle-Miles, 2009). Indeed, the broad definition of institutions includes cognitive and normative pillars, and national culture clearly is related to cognition and norms. Because of this, we believe that the more narrow definition of institutions in the tradition of North (1990)—which emphasizes the ‘hard’ governance embodied by political, economic, and regulatory institutions—is a more useful complement to the cultural dimension.

  2. 2

    Economic theory has traditionally held that tariffs could increase FDI if they led firms to substitute exports for local production (so-called horizontal ‘tariff-jumping FDI,’ see, e.g., Smith, 1987). However, more recent research suggests that tariffs in fact inhibit FDI, much of which is vertical (e.g., ‘offshoring’ or sales-related investments) and, therefore, more of a complement to than a substitute for trade (Cuevas, Messmacher, and Werner, 2005).

  3. 3

    Interestingly, the correlation between our institutional development score and the logarithm of GDP per capita is 0.86. This suggests that it is impossible to separate governance from economic development and, indeed, we can consider these two aspects as being key to the distinction between emerging and developed markets.

  4. 4
  5. 5
  6. 6

    Since we have the same independent variables in all three regressions, the estimated coefficients in the SUR are identical to the ones resulting from three independent regression models. However, due to the error correlation in SUR, the standard error, significance of the estimates, and significance of the model itself (as reported in the table) are different.

  7. 7

    Note that the SUR and shared components regressions are all significant, whereas two of the orthogonal component regressions are not. However, this can be ascribed to a combination of our limited sample size and the large number of insignificant control variables in these two equations—indeed, when backward elimination is applied, both equations become significant and their significant regressors remain. We nevertheless report the full models to retain the identity by which each SUR coefficient is the sum of the two regression coefficients.

  8. 8

    For an overview of this literature, see Bausch & Krist (2007).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Conceptualizing and Operationalizing Firm Globalization
  5. Internationalization, Proprietary Assets, and the Liability of Foreignness
  6. Methodology
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgments
  11. References
  • Asmussen CG. 2009. Local, regional, or global? Quantifying MNE geographic scope. Journal of International Business Studies 40: 11921205.
  • Asmussen CG. 2012. Foundations of regional versus global strategies of MNEs. In Handbook of Research on International Strategic Management, Verbeke A , Hemant H (eds), Edward Elgar: Cheltenham, U.K.
  • Asmussen CG, Foss N, Pedersen T. Knowledge transfer and accommodation effects in multinational corporations: evidence from European subsidiaries. Journal of Management. Forthcoming. Available at: http://intl-jom.sagepub.com/content/early/2011/10/13/0149206311424316.full.pdf+html (accessed 12 November 2012).
  • Banalieva E, Dhanaraj C. Home region orientation in international expansion strategies. Journal of International Business Studies. Forthcoming.
  • Barney J. 1991. Firm resources and sustained competitive advantage. Journal of Management 17: 99120.
  • Bartlett C, Ghoshal S. 1989. Managing Across Borders: The Transnational Solution. Harvard Business School Press: Boston, MA.
  • Baum JAC, Calabrese T, Silverman BS. 2000. Don't go it alone: alliance composition and startups' performance in Canadian biotechnology. Strategic Management Journal 21(3): 267294.
  • Baum JAC, Oliver C. 1991. Institutional linkages and organizational mortality. Administrative Science Quarterly 36: 187218.
  • Bausch A, Krist M. 2007. The effect of context-related moderators on the internationalizationperformance relationship: evidence from meta-analysis. Management International Review 47: 319347.
  • Benito G, Gripsrud G. 1992. The expansion of foreign direct investments: discrete rational location choices or a cultural learning process? Journal of International Business Studies 23: 461476.
  • Berry H, Guillén M, Zhou N. 2010. An institutional approach to cross-national distance. Journal of International Business Studies 41: 14601480.
  • Beugelsdijk S, McCann P, Mudambi R. 2010. Place, space and organization: economic geography and the multinational enterprise. Journal of Economic Geography 10: 485493.
  • Bhagwati J, Panagariya A, Srinivastan T. 2004. The muddles over outsourcing. Journal of Economic Perspectives 18: 93114.
  • Brouthers K, Brouthers L. 2003. Why service and manufacturing entry mode choices differ: the influence of transaction cost factors, risk and trust. Journal of Management Studies 40: 11791204.
  • Buckley P, Casson M. 1976. The Future of the Multinational Enterprise. Macmillan: London, U.K.
  • Burt R. 1992. Structural Holes: The Social Structure of Competition. Harvard University Press: Cambridge, MA.
  • Cantwell J, Janne O. 1999. Technological globalization and innovative centers: the role of corporate technological leadership and locational hierarchy. Research Policy 28: 119144.
  • Chang S. 1996. International expansion strategy of Japanese firms: capability building through sequential entry. Academy of Management Journal 38: 383418.
  • Crawford J, Laird S. 2001. Regional trade agreements and the WTO. North American Journal of Economics and Finance 12: 193211.
  • Cuervo-Cazurra A, Genc M. 2008. Transforming disadvantages into advantages: developing-country MNEs in the least developed countries. Journal of International Business Studies 39: 957979.
  • Cuevas A, Messmacher M, Werner A. 2005. Foreign direct investment in Mexico since the approval of NAFTA. World Bank Economic Review 19: 473488.
  • Daiwa Institute of Research. 1999. Analysts' Guide. Daiwa Institute: Tokyo, Japan.
  • Delios A, Beamish PW. 1999. Geographic scope, product diversification, and the corporate performance of Japanese firms. Strategic Management Journal 20(8): 711727.
  • Dikova D, Van Witteloostuijn A. 2007. Foreign direct investment mode choice: entry and establishment modes in transition economies. Journal of International Business Studies 38: 10131033.
  • Dow D, Karunaratna A. 2006. Developing a multidimensional instrument to measure psychic distance stimuli. Journal of International Business Studies 37: 578602.
  • Doz Y, Santos J, Williamson P. 2001. From Global to Metanational: How Companies Win in the Knowledge Economy. Harvard Business School Press: Boston, MA.
  • Dunning J (ed). 2000. Regions, Globalization and the Knowledge Based Economy. Oxford University Press: New York.
  • Dunning J, 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: 177199.
  • Eden L, Miller S. 2004. Distance matters: liability of foreignness, institutional distance and ownership strategy. In The Evolving Theory of the Multinational Firm: Advances in International Management, Hitt M , Cheng J (eds). Elsevier: Amsterdam, The Netherlands.
  • Fisch J, Oesterle M-J. 2003. Exploring the globalization of German MNCs with the complex spread and diversity measure. Schmalenbach Business Review 55: 221.
  • Frankel J, Rose A. 2002. An estimate of the effect of common currencies on trade and income. Quarterly Journal of Economics 117: 437466.
  • Galan JI, Gonzalez-Benito J. 2006. Distinctive determinant factors of Spanish foreign direct investment in Latin America. Journal of World Business 41: 171189.
  • Ghoshal S. 1987. Global strategy: an organizing framework. Strategic Management Journal 8(5): 425440.
  • Goerzen A. 2005. Managing alliance networks: emerging practices of multinational corporations. Academy of Management Executive 19: 94107.
  • Goerzen A. 2007. Alliance networks and firm performance: the impact of repeated partnerships. Strategic Management Journal 28(5): 487509.
  • Goerzen A, Asmussen CG. 2007. Geographic orientation of MNEs and implications for performance. In Regional Aspects of Multinationality and Performance, Rugman A (ed). Elsevier: Oxford, U.K.
  • Goerzen A, Beamish P. 2003. Geographic scope and multinational enterprise performance. Strategic Management Journal 24(13): 12891306.
  • Goerzen A, Beamish P. 2005. The effect of alliance network diversity on multinational enterprise performance. Strategic Management Journal 26(4): 333354.
  • Gómez-Mejia L, Palich L. 1997. Cultural diversity and the performance of multinational firms. Journal of International Business Studies 28: 309336.
  • Govindarajan V, Gupta A. 2001. The Quest for Global Dominance. Josey-Bass: San Francisco, CA.
  • Gupta V, Hanges P, Dorfman P. 2002. Cultural clusters: methodology and findings. Journal of World Business 37: 1115.
  • Gupta V, House R. 2004. Understanding leadership in diverse cultures: implications of project GLOBE for leading international ventures. In Leading in High Growth Asia: Managing Relationship for Teamwork and Change, Tjosvold D , Leung K (eds). World Scientific Publishing: Singapore.
  • Hamel G, Prahalad C. 1985. Do you really have a global strategy? Harvard Business Review 63(4): 139148.
  • Harbison J, Pekar P. 1998. Smart Alliances: A Practical Guide to Repeatable Success. Jossey-Bass: San Francisco, CA.
  • Henisz W. 2000. The institutional environment for multinational investment. Journal of Law, Economics, & Organization 16: 334364.
  • Hennart J-F. 1982. Theory of Multinational Enterprise. University of Michigan Press: Ann Arbor, MI.
  • Hennart J-F. 1991. The transaction costs theory of joint ventures: an empirical study of Japanese subsidiaries in the US. Management Science 37: 483497.
  • Hennart J-F. 2011. A theoretical assessment of the empirical literature on the impact of multinationality on performance. Global Strategy Journal 1(1/2): 135151.
  • Hennart J-F, Park Y-R. 1994. Location, governance, and strategic determinants of Japanese manufacturing investment in the United States. Strategic Management Journal 15(6): 419436.
  • Hitt M, Dacin T, Levitas E, Arregle J-L, Borza A. 2000. Partner selection in emerging and developed market contexts: resource-based and organizational learning perspectives. Academy of Management Journal 43: 449467.
  • Hofstede G. 1976. Nationality and espoused values of managers. Journal of Applied Psychology 61: 148155.
  • Hofstede G. 1980. Culture's Consequences: International Differences in Work-Related Values. SAGE Publications: Beverly Hills, CA.
  • Hummels D. 2007. Transportation costs and international trade in the second era of globalization. Journal of Economic Perspectives 21: 131154.
  • Hussler C. 2004. Culture and knowledge spillovers in Europe: new perspectives for innovation and convergence policies? Economics of Innovation and New Technology 13: 523541.
  • Hutzschenreuter T, Voll J. 2007. Performance effects of ‘added cultural distance’ in the path of international expansion: the case of German multinational enterprises. Journal of International Business Studies 39: 5370.
  • Hutzschenreuter T, Voll J, Verbeke A. 2011. The impact of added cultural distance and cultural diversity on international expansion patterns: a Penrosean perspective. Journal of Management Studies 48: 305329.
  • Hymer S. 1972. The multinational corporation and the law of uneven development. In Economics and World Order from the 1970s to the 1990s, Bhagwati J (ed). Collier-Macmillan: New York.
  • Hymer S. 1976. The International Operations of National Firms: A Study of Direct Foreign Investment. MIT Press: Cambridge, MA.
  • Jaffe A, Trajtenberg M, Henderson R. 1993. Geographic localization of knowledge spillovers as evidenced by patent citations. Quarterly Journal of Economics 108: 577598.
  • Javidan M, Dorfman P, Luque M, House R. 2006. In the eye of the beholder: cross cultural lessons in leadership from Project GLOBE. Academy of Management Perspectives 20: 6790.
  • 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 8: 2332.
  • Johanson J, Vahlne J-E. 2009. The Uppsala internationalization process model revisited: from liability of foreignness to liability of outsidership. Journal of International Business Studies 40: 14111431.
  • Kaufmann D, Kraay A, Mastruzzi M. 2004. Governance matters III: governance indicators for 1996, 1998, 2000, and 2002. World Bank Economic Review 18: 253287.
  • Khanna T, Palepu K, Sinha J. 2005. Strategies that fit emerging markets. Harvard Business Review 83(6): 6376.
  • Kim C, Chung J. 1997. Brand popularity, country image and market share: an empirical study. Journal of International Business Studies 28: 361386.
  • Kogut B, Singh H. 1988. The effect of national culture on the choice of entry mode. Journal of International Business Studies 19: 411432.
  • Kogut B, Zander U. 1993. Knowledge of the firm and the evolutionary theory of the multinational enterprise. Journal of International Business Studies 24: 625646.
  • Kostova T, Zaheer S. 1999. Organizational legitimacy under conditions of complexity: the case of the multinational enterprise. Academy of Management Review 24: 8298.
  • Leeson PT, Sobel RS, Dean A. 2006. Contagious capitalism. Working paper, West Virginia University, Morgantown, WV.
  • Levy D. 1997. Lean production in an international supply chain. Sloan Management Review 38: 94102.
  • Luo Y, Tung R. 2007. International expansion of emerging market enterprises: a springboard perspective. Journal of International Business Studies 38: 481498.
  • Mezias JM. 2002. Identifying liabilities of foreignness and strategies to minimize their effects: the case of labor lawsuit judgments in the United States. Strategic Management Journal 23(3): 229244.
  • Mezias SJ, Chen Y-R, Murphy P, Biaggio A, Chuawanlee W, Hui H, Okumura T, Starr S. 2002. National cultural distance as liability of foreignness: the issue of level of analysis. Journal of International Management 8: 407421.
  • Morck R, Yeung B. 1991. Why investors value multinationality. Journal of Business 64: 165187.
  • Morck R, Yeung B. 1992. Internalization: an event study test. Journal of International Economics 33: 4156.
  • Nachum L. 2003. Liability of foreignness in global competition? Financial service affiliates in the city of London. Strategic Management Journal 24(12): 11871208.
  • Nachum L. 2011. The home-based advantages and a hierarchy of location resources: foreign and local firms dependency on location resources. In Dynamics of Globalization: Location-Specific Advantages or Liabilities of Foreignness? Asmussen CG , Pedersen T , Devinney T , Tihanyi L (eds). Emerald Group: Bingley, U.K.
  • Narula R. 2012. Do we need different frameworks to explain infant MNEs from developing countries? Global Strategy Journal 2(3): 188204.
  • North D. 1990. Institutions, Institutional Change, and Economic Performance. Cambridge University Press: Cambridge, U.K.
  • Osegowitsch T, Sammartino A. 2008. Reassessing (home-) regionalisation. Journal of International Business Studies 39: 184196.
  • Ottaviano G, Peri G. 2006. The economic value of cultural diversity: evidence from US cities. Journal of Economic Geography 6: 944.
  • Page S. 2007. Making the difference: applying a logic of diversity. Academy of Management Perspectives 21(4): 620.
  • Peng M, Pleggenkuhle-Miles E. 2009. Current debates in global strategy. International Journal of Management Reviews 11: 5168.
  • Penner-Hahn J, Shaver JM. 2005. Does international research and development increase patent output? An analysis of Japanese pharmaceutical firms. Strategic Management Journal, 26(2): 121140.
  • Petersen B, Pedersen T. 2002. Coping with liability of foreignness: different learning engagements of entrant firms. Journal of International Management 8: 339350.
  • Redding G. 1976. Some perceptions of psychological needs among managers in South-East Asia. In Basic Problems in Cross-Cultural Psychology, Poortinga Y (ed). Swets & Zeitlinger: Amsterdam, The Netherlands.
  • Ronen S, Shenkar O. 1985. Clustering countries on attitudinal dimensions: a review and synthesis. Academy of Management Review 10: 435454.
  • Roth K, Morrison A. 1992. Implementing global strategy: characteristics of global subsidiary mandates. Journal of International Business Studies 23: 715735.
  • Rugman AM, Verbeke A. 1992. A note on the transnational solution and the transaction cost theory of multinational strategic management. Journal of International Business Studies 23(4): 761771.
  • Rugman AM, Verbeke A. 1993. Foreign subsidiaries and multinational strategic management: an extension and correction of Porter's single diamond framework. Management International Review 33: 7184.
  • Rugman AM, Verbeke A. 2004. A perspective on regional and global strategies of multinational enterprises. Journal of International Business Studies 35: 318.
  • Rugman AM, Verbeke A. 2005. Towards a theory of regional multinationals: a transaction cost economics approach. Management International Review 44: 517.
  • Rugman AM, Verbeke A. 2008. A new perspective on the regional and global strategies of multinational services firms. Management International Review 48: 397411.
  • Rugman AM, Verbeke A, Luxmore S. 1990. Corporate strategy and free trade agreement: adjustment by Canadian multinational enterprises. Canadian Journal of Regional Science 13: 307330.
  • Rugman AM, Verbeke A, Nguyen Q. 2011. Fifty years of international business theory and beyond. Management International Review 51: 755786.
  • Saxenian A, Hsu J-Y. 2001. The Silicon Valley-Hsinchu connection: technical communities and industrial upgrading. Industrial and Corporate Change 10: 893920.
  • Scherer F. 1983. The propensity to patent. International Journal of Industrial Organization 1: 107128.
  • Schilke O, Goerzen A. 2010. Alliance management capability and alliance portfolio performance: conceptualization and measurement. Journal of Management 36: 11921219.
  • Scott A. 2001. Global City-Regions: Trends, Theory, Policy. Oxford University Press: Oxford, U.K.
  • Shenkar O. 2001. Cultural distance revisited: towards a more rigorous conceptualization and measurement of cultural differences. Journal of International Business Studies 32: 519535.
  • Smith A. 1987. Strategic investment, multinational corporations and trade policy. European Economic Review 31: 8996.
  • Stahl G, Maznevski M, Voigt A, Jonsen K. 2009. Unraveling the effects of cultural diversity in teams: a meta-analysis of research on multicultural work groups. Journal of International Business Studies 41: 690709.
  • Stein E, Daude C. 2007. Longitude matters: time zones and the location of foreign direct investment. Journal of International Economics 71: 96112.
  • Stephan U, Uhlaner L. 2010. Performance-based vs socially supportive culture: a cross-national study of descriptive norms and entrepreneurship. Journal of International Business Studies 41: 13471364.
  • Stopford J, Wells L. 1972. Managing the Multinational Enterprise. Basic Books: New York.
  • Stuart T, Hoang H, Hybels R. 1999. Interorganizational endorsements and the performance of entrepreneurial ventures. Administrative Science Quarterly 44: 315349.
  • Tajfel H. 1982. Social psychology of intergroup relations. Annual Review of Psychology 33: 139.
  • Tihanyi L, Griffith D, Russell C. 2005. The effect of cultural distance on entry mode choice, international diversification, and MNE performance: a meta-analysis. Journal of International Business Studies 36: 270283.
  • Toyo Keizai. 1999. Kaigai Shinshutsu Kigyou Souran-Kuni Betsu. Toyo Keizai: Tokyo, Japan.
  • Tung R, Verbeke A. 2010. Beyond Hofstede and Globe: improving the quality of cross-cultural research. Journal of International Business Studies 41: 12591274.
  • Vermeulen F, Barkema H. 2002. Pace, rhythm, and scope: process dependence in building a profitable multinational corporation. Strategic Management Journal 23(7): 637653.
  • Williams K, O'Reilly C. 1998. Demography and diversity in organizations: a review of 40 years of research. Research in Organizational Behavior 20: 77140.
  • Yip G. 2002. Total Global Strategy: Managing for Worldwide Competitive Advantage. Prentice Hall: Upper Saddle River, NJ.
  • Zaheer S. 1995. Overcoming the liability of foreignness. Academy of Management Journal 38: 341363.
  • Zaheer S, Mosakowski E. 1997. The dynamics of the liability of foreignness: a global study of survival in financial services. Strategic Management Journal 18(6): 439463.