Institutional and resource based drivers as predictors of the degree of inter- and intra-regional liabilities of foreignness : the case of the mining industry

(1)

Institutional and resource based

drivers as predictors of the

degree of inter- and intra-regional liabilities of foreignness:

the case of the mining industry

Master Thesis

MSc. Business Studies – International Management

Supervisor: Dr. Johan P. Lindeque

Second reader: Francesca Ciulli MSc

Student: Rossana Playfair

Student ID: 5927595

Date: 20 June 2015

Word count: 18,513 (excl. tables)

(2)

Abstract

This study analyses the international strategies of four mining MNEs using data in the period from 2002 to 2012. Evidence is found that inter- and intra-regional liabilities of foreignness that prohibit the MNE in its internationalization can be circumvented through a combination of political resources, technological knowledge, and home country effects. These findings are consistent with previously studied concepts like institutional voids and firms specific assets; if the degree of institutional voids between two countries is roughly the same, this allows for a better understanding and thus more cooperation while doing business, with certain FSAs being the critical components in this process. When the degree of voids is unequal this leads to difficulties in undertaking business activities in a host environment, possibly increasing political tensions and discontinuity threats to existing activities. Here FSAs can be used to soften relations and gain access or provide technological knowledge to local partners as a means to build political resources and lower the degree of inter- and intra-regional liabilities of foreignness.

Keywords: [regionalization, institutions, LoF, assets, FDI motives, mining, MNE, emerging markets, home country]

Acknowledgements

My special thanks go to my supervisor Dr. Johan Lindeque for his guidance, encouragement and enduring patience. The thesis would not be what it is had it not been for his exquisite supervision. I would also like to thank my mother for her motivating words and the support she provided throughout the sometimes difficult process. I am very satisfied with the end result and realize I could not have done it without their help.

(3)

Statement of originality

This document is written by Student Rossana Playfair who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating

it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(4)

Index of Tables and Figures

Table 1: Sample cases………...27

Table 2: Collected documents and sources………...28

Table 3: Codebook……….…….….29

Table 4: Diagnosing the home country: Australia’s institutional context…….………..32

Table 5: Regionalization overview BHP Billiton………...….34

Table 6: Diagnosing the home country: England’s institutional context………….………...37

Table 7: Regionalization overview Rio Tinto Group………..….…40

Table 8: Diagnosing the home country: England’s institutional context………...….42

Table 9: Regionalization overview Anglo American………...……...45

Table 10: Diagnosing the home country: Brazil’s institutional context…………..………...48

Table 11: Regionalization overview Vale S.A………..……….….51

Table 12: Cross-case comparison……….56

Table 13: Results for the working propositions………...…..59

(6)

1. Introduction

During the internationalization of most firms a tendency exists to stick to strategies traditionally employed in the home country; a result of the insufficiency of domestically developed capabilities to generate equal value in all host countries (Khanna et al., 2005). Thus capabilities partly reflect the diffused skills and institutional strengths of the firms’ countries of origin (Kogut, 1991). International competition then results from comparing management and societal institutions of countries (Ghoshal, 1987). These factors also determine the organizational architecture of the firms inhibiting the respective institutional contexts (Peng, 2002). Thus when comparing the internationalization process of firms, similarities and differences exist as a result of home country legacies that can hinder cross-border activities of firms in present times.

For this reason the regionalization literature has introduced the notion of inter- and intra-regional liabilities of foreignness (LoF) (or the degree to which a multinational enterprise is prohibited in taking its activities across borders) (Rugman and Verbeke, 1994). Here inter-regional LoF is the difficulty of moving operations outside of the home region, while intra-regional LoF concerns the difficulty of moving operations outside the home country into neighboring countries that often share institutional similarities or have entered trade agreements (Rugman and Verbeke, 1994). Sethi and Judge (2009) have in turn introduced a framework that not only takes the (degree of) liabilities into account but also the assets (or advantages) a foreign company encounters when setting up activities in a host location. The implication here is that the types of costs and advantages that arise in a host country, are determined by differences in the institutional frameworks of the home and host country and the firm specific advantages (FSAs); those unique resources owned and/or controlled by the firm allowing it to overcome external disadvantages (Rugman and Verbeke, 1992) and thereby circumvent institutional differences between countries and ultimately operate profitably in its host locations.

Rugman and Oh (2012) however do not support host country effects as a critical explanation of variations in performance. They claim that it concerns a small effect and therefore region and industry factors are better suited to explain most of a firm’s international activity. So although it is acknowledged that more external influences need to be taken into account when studying the regionalization of firms, they suggest managers do this “by not only looking at the firm’s capability and national competitiveness, but also at its industry

(7)

peers and the regional conditions to determine its geographic venue” (Rugman and Oh, 2012, p. 12).

The findings of Makino et al. (2004) on the other hand show that national contextual factors do indeed influence firm behavior and economic performance, with host country and industry effects found to be of equal importance in determining foreign affiliate’s performance. The relevance of the effects is however tied to the level of development of the particular country. Thus corporate effects tend to be more critical in explaining subsidiary’s performance variations in developed countries, while country and industry effects are proven to be more salient in less developed countries or emerging markets. Here an emerging market is defined by how well an economy helps buyers and sellers come together (Khanna and Palepu, 1997). So in order to account for the performance of emerging market MNEs (EM-MNEs), country effects should also be taken into consideration as the country of origin may exhibit institutional voids (Khanna et al., 2005) that correspond with those found in host locations, lowering the degree of inter- and intra-regional LoF as a result. It is in this assumption that differences in the internationalization process of EM-MNEs and developed countries MNEs become apparent.

While companies have proved to be capable of successfully exploiting activities outside the boundaries of the home country or region, international scholarship will need to continue studying what prevents so many MNEs from achieving the same success outside of the home region to become truly global. Currently, the prohibitions to achieve this global stature are not completely understood, although transaction costs have been identified as key drivers inhibiting firms from profitably operating in a host market (North, 1991; Rugman and Verbeke, 1992; Williamson, 2000). In the international business (IB) literature transaction costs have been described as all costs (e.g. controlling performance, establishing a network of suppliers) a firm encounters in various aspects of the value added chain (Pan and Tse, 2000). The degree of inter- and intra-regional LoF for a large part captures these transaction costs found within and across regions.

Often times developed MNEs set up operating activities in host countries that are institutionally distant from the home country due to a lack of development in the ‘soft infrastructure’ (specialized intermediaries, regulatory systems, and contract enforcing mechanisms) (Khanna and Palepu, 2006). These institutional voids in developing host countries can then raise the transaction costs for MNEs, resulting in the need to change strategies and simultaneously determine the most suitable form to do business in a host location (e.g. joint venture, acquisition) (Pan and Tse, 2000).

(8)

Therefore, when analyzing the degree of inter- and intra-regional LoF between developed and developing economies, the institution based view has been broadly accepted as a reliable approach to study how differences between macro-level institutions (e.g. country-level legal and regulatory frameworks) influence the transaction costs MNEs experience (Meyer et al., 2009).

The inclusion of macro-level institutions is furthermore supported by Luo (2002) and Schlie and Yip (2000) under the guise that more studies need to incorporate macro-level analysis and regional dynamics which play a determining role in a firm’s abilities to succeed in foreign locations. Adding to this macro-level argument, home country effects in combination with FSAs make for a stronger company in the international arena (Rugman and Oh, 2012). Furthermore, in Porter’s (1990) home country diamond the international success of MNEs in a particular industry partly comes down to a home country context that is forward looking, challenging and dynamic. These home nations are said to gain their competitive advantage (CA) within an industry due to demanding buyers pressuring companies to innovate faster and thus achieve a more sophisticated CA than foreign rivals. Peng (2002) in addition highlights the importance of institutional influences on business strategies. Here, a dynamic interplay between the home country institutional context and the firms that emerge within this context, results in available strategies for the MNE.

Building upon these avenues for future research in the IB field, the current study applies the notion of inter-and intra-regional LoF and the encountered assets and liabilities (Sethi and Judge, 2009) in the mining industry. While the literature has made a distinction between difficulties experienced in either the service or manufacturing sector (Rugman and Verbeke, 2008), the mining sector, a capital intensive and mature industry characterized by its market seeking and natural resource seeking FDI motives, has not yet been studied in regards to the regionalization arguments. In this industry, regionalization and the degree of FSA transferability within the strategic business units (SBUs) (a single or a set of affiliates, located in various nations and organized in a geographic or product division (Rugman and verbeke, 2008)), determine the selected host countries and entry modes for these markets. This study therefore applies the inter- and intra-regional LoF notions to the conceptualization of this industry, where the transferability of FSAs, resource and market seeking FDI motives, high capital intensity and strong resource nationalism shape the industrial arena and influence whether and how these MNEs overcome their inter- and intra-regional LoF. The focus here is on how institutional factors influence the transaction costs of MNEs in distant host environments, thus determining not only their entry mode decision, but also the availability of

(9)

FSAs to overcome this distance thereby lowering the degree of inter- and intra-regional LoF. By combining these gaps and propositions above the following research question emerges:

“How do MNEs in the mining sector use FSAs to overcome the effects of institutional differences between the home and host countries inherent in the inter- and intra-regional dispersion of their operations?”

To address this question and make appropriate contributions to the IB literature, it is assumed that differences between home and host countries emerge from the respective institutional frameworks and FSA transferability and these differences affect the type of liabilities and

assets the MNE encounters thereby determining the degree of inter- and intra-regional LoF. This study attempts to contribute insight on what can be gained from having particular home

country advantages, and in case of a lack of these advantages, where in the MNE (network) these necessary skills can be developed. The managerial implications will shed light on which capabilities are needed in what location and how they can best be combined, thus providing firms with insights about their potential to overcome inter- and intra-regional LoF.

Theoretical lenses

To gain a more thorough understanding of the drivers and impediments of inter- and intra-regional LoF this study will adopt two theoretical lenses; the institution-based view (IBV) and the resource based view (RBV), conceptualized through FSAs (Rugman and Verbeke, 1992). These lenses respectively facilitate our understanding of the context in which MNEs operate, but also explain the opportunities and limitations MNEs face as a result of their unique histories, causal ambiguities, and social complexities (Barney, 1991). Incorporating FSA transferability (Rugman and Verbeke, 1992) allows an understanding of how MNEs use these FSAs built in the home country, to succeed in the international industry. By using both lenses the study acknowledges the complementarity of the perspectives, as the institutional lens accounts for the similarities and differences experienced in various host countries, while the FSAs having emerged in the institutional home context determine how well the MNEs are able to either overcome impediments or take advantage of drivers in the internationalization process and thus institutional barriers of host countries.

(10)

Research design and data

The study adopts a qualitative multiple-case study with embedded units of analysis as a research design. Three regions in the world, containing MNEs from both developed and developing countries are compared and contrasted. These regions are: Latin America, Asia Pacific, and Europe. In these regions, four MNEs active in the mining industry have been theoretically sampled.

This industry is an interesting one to study because the nature of the business dictates that differences between MNEs come down to the particular FSAs present and the degree to which these are able to be transferred to markets outside of the home country and region. But there is another interesting dynamic at play within this industry; a large part of the constraints MNEs face when internationalizing come from a geographic point of view as natural resources are unevenly distributed globally. The role of the government in host countries with natural resources deposits thus becomes an important source of LoF. Here we see the IBV at work and the degree of transferability of the FSAs providing challenges and opportunities that are unique to this industry.

The data collection is over a period of ten years, with 2002 as the base year. Furthermore, to increase the validity of the study, data triangulation in the form of newspaper articles, company documents, and database research is used.

The remainder of this study first presents the conceptual foundation. Here current and seminal academic work in the regionalization field is evaluated, compared and contrasted. This is followed by conceptualizing the mining industry and the emergence of working propositions that form the basic assumptions used in the data collection and analysis. The methodology chapter presents the theoretically sampled cases and their embedded units, the research philosophies, the quality criteria of the study, and a presentation of the data collection process. What follows is the analysis and discussion chapters where the findings are linked with the working propositions either proving or disproving them. Lastly, the conclusion summarizes the aim of the study, findings and the theoretical and managerial implications.

(11)

2. Literature review

The following section provides insight into past and recent work that contributes to the academic conversation around the globalization and regionalization debate.

The focus lies on the overall shift in IB literature from globalization and the resulting global strategies (Kogut, 1987; Ghoshal, 1989; Collins 1991) to the more recent regional strategies (Rugman and Verbeke, 2004; Rugman and Oh, 2012; Douglas and Craig, 2011). It is believed that the costs of doing business abroad (Hymer, 1976) or more specifically the degree of inter- and intra-regional LoF prohibit the MNE to achieve a truly global platform (measured by the distribution of sales and assets) and instead lead to MNEs mostly operating in a regional fashion (Rugman and Verbeke, 2004).

Zaheer’s (1995) study exploring the costs of engaging in foreign activity and how these differ between companies in different locations, concerns whether as industries globalize foreignness will continue to carry costs. Years later as a response to this inquiry, Sethi and Judge (2004) show that not only do these costs continue to exist but they have an even greater complexity than initially proposed. To account for this complexity, the liabilities but also the assets of being foreign and operating in the MNE network are drawn up in a new framework thus extending the conversation on the costs of doing business abroad.

Currently, the literature does not have a straightforward answer for which factors facilitate the MNE in overcoming inter- and intra-regional LoF. Meyer et al (2009) therefore state that questions on how country-level legal and regulatory frameworks influence transaction costs have been relatively unexplored. They argue that “markets for acquiring local resources may be suboptimal because of the institutional environment governing the transaction, but may also be suboptimal because of the characteristics of the sought resources” (Meyer et al., 2009, p 63) as in the mining industry. Before further inquiry into these factors, a closer look will first be taken in the development of the IB literature over the years.

2.1 Global and regional strategies

A natural starting point in the regionalization literature is the phenomenon of globalization. Garrett (2000) makes a strong argument for the presence of globalization in the current information age. He contends that contemporary globalization is driven by different factors, depending on which part of the system is under study. His overall conclusion is that scholars in the IB field have proposed the usage of global strategies in order for firms to deal with increasing market integration.

One of these scholars, Kogut (1985), introduced generic modes of international

(12)

competition, with the third mode stating that global strategies were determined through the interplay of comparative advantages of countries and the value added chains of firms. Solutions on how to implement a global strategy were also proposed: through the development of a broad product portfolio and a strong world-wide distribution system (Prahalad and Hamel, 1987) or by globalizing the international strategy through integration across countries (Yip, 1989). Furthermore, foundations of global strategies have been based on national differences and scale and scope economies (Ghoshal, 1989), analysis from an RBV perspective (Collins, 1991) and the perceived procedural justice of the process through which these strategies are generated (Kim and Mauborgne, 1991). Lastly, Kim and Hwang (1992) mapped global strategic considerations (global concentration, synergies and strategic motivations) to determine entry mode choice.

Rugman et al. (2011, p. 772) however contend that “global strategies and truly global non-location bound FSA bundles are likely to be extremely rare, as demonstrated by the regional nature of MNEs in terms of their (limited) sales and asset dispersion, and the regional elements in their governance structures”. Meanwhile Ghemawat (2003) departs from finding theoretical explanations to understand globalization and instead dives deeper into the practical outcomes of the phenomenon on a macro-level. He suggests that markets are not integrated enough to think of them in global terms and therefore fall into a state described as “semi globalization” (Ghemawat, 2003). Similar to one of Garrett’s (2000) drivers of globalization, the assumption is that political regulations may prohibit the complete convergence of markets and thus a theoretical extreme of integration allowing us to think of globalization as being trade and international investments without limitations has not been achieved.

In order to compete in multiple locations one must consider that certain key activities are location-specific, thus making it difficult to move into new home and especially host regions. Ghemawat (2003) calls this the ‘location-specificity’ of doing business abroad. The main point here is that the world should not be viewed as one large market. Foreign destinations should be evaluated on a case-by-case basis and operations located in different markets should be streamlined as efficiently as possible but are unlikely to share the exact same strategy.

Lending their support for the semi globalization argument, Rugman and Verbeke (2004) argue for a regional level of analysis to assess the MNE’s internationalization. Their study on MNEs in Forbe’s Fortune 500 shows that while firms are categorized as global companies, the majority of sales are in fact restricted to specific geographic locations, with a good amount of these sales concentrated in the home regions. These MNEs are somehow

(13)

unable to replicate their home region sales elsewhere. This then suggests that their business models and strategies are also home region based (Rugman, et al., 2011).

It is for this reason Rugman et al., (2011) believe that the country level of analysis needs to move towards a regional level, so that the impact of differences in for example culture, political regulation, economic and financial institutions can be better understood and may lead to a higher ability to overcome inter- and intra-regional LoF.

Explaining regionalization strategies

In the globalization and regionalization debate, Rugman and Verbeke (2004) believe that a truly global company is active in all three regions of the extended triad: European Union (EU), Asia pacific (Asia), and North America (NAFTA). To measure the degree of global capability of the largest MNEs they distinguish between four types of strategies: home region oriented, bi-regional, host region oriented and global. Most MNEs add value by using the benefits that accompany entering markets similar to the home country. Restrictions when accessing less familiar host environments are the result of inter- and intra-regional LoF (Rugman and Verbeke, 2004).

Liabilities of foreignness are explained as “the costs of doing business abroad that result in a disadvantage for the MNE subsidiary – all additional costs a firm operating in a market overseas incurs that a local firm would not” (Zaheer, 1995, p 342). Zaheer (1995) identifies four sources of LoF: costs due to spatial distance (transportation; coordination over distance/time zones), firm-specific costs (unfamiliarity, lack of roots), host country environment costs (lack of legitimacy, economic nationalism) and home country environment costs. The way these costs are dealt with by firms differs per industry (Henderson et al., 1995).

The inter- and intra-regional LoF thus refer to the difficulties firms encounter when transferring FSAs to host environments. These FSAs can be highly location bound and in order to deploy them in host countries will have to be recombined with for example country specific assets (Rugman and Verbeke, 1992). Because host countries in the home region usually have less compounded distance, intra-regional LoF will often be lower than inter-regional LoF due to fewer recombination investments as identified by Rugman et al. (2011).

As previously mentioned Sethi and Judge (2009) developed a framework based on Zaheer’s (1995) LoF and Hymer’s (1976) costs of doing business abroad concepts. The framework distinguishes between liabilities and assets by dividing them between those emerging within the host country environment (the liabilities or assets of foreignness) and

(14)

those originating outside of this context due to the dispersed MNE network (the liabilities or assets of multi-nationality). Costs that emerge in the host country environment are subject to local governmental and regulatory influences and can be divided into discriminatory and incidental LoF. Here discriminatory LoF are described as barriers used to discourage foreign firms entering a host country through hostility or governmental restrictions. Incidental costs entail the lack of knowledge of the foreign firm about the host market. These costs usually diminish after some experience in the host market (Sethi and Judge, 2009). The framework furthermore assumes that the costs and benefits of doing business abroad consist of firm specific ones and those common to a set of firms (e.g. within an industry).

The aforementioned concepts form the basis of the regionalization arguments in the rest of the study. But in order to find the roots of LoF in a wider practical setting the institution based view (IBV) and the resource based view (RBV) are used. Together they determine the available avenues for MNEs through their workings in the home contexts MNEs emerge from and their subsequent success due to an ability to circumvent certain institutional barriers by using their FSAs. These lenses are discussed in the following section. 2.2 Theoretical perspectives

The ability to operate in multiple foreign locations at once and increase efficiency throughout an MNE operational chain requires not only knowledge of the institutional environment of the host country, but also the ability of the firm to decrease its LoF (Zaheer, 1995). In the IBV this is established through familiarity with the formal and informal rules in a host country. The RBV accounts for the FSAs inherent in the SBUs forcing them to take different internationalization paths due to their diverse needs in natural resources and market seeking FDI motives (Dunning, 1998). By combining these perspectives the study accounts for the different levels of importance of the country, industry and corporate effects which depend on the economic development in home and host countries (Makino et al., 2004).

2.2.1 The Institution Based View

This study adopts North’s (1991) definition of institutions as it is deemed more suitable than Scott’s (2003) regulative, normative and cognitive framework. Important in North’s definition is the constrained interaction between humans and institutions to prevent wealth-maximizing tendencies and asymmetrical information. The transaction costs that firms encounter when doing business in foreign markets therefore become focal.

(15)

According to North (1991) transaction costs are determined by the institutional architecture and the effectiveness of rule enforcement. Institutions are described as “humanly devised constraints that create order and vow to reduce uncertainty in exchange” (North, 1991, p 97), and are elementary in the reduction of transaction and production costs so that potential gains from trade and FDI can be realized (North, 1991).

The IBV serves as the macro-level determinant deciding the degree of inter- and intra-regional LoF between home and host countries. This perspective considers the differences between legal and regulatory frameworks and how the transaction costs are affected, thereby determining the degree of inter- and intra-regional LoF. Peng (2002) in addition identifies possible indicators of an efficient institutional framework that serve as a basis for comparison, they are: a credible legal framework, stable political culture and functioning strategic factor markets.

Rugman and Verbeke (2008) furthermore propose that regional institutions that lower barriers to economic activity effectively lower the degree of intra-regional LoF relative to the inter-regional LoF. So, when evaluating MNEs attention must be paid to both the home and host country institutions, as well as those on a regional level (e.g. European Union).

Home country effects

Although the relevancy of the use of macro-level data in the global and regional debate is questioned, the country of origin’s institutional framework plays a determining role in shaping the MNE, albeit in differing degrees and depending on the economic development of the home country. Porter (1990) attests that the success of nations in particular industries comes down to a home country environment that challenges, looks forward and contains dynamism. Therefore, the lagging development of the economy in resource rich countries, supports the theory that the economy of a nation depends not on the inheritance of wealth but on the ability of industries to innovate and upgrade (Porter, 1990).

Rugman and Verbeke (2004) however argue that favourable diamond conditions in the home country may be insufficient in most cases to permit a truly global expansion. As a result factors supporting competitive advantages must be highly specialized (just schooling is not enough) and often require sustained investment to create. In turn when companies within an industry face a selective disadvantage (a lack of local raw materials) they have no choice but to upgrade and innovate, supporting the market seeking and natural resource FDI motives (Dunning, 1998). A connection then emerges between the home and host country, where the MNE increases its inter-regionalization by taking advantage of the developed home diamond

(16)

in the search of upstream assets (exploration and exploitation), and the host location which often lacks a strong home diamond but has an abundance of natural resources (Moon et al., 1998). And this lack of development can serve as an advantage to the MNE with market and natural resource seeking FDI motives (Dunning, 1998), as a host country becomes easier to access for these capital abundant outsiders due to a self-interested political state, no formal rules or informal check-up of faulty agreements (Auty, 2001).

Institutional voids

The role of institutions in a market economy is essential in supporting the effective functioning of the market mechanism, so that firms and individuals can engage in transactions without incurring undue costs or risks (Feinberg and Gupta, 2009). But not every country possesses an efficient institutional framework. These countries are found to have an increased importance of country level effects that influence the way transactions are done and by whom (Makino et al., 2004). When the MNE subsidiary moves into a host market with high compounded distance transaction costs will likely increase.

Firms that select new operational markets often use tools like country portfolio and political risk assessment, which focus on the potential profits from doing business in a host country but leave out essential information about the soft infrastructure; institutions necessary to support basic business activities (Khanna et al. 2005). And while emerging markets are hardly uniform, in varying degrees most fall short in providing these soft infrastructures (Khanna and Palepu, 1997). The absence of these infrastructures are referred to as institutional voids and can be identified in the capital, labor and product market, government regulations and openness (Khanna and Palepu, 1997). These institutional voids may lead to market failure for the developed MNEs due to the inability of handling these voids as a result of unfamiliarity, thus increasing the degree of inter-regional LoF. Although less familiar with institutional voids, the overall strength of the developed market MNE lies in its brand name, sophisticated technology, state-of-the-art innovation systems, access to vast reservoirs of talent and the ability to raise capital at a low cost due to well-established financial markets (Khanna and Palepu, 2006).

For the EM-MNEs these voids are easier to overcome due to a familiarity since inception; managers through years of experience know how to circumvent these voids and even use them as business opportunities (Khanna and Palepu, 2006). EM-MNEs have thus developed FSAs for overcoming the disadvantages associated with institutional voids. They are for instance able to raise money from the stock market on their good reputation and when

(17)

thy have demonstrated a degree of success, these firms can raise money by listing on foreign stock exchanges (e.g. NYSE) and tap into the capital and labor market of developed countries (Khanna and Palepu, 2006). This is in line with the authors suggestion that good governance is particularly valuable in emerging economies. (Khanna and Palepu, 2006).

Firms can map the institutional context of host locations and evaluate the extent to which strategies should be adapted to each nation’s institutional context (Khanna et al. 2005), because “firms that take the trouble to understand the institutional differences between countries are likely to choose the best markets to enter, select optimal strategies and make the most out of operating in emerging markets” ( Khanna et al. 200, p 5). After familiarity has been achieved Cantwell et al., (2010) suggest that MNEs can either engage in institutional adaptation, mostly seen in developed host markets or in institutional co-evolution as seen in developing host countries. This then increases their ability to overcome inter-regional LoF.

As mentioned, the EM-MNEs have fared better than their developed MNE counterparts in the circumvention of institutional voids and the tailoring of strategies to local markets. Thus it can be assumed that initially EM-MNEs take on foreign competitors by capitalizing on their ability to navigate their home turf and in doing so lower their degree of intra-regional LoF (Khanna and Palepu, 2006).

2.2.2 The Resource Based View

The RBV is conceptualized as the FSAs found in the SBUs (Rugman and Verbeke, 1992). Here both tangible and intangible resources are seen as ultimate sources of competitive advantage (Chang and Rhee, 2011). While tangible resources (plant, equipment, people) are held primarily at the subsidiary level, intangible resources (financial, organizational and reputational) are held at the firm level (Rugman et al., 2011). As a result the influence of FSAs in the MNE network supports Rugman and Verbeke´s (2004) arguments for the incorporation of firm-level determinants when explaining regional strategies.

Transferability FSAs

Due to transferability difficulties, the literature categorizes FSAs as location bound (LB FSA) and non-location bound (NLB FSA) (Rugman and Verbeke, 1992; 2004). The NLB FSAs are exploited globally and can lead to benefits of scale and scope or the exploitation of national differences (Rugman et al, 2011).

The LB FSAs can often times more easily be augmented to deploy in the home region, thus resulting in regional strategies lowering the degree of intra-regional LoF (Rugman and

(18)

Verbeke, 2007). Examples of these kinds of FSAs include: superior marketing and distribution skills, access to raw materials and intangible resources such as proprietary technology and the ability to achieve vertical and horizontal integration (Rugman et al., 2011).

But when market expansion is sought, LB FSAs will need to be combined with NLB FSAs for which so called ‘melding’ or ‘linking investments’ are needed (Rugman, 2005, p 225). These are the adaptation investments required to operate profitably in a host country environment (Rugman and verbeke, 2008).

However the degree to which these investments are necessary depends on the asset-specificity of SBUs as not all parts of the MNE share the same regional profile. As a result host location choices will be diverse and influence the LoF encountered. It is assumed that establishing part of the value chain in a host location, (resource seeking FDI motive), is institutionally more cumbersome than selling or exporting to a host location. But the demand in developing host locations is often not as sophisticated as in the developed world (Porter, 1990), thus affecting the market seeking FDI motive. The assumption therefore is that the difference between inter- and intra-regional LoF will be greater in upstream (exploration and exploitation) market seeking FDI than downstream market seeking FDI (selling).

Political FSAs

One of the latest developments in the internationalization theory is the concept of liabilities of outsidership instead of foreignness (Johanson and Vahlne, 2009). Outsidership here refers to the lack of access to resources and contends that relationships (a LB FSA) may be the missing factor in overcoming compounded distance and taking full advantage of host regions (Rugman et al., 2011).

Political resources are defined as “any firm attributes, assets, human resources, or any other resources that allow the firm to use the political process to improve its efficiency and profitability” (Frynas et al., 2006, p 324). They argue that firm-specific political resources probably bare a greater importance in emerging economies, but are still relevant for developed countries as they are difficult for rivals to match, and thus provide a source of competitive advantage. Political FSAs can thus mitigate risk in certain host markets, thereby lowering the degree of inter- and intra-regional LoF.

The resources are divided into physical capital resources (e.g. formal corporate nationality of the firm, home country context), human capital resources (managers familiarity with

(19)

emerging economies) and organization capital resources (relationship between managers and policy makers) (Frynas et al. 2006).

Frynas et al. (2006) furthermore find that governmental influence is considered a political resource. The argument here is that through interventions governments are able to create very effective (and sometimes the most effective) barriers to enter an industry. So, governmental differences are another way the costs and by extension the degree of regionalization are prone to either de- or increase. In addition Calhoen (2002) argues that levels of corruption explain the degree of institutional differences between countries and thus showcase the efficiency of governments. As a result of higher corruption levels, transaction costs are affected and the uncertainty of doing business within a country increases.

2.2.3 Integration of the Resource and Institution Based View

To better understand the degree of inter- and intra- regional LoF in capital intensive mature industries, the interaction between the IBV and RBV (conceptualized by FSAs) needs to be considered.

The IBV and RBV shed light on the interdependent factors in the mining industry. From an IBV perspective the MNE network may help to counteract idiosyncrasies of a weak institutional context, more frequently seen in emerging markets (Meyer et al, 2009). But it is the transferability of FSAs within the network that ultimately determines to what degree these idiosyncrasies are counteracted. Therefore the FSAs compliment the IBV by providing insight on how the MNE can leverage its own assets and the home country advantages to lower the degree of inter- and intra-regional LoF.

In contrast, while institutions in a host country can impede the entry of foreign MNEs through discriminatory Lof (Sethi and Judge, 2004), political FSAs can provide access into these institutional contexts through pre-established relationships between policy makers and firms, as political leaders are eager to work with companies (Khanna and Palepu, 1997). This lowers transaction costs and results in a decrease of the inter- and intra-regional LoF.

Furthermore developing host countries are often times characterized by institutional voids that increase the inter-regional LoF of the developed MNE and consequently affect the entry mode. Oviatt and McDougal (2005) for example argue that if market imperfections are present (resulting from institutional voids) MNEs will enter the market through wholly owned subsidiaries to protect against volatilities in the local environment. (Pan and Tse, 2000) in turn contend that the MNE is likely to enter a host market using equity investments (equity joint venture, Greenfields or acquisitions) to lower the transaction costs that high institutional

(20)

distance imposes. But the high amount of capital (a FSA) needed for these investments is more likely to be found in firms from developed markets (Khanna and Palepu, 2006). This shows that the two perspectives own a complementary dimension that explains the need to apply both perspectives to understand regional strategies in the focal industry.

Now that the theoretical foundation is in place, the following industry conceptualization identifies the factors affecting all mining MNEs and their strategies regardless of home region and FSAs owned.

2.3 Industry conceptualization

The conceptualization of the mining industry facilitates our understanding of the industry effects that determine the necessary FSAs to have successful operations. This conceptualization in combination with the previous academic conversation also provide a foundation from which the working propositions emerge. These propositions provide direction in gathering and organizing data to empirically establish if and how well the study is supported.

Because the industry deals with specific geographical restrictions companies are in competition over a piece of business limited in size. In order to achieve a market leader position then, companies must not only own critical (NLB) FSAs partly determined by the home country, but are dependent on additional factors such as political and regulatory frameworks in host markets that facilitate lobbying with governments and local groups.

To enhance the understanding of how MNEs manage the same variables within an industry in different ways (due to the country of origin’s institutional background providing advantages or liabilities to its firms operating across borders and the individual MNE’s LB and NLB FSAs) a characterization of the mining industry is presented. These characteristics serve as the conceptualization of the industry. Jamasmie (2014a) identifies the following characteristics of the mining industry,

(1) resources and market seeking FDI motives, (2) high capital intensity,

(3) resource nationalism, and

(4) tight regulation vs. ineffective environmental and government policy.

(21)

Natural resource seeking and market seeking FDI motives:

Developed MNEs are often from home countries lacking in natural resources but with access to sophisticated demand markets (market seeking FDI), thereby providing the MNE with an incentive to find these resources outside of the home country (Auty, 2006). Natural resources are often times found in less developed host countries with high distance to the home country, while demand markets are predominantly located in developed countries (Auty, 2001). This upstream resource seeking FDI (exploration and exploitation of new input markets) then depends on the asset specificity that exists within the SBUs. As a result, host location choice differs for each SBU, leading to different degrees of LoF experienced. This results in the first working propositions:

WP1a: The global distribution of natural resources and lower level of development in the host country results in the upstream resource seeking FDI exhibiting a higher degree of intra-regional LoF compared to the market seeking FDI for developed MNEs

WP1b: The global distribution of natural resources and lower level of development in the host country results in the upstream resource seeking FDI exhibiting a lower degree of inter-regional Lof compared to the market seeking FDI for developed MNEs

High capital intensity:

MNEs are more likely to have managerial and financial resources that allow them to invest in majority-owned subsidiaries (Ramamurti, 2001). These resources not only enable them to undertake drawn out negotiations with host country governments, but also allows complete control over the operations by choosing an equity entry mode (Pan and Tse, 2000) while simultaneously leveraging their influence on the institutional arena these operations are located in (Cantwell et al., 2010). Thus the following proposition emerges:

WP 2: Due to their size MNEs will enter host markets through wholly owned subsidiaries and use institutional co-evolution to lower the degree of inter-regional LoF

Resource nationalism and institutional voids:

Due to the nature of the industry, high capital intensity and FDI motives are relevant to the success of firms. The literature shows that regional and industry effects also matter in determining foreign affiliates’ performance; while Makino et al. (2004) argue that country

(22)

effects also still make a difference, especially in emerging economies. Differences between the institutional home context and the host context due to institutional voids (Khanna et al., 2005), can then affect the degree of intra- and inter-regional LoF. Political resources (Frynas et al, 2006) however can lessen the degree of intra- and inter-regional LoF. This leads to the following propositions:

WP 3: Less distance between the home and host country lowers resource nationalism, leading to local market access and lower intra-regional LoF relative to inter-regional LoF WP 4: Political resources can facilitate a lower degree of inter-regional LoF resulting from

high institutional distance between the home and host country Tight regulation vs. ineffective environmental and government policies:

Firms in saturated markets (e.g. Australia) deal with tight and clearly developed regulation. Emerging markets on the other hand encounter ineffective home policies and government interventions that not only increase the inter- and intra-regional LoF of the developed MNE, but also hinder the EM-MNEs flexible reactions to the economic conditions in the country, making regulatory decisions harder to predict (Khanna and Palepu, 1997). This increasing the inter-regional LoF. In contrast, countries in the home region are often part of the same trading block and abide by the same environmental agreements through regional institutions that lower the barriers to economic activity, thus effectively lowering the degree of LoF (Rugman and Verbeke, 2004). Thus the last working proposition emerges:

WP 5: Due to a lack of similarity in the formal institutional environments between the home and host country, as a result of environmental regulations and the amount of government intervention, the degree of intra-region LoF will be lower than the inter-regional LoF

2.3.1 Conclusion

As the working propositions have been established to organize and guide the collection and analysis of the data on mining MNE’s international strategies, the following chapter sets out the research methods used. Here the philosophies that frame the study are explained followed by a presentation of the theoretically sampled cases and the analyzing strategy used for the collected data.

(23)

3. Methodology

To understand the perspective from which the study is undertaken this chapter describes the assumptions held by the researcher about the field of study and the used methods of data collection. Furthermore a general introduction to the cases and their embedded units of analysis that make up the multiple case study design is provided. These form the empirical basis for the data analysis.

3.1 The ontological and epistemological assumptions

The ontological and epistemological assumptions of the researcher determine what is seen as valid, legitimate contributions to the existing theory and facilitates the researcher’s understanding of the best way to approach the study of a particular field of activity (Brannick and Coghland, 2007).

Ontology refers to the nature of social reality. It is concerned about “what exists, what it looks like, what units make it up and how these units interact with each other (Grix, 2002). In this study an objectivist view of ontology is adopted. This perspective assumes that social and natural reality can exist independently of human cognition (Brannick and Coghlan, 2007).

The epistemological positioning of a study addresses what constitutes acceptable knowledge in a field of study (Saunders et al, 2009). An objectivist view of epistemology facilitates objective access to the external world (Brannick and Coghlan, 2007). The goal is to achieve an explanation of the complexity of a phenomenon. And by taking an objective stance, research can be done from a detached position, describing the social phenomena observed. This is also termed a post-positivist perspective (Brannick and Coghlan, 2007).

By taking objective stances in the research philosophies the presence of an independent value-free researcher examining reality is achieved (Brannick and Coghlan, 2007).

3.2 Qualitative multiple case study design

The case study design allows for an in-depth exploration of a particular system policy or institution in a ‘real-life’ context by using a multitude of data sources to provide different perspectives on the same topic (Yin, 2009). Qualitative research methods seek to provide conclusions that account for the particulars in each case, with the goal to expand and generalize theory (Hyde, 2000). The cases and embedded units of analysis in this study therefore have a broad geographic dispersion to account for generalizability issues that often

plague case study research (Yin, 2009).

(24)

The case study also advocates the development of a theoretical framework or structure before the empirical data is collected, which corresponds with the deductive bottom-up theorizing approach (Hyde, 2000; Shepherd and Sutcliffe, 2011). Here, inductive and deductive research approaches are combined through predetermined propositions and categories which serve as the theoretical directions guiding the case study while allowing the unexpected insights to emerge from the materials (Shephard and Sutcliffe, 2011). These propositions and categorical divisions facilitate the reduction of the volume of data preventing the researcher to become overwhelmed by what the data shows (Yin, 2009; Shepherd and Sutcliffe, 2011). Problems with this approach however may stem from the inability to link concepts to reality. It is for this reason that great importance is placed on the validity, reliability and accurate measurements of concepts before findings can be accepted as knowledge (Brannick and Coghlan, 2007; Yin, 2009).

3.3 Quality criteria Internal validity

The internal validity of a study is concerned with whether a plausible causal argument is constructed that is convincing enough to defend the research conclusions (Gibbert and Ruigrok, 2010).This can be achieved by comparing conflicting literature and search for evidence to the ‘why’ behind relationships (Eisenhardt, 1989). This study will therefore use the working propositions in the previous chapter to serve as the basis from which conclusions are made regarding the empirical observations.

To ensure a clear and systematic process research should use “the refutation principle by using the constant comparative method, comprehensive data treatment, and deviant-case analysis” (Gibbert and Ruigrok, 2010, p 714). Here, the constant comparative method suggests that an attempt must be made to find another case through which the working propositions can be tested. Through the use of different units of analysis within the cases this study strives to account for this method. The comprehensive data treatment, states that generalization should be able to apply to every part of the collected data.

The data found in the annual reports will be entered into a spreadsheet where patterns, peculiarities, and differences are assessed. News articles and company documents will be assessed using the analytical tool NVivo.

(25)

Construct validity

Construct validity is concerned with the extent to which a procedure leads to an accurate observation of reality (Gibbert and Ruigrok, 2010). As this form of validity is relevant mainly during the data collection phase, it is important that a clear chain of evidence is established to allow the reader to reconstruct the process from the initial research questions to the final conclusions (Eisenhardt, 1989).

In order to provide a clear chain of evidence the remainder of this chapter highlights the characteristics on which these cases were selected, the different data sources from which empirical observations are made, and the process of coding the collected data based on the literature and empirical observations. The use of different data sources is one of the triangulation methods through which the grounding of theory is strengthened thereby further increasing the construct validity (Eisenhardt, 1989; Yin, 2013).

External validity

The external validity or generalizability deals with accounting for phenomena not only being valid in the setting in which they are studied but also in others (Saunders et al, 2009). The focus here lies on analytical generalization; a process that refers to the generalization from empirical observations to theory, rather than a population (Yin, 2013). Eisenhardt (1989) argues that case studies can be a starting point for theory development and suggests a cross-case analysis involving 4 to 10 cross-case studies that may provide a sound basis for analytical generalization. This study uses four cases that have been theoretically sampled and serve as a basis to make these empirical observations. From the academic literature and industry characteristics propositions have emerged which will guide the recognition and categorization of empirical observations to make meaningful generalizations about the literature. By using the home country context, in which all organizations therein have to interact with the institutional framework, the findings can increase in their generalizability for other industries within the same institutional context.

Reliability

Reliability refers to the absence of random error, enabling subsequent researchers to arrive at the same insights if they conducted the study along the same steps (Gibbert and Ruigrok, 2010). The main goal is to document the taken steps as clear as possible to facilitate future replication and enhance transparency. Reliability is furthermore enhanced by employing a clear and systematic documentation of the research process in the form of a case study

(26)

database to make as many steps operational as possible (Yin, 2009). This is done through the collection of data as part of ongoing research at the Amsterdam Business School on the regional presence of MNEs.

Another way to increase reliability is through the use of low inference descriptors (Gibbert and Ruigrok, 2010). This emerges in the form of quotes on the respective cases taken directly from company documents and news articles presented in the results chapter.

3.4 Research context: the mining industry

Regional strategies differ per industry, the industry of focus in this study calls for a great amount of business activities to be performed in host regions outside of the triad, in contrast to the focus of Rugman and Verbeke’s (2004) focus on triad regions inhibited by developed economies. The mining industry is an interesting field of study in the regionalization literature due to its asset- and industry specificity. The nature of the business makes the assumption that the differences between MNEs are in part determined by the FSAs inherent in the SBUs of each company. Furthermore, institutional factors may also play a significant role in the success of a company’s internationalization. Transferability of FSAs in and beyond the home region, high capital intensity, a mature industry and a need to move beyond regional markets are the factors at the foundation of this sector. But despite the industry’s inherent need for geographic dispersion, none of the four MNEs in this study are described as truly global companies. And while this could be attributed to the fact that MNEs pursuing sourcing strategies in host countries to facilitate sales in the home region are likely to have higher inter-regional dispersion of assets than sales (Rugman and Verbeke, 2008), this could change with the rapid development of emerging markets serving as upcoming demand markets.

3.4.1 Case selection

The sampling rationale is based on a selection of typical cases from the Forbes Global 500 list of largest multinational companies in 2012 (Forbes, 2014), where internationalization and home and host country influences are expected to play a significant role. Based on the theoretical sampling logic (Eisenhardt, 1989; Yin, 2009), cases are selected on their theoretical usefulness: “those that replicate or extend theory by filling conceptual categories” (Eisenhardt, 1989, p 533). The MNEs hail from a variety of home regions with different levels of economic development, which influences the available resources and consequently how and where activities are undertaken. This industry- and asset-specificity leads to contributions to existing theory (Rugman and verbeke, 2008; Barney, 1991).

(27)

The firms have furthermore been categorized on the basis of their regional orientations (Rugman and verbeke, 2004). The requirement for all cases under study is that they actively participate in the actual mining of natural resources such as iron ore (this requirement facilitates a general basis from where further analysis can be undertaken), excludes several units initially classified as mining companies. This exclusion corresponds with Eisenhardt (1989) who argues that in theoretical sampling the goal is to select those cases that are likely to replicate or extend the emergent theory.

The selected cases are BHP Billiton, Rio Tinto Group, Anglo American and VALE S.A. A brief overview of these firms is presented in table 1. The home regions consist of Latin America, Asia Pacific and European with two separate cases from the same home country. This however is not seen as a disadvantage because it allows more insight in how institutionally similar MNEs create, use, and locally recombine FSAs to overcome their LoF.

BHP Billiton hails from home country Australia. Characterized by its developed economy and abundance of natural resources, this country is a rare sight in the natural resources literature where the focus lies on resource rich countries with low economic development (Auty, 2006;). As a result BHP is not only able to exploit and extract within the home institutional context, but also enjoys the presence of a sophisticated demand market.

England as the home to Anglo American and Rio Tinto owns some natural resources (e.g. iron ore) but lacks others.. As a developed economy, it is able to support the cross-border activities of its firms. Rio Tinto as a British-Australian entity has the added advantage of attracting its main investors from two well developed nations in addition of two established demand markets.

(28)

Table 1. Sample cases

1

Home region: at least 50 percent of sales in the home region (Rugman and Verbeke, 2004). 2

Host region: more than 50 percent of sales in a region other than the home region (Rugman and Verbeke, 2004). 3

Bi-regional: at least 20 percent of sales in each of two regions, but less than 50 percent in any one region (Rugman and Verbeke, 2004). Year 2012 Home

Region

Country of Origin

Regional

Orientation Business Units Case description

Case 1: BHP Billiton

Asia

Pacific Australia Home region

1

Iron Ore Petroleum Base Metals

Diamonds Metallurgical Coal

Energy Coal

BHP Billiton is characterized as a diversified natural resources company. The corporate strategy places its focus on diversification through commodities, market and geography. As of June 30, 2012, the Company was working in more than 100 locations worldwide (BHP Billiton, 2014).

Case 2: Rio Tinto Group

Europe England Host region2

Iron Ore Aluminum Copper Coal Gold Diamonds and Minerals

Rio Tinto is described as a British-Australian international mining company with worldwide activities. As one of the biggest players in the mining industry Rio Tinto claims its position in Australia and North America with other businesses located in Asia, Europe, Africa and South America (Rio Tinto Group, 2014)

Case 3: Anglo American

Europe England Bi-regional3

Iron Ore and Manganese Metallurgical Coal Thermal Coal Nickel Copper Platinum Diamond

As the second mining company in England, Anglo American operates in the following host countries: Africa, Brazil, Chile, North and South America, Australia, China, India, Japan, and Europe (Anglo American, 2014)

Case 4: Vale S.A.

Latin

America Brazil Host region

Bulk materials Base materials

Fertilizers Logistics

Vale S.A. transferred from governmental control to the private sector. A publicly listed company whose shares are traded onthe stock markets of São Paulo, NewYork, Hong Kong, Paris and Madrid.

27

(29)

As a region rich in natural resources, Latin America has been an attractive destination for many firms. Brazil, which is currently one of the fast rising emerging economies, is bound to join the global arena, but per the literature policies within the country will determine if other firms will be able to follow Vale’s example, one of the few Brazilian MNEs in Forbe’s Fortune 500 list.

3.5 Data collection and data sources

To facilitate the quality of the study, archival analysis one of Yin’s (2009) research methods is used. Here archival analysis is seen in the form of newspaper articles on the theoretically sampled companies using the LexisNexis database, public (company) documents (e.g. sustainability reports, yearly accounts to identify greenfield activities and asset dispersion) and the Orbis database providing data on mergers, acquisitions and joint ventures from its Zephyr section (table 2). The data collection is based on a time-series approach over a period of twelve years, with 2002 serving as the base year.

Document sources

Documents Collected and Used

Annual accounts LexisNexis Orbis (Zephyr)

Available Used Available Used Available Used

Total BHP Billiton Rio TintoAnglo American Vale SA 44 12 10 12 10 32 8 9 10 5 2.801 672 650 802 677 111 30 31 28 22 1.642 472 562 436 172 264 36 47 71 110

Table 2. Collected documents and sources

The archival analysis supports the constant comparative method suggested by Gibbert and Ruigrok (2010) to test and compare the working propositions, while the four cases are the minimum number to provide a sound basis for analytical generalization increasing the external validity (Eisenhardt, 1989). Lastly, the varying data sources, the coding strategy and the collected data enhance the construct validity of the study (Yin, 2009).

3.6 Analytical strategy

Firstly, the individual cases will be analyzed. Here, the embedded units of analysis as focal host markets of the four mining firms are compared to develop the within-case analysis. The reason to analyze every case separately and withhold cross-comparisons until a later stadium

(30)

is to prevent making pre-mature conclusions (Eisenhardt, 1989). Subsequently the cases are compared and contrasted in the cross-case analysis, providing rival explanations, potential new insights (Eisenhardt, 1989) and possible support for the working propositions. This procedure of doing a multiple case study ascribes to Yin’s (1994) development of a ‘replication logic’.

In order to categorize the collected data the coding strategy is on a priori basis. This means that the choice of thematic codes is driven by the research question and based on the conceptualization of the industry, the institutional path dependency of MNEs in their home country and those institutional contexts they have chosen to operate in (table 3). This ensures a systematic approach to organizing and analyzing the data.

Nodes and subnodes Working Propositions Descriptions

Market seeking FDI

Demand New markets

Resource seeking FDI

Exploration Exploitation Resource nationalism WP1a WP1b WP4

The need for new market exploration and exploitation; with ease of access being determined by regional ties and FDI motivation. Entry modes Greenfield Acquisition Joint venture Merger WP2 WP4

The most adequate entry mode gauged by a nation’s resource nationalism and the MNEs prowess.

FSAs

Political resources Technological know how Bargaining power

WP2

WP3

The size and strength of the MNE can be leveraged in order to circumvent LoF in host markets.

Institutional voids

Markets (capital, labor)

Openness

Political regulations

WP2 WP3 WP5

To counteract inter-regional LoF MNEs can use their FSAs, but the similarity between countries in the region provides greater ease of access.

Table 3. Codebook

The thematic codes furthermore ensure the categorization of a vast amount of narrative texts (e.g. newspaper articles, annual accounts). To facilitate this the analytical software tool NVivo is usedfor a systematic analysis and increasing reliability. This tool allows data to be reduced for efficient analysis, while also allowing patterns to emerge leading to potentially new ideas on the interaction between home and host country institutions (Miles and Huberman, 1984).

(31)

The following chapter presents the within-case findings. Based on their regional orientations, the MNE’s strategy becomes clear. By analysing a selection of events in the focal host markets patterns, peculiarities, deviations from initial assumptions and differences become clear. From these forced comparisons new categories and concepts which were not anticipated can emerge (Eisenhardt, 1989).

Institutional and resource based drivers as predictors of the degree of inter- and intra-regional liabilities of foreignness : the case of the mining industry