THE END JUSTIFIES THE MEANS:

MASTER’S THESIS

THE END JUSTIFIES THE MEANS:

THE SPRINGBOARD PERSPECTIVE AND THE UNIMPORTANCE

OF BOUNDARY CONDITIONS

BY

PETER VAN DIJK

SUPERVISED BY

G. LE

S. R. GUBBI

UNIVERSITY OF GRONINGEN

MSC INTERNATIONAL BUSINESS AND MANAGEMENT

ABSTRACT

EMNEs that adopt risky and aggressive internationalization strategies, where they pursue knowledge-based assets to overcome latecomer disadvantages and technological deficiencies, ignore unfavorable boundary conditions: the end (acquisition of knowledge-based assets) justifies the means (high-risk ventures).

The empirical research consisted of regression analysis with moderation. The core of the model, which is the relationship between knowledge-based assets and its effect on OFDI Stock, was moderated by common spoken language, colonial history and legal systems, and preferential trade agreements. The sample consisted of 447/445 country pairs, with Brazil, The Russian Federation, India, China, and South Africa (BRICS) as home countries and 36 developed countries as host countries. Country pairs were included, based on the availability of the data. The data was collected over a period of five years (2008-2012), with all variables lagging one year behind on the dependent variable.

The results of my empirical work are clear: only common spoken language was significant but negative in relation with knowledge-seeking behavior and OFDI of the BRICS countries, and all the other boundary conditions yielded insignificant results. In retrospect, I acknowledge that common spoken language was probably negative, due to the characteristics of the data. China, economically one of the most active countries in the world, scores extremely low on common spoken language, while educated Chinese who are active in international business, do speak English.

The boundary conditions did not have the hypothesized effects, which leads me to conclude that boundary conditions are unimportant for EMNEs pursing knowledge-based assets. For future research, I suggest finding empirical proof for the explanations offered in this thesis, as this thesis did not offer the scope to test these explanations empirically. Moreover, I think it would be valuable to study the BRICS countries individually, as there are many differences between the five countries this research focused on. Lastly, the most recently gathered data is from 2012, and I believe it is interesting to see whether more recent data will show different results.

TABLE OF CONTENTS

I. Introduction ... 4

II. Literature Review... 5

General theories on FDI ... 6

Emerging market internationalization frameworks ... 7

The Springboard Perspective ... 8

Research gap and Research question ... 10

III. Hypotheses Development ... 11

Knowledge-based assets and Springboard behavior ... 11

Language ... 12

Colonial History and Legal Systems... 13

Preferential Trade Agreements ... 15

IV. Methodology ... 16 Data ... 16 The model ... 17 Variables ... 17 Dependent variable ... 17 Independent variable ... 17 Moderating variables ... 18 Control variables ... 18

Testing for assumptions ... 19

V. Results ... 20

VI. Discussion and Conclusion ... 22

Limitations, Further research ... 25

Acknowledgments... 26

VII. References ... 26

I. INTRODUCTION

After significant growth in 2016, the flows of foreign direct investment (FDI) from emerging economies towards developed economies kept rising from 5 percent to $1 trillion (UNCTAD, 2017). Nowadays, this growth is contrary to the overall decline in FDI, with a global drop of 2 percent to $1.75 trillion. The share of FDI towards developed countries accumulates to 59 percent of total FDI. One of the primary influencers of these trends are the internationalization strategies of emerging market multinational enterprises, referred to in the literature and in this thesis as EMNEs.

The relationship between EMNEs and investments in developed markets can be explained by the springboard perspective (Luo & Tung, 2007). According to the springboard perspective, EMNEs use outward FDI (OFDI) as a springboard to acquire strategic assets needed to compete more effectively against global rivals and to avoid the institutional and market constraints they face at home. Luo and Tung’s work follows the rationale of Mathews (2002), who argues that EMNEs behavior cannot be explained by existing theory but by theories that acknowledge that EMNEs are different from advances market multinationals (AMNEs).

Although the springboard perspective has generated many follow-up studies, such as Li, Li, and Shapiro (2012) and Gubbi, Aulakh, Sougata, Sarkar, and Reveendra (2010), few studies have focused on the boundary conditions of the relationship between EMNEs’ knowledge seeking behavior and their springboard FDI. This thesis will focus on the influence of common spoken language, the effects of colonial history and common legal systems, and preferential trade agreements on the relationship between the availability of knowledge-based assets and FDI from the BRICS countries. By focusing on this gap in the literature, I aim to expand the existing knowledge on EMNEs behavior and its ties to certain cultural aspects.

Contrary to my initial hypothesis that these boundary conditions would have a positive influence on the relationship between the knowledge-based assets in the host country and FDI from EMNEs, it turned out that the boundary conditions can be classified as unimportant. Springboard behavior is qualified as a risky and aggressive strategy; therefore, I argue that due to the nature of this internationalization strategy, the end justifies the means. EMNEs are willing to accept the risks of Springboard behavior because the outcome is highly rewarding.

For EMNEs, it is all about the result; the acquisition of knowledge, and not about hedging risks by incorporating boundary conditions in the decision-making process. The EMNEs approach, which is vastly different from Western firms (Johanson & Vahlne, 1977), requires new theories and deserve the attention of scholars to fully understand EMNEs’ behavior. Where the boundary conditions were previously unresearched, I can now provide the answers that this was done rightfully so. This paper allows the research community to focus on the internationalization strategy of the EMNEs itself and not the boundary conditions.

Thirdly, I will present the methodology and its results. The last section will be a discussion and conclusion.

II. LITERATURE REVIEW

The three main pillars of the International Business (IB) literature are the industry-based view (Porter, 1979), the resource-based view (Barney, 1991), and the institutional-based view (Peng, 2002). This paper follows the rationale of the most recent theoretical avenue, arguing that institutional factors play a role in the strategy of a firm. First, the IBV will be reviewed, followed by the springboard perspective, and literature on the moderating effects.

The institution-based view provides a framework to understand why strategies of firms from different countries and regions differ (Peng, 2002: 251). The IBV builds on the thought process of certain scholars, who argued that firms have to take into account broader influences from governments and society when developing their strategy (DiMaggio & Powell, 1991; Oliver, 1997). These broader influences are more generally referred to as the institutional frameworks of a country or region (North, 1990), where the institutions ‘are the rules of the game in a society or, more formally, are the humanly devised constraints that shape human interaction’ (North, 1990: 3). The IBV is relevant for all firms since every firm is affected by the institutions in which it is embedded (Peng, 2002). The urge for the legitimacy of MNEs motivates the firms to isomorph with the local environments, without increasing efficiency (Yiu & Makino, 2002). Sethi, Guisinger, Ford, & Phelan (2002) even argue that the need for institutional factors in the FDI theory can hardly ever be exaggerated.

There are three dominant institutional approaches in the IB literature: New institutional economics, new organizational institutionalism, and comparative institutionalism (Hotho & Pedersen, 2012). The third domain, comparative institutionalism, represents the domain that fits this research; which consists of socioeconomic differences and the configuration of societal institutions. Typical applications are the home country effects on MNE control mechanisms, work practices and organizational capabilities, and interaction with host county specificities (such as language). Although scholars have shown interest in this field of study, it remains underrepresented compared to the other two domains. The main reason is that the lack of data makes it difficult to construct a comparative institutional measure (Hotho & Pedersen, 2012). So far, cultural difference measures are the only exception.

The IBV applies to two sides of the argument in this research: the weak institutional environment of developing countries force firms to adopt springboard behavior, and the moderators are part of the institutions of a country.

The differences between AMNEs and EMNEs are significant. Madhok and Keyhani (2012) argue that besides Liability of Foreignness (LOF), EMNEs also face Liability of Emergingness (LOE), where LOE is the additional disadvantage that EMNEs tend to suffer over AMNEs by being from an emerging market (Madhok & Keyhani, 2012: 28). According to Mathews (2006), EMNEs differentiate themselves through very rapid internationalization, achieved not through technological innovation as AMNEs did, but through organizational innovations, and EMNEs were able to exploit their latecomer and peripheral status through strategic innovations. Moreover, EMNEs are considered less path-dependent and less risk-averse (Luo & Tung, 2007).

There are two opposing views among the IB scholars: “one is that EMNEs are a new species of MNEs that can be understood only with new theories (Mathews 2002); the other is that existing theory is quite adequate to explain EMNEs (Narula, 2006)” (Ramamurti, 2012: 41). The most prolific FDI theory, the eclectic paradigm, argues that EMNEs must possess ownership advantages to engage in FDI successfully. However, EMNE-focused theories argue that EMNEs use FDI to obtain ownership advantages (Luo & Tung, 2007; Mathews, 2002).

GENERAL THEORIES ON FDI

Ever since Krugman (1979) introduced the New Trade Theory, FDI has been a widely researched topic. Trade costs, such as tariffs and transportation costs, play a crucial role in the development of these early internationalization theories. When the trade costs are high enough, as argued by the Proximity-Concentration hypothesis (Brainard, 1993), firms would opt for an FDI strategy, as opposed to creating economies of scale and exporting their produced goods. More recent research has proven the opposite, however; high trade costs have a negative influence on FDI (Cuevas, Messmacher, & Werner, 2005).

LOF is a disadvantage that MNEs face when operating abroad compared to domestic competitors (Zaheer, 1995). Therefore, FDI would only be interesting for firms if the firm possesses firm-specific advantages (FSA) that compensate for the LOF. Internalization Theory suggests that FDI is only relevant due to market imperfections (Buckley & Casson, 1976). In other words, if the principle-agent problem and opportunistic behavior would not exist, firms would license domestic firms, who are not hampered by LOF, to sell their product in the host market (Nielsen, Asmussen, & Weatherall, 2017).

lowers transaction costs. An FDI strategy cannot be considered the best available strategy when one of the three criteria is not met.

EMERGING MARKET INTERNATIONALIZATION FRAMEWORKS

In their review article, Luo and Zhang (2016) found four major EMNE internationalization frameworks. The springboard perspective (Luo & Tung, 2007), the LLL framework (LLL) (Mathews, 2002, 2006), the Ownership advantage logic (OA) (Dunning, Kim, & Park, 2007; Ramamurti, 2009, 2012; Rugman, 2009), and the Institutional arbitrage logic (IA) (Boisot & Meyer, 2008; Cuervo-Cazurra & Genc, 2008; Witt & Lewin, 2007; Yamakawa, Peng, & Deeds, 2008). This section will compare these the 3 latter frameworks by their main points, their distinctions, and complementarities to the springboard perspective.

The springboard perspective (Luo & Tung, 2007) describes the internationalization process of EMNEs. Their framework revolves around the argument that EMNEs use OFDI as a springboard to acquire strategic assets needed to compete more effectively against global rivals and to avoid the institutional and market constraints they face at home (Luo & Tung, 2007: 482). The recursive nature of the springboard perspective allows EMNEs to apply this strategy to multiple flaws in their business. Moreover, international expansion is not the goal of the EMNEs, but a means to success in global competition (Luo & Tung, 2007).

Linkage, leverage, and learning define LLL. Linkage refers to how ‘dragon multinationals’ (Asian MNEs) accessed excellent external resources through partnerships to overcome resource deficiencies. Leverage refers to how the dragon MNEs managed to leverage links with incumbents. Lastly, learning refers to the fact that dragon multinationals should be able to learn from their linkage and leverage, in order to reproduce these activities effectively in other markets.

There are five main distinctions between LLL and the springboard perspective; (1) LLL is less focused on institutional conditions in both the home and host markets, (2) the domestic market growth rate is less critical in LLL, (3) in LLL, the acquiring of assets is less relevant, (4) LLL does not explain the radical M&A strategy, and (5) LLL ignores the continuous upward spiral link between home country and foreign expansion (Luo & Tung, 2017).

The springboard perspective and LLL are complementary to each other on three aspects. First, both concepts are concerned with how EMNEs try to become competitive in the global market through international expansion. Second, the influential monopolistic advantage logic may not apply to these firms. Third, they both reason that the network and partnerships are imperative to the success of the firm (Luo & Tung, 2017).

There are three main distinctions between OA and the springboard perspective. First, every firm has FSAs. Otherwise, it would be unable to even operate in the domestic market. The explanation of what specific advantages EMNEs have should be more elaborate. Moreover, the extent to which these FSAs are valuable and transferable abroad is unclear. Second, OA should adopt the springboard act, as it is a unique quality of EMNEs, supported by FSAs. Third, OA only focuses on advantage leveraging for market-seeking, where it should also focus on disadvantage compensation via capability acquisition and strategic assets (Luo & Tung, 2017).

The complementarities between OA and the springboard perspective are threefold. When EMNEs are showing opportunity-seeking behavior, both the springboard perspective and OA agree that the advantages of the product, marketing, price, and technology are crucial. The primary goal of most EMNEs is to become competitive in both the domestic and foreign markets. The springboard perspective allows for more ownership advantages than OA. Lastly, the springboard perspective, OA, and other frameworks acknowledge the importance of relational capability as ownership advantage (Luo & Tung, 2017).

The last EMNE internationalization framework to be discussed is IA. There are two views of the IA framework: escapism and exploitation. The first view argues that EMNEs go global to dissociate themselves from the weak institutional environment of the domestic market. The dissociate behavior is amplified when EMNEs go to markets with enhanced intellectual property protection. The second view, exploitation, reasons that EMNEs will venture to other emerging markets because they developed capabilities to survive and thrive in such locations.

There are four distinctions between the springboard perspective and IA. First, IA reasons from the point of view where EMNEs can leverage institutional arbitrage because they are well accustomed to the capricious environment of emerging markets, something which AMNEs are not. Second, IA focuses only on the institutional arbitrage argument, whereas the springboard perspective argues that EMNEs have various motives to internationalize. Third, a weakness of IA is that it must accept the coexistence of escapism and exploitation to explain why EMNEs invest in both emerging and developed markets. Fourth, by adding unique insights, IA is a terrific addition to the literature. However, it warrants to integrate this with capability seeking and market-seeking (Luo & Tung, 2017).

IA and the springboard perspective have three complementarities: (1) both frameworks define link between the home and host market conditions, (2) both frameworks stress the importance of home governments and their policies, and (3) both frameworks should be able to offer different insights depending on differences between EMNEs (Luo & Tung, 2017).

THE SPRINGBOARD PERSPECTIVE

country, supporting the idea that FDI is used as a means to gain access to comparatively advantaged country-specific assets in the host country (Makino et al., 2002: 418).

Besides asset-seeking motivations, EMNEs also try to escape the institutional voids of their domestic market (Witt & Lewin, 2007), by investing in foreign countries with strong institutions (Luo, Xue, & Han, 2010). However, a recent study by Yoo and Reimann (2017) shows that weak intellectual property rights protection in developed countries lead to an increased FDI flow from emerging markets, suggesting that the most reliable institutions may not necessarily be the most significant motive to invest in a country, but rather the chance of knowledge spillovers in order to gain knowledge-based assets. These all fall under opportunity-seeking behavior.

Not only the fact that EMNEs started to internationalize for different reasons is noteworthy, but EMNEs also ignored the traditional internationalization theories. The evolutionary process theory (Johanson & Vahlne, 1977) showed that firms chose to internationalize gradually to neighboring countries, based on FSAs and to countries that are institutionally similar. EMNEs often choose a much faster-paced internationalization strategy (Guillen & Garcia-Canal, 2009; Mathews, 2002) or choose countries that are physically and psychologically much further away (Ramamurti, 2004). According to Ramamurti (2012), the increased speed of internationalization is mostly due to a changing international business environment, supported by the ‘born-global’ strategy (Knight & Cavusgil, 2004). The latter is explained by theories such as the springboard perspective, where EMNEs pursue differences between home and host countries, instead of similarities (Ramamurti, 2012). Deng (2009) reasons that EMNEs use FDI to leapfrog to more value-adding activities, acquiring foreign knowledge-based assets.

The springboard perspective has emerged as one of the leading theoretical frameworks to analyze the internationalization strategy of EMNEs. Next is an anthology of academic work that has reinforced and corroborated the springboard perspective in the last decade.

Through analyzing 425 cross-border acquisitions by Indian firms between 2000 and 2007, Gubbi, et al. (2010) established that the internalization of resources, both tangible and intangible (which are difficult to trade through the market and require time to develop), is enabled by cross-border acquisitions. This effect was amplified if the target firm’s location is in a developed economy with advanced institutional and economic environments.

Another research found that the ability of EMNEs to overcome latecomer disadvantage was mainly dependent on the capability of the firm to seek knowledge in foreign countries through OFDI (Kedia, Gaffney, & Clampit, 2012). Moreover, such knowledge-seeking behavior was not based on asset-exploitation, but rather asset-augmentation.

Li, Li, and Shapiro (2012) found empirical proof for the theory that EMNEs invest in developed economies with industry-specific advantages to catch up on technological backwardness.

necessary resources to be able to compete on a global level. Moreover, the research confirmed Luo and Tung’s (2007) notion that EMNEs use reputable market intermediaries to overcome the reputationally poor perceptions of governance, transparency, and accountability.

That EMNEs have asset-seeking and opportunity-seeking motives to adopt springboard behavior was confirmed by multiple types of research (Kotabe & Kothari, 2016; Kothari, Kotabe, & Murphy, 2013). Asset-seeking motives are defined as a foreign expansion that focuses on the absorption of assets into the EMNEs’ own firm to create new FSAs. Opportunity-seeking motives can be categorized as EMNEs looking for different niche markets which are neglected by MNEs. Satta, Parola, and Persico (2014) found that the adverse effect of physic distance is inapplicable to EMNEs and their venture for foreign resources. Moreover, they found that inward FDI (IFDI) has a robust boosting effect on an EMNEs’ OFDI.

By comparing MNEs with EMNEs, De Beule et al. (2014) found empirical proof that during acquisitions, EMNEs take less control compared to MNEs. This effect was amplified in high-tech industry takeovers. The institutional distance and freedom of investment increased the chance that an EMNE participated in a full acquisition, an effect which was not found for MNEs. Wang et al. (2014) discovered that EMNEs often used decentralization as a means to facilitate springboard behavior. They conducted their research by focusing on autonomy delegation to foreign subsidiaries by Chinese MNEs.

Gubbi and Elango (2016) researched cross-border acquisitions by Indian firms from 2000 to 2010 and discovered that EMNEs’ quest for critical assets is one of the primary motivations for firms to acquire foreign companies. Acquisitions of firms in advanced markets amplify this effect. Lastly, by making a comparison between EMNEs from the BRIC countries and UK-based MNEs, Gaffney et. al (2016) found empirical evidence that in markets that are institutionally distant from the home country in terms of knowledge protection, BRIC EMNEs opted for equity participation or full acquisition, to ensure a successful transfer of tacit assets.

According to Luo and Tung (2017), the springboard perspective and the OLI-framework differ on one important aspect: that ownership advantages are a precondition for an MNE to internationalize. The springboard perspective sees internationalization as a means to overcome the firm-specific disadvantages, but besides that, the OLI-framework and the springboard perspective are mostly complementary to each other (Luo & Tung, 2017).

RESEARCH GAP AND RESEARCH QUESTION

RQ: What boundary conditions shape the location choice of EMNEs pursuing

knowledge-based assets?

III. HYPOTHESES DEVELOPMENT

I have established that institutions are paramount to the strategy of a firm and that EMNEs try to escape the institutional voids of their domestic market by adopting springboard behavior. Moreover, EMNEs try to acquire assets that are scarce in their domestic markets. This behavior fits the new paradigm of internationalization, where slow progression has made way for the fast accumulation of international assets. It is time to shift the focus towards the aim of this research: the boundary conditions on the relationship between knowledge assets in the host country and FDI location choice of EMNEs in the BRICS-countries. This thesis researches four moderating effects: language, colonial history, legal systems, and preferential trade agreements (PTAs).

KNOWLEDGE-BASED ASSETS AND SPRINGBOARD BEHAVIOR

The springboard perspective argues that EMNEs internationalize to compete with foreign firms entering their domestic market, to overcome latecomer disadvantages and to find resources that are scarce or nonexistent in the domestic market (Luo & Tung, 2007). Where an abundance of knowledge-based assets characterizes the majority of AMNEs, most of the EMNEs are not characterized by this abundance (Dunning et al., 2007; Zahra, 1999). Dunning (1993) argued that EMNEs could access knowledge-based assets that are scarce in the domestic market by investing in markets where these are plentiful. These knowledge-based assets are critical for the competitiveness of firms, as they provide the foundation for technological innovation (Freeman, 2004; Tassey, 1991).

Two concepts are heavily linked with the acquisition of knowledge-based assets: absorptive capacity and knowledge spillovers. Absorptive capacity refers to the ability of a firm to absorb, acquire, and internalize external knowledge in foreign locations (Cohen & Levinthal, 1990). In general, EMNEs’ absorptive capacity is lower compared to their home-country development level (Hoskisson, Eden, Lau, & Wright, 2000). IB scholars have acknowledged that absorptive capacity can be split up in two types of abilities: the ability of assimilation of the knowledge (potential absorptive capacity) and the ability to apply the knowledge (realized absorptive capacity) (Lane, Salk, & Lyles, 2001; Zahra & George, 2002). The assimilative capacity, which is the combination of the two, is heavily dependent on the investments made by the firm to either develop in-house capabilities to recognize and value external knowledge and to tap into diverse sources of knowledge (Gubbi & Elango, 2016: 362). The other concept, knowledge spillovers, is also a popular research phenomenon among IB scholars. EMNEs could benefit from knowledge spillover through linkage with consumers, competitors and suppliers, movement of skilled labor, and demonstration effects (Blomstrom & Kokko, 1998).

In countries that possess more knowledge-based assets, there are more potential acquisition targets with superior technological knowledge assets. The more potential acquisition targets there are, the more interesting the host country becomes to EMNEs.

In advanced markets, there are many learning opportunities increasing knowledge spillover potential. The greater the availability of knowledge-based assets, the more often firms will find themselves in positions to benefit from knowledge spillovers in the host country. These learning opportunities will consist of interaction with more capable competitors and better-qualified suppliers and engagement with more demanding customers. All will push EMNEs to perform better and to learn from these exchanges in the host country, enhancing their knowledge base and improving their home market competitive position.

H1. The greater the availability of knowledge-based assets in the host country the greater the FDI from BRICS countries.

LANGUAGE

Language is deeply rooted in culture and might be a more critical influencer of daily habits than one might think. For example, in languages where a future distinction is grammatically necessary (it will rain tomorrow, instead of ‘it rains tomorrow’) to indicate a future event, speakers are less inclined to engage in future-oriented behavior (Chen, 2013). Another example, by Carriere & Eco (2010), who identified that in movies, traveling shots moved by the writing direction: in western movies from left to right and in Iranian movies from right to left. Both examples show that language affects perception and decision-making of individuals and groups of people, justifying that more in-depth research on second language and its effect on bilateral trade is needed. Language is one of the most used variables in the gravity equation, and Egger and Lassmann (2012) performed a meta-analysis on 701 coefficients from 81 articles on the estimated effect of common (spoken or official) language. The channels through which common language affects trade are well understood, and there is more than enough evidence that it increases bilateral trade, but the coefficients are used in different ways. Their paper found that the estimated effect of common language is mostly affected by the sample period and the control variables that are used. When cultural proximity variables were included, the effect of language shrunk. They also find that over time, the language effect increases. Finally, they argue that one of the reasons for the variability of the variable is due to an inconclusive definition of common official language. Looking at the data set, 92% of the coefficients (642/701) are language dummies. Unfortunately, this indicates that most of the literature is only considering the (limited) binary variable of common official language and not the probability variable of common spoken language.

with which they can communicate. Based on that narrative, Melitz and Toubal (2014) constructed a variable that combined native languages, official languages, and secondary languages, allowing to see what the chance is that two random people from two countries will be able to communicate with each other in any language and defined this as common spoken language (CSL).

That language has a significant impact on the transfer of knowledge-based assets between the EMNE and its advanced market acquisition target is based on the premises that knowledge can be transferred. In the knowledge transfer literature, the discussion mostly ignores the transferability of knowledge but focuses on the transfer of knowledge (Bresman, Birkinshaw, & Nobel, 1999; Hedlund, 1994; Madhok, 1997). There are two paradigms in the literature, based on different theoretical perspectives on international business: static and dynamic. The static paradigm argues that all knowledge can be transferred and that technological knowledge is the essential knowledge a firm can possess since it is the core of all competitive advantages (Mansfield & Romeo, 1980). The dynamic paradigm argues that not only the characteristics of the knowledge has an effect, but also the relationship between the firms that exchange knowledge is critical (Choi & Johanson, 2012). The reason is that all knowledge consists of three elements: tacit, sticky, and explicit. The transferability of tacit knowledge is limited because understanding the background of the knowledge is needed, which can only be provided through continuous interaction over a period (Bresman et al., 1999).

In this line of reasoning, I argue that under the situation that the home and host country have a high degree of common spoken language, speaking the same language will enhance EMNEs’ acquisition of knowledge-based assets through interacting with customers, suppliers and competitors in the host country. Therefore, a high degree of CSL between home and host country strengthens the positive relationship between a host country’s knowledge-based assets and FDI from BRIC countries.

H2. Greater degrees of CSL between home and host country strengthen the positive relationship between the availability of knowledge-based assets in the host country and FDI from BRICS countries.

COLONIAL HISTORY AND LEGAL SYSTEMS

Imperialism is an integral part of our global history. It all started with voyages of the Portuguese and Spanish towards the Americas, followed by the Dutch Empire and British Empire towards the Far East, and the French Republic to Africa. The time that European powers had overseas colonies lasted until 1999, when the last colony, Macau, was gifted to China by the Portuguese. First, I will present an overview of the colonial history of the BRICS countries.

colonizers on Chinese soil. However, unlike the other countries, it is hard to see a shared history between the Chinese and its colonizers. South Africa was first colonized by the Dutch Empire from 1652 until 1795 and again from 1803 until 1806. In 1814, South Africa was also ruled by the British Crown, which lasted until 1931. Of the BRICS countries, The Russian Federation is the only one who colonized, instead of being colonized.

Klerman, Mahoney, Spamann, and Weinstein (2011) argue that the colonial history and legal systems of a country are highly correlated. In other words, the legal system of a country exogenously adopted originates most probably from its former colonizer. The origins of legal systems and its influence on the financial development and economic outcomes can be categorized into two main traditions: French civil law and English common law (La Porta, de Silanes, Shleifer, & Vishny, 1998; La Porta, Lopez-De-Silanes, Shleifer, & Vishny, 1997). Legal system based on common law is a result of the English law, while French civil law is based on Roman law. The research of La Porta et al. (1998; 2000; 1997) showed that the content of the laws is closely correlated with the legal families. In turn, these legal families influence the financial markets, the financial rules and regulations, and economic outcomes.

The ‘colonial power’ theory argues that the influence of former colonial powers on the judiciary system and institutional quality cannot be underestimated in the development of the country its economic development and growth (Klerman et al., 2011). The argument they make makes sense according to Kam (2018): colonized countries would become embedded in the cultural, political, and economic conditions of the colonizer and these conditions would have had a significant impact on the financial regulations and economic growth. This embeddedness is also in line with the argumentation regarding common spoken language, which can be a result of a former colonizer.

The embeddedness in the similar cultural, political, and economic systems tends to generate more interactions and eliminate communication barriers. Under the situation that there is colonial history between the home and host country, EMNEs more likely to benefit from knowledge spillover, and absorb useful knowledge. Therefore, I argue that the positive relationship between a host country’s knowledge assets and FDI from BRICS countries is strengthened by the colonial history between home and host country.

H3. The colonial history between home and host country strengthens the positive

relationship between the availability of knowledge-based assets in the host country and FDI from BRICS countries.

The reasoning above also applies to the situation where home and host country share legal systems.

PREFERENTIAL TRADE AGREEMENTS

FDI and PTAs are closely linked together, as an increase of FDI is often seen as one of the main reasons that countries participate in a PTA. The NAFTA, a PTA between Canada, Mexico, and the United States, provides a good example, where research suggests that Mexico’s IFDI would have been 40% less if the agreement would not have been signed (Lederman, Maloney, & Servén, 2003).

Preferential liberalization, which is the reduction of governmental intervention on trade through tariffs, quotas, and other regulations that obstruct the free flow of goods and services, has four different types of effects on FDI: an effect on investment and other non-trade provisions, an effect on changes in trade flows, creation of an extended market, and long-term growth effects (Medvedev, 2012: 49).

Looking at the first type of effect, the effect on investment and other non-trade provisions, Adams, Dee, Gali, and McGuire (2003) studied the impact of PTAs on stocks of OFDI. They found that the “deep integration”1 _{aspects of PTAs had a significant impact on FDI, rather than an impact}

on trade provisions. Various other research found that certain areas of “deep integration” have a positive effect on FDI, such as IPR protection and lower political risks (Kolstad & Tondel, 2002; Lee & Mansfield, 1996; Maskus, 1998).

In the second category, the effects on changes in trade flows, there is a notable difference between early literature on the matter and more recent literature. In a report of the World Bank (2002: 3) three studies are quoted that demonstrate that high tariffs on imported goods have a positive effect on IFDI (Brash, 1966; Dunning, 1958; Horst, 1972), meaning that PTAs would hurt FDI. More recent literature argues the opposite, emphasizing that there are many complementarities between trade and investment, due to the MNE production network and intra-industry and intra-firm trade (Caves, 1996; Globerman, 2002; Markusen, 2002), which leads to more openness to trade, which has been the most reliable indicator of FDI, only after market size (Chakrabarti, 2001). In a review article, Lim (2001) also find that market size is the most reliable indicator for FDI. This effect is paramount for the relationship between FDI and PTAs.

Lastly, PTAs are also positively correlated with economic growth (UNCTAD, 1998; World Bank, 2001). However, the direction of causation in this relationship is unclear. Much of the literature argues that FDI allows economies to grow (Lim, 2001), other literature suggests that FDI is attracted through growth (Rodrik, 1998).

According to Medvedev (2012), the problem with these relationships is that the sign and strength are not empirically tested. In his research, Medvedev finds that there is a positive relationship between PTAs and IFDI, which is amplified by increasing the size of partners and proximity to the host country.

To summarize, a positive relationship between FDI and PTAs has been extensively demonstrated. However, there is no literature that combines this relationship with the springboard perspective. Moreover, Luo and Tung (2017) acknowledge that the springboard perspective has ignored soft power deficits, through host country nationals’ attitudes and policies toward investors from emerging markets. I expect that under the impact of PTAs, host country nationals may possess a positive view towards foreign investors belonging to the same PTA system. So, under a situation that the home and host country belong to the same PTA, EMNEs may receive more help from host country firms and may be more capable of absorbing superior knowledge assets in the host country market. I suggest that the positive relationship between the availability of knowledge-based assets in the host country and FDI from BRICS countries can be strengthened by the number of PTAs.

H5. The amount of preferential trade agreements between home and host country

strengthens the positive relationship between the availability of knowledge-based assets in the host country and FDI from BRICS countries.

IV. METHODOLOGY

DATA

The central hypothesis is the effect of knowledge-seeking behavior on FDI. All the other hypotheses that represent a moderating effect are captured by the box in the middle. To test these hypotheses, I collected OFDI Stock data through the United Nations Conference on Trade and Development (UNCTAD). They provide these figures from 2001 up to and including 2012, where I opted to use the data from 2009-2012. The main reasons to start in 2009 are incomplete data before 2009 and the beginning of the recession that year. Other data is collected from various sources, of which an overview can be found in Table 1.

THE MODEL

I suggest the following estimation model for the locational choice of the BRICS countries’ OFDI in developed countries, with knowledge-seeking motives:

𝑂𝐹𝐷𝐼_𝑖𝑗𝑡= _𝛾1𝐾𝐵𝐴 𝐻𝑜𝑠𝑡_𝑡−1+ _𝛾2(𝐾𝐵𝐴 𝐻𝑜𝑠𝑡_𝑡−1∗ 𝐶𝑆𝐿_𝑡−1) + _𝛾3(𝐾𝐵𝐴 𝐻𝑜𝑠𝑡_𝑡−1∗ 𝐿𝑒𝑔𝑎𝑙𝑆𝑦𝑠_𝑡−1) +

𝛾4(𝐾𝐵𝐴 𝐻𝑜𝑠𝑡𝑡−1∗ 𝐶𝑜𝑙𝑜𝑛𝐻𝑖𝑠𝑡𝑡−1) + 𝛾5(𝐾𝐵𝐴 𝐻𝑜𝑠𝑡𝑡−1∗ 𝑃𝑇𝐴𝑠𝑡−1) + 𝛾6𝐶𝑆𝐿𝑡−1+ 𝛾7𝐿𝑒𝑔𝑎𝑙𝑆𝑦𝑠𝑡−1+ 𝛾8𝐶𝑜𝑙𝑜𝑛𝐻𝑖𝑠𝑡𝑡−1+ 𝛾9𝑃𝑇𝐴𝑠𝑡−1+ 𝑋𝑖𝑗𝑡−1𝛽+ 𝜀𝑖𝑗

where the dependent variable is the logarithm of the OFDI Stock of the BRICS countries i in the developed countries j in year t; 1, 2, 3, 4, and 5 are the parameters of interest; 𝑋𝑖𝑗𝑡−1 is a

vector of the control variables; 𝛽 is a vector of all other parameters; and 𝜀_𝑖𝑗 is the error term. VARIABLES

DEPENDENT VARIABLE

The dependent variable in this study is the log of the stock of OFDI of the BRICS countries in the developed countries. For every research on FDI, it is essential to make the right decision between FDI flow or FDI stock. Where flow data shows how MNE investments change over time, stock data shows the static number (Nielsen et al., 2017). Accordingly, flow data can be used to analyze entry decisions, where stock data allows us to analyze the correlation between country characteristics and the location of FDI (Nielsen et al., 2017). The latter is relevant for this study and in line with other studies (Filippaios, Papanastassiou, & Pearce, 2003; Kang & Jiang, 2012). As non-linearities can occur in FDI related models (Buckley et al., 2007), I transformed OFDI Stock in a natural log value.

INDEPENDENT VARIABLE

As a proxy for the availability of knowledge-based assets in the host country, I use the Global Innovations Index (GII) compiled by Cornell University, INSEAD, and the World Intellectual Property Organization (Yoo & Reimann, 2017). The GII aims to capture the multi-dimensional facets of innovation (Cornell University, INSEAD, & WIPO, 2013). The score that countries receive shows the overall technological development and innovation environment in a country, based on innovation input, such as institutions and market/business sophistication, and innovation output, such as knowledge and technological outputs and creative outputs. The standardized variable was used in the research.

MODERATING VARIABLES

Four moderators were used, and all moderating variables were standardized. As discussed earlier, the research by Melitz and Toubal (2014) serves as the critical foundation for the moderating effect of common spoken language. The variable that will be used for this research, Common Spoken Language, is published by Melitz and Toubal on the CEPII website. This variable shows the chance that two random individuals from a country pair can communicate with each other.

As for legal systems, the research of La Porta et al. (1999) is as relevant today as it was when first published in 1999. They group the origins of legal systems into five categories: English, French, German, Socialist, and Scandinavian. For all the country pairs that are grouped in the same legal system, the dummy variable will be set to 1. For all the country pairs that are not grouped, this value will be set to 0.

The data on colonial history is retrieved from the ICOW Colonial History Data website (Hensel, 2014). This database provides extensive information on all country pairs and their colonial history. This variable will be a dummy as well, where a one will be assigned to country pairs that share a colonial history and a two will be assigned to country pairs that do not share a colonial history. There are two ways that countries can share a colonial history. The most apparent history is the one where one of the developing countries has occupied one of the BRICS countries. However, when the same colonizer has occupied countries, the formerly occupied countries share a colonial history as well.

The last moderating effect that is research is whether country pairs are part of the same

PTAs. For this variable, the Design of Trade Agreements Database (Dür, Baccini, & Elsig, 2014)

will be used, which provides all information on all PTAs in the world. This variable will show the number of PTAs that both countries in the country pair share.

CONTROL VARIABLES

The control variables that are included will serve two motives. The first motive is to control for other important FDI motives. Secondly, it is crucial to control for country effects as well.

One of the essential alternative drivers of FDI is market-seeking behavior (Makino et al., 2002). To control for market size, I include the log of host country GDP per capita and GDP

growth (Li et al., 2012). Another driver of FDI could be the quest for natural resources that are

scarce in the home market (Deng, 2009), this is controlled for by including the ratio of ore and metal exports to

merchandise exports of the host country (Buckley et al., 2007). Lastly, the level of infrastructure development is also a determinant of host country attractiveness (Graf & Mudambi, 2005). To control for this motive, I use fixed and mobile phones per capita as a proxy (Bailey & Li, 2015; Cuervo-Cazurra & Genc, 2008).

It is also essential to control of country-level effects. Firstly, I control for geographical

distance, as markets tend to interact more with proximate markets, by using the weighted

acquisition literature, I control for cultural distance based on Hofstede’s cultural country scores (Rabbiosi, Elia, & Bertoni, 2012). Thirdly, as volatile and unpredictable inflation rates in a host-country discourage firms from engaging in FDI (Buckley et al., 2007), I include the host host-country’s

inflation rate. Lastly, as firms could engage in FDI to gain fiscal benefits (Kolstad & Wiig, 2012),

I control for tax havens. An influence like this can be controlled for by excluding the country pairs from the sample or by including a dummy variable marking the respective country pairs (Cheng & Ma, 2010). I opted to omit the tax havens from the sample, as most of the outliers were tax havens. The OECD and the International Monetary Fund define Cyprus, Ireland, Luxembourg, Malta, and Switzerland as tax havens.

Variable Source OFDI Stock UNCTAD

Knowledge-based assets Cornell University, INSEAD, and the WIPO Common Spoken Language (Melitz & Toubal, 2014)

Legal Systems (La Porta et al., 1999)

Colonial History The ICOW Colonial History Data Preferential Trade Agreements Design of Trade Agreements Database GDP per capita The World Bank

GDP Growth The World Bank Natural resources The World Bank Infrastructure The World Bank Geographical Distance CEPII

Cultural Distance CEPII

Inflation The World Bank Tax haven OECD & IMM

Table 1 - Data Sources

TESTING FOR ASSUMPTIONS

Statistical models make assumptions about the characteristics of the data; therefore, it is essential that the data fulfill these assumptions before conducting statistical analysis. If these assumptions are ignored, the value of the statistical analysis is close to zero. Since I will be conducting a regression with moderators, I will test the four assumptions of moderating analysis: normality, outliers, homoscedasticity, multicollinearity, and linearity (Aguinis, 2004).

First, I check for normality by plotting a kernel density estimate against a normal distribution. As the kernel density estimation line shows some signs of normality, but also some deviations, I also performed an IQR test. As the test showed that there are no outliers, there is no reason to reject normality. This test also shows that there is no need to check for outliers.

Thirdly, I check for multicollinearity by performing a variance inflation factor (VIF). All values stay clear of the threshold, which is commonly accepted at 10 (Belsley, Kuh, & Welsch, 1980; Hair, Anderson, Tatham, & Black, 1998; Kutner, Nachtsheim, Neter, & Li, 1996), indicating that there are no multicollinearity concerns with the data.

Lastly, I check for linearity. By plotting the dependent and independent variable against each other, I was not able to accept that there was no linearity. However, after taking the natural logarithm of OFDI Stock, the data was acceptable.

All figures and table can be found in Appendix I.

V. RESULTS

In Table 2, I present the descriptive statistics of all variables. Table 3 shows the correlation matrix of all variables in the research. The high correlation among several variables is not surprising, as some characteristics naturally correlate with each other, which is in line with comparable papers in this line of research (Cuervo-Cazurra & Genc, 2008).

Table 4 shows the regression results. Model 1 only includes the dependent variable and control variables, where model 2 also includes knowledge-based assets. In model 2, strong support is found for hypothesis 1. The presence of knowledge-based assets in the developed host countries made them more attractive for FDI from the BRICS countries. The higher the availability of

Variable Mean Std. Dev. Min Max

Dependent variable OFDI Stock 5.622 2.510 -0.616 11.076 Independent variables KBA host 53.041 10.296 30.286 82.857 CSL 0.103 0.164 0 1 LegalSys 0.252 0.362 0 1 ColonHist 0.110 0.313 0 1 PTAs 0.155 0.434 0 1 Control variables GDP per capita 10.292 0.629 8.8310 11.520 GDP Growth -0.204 4.148 -14.814 8.459 NatRes 6.115 8.280 1.074 41.998 Infrastructure 157.310 16.904 120.770 197.702 GeoDist 7714.341 3079.250 1537.558 17981.980 CultDist 62.101 20.569 11.747 111.508 Inflation 2.291 1.841 -1.353 12.006

knowledge-based assets in the host countries results in greater the amounts of OFDI Stock from the BRICS countries, which is significant at the p <0.001 level. The coefficient of this relationship is 1.1484, which shows that the Springboard perspective also applies to the sample of this research.

Model 3 shows the moderating effect of common spoken language on the relationship between knowledge-based assets and OFDI. The model shows no support for hypothesis 2, although it is the only significant moderating effect at p < 0.001. Where the relationship was expected to be positive, the model shows a negative relationship of -0.4188. Therefore, it could be stated that speaking a common language hampers the OFDI of BRICS countries in developed countries.

Model 4 shows the moderating effect of colonial history. Hypothesis 3, which argued that a shared colonial history would increase FDI, is not supported. The coefficient, -0.0959, is not as expected and the relationship is not significant.

1. 2. 3. 4. 5. 6. 7. 8. 9. 1. OFDI Stock 1,0000 2. KBA host 0,3185* 1,0000 3. CSL 0,1400* -0,0180 1,0000 4. LegalSys -0,0300 -0,1689* 0,3374* 1,0000 5. ColonHist 0,2014* 0,1213* 0,6634* 0,5102* 1,0000 6. PTAs -0,1281* -0,0248 0,1250* -0,1660* 0,0210 1,0000 7. GDP pc 0,3000* 0,6830* -0,0459 -0,2567* 0,0565 -0,0090 1,0000 8. GDP Growth 0,1225* -0,0921* -0,0285 0,0123 0,0345 0,0002 0,0951* 1,0000 9. NatRes 0,1127* 0,0455 0,0039 -0,0253 0,0548 -0,1028* 0,1108* 0,0600 1,0000 10. Infrastructu re 0,1203* 0,1061* 0,0080 -0,1687* -0,0894* 0,0725 0,1933* -0,0298 0,0320 11. GeoDist -0,0742 0,1942* -0,1868* -0,0955* 0,0565 0,1037* 0,1780* 0,0526 0,1449* 12. CultDist -0,0300 0,2493* -0,1661* -0,2287* -0,2166* -0,2241* 0,3070* -0,0002 0,0372 13. Inflation -0,0341 -0,3007* -0,0062 0,0653 -0,0025 0,0347 -0,2409* 0,1110* 0,3912* 10. 11. 12. 13. 1,0000 -0,1533* 1,0000 0,0400 -0,1498 1,0000 0,0745 -0,0361 -0,0104 1,0000

The moderating effect of legal systems on the primary relationship is shown in model 5. Hypothesis 4 is also not supported. However, the sign of the relationship is in line with the expectations, as the coefficient is 0.1269.

The last model, model 6, shows the moderating effects of PTAs. The model shows no support for hypothesis 5 that PTAs have a positive effect on the central relationship. Moreover, the effect was expected to be positive, but the model shows a negative relationship, namely, -0.1851.

From model 2-6, the adjusted R2 varies from 0.2297 to 0.2660.

VI. DISCUSSION AND CONCLUSION

In contrast to AMNEs, EMNEs have adopted a more straightforward internationalization strategy, where the internalization of knowledge-based assets is one of the primary goals (Luo & Tung, 2007). As the theory evolved in many directions, there was little consensus on the factors that influenced the EMNE’s location choice decision. This led me to ask the following question: what boundary conditions shape the locational choice of EMNEs pursuing knowledge-based assets?

In line with prior research, I found some proof that the Springboard model is an accurate theoretical concept for the internationalization strategy of EMNEs. However, the confirmation of an already statistically proven model is everything but revolutionary. This thesis focuses on the four boundary conditions: common spoken language, legal systems, colonial history, and preferential trade agreements.

I performed empirical research that used a sample of 447/445 country pairs, which consisted of the BRICS countries and the developed countries as identified by the UN. The data for this research was collected from 2008 through 2012. In line with the Springboard perspective, I researched the link between knowledge-seeking motivation and FDI. The moderating variables that were to be researched were lagging one year, as it takes time to collect information and to act accordingly.

Each home country, which was either Brazil, The Russian Federation, India, China, or South Africa, was paired with all the developed countries in the world. The data collected, which consisted of knowledge-based assets present in the host country, the moderating variables, and control variables, was used to investigate whether the hypothesized relationships existed.

1 2 3 4 5 6 Independent variable KBA host 1,1484*** (0,1632) 1,0945*** (0,1632) 1,0657*** (0,1647) 1,1384*** (0,1648) 1,1185*** (0,1623) Moderating variables CSL 0,0996 (0,1054) CSL x KBA host -0,4188*** (0,1051) ColonHist 0,3344*** (0,0806) ColonHist x KBA host -0,0959 (0,0721) LegalSys 0,2761* (0,1152)

LegalSys x KBA host 0,1269

(0,1090)

PTAs -0,2930**

(0,1124)

PTAs x KBA host -0,1851

(0,1209) Control variables GDP per capita 1,7283*** (0,2662) 0,4863 (0,3268) 0,6654* (0,3224) 0,5293 (0,3222) 0,6159 (0,3737) 0,5597 (0,3154) GDP Growth 0,0884* (0,0389) 0,0656 (0,0388) 0,0693* (0,0359) 0,0655 (0,0373) 0,0692 (0,0402) 0,0609 (0,0385) NatRes 0,0424*** (0,0146) 0,0586*** (0,0145) 0,0525*** (0,0143) 0,0487*** (0,0162) 0,0543*** (0,0155) 0,0515*** (0,0149) Infrastructure 0,0046 (0,0065) 0,0068 (0,0063) 0,0076 (0,0061) 0,0084 (0,0062) 0,0091 (0,0064) 0,0100 (0,0064) GeoDist -0,0002*** (0,0000) -0,0002*** (0,0000) -0,0001*** (0,0000) -0,0002*** (0,0000) -0,0002*** (0,0000) -0,0002*** (0,0000) CultDist -0,0159*** (0,0049) -0,0184*** (0,0047) -0,0193*** (0,0047) -0,0141*** (0,0049) -0,0140** (0,0050) -0,0241*** (0,0048) Inflation 0,1451 (0,0951) 0,1690 (0,0900) 0,2288** (0,0838) 0,1570 (0,0872) 0,1535 (0,0896) 0,1889* (0,0895) Observations 447 445 445 445 445 445 Adj.R2 _{0.1687 0.2297 0.2660 0.2481 0.2411 0.2491}

Standard errors are in parentheses. Year dummies included. _{*** p < 0.001 ** p < 0.01 * p < 0.05}

to the control model. The first hypothesis shows that the Springboard perspective, where EMNEs use FDI to increase their knowledge, finds support in this model.

Now it is time to look at the leading academic focus of this paper, the moderating variables. Of the four, only common spoken language was significant, but it did not have the sign that was expected. In the following sections, I will discuss the four moderators.

It was hypothesized that the degree of common spoken language would positively influence the relationship between knowledge-based assets and OFDI Stock. However, the opposite was found to be true: the higher the ability to speak in a common tongue, the more negative the impact was. For me, this defies all logic. However, there are a few explanations that I would like to share. First, the proxy used for CSL may be at fault, because the variable measures the number of people that speak the same language for each country pair. However, not everyone in the country is involved with business activities. For all country pairs, the Chinese average CSL score is 0.0066, for comparison, India scores on average 0.11 and Brazil scores 0.036 on average. In hindsight, perhaps it is irrelevant to measure the linguistic capabilities of the entire population because the people who participate in international business are more likely to be highly educated and are more likely to master relevant foreign languages, such as Chinese businessmen that speak English. The same can be said for Brazilian managers, who’s domestic population is limited in foreign languages, while educated managers can be expected to speak Spanish and English. Secondly, the reason why a common language could have had a positive effect might also be an explanation for why it has an adverse effect. I argued that speaking the same language made it easier to cooperate during an acquisition.

The third and fourth hypothesis dealt with colonial history and shared legal systems. A colonial history would mean that countries had tied for centuries, shared cultural characteristics, and shared institutional characteristics. These similarities were expected to have a positive effect on the locational choice. A shared legal system, sometimes part of shared colonial history, was expected to have a positive effect as well; lower contractual cost was expected to have a positive effect on the acquisition targets of EMNEs. Both moderating effects were insignificant but only shared legal systems had the expected positive impact.

not on the choices based on knowledge-seeking motivation. Additionally, looking at only colonial ties may have been too narrow of a scope. Consider for instance the effect of a shared governmental system (socialism, capitalist, communist) may have on knowledge seeking motivation.

The empirical results show the same expected sign for legal systems, however; the interaction term is insignificant. The standard variable, which shows the direct effect on OFDI Stock, is significant and has a positive effect on it. The same logic that I used for colonial history may apply here as well. For FDI, legal systems matter. However, when EMNEs are pursuing knowledge-based assets, it seems that it is only the presence of the knowledge-based assets that matters and not the boundary conditions.

Lastly, this paper looked at the PTAs between the BRICS countries and the developed countries. For both the interaction term, which was insignificant and the independent variable, which was significant, the sign is negative. This is contrary to the hypothesized relationship that was expected: a positive relationship between the two. Besides the more common theory that boundary conditions might be irrelevant for EMNEs, I believe there is another explanation here. I expected that PTAs would foster trade and therefore would increase the amount of OFDI Stock originating from the BRICS countries. However, there are reasons why countries agree to a PTA, and one of them is that governments feel that firms do not fully benefit from the potential that the two countries have. Therefore, governments may try to incentivize firms to engage in OFDI to a particular location, but this might not be effective. This, combined with the lay theory of these discussions, is possibly the reason for the negative relationship.

So, what boundary conditions shape the locational choice of EMNEs pursuing knowledge-based assets? Contrary to common practice, the lack of evidence for most of the hypotheses answers this question, instead of having accepted hypotheses. Where it was expected that a common tongue, colonial history and shared legal system, and PTAs would influence the way EMNEs pursued knowledge-based assets, the empirical results showed that there was no significant effect at all. This leads me to conclude that the end justifies the means: if the EMNEs can acquire knowledge-based assets, it does not matter what boundary conditions are present.

This rhymes with some of the previous research done in this field of study. For instance, the harmful effects of Psychic Distance were already proven to be inapplicable to EMNEs adopting Springboard behavior (Satta et al., 2014) and Gubbi & Elango (2016) found that it is the quest for critical assets that is the primary motivation for EMNEs. Both papers show that EMNEs are willing to pursue knowledge-based assets in riskier locations, also underlining the riskier strategy have adopted anyhow. Moreover, as Gaffney et al. (2016) found, EMNEs prefer to mitigate these risk by changing the style of acquisition, not the location.

LIMITATIONS,FURTHER RESEARCH

evidence can support the conclusions based on lack of proof. Moreover, the data used is relatively old as the period spans from 2008 until 2012, which had to do with the availability of the data. This leads to the second and third avenue for future research. There are significant differences between the BRICS countries regarding common spoken language, colonial history, and other characteristics. Therefore, it would be interesting to analyze the BRICS countries separately and to discover whether there are differences among EMNEs from different emerging economies. Moreover, an analysis of more recent data would be interesting as well, given how the world has changed from 2012 onwards.

ACKNOWLEDGMENTS

I want to thank my supervisor Gary Le for his continuous support and feedback. I really appreciated that the door of your office was always open for me for all kinds of questions. I would also like to extend my gratitude to dr. Gubbi, who co-supervised my thesis. It means a lot to me that an expert in FDI and emerging markets took the time to read my thesis. Lastly, I would like to thank my sister, Marijke, for reading my thesis and providing me with essential critiques.

On a personal note, I would like to thank my study friends; Marc, Rafael, and Jeroen. Without being able to discuss the do’s and don’ts of a writing thesis, ask for help whenever needed, and enjoy the moments when we did not work on our theses in the University Library, it would have been a lot harder. I would like to thank my friends, who were able to supply me with an abundance of distractions: David, Teddy, Jan Paul, and Vincent. Naturally, also a word of gratitude towards Larissa. Moreover, finally, I want to dedicate my last words to my parents: in these incredibly tough times, I am grateful to have your support, and I am happy to have made you proud.

VII. REFERENCES

Adams, R., Dee, P., Gali, J., & McGuire, G. 2003. The trade and investment effects of preferential trading agreements - old and new evidence. Canberra.

Aguinis, H. 2004. Regression analysis for categorical moderators. Regression analysis for categorical moderators.

Bae, K. H., Purda, L., Welker, M., & Zhong, L. 2013. Credit rating initiation and accounting quality for emerging-market firms. Journal of International Business Studies, 44(3): 216– 234.

Bailey, N., & Li, S. 2015. Cross-national Distance and FDI: The Moderating Role of Host Country Local Demand. Journal of International Management, 21(4): 267–276.

Barney, J. 1991. Resources and Sustained Competitive Advantage. Journal of Management, 17(1): 99–120.

Belsley, D. A., Kuh, E., & Welsch, R. E. 1980. Regression Diagnostics. Wiley Online Library. Blomstrom, M., & Kokko, A. 1998. Multinational Corporations and Spillovers. Journal of

Economic Surveys, 12(3): 247–277.

365.

Brainard, S. L. 1993. A Simple Theory of Multinational Corporations and Trade with a Trade-Off Between Proximity and Concentration. National Bureau of Economic Research Working Paper Series.

Brash, D. T. 1966. American investment in Australian industry. Canberra: Australian National University Press.

Bresman, H., Birkinshaw, J., & Nobel, R. 1999. Knowledge transfer in international acquisitions. Journal of International Business Studies, 30(3): 439–462.

Buckley, P. J., & Casson, M. 1976. A long-run theory of the multinational enterprise. The Future of the Multinational Enterprise, 32–65.

Buckley, P. J., Clegg, J. L., Cross, A. R., Xin, L., Voss, H., et al. 2007. The determinants of Chinese outward foreign direct investment. Journal of International Business Studies, 38(4): 499– 518.

Carriere, J., & Eco, U. 2010. N’espérez pas vous débarrasser des livres. Bulletin Des Bibliotheques de France, 3: 97–98.

Caves, R. E. 1996. Multinational enterprise and economic analysis. New York: Cambridge University Press.

Chakrabarti, A. 2001. The Determinants of Foreign Direct Investment : Sensitivity Analyses of Cross-Country Regressions. Kyklos, 54(1): 89–114.

Chen, M. K. 2013. The effect of language on economic behavior: Evidence from savings rates, health behaviors, and retirement assets. American Economic Review, 103(2): 690–731. Chen, V. Z., Li, J., & Shapiro, D. M. 2012. International reverse spillover effects on parent firms:

Evidences from emerging-market MNEs in developed markets. European Management Journal, 30(3): 204–218.

Cheng, L., & Ma, Z. 2010. China’s outward foreign direct investment. China’s Growing Role in World Trade: 545–578. Chicago: University of Chicago Press.

Choi, S. G., & Johanson, J. 2012. Knowledge translation through expatriates in international knowledge transfer. International Business Review, 21(6): 1148–1157.

Cohen, W. M., & Levinthal, D. A. 1990. Absorptive Capacity: A New Perspective on Learning and Innovation. Administrative Science Quarterly, 35(1): 128.

Cornell University, INSEAD, & WIPO. 2013. The Global Innovation Index 2013. Geneva, Ithaca, and Fontainebleau.

Cuervo-Cazurra, A., & Genc, M. 2008. Transforming disadvantages into advantages: Developing-country MNEs in the least developed countries. Journal of International Business Studies, 39(6): 957–979.

Cuevas, A., Messmacher, M., & Werner, A. 2005. Foreign direct investment in Mexico since the approval of NAFTA. World Bank Economic Review, 19(3): 473–488.

De Beule, F., Elia, S., & Piscitello, L. 2014. Entry and access to competencies abroad: Emerging market firms versus advanced market firms. Journal of International Management, 20(2): 137–152.

Journal of World Business, 44(1): 74–84.

DiMaggio, P., & Powell, W. 1991. The New Institutionalism in Organizational Analysis. The International Encyclopedia of Organization Studies. Chicago: University of Chicago Press. https://doi.org/10.1002/asi.

Dunning, J. H. 1958. American investment in British manufacturing industry. London: George Allen and Unwin.

Dunning, J. H. 1977. Trade, Location of Economic Activity and the MNE: A Search for an Eclectic Approach. The International Allocation of Economic Activity: Proceedings of Nobel Symposium Held at Stockholm: 395–418.

Dunning, J. H. 1993. Internationalizing Porter’s Diamond. MIR: Management International Review, 33(2): 7–15.

Dunning, J. H., Kim, C., & Park, D. 2007. Old Wine in New Bottles: A Comparison of Emerging Market TNCs Today and Developed Country TNCs Thirty Years Ago.

Dür, A., Baccini, L., & Elsig, M. 2014. The design of international trade agreements: Introducing a new dataset. Review of International Organizations, 9(3): 353–375.

Egger, P. H., & Lassmann, A. 2012. The language effect in international trade: A meta-analysis. Economics Letters, 116(2): 221–224.

Filippaios, F., Papanastassiou, M., & Pearce, R. 2003. The evolution of US outward foreign direct investment in the Pacific rim: A cross-time and country analysis. Applied Economics, 35(16): 1779–1787.

Frankel, J., & Rose, A. 2002. An Estimate of the Effect of Common Currencies on Trade and Income. The Quarterly Journal of Economics, 117(2): 437–466.

Freeman, C. 2004. Technological infrastructure and international competitiveness. Industrial and Corporate Change, 13(3): 541–569.

Gaffney, N., Karst, R., & Clampit, J. 2016. Emerging market MNE cross-border acquisition equity participation: The role of economic and knowledge distance. International Business Review, 25(1): 267–275.

Globerman, S. 2002. Trade, FDI and regional economic integration: Cases of Noth America and Europe. Enhancing Investment Cooperation in Northeast Asia. Honolulu.

Graf, M., & Mudambi, S. M. 2005. The outsourcing of IT-enabled business processes: A conceptual model of the location decision. Journal of International Management, 11(2 SPEC. ISS.): 253–268.

Gubbi, S. R., Aulakh, P. S., Sougata, R., Sarkar, M. B., & Reveendra, C. 2010. Do International Acquisitions by Emerging Economy Firms Create Shareholder Value? The Case of Indian Firms. Journal of International Business Studies, 41(3): 397–418.

Gubbi, S. R., & Elango, B. 2016. Resource Deepening Vs. Resource Extension: Impact on Asset-Seeking Acquisition Performance. Management International Review, 56(3): 353–384. Guillen, M. F., & Garcia-Canal, E. 2009. The American Model of the Multinational Firm and the

“New” Multinationals From Emerging Economies. Academy of Management Perspectives, 23(2): 23–35.

Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. 1998. Multivariate Data Analysis.