The factors influencing EMNEs’ knowledge-seeking FDI
Louise Geertzema
Master Dissertation
Dual Award
Advanced International Business Management and Marketing
December 2017
Newcastle University Business School
University of Groningen Faculty of Economics and Business
Newcastle Upon Tyne Groningen
Dr. Elizabeth Alexander Dr. Sathyajit Gubbi
1
ABSTRACT
2
TABLE OF CONTENTS
1. Introduction ... 4
2. Literature review ... 6
2.1. The OLI Paradigm... 6
2.2. Motivations for investing abroad ... 8
2.3. Knowledge-seeking FDI ... 9
2.4. Absorptive capacity ...12
2.5. Industry differences ...14
3. Research design ...17
3.1. Empirical context ...17
3.2. Data collection and sample ...17
3.3. Estimation model ...19
3.4. Variables and measurements ...20
4. Analysis and results ...23
4.1. Conditions and assumptions ...23
4.2. Descriptive statistics and correlations ...24
4.3. Binary logistic regression analysis ...24
4.4. Additional analysis and robustness checks ...27
5. Discussion and conclusion ...28
5.1. Discussion ...28
5.2. Limitations and recommendations ...30
5.3. Theoretical contribution ...31
5.4. Practical implications ...32
References ...34
Appendix A. CBA description by target firm ...42
Appendix B. List of research methods ...43
Appendix C. Industry classification ...44
Appendix D. Assumption testing ...46
3
LIST OF TABLES
Table 2.1 Different concepts describing strategic asset-seeking FDI 9
Table 3.1. Sample distribution by firm and CBA 19
Table 4.1 Summary of descriptive statistics and correlations 24
Table 4.2 Binary logistic regression analysis 26
Table A.1. CBA by target firm 42
Table B.1 Overview of variables, measures, and data sources 43
Table C.1 Technology intensity industry classification 44
Table C.2 Knowledge intensity industry classification 45
Table D.1 Independence of errors test 46
Table D.2 Collinearity statistics 46
Table D.3 Collinearity diagnostics 46
Table E.1 Complete correlation matrix 47
LIST OF FIGURES
Figure 2.1 Theoretical model 16
LIST OF ABBREVIATIONS
AMNE Advanced market multinational enterprise
CBA Cross-border acquisition
CFA Country-specific advantage
EMNE Emerging market multinational enterprise
FDI Foreign direct investment
FSA Firm-specific advantage
GDP Gross domestic product
KBR Knowledge-based resources
LOE Liability of Emergingness
MNE Multinational enterprise
OR Odds Ratio
OEM Original Equipment Manufacturer
R&D Research & Development
4
1. INTRODUCTION
Foreign direct investment (FDI) by multinational enterprises (MNEs) has been a central factor in shaping the contemporary global economy. In the past three decades, emerging market multinationals (EMNEs) have experienced increasingly rapid advances in international expansion by means of outward FDI (OFDI). Emerging markets reached the record level of $468 billion of outward investment in 2014, corresponding to 35% of global FDI outflows; and is increasingly towards developed countries (UNCTAD, 2015).
The EMNEs’ FDI boom has attracted the attention of academics, practitioners, and media alike. In fact, in challenging the underlying assumptions of conventional theory, EMNEs’ international expansions have led to an ongoing scholarly debate (Cuervo-Cazurra, 2012). Specifically, the (lack of-) traditional ownership advantages arguably have led to EMNEs’ dissimilar FDI patterns (Mathews, 2002). That is, as opposed to exploiting assets, EMNEs seek to augment those abroad, and desire to acquire the competitive advantages which they as latecomers, often lack (Luo and Tung, 2007). Accordingly, FDI flows from developing to developed countries are generally explained by the rationale of strategic asset-seeking FDI; which is aimed towards intangible assets such as technology, knowledge and brands (Dunning, 1993; 2006; Makino, Lau and Yeh, 2002). This widely recognized list of intangible assets sought are in a way, all knowledge-based resources.
Acquiring or owning knowledge-based resources is a crucial element for firms to survive, as it is an essential source for a sustained competitive advantage (Barney, 1991). In fact, in the increasingly globalized world economy, knowledge is ever more taking on greater importance as compared to labour, physical capital and natural resources (OECD, 1996). As knowledge increasingly becomes the preeminent resource of the firm, it is expected that knowledge will become the overarching motivation of FDI (Kedia, Gaffney, and Clampit, 2012). This seems to be even more crucial for EMNEs, given that they are latecomers to the international market, typically not own knowledge-based resources, and these are usually not available in their domestic market (Mathews, 2002, 2006).
5 challenged by Berry (2006) for addressing the issue on the country-level, thereby “making the problematic assumption that all firms within a country are of the same technological type” (Berry, 2006, p. 153). Other scholars have made calls for further advancement with a broader adoption of knowledge rather than solely considering technology (e.g. patents) (Kedia, Gaffney and Clampit, 2012)
This paper aims to address this lack of understanding in knowledge-seeking FDI literature, as well as bring some clarity in the EMNE scholarly discipline. Specifically, it is the purpose of this research to investigate the factors influencing EMNEs’ knowledge-seeking FDI. Thus, the main objective or research question of this study is:
RQ: What factors shape knowledge-seeking FDI by emerging market multinationals? The study addresses this question by adopting a lens of the organizational learning perspective. Following Makino and Inkpen (2003) EMNEs’ knowledge-seeking FDI includes: FDI
conducted through cross-border acquisitions aimed at acquiring knowledge from sophisticated target firms, which facilitates cross-border learning, and enhances local and global competitiveness. When knowledge-seeking and deciding for a potential location, the relative
economic development of the host country compared influences how the EMNE perceives the potential knowledge stock (Gupta and Govindarajan, 2000). Thus, it is expected that knowledge-driven FDI from EMNEs will be to the (economically) developed countries. Hence, this study researched EMNEs’ knowledge-seeking FDI by means of cross-border acquisitions (CBAs) to the developed (OECD) countries. The hypotheses were developed considering the knowledge based view of the firm and considering the dynamic capability of absorptive capacity (Zahra and George, 2002). Indian firms were analysed for the period 2008 to 2012 by means of a binary logistic regression analysis. The study finds that knowledge-seeking FDI is shaped by an interplay of country-, industry- and firm-level factors. Specifically, the research finds that EMNEs are more likely to pursue knowledge-seeking FDI when the host country has more industry-specific knowledge-based resources. Further, it finds that international experience increases EMNEs’ likelihood to conduct knowledge-seeking FDI, and that there is a moderating effect of technological capability with a host country’s KBR; in line with the absorptive capacity research stream (Cohen and Levinthal, 1990; Zahra and George, 2002).
6 research design and methods of data collection, measurement and analysis. The final chapters include the empirical results, discussion and conclusion.
2. LITERATURE REVIEW 2.1. The OLI Paradigm
Combining the notions of competitive advantage, trade-theory benefits, and internalization theory, Dunning (1980) created one of the most influential theories to explain international expansion by MNEs. This framework is widely known as the eclectic OLI paradigm, which identifies Ownership (O), Location (L), and Internalization (I) advantages that shape a firm’s FDI. That is, firms’ decisions towards international investments are made through an interplay of country- industry- and firm-level characteristics identified as the O, L and I advantages (Dunning, 1980). The ownership advantages are firm-specific factors and arise from a firm’s resources, which include all assets, capabilities and other firm attributes (Barney, 1991) that a firm may exploit internationally. Location advantages are country-specific factors as favourable institutional environments and capital markets; making foreign (i.e. local) operations more attractive than domestic. Lastly, the internalization advantages include the benefits of conducting operations internally in the firm through FDI (e.g. lower transaction costs, greater control) rather than relying on the market through arms-length transactions (Buckley and Casson, 1976). Furthermore, a key prerequisite of the framework is that all three (O, L, I) advantages must be present to successfully undertake FDI. However, EMNEs seem to not always satisfy these conditions, thereby posing challenges on the explanatory power of the OLI framework (Hennart, 2012), which has led to a scholarly debate.
2.1.1. The Goldilocks debate
7 and García-Canal, 2009; Li, 1998; Luo and Tung, 2007; Madhok and Keyhani, 2012; Mathews, 2002, 2006). On the other side, academics argue that despite observed differences in EMNEs’ internationalization, it neither demonstrates they are a new species of MNEs, nor does it generate a need for new theory; as current establishments are adequately equipped to explain EMNEs behaviour (Dunning, Kim and Park, 2008; Narula, 2012; Rugman, 2010). The third view takes on an approach somewhere in the middle; where scholars advocate for extending traditional theories (Cuervo-Cazurra, 2012; Ramamurti, 2012). Regardless of their point of view, academic are in consensus about the noteworthy observed differences in EMNEs’ sources of competitiveness (Amighini et al., 2015).
2.1.2. Competitive advantages
8
2.2. Motivations for investing abroad
EMNEs’ different motivations for investing abroad is a central issue in its internationalization, as it is dependent on, and constrained by, the resources that they hold or lack (Losada Otalora and Casanova, 2012). Traditional theory categorizes the four major motivations for FDI as efficiency-, market-, natural resource-, and strategic asset-seeking (Dunning, 1993, 2006). The market- and strategic resource-seeking are most widely accepted for EMNEs (Jain, Lahiri and Hausknecht, 2013; Luo and Tung, 2007). Recently, strategic-asset seeking motivations are increasingly dominating literature to explain EMNEs’ international expansion, and are arguably the main motivator for its FDI (Mathews, 2002, 2006). The strategic asset-seeking motivation reflects FDI intended towards asset-augmenting rather than asset exploiting (Makino, Lau and Yeh, 2002). It is aimed at acquiring strategic assets (e.g. technology, marketing and management expertise) in the host country, potentially by CBAs of foreign firms (Dunning, 2008; Makino, Lau and Yeh, 2002). Dunning (2001) describes strategic asset seeking as “to create or gain access to resources and capabilities that complement their existing core competencies”(Dunning, 2001, p. 183). In the context of EMNEs, literature has emphasized the impact of the acquired strategic assets beyond this scope: as ‘offsetting competitive disadvantages’ (Rui & Yip, 2008; Luo and Tung, 2007), and ‘enable or enhance global competitiveness’ (Cui, Meyer and Hu, 2014; Luo and Tung, 2007). As noted by Meyer (2015), the notion seems to capture a different implication when applied to EMNEs’ foreign investments. Moreover, literature has produced a variety of different conceptualizations essentially describing the same: FDI aimed at acquiring intangible assets (table 2.1). Therefore, it is useful to revisit EMNEs’ motivations for investing abroad from the organizational learning perspective.
2.2.1. EMNE motivations and organizational learning
9 2008). Moreover, EMNEs exploit their current FSAs, as for instance their expertise in mass production manufacturing, by expanding to other emerging countries (Luo and Tung, 2007). The strategic asset-seeking FDI as described in EMNE literature, reflects explorative learning; it involves acquiring intangible resources outside the boundary of the firm (Cui, Meyer and Hu, 2014) and is characterised by high risk and uncertain outcomes (Deng, 2010). To illustrate, as latecomer MNEs, they must accelerate the pace of internationalization to catch up with incumbents (Mathews and Zander, 2007), and have been expanding abroad via risk, high-control entry modes (Luo and Tung, 2007). In addition, there is a third type of foreign expansion, which captures both exploitative as explorative learning; as described in the concept of ‘Knowledge-seeking FDI’ (Makino and Inkpen, 2003).
2.3. Knowledge-seeking FDI
Adopting a more balanced view of the motivation of the firm as captured in the concept of knowledge-seeking FDI, it is proposed that firms invest abroad to acquire new knowledge, but simultaneously further exploit existing capabilities that facilitate cross-border learning (Makino
Table 2.1. Different concepts describing strategic asset-seeking FDI
Authors Concept Definition
Dunning 1991 Strategic asset seeking FDI “to create or gain access to resources and capabilities that complement their existing core competencies” (p. 183) Cui, Meyer & Hu,
2014
Strategic asset seeking FDI “to pursue long-term strategic objectives - especially that of sustaining or advancing global competitiveness” (p.490) Makino, Lau &
Yeh, 2002
Asset seeking FDI “to acquire strategic assets (i.e. technology, marketing and management expertise) available in a host country” (p.404) Dunning & Narula
1995
Strategic asset seeking R&D
“R&D activities … aimed at monitoring or acquiring competitive advantages - particularly in the technology and information-intensive sectors- which are complementary to those already possessed by the MNE” (p.42)
Rui & Yip, 2008 Strategic intent perspective of FDI
“to achieve specific goals, such as acquiring strategic
capabilities to offset their competitive weakness and leveraging their unique ownership advantages while making use of institutional incentives and minimizing institutional constraints.” (p. 214)
Kogut & Chang, 1991
Technology Seeking FDI “the seeking of new technologies resident in the USA” (p.401) Chung & Alcacer,
2002
Knowledge seeking FDI “expand abroad in search of capabilities that are not available in their home countries” (p.1534)
Li, Li & Shapiro, 2012
Knowledge seeking FDI “FDI that is geared … to augmenting firm-specific advantages through acquisition or partnering arrangements with local firms” (p. 278)
Luo & Tung, 2007 Springboard FDI “to acquire strategic assets needed to compete more effectively against global rivals and to avoid the institutional and market constraints that they face at home.” (p.482)
Hennart, 2012 Intangible-seeking investments
10 and Inkpen, 2003). That is, the acquisitive learning process enables experimental learning; in which the firm integrates and exploits the acquired strategic assets to create FSAs (Zahra, Nielsen, & Bogner, 1999). Hence, in knowledge-seeking FDI, the processes of exploitation and exploration are dynamically linked (Noorderhaven, Koen and Sorge, 2015). As stated by Makino and Inkpen (2003), in knowledge-seeking FDI the firm aims to acquire strategic assets such as technological, innovative, management and marketing expertise, or other intangible resources owned by other firms in locations abroad. This is especially relevant for EMNEs, due to the lack of essential knowledge-based FSAs (Rugman, 2010) and weak locally available intangible assets (Dunning and Lundan, 2008). Furthermore, EMNEs are confronted with a strong presence of global rivals in their home market (Luo and Tung, 2007) and have to compete with the typically more experienced and resource-richer MNEs (Elango and Pattnaik 2011; Hennart, 2012). Moreover, because EMNEs face a knowledge gap, many typically lack a sustainable competitive advantage in the global market (Mathews, 2002, 2006; Kedia, Gaffney and Clampit, 2012; Ramamurti, 2012).
However, when conducting CBAs in search for strategic assets, EMNEs have been found to acquire firms in developed countries which are more advanced in terms of technology and managerial skills, helping them to overcome their resource hurdles (Cui, Meyer and Hu, 2014; Deng, 2009; Rui and Yip, 2008). Additionally, these assets are strategic in such that they increase the EMNEs’ capabilities not only in the foreign market, but also in the home market and global operations, strengthening their competitive position (Chen, Li and Shapiro, 2012; Lancheros, 2016; Nair, Demirbag and Mellahi, 2015). Moreover, in the process of knowledge-seeking FDI, expanding abroad simultaneously helps EMNEs to achieve competitive advantages (Elango and Pattnaik 2011; Luo and Tung, 2007; Nair, Demirbag and Mellahi, 2015). This is because international markets could serve as learning laboratories in which firms’ ability to compete internationally may grow through experiential learning (Hitt, Hoskisson, and Kim, 1997). Knowledge-seeking FDI could hence serve as an essential mechanism to overcome EMNEs’ latecomer disadvantages (Elango and Pattnaik, 2011; Mathews, 2002). In fact, for the latecomer EMNE, international expansions forms a basis for organizational learning and allows for accelerated cross-border learning trajectories (Bonaglia, Goldstein and Mathews, 2007; Li, 2010).
2.3.1. Knowledge-based resources
11 capabilities it seeks. This implies that the EMNE will be more likely to invest in countries that have the richest endowments of knowledge-based resources (KBR) for a variety of reasons. First, the amount and quality of available knowledge depends on knowledge-generating institutions as innovation networks, educational institutions and R&D labs, which are location bound (Almeida and Kogut, 1999). Second, when a country has more KBR, more potential acquisition targets will be available that are likely to possess superior knowledge and competencies. This is because these firms have been able to benefit from large pools of specialized researchers and academia and state-of-the art technologies from research labs (Kalasin, Dussauge and Rivera-Santos, 2014). Third, the EMNE will consider the potential reverse knowledge transfer, as it is a key objective to transfer the acquired knowledge back home and improve its technological capabilities (Chen, Li and Shapiro, 2012). Research has found that the acquired foreign subsidiary plays a critical role in reverse knowledge transfer, and is essential transforming the host country location-bound resources into FSAs for the entire EMNE (Nair, Demirbag and Mellahi, 2015). Hence, the target firm’s human resources and communication skills are essential, especially for knowledge transfers in CBAs (Bresman, Birkinshaw and Nobel, 2010), and for subsequent cross-border knowledge exchange (Berry, 2014). As communication between R&D workers is accompanied by local innovation networks and national infrastructure (Globerman, Shapiro and Vining, 2005), firms from countries with high KBR are more likely to be competent.
12 the decision of knowledge-seeking FDI. For instance, research has shown that EMNEs investing in developed countries in search for localized knowledge and innovation, are drawn to industry-specific technologies (Crescenzi, Pietrobelli and Rabellotti, 2016). Although all developed countries have a relatively high technological development, their predominance and sophistication vary within different industries (Cantwell and Janne, 1999). This is reflected in R&D and industry clustering; and as a result, countries differ in knowledge profiles (Chung and Yeaple, 2008). Thus, when a country has more industry-specific KBR, it is more likely the EMNE will pursue knowledge-seeking FDI. Hence, it is hypothesized that:
Hypothesis 1 (H1): A host country’s industry-specific knowledge-based resources
increases the likelihood of an EMNE pursuing knowledge-seeking FDI.
2.4. Absorptive capacity
13
2.4.1. The ability to identify and understand knowledge
The essential idea of absorptive capacity is that prior related knowledge facilitates learning or absorption of new knowledge (Kim, 1998). Cohen & Levinthal (1990) further suggest that the prior related knowledge of the firm essentially reflects knowledge and skills of the individual employee; and that these development process require time and intensity effort. As emphasized by Van Den Bosch, Van Wijk, and Volberda (2003), the extent of prior knowledge might be classified as a depth and breadth dimension. When confronted with new knowledge, the unit first needs to acquire the requisite breadth of knowledge (Cohen and Levinthal, 1990). The breadth enables absorptive capacity in unrelated fields, thereby increases the potential of the firm to explore new knowledge (March, 1991). It is thus related to knowledge diversity, which enables the employees to make novel associations and linkages, and increases the identification dimension (Cohen and Levinthal, 1990). EMNE literature has commonly adopted the firm’s international experience as the distinct organizational measure for knowledge diversity (Deng, 2010; Elango and Pattnaik, 2011; Gubbi and Elango, 2016). International experience deepens EMNEs understanding of international markets and enable access to diverse, locally embedded ideas (Chittoor et al., 2009). Through corporation with foreign competitors and suppliers and customers, the EMNE accumulates diverse knowledge of foreign markets (Luo and Tung, 2007) and thereby increases its ability to recognize the external knowledge (Deng, 2009). Thus, EMNEs with international experience are more likely to pursue knowledge-seeking FDI. Hence, it is hypothesized that:
Hypothesis 2 (H2): EMNEs’ international experience increases the likelihood of
EMNEs pursuing knowledge-seeking FDI.
14 technological capabilities facilitate its understanding of the superior intangible resources held by the target firm. This specific deep R&D knowledge helps the EMNE understand the industry, product, processes and customers from the target firm, facilitating its ability to understand and value the new knowledge (Deng, 2010). Thus, EMNEs with technological capabilities are more likely to pursue knowledge-seeking FDI. Hence, it is hypothesized that:
Hypothesis 3 (H3): EMNEs’ technological capabilities increase the likelihood of
EMNEs pursuing knowledge-seeking FDI.
2.5. Industry differences
The type of knowledge sought (i.e. intangible resources as technological, innovative, management and marketing expertise) are resources with a rather high knowledge complexity, which contributes to the ‘stickiness’ or tacit nature of the required knowledge (Grant, 1996). When firms aim to acquire this type of knowledge, transferring its tacit elements poses difficulties (Kogut and Zander, 1992). Knowledge transfer is furthermore hampered if the home and host-country environments are different (Chen, Li and Shapiro, 2012). Moreover, international knowledge transfers often require adaptation, and transfer is subject to distance; as communication and quality diminishes with large differences in infrastructure between the foreign and receiving firm (Teece, 1977). Until this point, knowledge-seeking FDI has been described for all EMNEs’ industries. However, the necessity for knowledge is dependent upon the industry context (Kedia, Gaffney and Clampit, 2012), as EMNEs’ strategies and investment paths have been found to be industry-specific (Hobdari et al., 2017).
2.5.1. Manufacturing industries
15 Alvstam, 2011). Likewise, OEM arrangements with AMNEs improved EMNEs’ manufacturing expertise, enabling them to undertake subsequent FDI to further augment their resource base (Child and Rodrigues, 2005; Luo and Tung, 2007). Moreover, pursuing knowledge-seeking FDI is likely to be most important for the high-technology manufacturing EMNE; which is characterized by high levels of R&D investments (Nair, Demirbag and Mellahi, 2015). Knowledge in such industries is often embedded in host country local networks (Almeida and Kogut, 1999). Additionally, such industries are innovative in nature, and requires frequent upgrading of knowledge (Kumar, 2007). Taken together, it is expected that manufacturing EMNEs with more R&D investments, when combined with more host country industry-specific KBR, are more likely to engage in knowledge-seeking FDI. Hence, it is hypothesized that:
Hypothesis 4 (H4): For manufacturing sector EMNEs, technological capabilities
positively moderate the relationship between host country industry-specific knowledge-based resources and the likelihood of pursuing knowledge-seeking FDI.
2.5.2. Services industries
16 capabilities to acquire advanced target firms (Deng, 2010). Moreover, these elements seem to be reinforcing; the departure of the knowledge base promotes experimentation, facilitates explorative learning, and fosters the ability to comprehend new opportunities (Luo and Peng 1999). Taken together, it is expected that service EMNEs with more international experience, when combined with more host country industry-specific KBR, are more likely to engage in knowledge-seeking FDI. Hence, it is hypothesized that:
Hypothesis 5 (H5): For service sector EMNEs, international experience positively
moderates the relationship between host country industry-specific knowledge-based resources and the likelihood of pursuing knowledge-seeking FDI.
17
3. RESEARCH DESIGN
As stated by Mack (2010), the choice of research questions and assumed adequacy of methods addressing these, are inextricably linked with one’s prior assumptions relating to ontology, epistemology, and views on human nature. In line with the positivistic paradigm, the author assumes knowledge can be observed externally, which aligns with a methodology that focuses on an examination of relationships (Kuada, 2012) in which explanations of a causal nature could be provided (Mackenzie and Knipe, 2006). In accordance, the research is aimed to study a cause-and-effect relationship; between EMNEs’ FDI to developed countries and (1) host country knowledge-based resources, and (2) firm absorptive capacity.
3.1. Empirical context
The empirical context of this study is CBAs by Indian multinationals. India provides for an appropriate research setting since it is the second largest emerging economy, and has experienced immense growth rates in OFDI; including a five-fold growth jump in OFDI flows in 2014 (UNCTAD, 2015). Furthermore, India’s OFDI is increasingly geared to developed countries and dominated by knowledge-based industries (Sauvant and Pradhan, 2010). These patterns are reflected in Indian EMNEs’ strategic asset-seeking motives and desire to achieve competitive advantages in global markets (Elango and Pattnaik 2011; Ramamurti 2009). Increasingly, FDI occurs through acquisitions to acquire technological advancements and innovative capabilities residing in the developed economies (Kumar 2007; Pradhan, 2005, 2008). Thus, India’s CBAs in developed countries fits the overall research setting very well. Moreover, by examining CBAs it reduces the scope, enabling to exclude for market-seeking motives. CBAs were analysed for the period 2008 to 2012. This period covers new investment patterns for the post-financial crisis era following the financial crisis in 2008 (Yoo and Reiman, 2017). Additionally, the event window of five years enables for excluding the effects of any macro-economic shocks or other exogenous factors that could influence the results. At last, the selected period is apt for constructing a representative sample size and allows for an adequate analysis of answering the research questions.
3.2. Data collection and sample
18 of Indian firms (e.g. Bhaumik, Driffield and Zhou, 2017; Gubbi and Elango, 2016), and allowed for a wide inclusion of the desired control variables. Furthermore, both the Zephyr and ORBIS data were once more cross-referenced with the rich Prowess database maintained by the Centre for Monitoring the Indian Economy (CMIE), to filter out any inconsistencies and assure authenticity of the data. The Prowess database is widely adopted by previous research analysing Indian FDI (e.g. Buckley et al., 2016; Kumaraswamy et al., 2012; Lancheros, 2016), and provided essential firm-specific financial information. At last, country-industry information was collected from the OECD Science, Technology and R&D Statistics database, and aggregate macro-level information on the host countries was extracted from the World Bank’s World Development Indicator (WDI) database, and data provided by the Centre d’E-tudes Prospectives et d’Informations Internationales (CEPII).
19
3.3. Estimation model
The study aims to investigate location- and firm-specific factors affecting the propensity of EMNEs to invest abroad, in the context of knowledge-seeking FDI. The hypotheses will be tested with a binary logistic regression analysis based on a dichotomous dependent variable, which predicts the likelihood of Indian firms to undertake a CBA in a specific developed country. Logit models are largely employed in location and FDI literature (Jain, Kundu and Newburry, 2015) and are the most appropriate analysis when the dependent variable is dichotomous, and the predictor variables are continuous, categorical and/or binary (Leech, Barret and Morgan, 2015).
Like other regression analyses, the logistic regression allows for assessing the impact of a set of predictors variables on an outcome; which is in this case, the probability to invest (Pallant, 2013). Thus, the regression coefficients will estimate the impact of the variables (i.e. host country KBR and firm absorptive capacity) on the probability of EMNEs for conducting a CBA in a specific
Table 3.1. Sample distribution by firm and CBA
Acquiring firm industry1 Number of firms Target country Number of CBAs
Manufacturing sector (67) Australia 13
Food, beverages 4 Belgium 1
Textiles, leather 5 Canada 3
Chemicals 5 Czech Republic 1
Pharmaceuticals 14 Denmark 1
Rubber, plastic, metals 10 Finland 1
Electronics, electric equipment 6 France 9 Other machinery and equipment 8 Germany 10 Motor vehicles, other transport 6 Greece 1 Other manufacturing, construction 9 Hungary 1
Services sector (48) Ireland 1
Information, communication 8 Italy 7
Computer programming 33 Japan 2
Financial, insurance 3 Netherlands 5
Scientific, technical 2 Portugal 1
Administrative, support 2 South Korea 1
Total firms 115 Spain 9
By technology & knowledge intensity2 Turkey 1
High-tech manufacturing 16 United Kingdom 33 Medium-tech manufacturing 33 United States 43 Low-tech manufacturing 18 Total CBAs 144
High-tech knowledge-intensive services 37
Knowledge-intensive services 5 By Year
Low knowledge-intensive services 6 2008 57
Total firms 115 2009 17
By ownership type 2010 27
Family-owned 7 2011 24
Foreign-owned 8 2012 19
Business group 58
Total firms 115 Total CBAs 144
1For target firm industry distribution see appendix A
20 country. The logistic function for estimating the probability of an event happening is given by
the log of the odds ratio, 𝑙𝑛( 𝑝̂
1−𝑝̂), where 𝑝̂ denotes the probability of an acquisition being
undertaken, i.e. A=1; and (1 − 𝑝̂) gives the probability that A=0 (Verma, 2009). In effect, this logit is the dependent variable Z against which the independent variables are regressed. Specifically, the logistic function is as follows:
𝑓(𝑍) = 𝑒
𝑍
1 + 𝑒𝑍
Where Z denotes the logit of A, and is the algebraic conversion from the combined linear regression equation:
Ait = α + β1HostKBRij+ β2(HostKBRij) + β3(ACkjt) + β4(HostKBRij∗ ACkjt) + Xijt+ ε In which the acquisition 𝐴 from an Indian EMNE to a developed country i in year t is predicted; where α is a constant; 𝛽1-𝛽4 are the parameters (i.e. regression coefficients); HostKBR represents the value of knowledge-based resources of country i in industry j; 𝐴𝐶 reflects the elements of absorptive capacity of firm 𝑘 at time 𝑡; 𝛸𝑖𝑗𝑡 is a vector of the control variables; and ε is the error term.
3.4. Variables and measurements
An overview of the variables, measures and its data sources can be found in appendix B (table B.1). Following extant literature on FDI-modelling, all variables except for the dependent and dummy variables are transformed into natural logarithms (Buckley et al., 2007; Kalotay and Sulstarova, 2010). The log transformations also substantially improved the regression model. Moreover, in accordance with previous research on EMNEs’ FDI, average values were calculated for all country-level variables (Buckley et al., 2007). In calculating the mean from 2008 and 2012, effects from extreme between-year differences are minimized and controlled for, and more robust variables are obtained (Drogendijk and Blomkvist, 2013).
Dependent variable. The dichotomous dependent variable is constructed for every combination of all firms and all host countries in the sample (115 x 20 = 2300), which receives a value of 1 when a firm undertakes a CBA in the country, and 0 if not.
21 building technological networks. It is a rich measure as it reflects business’ total efforts in R&D, including investments that generate valuable knowledge creation but fail to result in registered patents (Gornik-Tomaszewski and Millan, 2005). Since technological knowledge is largely industry-specific, country-industry measures are preferred over country-level measurement (Nair, Demirbag and Mellahi, 2015). Hence, only the R&D investments in the relevant industry of the investing firm were taken into account. Specifically, the 2-digit ISIC primary industry code of the investing firm was matched with that of the host country’s R&D investments. Next, only the R&D expenditures of that industry was taken as the value. Mean values for 2008 to 2012 were calculated to achieve a more representative measure, thereby minimizing year-related outliers and missing values due to a transformation of the ISIC Rev.3 to Rev.4 classification system in 2008.
Following extant literature, absorptive capacity is analysed by considering two variables: international experience and technological capability (Deng, 2010). The variable international
experience is measured using export intensity, calculated as the ratio of export sales to total sales.
The variable technological capability is measured by R&D intensity which is calculated as the ratio of R&D investments to net sales. Both measures have been used in extant literature (Anand and Delios, 2002; Elango and Pattnaik, 2011; Gubbi and Elango, 2016).
Control variables. Several control variables have been included in line with previous literature. First of all, it is controlled for alternative motives for making investments overseas; for instance, market-seeking could be an alternative driver for investment (Makino, Lau and Yeh, 2002); Host
country GDP per capita and GDP Growth are used to control for the size of the market (Li, Li
22 context of knowledge-seeking FDI (Chen, Li and Saphiro, 2012). Hence, a dummy variable is included to indicate if the host country shares the same official language.
With regards to firm specific factors, various controls are accounted for. Firm size, calculated by a firm’s total assets in the respective year of the acquisition, is found to influence FDI (Bhaumik, Driffield and Pal, 2010). Moreover, firm age could influence the decision to invest abroad, and is measured by substracting the year of the acquisition by the year of incorporation (Buckley et al., 2016). For the ‘no-acquisition’ observations, the 2008 are taken to avoid erroneous conclusions due to reverse causation; as increased assets could also arise after or from an acquisition. Moreover, several ownership dummies were included typical for Indian MNEs:
state-ownership when the firm was state-owned, foreign-ownership when the global ultimate
owner was a foreign firm, and family-owned when it was owned by a family. Furthermore,
business group affiliation is found to influence outward investment patterns of Indian MNEs
23
4. ANALYSIS AND RESULTS
Exploratory data analysis led to a few alterations of the data. For instance, missing values for controls (e.g. firm size) were estimated by values of the closest year available. Additionally, by generating Z-values some extreme outliers (Z-values > 7) were detected and winsorized (Tukey and McLaughlin, 1963). All in all, 1.6 percent of the dataset was modified.
4.1. Conditions and assumptions
To assure statistical correctness, the mandatory conditions have been considered. Two were met through the study design; (1) the dependent variable (DV) is dichotomous with mutually exclusive categorical outcomes, and (2) the sample size meets the minimum of 60 total cases and the threshold of least 15 cases per IV (Leech, Barrett and Morgan, 2015). Further, several assumption tests were conducted; the results can be found in appendix D. First, logistic regression assumes linearity between continuous predictors and the logit of the DV, which was assessed by the Box-Tidwell procedure. That is, through examining significance of interaction terms between variables with its log transformations (Wilson and Lorenz, 2015), linearity was confirmed. The Independence of errors assumption holds that the errors of observations should not be correlated. Hence, the Durbin-Watson test was conducted, which varies between 0 and 4; where 2 means the residuals are uncorrelated (Field, Miles and Field, 2013). The statistic has a value of 2.18, and thereby shows negligible serial correlations between errors. The
Multicollinearity assumption was assessed by examining collinearity statistics; including the
24
4.2. Descriptive statistics and correlations
Additionally, the correlation matrix was analysed as summarized below (table 4.1); the complete matrix can be found in appendix E. Acquisition significantly correlates with Host country KBR (p <.01) in accordance with hypothesis 1. As expected, the DV significantly correlates to many of the control variables as geographic distance, GDP, natural resources and a shared language. Firm R&D and export intensity is not correlated to the DV but is significantly correlated to host country KBR and almost all firm control variables. Additionally, many significant correlations are observed among the variables, however are not surprising as they largely reflect within country-level and firm-level relations. Moreover, these correlations are all very low and meet the conditional threshold of 0.7 by far; and are therefore no concern for posing any problems with the statistical analysis (Pallant, 2013).
4.3. Binary logistic regression analysis
The analysis was conducted with SPSS 25 by means of a binary logistic regression. As the data is multilevel pooled data, including repeated country-, industry-, firm-, factors, it is likely that
Table 4.1. Summary of descriptive statistics and correlations (N=2300)1
Mean SD Min Max 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1 CBA (DV) 0.06 0.24 0 1 1 2 Host country KBR 19.57 1.86 12.86 24.58 .22** 1 3 R&D intensity 0.013 0.030 0.000 0.247 .03 .13** 1 4 Export intensity 0.418 0.382 0.000 1.00 .05* .07** .11** 1 5 Cultural distance 63.02 9.50 44.92 79.39 -.05* 0.02 -.00 -.01 1 6 Geographic distance 8.85 0.25 8.44 9.48 .23** .32** .00 .01 -.10** 1 7 GDP 27.56 1.23 25.64 30.35 .27** .71** .01 .01 -.06** .41** 1 8 GDP growth 0.03 1.94 -5.31 4.11 .04* .32** .00 -.00 -.06** .05* .30** 1 9 Natural resources 4.74 6.47 0.79 31.81 .06** -.00 -.00 -.00 .02 .34** .08** .26** 1 10 Inflation 2.58 1.63 -0.20 8.12 -.03 -.23** -.00 -.00 -.28** -.42** -.27** .39** .05** 1 11 Shared language 0.25 0.43 0 1 .24** .20** .00 .01 -.14** .74** .31** .16** .45** -.17** 1 12 Firm size 19.03 1.95 9.98 23.99 .06** -.05* .19** -.07** -.00 .01 .01 .00 .00 .00 .01 1 13 Firm age 28.85 20.72 2 90 .04 -.12** -.04* -.26** -.00 .01 .01 .00 .00 .00 .01 .35** 1 14 Family owned 0.06 0.21 0 1 -.00 -.04* .06** -.03 .00 .00 .00 .00 .00 .00 .00 -.01 -.22** 1 15 Foreign owned 0.07 0.24 0 1 -.01 -.08** -.10** .20** .00 .00 .00 .00 .00 .00 .00 .07** -.01 .12** 1 16 BG affiliation 0.50 0.50 0 1 .01 -.05* .04 -.25** .00 .00 .00 .00 .00 .00 .00 .41** .43** -.04 -.18** 1
25 both relations as variability exists between the groups (e.g. an industry) as in the units nested within the groups (e.g. firms in same industry). Hence, a hierarchical multi-level regression was conducted (table 4.2), where variables are included in the model at different levels (i.e. ‘building blocks’) at a predetermined order (Wilson and Lorenz, 2015). The first model1 contains all the control variables2, which significantly explain the DV, 𝛸2 (10) = 231.204 (p < .01). The Hosmer
and Lemeshow (HL) test evaluates the fit of the model by assessing if the model is significantly different from a good fitting model, and thus, should be insignificant (p > .05), which is the case. Furthermore, the Nagelkerke 𝑅2 of .260 shows that 26 percent of the variance in the DV is explained by the model; and 93.6 percent is correctly classified. GDP, inflation, shared language and firm size are highly significant (p <.01) with an odds ratio (OR) larger than 1, meaning that these factors increase the likelihood of an EMNE undertaking a CBA. GDP growth reduces the likelihood of a CBA with an OR of 0.78 (p <.01); as does BG affiliation, although it is weakly significant (OR=.70, p <.10). Natural resources shows a small and weakly significant effect by an OR of 1.03 (p <.10).
Model 2 shows that adding Host country KBR significantly improved the ability of the model to predict a CBA, 𝛸2 (11) = 238.457 (p < .01); a larger variance explained (Nagelkerke 𝑅2 of .267),
and 93.7% correctly classified. Host country KBR is highly significant with an OR of 1.25 (p <.01), confirming that it increases the likelihood of EMNEs pursuing a CBA; supporting hypothesis 1. Thus, when a (potential) host country has more industry-specific knowledge-based resources, EMNEs are more likely to conduct a CBA in that country. In model 3, R&D intensity and export intensity were added, which significantly improved the model in predicting the DV, 𝛸2 (13) = 241.753 (p < .01). The HL-statistic shows the model is a good fit for the data (p >.05)
and as observed from the Nagelkerke 𝑅2 of .271, a greater variance explained. Export intensity
shows a significant and positive effect, with an OR of 1.55 (p < .05), supporting hypothesis 2. That is, EMNEs with more international experience are more likely to conduct an CBA. R&D intensity does not show a significant effect, which indicates that firms with a higher R&D intensity are not necessarily more likely to conduct a CBA. Therefore, hypothesis 3 is not supported. The result might come from the large variance of R&D intensity of the firms in the sample, as firm R&D investments are often partly industry dependent (Lancheros, 2016).3
1 The original first models also included Cultural distance, but had to be omitted due to collinearity issues.
2 Additional models were tested with the control variable Family ownership, and the industry controls HT-KIS, KIS, HTM and MTM; but showed no significant differences or improvements of the models.
26 Next, the sample was divided into manufacturing and service industries. Model 4 to 6 (manufacturing) show a lightly lower variance explained (Nagelkerke 𝑅2 = .243) but the models
are still a good fit for the data (HL statistic p>.05). First, model 3 was repeated for the manufacturing sample in model 4, which shows similar results. Host country KBR has an OR of 1.25 (p <.05), R&D intensity is not significant, and export intensity has an OR of 2.10 (p >.05). In model 5, the interaction effect of R&D intensity and Host country KBR is added, which significantly improved the base model (model 1) with a 𝛸2(11) of 130.117 (p < .01). The
interaction term shows an OR of 1.58 (p <.05), supporting hypothesis 4. Model 6 includes all variables, and shows that only Host country KBR and export intensity are significant. The fact that some variables turn insignificant is not uncommon, since the variance is being shared through double inclusion. In sum, the results indicate that R&D intensity on its own, does not have an effect, whilst Host country KBR does. However, when combined, it results in a significantly larger likelihood (i.e. a higher OR). When a (potential) host country has more
TABLE 4.2. Binary logistic regression (DV: CBA)1
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8 Model 9 Predictors OR2 OR OR OR OR OR OR OR OR Constant 0.00*** 0.00*** 0.00** 0.00*** 0.00*** 0.00*** 0.00*** 0.00*** 0.00*** Geographic distance 0.56 0.51 0.49 0.17** 0.20** 0.20*** 1.16 1.59 1.20 GDP 2.72*** 2.14*** 2.16*** 2.05*** 2.62*** 2.04*** 2.07** 3.13*** 1.99** GDP Growth 0.78*** 0.74*** 0.74*** 0.74*** 0.73*** 0.74*** 0.69** 0.73** 0.68** Natural resources 1.03* 1.04* 1.04* 1.05** 1.04** 1.05** 1.05 1.03 1.05 Inflation 1.30*** 1.35*** 1.34*** 1.37*** 1.32*** 1.37*** 1.26 1.27 1.29 Shared language 3.66*** 3.81*** 3.83*** 5.96*** 4.91*** 5.98*** 2.15 2.67* 2.06 Firm size 1.17*** 1.17** 1.16** 1.09 1.10 1.09 1.20** 1.18* 1.20* Firm age 1.26 1.14* 1.15** 1.16** 1.48** 1.66** 1.67 1.47 1.76 Foreign-owned 0.62 0.70 0.62 0.74 0.59 0.74 0.55 0.49 0.55 BG affiliation 0.67* 0.67* 0.71 0.77 0.70 0.78 0.59 0.65 0.57 Host country KBR 1.25*** 1.24** 1.25** 1.23** 1.65** 1.43 R&D intensity 0.74 17.18 0.00 0.13 0.12 Export intensity 1.17 ** 2.10** 2.09** 2.31** 0.01 Host KBR * R&D intensity 1.58 ** 1.65 Host KBR * Export intensity 1.05 * 1.35 Observation (N) 2300 2300 2300 1340 1340 1340 960 960 960 Chi-square 231.204*** 238.467*** 241.753*** 124.419*** 130.117*** 124.475*** 137.074** 133.336** 138.166** -2 Loglikelihood 829.455 844.420 818.524 487.233 494.437 487.177 290.134 293.872 289.041 Nagelkerke 𝑅2 .260 .267 .271 .243 .240 .243 .354 .363 .366 HL-statistic .961 .878 .521 .863 .806 .822 .844 .895 .993 % classified 93.6 93.7 93.8 93.6 93.5 93.6 94.4 94.9 94.4 *** significant at.01, **significant at .05, *significant at .10
1 Model 1-4: full sample, Model 4-6: manufacturing sector, Model 7-9: service sector
27 knowledge-based resources, all EMNEs are more likely to do a CBA; but EMNEs with abundant R&D intensity are even more likely to do so.
Models 7 to 9 relate to the service subsample, which are all a good fit for the data (HL-statistic p>.05) and a larger variance explained (Nagelkerke 𝑅2 between .35 and .37). Model 7 is a repetition of model 3, showing similar results; Host KBR is positive and significant (OR = 1.65, p <.05), as is export intensity (OR = 2.31, p <.05), and R&D intensity does not have any effect. The negligible effect is likely to be expected, as almost none of the firms invest in R&D. Model 8 shows the interaction effect of Host KBR with export intensity, with an OR of 1.04 (p <.1). The OR being higher than 1 indicates it increases the likelihood of an CBA, although this effect is quite minimal and weakly significant. Therefore, hypothesis 5 is partly supported. Model 9 includes all variables and shows that none of the variables are significant. Apparently, both international experience as a host country’s knowledge-based resources increase the likelihood of a CBA, but there is not so much a combined increasing effect.
4.4. Additional analysis and robustness checks
28
5. DISCUSSION AND CONCLUSION 5.1. Discussion
This study investigated knowledge-seeking FDI by EMNEs. Specifically, it was aimed at exploring the factors that shape EMNEs’ knowledge-seeking FDI by means of five hypotheses addressing host country and firm specific factors. The research setting was designed to capture knowledge-seeking FDI; by analysing cross-border acquisitions (CBAs) by Indian multinationals in OECD countries. The findings have several implications. First, the paper finds strong support for the first hypothesis: when a country has more industry-specific KBR, EMNEs are more likely to pursue knowledge-seeking FDI. This is in line with the theoretical framework of Kedia, Clampit and Gaffney (2012), which states that the type of knowledge sought influences EMNEs’ location choice and entry mode. That is, when seeking technology, R&D and/or management expertise, FDI will be towards developed countries through CBAs or joint ventures (Kedia, Clampit and Gaffney, 2012). It also supports the strategic asset-seeking EMNE literature, that theorize EMNEs invest abroad to acquire intangible, strategic assets via acquisitions, as captured in the ‘Springboarding perspective’ (Luo and Tung, 2007).
The findings also show that EMNEs with international experience are more likely to engage in knowledge-seeking FDI. This is in line with the theories outlined in the absorptive capacity literature (Cohen and Levinthal, 1990; Zahra and George, 2002). Specifically, it corresponds with the notion of knowledge diversity; when confronted with new knowledge, the firm first needs to acquire the requisite breadth of knowledge (Cohen and Levinthal, 1990). This breadth enables absorptive capacity in unrelated fields, thereby increase the potential of the firm to explore new knowledge (March, 1991). It finds correspondence in Madhok and Keyhani’s (2012) statement that the LOE resulted in EMNEs developing essential capabilities as entrepreneurial drive and learning agility.
29 the foreign knowledge, as it influences the ability to analyze understand the information obtained from external sources (Kim, 1998).
Moreover, although the results indicate that initial technological capabilities do not increase EMNEs likelihood to pursue knowledge-seeking FDI, the effect of technological capabilities when combined with industry-specific host KBR, does. One could also argue that technological capabilities are only pertinent for the EMNE when the host country has knowledge relevant to the firm; i.e. ample industry-specific host KBR. Moreover, it suggests that high R&D-intensive firms are more likely to pursue knowledge-seeking FDI, as knowledge in such industries is often embedded in host country local networks (Almeida and Kogut, 1999). The results also show that host KBR itself increases EMNEs’ likelihood to pursue knowledge-seeking FDI, and EMNEs are more likely to do so when they have developed absorptive capacity (Cohen and Levinthal, 1990). Thus, although firms might invest abroad due to certain locational factors, they are more likely to do so when they have ample dynamic capabilities (Zahra and George, 2002).
At last, the results do not show strong support for the hypothesis of positive moderation of international experience with host country KBR. Increased international experience gives the EMNE an understanding of the potential of a host country’s KBR, and facilitates the capabilities to acquire advanced target firms (Deng, 2010). The results might indicate that for services firms, EMNEs already have developed substantial international experience, so that it does not reinforce the effect of host-country industry specific KBR. As noted in literature, Indian EMNEs have geared their FDI towards developed countries to tap into strategic resources since the turn of the century (Kumar 2007; Pradhan, 2005). Also, literature states that especially Indian IT firms are expanding abroad aggressively and repeatedly towards the United States to develop onsite client facilities (Pradhan, 2008). Their locations might reflect a motive to locate close to their customers rather than seeking intangible knowledge.
face-30 to-face interaction, since it is difficult to codify and transfer across borders (Berry, 2014). The results suggest that when countries are geographically too far, these constraints overweigh the benefits in EMNEs’ likelihood to conduct knowledge-seeking FDI.
5.2. Limitations and recommendations
As with every research, the study has some limitations. With regards to data, as all information was publicly available and included secondary data, therefore no ethical issues were violated. However, the sample has some limitations with regards to generalization for EMNEs, as only Indian firms were analysed. With regards to methodology, the study also has some limitations. First, as the data was pooled over five years, most locational factors included average values. On the one hand, in calculating the average values, effects from extreme between-year differences are minimized and controlled for, and more robust variables are obtained, following extant literature (Drogendijk and Blomkvist, 2013). However, as the CBAs were spread over these five years, the individual CBAs were not exactly matched with the host country variable value of that particular year. For the firm variables values, there were also some limitations. For the ‘no-CBAs’ or zeros in the observations, international experience and technological intensity reflected the 2008 values, whilst for the actual CBAs, it reflected the value of that particular year. It is therefore suggested that this study could be repeated by using panel data, to reflect the true values of these observations.
Further, there might be many other reasons aside from knowledge-seeking FDI to expand abroad; even though many issues are controlled for, it cannot be excluded that the observed patterns reflect other factors. In fact, research shows that EMNEs expand abroad for a variety of motives simultaneously (Jain, Jain and Newbury, 2015). Also, EMNEs could opt with various other types of entry modes besides CBAs when expanding abroad. However, knowledge-seeking FDI as proposed by Makino and Inkpen (2003) is particularly done through CBAs, and as Chung and Alcácer (2002) state, in the context of knowledge-seeking FDI, a desired location most likely offers opportunities for greenfield investments as acquisition targets alike.
31 CBAs. It seems unlikely that corporate executives stand before a binary decision of conducting a CBA or not, as the costs of deal (which depend on a variety of factors) would most likely also play a role. It is therefore suggested that future research advances this area by examining the extent of knowledge-seeking FDI, in terms of the frequency and monetary value.
5.3. Theoretical contribution
Theoretically speaking, the findings support several streams of literature. First, it provides empirical support for the theoretical framework of Kedia, Clampit and Gaffney (2012), which propose in knowledge-seeking FDI, the type of knowledge sought influences the location choice and entry mode. Second, it answers calls from Berry (2006) that challenges extant knowledge-seeking FDI research for comparing R&D intensities solely on the national level; thereby making “problematic assumption that all firms within a country are of the same technological type” (Berry, 2006, p. 152). By analysing R&D intensities on the industry as the firm level, this paper complements current empirical country-level knowledge-seeking FDI literature (e.g. Chung and Alcácer, 2002; Jindra, Hassan and Cantner, 2016; Yoo and Reimann, 2017). Specifically, it highlights that knowledge-seeking FDI is influenced by an interplay of firm, industry and host country attributes. Also, by comparing the manufacturing sector with the services sector, it gives insight in the different roles and implications these factors have among firms from different industries. Third, this study addresses suggestions laid out by Kedia, Clampit, and Gaffney (2012), that advocate for a broader adoption of the concept and measure of knowledge as opposed to solely considering technologies (i.e. patents).
32 do have initial FSAs which influence its foreign investments (Bhaumik, Driffield and Zhou, 2016; Cuervo-Cazurra, 2012; Dunning and Lundan, 2008; Ramamurti, 2012)
Fifth, the findings correspond with that of organizational learning perspective (March, 1991). The results indicate that international experience positively affects the likelihood of EMNEs pursuing knowledge-seeking FDI. Therefore, it finds correspondence with Makino and Inkpen’s (2003) view that firms invest abroad to acquire new knowledge, but simultaneously further exploit existing capabilities could facilitate cross-border learning. The acquisitive learning process could thus facilitate experimental learning (Zahra, Nielsen, & Bogner, 1999). Technological capability on its own does not significantly influence EMNEs likelihood to pursue knowledge-seeking FDI, which might indicate EMNEs are balancing exploiting existing capabilities with exploring future technologies and capabilities (Luo and Rui, 2009; March, 1991).
Sixth, the study offers support for the absorptive capacity literature (Cohen and Levinthal, 1990; Zahra and George, 2002). The findings show that even though host country’s KBR does influence EMNEs’ likelihood to pursue knowledge-seeking FDI, they are more likely to do so when the firm has technological capabilities. Hence, it provides support for the view that although firms are likely to invest abroad due to certain locational factors, they are more likely to do so when they have ample dynamic capabilities (Zahra and George, 2002). Also, technological capability alone does not increase EMNEs likelihood to conduct knowledge-seeking FDI, but when combined with a host country’s industry-specific KBR, it does. This is in line with the idea that the decision of acquiring intangible resources abroad is partly dependent on the firms’ technological capabilities; i.e. absorptive capacity, of the ability to identify and understand the knowledge (Cohen and Levinthal, 1990).
5.4. Practical implications
34
REFERENCES
Assalahi, H. (2015) ‘The philosophical foundations of educational research: A beginner’s guide’,
American Journal of Educational Research, 3(3), pp. 312–317.
Almeida, P. and Kogut, B. (1999) ‘Localization of knowledge and the mobility of engineers in regional networks’, Management Science, 45(7), pp. 905–917.
Amighini, A., Cozza, C., Giuliani, E., Rabellotti, R. and Scalera, V. G. (2015) ‘Multinational enterprises from emerging economies: What theories suggest, what evidence shows’, Economia
e Politica Industriale, 42(3), pp. 343–370.
Athreye, S. and Godley, A. (2009) ‘Internationalization and technological leapfrogging in the pharmaceutical industry’, Industrial and Corporate Change, 18(2), pp. 295–323.
Aulakh, P. S. (2007) ‘Emerging multinationals from developing economies: Motivations, paths and performance’, Journal of International Management, 13(3), pp. 235–240.
Awate, S., Larsen, M. M. and Mudambi, R. (2014) ‘Accessing vs sourcing knowledge: A comparative study of R&D internationalization between emerging and advanced economy firms’, Journal of International Business Studies, 46(1), pp. 63–86.
Barney, J. B. (1991) ‘Firm resources and sustained competitive advantage’, Journal of
Management, 17(1), pp. 99–120.
Berry, H. (2006) ‘Leaders, laggards, and the pursuit of foreign knowledge’, Strategic
Management Journal, 27(2), pp. 151–168.
Berry, H. (2014) ‘Global integration and innovation: multicountry knowledge generation within MNCs’, Strategic Management Journal, 35, pp. 869–890.
Bhaumik, S. K., Driffield, N. and Pal, S. (2010) ‘Does ownership structure of emerging market firms affect their outward FDI? The case of the Indian automotive and pharmaceutical sectors’,
Journal of International Business Studies, 41(3), pp. 437–450.
Bhaumik, S. K., Drieffield, N. and Zhou, Y. (2016) ‘Country specific advantage, firms specific advantage and multinationality - Sources of competitive advantage in emerging markets: Evidence from the electronics industry in China’, International Business Review, 25(1), pp. 165– 176.
Blomström, M. and Sjöholm, F. (1999) ‘Technology transfer and spillovers: Does local participation with multinationals matter?’, European Economic Review, 43(1), pp. 915–923. Bonaglia, F., Goldstein, A. and Mathews, J. A. (2007) ‘Accelerated internationalization by emerging markets’ multinationals: the case of the white goods sector’, Journal of World
Business, 42(4), pp. 369–383.
Bresman, H., Birkinshaw, J. and Nobel, R. (2010) ‘Knowledge transfer in international acquisitions’, Journal of International Business Studies, 41(1), pp. 5–20.
Buckley, P. J. and Casson, M. (1976) The future of the multinational enterprise, London: Macmillan.
35 Buckley, P.J., Clegg, J. and Wang, C. (2007) ‘Is the relationship between inward FDI and spillover effects linear? An empirical examination of the case of China’, Journal of International
Business Studies, 38(3), pp. 447–459.
Buckley, P. J., Munjal, S., Enderwick, P. and Forsans, C. (2016) ‘Cross-border acquisitions by Indian multinationals: Asset exploitation or asset augmentation?’, International Business
Review, 25, pp. 986–996.
Cantwell, J. and Janne, O. (1999) ‘Technological globalization and innovative centres: The role of corporate technological leadership and locational hierarchy’, Research Policy, 28(2), pp. 119– 144.
Cantwell, J. A. and Mudambi, R. (2005) ‘MNE competence-creating subsidiary mandates’,
Strategic Management Journal, 26(12), pp. 1109–1128.
Chen, V. Z., Li, J. and Shapiro, D. M. (2012) ‘International reverse spillover effects on parent firms: Evidences from emerging-market MNEs in developed markets’, European Management
Journal, 30(3), pp. 204–218.
Child, J. and Rodrigues, S. B. (2005) ‘The internationalization of Chinese firms: A case for theoretical extension?’, Management and Organization Review, 1(3), pp. 381–410.
Chittoor, R. and Ray, S. (2007) ‘Internationalization path of Indian pharmaceutical firms – A strategic group analysis’, Journal of Management, 13, pp. 338–355.
Chittoor, R., Sarkar, M. B., Ray, S. and Aulakh, P. S. (2009) ‘Third-world copycats to emerging multinationals: Institutional changes and organizational transformation in the Indian pharmaceutical industry’, Organization Science, 20(1), pp. 187–205.
Chung, W. and Alcacer, J. (2002) ‘Knowledge seeking and location choices of foreign direct investment in the United States’, Management Science, 48(12), pp. 1534–1554.
Cohen, W. M. and Levinthal, D. A. (1990) ‘Absorptive Capacity: A New Perspective on Learning and Innovation’, Administrative Science Quarterly, 35(1), pp. 128–152.
Collis J. and Hussey R. (2013) Business research. A practical guide for undergraduate and
postgraduate students, 4th edn, New York: Palgrave Macmillan.
Crescenzi, R., Pietrobelli, C. and Rabellotti, R. (2016) ‘Regional strategic assets and the location strategies of emerging countries’ multinationals in Europe’, European Planning Studies, 24(4), pp. 645–667.
Cuervo-Cazura, A. (2012) ‘Extending theory by analyzing developing country multinational companies: solving the Goldilocks debate’, Global Strategy Journal, 2(3), pp. 153–167.
Cuervo-Cazurra, A. and Genc, M. (2008) ‘Transforming disadvantages into advantages: developing-country MNEs in the least developed countries’, Journal of International Business
Studies, 39(6), pp. 957–979.
Cui, L., Meyer, K. E. and Hu, H. W. (2014) ‘What drives firms’ intent to seek strategic assets by foreign direct investment? A study of emerging economy firms’, Journal of World Business, 49(4), pp. 488–501.
Delios, A. and Henisz, W. J., (2000) ‘Japanese firms’ investment strategies in emerging economies’, Academy of Management Journal, 43(3), pp. 305–323.
