The internationalization of emerging market multinationals : a subsidiary-level investigation of foreign equity commitment across industries

(1)

The internationalization of emerging market multinationals:

A subsidiary-level investigation of foreign equity commitment

across industries.

Faculty of Economics and Business

Master Thesis – Final version

MSc. Business Administration: International Management

Karim Breeveld

10105476

Date: 24

th

of June 2016.

Supervisor: Dr. Niccolò Pisani

Second reader: Dr. Ilir Haxhi

(2)

STATEMENT OF ORIGINALITY

This document is written by Karim Breeveld who declares to take full responsibility for the contents of this document.

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

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

(3)

Abstract

In recent years emerging market multinational enterprises (EM MNEs) have become increasingly internationalized as they expand abroad in search of new markets (UNCTAD, 2014). The purpose of this thesis is to gain a better insight into these EM MNEs’ equity commitment on an industry level when expanding abroad. In this research two classic industry classifications have been used, being primary versus secondary and product versus service industries. Subsequently, a sample of 500 of the largest BRIC MNEs has been used in order to test various hypotheses. The main findings are that primary sector EM MNEs are more likely to establish a wholly owned subsidiary than secondary sector EM MNEs when expanding abroad. Also, contrary to the relation hypothesized, service EM MNEs are more likely to establish a foreign WOS than their producing counterparts. Lastly, both these proven relations are positively moderated by the degree of internationalization, measured by dividing a firm’s amount of foreign sales by its total sales. These results suggest that specific industry level characteristics such as knowledge-intensity, institutional influences and nature of the product or service play a vital role in an EM MNE’s entry mode choice and degree of foreign equity commitment.

Keywords: emerging market multinationals, ownership structure, foreign equity

commitment, degree of internationalization, product versus service, primary versus product.

(5)

1. Introduction

In 2014, the worldwide foreign direct investment (FDI) inflows reached $1.2 trillion (UNCTAD, 2015), with the global economic becoming increasingly interconnected and the business environment becoming increasingly international. Currently, 40 % of these global FDI inflows are already directed towards emerging markets, up from 12% in 2000, and this growth trend is expected to continue into the coming years (UNCTAD, 2015). Both FDI inward and outward flows pertaining to developing countries reached record levels in 2014, with half of the top 20 FDI-receiving countries considered as developing (UNCTAD, 2014).

When researching emerging markets the BRIC countries make for an especially interesting case. The term BRIC refers to Brazil, Russia, India and China and was first coined by Goldman Sachs Head Economist Jim O’Neill in 2001 to highlight a group of large emerging markets that had the potential of becoming the economic powerhouses of the future (O’Neill, 2001). Nowadays, the BRIC countries live up to their name as FDI outflows from these countries have significantly increased since the turn of the century and the group plays a vital role when assessing worldwide FDI flows (Holtbrugge & Kreppel, 2012). Throughout the years the BRIC countries have become popular destinations for attracting FDI, but multinational enterprises (MNEs) from these countries have also increasingly been investing abroad themselves with the rates of outward FDI by developing countries growing at a faster pace than outward FDI by developed countries over the past years (UNCTAD, 2014).

(6)

Due to the relatively recent emergence of outward BRIC FDI, up until now most literature on the BRICs has been with a focus on these countries’ inward FDI (O’Neill, 2001; Holtbrugge & Kreppel, 2012; Kaplinsky & Morris, 2009 and Gammeltoft, 2008). This research will provide a subsidiary-level investigation of the outward foreign equity commitment of emerging market multinational enterprises (EM MNEs) by focusing on these firms’ degrees of foreign equity commitment and entry mode choices. Previous studies have brought forth three main streams focused on explaining the concept of entry mode choice. The first group focuses on the trade-off between level of ownership and control (Hennart & Reddy, 1997). Second, Cho & Radmanabhan (1995) hypothesize that entry mode choice is mainly made according to host country risk and economic development. Third, Gatignon and Anderson (1988) explain entry mode choice according to the cultural distance between home and host country. This thesis will apply concepts from all three of these streams in order to assess EM MNE entry mode choice and foreign equity commitment. More specifically, the focus of this thesis is on how the foreign equity commitment of EM MNEs relates to the industry in which these firms operate and how certain industry-level attributes determine foreign expansion patterns. Furthermore, because EM MNEs are becoming increasingly internationalized (Lu & Beamish, 2004; Mahadeo et al., 2012) the degree of internationalization will be considered as a moderator as this has an effect on the way in which the firms structure their foreign entry mode choice and degree of equity commitment (Johanson & Vahlne, 1977). The key contribution of this work is thus to offer a better understanding of which factors drive outward EM MNE FDI and the how industry-specific characteristics affect entry mode choice and degree of equity commitment in foreign subsidiaries.

(7)

In the following section the relevant literature will be reviewed to shed light on concepts such as BRIC countries characteristics, FDI determinants and the different forms of equity commitment. Subsequently, in the theoretical framework the hypotheses are presented, followed by the methodology section, which focuses on the data collection methods. Lastly, the results from the statistical analysis are presented and discussed and a conclusion is formed with final statements.

2. Literature review

In this section of the thesis a review is presented containing relevant literature that has been collected up to present date on the topic. The literature review is divided into separate short paragraphs which each shed light on a particular concept related to the internationalization patterns of emerging market firms.

2.1 Emerging market firms

Emerging market multinational enterprises are defined as “international companies that originated from emerging markets and are engaged in FDI, where they exercise effective control and undertake value-adding activities in one or more foreign countries” (Luo & Tung, 2007). Even though there is much debate over when a market can be considered emerging, the general agreement is that every economy that has experienced drastic structural changes in recent years and continues to rapidly grow and develop can be categorized as such an emerging market (Luo & Tung, 2007). As the BRICs are considered to be leading emerging markets both in size and growth perspective this group will be used as a general representation of emerging markets. Emerging markets exist in many different configurations with diverse population sizes and other macro-economic characteristics and should therefore be

(8)

investigated which much care and consideration. Compared to their developed counterparts EM MNEs do generally display some key differences. First, compared to MNEs from developed countries EM MNEs are often more closely tied to the home country’s government (Gammeltoft, 2010). This can be reflected either in the firm adhering to wishes or objectives set by the government or in the more extreme form of the firm even being a fully state-owned enterprise. Close ties with government can provide these EM MNEs with preferential treatment when searching valuable resources, inputs and support. Governments are in this way also given the opportunity to for example focus on internationalization and increasing the country’s international competitiveness. Secondly, the institutional environment in emerging markets plays an important role in the shaping of its MNEs structures. Because emerging markets have generally less developed institutions EM MNEs are often more integrated into business groups (Gammeltoft, 2010; Khanna & Yafeh, 2007) that support each other. This is due to these firms having to cope with less readily available managerial skills, technological and economic resources. These structures are focused on achieving efficiency. Third, due to swift growth in recent decades and much emerging markets only opening up recently EM MNEs have less experience with crossing cultural borders and dealing with international partners. This suggests that EM MNEs often foster more closed networks with more distinct ways of governing compared to MNEs from developed regions (Dunning & Narula, 2004). Lastly, it is a known theory that EM MNEs often specifically focus on expanding into more mature industries (Dunning et al., 1998). Dunning reasons that this focus is due to EM MNEs looking to acquire cost-competitive products and themselves becoming increasingly competitive contenders in mature global industries (Gammeltoft, 2010).

(9)

Dunning’s OLI model suggests that MNEs in general expand outside of their home country to exploit firm specific advantages (FSAs) which are unique to the company and therefore provide a competitive position even outside of home borders. Rugman (2009) however states that this usage of FSAs does not apply to EM MNEs as they have not developed sustainable FSAs (such as managerial knowledge) and are therefore more dependent on country-specific advantages (CSAs) such as natural endowments and labor costs.

2.2 The BRICs

Simply looking at the sheer demographic and geographical numbers pertaining to the BRICs provides a clearer image of approximately how big of a role these countries play in the global economic landscape. Brazil, Russia, India and China’s populations account for 42 % of the world’s population and the countries’ combined landmasses cover up almost one third of the world’s surface (Kumar, 2013). Often the BRICs are seen as the frontrunners and the most promising of the world’s emerging markets. China’s outward FDI flows by itself for example accounted for almost one third of all global outward FDI flows (UNCTAD, 2010).Therefore these countries have received much attention over the years and are seen as the next big economies of the future (O’Neill, 2001). Growth rates further strengthen this statement as from 2006 to 2008 the BRIC countries experienced an average annual growth of 10.7 %, which is considerably larger than the growth rates of developed countries (Xing & Xin, 2009).

From the year 2000 to 2014 the FDI outflows from developing countries as a share of the world FDI outflows have grown from 10% to almost 35% (UNCTAD, 2015). Thus, for the BRICs specifically FDI outflows have also greatly increased (Holtbrugge & Kreppel, 2012).

(10)

As mentioned before, a large part of the rapid economic growth of the BRIC countries has been accounted for by inward FDI. China was the largest recipient of FDI amounting to $147.79 billion in 2008, followed by Russia, Brazil and India with inward investments of respectively $75, $45 and $41 billion (Ranjan & Agrawal, 2011). Some of the most prominent attributes of the BRIC countries attracting these foreign investments are market size by GDP, labour cost, economic stability, trade openness and growth prospects (Ranjan & Agrawal, 2011). Jadhav (2012) also concluded that the market size by GDP has a strong effect on the amount of inward FDI. This further consolidates that in many cases FDI into BRICs is driven by a market-seeking motive. In the case of outward FDI the BRIC countries have also shown tremendous growth. This is in large part due to their relatively recent advancements in trade regulation and economic liberalization (Hoskisson, 2000). China is especially promising due to its large manufacturing power and population, while India is known as having huge potential in the IT-services sector. Pertaining to FDI outflows, it is predicted that China and India shall thrive most on supplying manufactured goods, while Brazil and Russia’s FDI outflows will be mostly accounted for by the raw material industry (Ranjan & Agrawal, 2011).

In 2009 the BRIC group became more than simply an informal abbreviation for some of the world’s fastest-growing economies and was deemed an official initiative for economic and political cooperation. Yearly occurring BRIC summits have been established as a way of forming a collective voice for non-Western emerging markets and to thereby raise their bargaining power, influence and economic competitiveness in the global community (Stuenkel, 2014). This further consolidates the countries’ governmental commitments to increase cooperation and

(11)

further increase the participation of developing countries in stimulating technological advancement, economic stability and governance structures across all markets.

Apart from appraise the BRICs have also received some criticism over the formation of the group as an institution. One of the main arguments has been the obvious differences between the countries, which supposedly outweigh the shared characteristics (Hurrell, 2006). China and India for example have considerably larger populations than Russia and Brazil. From a macro-economic perspective there are also large differences between for example the countries’ GDP (per capita), as seen below in table 1.

Table 1. Macro-economic data of the BRIC countries. Source: Kumar, 2013.

Brazil Russia India China Population (2009) 194 million 142 million 1,15 billion 1,33 billion

GDP (billion US$, 2009) 1,573 1,232 1,310 4,985

GDP per Capita (PPP, Intl. $, 2009)

$10,499 $14,913 $3,015 $6,778

Furthermore, the fact that the countries each have their own characteristic economic strengths and weaknesses also complicates the grouping of all four nations into one term. For example, Brazil and Russia as large energy exporters have their own contradictory preferences when it comes to energy prices compared to India, which is a large energy importer (Sharma, 2012). In political terms, China and Russia can already be considered political powerhouses in the global community, while Brazil and India are lagging behind when it comes to international bargaining power. The

(12)

“authoritarian governments” (Stuenkel, 2014) contrast the more freely governed nations of Brazil and India. To further highlight the characteristic attributes of each one of the BRIC constituents, the following paragraphs provide a short overview of each one of the member countries.

To gain a better understanding of Brazilian MNEs and their motives to invest in the form of outward FDI, three common Brazilian types of multinationals have been categorized by Cuervo-Cazurra (2007) according to the type of subsidiary they initally invest abroad with (Cuervo-Cazurra, 2007; Bertoni, Elia & Rabbiosi, 2008). Firms from the first type have strong locational advantages in the home country and invest abroad for sales and distribution advantages. Firms from the second type invest abroad in hopes of being able to take advantage of locational advantages in the host country. Especially resource-seeking is an important motive for these types of Brazilian companies to expand to developing countries, such as for example other BRICs. Lastly, firms from the third type are considered a hybrid of the first two types as these companies invest abroad seeking for resources as well as being able to establish new sales and distribution facilities (Cuervo-Cazurra, 2007). Brazil is the BRIC country with the most globally spread FDI patterns, although the South American giant also remains the least open economy with regards to inward trade liberalization among all the BRICs (Freemantle & Stevens, 2010).

For Russian MNEs FDI is generally more homogeneous compared to the internationalization patterns of firms from the other BRICs as the Russian economy is highly dependent on commodities and natural resources such as oil. Therefore Russian MNEs often cross borders to seek new resources to support existing ventures and most Russian acquisitions abroad since the 1990’s have been in the gas, petroleum and mining industries (Kalotay, 2008). Especially in developing countries

(13)

Russian MNEs invest heavily in upstream coordinating natural resources (Kalotay, 2010). This dominance of the oil companies is also clearly seen when ranking the largest 25 Russian outward investors. Out of the 25 MNEs only four were active in non-natural resource industries. The three largest outward investors, all three oil and gas companies, comprised 44 % of the outward FDI flows in 2007 (Kalotay, 2010). An unstable home environment with regards to the economy also pushes Russian MNEs to expand and seek opportunities abroad. Also, the FDI dominance in general of state-owned MNEs is becoming less as private firms have shown a steady increase in outward investments, even though this is not the case for state-owned Russian MNEs (Kalotay, 2008). Russia was the last BRIC country to become a member of the World Trade Organization, with its accession in 2012. Even though this has led to trade liberalization for Russian MNEs it had little overall effect on FDI numbers as the largest part of Russian FDI flows are accounted for in the export of extractives, which previously already enjoyed less strict trade barriers (Sally & Sen, 2011). Taking the large size of the Russian economy into account the country has a relatively low level of foreign shares the number of Russian and foreign jointly controlled ventures only accounted for 7,2% of all Russian companies in 2009 and more than 90% of capital investments are under Russian control (Kuznetsov, 2010).

In the case of the Indian economy, traditionally closed policy was followed in the past with regards to foreign direct investment, although this has changed since economic reforms. These economic reforms occurred over three different waves of economic liberalization in 1991, 1995 and 2000 which resulted in trade barriers being lifted, and the Indian economy truly opening up to foreign investments (Zheng, 2013). From 2000 to 2010 Indian inward FDI flows increased by tenfold to a total of $198 billion (UNCTAD, 2011). Still, the Indian business landscape can be characterized as

(14)

rather homogenous with regards to ownership nationality as it has a relatively low percentage of foreign invested capital as a percentage of GDP, even compared to other developing markets (Zheng, 2013). The internationalization patterns of Indian MNEs indicate that the focus is mostly on long-term rather than short-term profitability, with the companies investing heavily in the extractive sectors to fuel strong economic development (Khan, 2012).

Lastly, the Chinese economy is a prime example of an emerging market economy as it has undergone extensive structural changes in economic terms over the past decades. In the 1970’s, Chinese leaders decided to shift from a traditionally closed communist economic policy to a more open policy in which foreign investment was welcomed and created seven Special Economic Zones (SEZs) designed to stimulate regional economic growth throughout the country with the help of foreign trade. These changes brought forth economic development as the value of imports and exports increased from 15 % of national income in the 1970’s to 35 % in 1986 (Whalley & Zhao, 2013). Even though acquisitions by Chinese MNEs mostly take place in developed economies with the goal of entering these markets and acquiring strategic assets (Rui & Yip, 2008), there is a growing trend of Chinese outward FDI into other developing countries as they are in these cases able to operate in similar institutional environments (Morck et al., 2008) and experience advantages in cost-efficiency terms (Morck et al., 2008; Rui & Yip, 2008; Bertoni, Elia & Rabbiosi, 2008). Nowadays, China is the largest country exporter in the world accounting for almost one third of all global outward FDI flows (UNCTAD, 2010).

(15)

2.3 Emerging market FDI determinants

To gain a better insight into how EM MNEs’ foreign equity commitment structures are formed this section will take a closer look at which FDI determinants drive outward FDI. Much research has been done concerning international transactions, or more specifically trade flows between different countries. One of the first frameworks to attempt to explain this phenomenon was Vernon’s (1966) product life cycle framework, which states that technologically advanced products are first made available in the most developed countries, after which the production is gradually relocated to less developed countries and regions.

Buckley and Casson (2009) extended this work and presented the internalization theory. According to internalization theory product markets are imperfect, leading to high transaction costs. Therefore firms are able to lower transaction costs by internalizing markets and basically “do it themselves” instead of having third parties control transactions. MNEs looking to internalise intermediate product markets therefore often leads to foreign direct investment (Buckley & Casson, 2009). This can be either in the form of (tacit) knowledge internalization or operational internalization, which is more aimed towards streamlining processes in the production and distribution.

Dunning’s (1998, 2000) eclectic paradigm (OLI-paradigm) is in a certain way related to Buckley and Casson’s (2009) internalization theory. Dunning (2000) provides a framework in which three variables are presented as determinants of FDI. These three variables are:

1) Ownership specific advantages: MNEs possessing superior resources such as knowledge, monopoly power or endowment of scarce resources can attain competitive advantages over other firms in a host country.

(16)

2) Locational attractions: If a certain country possesses for example immobile natural resources and economic and social infrastructure these are location-bound advantages that attract FDI.

3) Internalization: If above two variables are present in a country or firm the act of internalizing these advantages can lower transaction costs and make FDI even more viable.

In 1980 Dunning also provided a foundation for grouping MNE motivations for FDI into 4 different categories related to what kind of asset they seek across borders. Market-seeking MNEs cross borders in order to gain access to new large markets, consumers, to achieve economies of scale or for example to enter more stable markets. Efficiency-seeking FDI is driven by an MNE’s desire to find markets that provide more efficient circumstances, usually in financial terms. This can be either through lower costs of labor or lower costs of attaining tangible resources. Strategic-asset seeking firms invest abroad in order to gain access to usually intangible Strategic-assets such as technology and knowledge. Lastly, resource-seeking MNEs engage in FDI in order to gain access to for example natural resources or raw materials that are not available in the home country (Dunning, 1980).

Apart from a firm’s internal characteristics institutions should also be taken into account when looking to invest abroad. North (1991) defines institutions as “the rules of the game in a society or more formally, they are the humanly devised constraints that shape human interaction”. These can be either informal institutions such as culture, norms and religion or more formal institutions such as law, policy or trade regulations. For this reason formal institutions can change relatively fast, while informal institutions are shaped over hundreds of years (Williamson, 2000). Yet, both have a clear influence on economic activity in a country and therefore also on FDI

(17)

activity of these countries’ MNEs. Whitley (1992) managed to apply North’s (1991) institutional view specifically to the business context by stating that the institutional characteristics of a country shape its business environment, leading to a particular

national business system that is formed around “demographic, geographic, cultural

and political institutions” of said country (Whitley, 1992).

Economic performance is directly influenced by institutions (Cantwell, Dunning and Lundan, 2010). These institutional factors are especially important to consider when MNEs attempt to expand into emerging markets (Luo, Xue & Han, 2010). In order to be succesful in such markets that are characterized by complexity and instability, MNEs should therefore embrace changes in strategy and organizational structures (Cantwell, Dunning & Lundan, 2010). As governments are those that construct the formal institutions, such as law and regulation, they also play a vital role in either stimulating or slowing down FDI flows. This government backing is especially important for EM MNEs, as they have to overcome late-mover disadvantages and lack of certain capabilities (Luo, Xue & Han, 2010). According to Luo and Tung (2007), EM MNEs’ late-mover disadvantages have led to a springboard perspective when expanding abroad. The springboard perspective (Luo & Tung, 2007) explains internationalization by EM MNEs as a consequence of these companies seeking specific strategic assets outside of the home country and also crossing borders in order to avoid home country institutions that are not favorable for business. The way they acquire these new strategic resources is by sometimes aggressively buying stakes in competitive companies, often in developed countries. EM MNEs are considered to have a late-comer disadvantage due to lagging economic development in the past and in this way attempt to compensate for this disadvantage. This means that these firms are often less constrained to the classic

(18)

internationalization theories, such as for example Johanson and Vahlne’s (1977) Uppsala model. The Uppsala model states that companies expand internationally in incremental steps that range from starting more close to the home country to expanding distantly further. This distance can be both in geographical or cultural distance and by doing so step-by-step the MNE is able to acquire skills in overcoming specific country specific advantages (CSAs) during each expansion. Once the MNE becomes more experienced in overcoming CSAs it will therefore continue expanding more distant. Concluding, knowing that the institutional environment of a country plays a large role in its FDI behaviour, it should also be taken into account that this causes different strategies, and motives for expansion in the four BRIC countries.

2.4 Entry mode choice and foreign equity commitment

As the aim of this research is to gain a better insight into the relation between EM MNEs’ foreign equity commitment and the industry in which these firms operate this section will focus specifically on EM MNE entry mode choice. The entry mode choice determines to what degree a firm is committed to the venture and can be defined as the way in which a firm decides to enter a new foreign market. The main categorization of entry modes is that of equity-based versus non-equity-based (Pan & Tse, 2000). Equity-based modes are entry modes that involve acquiring a stake in a venture, for example a wholly owned operation or equity joint venture. Because of this commitment to set up a new entity this entry mode also requires constant mangement, interaction and coordination with local firms in order to work (Hennart, 1988). Non-equity modes on the other hand require a lower degree of commitment as these do not involve acquiring a stake in a new venture but are rather based on agreements between two parties that remain separate. Examples of non-equity modes

(19)

of entry are contractual agreements and export (Pan & Tse, 2000). This categorization by Pan and Tse is defined as the hierarchical model of market entry modes, meaning that MNEs first must make the choice between equity-based versus non-equity based modes based on critical factors such as host country institutions, industry and entry motives. In the second stage of the hierarchical model the MNE chooses in which equity mode precisely to engage. This stage is once again influenced by distinctive factors. The main finding of Pan and Tse was that the country risk of the host country was the most important determinant in the decision between equity-based versus non-equity based modes but played an insignificant role in the second-tier choice between wholly owned subsidiary (WOS) or equity joint venture (EJV). Rather, at the second tier decisions are based on micro-level factors specific to the firm itself (Vanhonacker, 1997). As this research is to gain insight into foreign equity commitment of EM MNEs the analysis will be specifically focused on the two equity-based entry mode forms of equity joint ventures and wholly owned subsidiaries. Harrigan (1988) described the structure of an equity joint venture as “a business agreement whereby two or more owners create a separate entity”. A majority EJV is any joint venture in which one party owns a controlling stake of more than 50%. Wholly owned subsidiaries are a parent firm’s sudsidiaries in which it owns 100% of equity (Pan & Tse, 2000) and therefore has full independent control over its operations. These wholly owned subsidiaries have either been established as a greenfield investment and built from the ground up or through acquisition from another company (Dikova & Witteloostuijn, 2007). In the case of acquisitions by emerging market firms in more developed countries, prior research has shown that these ventures have a positive effect on the market valuation of the parent firms (Gubbi et. al, 2010). In this way EM MNEs are able to acquire intangible strategic

(20)

assets such as problem-solving skills and managerial expertise that normally require much time to accumulate. Transaction cost theory hypothesizes that firms constantly consider the trade-off between control and the transaction costs associated with resource commitment (Anderson & Gatignon, 1986; Erramilli & Rao, 1993) and has shown that as transaction costs for a venture increase, the likelihood of choosing a WOS over an EJV as entry mode also increases (Brouthers, 2002).

From an institutional perspective other factors play a significant role in the choice between entry modes. When the institutional environment points towards a stable and well developed market MNEs are more likely to choose a greenfield investment over an acquisition (Dikova & Witteloostuijn, 2007). When entering countries with highly regulative environments subject to many legal restrictions, joint ventures are preferred in order to make use of local partners’ experience. This also is the case when MNEs perceive high investment risk as a joint venture offers a structure to spread the risk between the participating firms.

As cultural distance and state influence from the host country increase, an MNE tends to prefer an EJV to WOS (Makino & Yiu, 2002). For this reason entry mode choice into emerging markets can be greatly influenced by institutions as these countries are often characterized by high levels of state intervention (Chao, Lo & Yu, 2010). Inefficient markets in emerging economies also affect the entry mode choice as these lead to transaction costs, which differ between joint ventures and wholly owned subsidiaries. Furthermore, as the level of R&D intensity increases, which is characteristic for certain industries, the likelihood to choose a WOS over an EJV increases (Makino & Yiu, 2002). This is due to the positive relation between asset specificity and R&D intensity. Asset specificity refers to the degree to which a certain asset can be used for only one purpose (Williamson, 1991). As partners in a joint

(21)

venture require much coordination to optimally utilize these specific assets transaction cost theory states that this also increases the transaction costs of the venture. Therefore R&D intensive MNEs shall opt for the more efficient entry mode and are more likely to engage in a wholly owned subsidiary. Another reason that firms with many firm-specific assets prefer a WOS entry mode is in order to have more control over the subsidiary and protect themselves against theft of strategic assets such as technology (Anderson & Gatignon, 1986). Therefore the level of ownership is positively linked to the level of firm-specific assets (Chiao, Lo & Yu, 2010).

As companies differ in the strategy with which they approach operations this also has an influence on how subsidiaries are managed and thus also on the level of ownership and equity commitment (Birkinshaw & Morrison, 1995). Apart from the strategy of the parent firm subsidiary structure is also built upon the external factors present in the host country. In some cases this is causing subsidiaries to become increasingly autonomous and take own initiative (Birkinshaw, Hood & Jonsson, 1998). In this way subsidiaries transform from simply units of the firm that contribute to developing FSAs to more autonomous parties that actually drive the development of new FSAs. Therefore a firm’s equity commitment in a foreign subsidiary is especially important to determine the amount of control the firm will be able to exercise throughout the operations of the new venture. In majority owned subsidiaries the firm with a stake of at least 50% is able to exercise greater control over the different operations. In a wholly owned subsidiary the parent firm has full control (Osland, Taylor & Zou, 2001) and also is given the opportunity for the highest return on investment but is required to make the highest possible degree of commitment and thereby also bears the highest level of risk (Ekeledo & Sivakumar, 2004). Therefore,

(22)

the firm needs to consider what the optimal degree of equity commitment should be for each foreign subsidiary so it best fits institutional and industry-level factors.

2.5 Research gap

As we have now explored relevant literature on EM MNEs, how they differ from developed country firms and their motives to invest abroad, a clearer research gap can be defined. Literature has shown us that EM MNEs are nowadays not only recipients of large amounts of inward FDI but are also becoming large outward investors themselves by expanding abroad. When engaging in these foreign ventures EM MNEs have a broad choice of entry modes ranging from non-equity to equity-based modes. In the case of equity-based entry ventures can be either wholly owned or jointly owned with a certain stake. This translates into the foreign equity commitment that the parent firm has in the venture and is influenced by industry-level factors and institutional environment among others. Motive wise, EM MNEs often invest abroad to gain valuable strategic assets, seek new markets and overcome home country institutional difficulties. Yet, on an industry-level little research has been conducted specifically for EM MNEs and how the industry in which they operate affects the level of equity commitment abroad. The following chapters will therefore further explore this research gap.

3. Theoretical Framework

As mentioned before, this research has a specific industry-level focus on EM MNEs as they invest abroad through equity-based modes. The previous chapter presented relevant literature regarding these topics. A firm’s industry classification plays a key role in determining entry mode choice and equity commitment (Erramilli & Rao,

(23)

1993) as characteristics such as possible perishabilitity, intangibility and production methods can greatly differ between industries (Ekeledo & Sivakumar, 2004). In this chapter several hypotheses will be developed in order to test the relationship between EM MNE industry classification and foreign equity commitment. The industry classification has been divided into two separate groups; product versus service and primary versus secondary industries. Furthermore, it is proposed and tested that an EM MNE’s degree of internationalization positively moderates the abovementioned relationships.

3.1 Product and service industries

When categorizing on an industry-level perhaps the most basic distinction is the one made between product and service industries. MNEs that operate in a product industry offer their clients a tangible product. This can be anything ranging from natural resources to for example more complex products such as machinery or physical technology infrastructures. The product industry can be further divided into the primary and secondary (or manufacturing) sector. In the primary sector companies offer a product that is in its original form, like at its time of extraction, and can thus be any natural resource such as minerals, petroleum, wood or agricultural product. The manufacturing sector is often the next complimentary step in a product’s value chain as companies in this sector manufacture natural resources into finished goods that can subsequently be used for different business purposes. Examples of manufacturing industries are retail, processed foods and beverages, machinery or any other manufacturers of equipment. The service industry is different from the primary and secondary sectors in the sense that service-centred MNEs offer something that is intangible and sometimes called a soft part of the economy. Examples of service

(24)

industries are banking, consulting, education and construction. As a country becomes more developed this also has an effect on the composition of outputs per sector (World Bank, 2000). In early stages of a country’s development agriculture, being the primary sector, is the most important sector. As GDP rises the manufacturing sector gains importance. The subsequent transition as GDP keeps rising is the emergence of the service sector as the most vital part of the economy. These industrialization and postindustrialization transitions have an effect on a country’s labor force, the consumer landscape and general composition of its economy. These transitions can also be seen in the BRIC countries as in China for example economic output from industrial activities has annually grown an average of 11,4% since economic reforms in the 1970’s (Han & Zhang, 2006).

Product development stages are less abstract than service processes with the latter being intangible. This affects the way in which EM MNEs will structure their entry into a new market. As mentioned before, EM MNEs often expand abroad with a springboard perspective, in which the goal of international expansion is to gain strategic assets which are often intangible, hard to accumulate and therefore not easily replicable when staying in the home country (Luo & Tung, 2007). As product EM MNEs are less dependent of intangible assets than service EM MNEs, it is also expected that these firms are less likely to expand abroad by applying the springboard perspective, which is mostly focused on intangible strategic assets. Instead, product EM MNEs are more likely to be oriented towards market-seeking or efficiency-seeking motives that are mostly concerned with attaining tangible assets such as cheap labor and material resources (Dunning, 1980). Cohen and Levinthal (1990) state that firms that are after tangible assets have less need to enter joint ventures as these assets are also obtainable without local partnership.

(25)

Compared to product firms, service firms are more often subject to adapt to local cultural aspects such as language and national business system (Ramasamy & Yeung, 2010) as many services are location bounded and are conveyed through face-to-face contact with the customer (Boddewyn et al., 1986). By entering the foreign market through a joint venture with a local partner a service is better able to make use of the local partner’s knowledge of the host country market and culture (Cohen & Levinthal, 2007).

Furthermore, services are distinctively different from products largely due to their inseperability, meaning that they are produced and delivered at the same time (Habib & Victor, 1991). Thus, the assets of a service firm’s value chain are entangled with each other and inseperable. When a firm is interested in acquiring certain assets from another service firm through a full acquisition it would therefore have to acquire the entangled nondesired assets as well and incur high costs in doing so (Hennart & Reddy, 1997). In such a situation, the solution to gaining access to desired assets, while not having to incur unneccesarily high costs could be a joint venture in which the partner firm remains responsible for part of the operation. Taking all above findings into consideration the following hypothesis is formulated:

H1: EM MNEs that belong to a product industry are more likely to establish a wholly owned subsidiary when investing abroad than EM MNEs that belong to a service industry.

3.2 Primary and secondary industries

When conducting an industry-level analysis of foreign equity commitment it is also important to make the distinction within sectors. As mentioned before, the product

(26)

industry can also be further divided into the primary and secondary (or manufacturing) sector. The primary sector is concerned with the extraction of natural resources and the secondary sector often accomodates the next step in the value chain by processing natural resources into finished goods.

Characteristic for the BRICs is that certain countries are highly dependent of natural resource related FDI, with some of the largest companies being in the extractives sector. A good example of this is the Russian economy in which the three largest outward investors, all three oil and gas companies, together accounted for 44% of the national outward FDI flows in 2007. Since Brazil, Russia, India and China’s combined landmasses account for almost one third of the world’s surface (Kumar, 2013) it is no surprise that these countries possess many natural resources and are therefore active in the primary sector through oil, gas and other resource extraction; as seen in table 2.

Table 2. Main outward investing industries per BRIC country. Source: Gammeltoft, 2008

Country Industries

Brazil Energy, mining, services

Russia Oil, gas, metal, manufacturing and telecommunication

India Pharmaceuticals, agricultural inputs, software, IT and broadcasting

China Trade and services, manufacturing, resource extraction (oil, gas, minerals), IT.

A big difference between the primary and secondary sector is that natural resources do not need to be configured from their original form while manufacturing firms develop the products themselves into finished goods. As culture has a large

(27)

effect on consumer demand and preferences (Hofstede, 1983) manufacturers are thus more likely to have to alter their products to match these local preferences when expanding abroad. Petroleum extraction for example is usually conducted in only a few different ways no matter where in the world, while offering consumer products requires the manufacturer to often take local tastes and culture into account. These institutional factors have been proven to have an effect on the degree of foreign equity commitment in an overseas subsidiary (Makino & Yiu, 2002). As institutional uncertainty increases, MNEs are more likely to engage in an equity joint venture over a wholly owned subsidiary. Culture increases a manufacturing MNE’s possible exposure to such uncertainty (Hofstede, 1983) and will therefore lead to manufacturing EM MNEs being more likely to enter new markets through a joint venture instead of through a wholly owned subsidiary. These industries that require effective communication with consumers are also referred to as advertising intensive industries and can adapt to culture in the host country by partnering up with a local firm in order to gain access to for example market knowledge (Demirbag, Glaister, Tatoglu, 2007). Thus by entering through a joint venture with a local partner a manufacturing firm will be able to make use of the local partner’s complementary knowledge of the host country market (Cohen & Levinthal, 2007; Chang, Chung & Moon, 2009).

Many EM MNEs in the primary sector possess superior resources that remain scarcely available and cannot be imitated. From a resource-based view these assets, such as the ability to exploit petroleum supplies (as is the case with many BRIC MNEs) are therefore firm specific advantages (FSAs) that can lead to competitive advantage (Peteraf, 1993). In the case of oil companies these competitive advantages are sustained due to ex post limits to competition such as high entry barriers and

(28)

resource heterogeneity (Peteraf, 1993). Firms with these unique natural resources that form FSAs have enough monopolistic power generated by their unique resources in order to mitigate environmental uncertainties and “streamline indigenous activities” (Brouthers & Brouthers, 2003). When this is the case, the degree of an MNE’s resource commitment is more driven towards simply conforming to transaction cost theory. (Luo, 2004). Thus, as these firms are less in need of a local partner to mitigate environmental uncertainties we expect them to opt for a WOS as this lowers the costs of coordination between partners and leads to transaction cost savings (Brouthers & Brouthers, 2003). As the number of FSAs increases so does the likelihood of entering new markets through internalization (Williamson, 1975) in order to exploit FSAs internally and protect core competencies that are unique to the firm (Chao, Lo & Yu, 2010). We expect this to lead to an increase of ownership in foreign subsidiaries established by EM MNEs emerging from the primary sector and possessing many FSAs.Thus the following hypothesis is formulated:

H2: EM MNEs that belong to a primary industry are more likely to establish a wholly owned subsidiary when investing abroad than EM MNEs that belong to a secondary industry.

3.3 Degree of internationalization

Lastly for the hypothesis development, the degree of internationalization of EM MNEs will be taken into account. Rugman and Verbeke (2004) conducted perhaps one of the most widely referenced studies on the internationalization of MNEs. To gain a better insight into how some of the world’s largest MNEs invest abroad Rugman and Verbeke analyzed geographical dispersion of sales across the Triad

(29)

Region, consisting of North America, Europe and Japan. Results showed that the majority of the companies was more regionally focused than globally focused, meaning that business activities were mostly located in geographically nearby countries as opposed to evenly located across all regions. According to Johanson and Vahlne (1977) firms incrementally gain experiental knowledge as they become more committed to foreign ventures while also expanding increasingly geographically widespread. As hypotheses 1 and 2 state, we expect EM MNEs operating in the primary and product industries to be more likely to establish wholly owned subsidiaries than their counterparts when expanding abroad. Sole ownership has been proven to positively influence a firm’s level of international business experience (Davidson, 1982; Ekeledo & Sivakumar, 2004), which is similar to a firm’s internationalization learning ability as described in the Uppsala model (Johanson & Vahlne, 1977). As mentioned in previous sections, primary and product sector MNEs in general conduct business that is less exposed to host country specific institutional factors (Ramasamy & Yeung, 2010; Boddewyn et al., 1986; Habib &Victor, 1991). Therefore these firms will be able to adapt quicker to distant new business environments and we expect, in line with Johanson and Vahlne (1977), that these same firms are able to make larger incremental steps while internationalizing. EM MNEs operating in the manufacturing and service sectors on the other hand, experience a more complex internationalization process and will therefore be longer inclined to opt for a joint venture structure as they will still have need for local partners that can help them acquire necessary knowledge of the host country market (Chang, Chung & Moon, 2009). For this reason we expect that as the degree of internationalization increases primary and product industry EM MNEs will establish

(30)

even more majority-owned affiliates compared to their counterparts in the service and manufacturing industries.

H3: An EM MNE’s degree of internationalization positively moderates the relationship hypothesized in H1.

H4: An EM MNE’s degree of internationalization positively moderates the relationship hypothesized in H2.

The hypotheses and their underlying relations are shown in figure 1.

Figure 1. Conceptual framework.

Primary vs secondary industry Foreign equity commitment Degree of internationalization Product vs service industry H3 H1 H4 H2

(31)

4. Methodology

4.1 Sample and data collection

In this part of the research the means with which the hypotheses are tested will be discussed, as well as which variables will be used. Concluding, the results of the statistical analysis will be presented. The goal of the analysis is to statistically research the relation between the industries in which EM MNEs operate and their degree of foreign equity commitment in the form of wholly owned overseas ventures.

The study is a cross-sectional research in which the sample is comprised of the 500 largest MNEs originating from Brazil, Russia, India and China as a representation of emerging market firms. All five BRIC countries are equally represented in the sample as it includes 100 of the largest MNEs from each country. In this research the largest firms are considered those with the largest revenues in the fiscal year of 2014. The data was collected through Bureau van Dijk’s online database Orbis. Orbis provides extensive information on companies ranging from revenues to year of incorporation, subsidiary information and ownership information and therefore forms a reliable source for information on the 500 largest BRIC MNEs. Furthermore the 2014 annual report of each firm has been used as a primary source to collect data on the firm’s expenses and geographical dispersion of sales and assets. For this research the most important information collected through the Orbis database and the annual reports are the BvD major sector classification of each firm, geographical dispersion of sales and subsidiaries and the degree of equity commitment in each subsidiary, which can be measured by looking at the total percentage owned by the company in the foreign venture. After excluding all the companies that did not disclose full information on the amount of foreign sales, a final

(32)

sample of 129 EM MNEs remained. The amount of companies per BRIC country is displayed in table 3.

Table 3. Sample composition by companies’ country of origin.

4.2 Measures

In order to conduct the statistical analysis the following set of variables will be used to research the relation between EM MNE industry classification and foreign equity commitment as accurate as possible.

Dependent variable

The dependent variable used for the statistical analysis is EM MNE degree of foreign equity commitment. This refers to the degree of ownership of EM MNEs in ventures abroad. These ventures solely include equity based entry modes in which a new business entity is formed in the host country. In order to determine the amount of wholly owned affiliates a dummy variable is created for which 1 indicates a wholly owned affiliate (100% ownership) and 0 indicates a non-wholly owned affiliate (<100% ownership).

Home country No. of companies Brazil 30

Russia 27

India 30

(33)

Independent variable

As an independent variable the research uses the sector classification of the EM MNEs. The classification is sourced directly from the Orbis database in which each firm has been categorized into one of the nineteen BvD major sectors. These sectors range from primary sector and education and health all the way to construction and insurance. A full overview of all nineteen BvD major sectors can be found in appendix A. In order to conduct the statistical analysis with a focus on product versus service industries and primary versus secondary industries the BvD major sectors as found in Orbis need to be grouped. The service sector MNEs used for the analysis for example will be a grouping of all MNEs belonging to the BvD major sectors of “Construction”, “Banks”, “Insurance companies” and “Education” amongst others as these all belong in the service sector. For the complete overview of groupings of the BvD major sectors please refer to appendix B. For the analysis the different sector variables (product, service, etc) will be defined as dummy variables. For the two respective groupings a firm that belongs to the poduct or primary sector is classified as “1” and a firm that belongs to the service or secondary sector is classified as “0”.

Moderator variable

As a moderator variable the EM MNEs’ degree of internationalization will be used. As mentioned before much research has been conducted regarding mapping the degree of MNE internationalization by for example analyzing its geographic dispersion of sales. This resulted in the classification of regionally and globally focused MNEs (Rugman & Verbeke, 2004). The degree of internationalization will be used in this research as a means of determining its effect on the degree of foreign equity commitment by EM MNEs. The variable will be measured by dividing a firm’s

(34)

amount of foreign sales by the firm’s amount of total sales. Thus, the degree of internationalization is defined by a firms’s ratio of foreign to total sales.

Control variables

For the statistical analysis several control variables will be used. The first one is the commonly used control variable of firm size. The size of a company has an effect on its performance, as larger firms are often able to invest in larger projects, create economies of scale and therefore generate larger returns (Chao & Kumar, 2010). Firm size will be defined by a firm’s total number of employees. The second control variable used is firm performance and will be measured by a firm’s return on assets (ROA). As higher performing firms are expected to possess more abundant resources they are also expected to be more likely to expand more geographically dispersed and have a higher degree of internationalization. The third control variable is firm age and will be calculated by substracting the company’s year of incorporation from the year 2014. Johanson & Vahlne’s Uppsalla model shows that companies tend to first internationalize regionally in order to learn from experiences and incrementally expand more geographically and culturally distant. Therefore, firm age should be accounted for when comparing firms’ degrees of internationalization.

5. Statistical analysis and results

The descriptive statistics for the dependent and independent variables are detailed in table 4. The statistics for the moderator and control variables have also been added in order to investigate the correlations between the various variables. Any correlation above 0.7 can be considered harmful for the interpretation of results (Pallant, 2011). As such high correlations are not present in this table we can conclude that all

(35)

variables can be used for further analysis. We should note that the product versus service variable and primary versus secondary variable display a rather large correlation of -0.518 with each other. However, this does not form an issue as these two variables will be applied in separate regression models. The mean value of 0.728 for foreign equity commitment suggests that EM MNEs in general invest more often in the form of a wholly owned subsidiary as opposed to a non-wholly owned subsidiary. Also, the degree of internationalization seems rather low with EM MNEs on average generating only 27.2% of sales outside of the home country.

Table 4. Means, standard deviations and correlations.

Variables Mean SD 1 2 3 4 5 6 7 1. Foreign equity commitment (100%=1; <100%=0) 0.728 0.445 − 2. Primary (1) & Secondary (0) industries 0.12 0.331 0.058 − 3. Product (1) & Service (0) industries 0.62 0.488 0.197* -0.518** − 4. Degree of internationalization 0.272 0.288 0.128 -0.259** 0.321** − 5. Firm age 35.08 27.450 0.046 0.106 0.109 0.076 − 6. Firm size 65,650 98,597 0.006 0.547** -0.286** -0.028 0.041 − 7. Firm performance 4.709 10.048 0.016 -0.044 -0.007 0.061 0.189* 0.048 − *correlation is significant at the 0.05 level **correlation is significant at the 0.01 level

In order to test the hypotheses a hierarchical regression model is used. The first step in the regression model is shown in model 1 in table 5 and is executed in order to test the effect of the control variables firm size, firm age and firm performance on the

(36)

dependent variable foreign equity commitment. In models 2 and 3 the independent variable of industry classification is introduced in a binary logistic regression model. A binary logistic regression model is chosen because the goal is to test the interaction between a dependent and independent variable, which are both dichotomous. For this type of regression model assumptions differ from a typical OLS regression as no linearity, normal distribution or homescedasticity is assumed (Ho, 2013). Foreign equity commitment is expressed as either 1 for wholly owned subsidiaries or 0 for subsidiaries in which the EM MNE has a less than 100% stake. In model 2 the independent variable industry classification is expressed as either 1 or 0; with 1 indicating product sector and 0 indicating service sector. In model 3 the independent variable industry classification is also expressed as either 1 or 0; with 1 being primary sector and 0 being secondary sector. In models 4 and 5 the moderator variable degree of internationalization and the respective interaction terms are added to the regression. The interaction terms were calculated by multiplying the moderator variable with the independent variable.

In order to interpret the results of the regression model three values are the most important to look at. First, the beta standardized coefficient shows the change in the dependent variable as a result of a change in the independent variable. The beta standardized coefficient can be either positive or negative and can therefore also indicate a positive or negative relationship between the different variables. The second and third values important for interpretation of the results pertain to the goodness of fit of the model and are the Hosmer and Lemeshow test statistic and the Nagelkerke R2 statistic. The Nagelkerke R2 statistic indicates the goodness of fit of the model and the Hosmer and Lemeshow test measures the significance of the logistic regression model.

(37)

Model 2 indicates a beta of -0.342 and a significance of 0.005. This model tests hypothesis 1, which states that product EM MNEs are more likely to establish a WOS when investing abroad than service EM MNEs. As the p-value is significant (0.005) and the beta is negative we can conclude that service EM MNEs are actually more likely to establish a WOS abroad than their producing counterparts. Thus H1 is rejected. As for the goodness of fit, the Nagelkerke R2_{statistic increases from 0.063}

to 0.088 and indicates that after including the independent variable the data fits the model better. As for hypothesis 2, the results (b=0.513; p=0.045) indicate that primary industry EM MNEs are indeed more likely to establish a WOS when investing abroad than secondary industry EM MNEs. Thus H2 is supported. By including the independent variable in the regression model for H2 the Nagelkerke R2_statistic

slightly increases from 0.063 to 0.065 and therefore fits the model slightly better. When testing the moderator effect it seems that the degree of internationalization does not positively moderate the relationship hypothesized in H1. A moderating role can only be proven when the coefficient is both positive and significant. Thus H3 is rejected (b= -2.804; p= 0.000) and actually proves a relationship opposite to the one hypothesized. Ergo, as the degree of internationalization increases the likelihood of service EM MNEs to establish a foreign WOS increases even faster than the likelihood of product EM MNEs to establish a foreign WOS. Furthermore, model 5 is significant at a p < 0.1 level (b=2.893; p=0.056) and therefore indicates that H4 is supported and the degree of internationalization positively moderates the relationship hypothesized in H2. The regression model also indicates that all three control variables are significantly related to EM MNE foreign equity commitment (p < 0.01), with firm performance being the strongest predictor.

(38)

Table 5. Results of the hierarchical regression model.

Dependent variable: Foreign equity commitment

Controls

Model 1 H1 Model 2 H2 Model 3 H3 Model 4 H4 Model 5

Control variables Beta Sig. Beta Sig. Beta Sig. Beta Sig. Beta Sig.

Firm size 0.000*** 0.000 0.000*** 0.000 0.000 *** 0.000 0.000*** 0.000 0.000*** 0.000 Firm performance 0.043 *** 0.002 0.052 *** 0.002 0.044 *** 0.002 0.045*** 0.002 0.050*** 0.002 Firm age -0.004 *** 0.005 -0.004 *** 0.008 -0.003** 0.023 -0.006*** 0.000 -0.002 0.240 Independent variable Product vs. service sector -0.342*** 0.005 1.189 0.000 Primary vs. secondary sector 0.513** 0.045 -0.400 0.425 Moderator variable Degree of internationalization 1.144*** 0.007 0.608 ** 0.018 Interaction terms Product vs. service sector x degree of internationalization -2.804*** 0.000 Primary vs. secondary sector x degree of internationalization 2.893* 0.056 Constant 1.209 0.000 1.374 0.000 1.159 0.000 1.001 0.000 1.417 0.000 R2 0.063 0.088 0.065 0.127 0.074 Hosmer & Lemeshow (Sig.) 0.000 0.000 0.000 0.000 0.000 Change in R2 0.025 0.002 0.064 0.011 * p < 0.10 ; **p < 0.05 ; ***p < 0.01.

(39)

(40)

6. Discussion

Previous literature has proven that institutional and industry-specific factors influence the entry mode choice and foreign equity commitment of MNEs in general (Erramilli & Rao, 1993; Ekeledo & Sivakumar, 2004). With the statistical analysis we focused specifically on the relation between industry classification and foreign equity commitment by EM MNEs. The results show that in the case of product versus service firms a significant relation was found in a direction opposite from the one hypothesized. Thus we can conclude that service EM MNEs are more likely to establish wholly owned subsidiaries than product EM MNEs when expanding abroad. Also, the degree of internationalization has been proven to be a moderator for this relationship. Possibly, these outcomes are due to the knowledge-intensive nature of the service industry. Often acquisitions in the service industry are made with the goal of gaining intangible strategic assets. This is especially the case with service firms originating from emerging markets as they seek to overcome late-mover disadvantages by acquiring knowledge-intensive firms from developed countries (Luo & Tung, 2007). As intellectual property rights are often more delevoped in these host countries (Chen, 2013) a joint venture might possibly not offer the opportunity to gain access to these desired strategic assets while a wholly owned subsidiary does (Kogut and Singh, 1988).

Furthermore, industries with high levels of intangible assets are often more exposed to rapid technological changes and fierce competition, which require them to quickly adapt to these changes and implement fast decision-making (Chang, Chung, Moon, 2013). According to transaction cost theory (Anderson & Gatignon, 1986; Williamson, 1991) this is done most efficiently through sole ownership, which might make a wholly owned subsidiary the more attractive entry mode choice for service

(41)

EM MNEs. Also, a possible explanation for service EM MNEs’ foreign equity commitment being positively moderated by their degree of internationalization is related to the relatively low level of initial investment (Erramilli & Rao, 1993) needed to establish an overseas WOS. Contrary to producers, service firms have considerably less investment obligations in the form of tangible assets such as production facilities and machinery when opening new subsidiaries, making it a more complex process.

In the case of primary versus secondary EM MNEs the regression model shows that EM MNEs from the primary sector are indeed more likely to establish a WOS abroad than their secondary sector counterparts. This relation is positively moderated by the degree of internationalization. In the following sections the academic relevance, mangerial implications and suggestions for future research will be presented.

6.1 Academic relevance

This research contributes to existing literature on foreign direct investment by emerging market firms and the factors that determine the degree of equity commitment in these foreign subsidiaries. Specifically, this research has contributed to research in this field by adopting an industry-level focus. Results suggest that EM MNEs operating in service industries are more likely than producing EM MNEs to expand abroad by establishing a wholly owned subsidiary. When dividing industries into primary and secondary, results suggest that primary industry EM MNEs are more likely to establish a WOS than their secondary industry counterparts.

Firms originating from developing countries have received increasing academic attention in recent years (Hoskisson et al., 2000; Gammeltoft et al., 2010) due to their rising importance in the global economy and their increasing outward FDI

The internationalization of emerging market multinationals : a subsidiary-level investigation of foreign equity commitment across industries