The Influence of State-ownership on the
Internationalization Patterns of Multinationals
Thijs Gussenhoven
Student number: 10683690
Date: June 26 2015
MSc Business Administration: International Management Master Thesis
University of Amsterdam Supervisor: Dr. Niccolò Pisani Second supervisor: Dr. Rene Bohnsack
Statement of originality
This document is written by Student Thijs Gussenhoven 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.
ABSTRACT
State-owned multinational companies (SOMNCs) expanded their global reach in the last two decades. Extensive research has been conducted about the objectives and performance of SOMNCs. However, the role of state-ownership on the globalization patterns of multinationals is relatively unexplored. It is suggested that the globalization strategies of SOMNCs differ from private-owned multinational companies (POMNCs). SOMNCs have to take political objectives of the government into account, in contrast to their private counterparts whereby decisions are for the most part driven by business goals underlying the creation of economic value. This thesis examines the effect of state-ownership on the internationalization path of multinationals by looking at the distribution of sales of the Fortune Global 500 companies as listed in 2014. The presented research investigates firm’s level of multinationality and firm’s global (versus regional) focus. Furthermore, we study the moderating effect of the development of the home country. The findings suggest that SOMNCs have a low level of multinationality compared to POMNCs. This significant difference is even stronger for developing countries than for developed countries, indicating the effect of the moderation.
Keywords: State-owned multinationals; Private-owned multinationals; internationalization path; globalization; home country; development.
TABLE&OF&CONTENTS& 1.&INTRODUCTION&...&5! 2.&LITERATURE&REVIEW&...&7! 2.1!SOMNCS!...!7! 2.1.1$Reasons$for$the$existence$of$SOMNCs$...$7! 2.1.2$A$changing$view$on$SOMNCs$...$9!
2.2!INTERNATIONALIZATION!OF!SOMNCS!VIS6À6VIS!POMNCS!...!10!
2.3!INTERNATIONAL!STRATEGIES!OF!SOMNCS!VIS6À6VIS!POMNCS!...!12!
2.4!MNCS!AND!THE!INFLUENCE!OF!THE!HOME!COUNTRY!...!14! 2.5!RESEARCH!GAP!...!15! 3.&THEORETICAL&FRAMEWORK&...&15! 3.1!STATE6OWNERSHIP!AND!THE!DEGREE!OF!INTERNATIONALIZATION!...!15! 3.2!DEVELOPING!VERSUS!DEVELOPED!MARKET!MNCS!...!19! 4.&METHODS&...&21! 4.1!SAMPLE!AND!DATA!COLLECTION!...!21! 4.2!MEASURES!...!22! 4.2.1$Dependent$variables$...$22! 4.2.2$Independent$variable$...$23! 4.2.3$Moderating$variable$...$23! 4.2.3$Control$variables$...$23! 4.3!STATISTICAL!ANALYSIS!AND!RESULTS!...!25! 5.&DISCUSSION&...&31! 5.1!ACADEMIC!RELEVANCE!...!32! 5.2!MANAGERIAL!IMPLICATIONS!...!33! 5.3!POLICY!IMPLICATIONS!...!34! 5.4!LIMITATIONS!AND!SUGGESTIONS!FOR!FUTURE!RESEARCH!...!34! 6.&CONCLUSION&...&34! ACKNOWLEDGEMENT&...&36! LITERATURE&...&37! ! ! LIST&OF&FIGURES& Figure!1:!Conceptual!model!...!20! Figure!2:!The!relationship!between!state6ownership!and!internationalization!scale!...!31! ! LIST&OF&TABLES& Table!1:!Distribution!of!corporation's!home!country!and!type!of!ownership!...!22! Table!2:!Results!of!Hierarchical!Regression!analysis!for!scale!of!internationalization!...!29! Table!3:!Results!of!Hierarchical!Regression!analysis!for!scope!of!internationalization!...!30! ! !
1. INTRODUCTION
In a free market economy, the private-ownership of business is the norm while the state-ownership is rare. This is especially the case since the global privatization wave of the 1980s and 1990s (Estrin, Meyer, Nielsen & Nielsen, 2012). However, state-owned multinational companies (SOMNCs) have expanded their global reach in the last two decades (Economist, 2012). In 2013, there were at least 550 SOMNCs with more than 15,000 foreign affiliates and the estimated amount of foreign assets exceeded the $2 trillion. Some of the SOMNCs, both from developed and developing countries, are among the largest multinational companies (MNCs) in the world. Despite the fact that the number of SOMNCs is less than 1% of the total amount of MNCs, they are responsible for more than 11% of global FDI flows (UNCTAD, 2014).
SOMNCs tend to be extremely large in size; in 2014 103 SOMNCs were ranked on the Financial Times Fortune Global 500, which is the annual snapshot of the world’s largest companies. While considerable research has examined why SOMNCs become multinationals (e.g., Li, Cui & Lu, 2014; Choudhury & Khanna, 2014; Cuervo-Cazurra, Inkpen, Mussacchio & Ramaswam, 2014), relatively less attention has focused on their patterns of globalization. This gap in the IB literature could be explained by the fact that the internationalization of SOMNCs on a large scale is a rather new phenomenon. This is also emphasized by Mussacchio and Lazzarini (2012), who examined the widespread influence of the government in the economy, either by owning majority of minority equity positions in firms. The authors presented systematic, cross-country evidence showing that the current form of SOMNCs in the twenty-first century is different from what is observed in the second half of the twentieth century (Mussacchio & Lazzarini, 2012). In 2005 there was no single SOMNC among the top-10 firms of the Fortune Global 500 list while there were three in 2014.
SOMNCs have usually had a domestic focus. Therefore, much of the existing literature in international business (IB) has tended to overlook SOMNCs and their operations. Other academic fields such as the political economy and developmental economics have developed important views concerning globalizing patterns of SOMNCs. To this end, the IB literature could gain insights from these studies and enhance its focus on the increasingly international activities of SOMNCs.
This thesis sheds light on the relationship between state-ownership and globalization patterns. Existing research suggests that SOMNCs differ from private-owned multinational companies (POMNCs) in their globalization patterns (Estrin et al., 2012; Cuervo-Cazurra et
al., 2014). However, no previous study has investigated the specific internationalization paths of SOMNCs by looking at both the scale and scope of their expansion. Although considerable evidence highlights the existence and increasingly important role of regional strategies for MNCs (Rugman & Verbeke, 2004; Gilbert & Heinecke, 2014), theoretical work in this area is still in its infancy. Moreover, this study contributes to this growing literature by investigating whether the development of MNCs’ home country has a moderating role on their internationalization path.
The role of SOMNCs varies across countries because SOMNCs are owned by the government and hence subject to multiple political and economic pressures (Estrin, 2012; Cuervo-Cazurra et al., 2014). SOMNCs may be motivated by objectives other than profits. It is therefore likely that conflicts arise between the government as owner and managers of SOMNCS. At the same time, private shareholders create pressures for the SOMNC to act more like a profit-oriented POMNC. This constellation can lead to so-called agency problems that affect the international strategies of SOMNCs.
It is hard to generalize the findings of existing research since they only focus on SOMNCs that are active in one industry or belong to the same home country. For instance, Choudhury and Khanna (2014) recently examined the global footprint of 42 Indian State-owned laboratories. Another example is the paper of Wolf (2009) that examined the existence of ownership effects in the global oil and gas industry. To fill this gap in the literature this thesis focuses on the 500 largest MNCs around the globe, as included on the Fortune Global 500 list in 2014. Consequently, it provides a better understanding of how the internationalization path is different between the most relevant SOMNCs and POMNCs.
To assess the impact of state-ownership on the internationalization path of MNCs, we develop the literature on SOMNCs to address three questions. First, how does state-ownership relate to firm’s level of multinationality? Second, how does state-ownership relates to firm’s global (versus regional) focus? Third, how does the developing (versus developed) market status of the home country moderate the relation between state-ownership and internationalization path? We find that SOMNCs, on average, have a lower level of multinationality than POMNCs. This effect is even stronger when the home country of a SOMNC has a developing market status.
We thus contribute to the IB literature in several ways. First, we integrate different perspectives to advance the theory of the relationship between state-ownership and path of internationalization of large MNCs. Second, this thesis makes a distinction between scale and scope and therefore offers greater insights in the actual internationalization paths of both
SOMNCs and POMNCs. Third, this is as far as known the first study that identifies the relationship between state-ownership and the internationalization path by looking at the influence of the development of the home country. Fourth, our empirical results grounded in a large unique dataset point to behaviors of large SOMNCs and POMNCs that are of potential interest to managers.
This thesis is structured as follows. First, the relevant literature about the SOMNCs and its globalization patterns is discussed in the second section. Subsequently, a theoretical framework with the development of the hypotheses is stated in the third section. The fourth methodology section explains the data collection, variables and the method that is used to carry out this research. This section also presents the results from the statistical analyses. The fifth section discusses the interpretation, implications and limitation of the findings. This thesis ends with concluding remarks in the sixth section.
2. LITERATURE REVIEW
2.1 SOMNCsThis paper uses the following definition of a SOMNC: “A SOMNC is a legally
independent firm with direct ownership by the state that has value-adding activities outside its home country” (Cuervo-Cazurra et al., 2014, p. 925). These value-adding activities can be
downstream activities, which are more related to the buyer and often tied to where the buyer is located. An example is the production of advertising material. The value-adding activities can also be upstream activities such as inbound logistics, which can at least conceptually be decoupled from where the buyer is located (Porter, 1986). The amount of control a state has in the SOMNC varies, whereby it could be a full, majority, or minority ownership. The International Monetary Fund (2003) considers a corporation as state-owned if the state owns 10% or more of the ordinary shares. Our study uses this definition as well and will not make any further distinction in the level of ownership, unless stated otherwise.
2.1.1 Reasons for the existence of SOMNCs
The existence of SOMNCs could be generally explained by two different views. The first view is the idea that state-ownership could be a solution to market failures. For several decades the theory of market failures takes a central place in the study of public policy (Wu & Ramesh, 2014). The theory describes the situation whereby the pursuit of private interests leads to inefficient outcomes andthe market is unable to allocate resources or products in the
most welfare-enhancing way. Market failures could be associated with limited public goods, natural monopoly, externalities and information asymmetries. Governments can intervene to correct these problems by using regulations, subsidies or taxations. Governments can also use the instrument of direct ownership, which lead to the creation of SOMNCs.
Scholars such as Wolf (1987) criticized the theory of market failures and said that the efficacy of governments in solving problems cannot be taken for granted. Governments suffer from government failures whereby interventions may result in inefficient outcomes. This is due to a lack of competition in the public sector. Other criticizing scholars pointed to principal-agent problems, which are costs that emerge as a result of the division of labor between principals and agents (Kistruck, Sutter, Lount & Smith, 2012). There is no clear evidence that government intervention would worsen the situation of market failures. Some scholars still argued that society is better off when the market is left alone (Friedman, 2002). Wu and Ramesh (2014) emphasized that the disastrous impact of the recent financial crisis suggests otherwise and that the consequences of ignoring market failures can have catastrophic consequences for societies.
Another explanation for the existence of SOMNCs is related to ideological and political strategies of governments. This alternative view argues that four different ideologies and political strategies contribute to the creation of SOMNCs. First is the ideology of communism, which is the economic and political doctrine that seeks to replace a profit-based economy and private property with public ownership (Dagger, 2014). Communism, represented by the schools of thought of Marxism, supports the idea of communal control of the natural resources of a society and at least the major means of production, such as mines and factories. The ideology of communism justifies the nationalization of private firms and the creation of SOMNCs. The view supports the idea that the state, in the name of the country’s citizens, should be the owner of companies that are important for societies. Second is the social ideology, also mentioned by Musacchio and Lazzarini (2012), whereby governments have “non-commercial” objectives that go beyond profitability. Governments use SOMNCs to promote socially desirable objectives such as job creation, regional development, healthcare, and income redistribution. Third is the nationalist ideology that encourages the creation of SOMNCs if it accelerates the development of the country. Especially in poorly developed financial markets, local entrepreneurs often do not have the capacity or interest to invest in large-scale projects that stimulate the national economy (Musacchio and Lazzarini, 2012). Fourth is the strategic ideology that argues that the government needs to invest in SOMNCs when they are strategic for the country
(Cuervo-Cazurra et al., 2014). It depends on the political strategy and the country-specific situation to determine if an industry, like defense, is important enough for the creation of a SOMNC. 2.1.2 A changing view on SOMNCs
Much of the current literature still has a classical view of SOMNCs, in which firms are poorly managed without clear strategy and resource allocation decisions. The main focus of POMNCs is profit maximization. However, SOMNCs also need to take political objectives into account (Cuervo-Cazurra et al., 2014). Megginson and Netter (2001) supported this classical view and examined the extant literature on privatization, published in the 1980-2000 period. The authors concluded that POMNCs are more efficient and more profitable than otherwise-comparable SOMNCs. Dewenter and Malatesta (2001) agreed with this statement and examined whether firm performance of SOMNCs listed among 500 largest non-US firms in 1975, 1985, and 1995 differ from POMNCs on the same list. The authors concluded that POMNCs are significantly more profitable, have significantly less debt, and less labor-intensive production processes than SOMNCs. More recently, similar findings are published by Wolf (2009), who investigated the existence of ownership effects in the global oil and gas industry by looking at differences of performance and efficiency between national oil companies and international oil companies. The findings showed that “ownership matters” in the sense that private ownership leads to higher performance and more efficiency than state ownership.
This said, Cuervo-Cazurra et al. (2014) argued that it is time to revise this classical view of shortcomings of SOMNCs. The companies are becoming wealthier and more powerful, even as the overall state sector shrinks (The Economist, 2012). Due to pro-market reforms in Latin America, Europe and Asia, SOMNCs have increasingly expanded their presence in many market economies. Although SOMNCs already exist for a long time, these developments led to a new sort of SOMNCs that emphasize the shortcomings of their predecessors since they have a stronger international focus (Cuervo-Cazurra et al., 2014). This is especially the case for SOMNCs who are active in extractive industries, such as transport, electricity, telecommunication, and water. Those companies are increasingly venturing abroad in order to lock up future energy supplies or forming alliances with private-sector specialists. In doing so, SOMNCs are able to increase their access to expertise and ideas. The Chinese oil company Sinopec bought a huge Angolan oil well for $692 million in 2006 and it is striking deals across Africa ever since. Gazprom did the same across Eastern Europe and Asia. In 2006, the Russian SOMNC bought a 51% stake the Serbian energy giant
Naftna Industrije Srbije (The Economist, 2012). Aligned with this reconceptualization, Li et al. (2014) emphasized that the international expansion of SOMNCs is increasingly promoted in developing countries. The government of these countries owns some of the world’s largest MNCs, especially in the oil and gas industry (UNCTAD, 2013). The rapid globalization asks for a changing view on SOMNCs and increases the relevance to investigate the internationalization activities of such firms.
2.2 Internationalization of SOMNCs vis-à-vis POMNCs
The following section describes how the internationalization behavior of SOMNCs differs from the one of POMNCs. First, it explains the different internationalization motives of the two types of MNCs. Second, it describes how the internationalization strategies of SOMNCs differ from POMNCs.
It is important to clarify what is meant with a MNC. In this paper, a MNC is: “An
enterprise that engages in foreign direct investment and owns or, in some way, controls value-added activities in more than one country” (Dunning & Lundan, 2008, p. 3). This is the
threshold definition of a MNC and one that is widely accepted in the IB literature, by data-collecting agencies such as UNCTAD’s division on Investment and by most national governments. Moreover, this definition fits with the definition of the SOMNC, mentioned before. This thesis considers POMNCs as those MNCs that are not SOMNCs.
Generally, POMNCs internationalize and use foreign direct investment (FDI) when their profitability can be increased. Dunning (2000) identified four main types of foreign-based activities. First are those designed to satisfy a particular foreign market (market seeking FDI). Second are the activities designed to gain access to natural resources, like minerals or unskilled labor (resource seeking FDI). The third type of activities promotes a more efficient division of labor or specialization of domestic and foreign assets by MNCs (efficiency seeking FDI). At last are the activities designed to protect the existing ownership specific advantages of the MNC or to reduce those of the competitors (strategic asset seeking FDI). These activities fit perfectly well with the concept of a POMNC.
The profit-maximizing FDI motives, mentioned by Dunning (2000), are typical for POMNCs but not for SOMNCs. For POMNCs the success of FDI is based on the contribution to the profit of the firm, like return on investment. But the measure for success is more complicated for SOMNCs. This is due to different - and often conflicting - demands of citizens, managers, and policymakers (Globerman & Shapiro, 2009). Part of the traditional logic of the existence of state-owned companies is to enhance the welfare of its citizens by
solving market failures in the home country. In a domestic setting, the government has the ability to set rules to improve the economy or welfare of its citizens. However, the companies owned by the state do internationalize and become SOMNCs (Li et al., 2014). This is causing a dilemma for traditional explanations of state-owned companies since the FDIs are made outside the home country where the government does not have the power to set laws and regulations. The international dimension of SOMNCs might not go hand in hand with this traditional logic behind state-owned companies. However, three motives can be described to clarify how the government acts to help its home country citizens when the SOMNC invest abroad: (1) overcome market failures, (2) internalization theory, and (3) ideological and political strategies.
Market failures can exist in a global context. In such a case, extraterritorial state ownership is required to ensure the protection of global public goods and to come up with global solutions (Maskus & Reichman, 2004). Global market failures, or failures in other countries, could also negatively affect the welfare of the home country’s citizens. In such a case, SOMNCs might need to invest abroad to reduce the dependence of the home country on imports by private-owned corporations or to obtain raw materials and supplies (Cuervo-Cazurra et al., 2014). Thus it appears that SOMNCs can invest abroad to help its home country citizens.
A radical different explanation for the internationalization of SOMNCs is the internalization theory. Here, the existence of the MNC is caused by its efficiency properties, i.e., its capacity to reduce transactions costs when replacing an inefficient transaction in the market by an internal transaction within the firm (Rugman, Verbeke & Nguyen, 2011). This is especially the case in the context of transferring intermediate outputs across borders. The international element SOMNCs typically enhances rather than reduces consumer welfare because efficiently coordinated transactions substitute for inefficient ones.
At last, there are ideological and political motives for SOMNCs to internationalize. When governments invest abroad by using SOMNCs, they are able to impose their ideologies towards other countries. SOMNCs can thus become indirect extraterritoriality mechanisms to transfer policy or ideology predilections (Cuervo-Cazurra et al., 2014). Globerman and Shapiro (2009) investigated this mechanism by looking at the growing FDI from China. Policymakers in the US are concerned that acquisitions of US-companies by Chinese SOEs may be motivated by non-commercial objectives. The rationale for this concern is that political objectives dominate economic objectives in the case of foreign acquisitions of Chinese SOMNCs, whereas this is less obvious for acquisitions based in other countries.
However, Globerman and Shapiro (2009) emphasized that it is not possible to confirm this statement due to a lack of evidence. In line with the US policy concerns, Amighini et al. (2013) argued that for SOMNCs investing abroad, political objectives are at least – or even more - relevant than economic objectives. Managers of SOMNCs can use FDI to claim credit for themselves and their organization for undertaking activities that are in line with national interests. The next section describes to what extent the strategies of SOMNCs differ from POMNCs.
2.3 International strategies of SOMNCs vis-à-vis POMNCs
As discussed in the previous section, POMNCs have different motives for internationalization than SOMNCs, since the latter has non-business motives for internationalization. The success of FDI is measured in a different way, which means that international strategies also differ between POMNCs and SOMNCs. Cuervo-Cazurra et al. (2014) called this the “non-business internationalization argument” in order to explain these differences.
There have been several theoretical explanations within the IB literature that seek to explain the international behavior of MNCs (Rugman et al., 2011). One influential theory is that foreign firms must possess a countervailing advantage over local firms to make investments viable and the market for selling this advantage must be imperfect. It is also stated that there is a trade-off for MNCs between the financial benefits of entering a new country, e.g. market seeking or efficiency seeking, and the costs incurred in doing so (Hymer, 1976; Rugman et al., 2011). Interestingly, these standard logics do fit among POMNCs but are less applicable for SOMNCs since it does not take into account that the objectives of the latter might not be driven by profit. Whether or not a POMNC decides to internationalize is based on the potential economic value of host countries. On the other hand, SOMNCs might internationalize to facilitate a political relationship with the host country’s government. Actions like these have more to do with politics between countries than with the business or financial benefits of the SOMNC.
SOMNCs and POMNCs also differ in the way they select countries for FDI. The Uppsala model proposes that the international expansion pattern of a MNC is a function of its past international experience and knowledge base. It predicts a movement from “close” to more “distant” markets (Rugman et al., 2011). The economic theory on the other hand, does not predict a general expansion pattern of FDI. The theory predicts that location choices are discrete rational choices, and not cultural learning processes. Both theories agree on the idea
that the decision to undertake a FDI in a specific country is the outcome of a decision process where predicted costs and revenues are evaluated (Benito & Gripsrid, 1992). Both theories are therefore applicable for POMNCs. However, this is not the case for SOMNCs, whereby the choice of investment location might not be driven by profit. An example is the $12 billion deal between the state-owned China Railway Construction Corporation and the government of Nigeria in 2014, which is the biggest FDI contract in the history of China (Foreign Policy, 2014). This deal will not only boost the Chinese manufacturing sector, it will also support the political relationship between the nations. At last, it gives Chinese SOMNCs, and therefore the government, more of a foothold in Africa’s biggest economy.
Another difference between POMNCs and SOMNCs becomes clear when analyzing the appropriate mode of entry for both types. In the IB literature entry modes are closely associated with varying degrees of resource commitment, control, risk exposure, and profit return (Pan & Tse, 2000). Traditional models assume that firms try to increase their long-term profit while keeping risk-taking at a low level (Johanson & Vahlne, 1977). When considering equity modes (i.e., equity joint ventures or wholly owned subsidiary) or non-equity modes (i.e., export or contractual agreements), MNCs need to assess the risks and potential returns by looking at multiple host country characteristics, such as government restrictions or market conditions (Pan & Tse, 2000). However, SOMNCs select entry modes that serve the national interests even if such methods incorporate more risk and less profitability. Thus for SOMNCs, the selected entry mode is a product of the political objective, rather than economic consideration (Cuervo-Cazurra et al., 2014).
At last, SOMNCs have to deal with agency problems, which arise because contracts are not written and enforced without costs since agents have conflicting interests (Fama & Jensen, 1983). Traditionally, a POMNC has a single agency relationship whereby the interests of the managers could differ from those of the shareholders. SOMNCs, on the other hand, deal with two agency relationships, which increases the complexity. Citizens of the country (as principles) own the focal firm and ask the politicians (as agents) to achieve the objectives that serve national interests. In the second agency relationship the politicians now act as principals by giving tasks to the managers of the SOMNCs, who act as agents that are appointed to the politicians. The problem for SOMNCs is that these three parties all have their own interests. SOMNCs need to take all these into account in their international strategies.
Taken together, it appears that globalization strategies of SOMNCs differ from POMNCs. SOMNCs have to take political objectives of the government into account, in contrast to POMNCs whereby decisions are for the most part driven by business goals
underlying the creation of economic value. Managers of SOMNCs are therefore more constraint in their internationalization decisions than their private sector counterparts. The decision to internationalize, the selection of the country in which to invest, and the choice of entry mode are all influenced by political objectives of the home country (Cuervo-Cazurra et al., 2014).
2.4 MNCs and the influence of the home country
The institutions of the MNCs’ home country influence the internationalization patterns of MNCs. The importance of the institutions is emphasized by the New Institutional Economics. This economic perspective focuses on the legal rules and social norms that underlie economic activity (Rutherford, 2001). It explains why a MNC from a developing market, such as Brazil, has to face different challenges compared to a MNC from a developed market, for instance the US.
MNCs from developed countries remain the major source of outward FDI today. However, outflows from developing country MNCs have significantly risen from a negligible amount in the early 1980s to $83 billion in 2004. It even reached a record level of $553 billion in 2014, which is 39% of global FDI outflows (UNCTAD, 2014).
Developing country MNCs use outward investments as a springboard to acquire strategic assets. These assets are necessary in order to compete against global rivals and to avoid the institutional and market constraints developing country MNCs face at home. This ‘springboard behavior’ is characterized by rapid internationalization whereby critical assets from mature MNCs are acquired and aggressive risk-taking measures are used (Luo & Tung, 2007). This behavior is radically different from the traditional MNC model whereby international expansion occurs at a relatively slow pace (Chang & Rhee, 2011).
Vermeulen and Barkema (2002) argued that huge foreign expansions in a short period of time could have negative consequences for the company’s performance. Experience that comes too fast can overwhelm the management, which leads to an inability to transform experience into meaningful learning. However, particularly developing country MNCs have been able to capture the positive associations of rapid internationalization during the last two decades. Chang & Rhee (2011) emphasized that rapid FDI expansion can be a valuable strategy for firms that internationalize late, especially MNCs from developing markets, which must internationalize rapidly to compete in global industries.
Developing country MNCs also differ from traditional MNCs in their location choice decisions. For instance, the signs of China’s expansion are becoming increasingly manifest in Sub-Saharan Africa. MNCs from China built a network of trade, aid, and investment links with close to 50 African countries (Zafar, 2007). MNCs from developed countries often consider these countries as undesirable due to its political climate or low economic potential.
2.5 Research gap
SOMNCs have become more powerful in the last two decades. SOMNCs expanded their global reach and their globalization patterns are therefore becoming more relevant in the IB literature. SOMNCs need to take political objectives into account that serve the national interests. Their international behavior and strategies are therefore different from POMNCs. Most research conducted so far has focused on internationalization patterns of MNCs in general (Vermeulen & Barkema, 2002; Rugman & Verbeke, 2004; Luo & Tung, 2007). However, hardly any research has been done on the internationalization patterns of SOMNCs. This thesis tries to fill in this gap by looking at the internationalization scale and scope of SOMNCs.
The influence of the home country on the international patterns of MNCs is often analyzed within the IB literature (Lu & Beamish, 2004; Lee, Lin & Tsui, 2009; Rugman et al., 2011; Estrin et al., 2012). However, there has been less focus on how internationalization patterns are influenced by characteristics of the corporations’ home country, especially in relation to the distinction between SOMNCs and POMNCs. This thesis aims to shed light on whether the developing (versus developed) market status influences the relationship between state-ownership and internationalization path.
3. THEORETICAL FRAMEWORK
3.1 State-ownership and the degree of internationalization!
SOMNCs are economic as well political actors who are owned by the state and hence subject to multiple, and often conflicting, political and economic pressures. As a consequence, agency problems arise between the SOMNCs’ managers and the government as owners (Fama & Jensen, 1983). These problems are influenced by the specific political economy and institutional arrangements. At the same time, private shareholders push SOMNCs to act more like profit oriented MNCs. These conflicts affect the performance of the SOMNCs and their ability to act in an international environment (Estrin et al., 2012).
In general, governments aim to satisfy social objectives, such as the creation of jobs within their home country (Putterman & Dong, 2000). Political actors are thus particularly opposed to activities that contribute to the situation abroad rather than at home. Political pressures thus reduce the drive of SOMNCs to internationalize. However, there are also different reasons why internationalization of a SOMNC can be a strategy for development and a solution to market failures (Wu & Ramesh, 2014). Political pressures to attain development targets, such as access to natural resources, may offset pressures for social objectives, such as the creation of jobs. Even so, POMNCs are in most cases more efficient than SOMNCs (Boardman & Vining, 1989; Goldeng et al., 2008), and therefore the ones most able to compete in a global market. SOMNCs are thus unlikely to be favored instruments for the state to internationalize the economy. The pursuit of social objectives will therefore, on balance, not be offset by the governments’ objective of development through trade (Estrin, et al., 2012). That is why SOMNCs have a reduced drive to internationalize compared to POMNCs. Corporations who receive great institutional support face higher bureaucratic hindrances and political intervention (Luo & Tung, 2007). Risk-taking behavior and investment-strategies may vary significantly between SOMNCs and POMNCs. SOMNCs usually have less discretionary power in certain internationalization decisions (such as choice of foreign location) than POMNCs. Decisions made by SOMNCs might therefore be sub-optimal, with misalignments between optimal strategic options and actual choices under governmental influences. This suggests a fundamental difference whereby the SOMNCs are influenced by the political landscape and therefore restricted in their options for international expansion.
Restrictions from the host country can also reduce the international expansion options of SOMNCs. Governments around the world seek to protect their national interests and domestic industries. Restrictions on inward FDI still exist in different forms that discriminate against foreign investing firms in general (Cui & Jiang, 2012). According to Globerman and Shapiro (2009), the US government now restricts the FDI of Chinese SOMNCs because they are owned by the state. The authors emphasized that Chinese SOMNCs are motivated by non-commercial objectives that might threat the hosts’ national security. SOMNCs are therefore more likely to face hostility in their FDI compared to POMNCs. Host governments might block SOMNCs when they bid for strategic assets, such as infrastructure, utilities or natural resources. One notable example is that the Committee on Foreign Investment in the United States intended to block a deal between China’s Huawei Technologies and 3Com. The US Defense Department used the telecommunications equipment maker’s technology and there
were concerns that Huawei would gain access to this if the deal went through. The Chinese government does not own Huawei. However, the US House Intelligence Committee still argues that Huawei cannot be trusted to be free of foreign state influence (Financial Times, 2014).
At last, decision makers in SOMNCs may have incentive schemes and highly bureaucratic decision-making processes that discourage the taking of risk (Estrin et al., 2012). The internationalization process is however a high-risk activity. Thus, the risk-averse behavior of SOMNCs can reinforce the negative effect of the lack of capabilities on the scope of internationalization strategies, which decreases the chance on international expansion. All things considered, it is expected that state-ownership negatively affects the level of multinationality. This results in the following hypothesis:
Hypothesis 1: State-ownership is negatively related to firm’s level of multinationality.
This thesis is partly a response to calls for more research into the regional- as opposed to global-level of international business operations (Rugman & Verbeke, 2004; Fastoso & Whitelock, 2010; Rugman, Oh & Dominic, 2012). It does not only look at SOMNCs’ level of multinationality but also examines the scope of internationalization by looking at the distribution of sales across different regions.
Rugman and Verbeke (2004) investigated the key drivers in the globalization process, namely the MNCs that drive this process. The scholars examined geographic sales data of the world’s largest MNCs and concluded that very few are successful globally. In their research they showed that of the 365 with data, only nine MNCs were truly ‘global’, defined as having sales of 20% or more in each of the three regions of the ‘triad’ (i.e., North American Free Trade Association, the European Union or Asia), but less than 50% in any region. An example of a global MNC is Coca-Cola. Out of the 365 firms, 11 MNCs were ‘host region oriented’ (with 50% or more of their sales in a triad region that is not their home region) and 25 MNCs ‘bi-regional’ (with at least 20% of their sales in each of the two regions, but less than 50% in any one region). In all, 320 MNCs (80.3%) were home region oriented. Those corporations had at least half of its sales in their home region. As a consequence, Rugman & Verbeke (2004) questioned the academic focus on globalization and called for future research on the regional rather than global level of IB strategy.
The work of Rugman and Verbeke (2004) has not remained uncontested. The paper of Dunning et al. (2007) showed that much of the explanation for the regional concentration of
FDI and MNC activity reflects that of trade and the gross domestic product of the nations
concerned, rather than a specific strategy on the part of the investing corporation. However,
the authors were more than willing to concede the trend towards an increase in regional trading activities. Conversely, Osegowitch and Sammartino (2008) retested the data of Rugman and Verbeke (2004) using different schema and found that the original results were far from robust. The analyses of Osegowitch and Sammartino (2008) showed a significant share of firms attaining bi-regional or global status. The authors further argued that large MNCs increasingly are extending their sales beyond the home region.
Rugman et al. (2012) argued that most MNCs are not able to operate beyond their home region because they are not able to transfer national or home competiveness to foreign regions. This is due to their firm specific advantages such as firm size, technological knowhow, marketing capability, managerial capability and financial capability. Country specific factors such as physical distance, cultural distance, institutional distance and economic distance also play a role. The home region of MNCs may be particularly different from all other regions where a corporation operates. An example of a well-known global brand is Levi Strauss blue jeans, which is advertised based on strongly different brand positioning and pricing strategies in Europe vis-à-vis the US home market (Fastoso & Whitelock, 2010). This example supports the relevance of strong firm specific advantages (FSAs) for global success.
For POMNCs we argue that it is relatively less difficult to deal with global challenges (e.g., competition or difference in consumer behavior) since they only have to focus on their costs and benefits or building FSAs necessary to compete. On the other hand, SOMNCs have a strong domestic focus since they need to take political objectives into account that serve the home country’s needs (Maskus & Reichman, 2004; Cuervo-Cazurra et al., 2014). That is why it is expected to be even more difficult for SOMNCs to achieve a global distribution of sales than for POMNCs.
As a MNC expands beyond its domestic market, it is likely to enact a greater diversity of customers, competitors, cultures and regulations (Hofstede, 1994; Sanders & Carpenter, 1998). Rugman and Verbeke (2004) emphasized that widespread geographic diversification may have managerial pitfalls similar to those that arise with product diversification. MNCs must therefore take into account the costs and risks of inter-regional ‘distance’ and the liability of inter-regional foreignness (Rugman, Oh & Dominic, 2012). SOMNCs are risk-averse compared to POMNCs. It is therefore expected to be more likely that SOMNCs are even more region-oriented than POMNCs.
Strong firm specific advantages are required for a MNC in order to become competitive on a global scale (Fastoso & Whitelock, 2010). Firms need to develop superior products or very low cost products, which can be readily accepted by global customers. However, such products rarely exist because competition, technology and customer taste changes rapidly (Rugman, Oh & Dominic, 2012). SOMNCs often face less competition in their home country compared to POMNCs, for instance due to subsidies (Wolf, 2009). Lack of competition reduces the opportunity for SOMNCs to learn and to engage in competitive environments. SOMNCs are therefore less likely to expand to foreign regions since they lack suitable capabilities that are necessary to compete in the global economy (Estrin et al., 2012). Hence:
Hypothesis 2: State-ownership is negatively related to firm’s global (versus regional) focus.
3.2 Developing versus developed market MNCs
!
The international activities of MNCs have been intensively studied over the last decades, but less is known about the internationalization patterns of SOMNCs, and especially the distinction between SOMNCs from developed and developing countries. Knutsen et al. (2011) examined whether state-ownership systematically affects the relation between host country institutions and FDI. The authors found that SOMNCs invest relatively more than POMNCs in countries with high level of corruption and weak rule of law. However, there is a lack of information concerning the influence of the home country on the FDI of SOMNCs. This thesis therefore investigates whether the development of the SOMNCs’ home country influences the degree of internationalization.
SOMNCs may enjoy an advantage over POMNCs since they may have access to financial resources provided by the government (Wolf 2009; Estrin et al., 2012; Wu & Ramesh, 2014). These critical resources are important for the international expansion possibilities of SOMNCs. The extent to which SOMNCs are able to make use of this unfair advantage depends on the ability of the government to provide this important support. For SOMNCs, this can include access to subsidies and bank loans, beneficial regulations, or preferential access to protected domestic markets to generate big cash flows that can be used for international expansion (Estrin et al., 2012; Cuervo-Cazurra et al., 2014; Li et al., 2014).
The extent to which a government is able to support a SOMNC depends on the resources it has at its disposal. In line with this statement, Estrin et al. (2012) argued that a government is in a much stronger position to use SOMNCs when it has high financial
reserves. Governments that have strong currencies and accumulated reserves, as a result of past trade, are more able to use their domestic resources to acquire assets abroad. For those governments, acquisition targets appear relatively cheap, which increases the options to expand abroad.
Williamson (2000) explained the important role of the institutional environment for corporations. Institutions can be described as the formal rules of the game, such as constitutions, laws, property rights, etc. Developed countries are often characterized by stronger institutions with a well-developed financial market (Knight, 2000). A well-developed financial market means that more external agents evaluate the performance of a corporation. Moreover, governance arrangements are likely to be more effective and the competition in the market will increase. This pressures SOMNCs to become more efficient and to develop resources in order to compete in a competitive market (Estrin et al., 2012). For reasons described by Dunning (2000) (i.e., market seeking, resource seeking, efficiency seeking, and strategic asset seeking), SOMNCs are likely to internationalize in order to improve their competitiveness. Hence, the development of the SOMNCs home country, and thus the financial market, is likely to reduce the negative relationship between state-ownership and the degree of internationalization. Accordingly, this results in the following hypothesis:
Hypothesis 3: The developing (versus developed) market status of the home country negatively moderates the relationship hypothesized in H1 and H2.
The previously discussed hypotheses result in the conceptual model as shown in figure 1. Figure 1: Conceptual model
(-) (-)
State-ownership Internationalization path 6 Scale (H1)
6 Scope (H2)
Developing vs. developed countries (H3)
4. METHODS
4.1 Sample and data collectionThis study uses a cross-sectional research design to examine the effect of state-ownership on the internationalization path. Both the scale and scope of the internationalization patterns of SOMNCs and POMNCs are investigated. This study also examines the moderating effect of the developing (versus developed) market status of the home country on the relationship between state-ownership and internationalization path.
The sample for this study is based on the Fortune Global 500 corporations as listed in 2014. This is an annual ranking, published by Fortune magazine, of the top 500 corporations worldwide as measured by revenue. Together they account for over 90% of the world’s stock of foreign direct investment. The majority of the world’s largest corporations are MNCs, which means that they produce and/or distribute products and/or services across national borders (Rugman & Verbeke, 2004). Yet, the internationalization paths of the world’s largest 500 MNCs, of which 103 are state-owned, are different. Moreover, the home countries vary, just as the industries in which the MNCs operate. The use of this sample therefore offers great insights in the relationship of state-ownership on internationalization paths.
The Fortune Global 500 corporations stem from 38 different countries. Most originate from the United States (25.6%), followed by China (19.0%), Japan (11.4%) and France (6.2%). The government of China owns more than half (54.4%) of the Fortune Global 500 corporations that are owned by the state. Table 1 illustrates the 10 leading home countries of the Fortune Global 500, together with the amount of private-owned and state-owned corporations. The Chinese state owns more than half (54,4%) of all the SOMNCs listed on the Fortune Global 500 list. Data on state- versus private-ownership is missing for 24 corporations. Table 1 therefore excludes those companies that do not have ownership information.
&
Table 1: Distribution of corporation's home country and type of ownership
FG 500 POMNCs* SOMNCs*
Country Frequency % Frequency % Frequency %
U.S. 128 25,6 120 32,2 6 5,8 China 95 19,0 23 6,2 56 54,4 Japan 57 11,4 54 14,5 2 1,9 France 31 6,2 26 7,0 5 4,9 Britain 28 5,6 27 7,2 1 1,0 Germany 28 5,6 24 6,4 3 2,9 South Korea 17 3,4 14 3,8 2 1,9 Switzerland 13 2,6 12 3,2 1 1,0 Netherlands 12 2,4 9 2,4 2 1,9 Canada 10 2,0 9 2,4 1 1,0 Others 81 16,2 55 14,7 24 23,3 Total 500 100,0 373 100,0 103 100,0
*Data on state- versus private-ownership is missing for 24 corporations (N=476)
This study uses the Bureau van Dijk’s Orbis database and the company’s annual reports of 2013 as a primary data source to gather information on the Fortune Global 500. The use of Orbis to gather firm-level data is an appropriate method since it is one of the most comprehensive databases containing detailed information about many public and private companies (De Jong & Van Houten, 2014).
The firm specific data is also obtained from companies’ annual report and information generated from Internet searches. Some Asian companies don’t have English written annual reports. The sample is therefore limited to companies with available information, arriving at a final sample consisting of 333 MNCs.
4.2 Measures
4.2.1 Dependent variables
The dependent variable in this study is the level of internationalization, which is measured by looking at both scale and scope. To assess the scale of internationalization, the share of foreign sales in total sales is used. The scope of internationalization is also examined, which is measured by the geographical dispersion of total sales. Rugman and Verbeke (2004) also used this method by looking at the relative sales in host triad regions, vis-à-vis the home triad region. This study follows Rugman and Verbeke (2004), but makes two modifications. First, it will not only focus on the triad regions, like North America, the EU, and Asia, but
also examines South and Central America, Oceania and other regions. Second, this thesis focuses on the globalization patterns of MNCs by looking at both scale and scope of internationalization. The distinction of scale and scope offers greater insights in the actual internationalization paths. For example, a corporation is considered to be international when it has 90% foreign sales. But all of these foreign sales could take place in one country within the same region, which means that the MNC is not global but regionally based. Therefore, both the scale and scope are examined to assess the degree of internationalization.
4.2.2 Independent variable
The independent variable is state-ownership. This study is concerned with entities where the state has a minority of majority stake. State-ownership is therefore operationalized as enterprises with more than 10% state ownership. The international Monetary Fund (2001) also used this definition. This thesis uses dummy variables to indicate if it is a SOMNC, coded as 1, or a POMNC, coded as 0.
4.2.3 Moderating variable
This thesis further identifies the relationship between state-ownership and the internationalization path by looking at the influence of the development of the home country. Different lists such as International monetary Fund’s World Economic Outlook Report (2012), MSCI list (2015) and Dow-Jones list (2011) are used to determine if the home country of a corporation belongs to a developed or developing country. These lists measure development by looking at statistical indexes such as GDP per capita, rate of literacy and life expectancy. There are 23 countries that are commonly recognized as developed countries: Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Ireland, Israel, Italy, Japan, Netherlands, New Zealand, Norway, Portugal, Singapore, Spain, Sweden, Switzerland, United Kingdom and United States. Those countries other than these 23 countries are in this study considered as developing countries. Deng and Yang (2015) also used this qualification method.
4.2.3 Control variables
The influence of state-ownership on the degree of internationalization is controlled by four different variables. The first variable that is controlled for is firm size, which is a commonly used variable due to its impact on the internationalization of a MNC (Knight,
1996). Large MNCs usually have more financial, human, and tangible resources than small firms. Large MNCs are typically more capable of exploiting economies of scale and scope and consequently achieve better firm performance (Chao & Kumar, 2010). On the other hand, small firms can be more flexible and open to change (Knight, 1996). To determine the size of a corporation the logarithm of the total number of employees is used as well as the total turnover in millions of US dollar.
The age of a firm can influence the internationalization path and is therefore included as the second control variable. The age is measured as the number of years since the firm was founded until the year of reference for this thesis (2015). Older MNCs might have a higher degree of internationalization since they had more time to develop the required resources and experience to internationalize. However, this does not adequately explain why new small MNCs, such as “born globals”, internationalize at a rapid pace (Brush, 2012).
The third control variable is firm performance. Well-performing MNCs are able to take risky decisions and often have the required resources and capabilities to internationalize (Hitt, Hoskisson & Kim, 1997). This thesis operationalizes firm performance as return on assets at the end of 2013 financial year.
The type of industry influences the expanding motives and the degree of internationalization. For instance, SOMNCs involved in extractive industries are driven by a wish to secure mineral or strategic energy resources for the home country (The Economist, 2011). The UNCTAD (2014) report showed that these types of SOMNCs are increasingly active in cross-border acquisitions. Industry effect is therefore used as the last control
variable. This thesis uses the SIC codes of the Fortune Global 500 companies. The Orbis database is used for this purpose. The types of industries are divided into 5 categories, which were then coded into dummy variables taking on the value of 1 if the corresponding company belongs to the industry considered and 0 otherwise:
1. Mining, utilities and construction
2. Automotive, machinery and transport 3. Professional and information services 4. Wholesale and retail
5. Manufacturing and other services
4.3 Statistical analysis and results
Table 2 presents the descriptive statistics of the dependent, independent, and control variables. Multicollinearity can exist when there are high levels of correlation between the variables, which means a value above 0.7. This could indicate the presence of problematic constructs (Pallant, 2005; Field, 2009). As expected, table 2 shows that the internationalization scale and scope are strongly related. This creates no harm since both operationalizations for the path of internationalization are used in different models. All other variables were retained as they have values below 0.7.
The descriptive statistics (table 2) indicate that the average Fortune Global 500 firm is 59.28 years old and employs 120,752 workers. The firms have an average annual turnover of 61,028.95 millions of US dollar and a return on assets of 4.67%. When looking at the firm’s home countries it shows that 31.00% of the Fortune Global 500 stem from developing countries. 21.64% of the surveyed firms are owned by the state. Most firms are active in wholesale and retail (27.00%) followed by mining, utilities and construction (26.20%).
The internationalization path is both measured in scale and scope. Data on the scale of internationalization (i.e. foreign sales to total sales) is available for 246 firms. The average level of internationalization of these firms is 46.83%. This suggests that the average Fortune Global 500 firm is quite international, with almost 47% of their total sales outside their home country. The sample is therefore suitable for this thesis. Data on the geographical dispersion of foreign sales is retrieved for 171 firms to examine the scope of internationalization. The average scope of internationalization is 0.39. This indicates that 39% of the sales of an average Fortune Global 500 firm is made in a region that is not its home region.
This thesis uses a hierarchical regression analysis to test if the scale and scope of internationalization of MNCs is affected by state-ownership. To test the impact of the moderator variable (home country development), interaction terms are used by computing the product between the independent and moderating variable. This is done after standardizing the variables. The first step in the regression analysis is to introduce the control variables so that their effect on the dependent variable can be observed. Second, the independent variable and the interaction terms are added one at a time. This makes it possible to analyze their additional explanatory power. Accordingly,four models are carried out for both the scale and scope of internationalization. The first model only runs the control variables (age, industry, size and performance). Model 2a tests hypothesis 1 and model 2b tests hypothesis 2. The
influence of the development of the home country is illustrated in model 3a and 3b. Model 4a and 4b assess the third moderating hypothesis.
Table 3a and 3b illustrate the results of the hierarchical regression analysis for respectively the scale and scope of internationalization. The Beta standardized coefficient indicates if the variable included in the model contributes to the prediction of the dependent variable. The significance value indicates if the variable is making a significant unique contribution to the prediction of the dependent variable. At last, the R2 measures the goodness of fit and explains how much of the total variance in the dependent variable is uniquely explained by the explanatory variable (Pallant, 2005).
The results of the hierarchical regression show that not all the included factors have a significant value below the 0.05. This means that not all the variables have additional power. Hypothesis 1 states that state-ownership is negatively related to firm’s level of multinationality. The coefficient of state-ownership is significant (b=-0.163, p=0.033) and negatively related to scale of internationalization. Hypothesis 1 is therefore supported. The R2 improves from 0.079 in model 1a to 0.102 in model 2a. This indicates that the second model, with the independent variable, fits the data better and explains 10.2% of the variance. Hypothesis 2 states that state-ownership is negatively related to firm’s global (versus regional) focus. Model 2b indicates that state-ownership negatively relates to the scope of internationalization. However, the contribution of this variable is insignificant (b=-0.166,
p=0.107). Hypothesis 2 is therefore rejected. After the state-ownership variable is included,
the model as a whole improved and explains 9.7% of the variance of the scope of internationalization. The results in Model 3a and 3b indicate the negative relationship between the developing (versus developed) market status of the home country and the scale and scope of internationalization. The coefficient of the developing market status is significant for the scale of internationalization (b=-0.201, p=0.015), but insignificant for the scope (b=-0.133, p=0.188). Model 4a and 4b indicate the results of the interaction between state-ownership and home country market status on the scale and scope of internationalization. Model 4a indicates that the interaction is negatively related to the scale of internationalization, as expected. The contribution is insignificant at a 95% confidence level (b=-0.139, p=0.079). However, the impact of the interaction is significant at a 90% confidence level. After the interaction variable is included, the model as a whole explains 14.5% of the variance. Model 4b illustrates that the interaction has no influence on the scope of internationalization (b=0.000, p=0.997). The fit between the model and the data remains at 11.0% and did not improved after the interaction was included. Thus hypothesis 3 is partly
supported for the scale of internationalization and rejected for the scope of internationalization.
As for the control variables, no support is found for age, firm size (measured by turnover as well as amount of employees), and firm performance. These results suggest that company characteristics and resources do not influence the degree of internationalization of a firm. The automotive, machinery and transport industries have a significant effect on firm’s level of multinationality (b=0.181, p=0.034), in comparison to the wholesale and retail industries of reference. MNCs active in automotive, machinery and transport industries tend to internationalize more than other industries. MNCs in professional and information services are significantly less globalized than other industries. However, this can only be concluded at a 90% confidence level.
Table 2: Descriptive statistics: means, standard deviations and correlations
Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12
1. Ownership (state vs. private) 0.21 0.41
2. Internationalization scale 0.47 0.30 -.24**
3. Internationalization scope 0.39 0.27 -.23** .81**
4. Home country development 0.31 0.46 -.48** -.23** -.29**
5. Firm age 59.28 52.00 -.15** .15* .14 -.28**
6. Mining, utilities and construction 0.26 0.44 .26** -.08 -.11 .33** -.12**
7. Automotive, machinery and transport 0.18 0.39 -.09 .20** .18* -.05 .03 -.28**
8. Professional and inform. services 0.13 0.33 .02 -.16* -.21** -.01 .07 -.23** -.18**
9. Wholesale and retail 0.27 0.44 -.15** -.07 .03 -.23** .03 -.36** -.29** -.23**
10. Manufacturing and other services 0.16 0.37 -.05 .09 .04 -.06 .01 -.26** -.21** -.17** -.27**
11. Firm size: annual turnover 61,028.95 58,050.75 .03 .11 0.07 -.04 .02 .12** .00 -.09* -.01 -.05
12. Firm size: employees 120,752 161,694 .06 .04 .03 .01 .02 -.09 .05 -.01 .05 -.01 .47**
13. Firm performance 4.67 6.49 -.11* .11 .11 -.04 -.05 -.05 .02 -.11* .11* .01 .16** .11*
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Table 2: Results of Hierarchical Regression analysis for scale of internationalization
Controls H1 H3 H3
Dependent variable: Scale of internationalization Model 1a Model 2a Model 3a Model 4a
Control variables Beta Sig. Beta Sig. Beta Sig. Beta Sig.
Firm age .116 .106 .096 .178 .057 .433 .067 .354
Firm size: annual turnover .094 .286 .087 .320 .082 .340 .091 .291
Firm size: employees -.049 .568 -.038 .654 -.041 .629 -.032 .702
Firm Performance .070 .331 .062 .388 .054 .449 .060 .392
1. Mining, utilities and construction -.003 .970 .059 .523 .081 .373 .074 .417
2. Automotive, machinery and transport .181 .034* .187 .027* .203 .016* .193 .021*
3. Professional and information services -.064 .412 -.046 .558 -.012 .879 -.008 .921
5. Manufacturing and other services .127 .118 .130 .107 .151 .059† .149 .061†
Independent variable
State-ownership -.163 .033* -.077 .348 -.020 .824
Moderator variable
Home country (developing vs. developed) -.201 .015* -.182 .028*
Interaction terms
State-ownership x Home country -.139 .079†
N 196 196 196 196 R2 .079 .102 .130 .145 Adjusted R2 .040 .058 .083 .093 Change in R2 .079 .022 .028 .014 Sig. F Change .046 .033 .015 .079 p†<0.10; *p<0.05; **p<0.01; ***p<0.001.
Table 3: Results of Hierarchical Regression analysis for scope of internationalization
Controls H2 H3 H3
Dependent variable: Scope of internationalization Model 1b Model 2b Model 3b Model 4b
Control variables Beta Sig. Beta Sig. Beta Sig. Beta Sig.
Firm age .084 .340 .100 .253 .078 .380 .078 .382
Firm size: annual turnover .022 .850 -.007 .950 -.008 .947 -.008 .948
Firm size: employees -.006 .959 .005 .964 .006 .957 .006 .957
Firm Performance .075 .389 .063 .471 .055 .523 .055 .526
1. Mining, utilities and construction -.022 .839 .077 .538 .099 .429 .099 .432
2. Automotive, machinery and transport .124 .220 .124 .219 .135 .179 .135 .182
3. Professional and information services -.175 .061† -.171 .065† -.138 .151 -.138 .153
5. Manufacturing and other services .068 .481 .057 .553 .058 .548 .058 .551
Independent variable
State-ownership -.166 .107 -.116 .287 -.116 .289
Moderator variable
Home country (developing vs. developed) -.133 .188 -.133 .217
Interaction terms
State-ownership x Home country .000 .997
N 137 137 137 137 R2 .079 .097 .110 .110 Adjusted R2 .021 .033 .039 .031 Change in R2 .079 .019 .012 .000 Sig. F Change .218 .107 .188 .997 p†<0.10; *p<0.05; **p<0.01; ***p<0.001.
Examination of the interaction plot in figure 2 shows that POMNCs have a high level of multinationality compared to SOMNCs. This difference is stronger for developing countries than for developed countries, indicating the effect of the moderation.
Figure 2: The relationship between state-ownership and internationalization scale
5. DISCUSSION
The empirical findings and conceptual framework of this study shed novel insights on the phenomenon of globalization of SOMNCs. While previous research has concentrated and found support mainly for factors explaining globalization patterns of POMNCs, this study provides clarity on the internationalization path of SOMNCs. More specifically, the main results of this research show that state-ownership negatively influences the level of multinationality, also described as the scale of internationalization. The results do not provide support for a negative relationship between state-ownership and the firm’s global (versus regional) focus, described as the scope of internationalization. The analysis of the market status of the home country and its moderating effect on the key relationship observed deserves great attention. When the developing (versus developed) market status of the home country is used a significant negative moderating effect appears on the negative relationship between state-ownership and level of multinationality. This suggests that SOMNCs have a relatively lower level of multinationality compared to POMNCs, especially when the SOMNCs’ home country has a developing market status. However, there is no significant moderating effect of
0 0,05 0,1 0,15 0,2 0,25 0,3 0,35 0,4 0,45 0,5 POMNCs SOMNCs S cal e of i n te rn ati on al iz ati on Developed countries Developing countries
