Level of multinationality and Firm Performance: The
moderating role of different ownership structures
Rosalie Lathouwers
10886338
30 August 2015
MSc Business administration: International Management University of Amsterdam
Final Version Master Thesis Supervisor: Dr. Niccolò Pisani
STATE OF ORIGINALITY
This document is written by Student Rosalie Lathouwers, 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
TABLE OF CONTENTS
1. INTRODUCTION...5
2. LITERATURE REVIEW: MULTINATIONAL ENTERPRISE, SCALE OF INTERNATIONALIZATION, SCOPE OF INTERNATIONALIZATION, LEVEL OF MULTINATIONALITY AND FIRM PERFORMANCE, AND DIFFERENT OWNERSHIP STRUCTURES………...8
2.1 MULTINATIONAL ENTERPRISE………...8
2.2 LEVEL OF MULTINATIONALITY………...9
2.2.1 Scale of internationalization………...9
2.2.2 Scope of internationalization………....12
2.3 LEVEL OF MULTINATIONALITY AND FIRM PERFORMANCE……...16
2.4 OWNERSHIP STURCTURES AND INTERNATIONALIZATION...19
2.4.1 Family versus non-family owned firms………...….19
2.4.2 State versus private owned firms………...…20
2.5 RESEARCH GAP………...…21
3. THEORETICAL FRAMEWORK………...22
3.1 LEVEL OF MULTINATIONALITY AND FIRM PERFORMANCE...22
3.2 DIFFERENT OWNERSHIP STRUCTURES………...25
4. METHODS………...28
4.1 SAMPLE AND DATA COLLECTION………...……28
4.2 MEASURES………...28
4.2.1 Dependent variables………...28
4.2.2 Independent variable………...…29
4.2.3 Moderating variables………...29
4.2.4 Control variables………...29
4.3 STATICAL ANALYSIS AND RESULTS………...30
5. DISCUSSION………...38
5.1 ACEDEMIC RELEVANCE………...39
5.2 MANAGERIAL IMPLICATIONS………...40
5.3 LIMITATIONS AND SUGGESTIONS FOR FUTURE RESEARCH…………...41
6. CONCLUSION………...42
ACKNOWLEDGEMENT………...45
REFERENCES………...46
LIST OF FIGURES Figure 1. Conceptual model………...27
LIST OF TABLES Table 1. Descriptive statistics; means, standard deviation and correlation………...35
Table 2. Regression results (OLS), Scale of internationalization and firm performance………...36
ABSTRACT
Many scholars have investigated the relationship between level of multinationality and firm performance. The literature on this topic has shown positive, negative, U-shaped, S-shaped or even no relationship between both variables. The first test in this thesis investigated the S-shaped relationship between the scale and scope of internationalization and firm performance measured in the average return on assets. This relationship has already been investigated in previous research but this study will provide a greater degree of novelty by testing the moderation effects of different ownership structures in the second test. The moderation effects of different ownership structures contribute to the literature since no previous study has done that.
This study used a sample of the Fortune Global 500 companies, listed in 2014. The findings do no support a S-shaped relationship between the level of multinationality and firm performance. Also there is no evidence found for the moderation effects of different ownership structures, family versus non- family owned firm and state versus privately-owned firms on the relationship between level of multinationality and firm performance.
Keywords: Level of multinationality, Scale of internationalization, degree of
1. INTRODUCTION
Globalization is a common definition in today’s world, which influences social,
cultural, political, economic, and other aspects. It is defined as the growing interconnectedness of the world through cross-border flows of people, information and capital (Held, McGrew, Goldblatt, and Perraton, 1999). Due to the fact that multinational enterprises (MNE’s) support economic interdependence among national markets and have
played a central role in the global economic, social and politic al changes, MNE’s are considered to be the key drivers of globalization (Held & Mcgrew, 2000). In the past several decades, the possible effects of globalization have become more visible to all kinds of firms since they create new opportunities and challenges for their business. These opportunities, such as access to new markets that were previously closed due to different barriers and the challenge to become more efficient by extending the geographic reach have contributed to the fact that firms are becoming more internationally orientated and expand their business abroad.
The literature underlines the importance of international diversification since it is based on the exploitation of foreign market opportunities and imperfections through internalization. Due to the international changes as mentioned above, managers had to think of international opportunities to gain competitive advantages. Each strategy of international expansion of a firm provides insight into the level of multinationality, which refers to the scale and the scope of internationalization and variety among the companies. The scale of internationalization refers to the degree of internationalization, which relates to the extent of a firm’s international operations. On the other hand, the scope refers to the geographical
dispersion of a firm’s international activities. To show the effect of the scale and the scope of internationalization many international business scholars have explored the relationship
positive, negative, U-shaped, S-shaped or even no relationship between both variables. The positive relationship on the one hand, is based on a variety of reasons, such as the possibility of exploiting scale of economies (Grant, 1987; Porter, 1985), better access to resources (Contractor, Kundu, and Hsu, 2003) and the learning opportunities (K im, Hwang, and Burgers, 1993). O n the other hand, the negative relationship is based on the increased costs, human resources and complicated systems (Geringer, Beamish, and DaCosta, 1989). Other scholars suggest a non- linear relationship since the first period consists of an expansionary growth where benefits exceed costs followed by an increasing in costs, which will exceed the benefits (Gomes & Ramaswamy, 1999). Thus, the results of the many studies that address the relationship between the level of multinationality and the firm’s performance provide mixed conclusions.
Until now, research mainly focused on the relationships with their many outcomes. To fill the gap in the existing literature, this thesis will shed more light on the S-shaped relationship between both variables and the moderation effects of different ownership structures of the firms. No previous studies have investigated the possible moderation effects of different ownership structures on the relationship between level of multinationality and firm performance. It can be expected that differences in ownership structures of the firm do matter to the degree of internationalization and firm performance. In this thesis, a distinction in ownership structures is made between family versus non- family owned firms and state versus private firms.
According to Dunning (2001), the Fortune Global 500 companies are very important because these firms engage in the highest percentage of foreign direct investment (FDI) around the world. Given the fact that multinationals exist because of their business outside of their home country, the scale and scope of internatio nalization of these firms are easier to identify. Due to above- mentioned reasons this study focuses on the 500 largest multinational
enterprises in the world, listed in the Fortune Global 500 companies of 2014. By focussing on these firms a better understanding of the internationalization pattern of large multinationals and their effect on firm performance can be gained.
Although recent studies showed evidence for a horizontal S-shaped relationship (Lu & Beamish, 2001; Contractor, K undu, and Hsu,2003), the findings of this study do no support a S-shaped relationship between the level of multinationality and firm performance. Also there is no evidence found for the moderation effects of different ownership structures, family versus non-family owned firm and state versus privately-owned firms on the relationship between level of multinationality and firm performance.
This thesis is structured as follows: first, the relevant literature about the level of multinationality is provided. This consists of the scale and scope of internationalization and their effect on firm performance and the different ownership structures is discussed. The third chapter provides a theoretical framework with the development of the hypotheses and propositions. The fourth chapter contains the methodology section in which the different variables, the data collection, and the method of analysis used to perform this research are discussed. This chapter concludes by providing the results from the statistical analysis. The following chapter discusses the findings and mentions their implications and limitations. The final chapter, chapter 6, provides the overall conclusion of this research.
2. LITERATURE REVIEW: MULTINATIONAL ENTERPRISE, SCALE OF INTERNATIONALIZATION, SCOPE OF INTERNATIONALIZATION, LEVEL OF MULTINATIONALITY AND FIRM PERFORMANCE, AND DIFFERENT
OWNERSHIP STRUCTURES
2.1 Multinational enterprise and internationalization
An MNE can be defined as an enterprise that engages in foreign direct investment (FDI) and owns or controls activities in two or more different countries (Caves, 1996). The existence of an MNE is explained through the modern theory of FDI, which s uggest that an MNE originated as a response to imperfections in markets. Hymer (1960) as well as Vernon (1966) are two of the many scholars that emphasized the emergence of the multinational corporation and recognized the market imperfections. According to Hymer (1960), MNE’s possess firm specific advantages (FSA) and need to exploit them by undertaking foreign direct investment to gain monopolistic advantages. Vernon (1966) continues by stating that firms can also have country specific advantages (CSA) that can be embedded in the MNE and can lead to other benefits of foreign ownership. Rugman (1979, 1981) explains this further by emphasizing the importance of internalization of these advantages because it is based on exploiting foreign market opportunities and imperfections. The eclectic theory of Dunning (1979, 1995) then shows a combination of the firm specific advantages with the factors that already exist in the foreign country to create more potential benefits o f the international expansion.
As the above- mentioned scholars indicated, international expansion is important and is based on the capability of the firm to exploit its local advantages in foreign markets. Because of the possible benefits of international expansion, international diversification plays a key role in the strategic behavior of large firms (Hitt, Hoskisson, and K im, 1997). International diversification can be defined as “the expansion of a firm into different geographic locations or markets across the borders of global regions ” (Hitt et al., 1997). The
level of multinationality, which refers to the scale and scope of internationalization, provides insights into the internationalization path of firms.
2.2 Level of multinationality
2.2.1 Scale of internationalization
The internationalization path of firms is an interesting research subject in the international business studies and refers to the process in which the firms gradually increase their international involvement. The degree of internationalization or in other words the scale of internationalization relates to the extent of a firm’s international operations. Firmness of
scale can be assessed from two different points of view; first, the amount of turnover received from international markets (K night & Cavusgil, 1996, 2004; Moen, 2002) and second, to the number of markets into which the firm enters (Crick, 2009; K uivalainen, Sundqvist, and Servais, 2007). According to existing literature, the degree of internationalization of a firm has three components: attitudinal, performance, and structural. According to Perlmutter (1969), attitudinal refers to the international orientation of the top management of a firm. The second component performance refers to the activities overseas (Vernon, 1971), whereas the third component refers to the available o verseas resources (Stopford & Wells, 1972). Although there are many studies about the internationalization process, the literature does not provide specific theories that explain the optimal degree of multinationality of a firm’s operations.
According to Harris, Marr, and Spivey (1991) and Mitchell, Shaver, and Yeung (1993), international expansion could have potential benefits for MN E’s, including intelligence gathering, volume economies, improvement of products, operational flexibility and stability, tax arbitrage, and organizational advantages. The fundamental assumption is that diversification across national boundaries increases the stabilization of profits (Madura &
Whyte, 1990). To obtain high profits firms need to exploit their firm-specific assets, in particular intangible assets, in international markets (Buckley, 1988).
Besides the profit aspects there are also different motives for MNE’s to expand internationally. Scholars have identified four main motives for MNE’s to expand into foreign
markets. The first motive is market seeking or demand oriented foreign direct investment and relates to satisfy a particular foreign market or set of markets. The second motive relates to gain access to natural recourses and called resource seeking or supply oriented foreign direct investment. The third motive relates to create a more efficient portfolio of foreign and domestic assets by MNE’s and a useful division of labor, rationalized or efficiency seeking. The fourth motive is strategic asset seeking, which covers the existing ownership advantages.
The eclectic or OLI paradigm of Dunning (1993) is the dominant analytic framework for testable economic theories thatare related to the foreign activities of MNE’s and foreign direct investment. The interaction of ownership advantages, location advantages and international advantages determine the geographical composition of internationalization undertaken by MNE’s. Ownership advantages consist out of firms’ superior assets and skills to engage in international expansion and to compete with firms who operate in their own markets. Superior assets are based on the size and experience of the multinational, whereas skills derive from differentiated products. In order to be successful in international expansion and to compete with host country firms, asset power is needed. According to Hood and Young (1979) resources are essential to achieve economies of scale, obtaining patents and contracts, and to absorb the high cost of marketing. To overcome these cost s the size of the firm does matter (Buckley & Casson 1976; Kumar 1984). Empirical evidence has showed that the size of the firm has a positive effect on foreign direct investment ( Yu & Ito 1988; Terpstra & Yu, 1988). Besides the size of the firm, the age of the firm is also considered as an important firm characteristic, which also influences internationalization. Empirical
evidence showed that the age of the firm has a positive effect on foreign direct investment since older firms have a greater breadth of geographic expansion (Julian & Castrogiovanni, 1995).
Besides the importance of the size and the age of the firm, other studies suggested that founder attributes are critical in the decision to operate abroad (Bijmolt & Zwart, 1994). However, more recent studies suggest that the firm’s knowledge intensity (Autio, Sapienza, and Almeida, 2000) or access to networks (Blomstermo, Eriksson, and Lindstrand, 2004) can be seen as key recourses for internationalization.
Besides the ownership advantages, firms enter into more attractive markets because these countries offer location specific advantages and offer a unique resource endowments to obtain higher profits. These location specific advantages can be classified into networks, resources, institutional structures, and other country specific advantages (Dunning, 1993; Singh & K undu, 2002). According to Kogut (1991), these location advantages ensure that firms are willing to establish subsidiaries abroad and increase their competiveness in the home and host country. According to Dunning (1988), the decision on location is based on the following important factors: infrastructure, country risk factors, and government policy. Market potential and investment risk of a specific country have also been considered as important determiners of overseas investments (Weinstein 1977; Terpstra and Yu 1988).
However, in the past twenty years the interest in location as a determining factor for internationalization has changed. This led to firms focusing on the organization of cross-border operations (Dunning, 2009). The shift in environment, which relates to changes in economic environment including global interconnectedness of activity and increasing importance of knowledge, has ensured that the interaction between location advantages and ownership advantages changed and beca me more complex (Cantwell, 2009). Nowadays, subsidiaries are a part of international networks that are uniformly centralized and less
singularly hierarchical (Dunning & Lundan, 2008). Because of the formation of these international networks, the use of location is organized differently in order to ensure competitive advantages (Cantwell & Mudambi, 2005; Nachum & Zaheer, 2005). These international networks have increased the attention towards the connectedness between local knowledge creation and exchange in each node of the network (Cantwell, 1989). Combined with the increasing importance of absorptive capacity (Cohen & Levinthal, 1989), MNE’s have to choose locations because they generate suitable business relationships instead of siting activities in a munificent location (Cantwell, 2009).
Looking at the measurement of the scale of internationalization, latest research shows the measurement of the degree of internationalization by criticizing that the 25 percent cut-off ratio for exports is to low (K night et al., 2004). The requirement is often cited as 25 percent of the total turnover of a firm (Knight et al., 1996, 2004; Moen, 2002). Because of the low cut-off, small knowledge-intensive firms with high export sales deriving from one single country can thus be labeled with a high degree of internationalization (K uivalainen et al., 2007). Besides the financial measurement of the degree of internationalization, it can also be measured with psychological and attitudinal measurements (Sullivan, 1994).
2.2.3 Scope of internationalization
The motives and determining factors of internationalization were discussed in the previous paragraph. This paragraph focuses on the analysis of the scope of internationalization. The scope of internationalization refers to the geographical dispersion o f MNE’s activities. Ansoff (1957) explains the corporate and business-level choices, which consist of the firm’s product, vertical and geographic scope. The last mentioned choice is
considered to be an important strategic choice for international expansion and relates to the choice between regionalization and globalization. International changes in economic,
political, technological and social such as, higher mobility of products and people, liberalization of capital markets and relative global regulatory harmo nization have ensured that MNEs internationalization strategies have changed (Dunning, 2001, 2002; Gatignon & Kimberly, 2004). Adopting a new strategy had caused a change in the geographical spread of MNEs’ activities in light of the global-scale transformations. In the late 1990s the nature and extent of regional versus global activity by MNE’s became more visible due to the fact that
firms were required to identify the broad regions in which their sales and assets took place (Rugman & Verbeke, 2000, 2004). As a result, a debate arose between international business scholars about the foreign location choices of MNE’s.
The choice between regionalization and globalization depends on the functioning of an MN E and is subject to five issues (Rugman and Verbeke 2004). The first issue relates to the accessibility and attractiveness of a product and its consumers around the world. When a product is more accessible or attractive to consumer in a specific region, MNE’s operations
will be spread across the globe. The second issue relates to the transferability of firm specific advantages of the firms. Limitations in the transferability of the firm specific advantages ensure a lack of global market success. According to Rugman and Verbeke (2004), many MNEs have their firm’s specific assets, which are location bound and cannot be implemented worldwide. Due to this, it could be a possible explanation for the fact that most MNE ’s are operating on a regional basis within the triad (Rugman, 2005; Rugman and Verbeke, 2002). The third issue has to do with the lack of market performance across regions, which do not allow companies to benefit from the location specific advantages of the foreign market. The fourth issue is related to the different position o f an MN E in a foreign market in comparison to their home-based market. The fifth issue is related to the governance of the MNE and includes the presence of multiple environmental circumstances. Due to these issues, market
familiarity (geographic and cultural proximity) with the overseas target market is an important influence on the location choice of an MNE according to Davidson (1980).
The debate between the geographic dispersion of MNE’s activities consists of two groups with their own justifications for regional or global operations of large multinationals. One party of scholars argue that MNE’s activities have become increasingly regional during the last two decades. They base their findings on the distribution of sales of the Fortune global 500 companies in 2001 (Rugman, 2003, 2005; Rugman and Verbeke, 2004, 2007). The results showed that 380 MN E’s had 80% of their sales within the same region as where the headquarters were established. The results support the idea that most of the MNE’s sell and produce on a home region basis instead of operating globally. According to Rugman (2005), the home MNE’s in North America, Europe, and Asia, dominate in their home region
of the triad. The results also show an unevenly spread of sales across different regions, which emphasizes the existence of different market positions (Rugman and Verbeke, 2004). These different market positions ensure that firms need to be aware that each region requires their own competitive strategy.
According to Mansfield and Reichman (2003), the rising regional integration is partly attributed to governmental actions taken to achieve better bargaining power for the region compared to countries which are part of a different region. Policies that ensure lower trade and investment barriers will help firms within a specific region to have greater access to regional markets. Through this access, firms will be stimulated to develop a strategy that is regional based as demonstrate for example in the retail sector (Rugman and Girod, 2003) and automotive sector (Rugman and Collinson, 2004). Besides these above- mentioned supporting governmental actions, most MNE’s are simply not capable of deploying and exploiting their
firm specific advantages to achieve a balanced distribution of their sales in North America, Europe and Asia (Rugman and Verbeke, 2007). The main reason for this lies in the fact that
the additional costs of doing business in other regions than the home triad region are often much higher. According to Rugman and Verbeke (2007), many FSA’s need complementary FSA’s to operate successfully when penetrating a host region. Another explanation for
expanding intra-regionally is the implications of ‘distance’ that relates to administrat ive, cultural, geographic, and economic differences between regions and ensures higher transaction costs (Ghemawat, 2001,2005). Firms who are operating in various regions will have to face greater difficulties by successfully overcoming the environmental complexities (Qian and Li, 2002).
The other party of scholars underlines precisely the growing significance of global activities by means of four arguments (Bird and Stevens, 2003; Clarke and K nowles, 2003). First, they argue that globalization goes be yond economic events and trade. The second argument focuses on the weakness of the dependent variable “firms revenue’’ which is used in the research of Rugman and Verbeke (2001). This variable does not automatically contain MNE activities abroad (Clark et al., 2003, Bird et al., 2004). Third, they criticize the validity of the first party research and argue that country- level data is crucial (Dunning et al., 2007). The final argument refers to the regional categorization of first party research and criticized this distribution because in some cases countries are included in other regions (Bird et al., 2004).
However, recent study assumed that large MNE’s are neither purely global nor purely
regional but have a focus on their home region combined with strong national biases (Asmussen, 2009). According to this research, international expansion within regions would also ensure significant barriers, which include that regional integration would be less effective than believed in previous studies.
2.3 Level of multinationality and firm performance
The relationship between the level of multinationality and firm performance is a well-known research topic in international business literature. This relationship refers to the level of which a firm undertakes value-adding activities in different foreign markets and their effect on the performance of the firm. However, empirical evidence on the impact of a firms’
international expansion on its profitability is decidedly mixed (Tallman and Li, 1996; Hitt et al., 1997; Geringer, Tallman, and O lsen, 2000). The literature has shown positive, negative, U-Shaped, S-shaped or even no relationship between both variables. Possible explanations for these different outcomes are related to the incompleteness of the theory and the study of the relationship in different parts instead of studying the whole relationship.
According to Hennart (2007), the literature of the above- mentioned relationship has two main predictions deriving from the theory of portfolio diversification in finance, where Markowitz (1976) showed that an investment in assets whose returns were uncorrelated, could cause a possible reduction in the risk of a portfolio of securities. The first prediction relates to the relationship between MNE’s international d iversification and risk, where larger
number of countries reduce the risks of an equal level of profit (K im, Hwangan and Burgers, 1993). Besides the number of countries it was also possible to reduce risk if firms operate in a portfolio of countries that were not economically integrated (Shapiro, 1978). After focusing on the effect of international diversification on the possible risks for firms, the emphasis was placed on achieving higher profits. Therefore, according to Hennart (2007), the second prediction states that if a MNE is more internationally diversified the firm’s profitability will be greater.
Based on the second prediction, the positive relationship between the level of multinationality and firm performance is expected to be positive on the basis of the following arguments. De first argument for a positive relationship is the possibility of exploiting scale
economies (Grant, 1987; Porter, 1985). This allows firms to spread their fixed costs over a larger market, which results in an increase of their profitability (Hitt et al., 1997; Contractor et al., 2003). The second argument which underpinned a positive relationship has to do with the fact that international diversification allows firms to provide better and more flexible access to resources (Contractor et al., 2003). The third reason why internationally diversified firms should be more profitable refers to the expansion of opportunities to learn (K im et al., 1993; Hitt et al., 1997; Contractor et al., 2003). Because of the transaction cost/internalization (TCI) model the above- mentioned arguments can be used to evaluate these arguments or to come up with possible counterarguments. According to Hennart (2007), the first argument is not always valid because of exploitation of economies of scale. It is not always necessary to go abroad since a spread of fixed costs could also be achieved through a large market within the home country. The second argument that is supposed to achieve better access to resources can be refuted through the TCI model because it is only true in specific cases with high transaction costs (Williamson 1985, Hennart, 2001). The third argument can be refuted by means of the fact that learning from abroad is not a real common driver of international expansion (Hennart, 2007). Besides this, Hennart (2007) argues that firms can also learn from their headquarters or their own market instead of learning from operations abroad.
Instead of a positive relationship between the level of multinationality and firm performance, some scholars assume a negative relationship (Denis, Denis, and Yost, 2002; Click & Harrison, 2000) or a very weak relationship (Tallman & Li, 1996; Bodnar, Tang and Weintrop, 1997). The negative effects of multinationality are based on the increased costs, arising out of control and coordination, human resources, and complicated systems that are required to manage culturally various markets (Geringer et al., 1989). The increased costs may be limited by taking into account market familiarity in overseas target markets. Market familiarity provides relatively similar administrative mechanisms compared to geographically
distant locations (Davidson, 1983). Therefore, firms would be able to leverage their home country competencies in these locations more easily (Davidson, 1983; Papadopoulos and Denis, 1988).
Besides the positive or negative relationship, scholars claim there is a curvilinear relationship between the level of multinationality and firm performance. These scholars argue that in the initial phase of overseas expansion high levels of marginal performance could be achieved because of the higher efficiencies, which originated from the skill and resources embedded in the firm (Bartlett and Ghoshal, 1989). However, during the later stages of international expansion firms may have to deal with increased costs because of the coordination and control of managing a higher number of locations where operations are established, especially in different cultural environments (Gomes et al., 1999). Thus, the advantages of performance of multinationality will decrease when “internal governance costs exceed the benefits provided by the economies achieved and thus, the range of resources used and scope of governance exceeds managerial capabilities” (Hitt et al., 1997). Based on the findings of this U-shaped relationship, the idea arose that an S-curve also might exist (Lu & Beamish, 2001; Contractor, Kundu, and Hsu, 2003).
Besides the different argumentation of the relationship, there are also methodological issues in measuring the relationship between the level of multinationality and firm performance. According to Henart (2007) scholars measure the level of multinationality by dividing the foreign sales by the total sales ratio, but the problem is that there is no general agreement on how to measure the level of multinationality. The most problematic issue relates to the fact that measurements of the relationship neglect the differences between overseas markets (Q ian et al., 2008). Other comments on the measurement are related to the argumentation that the global configuration of a firm’s value chain is not included in the
measurement of this relationship (Amussen et al., 2007). Another methodological issue is the lack of appropriate control variables.
It can be concluded that there are many different assumptions about the relationship between the level of multinationality and a firm’s performance. An important factor has to do with the fact that the relationship depends on different features of the international expansion such as the pace, rhythm, and speed of the internationalization and the phase of internalization (Vermeulen and Barkema, 2002).
2.4 Ownership structures and internationalization
A firm’s governance structure includes the ownership and the board (Brunninge,
Nordqvist and Wiklund, 2007; Rediker & Seth, 1995). A firm’s governance structure forms an important factor when it comes to the firm’s capability to gain access to the resources required for internationalization (Melin, 1992; Tihanyi, Hoskisson, Johnson and Hitt, 2003). Besides the difference in gaining access to recourses also the motives of MNE’s for
internationalization can differ due to the difference in ownership structure.
2.4.1 Internationalization paths of family versus non-family owned firms
An important motive for international expansions of family firms refers to the altruistic behaviors of family members (Schulze, Lubatkin, Dino, and Buchholtz, 2001). Because of this setting, family firms are willing to take high risks if long-term success and an increase of the employment for family members can be achieved. Scholars showed mixed evidence concerning the factors that constrain or facilitate the internationalization of family owned firms (Fernández and N ieto, 2006; Kontinen and Ojala, 2010). Family ownership would stimulate internationalization because family members act as a good agent of their existing resources and possess attributes as long-term orientation, unique knowledge, and
commitment (Zahra, 2003). Other scholars argue that family firms show lower levels of internationalization to refrain from loss of their socio-emotional wealth (Gomez-Mejia, Makri, and K intana, 2010). Socio-emotional wealth is the “single most important feature of a family firm’s essence that separates it from other organizational forms” (Berrone, Cruz, and
Gomez-Meija, 2012).
Another explanation of a lower inclination of international expansion is the limited access to resources of family-controlled firms (Gomez-Mejia et al., 2010; Schulze et al., 2001; Sciascia, Mazzola and Astrachan, 2012). Scholars have noted that the lack of managerial resources (Graves & Thomas, 2008) and financial resources (Gallo & Pont, 1996) limited the international expansion of family firms. To gain access to resources that are unavailable within the family owned firms, it can be helpful for family firms to create external links or tactics with actors in the external environment in order to provide family firms in additional resources (Arregle, Hitt, Sirmon and Very, 2007; Bammens, Voordeckers and Van Gils, 2011). Besides the lack of essential recourses, Okoroafo (1999) noted that the lack of knowledge of international markets constrains the internationalization of family firms.
Besides the positive or negative relationship, scholars also found an inverted U-shaped relationship between family ownership and international intensity (Sciascia et al., 2012) or no difference between family and non-family businesses and internationalization practices (Cerrato & Piva, 2012; Carlos Pinho, 2007).
2.4.2 Internationalization paths of private versus state owned firms
A state-owned MNE is defined as “A legally independent firm with direct ownership by the state that has value-adding activities outside its home country” (Cuervo-Cazurra, Inkpen, Musacchio, and Ramaswamy, 2014). The literature shows that the internationalization behavior of state owned MNE’s differ from privately owned MN E’s. As
discussed in the chapter “Scale of internationalization’’, Dunning (2000) described four motives for MNE’s activities, which all have the aim to maximize the profit of the firm.
However, the motives of state-owned firms in general are focused on improving the welfare of the their citizens by responding to the market failures of the home country (Amighini, Rabellotti, and Sanfilippo, 2013). Two general motives are at the basis of international expansion in order to contribute to the improvement of the welfare of their citizens. The first motive for state-owned firms to expand international contains political and ideological aspects. The second motive refers to global market failures, where state-owned MN E’s can intervene to project global goods or services and to provide global solutions (Maskus & Reichman, 2004).
Besides the different motives for internationalization, the literature also provides information as to why the foreign expansions strategies of state-owned MN E’s and privately owned MNE’s could be different (Song, Yang, and Zhang, 2011). The first reason is based on
the different uses of business and economic conditions in the domestic market, which could result in different motivations for international expansion. According to Song et al. (2011), state-owned MNE’s will benefit more from their access to strategic resources and capital. Ramasamy (2012) underlines this argument and stated that strategic asset seeking motivation is more important among state-owned firms. The second reason for the difference in strategies between the state and privately owned firms refers to the several objectives of state-owned firms, which cover both economical and political aspects.
2.5 Research gap
Summarizing, a lot of research is done on the relationship between level of multinationality and firm performance. However, in the field of the relationship between the level of multinationality and firm performance, the possible influences of different
ownerships structures is still unexplored. Although in the existing literature, there is some attention paid to the difference in internationalization paths of family versus non-family and state versus non private-owned firms. The impacts of these different ownership structures on the relationship between the level of multinationality and firm performance is not included in the literature. To build upon the latest research, which proposes a non-linear relationship, the first part of the research will test the already investigated S-curve relationship. To make a contribution to the literature this paper implies a greater degree of novelty in this S-curve relationship by testing the moderation effects of the different ownership structures since no previous study has done that.
3. THEORETICAL FRAMEWORK 3.1 Level of multinationality and firm performance
As discussed in the literature review, most recent scholars have provided findings that propose a horizontal S-curve instead of a linear relationship between level of multinationality and firm performance. The curvilinear relationship arose because these studies started to address costs involved in internationalization. Based on the findings of the U-shaped relationship, the idea arose that an S-curve also might exist (Lu and Beamish, 2001, Contractor et al., 2003).
The benefits of the level of multinationality include both exploration and exploitation. According to Caves (1971) the main driver of internationalization refers to exploiting market imperfections in the cross-border use of firm’s intangible assets. Also the exploration be nefit of internationalization through the organizational learning perspective can help to overcome competitiveness and expand capabilities and knowledge based to enhance above- normal profits.
Besides the various benefits, there are also costs involved in the process of internationalization, which can be attributed to the liabilities of newness and foreignness (Hymer, 1976), which will influence the firm performance. The costs are building upon the challenges that firms need to overcome by starting new operations across their national borders. Possible challenges could include access to external business networks, human capital and certain features that need to be purchased by starting a business abroad. A possible consequence of these liabilities could be that the competitiveness of these new firm decreases compared to established firms in the host country. Due to the liability of foreignness (Hymer, 1976), foreign subsidiaries cannot conduct business activities as effectively as local firms do (Lu & Beamish, 2004). This could result in higher costs and possible mistakes in their activities, which affects the firm performance. However, these consequences will decrease by learning-by-doing process (Barkema & Vermeulen, 1998; Vermeulen & Barkema, 2002). According to Barkema, Bell and Pennings (1996), if firms build and improve reputations and legitimacy in the target market, it could decrease these liabilities.
Besides the above- mentioned costs, firms also need to face costs that relate to the transaction and coordination aspects (Denis et al., 2002; Harris, Kriebel, & Raviv, 1982). In contrast to the costs of liabilities of newness and foreignness, the transaction and coordination costs will increase with the degree of internationalization (Jones & Hill, 1988). An increase in transaction could results in higher governance costs. Governance costs will further rise if subsidiaries face environmental uncertainties, which arise from operating in dissimilar markets (Bergh & Lawless, 19989; Hill & Hoskisson, 1987; Jones & Hill, 1988). A possible consequence would be that the governance costs rise rapidly to a point where these costs exceed any benefit of internationalization (Hitt el al., 1997; Tallman & Li, 1996).
Through combining the exploitation and exploration benefits and the possibility of the increased costs of internationalization a nonlinear relationship is expected between the level of
multinationality and firm performance. Existing research shows that this S-shaped relationship derives from the expectation that in the initial stages of international expansion firms will encounter different costs, which do not outweigh the benefits. This results in the benefits increasing with respect to the costs, but once the costs rise due to high coordination and control costs for example, these costs will exceed any benefit of internationalization (Lu & Beamish, 2004). However, this study will not test the level of multinationality as a whole but will make a distinction between the scale and the scope of internationalization.
The scale of internationalization refers to the degree of internationalization. This study predicts an S-shaped relationship between the scale of internationalization and firm performance. At first increasing internationalization will ensure a declining in firm performance, followed by a positive relationship between scale of internationalization and firm performance. At a certain point the higher degree of internationalization will not ensure higher firm performance but instead the firm performance will decline. This led to the following hypothesis:
H1: The relationship between the scale of internationalization and firm performance is S -shaped
The scope of internationalization refers to the global activities versus the regional activities of an MN E. According to Stopford and Wells (1972), the extent of area diversification is one of the two critical determinants that affect the success of a firm’s growth. As discussed in the literature review, the fact that the nature and extent of global versus regional activity has become more visible in the past years made it possible to get better insides into the international activities of many MNE’s across the globe. Besides the expected S-curve for the
firm performance predicts to be same. It is expected that for the relationship between the scope of internationalization and firm performance at the initial stages, performance decline with increasing global activities, followed by a positive relationship between increas ing scope of internationalization and firm performance. Then at very high levels of globalization, the performance of the firm will decline. This led to the following hypothesis:
H2: The relationship between the scope of internationalization and firm perf ormance is S-shaped
3.2. Different ownership structures Family versus non-family owned firms
According to the literature, the path of internationalization of family versus non- family owned firms could differ. As mentioned in literature, family firms are dealing with a limited access to resources, which ensure a lower inclination of international expansion (Gomez-Mejia et al., 2010). The financial resources of family firms ar e found to be lower in comparison to non- family business, because of the reluctance to rely on banks or to put the firm into debt (Pukall et al., 2014). Due to the possibility of missing financial firepower of family firms, a negative relationship was found between family firms and their impact on internationalization efforts (Graves & Thomas, 2008). The combination of these restricted financial resources and the lack of expertise in international management has a negative impact on the internationalization of family firms (Graves et al. 2008). The lack of expertise can be derived from a shortage of managerial capabilities and internationalization knowledge of family members working in the business (Okoroafo, 1999).
Besides the resources and capabilities view on family versus non-family firms, Banalieva et al. (2011) argue that family firms can be more profitable with a strong focus on
home region than a more global strategic focus. When family firms would like to focus on a more global strategy, external leadership is assumed to be a better choice for family firms (Banalieva & Eddleston, 2011).
Most theories of firm internationalization stress the importance of having different types of resources in order to be successful in international expansion. Dunning (1988) as well as Hitt et al. (1997) emphasizes the importance of the presence of strategic capabilities for internationalization. To provide competitive advantages over local firms, firms need to have strategic resources and knowledge. In observance of above- mentioned shortage of family firms, the effect on firm performance is expected to be negative. Straight from the beginning phase, family firms have to face high costs of gaining access to recourses. Therefore, it will be assumed that this will result in lower firm performance. Therefore it is expected that fami ly firms have overall lower performance as compared to non-family firms, regarding internationalization. This led to the following hypothesis:
H3A: Family ownership negatively moderates the relationships hypothesized in H1 and H2
State versus private-owned firms
Besides the difference between family and non- family firms, the literature also provides information on the difference in the internationalization path of state versus private firms. An important distinction between state and private firms refers to the different motives for international expansion. State-owned firms are generally focused on political and ideological aspects to provide global solutions; where private firms overall have the aim to maximize the profit of the firm. Despite the fact that state-owned firms will benefit more from their access to strategic resources and capital (Song, 2011), state-owned firms are more likely to use this in order to improve the welfare of their citizens instead, they can be motivated by
non-commercial objectives. Because of the fact that state-owned firms give the same or even more preference to political than economic objectives (Amighini et al., 2013), it is expected that there will also be difference in the firm’s performance in both different ownership structures.
Besides the difference in motives for international expansion between state and private-owned firms, the literature tends to view state-private-owned firms as inefficient, with a lack of coherence in their strategy and poorly managed bureaucratic entities (Arocena & Oliveros, 2012). According to Megginson and Netter (2001), because of the above-mentioned characteristics firms are less efficient in state than in private hands.
Because of the different motives of operating abroad and in addition the o verall view of the characteristics of state-owned firms it is expected that state-owned firms have an overall lower performance when compared to private firms, regarding internationalization. This led to the following hypothesis:
H3B: State ownership negative moderates the relationships hypothesized in H1 and H2
The above-mentioned hypotheses constitutes the conceptual model as shown in Figure 1. Figure 1. Conceptual model
Firm performance -Average ROA Level of multinationality -Scale H1 foreign/total sales -Scope H2 -global/total sales Ownership structure
-Family versus non-family H3A -State versus Private H3B
4. METHODS 4.1 Sample and data
In this thesis an analysis was made of the relationship between the level of multinationality and firm performance of the Fortune Global 500 companies. These companies include worldwide firms, arranged according to their revenue. The sample used in this study was a selected frame, comprised of all the firms included in the Fortune Global 500 in the year of 2014. The listed firms included large multinationals, which provided insight into the level of multinationality because of their activities outside their home country. Another important feature of the Fortune Global 500 was the fact that there is no focus on specific countries, which ensure no specific country bias; this resulted in a better generalizability of the results.
This research is based on quantitative data and used a cross-sectional research design. The data is based on the financial performance of these firms in the year 2013, as the results of last year not yet been communicated by all firms. In order to construct a data set of the Fortune Global 500, data was collected from one database and was established by the use of primary sources. The obtained data came from the database ORBIS and also derived from the annual reports of the listed companies. The firms for which not enough information could be gathered were discarded, leaving a final sample consisting of 246 firms for the scale of internationalization and 173 for the scope of internationalization.
4.2 Measures
4.2.1 Dependent variable
The dependent variable in this research was the performance of the firm. In order to measure the firm performance the Return on Asset (ROA) was used. The ROA is defined as the ratio of earnings before taxes and is a common used variable to measure firm performance
are commonly used to measure the firm performance are the return on equity (RO E) and return on sales (ROS) (Peng, 2004).
4.2.2 Independent variable
The independent variable in this research was the level of multinationality. This level is determined by means of the scale and the scope of internationalization of a firm. The first measurement of internationalization was the scale, which included the percentage of foreign sales relative to total sales. The second measurement of internationalization was the scope, which can be divided in two different types, namely the geographical and product scope. In this research the geographical has been taken into account, which corresponds to global sales over total sales (Barkema and Vermeulen, 2002).
4.2.3 Moderating variables
The moderating variable used in this research was the different ownership structure. The Fortune Global 500 consists out of companies with different ownership structures. In the thesis a distinction was made between FAMILY versus NON-FAMILY owned companies and
STATE versus PRIVATE owned companies.
4.2.4 Control variable
The control variables used in this research were the size of the firm, age of the firm and the
industry in which the industry operates. The Fortune Global 500 consists of firms that vary in
size, age and industry.
The first control variable at firm level was the size of the firm. According to Gomez-Mejia and Palich (1997) the size of the firm has an impact on the firm performance. Due to the fact that large firms are more capable to deal with uncertainties and exploiting the economies of
scale, it results in higher return on assets (Chao & Kumar, 2010). However, Tihanyi et al. (2000) suggest that smaller firms are more pliable and unclosed to change. The size of the firm was measured by looking at the total numbers of employees within the firm.
The second control variable at firm level was the age of the firm. This variable was used as a control variable since it could have an effect on the short and long-term performance of a firm. According to De Jong & Van Houten (2014) young firms are in favor since they make use of more up-to-date technology and/or modern manage ment techniques than the outdated techniques of older firms. In contrast, Julian & Castrogiovanni (1995) found a positive effect between older firms and foreign direct investment because of their greater breadth of geographic expansion. The age of the firm was measured by looking at the number of years since incorporation.
The last control variable was at industry level and related to the different industries in which the firm operates. Since De Jong & Van Houten (2014) suggest that external industry factors could have an effect on the firm performance this control variable was taken into account. In order to get a better overview of the industries in which the firms of the Global Fortune 500 operates, the industries were classified into five categories. On the basis of the SIC code the following categories were made: 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
In this chapter the descriptive statistics and the regression model of the independent, dependent, moderator and control variables are displayed. In the first table the descriptive statistics and the corresponding correlations between the variables are presented. According to Pallant (2011) it is important to view if the correlations have a value above 0.7, should this
be the case there has been presence of problematic constructs. However the scale and the scope of internationalization were highly correlated (r = .81). The possible explanation for the high correlation between those variables could be the fact that firms with a higher degree of internationalization also provide a higher global focus. Besides evaluating the correlations for multicollinearity it is also important to check the variance inflation factors (VIF) and the tolerance levels. According to Field (2009) a VIF value of 10 and tolerance levels below 0.2 are critical values. All the VIF values of the variables are between 1.1 and 9.7 and almo st all tolerance levels are higher than 0.2 with the exception of the scale and the scope of internationalization. In response to the VIF values and the tolerance levels we can conclude that multicollinearity was not problematic in this thesis.
On the basis of the mean of the scale of internationalization (0.47) it can be concluded that the firms that are listed in the Fortune Global 500 are quite international. Slightly less than the half of the firms, namely 47%, had most of their sales outside the home country. The other independent variable, the scope of internationalization, suggests that 39% of the firms were more globally orientated than regional. The average of the firm’s performance, which is
the dependent variable and was measured by the ROA, give s a 4.67. This average refers to the efficiency of the management of a firm by using its assets to generate earnings. O ne of the moderators, family ownership, showed that only 4% of the firms were family owned. The other moderator, state ownerships, indicated that 21% of the firms listed in the Fortune Global 500 were owned by the state. If we consider the first control variable, the size of the firm, the average is 103849.75 employees. This high average can be explained, as it is about the largest companies worldwide. The second control variable, the age of the firm, showed that the firms exist almost 60 years on average. The final control variable, the firm industry, has showed that most of the firms are active in the automotive, machinery and transport industry.
To test the relationship between the scale and scope of internationalization and firm performance a hierarchical regression analysis was used. The hypothesis consisted out of a dependent variable firm performance, measured by return on assets, wit h continuous values wherein it was expected that there would be an S-shaped relationship. The model, which is best suited to test this relationship, is the Ordinary Least Squares (O LS). The regression analyses consists out of two-separated analysis, the first analysis the scale of internationalization and the second the scope of internationalization. The first table shows the hypotheses, which are related to the scale of internationalization. The second table shows the hypotheses, which are related to the scope of internationalization. The first model in both tables is the baseline model that includes only the control variables. The second, t hird en fourth model test the S-shaped relationship that are stated in hypotheses 1 and 2. Model 2 adds the linear term of internationalization, model 3 uses its squared term and model 4 its cubic term. The fifth and the sixth model test the moderation effects of family versus non-family firms, and state versus privately owned firms.
The results of the regression analysis are shown in table 2 and 3. The three most important values to be scrutinized are the significance, the unstandardized b-coefficient and the R-squared measures. The significance relates to whether or not to support the hypotheses that were drawn up. The unstandardized b-coefficient refers to average change of the dependent variable, given that the independent variable goes up one unit. The last important value, R-squared, measures how close the data is to the fitted regression line and thus measure the goodness of the fit of the model. The R-squared values of both regressions showed low values that were not significantly higher than 0, as the significant F. change shows. This can be explained through the dependent variable ROA that can be influenced by a lot of different factors, amongst which some are not included in this model.
First, the results of the relationship between the scale of internationalization and firm performance are discussed. Table 2 shows that the results of the OLS regression did not provide any explanatory power for the variance in firm performance, since most variables showed a significant result above .05 thresholds. Only the control variable professional and
information services, which is part of the firm industry, showed significant effects in all the
models. This variable had a negative effect, indicating that firms in the professional and
information service deteriorate the performance of the firm when compared to the reference
category wholesale and retail. All the other control variables did not have a significant impact on firm performance. As stated in hypothesis 1, an S-shaped relationship was expected between scale of internationalization and firm performance. The results in model 4 do no support this prediction since the scale of internationalization cubed is not significant (b = 14.359, p = .497). When testing for the first interaction term Family vs. Non- family, model 5 did not bring any support for the third hypothesis. As expected, the family firms had a negative effect (b=-1.181, p= .675) on firm performance; the interaction term showed a negative effect as well (b = -.112, p = .776). Since both results did not provide significant results, there is no support that family ownership moderates the relationship hypothesized in H1 in a negative manner. When looking at the other ownership structure, state vs. private firms, there is also a negative interaction effect visible (b=-.172, p=. 752). Since the interaction term was not significant, no negative moderating effects were found for this ownership structure on the relationship between scale of internationalization and firm performance, as stated in H4. Besides testing the hypotheses, the results on the F change showed that the improvement of the model cannot be considered significant. The results of the adjusted R-squared showed that the fit of the model declines when adding variables to the model.
The results of the scope of internationalization on firm performance are shown in table 3. All the control variables, except the firm size, have a negative effect on the firm performance. However, none of these variables have additional explanatory power since their significance is above the .05 threshold. Hypothesis 2 predicted an S-shaped relationship between the scope of internationalization and firm performance. The results do not support this relationship since the scope of internationalization cubed is not significant (b = 29.695, p= .304). This includes that hypothesis 2 will be rejected. Hypothesis 3 predicted that the variable family ownership had a negative moderation effect on the relationship between scope of internationalization and firm performance. Model 5 tested hypothesis 3 by entering the interaction of family ownership structures. The interaction between both variables is negatively signed (b = -.329, p = .517) but not significant, suggesting that the relationship between scope of internationalization and firm performance is not affected by the family ownership structures of a firm. Therefore, hypothesis 3 is rejected. Hypothesis 4 predicted that state ownership also has a negative moderating impact on the S-shaped relationship as stated in H1 and H2. The interaction is negatively signed (b=-1.757, p=. 334) and insignificant in model 6, where scope was the independent variable. No evidence was found for the effects of interaction term between scope of internationalization and firm performance, therefore hypothesis 4 is not supported.
Table 1: Descriptive statistics: means, standard deviations and correlations
Variables M SD 1 2 3 4 5 6 7
1. Scale of internationalization 0.47 0.30
2. Scope of internationalization 0.39 0.27 .81**
3. Firm performance 4.67 6.49 .11 .11
4. Ownership (family vs non-family) 0.04 0.04 .16* .10 .02
5. Ownership (state vs private) 0.21 0.40 -.24** -.23** -.11** 1
6. Firm size 103849.75 161693.52 .06 .04 .14** .20** .01
7. Firm age 59.28 52.00 .15* .14 -.05 .03 -.15** .03
8. Industry effects 2.88 1.50 .01 .04 .06 .10* -.20** .08 .08
**. 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 SCALE on firm performance Controls
Dep. variable: firm performance Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Control variables B Sig. B Sig. B Sig. B Sig. B Sig. B Sig
Firm size: number of employees .003 .170 0.003 .180 .003 .158 0.003 .167 0.004 .125 .003 .157
Firm age -.005 .558 -.007 .466 -.006 .472 -.006 .442 -.006 .544 -.006 .491
1. Mining, utilities and construction -1.551 .212 -1.578 .203 -1.458 .220 -1.394 .353 -1.297 .307 -1.033 .444 2. Automotive, machinery and transport -1.312 .261 -1.581 .181 -1.455 .188 -1.427 .196 -1.370 .255 -1.383 .253 3. Professional and information services -3.852 .021 -3.629 .020* -3.784 .035* -3.587 .043* -3.514 .042* -3.498 .045*
5. Manufacturing and other services -.172 .903 -.357 .800 -.298 .797 -.222 .785 -.222 .884 -.228 .873
Independent variable
Scale of internationalization 2.082 .178 -1.738 .153 6.305 .629 5.910 .653 5.723 .664
Scale of internationalization squared 4.025 .467 -17.127 .588 -16.928 .595 -16.987 .593
Scale of internationalization cubed 14.359 .497 14.881 .485 14.532 .493
Moderator variable
Ownership structures: Family vs Non-family -1.181 .657
Ownership structures: State vs Private -1.312 .399
Interaction terms
Scale of internationalization x ownership: family vs non
family -.112 .776
Scale of internationalization x ownership: state vs private -.172 .752
Constant 6.146 .000 5.311 .000 5.770 0.000 5.203 0.000 5.156 0.001 5.417 0.001 R2 .042 .050 .053 .055 0.055 0.058 Adjusted R2 .014 .018 .016 .013 0.013 0.007 Change in R2 .0042 .008 .002 0.002 0.002 0.003 Sig. F Change .175 .178 .467 .497 .497 .699 p†<0.10; *p<0.05; **p<0.01; ***p<0.001.
