1
Home country effects and home region policy integration
on the internationalization patterns of Emerging Market
Multinationals: The case of BRICS
Elly Overdiek 10001255 June 24th 2016
University of Amsterdam
Master of Business Administration: International Management Supervisor: dr. Niccolò Pisani
2
Statement of originality
This document is written by Student Elly Overdiek who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Abstract
Extensive research has been conducted about the effect of home country institutions and regional policy integration on the internationalization of firms. However, the existing literature leaves three gaps. First, the effects of several types of home country institutions have not been differentiated before. Second, only the direct effect of regional integration has been examined. It has never been related to home country institutions before. Third, the main focus of research on EM MNE internationalization has been on Asian countries like China and India, and has not been broadened to other emerging market. This thesis attempts to fill these gaps by investigating the internationalization patterns of the 100 biggest firms from each of the five BRICS countries. Findings suggest that regulative- and governmental institutions have a positive effect on the scale and scope of internationalization of firms. Financial- and trade institutions have a negative effect on both the scale and scope of internationalization of firms. Legal institutions do not have a significant effect on the scale and scope of internationalization. Regional policy integration negatively moderates the relationship between financial institutions and scale- and scope of internationalization. Regional policy integration also negatively moderates the impact between trade institutions and scope of internationalization. It does not impact the relationship between trade institutions and scale. No moderating effect was found for all other types of institutions.
Keywords: Internationalization, EM MNEs, Home country institutions, Regional policy
4
Table of Content
1. Introduction page 6
2. Literature Review page 9
2.1 Internationalization of firms
2.2 Internationalization of Emerging Market Multinationals
2.3 The role of Home Country Institutions in foreign expansion of firms 2.3.1 Formal Institutions
2.3.2 Informal Institutions
2.4 The role of Regional Policy Integration in foreign expansion of firms
3. Theoretical Framework page 19
3.1 Home Country Institutions and Internationalization 3.2 The moderating effect of Regional Integration 3.3 Model
4. Methods page 27
4.1 Sample and Data collection 4.2 Measures
5. Analysis page 32
5.1 Descriptive Statistics
5.2 The effect of Institutions on Internationalization
5.3 The moderating effect of Regional Integration on Institutions and Internationalization
6. Discussion page 41
6.1 Academic Contribution 6.2 Managerial Implications
6.3 Limitations and suggestions for future research
7. Conclusion page 46
Acknowledgements page 48
5
List of Figures
Figure 1: Conceptual Model Page 26
List of Tables
Table 1: Regional Policy Integration Page 30
Table 2: RTA Scores Page 31
Table 3: Descriptive statistics and Correlations Page 33 Table 4: The effect of Home Country Institutions Page 35 Table 5: The moderating effect of Regional Integration Page 38
6
1. Introduction
With the rise of Emerging Market Multi-National Enterprises (EM MNE’s) the global market place has become more differentiated. When it comes to internationalization, it must be recognized that EM MNEs distinguish themselves from multinational enterprises that are based in established markets (Luo and Tung 2007). Therefore, existing theories on internationalization of multinationals must be re-evaluated for EM MNEs. Several factors have already been linked to the internationalization patterns of firms (Child and Rodrigues 2005, Buckley et al. 2007, Luo and Tung 2007, Kolstad and Wiig 2012, Wang et al. 2012, Wu and Chen 2014). Amongst other things, home country institutions play an important role in the internationalization of EM MNEs (Wang et al. 2012, Wu and Chen 2014). Another important factor that has a direct effect on the internationalization of firms is regional policy integration (Banalieva et al. 2008, Mansfield and Reinhardt 2008, Chen 2009). This thesis examines whether different types of institutions can have a specific impact on the internationalization of EM MNEs and if regional policy integration moderates the impact of home country institutions on firm internationalization.
Additionally, over the last few years the focus of research on internationalization patterns of emerging market multinational firms has been mostly on Asian countries (Kolk and Rivera-Santos 2015). However, there are also a lot of non-Asian countries that distinguish themselves from other developing economies by their fast growth and increasing global significance. An interesting example is the BRICS countries, consisting of Brazil, Russia, India, China and South Africa. Nevertheless, the BRICS have yet received less scholarly attention. Therefore, the BRICS countries make an interesting case to study. Characterized by their fast economic growth and significant influence on regional and global affairs, the BRICS countries are important home bases for a lot of EM MNEs.
To fill these gaps in the existing literature, this thesis attempts to make three
contributions. The first main contribution of this thesis is to investigate the role of several types of home country institutions in internationalization of firms. It is already known that home country supporting institutions have a positive effect on the internationalization of firms (Wang 2012) and that home country institutions have a negative impact on the internationalization of firms when they impose more rules and provide less freedom (Witt and Lewin 2007, Wu and Chen 2014). However, it remains unclear whether the effects of different types of institutions, like regulative-, governmental-, financial-, legal- and trade institutions can each have a different
7
impact on the internationalization of firms. This thesis attempts to provide insights into the effects of different types of institutions.
The second contribution of this thesis, is to examine whether regional policy integration can have a moderating effect on the relationship between institutions and the internationalization of EM MNE’s. Regional policy integration is predominantly regulated through Regional Trade Agreements (RTA’s). Mansfield and Reinhardt (2008) established that preferential trade agreements have a direct effect on the trade between countries by lowering trade barriers and lowering market volatility. Furthermore, Chen (2009) found that firms that are based in countries that are integrated into large RTAs engage more in Foreign Direct Investment (FDI) and are more international. Although the growing evidence exist that regional policy integration has a direct effect on the internationalization of firms, it has never been related to the home country institutions before. This thesis attempts to fill this gap by examining if regional policy integration can moderate the relationship between home country institutions and internationalization.
The third contribution of this thesis is to include non-Asian EM MNES in the research. When it comes to internationalization patterns of EM MNEs, substantial focus has been on vastly upcoming Asian countries like China and India (Kolk and Rivera-Santos 2015). It is established that they internationalize more aggressively and faster than firms that are based in established markets (Luo and Tung 2007), they are more attracted to countries with a similar institutional environment (Buckley et al. 2007, Kolstad and Wiig 2012) and that internal resources, industry and institutions all play a role (Wang et al. 2012). There is also evidence that home country institutions play an important role in those countries (Wang et al. 2012, Wu and Chen 2014). However, it has never been tested if these effects also remain standing for other empirical
settings. This thesis attempts to fill this gap by adding firms from the other BRICS countries into the research.
This thesis takes a Transaction Cost Economics (TCE) perspective (Williamson 1989). Although Johansson and Vahlne (1977) argue that internationalization of firms is a gradual learning process, a growing body of evidence supports an opposing view (Williamson 1989, Benito and Gripsrud 1992, Banalieva et al. 2008). Transaction cost economists argue that firms do not internationalize gradually and with increasing commitment strategies, but that
internationalization is a choice based on a rational cost-benefit analysis. Benito and Gripsrud (1992) compared both views and found evidence that lends support for the transaction cost
8
perspective. This evidence supports the view that internationalization is not a progressive
learning process, but a conscious and calculated choice. Furthermore Luo and Tung (2007) argue that firms from emerging markets do not internationalize gradually, but in a faster and more aggressive way than firms from established markets. This also supports the view that internationalization of firms is based on a rational choice. Therefore, the TCE perspective especially suits the case of firms from emerging markets, and is consequently adopted in this thesis.
This thesis is structured as follows. First, the relevant on internationalization of firms, and more specifically the effects of institutions and regional integration, will be discussed in the next section. Subsequently, a theoretical framework and the development of hypothesis is given in section 3. The methodology, measures and variables are discussed in section 4. In the following section, the results of the statistical analysis are presented. The results are discussed in section 6, with the interpretation of the results, the managerial implications and limitations. Finally, section 7 consists of concluding remarks.
9
2. Literature Review
This literature review consists of four sections. In the first section, I will review the literature on internationalization patterns of firms. In the second section, I will discuss the literature on how EM MNEs differ from other MNEs when it comes to internationalization patterns. In the third section I will discuss the literature on the effects of institutions on the internationalization patterns of firms, and more specifically on EM MNEs. I will review the literature on regional policy integration in the fourth and final section of this literature review.
2.1 The Internationalization of firms
The internationalization of firms is a topic that has been well researched in the past. One of the leading theories about the internationalization of firms is John Dunning’s OLI paradigm (Dunning 1988). This theory argues that, through internationalization, firms can enjoy three types of advantages. The first type is Ownership advantages (O), which means that a firm can extend or gain the possession of tangible or intangible assets abroad. The second type is Location advantages (L) which means that firms can enjoy benefits of the location they are situated in, such as low cost or positive knowledge spillovers. The third and final type is Internalization advantages (I), which means that firms can enjoy advantages when they internalize economic transactions which are imperfectly organized by the market. This theory is based on the internationalization patterns of traditional MNEs in developed countries. Furthermore, a lot of research has been conducted on the risk of doing business abroad and the cost that come with it (Zaheer 1995). Although international expansion brings high cost with it, firms are still
expanding. This seems like a contradiction to the TCE perspective, which argues that firms make a rational cost-benefit analysis when it comes to making choices. However, Sethi and Judge (2009) argue that there are, next to the cost of doing business abroad, also benefits of doing business abroad. Several benefits can be incentives for companies to go abroad, even if there are cost involved. Four distinct motivations of doing business abroad can be identified: strategic assets seeking, natural resource seeking, market seeking and efficiency seeking (Dunning 1988). These motives yield benefits for firms that outweigh the cost of foreign expansion.
However, the process of globalization is far from complete (Ghemawat 2003).
10
insulation of countries, and total international integration of countries. Therefore, Ghemawat (2003) argues that the world is semi-globalized. Rugman and Verbeke (2004) support this contention. They argue that very few firms accomplish to be a global player. Instead, most firms employ a regional strategy, and are mostly active in their home region leg of the triad. The triad used to consist of the United Stated, the European Union and Japan. However, with the rise of emerging economies, the triad can now be extended to members of the NAFTA, the expanded European Union and Asia. The strategy where firms are mostly active in the home region leg of the triad can be called ‘home region strategy’ and is the most common strategy. Most of the 500 firms that were investigated by Rugman and Verbeke (2004) had 80% of their sales in their home region leg of the triad. From the 500 firms, only 9 were unquestionably global. Rugman and Verbeke (2004) argue that the fact that firms are mostly regional rather than global, is the result of a simple cost-benefit analysis firms consider when deciding on a strategy. However, Rugman and Verbeke (2004) propose a differentiation between two more types of international strategies: ‘bi-regional strategy’, where firms are active in the home-region and one other region, and ‘host-region strategy’, where firms are active in only one of the other two ‘host-regions than the home-region. However, these strategies are quite uncommon.
Since the process of globalization is not complete, firms have to make choices on the locations where they want to invest, the degree of commitment of their investments and the timing of their investments. Rugman and Verbeke (2004) established that firms often choose to remain regional instead of becoming global. Two perspectives exist on how firms make these choices. The perspective of Rugman and Verbeke (2004) is based on Transaction Cost
Economics (TCE), which means that firms make rational choices, based on calculations of the cost of all the choices available to firms (Williamson 1989). In contrast with TCE theory, Johanson and Vahlne (1977) argue that firms gradually intensify their activities in foreign markets. Firms start their international expansion in countries that are culturally close and start with low commitment. As their experience with foreign markets increases, they gradually move towards more distant countries and increase their commitment. However, a growing body of research supports the TCE perspective. Benito and Gripsrud (1992) found that Norwegian firms made discrete rational choices instead of following a cultural learning process, which is based on the calculation of transaction cost. Li (2005) found that home region strategies are more effective than global strategies in terms of performance. Firms that followed a more regional strategy
11
rather than an international strategy outperformed the firms that pursued global strategies. Banalieva et al. (2012) found that a high home region focus lead to more technical efficiency performance due to cost advantages. When the home region focus was aligned with regional integration, the effect was even bigger.
2.2 Internationalization of Emerging Market Multinationals
With the rise of EM MNEs, the existing internationalization theories must be
re-evaluated and there must be assessed whether the existing theories are also applicable to MNEs from emerging markets. The internationalization of firms from emerging markets (EM MNE’s) is a recent subject of interest amongst scholars. Hennart (2012) argues that the OLI-theory does not equally apply to emerging market firms, since the theory assumes that resources in a host country are equally available to everyone through market transactions which, in reality, is not the case. Established MNEs from developed countries have competitive advantages which makes certain resources easier to access than it is for new players on the market.
Furthermore, Luo and Tung (2007) describe that EM MNE’s internationalization patterns differ from the traditional path of internationalization that firms from developed markets went through. They propose the springboard perspective: emerging market firms have profited from the years of inward FDI and cooperating with big global players, and use this as a springboard to internationalize quickly and aggressive, and catch up with other large MNE’s. Through the inward FDI, they can acquire strategic resources, thereby reduce institutional constraints and overcome their latecomer disadvantage. This results in different internationalization patterns from MNEs based in established markets, since the MNEs from established markets did not have the same opportunities to profit from inward FDI. Therefore, EM MNEs are able to
internationalize much faster and aggressive.
However, the previous discussed research does not yet clarify which factors determine the degree of FDI of EM MNEs. Scholars have come up with multiple explanatory factors on several levels of analysis: firm level, industry level and institutional level. Kolstad and Wiig (2012) investigated what factors determined Chinese FDI on the institutional level. They tested a series of host country characteristics and found that Chinese FDI is attracted by large markets, countries that are rich of natural resources and have poor developed institutions. Chinese firms
12
were attracted to countries that were culturally and institutionally close to their home country. Child and Rodrigues (2005) investigated firm level factors and found that seeking of
technological and brand assets was the main motivation for Chinese firms to internationalize. Wang et al. (2012) combined the resource based view, industry based view and institution based view to show that a firm’s resources, foreign presence of the industry and home country
supporting institutions are all positively related with the degree of OFDI of Chinese companies. Recently, a growing body of evidence uncovered the relationship between institutional environments and business strategy. Buckley et al. (2007) argue that Chinese firms are more attracted to host countries with higher political risk and are therefore culturally closer. Because the firms are used to the volatile environment of the emerging market they are based in, they are more comfortable operating in the volatile environments that countries with higher political risk are. It can even provide more opportunities since there are less regulations. Regulations can provide security, but also impose constraints on the way the firms operate and the choices they can make. Although a lot of different levels of analysis might provide explanations for the way firms internationalize, this thesis focusses on the institutions based view. The previous discussed research focusses mainly on host country institution. This thesis focusses on the institutions of the home countries firms are based in. The literature on home country institutions will be discussed in the following section.
2.3 The role of Home Country Institutions in Foreign Expansion of Firms
Institutions play an important role in the internationalization of firms, since firms have to deal with multiple different institutional environments when they operate in multiple countries. It has therefore been an important subject of interest in international business research (Rugman et al. 2011). As the research of Wang et al. (2012) shows, home country influences play a
significant role in the way firms choose to internationalize. A growing body of evidence supports the institution based view. North (1991, pp 97) defines institutions as “humanly devised
constraints that structure political, economic and social interaction. They consist of both informal constraints (sanctions, taboos, customs, traditions and codes of conduct), and formal rules
(constitutions, laws, property rights)”. North argues that institutions are designed to reduce uncertainty. When certainty increases, cost can be reduced. Therefore, formal- and informal
13
institutions play an important role in the decision making of firms. However, as Buckley et al. (2007) argue, EM MNEs are more comfortable with uncertain and volatile environments than MNEs from established markets, since this is more similar to the conditions on their home markets. Therefore, the effect of institutions on EM MNEs must be investigated apart from other MNEs. In the next section, the literature on the effect of institutions for EM MNEs is discussed.
2.3.1 Formal Institutions
A lot of research has been done on the effect of home country formal institutions on the internationalization patterns of firms. He and Ciu (2010) examined a panel of 511 firms from 38 countries. They found that firms that were based in countries where the quality of home country institutions was higher were more international than firms that were based in countries with a lower quality of home country institutions. Although He and Ciu (2010) did include EM MNEs in the analysis, they did not differentiate between EM MNEs and firms from developed markets. Therefore, it still remains unclear whether the positive effect of home country institutions on the internationalization of firms is the same for firms from developed markets as for firms from emerging markets.
Several studies researched the impact of home country institutions on internationalization in emerging markets. Wu and Chen (2014) looked into the impact of formal home country institutions on international expansion of EM MNEs and used China as an empirical setting to investigate this. They used the impact of institutional development in the home country and institutional stability in the home country as proxies to measure home country institutions in their research. They found a positive impact of development and stability of home country institutions on the propensity of EM MNEs to expand internationally for both proxies. They argue that developed institutions increase market efficiency and reduce transactions cost of cross-border transactions. Furthermore, they found a positive moderation of government ownership of firms on the positive relationship between institutional development in the home country and
propensity of EM MNEs to expand internationally. Luo et al. (2010) argue that the increasing outward foreign direct investment of countries like China, Brazil and India is mainly due to the supporting institutions and government promotion to expand internationally. The supporting institutions might offset the disadvantages that EM MNEs have compared to MNEs from
14
established economies. Witt and Lewin (2007) argue that firms internationalize to escape the institutional constraints of the home market. If home country institutions are more developed, they impose more regulations and constraints upon firms to do business in the home market. Therefore, firms are more likely to do business abroad as a form of escapism.
Wu and Chen (2014) measured the home country institutions by calculating the stability and development of home country institutions by calculating an index based on numbers from China’s National Economic Research Institute. Based on their results, it can be concluded that formal institutions play a significant role in the internationalization patterns of firms. However, conclusions on which types of institutions are more important than others cannot be drawn using this measure. Gwartney et al. (2014) developed an instrument to differentiate between several types of institutions. They scored the quality of institutions in over 160 countries on five different areas: regulative institutions, legal institutions, financial institutions, trade institutions and governmental institutions. Gwartney et al. (2009) found that market volatility increases when the quality of the institutions decrease, which means that economic freedom becomes lower. Gwartney et al. (2009) did not yet relate the several types of institutions to the
internationalization patterns of firms. They also did not yet check for different effects of the institutions on developed markets and emerging markets.
Although there is a growing body of evidence that formal home country institutions play an important role in the way firms make choices and the way they internationalize (North 1991, Luo et al. 2010, Wang 2012, Wu and Chen 2014), it still remains unclear which type of
institutions are more important than others. Especially the effect of different types of institutions for firms from emerging markets remains unexplored.
2.3.1 Informal Institutions
According to North (1991) informal institutions can be customs, codes of conducts and so forth. They can for instance be understood in terms of national culture or local informal networks of firms. Peng (2002) argues that formal institutions of home countries of EM MNEs in Asia tend to fall short on three areas: a credible legal framework, a stable political framework and well-functioning factor markets. Therefore, Asian firms tend to rely more on the informal
15
linking executives and key stakeholders (especially government) and the reputation of conglomerates. Peng (2002) argues that this is an explanation why Asian firms tend to use networks and alliances for expansion, since these are growth strategies that can be established through informal networks. In countries where formal institutions are stronger and better developed, firms tend to use growth strategies like generic expansion and mergers and acquisitions. This research implies that informal institutions are especially important for EM MNEs, since they form a replacement of the formal institutions. Although it is known that the informal networks influence the way EM MNEs choose to expand, it still remains unclear how informal networks influences the degree and scope of internationalization of EM MNEs. Another way to measure informal networks is through culture. Most research on culture has been focused on the distance between home and host country and the impact this had on firms’ market selection, entry mode decisions, knowledge exchange patterns, subsidiary
management and the relationship between internationalization and firm performance (Drogendijk and Slangen 2006, Hutzschenreuter et al. 2015). This research has been mainly directed to MNEs from developed markets.
Many scholars have been pre-occupied with the measurement of culture. The first one to develop an instrument to measure culture was Hofstede (1984). He defines culture as “the collective programming of the mind which distinguishes the members of one group from another” (Hofstede 1994, pp 1). Five dimensions of national culture were identified. The first one is power distance. This dimension encompasses the degree to which less powerful member of a society accept that the power is distributed unequally. When power distance is high,
hierarchy is important and people are not allowed to make their own decision without consulting superiors. When power distance is low on the other hand, subordinates expect to be consulted and roles are distributed equally. The second dimension is individualism, which is the degree to which individuals in a society are very loosely tied, or when the society is more collectivist and individuals are integrated into strong, cohesive groups. The third dimension is masculinity. The fourth dimension is uncertainty avoidance, which is the degree to which member of the society are comfortable with unstructured situations. The fifth and final dimension is Long term orientation, which is the degree to which members of the society are focused on the short- or long term. Long term orientated societies are characterized by thrift and perseverance, whereas short term orientated societies are more focused on tradition and fulfilling social obligations.
16
There has been a lot of critique on Hofstede’s five dimensions. Scholars think the model is over simplistic, the sample of the research has been limited to one single MNE, it assumes that culture is stable over time which is not true in reality and it assumes culture is heterogenic within a country (Kirkman et al. 2006). Furthermore, it is unclear whether people across different cultures understand all five dimensions in the same way, and can therefore not be compared across countries (Schwartz 1994). Several scholars developed alternative measurements methods for culture (Drogendijk and Slangen 2006). For instance Schwartz (1994) identified 56 values across cultures, and simplified these to seven dimension of culture. Schwartz studies overcome a lot of the limitations of Hofstede’s research by systematic sampling and proposing model that is less simplistic (Brett and Okumura 1998). However, Schwartz’s framework has not yet been used in research and has therefore less empirical confirmation than Hofstede’s framework
(Drogendijk and Slangen 2006). Several authors have proposed an alternative to measure culture. For instance, multiple authors have suggested to use individual perceptions of members of a society so measure culture, and more specifically, managerial perception (Drogendijk and Slangen 2006). After all, managers are the ones behind strategic decisions and behavior. Other alternatives are the GLOBE project (House, Hanges, Javidan, Dorfman and Gupta 2004) and the World Values Survey (Inglehart, Basañez, Diez-Medrano, Halman and Luijkx 2004). However, all of these studies have yet received less empirical confirmation than Hofstede’s five
dimensions.
2.3 The role of Regional Policy Integration in Internationalization of firms
Another important factor that influences the way multinationals internationalize is Regional Policy Integration. Banalieva et al (2008) and Hill (2011) argue that the development of a Regional Trade Agreement (RTA) is a gradual process. This means that there are several levels of policy integration within RTAs with a gradually increasing degree of integration (Hill 2011). For instance APEC is still in the earliest stage of integration, and does not have a free trade agreement yet. Others are more integrated, which means that trade barriers amongst member are eliminated. This is a Free Trade Area. The next step is establishing common trade barriers to non-members of the agreements. When this happens, the RTA is a Customs Union. After that, free movement of goods, labor and capital is realized. The RTA becomes a common
17
market. In the next stage, there is a common economic policy for all members. In this stage the RTA is an Economic Union. In the final and most extreme case sovereignty is being transferred. In this case, the RTA becomes a Political Union. However, a real world example of e Political Union does not yet exist.
One of the benefits of high policy integration within RTAs, is that institutions and regulations are developed to serve the interest of the RTAs’ members and to facilitate trade within the RTA (Krueger 1993). This means that MNEs in countries that are more integrated in terms of regionals policy, can enjoy benefits, mostly in terms of cost savings (Banalieva et al. 2012). Conversely, lower policy integration makes it more difficult to resolve conflict amongst member because of unclear defined enforcement mechanisms and members tend to rely more on their veto-power (Anson et al. 2005). Furthermore, cooperation amongst member can be more troublesome because there are more trust issues, and national pride and disparity in national incomes play a role (Schiff and Winters 2003).
Reinhardt and Mansfield (2008) established that international trade agreements not only directly lead to more trade because trade barriers are lowered, but also because markets become less volatile. When the volatility of cross border transaction reduces, it becomes more attractive for firms to engage is these types of transactions. However, it is still unclear if this home region integration can also strengthen, or substitute the effect of home country institutions on the integration of firms. It might, for instance, be interesting to see if Regional Trade Agreements (RTAs) encourage firms to internationalize, even when home country institutions are weaker. It might also be interesting to see if RTAs are a necessary condition to encourage firms to
internationalize, even when home country institutions are strongly supporting international expansion of firms.
Banalieva, Santoro and Jiang (2012) studied regional policy integration more specifically. They found that regional policy integration had a positive moderation effect on the relationship between home region focus and technical efficiency. They argue that regional policy integration reduces cost and therefore leads to more technical efficiency. Reasoning from the TCE
perspective, it might be interesting to see whether regional policy integration reduces cost enough to substitute for weak home country institutions. It is also interesting to see whether regional policy integration can also reduce enough cost to encourage firms to internationalize, if when home country institution are already supporting to internationalization of firms based in the
18
country. Chen (2009) already found that firms that are based in countries that are integrated into larger RTAs are engaging more in foreign direct investment. This means that firms that are based in countries that are integrated into large RTAs are more international than firms that are based in countries that are not regionally integrated. It can be concluded from this research that regional policy integration has a positive direct effect on the internationalization of firms. However, it still remains unclear how the RTAs relate to home country institutions.
Banalieva et al. (2012) follow Balassa (1961) and Nye (1968) in measuring regional policy integration in assigning scores to the RTAs that home countries of firms are participating in. However, they only differentiate between pre-regional integration and the first two levels of integration. These are the pre-free trade areas, the free trade area and the customs union. Nevertheless, as Hill (2011) argues, three more types of regional integration can be differentiated: common markets, economic unions and political unions.
Three gaps can be identified in the existing literature. First, the main focus of research on EM MNEs has been directed to Asian countries, while other EM MNEs from the BRICS
countries also play an increasingly important role in the world economy. Second, several areas of home country institutions have never been related to the internationalization patterns of firms before. Finally, several authors established that membership of RTAs plays an important role in the internationalization of firms. However, it has never been investigated whether RTAs also relate to home country institutions. If they moderate the effect of home country institutions, it might be possible that they can make up for the lack of quality of home country institutions or strengthen the relationship between home country institutions and internationalization.
19
3. Theoretical Framework
3.1 Home Country Institutions and Internationalization
As North (1991) argues, institutions are humanly devised constraints that shape any kind of interaction. It influences the way people make choices and therefore also influences economic transactions. He and Ciu (2010) found that MNEs that are based in countries where home
country institutions are of better quality are more internationalized than firms that are from countries where home country institutions are of lower quality. However, Luo and Tung (2007) argue that firms from emerging markets do no internationalize the same way as MNEs that are based in established economies. EM MNEs use a much more aggressive strategy to acquire critical assets and internationalize much faster to overcome their latecomer disadvantage and overcome market constraints at home. This raises questions whether or not home country institutions still play a role in the way EM MNEs internationalize. Wu and Chen (2014) argue that home country institutions have an important effect on the internationalization of EM MNEs. Their findings suggest that the more developed and stable the institutions are, the more the firms in that country will internationalize. So if the quality of home country institutions increases, firms are encouraged to internationalize more. Wu and Chen (2014) argue that this is caused by the reduction of cost of cross border transaction, and increase in market efficiency.
As Witt and Lewin (2007) argue, the development of home country institutions can also impose constraints upon firms, which makes them more eager to do business across borders as a way to escape these constraints. On one hand the home country constraints, which increase with the development of institutions, work as a push factor to internationalize. On the other hand, home country institutions provide more support to internationalize if the quality increases. For this reason, it can be argued that the quality of home country institutions has a positive effect on the internationalization of firms. Wang et al. (2012) found that support of home country
institutions specifically is an important factor for firms to internationalize. Institutions that are of better quality might also provide more support for firms to create a stable foundation and make it therefore easier to expand across the borders of the home country. This also means that the higher the quality of the home country institutions, the better the foundation is to expand internationally. As Williamson (1989) argues, doing international business consist of rational choices based on a cost-benefit analysis. This means that firms would do business in countries
20
where to cost are relatively low and the benefits are relatively high.
Summarizing, my expectation that the quality of home country institutions has a positive effect on the scale and scope of internationalization is based on three arguments. First, if quality of home country institutions increases, firms enjoy more support to expand internationally (Wang et al. 2012). Second, if the quality of home country institutions increases, regulations increase which form constraints to operate on the home market. Witt and Lewin (2007) argue that this works as a push effect. Firms would be more eager to look for opportunities in markets where they have more freedom to make choices and evade the constraints of the home market. Third, firms are more likely to internationalize if cost of doing business in the home country increase. If institutions become of better quality, firms have to ante up to more regulations which might increase cost. Therefore, my expectations for each type of institutions are based on the degree of support for internationalization the type of institutions offers, the degree to which the type of institutions imposes constraints upon firms and the home market and the degree to which home country institutions impose extra cost on firms.
Regulative institutions are expected to have a positive effect on the scale and scope of internationalization of EM MNEs. He and Cui (2010) already found that the quality of regulative institutions had a positive effect on internationalization of firms. Although this effect is based on MNEs from developed markets, the quality regulative institutions are still expected to have a positive effect on internationalization of EM MNEs. Regulative institutions consist of three broad areas: credit market regulations, labor market regulations and business regulations. All these regulations can impose constraints on the home market and the operational freedom of firms. They can also increase cost. For instance, labor market regulations consist of minimum wages, hiring regulations and centralized bargaining amongst other things. As firms are obliged pay at least minimum wage as a result of these regulations, labor becomes more expensive for firms. They have less freedom to hire and fire people. Firms also have to be more engaged in stakeholder management since they are more likely to encounter strikes as bargaining becomes centralized. Business regulations can also reduce freedom and increase cost of doing business in the home market. For instance administrative requirements can make it more difficult for firms to do business. Also licensing restrictions constrain the freedom of firms. Tax compliance
increases cost, since it is likely firms have to pay more taxes. Regulative institutions do not really consist of supporting institutions to internationalize. Thus, regulative institutions are made up out
21
of a series of institutions that impose constraints and increase cost to do business in the home country, which makes it more likely that firms would evade the home market, and engage more in direct investment outside of the home country. Hence:
Hypothesis 1a: The quality of the home country regulative institutions is positively related to the scale of internationalization of EM MNEs
Hypothesis 1b: The quality of the home country regulative institutions is positively related to the scope of internationalization of EM MNEs
For legal institutions I expect a positive effect on the scale and scope of
internationalization of firms. As stated before, it can be expected that firms tend to be more international if home country institutions are of better quality, because they lend more support to internationalize, the constraints on the home market work as a push factor and firms are more eager to expand across borders if cost increase (Witt and Lewin 2007, Wang et al. 2012,
Williamson 1987). This is also the case for legal institutions, especially in the case of firms from emerging markets. As Buckley et al. (2007) argue, firms from emerging markets are used to volatile environments and therefore more comfortable to operate in these types of environments. Therefore, they are even more likely to avoid institutions which decrease volatility, but at the same time increase constraints. There are also other factors that are part of the calculation for legal institutions which might create a less volatile environment, but at the same time impose more constraints on firms. Some examples are the protection of property rights and the regulatory restrictions on the sale of real property. Military interference and reliability of the police are also part of the calculation for legal institutions, which also reduces the freedom of firms to make any choice they want. Thus legal institutions provide constraints which might encourage firms to invest across borders. Therefore, EM MNEs are expected to engage more in foreign direct investment as legal institutions become stronger. Hence:
Hypothesis 2a: The quality of the home country legal institutions is positively related to the scale of internationalization of EM MNEs
22
Hypothesis 2b: The quality of the home country legal institutions is positively related to the scope of internationalization of EM MNEs
Financial institutions are expected to have a positive effect on the scale and scope of internationalization. If home country financial institutions increase in quality, the access to sound money increases. The access to sound money might lend firms support to internationalize and supply them with the necessary financial resources to make it possible to internationalize. Furthermore, if home country financial regulations increase, they might put more constraints on the home market. The freedom to own foreign currency bank accounts are also part of financial institutions. This freedom makes it easier for firms to engage in cross border activity and therefore lends support to internationalize. This institutional support and home country market constraints encourage firms to internationalize (Wang et al. 2012, Wu and Chen 2014).
Therefore, home country financial institutions are expected to have a positive effect on the scale and scope of internationalization of EM MNEs. Hence:
Hypothesis 3a: The quality of the home country financial institutions is positively related to the scale of internationalization of EM MNEs
Hypothesis 3b: The quality of the home country financial institutions is positively related to the scope of internationalization of EM MNEs
I expect trade institutions to have a positive effect on the internationalization patterns of firms. Trade institutions score higher on the quality measure when they increase in terms of freedom. More freedom does not lead to higher cost or more constraints. More freedom also does not necessarily mean that the institutions lend more support to internationalize. Therefore, the arguments of higher cost of doing business in the home market, higher constraints on the home market and more support to internationalize do not apply to trade institutions. However, as stated before, a positive effect for trade institutions is still expected. Mansfield and Reinhardt (2008) found that the lowering of trade barriers leads to more trade between two countries. If the trade barriers are lower, more FDI will enter the country. Luo and Tung (2007) argue that EM MNEs can profit from inward FDI by learning from successful MNEs. This way, EM MNEs can use the
23
inward FDI as a springboard to internationalize much faster and more aggressively than firms from developed market. So if trade barriers are lower, the more trade will enter a country, the more EM MNEs can profit for it. This way, they can internationalize more and diversify into more countries. It also makes it easier to enter more distant markets. Therefore it can be expected that trade institutions that score higher on quality in terms of economic freedom have a positive effect on the scale and scope of internationalization of EM MNEs. Hence:
Hypothesis 4a: The quality of the home country trade institutions is positively related to the scale of internationalization of EM MNEs
Hypothesis 4b: The quality of the home country trade institutions is positively related to the scope of internationalization of EM MNEs
Governmental institutions are expected to have a positive effect on the scale and scope of internationalization. Home country governmental institutions can provide firms support to internationalize, and might also put constraints on the home market that make it less interesting for firms to just stay within borders. For example, governmental institutions can provide
investments in firms or provide subsidies which can function as support to internationalize. Furthermore, taxes on the home market form constraints and impose cost on firms, which works as push factors to search for other markets where taxes are lower. If the quality of home country governmental institutions increase, investments and taxes also increase. Therefore, home country governmental institutions are expected to have positive effect on the internationalization of firms. Hence:
Hypothesis 5a: The quality of the home country governmental institutions is positively related to the scale of internationalization of EM MNEs
Hypothesis 5b: The quality of the home country governmental institutions is positively related to the scope of internationalization of EM MNEs
24 3.2 The moderating effect of Regional Integration
As Chen (2009) argues, firms that are based in countries that are integrated into large RTAs are more engaged in foreign direct investment. Mansfield and Reinhardt (2008) also argue that RTAs lead to more trade amongst members. This happens in two ways: directly, by lowering trade barriers and thus making it easier to trade. And secondly, by creating a less volatile
environment because the same rules apply to all members. The stronger the regional integration of the RTA, the more the regulations are aligned. Accordingly, RTAs have a positive effect on internationalization of firms within the home region. If home country institutions are not
supporting internationalization of firms that are based in that country, RTAs might complement the shortcomings of the home country institutions by making it easier to cross borders. Also, if home country institutions are already sufficiently supporting internationalization of firms, RTA membership might strengthen this relationship by making it even easier to trade with other countries that are member of the RTA. Furthermore, firms can make a simple cost benefit analysis (Williamson 1987). If home country institutions support internationalization, the cost of doing business abroad are reduced which makes it more attractive. If regional integration
increases, cost of doing business abroad, in the home region, are reduced even more (Banalieva et al. 2011). This makes it even more appealing to firms to internationalize within the home region. It also increases cooperation between countries within the home region because conflict resolution is simplified (Anson et al. 2005). This can have a complimentary effect on the direct effect of institutions: if firms are already a little bit encouraged to internationalize, they will be even more encouraged when it is also easier in terms of trade relationships across borders. So the cost reduction and better relationship building of regional integration might complement the positive effect of the institutions. Therefore, a positive moderating effect of Regional Policy Integration on relationship between home country institutions and scale internationalization of EM MNEs is expected. Hence:
Hypothesis 6: Regional Policy Integration has a positive moderating effect on the relationship between home country institutions and the scale of internationalization of EM MNEs. Thus, regional policy integration positively moderates the relationships as hypothesized in hypotheses 1a, 2a, 3a, 4a and 5a.
25
Regional integration is expected to have a negative moderating effect on the scope of
internationalization. If institutions improve and countries become more regionally integrated, they are encouraged to engage more in foreign direct investment within the home region. This is a result of a simple cost benefit analysis: the cost of doing business within the home region decreases with regional integration (Reinhardt and Mansfield 2008). However, firms will
become less eager to invest outside of the home region. The cost of doing business outside of the home region increases relative to doing business inside the home region when regional
integration increases. Furthermore, since trade amongst member increases, there is less need for firms to seek for markets outside of the home region. Firms are therefore expected to invest more inside the home region, and less outside the home region when regional integration increases. Furthermore, regional integration improves the relationship amongst members and reduces risk. The firms are not regionally integrated with firms outside of the home region which makes conflict resolution more difficult (Anson et al. 2005). This makes it more likely for firms to trade within the home region, because cost and risk are relatively lower than doing business outside the home region and because the market in the home region is growing, it is not necessary for them expand outside of the home region. Even if home country institutions increase the degree of internationalization of firms, this internationalization is expected to be directed towards the home region instead of being directed outside of the home region. Hence:
Hypothesis 7: Regional Policy Integration has a positive moderating effect on the relationship between home country institutions and the scope of internationalization of EM MNEs. Thus, regional policy integration positively moderates the relationships as hypothesized in hypotheses 1b, 2b, 3b, 4b and 5b
26 3.3 Conceptual Model
The previously discussed hypotheses result in the conceptual model as shown in figure 1. Home country institutions (H1 to H5+) Scale and Scope of Internationalization
Regional Policy Integration (H6+ and H7-)
27
4. Methods
4.1 Sample and Data Collection
This study uses a cross-sectional design to test the effect of home country institutions on the internationalization patterns of EM MNEs and whether or not regional integration has a moderating effect on this relationship. The sample consists of emerging market firms from the five BRICS countries, which includes Brazil, Russia, India, China and South Africa. In each country, the 100 biggest firms in terms of annual turnover were selected, which make for a total sample size of 500. For each country, the selected firms must be based in this specific country, which means that the international headquarter is located there, and that the firm is Global Ultimate Owner (GUO). The biggest firms were selected based on turnover in the most recent year reported, but no later than 2013.
I selected the BRICS countries as a population because of the rising importance of the BRICS economies. The 100 biggest firms of each country are selected because these are the firms that make the biggest contribution to the economy in those countries. This means that a purposive sampling technique is used. If firms would have been randomly selected, it is possible that a large part of these firms are not internationally active, which would make them less
interesting in the context of this research. Also, bigger firms are more likely to have public available data than smaller firms. Therefore, the purposive sampling method is most suitable for this research question.
The data collection method is an archival research, based on secondary data on the annual turnover, home country sales, foreign sales, R&D budgets, number of foreign subsidiaries and other numbers. I used this method because it is a relatively fast and low cost method to collect large amounts of relatively reliable data. Part of this data was collected from the ORBIS database (Bureau van Dijk 2016). The other part of the data was obtained from the annual integrated reports published by the firms. In cases where values were reported in the home country currency they were converted to US dollars. For determining the home country institutions for the five countries, data from the Economic freedom dataset are used (Gwartney, Lawson and Hall 2014). I used the dataset of 2014 to match the figures of financial performance from fiscal year 2014. It contains information on the following five broad areas: size of government (governmental
institutions), legal structure and security of property rights (legal institutions), degree of access to sound money (financial institutions), freedom to trade internationally (trade institutions), and
28
regulation of credit, labor, and business (regulative institutions).
The research uses a deductive approach, because hypotheses derived from previous research will be tested. An Ordinary Least Squares (OLS) regression method is applied to test whether or not the variables are related to each other. This is a commonly used method, suitable for estimating parameters of quantitative datasets (Field 2009). The data was tested for all the assumptions of regression analysis. The five institutions are not used in one final and complete model because some multi-collinearity issues occurred. Some variables are transformed to correct for the spread in standard errors. This is specified in the following section. Some
variables were not normally distributed. After a sensitivity analysis, transformation turned out to be very complicated. However, since the N is sufficiently large I went by the central limit theorem which states that all variables are approximately normally distributed when certain conditions are met (Rice 2006).
4.2 Measures
Home country institutions (IV) is measured using the Economic Freedom Index
(Gwartney et al. 2015). Five types of institutions are distinguished: regulative-, legal-, financial-, trade- and governmental institutions. For each type of institutions, a country can obtain a score from 0 to 10, with 0 being low quality and 10 being high quality. The score for regulative institutions is composed of several credit market regulations, labor market regulations and business regulations. The score for legal institutions is composed of several items, including legal enforcement of contracts, protection of property rights, integrity of the legal system and reliability of the police. The score for financial institutions is composed of money growth, standard deviation of inflation, inflation in the most recent year and freedom to own foreign currency bank accounts. The score for trade institutions are composed of tariffs, regulatory trade barriers, black market exchange rates and controls of the movement of capital and people. Finally, the score for governmental institutions is based on transfers and subsidies, government enterprises and investments, government consumption and top marginal tax rate.
Internationalization of EM MNEs is the Dependent variable (DV). Baum et al. (2015) describe three types of measures of internationalization patterns they uncovered in previous research: timing, scale and scope of internationalization. However, since the dataset used in this
29
study is cross-sectional and not longitudinal, it was not possible to measure internationalization timing of the emerging market firms. Therefore, I will measure this variable in two ways: the degree, or scale of internationalization of EM MNEs and the scope of internationalization of EM MNEs. Scale will be measured by comparing the foreign sales of a firm with the total sales. Foreign sales are defined as all sales outside of the country, in other words; total sales minus domestic sales. The percentage of foreign sales over total sales is used as final measure of scale. Scope will be measured by comparing the firms’ global sales with the total sales. Global sales are defined as sales outside of the home region, in other words; total sales minus regional sales. The scope measures to what degree firms are actually international, instead of regional. The percentage of global sales over total sales is used as final measure of scope.
Regional Policy Integration is the moderating variable (MV). According to Hill (2011), the progression of policy coordination increases gradually with each level of Regional Trade Association (RTA). Each country got a score assigned for the membership of a RTA and to which degree this RTA demands policy integration among its members. Thus a higher value for regional policy integration indicates a higher level of policy coordination and integration among the home countries of firms that are members of the RTAs. The ascribed scores are shown in table 2.This table includes all RTAs that the BRICS countries are participating in according to the Regional Trade Agreement Information System (WTO 2016). The bi-lateral agreements are excluded. If the RTA is a pre-Free Trade Area, the score for regional integration is 0. For a Free Trade area the score is 1, for a Custom Union 2, for a Common Market 3, for an Economic Union 4 and for a Political Union 5. This is also shown in table 1. The RTA with the highest score for each country is used in the analysis, because this indicates the highest degree of integration for the country. Even though South Africa and Russia both have a customs union as the highest level of regional integration, the score for Russia on regional integration is 2 and the score for South Africa has been adjusted down to 1. Since South Africa is only integrated into one custom union, and Russia is, next to a customs union, also integrated into two more trade agreements, South Africa is less regionally integrated in total and the score for regional integration of South Africa should be lower. Therefore, the score for South Africa has been adjusted to 1. China, India and Brazil are all integrated in a common market and thus score 3 on regional integration.
30
is the age of the firm. Age of the firms is measured as the natural logarithm of the number of years the firms has been active since the year of incorporation, following the measure of Li (2008). The natural logarithm was used in the analysis to correct for the spread in standard errors. The second control variable is firm size, which is measured in two ways. It is measured by the number of employees, and by the total turnover. The two variables for measuring size are also corrected by taking the natural logarithm of the variables. The final control variable is gearing, which is the percentage of debt related to the company’s equity capital.
Table 1: Regional Policy Integration
Source: Adapted from Hill (2011)
Type Barriers to Trade Amongst Member Barriers to Trade for Non-Members Free Movement of Goods, Labor and Capital Member Economic Policy Transfer of Sovereignty Score for regional integration Free Trade Area Removed Member specific No Member Specific No 1 Customs Union Removed Common External Barrier No Member Specific No 2 Common Market Removed Common External Barrier Yes Member Specific No 3 Economic Union Removed Common External Barrier Yes Harmonized No 4 Political Union Removed Common External Barrier
31 Table 2: RTAs and Scores for Regional Integration
Source: WTO (2016)
Country Country Scores
RTAs and Scores for Regional Integration
Brazil 3 Mercosur (3) GSTP (1) LAUA (0) PTN (0)
Russia 2 EAEC/EAEU (2) CIS (1) CEZ (1)
India 3 ASEAN (3) GSTP (1) SAPTA (0) SAFTA (0) APTA (0)
China 3 ASEAN (3) APTA (0)
South Africa
32
5. Analysis
5.1 Descriptive statistics
The descriptive statistics of the dependent, independent and control variables are
presented in table 3. As the table shows, some of the predictor variables are correlated with each other. Some of the institutions are correlated with each other and with regional integration. A series of tests for the assumptions of linear regression analysis have been conducted. Since the development of several types of institutions within a country are interrelated, multicollinearity was expected. This turned out to be the case. Some of the institutions and the interaction effects between the regional integration and institutions scored a tolerance of lower than 0.20. This means there is some multicollinearity in the data. Therefore, the five institutions and the five interaction effects between institutions and regional integration are not used in one complete model. Although regional integration was also correlated with some of the institutions, there was no multicollinearity between those variables. Therefore, there was no harm in using the regional integration variable together in a model with each type of institutions separately.
The mean of scale of internationalization measures the degree of internationalization. The scale of internationalization was 0.28 which means that the percentage of international sales was 28% on average, with an N of 196 and a standard deviation of 0.282. The scope of
internationalization measures the dispersion of internationalization. The mean of scope of internationalization was 0.32 which means that the percentage of global sales was 32% on average with, with an N of 113 and a standard deviation of 0.316.
Out of the institutions, the financial institutions scored the highest on average with a 7.98 out of 10. The financial institutions were followed by trade institutions with a 6.57, regulative institutions with a 6.36 and governmental institutions with a 6.28. Legal institutions scored lowest with a 5.45. The mean of regional integration scored a 2.4 out of five, although scores 4 and 5 never occurred in this dataset. Since the scores for each country where applicable to all 100 firms for each country, the independent variables where not depending on the willingness of firms to report on several figures. Scores where available for all 500 firms. Therefore, the N for all of these variables is 500.
33 Table 3: Descriptive statistics and correlations
Variables Mean Std. Dev. N 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 1. Scale 0.28 0.282 196 1 2. Scope 0.32 0.316 113 0.950*** 1 3. Regulative Institutions 6.36 0.749 500 0.152** 0.246*** 1 4. Legal Institutions 5.45 0.364 500 -0.110 -0.013 0.767*** 1 5. Financial Institutions 7.98 0.656 500 -0.157** -0.396*** -0.003 0.272*** 1 6. Trade Institutions 6.57 0.410 500 -0.259*** -0.276*** -0.274*** 0.190*** 0.372*** 1 7. Governmental Institutions 6.28 0.909 500 0.261*** 0.299*** -0.287*** 0.086*** -0.631*** -0.605*** 1 8. Regional Integration 2.40 0.801 500 -0.119* -0.132 -0.665*** -0.483*** -0.393*** -0.272*** 0.344*** 1 9. Ln Firm Age 3.31 0.814 479 0.125* 0.128 -0.058 -0.143** -0.241*** 0.106** 0.178*** -0.045 1 10. Ln Firm Size: Employee’s 9.47 1.727 404 -0.064 -0.051 -0.121** 0.063 -0.058 -0.078 -0.105** 0.366*** 0.011 1 11. Ln Firm Size: Sales 8.06 1.557 491 -0.152** -0.127 -0.170*** 0.088* -0.116*** -0.090** -0.133*** 0.527*** -0.036 0.666*** 1 12 Gearing 139.53 274.11 416 -0.026 -0.087 -0.061 -0.127*** -0.049 -0.125** 0.143*** 0.086 -0.018 -0.005 0.046 1 *** p > 0.01, ** p > 0.05 * p > 0.1
34
The natural logarithm of firm Age had a mean of 3.31. The mean of firm age before transformation of the variable was 37.49, which means firms where 37.49 years old on average. Out of the 500 firms, 479 reported their age. The natural logarithm of firm size measured by employees had a mean of 9.47. Before transformation, the mean of the number of employees that worked for the firms was 43997.13. Out of the 500 firms, 404 reported their number of
employees in fiscal year 2014 or 2013. The natural logarithm of firm size according to sales was 8.06. Before transformation of the variable, the mean for firm size measured by sales was 12585.36, which means that the mean of turnover for the fiscal year of 2014 for the companies was 12.58536 billion dollars. Out of the 500 firms, 491 reported on their annual turnover for the fiscal year of 2014 or 2013. Gearing scored a mean of 139.53.
5.2 The effect of Institutions on Internationalization
The base effects of home country institutions on the internationalization of EM MNEs in the BRICS countries are presented in table 4. Model 1 shows the effects of the control variables. For both scope and scale the models where not significant, and none of the control variables had a significant effect. This is in line with the theory of transaction cost economics (Williamson 1989) that internationalization is a rational choice based on a cost benefit analysis, rather than a gradual process where internationalization occurs in stages and firm size and age would play a role. It also supports the theory of Luo and Tung (2007) that EM MNEs internationalize much faster and more aggressively than MNEs from developed market where age and size do play a role.
As model 2 shows, regulative institutions have a positive and significant effect on the scale (b = 0.093, p = 0.006) and scope (b = 0.105, p = 0.013) of internationalization of EM MNEs. The scale of internationalization increased with 9.3 percentage points for each point the regulative institutions improved. Model 2 including regulative institutions explains 5.8% of the variation in in scale of internationalization. With a p-value of 0.026 the model is significant. The scope of internationalization increased with 10.5 percentage points for each point the regulative institutions in the home country improved. The model containing regulative institutions explain 5.6% of the variation in scope of internationalization. With a p-value of 0.099, this model is also
35 Table 4: The effect of Home Country Institutions on Internationalization
*** p < 0.01, ** p < 0.05 * p < 0.1
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Scale Scope Scale Scope Scale Scope Scale Scope Scale Scope Scale Scope
Constant 0.009 (0.241) -0.161 (0.364) -0.74 ** (0.355) -0.954** (0.470) -0.033 (0.502) -0.597 (0.678) 0.760* (0.415) 2.406*** (0.731) 1.475*** (0.545) 1.349* (0.758) -0.341 (0.275) -0.548 (0.418) Dependent Variables Regulative 0.093*** (0.033) 0.105** (0.042) Legal 0.007 (0.071) 0.073 (0.095) Financial -0.088** (0.040) -0.276*** (0.070) Trade -0.219*** (0.059) -0.209** (0.093) Governmental 0.069** (0.028) 0.082* (0.045) Control Variables Firm Age 0.048 (0.034) 0.063 (0.048) 0.062* (0.033) 0.069 (0.046) 0.049 (0.036) 0.069 (0.049) 0.027 (0.035) 0.019 (0.045) 0.058* (0.032) 0.083* (0.047) 0.019 (0.035) 0.046 (0.048) Firm Size: Employees 0.042 (0.027) 0.050 (0.045) 0.049* (0.027) 0.051 (0.043) 0.042 (0.027) 0.055 (0.045) 0.058** (0.028) 0.031 (0.041) 0.042 (0.026) 0.018 (0.046) 0.043 (0.027) 0.029 (0.046) Firm Size: Turnover -0.037 (0.024) -0.028 (0.045) -0.032 (0.024) -0.018 (0.043) -0.037 (0.025) -0.032 (0.045) -0.052** (0.025) -0.028 (0.041 -0.042* (0.023) -0.010 (0.044) -0.034 (0.024) -0.007 (0.046) Gearing 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) 0.000 (0.000) Model fit N 133 78 133 78 133 78 133 78 133 78 133 78 R2 0.038 0.039 0.094 0.117 0.038 0.047 0.079 0.208 0.133 0.101 0.082 0.080 Adj R2 0.008 -0.012 0.058 0.056 0.000 -0.019 0.037 0.153 0.099 0.040 0.046 0.017 F-stat 1.260 0.750 2.647 1.932 1.002 0.713 2.014 3.825 3.913 1.646 2.292 1.272 P-value 0.289 0.561 0.026 0.099 0.419 0.616 0.081 0.004 0.002 0.159 0.049 0.285
