Corporate Diversification Strategies and Optimal Firm Performance:
Empirical Evidence for Hybrid Environmental Munificence
Master’s Thesis, International Business and Management (MSc) University of Groningen, Faculty of Economics and Business
This paper analyzes the relationship between corporate diversification strategies and firm performance in hybrid home country environments. These types of home country environments are classified as nations that possess sufficient factors and deficient institutions or vice versa. The analysis is conducted using a sample of 400 firms in 8 different countries, distributed evenly amongst the 2 distinct types of hybrid home country environments. The results provide evidence for a relationship between firm performance and inbound international diversification in hybrid home country environments. However, the study does not find any significant results that support the presumed relationship between the other forms of corporate diversification strategies and firm performance.
Corporate Diversification Strategies and Optimal Firm Performance:
Empirical Evidence for Hybrid Environmental Munificence
ABSTRACT
Keywords: Corporate diversification strategies, home country environments, firm performance, hybrid environments
Research theme: Multinational Corporations and Internationalization
ACKNOWLEDGMENTS
The author would like to thank Drs. Hans van Polen from the University of Groningen for his helpful comments and insights on earlier versions of this paper. Furthermore, the author would like to thank his family, friends and girlfriend for their support during the writing of the thesis.
TABLE OF CONTENTS
1. INTRODUCTION ... 5
2. LITERATURE REVIEW AND HYPOTHESES ... 8
2.1 Home country environments ... 8
2.1.1 The role of factors ... 9
2.1.2 The role of institutions ... 11
2.2 Corporate diversification strategies ... 13
2.2.1 Product diversification ... 14
2.2.2 Inbound international diversification ... 16
2.2.3 Outbound international diversification ... 17
2.3 Conceptual model ... 19
3. RESEARCH METHODS ... 20
3.1 Sample ... 20
3.1.1 Selecting the country sample ... 20
3.1.2 Selecting the firm sample ... 22
3.2 Data and measurement ... 23
3.3 Method of analysis ... 26
3.4 Linear regression assumptions ... 26
4. RESULTS ... 28
4.1 Descriptive statistics ... 28
4.2 Correlations ... 28
4.3 Regression analysis results ... 31
4.3.1 Product diversification ... 31
4.3.2 Inbound international diversification ... 31
4.3.3 Outbound international diversification ... 32
4.3.4 Brief overview and remarks ... 32
5. DISCUSSION ... 35
6. CONCLUSION ... 38
6.1 Limitations and suggestions for future research ... 39
REFERENCES ... 40
APPENDIX A ... 46
APPENDIX B ... 47
1. INTRODUCTION
An essential element of business is the strategic orientation of the firm (Morgan & Strong, 2003). Previous research has shown the role of corporate strategy in explaining profitability differences between organizations (Bowman & Helfat, 2001). Strategic decisions entail the unique and sustainable means by which firms aim to create value (Kaplan & Norton, 2001). For a strategy to be successful, it is important that there exists a strategic fit between the strategic orientation of the firm and its specific organizational and environmental factors (Zajac, Kraatz & Bresser, 2000). The multinational corporation (MNC) is confronted with additional challenges when making strategic decisions, such as the international transferability of strategic resources (Cuervo-Cazurra, Maloney & Manrakhan, 2007), adding to the
complexity of cross-border firm activity. Strategies that have proven to be successful in the domestic market often cannot be translated directly to foreign markets (Huo & McKinley, 1992). Such international markets generally demand modified strategies, different from the approach used in the home country environment (Svensson, 2001). The author Ghemawat (2003, 2013) refers to this as semi globalization, where successful cross-border strategies take advantage of international similarities, but also address the critical differences. In other words, firms are often required to adapt their behavior to the international context in which they operate (Calof & Beamish, 1995). Common examples of challenges in foreign markets are: distinct customer needs (Javalgi & White, 2002), differences in the institutional environment (Peng, 2002) and conflicting business cultures (Trompenaars, 1996).
However, it is not only the foreign market or host country environment that determines the effectiveness of corporate strategies. Scholars have examined the role of the home country environment of firms as well. In a recent study, Wan & Hoskisson (2003) describe the
relationship between home country environments, corporate diversification strategies and firm performance. In the paper, it is argued that the diversification strategies that lead to optimal firm performance (measured as both the return on assets and earnings before interest and taxes divided by assets), depend on the home country environment in which the firm is active. Corporations in specific home country environments need to adapt certain strategies in order to achieve optimal firm performance. Thus, according to the paper, the home country
environment of the firm determines the set of efficient strategies that lead to optimal firm performance. According to the study, home country environments consist out of two
elements: factors and institutions. Factors are the elements used for transformational activities and are divided into: endowed factors, advanced factors and human factors. Institutions are the elements that foster transactional activities and consist out of: political institutions, legal
institutions and societal institutions. The literature often claims institutions to be the rules of the game (e.g. Gertler, 2010). Using the terminology of the authors: “Factors are used to produce goods and services, whereas institutions are used for the exchange of inputs and outputs with other firms (Wan & Hoskisson, 2003)”. Because nations or home country environments differ in the amount and nature of factors and institutions they possess, there is a variance in the most efficient strategies that firms can apply (Wan & Hoskisson, 2003). According to Wan & Hoskisson (2003), there are three possible corporate diversification strategies: product diversification (i.e. diversification into more than one product-market combination); inbound international diversification (i.e. cooperation with foreign firms to improve the domestic position); and outbound international diversification (i.e. diversification into foreign markets). To illustrate, Wan & Hoskisson (2003) demonstrated that companies in environments with insufficient factors and institutions are unlikely to benefit from outbound diversification strategies, as the lack of factors and institutions in the domestic environment limits the firms in developing a competitive position in the international market. On the other hand, firms in home country environments with sufficient factors and institutions did have the opportunity to develop a world-class competitive position and thus benefitted from outbound diversification strategies. These examples show the importance of home country
environments in shaping the most efficient strategic route for organizations. Despite the contributions that the paper of Wan & Hoskisson (2003) made to the
international business literature, there still exists a significant research gap that needs to be addressed. The work of the authors focused exclusively on home country environments with either: sufficient factors and institutions or insufficient factors and institutions. Possible hybrid situations such as: sufficient factors but insufficient institutions or vice versa, were not addressed in the study. Wan & Hoskisson (2003) claim to have left this area unexplored, because the authors wanted to test their assumptions in the least complicated manner before addressing more complex situations. Still, the possibility of hybrid home country
environments is a plausible situation. For example, Wan (2005) describes South Korea as a nation with relatively developed factors, in comparison to its relatively underdeveloped institutions. Researching such hybrid environments is important as there exists no empirical evidence for the relationship between corporate diversification strategies and firm
performance in such environments, but there are numerous nations that fit the hybrid home country environment profile (e.g. Hoskisson, Wright, Filatotchev & Peng, 2013). This paper discusses the research performed on the relationship between hybrid home country environments, corporate diversification strategies and firm performance. What is the
relationship between corporate diversification strategies and firm performance in hybrid home country environments? The study has the potential to contribute to the current literature in the following ways. First, to our knowledge, this is the first study to provide empirical evidence for the relationship between corporate diversification strategies and firm performance in hybrid home country environments. Such evidence has both academic and managerial value. From an academic viewpoint, it contributes to our understanding of the moderating effect of home country environments on firm performance. From a managerial perspective, it provides managers with new insights on the most efficient strategies for their specific environment. This is important as the role of the environment is often assumed away in strategic research (Wan, 2005). Second, because we focus on hybrid environments, this means that either the factors or the institutions in the nation are underdeveloped. Understanding the effects of corporate diversification strategies in such settings contributes to our knowledge on firm performance in emerging or rather mid-range emerging economies, the latter being economies to be considered between traditional emerging and developed economies (Hoskisson et al., 2013). Finally, as stated by Wan & Hoskisson (2003), researching hybrid environments is an important intermediate step in understanding and dissecting the individual effects of the variables that together form factors and institutions. Furthermore, it contributes to our understanding by investigating the possible complementary or conflicting roles that factors and institutions can have on each other.
To test our hypotheses, we used secondary data on 400 firms located in one of the two possible hybrid home country environments. Of the sample, 200 firms are located in
environments with sufficient factors, but insufficient institutions. The remaining 200 firms are located in environments with insufficient factors, but sufficient institutions. The companies used in the sample are situated in 8 different countries, evenly distributed among the two types of hybrid home country environments. Ordinary least squares (OLS) regression is used to test the presumed relationship between hybrid home country environments, corporate diversification strategies and firm performance.
The remainder of the paper is constructed as follows. The next section contains a literature review on home country environments, corporate diversification strategies and firm
performance. Furthermore, hypotheses in the case of hybrid home country environments are formulated. Section 3 discusses the research methodology and sample used in the paper. In section 4 we review the results of the study. Subsequently, the discussion is presented in section 5. The final section of the paper contains a conclusion on the performed research.
2. LITERATURE REVIEW AND HYPOTHESES
This section will present a literature review on home country environments, corporate diversification strategies and firm performance. Subsequently, the hypotheses will be formulated. At the end of the section, the conceptual model is presented.
2.1 Home country environments
The literature has identified external market factors and organizational factors and their fit with the environment as the major determinants of firm performance, emphasizing the role of the wider environment in business (Hansen & Wernerfelt, 1989). Furthermore, previous research has shown the impact of the home country environment on strategic choices, such as: entry-mode decisions (Mayrhofer, 2004), locational strategies (Le Bas & Sierra, 2002),
foreign direct investments (FDI) (Luo & Wang, 2012) and even bribery (Martin, Cullen, Johnson & Parboteeah, 2007). In his paper, Cuervo-Cazurra (2011) states that the home countries of firms impact their global strategies in a direct manner by acting as resources and in an indirect manner by inducing firms to create specific resources. The scholars Wan & Hoskisson (2003) emphasize the moderating role of the home country environment on the relationship between corporate diversification strategies and firm performance. It is noted that the home country environment of firms determines which diversification strategies lead to optimal firm performance.
As stated earlier in the paper, home country environments consist out of factors and institutions (Wan & Hoskisson, 2003). The scholars North (1990) and Wan & Hoskisson (2003) refer to this as the opportunity set: the availability of production factors and
institutions in an environment that either enable or constrain firms in their actions. In essence, the competitive advantage and performance of firms will be determined by the degree to which a firm has access to and can exploit its environmental resources (Wan, 2005). Firms can use different corporate diversification strategies to do so and to optimally benefit from their domestic environment (Wan & Hoskisson, 2003).
Countries differ in the amount and nature of factors and institutions they possess. For example, nations can be munificent or deficient in both the factors and institutions they possess (Wan & Hoskisson, 2003). However, there are also examples of countries with developed factors while simultaneously possessing underdeveloped institutions or vice versa (Hoskisson, Wright, Filatotchev & Peng, 2013). In this paper, the focus is on the latter and we distinguish between two types of hybrid home country environments: environments with sufficient factors, but deficient institutions and environments with deficient factors, but
sufficient institutions. The following subsections will discuss the role of the above-mentioned factors and institutions, regarding corporate strategies and firm performance, in more detail.
2.1.1 The role of factors
Factors determine the effectiveness of transformational activities in an environment (Wan & Hoskisson, 2003) and are an important determinant of economic wealth (Hirschman, 1981). That is to say, factors allow for the efficient production of goods and services (Wan & Hoskisson, 2003). The role of factors in the environment and in firms finds its roots in the neoclassical economic growth theory (Wan, 2005). This theory states that firms are economic actors that combine physical capital (i.e. production inputs) and labor (i.e. human capital), both sub-elements of factors, to produce goods or services (Das, Mourmouras & Rangazas, 2015). Based on the paper of Wan & Hoskisson (2003), factors are divided into three categories: endowed factors, advanced factors and human factors. We will now discuss the individual effects of these three categories of factors on the actions of corporations.
Endowed factors are resources derived from the natural environment (Wan & Hoskisson, 2003). Although the presence of natural resources does not per se lead to economic growth (Gylfason, 2001; Sachs & Warner, 2001), their role is important as firms need access to these resources in order to engage in transformational activities. The natural-resource-based view of the firm (NRBV) states that the strategic actions of firms and their links with the natural resource environment can influence firm performance (Hart, 1995; Hart & Dowell, 2011). An important recurring topic in the literature on environmental resources is the curse of natural resources (e.g. Sachs & Warner, 2001; Van der Ploeg, 2011). This phenomenon states that countries that are abundant in their natural resources at times grow more slowly than resource-poor countries (Sachs & Warner, 2001). The scholars Mehlum, Moene & Torvik (2006) claim that the institutions in an environment are responsible in determining if an abundance of natural resources leads to economic growth or not. This finding indicates an interaction effect between factors and institutions. Taking economic factors, such as economic growth, into account is important as these elements partly determine firm performance
(Hansen & Wernerfelt, 1989). In sum, a munificence of endowed factors in the home country environment should benefit firms as these resources allow firms to engage in transformational activities and provide firms the ability to create value (Wan & Hoskisson, 2003).
Advanced factors are defined as the physical infrastructure, capital goods accumulation and financial resources in an environment (Wan & Hoskisson, 2003). Advanced factors have an important role in fostering transformational activities. Investments in the physical
infrastructure of an environment have shown to increase industry-level productivity (Yeaple & Golub, 2007; Démurger, 2001). Furthermore, infrastructure is an important determinant of transport costs which in turn impact the trade flows of nations (Limão & Venables, 2001). In other words, poor infrastructure can put firms in a competitive disadvantage by increasing their costs and consequently decreasing their attractiveness in for example international markets. Investments in capital goods in the form of increased devotement to research and development (R&D) have shown to increase firm-level productivity and has the potential to increase export profits (Aw, Roberts & Winson, 2007). The latter can be explained when looking at the importance of innovation as a source of competitive advantage in international markets (Chadee & Kumar, 2001). Another advanced factor is the financial market. Financial markets are important mechanisms in the efficient allocation of capital (Wurgler, 2010). Financial resources are relevant as firms need access to these resources to make investments and engage in transformational activities, that is to produce goods and services. The shortage of finance is mentioned as one of the key resource constraints that limit firms in their strategic actions (Mahoney & Pandian, 1992). Furthermore, financial constraints have shown to cause underinvestment (Li, Su & Yang, 2012). The author Barney (1986) emphasizes the role of financial resources when engaging in specific strategic behaviors, where insufficient access to finance might limit firms in the strategies they are able to execute. Furthermore, efficient local financial markets stimulate the gains from FDI, this in turn has a positive effect on economic growth (Alfaro, Chanda, Kalemli-Ozcan & Sayek, 2004).
Human factors are reflected by the available labor quality in an environment (Wan & Hoskisson, 2003). Access to superior human capital leads to increased total factor
productivity (Miller & Upadhyay, 2000) and is a necessary condition for innovative capacities (Teixeira & Fortuna, 2004). The educational attainment of the workforce has shown to
influence the absorptive capacity of firms, that is the ability of firms to learn (Vinding, 2006). The learning orientation of organizations is of significant value as it impacts firm performance in a second-order construct, the learning orientation of firms influences innovation which in turn determines firm performance (Calantone, Cavusgli & Zhao, 2002). In addition, human capital is an important determinant of firm performance and can be a source of sustainable competitive advantage because of its inimitable characteristics (Crook, Todd, Combs & Woehr, 2011; Hatch & Dyer, 2004). Finally, the literature has identified the moderating effect of human capital on the relationship between diversification strategies and firm performance, where human capital plays an important role in the implementation and feasibility of
corporate strategies (Hitt, Biermant, Shimizu & Kochar, 2001).
To summarize, building on the abovementioned literature and based on the
conceptualization of Wan & Hoskisson (2003), it can be stated that the munificence of factors in an environment determines how efficient a firm can engage in transformational activities, thereby influencing their strategic options and performance.
2.1.2 The role of institutions
The influence of institutions on economic performance has been well-documented in the literature, most notably in the work of North (1990), and is derived from the theory of institutional economics (e.g. Williamson, 2000). Institutions are often termed to be the rules of the game (e.g. Gertler, 2010) and consist out of formal rules and informal constraints (Scott, 1995). In this sense, institutions define the playing field of firms and determine what they can or cannot do. Institutions facilitate the exchange of inputs and outputs between different parties (Wan & Hoskisson, 2003). The presence of reliable institutions allows firms to interact with unfamiliar, but more efficient parties and permits organizations to engage in more complex transactions (Clauge, 1997; Greif, 1993; Wan & Hoskisson, 2003).
Furthermore, institutions enable specialization benefits because of the presence of market transaction mechanisms (Wan, 2005). Thus, institutions are an important condition for efficient transactional activities. Institutions contribute to this process, for example through better contractual assurance and enforcement (Zhou & Peng, 2010).
In the strategic management literature, the institutional environment is argued to be the third leg of the strategy tripod, together with firm-specific resources and capabilities, and industry-based competition, that together determine strategic outcomes and firm performance (Peng, Sun, Pinkham & Chen, 2009). As an example, the paper of Voss, Buckley & Cross (2010) finds that institutional imperfections in the domestic context influence the internationalization strategies of firms. Here, larger firms take advantage of such institutional flaws whereas smaller firms internationalize to escape domestic institutional constraints. Furthermore, the quality of formal institutions is related to FDI. More specifically, countries with low-quality institutions are more likely to invest in nations with high-quality institutions than vice versa (Seyoum, 2009). The latter indicates that firms in environments with poor institutions might have trouble attracting resources from nations with high-quality institutions. In sum, it can be stated that institutions influence the strategic direction and performance of firms by either enabling or constraining them in their actions (North, 1990; Wan & Hoskisson, 2003). Building on the work of Wan & Hoskisson (2003), in this paper, institutions are divided into three categories: political institutions, legal institutions and societal institutions. We will now
discuss the individual effects of these three categories of institutions on the actions of firms. “Political institutions are the main exogenous force shaping the incentive structure that determines both corruption and the implementation of specific policies (Lederman, Loayza & Soares, 2005)”. Focusing on the latter, the author Persson (2002) wrote an article on how political institutions have proven to shape economic policies. Political institutions also have an effect on FDI inflows, indicating their importance to multinational corporations. Amongst other factors: government stability, corruption and the quality of bureaucracy have shown to be significant determinants of FDI inflows (Busse & Hefeker, 2007). These findings have important implications, as nations with deficient political institutions might find it difficult to attract FDI. Another issue related to political institutions that directly influences firms is corruption. Previous research has indicated the negative effect of corruption on firm performance (Gaviria, 2002) and the foreign investment climate (Davis & Ruhe, 2003). Corruption in the form of bribery undermines the ability to create a competitive private sector (Hardoon & Heinrich, 2011). In turn, the literature has shown that new firm innovation is greater in competitive markets (Katila & Shane, 2005). Thus, firms in corrupt environments might find it difficult to create an international competitive position through innovation. To conclude, political institutions are relevant to transactional activities as they create stable, fair and predictable business environments, therefore impacting interfirm exchanges.
Legal institutions can be defined as the formal rules that determine the process of business transactions (Wan & Hoskisson, 2003). Legal institutions serve to protect the legitimate interest of corporations and parties (Weber, 1978). Furthermore, legal institutions are an important mechanism in ensuring investors’ protection, for example through patent protections. Countries with strong patent protections are more likely to invest in R&D (Varsakelis, 2001), innovation (Allred & Park, 2007) and to capitalize on their investments (Fan, Gillan & Yu, 2013). Research has also shown that if the legal institutions in an
environment are deficient, personal ties and relational contracting can function as substitutes. However, this might restrict business opportunities and negatively impact competition (Zhou, Poppo & Yang, 2008). In sum, legal institutions are an essential element in fostering
transactional activities by providing legal protection and guarantees, acting as a safeguard when dealing with other parties.
Societal institutions reflect the level of trust and interaction in an environment (Knack & Keefer, 1997; Wan & Hoskisson, 2003). Trust is an important element for transactional activities (Schurr & Ozanne, 1985), especially when facing uncertainty (Krishnan, Martin & Noorderhaven, 2006). Moreover, trust complements formal contracts (Poppo & Zenger,
2002). Thus, societal institutions reinforce the formal function of legal institutions. When firms interact and cooperate, trust has proven to be a key determinant of the success of the collaboration (Fink & Kessler, 2010; Cullen, Johnson & Sakano, 2002). Interacting with other firms has several advantages as these firms can act as resources or help to access additional resources and capabilities (Gulati, Nohria & Zaheer, 2000). The advantages of interfirm collaborations finds its support in the relational view (Dyer & Singh, 1998). To conclude, societal institutions foster transactional activities by providing an environment of trust and encouraging interfirm interaction.
To summarize, largely building on the work of Wan & Hoskisson (2003), the overall quality of the institutional environment in which firms are located determines how effective firms can engage in transactional activities with other, even unfamiliar, parties. Furthermore,
institutions can act as enabling or constraining devices that influence the strategic actions of firms. Together, factors and institutions determine the opportunity set that firms can exploit in order to achieve optimal performance. Because firms differ in their opportunity sets, that is the availability and nature of factors and institutions in their home country environments, the optimal strategic decisions of firms differ (Wan & Hoskisson, 2003). The next section will review how firms can apply distinct corporate diversification strategies in order to optimally benefit from their specific home country environment. Furthermore the presumed
relationships between hybrid home country environments, corporate diversification strategies and firm performance will be discussed.
2.2 Corporate diversification strategies
Using the terminology of Wan & Hoskisson (2003), we define corporate diversification strategies as strategic actions that organizations take in order to exploit or compensate for the presence or absence of factors and institutions in the home country environment. Thus,
corporate diversification strategies are strategic decisions that allow firms to optimally benefit from their specific home country environment (Wan & Hoskisson, 2003). Following the literature (Wan & Hoskisson, 2003), we distinguish between three types of corporate diversification strategies: product diversification, inbound international diversification and outbound international diversification. The upcoming subsections will discuss each distinct type of corporate diversification strategies and will formulate hypotheses regarding their effects on firm performance in the case of hybrid home country environments.
2.2.1 Product diversification
Product diversification refers to the degree to which a firm diversifies itself into distinct product-market combinations (Wan & Hoskisson, 2003). According to these authors, product diversification can lead to optimal firm performance in two ways. First, firms can engage in high degrees of product diversification and achieve performance by being active in multiple product-market combinations. Second, firms can engage in low degrees of product
diversification and achieve performance through focus and specialization benefits. The paper of Wan & Hoskisson (2003), which we are extending with hybrid home country
environments, has already shown that the performance of firms in environments with
munificent factors and institutions is negatively impacted by product diversification strategies. Firms in such environments benefit from low levels of product diversification which leads to specialized product-market knowledge and the development of unique capabilities, required to outperform their competitors. Thus, firms in munificent environments achieve optimal firm performance when not engaging in product diversification strategies, but when specializing. On the other hand, the paper of Wan & Hoskisson (2003) shows that firms in environments that lack both factors and institutions do benefit from product diversification strategies. Firms in these environments use product diversification strategies to compensate for the absent factors and institutions in their domestic environment. As an example, the authors mention that by diversifying into multiple product-market combinations firms can replace the lack of external finance by efficiently allocating funds among their multiple internal business units or are able to attract talented human capital because of their broad business scope. But what are the implications of these findings for firms in hybrid home country environments?
In environments with munificent factors and deficient institutions, firms will find it difficult to achieve optimal firm performance when using low degrees of product diversification. To specialize firms are likely to need sufficient access to both factors and institutions. Factors are required to engage in specialized transformational activities. For example, access to superior human capital is needed to engage in innovation (Teixeira & Fortuna, 2004) and as stated earlier functions as a conduit for the learning ability of firms (Vinding, 2006). The latter is important as adequate human capital can foster the acquirement of specialized product-market knowledge by firms. Institutions are needed to protect firms that specialize and innovate in safeguarding their competitive advantages, for example through patent protections (Allred & Park, 2007). If the institutional context does not provide such securities, it is unlikely that firms are willing to invest in innovation and specialization (Varsakelis, 2001). In this case, competitive firms could reap the benefits of the investments made by the innovating firm
without facing the costs, for example through unintended spillovers or imitation (Fan et al., 2013). Thus, in these environments maintaining a specialization strategy might even hurt firm performance as specialization gains might be lost to rival firms without them investing any resources. We expect product diversification to be a more suitable strategy in these
environments. Following the logic of Wan & Hoskisson (2003), firms can use product
diversification strategies to compensate for the institutional deficiencies in their environment. As an example, the authors mention that product diversification can lead to close government ties, as diversified firms influence a wide range of economic sectors. In turn, it is stated that a close relationship with the government allows firm to benefit from state favors such as the ability to introduce institutional entry barriers. Using this strategy, these firms can engage in product diversification and create an advantage by hindering other firms in entering their product-market combinations (Wan & Hoskisson, 2003). Thus, based on the reasoning above, we hypothesize that:
Hypothesis 1a. In hybrid environments with relatively sufficient factors and relatively deficient institutions, product diversification is positively related to firm performance. When looking at firms with home country environments that are deficient in factors but munificent in institutions, organizations face another barrier that hinders them from reaping the benefits of specialization strategies. Although the firms have access to the institutions that stimulate focus and specialization, for example by providing a stable and secure business environment, the lack of factors hampers these firms in engaging in specialized
transformational activities. Again, we expect product diversification strategies to positively influence the performance of firms in such environments. Product diversification enables these firms to compensate for the lack of factors in their home country environment, such as the earlier mentioned substitution of external financial resources or the ability to attract human capital (Wan & Hoskisson, 2003). Maintaining a specialization strategy is unlikely to increase firm performance as the lack of factors hinders the firms to successfully execute this strategy. Thus, following the abovementioned logic, we hypothesize that:
Hypothesis 1b. In hybrid environments with relatively deficient factors and relatively sufficient institutions, product diversification is positively related to firm performance. The hybrid home country environments of firms seem to create an ironic situation when looking at product diversification strategies. Firms that have access to the factors that allow for specialization lack the institutional incentives to do so, whereas firms that do have the
institutional assurance lack the factors to pursue such strategies. In sum, both categories of hybrid home country environments are expected to benefit from high degrees of product diversification, rather than pursuing specialization strategies.
2.2.2 Inbound international diversification
Inbound international diversification refers to the degree to which a firm cooperates with foreign partners in order to improve its domestic competitive position (Wan & Hoskisson, 2003). Thus, firms that maintain high degrees of inbound international diversification have a substantial network of cross-border partners. Inbound international diversification strategies can increase firm performance as they enable firms to: obtain valuable resources it lacks in its domestic environment (Das & Teng, 2000), create complementary resource endowments (Dyer & Singh, 1998) and develop a competitive advantage (Elmuti & Kathawala, 2001). When analyzing home country environments that are munificent in factors but deficient in institutions, we do not expect firms to benefit from inbound international diversification strategies. As shown by Wan & Hoskisson (2003), the performance of firms in factor and institution munificent environments is negatively impacted by this type of corporate
diversification strategy. Organizations in munificent home country environments are unlikely to benefit from their foreign partners. For firms in munificent home country environments it is safer to develop the required resources or capabilities on their own, using the factors available in their domestic context, rather than with foreign firms (Wan & Hoskisson, 2003). Because the institutional environment is deficient, the risk of unintended spillovers to the foreign partner is high (Fan et al., 2013). Thus, when cooperating with unknown parties, firms in environments with sufficient factors but insufficient institutions are in a vulnerable position. In essence, these firms have more to lose to their partner than to gain from their partner, especially because the factors needed for transformational activities are widely available in their domestic environment (Wan & Hoskisson, 2003). The risk that the foreign partner will parasite on the home country firm to extract their resources is substantial due to the weak institutional environment. It is likely that the coordination costs of the strategic alliance, which are amplified by the institutional uncertainty (Artz & Brush, 2000), exceed the potential gains from the partnership. Thus, we expect that:
Hypothesis 2a. In hybrid environments with relatively sufficient factors and relatively deficient institutions, inbound international diversification is negatively related to firm performance.
In hybrid home country environments with deficient factors and sufficient institutions firms can use inbound international diversification strategies to improve firm performance. Firms can use strategic alliances with foreign partners in order to access the resources that they lack in their domestic context (Dyer & Sing, 1998). Furthermore, the institutional environment in these nations stimulates interaction and transactional activities between organizations as they provide the partners with transactional security, for example through contractual reassurance (Zhou & Peng, 2010). Thus, firms in these environments can use foreign partners to
compensate for the lack of factors in their domestic environment while the institutional environment provides the transactional assurance. This is important as foreign partners might want to prevent unintended spillovers and the institutional environment limits this option as firms face higher legal risks when undertaking activities outside the scope of the alliance (Fan et al., 2013). So why would foreign partners be willing to cooperate with a factor-poor firm? Foreign partners can benefit from the cooperation by for example gaining access to valuable local market knowledge, while the domestic partner benefits through the access to the foreign partner’s resources and technologies (Inkpen & Beamish, 1997; Wan & Hoskisson, 2003). Hence, we expect that:
Hypothesis 2b. In hybrid environments with relatively deficient factors and relatively sufficient institutions, inbound international diversification is positively related to firm performance.
2.2.3 Outbound international diversification
Outbound international diversification refers to the degree to which a firm diversifies itself into foreign or international markets (Wan & Hoskisson, 2003). Outbound international diversification can increase firm performance through for example additional
internationalization profits, such as: export profits (Sousa, 2004) or supplementary profits generated by foreign subsidiaries. In home country environments with sufficient factors and deficient institutions, we expect outbound international diversification to have a negative impact on firm performance. As stated before, firms in these environments have trouble innovating and specializing due to the deficient institutions in their home country
environment. Since innovation (Chadee & Kumar, 2001) and the level of technology (Zeng, Xie & Tam, 2008) are an important source of competitive advantage in international markets, it is unlikely that these firms are able to outcompete their competitors in foreign markets (Wan & Hoskisson, 2003). Hence, we hypothesize that:
Hypothesis 3a. In hybrid environments with relatively sufficient factors and relatively deficient institutions, outbound international diversification is negatively related to firm performance.
The negative effect of outbound international diversification on firm performance is expected to be applicable to home country environments with deficient factors and sufficient institutions as well. As stated earlier, firms in these environments lack the factors to engage in specialized transformational activities. These firms are unlikely to have developed an
international competitive position which allows them to outperform their cross-border rivals, a situation comparable to the less munificent home country environments of Wan &
Hoskisson (2003). Thus, we hypothesize that:
Hypothesis 3b. In hybrid environments with relatively deficient factors and relatively sufficient institutions, outbound international diversification is negatively related to firm performance.
2.3 Conceptual model
The research problem and presumed relationships are visualized in Figure 1. As illustrated, firms can use three different corporate diversification strategies: product, inbound
international and outbound international. The use of corporate diversification strategies affects firm performance. The relationship between corporate diversification strategies and firm performance is moderated by the home country environment of the firm. In this research, the home country environment of the firm can take on one out of two forms, it can possess: sufficient factors and deficient institutions or deficient factors and sufficient institutions. The upcoming section will discuss the research methods used in this paper.
FIGURE 1 Conceptual Model
3. RESEARCH METHODS
In this section, the country and firm samples and their selection methods are discussed. Subsequently, the data sources and measurement of the variables is presented. Finally, the method of analysis and linear regression assumptions are discussed.
3.1 Sample
The sample selection process used in this paper is twofold. First, the country sample that represents both forms of our hybrid home country environments is selected. Second, a firm sample from these nations or environments is determined. The following subsections will discuss both sampling strategies in detail.
3.1.1 Selecting the country sample
For our initial country sample, we draw on the paper of Hoskisson et al. (2013) on emerging multinationals from mid-range economies. In this paper, the authors define mid-range
economies as environments with relatively well-endowed factor markets and infrastructure, while simultaneously possessing relatively underdeveloped institutions or vice versa. Thus, this definition is comparable to our conceptualization of hybrid home country environments. The scholars use cluster analysis, more specific an adaptive version of K-means clustering, to identify which nations belong to the mid-range economies using a sample of 60 countries. However, because we have no reason to assume that nations are static in their development of factors and institutions and because the measurements used by Hoskisson et al. (2013) differ from our operationalization, we recalculate the factor and institution scores to determine our hybrid home country environments. Due to data limitations 6 nations were filtered from the initial sample of 60 nations used by Hoskisson et al. (2013). Thus, our initial country sample consists out of 54 countries.
The variables used to determine the hybrid home country environments are measured using the following operationalization, which is largely based on the paper of Wan & Hoskisson (2003). Endowed factors are measured using the natural resource rents published by The World Bank in the Wealth Accounting dataset. The natural resource rents represent the contribution of natural resources to the wealth of a specific nation, thereby indicating the abundance of natural resources in an environment and how efficient a nation uses its natural resources. The advanced and human factors are determined using the Global Competitiveness Report 2014-2015 published by the World Economic Forum. For each of these factors, we selected a group of indicators that represented them. The advanced factors are represented by the indicators: infrastructure, financial market development, innovation and technological
readiness. The human factors are measured using the indicators: higher education and training and labor market efficiency. For both the advanced factors and the human factors we calculate a single index score by calculating the average of the individual indicators. To illustrate, the final score of advanced factors is the mean of the pillars: infrastructure, financial market development, innovation and technological readiness. Political institutions are measured using the Corruptions Perceptions Index 2014 published by Transparency International. This index measures the level of public sector corruption in a nation. Legal institutions are captured using the International Property Rights Index 2013, which measures the legal and political
environment, physical property rights protection and intellectual property rights protection in a country. Finally, societal institutions are measured using the Global Competitiveness Report 2014-2015 and more specifically the pillar: business sophistication. This index measures the quality of business networks in a country and the extent of their (social) interaction. The variables and their data sources used for the classification of hybrid home country environments are summarized in Table 1 below.
TABLE 1
Variables Used to Determine Hybrid Home Country Environments
Variables Description Data sources
1. Factors
1a. Endowed Natural resources Wealth Accounting 2012
1b. Advanced Infrastructure, capital goods and financial resources
Global Competitiveness Report 2014-2015
1c. Human Workforce quality Global Competitiveness Report
2014-2015
2. Institutions
2a. Political Public sector corruption Corruption Perceptions Index 2014
2b. Legal Legal efficiency International Property Rights
Index 2013
2c. Societal Trust and interaction Global Competitiveness Report
2014-2015
Because the variables are measured using different scales and because we want to create a single composite variable for factors and another composite variable for institutions, we first standardize the variables using Z-scores. After the standardization process, we create the two
composite variables. The factor composite variable consists out of: endowed factors, advanced factors and human factors. The institution composite variable consists out of: political institutions, legal institutions and societal institutions. The composite variables are then analyzed in order to determine which nations possess a hybrid home country
environment. To determine which factor and institution scores are classified as sufficient and which as deficient, we follow the work of Wan & Hoskisson (2003). In their paper, the authors use mean- and median split criteria to determine which environments score sufficient and which score deficient. Using this method, 10 nations out of the 54 nations were identified as having hybrid home country environments. To validate this sample, we also followed the K-means clustering method used in the article of Hoskisson et al. (2013). The hybrid home country environment classification of the abovementioned nations were confirmed using an additional K-means cluster analysis. The first group of nations with sufficient factors and deficient institutions consists out of: Azerbaijan, Kazakhstan, Russia and Trinidad and Tobago. The second group of nations with deficient factors and sufficient institutions is composed out of: Botswana, Ghana, India, Slovakia, Sri Lanka and Turkey. Due to time limitations and to test our hypotheses in more detail, the final sample size has been limited to 8 nations, evenly distributed amongst the two types of hybrid home country environments. Following the work of Wan & Hoskisson (2003) the final nations were selected based upon: availability of firm level data and unambiguous group classification. Thus, our final country sample consists out of 8 nations of which 4 nations have sufficient factors and deficient institutions whereas 4 nations possess deficient factors and sufficient institutions. Based on the abovementioned criteria, the nations selected for home country environments with
sufficient factors and deficient institutions are: Azerbaijan, Kazakhstan, Russia and Sri Lanka. The selected nations for environments with deficient factors and sufficient institutions are: Botswana, Ghana, India and Turkey. An overview of the cluster analysis output and the classification of nations using mean- and median split criteria can be found in Appendix A.
3.1.2 Selecting the firm sample
The firms sample consists out of the largest firms (according to operational revenue in USD) in each of the selected nations, obtained via the Orbis database. To control for accounting fluctuations, we only included the firms that documented at least 3 years of financial data in the Orbis database. The total sample size will consist out of 400 firms, where 200 firms are located in nations with sufficient factors and deficient institutions and the remaining 200 firms are situated in environments with deficient factors and sufficient institutions. For
environments with sufficient factors and deficient institutions the distribution is as follows: Azerbaijan (n=27), Kazakhstan (n=61), Russia (n=62) and Sri Lanka (n=50). For
environments with deficient factors and sufficient institutions the firms are distributed as follows: Botswana (n=29), Ghana (n=38), India (n=50) and Turkey (n=83). The distribution of firms is not equal amongst the nations as we only included the firms that documented at least 3 years of financial data in order to control for accounting fluctuations. The total sample size of 400 units is sufficient according to Thomas (2004) who suggests a threshold of 200 units when performing data analysis.
3.2 Data and measurement
The variables used to test the hypotheses are summarized in Table 2. The operationalization is largely based on the work of Wan & Hoskisson (2003). We use secondary data sources because the data required for the analysis is available in online databases and reports.
Collecting primary data would consume more time and resources without adding significantly more value than the secondary data. The dependent variable firm performance is measured as the return on assets (ROA) and as the log of profit and loss before taxes in USD1 (Baum & Rowley, 2008; Bogan & Sandler, 2012; Booth, Cornett & Tehranian, 2002; Cumming, 2010; Faria, Martins & Brandão, 2013). This data is found in the Orbis company database.
Following the authors Wan & Hoskisson (2003) we average the last 3 years of financial performance available in the database to avoid fluctuations in the accounting data. We use the term optimal firm performance in our thesis as we want to analyze which corporate
diversification strategies firms need to use or avoid in order to achieve maximum
performance. The independent variable corporate diversification strategy is measured using data from the Orbis and Zephyr company databases. Product diversification is captured using industry codes, to be more specific NACE Rev. 2 codes, representing the diversification into distinct product-market combinations. We use the number of primary and secondary industry codes linked to the firm to determine the degree of product diversification. The data for this variable is found in the Orbis database. Inbound diversification in measured as the number of cooperative ventures the domestic firm has with foreign partners. We do so by counting the number of international joint ventures the firm has, using the Zephyr database. Finally,
outbound diversification is measured as the number of foreign subsidiaries the firm possesses. The number of foreign subsidiaries of a firm can be found in the Orbis database. For the
1 Adding a constant prior to the log transformation (Norman & Streiner, 2008) did not change the outcome of the hypotheses
testing.
moderator variable we use a dummy variable to represent the two distinct groups of hybrid home country environments. We add two control variables to our research. First, firm size is controlled for in our analysis. Firm size is measured as the log of the total operational revenue of the firm in USD, this data is obtained from the Orbis database. We control for firm size because of its effects on firm performance and profitability rates (Hall & Weiss, 1967). We use the log because revenues and profit and loss before taxes are not normally distributed. Finally, we control for industry effects using the BVD major sector codes classifying the core industry of the firm. This data is found in the Orbis database. We control for industry effects as firm performance standards can vary across different industries (Mauri & Michaels, 1998). Dummy variables are used to control for this element. To avoid a perfect linear relationship with the constant, we omit one category for the industry control variable. In the paper, the industry category: primary sector, is used as the reference group and is therefore the omitted industry category. For the industry control variable we have deleted the categories: education and health, public administration and wood as no observations of these categories were found in our sample. The remaining industry categories used in the paper are: food, textiles,
publishing, chemicals, metals, machinery, gas, water and electricity, construction, wholesale, hotels, transport, communications, banks, insurance and other services.
TABLE 2
Overview of the Variables and Data Sources
Variables Description Data sources
Dependent
1. Firm performance Return on assets (ROA) and profit and loss before taxes in USD (log)
Orbis
Independent
2. Diversification strategies
2a. Product Product-market combinations
measured using NACE Rev. 2 primary and secondary industry codes
Orbis
2b. Inbound international Cooperation with foreign partners measured as the number of international joint ventures
Zephyr
2c. Outbound international Diversification into foreign markets measured as the number of foreign subsidiaries
Orbis
Moderator
3. Home country environment Factors and institutions Composite scores
Control
4. Firms size Total operational revenue of
the firm in USD (log)
Orbis
5. Industry effects Core industry of the firm using BVD major sector codes
Orbis
3.3 Method of analysis
Building on the work of Wan & Hoskisson (2003), we use ordinary least squares (OLS) regression analysis to test our hypotheses. There are four regression models run in this study. The first and second model are used for the nations with sufficient factors and deficient institutions using ROA and profit and loss before tax as performance measures respectively. The third and fourth model are used for the nations with deficient factors and sufficient institutions using ROA and profit and loss before tax as performance measures respectively. The main reason why we use multiple performance measures is because there is no agreement in the literature on which performance measure is most suitable. Furthermore, although ROA is considered to be an appropriate performance measure it has its drawbacks. Most notably, ROA does not account for intangible assets (e.g. patents, knowledge or reputation), that can be valuable to companies, when analyzing assets. Hence, we have chosen to add an additional performance measure to our analysis. The analysis will be performed using the IBM SPSS Statistics 20 software. The regression equation (1) is as follows:
3.4 Linear regression assumptions
The assumptions of linear regression are tested to ensure the reliability of the results. The output for both dependent variables can be found in Appendix B. The distribution of the residuals for the dependent variable ROA does not perfectly follow a normal distribution. The P-P plot shows a slight S-shape distribution for this variable. This means that the normality assumption for ROA is not perfectly met. The most obvious step in this case is to transform the non-normal data. However, unlike the other dependent variable, we could not find support in the literature to transform the performance measure ROA using for example a log
transformation. Hence, we have decided to accept the data limitation. For the dependent variable profit and loss before tax, the observations in the P-P plot lie reasonably close to the diagonal line. Hence, it can be assumed that the residuals are fairly normally distributed. The histogram also has a bell-shaped distribution, suggesting an approximate normal distribution
Performancei = α + β(DIVi) + γ(Di)+ γc(Xi) + εi (1) Performance = Firm performance γ = Coefficient of dummy variable
i = Country γc = Coefficient of control variable
α = Intercept X = Vector of control variable
β = Coefficient of independent variable ε = Error term
DIV = Diversification strategy D = Dummy variable
of the residuals. Furthermore, the scatter plot shows a fairly rectangular distribution of the residuals. This does not indicate severe violations of the homoscedasticity assumption.
4. RESULTS
In this section, the results of the data analysis are presented. First, the descriptive statistics will be discussed. Second, the correlation matrix will be analyzed. Finally, the output of the OLS regression analysis is presented and the hypotheses are briefly discussed.
4.1 Descriptive statistics
When we compare the descriptive statistics of the two different types of hybrid home country environments, summarized in Table 3, some interesting facts can be noted. First, on average, the firms in environments with sufficient factors and deficient institutions perform better (ROA mean: 10.45 and P/L mean: 11.65) than the firms in environments with deficient factors and sufficient institutions (ROA mean: 6.19 and P/L mean: 10.30) when using either performance measure. Second, firms in home country environments with sufficient factors and deficient institutions are on average larger (firm size mean: 13.86) than the firms in environments with deficient factors and sufficient institutions (firm size mean: 12.54). Finally, when analyzing the different diversification strategies, it can be noted that product diversification and inbound international diversification is on average used more in
environments with sufficient factors and deficient institutions (product diversification mean: 4.29 and inbound international diversification mean: .85) than in environments with deficient factors and sufficient institutions (product diversification mean: 2.29 and inbound
international diversification mean: .46). Outbound international diversification, however, is on average more common in environment with deficient factors and sufficient institutions
(outbound international diversification mean: 6.44) than in environments with sufficient factors and deficient institutions (outbound international diversification mean: 3.28).
4.2 Correlations
The Pearson correlations for the variables of the different types of hybrid home country environments are also summarized in Table 3. The correlation matrix does not indicate any issues with multicollinearity. The values used for this issue range from >.50 to >.70 and even the value >.90 is used by some researchers (i.e. Field, 2009). This paper maintains the value of >.70 to see if there are multicollinearity issues and it can be concluded that such issues are not observed in Table 3. In the correlation matrix only the control variable firm size correlated with the dependent variable profit and loss before taxes exceeds this value in both hybrid home country environments with a value of .79 for environments with sufficient factors and deficient institutions and a value of .80 for environments with deficient factors and sufficient institutions. The VIF values (variance inflation factor) also did not exceed the threshold of 10.
TABLE 3a
Descriptive Statistics and Correlations
(3a) Home country environments with sufficient factors and deficient institutions
Variable Mean s.d. n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 1. ROA 10.45 16.02 200 2. P/L 11.65 2.16 179 .28 3. Firm size 13.86 2.18 200 .03 .79 4. Food .04 .18 200 .24 -.10 -.13 5. Textiles .01 .07 200 .00 -.07 -.06 -.01 6. Publishing .00 .00 200 . . . . . 7. Chemicals .07 .25 200 .09 -.07 .08 -.05 -.02 . 8. Metals .07 .25 200 -.05 .11 .14 -.05 -.02 . -.07 9. Machinery .03 .17 200 -.04 -.12 -.03 -.03 -.01 . -.05 -.05 10. Gas, water & electricity .03 .17 200 .00 .05 .02 -.03 -.01 . -.05 -.05 -.03 11. Construction .02 .12 200 .01 -.01 .02 -.02 -.01 . -.03 -.03 -.02 -.02 12. Wholesale .12 .32 200 .03 -.10 .08 -.07 .03 . -.10 -.10 -.06 -.06 -.04 13. Hotels .01 .07 200 -.01 -.04 -.04 -.01 -.01 . -.02 -.02 -.01 -.01 -.01 -.03 14. Transport .06 .24 200 -.09 .07 .17 -.05 -.02 . -.07 -.07 -.04 -.04 -.03 -.09 -.02 15. Communications .04 .18 200 .05 .10 .08 -.04 -.01 . -.05 -.05 -.03 -.03 -.02 -.07 -.01 -.05 16. Banks .28 .45 200 -.36 -.18 -.37 -.12 -.04 . -.16 -.16 -.11 -.11 -.08 -.22 -.04 -.16 -.12 17. Insurance .01 .10 200 -.01 -.05 -.08 -.02 -.01 . -.03 -.03 -.02 -.02 -.01 -.04 -.01 -.03 -.02 .06 18. Other services .08 .27 200 -.04 -.10 .04 -.06 -.02 . -.08 -.08 -.05 -.05 -.04 -.11 -.02 -.08 -.06 -.18 -.03 19. Product diversification 4.29 3.37 200 .01 .18 .18 -.06 -.07 . .08 .03 .03 .01 .04 .00 -.05 -.07 -.04 -.01 -.07 -.05 20. Inbound diversification .85 3.13 200 -.04 .39 .36 -.05 -.02 . .05 .16 -.04 .03 -.03 -.09 .00 .22 .03 -.12 -.01 -.03 -.09 21. Outbound diversification 3.28 11.28 200 -.07 .31 .32 -.05 -.01 . -.04 .05 -.04 -.01 -.04 -.08 .07 .04 .03 -.04 -.02 .01 -.10 .59
a Primary sector is the omitted industry category for both hybrid home country environments.
Primary sector is the omitted industry category for both hybrid home country environments.
Due to data limitations the number of observations for the two performance measurements is different. Furthermore, Table 3b has no observations of transport or P/L for construction firms in the sample. Correlations in bold are significant at p < .05 (two-tailed).
(3b) Home country environments with deficient factors and sufficient institutions
4.3 Regression analysis results
The results of the OLS regression analysis are given in Table 4. Below, the regression output will be discussed and will be linked to the earlier formulated hypotheses. As noted earlier, there are four regression models which are run in this study. In the analysis, either return on assets (ROA) or profit and loss before taxes in USD (P/L) is taken as the
performance measure for the dependent variable. The models include all control variables that are used in the study.
4.3.1 Product diversification
The first independent variable to be discussed is product diversification, which refers to the degree that a firm diversifies itself into distinct product-market combinations. Hypothesis 1a predicted that the performance of firms in environments with sufficient factors and deficient institutions is positively impacted by product diversification. Contrary to our prediction, the models A and B both show that product diversification has a negative relationship with firm performance in such home country environments. The standardized beta coefficients are -.013 for model A and -.019 for model B. However, the results are not statistically significant. Thus, hypothesis 1a is not supported.
Hypothesis 1b stated that the performance of firms in environments with deficient factors and sufficient institutions is positively impacted by product diversification. In model C, product diversification has a negative relationship with firm performance (ROA measure). In model D, product diversification has a positive relationship with firm performance (P/L measure). The standardized beta coefficients are -.037 for model C and .066 for model D. However, both results are not statistically significant. Hence, hypothesis 1b is not supported.
4.3.2 Inbound international diversification
The second independent variable to be discussed is inbound international diversification, which refers to the degree that a firm cooperates with foreign partners in order to improve its competitive position at home. Hypothesis 2a stated that the performance of firms in
environments with sufficient factors and deficient institutions is negatively impacted by inbound international diversification. Contrary to our expectation, the output shows that inbound international diversification has a positive and statistically significant (p <.05) relationship with firm performance as illustrated in model B (P/L measure). The standardized beta coefficients are .032 for model A and .132 for model B. Thus, hypothesis 2a is not supported as the coefficient sign is in the opposite direction.
factors and sufficient institutions is positively impacted by inbound international diversification. As predicted, the regression table shows that inbound international diversification has a positive and statistically significant (p <.10) relationship with firm performance as illustrated in model D (P/L measure). The standardized beta coefficients are -.003 for model C and .113 for model D. Hence, hypothesis 2b is supported.
4.3.3 Outbound international diversification
The third independent variable to be discussed is outbound international diversification, which refers to the degree that a firm diversifies itself into foreign or international markets. Hypothesis 3a predicted that the performance of firms in environments with sufficient factors and deficient institutions is negatively impacted by outbound international diversification. The coefficients of outbound international diversification are indeed negative in models A and B. The standardized beta coefficients are -.087 for model A and -.050 for model B. However, the results are not statistically significant. Thus, hypothesis 3a is not supported.
Hypothesis 3b stated that the performance of firms in environments with deficient factors and sufficient institutions is negatively impacted by outbound international diversification. The coefficient of outbound international diversification is positive in model C and negative in model D. The standardized beta coefficients are .024 for model C and -.010 for model D. However, the results are not statistically significant. Hence, hypothesis 3b is not supported.
4.3.4 Brief overview and remarks
To summarize, we have found a statistically significant and positive relationship for
hypothesis 2a. However, for this case we predicted a negative relationship with the dependent variable. We found support for our hypothesis 2b with a statistically significant and positive relationship with the dependent variable. We did not find statistically significant results for our hypotheses 1a, 1b, 3a and 3b. When analyzing the control variables, the variable firm size seems to be the strongest and most significant predictor as illustrated in the models B, C and D. The effects of the control variable industry varies amongst the different types of hybrid home country environments and performance measures. Furthermore, when looking at the R-squared values, the models that use profit and loss before taxes in USD as the dependent variable are better in predicting the proportion of variance from the independent variables on the dependent variable than the models that use ROA as the dependent variable in both types of hybrid home country environments.
TABLE 4a
Results of the OLS Regression Analysis of Home Country Environment and Corporate Diversification Strategy Effects on Firm Performance
Home Country Environments with Sufficient Factors and Deficient Institutions Home Country Environments with Deficient Factors and Sufficient Institutions
Dependent Variable Model A: ROA Model B: P/L Model C: ROA Model D: P/L
Product diversification -.063 -.013 (.317) -.012 -.019 (.029) -.256 -.037 (.600) .104 .066 (.085) Inbound international diversification .162 .032 (.432) .087** .132 (.037) -.028 -.003 (.765) .198* .113 (.103) Outbound international diversification -.123 -.087 (.117) -.009 -.050 (.010) .010 .024 (.036) -.001 -.010 (.005) Firm size -.481 -.066 (.575) .862*** .840 (.056) .902** .195 ( .423) .813*** .762 (.060) Food 7.157 .083 (6.033) -.383 -.034 (.515) 7.405 .126 (6.289) .825 .050 (.979) Textiles -12.179 -.057 (14.577) -1.067 -.037 (1.238) -7.866 -.078 (8.531) -.593 -.023 (1.314) Publishing . . . . . . .412 .004 (9.100) -.473 -.018 (1.385) Chemicals -5.982 -.093 (4.687) -.726* -.088 (.399) -5.096 -.099 (5.838) -.938 -.071 (.880) Metals -14.060*** -.219 (4.715) -.769* -.082 (.444) -8.025 -.136 (6.424) .009 .001 (.899) Machinery -15.319** -.165 (6.347) -1.288** -.099 (.586) -2.680 -.054 (5.994) -.752 -.069 (.838) Gas, water & electricity -11.936*
