The moderating role that psychic distance plays in explaining the relationship between international ambidexterity and firm performance

(1)

The moderating role that psychic distance

plays in explaining the relationship between

international ambidexterity and firm

performance

Sebastiaan Draijer 5904699 August 15, 2014 Academic year: 2013-2014 Final version

First supervisor: C. Gelhard

Second supervisor: M. Westermann-Behayl

MSc in Business Studies - International Management Faculty of Economics and Business

University of Amsterdam Number of words: 12728

(2)

Abstract

Explorative and exploitative FDI activities determine the location choice of a firm. This study investigates how psychic distance – the distance between the home market and a foreign market, resulting from the perception of both cultural and business differences – influences the effect of international ambidexterity - the simultaneous execution of exploration and exploitation activities - on firm performance. Psychic distance is measured along four

dimensions: cultural, political, geographic and economic. Research suggests that firms with a tendency to international ambidexterity in their strategic decision-making perform better. Furthermore, scholars suggest that distance creates barriers – and thus risks and costs - for firms when doing business abroad. It could therefore be that psychic distance moderates the effect of international ambidexterity on firm performance.

To investigate the hypotheses, data is collected of 2.636 subsidiaries, related to 50 Japanese MNEs listed at the Nikkei-225 in 7 industry sectors, and in 105 host countries. The distance between home country, Japan, and the host countries (locations of the subsidiaries) is measured for each firm. Further quantitative analysis is done with SPSS.

Results give support for the expected positive effect of international ambidexterity on firm performance. Although, this is only the case for effective firm performance, international ambidexterity has a negative effect on efficient firm performance. Economic distance

negatively influences this effect on efficient firm performance; while cultural and political distance even abolish the positive effect on effective performance. Thus, it shows that international ambidexterity is vulnerable for unfamiliarity with the environment.

(4)

1

Chapter 1: Introduction

Conventional theories in the research field of international management view

internationalization as the exploitation of a firm’s competitive advantages in foreign markets. Several scholars state that internationalization is not only driven by its exploitation of existing advantages, but also by its will to explore and capture resources in foreign markets that strengthen long-term performance and corporate global competitiveness (Hsu et al, 2013; 58). The Uppsala model states that internationalization is an incremental process due to

differences between the home country of the firm and host countries. The process is

incremental because firms miss the knowledge for foreign expansion and this knowledge can only be acquired through operations abroad. This also means that the pace and pattern of internationalization are determined by business and economic factors (Johanson and Valhne, 1977; 23). This implies that firms who have more experiential knowledge of foreign markets expand quicker (Oviatt and McDougall, 1977; 23).

These factors can be captured in the concept of psychic distance. This is the distance between the home market and a foreign market, resulting from the perception of both cultural and business differences (Evans and Mavondo, 2002; 517).Greater unfamiliarity with the environment, from political, economic and cultural differences, between other factors, can reduce the efficiency of the internationalization of a firm’s competitive advantages. These differences, described as the liability of foreignness, create barriers to entry for foreign firms - which results in costs and risks. Firm-specific advantages as organizational capabilities could help multinationals (MNEs) to overcome the liability of foreignness, and make it possible to compete successfully with local firms (Zaheer, 1995; 341; Ghemawhat, 2001; 3).

These firm-level advantages have put emphasis of international business research on the resource-based view (RBV). The resource-based view is originating from strategic management literature. The model states that heterogeneous and immobile resources can become sources for establishing competitive advantages. A resource needs to fulfill four requirements – valuable, rare, imperfectly imitable, and not substitutable – to give a competitive advantage (Barney, 1991; 101, 102, 105, 106). Firms who have resources or capabilities superior to their competitors, can establish a competitive advantage if they match properly to the environment (Peteraf, 1993; 179). According to the dynamic capabilities perspective, firms will try to build, integrate, and to handle rapid changing environments (Hsu & Chen, 2009; 589; Teece et al, 1997; 510). Firm-level variables are related to the concept of transaction costs. These costs exists when a firm is interrupted by market failures to transfer

(5)

2

specialized assets towards its subsidiaries. The process of transferring a firm’s specific advantages between borders can be viewed as exploitative foreign direct investment; while acquiring strategic resources in a host country can be viewed as explorative foreign direct investment (Hsu & Chen, 2009; 589).

Contemporary multinational enterprises show a significant tendency towards

international ambidexterity in their strategic decisions. International ambidexterity means that firms strive to balance their exploitative and explorative foreign direct investment activities during the internationalization process (Hsu et al, 2013). Previous research showed that firms who are capable of performing exploitive and explorative activities simultaneously during the internationalization process, perform better than firms focusing on one at the expense of another (Luo & Rui, 2009; Prange & Verdier, 2011; Tushman and O’Reilly, 1996; Hsu et al, 2013).

Prior research showed that for firms from small emerging economies, international ambidexterity is highly vulnerable for environmental complexity, sensitive to international experience and international capability (Hsu et al, 2013). Hsu et al (2013) measured the influence of the degree of internationalization on the relationship between ambidexterity and firm performance. The degree of internationalization was measured with use of the

geographic FDI distribution and the number of FDIs. An explanation for this is that a high degree of internationalization weakens the institutional relatedness - network of ties with a nation’s dominant institutions - of a firm, consequently the advantage of practicing

ambidexterity declines. Although, this may be because of specific features of firms from small emerging economies and their particular FDI patterns (Hsu et al, 2013; 66). Also it doesn’t make clear how different dimensions of distance influence the effect of international ambidexterity.

This research addresses this gap and isolates these dimensions through the empirical analysis of 2.636 subsidiaries related to 50 Japanese multinational enterprises, from different industries, ages and sizes. More specifically, the purpose of this study is to investigate if psychic distance, caused by cultural, political, geographic, and economic differences between countries, moderates the effect of international ambidexterity on firm performance.

To do this, I have constructed a database which contains information about the subsidiaries (exploration/exploitation/combined), the different dimensions of distance and firm

performance. International ambidexterity is measured through the extent of imbalance between the number of explorative and explorative subsidiaries. The average extent of distance a multinational faces for all his international locations is also calculated. The sample

(6)

3

consists of Japanese firms because of three reasons. First, Japan differs from countries used by previous research (small and emerging versus large and developed) on international ambidexterity. Second, it’s likely that some of the Japanese multinationals have an

ambidextrous structure. Third, Japan is probably relatively distant from other countries in a psychic sense - which makes it easier to measure the effect of distance on the entry mode choice of multinational enterprises.

The remainder of this thesis is structured as followed. First, the key literature of this research will be discussed. The resource-based view (RBV) is reviewed because it helps to determine how firms exploit their existing resources and explore new resources in host countries through FDI. Discussing internationalization will help to understand the process of foreign expansion. The attractiveness of host countries will be considered with use of the concept distance. The literature review will end with the theoretical foundation for the relationship between firm performance and international ambidexterity. Subsequently, the hypotheses are formed. The next section discuss the data and methods used, a sample of 50 Nikkei-225 listed MNEs were selected. Qualitative data is collected through secondary sources about (a) the subsidiaries related to the MNEs, in order to measure international ambidexterity, (b) the firm performance of the MNEs, and (c) the differences between the host countries of the subsidiaries and home country Japan, to measure distance. The validity of the hypotheses is reviewed in the results section. This thesis concludes with a summary of the key findings, followed by the limitations, scientific relevance and managerial

(7)

4

Chapter 2: Literature Review

In the research field of international management several scholars have studied the

motivations for multinationals (MNEs) to invest in foreign markets. Conducting a business abroad can be attractive, for example, because of access to new resources or to promote a more efficient division of labor (Dunning, 2000; 164, 165). The resource-based view (RBV) helps to address how firms explore new resources and exploit their existing resources in host countries through FDI (Makino et al, 2002; 404). The attractiveness of foreign market opportunities will be considered with use of the different dimensions of psychic distance.

This section starts with the process of internationalization. After that I will describe the resource-based view, followed by how distance between countries can affect foreign direct investment. Next, I explain how balancing explorative and exploitative activities influence firm performance. To end, I will develop my hypotheses.

2.1 Internationalization process

International business and global strategy research have long been focusing on strategies of internationalization (Prange & Verdier, 2000; 126). Most researchers have described

internationalization as the outward movement in a firm’s international operations (Andersen, 1997; 28). Thence, internationalization is a major dimension of the ongoing strategy of most business firms (Melin, 1992; 101). Internationalization differs from other types of strategy processes (or growth strategies) because of two reasons. First, the firm transfers resources, products and services across national boundaries. This implies that the firm has to select in which countries the transactions should take place. Second, the firm has to select the foreign market entry strategy. Both dimensions, choice of entry mode and international market

selection, represent the key strategic decisions with regard to the internationalization of a firm (Andersen, 1997; 29).

The pace and the pattern of this process is influenced by business and economic factors. Johanson and Valhne (1977; 23) developed a model to consider the firm’s

internationalization process, known as the Uppsala model, that “focuses on the development of

the individual firm, and particularly on its gradual acquisition, integration, and use of knowledge about foreign markets and operations, and on its successively increasing commitment to foreign markets”. The model’s basic assumptions are that lack of such

knowledge is an important barrier to the development of international operations and that this necessary knowledge can be acquired chiefly through operations abroad (Johanson and

(8)

5

Valhne, 1977; 23). According to the model internationalization is an incremental process depending on the firm’s experiential knowledge of foreign markets (Oviatt & McDougall, 1997; 87). This process is incremental due to the lack of market knowledge and the resulting uncertainty (Johanson and Valhne, 1977; 26). Because of the high uncertainty, firms begin their foreign expansion with the country that is most similar and with only small engagements of resources. As a result of more foreign experience, and following improved understanding of foreign markets, firms enter host countries that are more dissimilar to their home country, which results in an increase of foreign direct investments. For example, a Swedish firm deploys its FDI activities in Germany and Norway before it deploys their FDI activities in the United States or Japan (Oviatt & McDougall, 1997; 87). Barkema & Drogendijk (2007; 1133) offer a new perspective on internationalization while retaining some important notions from the Uppsala model. The key assumption of their perspective is that learning how to deal with foreign cultures is a requirement in order to operate successfully abroad (Barkema &

Drogendijk, 2007; 1134). In the following paragraph I will explain why psychic distance may affects firm performance.

2.2 Dimensions of distance

Internationalization is within international business viewed as the exploitation of the competitive advantages of a firm in overseas markets. The Uppsala theory states that the location choice of an expanding firm is sequentially arranged to ‘perceived proximity’ to the home environment. This is because greater cultural distance and environmental differences can reduce the efficiency of a firm’s internalization of competitive advantages (Hsu et al, 2013; 58). The ‘liability of foreignness’ is used to describe the unfamiliarity with the environment, from political, economic and cultural differences, between other factors (Zaheer, 1995; 341). Firms have to overcome this liability in order to successfully compete with local firms. Hymer (1976) states that firms will only go abroad when the expected returns are greater overseas than in their home markets. This is based on their subjective assessment of the risks and costs involved. Most of these risk and costs results from barriers created by psychic distance (Ghemawhat, 2001; 3). This is distance between the home market and a foreign market, resulting from the perception of both cultural and business differences (Evans and Mavondo, 2002; 515). Internationalization theories make the assumption that psychically close markets are easier to learn and understand than distant ones. However, the psychic distance paradox states that conducting business in “psychically close markets is not

(9)

6 necessarily easy to manage, because assumptions of similarities can prevent managers from learning about critical differences” (O’Grady and Lane, 1995; 309). Several studies support

this paradox and suggest that psychic distance is positively related to firm performance (Evans and Mavondo, 2002; Tihanyi et al, 2005). Psychic distance between two countries can manifest itself along four basis dimensions: cultural, political, geographic and economic. Each of these dimensions of distance include many different factors, some of which are quite subtle, others are apparent. In the following part I will give a brief description of the four dimensions of distance, starting with the two overlooked the most – cultural distance and political distance.

Cultural distance. A country’s cultural attributes determine how people interact with each other and with institutions and companies. Difference in race, language, social norms and religious beliefs are all able of creating distance between countries. Actually, this

attributes can have an enormous impact on trade: for example, trade between countries with a common language will be three times greater than among countries who don’t share the same language – if all other attributes are held equal (Ghemawat, 2001; 3, 4). However, the study of Evans and Mavondo (2002; 528) indicates that language differences have a significant

positive effect on firm performance in the close market. A possible explanation is that firms make a greater effort to effective communication if the language differs, even if they identify the foreign market as similar to their own. In addition, the study of Tihanyi et al (2005; 278) showed that high cultural distance leads to higher firm performance when multinationals operate in other developed markets. This relationship may be due to their associated

institutions across developed markets and similarities in market conditions. An explanation for this is that firms, who expand into culturally diverse but developed markets, acquire new resources and knowledge that can lead to improved performance.

Political distance. Political and historical links between countries can have a great effect on trade. The leading example of deliberate efforts to decrease political distance is the integration of the European Union. Unilateral measures can also create political distance between countries. Indeed, the most common barriers to foreign competition are policies of individual governments – trade quotas, restrictions on foreign direct investment, tariffs, and preferences for domestic competitors in the form of favoritism and subsidiaries in

procurement and regulation. Although, social conflicts and corruption decrease investments and trade much more than any explicit administrative restriction or policy (Ghemawhat, 2001; 5, 6; Schneider and Frey, 1985; Busse & Hefeker, 2007).

(10)

7

Geographic distance. Generally, how greater the physical distance between two countries, the more difficult it is to conduct business. However, geographic distance is not only based on distance in kilometers. Other attributes that must be considered encompass the average within-country distances to borders, topography, the physical size of the country, and access to ocean and waterways. Also, human-made geographic attributes, such as

communications infrastructures and country’s transportation, must be taken into account. Needless to mention, that geographic attributes affect the cost of transportation. However, also intangible goods and services, are influenced by geographic attributes. Research shows that if the geographic distance increases, the cross-border equity flows decreases

(Ghemawhat, 2001; 7).

Economic distance. Research also shows that rich countries relatively more engage in cross-border economic activity - in comparative to their economic size - than poorer

countries. Of course, economic disparities also arise as a result of cost and quality of

resources. Firms that rely on economies of scale, standardization and experience should focus more on countries with similar economic profiles. The reason for this is that firms exploit their competitive advantage by replicating their existing business model, which difficult to conduct in a country where consumer incomes, not to mention the quality and costs of resources, are completely different (Ghemawhat, 2001; 8). Furthermore, a consistent

macroeconomic policy increases the attractiveness of a foreign market. For example, a good macroeconomic policy includes (or leads to) low inflation, which is likely to reduce the risk premium for foreign direct investment, decreases transaction costs, and may thus increase foreign direct investment (Busse and Hefeker, 2007; 404). In contrast, a bad macroeconomic environment with high inflation has a significant negative effect on firm performance (Li and Lui, 2005; 401).

(11)

8 Table. 1: Overview of distance’s four dimensions (Ghemawhat, 2001; 1)

Cultural distance Political distance Economic distance Geographic distance More distance between two countries is created as a result of: - Lack of connective social or ethnic networks - Different ethnicities, social norms, languages, and religions - Weak financial and legal institutions - Absence of shared political and monetary associations - Political hostilities - Different knowledge or information - Differences in consumer incomes - Different quality and costs of human, financial and natural resources - Physical remoteness - Different climates - Lack of waterway access, communication links, adequate transportation or a common border

As I mentioned before the liability of foreignness is used to describe unfamiliarity of the environment as a result of psychic distance. To overcome the liability of foreignness, MNEs must provide their overseas subsidiaries with some firm-specific advantages, often in the form of managerial or organizational capabilities. However, research suggests that cultural distance also can improve firm performance. In that case a firm needs the explorative and learning capabilities which make it possible to acquire new knowledge and resources. Also the resource-based view of strategy have emphasized the importance of firm-specific

advantages and organizational capabilities in establishing a sustainable competitive advantage to firms (Zaheer, 1995; 341). The next paragraph will describe how firm-specific advantages can lead to a competitive advantage.

2.3 Resource-based view

Numerous studies state that the internationalization of a firm is not only driven by the

exploitation of its existing advantages, but also by its wish to capture and explore resources in foreign markets that strengthen corporate global competitiveness and long-term performance (Hsu, et al, 2013; 58). The emergence of the resource-based view (RBV) helps to understand how multinationals could achieve a sustained competitive advantage (Peng, 2001; 803). The resource-based view makes two assumptions in analyzing sources for competitive advantage. First, the model assumes that firms within an industry (or group) are heterogeneous, in terms of their internal capabilities and resources. Second, the model assumes that resources may not

(12)

9

be perfectly mobile between firms, and thus heterogeneity may be long-lasting (Barney, 1991; 100, 101). Firm resources include all capabilities, knowledge, information, organizational processes, firm attributes, assets, etc. controlled by a firm that enable the firm to implement and/or conceive strategies that improve its effectiveness and efficiency. Obviously, not all aspects of a firm’s human capital, organizational capital, and physical capital held the

potential of a competitive advantage. Some of these firm attributes may prevent to implement or conceive valuable strategies. Others may lead to strategies that reduce its effectiveness and efficiency (Barney, 1991; 102). To capture this potential, firm resources must have four attributes: (a) the resource is valuable, which means that it neutralizes threats and/or exploit opportunities in a firm’s environment; (b) the resource is rare, which means that competitors don’t possess it; (c) the resource is imperfectly imitable, and (d) the resource is not

substitutable (Barney, 1991; 105, 106). Firms who have these resources or capabilities, which are superior or distinctive comparative to those of their rivals, can achieve a competitive advantage if they are adjusted properly to the environment (Peteraf, 1993; 179).

A key insight of conventional international management literature is that MNEs face a substantial ‘liability of foreignness’ which leads to risks and costs (Zaheer, 1995). To

overcome this liability, both transaction cost economics and the eclectic paradigm argue that MNEs need to equip their foreign subsidiaries with firm-specific advantages. The RBV enlarges these perspectives by specifying the nature of these resources and capabilities, such as organizational practices (Peng, 2001; 810). Buckley and Casson (1998) view the firm as an internalized bundle of resources which can be distributed between product groups, and

between foreign markets. Firm-level variables are linked to the concept of transaction costs, whereby the transfer of specialized assets among firms is obstructed by market failures, which makes international expansion necessary in order to internalize the transfer (Kogut and Singh, 1988; 413). The eclectic paradigm, a framework to test determinants for foreign production, states that “the greater the net benefits of internalizing cross-border intermediate product

markets, the more likely a firm will prefer to engage in foreign production itself” (Dunning,

2000; 164).

Within this view foreign expansion is considered as a way for firms to appropriate rents in foreign markets by exploiting and exploring valuable resources, such as scientific knowledge, brand names, and technological capabilities. The resources already owned by the firm can be implemented in multiple markets, which helps to balance the risks and costs as result of liability of foreignness, and achieve economies of scope (Wang et al, 2012; 426). Several conceptualization of international expansion are compatible with the RBV, because

(13)

10

value-generating resources are conceptually similar to ownership advantages (Dunning, 2010; 172). Ownership advantages includes “the competitive advantages of the MNEs seeking to

engage in FDI (or increase their existing FDI), which are specific to the ownership of the investing MNEs” (Dunning, 2000; 164). As a consequence, exploitative and explorative FDI

activities rest upon the resource-based view.

2.4 Exploration and exploitation

How firms explore new assets and exploit their existing assets in host countries through foreign direct investment (FDI) is one of the key issues in the field of international business research (Makino et al, 2002; 405). The balance between exploration and

asset-exploitation can be discussed with the use of a theory of rational search. This theory makes the assumption that there are various alternative investment opportunities. Each opportunity is characterized by a probability distribution over returns that is initially unknown. Over time there is an accumulation of information about the distribution, however the firm still has to choose between two options: exploration or exploitation.

The exploitation perspective “views FDI as the transfer or exploitation of firm-specific

advantage assumes that firms should possess certain forms of rent-yielding resources when investing in a host country” (Makino et al, 2002; 405). This perspective proposes that when a

firm possess resources and skills which give rise to a monopolistic (or competitive) advantage in a host country, FDI would occur. The main thought about exploitation is that it is possible to establish a good position in the market by using sufficient organization’s assets to protect the firm against its competitors (Auh & Menguc, 2005; 1653). The exploration perspective states that “firms invest in foreign countries not only to exploit but also to develop their

firm-specific advantages or acquire necessary strategic assets in a host country” (Makino et al,

2002; 406). This perspective suggests that the possession of critical capabilities and resources also results from the firm’s acquiring capacity, or from the efficient coordination of

complementary assets owned by other firms in a host country. So, in order to build

advantages, firms invest in a host country in which their needed strategic assets are available (Makino et al, 2002; 406). The study of Hsu & Chen (2009; 600) gives more insight in FDI patterns. Their research states that explorative FDI, in the form of R&D and marketing

activities, used for achieving further development of marketing and technological capabilities, is more likely to developed countries, while exploitative FDI, in the form of manufacturing

(14)

11

activities, used for achieving further development in manufacturing capability is more likely to less developed countries.

From the resource perspective, the success of exploration is determined by the firm’s capabilities to seek, acquire and attract external resources. In contrast, the success of

exploitation requires an effective and efficient process to extend and increase the use of existing resources ‘internal’ to the firm. As a result, exploration and exploitation demand different behaviors facilitated by the firm’s structure and culture (Ireland and Webb, 2007; 54). Auh and Menguc (2005) found that exploitation was more positively associated to

efficient firm performance, whereas effective firm performance was more positively related to exploration. Although trade-offs between exploration and exploitation are surely necessary because they compete for scarce resources, both options are essential for organizations (He and Wong, 2004; 482). Levintahl & March (1993; 105) conclude that firms have to execute sufficient exploitive activities to ensure the organization’s current viability, and execute enough explorative activities to ensure future viability. As a result, organizations make implicit and explicit choices between exploration and exploitation. The explicit choices are based on calculated decisions about competitive strategies and alternative investments. The implicit choices can be found in many features of organizational forms and customs, for example, in incentive systems and in the ways in which targets are changed. The problem with making explorative and exploitative choices is that when new alternative investment opportunities appear, the probability distributions may not remain stable, or that the probability depends on the choices of others. It’s obvious that the exploration of new alternatives declines the speed of the improvement of already existing skills, and that competence improvements of existing procedures reduces the attractiveness of experiments with others (March, 1991; 71, 72).

However instead of choosing one, successful firms maintain a balance between exploitation and exploration on the long-term. Their expansion may more deeply into well-known cultural blocks, in order to exploit the firm’s knowledge base and maximize short-term performance. Or the firm may enter new cultural blocks, which implies lower short-term performance but the ability to learn in order to enhance future success. Depending on the firm’s strategy and resources, a firm does either both, or does them sequentially. The optimal balance depends on the industry conditions, for example, the extent to which the firms in the industry are involved in a struggle for global dominance (Barkema & Drogendijk, 2007; 1143). It suggests that a firm needs the ability to adapt to its environment.

(15)

12

To summarize, exploitation and exploration are fundamentally different concept that lead to tensions. Both concepts compete for scarce resources, which require firms to manage trade-offs between the two. Although, also synergy can exist between the two, therefore it’s necessary to balance explorative and exploitative activities. The ambidexterity perspective states that if these two activities are combined a synergistic effect exist between exploration and exploitation, and their balance and integration are preconditions to ensure the firm’s success and long-term survival (Hsu et al, 2013; 59). In the next paragraph this will be explained.

2.5 International ambidexterity

The process of choosing between exploration and exploitation can be considered from an organizational learning perspective. March (1991; 71) has considered the relation between the exploitation of old certainties and the exploration of new possibilities in organizational learning. He concluded that “the distance in time and space between the locus of learning and

the locus for the realization of returns is generally greater in the case of exploration than in the case of exploitation, as is the uncertainty” (March, 1991; 85). In other words, exploration

might be effective but as a result of its long-term character, it might miss a high degree of efficiency (Auh and Menguc, 2005; 1653). If a firm relies too much on exploration it reduces possibilities to learn new capabilities and result in an organizational myopia, making the firms’ core capabilities to become a ‘failure trap’ (Levitt and March, 1988; 322). This trap places a continuous drain on the organization’s resources with no direct financial reward. On the opposite, if a firm relies too much on exploitation it can lead to a ‘success trap’: focusing on the more certain short-term returns and overlooking the uncertain, potentially profitable, outcomes associated with exploration (Auh & Menguc, 2005; 1654). As a result, the speed at which existing capabilities are refined and improved is reduced.

Tushman and O’Reilly (1996) conceptualized ambidexterity in order to crystallize the need for an appropriate balance between exploration and exploitation. According to the ambidexterity perspective, it’s better for a firm to balance between both options, rather than focusing on one at the expense of the other (He and Wong, 2004; 483). To facilitate

ambidexterity, an organizational structure should be adapted in such a way that this balance is cultivated. In addition, the organization’s capacity to absorb spillovers from exploration to exploitation determines the ability to perform ambidextrous activities (Hsu et al, 2013; 61).

(16)

13

Such ambidextrous implications are in fact implicit in existing literature about strategies and organizations (Hsu et al, 2013; 59). For example, organization literature suggests that firms should conduct an ambidextrous structure to combine two different organizational goals, such as the exploring new capabilities, while exploitation the existing capabilities. Katila and Ahuja (2002; 1192) support the idea that the dynamic problem-solving capabilities of a firm can be an important source of resource heterogeneity. Also strategy literature highlights the concept of international ambidexterity. Scholars argue that the development of dynamic capabilities demand for an effective mix of exploration and exploitation (Eisenhardt and Martin, 2000; 1107). Dynamic capabilities are needed to

facilitate international ambidexterity because they effectively match the firm’s activities with its environment (Prange and Verdier, 2011). In return, defending and sustaining the efficiency of ambidexterity supports long-term firm performance (Hsu et al, 2013; 49).

2.6 Development of hypotheses

In this paragraph I will develop testable hypotheses, to investigate the moderating role of distance for the relationship between firm performance and international ambidexterity, in the specific context of foreign direct investment. Several scholars argue that multinational enterprises should strive for establishing an ambidextrous organizational structure that allows a firm to simultaneously perform exploitative and explorative tasks. For example, Han (2005; 12) who state that “an ambidextrous organization is the best supporting platform to embrace

alignment among structure, people, culture, and strategy’’. Han (2005; 72) adds that

ambidexterity not only leads to more effective and efficient financial performance, but also to more knowledge enhancement. Han and Celly (2008; 347) show that firms who have the capability to simultaneously and actively pursuing, managing, and executing paradoxical strategies - which refers to strategies that demand different resources and capabilities – achieve superior performance. Their study use the ambidexterity concept as a firm capability that allows firms to decide ‘how’ and ‘why’ paradoxically strategies can be executed to achieve superior performance. These strategies are paradoxically because executing them simultaneously causes tradeoffs and tension. Although, when a firm has the ambidexterity capability to transform this tension into complementary effects for long-term sustainability it can achieve superior performance.

The study of Luo and Rui (2009; 68) was the first to apply the ambidexterity

perspective on international business research. Luo and Rui (2009; 49) argue that every firm needs and maintains some degree of ambidexterity, although multinationals from emerging

(17)

14

markets (EM MNEs) have stronger abilities and motives to build and leverage such

ambidexterity to compensate their late-mover disadvantages. Their study suggests alternative dimensions for measuring international ambidexterity: co-orientation which captures the cultivation of short-term profitability and long-term growth, co-competence involves the possession of both relational (capabilities for dealing with external business stakeholders) and transactional (capabilities that are purely economic and market-based), co-opetition refers to simultaneous cooperation and competition between MNE and its external stakeholders, and co-evolution encompasses the simultaneous cultivation of proactive influencing the

environment and organizational adaptation (Luo and Rui, 2009; 53, 54, 55, 56). Luo and Rui (2009; 68) conclude that organizational ambidexterity can help EM MNEs overcome late-mover disadvantages and set a path towards new growth.

Prange and Verdier (2011; 131) also relate ambidexterity to the internationalization process. Their study argues that both explorative and exploitative activities are required to obtain different sets of dynamic capabilities and increase internationalization performance. Foreign expansion, both exploitation and exploration, gives firm the ability to accumulate learning or create new knowledge. Knowledge creation and continuous learning are important tasks, and distinct culture, knowledge, and skills can lead to tensions in a not properly

structured organization. Firms need to emulate an ambidexterity internationalization process which leads to obtaining different sets of dynamic capabilities for long-term performance (Prange and Verdier, 2011; 131). Hsu et al (2013) have investigated how international ambidexterity affects firm performance. Their study shows that “firms with a balanced

ambidextrous configuration of exploitative and explorative FDIs outperform others” (Hsu et

al, 2013; 63). Therefore the following is hypothesized:

Hypothesis 1: The higher the extent of international ambidexterity, the higher the firm performance

Complexity and uncertainty can diminish returns from executing international

ambidexterity during the international process. Hsu et al (2013; 61) showed that for firm from small emerging economies, international ambidexterity is vulnerable for environmental complexity, sensitive to international experience and international capability. Their empirical results show that the degree of internationalization negatively moderates the relationship between international ambidexterity and firm performance (Hsu et al, 2013; 63). The degree of internationalization is measured as geographic distribution across the global market and the

(18)

15

number of FDIs. An explanation for this is that institutional relatedness – the network of ties with dominant institutions in a country – declines as a result of foreign expansion. As a result of the expansion more institutional variance exist and thus less ties remain (Hsu et al, 2013; 61). However, this could be to specific features of firms from small emerging economies and their particular FDI patterns. In extension of Hsu et al (2013) this study isolates the separate dimensions of psychic distance for developed large economies. As discussed previously, psychic distance can increase – as a result of new learning opportunities – and decrease – as a result of the liability of foreignness – firm performance. Although, executing international ambidexterity during the internationalization process is quite challenging. Improved complexity and uncertainty in the firm’s operational domain can further stimulates a conservative tendency to resist exploration – and thus to acquire and develop resources and knowledge - and degrade the already weak transition between exploitative and explorative activities (Hsu et al, 2013; 61). Needless to mention that this abolishes the possible positive influence of cultural distance. These ideas suggests that international ambidexterity can be negatively moderated by psychic distance, which leads to the following hypothesis:

Hypothesis 2: The higher the extent of distance, the more the effect of international ambidexterity on the firm performance

(19)

16

Chapter 3: Data and method

In this chapter I will discuss the data and methods of this study, beginning with the research design and the different variables used. This will be followed by an explanation of the data and sample selection. After that, I will describe the several measures that were used. To conclude, I will explain the analyses performed.

3.1 Research design

The research design is built to investigate if psychic distance moderates the relationship between international ambidexterity and firm performance. Previous research already focused on the relationship between international ambidexterity and firm performance (Hsu et al, 2013; Luo & Rui, 2009; Prange and Verdier, 2011). This studies show that balancing between exploration and exploitation has a positive effect on performance by firms from emerging economies, in the specific context of foreign direct investment. No research have measured the influence of distance this relationship. It may be that distance moderates the effect of international ambidexterity on firm performance. In order to investigate the role of psychic distance, I measure seven variables and three control variables. International ambidexterity as an independent variable; firm performance as a dependent variable; and the four different dimensions of distance as moderating variables. In order to test the hypothesis a quantitative research method will be used. I will use the following hypothesized model:

(20)

17

3.2 Data and sample

In order to test the hypothesized model, I have conducted a comprehensive database. The database consist of 2.636 subsidiaries, established by 50 Japanese MNEs in 7 industry sectors, in 105 host countries, over an one-year period. Data of this study is collected through primary sources, e.g. annual reports and corporate websites. The 50 Japanese MNEs are selected from the Nikkei-225 Stock Average, which is listed in the First Section of the Tokyo Stock

Exchange. The Nikkei-225 Stock Average is a price-weighted average of 225 top-rated Japanese companies (Bloomberg, 2014). A criteria for selection was that the MNEs provided sufficient information, in their annual reports and/or on their corporate websites, about their subsidiaries types. This was needed in order to categorize the subsidiaries. Categories were based on the business activity, marketing/R&D/manufacturing, of the subsidiary. Another criteria was that the firm provided complete financial information for revenue or return-on-sales over the years 2012 and 2013.

To avoid home country bias this study focused on a single country. Host country bias can exist when MNEs with headquarters in different countries (and thus in different home countries) are compared. Another advantage of using a single country is to help ensure that variations in international ambidexterity are more likely associated with strategic choice than with variations in market favorability (Autio et al, 2000; 910). A sample drawn from one single country also ensures an easier comparison with prior research and easier interpretation (Gande, Schenzler & Senbet, 2009; 1519).

The sample is drawn with Nikkei-225 listed MNEs for three reasons. The first reason are the differences between Japan and the countries used by previous research on international ambidexterity. The studies of Hsu et al (2013; 62) and Luo and Rui (2009) tested for

multinationals from emerging countries. The sample of Hsu et al consists of MNEs on the Taiwanese Stock Exchange, listed in the Electronics Information Technologies category. Hsu et al. have picked Taiwan because it showed high levels of internationalization. This can be explained with the fact that Taiwanese firms have a very small home market and are thus heavy relying on their foreign activities. Japan is more or less the opposite, it’s an established country with a very large home market and thus there is less need to globalize. This is

supported by the differences between ranking in the EY Globalization Index 2012 which shows that Japan is ranked 43th and Taiwan is ranked 17th. However, in terms of national competitiveness Taiwan and Japan are comparable. The Global Competiveness Index 2013 (World Economic Forum, 2014) shows that Japan is ranked 9th and Taiwan is ranked 12th.

(21)

18

Second, it’s likely that some of the Japanese multinationals have an ambidextrous structure. This is because Japan has a developed economy, which means that Japanese MNEs probably have the resources and capabilities to compete on a global scale. The previously mentioned Global Competitive Index supports this. Ghoshal and Bartlett (1994) suggest that ambidexterity exists when leaders in a business unit develop a supportive organization context. Firms from less developed markets lack knowledge in the form of technological capabilities, strategy, and management skills. A lack of internal managerial capabilities makes it thus more difficult to develop dynamic capabilities such as ambidexterity (Rugman and Li, 2007).

Third, Japan is probably relatively distant from other countries in a psychic sense - which makes it easier to measure the effect of distance on the entry mode choice of multinational enterprises. Japan is a religiously and ethnically homogenous society with a feudal past of strong nationalistic and internal allegiances. Its culture differs from other countries, which makes is interesting to use. Hofstede model’s (1980) gives a good overview of the four deep drivers of Japanese culture in comparison with other cultures. First, it shows that Japan is a borderline hierarchical society but less hierarchical than other Asian cultures. Also it shows that Japan is a collectivistic society but less collectivistic than other Asian countries. Third, it shows that Japan can be considered as one of the most masculine societies in the world. However, in combination with their mild collectivism, competitive and assertive individual behaviors which are often associated with masculine culture aren’t visible. To conclude, Japan is one of the most uncertainty avoiding countries in the world. Also Japan is a ‘special case’ in terms of economic status, it’s an Eastern country equivalent to the United States. Its interpretation of capitalism is not identical to the model of the United States (Ralston, et al., 1997; 185).

3.3 Measures

The next section describes how the different variables are measured, how the data is collected for each variable and the descriptive statistics. I will start with the dependent variables, continued with the independent, moderator and control variables.

3.3.1 Dependent variables

The dependent variables of this study are ‘efficient firm performance’ and ‘effective firm performance’. As previously mentioned, exploration is more positively related to efficient firm performance, while exploitation is more positively associated with effective firm

(22)

19

performance (Auh and Menguc, 2005). Consequently, using an one-dimensional indicator of firm performance, such as sales growth, might lead to the risk of producing biased estimations of international ambidexterity’s contributions to the firm’s overall success (Raish and

Birkinshaw, 2008; 400). In this study firm effectiveness will be measured in terms of revenue growth; while firm efficiency will be measured in terms of return-on-sales (Auh & Menguc, 2005; 1654). A firm can excel in both perspectives, in only one, or neither. However, the best-performing firms combine efficiency and effectiveness (Ostroff & Schmitt, 1993; 1345).

Information about revenue growth is collected from the annual reports. The current amount of revenue will be compared with the amount of previous year. The percentage difference will be used in the dataset. Information about return-on-sales will also be collected from the annual reports. In most cases the annual reports provided the value of the return-on-sales. In other cases, return-on-sales is calculated through ‘net income before interest and tax divided by total revenue’.

3.3.2 Independent variable

The independent variable of this study is international ambidexterity. Studies have suggested several methods for the measurements of exploration and exploitation (Auh & Menguc, 2004; Barkema & Drogendijk, 2007; Makino et al, 2002). In this study, international ambidexterity, will be measured following He and Wong (2004) and Cao, Gedajlovic, and Zhang (2009) by using the absolute difference between the numbers of its explorative and exploitative FDI activities, which gives high reliability and validity. The number of overseas manufacturing subsidiaries represents the exploitation FDI activities; while the explorative FDI activities are represented by the number of overseas R&D and marketing subsidiaries. “The difference

between these numbers represents the extent of imbalance between exploitation and

exploration for the company i in the year t” (Hsu et al, 2013; 62). Although, for convenience

of interpretation, this measure is reversed thus a higher value indicates greater ambidexterity.

In addition, some subsidiaries can be considered as both manufacturing and

marketing/R&D. Following Hsu et al (2003), in that case there is no imbalance between exploration and exploitation. This could lead to the assumption that foreign subsidiaries who combine exploitative and explorative business activities can be considered as international ambidextrous. Although, these type of subsidiaries still need to be part of the calculation in order to prevent a bias between explorative and exploitative activities. Therefore the dataset contains of three subsidiary types: manufacturing (exploitation), R&D/marketing

(23)

20

(exploration), and ‘combined’ (international ambidexterity). In order to measure international ambidexterity, the first step is to calculate the difference between the explorative and

exploitative subsidiaries. The second step is to calculate the ‘weighted difference’ through dividing this difference by the total of ‘explorative’, ‘exploitative’ and ‘combined’

subsidiaries. The final step is to reverse the measure, by ‘1 minus the weighted difference’. If the outcome is closer to 1, it indicates greater international ambidexterity. I have captured the three steps in the following formula: 𝐼𝐴 = 1 − ( 𝐸𝑥𝑜𝑟−𝐸𝑥𝑜𝑖

𝐸𝑥𝑜𝑟+𝐸𝑥𝑜𝑖+𝐵𝑜𝑡ℎ). In which, IA stands for the

extent in which a firm can be characterized as international ambidextrous, Exor stands for the number of subsidiaries with explorative FDI activities, Exoi stands for the number of

subsidiaries with exploitation FDI activities, and Both stands for the number of subsidiaries who combine FDI exploitative and FDI exploitative activities.

Information about the different subsidiary types is collected out of annual reports and corporate websites. All multinationals included in this study provide in their annual report or on their corporate website a list with their overseas subsidiaries. Most companies also mentioned if the subsidiary refers to manufacturing, marketing or R&D. If this was not the case, corporate statements and newspaper articles were used to determine the subsidiary type. Subsidiaries who only refer to other business activities then marketing/R&D and/or

manufacturing, for example financial holdings, will be excluded from this research.

3.3.3 Moderating variables

The moderating variable of this study is distance. Barriers created by distance could create risks and costs. Within international business research there is a broad selection of indicators for distance (Dow & Karunaratna, 2006; 578). Distance between two countries can be divided into four basic dimensions: cultural distance, economic distance, geographic distance and political distance.

Cultural distance can be created by differences in race, social norms, religious beliefs and language (Ghemawhat, 2001; 3). It reflects how country cultures differ from each other. Cultural distance can, for example, affect consumer preferences. Hofstede’s (1983)

dimensions are used to measure cultural differences between countries. The Hofstede dimensions of national culture can be statistically categorized into four groups: Power

Distance, Individualism versus Collectivism, Masculinity versus Femininity, and Uncertainty Avoidance. Power Distance “expresses the degree to which the less powerful members of a

(24)

21

Collectivism reflects if there is a preference in society for “a loosely-knit social framework in

which individuals are expected to take care of only themselves and their immediate families”

or “for a tightly-knit framework in society in which individuals can expect their relatives or

members of a particular in-group to look after them in exchange for unquestioning loyalty”.

Masculinity versus Femininity shows if society members have a preference “for achievement,

heroism, assertiveness and material rewards for success” or for “cooperation, modesty, caring for the weak and quality of life’.Uncertainty Avoidance“expresses the degree to

which the members of a society feel uncomfortable with uncertainty and ambiguity” (The

Hofstede Center, 2014). The data for these different dimensions is collected from the website of The Hofstede Center.

Cultural distance is calculated with use of Kogut and Singh Index (1988; 422). This index is based upon the deviation between each of the four cultural dimensions of each country from Japan ranking, using the values of Hofstede. Kogut and Singh conducted the following index: 𝐶𝐷𝑗 = ∑ {(𝐼𝑖𝑗 − 𝐼𝑖𝑢)

2 4

𝑖.=1

∕ 𝑉𝑖}/4. In which, 𝐼𝑖𝑗 stands for the index for the

ith cultural dimension and jth country, 𝑉_𝑖 is the variance of the index of the ith dimension, u indicates Japan, and 𝐶𝐷𝑗 is cultural difference of the jth country from Japan.

Economic distance is used to examine differences in economic development between the home and host country. In this study economic distance is measured with the indicator ‘Host Country Inflation in % (Consumer Prices)’. Inflation affects the net worth of consumer incomes. Information on this variable will be obtained from the Development Indicators of the Word Bank (López-Duarte & Vidal-Suárez, 2013; 54).

Geographic distance affects MNEs because, in general, the farther the host country is from the home country, and the more difficult it will be to conduct business in that country (Ghemawhat, 2001). It will be measured through the flying distance in kilometers between the capital of the home country, Tokyo, and the capital of the host country (López-Duarte & Vidal-Suárez, 2013; 54). The data is collected with use of the website http://nl.distance.to/.

Political distance is measured with use of the indicator Political Stability and Absence of Violence/Terrorism. Political Stability and Absence of Violence/Terrorism “reflects

perceptions of the likelihood that the government will be destabilized or overthrown by unconstitutional or violent means, including politically-motivated violence and terrorism”

(WGI project of the World Bank). The data is collected from the Worldwide Governance Indicators (WGI) project of the World Bank. The WGI project reports individual and aggregate governance indicators for 215 economies over the period 1996-2012. It’s

(25)

22

assumable that MNEs are less likely to invest in unstable countries and/or countries with high violence/terrorism levels.

After data is collected about the different dimensions, the average distance needs to be calculated. The average distance is the extent of distance (for each dimension) a multinational enterprise faces for all its international locations. This means that I have to calculate the average distance per dimension: the distance between home and host country multiplied with the number of subsidiaries divided through the total number of subsidiaries.

3.3.4 Control variables

In this study have use control variables because of their potential impact on firm performance and international ambidexterity. The control variables that I have used are firm size, firm age and industry sector.

Firm size is a control variable because of two reasons. First, because of the effect of economies and diseconomies of scale at the corporate level (Hitt et al, 1997; 781). Second, because, some firms, even with the same home country and from the same industry sector, are more likely to engage in foreign direct investment than others. Sometimes, this is due to their size, in general, large firms tend to be more multinational than small firms (Dunning, 2000; 165). Firm size will be measured as the log transformation of the number of consolidated employees (Auh & Menguc, 2005; 1656). Data about the number of consolidated employees will be obtained through the corporate website or annual report. Quartiles helped to divide the number of consolidated employees into four equal (1: small – 4: large) groups.

The experiential effect will be controlled with the use of age. Oviatt and McDougall (1997) show that the accumulated experience influences the accessibility to foreign markets. It’s calculated as the number of years since the firm’s establishment to 2014. Data about the firm age will be collected through the corporate website or annual report. This study will control for industry sector differences because of three reasons. First, it may be hypothesized that some sectors, for example the healthcare and technology sectors, are likely to generate more foreign direct investment than others. This could be for several reasons: because their locations needs prefer production outside their home countries; because the net benefits of internalizing cross-border intermediate product markets are higher; and because the

characteristics of some sectors generate more unique ownership advantages (Dunning, 2000; 165). Second, there are differences between industries in the level of technological

opportunity (Hambrick & MacMillan, 1985). This reflects the extent to which the market demands or accepts product innovations. R&D expenditures differ between industries because

(26)

23

of the scientific base from which products are derived. Further, it can be that firms operating in fragmented industries are less willing to invest in product innovation than dominant firms in concentrated industries (Hoskisson & Hitt, 1988; 612). Data about the different industry sectors will be collected from the website of Bloomberg, who is leader in global business and financial information.

Industry sector is divided in the following sectors: technology and health care, consumer, materials, industrials, energy, financials, communications, and utilities. The categorization of industry sectors by Bloomberg have been used to classify each Nikkei listed MNE. Data shows that firms from the sample are from five different industry sectors (See Table 1). The technology and health care sector represents the biggest group (N = 17, 16.4%), second is the consumer sector (N = 13, 26.0%), followed by the materials sector (N = 13, 26.0%), industrials sector (N = 7, 14.0%) and the energy sector (N = 5, 10.0%). In addition to the sample, the Nikkei also consists of three other sectors: financials, communications, and utilities. These sectors are not represented in the sample because it was not possible to

categorize the subsidiaries out of these sectors into manufacturing, marketing or R&D. When comparing the representation of the sectors in the sample to the Nikkei in 2013, it can be concluded that the energy, technology and health care sectors are overrepresented in the sample: 44,0% compared to 20,0% of the Nikkei. An explanation may be that these sectors have a clearer distinction between their exploration and exploitation activities. The remaining industries (thus without financials, communications, and utilities) account for a total of 56,0% in the sample, in comparison to 63,1% of the Nikkei.

Table 2. Overview of Industry Sector

Industry Sector Sample Sample % Nikkei Nikkei % Difference Technology and Health Care 17 34.0% 37 16.4% 17.6% Consumer 13 26.0% 67 29.8% 3.8% Materials 8 16.0% 38 16.9% 0.9% Industrials 7 14.0% 37 16.4% 2.4% Energy 5 10.0% 8 3.6% 6.4% Financials 0 0.0% 25 11.1% 11.1% Communications 0 0.0% 8 3.6% 3.6% Utilities 0 0.0% 5 2.2% 2.2%

(27)

24

Industry sector is a categorical variable. The difficulty with using categorical variables as predictors is that there are often more than two categories (Field, 2009; 253). In this case there are five categories so these industry sectors can’t be distinguished using a single

variable coded with ones and zeros. In order to make this distinction dummy variables can be used. Dummy coding makes it possible to represent industries using only ones and zeros. Several variables therefore have to be created; the total number of variables is one less than the number of industries that are recoded. This is because one of the industry sectors is chosen as a baseline - the industry sector against which all other industry sectors are compared. In this research the baseline is the industry sector ‘technology and health care’ because it

represents the majority of multinationals, and thus might be interesting to compare with other sectors. An example of dummy coding for industry sectors is visible in Table 2. After the industry sectors are recoded they are placed in a regression analysis, which is more briefly explained in the results section (Field, 2009; 254).

Table 3. Example of dummy coding for industry sectors

Consumer Materials Industrials Energy

MNE A 1 0 0 0 MNE B 0 1 0 0 MNE C 0 0 1 0 MNE D 0 0 0 1 3.4 Data analysis

The information collected from the different data sources was saved in an Excel file. Before I could do the analysis I needed to calculate the revenue growth, as indicator of economic distance, and to calculate the average distance each MNE experiences.

Revenue growth is calculated as the percentage difference between the revenue in 2013 and the revenue in 2012. After that I have calculated the average distance a MNE experiences when operating abroad. To start, I have calculated the type of distance per country. The number of subsidiaries in a country determines the total global distance a MNE experiences. It would give a biased view if a country with more subsidiaries is weighted equally as a country with only one subsidiary. So, the number of subsidiaries in a country is multiplied with the distance of a country. I have conducted the following formula for the average distance between Japan and the countries where the MNE operates: 𝐷_𝑀 =

(28)

25

when internationally operating and C stands for the different countries where the MNE has subsidiaries. 𝑁_𝑐 is the number of subsidiaries in a country, 𝐷_𝑡 stands for the type of distance per country, and 𝑁𝑠 is the total number of subsidiaries per MNE.

The calculated values from the Excel file were put into SPSS for a further analysis. Both correlation and regression analysis were used. The results of the analyses will be discussed in the following chapter.

(29)

26

Chapter 4: Results

In this chapter I will report on the analysis of quantitative data. The aim of this chapter is to show which strategies for analyzing empirical data are applied and how the research results are reported. To research the hypotheses I have used both correlation and regression analysis.

4.1 Analysis of empirical data

In order to test the hypotheses I have ran a correlation analysis and a multiple hierarchal regression analysis. The empirical data is derived from the dataset that I have conducted. Before I was able to run an analysis with use of SPSS (Statistical Package for the Social Sciences), I had to check for the missing values. To do this, I performed a frequencies analysis. It showed that for the dependent variable ‘firm performance efficiency’, there were missing values for 3 of the 50 MNEs (6%). To solve this, I performed a ‘missing values analysis’. As a result, the missing values were replaced with estimated values. This is a reasonable approach because the amount of missing values is small and the data is randomly missing.

4.1.1 Correlation

In this section I will describe the relationships, or correlations, between the independent variable and dependent variables. A correlation shows how two variables are related, that could be in three ways: positively related, meaning that if the value of variable A increases, the value of the variable B also increases; negatively related, meaning that if the value of variable A decreases, the value of the variable B also decreases; not related, meaning that if the value of variable A decreases or increases, variable B isn´t affected.

There are two ways to measure correlation: covariance and the correlation coefficient. Covariance is based upon the concept of variance. Variance shows the average amount that the data vary from the mean (Field, 2009; 167). The mean is the average value in the dataset (Field, 2009; 21). There is covariance, when one variable deviates from its mean, the other variable also deviates from its mean in a similar manner (Field, 2009; 167). The problem with covariance, as the measure of the relationship between variables, is that it depends upon the scales of the measurements used. To overcome this problem, standardization is needed. This means that the covariance must be converted into a standard set of units (Field, 2009; 169). The unit of measurement used, in order to do this, is standard deviation. The standard

(30)

27

deviation is the average error in the data expressed in the same units as the original measure (Field, 2009; 37).

The standardized covariance is known as the Pearson correlation coefficient, r. The correlation coefficient ends up with a value between -1 and 1. A coefficient of -1 or less indicates that the two variables are perfectly negatively correlated; a coefficient of 1 indicates that the two variables are perfectly positively correlated. On basis of the correlation

coefficient the size of the effect can be measured: a small effect has a value of ±.1, a medium effect a value of ±.3, and a large effect a value of ±.5. In order to have an effect the correlation has to be significantly meaningful. The statistical model used will be non-directional, thus a two-tailed test. A non-directional hypothesis states that an effect will happen, but it doesn’t states the direction of the effect (Field, 2009; 27). It is an appropriate method, because it can be that higher international ambidexterity increases or decreases firm performance.

In this thesis I have used a bivariate correlation: a correlation which looks at the relationship between two variables (Field, 2009; 175). Correlations between all the measures used will be discussed. First, I will describe the correlations between the independent and the two dependent variables, followed by the control variables.

Table 4. Descriptive statistics and correlation matrix.a

Variables Mean S.D 1. 2. 3. 4. 5. 6. 7. 8. 1. Effectiveness ,01 ,17 2. Efficiency ,05 ,08 -,06 3. Firm Size 2,5 1,11 -,03 -,17 4. Firm Age 2,6 1,14 ,15 -,15 -,06 5. Consumer ,26 ,44 ,23 -,01 ,19* ,06* 6. Energy ,10 ,30 ,19 ,35** _-,09 _,12 _-,20 7. Materials ,16 ,37 -,05 -,04 -,15 -,27* -,26* -,15 8. Industrials ,14 ,35 ,05 ,03 -,13 ,25* -,24* -,13 -,18 9. International Ambidexterity ,64 ,23 ,32** _-,21 _,17 _-,03 _,18 _,06 _,28* _,11

Correlations between the independent and the dependent variables show something interesting. On basis of previous studies it is expected that firm performance positively correlates with international ambidexterity (Hsu et al, 2013; Luo and Rui, 2009; Prange and

a _{N = 50}

* Significant at 0.1 level ** Significant at 0.05 level *** Significant at 0.01 level

The moderating role that psychic distance plays in explaining the relationship between international ambidexterity and firm performance