National Cultural Distance – a predictor for a firm’s
international expansion strategy?
Evidence from German MNEs, tested via current
international business activities
National Cultural Distance – a predictor for a firm’s
international expansion strategy?
Evidence from German MNEs, tested via current
international business activities
Master Thesis
University of Groningen
Faculty of Economics and Business M.Sc. International Business and Management
13th June 2016
First Supervisor: Olof Lindahl, PhD Second Supervisor: prof. dr. A.R. (Alan) Muller
Sandra Rostek Sabangplein 6A 9715 CX Groningen Email: s.rostek@student.rug.nl
I Studies about the influence of national cultural distance on firm behavior have found that in some cases firms follow international expansion strategies of exporting whereas in other cases, firms prefer to internationalize via the establishment of wholly owned subsidiaries in the target markets. This study further explores the impact of national cultural distance on firms’ international expansion behavior and the moderating effect of the host country’s political stability. The research is based on firms’ current international activities and comprises 57 German large and very large firms. The findings provide significant support for the moderating influence of political stability on the relationship between national cultural distance and a firm’s international expansion strategy. Moreover, it is demonstrated that the actual influence of higher national cultural distance depends on the different national cultural dimensions.
Keywords: (National) cultural distance, international expansion, exporting, wholly owned
II I would like to take the chance to thank everybody who has helped me finalize this project and without whom I could not have mastered this academic challenge. First of all I would like to express my gratitude to my supervisor Olof Lindahl. He has been a great supervisor by providing constructive feedback and motivating me to continuously improve my work. Furthermore I would like to thank my family for their encouraging and motivating words throughout the whole process of creating my thesis, as well as my friends who made my time here in Groningen special.
Groningen, 13th June 2016
III
Table of Contents
List of Figures ... V List of Tables ... VI List of Abbreviations ... VII
1 Introduction ... 1
2 Theoretical Background ... 4
2.1 Cultural Distance, Liability of Foreignness, and Institutional Theory ... 4
2.2 Hofstede’s Cultural Dimensions ... 6
2.2.1 Power Distance ... 7
2.2.2 Long-term Orientation vs. Short-term Orientation ... 7
2.3 International Expansion Alternatives ... 8
2.3.1 Exporting ... 10
2.3.2 Foreign Direct Investment: Wholly Owned Subsidiary ... 10
2.4 Findings and Limitations from previous studies ... 11
2.5 Hypotheses... 15 2.6 Research Model ... 17 3 Methodology ... 18 3.1 Procedure ... 18 3.2 Variables ... 18 3.3 Data Collection ... 19 3.3.1 The Sample ... 20 3.3.2 Dependent Variables ... 22 3.3.3 Independent Variables ... 22 3.3.4 Moderator Variable ... 23 3.3.5 Control Variables ... 24
3.4 Statistical Method and Statistical Model ... 24
IV
5 Discussion ... 37
5.1 Conclusion ... 37
5.2 Managerial Implications ... 39
5.3 Limitations and Future Research ... 40
References ... 42
V
List of Figures
Figure 2.1: Key characteristics of internationalization alternatives, own representation based
on Osland, Taylor & Zou, 2001, p. 155. ... 9
Figure 2.2: Research model ... 17 Figure 3.1: Cultural distance between the target markets and Germany, own representation
based on The Hofstede Centre, 2016. ... 23
Figure 3.2: Political stability index of the target markets, own representation based on The
Global Economy, 2016. ... 24
Figure 4.1: The impact of national cultural distance on the international expansion strategy of
firms ... 35
Figure 4.2: The impact of the moderating effect of political stability on the international
VI
List of Tables
Table 2.1: Overview of previous studies about the influence of national cultural distance on
international expansion behavior of firms ... 13
Table 3.1: Variables and measures ... 19
Table 3.2: Overview of the selected target countries, own representation based on The World Bank, 2016; Central Intelligence Agency, 2016; OECD library, 2016... 21
Table 3.3 Cultural distance between the target markets and Germany, own representation based on The Hofstede Centre, 2016. ... 22
Table 3.4: Coding of variables in IBM SPSS ... 25
Table 3.5: Test for multicollinearity ... 26
Table 4.1: Descriptive statistics for continuous data ... 27
Table 4.2: Frequency statistics for categorical data ... 28
Table 4.3: Logistic regression results: Power Distance ... 30
Table 4.4: Logistic regression results: Long-term Orientation ... 33
Table 4.5: Overview results of hypotheses testing ... 34
List of Tables: Appendices
Table A: Overview of previous studies about the influence of national cultural distance on international expansion behavior of firms………..………...50Table C: Checking for outliers………..………….53
Table D: Cox & Snell R2 values……….……53
Table E: Hosmer and Lemeshow test………..………...53
VII
List of Abbreviations
approx. Approximately bn Billion cf. Confer df Degree(s) of freedom et al. Et alliExp(B) Odds ratio
FDI Foreign direct investment
ib. Ibidem
i.e. Id est
IJV International joint venture
IV Independent variable
JV Joint venture
m Million
MNE Multinational enterprise
MV Moderator variable
VIF Variance inflation factor
vs. Versus
1
1 Introduction
The pace of globalization has increased significantly since the late 1980’s and also continuous today, in the 21st century (Johnson, Lenartowicz & Apud, 2006). Globalization can be described as the process of increased global interdependence, where multinational enterprises (MNEs) play the major role. By operating in foreign countries, in order to gain access to new markets and resources, MNEs strengthen the economic interdependence among numerous national economies (Rugman & Verbeke, 2004). However, internationalization of the firm is not an easy process and strategy formulation is strongly influenced by national cultural patterns. In international settings, firms face various insecurities and managers have a variety of business strategies to choose from in order to mitigate risks, maximize benefits, and create legitimacy in the host market. Legitimacy creation depicts the adjustment to rules, norms, and standards of the host market (Lee, Shenkar & Li, 2008). It not only depends on formal rules and institutions but also on informal influences such as national culture. Since an MNE’s operations in foreign countries have brought managers into face-to-face contact with unknown cultures (Miroshnik, 2002), this is an important factor to consider when defining a business strategy.
Different international expansion strategies have extensively been evaluated in the existing management literature and in today’s business environment. Among the most famous one is the Uppsala Model. This model suggests that companies do not internationalize in big steps, but rather incrementally. In this way, firms first expand to foreign markets within the same geographical region in order to reduce the risk of lack of knowledge and unfamiliarity of local market practices. As a consequence, the cultural distance between home and host market is limited. Later, when firms learned from these first steps, they can expand to more distant markets (Johanson & Wiedersheim-Paul, 1975). Moreover, Johanson and Wiedersheim-Paul (ib.) claimed that different national cultures may threaten a firm’s success in a geographically far-away market. Barkema and Drogendijk (2007) confirmed that firms which internationalize first via exporting achieve higher success when they later internationalize in more distant markets via foreign direct investment (FDI).
2 MNEs in such a globalized environment thus is: Which international expansion behavior should be followed in order to survive successfully in a chosen target market? For this reasoning, it is important for MNEs to be aware of the abovementioned differences, especially the differences in national culture, in order to make efficient decisions concerning the firm’s international activities (Hitt, Keats & De Marie, 1998). Beyond that, it is argued that the political stability of the target market has an effect on the link between cultural distance and a firm’s international business operations.
3 target market’s political stability on the abovementioned relation is tested. Hence, the research question of this academic study is:
“Does cultural distance influence a firm’s international expansion behavior? How is this relationship moderated by the target market’s political stability?
Attention is paid to the influence of the cultural dimensions identified by Geert Hofstede (1980; 2001). To that end, data is gathered from German firms from two different industries: (1) the automotive industry and (2) the mechanical engineering industry. The here investigated target markets are located in North America, Oceania, Latin America, Asia, and Africa (see also chapter 3.3.1).
The aim of this study is to supplement existing international business and cultural distance literature by shedding further light on the influence of national culture characteristics on firms’ international expansion strategies. MNEs keep exploring more cultural distant markets and, considering that existing results are inconsistent, the underlying issue of this paper is of great importance for successful firm performance in international business. If it is known how exactly differences in national culture influence a firm’s decision making concerning internationalization, firms can make more efficient choices which might decrease the internationalization failure rate. In addition, political stability of the target country is incorporated into this study as a moderator variable. It is believed that this aspect has the ability to moderate the impact of national cultural distance on firms’ expansion behavior. By examining the moderating role of target market political stability, the contribution of this study to existing literature is twofold. First, it adds to the clarification of existing (conflicting) results. Second, it illustrates how the relationship between culture and firms’ international expansion patterns is influenced by a specific exogenous, dynamic, and non-controllable factor. Political stability represents a framework condition of the target market which needs to be accepted by the firm. However, firms need to keep in mind that political stability is not stable over time, as country situations are subject to continuous change.
4
2 Theoretical Background
When expanding internationally, firms often lack crucial knowledge about the target country’s cultural and institutional patterns (Weber & Tarba, 2010). Hence, firms face additional costs which arise due to the liability of foreignness (Zaheer, 1995). What is more, entering new markets can be accompanied by so-called “culture-shocks” which, in reverse, affect the management’s decisions on international expansion behavior and ownership structures (Barkema & Vermeulen, 1997).
In this following section, all relevant theories and concepts used within this analysis are introduced. Section 2.1 explains the notions of cultural distance, liability of foreignness, institutional theory, and how all these concepts are related to national cultural distance and a firm’s international expansion patterns. Section 2.2 presents an overview of Hofstede’s cultural dimensions, and section 2.3 introduces different international expansion alternatives for firm internationalization. Furthermore, findings from former studies within this research area are presented in section 2.4. Together, this leads to the formation of hypotheses in section 2.5 and the construction of the research model in section 2.6.
2.1 Cultural Distance, Liability of Foreignness, and Institutional Theory
Cultural distance indicates to what extent national cultures of home and host countries differ from each other and, accordingly, to how much newness and unfamiliarity the expanding firm is exposed (Shenkar, 2001). A substantial cultural distance between two countries can hamper the internationalization of a firm as the firm faces a variety of challenges (Collinson & Rugman, 2008). This illustrates that higher cultural distance increases the difficulties of international expansion processes and its associated risk to fail (Johnson, Lenartowicz & Apud, 2006). In the field of international business research, only few constructs have been more extensively explored than cultural distance. As a consequence, the concept has been applied to almost every discipline in business administration, such as strategic management or organizational behavior. Within strategic management, the most explored area is the one of FDI, where it was, among others, used to explore its impact on a firm’s international expansion and market entry mode choice (Shenkar, 2012).
5 (Zaheer, 1995). Hence, it comprises all additional expenses a firm faces when operating in an overseas market, for which local firms do not have to pay. Examples are increased organization costs and the managerial inability to understand culturally distant customers’ mindsets with the associated possibility for consumer discrimination. As a result, liability of foreignness, which arises through cultural distance, enhances a firm’s risk to fail in the foreign market.
The resource-based view of the firm suggests that, in order to overcome the liability of foreignness, MNEs should import their firm-specific resources into the host-market for the purpose of the sustainment of the firms’ competitive advantages in the target market (Barney, 1991). Correspondingly, full ownership is preferred. Here it can be highlighted that knowledge and awareness of cultural differences and cultural distance is crucial for business success.
Another related concept to cultural distance and liability of foreignness is the institutional theory. Each country has its own institutional environment which enables and constrains behavior. It consists of three pillars: (1) regulative (official rules, monitoring), (2) normative (unwritten rules about how things are done, related to social life), and (3) cognitive pillar (subjective rules and behavior that derive from culture and social groups) (Scott, 2008). Firms are embedded into their home country institutional environment (Granovetter, 1985) and, when entering new markets, firms need to adapt to the new institutional environment in order to overcome the liability of foreignness. The achievement of such embeddedness within the target country, which means to achieve a fit between the firm itself and its external environment, is crucial in order to establish legitimacy which, in turns, enables the firm to operate its business successfully. Moreover, coercive isomorphic pressures (political and cultural pressures to adapt) raise the importance of a firm’s embeddedness in the host market (Di Maggio & Powell, 1983). Only if the firm is willing to adapt to the new, unknown institutional environment such a fit can be generated (Mattes, 2013). This fit enables firms to enhance their legitimacy and it facilitates successful business operations in the foreign market by bridging cultural distance (Di Maggio & Powell, 1983).
6
2.2 Hofstede’s Cultural Dimensions
According to Geert Hofstede (1980), culture is the “collective programming of the mind which distinguishes the members of one human group from another” (Hofstede, 1980, p. 25), assuming that entire societies share the same “programming”. What is more, each culture inherits a certain set of values which determines a society’s desires, preferences, and decision making. Culture functions as a means that provides individuals with standards and norms, which help people to distinguish between what is right and what is wrong (Miroshnik, 2002). Previous research revealed that differences in national culture and the associated lack of cross-cultural knowledge are the primary reason for firm’s failure abroad, as people fail to understand each other (Miroshnik, 2002; Johnson, Lenartowicz & Apud, 2006).
The present study is based on Hofstede’s cultural dimensions. In an intensive research within the company IBM (which originally covered 40 different countries) between 1967 and 1973, Geert Hofstede (1980) found that cultural differences between countries can be captured via four major dimensions, out of which each single dimension is unique to every country. These dimensions are: (1) Power Distance, (2) Individualism, (3) Masculinity, and (4) Uncertainty Avoidance. Later, in 1985, Hofstede (2001) added a fifth cultural dimension to the national culture framework: (5) Long-term Orientation. This dimension is independent from the original study of IBM. Nowadays, scores on all cultural dimensions are available for 76 countries (The Hofstede Centre, 2016). Although this approach has been exposed to criticism in the last years (i.e. Shenkar, 2001; Sivakumar & Nakata, 2001; Ramsey, Barakat & Monteiro, 2013), it is still the most widely applied concept when measuring the influence of national culture. Debates arose about whether Hofstede’s cultural framework can still be applied today. Yet, instead of proofing Hofstede’s approach as being outdated, scholars noticed that many studies confirm and strengthen Hofstede’s conclusions (Kirkman, Lowe & Gibson, 2006). Moreover, Beugelsdijk at al. (2015) recently found that national cultures are indeed subject to ongoing changes. Interestingly, all countries move constantly into the same direction which implies that cultural differences between nations remain relatively constant. Consequently, cultural change needs to be seen as being absolute and not relative. Country scores on the cultural dimensions might have changed over time. However, as scores from all countries changed in relation to each other, the cultural framework of Hofstede is still relevant today when investigating cross-cultural influences (Beugelsdijk, Maseland & Van Hoorn, 2015).
7 in its national cultural system (Granovetter, 1985), every single dimension highly influences and shapes a society. Nevertheless this research paper focuses only on two out of the five presented cultural dimensions: (1) Power Distance and (2) Long-term Orientation. Different motivations led to this decision and the reasoning for this choice can be found on page 8.
2.2.1 Power Distance
The first dimension, power distance, explains how less powerful people expect and/or accept unequal power distribution among individuals. People from a country with a high power distance score are more likely to accept hierarchies in the business environment than people from a country with a low power distance index (Hofstede, 1980; 2001). Therefore, this dimension measures less powerful peoples’ perception of interpersonal power between two people. From the organizational point of view, a certain level of power distance is appreciated in such as it facilitates control and decision-making. However, too much power can discourage trust and work motivation (ib.), and decision making becomes autocratic (Van der Vegt, Van de Viert & Huang, 2005). Hussler (2004) found that a higher level of power distance between countries leads to decreased levels of innovation and change within firms, since new ideas are seen as a potential threat to the established power.
2.2.2 Long-term Orientation vs. Short-term Orientation
The fifth cultural dimension, long-term orientation, has been added to the national culture framework by Geert Hofstede in 1985, and it is independent from his original study of IBM. It was developed via the so-called “Chinese Value Survey” which investigated cultural values among 23 new countries, by sending questionnaires to 100 students of each investigated country. The results showed high overlaps with Hofstede’s original four cultural dimensions. Nevertheless, one additional dimension crystallized out of this survey. This new dimension was labeled “Long-term Orientation”. Originally, it was found to be correlated to economic growth (this correlation does not exist anymore today) and exactly for that reason Hofstede decided to include this dimension into his cultural framework (Hofstede & Minkov, 2010).
8 rights, and freedom. The focus of firms is on profits within the current year (Hofstede & Minkov, 2010).
It was decided to only concentrate on the two above presented cultural dimension for different motives: Overall, existing studies focused on the dimension of individualism vs. collectivism. However, by focusing only on a single dimension, results remain narrow. For this reason, investigating the influence of power distance and long-term orientation can complement preceding research. Power distance is an important dimension as discrepancies about how to manage the company in terms of authority and decision making are directly connected to firm performance (Walter, Lechner & Kellermanns, 2008). Long-term orientation focuses on people’s ability to learn and adapt to new and unpredictable situations. With that said, it becomes clear that this dimension has high influential power on how people from different countries work together. Due to the fact that this dimension was later added to the original four cultural dimensions, studies which applied the Kogut and Singh index all neglected long-term orientation and, hence, this dimension remained untended. It was decided to neglect masculinity vs. femininity since it is believed that its influence on international business strategies is not significant. The nature of this dimension is not only concerned about business but also about private life and, accordingly, it might gain in importance after, i.e., a subsidiary has been successfully established in the target market and aspects such as work-life balance become an issue. Lastly, uncertainty avoidance is also not included into this study. Although this dimension might influence a firm’s decision, it is claimed that parts of its nature are also visible in the dimension of long-term orientation, such as the willingness to prepare for unknown, future events.
2.3 International Expansion Alternatives
9 exporting (arm’s length transactions), which represent low commitment international activities, and (2) wholly owned subsidiaries (hierarchy), which indicates high commitment of the firm (figure 2.1).
Firms, expanding to developed and to developing markets, need to keep major differences in mind. Both economical contexts offer different resources to the firm, and both contexts are embedded in different institutional environments. These discrepancies are important to remember when firms expand internationally (Child & Yan, 2003). Respectively, this study not only focuses on the influence of cultural distance on international expansion decisions but also on the impact of political stability of the target market on internationalization behavior.
According to previous studies in this filed, higher cultural distance between home and host countries increases the likelihood that the expanding firm will use a business activity which ensures high control (Padmanabhan & Cho 1996). On the contrary, other scholars found that high differences in national culture lead to shared ownership as a preferred mode of doing business (Kogut & Singh, 1988). With that said, it becomes clear that previous findings are inconsistent. Hence, the aim of this study is to investigate the influence of cultural distance on decisions concerned with the firm’s international expansion patterns once again in order to be able to understand the previous, contradictory results better. In addition, the moderating effect of the target country’s political stability is taken into account. By adding
Figure 2.1: Key characteristics of internationalization alternatives, own
10 political stability to this scientific investigation, it is hoped to complement results from preceding studies.
2.3.1 Exporting
Exporting represents the most basic mode of international firm expansion. Using exporting, firms only send their products from one country to another but the firm’s knowledge stays within the home country (Ramsey, Barakat & Monteiro, 2013). Partners set up their business relationships via specified contracts. Such partnerships are often only short-term and exporting can be considered as a form of arm’s length transactions, where partners can exit the affiliation easily. Neither knowledge, nor equity is shared (Oxley, 1999) which illustrates that the firm’s commitment is low.
Exporting enables firms to spread to foreign markets without having to make major investments in the target market. Usually, firms that expand to a foreign market for the first time use exporting as it neither involves high risks nor high investments (Inkpen & Beamish, 1997), as suggested by Johanson and Wiedersheim-Paul’s (1975) Uppsala model. This so-called “incremental-approach” implies that, after having learned from exporting in one country, firms might expand their operations, within a specific market, through equity ownership modes. On the other hand, some firms might not be willing to increase their commitment at any time and, hence, do not change their exporting activities to other activities (Young, 1987). The downside of exporting, however, is that it only creates limited profits and the exporting firm only has little marketing control which is essential for penetrating the market (Agarwal & Ramaswami, 1992).
2.3.2 Foreign Direct Investment: Wholly Owned Subsidiary
In contrast to exporting, some firms prefer international business activities in the form of foreign direct investment (FDI). Expanding through FDI, firms can expect higher returns on investments. In spite of these financial benefits, due to a high level of firm commitment, firms face significant higher levels of risks and management complexity. Among the most common FDI modes are wholly owned subsidiaries (Chung & Enderwick, 2001).
11 there is no foreign partner involved, which limits the hazard of opportunistic behavior and uncertainty associated with a partnership, such as leakage of intellectual property. Furthermore, initiation costs, namely search and information seeking about potential partners, as well as negotiation costs are avoided (Gulati, 1995). What is more, since the control over the subsidiary remains solely in the hands of the home country managers, no conflicts over the use of assets can arise (Kogut & Singh, 1988) and it is easier to implement and manage the business processes abroad (Child & Yan, 2003). Hierarchies pursue long-term oriented goals. Knowledge and innovations are generated and shared within the firm, but not with outsiders (Barney, 1991). On the other hand, the expanding firm might lack crucial knowledge of the local market due to the absence of local partners.
2.4 Findings and Limitations from previous studies
Since 1980, when Geert Hofstede first published his work “Culture’s consequences: International differences in work-related values”, his framework with the four (nowadays five) cultural dimensions has widely been applied in cross-national research. The construct was most applied in the field of international business where it has been used, among others, to investigate headquarters-subsidiary relations, expatriate management, and international expansion decisions (Shenkar, 2001). Nonetheless, Hofstede’s work has been exposed to criticism. Sivakumar and Nakata (2001) argued that, within the national cultural framework, culture is simplified due to the fact that it is reduced to only four (nowadays five) dimensions and, the original study which discovered the cultural dimensions is only based on one single company, namely IBM. Moreover, Hofstede did not take within-nations cultural differences into account. Other criticisms were voiced by Shenkar (2001). He claimed that most research in international business applied Kogut and Singh’s (1988) index which compares the differences between a subsidiary’s country score and a home country score (Hutzschenreuter, Kleindienst & Lange, 2016). This index, however, has not been updated as Hofstede’s work continued. Consequently, the index does not include the later developed, fifth dimension, of long-term orientation. Most importantly, it makes invalid assumptions of equivalence of the cultural dimensions, although Hofstede stated that some cultural gaps might be less problematic than others (Shenkar, 2001).
12 that, with increased cultural distance, the tendency to prefer an international joint venture over an acquisition increased as well (Kogut & Singh, 1988; Chang & Rosenzweig, 2001). Other academics showed that, as cultural distance increased, a firm’s preferred international expansion mode was a WOS (Padmanabhan & Cho, 1996). Opposed to that, Barkema & Vermeulen (1998) stated that, in such situations, WOSs were generally less preferred than shared-equity options. In cases where exogenous factors were included, the majority of scientists focused on country risk. Country risk was mostly defined as a multidimensional construct which includes political, legal, cultural, and economical environmental aspects. It was argued that country risk affects the stability of doing business and, thus, it influences firms’ decision making and some expansion alternatives might be preferred over others. Generally speaking, within all these analyses, countries can be classified into different categories such as safe and risky (cf. Anderson & Gatignon, 1988; Barkema & Vermeulen, 1998). Other scholars, on the other hand, tried to assess country risk via the perceived risk of foreign bankers (i.e. Pan, 1996). Findings from these studies highlighted that in high risk countries, firms try to avoid high levels of ownership (cf. Anderson & Gatignon, 1988; Kim & Hwang, 1992; Arora & Fosfuri, 2000). However, such effects differ between equity and non-equity modes (Pan & Tse, 2000). As it can be seen, the just mentioned studies were inconsistent in the way they defined country risk. Besides, the notion country risk included various sources of risk, which can be broken down to different influencing aspects. Therefore one component of the former original country risk measure is included into this study as a moderator variable: Political stability (see also chapter 3.3.4). This dimension is evaluated and documented by The Global Economy and exists for 191 countries in the world. Hence, here it is focused on scores that are accepted by institutions and firms worldwide. Opposed to that, former studies constructed and measured country risk via self-defined empirical measures. Cluster-analysis, which grouped countries into three different clusters, indicating low, medium, and high country risk, is one example for such a self-invented measurement (c.f. Barkema & Vermeulen, 1998).
13 CEE countries (i.e. Barkema & Vermeulen, 1997; 1998; Hennart & Larimo, 1998; Brouthers & Brouthers, 2003; Barkema & Drogendijk, 2007). Furthermore, the most investigated economic sectors are the manufacturing sector and the service sector (i.e. Erramilli, 1991; Erramilli & Rao, 1993; Shane, 1994; Brouthers & Brouthers, 2003). With that said, widespread limitations are the limited number of target markets, the small variety of explicitly studied countries and the therewith associated small sample size. Moreover, studies took place within one or sometime two continental regions and mostly, exclusively equity-based entry modes were examined. These limitations suggest to further examine the influence of national cultural distance on international expansion strategies in a new context, which comprises other sectors and other home and target countries. Additionally, non-equity expansion patterns, such as exporting and the influence of national cultural distance on a firm’s current international business operations should be explored.
Table 2.1: Overview of previous studies about the influence of national cultural distance on international expansion behavior
of firms
Author(s) Sample Main Findings
Kogut and Singh (1988)
Market entries into the USA
Increased cultural distance between countries leads to an increased tendency to choose an IJV over an acquisition.
Erramilli (1991) American service firms engaged in international operations
Firms are more likely to enter countries which have a high cultural distance from the home country, when they already gained international experiences.
Kim and Hwang (1992) Investigation of the preference of American MNEs
Increased cultural distance between the markets leads to licensing rather than to an IJV or a WOS.
WOS are preferred when country risk is low.
Erramilli and Rao (1993)
Entry mode preferences of American service sector firms
A firm’s entry mode choice is moderated by its asset specificity. Such costs influence the decision for either full or shared ownership.
Shane (1994) Entries of American
MNEs in the
manufacturing sector
Cultural distance is used as a control variable: A firm’s preferred market entry mode is via FDI.
Agarwal (1994) American firms that set up foreign
manufacturing subsidiaries
Higher cultural distance will lead to partnerships such as IJVs.
High external uncertainty leads to a preference of higher control entry modes.
Padmanabhan and Cho (1996)
Japanese market entries into 36 countries
Japanese firms prefer WOSs as market entry mode over shared ownership if cultural distance is high.
Erramilli et al. (1997) Subsidiary ownership preferences of South Korean MNEs
In case of an IJV, equity partners aim at either an equal or a majority share when cultural distance is high.
Barkema and Vermeulen (1998)
Non-financial large Dutch firms
Firms prefer partial ownership when expanding into countries with highly different cultures or when country risk is high.
Hennart and Larimo (1998)
Japanese and Finish companies entering the USA
With increased cultural distance, the chances for WOSs decrease and firms prefer to choose shared-equity ventures.
Makino and Neupert (2000)
Japanese firms investing in the USA and vice versa
Firms from home countries with high scores on power distance and uncertainty avoidance prefer majority ownership to enter new markets.
Chang and Rosenzweig (2001)
European and Japanese entries into the USA
Increased cultural distance between countries leads to an increased tendency to choose an IJV over an acquisition.
14
Brouthers (2003) mode choices between Dutch, British, and German service and manufacturing firms doing business in CEE countries
uncertainties lead to the formation of an IJV. With increasing international experience, the preference for a WOS increases as well.
Service firms: High perceived uncertainty of the target
market leads to the formation of IJV. If high asset specific investments are involved, firms prefer WOSs as entry mode. Larger firms and firms with prior international experience prefer WOSs. Higher legal restrictions lead to the preference of IJVs.
Gaur, Delios and Singh (2007)
Foreign subsidiaries of Japanese firms
When entering a foreign market, firms rely on managers from the home country if cultural distance is high.
Barkema and Drogendijk (2007)
Foreign market entry of Dutch firms into CEE countries
Incremental internationalization enhances a firm’s knowledge base which leads to improved performance. Although firms which directly enter markets via FDI suffer from low performance in the beginning, they are able to catch up after a certain time period.
Wang and Schaan (2008)
Japanese foreign direct investment overseas
Higher cultural distance increases the likelihood for WOSs.
Chari and Chang (2009)
Cross-border
acquisitions of American firms
In cultural distant markets, firm prefer to only partially acquire a partner via an equity stake instead of acquiring the target firm completely.
Firms prefer low-equity entry modes in case of high exogenous country risk.
Arregle, Miller, Hitt and Beamish (2013)
Japanese MNEs
operating in 45 countries
Firms tend to expand regionally via FDI into one country first and later into more countries within this region. Firms evaluate the institutional environment of regions rather than of single countries.
15
2.5 Hypotheses
Previous studies have focused on different national cultural dimensions. Many followed the Kogut and Singh (1988) index, which combines all initial four dimensions (power distance, individualism, masculinity, and uncertainty avoidance) into one formula. Interestingly, and also surprisingly, only few researchers investigated the impact of different national cultural dimensions separately.
This study investigates two out of Hofstede’s five national cultural dimensions in isolation from each other. By doing so it is hoped to find answers to whether one cultural dimension has more influence on a firm’s decision making than others. In the analysis, it is focused on the national cultural dimensions of power distance and long-term orientation.
Power distance has significant impacts on the way firms are structured, and how power is allocated within the firm (Erramilli, 1996). In countries which show a low score on the power distance dimension, power is equally distributed within a firm and hierarchies are flat. Managers tend to consult subordinates before decision making and authority (in general) is decentralized. On the contrary, in high power distant countries, employees on top of the hierarchy of the firm have more power than others, which results in a more autocratic leadership style (Hofstede, 1980; Erramilli, 1996). This leads to the conclusion that it is difficult for managers from low power distant countries to work together, and to cooperate with managers from high power distant countries. People from countries were power distance is high would feel uneasy when working together with people from low power distant countries since they would lack the instructions and guidance from a superior managers. On the other hand, people who are used to a working environment where power is equally distributed would not be able to accept to be treated as a subordinate. Thus, in order to prevent such conflicting situations, it is hypothesized that:
Hypothesis 1: The higher the power distance between the home and the host market,
the greater the propensity that a firm pursues WOSs as international business operations.
16 Accordingly, this cultural dimension has important influences on a firm’s behavior regarding strategic decisions, such as how to operate in foreign countries. Short-term oriented societies are more concerned about the present. Firms prefer to keep long-proven traditions from the past instead of adapting new, future oriented aspects as they are convinced that what was good in the past will also be applicable in the future (The Hofstede Centre, 2016). The fact that firms from long- and short-term oriented countries pursue completely different firm goals let one belief that employees and managers from these different backgrounds can hardly manage to conduct successful business together. This leads to the following hypothesis:
Hypothesis 2: The higher the long-term orientation difference between the home and
the host market, the greater the propensity that a firm pursues WOSs as international business operations.
In addition to the two independent variables, the influence of a moderator variable is tested. Thunnell (1977) examined the influence of a host country’s political condition on the investment behavior of international firms. He found that, due to the fact that political instability is more present in Latin American countries, political instability has different effects in Europe than in Latin America. Hence, firms planning to expand into more risky countries need to make a thorough assessment of country characteristic. Aside from that, it was found that firms tend to avoid full ownership modes in high risk countries (Anderson & Gatignon, 1988). Existing literature states that an instable institutional environment of the target market increases uncertainty and investment risk (Dhanaraj & Beamish, 2009; Kotabe, Jiang & Murray, 2014; Barros, 2015). Such an external uncertainty can be explained as the instability of the firm’s external business environment (Anderson & Gatignon, 1986). On account of the increased costs associated with greater risk and uncertainty, it is assumed that the target country’s political stability moderates the relationship between national cultural distance and a firm’s preference for carrying out its international expansion activities either via exporting or via a WOS. Therefore, it is hypothesized that:
Hypothesis 3: The higher the political stability of the target country, the lower the
impact of power distance on the international expansion pattern of a firm.
Hypothesis 4: The higher the political stability of the target country, the lower the
17
2.6 Research Model
Relating the presented constructs to the defined hypotheses, a research model can be drawn (figure 2.2) which represents the expected influence of national cultural distance on a firm’s international expansion behavior. Additionally, following Shenkar’s (2012) suggestion, in this study it is controlled for the firm’s size, age, and international experience. These variables can be seen as mechanisms on the firm level that can enable firms to close cultural distance. Firm size determines the amount of resources available when entering a foreign market. As a consequence, it influences, together with a firm’s international experience, the knowledge transfer between countries, intercultural trainings, and the adaption process in the target market (Shenkar, 2012). Likewise, with increasing international experience, firms have additional acquired knowledge and established relationships, which will affect future market entries (Anderson & Gatignon, 1986). Moreover, by adding political stability of the target country as a moderator variable, it is hoped to learn whether the affiliation between national cultural distance and the decision about how a firm pursues its current international business operations is influenced by an additional, external factor.
18
3 Methodology
As the underlying question of this paper, “Does cultural distance influence a firm’s international expansion behavior? How is this relationship moderated by the target market’s political stability?”, is of closed nature, this research follows a deductive approach and quantitative research is conducted. The hypotheses developed in the previous part are based on existing theory and in chapter four it is aimed to test these hypotheses and to draw generalizable conclusions from the results.
3.1 Procedure
Since this research concentrates on constructs which have already been widely discussed in the literature, a quantitative study is preferred over a qualitative study. The underlying goal of quantitative research is to measure a specific phenomenon. As such, quantitative research is used when a particular question should be explained and hypotheses should be tested with numerical data. This kind of research typically draws from a large sample size (Cooper & Schindler, 2006), in this particular case from 57 German MNEs with business activities in 19 different target countries. After having excluded the incomplete observations, 653 cases were left.
3.2 Variables
Referring to the research model, the variables displayed in table 3.1 can be defined for this analysis.
19 with their resource stocks. Secondly, the age of the firm needs to be controlled for. Long-established multinationals might have already gathered international experiences and developed an internationalization routine which would lead to biased results. The third control variable is the firm’s international experience. This variable is important for the same reason as the second control variable, firm age (c.f. Shenkar, 2012). The more experienced a firm is, the less it might be influenced by national cultural distance since managers and employees most likely developed intercultural skills and understandings in previous transactions. Lastly, the model also includes a moderator variable: Political stability of the target country.
Table 3.1: Variables and measures
Type Variable Description Data
Source Dependent
Variable
The firm’s current international business operation: Wholly owned subsidiary (hierarchical firm, high commitment)
Indicates whether a firm has more current international activities in the target markets in form of wholly owned subsidiaries (coded as 1). The German chamber of commerce Dependent Variable
The firm’s current international business operation: Exporting (arm’s length transaction, low commitment)
Indicates whether a firm has more current international activities in the target markets in form of exporting (coded as 0). The German chamber of commerce Independent Variable
Hofstede’s national cultural dimension: Power Distance
The difference between the host and the home country score on the cultural dimension of power distance.
The Hofstede Center
Independent Variable
Hofstede’s national cultural dimension: Long-term Orientation
The difference between the host and the home country score on the cultural dimension of long-term orientation.
The Hofstede Center
Moderator Variable
Political stability of the target country
The political stability of each investigated country according to the index of The Global Economy.
The Global Economy
Control Variable
Firm size The size of each firm measured via the firm’s total amount of employees.
The German chamber of commerce, firms’ website Control Variable
Firm age The age of each firm measured in years from the year of establishment to the year 2014.
Firms’ website
Control Variable
Firm’s international experience The total number of a firm’s
international subsidiaries based on the number of current subsidiaries.
The German chamber of commerce
3.3 Data Collection
20 German chamber of commerce databases of firms located in the federal states of Baden-Württemberg and Bavaria (BaWü IHK, 2016; BIHK e.V., 2016). Data about a firm’s international experience and firm size, two of the control variables, were taken from the German chamber of commerce databases as well. Firm age, the remaining control variable was directly obtained from the firms’ websites. Data on the moderator variable were available on the website of The Global Economy (The Global Economy, 2016). Finally, the independent variable used data published by Geert Hofstede (The Hofstede Centre, 2016).
3.3.1 The Sample
In the scope of this research, all investigated firms are of German origin. The sample of this study consists of 57 German large and very large firms (MNEs). Large and very large firms are defined as organizations that have at least 500 employees and a minimum annual turnover of 50m Euros (Gabler Wirtschaftslexikon, 2016). The decision to only take firms originally located in Germany into consideration is based on various arguments: First of all, it is advantageous to examine MNEs of only one country since it ensures that the data is more harmonized and comparable. This is based on the underlying fact that all firms operate under the same home country conditions. Moreover, munificence varies among countries and, hence, only when comparing firms of the same home country, information about international expansion strategies is reliable. Therefore it is hoped that, by focusing only on one single home-market, results will reveal deeper insights into the influence of national cultural distance on firm’s international expansion behavior.
Germany is the fifth largest economy in the world and is, among European countries, the biggest participant in terms of outward FDI (approx. 8.3% of the world’s total outward FDI in 2014) (UNCTAD, 2013). Besides, Germany is a leading exporter of vehicles and machinery (Central Intelligence Agency, 2016). The economy functions well and developments have been stable over the past years (Statista, 2016a). In short, Germany is the most important economy in Europe with a GDP of 2.91bn Euros (in 2014) (Statista, 2016b) and, thus, findings on German firms’ strategies have important implications for other countries. For these reasons, it is believed that investigating German firms can deliver important results that might reveal the secret of Germany’s well-functioning economy. The selected firms operate in two different sectors: (1) the automotive industry and (2) the mechanical engineering industry, which are among Germany’s leading industries (Make it in Germany, 2016).
21 different continents, the variety of different cultures increases (Stanton, Ramsamy, Seybolt & Elliott, 2012). Second, all countries show interesting varieties in the cultural dimensions of power distance and long-term orientation (table 3.3 and figure 3.1). This, in turn, can lead to interesting result of this study. Thirdly, all target markets are located outside of Europe in order to test firms’ internationalization behavior in far-away markets. Far-away markets are particularly of interest as such countries show significantly higher cultural distances and also the political stability is different outside the European Union, which generally increases the risk of internationalization.
To further illustrate the differences of the markets a brief overview of the selected target countries is given in table 3.2.
Table 3.2: Overview of the selected target countries, own representation based on The World Bank, 2016; Central
Intelligence Agency, 2016; OECD library, 2016.
All mentioned data is from the fiscal year of 2014.
This short presentation of each target market, the government form, and the GDP shows that there are not only large differences in national culture itself but also in the formal institutional environment. From this, it can be concluded that the target markets show a
Region Country Population
(people) GDP (in USD)
GDP per capita
(in USD) Government form
North America
USA 328.9 million $17.42 trillion $54,400
Constitution-based federal republic with a democratic tradition Canada 35.54 million $1.785 trillion $45,400
Parliamentary democracy with a constitutional monarchy
Oceania
Australia 22.75 million $1.454 trillion $64,700
Federal parliamentary democracy
(commonwealth) New Zealand 4.4 million $162.4 billion $35,700 Parliamentary democracy
(commonwealth)
Latin America
Mexico 125.4 million $1.295 trillion $18,100 Federal republic Brazil 200.3 million $3.265 trillion $15,800 Federal republic Venezuela 30.69 million $546.2 billion $17,900 Federal republic Colombia 47.79 million $377.7 billion $13,600 Republic Chile 17.76 million $258.1 billion $23,300 Republic
Asia
Japan 127.1 million $4.601 trillion $36,400
Parliamentary government with a constitutional monarchy China 1.364 billion $10.35 trillion $13,400 Communist state India 1.2 billion $7.484 trillion $5,900 Federal republic Singapore 5.470 million $307.9 billion $83,900 Parliamentary republic South Korea 49.1 million $1.410 trillion $35,700 Republic
Africa
22 variety in political stability. Consequently, all countries bring different risks and opportunities with them as a target market, and the sample represents a large number of differences.
3.3.2 Dependent Variables
The dependent variable is a categorical one. Each firm can fall into one out of the two categories: Exporting or WOS. For this reason, a dummy variable was created for the analysis in IBM SPSS where 0 represents the group of exporting and 1 depicts WOSs (table 3.4).
In order to be able to assign each firm to the corresponding category, data about current international activities in the form of either exporting or a WOS were gathered for all 57 firms within the 19 target countries. Therefore, each firm was paired with each target market separately. In cases where the firm exports goods and services to a country it was assigned to the exporting group. If the firm operates a WOS in one country, it was allocated in the category of WOSs. Some firms follow both strategies in one country at the same time. Since it is believed that firms first exported to that country and later established a WOS, which then represents the current international expansion behavior, all these cases were assigned into the WOS category. In order to ensure the greatest possible reliability of the data, numbers about international activities from the two databases of the German Chamber of Commerce were compared to data available on the firm websites. In case of large discrepancies, the numbers available on the German Chamber of Commerce websites were adjusted according to the firm websites.
3.3.3 Independent Variables
The independent variable, namely national cultural distance, was evaluated via Hofstede’s cultural dimensions. These data were, thus, retrieved from Geert Hofstede’s website (table 3.3 and figure 3.1). Cultural distance was calculated for each target market separately by taking the margin between the respective target market’s country score and the score of Germany. This was done for both investigated dimensions (power distance and long-term orientation).
Table 3.3 Cultural distance between the target markets and Germany, own representation based on The Hofstede Centre,
2016.
Country Power Distance Long-term Orientation
23 Japan 19 5 China 45 4 India 42 -32 Singapore 39 -11 South Korea 25 17 Angola 48 -68 Egypt 35 -76 South Africa 14 -49 Nigeria 45 -70 Morocco 35 -69
Figure 3.1: Cultural distance between the target markets and Germany, own representation based on The Hofstede Centre,
2016.
3.3.4 Moderator Variable
Political stability of the target country, the moderator variable, was measured via the political stability index. This index measures a country’s political stability from the range of -2.5, which indicates a weak political stability, to +2.5, which stands for high political stability (The Global Economy, 2016). The index consolidates seven different aspects which represent together a country’s political stability. These aspects are: Uncontrolled transfer of government power, armed conflicts, violent demonstrations, social unrest, international tensions, terrorism, and conflicts that arise from an ethnical, religious or regional basis (The Global Economy, 2016).
The moderator variable was included into the model with its actual index-scores and, hence, as a continuous variable (figure 3.2). Data were taken from The Global Economy (2016). Index scores are from the year 2014.
-57 -47 -62 -50 -59 -39 -67 -70 -52 5 4 -32 -11 17 -68 -76 -49 -70 -69 5 4 1 -13 46 34 46 32 28 19 45 42 39 25 48 35 14 45 35 -80 -60 -40 -20 0 20 40 60 Cultura l D is ta nce
Cultural Distance between the target markets and Germany
24
Figure 3.2: Political stability index of the target markets, own representation based on The Global Economy, 2016. 3.3.5 Control Variables
Lastly, the control variables were measured as followed: Firm size was measured in terms of the total number of employees per firm. There are four different groups to determine firm size: 500-999 employees, 1,000-4,999 employees, 5,000-9,999 employees, and more than 10,000 employees. These data were collected from the databases of the German chamber of commerce and from the firms’ websites. The age of each firm was measured in years from the date of establishment to the year 2014. This information was directly taken from the firms’ websites. Thirdly, a firm’s international experience was analyzed by counting the total number of countries a firm currently operates foreign subsidiaries in. These numbers were available in the databases of the German chamber of commerce.
3.4 Statistical Method and Statistical Model
Since this study follows the aim of analyzing the relationship between national cultural distance and a firm’s international expansion patterns, regression analysis is an appropriate research method (Field, 2013). More precise, binominal logistic regression analysis is applied. Logistic regression analysis enables researchers to assess the independent variables’ strength in explaining the categorical dependent variables. Consequently, this type of statistical analysis can indicate the relative importance of each predictor variable with regard to the dependent variable (Pallant, 2013). Moreover, in the case of the present research, the regression coefficient assesses the impact of national cultural distance on mechanical engineering and automotive firms’ propensity to pursue international expansion activities
0.62 1.18 1.08 1.49 -0.76 -0.01 -0.83 -1.12 0.49 1.02 -0.46 -0.96 1.23 0.19 -0.27 -1.58 -0.08 -2.11 -0.39 -2.50 -2.00 -1.50 -1.00 -0.50 0.00 0.50 1.00 1.50 2.00 2.50 P o litic a l Sta bil it y
25 either via exporting or via the establishment of WOSs (Field, 2013). The test is run for each of the abovementioned independent variables, one at a time (Pallant, 2013).
For testing the proposed hypotheses the following statistical model is used: (I) Probability (Y = WOS) = 1+𝑒1−𝑔(𝑥)
(II) while g(x) = ln ⌈(1−𝜋)𝜋 ⌉ = 𝛼 + 𝛽1𝑋1+ ⋯ + 𝛽𝑘𝑋𝑘
The left side of equation (II), g(x), stands for the dependent variable, where π/((1-π)) is the ratio of probabilities and is called the “odds”. The odds, also called odds ratio, states the change in probabilities of the dependent variable if the value of one of the independent variables increases by one unit (Pallant, 2013). The right side of the equation contains α, which is the constant of the model, β, the regression coefficient, and X, the respective variables. The predicted probability of exporting being chosen as an international business
strategy is: 1− Probability (Y = WOS). Equations for the models, which have been run for
testing the hypotheses, are displayed in Appendix B.
Within the IMB SPSS analysis the variables were defined as indicated in table 3.4.
Table 3.4: Coding of variables in IBM SPSS
Type Variable SPSS-Coding
Industry Criteria in the sample selection process 3 categories:
1: Automotive industry
2: Mechanical engineering industry 3: Both industries
Dependent Variable The firm’s current international business operation evaluated for each of the selected target markets separately: Exporting or WOS
2 categories: 0: Exporting
1: WOS (reference group)
Independent Variable Hofstede’s national cultural dimension:
Power Distance
Continuous variable: absolute number Power Distance was calculated for each target market separately by taking the margin between the respective target market’s country score and the score of Germany.
Independent Variable Hofstede’s national cultural dimension:
Long-term Orientation
Continuous variable: absolute number Long-term Orientation was calculated for each target market separately by taking the margin between the respective target market’s country score and the score of Germany.
26
Control Variable Firm Size 4 categories:
1: 500-999 employees (reference group) 2: 1,000-4,999 employees
3: 5,000-9,999 employees 4: 10,000+ employees
Control Variable Firm Age Continuous variable: absolute number
Measured in years from the date of establishment to the year 2014.
Control Variable Firm’s International Experience Continuous variable: absolute number The total number of countries a firm currently operates foreign subsidiaries in.
First of all, the variables were checked for outliers by looking at the boxplots. This revealed that four outliers exist among the control variables firm age, firm size, and firm international experience. By comparing the variable’s original mean value with the variable’s 5% trimmed mean value, it was ascertained that the extreme values do not have a strong influence on the mean value (Pallant, 2013). Therefore, all outliers can be kept within this analysis (Appendix C).
Logistic regression implies a necessary assumption that needs to be fulfilled in order to ensure unbiased and reliable outcomes: Multicollinearity (Sreejesh, Mohapatra & Anusree, 2013). The non-existence of multicollinearity requires that both predictor variables are independent from each other. To test whether one independent variable has a strong relationship with the other independent variable, the variance inflation factor (VIF) was calculated. Multicollinearity is not given when the VIF values are less than 10 and the tolerance values (1/VIF) are not less than 0.1 (Field, 2013). Table 3.5 confirms that no multicollinearity among the independent variables exists.
Table 3.5: Test for multicollinearity
Tolerance (1/VIF) VIF
Power Distance .343 2.916
Long-term Orientation .662 1.512
Political Stability .317 3.150
Firm Size .796 1.257
Firm Age .958 1.004
Firm International Experience .775 1.290
Mean VIF 1.855
27
4 Empirical Results
After having collected all the required data from the resources described above, the entire dataset was analyzed using IBM SPSS. First of all, the assumption of multicollinearity is met and no disturbing outliers exist. On these grounds, the descriptive statistics (Pallant, 2013) and the main analysis were performed. Lastly, this chapter discusses the findings of the analyses. As the sample consists of 57 firms and 19 target countries, a total of 1,083 observations was achieved. However, not each firm has business operations in each of the 19 foreign markets. After having deleted all cases where firms did not pursue business activities in certain countries, 653 observations were left.
4.1 Descriptive Statistics
Table 4.1 provides an overview of the descriptive statistics for the continuous data. Descriptive statistics for the categorical data are displayed in table 4.2 (frequency statistics). From the descriptive statistics table can be concluded that the mean age of all firms is 101.33 years, with a minimum age of 26 and a maximum age of 297 years. On average, a firm has international activities in 15 foreign markets (firm international experience). The cultural distance of the dimension of long-term orientation varies from -76 to 17 and for power distance it varies from -13 to 48. Furthermore, the mean political stability of the target markets is 0.0524, with a minimum political stability of -2.11 and a maximum of 1.49.
Table 4.1: Descriptive statistics for continuous data
Variable Minimum Maximum Mean Std. Dev.
Firm Age 26.00 297.00 101.33 43.85
Firm International Experience 1.0 52.00 15.34 12.01
Long-term Orientation -76.0 17.00 -.025 1.227
Power Distance -13.0 48.00 26.04 17.56
Political Stability -2.11 1.49 .0524 .93
Number of observations: 653
28
Table 4.2: Frequency statistics for categorical data
Variable Frequency Percent (%)
Industry Automotive 267 40.9 Mechanical engineering 307 47.0 Both industries 79 12.1 International Expansion Strategy Export 402 61.6 WOS 251 38.4 Firm Size 500-999 181 27.7 1000-4999 292 44.7 5000-9999 38 5.8 10000+ 142 21.7 Number of observations: 653 4.2 Hypotheses Testing
In order to analyze the in the literature section established hypotheses, a binominal logistic regression analysis was performed. As mentioned above, there are two independent, one dependent, and one moderator variable to test. Within the entire analysis, WOS was used as the reference category for the dependent variable.
Four models were run. The first model included all control variables. This was done to estimate the firm level control variables’ effects on the probability to choose one international expansion strategy over another, namely the establishment of a WOS over exporting activities. In the second model the independent variable was introduced and in the third model, the moderator variable (political stability) was added. Lastly, in the fourth model, the moderating effect was tested by having included the interaction effect between the independent and the moderator variable into the model (Jose, 2013). These four models have been conducted for both of the independent variables, power distance and long-term orientation, separately. Table 4.3 gives an overview of the logistic regression results of the independent variable “Power Distance”. Results for the independent variable “Long-term Orientation” are displayed in table 4.4.
