Entry Mode Strategy of European Energy Companies: The Effect of Distances on a Choice of Entry Mode.

(1)

Entry Mode Strategy of European Energy Companies: The Effect of

Distances on a Choice of Entry Mode.

By

Alexander Malinovski

University of Groningen Faculty of Economics and Business

(2)

2

Abstract

The energy sector is constantly developing, and companies which are operating in the energy sector are always looking to diversify their portfolio by internationalizing to different countries. Therefore, this thesis aims to explore how institutions and cultures are affecting the entry mode choices of large European energy companies. In this research, entry mode strategy is the dichotomous dependent variable (partial/full ownership). Independent variables in this research cover all institutional and cultural pressures that face a company when internationalizing to a different country. The sample size of this research includes 230 entries from Europe, covering 23 home and 44 host countries. This research concludes that institutional distance is positively related to a collaborative entry mode. It also finds that culture affects entry mode strategy in a quite specific way. It confirms that not every Hofstede dimension is equally important when studying entry mode strategy. The uncertainty avoidance dimension has a significant influence on the choice of an entry mode.

Key words: Entry mode; Greenfield investment; Acquisition; Foreign direct; Investment; Culture; Hofstede culture dimensions; Energy Companies

Research theme: Entry mode strategy of energy companies.

Supervisor: Rudi de Vries Co-assessor: Visscher

(3)

3

Introduction and Central Research Question

Initial Motive

The energy market in Europe is huge but stagnant. The European energy market reached a value of 1,373.2 Billion dollars in 2015 (MarketLine, 2016). Oil and oil products are still the largest segment of the energy market. Oil consumption accounts for 36,8% of the sector’s value (MarketLine, 2016). However, the European wholesale gas and oil markets are suffering from uncertainty over the strengths of the economic recovery from the sovereign crisis of 2008 (McKinsey, 2014). Moreover, renewable energy sources are slightly affecting demand for oil and gas in Europe (Deloitte, 2015). In the graph below (Figure 1), we can see a clear pattern that renewables will gain a significant market share in the future. This can be explained by the Paris agreement, which was signed in 2015 (United Nations, 2015). Moreover, we see that fossil fuel market share will be suppressed in the near future from 82% to 76%. It is planned that renewables will fill this market share. In Europe, prices for the oil and gas are becoming more and more transparent, making it more difficult to earn revenue on the spreads (Deloitte, 2015). As a result, we see that energy companies start to diversify their portfolio structures. Investments into new markets can enhance chances of commercial enlargement (Raihan & Azeem, 2011). Thus, we can see that many large European companies have spread their operations through the entire world and are covering almost every continent (Shell, 2016). Therefore, it is quite interesting to research the internationalization policies of Europe’s large energy companies.

Figure 1 Total primary energy supply (Gazprom Marketing & Trading, 2016)

Internationalization of energy companies can happen in several ways. There are two common ways of summarizing entry mode strategies. The first way divides entry modes into direct investment and indirect investment (Raihan & Azeem, 2011). Direct investments

(7)

7

include investments such as Greenfield investment, International Joint Venture, Strategic Alliance. Under the heading of indirect investment are considered investments such as Exporting and Licensing (Raihan & Azeem, 2011).

The second way summarizes entry mode strategies into two categories; full ownership entry mode and partial ownership entry mode. Each entry mode has its own distinct characteristics, but ownership control is the main dimension. Under full ownership, entry modes are considered acquisitions and Greenfield investments. Under partial ownership, entry modes are considered Joint Venture, Mergers and Strategic Alliances. All in all, a company should pick an entry mode very carefully, by evaluating the institutional environment and culture of the host country in order to ensure the proper ownership control.

(8)

8

There are many companies that are operating all around the world, especially in the energy industry. However, contrary to the popular belief that our world is becoming increasingly integrated, like “one big village”, few companies have become truly global. Among companies that are present in Fortune 500, only a small number can be identified as global companies (Rugman & Verbeke, 2004). This suggests that, when a company is internationalizing to a foreign country which does not share same culture, it will face “Liabilities of foreigners” problems (Barkema & Drogendijk, 2007). Since no two countries are alike, a company should be able to gain understanding of the host country’s culture and choose an appropriate entry mode strategy. Moreover, home country culture may affect the internationalization policy of the company (Hobdari, Gammeltoft, Li & Meyer, 2017). Therefore, this study will account for home and host country cultural differences. This will allow us to predict which entry mode strategy is more suitable for energy companies from Europe. National culture in this research will be measured and examined using dimensions of the Hofstede.

All in all, the energy industry is large and has an enormous influence on the society in general. Many different stakeholders are influenced when an energy company is internationalizing to a different region. There are no other industries which involve so many stakeholders. Companies in the energy industry should consider many externalities, and take into account the public interest. Companies are heavily influenced by the institutional pressures of the host country. Both formal and informal institutional factors may significantly affect operationability of the energy company. Therefore, it is quite interesting to research entry mode strategies of European energy companies under institutional theory lens.

Problem Description and Analysis

(9)

9

mode strategies of European energy companies. All relevant factors which directly or indirectly contribute to the choice of entry mode strategy will be considered.

The topic of entry mode strategy has been broadly discussed in the literature. There are many studies about the entry mode choice in different sectors and countries. However, the foreign market entry decisions of European energy companies hasn’t received much attention from the researchers. There is no research that delves into the entry mode in the context of energy firms. Therefore, this paper aims to fill in this research gap.

Central Research Question

The Central Research Question of this research paper focuses on factors which influence the choice of entry mode of large European energy companies.

“What is the effect of cultural distance and institutional distance on the choice of entry mode strategy of European energy companies?”

Structure of the thesis

(10)

10

Literature Review

This section will delve into the literature that already exists about the topic. First, the entry mode strategies will be introduced. Second, the institutional theory will be introduced and institutional distance concepts will be clarified. Third, the influence of the culture will be studied. Finally, a conceptual model will be drawn.

Entry mode strategy research has gained a lot of attention among scholars from all around the world. The most relevant theories, which help to analyse the choice of entry mode are transaction cost economics (Williamson, 1979), resource based view (Barney, 1991) and the institutional theory (Peng, Wang & Jiang, 2008; North, 1991; Wright, Filatotchev, Hoskisson & Peng, 2005). Transaction cost economics (TCE) theory has been the most applied theory to study entry mode strategy. Resource based view (RBV) theory has received less attention than TCE. Institutional theory (IT) since recent times has gained a lot of attention among scholars (Canabal & White, 2008). However, IT remains the most underrepresented theory in the entry mode literature. Therefore, in this thesis I will focus primarily on institutional theory to explain entry mode choice of the large European energy companies.

Entry Mode Strategies

(11)

11

foreign market, and the subsidiary delegates to the headquarters of the company. Therefore, the company maintains full ownership and does not share operations with anybody else (Hill W.C.L., et al. 1990).

When the company decides to pursue a partial ownership entry mode strategy, such as joint venture, strategic alliance or merger, the level of control over the company will depend on equity possession. Usually there is only one dominant partner in a partial ownership mode, the one with a higher stake of equity (Lavie, 2006).

When the company does not have good knowledge and sufficient funds, it will choose an indirect entry mode strategy, such as export. This entry mode restricts the company’s operational activities, and increases dependency on foreign agents (Gulati,1995).

Thus, in introducing entry mode strategies, they will be divided into three corresponding sub-groups; full ownership, partial ownership and indirect entry mode strategies.

Entry mode strategies of international companies have gained a lot of attention during the last decade. The entry mode strategy varies according to the company type and the country to which the company is internationalizing. When a company wants to internationalize, it should choose from a wide range of entry mode strategies, such as Greenfield, Acquisition, Direct export or Strategic alliance (Brouthers & Hennart, 2007). Table 1 below summarizes entry mode strategies and differentiates them according to the dimensions of control, resource commitment, and dissemination risk.

Table 1 Characteristics of different entry modes (Hill C.W.L et al. 1990) Indirect entry mode: Exporting/Licensing

(12)

12

is mostly used to enter markets that are in the early stages of development (Khemakhem, 2010). Exporting and Licensing are ranked as low control entry modes. These entry modes provide little control over the operations and limit the company’s ability to exert control over the foreign operations. Therefore, this entry mode is not recommended for companies that possess a lot of valuable assets and are internationalizing to markets where property protection and patent protection are low (Gao, 2004). However, this entry mode strategy does not require high investments, and allows a company to establish operations in a host country relatively quickly. (Hill W.C.L., et al. 1990).

Partial ownership entry mode: Joint Venture/ Strategic Alliance/ Merger

Partial ownership entry mode can be identified as the links formed between two or more independent companies which choose to carry out a project or specific activity jointly (Dussage & Garetter, 1999). The formation of strategic alliances in varying industries, including energy, has been analyzed by many researchers (Gulati, 1995; Flatten et al 2011; Lavie, 2006; Schan & Hamilton 1991). Strategic alliances are usually made for the purpose of: economy of scale, entering a new market, overcoming competition in a market, acquisition of a new skill, circumventing foreign market barriers, gaining competitive advantage, setting new standards for the technology or risk sharing (Gumus & Apak, 2011). All the authors mentioned above have argued that companies in resource-demanding industries benefit from forming strategic alliances. It is especially applicable for companies that don’t have sufficient knowledge about the industry.

Normally in a strategic alliance, the larger company acts as a foreign partner and the smaller company acts as a local company (Inkpen & Beamish, 1997). The larger company in the partnerships is usually the dominant partner, (Hitt, Levitas, Arregle & Borza, 2000). The larger company wants to stay in a dominant position because, usually, the larger partner possesses more resources and has better industry-specific knowledge (Oxley. 1999). The smaller company usually has greater knowledge of the foreign market and experience in dealing with the institutional environment of the host market (Oxley, 1999). The smaller partner in the strategic alliance usually plays a supportive role (Lavie, 2006). Δ

(13)

13

many large energy companies would form Joint Ventures to exploit opportunities and conduct operations which require significant initial money injection. An example of this can be Total and Lukoil Benelux, which share exploitation of the Zeeland refinery, with Total as the dominant shareholder with 51% of the equity (LITASCO, 2015).

A firm that enters a new market without good knowledge of the institutional environment and culture of the host market may suffer from the “Liability of foreignness”. Therefore, in order to minimize the “Liability of foreignness” issue, it is advisable for large MNEs to choose a partial ownership entry mode strategy (Tecee, 1985, Zaheer, 1995).

Joint Ventures (JV) are short-term cooperation between two companies. JV is usually chosen as an entry mode strategy when two companies want to achieve a common goal (Choi & Beamish, 2011). JV does not require full commitment, or sharing all information about the company. Both parties agree at the beginning about the amount of investments that each partner will make. JV is driven by the partners’ complementarity of resources, according to Resource Based View (Choi & Beamish, 2011). Resources complementarity not only drives developed and emerging market firms to enter into JV, but also influences the performance of the JV. JV performance is driven by how much value or synergy JV partners create by joining their complementary resources (Choi & Beamish, 2011). However, unlike alliances such as R&D alliance or licensing agreements, contracts between IJV partners are often executed under conditions of high uncertainty (Inkpen & Beamish, 1997).

Full ownership entry mode: Greenfield / Acquisitions

(14)

14

want to share. Moreover, when the company decides to acquire a company in the host market, it reduces competition for itself (Lehner, 2009). Obtaining strategic resources or locations in the host country is crucial for energy companies. This will allow the company to gain competitive advantage (Lehner, 2009).

Institutions

Institutional environment is defined as the set of formal and informal rules of social order and cooperation governing the behavior of two or more individuals (North, 1991). Companies are embedded into the institutional setting of the operational country. Therefore, if the company decides to internationalize, it should accept and follow the institutional settings of the host country.

Thus, institutional distance can be identified as the difference between the institutional profiles of the countries (Hutzschenreuter, Kleindienst & Lange, 2015). Institutional distance plays an important role in assessment of the host market’s attractiveness. A company that is trying to set up operations in a host country should follow both the formal and informal rules of the game of the host country (North, 1991; Dimaggio & Powell, 1983). Institutional based measurement includes a whole variety of host country dimensions, which include both formal and informal dimensions. According to Transaction Cost Economics, a decision on the entry mode strategy depends a lot on the institutional forces of the host country, because it structures the environment where the transaction takes place (North, 1990)

A company that enters a foreign market should strive for legitimacy (Dimaggio & Powell, 1983). Legitimacy was described by Suchman (1995:574) as “A generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions” (Suchman, 1995:574). Hence, in order to obtain legitimacy in the foreign market, a company should fulfil the requirements of isomorphic pressures of the host country (Dimaggio & Powell, 1983). Organizational legitimacy in the foreign country can be obtained by forming an alliance or acquiring a company in the host country or by mimicking operations of the companies in the host country. If a company is able to gain legitimacy in the foreign market, its chances of success in the foreign market would increase (Roth, Kostova & Dakhli, 2011).

(15)

15

normative pressure are imposed on the company that is trying to gain legitimacy (Dimaggio & Powell, 1983). Coercive isomorphic pressure can be summarized as rules and regulations of the host country (Dimaggio & Powell, 1983), the rules which every company that operates in that institutional environment should follow. For example, when the government is imposing new pollution regulations, every company should fulfill this new requirement. Cognitive pressures can be summarized as the cognitive structures and social knowledge shared by the people in a given country. The cognitive structures affect individual behavior as they shape to a great extent the cognitive programs, i.e. schemas, frames, and inferential sets, which people use when selecting and interpreting information (Dimaggio & Powell, 1983). If the company which is entering can’t interpret information correctly, it will not be tolerated in the host market institutional environment. Normative pressures can be summarized as pressures related to social norms, beliefs, values, professionalization and different assumptions regarding human behavior and human nature that are carried by individuals and are socially shared (Dimaggio & Powell, 1983). As can be seen from the graph below (figure 2) ownership structure is correlated with institutional development. Well-developed institutions are associated with full ownership entry mode. Low-developed institutions are linked to partial ownership.

Figure 2 Institutional development of the host country and ownership commitment (Griffioen, 2011)

(16)

16

Institutional development

This study is focused on European energy companies. As can be seen from the figure above, high institutional development is linked to the full ownership FDI. High institutional development is associated with clear rules and regulations. Therefore, the chances that a company will choose acquisition or greenfield investment are higher.

Hypothesis H1a

H1a An energy company will prefer to use a direct entry mode (full ownership)

strategy when the institutional development of the host country is high.

However, developing countries mostly have a poor score in the “WGI” index. Therefore, when the energy company is choosing an entry mode strategy, it will choose a collaborative entry mode in the foreign market (Delious & Beamish, 1999). This implies that a lack of knowledge about the institutional environment of the host country reduces performance and increases the certainty of failure in the host country (Hymer, 1975). Therefore, in order to overcome so called “Liability of foreignness”, an energy company will choose a collaborative entry mode strategy.

Hypothesis H1b

H1b An energy company will prefer to use a collaborative (partial ownership) entry

mode strategy when the institutional development of the host country is Low.

Institutional distance

(17)

17

Hypothesis H1c&d

H1c An energy company will prefer to use a collaborative (partial ownership) entry

mode strategy when the institutional distance is high.

H1d An energy company will prefer to use the non-collaborative (full ownership)

entry mode strategy when the institutional distance is low.

Economic distance

The economic distance variable plays a crucial role when assessing a host country’s economic development (Ghemawat, 2001). Therefore, the economic distance variable can provide a good indication of the probability of failure or success of the FDI. This variable can be identified as the level of economic development in the host country compared to the home country (Ghemawat, 2001; Cuervo-Cazurra, 2006). Economic distance is targeted to measure differences in economic development and macro-economic development (Berry et al., 2010). Economic development influences the growth of national technological capabilities. This allows companies to tap into the knowledge of foreign companies (Gulati, 1995). Therefore, it can be expected that, in the more developed countries, companies will focus more on exploration, whereas in developing countries, companies will focus more on exploitation (March, 1991). This suggests that companies that don’t want to exploit opportunities will go to countries with a low level of development, while companies that want to explore new opportunities will choose countries with a high level of development.

Hypothesis H1e

H1e High GDP per capita of the host country will trigger a company to choose a

collaborative entry mode strategy.

Culture

(18)

18

and “Liabilities of foreignness” (Barkema & Drogendijk, 2007). Therefore, it is crucial to account for cultural differences in the study of entry mode strategies of energy companies.

The idea of culture measurement has received a lot of attention from researchers from all around the world (Shenkar, 2001). However, there is a lot of criticism of measuring culture on a country level. Despite this fact, researchers such as Hofstede (2005), Schwartz (2006), Hall (2005) and Kaufmann (2010) tried to measure national cultural profiles of countries. Culture as a measurement tool has been widely used to explore foreign investment and entry mode choice (Shenkar, 2001).

It is important to first define national culture. National culture is defined as “values, beliefs and assumptions learned in early childhood that distinguish one group of people from another” (Beck & Moore, 1985:55). One of the most influential researchers of culture defines it as “the collective programming of the mind which distinguishes members of one human group from another” (Hofstede, 1980: 25). National culture is charged by the idea of nationalism. Each country wants to shape its identity, and influence people’s norms and values. Many researchers have used culture as a study tool, in order to measure key business patterns, economic activities of organizations, groups of people, regions and nations (Tabellini, 2010).

Many authors claim that national culture characteristics have an influence on the entry mode strategy of the company (Kogut & Singh, 1988). National cultures of the host and home countries both matter, when researching the entry mode strategy of MNEs. Many researchers have confirmed that the home culture of the international company plays a crucial role when deciding on the structure of the foreign operations (Tayeb, 1998; Kogut & Singh, 1988). In general, the higher the culture distance, the more likely that MNE will prefer to have a higher control over its foreign operations. Culture distance will create many impediments in operating in the host market, will bring ambiguity into operations and possibly have a negative effect on the performance of the company (Barkema & Vermeulen, 1997). Therefore, the MNE is most likely to choose an entry mode such as greenfield or acquisition. The more control MNE possesses over its foreign operations, the higher the chances to overcome uncertainties related with entry into the new market (Alpander, 1976).

(19)

19

created transaction costs theory (Wiliamson, 1985). Authors have linked culture distance with the transaction cost theory. They claim that higher distance will be associated with higher transaction costs. Moreover, they claim that the bigger the distance, the harder it is to transfer the core competencies of the company (Vachani, 1991). In addition, other researchers have claimed that agency costs related with monitoring the foreign operations will increase with the culture distance. Therefore, when the company is internationalizing into an unfamiliar cultural block, it will incur additional monitoring costs. The bigger the culture difference, the harder it is to control the subsidiary, the harder it is to obtain information about the operational industry and the more problematic monitoring becomes (Roth & O’Donnelle, 1996). But when two countries share similar values and behaviours, it results in a much better understanding of the host market’s business activities and customers. Thus, the commonalities between national cultures of two countries will positively influence operability of the company in the host country (Lucke & Eichler, 2016).

Another study found that, when the company already has experience with the local culture of the host country, it will not face many constraints, because consecutive foreign direct investment into the same country will increase chance of success in the host country (Barkema & Drogendijk, 2007). This internationalization strategy can be linked to the Uppsala internationalization strategy, where the company internationalizes first to countries with a similar institutional environment. However physical distance does not automatically imply a high cultural/ institutional distance. Some countries located far away from each other still may share a lot of commonalities in the culture and institutional environment. This can often be linked to the colonial past (Tabellini, 2010).

(20)

20

Hofstede’s dimensions

Hofstede is well known for creating a scale of 6 dimensions to differentiate national culture diversity. They serve for distinguishing the culture differences between different countries. Hofstede’s work, “Culture’s Consequences” (1980), has been cited more than 40,000 times. This section seeks to investigate the likely impact of the culture distance along the Hofstede national culture dimensions. Hofstede created 6 dimensions of cultural diversity; Individualism/Collectivism, Uncertainty Avoidance, Power distance, Masculinity/Femininity, Indulgence and Long-Term orientation. In order to simplify this research, only the most appropriate Hofstede dimensions will be used to research the entry mode strategies of European energy companies. The dimensions of Individualism/ Collectivism and Uncertainty Avoidance will be used in this research.

The prime reason for choosing to focus only on two Hofstede dimensions is to improve the validity of this research. Not every Hofstede dimension is equally important for the study of entry mode strategy choice. Therefore, only two out of Hofstede’s six dimensions were chosen. Hofstede’s dimension “Uncertainty Avoidance” was chosen because most of the literatures on strategic alliances use this dimension as a proxy to measure the extent to which a company is willing to share its resources and knowledge with a partner (Dyer & Singh, 1998; Gulati, 1995; Hitt et al, 2000). Uncertainty avoidance can facilitate both collaborative and non-collaborative entry modes. Hofstede’s dimension “Individualism/Collectivism” was chosen because this dimension has received a lot of attention in entrepreneurship literature (Rauch, Frese, Wang & Unger, 2013). They state that collectivistic culture provides a protective environment for operations abroad. Collectivism facilitates communication with the company and communication between stakeholders. Therefore, a company will prefer to choose collaborative entry mode in high collectivism countries. This research avoids using all dimensions of the Hofstede dimension, because the results may be inflated, and would fail to identify impacts of each dimension on the choice of the entry mode.

Individualism / Collectivism

(21)

21

strong and people prefer to cooperate in groups. Moreover, people are integrated into strong cohesive groups, which act as a protective tool against outsiders (Rauch et al., 2013)

In the individualistic society, personal values are ranked higher than collective values. Therefore, commitment to the group is low and the preference to operate for one’s own self-interest is high. Moreover, in the individualistic society, a high stress is on the survival unit, people strongly prioritize personal goals and achievements and value their self-sufficiency and independence from anybody else. They ignore group interests and look only after themselves, neglecting benefits from group formation (Hofstede, 1980)

In contrast to an individualistic society, a collectivistic society stresses the need to belong to the same group, and to keep the identity of your own group and develop a tight social framework. People in a collectivistic society believe in cooperation and they expect in-group welfare and care from each other. People do not primarily look out for themselves, but for the whole well-being of the group to which they belong, sometimes even disregarding their own personal desires. The stress on the common well-being of the group is higher than stress on the well-being of individuals in the group (Hofstede, 1980).

From the company aspect, managers from individualistic countries exert more control over operations (Crossland & Hambrick. 2011). Therefore, better control of the company can be expected in Individualistic countries. Moreover, they limit in-group communication and avoid group representation in the decision-making process (Dimitratos, Petrou, Plakoyiannaki & Johnson, 2011). Most of the time in an individualist society, people pursue their own self-interest and behave how they think is better for themselves. Therefore, a company in such a culture environment should co-align the interest of the employees and the company, in order to reduce misbehaving in the company, so the employees would not think of pursuing their own self-interest (North, 1991).

(22)

22

employees, the company will have more chances to create a sustainable competitive advantage. Moreover, it will be very hard for other firms to copy their resources. Only when two parties co-align, can they create a sustainable competitive advantage. Alone, each principal cannot copy and maintain a sustainable competitive advantage, so the partners have more motivation to stay with each other and to develop new sources of competitive advantage. In a collectivistic society, the motivation to form a Joint Venture is presumably higher than in individualist countries (Dyer & Nobeoka, 2000)

Hypothesis H2

H2a A company will prefer to use a collaborative entry mode strategy when

entering a country with a high score on collectivism

H2b A company will prefer to use a non-collaborative entry mode strategy when

entering a country with a high score on individualism

Uncertainty Avoidance

(23)

23

good institutional environment. However, companies tend not to be very structured in countries where the uncertainty avoidance is low and employees are more comfortable with flexible, ad hoc structures.

Literature stresses the fact that firms that are planning to invest in a high uncertainty avoidance country would be unlikely to expand through the wholly-owned subsidiary mode. Especially in the energy industry, where the establishment of operations requires huge investment, companies that are entering a country with high uncertainty avoidance would prefer to share the risks and to establish a joint venture.

Hypothesis H3

H3a A company entering a country with a high score on uncertainty avoidance will

choose a collaborative entry mode

H3b A company entering a country with a low score on uncertainty avoidance will

choose a non-collaborative entry mode

Culture Distance

(24)

24

However, high culture distance may have a positive influence on a choice of a direct entry mode. When a company possesses valuable knowledge which it does not want to share, the company will choose a direct entry mode, even though it may suffer from “Liability of foreignness”. However, for the sake of simplicity, in this research culture distance will positively relate to the collaborative entry mode.

In this research, culture distance will be measured using only two culture dimensions of the Hofstede scale, namely uncertainty avoidance and individualism-collectivism. However, in the Appendix D, the model using 6 dimensions of the Hofstede will be presented.

Hypothesis H4

H4a High cultural distance will force a company to choose a collaborative entry

mode strategy

H4b Low cultural distance will force a company to choose a non-collaborative entry

mode strategy

Conclusion on literature review

(25)

25

Conceptual model

Figure 3 Conceptual model

(26)

26

(27)

27

Methodology

This section will introduce the research methodology of the project. The methodology employed will be outlined. In this section of the thesis the sample of this research will be introduced. Afterwards, all the variables will be introduced in more detail. A measurement section will conclude this chapter.

Introduction to Methodology

In order to conduct this research, a deductive approach was selected. Based on this approach, hypotheses were formulated from the theory and variables were identified. The deductive approach allows the researcher to select the most appropriate variables. This research will be done as a quantitative study. In order to research the entry mode strategies of European energy companies, the quantitative method of Logistic Regression will be used. In order to select proper independent variables, past studies on entry mode strategy have been analyzed. The independent variables of Culture, Institutions, Economic development and control variables of FDI experience and Firm size have been selected for this research. All these independent variables have proven to be relevant in the past research on the entry mode strategy of international companies. A step by step approach was chosen in this research to build a full model. The first model will take into account only the control variables “FDI experience” and “company size”. The second model will include all independent variables. The third model will include all control variables and independent variables.

Sample

The sample frame of this research will contain all full and partial entries of energy companies from European countries. From the dataset Zephyr, it was identified that 3,570 deals were conducted in the Energy sector in Europe from 2000 to 2016.

This study focuses on the energy industry in general. In order to ensure the reliability of this research, the industries were chosen based on their International Securities identification Number (ISIN) code.

(28)

28

are prohibited for FDI, or have restrictions on the FDI. Therefore, these industries will be filtered out.

A funnelling strategy is used to determine which deals will be included in the final dataset, and which deals will be deleted. First, only deals from year 2000 to year 2016 will be researched. This will allow to control for the year of the FDI, and to ensure comparability of all the indices that were used for control and independent variables. Second, in this research I’m seeking to study how culture influences the entry mode strategy of European energy companies. However, the Hofstede data is not available for some countries. Therefore, some deals were deleted due to unavailability of the information. Third, many insignificant partial acquisitions happened during that time frame. Under insignificant deals fall all deals where the ownership structure didn’t change more than 10 percent: deals where the ownership structure didn’t change more than 10 percent were excluded from this research. In this way, only significant deals were chosen. The 10 percent equity limit was set in line with previous research, such as Benito & Gripsrud (1992), Dikova & Wittellostuijn (2007), Padmanabhan & Cho (1999) and Larimo (2003). Moreover, only listed companies will be researched. This will increase the validity of the research dataset. Companies that are listed provide full descriptions of the ownership, structure and revenue. In addition, all relevant information about the deal and the timeframe should be known. Therefore, the deals with some unknown values will be deleted from the dataset. After these deletions, the final sample size of this research will contain 230 entries of energy companies from European countries. Therefore, the sample of this study will consist of 230 deals.

The dominant home country in this research is Great Britain: 100 entries from this country were selected for this research. However, countries with a significant influence on the European energy market are substantially represented in this research as well: Germany (14), France (14), Ireland (31), Italy (6), Netherlands (9) and Russia (15) have a significant presence in this dataset. In total, 23 home countries from Europe are presented in this research.

(29)

29

Dependent variable

Entry mode strategy is the dependent variable in this research. By scanning the literature, I found that most studies use dichotomous variables and classify entry mode strategy as either full or shared (Delios & Beamish, 1999; Meyer, 2001; Uhlenbruck, Rodriguez, Doh & Ede, 2006). However, some studies distinguish between three different types of entry modes; Greenfield, Acquisition and Joint Ventures (Kogut & Singh, 1988). This study will focus on two types of ownership; full or partial. In general, the main difference between different entry modes is the level of control. In the joint venture and strategic alliance, the control is shared, whereas. in acquisitions and greenfield investment, ownership is full. In this study, dichotomous variables will be used. 0 will represent shared ownerships (collaborative entry mode) and 1 will represent full ownership (direct entry mode).

The data about entry mode strategies will be obtained from the Database Orbis/Zephyr and from the companies’ books. The main contribution of my thesis to the scientific academic world will come from the research of the entry mode strategies of the European energy companies, more precisely on how culture and institutions influence the entry mode strategy of such companies.

Independent variables

In order to research the entry mode strategy of European energy companies, independent variables of Culture, Institutional development, Institutional distance and economic distance were selected. By choosing these variables, this research will cover all institutional pressures that may influence entry mode strategy. Each independent variable corresponds to a certain institutional isomorphic pressure (Dimaggio & Powell, 1983; Kostova, 1997). The culture variable corresponds to a cognitive isomorphic pressure. The institutional variable corresponds to a regulative isomorphic pressure. Institutional distance and economic distance correspond to regulative and normative pressures. Each variable will be identified and explained below.

Cultural dimensions and culture distance

(30)

30

research, Hofstede’s cultural dimensions of Individualism/ Collectivism and Uncertainty Avoidance will be used. Hofstede’s dataset is publicly available and can be reached via the Hofstede web-site (Hofstede, 2017). The country scores on these two dimensions will be identified and analysed. Scores vary per country from 0 to 100, identifying the extent of belonging of the country to a certain cultural type.

Culture Distance between countries will be calculated according to Kogut and Singh (1998). They have developed a measurement technique to measure the distance between countries using Hofstede’s cultural dimensions.

ID𝑗𝑘 = , { 𝐷₍₎ − 𝐷₍₊ ,

(-. /𝑉(}/2

In the second model culture distance will be measured using only two dimensions of the Hofstede. In this formula 𝐷₍₎ reflects culture distance between two studied countries (host(k) and home(j)). 𝐷₍₎is the value for the home country j, 𝐷₍₊ is the value for the host country. Variance in the equation is 𝑉(. The total is divided by 2, because 2 measures of the culture are used in this research.

ID𝑗𝑘 = 4 { 𝐷₍₎ − 𝐷₍₊ ,

(-. /𝑉(}/6

In the Appendix D fourth model, culture distance will be measured using six dimensions of the Hofstede. In this formula 𝐷₍₎ reflects culture distance between two studied countries (host(k) and home(j)). 𝐷()is the value for the home country j, 𝐷(+ is the value for the host country. Variance in the equation is 𝑉(. The total is divided by 6, because 6 measures of the culture are used in this research.

Institutions and institutional distance Institutions development:

(31)

31

used. This measure comprises six different items: voice and accountability, political stability, government effectiveness, regulatory quality, rule of law and corruption control. The score on each item ranges from 0 to100, where a higher number means a more advanced institution. This gives a score for the host country WGI. (Hypothesis 1).

Institutional distance:

The data for this variable will be gathered from the WGI index. In order to measure Institutional Distance Euclidian distance method will be used, which follows the formula below.

ID𝑗𝑘 = 4 { 𝐷₍₎ − 𝐷₍₊ ,

(-. /𝑉(}/6

In this formula 𝐷₍₎ reflects institutional distance between two studied countries (host(k) and home(j)). 𝐷()is the value for the home country j, 𝐷(+ is the value for the host country. Variance in equation is 𝑉₍. At the end, the total is divided by 6, because 6 measures of the economic freedom index are used.

Economic distance:

This research will focus on the simplified version of the economic development measurement. In order to measure economic development of the host country, GDP per capita will be taken from The World Bank database (The World Bank, 2017). This measurement of economic development is targeted to measure differences in economic development and macro-economic development of the host country (Berry et al., 2010). Researchers such as Guillen (1999); Iyer (1997), Yeung (1997); Zaheer and Zaheer (1997) have used this dimension to identify economic development of the host country. Theoretical background for this measurement was developed by Whitley (1992) and Caves (1996).

Control variables

(32)

32

Firm size

The size of the company can influence heavily the entry mode strategy of the company, especially because larger companies have a stronger bargaining power to get incentives from the host government (Brewer, 1993). Larger companies have more resources and have a better backup from the main office (Inkpen & Beamish, 1999; Slangen, 2005). Therefore, we can expect that the larger firms would prefer to acquire firms when entering a new market. However, smaller firms are better able to respond to environmental changes and don’t have bureaucratic issues so they have higher chances to succeed in the foreign market (Hitt, Ireland & Hoskisson, 2003). Larger firms would prefer to enter a foreign market through Greenfield or Acquisition, because large firms have more resources and also have more experience. Moreover, some large companies acquire companies in the foreign market in order to reduce competition (Lehner, 2009). Almost all studies on entry mode strategy have used the firm size as the control variable (Brouthers, 2002; Nakos & Brouthers, 2002). Therefore, in order to improve the validity of this research, firm size will be controlled. In this research, firm size will be evaluated based on the company’s revenue. If the company has a mother company, the revenue of the mother company will be taken.

FDI experience

(33)

33

Measurement and measure

This is a quantitative study, so a quantitative technique, Logistic Regression, will be used to measure results. The dependent variable of this research is dichotomous. The choice of using dichotomous dependent variable came from the past research on entry mode strategy (Delios & Beamish, 1999; Meyer, 2001; Uhlenbruck, Rodriguez, Doh & Ede, 2006). The independent variables are continuous or nominal in this research: both contain only whole discrete numbers. Previous FDI experience, and size of the firm will act as control variables in this research. Previous FDI experience is a dichotomous variable, size of the company is a continuous variable.

First, descriptive statistics will be conducted. This will allow us to clarify the sample, and to see any patterns. After the descriptive statistics, a check on the assumptions of the logistic regression will be conducted. Before the regression, problems such as linearity, multicollinearity and independence of error will be tackled. Afterwards, the regression analysis will be conducted, and three separate models will be built. The first model will cover only the control variables. In the second model, all independent variables will be added, except the culture distance independent variable. In the third model, a full model will be presented, including all independent and control variables.

Data analysis

Binary Logistic regression

The dependent variable of this research is dichotomous, so the logistic regression will be used in this research. Logistic regression allows the researcher to test the impact of each independent variable. It will be possible to determine whether each of the key variables is important in determining the entry mode strategy of European energy companies. Different types of measurement used for independent and control variables (continuous & dichotomous) can be used simultaneously without any constraints (Hair, Anderson, Tatham & Black, 1995). Thus, all key variables will be analysed simultaneously. After checking for all assumptions which are associated with the logistic regression, data will be tested.

(34)

34

or partial (0). Partial ownership will be used as a base mode: 0 is assigned to partial ownership and 1 is assigned to full ownership, so the full ownership will be estimated with reference to the 0 value. A positive coefficient will indicate that independent variables tend to increase the probability that a full ownership entry mode will be chosen.

𝑝 𝑦 = e9

1 + 𝑒9

The probability of the full entry mode is calculated using the algebraic formula above, where P is the chance of Y. In this equation, Y is the independent variable. Z is the combined linear model of all control and independent variables.

𝑍 = 𝛽?+ 𝛽.×𝑥.+ 𝛽,×𝑥, + ⋯ 𝛽C×𝑥C In the equation above 𝛽_? is constant.

𝛽_.…𝛽_C are the regression coefficients.

𝑥.… 𝑥C are the explanatory variables (control and independent variables)

By the default in the SPSS the cut-off threshold is set at 0,5. The Appendix C, clarifies why the cut-off level of 0.5 suits this research. It was concluded that the model’s accuracy is the highest at 0.5 cut-off level.

Evaluation of Method Assumption

When doing Logistic Regression there are four critical assumptions that should be satisfied prior to the execution. In this section these assumptions will be presented

(35)

35

Multicollinearity

Multicollinearity occurs when one independent variable strongly correlates with another independent variable (Blumberg, Cooper, Schindler, 2014). In order to avoid the problem of multicollinearity, a correlation matrix and Variance Inflation Factor (VIF) analysis were executed. In order to check for multicollinearity, a collinearity test was executed. Variables “Individualism Host” score 2.853, and “WGI Host” score 2.951. These two variables score high on Collinearity statistics. However, none of them exceeds the VIF threshold of 3: values that score higher than 3 are suspected for multicollinearity (Belsley, Kuh & Welsch. 1980). Since all scores are below the threshold of 3, it can be concluded that no issues of multicollinearity are present in this dataset.

It is required that the tolerance level in the multicollinearity test should be as high as possible (Tabachnick & Fidell, 2001). The lowest accepted level of tolerance in the research is 0.10. Tabachnick & Fidell (2001) advise that values which exceed this threshold should be investigated for multicollinearity issues. However, other researchers claim that tolerance values should exceed a higher threshold, such as 0.20 (Menard, 1995) or 0.25 (Huber & Stephens, 1993). In order to increase the validity of this research, the 0.25 threshold was chosen. All independent and control variables in this research exceed the threshold of 0.25 tolerance. Two values in Table 2 are relatively low, i.e. IndividualismHost (0.351) and WGI Host (0.339). Such values are not critical, but should be interpreted with great caution. Therefore, I can pre-conclude that no issues of multicollinearity are present in this dataset. Further additional test on multicollinearity, a correlation matrix will be conducted.

Correlation matrix

(36)

36

Table 3 Correlation table

(37)

37

Heteroscedasticity

Heteroscedasticity is not a common issue in the Logistic regression analysis. Thus, heteroscedasticity check is not crucial when conducing a binary logistic regression. However, according to Davidson & MacKinnon (1984) test for the heteroscedasticity should be accomplished in order to demonstrate that the model is adequate, and does not suffer from complete separation issue. Heteroscedasticity occurs when a correlation is biased towards one end, so the regression farthest away from the true regression line (Blumberg et al., 2014). Davidson & Mackinnon (1984) have identified two tests which can be used to test for heteroscedasticity such as LM test and pseudo-LM test. However, these tests provide only partial explanation of the heteroscedasticity issue. The most common method which is used to identify heteroscedasticity is drawing a scatterplot (Blumberg et al. 2014). Thus in order to demonstrate that this research does not suffer from the heteroscedasticity, two scatterplots are presented in the appendixes (Appendix A; Appendix B). Even though, this research does not suffer from heteroscedasticity, there is no meaningful solution to correct this problem.

Sample size

The main assumption of the logistic regression is an adequate sample size (Blumberg et al., 2014; Long, 1997). The small sample size of the research may lead to an inaccurate conclusion. In this research there are 6 independent variables. Therefore, in order to assure the significance of the model and the right conclusion, according to Peduzzi et al. (1996), the minimum sample size in my research should be at least equal to 120.

𝑛 =10 𝑥 #𝑜𝑓 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠 0.50

(38)

38

Findings and Analysis

Descriptive statistics

This chapter will introduce the findings of this research. As described in the previous chapter, this research will identify which factors influence entry mode choice of European energy companies when entering new markets.

Table 4 Descriptive statistics

As can be seen from Table 4 above, a full model contains 230 entries. In this research 81.3% of the companies chose full ownership entry mode, and only 18.7% of companies chose collaborative entry mode. Thus, in this research most of the companies chose a full ownership entry mode. This can be explained by the fact that most of the companies in this sample have a lot of resources or are backed up by a strong parent company. The average revenue of the companies in this sample is 33.56 billion dollars. Companies with high profits usually prefer not to use a collaborative entry mode, and not to share information or strategic resources with another company. The average GDP per capita of the host country is 35.329 USD, so it can be concluded that most of the host countries are developed. Therefore, we can expect that corruption will not play a significant role in this research. High GDP per capita of the host country can be linked to a good institutional development of the host country.

(39)

39

Each culture dimension is represented equally in the host countries. The average score of the individualism/collectivism dimension is 68 and Uncertainty avoidance dimension is 57. So most of the countries in this sample are individualistic and lean towards uncertainty avoidance. The culture distance on average between two studied countries is 43. Most of the countries in the sample have a high score on WGI. The average WGI score in the sample is 78.57. This means that a high proportion of the companies in this sample have well-developed institutions. A high score in WGI correlates with a high average GDP per capita. In this sample, the average GDP per capita is 35,329.1304 USD, which means that most of the countries in this sample are wealthy. The high GDP per capita can be explained by the fact that 100 entries out of 230 entries come from Great Britain, which has one of the highest GDP per capita in the world. In the sample, most of the companies have experience with FDI. This can be explained by the fact that most of the companies are large European energy companies.

Binomial Logistic Regression

(40)

40

Table 4 Binominal Logistic Regression

The results of the binomial logistic regression can be found in Table 5 above. The fit of the model is represented in the R2 statistic, or in order to convey a more suitable statistic, the Nagelkerke R2.

The first model includes only the control variables “Experience” and “Revenue Size”. These two variables were proven to be constant predictors in the past research (Barkema & Drogendijk, 2007 Brouthers, 2002; Nakos & Brouthers, 2002;). Indeed, these two control variables based on the Model 1 Table 5 explain 6.5% (Nagelkerke R2) of the variation in the dependent variable. Revenue size is positive and very significant at 0.991***, and Experience is positive but not significant at 1.491.

(41)

41

The third model includes all independent variables and control variables. The Nagelkerke R2score of the model is 0.20%, which means that 20.0% of the model’s variance is explained by the variables in this model. Logistic regression is strictly about probabilities (the probability of the full ownership or shared ownership). Thus, the third model correctly identified 83,5% of cases overall, an improvement over the 80,9% of the base model.

In this research Log Likelihood is also identified. Log Likelihood is the probability with which observed values of the dependent variable may be predicted from the independent values (Blumberg, Cooper, Schindler, 2014). The Log Likelihood ranges from 0 to 1. The lower the score of the Log Likelihood, the better independent variables are able to predict the dependent variable. The baseline model -2LL in this research is 212.213. In Table 5, it can be seen that the -2LL value of the full model 191.249 is significantly lower than the -2LL value of the baseline model. Thus, confirming, the higher explanatory power of the full model.

(42)

42

Discussion

In this chapter, discussion of the results of the study will be presented. All hypotheses will be supported or rejected, based on the information provided in the previous chapter. If the hypothesis is rejected, an explanation will be provided.

Previous studies have confirmed that Culture, Institutions and other factors influence the entry mode choice of companies. According to Zaheer (1995), cultural distance has a negative effect on a choice of collaborative entry mode, because culture distance creates so-called “Liability of foreignness”, which decreases chances of success in a foreign market. Therefore, many researchers claim that a company should choose a collaborative entry mode when entering a culturally distant country (Kogut & Singh, 1988).

In this research, only two dimensions of Hofstede were used to evaluate culture distance, because not every Hofstede dimension is expected to be responsible for the choice of the entry mode. This can be seen in Table 5 model 2, which includes only two Hofstede dimensions has a higher explanatory power than model 4 (Appendix D), which includes all dimensions of the Hofstede. Institutional distance and development have a significant influence on the choice of entry mode (Van Hoorn & Maseland, 2015). It is predicted that if the institutional distance between two countries is high, the company will prefer to use a collaborative entry mode, and if the institutional development of the host country is high, a company will prefer to use a direct entry mode strategy (Spencer & Gomez, 2011; Hall & Gingerich, 2009). Control variables, which were proven to have an influence on choice of entry mode, i.e. size of the company and experience, were included in this research (Brewer, 1993). Below each hypothesizes will be discussed in more details. At the end of the chapter, a summary table will be shown.

Hypotheses

H1a An energy company will prefer to use the direct entry mode (full ownership)

strategy when the institutional development of the host country is high.

(43)

43

0.01% and negatively correlates with the choice of direct entry mode (-0,061***). However, the hypothesis H1a states that an energy company will prefer to use the direct entry mode

when the WGI score of the host country is high. In this research, it is the opposite: with a one-unit increase of the Host country’s WGI score, the chance that a company will choose a direct entry mode decreases by 0.061 percent. Therefore, hypothesis H1a is not supported. This

Hypothesis would have been confirmed if the high institutional development of the host country would coincide with full ownership. However, in this case high institutional development of the country is coincide with collaborative entry mode.

Discussion: An energy company will prefer to share ownership in countries with high

institutional development (Oxley, 1999). This can be explained by the fact that in well-developed economies, companies usually possess more intangible assets, such as knowledge and routines (Hitt et al., 2000). Therefore, an energy company will prefer to co-align with a company in the host market. Moreover, intellectual property protection is high in countries with well-developed institutions (Oxley, 1999) and corruption is low in countries with a high score in WGI (The World Bank Group, 2017). These factors contribute to collaborative entry mode strategy.

H1b An energy company will prefer to use a collaborative (partial ownership) entry

mode strategy when the institutional development of the host country is Low.

From the regression (Table 5, Model 3), we can see that the WGI of the host country is significant under 0.01% and negatively relates to the choice of direct entry mode (-0,061***). The hypothesis indicates that a company which is entering a country with low developed institutions will prefer to use a collaborative entry mode, in order to avoid the problem of “Liability of foreignness”. However, the results from Table 5 do not confirm this hypothesis. Based on the results from the regression (Table 5, Model 3), it can be argued that with one-unit increase of the Host country WGI score, the chance that a company will choose a direct entry mode decreases. Therefore, the hypothesis is not supported.

Discussion: This hypothesis is not supported. This can be explained by the fact that

(44)

44

with anybody. Therefore, energy companies will choose a direct entry mode even though they will face additional expenses, related to “Liability of foreignness” (Zaheer, 1995).

H1c An energy company will prefer to use the collaborative (partial ownership)

entry mode strategy when the institutional distance is high.

In general, distance is positively related to uncertainty and negatively related to availability of information (Jong, Dut, Jindra & Marek, 2015; Barkema & Drogendijk, 2007). In this research, institutional distance between two countries was calculated using Euclidian distance method. WGI distance score (Table 5, Model 3) is negative and significant under 0.01% (-0,039***). Therefore, this hypothesis is fully supported.

H1d An energy company will prefer to use the non-collaborative (full ownership)

entry mode strategy when the institutional distance is low.

When countries share similar institutions, it positively influences a company’s knowledge about the host country (Van Hoorn & Maseland, 2015). In most cases, companies prefer to internationalize first to countries with a similar institutional development (Zaheer, 1995). In this research, the institutional distance between two countries was calculated, using Euclidian distance method. According to Table 5 Model 3, the WGI distance score is negative and significant under 0.01% (-0,039***). Therefore, this hypothesis is fully supported.

H1e High GDP per capita of the host country will trigger a company to choose a

collaborative entry mode strategy.

Countries with a high GDP per capita usually have well-developed institutions, which positively influences FDI inflows (Cuervo-Cazurra, 2006; Ghemawat, 2001). Based on Table 5 model 3, GDP per capita is significant at 0.05; however, the effect is very small (0,000**). Indeed, host country GDP does influence a company’s perception of the country, but does not play a decisive role in choosing a foreign market. This hypothesis is supported, but does not have much explanatory power, due to its limited effect on the model.

H2a A company will prefer to use a collaborative entry mode strategy when

entering a country with a high score on collectivism

(45)

45

2013; Hayton & Cacciotti, 2013). However, based on the Table 5 Model 3, this variable is not significant (-0,021). Indeed, from the literature review, it was argued that the company will prefer to use a collaborative entry mode when entering a country which scores high on the collectivism dimension, but, according to the results of the Full model (Table 5 model 3), the value is not significant. Therefore, this hypothesis is rejected.

H2b A company will prefer to use a non-collaborative entry mode strategy when

entering a country with a high score on individualism

According to Table 5 Model 3, the value (-0,021) is negative, but insignificant. Therefore, hypothesis H2b is rejected. In countries with an individualistic culture, managers

tend to exert more control over the operations of the company (Crossland & Hambrick. 2011). Therefore, it was expected that companies would prefer to use a direct entry mode strategy when entering countries which score high on individualism. But, from the result it can be concluded that individualistic culture is not associated with a direct entry mode strategy.

H3a A company entering a country with a high score on uncertainty avoidance will

choose a collaborative entry mode

From Table 5 Model 3, it can be concluded that uncertainty avoidance is significant under 0.10% and negative (-0,022*). Therefore, this Hypothesis is fully supported at the 0.10% significance level. Uncertainty avoidance plays a crucial role when a company is choosing between the full ownership and partial ownership entry modes (Oxley, 1999; Hitt et al, 2000; Inkpen & Beamish, 1997) Uncertainty avoidance is associated with high transaction costs (Dyer & Singh, 1998; Gulati,1995). Companies prefer to use a collaborative entry mode when uncertainty avoidance in the host country is high. With one unit increase in Uncertainty avoidance of the host country, the chance that the company will prefer to use a collaborative entry mode increases by 0.022.

H3b A company entering a country with a low score on uncertainty avoidance will

choose a non-collaborative entry mode

(46)

46

avoidance independent variable is significant under 0.10% and negative (-0,022*). Therefore, it can be concluded that, when a country scores low on uncertainty avoidance, a company will prefer to use a non-collaborative entry mode strategy.

H4a&b High culture distance will force a company to choose a collaborative entry

mode strategy

From Table 5, it can be concluded that culture distance is not significant (0,000) and does not help to explain the choice of entry mode. This can be explained, from the fact that only H3a was supported and confirmed (-0,022*), whether H2a was not supported and was not

(47)

47

Summary of Hypotheses

Table 6 below provides an overview of which hypotheses are rejected and which hypotheses are not rejected. The hypotheses about institutional distance are all significant, but only hypotheses H1c ,H1d and H1e are supported. The hypotheses about culture are not all

significant, and only two hypotheses are significant and supported, namely H3a and H3b.

Table 5 Summary of hypotheses

Entry Mode Strategy of European Energy Companies: The Effect of Distances on a Choice of Entry Mode.