Cultural distance between home and host country and the impact on entry modes

(1)

Cultural distance between

home and host country and

the impact on entry modes

By: Anej Hvala Student number: 10827978

Date: 28 June 2015

MSc Business Studies: International Management Final version Master Thesis

Supervisor: Dr. Niccolò Pisani Second Reader: Francesca Ciulli

(2)

Statement of Originality

This document is written by Student Anej Hvala who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

Abstract

This research examines the relationship between cultural distance and the likelihood of a firm to establish a majority-owned (versus minority-owned) affiliate in a specific host country. Moreover, the moderating effect of ownership structure of the parent company is assessed. By using a sample of 100 largest companies in each of the six leading economies in Western Europe, the findings could not confirm the assumption that the higher the cultural distance between home and host country, the lower the likelihood of a firm based in the home country to establish a majority-owned affiliate in the host country. However, we can argue that this increases integration costs that the company might face when establishing a majority-owned affiliate. The findings suggest that public listing and foreign listing have a positive effect on the relationship between cultural distance and the likelihood of a firm based in the home country to establish a majority-owned (versus minority-owned) affiliate in the host country.

Keywords: Cultural distance, Entry modes, Ownership structure, Publicly listed company,

(4)

List of Figures

Figure 1. Research model ... 21

List of Tables

Table 1. Descriptive statistics: means, standard deviations and correlations ... 29

(5)

Introduction

In today’s business environment, companies need to take into consideration several aspects, which might have a direct or indirect impact on the firm’s performance. Cultural distance is among the most important aspects companies need to take into consideration while expanding to foreign markets.

Several international scholars tried to explore the relationship between cultural distance and entry mode (Kogut and Singh, 1988; Hennart, 1998; Shenkar, 2001; Brouthers & Brouthers, 2000; Drogedijk and Slangen, 2006). A question about the shareholders’ reaction while the company is expanding abroad arises, thus a closer look at the ownership structure of these companies is needed. Despite being publicly listed and having foreign listing, a company does not necessarily need to be globally oriented. Even those with public and foreign listing might have a higher ownership percentage in an affiliate, while their company remains focused on the domestic market.

Because of everything stated above, it is necessary to look more deeply into the concept of cultural distance. Further, an analysis of the ownership structure´s effect on different types of entry modes is needed, as these can vary in their significance and magnitude. Also, the context in which cultural distance is assessed may increase or decrease its effect. Despite conducting a substantial amount of research, literature has paid little attention to the moderating effect of ownership structure (e.g. wholly owned subsidiary (hereinafter: WOS), joint venture (hereinafter: JV), acquisition or greenfield (e.g. Kogut and Singh, 1988; Brouthers & Brouthers, 2000)) on the relationship between cultural distance and entry modes. Therefore, the following question still needs to be answered: Does the cultural distance of a company affect its entry mode into a

(6)

foreign market? If yes, what is the moderating effect of ownership structure on this particular relationship?

Building on the Kogut and Singh (1998) theory of how cultural distance and the difference in institutional environment between countries influence the entry modes of a company in a foreign country, this study takes into consideration the ownership structure as a moderator. By doing this, it tries to shed new light on the relationship between cultural distance and the likelihood of a firm based in a home country to establish a majority-owned (versus minority-owned) affiliate in the host country. Initially, Hofstede’s (1980) cultural distance study needs to be considered. Additionally, the Zhang, Li, Hitt and Cui (2007) study on ownership structure has to be taken into account. The study focuses on 400,000 foreign-invested firms founded in China during the period 1979-2000, while investigating the number of majority/minority-owned foreign companies. Other studies tried to research the effect of cultural distance on the entry mode with the moderating effect of different national environment (Hennart, 1998), physical distance (Johanson &Vahlne, 2009) or different organizational environment (Kogut & Singh, 1988). As cultural distance arguably influences the entry mode of a company into a foreign country, so has the ownership structure gained importance in business environment as it can help the managers in their decision making process in uncertain situations. In addition to its academic relevance, this thesis, which links the concepts of cultural distance, entry mode & ownership structure may be used as a reference for managers in their decision making process related to a new market entry, as well as mapping out the further strategy of their expansion to foreign markets.

This thesis is structured as follows: First, the relevant literature is reviewed and we introduce the concepts that are of interest for the purposes of this study. We identify that the concept of cultural distance is an important deterministic attribute and after discussing important

(7)

contextual factors, we propose our hypothesis. The methodology part follows, where we describe the data collection, the variable used and the type of analysis used to test the hypothesis. The following chapter we present and discuss the findings that emerged from this analysis. Finally, we propose the academic and managerial implications of this thesis and mention the limitations and suggestions for future research.

Literature review: Home/Host cultural difference and the effect on the

entry mode

Cultural difference

Cultural difference gained major recognition when in 1980 Hofstede analyzed work-related values among 117.000 IBM employees in 40 different countries and stated that cultural difference refers to a difference in cultural background among people form home and host country. The findings suggested that countries which are culturally more distant tend to have different norms and values form the ones where the firm is located. The outcome of this is an increased organizational divergence, customer preferences and communication style. The main findings of this research presented by Hofstede are the four statistically-independent dimensions that can explain the variation between countries.

First, power distance involves the problem of how the society reacts to the information that there is inequality between people. There are two different inequalities, physical and intellectual, and certain societies let these inequalities grow over time and transform to the inequality in power and wealth. Counter to that, other societies try to minimize the inequality in power and wealth. Of course they won’t reach complete equality but at least they will be more equal than others (Hofstede, 1983). Second, uncertainty avoidance describes how society reacts to the fact that time goes only in one direction – from past to present and afterwards to the future. People

(8)

need to live with the fact that uncertainty will always be present. On the one hand, certain societies push their members to live with the uncertainty, making their people appreciate each day as it is. Contrary to that, some societies push their members to find out what is going to happen. Due to the fact that the future is unpredictable, these societies will tend to be more nervous and aggressive (Hofstede, 1980). Third, individualism describes the relationship between an individual and its colleagues. On one side we find a society that supports individuals to have loose ties with other colleagues and therefore encourages them to just look after their own interest. On the other side there is a society supporting individuals to have tight ties with other colleagues and therefore to look at the interests of the group as a whole. In return the group will protect them when and if they find themselves in trouble (Hofstede, 1983). Finally, masculinity and the opposing femininity describe the acceptance of gender roles in society. Societies tend to augment or reduce the social division by gender. Therefore, those that maximize gender division are referred to as masculine, where values such as competitiveness, achievement and ambition are important. Other societies which minimize or lack gender division are referred to as feminine, where values such as helping others, putting relationships with people before money, not showing off, etc. dominate (Hofstede, 1980). The study conducted by Van Oudenhoven (2001) recommends that these dimensions can be used to assess countries by their national culture and determine the cultural distance between them.

Hofstede’s work had a particular impact on the later assessment of cultural distance. Kogut and Singh (1988) developed a formula how to calculate cultural distance. It assessed the deviation between the score of home country and host country, for each of the four cultural dimensions developed by Hofstede (1980). They developed this formula while trying to answer their hypothesis on how cultural distance affects the entry mode.

(9)

Despite the fact that Hofstede’s work helped to increase the understanding of differences between national cultures, Schwartz (1994) and other researchers pointed out a few concerns. The first concern emphasized that not all cultural dimensions are exhaustive. This is due to that fact that the survey designed by Hofstede was not meant to identify national cultural dimensions and therefore it might not be structured properly. The second concern addresses the sample. Schwartz (1994) argued that the sample of countries used by Hofstede does not reveal the full spectrum of national cultures; therefore the results can be different if another country is added. As a third concern, IBM employees assessed by Hofstede are clarified as not representative for all the population of a certain country. Schwartz (1994) explained that misrepresentations were different for each country. The forth concern was about the time when the research was made; Hofstede collected these data from 1967 until 1973 and since then many important discoveries happened (Ohmae, 1990). The last concern questioned if people from different cultures perceive others in the same way as they are perceived (Drogendijk, Slangen, 2006). After pointing out the limitations of Hofstede (1980), Schwartz (1994) assembled his own value survey. With this survey he was able to recognize significant values varying between cultures of different countries. From those, Schwartz (1994) tagged seven dimensions, respectively conservatism, intellectual autonomy, affective autonomy, hierarchy, egalitarian commitment, mastery, and harmony. Some researchers (Steenkamp, 2001) state that Schwartz’s dimensions are well-funded, but that they should be tested through empirical applications.

Apart from Schwartz, Shenkar (2001) also criticized Hofstede’s work. He based his critique on hidden assumptions that appear in two groups: one emerging from the conceptual properties which are the illusion of symmetry, the illusion of stability, the illusion of linearity, the illusion of causality, and the illusion of discordance, respectively, and the other emerging from its

(10)

methodological properties, the assumption of corporate homogeneity and the assumption of spatial homogeneity.

Hennart (1998) proposed two different reasons why the national origin of a MNE might have an impact on the ownership strategies of a subsidiary. Firstly, he states that national cultures of a country might have an impact on the strategies conducted by a MNE. This scenario can be explained when high power distance and low uncertainty are the two main cultural characteristics dominating in a specific country. Firms then might opt for an entry mode with a higher commitment (Hennart, 1998). Secondly, he focuses on the differences between national characteristics. For this second view, cultural distance is referred to as a difference between the characteristics of a national culture of the home and the host country. In this example MNEs cannot fully manage their subsidiaries, hence they should engage in a joint venture in order to enable them to regain access and make use of the help of a local partner for negotiations with other stakeholders (Hennart 1998). Therefore, it is deductible that if there is a big difference in national cultures between home and host country, MNEs will more likely tend to establish a partnership (JV) in that country (Hennart, 1998).

Shenkar (2001) and other researchers noticed that loosening of control in locations that are culturally more distant leads to reducing uncertainty and information costs. Goodnow and Hansz (1972, p. 46) put it as "degree of control declines as the environment becomes less favorable".

Entry modes

Firms expanding abroad need to be clear about the country of their planned before choosing the suitable entry mode for that market. The Uppsala model explains the characteristics of the internationalization process of the firm and suggests that firms usually start to internationalize in foreign markets that are close to the domestic market in terms of psychic distance and later

(11)

gradually enter markets that are further away from those already present in psychic distance terms (Johanson &Vahlne, 2009).

Companies undergoing international expansion get access to new market opportunities where they can sell their products. As companies increase their presence in foreign countries, they are able to leverage their skills and products in a broader range of markets, therefore getting a possibility to increase growth and profits (Zahra, Ireland and Hitt, 2000).

While expanding into new foreign markets, companies have to decide which ownership stake approach in an affiliate will be the most suitable to participate in the new market. Usually companies decide between WOS and JV. If a parent company chooses to enter a new foreign market through a WOS, they will retain control of this affiliate, but will also carry the costs and risks of full ownership. When a parent company does not want to carry all costs and risks alone, or when it cannot get the majority control, it decides to own a minority stake (Anderson and Gatignon, 1986). On the other side, a JV is a business agreement between two or more parties, where they pool their resources for the purpose of establishing a JV and set up a business project. The participants of a JV are not just entitled to its profits, but are also responsible for bearing the costs associated with it (Anderson and Gatignon, 1986).

As indicated above, companies can reduce their external uncertainty linked to WOS in foreign countries by establishing a JV with a local partner. The rationale behind this is that local partners possess the knowledge about the host country´s norms and values (Slangen and van Tulder, 2009). Because both partners share profits in a JV, local partners are willing to make use of their market-specific knowledge. On the other hand, since the parent firm does not transfer any profit rights to the local firm in the case of acquisition, the local firm’s willingness to use its

(12)

knowledge for the financial benefit of its parent firm is reduced (Hennart, 1988). If a company engages in a JV, it requires less equity commitment as in a WOS. Managers are usually unwilling to engage such a vast amount of resources in a culturally distant country; therefore a JV might present an advantage. Moreover, because of fewer resources engaged in a JV, the exit costs are lower as well. This increases the flexibility of a JV, which usually plays as an advantage in attracting foreign companies. Besides external uncertainty, companies also face internal uncertainty while trying to enter a foreign market. This happens mostly because JV partners come from dissimilar countries and they have different organizational cultures (Kogut & Singh, 1988).

Since cultural distance affects the entry mode of MNEs, companies willing to expand abroad should enter a new foreign market through greenfield or acquisition (Brouthers & Brouthers, 2000; Kogut & Singh, 1988). The larger the cultural differences between the countries, the larger managerial and organizational differences are among the two. (Kogut & Singh, 1988). Companies might find it more difficult to integrate their corporate culture in countries that are culturally more distant. Therefore, companies should do acquisitions in countries that are more similar to the home one. If the company was to be declared culture-specific, bad employees’ attitude and poor post-acquisition performance might be the consequence (Drogedijk and Slangen, 2006). Contrary, companies might find it easier to integrate with a greenfield investment in a culturally more distant country. With this move, companies can introduce their managerial practices without facing already existing ones and are able to hire people who fit the most into the company culture (Kogut & Singh, 1988; Drogedijk and Slangen, 2006).

Companies usually expand abroad in a so-called staged-internationalization (Fan and Phan, 2007). There are two perspectives affecting this decision: one being the learning capacity of the

(13)

organization, and the other being the uniqueness of international new ventures. On one side, the experience of the management might have an impact in the early internationalization decisions, therefore reducing uncertainty (Fan and Phan, 2007). On the other side, customers and suppliers from a niche market understand each other better than the ones in a mass-market.

Ownership

Claessen, Dajnkov and Lang (2000), divide corporations in two different categories, one having the widely spread ownership and the other being characterized by controlling owners (usually family companies). A corporation with a wide spread ownership has more equal owners and no one has a significant voting rights (majority). Furthermore, they are divided into four additional divisions, families, the state, widely held financial institutions (banks etc.) and widely held corporations, respectively. Claessen, Dajnkov and Lang (2000) also explained that firms can have more than one strategic owner.

Jensen and Meckling (1976) were among the first researchers that started to discuss the ownership structure and its influence on company’s willingness to expand abroad. They suggested that both, the cash flow and the voting rights are important while assessing the ownership structure. Jensen and Meckling (1976) explained that the separation between ownership and control leads to potential principal-agency conflicts and potential opportunistic behavior. Additionally to Jensen and Meckling, also Jaskiewiczn & Klein (2007) pointed out concerns regarding potential consumption of private funds, at the shareholders expenses. On the other side Jaskiewicz & Klein (2007) expressed their concerns about the potential extraction of private benefits by managers, at the expenses of shareholders. With this said, the funds destined for R&D may be spend in a different way. According to Tsao, Lin and Chen (2015), R&D represents a long-term strategic orientation for a firm and is therefore crucial for firm’s survival

(14)

and future growth. The R&D carries besides its benefits also additional risk because R&D investments do not result in any payoff and they might turn into profit just after many years. Baber, Fairfield, & Haggard (1991) also pointed out that if CEOs bonuses are tied to company’s earnings, they will try to improve the short-term accounting and stock performance, therefore cutting R&D expenses.

Usually large firms are managed in a more dispersed way, whilst small as well as non-financial companies are controlled by a single person or an entire family of owners (Faccio and Lang, 2002). Additionally, state-controlled companies – usually large companies from a country’s strategically important industries – exist. Certain strategies, such as dual class shares or pyramids, reduce the control of larger shareholders. Yet, there are still significant differences among countries when it comes to the relationship between ownership and control (Faccio and Lang, 2002).

It is common that a specific company’s largest shareholders have additional controlling interests in other companies from different countries. La Porta et al. (1999) examined the ownership structures of 30 firms in 27 different countries. The aim of this research was to examine the global ultimate owner (hereinafter: GUO) and the process of achieving such a significant ownership percentage. La Porta et al. (1999) found that companies located in Western Europe are either widely-held or family-owned. Widely-held, often large and/or financial companies exist mostly in the UK and Ireland, while family-controlled companies, usually small and non-financial, are predominant in continental Europe.

There are different techniques described by La Porta et al. (1999) of how larger shareholders get more voting rights thanks to multiple classes of shares, pyramidal structures,

(15)

holdings in exchange for dividends rights (also called cash-flow rights). Pyramiding occurs when an owner acquires majority shareholding of a firm and then controls another company with the combined shareholding, with the goal to eventually build a large holding company.

Firms are steered via multiple control chains (Faccio and Lang, 2002). Cross-holding happens, when a specific company holds directly or indirectly its own shares (La Porta, 1999). La Porta (1999) reported that in at least 66% of family-owned companies family members are part of the management team. On a similar note, Faccio and Lang (2002) found that families control most of the family-owned companies and that each family is in charge of a few different companies.

Faccio and Lang (2002) described techniques how larger shareholders gain control (voting) rights in exchange to dividend (cash-flow) rights. Fan and Wong (2002) reported in their research that in East Asia ownership is highly concentrated in the hands of a few large owners. Additionally they found that because of the complicated ownership structures, these owners possess more control over important companies than their ownership percentage would indicate.

A firm’s performance and its strategy can be significantly affected by the size and type of the shareholder (Thomsen and Pedersen, 2000). Different groups of shareholders might not have the same view of the firm’s development and investment strategy, since they may be facing increased risks and organizational complexity (Andriosopoulos and Yang, 2015). Haskisson et al. (2002) explained that different types of institutional investors have different preferences for corporate development and in this way affect the type of corporate governance. On the other hand, institutional investors provide effective external management monitoring (Chen et al.,

(16)

2007). Despite such monitoring being costly, large shareholders can sustain these costs in return to gain greater returns (dividends) (Gillan and Starks, 2000).

Another important aspect of ownership structure is foreign listing or cross-listing. In these circumstances, the size of the company plays an important role as big companies can carry the significant costs associated with cross-listing. Usually cross-listing is considered as an option by companies that need to sell large number of their shares (Pagano, Röell and Zechner, 2002). Large European companies in little need of financial investment are usually suitable candidates to have their stocks traded in a stock market, while the contrary is true for US-American companies. This is due to the fact that US-American stock exchange market is usually more liquid than the European one. Additionally to their more liquid market, the United States also have a bigger product market, which in the last decade has grown at a faster pace than the European one. In conclusion, this leads to cheaper cost of capital, which might be extremely important to companies with a need for large amounts of fresh capital (Pagano, Röell and Zechner, 2002).

Companies engaging in mergers and acquisitions (M&As) show that the shareholders’ value in such a company is significantly affected (Gaspar et al. 2005). Because of cultural distance and asymmetry of information, cross-border transactions require more skilled and experienced companies to engage in cross-border M&As (Dikova and Rao Sahib, 2013).

Dunning (1988) argues that in a country with domestic and foreign companies, the foreign ones possess advantages over the domestic ones. Despite that, FDI can still have an impact on the output of domestic companies. Mayer and Sinani (2009) reported that domestic firms in a host country still report positive spillovers caused by foreign companies. On the other side, as Agahion et al. (2009) argue, this might affect the output of domestic companies with negatively

(17)

related productivity spillovers. Larimo (2003) expressed that entering a host country via greenfield or acquisition, no matter if the ownership is partial or full, it still influences the final decision of the appropriate entry mode.

To sum it up, although numerous studies already dealt with the relationship between cultural distance and entry modes, the effect of ownership structure still calls for further investigation. Several studies showed how different entry modes, acquisitions, greenfields and JVs all have a different effect on the performance of a firm. Kogut & Singh (1988) focused in their research on how cultural distance between home and host country affects the entry mode. Moreover, Brouthers & Brouthers (2000) examined how different entry modes, WOSs or JVs, affect the parent firm’s performance. On the other side, studies such as the one conducted by Claessen, Dajnkov and Lang (2000), examine how ownership structure affects firm performance or the effect of the ownership structure on R&D expenditures (Baber, Fairfield, & Haggard, 1991). Nonetheless, the effect of ownership structure on the relationship between cultural distance and entry mode is yet unknown. This relationship can be influenced by several factors providing positive or negative effects. Therefore, this study will focus on how dispersed ownership can affect the entry mode as well as the effect of foreign listing on the entry mode.

(18)

Theoretical Framework

Cultural distance affects the entry mode of the parent company while establishing an affiliate in a host country

The cultural distance between home and host country has an important influence on the entry mode of the new affiliate. To calculate the cultural distance between two countries, Hofstede (1980) developed four main dimensions that allow assessing cultural distance. The higher the cultural distance between two countries, the more likely the companies will opt for higher control of the affiliate in a foreign country (Meyer, Estrin, Bhaumik, & Peng, 2009). In 1977 the two Swedish researchers Johnson & Vahlne developed the Uppsala model of Internationalization. It states that companies, which are expanding over national boarders, first try to succeed in neighbouring countries and then move further away. The rationale behind this is that the markets in neighboring countries are similar to the domestic one; therefore the companies should not be facing big challenges to succeed in them.

On the other side, Benito & Gripsurd (1992) explained in their research that if previous investments of a company were made in more distant countries, there is a trend that the next investment will be undertaken in a less distant country and vice versa. The results support the notion of location choice being a discrete rational choice and not a cultural learning process. The nature of business decides the possible location of the new affiliate, thus occasionally affiliates are being established in more distant locations, and at other times in countries closer to home (Benito &n Gripsurd, 1992).

While establishing new affiliates in a host country, companies also take into consideration the effect of the competitors in that location. While evaluating if undertaking an agglomeration would be a suitable choice, companies usually assess the net outward and inward knowledge (Mariotti et al., 2010). They are more willing to engage in agglomeration with another MNE

(19)

because they expect that the knowledge obtained from this interaction will be more balanced than the one obtained from a local foreign firm (Mariotti et al., 2010).

Companies prefer to enter the foreign market and establish their operations in an institutional environment that is similar to the home one (Frynas, Mellahi & Pigman, 2006). Rugman & Verbeke (2004) explained that companies prefer to locate their operations in their home region or triad regions (NAFTA, EU and Asia-Pacific), where they report most of their sales. Therefore, companies who were believed to be global are in reality regional, with their strategies being focused on their regional/local customers rather than on the global ones (Rugman & Verbeke, 2004).

Each level of equity participation in a newly formed affiliate corresponds to a different entry mode. Besides the important fact of the company already having an affiliate established in a host country, the level of this participation also explains how the environment in the host country supports foreign investments. If a country is culturally closer to the home one, the firm might opt for an acquisition, as the costs of such integration are lower. On the other hand, if the host country environment is different from the home one, then companies might opt for a greenfield because they find it easier to implement their philosophy through a greenfield than an acquisition (Meyer, Estrin, Bhaumik, & Peng, 2009). The size of the company undergoing the expansion is also relevant. Smaller companies may opt for the low commitment (partial acquisition or JV) (Ando, 2012). Despite facing cultural obstacles in understanding the foreign company, the host partner in a JV would have the advantage of understanding the local environment and its effects on the performance of the JV (Brouthers and Hennart, 2007). However, firms also face uncertainty when entering into institutionally advanced countries because institutions between the

(20)

home and the host country might not be similar and this could create additional adaptation costs for a foreign company (Ando, 2012).

Foreign markets might be predominated by informal rules and social contracts, which will prompt the MNC to establish a relationship with a local partner for better understanding of the foreign market (Zhang et al, 2007). Majority ownership in a newly formed foreign affiliate helps the company to mitigate future problems with the local partner. The first reason for this is that the company will provide a more advanced technology if it has a majority ownership; the second reason is that a majority ownership can act as a control mechanism in a partnership with the company overseeing the usage of the investment funds by the local partner. Therefore, equity stakes can provide the mechanism to enforce the distribution of returns in cases when the contractual agreements cannot be written. Furthermore, the equity ownership provides the company and its local partner with the rights over the JV assets (Zhang et al, 2007).

Hence, companies coming from countries with a higher cultural distance to the host country will be more willing to establish an affiliate with a majority ownership. This is influenced by the differences in business environment, which lead the company to enforce their mentality and not try to undertake the local one. Furthermore, the difficulty to trust a local partner in such an environment leads the company to consider an expansion into specific foreign market only in the case of a majority ownership. Consequently, this will affect the ownership structure of an affiliate, leading to the first hypothesis:

Hypothesis 1: Cultural distance between home and host country is positively related to the likelihood of a firm based in the home country to establish a majority-owned (versus minority-owned) affiliate in the host country.

(21)

International ownership structure

Besides entry modes, the level of internationality and cultural diversity in ownership structure might also have the influence over the choice of affiliates’ location. According to Claessen, Dajnkov and Lang (2000), companies have two types of ownership structure: family owned, usually non-financial companies and SMEs, or companies with a high variety in the ownership structure – usually associated with public companies and large MNCs. Large companies face a principal-agency relationship that might lead to a dispute between management and shareholders. This can be avoided by giving the management a small portion of the ownership and performance-related incentives (Faccio and Lang, 2002).

It is common for large companies to have more than one shareholder. Usually there is a dominant one among all shareholders who has a bigger portion of shares than others. In order to achieve this, companies might opt for a dual share system where bigger shareholders exchange their cash flow rights with a smaller shareholder for voting rights (La Porta et al., 1999). The amount of shareholders a company has depends on the size of the company, the number of outstanding shares and whether it is publicly traded or private (La Porta et al., 1999). Bebchuk (1999) suggested that in countries where large shareholders prevail, private benefits associated to them might be larger. With the separation of cash flow rights and voting rights, shareholders can maintain control even in a less efficient business environment.

Dwivedi and Jain (2005) found that higher performance is positively related to the presence of foreign shareholders. Additionally it was discovered that lower performance was occurring in companies with a more fragmented ownership structure. Performance declines when the number of shareholders increases, leading to a greater disconnection between ownership and control (Holm and Schøler, 2010). Companies with closely held ownership do not need to rely on

(22)

external publications in the same amount as companies with widely spread ownership. Companies with widely held ownership have to balance the lack of owner presence despite the high level of disclosure (Holm and Schøler, 2010).

Stock market liquidity also plays an important role in the ownership dispersion. Less liquid markets might capture larger shareholders and limit their possibility of selling their stake. This makes all the shareholders strategically invest, while looking at a long-term return and thus the number of shareholders remains limited. On the other side, in more liquid markets, shareholders can sell their ownership stake quite easily, leading to a widely spread ownership structure in more liquid markets (Maug, 1998).

Ownership structure also affects the internationalization strategy of a company. A more dispersed (widely-held) ownership structure can accelerate the pace of internationalization of a company. Especially family firms tend to be conservative and risk-averse and might find it difficult to obtain financial resources needed for the internationalization. Therefore, large companies and investors have minority ownership in other companies with the intent to help them expand at a faster pace as they were supposed to. The large companies and investors contribute to building the competitive advantage and acquiring resources (Fernández and Nieto, 2006).

Companies with a more dispersed ownership are able to gain substantial competitive advantage by switching from a domestic player with an outdated product portfolio to a global player with an updated portfolio (Rui and Yip, 2008).

Foreign investors are not inclined to invest in companies with lower past returns. The main difference between foreign and domestic investors is that foreign investors take more into

(23)

consideration the higher current rations, while domestic investors tend to invest more in companies with a lower turnover. Moreover, domestic firms tend to avoid the most risky firms, contrary to foreign investors (Dahlquist and Robertsson, 2001).

Investors’ protection plays an important role in attracting foreign investors. Franks, Mayer, and Miyajima (2008) acknowledged a high level of ownership dispersion between Japanese companies during the first half of the 20th century. A large amount of publicly listed companies with a large amount of shareholders were present (Franks, Mayer and Rossi, 2009).

Hypothesis 2: Ceteris paribus, geographically dispersed-ownership positively moderates the relationship hypothesized in H1

Foreign listing

Foreign investors prefer to invest into companies with more liquid stocks, companies with a low concentration of ownership and in companies with foreign listing (Dahlquist and Robertsson, 2001). Foreign investors tend to invest a high share of their funds into large companies, preferably with a foreign listing. Companies with large cash positions in their balance sheet and an intent to pay low dividends are also preferred. Market liquidity plays an important factor in assessing the chance for a foreign investor to invest. Moreover, foreign ownership is related to the current firms' presence in a global market (export or foreign listing). Usually firms with widely-spread ownership structure tend to attract foreign investors more than the ones held by a family (Dahlquist and Robertsson, 2001)

Generally, domestic investors are better informed than foreign investors, which allow them to have a significant advantage in stock trading (Andriosopoulos and Yang, 2015). On the other hand, foreign investors perform significantly better compared to domestic investors when

(24)

they trade shares of large firms in emerging markets (Seasholes, 2004). If both domestic and foreign investors have access to the same information, characteristics like investor sophistication and better processing of information give foreign investors the ability to gain higher returns (Chen et al., 2009).

International foreign listing allows companies to benefit from a lower cost of capital by making their shares available to foreign (non-resident) investors. Otherwise, they would find it more difficult to hold these shares due to occurring barriers in international investment. With such action companies achieve a lower cost of capital thanks to risk sharing with other foreign investors (Karolyi and Stulz, 2002).

As the company’s shareholders base increases with a foreign listing, greater risk sharing reduces the risk premium investors require to hold the shares of the company. Information asymmetries between home and foreign market create equilibrium among investors, signaling that a company that lists in foreign markets with high disclosure standards is of a high value. Moreover, the protection of a company’s investors increases with foreign listing as a company with foreign listing is subject to greater investigation and monitoring by the press and authorities (Doidge et al., 2004).

Bonding and monitoring benefits exist because large shareholders and their actions are controlled by authorities. By listing in a foreign country large shareholders’ costs of expropriation increase, leading to attract more minor foreign investors who provide fresh capital. This will lead to more foreign listing if companies have valuable growth opportunities (Doidge et al., 2004).

(25)

Hypothesis 3: Ceteris paribus, foreign listing of a company positively moderates the relationship hypothesized in H1

Figure 1: Research model

Methodology

The study uses a cross-sectional research design to analyze the effect of cultural distance on the entry mode of a firm based on two moderators, which are public listing and foreign listing. This database allows us to evaluate if an internationally diverse ownership structure has an influence over the entry mode of an affiliate in a company.

The sample collected for the purpose of this research is based on the 100 largest companies for each of the six leading economies in Western Europe (France, Germany, Italy, Spain, The Netherlands and United Kingdom). In general, almost all of each country’s 100 largest companies are multinational companies, meaning that they produce and/or sell products and/or services

Cultural distance Location choice of _{the affiliate}

Ownership structure

1. International ownership structure (H2)

2. Culturally diverse ownership structure (H3)

H2, H3 H1

(26)

beyond national borders (Rugman & Verbeke, 2004). Moreover, the companies included in the sample represent different industries. This sample gives us a clear insight into the financial position of a company, ownership structure and the entry mode of already established affiliates. The companies in this sample hail from six most developed countries of Western Europe. The database collected contains 114.290 companies from six the most developed countries of Western Europe. However, due to missing values in the database, the sample size used in this research is 15.260.

The data used in this research is secondary data. All data used, such as annual reports of companies and ORBIS database data, has been collected in 2013. The Bureau van Dijk’s ORBIS Database is used because it contains comprehensive information on companies worldwide. It can be used to research public and private companies. In it, it is possible to find information regarding the company’s financials, detailed corporate structure, industry research, etc. Moreover, for additional information needed, internet sources such as Capital IQ (capilatiq.com), Company Check (companycheck.co.uk) and Bloomberg (Bloomberg.com) were used.

Variables

Dependent variable

In this research the dependent variable is the ownership structure of an affiliate. This is due to the fact that we distinguish between a majority and minority equity share in a specific affiliate. The variable is a dummy variable because it only explains if a parent company has a majority or minority share. While expanding abroad, companies could enter a foreign market with a majority stake or a minority stake. Control rights associated with the ownership of a company can as well be used as a dependent variable (La Ports et al., 1999). Other researchers such as Ando (2011), Yamin and Golesorkhi (2011) used the ownership structure of an affiliate as a dependent

(27)

variable. The classification for the dummy dependent variable (Y) assumes that values of: Y=1, if the equity share in an affiliate is above 50%, Y=0, if the equity share in an affiliate is below 50%.

Independent variable

Cultural distance expresses the different aspects of distance between home and host country (Hofstede, 1980). This study assumes that cultural distance between home and host country, calculated on the Kogut and Singh formula, is the independent variable that correlates with the entry mode for the affiliates. Cultural distance was calculated based on the Kogut and Singh (1988) formula, which was calculated based on Hofstede’s findings. The cultural distance is calculated between two countries and expresses if a country is culturally more close or distant to the home one. The formula is structured as follows:

Cultural distance (CD_𝑗) = ∑ {(𝐼_𝑖𝑗− 𝐼_{𝑖6𝐸𝑈})2/𝑉_𝑖} /4 4

𝑖=1

Where CDj stands for cultural distance between home and host country, Iij is the country j's score on the i-th cultural dimension, Ii6EU is the score for the six countries form Western Europe (each separately calculated) and Vi is the variance of the score of the dimension. Similar studies such as Brouthers and Brouthers (2001), Drogendijka and Slangen (2006), used cultural distance as an independent variable as well but both with a slightly different formula how to calculate it. Therefore, for the purposes of this thesis, we can conduct that the higher the score of CD j, the higher is the cultural distance between home and host country.

Moderating variables

While further investigating the relationship between cultural distance and entry mode of a company into a foreign country, we identified two moderating variables that influence the relationship in a different way.

(28)

The first moderating variable is public listing of a company and it explains if a company is publicly listed or not. The thesis focuses on a distinction between a more closely held or more widely held ownership structure. The study of Drogendijka and Slangen (2006) reports how cultural distance has an effect on a specific entry mode, mostly focused just on greenfield and acquisition. Moreover, public firms tend to have their affiliates spread more over the world than the non-public ones (Anwar and Sun, 2015; Ando, 2015). The data collected for the purpose of this thesis also includes the ownership percentage in affiliates. The companies can have a majority or a minority stake in their affiliates. Publicly listed companies, despite having a wider network of affiliates, do not necessarily need to have a majority ownership in those affiliates. This shows that a company, which is more inclined to internationalization can still not benefit so much from an affiliate that is not in its full ownership.

Another moderator, which plays an important role in moderating the role between cultural distance and entry mode is the foreign listing of a company. As Ando (2011) explained, companies with foreign shareholders are more inclined to internationalize, therefore expanding abroad faster than companies with no foreign shareholders. Each company needs to know the market where they are expanding to; foreign shareholders might offer their knowledge to help a company co-operate with local foreign institutions.

Both moderating variables are dummy variables that equal 1 if a corresponding company is respectively publicly listed and has a foreign listing. If the company does not correspond with the criteria, then the dummy variable equals 0.

Control variables

At the firm level, the four controlling factors are selected. First is the firm size, which is commonly used and is linked to the outcomes generated by the company. Usually publicly listed

(29)

companies are larger than non-listed companies (Anwar and Sun, 2015). Larger companies can typically easily bear the costs of larger investments than smaller firms (Drogedijk and Slangen, 2006). Gross annual turnover in millions of US dollar is used to measure the size of the company. This study tries to examine the data without any external influences. Therefore, the logarithm was used for the firm size variable in order to prevent such influences.

The second factor is age of the company, which is measured from the incorporation until the year 2013. More mature firms have more knowledge about foreign markets since they are more experienced (Gaba et al., 2002) and as they have been established for a longer period, they might already have diversified, become publicly traded and increased their financial position, which made them more prone to further future expansion (Anwar and Sun, 2015).

The third factor is related to firm’s performance and can as well impact the firms internationalization process. Companies which are publicly traded usually perform better in a short-term as they need to take care of the investors’ satisfaction and focus on the return they achieve (Houmes and Chira, 2015). This leads publicly-traded companies to increase their financial position and increase the company’s possibility for future expansion into foreign markets. Firms’ performance is operationalized in terms of their gearing in the fiscal year 2013. Gearing is a financial instrument which measures the degree of which firms’ operations are funded by shareholders funds compared to the creditors’ ones. Therefore it is deductible: the higher the ratio, the higher the debt of a company.

The fourth and last control variable is the industry in which companies operate. Companies forming distinct types of industries might have different strategies for international expansion. The industry effects were seized into a categorical variable which escalades from 1 to 5 and was

(30)

later coded into 4 dummy variables. By collecting all the industries from the dataset gathered for the purpose of this thesis, 5 industrial categories were created. The SIC codes from the database were used to clarify the pertaining industry segments for companies. The categories are: 1-Mining and Construction (Division B and C), 2–Manufacturing (Division D), 3-Transportation, Communications, Electric, Gas, and Sanitary services and Services & Public Administration (Division E, I and J), 4-Wholesale traded and Retail Trade (Division F and G), 5-Finance, Insurance, and Real Estate (Division H).

Statistical analysis and results

As we aim to test the dependency between cultural distance as an independent variable and the likelihood of a firm based in a home country to establish a majority-owned (versus minority-owned) affiliate in the host country as a dependent variable with a discrete outcome, the logistic regression technique was applied for the purpose of this research. The logistic regression technique allows us to predict the membership of different groups. Therefore, the first group membership means that the parent company has a minority share in the ownership structure of an affiliate (below 50% of the total share). The second group membership means that the parent company has a majority share in the ownership structure of an affiliate (above 50% of the total share). For the prediction of this relationship, 6 controlling variables were used, namely: firm size (in terms of Gross Annual Turnover in mill. of US dollars), maturity of the company in 2013, firm’s performance (in terms of gearing in %) and industry in which the company operates (categorical variables from 1 to 5). The 2 moderating variables, public listing and foreign listing of the company (dummy variables), were hierarchically added for testing the moderating effect on the relationship stated in hypothesis 1.

(31)

The test for multi-collinearity resulted in all VIF values being under the value 2. All bivariate correlation results were below the value 0.7, hence all the values were retained. The descriptive statistics and correlation are presented in Table 1.

For testing the impact of the second moderator variables on the relationship between the cultural distance and the likelihood of a firm based in a home country to establish a majority-owned (versus minority-majority-owned) affiliate in the host country, the logistic regression analysis was run in more steps. First, only control variables were inserted into the model, in order to observe their effect on the dependent variable. Afterwards the independent variable and the interaction terms of independent variable with moderating variables were added for observing their additional explanatory power. Previously, the interaction terms were computed by multiplying the independent variable with each moderating variable. The moderating variables were standardized (Z-score standardization) to facilitate the interpretation of coefficients. The results from the logistic regression are presented in Table 2. The Model 1 results represent the ability of the control variables to explain the ownership structure of an affiliate. The Model 2 tests the Hypothesis 1. The Model 3 and Model 4 test the Hypothesis 2 and 3, respectively. The significance level, beta standardized coefficients and goodness of the model’s fit are presented. Hence, the Hosmer and Lemeshow test has been used as the measure of the model’s fit. Additionally, as a measure of logistics model fit, the Negelkerke pseudo R-square has been used.

The results of the analysis indicate that not all of the factors included have the further explanatory power, due to the fact that their significance is below 0,05. For hypothesis 1, the results show no support. There is no significant relationship between cultural distance and the likelihood of a firm to establish a majority-owned (versus minority-owned) affiliate in the host country (b= -0,36, p=0,244). When conducting the tests for the hypothesis 2, model 3 brings

(32)

significant support. The public listing of a parent company has a significant effect on a firm’s likelihood to establish a majority-owned (versus minority-owned) affiliate in the host country (b= 0,134, p= 0,000), and the interaction is significant. The last model supports hypothesis 3. The effect of foreign listing on the likelihood of a firm to establish a majority-owned (versus minority-owned) affiliate in the host country is highly significant (b= 1,571, p= 0,000). Consequently, the results suggest that both moderator variables, public listing and foreign listing, positively moderate the relationship between cultural distance and the likelihood of a firm to establish a majority-owned (versus minority-owned) affiliate in the host country.

As for the control variables, a strong support for firm performance (b= -0,721, p= 0,000) and firm size (b=-0,099, p=0,021) variable was found. Along with these, one industry dummy seems to be significant. The automotive and machinery industry results significant at p=0,021. Therefore companies from this group have are more likely to establish a majority-owned (versus minority-owned) affiliate in the host country.

(33)

Table 1. Descriptive statistics: means, standard deviations and correlations

Variable Me an S.D. 1 2 3 4 5 6 7

1. Ownership structure in affiliate 0.9110 0.28471

2. Cultural distance 1.0725 1.33506 .024**

3. Public listing of a company 0.86 0.346 .054** .106**

4. Foreign listing of a company 0.47 0.499 .104** .096** .379**

5. Firm performance 10.00 1.363 -.029** .086** .124** .530**

6. Firm size 132.74 116.021 -.241** -.110** -.255** -.112** .181**

7. Firm age 78.46 109.432 -.028** -.050** .056** -.115** .022** -.033**

8. Industry of the Parent Company 3.50 1.423 -,008 -.067** -.053** .134** .182** .034** .059** * p < .05, **p <.01, *** p < .001.

(34)

Table 2. Results of Logistic regression

Dependent variable : Ownership structure in affiliate Controls Model 1 H1 Model 2 H2 Model 3 H3 Model 4

Control variable s Beta Sig. Beta Sig. Beta Sig. Beta Sig.

Firm performance -,721 ,000*** -,711 ,000*** ,031 ,015* -,590 ,000*** Firm size -,099 ,021* -,089 ,041* ,045 0,00* -,534 ,000*** Firm age -,055 ,386 -,054 ,394 ,070 ,04* ,061 ,302 Industry of the Parent Company: ,000 ,000 0,00 ,000

1. Mining, utilities and construction ,272 ,484 ,292 ,455 ,394 ,22 -,213 ,583

2. Automotive and machinery -,847 ,021* -,822 ,025* ,378 ,30 -1,612 ,000

3. Professional and information services -,368 ,316 -,397 ,280 ,372 ,71 -,981 ,008**

4. Wholesale and retail ,676 ,150 ,656 ,162 ,475 ,04* ,247 ,596

5. Manufacturing and other services

Independent variable

Cultural distance -,036 ,244 ,134 0,00** -,113 ,001**

Moderator variables

Public listing of a company ,136 ,000***

Foreign listing of a company 1,571 ,000***

Interaction terms

Cultural distance × Public listing of a company ,138 ,000***

Cultural distance × Foreign listing of a company ,577 ,000***

Constant 3,456 ,000 3,439 ,000 ,377 ,000 3,759 ,000

Model χ2, at p<0.05 704.207 513.399 504.824 499.873

Pseudo R2 .137 .133 .143 .190

Total number of observations 15819 15260 15260 15260

(35)

Discussion

Drogedijk and Slangen (2006) in their work state that cultural distance has a significant influence on the choice of the entry mode, in a host country. The empirical findings and conceptual framework, suggested by this research, bring interesting implications regarding the influence of the cultural distance on the ownership structure of the affiliate in a host country. The findings of this research show that cultural distance is not positively related to the likelihood of a firm based in home country to establish a majority-owned (versus minority-owned) affiliate in the host country. Thus, the results show that the higher culturally distanced firms will more probably opt to establish a minority stake affiliate in the host country. Thus, the research on the relationship between ownership structure of foreign affiliates and its moderating effects deserves greater attention. The two moderating factors, public listing and foreign listing, both showed to have a positive moderating effect on the relationship between cultural distance and likelihood of a firm based in home country to establish a majority-owned (versus minority-owned) affiliate in the host country. More precisely, the results of this research have suggest that if a company is publicly listed and/or has a foreign listing as well, this leads to the establishment of a majority owned affiliate in the host country.

In the following sub-sections the academic relevance, the managerial implications and the limitations and suggestions for the future research are discussed in greater detail.

Academic relevance

This research contributes to the existing literature dealing with cultural distance and ownership structure of the affiliates in the host country by shedding new light on this relationship and assessing the moderating effects of public listing and foreign listing of the parent company. Contrary to our initial expectation, the results of this thesis suggest that the cultural distance has

(36)

an effect on ownership structure of the affiliate in the host country (majority vs. minority stake). This suggests that the higher the cultural distance between home and host country, the more firms will have to invest in terms of time and resources in the affiliate if they have a majority stake. The costs of having a majority stake in the affiliate are higher than if opting for a minority stake. If a company establishes a minority-stake affiliate, they will have a partner who will help them to adapt to the local environment and save time and funds which they would spend in case of establishing a majority stake. As a consequence, a higher cultural distance between a home and host country leads a company to establish a minority owned affiliate, because the costs of having a majority-owned affiliate exceed the benefits.

Existing research evidences that majority ownership of the affiliate results in higher commitment, level of complexity, and responsibility. By trying to respond to those challenges, companies reduce their ownership stake in the affiliate by accepting or undergoing an investment with a local partner (JVs) (Brouthers & Brouthers, 2000; Drogedijk and Slangen. 2006). Findings of this research support such outcomes, showing that both public listing and foreign listing are positively correlated to the cultural distance and the likelihood of a company to establish a majority-owned (versus minority-owned) affiliate in the host country. As shown in the Drogedijk and Slangen (2006) study, the ownership structure of the parent was not a significant characteristic of the cultural distance that would influence entry mode. Hence, this research extended the literature and tried to fill the gap by considering ownership structure of a parent company as a dimension of the cultural distance and by bringing evidence of the impact of the relationship between the ownership structure with a decision on the most suitable form of an entry mode (majority vs. minority).

(37)

Both moderators showed significant association for the relationship between cultural distance and the likelihood of a firm based in the home country to establish a majority-owned (versus minority-owned) affiliate in the host country. A possible explanation for this is that a company which is publicly listed has more knowledge of global markets, has more resources and is willing to risk more by investing in a foreign affiliate with a majority stake. As for foreign listing, companies usually decide to undergo this to gain more capital, gaining more resources and thus allowing the company to invest more. The increased knowledge of foreign markets that new investors bring, adds value to a company and makes it easier to evaluate a decision whether to expand with a majority or minority stake. Publicly-listed companies are usually bigger than the non-listed ones and can easier bear the costs of an unsuccessful investment decision. This suggests that negative relationship between cultural distance and the likelihood of a firm based in the home country to establish a majority-owned (versus minority-owned) affiliate in the host country is affected by the ownership structure of the parent company.

Managerial implications

Along with academic relevance, the findings of this research have managerial implications as well. Firstly, when a company wants to expand abroad it should consider if it is publicly listed, then expand with a majority stake in a newly formed affiliate or, in the case of a private company, expand with a minority stake. As publicly listed companies have more knowledge about global markets due to a more dispersed ownership, they may be facing less uncertainty in a risky, unknown environment outside their home country as compared to a private company. Moreover, publicly-listed and foreign-listed companies usually are financially better positioned compared to the private ones and can bear more financial exposure to culturally more distant countries.

(38)

Secondly, following the results of this study, higher cultural distance does not necessary mean higher ownership commitment in a new affiliate in a host country. Companies find it harder to enforce their corporate environment in culturally more distant countries and easier in those that are culturally closer. This leads to the conclusion that a company expanding into culturally more distant countries will go for a lower ownership commitment in the affiliate (JV) rather than entering a new foreign market through a WOS (majority ownership). Thus companies should be careful when deciding about the ownership structure of the new affiliate while expanding abroad.

Thirdly, companies can look at the findings of this research to analyze the future moves of their competitors. By analyzing the ownership structure of their competitor, they could anticipate whether their competitors are considering a future expansion and how they will enter new foreign markets (majority or minority). In today’s highly competitive global markets, this type of information is important and can help a company to obtain a competitive advantage. With globalization allowing an easier trade between countries, the advantage of the first mover can help a company to be in lead against its competitors.

Limitations and suggestions for future research

This research has also limitations that need to be acknowledged. Firstly, the sample used for the purposes of this research contains only European companies. This makes our findings only relevant for European companies, which are more conservative in foreign expansion than U.S. companies. The U.S market stimulates companies to expand faster and this can be helpful for the purpose of this research. As well, the time period used to observe these companies was a cross-sectional (snap shot), assessing the current situation and not conducting a research over time. Therefore, future research can expand this research by including foreign companies outside Europe (U.S. companies) and use panel data that evaluates a period of time.

(39)

Secondly, cultural distance is used as an independent variable, though it might have been affected since it was calculated based on the data collected by Hofstede (1980). Hofstede’s database might be obsolete and it does not contain all countries in the world. This can influence the results of this research since companies that opened an affiliate in a country without the data necessary to calculate the cultural distance are not included in the sample.

Thirdly, this study takes into consideration only publicly-listed companies. The rationale behind this is that publicly-listed companies have a more dispersed ownership. Foreign investors bring additional knowledge that can be used as additional information while assessing the possibility for foreign expansion. Usually publicly-listed companies are in lead compared to private companies because they have more investors. . Also private companies have foreign investors in their ownership structure. This implies that private companies may possess the same knowledge as publicly listed companies, a fact that was not taken into consideration in this research. Therefore, another possibility for future research is not to use only publicly listed companies, but use all companies, public and private, that have at least a certain amount of investors and/or more dispersed ownership (that is not concentrated).

In addition to the findings, limitations and opportunities for future research proposed above, another opportunity is to investigate other moderating variables such as entry mode (WOS vs JV) in the relationship between cultural distance and the likelihood of a firm based in the home country to establish a majority-owned (versus minority-owned) affiliate in the host country. Along with this, a multi-level approach and including new factors at different levels (firm, industry, country) might bring new light on this relationship.

Cultural distance between home and host country and the impact on entry modes