Distance and the choice between equity and nonequity on the African continent

Master Thesis Thomas van den Hof

Distance and the choice between

equity and nonequity on the

African continent.

Master in Business Administration

International Management Track

Semester 2, block 3 2014/2015

First thesis supervisor E. Dirksen MSc

Table of contents:

Cover p. 1 Table of contents p. 2 Introduction p. 3 § .1 Intro p. 3 § .2 Research question p. 5 § .3 Hypothesis p. 5 Chapter 1 Entry modes p. 7 § 1.1 School 1: The Uppsala model p. 8 § 1.2 School 2: Transaction Cost

Economics

p. 12 § 1.3 School 3: The Eclectic

Paradigm p. 16 § 1.4 Types of entry modes p. 17 Chapter 2 Distance p. 19 § 2.1 Hofstede & cultural distance p. 19 § 2.2 Kogut & Singh and the KSI p. 21 § 2.3 Ghemawat and everything

Hofstede forgot

p. 25 § 2.4 TheKAGE model p. 26 Chapter 3 Emerging Markets p. 28 § 3.1 Emerging markets explained p. 28 § 3.2 Bottom of the pyramid p. 32 Chapter 4 Case studies p. 35 § 4.1 Methodology p. 35 § 4.2 Heineken Case p. 37 § 4.3 Procter & Gamble Case p. 40 § 4.4 Nestlé Case p. 44 § 4.5 Cross-case analysis p. 47 Conclusion p. 50 Appendix p. 52 Sources Intro p. 53 Sources HDI p. 54 Sources Heineken p. 55 Sources Procter & Gamble p. 56 Sources Nestlé p. 57 Literature List p. 58

§ .1 Intro:

“Over the last ten years, through acquisitions and greenfield operations that we built, we have been shifting our footprint from let’s say, ten, twelve years ago, from 80 % in the developed markets: Europe & North America. And today, 60%, quasi two thirds of our volumes, come out of these high growth markets, from these emerging markets.”

Heineken CEO Jean Francois Boxmeer (2013) Over the past few years, western firms have realised that most European and American markets are saturated and have acknowledged that developing countries have the most growth potential (Heineken Full Year Report 2014). Backed by high GDP growth rates and a 16% annual raise in consumer spending between 2005 and 2008, Fast Moving Consumer Goods companies (FMCGs) are attracted by its potential (Fiorini and Russo 2010). This shift of multinational companies away from the developed world to new emerging markets in the undeveloped world is still relatively recent. Not just Heineken is shifting its focus, other FMCGs like Nestlé and Procter & Gamble have also recognised the opportunities in the emerging markets.

“Nestle is committed to unlock the business opportunities and to promote growth in Equatorial Africa. With 400 million people and an emerging middle-class with rising purchasing power, this region has major potential for Nestle,”

Nestlé CEO Paul Bulcke (2010) My gut is that there is still a tremendous opportunity in emerging markets.

Procter & Gamble CEO Alan George Lafley (2006) The academic world has recognised a turning point in African economic growth and the subject is slowly becoming more popular, a shift showcased in a few studies (Go, Arbache & Page 2008; Fosu 2009; Bloom & Sachs 1998)

The field of business studies has yet to catch up with these developments. The study of entry mode choice has been an integral part of business studies since the 1970’s. Hundreds of articles, academics and different schools of thought have

passed the revue over the years. Researchers have mostly focused on entry mode choices by multinational enterprises (henceforth referred to as MNEs) from developed markets entering developed markets. The only research done on firms entering emerging markets have concentrated on Central & Eastern Europe and China (Meyer et. al. 2009). This thesis aims to complement this research, address the aforementioned recent developments and fill the still existing gap by focusing on multinationals from markets with very high economic development from

Europe and the United States entering markets with low economic development on the African continent.

In order to successfully investigate these new developments certain

theoretical fields will need to be reviewed and researched. Each field of study also represents a chapter in this thesis. The first will be entry mode literature. A short history of the field will be presented which covers the three main schools followed by more recent work on the subject. Secondly literature about distance, either solely focussing on cultural distance (Hostede 1980; Ghewawatt 2001; Kogut & Singh 1988) or comprising multiple topics will be reviewed after which two different existing frameworks will be merged into this thesis’ own concept of distance. The third literature component will contain work on developing or emerging markets. After building a solid theoretical basis for conducting this research, chapter four contains three different case studies on Heineken, Procter & Gamble and Nestlé. Each case study will contain a within case analysis, comparing the situation in different markets with each other followed by a cross-case

analysis, comparing the three MNEs strategy with each other across the continent. Armed with the theoretical basis and the practical knowledge of the MNEs

operations interviews will be conducted with (former) employees of the companies. The results of the interviews will be presented in the fifth and final chapter after which the conclusions and results of the research will be presented in the concluding section.

§ .2 Research Question

Now that you, my dear reader, have been walked through the design of this thesis all that rests is the big question: What does this thesis then exactly research? This paper will focus on the effect of distance on the choice between equity and non-equity FDI by MNE’s from developed countries when entering undeveloped markets in Sub-Saharan Africa. Following this statement the research question is:

What is the effect of distance on entry mode choice (equity versus non-equity) by MNE’s from developed economies when entering markets in Sub-Saharan African emerging economies?

§ .3 Hypothesis

After carefully reviewing the literature, discussing its content and lifting it from its theoretical basis to a more practical view the author of this thesis argues the following: the higher the distance between the home and host country is, the lower the parent firm’s commitment is. The expectation is that firms will choose a non-equity entry mode like exporting, licensing or franchising.

Higher cultural distance will result in lower initial commitment in the form of an non-equity mode. So MNEs choosing a location for setting up production facilities, a high commitment equity mode of entry will look for countries in the region with a lower cultural distance. This might show in the form of choosing a market with a large English speaking contingent of the population.

This brings us to administrative or political factors as the English speaking population in the region results from the fact that a large part of Africa has a history of being an English colony. Lower political distance will increase MNEs commitment. This expectation is backed by the fact that former colonial ties increase trade up to 900% (Ghemawat 2001).

A higher geographical distance will have more effect for certain industries than for others. Heineken will opt for more equity modes of entry than for Procter & Gamble because of the perishable nature of the products they sell. Especially when one takes into account the regional climate.

A lower economical distance will result in higher commitment. All MNEs come from developed western markets, classified as very high human

development by the United Nations (see chapter 3). All markets in the region are classified as very low human development but there are still inter-country differences. The countries with a higher human development index (HDI) score will be more tempting for MNEs to enter in an equity mode considering that the return on investment will be higher and achieved quicker. Even so the differences between the host and home countries are enormous so the effect might be

Chapter 1: Entry modes

The literature on entry mode choice for MNEs when initiating foreign direct investment (henceforth referred to as FDI) into new markets has a long history. The volume of research done on entry mode choices throughout is huge. Its importance is noted by various scholars like Anderson & Gatignon (1986) who go as far as to claim that the success of foreign operations is decided by its mode of entry. There has been so much written on the subject that Myles Shaver (2013), a noted author on the subject wrote a commentary for the Journal of International Business Studies titled: Do we really need more entry mode studies? He argues that it is time to take a step back, assess the state of the field of study and its future. The essence of his commentary is that we do not need more of the same studies. Even though a distinguished scholar has openly questioned the need for more studies in the field in a peer-reviewed journal, the author of this thesis argues that there is still a research gap in the choice of entry mode literature. As stated in the intro most research has focused on firms from developed economies entering other developed markets. The author argues that undeveloped markets are significantly different from developed markets due to their different governance models, institutional frameworks, levels of development of the economy and infrastructure so that further research is needed and thus justifying the thesis subject. So it should be viewed not as more of the same, but more of something different. More recently Hennart & Slangen (2015), two noted scholars in this field of study provided Shaver with a reply on his provocative question firmly stating that the answer is yes, we need more studies on entry modes. After reviewing Shaver’s argumentation they continue with three questions that most definitely need more research. First is the question what determines the evolution of

operations for future entry mode research. Second is what causes the replication of past mode choices. The third question, the one that is most relevant for this study is the question of how the entry mode decision process is structured. Putting these questions at the center of future investigations will assure a bright future for entry mode research according to the authors. A bright future this researcher would like to contribute to in the form of partially answering the third question.

In order to successfully transfer the need for yet another research paper on entry modes from the author to the reader, one must provide a succinct review on the literature written up until this point in time. The research on entry modes has grown to be the cornerstone of business studies and the main focus of

international management. It can be divided into roughly three schools of thought. Each school builds upon the groundwork done by its predecessors, other fields of study, new insights and responds to changes in the institutional environments. Furthermore these schools are classifications or labels that are put upon certain authors and their works in hindsight and one should thus not see the distinctions too sharp.

§1.1 School 1: The Uppsala model of internationalization

The first school appeared in the 1970’s, originated in Scandinavia and essentially created the field of study. In their, at that time, groundbreaking article: ‘The internationalization process of the firm: A model of knowledge development and increasing foreign commitments’ (1977), the authors argue that firms

internationalise through an incremental pattern. Based upon empirical research on the internationalisation processes of Swedish firms in the pulp and paper- and the pharmaceutical industry they noticed certain patterns. Slowly and gradually the MNE’s gain more experience abroad in host-countries that they are culturally and/or geographically close to and move further and further from the home country (Johanson & Vahlne 1977).

This theory of incremental international expansion later became known as the Uppsala model, named after Johanson & Vahlne’s base of operations. Apart from moving further away from the home country, as time goes by and firms gain experience abroad, they not only expand to ever increasing more distant host countries, they also increase the level of resource commitment as they view doing business in the host country as less risky. Shown schematically in figure 1:

Figure 1: The Basic Mechanism of Internationalization:

Source: Johanson & Vahlne 1977.

The sets of variables are displayed, state variables on the left and change variables on the right. Both variables have an effect on each other and thus show the dynamic process of learning when expanding internationally. All firms which were researched showed the same pattern: no regular export  independent representative (a sales agent)  sales subsidiary  production facility. Note the gradual change from a non-equity to equity form of FDI (Johanson & Vahlne 1977). This increasing continuum in levels of resource commitment is supported by other scholars as for example in Chu & Andersons’ (1992) article, who link this increase with an increase in profits.

In the present day and age, the original Uppsala model seems less relevant because the gradual part of their model was based on not being able to obtain large amounts of knowledge about foreign markets in a short period of time. With the advent of new technologies like the Internet, the possibilities of obtaining

information about foreign markets seem endless. Furthermore the economy has changed, the regulatory environment has changed, companies have changed and new fields of study have contributed to the subject in the meantime (Johanson & Vahlne 2009). Fortunately Johanson & Vahlne are still active researchers

publishing new versions of the Uppsala model every few years.

In their 2009 paper they argue that the model needs to be revisited due to the changes in the environment. In modern times the business environment can be seen as a web of relationships or networks as opposed to a couple of decades ago when it contained more independent suppliers and customers that were less connected. Internationalization is now seen as a set of actions to strengthen the

firm’s networks position, actions which used to be described as improving or protecting market positions. Overcoming various barriers of entry when entering new markets, the traditional view has become less important compared to

internationalizing undertaken to strengthen the firm’s position. The authors claim that existing business relationships have a considerable impact on the particular geographical market to enter and which entry mode to use. The paper also

provides us with a revised model which bears a close resemblance in structure to the original model displayed in figure 1. The state aspect, resembling the status quo, and change aspects remain but the focus has shifted from knowledge about markets into knowledge about networks and then using this knowledge to improve the network position as shown in figure 2:

Figure 2: The business network internalization process model:

Source: Johanson & Vahlne 2009.

Again the variables influence each other, displaying a dynamic relationship of incremental learning, creating and network building. Furthermore the authors argue that a firm’s problems and opportunities in international business are not country-specific but network specific hence the paper’s subtitle from liability of foreignness to liability of outsidership.

In their most recent paper Vahlne & Johanson (2013) provide us yet again with an update on the Uppsala model in an effort to explain the evolution of the MNE into the multinational business enterprise or MBE. The model is again

dynamic, showing that when new knowledge is gathered it will have an impact on the commitment decisions. The commitment decisions will then influence the subsequent knowledge development. In turn this influences and enhances the MNEs position within a business network. Because of these cycles of growth, some MNEs have grown so big that they evolved into MBEs as shown in figure 3:

Figure 3: The Uppsala model of MBE evolution:

Source: Vahlne & Johanson 2013

In their concluding remarks the authors state that in an effort to better understand the strategic considerations and contextual aspects that are important in the processes of internationalization and globalization, case studies in real time have to be conducted. By conducting three different case studies this thesis will research if the firms involved have expanded in the gradual, incremental way this school proposes. Two authors who also have taken up Johanson & Vahlne on this call to action are Schuster & Holtbrügge (2012). In an effort to fill the knowledge gap on MNEs entering low income markets they conducted three case studies by analyzing Vodafone, Allianz and Unilever activities in Africa and Asia. The

theoretical foundation used for the analysis was Johanson & Vahlnes

internationalization process. A distinction is made between top of the pyramid (TOP) sections and bottom of the pyramid (BOP) sections within markets. The research then states that when MNEs are active in the TOP sections they tend to

expand to the BOP sections. Furthermore the authors argue that in these three cases, as time goes by, market knowledge and market commitment both increase like in the original Uppsala model. So here we find proof that in modern

circumstances with MNEs entering developing markets the Uppsala model still holds (Schuster & Holtbrügge 2012).

To conclude, the Uppsala model advocates a gradual involvement in the foreign market. To combat the risk of unknown factors, different markets and actors firms should enter new markets in a low commitment form like export. When more knowledge and experience is gained firms can gradually increase their resource commitment (Pan & Tse 2000). The Uppsala model goes a long way in explaining the international expansion of firms. It has received much praise, the original 1977 paper is one of the most cited ones in the history of the field of business studies. However the Uppsala model represents just one school of thought in the literature on entry mode decision. In order to have a full

understanding of the subject we move on to the next school of internalization. § 1.2 School 2: Transaction Cost Economics / Internalization

The second school of thought that has influenced thinking on entry modes has a more widespread use in economic theory and business studies. The theories behind it can be used as an explanation for a wide variety of occuring business phenomena and unsurprisingly this has been done for entry modes.

The basic mechanisms of transaction cost economics and internalization go back further than the Uppsala model. Transaction cost economics were first mentioned by Ronald Coase in his 1937 paper “The nature of the firm”. It was a groundbreaking theory and earned him a Nobel prize in economics later in life. At the time, the thinking about economics and markets was dominated by Adam Smith. According to Smith the invisible hand, or the price mechanism guides the market, providing everyone with the correct price. Following this logic, the market would organise itself through unconscious cooperation and there would be no need for firms. Everyone would buy and sell products at the right price as dictated by the market (Smith 1776). Coase goes on to argue that entrepreneurs also fulfill a coordinating function in the market and sets out to fill the gap in economic reasoning that is left because of this dichotomy. He argues that there are costs of

using the price mechanism. For example that the amount of effort needed to obtain a certain product may not always be worth the same as the market price. He

summarizes these costs as bringing to market- or marketing costs (not to be mistaken with the modern connotation). By organising market transactions the entrepreneur can eliminate certain market costs thus turning a profit. By adding more and more of these transactions the firm can grow, up until the return of adding extra transactions do not exceed the costs of carrying out the transaction (Coase 1937). Coase was the first to explain the existence of the firm and laid the groundwork for transaction cost economics, on which Buckley & Casson built their foundation in 1976.

Building upon Coase’s transaction cost economics, the two British scholars Buckley & Casson developed their internalisation theory of the firm. In their 1976 book The future of the multinational enterprise they provided a broad based analysis of the MNE, defining a few important principles. First was the fact that an MNE was an enterprise that owned and controlled activities in two or more different countries. Secondly the boundaries of the firm were set at the margin where further benefits of internalizing market transactions did not outweigh the added costs. A notion that clearly shows Coase’s influence. Thirdly firms strived to be at the lowest cost location for each of its activities. Fourth a firm’s profitability was determined by the non stop process of innovation from the firm’s research and development department. In hindsight these findings may seem obvious and logical but at the time they were quite remarkable and forward thinking. Keep in mind the attention paid to the firm’s location, something we will encounter in the next paragraph. Underlying these principles is the assumption that actors are driven by rational actions. On a more practical note, the authors distinguished two forms of internalisation: operational and knowledge. Operational internalisation involves products flowing through successive stages of production and

distribution and knowledge internalisation involves knowledge flows from research and development (Buckley & Casson 2009).

Anderson & Gatignon (1986) set out to combine transaction cost economics with entry modes. More specifically they provide a framework for investigating the entry mode decision as shown in figure 4. They argue that the most efficient entry mode choice is chosen by way of a trade off between control and the cost of

resource commitment. Furthermore they were the first to provide a scale for the entrants degree of control and a transaction cost framework for analysing the efficiency of entry modes. By internalising the control of foreign operations the firm diminishes the degree of risk and potential external threats that face the firm’s subsidiary.

Figure 4: A transaction cost framework for analyzing the efficiency of entry modes:

Source: Anderson & Gatignon (1986)

They provide four sets of constructs that determine the choice of entry mode as shown in figure 4. First are transaction-specific assets or investments in the subsidiary. Second is external uncertainty, the unpredictability of the

subsidiaries environment. Third is internal uncertainty, how well do the

measurable outputs of the subsidiary tell the story of how it is performing. Last is free-riding potential or the ability to receive benefits without making costs. The entry modes are classified on the basis of the degree of control the firm has on its subsidiaries.

The framework is then used for making nine propositions (see Table 1) on the long-term efficiency of the subsidiary with the main proposition being that a low level of ownership is preferable until proven otherwise. Apart from the main proposition one especially sticks out in consideration of the topic of this thesis. Proposition seven states that when socio cultural distance is large, firm’s should either go for a high- or low level of control, avoiding intermediate. Furthermore it

states that high levels of control only work if the mother firms’ way of doing

business is substantially better than the local version (Anderson & Gatignon 1986). Table 1: Anderson & Gatignons’ propositions:

P1: Modes of entry offering greater control are more efficient for highly proprietary products or processes.

P2: Entry modes offering higher degrees of control are more efficient for unstructured, poorly-understood products and processes.

P3: Entry modes offering higher degrees of control are more efficient for products customized to the user.

P4: The more mature the product class, the less control firms should demand of a foreign business entity.

P5: The greater the combination of country risk (e.g., political instability, economic

fluctuations) and transaction-specificity of assets (proprietary content, poorly understood products, customization, product class immaturity), the higher the appropriate degree of control.

P6: The entrant’s degree of control of a foreign business entity should be positively related to the firm’s cumulative international experience.

P7: When socio cultural distance is great:

a: Low-control levels are more efficient than intermediate levels; b: High-control levels are more efficient than intermediate levels;

c: High-control levels are more efficient only when there is a substantial advantage to doing business in the entrant’s way.

P8: The larger the foreign business community in the host country, the lower the level of control an entrant should demand.

P9: Entry modes offering higher degrees of control are more efficient the higher the value of a brand name.

Source: Anderson & Gatignon (1988).

In practice this comes down to the fact that when entering a new market, according to Anderson & Gatignons transaction cost framework, firms should start with exporting, franchising, licensing or acquiring minority equity positions before moving on to a higher level of resource commitment. This is a strategy of

incremental expansion similar to the Uppsala model. Only when the initial results are positive should firms switch to higher levels of control or commitment. These propositions will be taken into consideration when interviewing the case studies firm’s employees.

Anderson & Gatignon are not the only authors who are proponents of the transaction cost economics view within the literature. Other authors as Beamish & Banks (1987) explain the choice of a joint venture over a wholly owned subsidiary by using the internalization theory. Kogut & Singh (1988) agree with Anderson & Gatignon when they argue that when firms have the capability to perform

transactions, for example financial transactions or the transport of goods from one country to another, for a lower cost than when using the currently available

market mechanisms, firm’s create their own mechanisms to internalise these transactions (Pan & Tse 2000).

§1.3 School 3: The Eclectic Paradigm or the OLI framework

The third school of thought also builds upon internalisation theory, Dunning even refers to it as its near relative, thus continuing the findings of the second school as described in the previous paragraph (Dunning 2000). The eclectic paradigm is a construct based upon three types of advantages a firm can have over its competition: ownership (O), location (L) and internalisation (I). Hence the paradigm is also known as the OLI framework. Ownership advantages of the firm engaging in FDI are firm specific advantages giving it a competitive edge. The advantages can come from a variety of different resources as for example capital, innovative technologies or a talented workforce. Locational advantages create attractions for firms to invest in a subsidiary in an alternative location as opposed to the home location. This can either be the availability and or abundance of cheap resources, high quality modern infrastructure or a highly skilled or extremely low cost work force. The third section of advantages that concludes the framework comes forth from the way the firm is able to combine the first two sets of advantages, or internalise them (Dunning 2000). Dunning states that:

“the greater the net benefits of internalizing (sic) cross-border intermediate product markets, the more likely a firm will prefer to engage in foreign production itself, rather than license the right to do so, e.g. by a technical service or franchise agreement, to a foreign firm.” (Dunning

2000).

So according to the eclectic paradigm, firms will choose an equity based entry mode when entering new markets.

In more recent years the research on entry modes has not slowed down. Authors like López-Duarte & Vidal-Suárez (2010), López-Duarte & Vidal-Suárez (2013), Brouthers (2002), Chang et. al.(2012) and Chiao et. al. (2010) have

continued to produce an impressive stream of articles on the subject. These recent papers have proved to be essential in identifying the research gap for this article because the study of entry modes in undeveloped economies on the African continent is still too large a subject. As the more recent articles show it is

necessary to narrow down the different modes of entry, the geographical location and the moderating effect. For example López-Duarte & Vidal-Suárez (2013) research the effect of cultural distance on the choice between wholly owned subsidiaries and joint ventures. In their concluding remarks they state that the study provides conclusive results as to the questions to include other forms of distance when researching distance between countries. Furthermore they suggest that in depth analysis of FDI in markets with a lower political stability than their example, Spain, are needed. What these recent studies fail to do is look at the bigger picture. In order to do this, we have to consider Pan & Tse’s 2000 article on which the next paragraph will elaborate. They provide a hierarchical model for entry modes arguing that firms first make a decision between equity versus non-equity entry modes and subsequently within these two forms of FDI. No research has been done on the choice between these two for firms entering undeveloped markets.

§1.4 Types of entry modes:

This research makes a distinction between two sorts of entry modes: equity and non-equity. In the past, authors as Anderson & Gatignon (1988) have provided a classification of entry modes based upon the degree of control the parent company would have over its subsidiaries. Kumar & Subramaniam further develop this classification into a hierarchical model of entry mode decisions. They assume that managers use a hierarchical process when choosing entry modes, that at each level they consider a few critical factors and that the nature of these factors also change at each level (Kumar & Subramaniam 1997).

Building upon these three schools of thought that dominate the literature on entry modes are the scholars Yigang Pan & David Tse. In their 2000 article they

provide a hierarchical model of entry modes, shown schematically in figure 5. According to them the first level in the hierarchy is the choice between an equity mode of entry versus a non-equity mode of entry. Equity modes of entry are major resource commitments for MNEs. They require a lot of manpower, financial

resources and involve a lot of risk. Equity modes of entry are either equity joint ventures or wholly owned subsidiaries. Non-equity modes of entry include

exporting or contractual agreements. Non-equity modes require smaller resource commitments from firms and are thus less risky. As with most investment

decisions the less risky they are, the lower the return on investment is. Figure 5: A hierarchical model of choice of entry modes:

Chapter 2: Distance

Further complicating the matter is the distance between home countries of the MNE’s and the host countries of their subsidiaries, either be it cultural or geographic. The effect of distance between home and host countries has been described by different noted scholars like Hofstede (1994), Kogut & Singh (1988), Ghemawatt (2001) and Ronen & Shenkar (1985). Where some authors like

Hofstede focus on cultural differences, other authors like Ghemawat go for a more all-round approach listing multiple dimensions for distance.

§ 2.1 Hofstede & cultural distance

Geert Hofstede is widely acclaimed as an expert on the subject of cultural distance. His research among subsidiaries of IBM and later on amongst students in different countries have received praise from the scholarly community for decades.

First described in his groundbreaking book Culture’s Consequences: International Differences in Work-Related Values (1980), which he wrote after conducting his research at IBM which was quite an impressive one as 88,000 employees in over 40 countries participated in the survey. Hofstede states that national cultures differ on four aspects, each of which will be shortly explained and a workplace example provided.. The first, power distance, is the level of acceptance that less powerful members of organizations and/or institutions have concerning the unequal distribution of power. In the case of a business setting, in small power distance societies the ideal boss will be seen as a resourceful democrat who

consults with his subordinates. Large power distance societies normally see a benevolent autocrat who tells his or hers subordinates what to do as an ideal leader.

The second concerns individualism versus collectivism, which is mostly based upon the way one experiences family life. Either one identifies itself as being a part of the immediate family (nuclear) or part of the extended family. The

difference between collectivist or individualist societies is also shown in different things on the workplace. Individualists will find completing a task more important than the relationships forged, contrary to a collectivist.

The third difference is divided along gender lines: masculinity versus femininity. This does not imply some countries are ruled by women and other by men. The masculine societies are societies where assertiveness is appreciated, competition is viewed as something good and (study)careers are extremely

important. The feminine societies focus more on modesty and caring for each other with the emphasis on relationships, solidarity and intuition. In summary the

difference can be best explained as a masculine task orientation focus versus a feminine person-orientation.

Number four focusses on uncertainty avoidance. To what degree is a society at ease without clear, structured and formalized regulations? In weak uncertainty avoidance societies different things are viewed as curious, emotions are preferably not shown, a dislike of rules, formalization and standardization is evident. On the opposite side in a strong uncertainty avoidance society different things are dangerous, emotions can be shown, students are comfortable with strict time tables and precise objectives and there is an emotional need for rules, either written or unwritten. People in the latter societies also often claim a religious and philosophical absolute truth.

Hofstede’s last dimension entails the difference between long term and short term orientation and was found in one of his later studies. In this particular research Hofstede used a questionnaire devised by Chinese students. By using this study and finding a new dimension Hofstede acknowledged that his ideas and his research were culturally limited, a product from Western social science. His ideas on culture were heavily influenced by him being from the west and he needed the help of Chinese students to identify this last dimension (The Chinese Culture Connection 1987). The long term orientation incorporates values about

perseverance and thrift whereas the short term orientation has values associated with respect for tradition, fulfilling social obligations and protecting one’s face. The values on the dimension are oftentimes referred to as Confucian dynamism but can also be applied to countries without a Confucian heritage (Hofstede 1994).

Extensive research on Hofstede’s dimensions is conducted even to more recent times. In 2010 Michael Minkov published the results of his research on cultural differences in his book Cross-Cultural Analysis: the science and art of comparing the world’s modern societies and their cultures (2010). By doing this

research he extended the amount of countries that have been researched from around 40 in the 1970’s up to 93 this day. Furthermore he found more evidence for the fifth dimension, which he named normative vs. pragmatic and he found evidence for a sixth dimension: indulgence vs. restraint (Minkov 2010). § 2.2 Kogut & Singh and the KSI

Hofstede first published the results of his research in 1980. Using the results from the surveys, Hofstede not only provided us with four dimensions of cultural

distance but also with a score or index on each dimension for each country. Within a decade two fellow scholars Bruce Kogut & Harbir Singh set out to develop a formula which could accurately calculate the cultural distance between two countries which they published in their paper: The effect of national culture on the choice of entry mode (1988). By then scholars were already using Hofstede’s work in research on entry modes.

In response to numerous claims that characteristics of national cultures influence the selection of entry modes when entering foreign markets they set out to develop a theoretical argument for why this happens. To prove their arguments they created a formula to calculate the difference between cultures. The authors argue that the larger the cultural distance between the country of the investing firm and the host country, the more likely a firm will choose a Joint Venture (JV)or Wholly Owned Greenfields (WOG) over an acquisition. Furthermore they argue that the higher the uncertainty avoidance in the home countries firm’s culture is, the more likely it is that the MNE will choose a JV or WOG over an acquisition. The formula they created, the Kogut & Singh Index (from now on the KSI) uses the deviations of each of the four, at the time known, cultural dimensions of the home and host countries. The deviation is the difference between the countries score on a dimension minus the dimensions mean or average. They then corrected the deviations for differences in the variances of each dimensions and then averaged these. This leads to the following formula:

where Iij stands for the index of the ith cultural dimension and jth country, Vi is the variance of the index of the ith dimension, u indicates the home country and CDj is the cultural difference of the jth country from the home country. In the case studies done in chapter four, depending on whether the data for the relevant countries is available, the KSI will be put in practice.

The KSI has been adopted by a wide variety of scholars in various fields of study. As of 2014 it has been used in 310 empirical studies (Yeganeh 2014) and according to Google Scholar on the 15th of January 2015 has been cited 4137 times. Nothing in this world is perfect so neither is the KSI. An index with such a widespread use was bound to receive some critique, the most notable of which are commonly known as the illusions of the KSI by Shenkar (2001) as seen in table 2. Table 2: Illusions of the KSI:

1. Illusion of symmetry The implications of distance are often asymmetrical. Dutch firms may have less trouble operating in China than vice versa. 2. Illusion of stability The distance between two countries is not a

stable fact but always changing.

3. Illusion of discordance The difference doesn’t necessarily have to be problematic.

4. Assumption of equivalence Not all dimensions are equally important when doing business.

5. Illusion of linearity The KSI assumes a linear association between CD and bilateral business relations.

6. Illusion of causality The KSI implies that cultural difference is the main variable influencing relations.

7. Assumption of corporate homogeneity It neglects variance in corporate culture. 8. Assumption of spatial homogeneity It assumes homogeneity of culture withing

countries.

Sources: Yeganeh (2014) & Shenkar (2001).

Of his eight points of critique one stands out from the rest and that is the illusion of causality.

“The illusion of causality implies that cultural distance is the main variable that necessarily has a causal effect on bilateral business relations” (Yeganeh 2014)

So cultural distance may be one of the factors affecting bilateral business relations. Many other factors may be of an influence like political, economic or geographic factors as will be explained in the next paragraph. Fortunately, unlike other illusions stated by Shenkar, this inaccuracy does not stem from the way the index is constructed so the use of the KSI in this study is still warranted. Even though Shenkar has provided us with no less than nine points of critique on the KSI it is still widely used and even more so since the publishing of his 2001 paper. Pre 2001 a total of 62 studies used the KSI, post 2001 a total of 242 empirical studies employing the KSI have been published. Even after Kirkmans et. al. (2006) paper advising to avoid the KSI the number of studies using kept on rising. We take into account these critiques but still value the KSI as useful enough. Seeing that there are still dozens of articles published in peer reviewed journals using the KSI, the author of this thesis does not feel prohibited in using the KSI as a useful measure of cultural distance and will do so in the case studies in chapter four.

In a more recent paper on the KSI, Hamid Yeganeh (2014) provides an improved formula. His critiques are defined as conceptual and he provides methodological (mathematical) solutions which will be mentioned briefly. The conceptual problems build upon the work of Shenkar (2001). The formula unjustly assumes that the four dimensions used by the KSI are of equal importance, the KSI neglects correlations amongst cultural dimensions and it neglects cultural

asymmetry. In order to counter the assumption of equivalence Yeganeh uses the HDI because it is a broader measurement of socio-economic development. The second step uses a concept known as Mahalanobis Distance (MD) to calculate the cultural distance. The use of MD is recommended for the calculation of many cross-national measures by Berry et al. (2010). Mathematically the MD (in this use) is calculated as the following:

The third and last step counters the illusion of symmetry which, simply stated comes down to the distance from for example the Netherlands to China does not

necessarily be as high as the distance from China to the Netherlands. Inspired by Modernization Theory, Yeganeh argues that developed societies share cultural traits such as Individualism and Low Power Distance while developing societies share opposite traits like Collectivism and High Power Distance. Building upon this knowledge and Modernization theory, Yeganeh defines:

‘cultural distance from country j to country u as positive if country u is socio-economically more developed than country j, and negative if country u is socio-economically less developed than country j.’ (Yeganeh 2014).

When countries are at the same state of development, the cultural distance becomes symmetrical and the algebraic +- sign loses its relevance. Based upon these three changes Yeganeh presents the formula for his weighted, Mahalanobian and symmetrical approach to calculating distance:

Where:

WMAj  u stands for cultural distance from country j to country u.

Xj and Xu denote 4-dimensional vectors (weighted Hofstede’s four

dimensions) or 9-dimensional vectors (Weighted GLOBE’s nine dimensions) of X for jth and uth countries

C is the covariance matrix for X T is the transpose operator

HDIj and HDIu are respectively the long-term average of human development index

for countries j and u WMAj  u is positive (+) if HDIu > HDIj

WMAj  u is negative (-) if HDIu > HDIj

WMAj  u can be either negative or positive (neutral) if exceptionalyHDIj= HDIu

Fortunately the appendix of his paper provides the national cultural distance scores for all countries for which data are available making it unnecessary to calculate these ourselves and further investigate the workings of the adjusted formula.

Unfortunately not all Sub-Saharan countries have data available so we can not calculate the CD for all of the MNEs subsidiaries. This stems from the fact that in some countries the necessary data are not available (possibly due to weak governments, social or military conflicts etc.) and the fact that Sub-Saharan countries are not necessarily hot spots for academic interest. Hopefully in the future this situation will change as Western firms continue to seek value in these markets.

More recently five scholars undertook a project named GLOBE in which they took societies rather than countries as their measure of analysis for cultural differences. This new basis was chosen to account for big cultural differences within countries as for example South Africa. Nine different attributes or cultural dimensions were selected: assertiveness, institutional collectivism, in-group collectivism, future orientation, gender egalitarianism, humane orientation, performance orientation, power distance and uncertainty avoidance (House et. al. 2004). Even though the GLOBE approach is an addition to Hofstede’s work it still fails to account for a lot of different factors. These factors will be explained in the next paragraph.

§ 2.3 Ghemawat and everything Hofstede forgot

More recently the influential scholar Pankaj Ghemawat (2001) published an article on distance between countries which can either be seen as a critique or a

complementing study to Hofstede. Using the example of one of Rupert Murdoch’s failed attempts at starting a media company in China he argues that the concept of distance when doing business is made up out of four different dimensions:

cultural, administrative, geographic and economic. Ghemawat does not give a lot of attention towards cultural differences apart from lingual differences, social norms, religions and ethnicity. Perhaps because of Hofstede’s dominating work on the subject. This lack of attention towards cultural differences will have an impact on the framework used in this study. With the administrative or political distance his framework is starting to make more sense. It can be broken into three main components. First colonial ties between countries can boost trade by 900%, preferential trade agreements and political unions (for example the EU) can boost

inter-country trade by more than 300%. Secondly distance can also be created in both host and home countries on the administrative level. Home country distance is mostly created in the form of domestic prohibitions on bribery and health and safety standards. Host country distance is created by imposing tariffs, trade quotas, restrictions on foreign direct investment and subsidies or favoritism for domestic competitors. These last measures normally occur when certain big, national

industries that provide jobs for a majority of the people are seen to be in danger of international competition. Thirdly when a host country has a weak institutional infrastructure it can decrease international trade. Countries that are ridden with corruption and or subject to large scale social conflicts tend to be avoided by foreign MNEs.

Moving to the most easily comprehensible dimension of distance: geography. Ghemawat states that not only the absolute distance between home and host countries should be a factor but also size of the country, within-country distances to the border, access to waterways and oceans, transportation and communication infrastructures should also be included in the mix. Although geographical distance plays a more important role in some industries as compared to others, even intangible goods and services are affected by it.

The last dimension in Ghemawat’s framework is economic distance. The consumer wealth, income and GDP per capita is the most influential factor on levels of trade between countries. Rich countries engage in relatively more international

economic activities with other rich countries than poorer countries (Ghemawat 2004).

§ 2.4 The KAGE model

The work of Ghemawat is a welcome supplement to the work done by Hofstede and his followers. While not being perfect, Hofstede’s work on cultural distance has stood the test of time and to this day he is the most cited author on the subject. Kogut & Singh’s translation from a theory into a working equation to actually measure the distance between two countries has also greatly added to the practicality of it. Even though never claiming to be more than a measure for cultural distance, Hofstede’s work falls short when measuring the actual distance

between two countries when doing business as Shenkar (2001) pointed out to us. Enter Ghemawat’s four dimensions.

According to the author of this thesis these two models combined perfectly describe the distance between two countries. So I propose a merger of Hofstede’s cultural work, in the form of the revised KSI, as the culture dimension (K) of Ghemawat’s framework. Complemented by Ghemawat’s dimensions of

administrative (A), geographical (G) and economic distance (E). All these measures and dimensions together will form the concept of distance, the KAGE model for assessing distance between national markets as seen in figure 6:

Figure 6: The KAGE model for assessing distance:

During the case studies the KAGE model will be translated into questions about the decision making process behind the choice of entry modes of the relevant

companies and markets. Adjusted KSI • Uncertainty avoidance • Individuality vs. collectivism • Power distance • Masculinity vs. feminity Administrative • Trade Blocs • Colony / Colonizer relationship • Degree of corruption • Legal origin Geographical • Kilometers beweteen main cities • Adjecency • Land Area • Time Zone • Climate Zone Economical • GDP per capita • Real GDP growth rate • Human Development • Internet population

Chapter 3: Emerging markets

Last but not least this paper will review the literature on firms entering emerging markets like Meyer et. al. (2008) who combined institutional and resource based views on entry strategies in emerging markets.

§ 3.1 Country classifications

When writing a thesis on emerging markets one thing that comes to mind is what exactly are emerging markets? To provide a clear definition for the term emerging markets in this research the author dove into the terms history. It was first coined by Antoine van Agtmael when working for the World Bank’s International Finance Corp. (IFC) in the 1980’s in an effort to rebrand the large amount of countries that had enormous, yet unrealized economic potential. At the time all the countries not part of the first and second world were branded underdeveloped countries. In order to make a necessary distinction between the countries with and without potential (from an investor standpoint) the term emerging markets was born (Van Agtmael 2012).

Since then the term has been used by a wide variety of different global organisations. Some of these organisations issue annual reports where they sort every country worldwide into different classifications. Globally there are three major players when it comes to gathering socio-economical statistics: the

International Monetary Fund (IMF), the World Bank and the United Nations. There are also a few private firms who provide statistical data mostly for investors, investment banks and hedge funds of which the MSCI (formerly the Morgan Stanly Capital International) is the most notable one. Each of these organisations also provide their own classifications of countries worldwide.

The MSCI sorts the world into three classes or investment universes. The classification takes place based upon economic factors as economic development, size and liquidity requirements and market accessibility criteria. The first are the developed markets, consisting of the major western economies. The second are the emerging markets, with South Africa being the only country with our geographic scope. The third are frontier markets, containing a few Sub-Saharan countries like Botswana and Ghana. Due to the fact that there are a lot of countries missing in

their classification and that all measures are economic in nature we can dismiss the MSCI classification (Source MSCI website).

The World Bank also provides a classification based solely upon Gross National Income (GNI) per capita. They list four categories: low, lower-middle, upper middle and high were the bottom two are branded as developing only. This amounts to all countries with a GNI per capita of below $4,125 as developing. Because all measures are economic in nature we keep on looking for a more complete measure (source World Bank Website).

The International Monetary Fund or IMF maintains a simpler classification. To them the world consists of advanced economies on one side and emerging and developing economies on the other. The only distinctions they make within the emerging and developing economies are geographic of nature. Keeping in mind the shortcomings of such a classification and the lack of theoretical or statistical

backing we can dismiss this one as well (Source: IMF website).

Off all the classifications out there the one made by the United Nations seems to be the best. It is the most widely used index in academic circles and accounts for a multitude of different factors and is based upon the Human

Development Index or HDI. The HDI is made up out of three different dimensions believed to be the cornerstone of leading a decent life. It attempts to rank all

countries on a scale from zero (lowest development) to one (highest development) The first dimension is a long and healthy life or longevity measured by life

expectancy at birth. The second dimensions is knowledge which is measured by two facts, the average years of schooling by the nation’s population and the expected years of schooling at birth. The third measure is a decent level of living measured by the Gross National Income (GNI) per capita. The GNI is the “total domestic and foreign output claimed by residents of a country, consisting of gross domestic product (GDP) plus factor incomes earned by foreign residents, minus income earned in the domestic economy by nonresidents.” (Todaro & Smith, 2012, p. 44). To make the GNI comparable over time, cancelling the effect of changes in exchange rates and inflation, the researchers at the United Nations Development Programme (UNDP) used purchasing power parity (PPP) as a conversion factor. PPP is defined as the amount of units of a country’s currency needed to buy the identical amount of goods and/or services in the local developing country as would

1 dollar buy in the United States. This is an internationally recognised way of providing more accurate comparisons of living standards. (Todara & Smith 2012).

The technical notes which are published online alongside the report give us an insight in how the HDI is calculated. We use the 2013 version as the 2014 was not available at the time of writing. It takes two steps to calculate the HDI, the first one is creating the dimension indices. The researchers have set minimum and maximum values or goalposts to transform the indicators into indices between zero and one. The maximums are the highest observed values in the time period measured by the UN. So for example the highest life expectancy is 83,6 years in Japan and the highest GNI per capita is $87,478 in Qatar. The minimum values are set at twenty years for life expectancy, at zero years for the education variables and at $100 dollar for GNI per capita (keeping in mind the considerable amount of unmeasured economic transactions in certain parts of the world). We then

calculate the dimension as follows: Dimension index = actual value – min. value / max. value – min. value. In the next step we aggregate the subindices to produce the HDI which is the geometric mean of the three subindices:

Et voila, the HDI.

The UN has divided all the countries in the world (with a few exceptions for which no data can be found) into four classes as seen in table 3:

Table 3: Human Development Index

Very high human development (0.90 – 1) High human development (0.80 - 0.899) Medium human development (0.50 - 0.799) Low human development (0.0 – 0.499)

Source: UNDP Human Development Report 2014 One of the advantages of the HDI is that it shows that a country can do much better than might be expected at a low level of income and that gains in level of income don’t necessarily mean a gain in HDI. It also shows us that differences in

income are bigger than differences in health and education. Furthermore it shows that by development we look at more than just economic development (Todaro & Smith 2012). Looking at multiple factors when comparing different countries is also something taken into account when calculating the distance between the home and host countries so it serves us better to look at the bigger picture in this part as well.

Naturally the HDI is not perfect and Todaro & Smith provide us with a few critiques. No equal weight is given to all three dimensions in the formula. This would imply that all three factors are equally important which is somewhat debatable. Also there is no account for quality, for example quality of schooling differs in developed countries as compared to undeveloped countries. Finally there are of course better ways of measuring health and education than the ones chosen here but we have to keep in mind that these data need to be available for every country in the world (Todaro & Smith 2012). Making it more complex would make the HDI less complete.

As with all classifications, the one used for the geographical demarcation of this study is man-made and does not account for all the different cultures, nations, languages and religions which are prevalent throughout the region. Keeping this in mind, it is still necessary to create this demarcation for the sake of the time limit imposed on this study. The focus of this thesis is on countries in Sub-Saharan Africa but what specifically is Sub-Saharan Africa? The definition used by the United Nations seems most appropriate and thus will be used here as well. It states:

“The designation sub-Saharan Africa is commonly used to indicate all of Africa except northern Africa, with the Sudan included in sub-Saharan Africa.” (Source: United Nations Statistics

Division)

Please note that northern Africa in this case is made up out of: Algeria, Egypt, Libya, Morocco, Tunisia and the Western Sahara.

§ 3.2 Bottom of the Pyramid

Research on doing business in low HDI markets has been relatively new, but the subject seems to be receiving more attention in recent years. Scholars who played a major role in kicking off this new trend have been Prahalad & Hart (2002) by publishing their paper: The fortune at the bottom of the pyramid. This very

influential paper brought forth the BOP prophecy, a thoroughly criticised prophecy about making money from the world’s poor whilst helping them at the same time. A short review about Prahalad & Hart’s paper will follow, accompanied by a few critiques on it by Karnani (2007) and more recently Agnihotri (2014) and finally its presence in this thesis will be discussed. So who exactly are the people in the bottom of the pyramid, what pyramid are we talking about and what constitutes the bottom?

Prahalad & Hart (2002) argue that there is money to be made from the world’s poorest four billion people at the bottom of the World Economic Pyramid. Based upon data from the United Nations World Development reports (no year is given) they divide the world population in four tiers based upon the annual per capita income (adjusted for PPP). Tier one contains a hundred million people making more than $20,000 a year. Tier two and three make up about 1.75 billion people raking in an annual income between $1,500 and $20,000. Tier four, the bottom, contains four billion people who make less than $1,500 a year. In order to do this, MNEs have to radically rethink and reshape the way they produce their products and services and reexamine a set of assumptions shown in table 4. A set of assumptions that currently are, according to the autohor, viewed as normal by managers in MNEs around the globe.

Table 4: Assumptions about the BOP:

Assumption # 1 The poor are not our target consumers because with our current cost structures, we cannot profitably compete for that market.

Assumption # 2 The poor cannot afford and have no use for the products and services sold in developed markets.

Assumption # 3 Only developed markets appreci- ate and will pay for new technology. The poor can use the previous generation of technology.

Assumption # 4 The bottom of the pyramid is not important to the long-term viability of our business. We can leave Tier 4 to governments and nonprofits. Assumption # 5 Managers are not excited by business challenges

that have a humanitarian dimension.

Assumption # 6 Intellectual excitement is in developed markets. It is hard to find talented managers who want to work at the bottom of the pyramid.

Source: Prahalad & Hart 2002

Furthermore the commercial infrastructure at the BOP needs to be addressed. Buying power needs to be created providing access to credit and generate incomes. Aspirations of the consumers in the BOP need to be created by education and sustainable development. Access needs to be improved by creating distribution systems and setting up communication links. Last but not least the products need to be tailored to local solutions by targeted product development and bottom-up innovation. The authors then proceed with providing a list of arguments why specifically MNEs need to be the ones to take the lead in this struggle. MNEs have the resources available to do so, the possibility of leveraging knowledge from one market to another, MNEs are best suited to build bridges between all involved parties and they can transfer innovative solutions from tier 4 to tier 1 for their own benefit. Furthermore in order to build the organizational infrastructure they have to build a local base of support, conduct R&D focused on the poor, form new alliances, increase the employment intensity and reinvent cost structures (Prahalad & Hart 2002).

The most important critiques on the BOP proposition are by Aneel Karnani who argues that it is too good to be true. He suggests that instead of viewing the poor as consumers, we need to focus on the poor as producers and educate the

poor to buy from the poor. He then goes on showing flaws in Prahalad & Hart’s calculation of the value of the BOP market and criticizes their business cases. He then states that the only way to help the poor and alleviate poverty is to raise their income by lowering the prices of the goods sold to them or raise the income they earn. There are three options to reduce prices, reduce profits and reduce costs with or without reducing quality that are not perfect or economically viable. The only viable option he provides is focusing on the poor as producers via a

combination of microcredit, creating opportunities for steady employment, increasing productivity, involve local governments and create efficient markets (Karnani 2007). More recently Arpita Agnihotri (2014) argues that there is a fortune to be made at the bottom of the pyramid as long as the population is treated as suppliers, producers, co-owners and/or employees instead of just consumers. By setting up co-operations or employee stock ownership plans firms can help poor employees raise their income. When treating the poor solely as consumers, firms have a few options to generate profits while still reducing costs. First they can reduce the poverty premium. Poor people generally have to pay higher prices for basic necessities. Secondly, poor people should avoid the exploitation gap, poor people pay way higher interest rates than the rest of the market. In order to avoid this, microfinance should be available to the BOP. Firms also need to avoid the affordability trap, miscalculating the amount of income poor people can miss for certain products. Last but not least, firms should avoid the adaptability trap, trying to sell products that are not suitable to the environment in which they are sold.

The main reason the BOP prophecy has gained a place here is that one can not simply ignore such an influential and recent development in the field of emerging markets. It has also provided managers around the world with an academic justification for entering or not entering emerging markets and thus should be mentioned here.

Chapter 4: Case studies

This chapter contains three different case studies about MNEs activities on the African continent.

§ 4.1 Methodology § 4.1.1. Research design:

First of all I will do background research on the three companies selected: going through annual reports, newspaper articles and other company documents. Making use of these documents I will set up a table of all the MNEs African presence, stating modes of entry as done by Schuster & Holtbrugge (2012). To answer the research question I will use an embedded multiple case-design in sequential order conducting within- and cross-case analyses with the MNEs being the cases and the units of analysis the markets entered (Yin 2009). This approach lets us link empirical observations with existing theories, helps build empirically valid theories, reveals differences and similarities amongst cases and enables comparisons to clarify if the results are idiosyncratic or consistently replicated in multiple cases (Schuster & Holtbrugge 2012; Eisenhardt 1989; Eisenhardt 2007). This also adds to the construct validity. The within-case analysis will provide insights into the choice between equity and nonequity in different countries while the cross-case analysis will focus on the effect of distance.

§ 4.1.2. Case selection:

The three cases were selected using different criteria. Each case will feature an MNE with its home country in a developed economy. MNEs with a Dutch

subsidiary were preferred. Companies within the Fast Moving Consumer Goods industry were preferred because of their recent interest in expanding their operations in developing economies and more specifically in Sub-Saharan Africa. After careful consideration three MNEs were selected: Heineken, Procter & Gamble and Nestlé. The decision was based upon the MNEs activities in the region,

containing both equity and nonequity subsidiaries. The availability of information on recent regional activities, employees’ willingness to cooperate with interviews and concluding the MNEs had to contain people from the authors personal

§ 4.1.3 Data collection and method of analysis

Getting access to the companies via the official route, calling customer service and being redirected to HR or public affairs has proven difficult. In order to set up interviews use was made of the researchers personal network, contacting friends and acquaintances who occupy, or have occupied, various positions within these companies. For the Nestlé case a different approach was used. Because the company is a decentralized organization subdivided into different geographical regions, people working for Nestlé in the Netherlands normally do not come in contact with African employees or vice versa. In order to obtain interviews employees of Nestlé Eastern African Region (EAR) and Central & West African Region (CWAR) were approached using the professional social media website Linkedin. Interviews were conducted in person where possible, via Skype, Lync or in certain cases when the interviewees proved to be too busy via email. In some cases introductions were made to people in higher positions or positions that were more closely related to the subject of this study.

The aim of this thesis was to conduct at least three semi-structured interviews with employees of each the MNEs listed above. The questions were formulated after the initial research on the MNEs activities in the low HDI markets, specific questions for specific countries and entry modes will be assigned

depending on the results. The topics covered mostly stuck to distance, emerging markets, the hierarchy in the decision making process and the choices between equity and non equity. I made use of open questions (for example starting with why and how). When taking into account distance, while preparing for the interviews, use of Yeganeh (2014) tables of cultural distance was made. All interviews, except interviews via email, were recorded, after asking permission. The employees interviewed included a wide range of different employees: everything from former interns, brand managers, to supply chain managers, HR specialists, managing directors and a regional CFO. The nationalities of the interviewees were Dutch, French, Zimbabwean, Swiss and Italian. The

interviewees were posted in the Netherlands, Taiwan, Zimbabwe, Cameroon, Switzerland, United Arab Emirates, Ghana and Kenya. The names and specific positions are withheld from this study for privacy reasons. A request a few interviewees made in order to cooperate with the research. After conducting the

interviews they were transcribed for research purposes. To further analyse the information obtained from the interviews they were uploaded together with the secondary sources gathered into the Nvivo software program and coded. The different codes were created by myself between and after conducting the interviews.

Examples of codes used were: - economic factors - administrative factors - distance

- geographical distance - Uppsala model

- Transaction cost model

By conducting interviews and gathering secondary sources I will obtain new information and link this in a way not done previously, thus obtaining new insights and clarifying and testing theory in practice. After conducting the interviews we can answer the research question and thus provide a significant contribution to managerial science via analytic generalization.

Three within-case analyses were conducted to see the different factors of distance playing a role within the different markets on the continent. Afterwards a cross-case analysis was conducted to compare the MNEs strategies against each other.

§ 4.2 The Heineken case:

Heineken is the worlds most international brewer who’s brands are available in 178 countries worldwide. The importance of the Sub Saharan region is showcased by the fact that the Africa Middle East region is responsible for 14.6% of the companies total revenue with an operating profit of 700 million euros (Heineken Annual Report 2014).

In the Heineken case a within-case analysis is made of Heineken’s activities in various countries in Sub-Saharan Africa. Heineken first entered the African market via the acquisition in 1935 of a Belgian holding company by the name of S.A. Internationale de Brasserie (Interbra). Interbra maintained shares in various