Demerged multinational enterprises

(1)

‘Demerged Multinational Enterprises’

Master’s thesis

M.Sc. Business Studies – International Management Supervisor: Dr. Johan Lindeque

Second reader: Mr. Erik Dirksen M.Sc. Student: Wouter Merkestein Student ID: 10625550

Date: 26-06-2014

(2)

(3)

Abstract

This study conceptualizes the demerged multinational enterprise (DMNE) as a unique firm type within the internationalization literature, explaining their strategic responses, internationalization patterns and post-demerger performance. Four DMNEs from a diverse range of sectors and backgrounds have been theoretically sampled and studied. Due to the lack of explaining literature, the working propositions are based on a theoretical framework built upon a wide variety of proven internationalization related theories and firm types.

Based on a qualitative multiple-case study design, five years of post-demerger data has been gathered on four theoretically sampled firms. The primary data sources were annual reports and newspaper articles that were further reinforced with quantitative annual accounts and information from databases.

The results have shown that demerged multinational enterprises are unique firms with a great variety of post-demerger strengths, weaknesses and performance difficulties after demerging. Optimal strategic responses of these DMNEs have been found to be affected by previous history, dependency on their parent firm, geographic presence and ambitions.

Keywords: demerger, demerged multinational enterprise, internationalization,

regionalization, firm-specific advantages, dependency, international new venture, born global, born-again global

(4)

2

Acknowledgements

First and foremost I would like to thank Dr. Johan Lindeque for supervising me for the previous months. I felt very privileged to be chosen for this thesis subject and I have gotten many jealous messages from fellow students about me having this subject and Johan as a supervisor, which says a lot. Johan has been very patient with me and helped me through some rough patches along the way. His kindness combined with constructive feedback have been very valuable to me and made the whole process of writing this thesis a pleasant journey with a great result

I would also like to thank my fellow students for supporting me and helping me out whenever I had any questions or whenever I was in need of some support. My time at the Amsterdam Business School has led to many new friends and it gave access to a great network. Lastly I would like to thank my family and friends for their patience and support.

(5)

3.3.1 Construct validity ... 32 3.3.2 Internal validity ... 33 3.3.3 External validity ... 33 3.3.4 Reliability ... 34 3.4 Case selection ... 34 3.5 Data collection ... 36 3.6 Analytical strategy ... 37 4. Results ... 39 4.1 Within-case analysis ... 39 4.1.1 Orica Ltd. ... 39 4.1.2 Wincanton Plc. ... 47 4.1.3 Aviat Networks ... 54 4.1.4 Alumina Ltd... 59

(7)

4.2 Cross-case analysis ... 64

4.2.1 Dependency and FSA retention ... 64

4.2.2 Strategic orientation and governance dynamics ... 66

4.2.3 Pattern of internationalization - geographic scope ... 69

4.2.4 Pattern of internationalization - rhythm and pace ... 70

4.2.5 Theoretical fit ... 71

5. Discussion ... 72

5.1 Dependency and FSA retention ... 73

5.2 Strategic orientation and governance dynamics ... 74

5.3 Pattern of internationalization – geographic scope ... 75

5.4 Pattern of internationalization – rhythm and pace ... 77

5.5 Conceptualization ... 78

6. Conclusion ... 80

6.1 Scientific relevance and managerial implications ... 81

6.2 Limitations ... 82

6.3 Suggestions for future research ... 82

7. References ... 83

7.1 Annual reports and newspaper articles ... 89

(8)

6

Index of Tables and Figures

Table 1 Firm conceptualization ... 12

Table 2 Case selection characteristics ... 35

Table 3 Within-case analysis Orica Ltd. ... 44

Table 4 Within-case analysis Wincanton Plc. ... 51

Table 5 Within-case analysis Aviat Networks ... 57

Table 6 Within-case analysis Alumina Ltd. ... 62

Table 7 Cross-case analysis per topic of interest... 68

(9)

1. Introduction

For decades, internationalization has been a major research topic for international business (IB), strategic management and entrepreneurial articles involving firm types, processes of internationalization and performance influences. When it comes to internationalization a few topics are widely studied, like traditional MNEs that follow steps as described in the internationalization process theory (Johanson and Vahlne, 1977), but also Born Globals (BGs) (Knight and Cavusgil, 2004, Madsen and Servais, 1997, Rennie, 1993) and International New Ventures (INVs) (Oviatt and McDougall, 1994, Oviatt and McDougall, 1997). However, a great deal remains unclear when it comes to demerged multinational enterprises (DMNEs), especially looking at their positioning in IB literature. Demergers are expected to have great influence on organizations, stakeholders and shareholders due to the significance of change it has on both parent firm and DMNE. In this thesis the following research gap will be addressed: ‘Bringing

together the (early) internationalization literature and other relevant IB topics to conceptualize the understudied topic of DMNEs’.

Since DMNEs lack theoretical literature coverage it is important to have a strong theoretical framework build on similar theories to guide the research and its working propositions. Because these firms start off as a separate entity, it is interesting to find in what way they are similar to traditional MNEs that internationalize according to the internationalization process theory (Johanson and Vahlne, 1977) or whether they are more like INVs (Oviatt and McDougall, 1994) and BGs (Knight and Cavusgil, 2004). To truly understand what DMNEs look like it is important to study them while using common IB language and proven theories. The conceptual background will provide key internationalization dimensions, like scope of internationalization (Aggarwal et al., 2011), but also regionalization (Rugman and Verbeke, 2004) and theory on

(10)

8

mergers (Boateng et al., 2008) will provide information to which DMNEs will be related. The internationalization process theory (Johanson and Vahlne, 1977) will form the basis of the conceptual background with its traditional path dependence. In this theory, firms are expected to expand in a risk avoiding pace to countries similar to their home-country (Johanson and Vahlne, 1990). The wide variety of literature is used to justify the working propositions, which are on the topics ‘dependency and FSA retention’, ‘strategic orientation and governance dynamics’, ‘breadth and depth of geographic scope’ and ‘rhythm and pace of expansion’. The desire to conceptualize DMNEs based on their fit in existing literature can be seen by the research question, which is: ‘How do firms internationalize and perform after demerging?’.

Due to the nature of the study and the research question this study will be done as a post-positivist qualitative multiple-case study (Eisenhardt, 1989, Yin, 2003). Due to the specific nature of the study, sample firms and data availability was not extensive. Four DMNEs have been chosen based on their theoretical importance (Eisenhardt and Graebner, 2007) and five years of post-demerger data has been gathered from annual reports and newspaper articles, which was further strengthened by quantitative data from annual accounts and databases. The four quality criteria by Yin (2003), which are construct validity, internal validity, external validity and reliability, will be used to enhance the credibility (Tight, 2010).

Findings have shown that DMNEs do not always fit a specific theory due to their great variety of strengths and weaknesses. Whereas some demerge with a great loss of FSAs and broad geographic presence, another would demerge without any major FSA loss and a home-region orientation (Rugman and Verbeke, 2004). The path of internationalization that DMNEs take after demerging greatly varies on factors that influence the DMNE, like their previous history as an independent firm, dependency on their parent firm and geographic presence upon

(11)

demerging. A new finding is the effect of voluntarily versus forced change. DMNEs that were forced into demerging have shown completely different post-demerger strategies and performance.

This study adds a new characteristic to internationalization literature, which is a valuable addition to the discussion about using theories to define firms with unique characteristics. The results have shown that IB literature needs to be further extended, which is initiated by this study. Managers should realize that one size does not fit all when deciding on what to do after demerging. It is important to know their previous and current dependency, FSA losses, financial strength, geographic presence and ambitions before deciding on a post-demerger strategy. Besides this, industries, firm size and location can also influence post-demerger performance.

The structure of this study is adjusted to the broad literature coverage needed for this understudied topic. First of all the conceptual background will provide important literature and firm conceptualizations. This section will first show a firm conceptualization table built upon current IB literature that can be used as reference while reading the rest of the chapter. This literature, will give a broad overview of relevant theories. After this the DMNE will be conceptualized further based on this literature and the working propositions will be formed. The second section will be that of the methodology. This chapter will contain the research philosophy, choice for a qualitative case study, quality criteria, case selection, data collection and analytical strategy. The result is split up in a within-case analysis and cross-case analysis. In the fourth section the discussion will sum up the results and link them to the conceptual background to answer the WPs and research question. Lastly the conclusion will summarize the findings and address the scientific relevance, managerial implications, limitations and suggestions for future research.

(12)

10

2. Conceptual background

The conceptual background is specifically structured to give a clear and guiding overview of the broad range of internationalization related theories. Before getting into the specific theories, this chapter will start off with Table 1 Firm conceptualization showing the firm conceptualizations. This table gives a brief overview on internationalization literature and how DMNEs would fit into this literature. The theoretical base will be formed in the second paragraph by the traditional internationalization process theory with an emphasis on key internationalization dimensions. After this, INV and BG literature will be discussed building upon the first internationalization chapter and with references to its internationalization dimensions. The fourth paragraph will be about the focal firm in this study; the DMNE. It will combine the literature available with conceptualizations based on the conceptual material gathered in the previous paragraphs, which will lead to the working propositions of this study.

(13)

2.1 Firm conceptualization

D-MNE

Established literature

MNE INV BG

Degree of Internationalization

Performance effects The degree of dependency on its parent firm will influence the DMNE and the degree to which changes in performance are expected after the demerger.

Although diversification and expansion are expected to lead to greater performance, the actual performance increase of internationalization is said to be context dependent and depends on how a firm is able to cope with the additional costs of expansion (Bausch and Krist, 2007).

Internationalization is essential for INV performance and is expected to influence this positively (Oviatt and McDougall, 1994).

Internationalization is essential for BG performance and is expected to influence this positively (Knight and Cavusgil, 2004).

Breadth of geographic scope

DMNEs their geographic scope depends on their retained FSAs, ambitions and international presence.

Traditional MNEs are expected to, at least at first, expand into just a few countries avoiding too much risk (Johanson & Vahlne, 1977).

INVs internationalize in as many countries as possible, this is what they gain their main strength from (Han, 2008, Zettinig and Benson-Rea, 2008).

Although BGs expand into a lot of countries as well, they tend to focus mostly on expected core markets (Knight and Cavusgil, 2004).

Depth of geographic weight

The depth of international expansion depends on whether the DMNE wants to reach new markets or if it just wants to strengthen its position and its commitment to do so.

The depth of expansion by traditional MNEs varies based on things like their strategy, product and industry (Rugman et al., 2011).

INVs have a lightweight (non-equity) network related expansion and usually try to minimize internalization of processes so they can stay flexible (Han, 2008, Zettinig and Benson-Rea, 2008).

BGs are similar to INVs in a way that they try not to internalize processes so they can stay flexible. Their expansion is usually less committed and non-equity as well (Bell et al., 2003).

Measurement DMNE international presence can be expected to be measured in the same manner as traditional MNEs.

There are many measurement indicators for traditional MNEs, the mostly used ones are the foreign sales compared to the total amount of sales (%), the ratio of regional sales to total sales, foreign assets compared to the total amount of assets (%), the number of foreign subsidiaries and foreign profits compared to the total amount of profit (%) (Aggarwal et al., 2011, Rugman and Oh, 2010, Sullivan, 1994).

For INVs it is important to look at the value chain activities abroad and the amount of foreign markets entered (Oviatt and McDougall, 2004).

For BGs it is important to look at the sales growth and the value chain activities abroad (Knight and Cavusgil, 2004).

(14)

12

D-MNE

Established literature

MNE INV BG

Process of internationalization

Pattern The rhythm and pace depend on the retained FSAs and ambitions of the DMNE.

The rhythm and pace of internationalization for traditional MNEs varies based on their strategy, competition, product and industry. However, usually it is a gradual process (Johanson & Vahlne, 1977).

INVs internationalize early and into a lot of countries at once so their rhythm is irregular but their pace is high (Han, 2008, Zettinig and Benson-Rea, 2008).

BGs internationalize early and into a lot of core markets at once so their rhythm is irregular but their pace is high (Knight and Cavusgil, 2004).

Timing Timing of internationalization depends on how competitive the industry is.

Traditional MNEs are expected to expand step by step into closely related countries until they feel like they’re strong enough to expand to the next country, which is more distant from its home-country (Johanson & Vahlne, 1977).

Although different authors used different amounts of years, usually an INV is defined within a time horizon of 3 years after inception (Madsen, 2013).

For BGs there seems to be a more widely agreed upon time horizon of its internationalization within 3 years after inception (Madsen, 2013).

(15)

2.2 Internationalization

For many years now the Uppsala model (Johanson and Vahlne, 1977) has been very influential in entrepreneurial and IB literature by describing the process of internationalization. In their model, Johanson and Vahlne (1977) talk about how firms internationalize by following certain stages. The model describes how firms internationalize depending on the knowledge that they hold about specific foreign markets.

Because firms are expected to start off with little knowledge, they are also expected to go to psychically closer countries to minimize uncertainty. Things like language and cultural barriers combined with transportation difficulties have long been influential when it came to internationalizing (Oviatt and McDougall, 1994). Firms have to think about the trade-off between exploiting their firm-specific advantages (FSAs) and the risk they take by doing business in a country that they do not know well. Building on their experience with other countries abroad, firms slowly move towards countries that differ a lot from their home-country (Oviatt and McDougall, 2004). Many firms today still internationalize slowly following steps similar to the ones Johanson and Vahlne (1977) described many years ago, however, there is a different kind of firm that seems to be skipping these steps. These firms are often defined as International New Ventures (Oviatt and McDougall, 1994) or Born Globals (Rennie, 1993), which will be described later on this chapter.

Besides the process of internationalization, there is also a more FSA kind of advantage to internationalization describing monopolistic advantages (Hymer, 1976). Hymer (1976) says that firms need specific FSAs that help in imperfect markets by giving them monopolistic advantages, which are needed to outweigh their LOF. Product differentiation, having a strong brand name, a lot of assets, access to capital or economies of scale are the usual FSAs that allow firms to compete in foreign countries (Rugman et al., 2011). The recombination capability of

(16)

partially non-location-bound FSAs by a firm can also be of influence when overcoming LOF. The recombination of FSAs is more than just transferring them abroad, it’s about using partially transferrable FSAs and host-country advantages to overcome LOF in host-countries. It’s about the development of new and sometimes innovative capabilities based on linking current knowledge to differentiated local needs in host-regions (Luo, 2002). This way FSAs and CSAs get combined to enhance flexibility and create new opportunities.

When it comes to internationalization as a whole, there are certain choices to be made by a firm that will influence its future. Because of technological innovations firms are now able to differentiate more easily when it comes to degree, scope, timing and pace of internationalization (Cesinger et al., 2012).

2.2.1 Degree of internationalization and MNE performance

The degree of internationalization (DOI) has been widely studied, but the results have not always been consistent (Lu and Beamish, 2004). The DOI literature can be divided into several research interests like the (1) breadth of internationalization, (2) depth of internationalization into countries, (3) ways of measuring DOI and some that (4) describe firm performance due to internationalization in general (Aggarwal et al., 2011).

When measuring DOI, many researchers have used different internationalization related indicators. Widely used indicators to measure the DOI are the foreign sales compared to the total amount of sales (%), the ratio of regional sales to total sales, foreign assets compared to the total amount of assets (%), the number of foreign subsidiaries and foreign profits compared to the total amount of profit (%) (Aggarwal et al., 2011, Rugman and Oh, 2010, Sullivan, 1994).

When looking at the performance of firms it becomes clear that there is little consensus among scholars about the exact influence of the DOI on performance (Aggarwal et al., 2011, Hennart, 2007). Based on the arguments made by Markowitz (1952), internationalization is often

(17)

expected to lower risks due to its diversification advantages (Hennart, 2007). Internationalization is also often related to firm profitability (Brouthers et al., 2000, Grant, 1987, Kotabe et al., 2002). The increased profitability and reduced risks are expected to be realized due to the creation of economies of scale, learning effects and improved access to resources (Bartlett and Ghoshal, 1999, Contractor, 2007, Hennart, 2007). The actual performance increase however is context dependent (Bausch and Krist, 2007) and depends on how a firm is able to cope with the additional costs of expansion (Vermeulen and Barkema, 2002). Diversifying too much might actually influence performance negatively because of its complexity (Hitt et al., 1997, Lu and Beamish, 2004). Due to the lack of uniformity and mixed results it would be hard to have performance related conclusions (Aggarwal et al., 2011, Hennart, 2007).

2.2.2 Geographic scope of internationalization

Some firms have experienced a boost in performance due to expansion of their geographic scope (Goerzen and Beamish, 2003, Goerzen and Asmussen, 2007). This geographic scope, which can be split up in breadth and depth of scope, is often used by firms to gain what Porter (1993) calls a competitive advantage.

2.2.2.1 Breadth of geographic scope

The so called breadth (Aggarwal et al., 2011) of geographic scope is based on the amount of countries or regions that a firm internationalizes into. By looking at certain countries or regions, important aspects from internationalization process theory, like the path of expansion can be measured (Johanson and Vahlne, 1977). As mentioned before, the breadth of geographic scope is traditionally viewed as something that widens slowly in a risk avoiding pace by going to closely positioned countries (Johanson and Vahlne, 1977).

(18)

2.2.2.2 Depth of geographic scope

The depth (Aggarwal et al., 2011) or weight of internationalization refers to the amount of activities in a host-country and so its commitment. The weight of internationalization can be measured by looking at used entry modes, whereas non-equity modes (Pan and Tse, 2000) like exporting and franchising are considered to be lightweight expansion, and wholly owned subsidiaries are considered to be heavyweight expansion. An example firm with a strategy that focuses on the depth of expansion is an emerging market MNE (EM-MNE). These EM-MNEs are special in a way that they originate in emerging markets but focus heavily on outward FDI into developed markets. They do so to leapfrog stages to lower their institutional constraints and gain a lot of knowledge that their home-country might lack (Luo and Tung, 2007). Since they need knowledge that they do not necessarily poses, their expansion is usually equity-based via acquisitions (Pan and Tse, 2000).

2.2.2.3 Regionalization

With the concepts of regionalization (Rugman and Verbeke, 2004) and semiglobalization (Ghemawat, 2003, Ghemawat, 2005) there has been another addition to the understanding of internationalization. The regionalization literature states that most so called global firms are not that global but usually home-region bound, both in geographic scope of breadth and depth (Rugman and Verbeke, 2004). Most firms choose to stay close to their home-region due to the earlier performance reason that diversification can become too complex and hard to handle, which Rugman and Verbeke (2007) call the liability of inter-regional foreignness. According to Rugman and Verbeke (2004), a true global firm has 20% - 50% of its sales in the extended triad regions North America, Europe, Asia Pacific, Latin America, Middle East and Africa (Rugman, 2005). Firms that have more than 50% of their sales in a region are called host-region oriented, if they have between 20% - 50% in their home-host-region and more than 20% in a

(19)

host-region they are called bi-regional and if none of these options are applicable the firm is home-region oriented (Rugman and Verbeke, 2004). The reason for regionalization is expected to lie in the physical distance that has to be overcome by firms (Johanson and Vahlne, 1977). Due to this physical distance firms have to leverage on their country-specific advantages (CSAs) and FSAs, which usually causes them to end up internationalizing within the same region at first to minimalize risks but still enhance performance (Rugman and Oh, 2010). This shows that not all FSAs are equally transferable across international borders.

2.2.3 Pattern of internationalization

It is important for firms to realize how their internationalization patterns can influence the actual performance gains from having subsidiaries abroad. Vermeulen and Barkema (2002) talk about patterns of internationalization by looking at the rhythm and pace. The rhythm and pace are mainly about the choices that have to be made regarding how regularly and how fast the internationalization takes place. Some expansion patterns seem to boost performance stronger than others, which shows that it is important for firms to take the pattern of internationalization into account. Research shows that high speed and irregular expansion can negatively influence firm performance gains abroad (Vermeulen and Barkema, 2002) because of the creation of diseconomies of time compression (Dierickx and Cool, 1989). Diseconomies of time compression exist because firms are not always able to absorb all the information they gain from expansion and that they lack time to optimize their learning curve (Cohen and Levinthal, 1990). Firms need to be aware of the trade-offs when internationalizing and they have to prioritize the internationalization dimensions in a balanced way (Vermeulen and Barkema, 2002).

(20)

2.2.4 Timing of internationalization

It is important to realize why firms choose a certain timing of internationalization. There are pros and cons that have to be taken into account when deciding on the timing of internationalization. Firms have to think about their strategy and about the risks involved with each option. In their article about first mover advantages (FMAs), Lieberman and Montgomery (1988) talk about the reasons that firms might have to expand either early or late. With FMAs like the experience and learning curve that will build up, the preemption of assets and relationship building with customers, firms that internationalize early try to gain a solid market position. However, there are also late mover advantages (LMAs) to gain. By entering a market later on, firms can gain LMAs like free rider effects and incumbent inertia (Doh, 2000, Lieberman and Montgomery, 1988).

2.3 Relevant internationalizing MNE types

The conceptual background is focused on giving insights into internationalization patterns so these can be compared with those of a demerged firm. For this it is important to specify the relevant firms, their corresponding internationalization patterns and in what ways they try to gain strength from internationalization.

When looking at all the INV and BG related literature, it can be said that these early internationalizing firms are no longer understudied. Some argue that INVs and BGs are not a special breed of firms as they are sometimes considered but just new firms with an international focus (Madsen and Servais, 1997, Phan and Fan, 2007). They can be seen as a Small and Medium Enterprise (SME) with a global vision to gain fast growth (Gabrielsson et al., 2008). Based on this conceptual background it can be said that there is a broad interest when it comes to internationalization, with INVs and BGs in specific. Another interesting topic is the one of born-again globals by Bell et al. (2001). In their article they show that the early

(21)

internationalization theory are still not completely comprehensive and that extensions can be valuable. Born-again globals have some similarities with DMNEs in a way that they internationalize rapidly after not having internationalized for a longer period.

2.3.1 Traditional MNE

First of all it is important to clarify what type of firm will be used to compare special firm types with. The so called traditional MNE is one that internationalizes by somewhat following the patterns described in the Uppsala model by Johanson & Vahlne (1977). Since the theory itself has already been explained earlier this chapter, this firm type will only be explained briefly. Based on the Uppsala model, traditional MNEs are expected to go through certain stages of internationalization. These stages are dependent on the trade-off between the risks of expansion into new countries and the advantages gained by exploiting FSAs abroad. Based on this trade-off it is expected that these traditional MNEs will expand into countries with a similar environment and closely related CSAs (Rugman et al., 2011). Step-by-step they learn how to overcome their liability of foreignness (Zaheer, 1995), which gives them a relatively risk avoiding low pace of internationalization with a narrow geographic scope.

2.3.2 International New Venture (INV)

Oviatt and McDougall (1994) are among the first to talk about INVs and broadly define them as firms that are international from inception and leverage on this, meaning that they seem to skip the earlier mentioned internationalization stages and are able to go to virtually every country relatively fast and gain strength by doing so. Looking at the internationalization dimensions, INVs gain strength from a broad geographic scope, high pace and early timing of internationalization. There are several reasons for the emergence of these INVs. A lot of things have changed since the 1970s; there has been the emergence of more flexible governance models for firms which makes it easier to leverage their FSAs abroad, more and more people

(22)

have international experience, consumer preferences became more alike, international alliances and trade agreements are formed, cost and time of travel and communication have gone down and the information flows between countries have become richer (Autio, 2004, Knight and Cavusgil, 2004). By defining INVs, Oviatt and McDougall (1994) somewhat answered the critiques that many scholars had on the original internationalization stages theory (Johanson and Vahlne, 1977) and give room for a new way of strategic thinking about firms when it comes to internationalization (Zettinig and Benson-Rea, 2008). INVs do not only go against traditional internationalization process theory, but they also go against Hymer’s (1976) thoughts about scale. For INVs it’s not about competing at a broad and large scale but about being competitive by gaining strength from a narrow focus with unique FSAs.

There has been a lot of critique on how this internationalization process theory lacked in a variety of possible strategies that firms could have, the focus on large firms and on aspects like experience and uncertainty avoidance. Firms do not always have standardized paths, their path dependence usually depends on many different firm characteristics like their product nature and environment (Aspelund et al., 2007).

It is important to define the characteristics of these INVs. There have been many articles that either tested or helped conceptualizing INVs (Di Gregorio et al., 2008, Fernhaber et al., 2007, Hewerdine and Welch, 2012, Oviatt and McDougall, 2004, Oviatt and McDougall, 1994). In general, an INV is described as a relatively young firm that has few assets and that gains most of its strength in countries other than its home-country. They usually hold strong or unique FSAs that could give them FMAs when entering these foreign countries. Through networking INVs manage to gain knowledge about possible opportunities that they can exploit. Because they do not own a lot of tangible resources themselves they manage to be flexible and fast in their way of doing business, which can give them an advantage over other firms (Han, 2008, Zettinig and Benson-Rea, 2008).

(23)

Oviatt and McDougall (1994) identify four features that distinguish an INV from other MNEs, namely, (1) ownership of some assets that can be used for economic transactions, so they must have something of value, (2) the use of different structures of governance to cope with the lack of internalized assets, (3) they must hold locational advantages abroad to overcome local obstacles, and lastly (4) there must be something unique about what a firm is doing or selling compared to what other firms can do in order to keep a sustainable advantage (Oviatt and McDougall, 2004).

Since the publication by Oviatt and McDougall (1994), a lot of research has been done on INVs (Zahra, 2005). A regularly studied aspect when it comes to INVs is the influence of the founder(s) of these firms on their performance and decision-making (Evangelista, 2005, Ghannad, 2013, Kropp et al., 2008, Spence and Crick, 2009, Zahra et al., 2000). The founder(s) can be seen as some sort of embodiment of intangible FSAs that vary in strength based on educational background, connections and experience in similar industries or abroad seem to be of great influence when forming an INV. These aspects can give the founder(s) a head start when it comes to identifying and making use of market opportunities, which is of great value for the firm itself (Evangelista, 2005). Ghannad (2013) however questions this and claims it has to do with the whole lifespan of experiences that someone has including their childhood due to the great influence this has on the eventual entrepreneurship and creation of an INV.

2.3.3 Born Global (BG)

Besides INVs, there is another breed of firms that has been challenging traditional internationalization process theory. BGs are somewhat similar to INVs since they also have a high pace of internationalization soon after inception to gain FMAs (Lieberman and Montgomery, 1988) and they often have a relatively broad geographic scope. Rennie (1993) was among the first to define these firms, which he called Born Globals. What Rennie (1993)

(24)

saw was how relatively small Australian firms were able to compete with large MNEs on the global market within just a few years after they originated. The reasons that are described for the emergence of these BGs are similar to the ones mentioned about INVs. Changing consumer needs, new technologies, communication advancements and the need for firms to adapt quickly have given way for these BGs to emerge. These BGs are usually characterized by strategies and business cultures that stimulate innovation and accommodate a fast pace of internationalization (Knight and Cavusgil, 2004). Starting to internationalize soon after inception is seen as a FMA since the longer a firm waits the harder it will be to actually internationalize in ways BGs can (Sharma and Blomstermo, 2003). Intangible FSAs like the network a BG is part of are essential for their success, either by information exchange or working together they get unique market insights (Han, 2008). It is very unlikely that a BG invests time and money in market research. What is interesting about these networks is that BGs purposely do not invest too much in these relations because they need to stay flexible and able to move on whenever needed (Sharma and Blomstermo, 2003).

This all seems rather similar to INVs, however there is a difference. INVs are known to have a specific focus that is narrow but across many countries whereas BGs are determined to exploit their FSAs in expected core countries. This narrow focus is why INVs as described in the previous paragraph have a lot of flexibility, they are ‘lightweight’ firms that can pull out of certain markets relatively easy (Bell et al., 2003). A clear distinction between INVs and BGs is given by Madsen (2013), saying that BGs focus on the quantity of international sales by somewhat adapting their depth of internationalization to the market and INVs focus more on a broad scope of their international sales and really do their own thing.

Another interesting addition to the BG literature is the one that Bell et al. (2001) calls born-again global. These firms are those that still internationalize rapidly but do not do this right from inception. Born-again globals are usually very strong embedded in their home-market

(25)

which causes them to have less ambitions to internationalize (Graves and Thomas, 2008, Jantunen et al., 2008). Shrinking markets, following competitors or a management buyout are usually the causes for a turnaround within these firms. Most of the times it’s not a single reason but a combination of several ‘critical incidents’ that causes firms to rapidly internationalize years after inceptions (Bell et al., 2001). These born-again globals usually suffered from constraints like the lack of finances, experiences, foreign listings, patents and a strong international brand name which caused them not to internationalize. In many cases these constraints can even lead to the desire to get acquired or to merge with another firm that can provide the necessary resources to internationalize further (Bell et al., 2001, Bell et al., 2003). Another option is that of a subsidiary getting released from its local restrictions which gives the possibility for a further internationalized scope (Bell et al., 2001). The early internationalizing firms require 25% of foreign sales within the first three years after inception (Madsen, 2013). Born-again globals have to start their internationalization after at least three years, and then from the moment they start internationalizing they should gain the same 25% of foreign sales within three years.(Jantunen et al., 2008). It can even be argued that these INVs and BGs are not specific forms of organizations, but merely firms with a specific strategy to gain strength from rapid internationalization. This strategy can also be used later in time by different firm types to gain similar results (Bell et al., 2001).

Having discussed all these possibly relevant internationalization related firm types there is now enough background information to move on to the focal MNE type, which is the DMNE.

2.4 Demerged MNE (DMNE)

IB literature mainly focusses on mergers and acquisitions as entry modes, which are associated with firm growth (Boateng et al., 2008). However, demergers are also very interesting because of their influence on organizations, stakeholders and shareholders. A reasonable and mainly

(26)

unanswered question is: what do these DMNEs look like? Whereas mergers focus on acquiring FSAs, an interesting question is whether DMNEs are able to retain key FSAs like intangible knowledge after demerging. The only demerger related research involves performance changes or financial advantages from the parent firm perspective, not what happens with the DMNE. The abbreviation ‘DMNE’ adopted in this study has sporadically been used to describe other firm types like developed-country multinational enterprises (Govindarajan and Ramamurti, 2011) or dummy separating multinational enterprise affiliates (Kumar and Agarwal, 2006). For this study it will be used as the indicator for a demerged multinational enterprise (DMNE).

Demergers come in a variety of types and there are several reasons for firms to demerge (Singh

et al., 2009). Firstly, it can be done because the structure of the firm has become too expensive,

so the firm decides to remove certain activities that have a negative effect on firm performance. Secondly, it can be done to simply improve the organization of the firm and make room for a changing strategic focus. Thirdly, a demerger can be done to open up to the capital market and gain more and more specific capital for further expansion. And lastly, demergers happen to improve governance and give management a clearer role.

A demerger can be seen as a separation of a part of an existing firm into a new separate firm. By doing so firms can go back to their core business and narrow down their focus. The DMNE will become its own entity separate from the demerging firm (Singh et al., 2009). However a DMNE is not always completely independent, especially not right after demerging. What is really unique about DMNEs is that sometimes the parent firm still has a role in post-demerger processes (Hoare and Cartwright, 1997).

DMNEs are unique in a way that they do have a history but on the other hand they form a new entity with possible new strategies, new top management and more freedom. It is interesting to find out whether this history is crucial for post-demerger performance in a way like INV theory

(27)

describes the relevance of founders their lifespan (Ghannad, 2013). Reasons for mergers are usually to gain knowledge, to overcome liabilities or to expand rapidly (Boateng et al., 2008).

One of the most known demergers has been the one of Britain’s large manufacturing firm Imperial Chemistry Industries (ICI) out of which Zeneca demerged which instantly became one of the biggest biochemical firms. ICI was an example of a firm that became too big and too diversified so it became too hard to manage everything in one firm. Besides this, the department that later became Zeneca simply outgrew ICI (Kennedy, 1993, Owen and Harrison, 1995). The ICI demerger has proven that a demerger can be of high additional value for both businesses by enhancing responsiveness and narrowing foci.

2.4.1 Demerged MNE conceptualization

The DMNE is a relatively under conceptualized firm type, however drawing on the previous discussed literature about mergers and the internationalization of the three different firm types, further conceptualization of DMNEs is possible. Although all of the previous discussed dimensions and strategic decisions are relevant for DMNEs, there are certain key elements that are expected to most strongly define DMNEs.

2.4.1.1 Dependency and FSA retention

The difficulty in exactly conceptualizing DMNEs lies in the fact that every DMNE starts off differently. The changes in performance after demerging will highly depend on the degree of dependency of the DMNE within the parent firm. Dependency is a widely studied IB subject with regard to parent firm and subsidiary relationships (Birkinshaw et al., 1998, Monteiro et

al., 2008). Previous dependency on the parent firm is expected to be key in post-demerger

performance and will most likely influence decision-making. If a firm had already been performing individually within its parent firm, less change can be expected after demerging.

(28)

The degree of dependency is key in describing how FSAs were either retained or if they had to be regained. Had the DMNE been a relatively standalone subsidiary, the post-demerger process is expected to be less difficult. However, had the firm experienced significant dependencies within the parent firm on aspects like R&D and intangible FSAs such as brand name, the demerger will require additional adaptation and financial support. Drawn upon the early internationalization theories, intangible FSAs like the brand name and network are expected to be essential for post-demerger success (Han, 2008). These FSAs are seen as critical for MNE competitiveness so regaining them is expected to have a high priority (Rugman et al., 2011). It is important to realize until what extent the DMNE could still retain FSAs like brand name etcetera for their products. Depending on its loss of FSAs, the DMNE could choose an aggressive approach to quickly regain these FSAs like EM-MNEs do when expanding abroad (Luo and Tung, 2007).

WP1a DMNEs that heavily depended on their parent firm are more likely to suffer from the loss of key FSAs.

WP1b DMNEs that lose control of important intangible/tangible FSAs are expected to experience performance difficulties after demerging.

2.4.1.2 Strategic orientation and governance dynamics

The degree of change in strategic orientation and governance dynamics is also related to the dependency that the DMNE used to have. If the DMNE used to be a relatively isolated (Monteiro et al., 2008) part within the parent firm, the strategic change after demerging is expected to be minimal. In return for the potential FSA loss, a DMNE has more freedom since it is no longer restricted by corporate strategy so it could take on new challenges and seek for new opportunities that it might not have been able to while still being a part of the parent firm (Singh et al., 2009). A different and smaller form of governance is seen as one of the key FSAs

(29)

for early internationalizing firms (Oviatt and McDougall, 2004), this might also go for DMNEs with similar characteristics.

WP2a DMNEs have improved quality of governance due to the fact that they become smaller and gain more control.

WP2b DMNEs tend to take on a lot more challenges and opportunities because they are no longer restricted by corporate strategy.

2.4.1.3 Pattern of internationalization – geographic scope

Another effect of the demerger is that the DMNE might be forced to reconsider sustaining its international presence. Since the DMNE is no longer a part of the larger parent firm, it can be expected that it chooses to expand with a broad geographic scope and a heavyweight entry to strengthen its position in the market and avoid being acquired. This pattern of internationalization also depends on how the competitive environment is that the DMNE firm faces, since more competition would strengthen the uncertainty after demerging. It is also important to take notice of the global presence at the time of demerging and its influence to the dynamics of decision-making. High costs and previous investments in some locations might tie the firm down due to its commitment, which gives the DMNE fewer resources to exploit new opportunities. The previously mentioned loss of brand awareness can also lead to a renewed LOF, the firm would have to reestablish itself (Zaheer, 1995).

WP3 DMNEs are expected to adjust their degree of geographic international presence to more accurately reflect competitive dynamics instead of corporate dynamics to (re)gain a solid market position.

WP4 The greater the weight of the DMNEs global presence, the less dynamic initial decision-making and strategic responsiveness is expected to be.

(30)

WP5 The lower the amount of retained FSAs, the higher the renewed liability of foreignness and thus the lower the incentives are to internationalize outside home-region(s).

2.4.1.4 Pattern of internationalization – rhythm and pace

The rhythm and pace of further expansion is also expected to be influenced when demerging. Based on what FSAs were lost and which ones were kept, the firm might face new challenges like the need to fill certain FSA gaps. The need to fill these FSA gaps can force the DMNE to irregular but in a fast pace regain its FSAs by acquiring other firms. The irregular pattern and pace can explained by the fact that they are looking for specific FSAs which will be bought up quickly when the opportunity arises.

WP6 The loss of key FSAs when demerging leads to an irregular highly paced expansion to regain these FSAs.

(31)

3. Methodology

This chapter will describe the research methodology and clarify assumptions and decision-making with regard to the execution of the study to positively influence its credibility (Tight, 2010). First of all the research philosophy of the study will be described in terms of its ontology and epistemology. Next the choice for a qualitative multiple-case study will be motivated, after which the quality criteria for qualitative research and associated design choices are discussed. The fourth section will go through the case selection process. After defining the cases that will be studied, the fifth section will describe how the data for this study will be collected. Lastly the analytical strategy will clarify how the derived data will be used to eventually answer the working propositions and the research question.

3.1 Research philosophy

3.1.1 Ontology

Ontology is often described as the study of existence and its categorized relationships between social entities (Bryman and Bell, 2011, van Willigenburg, 2012). The ontology of research is a philosophical way of thinking about what, how and in what way entities exist. It involves thinking about whether social entities have a completely objective nature or whether they are the result of several perceptions and actions coming together (Bryman and Bell, 2011, Morgan and Smircich, 1980, Saunders et al., 2012). The interest of this study is not necessarily about how people structure their knowledge, which is a subjectivist perspective, but about the knowledge itself of which the researcher is capable of analyzing objectively, this is why this study will adopt a more objectivist perspective (Bryman and Bell, 2011).

(32)

3.1.2 Epistemology

Whereas ontology is the theory of existence, epistemology is the theory of knowledge, it addresses questions involving what we know and especially what we cannot know. Epistemology looks at knowledge and what can be justified as knowledge (van Willigenburg, 2012). Epistemological questions are of great importance because they are questions about the rationality of arguments that are based on theoretical assumptions. By understanding the nature of something, researchers tend to explain a phenomenon and possibly do predications based on these finding (Bryman and Bell, 2011). The epistemology of this study follows a post-positivist approach. This is seen as a critique on positivism, which says that claims have to be based on observable phenomena or experiences with a clear causal relationship (Bryman and Bell, 2011, Saunders et al., 2012, van Willigenburg, 2012). Post-positivists agree with positivist by saying that reality exists, however they also say that research can be influenced by the researcher’s experience and beliefs (Clark, 1998). Knowledge is therefore somewhat biased and not necessarily true, which is somewhat a realist vision saying that what we find only gets us closer to reality, it does not allow reality to be fully captured, which is seen as ontologically independent (van Willigenburg, 2012). This study is focused on getting enough fractions of information that when these are combined there should be a fuller more credible representation of the phenomenon being studied. This means that the findings and resulting knowledge claims will represent the most complete account of the phenomenon as possible given the available data sources and theoretical foundations of the study.

3.2 Qualitative multiple-case study design

When doing social science related research, a case study obviously is not the only option available, the decision for a specific research design has to be made based on several factors that define the nature of the research. Important aspects are the kind of research question, the

(33)

desire for control during the research and whether it is something historical or contemporary (Yin, 2003). When talking about demergers it is important to know what happened and how this happened, for this reason this study will be done by doing a post-positivist qualitative multiple-case study based on the work of Yin (2003) and Eisenhardt (1989) with a phenomenon-driven research question (Eisenhardt and Graebner, 2007). The amount of available data is so narrow that a deductive bottom-up approach is recommended when studying DMNEs (Shepherd and Sutcliffe, 2011). Yin (2003) will provide the deductive and explanatory foundations used in this study but with additional strengthening by a theoretical framework to guide the research, which is more of an inductive design approach mentioned by Eisenhardt (1989). The research question has a broad scope which gives flexibility for the use of theory and justification based on the phenomenon’s importance.

Case studies usually use several data sources to give a thorough empirical description of a specific phenomenon (Yin, 2003). They are excellent for getting a thorough understanding of phenomena by also looking at the context in which they happen (Eisenhardt and Graebner, 2007). Although there is a lack of theory when it comes to DMNEs, in the conceptual background a theoretical framework has been set up based on previous IB literature to give guidance for this study (Eisenhardt, 1989).

As can be seen in the conceptual background, there is a lack of theory explaining demergers and how the DMNEs internationalize afterwards. However, there is other relevant IB literature that can be used and this is why this multiple-case study will adopt an explanatory deductive design. The study remains open to new exploratory insights suggested by the data for establishing new insights and suggesting possibilities for future research; thus trying to broaden ways of thinking and find new insights with regard to demergers but based on existing theory.

(34)

existing early internationalization theories, not necessarily a complete renewal of theories (Jack and Kholeif, 2007). Important for this possible theory extension is to take note of the replication logic implied by the case study’s design (Eisenhardt, 1989). By establishing methodological choices that increase the reliability of the interpretation of the data, it is possible to eventually lead to an objective outcome that is precise and testable (Eisenhardt and Graebner, 2007). However, there has always been a lot of criticism of case studies, especially about biased views and the lack of generalizable results. Due to this reason, the quality criteria adopted by Yin (2003) are essential for the overall quality of the study, this is why the next chapter will be built around these criteria.

3.3 Quality criteria for qualitative multiple-case studies

There are four case study design quality criteria that are essential to enhance the quality of research (Yin, 2003). These four essential quality criteria are construct validity, internal validity, external validity and lastly the reliability. By applying these quality criteria throughout the whole research, the eventual result will be more credible (Yin, 2003). The basis for the quality criteria is the use of triangulation. The concept of triangulation means that there will be more than one data source to enhance the completeness and credibility of certain outcomes (Yeasmin and Rahman, 2012, Yin, 2003). In this case there will be data-, theoretical- and methodological triangulation (Denzin, 1970).

3.3.1 Construct validity

The first quality criteria is that of the construct validity when collecting data; does the measurement tool really measure what it should if it wants to conclude something about a phenomenon (Yin, 2003). When failing to the criteria of construct validity, the research can be seen as subjective with biased judgments, with a loss of credibility as a result (Yin, 2003). This study will have methodological triangulation (Hussein, 2009) to provide different confirming

(35)

data. The primary data source will be annual reports and newspaper articles gathered via LexisNexis Academic. As a secondary data source financial data gathered from annual accounts and databases like Zephyr and Orbis will be used to further strengthen conclusions and enhance the construct validity.

3.3.2 Internal validity

Internal validity is also of importance for explanatory case studies due to the fact that it is about the causal relationship between different concepts (Yin, 2003). The internal validity looks at the correctness of the stated results. The causal relationship needs to be proven by for example independent theory driven working propositions. For this study the working propositions are based on a theoretical framework built from existing IB literature to proof the researcher’s inference and validate causal relationships. The addition of quantitative data to the qualitative data brings the advantage that this data is not based on interpretation by the researcher and provides more than one explanation to a certain outcome (Yin, 2003).

3.3.3 External validity

External validity looks at theoretical generalizations of the results that go beyond the researched case (Bryman and Bell, 2011). Although there is triangulation, it still has to be said that due to the fact that it is a case study, the statistical generalizability is not one of the goals. It is some sort of particularization (Lee et al., 2007) instead of a generalization, which basically accepts the narrow focus and concentrates on a thorough understanding of a single case. This means that the triangulated data will have to be woven into current theory and show its additional importance based on in-depth high quality results. However, by making use of a wide variety of theories and data sources to strengthen the findings there is still a possibility to have an outcome that is precise and testable and generalizable to theory according to the replication logic (Eisenhardt and Graebner, 2007).

(36)

3.3.4 Reliability

Reliability lastly has to do with whether a replication of the study would lead to the same results. This will be enhanced by documenting all the steps that have been made with regard to the case selection, data gathering sources, codebook and data interpretation so others could replicate the research (Yin, 2003).

3.4 Case selection

A thorough case selection is important for the reliability of a qualitative multiple-case study. The cases in this study are chosen based on data availability, firm survival, uniqueness and theoretical sampling (Eisenhardt, 1989). For this study four surviving DMNEs have been chosen, allowing the study of a unique set of firms that are not necessarily representative for the whole population, but sampled based on their theoretical importance (Eisenhardt and Graebner, 2007) and thus allowing theoretical generalization. The cases have been chosen using the Zephyr database fordeal information on demergers. A firm introduction and demerger data can be found in Table 2 Case selection characteristics. The initial data set contained completed demergers of firms that were or are listed on any stock exchange worldwide since 1997. After this initial selection, the amount of firms was reduced by looking at whether they had been acquired within five years after demerging, the ones that did not get acquired within five years continued to the next step, on which research was done about the data availability in newspapers, annual reports and databases for a minimum of five years. Eventually four firms were identified as theoretically relevant and having a sound empirical basis to study, these firms are Alumina, Orica, Aviat Networks and Wincanton. Besides data availability, the uniqueness also plays a role, like how Aviat Networks kept using their parents name for a year or how Alumina is a unique joint-venture demerger. Each case has its own theoretical importance to provide unique insights in certain situations.

(37)

Background information Country Deal type Deal value Demerger date Orica Ltd. Orica Ltd, previously known as ICI Australia, had been an explosives supplier for almost 130 years when

it demerged, both independent and as a part of ICI The demerger, which was basically ICI selling their 62,4% shareholding, was unexpected and somewhat forced due to reorganizations at ICI The uniqueness of this case lies in the long history it has and the deal size of the demerger. With an extensive coverage of both post- and pre-demerger in both qualitative and quantitative data, there is a lot to be said about Orica.

AUS 62,4% demerger

€1,409,349,000.0* 05-02-1998

Wincanton Plc.

Wincanton is a firm that is specialized in providing solutions for supply chains. It demerged as a chilled food distribution firm from the food group Uniq. This demerger was initiated with mutual support with the expectation that both businesses would perform better individually. What is unique about this demerger is that Wincanton got demerged as a rather narrow focused firm, but with the ambitions to steadily grow its business both within its home-region and to new host-regions in Europe.

GB 100%

demerger

€658,450.7 18-05-2001

Aviat Networks

Aviat Networks is a nowadays California based communication network solutions provider with a global presence. What is interesting about this demerger is the post-demerger performance and FSA loss due to the dependency that Aviat Networks had on their parent firm Harris Corporation. The demerger was basically one sided in which Harris Corporation sold their 56% stake in Stratex.

USA 56% demerger €121,525.31 27-05-2009

Alumina Ltd. Alumina used to be part of WMC, a large mining company. They got demerged because they believed they could achieve a better shareholder value when separated. The demerger was mutually agreed upon due to the expected value creation for both firms. The uniqueness about this case is that Alumina’s only business is in a large joint-venture called AWAC. Interesting will be to see how de dependency on the joint-venture influences the dependency on Alumina’s parent firm WMC.

AUS 100% demerger

€2,894,064.39 4-12-2002

Table 2 Case selection characteristics

(38)

3.5 Data collection

The specific topics of interest will be broad since there can be a lot of reasons for internationalization. Getting a thorough description of the internationalization patterns and processes will be key in finding similarities of DMNEs with firm types like INVs and BGs. To find out whether demerging firms can be positioned in early internationalization theories, it is important to focus on similar characteristics. Because of this, most of the research will be about speed, scope, and timing of internationalization, which are the three main pillars for early internationalization theories (Aggarwal et al., 2011, Vermeulen and Barkema, 2002). Other interesting additions are changes in governance, previous dependency, the percentage of foreign sales, years of experience and size of the firm. To answer the research question and cover the research gap, there is a need to study DMNEs that can provide interesting insights based on these characteristics.

A systematic, transparent and proper data collection is essential for the quality of the research (Eisenhardt and Graebner, 2007). As mentioned, both qualitative and quantitative data will be used for this study to enhance the completeness and credibility of the results (Hussein, 2009, Yin, 2003). Case study data can come from several sources, in this case it will come from annual reports, newspaper articles and databases. The strength of documentation, in this case annual reports and newspaper articles as a primary data source is that it’s stable, widely covered and exact (Yin, 2003). On the other hand it is data that can get lost or blocked over time and it can be biased by for example the unidentified author so it should not always be accepted as factual unbiased information (Yin, 2003). These problems can be tackled by getting documentation from trusted newspapers and by finding evidence in other sources too (Yin, 2003). Five years of post-demerger newspaper articles will be derived using LexisNexis Academic while focusing mainly on business related journals with a strong record of publications when coverage is extensive. In Australia for example this would be Australian Financial Review. The search

(39)

terms are the focal firm names to get a wide variety of possibly interesting information. The annual reports will be gathered from the company’s website. The addition of databases like Zephyr and Orbis adds a more precise and quantitative source of data but is not always easy to access (Yin, 2003). Zephyr is a database that focusses on announced, rumored and completed company deals of which it provides both qualitative and quantitative information. Orbis is a large database in scope due to a combination of data sources. It holds 10 years of annual reports for extensive amount of companies worldwide.

3.6 Analytical strategy

After the right data has been collected, it still has to be categorized, examined and combined so statements with regard to the working propositions and research question can be done (Yin, 2003). The qualitative data will be used to find initial information with regard to the conceptual background related topics, to which the quantitative data can provide additional proof for certain findings and statements. Due to the desire to fully capture the understudied phenomenon of DMNEs, analytically this study is based on theoretically deducted themes, but there is a possibility that other themes arise which are inducted from the data after analyzing (Fereday and Muir-Cochrane, 2006).

The primary data that has been collected from newspaper articles and annual reports will be analyzed using NVivo software. The initial dataset contains all the information on the focal firms for the five years after demerging. This dataset will first be scanned for relevant information, after which the relevant information will be uploaded in NVivo. In NVivo thematic coded themes were chosen based on key topics found in the conceptual background and working propositions (Ryan and Bernard, 2003). Certain words, sentences or complete paragraphs can be assigned to the defined thematic codes to give a holistic and manageable overview of the complete data set per topic or working proposition so the data can be used for

(40)

both within-case and cross-case analysis. (Ryan and Bernard, 2003, Welsh, 2002). The higher level codes are related to the four topics around which the working propositions have been formed. Lower level codes go into depth more to specify exactly what part of the high level code is being discussed since they are rather general. Examples of this are the ‘Pattern of internationalization – geographic scope’ code which has its sub-codes like ‘breadth’ and ‘width’, which are also related to the working propositions and conceptual background. The full list of used codes can be found in Appendix A.

The secondary data has been exported into Microsoft Excel from the annual accounts and earlier mentioned databases. This data will be analyzed based on the needs to reinforce statements based on primary data. For example, when acquisitions after demerging are said to have a certain effect on the company, this should be visible in the financial data. Until some extent this also goes the other way around. When financial data shows unusual developments, the primary data can be used to motivate what exactly caused this mutation.

(41)

4. Results

4.1 Within-case analysis

This section will provide the individually addressed results for each of the focal firms. As described in the case selection chapter, every firm has its unique characteristics, which make them interesting for further conceptualizations of DMNEs. Based on the theoretical framework of this study, key topics of interest have been formulated. The relevance of these key topics will be tested by going through the first five years of post-demerger news- and financial coverage. For every firm a summarizing table has been provided with key qualitative data and references to prove certain statements, the addition of quantitative data is gathered from annual reports and databases. The qualitative data will provide the initial background information about the firm to find out what it has gone through and quotes to confirm this. The quantitative data has been selected based on their expected relevance, like the total revenues and total assets in both home- and host-regions and the amount of deals done to cover topics like regionalization and geographic scope.

4.1.1 Orica Ltd.

Throughout the five years after demerging Orica has gone through a lot of changes. Right after demerging they decided to strengthen their core business, which was explosives. As can be seen in Table 3, a stronger focus and broader reach across the globe positively influenced Orica’s post-demerger performance.

Dependency and FSA retention

Due to over-diversification and financial struggles, ICI demerged their pharmaceutical business (Kennedy, 1993, Owen and Harrison, 1995) and their explosives business, which was called ICI Australia (Orica 1998). Upon demerging, ICI Australia was forced to reconsider the value

Demerged multinational enterprises