The Good, the Bad and the Ugly of the Global Financial Crisis
An analysis of multinational enterprise’s joint venture decisions in
relation to institutions in the aftermath of the global financial crisis
Final Thesis
Laura Beijleveld - 10428291
Date:
31/08/2015
Course Code:
MSc BA
Track:
International Management
Thesis Supervisor:
Dr. Ilir Haxhi
Statement of Originality
This document is written by Laura Beijleveld who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Acknowledgments
I’d like to thank all those close to me who have been patient and loving in their support for me during my thesis and masters. I thank my supervisor Dr. Ilir Haxhi for his advice and for being so flexible and patient. I’d also like to express gratitude to my mother Marja and aunt Yo for their support and my partner Javier for his encouragement. Special appreciation goes to my sister and talented editor Anouk who proofread this thesis.
I dedicate this thesis to my grandfather Guus who deceased during my Master. He would have been a great scholar if he would have had the chance. From a very young age he has always encouraged me to do well in school. It was his belief that stimulating his grandchildren in their education would give them the best chance of becoming fulfilled and independent personalities. Thanks to him I reached high and came far.
Abstract
Recent research has focussed on the impact of the 2008 global financial crisis on internationalisation strategy of multinational enterprises (i.e. MNEs). We argue that the effect of the crisis on share decisions in international joint ventures endures after the crisis. Building on institutional and transaction cost theory, we propose that after the crisis joint venture share decision were different than before the crisis and that the time of the move moderates the relationship between institutions and joint venture decisions. We test these relationships on a sample of 358 Dutch, German and French joint ventures. Our results show that there is no difference between joint venture share before and after the crisis and no support is found for the moderating relationship. Also, we find that higher cultural distance leads firms to opt for lower share in international joint venture decision. Therefore, our findings show that governmental effectiveness does not affect joint venture share decisions while cultural distance does. This study contributes to the entry mode literature by showing that MNEs are resilient after the global financial crisis. Share decisions in joint ventures in the after the crisis are highly similar to those before the crisis. Practically, this proves to managers that potential joint venture partners do not deviate from their decision-making behaviour in joint ventures after a global crisis of such a magnitude. Also, this study shows that cultural distance plays a pivotal role in joint venture decisions while the formal institutional environment does not.
Table of Contents
Introduction ... 6
Chapter 1: Literature Review ... 11
1.1 Definitions and Theories ... 11
1.2 Linking Institutions, Multinational Enterprises and the Crisis ... 24
Chapter 2: Theoretical Framework ... 29
2.1 The Good and the Ugly ... 30
2.3 Direct Effects ... 32 2.4 Moderating Effects ... 35 Chapter 3: Methodology ... 37 3.1 Sample ... 37 3.2 Data Collection ... 39 3.3 Variables ... 40 3.4 Data Analysis ... 44 Chapter 4: Results ... 48 4.1 Descriptive Analysis ... 48 4.2 Inferential Analysis ... 53 Chapter 5: Discussion ... 58 5.1 Findings ... 58 5.2 Managerial Implications ... 64
5.3 Limitations and Future Research ... 64
Conclusion ... 67
List of Figures and Tables
Figure 1: Timeline and Development of the Global Financial Crisis ... 21
Figure 2: Timeline of the crisis ... 23
Figure 3: Average FDI growth for EU countries from 1990 to 2011 ... 27
Figure 4: Average AEX scores from 2003 to 2013 ... 28
Figure 5: Conceptual Framework ... 36
Figure 6: Variable Overview ... 44
Table 1: Setup of Hierarchical Regression ... 46
Table 2: Descriptive Statistics ... 48
Table 3: Correlation Matrix ... 51
Table 4: Multicollinearity Statistics – based on model 9 ... 52
Table 5: Independent Sample Statistics; Percentage of IJV Share ... 53
Table 6: Independent Sample T-test; Percentage of IJV Share ... 53
Introduction
In 2007 the world faced the most severe financial crisis since the great depression of 1930. The global financial crisis and its effect on the global business world has been a popular topic of recent research. The focus of scholars has, however, been primarily on the crisis itself. They have not looked at how the business world sought to recover after the crisis.
The 2007 financial crisis (hereafter “GFC” or “the crisis”) has affected business and countries worldwide (Lairson, 2011; Dornean & Sandu, 2012). That makes its relation to multinational enterprise’s (hereafter “MNE”) and international joint venture (hereafter “IJV”) decisions relevant to research. MNE internationalization decisions are crucial as they affect a firm’s competitiveness and performance, determining the success of companies in foreign markets (Cavusgil & Sarkar, 1996; Brouthers, 2002; Dhanaraj & Beamish, 2004; Greenaway, Guariglia & Yu, 2014). Recent research shows that MNE behaviour changed during the crisis, both in intensity and form (Chung, Lee, Beamish & Isobe, 2010; Lairson, 2011; Williams & Martinez, 2012; Essen, Engelen & Carney, 2013). Indeed, the GFC has caused a significant decrease in foreign direct investment (hereafter “FDI”) flows worldwide (Te Velde, 2008; Lairson, 2011) because of increased global risk (Frank & Hesse, 2009).
Although it is known that the GFC affected global business, it is still not fully clear what the effect of the aftermath of the crisis on business has been. The global nature of the GFC caused a more negative impact on FDI than any other financial crisis in the past decades (Poulsen and Hufbauer, 2011). As with the great depression of 1930, the economic impact of the financial crisis persisted after the initial two-year recession (Xiao-yan, Yan-lei & Peilong, 2011). Therefore, Xiao-yan et al. (2011) argue that there are different stages in the crisis. Indeed, in The Literature the official GFC is the two-year financial crisis. This financial crisis was followed by an economic crisis in the after-GFC period of which Xiao-yan et al. (2011; P.4796) state that this “does not mean that […] the crisis has completely passed”. Instead, it
is “a more stable state after the crisis” (P.4796). This is supported by Dornean, Işan & Oanea (2013) who show that the impact of crises on FDI intensity can continue after an actual crisis. Indeed, after the GFC, FDI increased compared to the level during the recession but was not up to the same level as before the crisis (Worldbank, 2012). The change in FDI intensity of MNEs suggests a difference in MNE internationalisation behaviour after the crisis.
This research draws on transaction costs theory (i.e. TCA) and the institutional theory (i.e. IBV). It combines these two theories in what is known as the extended transaction cost model (Brouthers, 2002), which treats institutions and transaction costs as intrinsically linked. Many a research has been dedicated to finding out what role different institutional factors play in MNE internationalisation decisions (Brouthers & Hennart, 2007). The institutional environment has been shown to shape MNE internationalisation decisions (incl. Brouthers, 2002; Berry, Guillén & Zhou, 2010). Also, research has shown that there is a relationship between institutions and the crisis (Williams & Martinez, 2012). Therefore, we need to look at all these factors together and investigate the relationship between the crisis, entry mode behaviour and institutions. Williams and Martinez (2012) find that the effect of formal institutions on internationalizations decisions of Dutch MNEs changed during the GFC period. Though the GFC affects the relationship between institutions and MNE behaviour, it remains unclear whether the effect of the crisis on MNEs’ need for control in international ventures has endured until after the official GFC and thus differed from that before the crisis (Williams & Martinez, 2012; Essen et al., 2013). Regarding this, William and Martinez (2012) state that “gathering an extended data set will provide more insights into whether enduring economic crises exercise a sustained effect on MNE market entry strategies” (P.75). Also, it remains unclear what role the distance between the home and host country plays in this context and how informal institutions affect internationalisation decisions.
Scholars have studied a range of entry modes and entry mode classifications in relation to the GFC (Brouthers & Hennart, 2007). None of these have a strong focus on joint ventures solely. Joint ventures are a subset of entry modes and an accepted representation of MNE cooperative internationalization strategy (Brouthers & Hennart, 2007). IJVs are a form of entry modes used by MNEs when they need to cooperate with host-country partners. The reason for cooperating is to access the partner firm’s resources and knowledge of the host-country market (Brouthers & Brouthers, 2001), and to gain legitimacy in the host-host-country context (Kostova & Zaheer, 1999). Though IJVs serve these purposes, ownership is shared by two or more parties, leading to higher complexity due to conflicting interests and approaches. This can cause synergies to become more difficult to reach (Gaur & Lu, 2007). Indeed, in order to create a mutually beneficial situation, two or more parties share control and therefore share power (Dhanaraj & Beamish, 2004). IJVs are especially applicable to this study because the global impact of the crisis could have affected decisions regarding power sharing. In the IJV literature, it is argued that MNEs make a trade-off between costs associated with the liability of newness and the cost of integration in the host-country (Slangen & Hennart, 2008). In more uncertain and distant host-countries MNEs have lower legitimacy leading to higher costs of integration. To lower these transaction costs, MNEs need to cooperate with partner firms to access their knowledge and legitimacy (Zajac & Olsen, 1993; Kostova & Zaheer, 1999; Brouthers & Brouthers, 2001).
The lasting negative impact of the crisis worldwide should also be visible in the effect of institutions on MNE joint venture behaviour after the GFC compared to before the GFC. Therefore, this study looks at the crisis as the moderator in the relationship between both formal and informal institutions and IJVs. To address the gap in the literature this research sets out to answer a research question divided into two sub questions:
How does the aftermath of the Global Financial Crisis affect the relationship between host-country institutional frameworks and international joint venture decisions of MNEs?
Are joint venture share decisions in the after-crisis period different from those in the pre-crisis period?
Does the presence of a pre-crisis and after-crisis period moderate the relationship between formal and informal institutions and joint venture decisions of MNEs?
Together with economic circumstances, formal and informal institutions make up the environment for MNEs and determine their choice set (North, 1991; Brouthers, 2002). Combining these findings with the facts that the crisis continued to have a global economic impact after the official crisis (Xiao-yan et al., 2011) and FDI behaviour was significantly different after the crisis (Essen et al, 2013) it is likely that the impact of the crisis on IJV share behaviour continued as well. Transaction cost theory dictates that transaction costs rise when uncertainty increases (Williamson, 1975). Combining transaction cost theory with institutional theory, scholars argue that institutions determine uncertainty in the MNEs environment (e.g. Brouthers, 2002; Peng, 2011). Also, the crisis caused increased financial barriers and uncertainty for MNEs (Frank & Hesse, 2009; Pixley, 2010). Therefore, economic circumstances are another important contextual determinant of uncertainty and MNE IJV behaviour (Brouthers, 2002; Williams & Martinez, 2012). In fact, by affecting both MNEs and governments directly, the global financial crisis worked as catalyst for uncertainty thereby affecting their internationalisation decisions. This is also expected to affected MNE IJV behaviour. We therefore expect to find that MNEs make different IJV share decisions after the crisis as compared to before the crisis.
The purpose of this study is to find out how resilient MNEs are in times of global crisis when it comes to their cross-border cooperative activities. It aims to clarify which institutional factors shape their decision-making and if this behaviour is different in the after-crisis period. To find the answers to the research questions this research analyses a sample of 358 IJVs of public MNEs from Germany, France and The Netherlands, taken from the Thomson One Database. Formal institutions are measured through a composite index in the form of the economic freedom index. Informal institutions will take the form of cultural distance measured through the cultural distance index of Berry, Guillén and Zhou (2010).
This research contributes to the international management (IM) literature since it is the first study to investigate the link between MNE cross-border cooperation and the after-crisis period in the globalized business world. Considering the unequalled global economic impact of the crisis, previous research into crises has not had the opportunity to show how such a dramatic shock affects MNEs and institutions in the longer term. Also, this thesis contributes to both institutional and transaction cost literature in the area of international joint ventures. It builds on influential theories and ads to our theoretical understanding of the cooperative branch of MOE decisions. Lastly, by using the more recent cultural distance measurement of Berry, Guillén and Zhou (2010) as opposed to the more conventional index by Kogut and Singh (1988), this study also contributes to the literature on institutional distance.
This study starts out with a review of the existing literature in Chapter 1. In order to test the research questions, matching hypotheses are developed in Chapter 2. In Chapter 3 the methodological aspects of this study are explained, followed by analyses and results in Chapter 4. In Chapter 5 the findings and implications are discussed and suggestions for future research are given. In the final section, the research is concluded.
Chapter 1: Literature Review
1.1 Definitions and Theories
The following sections define the constructs in both the fields of MOE and GFC research.
MNE Internationalization Behaviour
This research looks specifically at MNE entry mode behaviour in the form of joint ventures. The sections below define IJVs and explain why these were chosen for this study.
MNE FDI activity, as opposed to zero-control modes such as outsourcing and foreign portfolio investment (i.e. FPI), is considered to be the most stable cross-border activity. Indeed, this type of investments typically involves more commitment of resources and cannot be abandoned that easily without considerable loss (Te Velde, 2008). As soon as a firm engages in FDI it is defined as a Multinational Enterprise. Indeed “non-MNE firms can also do business abroad by exporting and importing, licensing and franchising, outsourcing, or engaging in FPI. What sets MNEs apart from non-MNEs is FDI” (Peng, 2011; P.165). Importantly, Dornean et al. (2013) find a strong negative relationship between the crisis and FDI. Therefore, MNE FDI decisions are strongly suitable for this specific study. The International Management (i.e. “IM”) literature has looked at FDI behaviour in the form of amount of FDI (incl. Feinberg & Gupta, 2009; Holmes, Miller, Hitt & Salmador, 2013) and the methods which firms use to internationalize (incl. Brouthers 2002). There are different ways to analyse how firms internationalize. Amongst others, scholars have studied entry mode decisions (incl. Brouthers, 2002), export channel selection (incl. He, Brouthers & Filatochev, 2013) and the type of value-adding activities companies move across borders (incl. Slangen & Beugelsdijk, 2010).Out of all options, non-sequential mode of entry decisions are the most used indicator of MNE internationalization behaviour (Brouthers & Hennart, 2007). The definition of entry modes is “a structural agreement that allows a firm
to implement its product market strategy in a host country either by carrying out only the marketing operations (i.e., via export modes), or both production and marketing operations there by itself or in partnership with others (contractual modes, joint ventures, wholly owned operations)” (Sharma & Erramilli, 2004; P.2). Entry modes are often thought to be the first move a specific MNE makes into a new country (e.g. Tihanyi, Griffith & Russell, 2005). This is, however, not the case as they entail all non-sequential entries into foreign countries (Erramili & Rao, 1990). As the definition shows, joint ventures are a subsection of MOE decisions. This study looks at this subsection of MOE decisions. Therefore, research in the other areas of firm internationalization decisions is not included unless directly relevant. The MOE literature is of high relevance as joint ventures are a part of MOE decisions. Therefore, both IJV and MOE literature are used.
Measurements of Entry Modes
There are many ways to classify MOE decisions. In fact, Brouthers & Hennart (2007) identify a total of sixteen different ways in which past research has classified entry modes. This study looks at joint ventures, which means it excludes the zero modes as well as the full acquisition modes, thereby leaving out the two extremes. Although that rules out the majority of the different mode of entry classifications, there is still one very important divide in the measurements of MOE behaviour (Brouthers & Hennart, 2007). On the one hand there is the continuum approach, in which joint venture decisions are measured according to the height of their percentage, ranging from low to high. On the other hand there is the categorical approach, usually in the form of a dichotomy, where joint venture decisions are divided into majority and minority ownership moves. In the TCA continuum approach, strategic internationalization options are arranged along a hierarchical continuum of increasing control, commitment and risk (Yiu & Makino, 2002; Brouthers & Hennart, 2007). In this view,
licensing and contracts are the options with minimum control, commitment and risk, joint venture (i.e. IJV) the middle-range option and wholly owned subsidiary (i.e. WOS) the option with the highest control, commitment and risk (Brouthers & Hennart, 2007). In contrast with the hierarchical continuum view on MOE choices, the dichotomy supporters consider entry modes as a fundamental choice between either cooperation (i.e. IJV and acquisition) and sole modes (e.g. Feinberg & Gupta, 2009; Dikova & van Witteloostuijn, 2007) or the choice between joint ventures (IJV) on the one hand and (wholly owned) greenfield and acquisition on the other (e.g. Kogut & Singh, 1988; Yiu & Makino, 2002; Slangen & Hennart, 2008; Slangen & Van Tulder, 2009). Others, including Williams and Martinez (2012), have reconstructed the hierarchical continuum into a dichotomy between a majority control (51% and up) and minority control (under 51%), thereby taking out the continuum aspect.
Based on the principles of transaction cost theory and the extended transaction cost model, this research supports the view that risk is directly related to amount of control. Therefore, this study adopts the continuum approach.
Theories in the Entry Mode Literature
A variety of research has been done into entry mode decisions. Considering the vastness of this field of literature, it is imperative to specify what will be included in this study.
There are four main theoretical perspectives in the entry mode literature (Brouthers & Hennart, 2007). Aside from the IBV and TCA theories, the third influential perspective in the MOE literature is the resource-based view (i.e. RBV), which argues that a firm’s internal resources and capabilities determine its competitive advantages and thereby shape its strategic options and decisions (Barney, 1991). The fourth influential view is Dunning’s eclectic OLI paradigm (Dunning, 1988). In each of these four areas, relationships of their respective antecedents with MOE decisions have been found (Brouthers & Hennart, 2007).
Even though the four theoretical views are frequently viewed as separately, they are highly interrelated and complementary. Indeed, there is much contradiction and discussion in the field of MOE research on the relationships between the different theoretical perspectives (Delios & Beamish, 2001; Brouthers & Hennart, 2007; Canabal & White, 2008; Dunning & Lundan, 2008). This research does not aim to meddle in the theoretical discussions in the entry mode field of study. The interrelations are not denied though they do lie outside of the scope of this study. This research focuses exclusively on institutional theory and transaction cost theory combined into an extended transaction cost model (Brouthers, 2002). Research around the OLI and RBV theories is therefore excluded, unless relevant. The complementarity of institutional theory and transaction costs theory in the context of this research needs further explanation. This is done in the following sections.
Institutional Theory
This research looks at the relationship of both informal and formal country-level institutions with MOE decisions. The following sections are aimed at defining institutions.
Defining Institutions
There are many different definitions of institutions (Berry et al., 2010). In this research, Douglas North’s (1991) definition of institutions is adopted. According to North (1991) institutions are “humanly devised constraints that structure political, economic and social interaction” (P.91). An institutional framework is made up of institutions divided into formal constraints (i.e. formal institutions) and informal rules of the game (i.e. informal institutions) that, together with economic constraints, determine the choice set of companies and individuals (North, 1991; Dikova & Van Witteloostuijn, 2007). The institutional environment consists of a variety of institutional arrangements that make it dynamic and unpredictable (North, 1991).
The Institutional Framework
As mentioned before, the institutional framework is made up of formal and informal institutions. Formal institutions are tangible formalized rules of the game (Yiu & Makino, 2002; Peng, 2011; Holmes et al., 2013). They shape interaction in societies through codified, and frequently enforceable, standards and rules (North, 1991). Informal institutions reflect the social norms and values of specific societies that determine the interactions of individuals within a society (Holmes et al., 2013; Holmes et al., 2013). In contrast to formal institutions, informal institutions are ‘soft’, meaning that they are not documented and codified (Scott, 1995; Holmes et al., 2013). In the institutional literature, there is a divide between those that include cultural context in the institutions (e.g. Holmes et al., 2013) and those that treat culture as a construct separate from institutions (e.g. Brouthers, 2002). This research adopts the first approach and treats culture as an informal institution. Culture is a highly complex and diverse concept, making it difficult to define and to establish its effects on other variables (Brouthers & Brouthers, 2001; Tihanyi et al., 2005; Shenkar, 2012). A country’s national culture entails the “created and learned standards for perception, cognition, judgment, or behaviour shared by members of a certain group” (Te Velde, 2008; P.288). Culture is formed by societal values, beliefs, norms and behavioural patterns (Hofstede, 1980).
The relationship between informal and formal institutions is highly complex. Holmes et al. (2013; P.534) state that “culture is expressed through the resulting formal institutions”. Indeed, informal institutions form the logic and toolkit behind formal institutions. Formal institutions are a codified and specified expression of informal rules of the game (Holmes et al., 2013). In agreement with Williamson’s hierarchical model of institutions (2000) and Hofstede’s view on institutions (Hofstede, 1983), informal institutions have been found to shape, in part, a country’s formal institutions (Holmes et al., 2013). Indeed, a country’s culture shapes its members’ beliefs, norms and values which in turn determines their
behaviour and decisions (Hofstede, 1983; Holmes et al., 2013). However, Holmes et al. (2013; P.555) state that “we lack the data to show definitively that informal institutions are the source of formal institutions”. In addition to that, scholars argue that in the absence of strong formal institutions, informal institutions become more important and act as a substitute for formal institutions (North, 1991; Abdi & Aulakh, 2012). In systems with strong formal institutions, therefore, informal institutions take a smaller role. Outstandingly, it is Hofstede (2002) himself who claims that it is useless to claim causality between informal and formal institutions. Although the hierarchical relationship between informal and informal institutions is not denied, examining their causality is outside of the scope of this research. Therefore, this study follows Hofstede’s advice.
The literature on institutions in relation to firm strategic decision-making shows “little agreement on what factors make up the institutional environment or how to measure it” (Brouthers, 2013; P.17). The most widespread and accepted way of measuring a countries’ culture in the IM Literature is Hofstede’s cultural values (1980). These values consist of several dimensions which each have a score on a scale. Hofstede determined that the most significant indicators of a country’s culture are made up of four dimensions. These four dimensions are power distance (PDI), uncertainty avoidance (UAV), individuality (IDV), and masculinity (MAS). Power distance measures to what extent a country has hierarchical power structures while uncertainty avoidance shows how comfortable individuals are with taking risks. Individuality records to what extent individuals in a country identify themselves as a group or as individuals. Lastly, masculinity measures to what extend people value status and possessions as opposed to quality of life and personal relationships (i.e. femininity).
Measurements of Institutions
There have been two ways of measuring institutions in relation to MOE variables. The first entails the categorization of formal and informal institutions in numerous well-known indexes, such as Hofstede’s four dimensions of culture. The more recent trend is the usage of institutional distance measurements between the home and host country (e.g., Xu & Shenkar, 2002; Zhao, Luo & Suh, 2004; Berry et al., 2010; Schwens, Eiche & Kabst, 2011).
Literature on cultural distance versus that of cultural dimensions has caused many debates (Slangen and Van Tulder, 2009). Scholars have found that cultural distance has a direct impact on mode selection (incl. Kogut & Singh, 1988; Brouthers, 2002; Xu & Shenkar, 2002; Zhao et al., 2004; Brouthers, 2013). In other studies this was not the case and no relationship was found (e.g., Tihanyi et al., 2005; Slangen & Van Tulder, 2009). Also, scholars such as Brouthers & Brouthers (2001) and Haxhi and Van Ees (2010) found that not all dimensions of culture have equal effects on firm-level decisions and processes, suggesting that the relationship is more complex. This shows that there is still much to learn about cultural dimensions and cultural distance in relation to MOE decisions.
Cultural distance has been measured in different ways (Slangen and Van Tulder, 2009) and there is little consensus on which is the best measurement (e.g. Drogendijk & Slangen, 2006). Commonly used cultural distance measurements are, amongst others, the Kogut and Singh (1998) index (incl. Erramilli & Rao, 1993), perceived similarity in cultures (incl. Brouthers, 2002) and the Schwartz index (incl. Drogendijk & Slangen, 2006). One of the most recent CDI indexes, namely the Berry, Guillén and Zhou (2010) index takes into calculation a whole range of influencing factors. As with Kogut & Singh (1988) they calculate cultural distance based on the four dimensions of Hofstede (1980). However, unlike traditional Euclidean distance measurements, the distance measurement of Berry, Guillén and Zhou (2010) is based
more accurate as it takes into account the variance-covariance matrix and controls for multicollinearity.
Combining TCA with IBV, Zhao et al. (2004) suggest that external uncertainty within transaction cost theory consists of the two constructs of cultural distance and country risk. Indeed, this research will take the actual quality of formal institutions of the host country alongside cultural distance. Within the context of this research, the reason for choosing the actual quality of formal institutions is that there is a stronger relationship between the GFC and formal institutional effectiveness in the host-country environment (Williams & Martinez, 2012). Distance measurement of formal institutions in this context would be less appropriate as these focus more on the likeness of institutions rather than the actual form and effectiveness of these institutions. It can be assumed that these were less affected by the GFC than formal institutions as informal institutions change slower and incremental (Hofstede, 1980; Holmes et al., 2013).
Cultural distance has been used successfully by many scholars in the past, and should thus be taken serious as a measurement. Also, findings suggesting that relationships can differ per cultural dimension of Hofstede should be taken serious. Therefore, both these measurements are included in this research. The only difference is that the cultural dimensions of Hofstede are included as control variables and not predictor variables. If they would be included as predictors as well, the model would become too complex and the results too tainted.
For the dimensional measurement, Hofstede’s values are adopted as they are such a widely accepted measurement which covers almost all countries in the world. For the distance measurement there are different possibilities based on Hofstede’s dimensions. The Kogut and Singh (1988) index is most used in the literature thus far and is therefore most accepted. Shenkar (2001) heavily criticizes the cultural distance measurement of Kogut & Singh (1988) as he believes that the cultural distance measurement is based on different faulty assumptions.
In relation to FDI and MOE, he states that “the Kogut & Singh (1988) index is a rather simplistic aggregate of Hofstede’s (1980) dimensions” (P.525). He poses that this cultural distance measurement contains many biases, making it an unreliable measurement. Indeed, Brouthers and Brouthers (2001) and Brouthers (2013), Hennart and Slangen (2008) and Slangen and Van Tulder (2009) argue that cultural distance, as measured by Kogut & Singh (1988), does not have enough explanatory power in relation to entry mode decisions. To avoid as much as possible these biases of the Kogut and Singh (1988) index, this research adopts the index of Berry, Guillén and Zhou (2010) as it is more accurate and recent.
Transaction Cost Theory
Transaction cost theory (Williamson, 1975) is one of the most influential theories in the MOE literature (Shelanski & Klein, 1995; Zhao et al., 2004). This theories captures the notion that companies make balanced strategic decision based on costs and benefits. Transaction costs arise when negotiating and cooperating with partner firms in the host country, or alternatively, with internalization of a new market venture (Williamson, 1975; Brouthers, 2002). Transactions costs are based on two types of uncertainty faced by MNEs, namely behavioural uncertainty and contextual uncertainty (Williamson, 1979; Yiu & Makino, 2002; Zhao et al., 2004). Respectively, the first is embodied by risk of opportunism of a partner firm and the latter entails risks caused by uncertainty in the host country context, such as political instability and ineffective regulations (Yiu & Makino, 2002). MNEs make a trade-off between the need to cooperate with partner firms and the need to protect their assets in the host country environment (Brouthers, 2002). Putting this in transaction costs terms, MNEs make a choice between costs associated with the liability of newness and the cost of integration (Slangen & Hennart, 2008). Cooperating with a host-country partner can be beneficial for new entrants as the partner has both resources and (local) market-knowledge
that can be valuable for the venture. Also, cooperation ensures a division of costs and financial risks among partners (Zajac & Olsen, 1993; Kostova & Zaheer, 1999). The downside of cooperation is the costs that arise in monitoring the cooperation with the partner firm (Brouthers, 2002). For example, in its annual report of 2006, Ahold states about its majority stake in the joint venture with ICA that “the 60 percent shareholding stake in ICA does not entitle Ahold to unilateral decision-making authority […] The Company concluded that it has no control over ICA” (P.90). The cooperation costs increase when the host-country context and the trustworthiness of the partner firm pose a threat to the MNEs control over its assets (Henisz, 2000).
Linking Institutional Theory and Transaction Costs Theory
Transaction cost theory is the eldest and most researched theoretical perspective in the entry mode literature. The more recent institution-based view has long been considered to fall within contextual uncertainty within transaction cost theory (Yiu & Makino, 2002; Dunning & Lundan, 2008; Slangen & Van Tulder, 2009). The institutional context of MNEs strongly affects uncertainty, and with that potential transaction costs (Zhao et al., 2004). In past TCA research the institutional contextual factors “tend to be specified as moderators rather than direct effects” (Yiu & Makino, 2002; P.669), or are limited to several specific aspects of formal institutions. TCA critics say that the decision between the different strategic MOE options is determined by more than the factors the TCA scholars emphasize (Brouthers, 2002; Yiu & Makino, 2002; Brouthers & Hennart, 2007; Benito, Petersen & Welch, 2009). Indeed, the institutional environment shapes contextual uncertainty (Slangen & Van Tulder, 2009) resulting in transaction costs. Brouthers (2002; P.9) finds that “explanatory power of transaction cost models of mode choice could be improved by including aspects of both the institutional and cultural context”. In line with his findings, he argues for an extension of the
TCA with the IBV in what he calls the “extended transaction costs model” (P.203). This study supports Brouthers (2002) notion of combining transaction cost theory with both formal and informal institutions. It uses a variation of the extended transaction cost model in which both formal and informal institutions are included.
The Global Financial Crisis
The global financial crisis was a large and complex crisis. The following sections aim to explain the GFC in further detail.
History and Stages of the Global Financial Crisis
Figure 1: Timeline and Development of the Global Financial Crisis
The global financial crisis is the worst the world has seen since the great depression of 1930 (Xiao-yan et al., 2011). Past crises have affected various countries and regions simultaneously but never before has a crisis spread on such a scale and intensity as the global financial crisis (Calderon & Didier, 2009). Figure 1 shows that the US financial crisis began in 2007 with the US subprime mortgage crisis (Moshirian, 2011). On September 15, 2008 the crisis reached its breaking point with the fall of Lehman Brothers (Leschke & Watt, 2010) shortly followed by the collapse of Icelandic banks in October 2008 (Sigurjonsson, 2010; Throstur, 2010). Considering the globalized nature of the world economy with strong interdependent economies and sectors (Lairson, 2011), the financial crisis rapidly transcended
Stage 1 •2007/08 •Financial Sector crisis US Stage 2 •2008 •General Financial crisis Stage 3 •2009 •Public Finance crisis Stage 4 •2010 - on •Social and Economic crisis
national boundaries and transformed into the global financial crisis, affecting all countries (Lairson, 2011; Napolitano 2011; Williams and Martinez, 2012). Governments increasingly had to step in to bail out financial firms to prevent national economies from collapsing. In 2010, the economy began recovery and started to stabilize. Having been forced to bail out many companies many governments came into high debt. This stage is commonly referred to as the public finance crisis. The intensity of this interdependence between institutions, sectors, regions and countries and its repercussions has resulted into both national and global political and social crises that endured long after the public finance crisis.
Despite its global impact, the vastly different circumstances worldwide have caused the GFC to impact countries and regions with varying degrees of intensity (Te Velde, 2008; Zagelmeyer & Gollan, 2012; Berkmen, Gelos, Rennhack & Walsh, 2012; Dornean and Sandu, 2012). Indeed, Berkmen et al. (2012; P.42) find that “there is also some evidence that countries with a stronger fiscal position prior to the crisis were impacted less severely”. Emerging markets had a less strong fiscal position at the start of the crisis and therefore the impact in this respect might have been more severe for such countries. Some markets endured a less negative effect of the GFC and some markets recovered sooner (Dornean et al., 2013). Leschke and Watt (2010) point out that different countries either have or lack buffer mechanisms and thus differ in their ability to buffer the effect of the GFC. Also, several emerging markets were less affected by the GFC (Frank & Hesse, 2009; Lairson, 2011) while other regions and countries suffered more from the GFC. Different countries, regions, and MNEs have highly dispersed value chain activities and thus stronger cross-border interdependencies (Coe, Hess, Yeung, Dicken & Henderson, 2004). Such interdependencies typically result in a stronger transition of cross-border financial impact (Aloui, Aïssa & Nguyen, 2011). Indeed, Xiao-yan et al. (2011) state that “some countries are recovering […] but this does not mean the entire world economy [is] (sic) over the financial crisis” (P.4796).
Despite the varying impact, both the intensity of the global financial crisis and the vast global interdependencies have caused a global financial contagion effect, thereby negatively affecting all countries worldwide (Calderon & Didier, 2009; Frank & Hesse, 2009).
The Official Crisis Period
There is much discussion about the official timeline of the crisis (Xiao-yan et al., 2011). Where Essen et al. (2013) state that the crisis started may 2007 and ended beginning 2009, GPD data, however, shows that the crisis in Europe began in the first quarter of 2008 (Leschke & Watt, 2010). Indeed, Williams and Martinez (2012) have taken the full years of 2008 and 2009 as the GFC period. This period was in fact the recession which finished in 2009. In that line of thinking, Lairson (2011) specifies the GFC period as the period from 2008 to 2010. To bring clarification to this discussion Xiao-yan et al. (2011) introduce a dual conceptualization of the period after the crisis. They refer to the earliest period after the public finance crisis (i.e. around 2009) as the post-financial crisis period. The period after the post-financial crisis period, starting after 2010, is referred to as the post financial crisis era.
This study looks at the three-year period after the crisis. Specifically, this period entails the post-GFC period from the end of 2009 to 2011 and a part of the post-GFC era after 2010. This also coincides with the AEX index scores, which started to show some negative developments in 2007 and were lowest from mid-2008 to the beginning of 2009 while going up again around 2010 (Euronext, 2014). This will be discussed in more detail later.
Pré-GFC Start GFC End GFC Post-GFC Post-GFC era
Jan 2005 Dec 2007 Dec 2009 Dec 2011 Dec 2012
1.2 Linking Institutions, Multinational Enterprises and the Crisis
Having discussed the theories and definitions in the previous section, this section focuses on linking them all together.
Institutions and MNE Behaviour
Cantwell, Dunning & Lundan (2010) show that MNEs and institutions are strongly interdependent. A foreign market’s institutional characteristics are considered of great importance when deciding on an entry mode (Henisz, 2000; Yiu & Makino, 2002). Every country has a unique institutional environment that determines the way in which firms can arrange their business activities (North, 1991). The quality of market-supporting institutions affects market effectiveness (Meyer, Estrin, Bhaumik, & Peng, 2009). Market effectiveness, in turn, affects the ability of firms to execute their strategies and use their resource base to create value in the host country context (Dikova & van Witteloostuijn, 2007). Indeed, North (1991; P.98) states that “effective institutions raise the costs of defection or the benefits of cooperative solutions”. As a consequence of this relative market effectiveness, firms decide on the most appropriate entry strategy (Meyer et al., 2009). This way “global strategy is fundamentally shaped (at least in part) by the formal and informal institutions” (Peng & Pleggenkuhle-Miles, 2009; P.53).
Many scholars have researched the effect of either formal or informal institutions on MNE behaviour (e.g. Williams & Martinez, 2012). Overall, formal institutions have more impact on MOE behaviour than informal institutions (Yiu & Makino, 2002; Zhao et al., 2004; Hennart & Slangen, 2008; Williams & Martinez, 2012). Indeed, Slangen and Hennart (2008) argue that dimensions of formal institutional distance, such as political and governance distance have stronger explanatory power than those of informal institutions. Slangen and Beugelsdijk (2010) find that formal institutions are more difficult to go around, as they are
often enforceable. That while informal institutions can be more easily circumvented (Slangen & Beugelsdijk, 2010). MNEs are better able to circumvent any specific cultural issues and instead focus on changing or adapting to formal institutions. Formal and informal institutions are interdependent and have a combined effect. This also means that they complement each other in their effect on MNE internationalization behaviour.
The Global Financial Crisis and Institutions
The relationship between institutions and economic factors is strongly present yet not straightforward. In the very definition of institutions, Douglas North (1991) suggests that economic constraints complement institutional constraints. Although this definition suggests a simple co-effect, reality shows that the relationship is less straightforward.
Scholars claim that the GFC is the direct result of failing institutions (Duoqi, 2011; Cabral, 2013). Indeed, Duoqi (2011) claims that the financial crisis was the result of financial corruption. Similarly Sigurjonsson (2010) notes that the failure of the Icelandic government’s regulations and risk assessment led to the Icelandic banking crisis. However, although failing institutions have been a major contributing factor, it is questionable whether it was solely failing institutions that caused the GFC Moshirian, 2011). MacNeil (2010) points out that “the interplay of different causes of the crisis makes it difficult to attribute causality in any precise way to different factors” (P.485). Next to being caused, at least in part, by failing institutions, the GFC has also been the cause of dramatic institutional changes. Indeed, Moshirian (2011; P.502) states that “crises often lead to the emergence of new national and international institutions”. The GFC has induced changes in both formal and informal institutions. Essen et al. (2013: P.102) note that “the efficacy of governance mechanisms (institutions) may be contingent upon organizational and environmental circumstances”. The GFC resulted in environmental and organizational circumstances that posed significant
problems, thereby inducing change in existing institutions and necessitating the creation of new market-governing institutions. “Formal institutions begin as solutions to problems in society” (Holmes et al., 2013; P.533). Indeed, the societal problems of the GFC has induced change on a firm level (Zagelmeyer & Gollan, 2012), an industry level (Zagelmeyer & Gollan, 2012), and a country level (MacNeil, 2010; Moshirian, 2011). The degree of direct government involvement in this GFC is unprecedented as many governments have had to bail-out banks and other companies (Zagelmeyer & Gollan, 2012). Governmental bodies are strongly linked to formal country-level institutions and therefore the bail-outs by governments are already an indirect cause of change in formal institutions. Considering that the failure of institutions has contributed to the GFC, it is not difficult to imagine why formal institutions have changed. The GFC may also have affected the role of informal institutions as well. Indeed, where formal institutions fail or lack effectiveness, informal institutions become substitutes (North, 1991; Abdi & Aulakh, 2012; Holmes et al., 2013). Considering that the GFC was caused, in part, by failing formal institutions, informal institutions may have gained a more important role (Napolitano, 2011). Although this is a fruitful area of research, the study of causality of the GFC and changes in institutions lie outside of the scope of this research. Its interrelation, however, makes it even more appropriate to research the link between the after-GFC and institutions.
The Global Financial Crisis and Multinational Enterprises
The global nature of the GFC has universalised certain types of risks and financial influences faced by MNEs (Lairson 2011). Figure 4 shows the FDI development of European countries in the period from 1990 to 2011. It shows that the GFC caused companies to drastically reduce their FDI and it did not recover after a mere two years. Indeed, in 2011, it was still at least three times as low as it had been before the crisis.
Figure 3: Average FDI growth for EU countries from 1990 to 2011
Based on data from Worldbank, available at www.data.worldbank.org (2012)
Publicly listed MNEs in times of crisis
The European stock indexes partially reflect the economic circumstances (Euronext, 2014). Having been impacted by the GFC, the stock indexes affect MNE behaviour as well. For example, Figure 3 shows the AEX scores from 2003 to 2013 (Euronext, 2014). It shows the development of the AEX before and during the crisis period. As expected, the start of the US subprime mortgage crisis during 2007 did not directly impact the AEX scores drastically, though there is a relatively steep downfall around September 2007. The real downfall in the AEX scores did not come until 2008, coinciding with the Lehman Brothers bankruptcy in September, shortly followed by the collapse of the Icelandic banks in October. This was the point at which the AEX started responding more intensely, as the Icelandic bank crisis was the most intense transfer of the US banking crisis to Europe. In 2009, when the economic crisis was at its peak after months of continuing crisis, the AEX reached its lowest point. From there on, the market started to pick up slowly, coinciding with the official end of the official two year crisis period. Indeed, in 2010 the FDI level was comparable to that during the 2002 economic downturn. Other European stock exchanges have shown similar patterns.
Figure 4: Average AEX scores from 2003 to 2013 (Source: Euronext, 2014)
Linking Institutions, Multinational Enterprises and the Crisis
Institutions change in reaction to changes in MNE behaviour and MNE behaviour changes in reaction to changes in the institutional environment (Cantwell et al., 2010). Change in country-level institutions and MNE MOE behaviour can be induced by major environmental shocks such as the crisis (Lairson 2011, Moshirian 2011, Williams & Martinez, 2012). The uncertainty caused by the crisis was universally diffused worldwide and its impact on institutions and IJV behaviour of MNEs was omnipresent (Dornean et al., 2013). Indeed, Essen et al. (2013) state that “[MNE] governance mechanisms operate differently in crisis and non-crisis periods” (P.201).
Chapter 2: Theoretical Framework
The relationship between host-country institutions and entry mode decisions of MNEs has been researched before. Also, the relationship between economic circumstances and entry modes decision has been the subject of past research. However, never before has the relationship been put together into a framework where informal institutional distance and formal institutional effectiveness together were proposed to be predictor variables of joint venture share decisions. More importantly, past research has not linked these relationships together in times of economic shock. This research theorises that only a clear picture of the effect of institutions on joint venture behaviour can be derived when both formal and informal institutions are taken together as predictors (North, 1991; Brouthers, 2002). Indeed, informal institutions could have become more important than formal institutions due to the connection between failing formal institutions and the crisis.
The institutional environment seen through a lens of transaction cost theory indicates that institutional factors determine environmental uncertainty of the MNE. Indeed, Peng (2011) points out that “Attempting to reduce (such) transaction costs, institutional frameworks increase certainty by spelling out the rules of the game so that violations (such as failure to fulfil a contract) can be mitigated with relative ease (such as through formal arbitration and courts).” (P.34). This in turn determines share decisions of MNEs. From a transaction cost perspective, the global financial crisis was simply another environmental factor which increases uncertainty for MNEs and adds to their transaction costs. Economic circumstances that take place on a transnational scale, such as the GFC, cannot be controlled by firms. Indeed, the GFC directly caused financial and institutional barriers for global businesses (Frank and Hesse, 2009), forcing them to adapt their firm-level strategy to these global-level circumstances. In these times the revenues of MNEs lowered and equity prices decreased. Also, credit and lending condition became stricter, making it more difficult to secure
financing for ventures. This in turn makes the prospects bleaker as potential joint venture partners are less promising and there’s a higher chance of failure in the partnership. Also, the crisis adds to the transaction costs as it is even more difficult for MNEs to search for partners, negotiate partnerships and ensure commitment (Dyer, 1997). Lastly, in times of crisis MNEs expect less support from the government in their co-ownership transactions as the governments are more occupied with the crisis. Therefore the global financial crisis works as a catalyst for environmental uncertainty. After the official global financial crisis period the social and public finance crisis ensued. This continued to be the cause of higher uncertainty and transaction costs in the environment and continued to affect MNE share decisions. The connection between formal institutions and the crisis, the evidence of endurance of the crisis, alongside the importance of cultural distance in joint venture decisions shapes the theoretical framework as proposed in this study. This theoretical framework can be found in Figure 5. The following sections will go deeper into this framework.
2.1 The Good and the Ugly
Entry mode decisions are influenced by economic circumstances (Brouthers, 2002; Dornean et al., 2013). The overarching implication of the global financial crisis for the business sector is the increased uncertainty and risk in global businesses activities (Pixley, 2010; Zagelmeyer & Gollan, 2012). The general tendency for firms is to return to a more steady-state after the initial crisis has passed. This shows, for example, from a statement that AEX member ArcelorMittal (n.d.) made on its corporate website, namely “Post-crisis, ArcelorMittal has cautiously restarted certain projects to capture growth in key emerging markets and mining.” This return to steady-state conditions does not mean that MNE internationalization behaviour returns to exactly the same decision behaviour as before the crisis. This shows evidently from the fact that FDI in the post-crisis period was significantly different from that before the crisis
(Essen et al., 2013). Although research shows a steep decrease in FDI during the crisis, not all firms have decreased their internationalization behaviour during the GFC. For example, Chung et al. (2010) find that certain firms internationalized more during the period of the GFC. In fact, both Lipsey (2001) and Chung et al. (2010) propose that internationalization can be a way to spread risk or even to counterbalance financial setbacks in the home country environment. The diminishing economic circumstances and changing exchange rates can provide opportunities for MNEs to invest in advantageous international ventures as more host-country firms may be undervalued (Agrawal & Jaffe, 2003). On the other hand, Frank and Hesse (2009) argue that investors were more prone to invest in less risky assets in their home country environment and withdraw or avoid investment in higher risk countries. This is supported by FDI data and the findings of Williams and Martinez (2012).
Indeed, Williams and Martinez (2012) proved that during the time of the global financial crisis MNEs tend to opt for higher control in entry modes. MNE behaviour changed during the period of economic crisis (Williams & Martinez, 2012) and continues to be affected by the repercussions following the initial global downturn. This should also express itself in a difference in MNE behaviour in the post-GFC period. I pose that this difference is not only visible in FDI quantity but also quality, i.e. in IJV share decisions. It would be logical to infer the same difference in IJV behaviour after the crisis as during the crisis. Although FDI data shows an increase as compared to during the crisis after the initial two year recession (Worldbank, 2012) it is still relatively low compared to before the crisis. That means that MNE joint venture share behaviour could be argued to have gone in either direction.
For example, FDI increased as opposed to its decrease during two- year recession (Worldbank, 2012). This could mean that MNEs opted for opposite behaviour as well, namely choosing lower control instead of higher control. On the other hand, the fact that FDI data is still relatively lower than in the before-GFC period could mean that the preference for
higher control modes by MNEs persisted during the after-GFC period. Alternative hypotheses are needed to capture all possible changes in MNE IJV behaviour in the after-crisis period.
Hypothesis 1: After the crisis MNEs opt for higher control in joint ventures than before the crisis
Hypothesis 1-alt: After the crisis MNEs opt for lower control in joint ventures than before the crisis
2.3 Direct Effects
Formal Institutions
The effect of formal institutions on strategic mode decisions in general shows that results are significant yet contradicting (Morschett, Schramm-Klein & Swoboda, 2010). On the one hand, scholars have found that decreased institutional effectiveness leads firms to opt for less control. The reason for this is that the host country partner is familiar with the host country environment and its regulatory structure and thereby reduces uncertainty (e.g., Morschett et al., 2010). On the other hand, another group of scholars finds opposite evidence and argue that firms opt for more control in these environments as there are less effective regulations. That causes the risk of opportunistic behaviour of the partner firm to increase (Zhao et al., 2004; Brouthers & Hennart, 2007). In addition, countries that have less strong formal institutions oftentimes have high corruption and political instability (Investopedia, n.d.). This also causes increased risk of, for example, expropriation (Dadasov & Lorz, 2013). Holmes et al. (2013) find a possible explanation for the opposing findings. They look at tightness of control of MNE behaviour by government and find that MNEs tend to use lower investment (i.e. control) when government involvement is high. In addition to proving that MNEs let go of control when institutional effectiveness is high, Holmes et al. (2013; P.540) state that
“managers seek to minimize government involvement in MNE affairs”. Indeed, in their meta-analysis of 72 empirical studies into entry mode antecedents they find that MNEs opt for lower control in entry modes when institutional effectiveness is high.
Transaction costs theorist argue that increased regulatory effectiveness decreases uncertainty and thereby transaction costs (North, 1991; Brouthers, 2002; Meyer et al., 2009). As the effectiveness of formal institutions increases, the host-country conditions become safer. Indeed, both governmental bodies and the IJV partner firms have less space for opportunism. In line with transaction cost theory, this study poses that when formal institutions are effective, firms opt for less control.
For example, Williams and Martinez (2012) analysed the effect of formal institutions on entry mode decision in the pre- and during-crisis periods on a sample of Dutch publicly listed MNEs. They found a significant negative relationship.
Hypothesis 2: Formal institutional effectiveness has a negative relationship with joint venture share
Informal Institutions
Research has shown opposing results when linking internationalization decisions with cultural distance (incl. Slangen and Van Tulder, 2009). The main explanation of these differing results is the vastly different categorizations of entry modes (Brouthers & Hennart, 2007). For example, Brouthers and Brouthers (2001) find that higher cultural distance leads to a preference for cooperative modes. Interestingly, when researching a sample of publicly listed Dutch MNEs, Slangen and van Tulder (2009) find no effect of cultural distance on entry mode decisions. Where the relationship was found between cultural distance and entry modes there are opposing views in this stream of literature (Shenkar, 2001). The first group
of scholars argues that higher cultural distance increases the likelihood that a low-control mode is used (Tihany et al., 2005; Brouthers & Brouthers, 2001; Greenaway et al., 2014) because in such a situation the MNE needs a partner who possess relevant knowledge (e.g., Brouthers and Brouthers, 2001) and legitimacy in the host country context (Zajac & Olsen, 1993; Kostova & Zaheer, 1999). Therefore, increased cooperation with the partner is preferred in the face of higher cultural distance. On the other hand, scholars find that MNEs opt for higher control in the face of more cultural distance (e.g. Padmanabhan and Cho, 1999, Xu & Shenkar, 2002; Drogendrijk & Slangen, 2006). The reason for this is that a higher cultural distance increases the costs associated with cooperation as firms are fundamentally different and more difficult to integrate (Gaur & Lu, 2007). This increases the likelihood of failure and opportunistic behaviour of partner firms. In the worst case such cultural distance can cause conflicts and dissolution of the partnership (Vaaland, Haugland & Purchase, 2008). This study poses that high cultural distance leads to a higher need to cooperate with the partner firm. As the partner firm is oftentimes a local business, they know the local culture better (Brouthers & Brouthers, 2001). Also, they are already established in the host-country market and therefore have higher legitimacy (Kostova & Zaheer, 1999). Thus partner firms are given higher control in IJVs when cultural distance is high.
For example, Greenaway et al. (2014) show that IJVs are a very common type of FDI when companies go abroad to less developed and more distant countries. Also, in their analysis of US public MNEs Berry et al. (2010) find a negative relationship between their cultural distance measurements and amount of control in entry modes. These findings are relevant for this study as the same distance measurement is used.
2.4 Moderating Effects
Essen et al., (2013) showed that the quality of formal country-level institutions (i.e. quality of legal framework) impacted MNE performance during the GFC. In a similar vein, Williams and Martinez (2012) find that in the face of the crisis, MNEs tend to opt for relatively higher control. This shows that the crisis conditions work as a catalyst of uncertainty.
The crisis overall increases transaction costs associated with finding and negotiating with potential partners (Dyer, 1997). During the crisis period MNEs can rely less on governments to mediate the partnership as governments are preoccupied with the economic challenges. Overall, transaction costs related to uncertainty increased, making MNEs more considerate of their joint venture decisions. This made MNEs more inclined to choose higher share in order to control the partner firms more. Considering that MNEs could rely less on formal institutions to regulate the partnerships with joint venture partners, cultural distance could have become a bigger obstacle. Indeed, with lower involvement of regulations to guide the joint venture cultural distance could lead to problems with reaching effective cooperation in the partnership (Gaur & Lu, 2007) and in worst case failure of the joint venture (Vaaland, Haugland & Purchase, 2008).
Martinez & Williams (2012) included cultural distance as a moderator in their model. In a sample of MOEs from before and during the crisis, they find that cultural distance is negatively related to majority control. This means that the higher the distance between the home and host country, the lower the control MNEs opt for in international ventures. Nevertheless, Williams and Martinez (2012) did not look at cultural distance as a predictor variable and therefore information lacks about its effects in relation to the global financial crisis. What is known is that FDI was different in the after-crisis period as compared to before the crisis (Dornean et al., 2013).
This study poses that the relationship between institutions and IJV share decisions is affected by the post-crisis situation. According to the principles of transaction cost theory, increased risk caused by the GFC, due to several complications, encourages MNEs to intensify their uncertainty-controlling behaviour. Therefore, both of the direct effects are theorized to be magnified by the after-crisis circumstances.
For example, the findings of Williams and Martinez (2012) show that this is the case during the crisis. Therefore, it is assumable that this is also the case in the after-crisis period.
Hypothesis 4: Compared to before the crisis, there is a stronger negative relationship between formal institutional effectiveness and the percentage of joint venture share in the after-crisis period.
Hypothesis 5: Compared to before the crisis, there is a stronger negative relationship between cultural distance and the percentage of joint ventures share in the after-crisis period.
Economic Freedom Cultural Distance
Time: Pre/Post Global Financial Crisis H2 - H3 - H1 H4 H5 IJV Share (Percentage)
Chapter 3: Methodology
To test the hypotheses, this research entails a statistical analysis of secondary database data. Considering the nature of the variables, the statistical technique chosen is a multiple logistics regression. The following sections are aimed at explaining the data and methodology.
3.1 Sample
The sample consists of 358 non-domestic cooperative outward FDI moves in a wide range of countries of publicly-trading MNEs listed in the Dutch, German and French stock exchange in the periods of 2005 to 2007 and 2009 to 2012. The IJV moves are collected from the Thomson One Database. As is practise in this field, joint ventures are all cooperative non-sequential moves into a foreign country (e.g. Erramili and Rao, 1990).Therefore, moves into foreign countries where the MNE is already present are also included while sequential moves are excluded. All investments under 5% are excluded because they are small and inconsequential equity holdings (Kuo, Kao, Chang & Chiu, 2012). In extent of that, joint ventures are all investments up to 95% (Hennart, 1990). Williams and Martinez (2012) took all AEX companies listed during 2004, the first year in their sample-period. Considering that this study compares the after-GFC period with the pré-GFC period, a different approach is taken. The dataset includes IJVs of MNEs that were publicly listed throughout the sample period (i.e. from 2005 to 2012). The companies not publicly listed in all the sample years are not representative of the sample and thus excluded.
The GFC impacted countries differently worldwide. This creates differences in contextual influences on MNE behaviour and makes it more difficult to research causality. The different circumstances and effects in the region-of-origin should be controlled for by taking out the region-of-origin bias. It was decided to take a sample of MNEs from three developed countries in a highly similar e economic region (EU), namely the MNEs listed in the Dutch,
