Directional institutional distance, host country
experience and subsidiary performance
To what extent does host country experience moderate the effects of upward- and
downward institutional distance on subsidiary performance?
Master Dissertation
by Elles Faber
DDM MSc Advanced International Business Management & Marketing Student no. University of Groningen: S2970171
Student no. Newcastle University Business School: 190168759
Supervisors: Dr. Stefanie Reissner
Dr. Rieneke Slager
1 ABSTRACT
Inspired by the increasing importance of the asymmetry of institutional distance, this study investigates the moderating impact of host country experience on the relationship between institutional distance and subsidiary performance, taking both the magnitude and the direction of distance into account. This relationship is evaluated based on a panel data-set of 1757 foreign subsidiaries in 64 host countries and from 65 home countries, over the four-year period of 2013-2016, collected from Bureau van Dijk’s ORBIS database. The findings demonstrate a strong positive effect of institutional distance on subsidiary performance in both the upward and the downward direction, which contrasts with existing literature. Moreover, the findings show a significant positive moderating effect of host country experience, strengthening the relationship between institutional distance and subsidiary performance. This was only the case, however, for firms moving in an upward direction, i.e. to a stronger institutional environment compared to their home country. This study makes an important contribution to the existing literature by accounting for both the magnitude and the direction of institutional distance in examining its effect on subsidiary performance, which is a recent phenomenon. Additionally, the findings regarding the moderating effect of host country experience contribute to the literature as they confirm suggestions that increased host country experience reduces costs associated with operating abroad, and, more importantly, they demonstrate that the moderating effect of host country experience varies depending on the direction in which the firm moves. Thus, this provides important new theoretical insights regarding the effects of host country experience on institutional distance and subsidiary performance, which had not been clearly established by previous research.
KEYWORDS – institutional distance · directional institutional distance · subsidiary performance · host
2 ACKNOWLEDGEMENTS
3 TABLE OF CONTENTS
CHAPTER 1 – INTRODUCTION ... 5
1.1 – Introduction ... 5
1.2 – Theoretical background ... 5
1.3 – Literature gap and research questions ... 7
1.4 – Research approach, findings and contributions ... 9
1.5 – Structure ... 10
CHAPTER 2 – LITERATURE REVIEW ... 11
2.1 – The costs of doing business abroad ... 11
2.2 – Institutions and institutional distance ... 11
2.2.1 – The institutional environment ... 11
2.2.2 – Institutional distance ... 13
2.3 – Insights from institutional distance studies ... 15
2.4 – Institutional distance and performance ... 16
2.4.1 – Negative effects of institutional distance ... 17
2.4.2 – Positive effects of institutional distance ... 17
2.4.3 – Assets of foreignness ... 18
2.4.4 – Adaptation and arbitrage ... 19
2.5 – Asymmetry of distance ... 20
2.6 – Host country experience ... 22
CHAPTER 3 – DATA AND METHODS ... 27
3.1 – Research philosophy statement ... 27
3.2 – Data collection and sample development ... 27
4
4.2 – Regression models ... 38
4.2.1 – Results Model 1 ... 42
4.2.2 – Results Models 2 and 3 ... 42
4.2.3 – Results Model 4 ... 43
4.2.4 – Results Model 5 and 6 ... 43
4.2.5 – Results control variables ... 44
4.3 – Alternative tests ... 44
4.3.1 – Direction of institutional distance as a dummy ... 44
4.3.2 – Alternative performance measure ... 45
CHAPTER 5 – DISCUSSION ... 50
5.1 – Implications ... 50
5.2 – Theoretical and methodological contributions ... 52
5.3 – Managerial implications ... 54
5.4 – Limitations and suggestions for future research ... 54
CHAPTER 6 – CONCLUSION ... 56
REFERENCES ... 57
5 CHAPTER 1 – INTRODUCTION
1.1 – Introduction
As multinational enterprises are operating in multiple countries, they are subject to different institutional environments and affected by the various institutions within which they are embedded (Peng et al., 2008). In order to operate effectively in the host country, MNEs are required to adjust their practices to the foreign institutional environment, and have to be prepared for challenges such as differences in laws, regulations, and policies (Chao et al., 2012). Institutional differences can present significant challenges, however, by carefully adapting to the differences between home and host countries, and by developing extensive country-specific knowledge, MNEs may be able to use these differences to their advantage. For example, the American media-services provider Netflix has expanded to 190 countries within the past decade, by focusing on developing broad and deep knowledge about its host countries, which allowed them to respond to differences and to build credibility, even in countries very distant from their home country (Brennan, 2018). This highlights the importance of institutional differences and host country experience for international business. The next section provides a summary of the theoretical background to set the stage of this research. Thereafter, the literature gap is identified and research questions to address this gap are presented, followed by an overview of the research approach and a summary of the findings. This introductory chapter concludes with an overview of the structure of this dissertation.
1.2 – Theoretical background
6
institutional distance considers differences and similarities in institutional profiles of home and host countries, including regulatory, cognitive, and normative institutions.
Over the past decades, the concept of institutional distance has gained significant attention in the literature, and it is regarded as an important factor that influences MNE decision-making and subsequently affects subsidiary performance (Ambos and Håkanson, 2014; Gaur and Lu, 2007; Konara and Shirodkar, 2018). There are varying results in the literature regarding the relationship between institutional distance and subsidiary performance, ranging from negative to positive. For example, it has been argued that a larger institutional distance hinders the MNE from establishing legitimacy in the host country and from transferring strategic routines to its foreign subsidiaries (Kostova and Zaheer, 1999). Additionally, a larger institutional distance may trigger conflicting demands for external legitimacy in the host country and global integration within the MNE system (Xu and Shenkar, 2002). Moreover, Chao and Kumar (2010) demonstrate a negative relationship between institutional distance and performance as a firm continues its international expansion process. Chao et al. (2012) underline these findings and argue that regulatory institutional distance creates the greatest impediments for effective operations abroad.
7
literature regarding the relationship between (institutional) distance and performance are quite mixed.
A potential explanation for these contradictory results resides in the way institutional distance is operationalised. Traditionally, research on regulatory institutional distance has conducted its analysis in absolute terms, considering only the magnitude of the distance and assuming a symmetrical effect (Hernández and Nieto, 2015). The magnitude of institutional distance describes to what extent institutional differences are present between a particular home and host country. However, Konara and Shirodkar (2018) demonstrated that the ‘direction’ of institutional distance should also be taken into account, as the magnitude and the direction of distance have varying effects on subsidiary performance. The extent of distance depends on the MNE’s relative position in terms of the level of development of the home country institutional environment, compared to that of the host country (Hernández and Nieto, 2015). It can be argued that particular institutional environments are stronger, better regulated than others, as levels of development differ. The ‘direction’ of the distance concerns whether a firm moves from an institutional environment with stronger, more developed institutions, to an institutional environment with weaker, less developed institutions, or vice-versa. According to Konara and Shirodkar (2018), this direction determines whether the magnitude of institutional distance affects subsidiary performance negatively or positively. That is, MNEs moving in an ‘upward’ direction, to a host country with stronger institutions compared to their home country, are likely to face negative effects on subsidiary performance as the result of a negative outcome of arbitrage benefits and adaptation costs (Konara and Shirodkar, 2018). On the contrary, MNEs moving in a ‘downward’ direction, to a host country with weaker institutions compared to their home country, can face positive effects on subsidiary performance due to a positive outcome of arbitrage benefits and adaptation costs (Konara and Shirodkar, 2018).
1.3 – Literature gap and research questions
8
by conducting extensive pre-entry market research, the most important knowledge about the local business environment is inherently experiential and market-specific (Pedersen and Pedersen, 2004). With increased experience in the host country, firms develop a better understanding of the local culture and the institutional environment, which makes MNEs’ subsidiaries better equipped to address institutional differences and to reduce their liability of foreignness (Luo, 1997; Gaur and Lu, 2007; Shirodkar and Konara, 2017). This affects adaptation costs and arbitrage benefits, which in turn influences the subsidiary’s performance. As the levels of potential arbitrage benefits and adaptation costs differ for firms moving in an upward direction and in a downward direction (Konara and Shirodkar, 2018), the moderating impact of host country experience may also differ depending on the direction of institutional distance.
This dissertation contributes to filling the literature gap by examining whether there is a moderating impact of host country experience on the relationship between institutional distance and subsidiary performance, taking both the magnitude and the direction into account. To address these questions, the following research questions were developed:
1) To what extent does the direction of regulatory institutional distance influence subsidiary performance?
2) To what extent does host country experience moderate the effects of upward- and downward institutional distance on subsidiary performance?
9
1.4 – Research approach, findings and contributions
In order to test the relationship between directional institutional distance, host country experience, and subsidiary performance, I developed several hypotheses based on an extensive review of the current literature (Chapter 2). The hypotheses are evaluated based on a panel data-set of 1757 foreign subsidiaries over the four-year period of 2013-2016, collected from Bureau van Dijk’s ORBIS database. The sample consists of subsidiaries operating in 64 host countries, with parents originating from 65 home countries. A detailed explanation of the data and methods used in the analysis is provided in the methodology chapter (Chapter 3). The results provide evidence for a positive relationship between institutional distance and subsidiary performance, in both the upward- and the downward direction. This contrasts with findings of Konara and Shirodkar (2018), who provided evidence for a negative effect of upward institutional distance on subsidiary performance. Furthermore, host country experience showed a strong positive moderating effect on the relationship between institutional distance and subsidiary performance. While this moderating effect was not statistically significant in the downward direction, the results show a significant positive moderating effect of host country experience in the upward direction. This suggests that host country experience plays a larger role for firms moving in the upward direction than for firms moving in the downward direction. It is important to note, however, that the significance of the results depended on the performance measure used, suggesting that some aspects of performance are affected more significantly. This provides interesting opportunities for future research to investigate the effect of institutional distance and host country experience on both objective and subjective measures of performance. A detailed discussion of the findings and implications is provided in Chapter 5.
10
which the firm moves and the level of host country experience. Moreover, the findings regarding the moderating effect of host country experience contribute to the literature in two ways. First, they confirm suggestions that increased host country experience reduces costs associated with operating abroad. Second, they demonstrate that the moderating effect of host country experience varies depending on the direction in which the firm moves. Thus, this provides new theoretical insights regarding the effect of host country experience on institutional distance and subsidiary performance, which had not been clearly established by previous research. Additionally, the findings provide important considerations for international business managers to take into account when deciding upon a location for international expansion, because it has important implications for subsidiary performance.
1.5 – Structure
11 CHAPTER 2 – LITERATURE REVIEW
2.1 – The costs of doing business abroad
From the mid-1970s, scholars have started to recognise that multinational enterprises operating abroad face economic and social costs, which are not experienced by local firms in the host country. Hymer (1976) was the first scholar that introduced these ‘costs of doing business abroad’, and argued that firms looking to expand to a foreign environment faced barriers to entry because of their foreignness. That is, foreign firms face several types of disadvantages compared to local firms, which result in additional economic and social costs (Hymer, 1976). The economic costs can be termed activity-based costs and regard costs that are easily anticipated and quantified, such as transportation and communication costs, trade barriers, and costs associated with foreign exchange transactions (Eden and Miller, 2004). In addition to these economic costs, a firm operating in a foreign environment also faces social costs associated with its foreignness. These social costs are best captured by the concept of ‘liability of foreignness’, which was introduced by Zaheer (1995), who defined as it “all additional costs a firm operating in a market overseas incurs that a local firm would not incur” (Zaheer, 1995, p. 343). These additional costs arise from the unfamiliarity with the foreign business environment and from relational and discriminatory hazards that firms face in the host country (Eden and Miller, 2004). In sum, the liability of foreignness involves costs associated with an MNE’s (lack of) network linkages in the host country as well as the institutional distance between the home and host country (Zaheer, 2002). Eden and Miller (2004) argue that the key driver behind liability of foreignness is institutional distance.
2.2 – Institutions and institutional distance
2.2.1 – The institutional environment
12
enforcement, and the system of government (Gaur et al. 2007; North, 1990). Moreover, the normative and cognitive pillars refer to informal aspects of the institutional environment, meaning legitimate means of pursuing goals and the beliefs and value system of a society (Gaur et al. 2007; Scott, 1995; DiMaggio and Powell, 1983). Examples of these are the responsiveness of political systems to economic challenges, governance transparency, the importance of business networks and connections, and cultural differences (Gaur et al., 2007). Furthermore, Scott (1995) states that the normative and cognitive pillars are largely similar to each other, and can therefore be grouped together into one concept, as is also done in some more recent studies (e.g. Gaur and Lu, 2007; Chao and Kumar, 2010). In sum, institutional differences between countries are determined by differences in their regulative and normative structures.
13
countries, MNEs are subject to different institutions that can both facilitate and constrain their operations, affecting decision-making and performance.
2.2.2 – Institutional distance
14
Institutional distance is sometimes used interchangeably with psychic distance (e.g. Konara and Shirodkar, 2018). Psychic distance can be defined as “the sum of factors preventing or disturbing flows of information between firm and market” (Johanson and Valhne, 1977, p. 24). This concept was first introduced by Beckerman (1956), who identified it as a perceptual obstacle to trade in addition to geographic distance. Whereas institutional distance is concerned with regulative, normative, and cognitive differences in countries’ institutional profiles, psychic distance is a much broader concept that encompasses a larger set of factors that affect individuals’ perceptions, such as culture, language, religion, education, legislation, politics, economic conditions, market structure, and business practices (Ambos et al., 2019; Yildiz and Fey, 2016). According to Ambos et al. (2019), institutional distance is merely an antecedent of psychic distance perceptions. Moreover, psychic distance is manifested at the individual-level; it concerns the perceptions of individuals such as managers regarding the distance between the home and the host country in terms of culture, language, religion, business, and other factors (Yildiz and Fey, 2016; Ambos et al., 2019). In contrast, institutional distance is operationalised at the country-level and concerns differences in the institutional profiles of two countries. Differences in institutional profiles, however, affect psychic distance perceptions of managers. That is, the greater the institutional differences between the manager’s home and host country, the greater the perceived psychic distance will be (Ambos et al., 2019).
15
2.3 – Insights from institutional distance studies
Various studies have demonstrated how institutional distance affects MNE decision-making. For example, Gaur et al. (2007) investigated the relationship between institutional distance and subsidiary staffing decisions. Subsidiary staffing is a primary strategic means for MNEs to share knowledge, coordinate activities, and to exercise control over their subsidiaries (Gaur et al., 2007). As institutional distance between the home and the host country increases, the coordination and control of the subsidiary is affected, as well as the knowledge transfer between the MNE and its subsidiary. Gaur et al. (2007) found that with greater institutional distance, MNEs are more likely to employ parent country nationals as general managers of subsidiaries. This in turn affects subsidiary performance, as the employment of parent country nationals provides performance benefits through improved control and coordination (Gaur et al., 2007). However, it was also shown that this positive relationship between the employment of parent country nationals and subsidiary performance eventually decreases when the proportion of parent country nationals in the workforce becomes too high (Gaur et al., 2007).
16
over their operations, which can mitigate costs associated with distance (Gaur and Lu, 2007). Thus, Hernandez and Nieto (2015) contribute to the literature as they find a plausible explanation for the inconclusive results of existing studies by accounting for the direction of institutional distance in addition to the magnitude.
Furthermore, Xu and Shenkar (2002) proposed that regulative institutional distance negatively affects ownership strategy. That is, an MNE is more likely to choose higher levels of ownership when entering a host country that is less distant from the home country, whereas it will choose lower levels of ownership when the institutional distance is large. Moreover, Gaur and Lu (2007) suggest that institutional distance acts as a moderator in the relationship between ownership strategy and subsidiary survival, such that higher levels of ownership positively affect subsidiary survival when the institutional distance between the home and the host country is large (Gaur and Lu, 2007). That is, with greater institutional distance, MNEs can operate more effectively when they have tight control over their subsidiaries, i.e. high levels of ownership.
These findings demonstrate how institutional distance affects MNE decision-making and subsequently subsidiary performance. Additionally, the coordination and control of subsidiaries, as well as the transfer of knowledge between the parent and the subsidiary can also be affected by the degree of institutional similarity between an MNE’s home and host country (Kostova, 1999; Gaur, Delios, and Singh, 2007). Hence, institutional distance can present both opportunities and challenges to MNEs and their foreign subsidiaries, which in turn influence their performance in the host country (Gaur and Lu, 2007). The next section further elaborates on the relationship between institutional distance and subsidiary performance.
2.4 – Institutional distance and performance
17
2.4.1 – Negative effects of institutional distance
Traditionally, it was assumed that the larger the institutional distance, the more costs the MNE would incur, and the more difficult it would be for the subsidiary to operate effectively in the host country. For example, Kostova and Zaheer (1999) asserted that a larger institutional distance makes it more problematic for the MNE to establish legitimacy in the host country and to transfer strategic routines to foreign subsidiaries. Additionally, a larger institutional distance between the home and the host country may trigger conflicting demands for external legitimacy in the host country and global integration within the MNE system (Xu and Shenkar, 2002). Chao and Kumar (2010) suggested that, because of these difficulties, institutional distance negatively affects a firm’s performance as it continues its international expansion process. Likewise, Chao et al. (2012) researched the moderating effect of institutional distance on the international diversity-performance relationship and highlighted that regulative institutional distance creates the greatest obstacles for operating abroad, whereas normative institutional distance was shown to positively moderate the relationship between international diversity and performance. Moreover, Shirodkar and Konara (2017) investigated the relationship between regulatory institutional distance and subsidiary performance in emerging markets and found a significant, negative relationship. These studies show a negative relationship between institutional distance and performance.
2.4.2 – Positive effects of institutional distance
18
findings are in line with this psychic distance paradox, and with the findings of Evans and Mavondo (2002). They argued that firms moving to distant foreign environments face high levels of uncertainty, and therefore these firms are prompted to undertake extensive research and planning before entering the host country, which in turn positively affects their performance abroad.
As explained earlier, psychic distance is a broader concept than institutional distance concerning a wider range of home-host country differences. However, as institutional differences are an important antecedent of psychic distance (Ambos et al., 2019), the findings regarding the relationship between psychic distance and subsidiary performance provide some insight in the effects of institutional differences as well. Thus, these findings demonstrate that there might also be positive performance effects as a result of institutional distance between the home and the host country. However, these ‘positive’ findings all indicate that managers must prepare for the institutional distance and gain extensive knowledge about the host country, i.e. pre-entry experience, to compensate for the liabilities of foreignness and thereby perform well. Nonetheless, this emphasis on liabilities of foreignness has increasingly been balanced by studies suggesting that foreignness may actually provide MNEs’ subsidiaries with unique advantages (Edman, 2016), which will be elaborated on next.
2.4.3 – Assets of foreignness
19
institutional asymmetries (Martin, 2014; Mallon and Fainshmidt, 2017). If local competitors cannot easily access or imitate the benefits arising from these institutional asymmetries, this may provide the MNE with unique assets of foreignness (Mallon and Fainshmidt, 2017). For example, if the subsidiary has gained access to particular technological infrastructure as a result of advanced technological conditions in the MNEs’ home country, it can develop firm-specific technological resources that local firms in the host country cannot duplicate because they do not have access to the required infrastructure (Mallon and Fainshmidt, 2017).
Thus, different institutions support the development of different capabilities and competencies. The more different the institutional systems of two countries, i.e. the larger the institutional distance, the more different the capabilities of firms from these two countries, and the larger the potential for creating assets of foreignness through institutional arbitrage (Martin, 2014; Mallon and Fainshmidt, 2017; Jackson and Deeg, 2008). It is important to note, however, that assets of foreignness do not necessarily replace liabilities of foreignness. That is, MNEs’ subsidiaries can simultaneously face costs related to liabilities of foreignness, as well as benefits related to assets of foreignness (Mallon and Fainshmidt, 2017). Moreover, while assets of foreignness may be used to offset liabilities of foreignness, they can also be utilised to create value for the MNE’s subsidiary in the host country, which in turn can affect the subsidiary’s performance. This demonstrates that foreignness and institutional distance do not necessarily have to affect performance in a negative fashion, but that there can also be positive performance implications as the result of foreignness due to opportunities for arbitrage. Based on this discussion, I argue that institutional distance positively affects subsidiary performance, which leads to the development of the following hypothesis:
Hypothesis 1: Institutional distance positively influences subsidiary performance.
2.4.4 – Adaptation and arbitrage
20
other hand, as institutional distance increases, adaptation costs increase as well, which implies that subsidiaries in the host country bear costs as they need to learn about the host country environment, while at the same time ‘unlearning’ certain aspects of embedded knowledge that is imprinted from the home country environment (Konara and Shirodkar, 2018). Moreover, Gaur and Lu (2007) investigated the relationship between institutional distance and subsidiary survival and found an inverted U-shaped relationship, meaning that the chances of survival initially increase, but as the institutional distance between the home and host country further increases, the chances of survival eventually decline. They argue that institutional distance can present the MNE with opportunities for institutional arbitrage, as the host country institutional environment may provide better opportunities for certain firm-specific capabilities. Simultaneously, institutional distance presents the MNE with additional costs due to the unfamiliarity with the new environment, which causes competitive disadvantage (Gaur and Lu, 2007). When institutional distance is small to medium, the potential benefits from institutional arbitrage should outweigh the costs of doing business. However, as the institutional distance between the home and the host country increases, the costs accelerate and may eventually exceed the benefits, which affects the continuity of the MNE’s operations (Gaur and Lu, 2007). This discussion suggests that the performance implications of institutional distance depend, to a certain extent, on the interplay between adaptation costs and arbitrage opportunities.
Nonetheless, the results in the literature vary significantly, and the findings are inconclusive. These contradictory results raise questions about the determinants of subsidiary performance and indicate that there may be other factors at play that may be affecting the way in which institutional distance influences subsidiary performance.
2.5 – Asymmetry of distance
21
regulatory institutional distance has conducted its analysis in absolute terms, considering only the magnitude of the distance without taking the direction into account (Hernández and Nieto, 2015). The magnitude refers to the extent of institutional distance, whereas the direction indicates whether MNEs move from a stronger, better developed institutional environment to a relatively weaker, less developed institutional environment, or vice-versa. Konara and Shirodkar (2018) argue that the magnitude and direction of regulatory institutional distance have varying effects. They demonstrate that MNEs moving in an ‘upward’ direction, to a stronger institutional environment compared to their home country, face negative effects on subsidiary performance. In contrast, MNEs moving in a ‘downward’ direction, to a weaker institutional environment compared to their home country, are likely to face positive effects on subsidiary performance due to a positive outcome of arbitrage effects and adaptation costs (Konara and Shirodkar, 2018).
In line with these findings, they argue that subsidiaries of MNEs from a stronger institutional environment are ‘imprinted’ with better market- and non-market-based capabilities, which is advantageous when competing with other subsidiaries in the host country. That is, with greater institutional distance, these subsidiaries can benefit from a greater positive home-imprinting effect (Konara and Shirodkar, 2018). In contrast, subsidiaries of MNEs from relatively weaker, less developed, institutional environments may face a competitive disadvantage compared to subsidiaries from relatively stronger institutional environments, as weak home institutions cannot provide the necessary resources and infrastructure to be able to compete effectively internationally (Gillespie and Teegen, 1996). Thus, the arbitrage opportunities are greater for subsidiaries operating in weaker institutional environments compared to their home country than for subsidiaries operating in stronger institutional environments compared to their home country.
22
effectively. On the other hand, subsidiaries operating in weaker environments compared to their home country, face these unlearning costs to a lesser extent, as their market- and nonmarket capabilities (e.g. superior management techniques and advanced technologies) are generally viewed as positive capabilities. That is, they are to a lesser extent expected to unlearn these capabilities (Cuervo-Cazurra and Genc, 2011). Thus, the adaptation costs are greater for subsidiaries operating in stronger institutional environments compared to their home country than for subsidiaries operating in weaker institutional environments compared to their home country. This discussion leads to the development of the following hypotheses:
Hypothesis 2a: Upward institutional distance negatively influences subsidiary performance.
Hypothesis 2b: Downward institutional distance positively influences subsidiary performance.
2.6 – Host country experience
23
experience, as “international expansion is inhibited by the lack of knowledge about markets” (Forsgren and Johnson, 1992, p. 10).
Moreover, according to the mere exposure effect, or the ‘familiarity principle’, people become more attracted to a certain object the more they are exposed to it (Zajonc, 2001). In line with the familiarity principle, it could be argued that the longer managers are exposed to a particular foreign setting, the more they will perceive it as attractive. Ambos et al. (2019) demonstrate that managers’ distance perceptions will diminish as they are increasingly exposed to a certain host country environment. As the managers spend more time in the foreign environment, they are more likely to understand its customs, they will be more familiar with the local institutions and they will possess a deeper knowledge of how business is done in that country (Ambos et al., 2019). As a consequence, they will perceive the institutional distance between the home and the host country to be smaller, and they will likely face less hurdles when continually operating in the host country, which in turn may affect subsidiary performance. Additionally, Ambos et al. (2019) also demonstrated that general international experience was not sufficient; country-specific experience is essential in order to mitigate the perceived home-host country distance. Therefore, I argue that host country experience moderates the relationship between institutional distance and subsidiary performance. This leads to the development of the following hypothesis:
Hypothesis 3: Host country experience positively moderates the relationship between institutional distance and subsidiary performance such that this relationship is strengthened, i.e. becomes more positive with greater host country experience.
24
knowledge in a firm’s internationalisation process, as it implies that the firm’s psychic distance is reduced as a result of an increased understanding of the specific market and its characteristics.
Furthermore, the local knowledge base that is accumulated through experience in the host country environment can be very valuable to a subsidiary and its corporate headquarters in order to address the dual pressures of local adaptation and global integration (Gaur et al., 2007). That is, experience plays an important role in balancing legitimacy and fostering coordination, control, and knowledge transfer, which in turn reduces the effects of distance. With experience, the subsidiary will develop better capabilities to overcome unfamiliarity and relational hazards arising from a lack of knowledge of the local environment and difficulty in managing relationships at a distance (Henisz and Williamson, 1999; Gaur et al., 2007). As a result, subsidiary legitimacy is enhanced. Additionally, the longer the subsidiary operates in the host country environment, the greater its knowledge about the local culture and the institutional environment of that country will be. This makes it easier to adapt the strategic organisational practices to the local environment, as the MNEs’ ease of adjustment depends on a country’s institutional profile (Gaur and Lu, 2007; Ionascu et al. 2007; Xu and Shenkar, 2002). Thus, with increased host country experience, it is likely that adaptation costs decrease as the result of increasing host country knowledge.
25
the value-creating potential of these assets of foreignness will decline over time (Mallon and Fainshmidt, 2017). This in turn may affect the subsidiary’s performance.
Shirodkar and Konara (2017) investigated the moderating impact of host country-experience on the relationship between institutional distance and subsidiary performance in emerging markets, and found that with greater host-country experience, subsidiaries were able to mitigate the negative effect of institutional distance on performance. However, this study did not take into account the direction of the distance, and focused solely on emerging markets. Hernández and Nieto (2015) and Konara and Shirodkar (2018) have highlighted the importance of an asymmetrical effect of institutional distance. Nonetheless, in the current literature, there is still a lack of studies that investigates the combined effect of the magnitude and the direction of institutional distance on subsidiary performance. Additionally, Konara and Shirodkar (2018) suggested that host country experience might affect institutional distance in both directions. However, this has not been clearly established by previous research yet. Therefore, this dissertation investigates whether there is a moderating impact of host country experience on the relationship between institutional distance and subsidiary performance, considering both the magnitude and the direction of the distance. Additionally, a wide variety of home and host countries is included. This provides a deeper insight in the relationship between host country experience, institutional distance, and subsidiary performance.
26
costs by obtaining the relevant knowledge required to compete effectively with local firms and by unlearning their home-based practices through ongoing activities in the host country environment.
In contrast, subsidiaries of MNEs moving to a host country environment with weaker institutions compared to their home country environments face less adaptation costs, as they possess superior capabilities and technologies which they do not have to ‘unlearn’. Additionally, the level of potential arbitrage benefits is much higher (Konara and Shirodkar, 2018). While greater experience may decrease the marginal benefits from institutional arbitrage (Gaur and Lu, 2007), it also decreases adaptation costs, meaning that there will still be positive performance implications. This discussion leads me to the development of the following hypotheses. The hypothesised relationships can be seen in the Figure 1 below.
Hypothesis 4a: Host country experience positively moderates the relationship between upward institutional distance and subsidiary performance such that this relationship is weakened, i.e. becomes less negative with greater host country experience.
Hypothesis 4b: Host country experience positively moderates the relationship between downward institutional distance and subsidiary performance such that this relationship is strengthened, i.e. becomes more positive with greater host country experience.
27 CHAPTER 3 – DATA AND METHODS
3.1 – Research philosophy statement
According to Johnson and Clark (2006), it is important in business and management research to be aware of the research philosophy used throughout the research as this has an impact on the research process. This research is based on the ontological position of objectivism, which asserts that “social phenomena and their meanings have an existence that is independent of social actors” (Bryman, 2001, pp. 16-18). This implies that more focus is placed on objective facts rather than subjective meanings behind phenomena. Furthermore, this research takes the epistemological position of positivism, which “advocates the application of the methods of the natural sciences to the study of social reality and beyond” (Bryman, 2001, pp. 12-13). This is also reflected in the methodology and methods; a deductive approach is used, with a focus on observable and measurable facts and regularities. In order to explain and predict the relationship between directional institutional distance, host country experience, and subsidiary performance, hypotheses were developed based on existing theory (Chapter 2), which were tested to be able to draw conclusions. Thus, a quantitative method was used. During the process bias was eliminated as far as possible. This was done, for example, by using simple random sampling, by including an appropriate amount of control variables, by transforming highly skewed variables, and by using robust standard errors to overcome heteroscedasticity. The next paragraphs outline in detail how the sample was developed, how the data were collected, and which methods were used to test the hypotheses.
3.2 – Data collection and sample development
28
World Development Indicators, following previous literature (e.g. Konara and Shirodkar, 2018; Shirodkar and Konara, 2017; Ambos et al., 2019; Dikova, 2009; Cuervo-Cazurra and Genc, 2008).
3.2.1 – Sample development
To create the sample, I executed various search steps to ensure that several criteria were met. That is, I sought for firms that (1) are ultimately owned by a corporate parent company (2) that is from a country different than the operating country of the subsidiary. It was assumed that a firm with ultimate ownership (> 51%) has the authority for the final decision making, and therefore this can be considered as the parent firm. Additionally, (3) the parent firm should be from a different country than the subsidiary to ensure that there is some degree of institutional distance between the home and the host country, as different countries have different institutional characteristics and different levels of institutional development. Furthermore, (4) I only included firms that have a known value for Return on Equity and Return on Assets for the years 2013-2016. This was done because these ratios were required for the dependent variable, as discussed in the next paragraph. The years 2013-2016 were chosen because it yielded the highest amount of available data. Finally, (5) I only included firms with a size classification of ‘large’ or ‘very large’ (defined by Bureau van Dijk based on total sales, operating revenue and employment). Small- and medium-sized firms (SMEs) were ruled out because it is shown that these firms generally face additional constraints in the internationalisation process because of a lack of managerial expertise and competence, which affects their performance (Pangarkar, 2008). Executing these search steps eventually yielded 48.386 firms. To create an appropriate sample to work with, I randomly sampled 2.000 firms from this search ‘population’ using the random sampling function in the ORBIS database. After cleaning the dataset, the final sample consisted of 1.757 subsidiaries distributed over 64 host countries and originating from 65 home countries. An overview of this distribution can be seen in Table A1 in the appendix.
3.3 – Measures
3.3.1 – Dependent variable
29
(ROA) as a measure of subsidiary performance (e.g. Chao and Kumar, 2010; Chao et al., 2012). Asset turnover depends on the market value of assets, which may vary because of cross-country differences in the market value of assets (Chan et al., 2010). Therefore, ROE is a slightly more appropriate measure of performance in cross-country settings. Nonetheless, both ratios are commonly used measures of subsidiary performance. For the purpose of supplementary analysis, I also obtained the ROA figures for each subsidiary. The ROE and ROA ratios for the years 2013-2016 were extracted from the ORBIS database.
3.3.2 – Independent variables
The main independent variable used in the model is the regulatory institutional distance between the host country, where the subsidiary is operating, and the home country, where the parent originates from. Various measures of institutional distance have been developed, including the Dow index (Dox and Karunaratna, 2006), Hotho’s indices (Hotho, 2009), and indicators from the World Competitive Yearbook and the International Country Risk Guide (Hahn et al., 2009) as used by Gaur and Lu (2007). Following previous literature (e.g. Konara and Shirodkar, 2018; Shirodkar and Konara, 2017; Ambos et al., 2019; Dikova, 2009; Cuervo-Cazurra and Genc, 2008), institutional distance was operationalised using the Worldwide Governance Indicators, produced by Kaufmann et al. (2007). The Worldwide Governance Indicators provide estimates regarding six dimensions of governance for more than 200 countries and territories over the period of 1996-2018. The six dimensions of governance are Voice and Accountability, Political Stability and Absence of
Violence, Government Effectiveness, Regulatory Quality, Rule of Law, and Control of Corruption.
30
Table 3.1 – Correlation analysis on the six dimensions of governance
(1) (2) (3) (4) (5) (6) (1) Voice and Accountability 1.0000 (2) Political Stability Abs. of Violence 0.7344 1.0000 (3) Government Effectiveness 0.7289 0.6970 1.0000 (4) Regulatory Quality 0.7313 0.6287 0.9233 1.0000 (5) Rule of Law 0.8171 0.7892 0.9239 0.8921 1.0000 (6) Control of Corruption 0.7925 0.7900 0.9023 0.8430 0.9425 1.0000
Table 3.2 – Factor analysis on the six dimensions of governance
Variable Factor loading Cronbach’s alpha
31
The composite variable that was computed as a result from the factor analysis indicates the institutional profile score for each home and host country. Consequently, following Konara and Shirodkar (2018), I calculated the institutional distance (ID) between the home and the host country by subtracting the host country institutional profile score from the home country institutional profile score (ID = INPhome – INPhost). Institutional distance can also be calculated using other methods, such as the Mahalanobis method, which is a popularly used method to construct composite distance variables. However, this method drops the information on the directionality in the process (Konara and Shirodkar, 2018). Because I am interested in the differential effects of upward- and downward institutional distance, I used the method proposed by Konara and Shirodkar (2018).
The resulting institutional distance values range from -2.687628 to 3.184264. A positive institutional distance score represents firms moving in the downward direction, i.e. moving to a relatively weaker institutional environment compared to their home country. A negative institutional distance score represents firms moving in the upward direction, i.e. moving to a relatively stronger institutional environment compared to their home country. Moreover, following Konara and Shirodkar (2018), I partitioned the institutional distance measure into two vectors in order to capture the effect of these two directions of institutional distance. The two vectors represent the upward and the downward direction as follows:
IDu = |ID| if ID < 0 IDu = 0 otherwise
IDd = |ID| if ID > 0 IDd = 0 otherwise
32
3.3.3 – Moderator
In order to examine whether host country experience affects the relationship between institutional distance and subsidiary performance, host country experience was included as a moderator. Following Shirodkar and Konara (2017), host country experience was measured using the age of the subsidiary. This refers to the number of years it has been operating in the host country, i.e. the number of years since its incorporation in the host country. Subsidiary age is also an appropriate measure of host country experience from the subsidiary perspective across different investment types (Shirodkar and Konara, 2017). When an MNE enters a host country by acquiring a local firm, the local firm’s age contributes to the subsidiary’s host country experience because the local firm is already deeply embedded in the local institutional context (Shirodkar and Konara, 2017). However, when an MNE enters the host country through a greenfield investment, i.e. by establishing a new subsidiary from scratch, the subsidiary is relatively new to the foreign institutional environment (Shirodkar and Konara, 2017). Therefore, the number of years since its incorporation in the host country effectively represents the subsidiary’s host country experience because it demonstrates its degree of embeddedness in the host country.
3.3.4 – Control variables
The aim of this study was to discover the moderating effect of host country experience on the relationship between institutional distance and subsidiary experience. In order to capture this relationship effectively, several control variables were included, guided by previous literature and empirical evidence, as will be further discussed below.
Firm-level control variables
33
MNE internationalisation, i.e. international experience, is included as a control variable, because knowledge obtained abroad is generally considered as a firm-specific resource that may lead to better performance. A proxy for international experience used in previous literature (Dikova, 2009) is the number of subsidiaries an MNE has established worldwide. I measured this using data from the ORBIS database regarding the number of firms in the corporate group, which is a count variable that includes both foreign and domestic subsidiaries. This control variable would have been more accurate if domestic subsidiaries were excluded, however, due to a lack of data availability (e.g. regarding the ratio of total sales/foreign sales), I used this measure as an alternative.
Country-level control variables
To control for (socio-)economic conditions in the host country that may inhibit or foster the subsidiary’s performance, I included some country-level control variables. First, to control for the standard of living and the economic growth of the host country, I included control variables for GDP per capita and GDP growth, obtained from the World Bank’s World Development Indicators (World Bank, n.d.). Additionally, following previous research (e.g. Konara and Shirodkar, 2018; Shirodkar and Konara, 2017), I controlled for communications infrastructure and human capital. Communications infrastructure was measured using the number of mobile connections per 100 people in the host country. Human capital was measured using the gross percentage of secondary school enrolments of the host country. Both measures were provided by the World Bank’s World Development Indicators.
Inter-country control variables
34
Moreover, to control for normative institutional distance, I also included a control variable for common language and colonial history, measured using CEPII data. Colony is a dummy variable that measures whether the two countries have ever been in a colonial relationship with each other, that takes on the value of 1 if this is the case, and 0 if not (Mayer and Zignago, 2011). Common language measures whether a particular language is spoken by at least 9% of the population in both countries, taking on the value of 1 if this is the case, and 0 if this is not the case (Mayer and Zignago, 2011).
3.4 – Descriptive statistics
Descriptive statistics of all variables included in the model can be seen in Table 3.3 below.
Table 3.3 – Descriptive statistics
Obs. Mean SD Min Max
35
3.5 – Analysis approach
To test the relationship between directional institutional distance and subsidiary performance, and the moderating effect of host country experience on this relationship, a generalised least squares (GLS) regression analysis was conducted in a panel data framework, while controlling for individual-specific unobserved effects. To determine which estimator to utilise in the regression model, I performed a Hausman specification test, which tests the similarity between fixed and random effects estimators (Hausman, 1978). The test showed a Chi-square value of 38,97, which was significant at the P<0.05 level. This implies that the null hypothesis of no correlation is rejected, suggesting that there is some degree of correlation between individual-specific effects and some of the explanatory variables, which means that the fixed effects model is the preferred model (Park, 2011; Hausman, 1978). However, I am also interested in the impact of several time-invariant control variables; management size, MNE internationalisation, geographic distance, colonial history, and common language, which are omitted in the fixed effects model. The use of a random effects GLS model is sometimes preferred in this case (Longi and Nandi, 2017). A random effects GLS model reduces the number of parameters to be estimated, which makes it more efficient, but it produces inconsistent, biased estimates when individual-specific effects are correlated with explanatory variables (Greene, 2008; Mundlak, 1978).
36
institutional distance variables. The moderator variable, subsidiary age, i.e. host country experience, enters the regression models in two ways; individually and as an interaction with the institutional distance variables. A detailed description of the variables that are included in the regression models can be found in Table 3.4 below.
Table 3.4 – Description of variables included in the model
Variable Description/measurement
ROE Return on equity, representing firm performance ROA Return on assets, representing firm performance
Institutional distance (ID) Composite measure of institutional distance, computed using institutional profile scores resulting from factor analysis on six indicators of governance from Worldwide Governance Indicators Downward institutional distance
(IDd)
Downward directional vector of institutional distance
Upward institutional distance (IDu)
Upward directional vector of institutional distance
Subsidiaryage Age of the subsidiary, i.e. the number of years the subsidiary has been operating in the host country, representing host country experience
Subsidiaryage*ID Interaction of subsidiary age with the composite institutional distance variable
Subsidiaryage*IDd Interaction of subsidiary age with the downward directional institutional distance variable
Subsidiaryage*IDu Interaction of subsidiary age with the upward directional institutional distance variable
Firmsize Firm size, measured using the log value of the firm’s total assets lnManagementsize Log value of the number of directors and managers of the firm lnMNEinternationalisation Log value of the number of firms in the corporate group,
representing the international experience of the MNE
GDPpc GDP per capita in the host country
GDPgrowth GDP growth percentage in the host country
37
Table 3.4 (continued)
Humancapital Human capital, measured using the gross percentage of secondary school enrolments of the host country.
Geodistance Geographic distance between the home and the host country, measured using bilateral country distance between the most populated cities of two countries.
Colony Colonial history, measures whether the two countries have ever been in a colonial relationship with each other, takes on the value of 1 if this is the case and 0 if this is not the case.
38 CHAPTER 4 – RESULTS
4.1 – Correlation
Correlation coefficients of the variables included in the model are presented in Table 4.1 below. As expected, the correlation between the composite institutional distance variable and the two directional institutional distance variables is moderate to high, with a correlation coefficient of 0.665 for upward institutional distance and 0.929 for downward institutional distance. This is not an issue, because I included them in separate models. Moreover, some control variables show correlation coefficients that indicate moderate correlation. However, after calculating variance inflation factors I concluded that there was no evidence of multicollinearity, as the values were all well below 4, which is a threshold often used by researchers as an indication of multicollinearity (O’Brien, 2007).
4.2 – Regression models
36
Table 4.1 – Correlation between the variables included in the models
36
Table 4.2 – GLS regression results on dependent variable ROE (return on equity)
VARIABLES Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
41
Table 4.2 (continued)
Constant -0.696 -0.108 -0.611 -0.515 -0.910 -0.440
Observations 6616 6616 6616 6616 6616 6616
Prob. > chi2 (Sig.) 0.000 0.000 0.000 0.000 0.000 0.000
R2 0.630 0.630 0.630 0.630 0.630 0.630
42
4.2.1 – Results Model 1
In Model 1, the effect of institutional distance on return on equity (ROE) was tested using the composite variable of institutional distance. The variable ‘ID’ showed a coefficient of 12.461, significant at the p<0.1 level after controlling for firm size, management size, MNE internationalisation, GDP per capita, GDP growth, infrastructure, human capital, geographic distance, colonial history, and common language. This indicates that there is a significant positive relationship between institutional distance and return on equity, i.e. subsidiary performance. Moreover, most control variables had insignificant coefficients, however, firm size showed a coefficient of 9.322, significant at the p<0.01 level. MNE internationalisation showed a significant coefficient of 0.049 (p<0.1), and GDP growth showed a coefficient of 0.791, significant at the p<0.01 level. Finally, geographic distance showed a significant coefficient of -0.000 (p<0.1). In sum, these results provide significant evidence for a positive relationship between institutional distance and subsidiary performance, supporting Hypothesis 1.
4.2.2 – Results Models 2 and 3
43
4.2.3 – Results Model 4
In Model 4, the moderating effect of host country experience (subsidiary age) was tested on the relationship between institutional distance and return on equity (ROE), using the composite variable of institutional distance (ID). The variable ID showed a positive coefficient of 6.251 however, this coefficient was insignificant. Subsidiary age showed an insignificant coefficient of 0.529, and the interaction between subsidiary age and ID showed a coefficient of 0.294, also insignificant. This indicates that, after including the variable subsidiary age in the model, the relationship between institutional distance and ROE turns insignificant. While the sign of the coefficient of subsidiary age is in line with Hypothesis 3, there is not sufficient evidence to support this hypothesis. Thus, there is no statistically significant moderating effect of subsidiary age on the relationship between institutional distance and ROE. Again, most control variables showed insignificant coefficients. However, firm size showed a coefficient of 9.497, significant at the p<0.01 level. MNE internationalisation showed a significant coefficient of 0.046 (p<0.1), and GDP growth was significant at the p<0.05 level with a coefficient of 0.671. Finally, geographic distance showed a coefficient of -0.000, significant at the p<0.1 level. Overall, these results provide insufficient evidence to support Hypothesis 3. That is, there is no statistically significant moderating effect of host country experience on the relationship between institutional distance and subsidiary performance.
4.2.4 – Results Model 5 and 6
44
performance. Furthermore, in Model 5, firm size showed a significant (p<0.01) coefficient of 9.406. GDP growth showed a coefficient of 0.682, significant at the p<0.05 level. The control variable colony showed a significant (p<0.1) coefficient of 0.231. In Model 6, firm size showed a coefficient of 9.421, which was significant at the p<0.01 level. MNE internationalisation showed a coefficient of 0.044, and geographical distance showed a coefficient of -0.000, both significant at the p<0.1 level. Finally, GDP growth showed a significant coefficient of 0.667 (p<0.05). The remaining control variables had insignificant coefficient in both models. In sum, the results do not provide sufficient evidence to support Hypotheses 4a and 4b, suggesting no significant moderating effect of host country experience on the relationships between the directional institutional distance variables and subsidiary performance.
4.2.5 – Results control variables
The results regarding the control variables demonstrate that firm size significantly affects the performance of subsidiaries in a positive way, meaning that a larger firm size results in a larger return on equity. This suggests that larger firms generally perform better as compared to smaller firms. Additionally, in most models, MNE internationalisation positively affected return on equity, suggesting that subsidiaries of a highly internationalised MNE perform better. Moreover, the percentage of GDP growth in the host country significantly affected the return on equity of subsidiaries. This suggests that in host countries with a larger economic growth, subsidiaries perform better. Finally, in most models, geographic distance significantly affected return on equity in a negative way, which suggests that a larger geographic distance between the home and the host country slightly reduces subsidiaries’ performance.
4.3 – Alternative tests
The results from the initial models do not provide sufficient evidence to support the hypotheses, however, many of the coefficients’ signs are in line with the hypotheses. Therefore, I performed some alternative tests to examine the robustness of the results.
4.3.1 – Direction of institutional distance as a dummy
45
between institutional distance and subsidiary performance, depending on the direction of the distance. However, a more simplified approach to showing the effect of the direction of the distance is to compute a dummy variable for institutional distance direction, taking on the value of 1 when institutional distance is positive (i.e. downward direction), and the value of 0 when institutional distance is negative (i.e. upward direction). This measure demonstrates whether the direction of the distance affects subsidiary performance, however, it does not provide an overview of the differential effects of upward- and downward institutional distance. The regression coefficients following from this test can be seen in Table A2 in the appendix. Models 1 and 2 test the relationship between the composite variable of institutional distance, subsidiary age, and subsidiary performance and provide the same results as in the main models (Models 1 and 4) described above. Model 3 tests the relationship between the institutional distance directional dummy and subsidiary performance (return on equity), and Model 4 tests the moderating effect of host country experience on this relationship. In Model 3, IDdirectiondummy shows a coefficient of 3.222, which is insignificant. Furthermore, firm size has a significant coefficient of 9.298 (p<0.01). GDP growth has a coefficient of 0.783, significant at the p<0.01 level. Geographical distance has a significant coefficient of -0.000 (p<0.1), and colony has a significant coefficient of 0.213 (p<0.1). In Model 4, IDdirectiondummy shows a coefficient of 0.318, which is insignificant as well. Subsidiary age shows an insignificant coefficient of 0.559, and the interaction between the institutional distance direction dummy and subsidiary age is insignificant as well, with a coefficient of 0.113. Moreover, firm size shows a coefficient of 9.402, significant at the p<0.01 level. GDP growth shows a significant coefficient of 0.669 (p<0.05), and colony shows a coefficient of 0.218, significant at the p<0.1 level. Overall, the results of this alternative test are not notably different from the results of the initial regression models.
4.3.2 – Alternative performance measure
46
Model 1 tested the relationship between the composite institutional distance variable (ID) and return on assets. ID showed a coefficient of 3.716, significant at the p<0.05 level, suggesting a significant positive relationship between institutional distance and return on assets, i.e. subsidiary performance. This supports Hypothesis 1. Models 2 and 3 tested the relationship between upward institutional distance (IDu) and ROA and downward institutional distance (IDd) and ROA, respectively. Model 2 showed a coefficient of 4.888 for IDu, significant at the p<0.1 level. While this coefficient was significant, the sign contrasts with Hypothesis 2a, which predicted a negative effect. Therefore, Hypothesis 2a was not supported. Nonetheless, the results indicate that there is a significant positive relationship between upward institutional distance and subsidiary performance. Moreover, Model 3 showed a coefficient of 3.283 for IDd, also significant at the p<0.1 level. This indicates that there is a significant positive relationship between downward institutional distance and subsidiary performance, supporting Hypothesis 2b.
47
distance and ROA. In this model, IDd had an insignificant coefficient of 0.878. Subsidiary age had a coefficient of 0.248, significant at the p<0.1 level, suggesting a positive effect of subsidiary age on return on assets. However, the interaction of subsidiary age and downward institutional distance had an insignificant coefficient of 0.116, providing insufficient evidence to support Hypothesis 4b.
48
Table 4.3 – GLS regression results on dependent variable ROA (return on assets)
VARIABLES Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
49
Table 4.3 (continued)
Constant 0.027 -0.146 0.098 0.113 -0.063 0.17
Observations 6616 6616 6616 6616 6616 6616
Prob. > chi2 (Sig.) 0.000 0.000 0.000 0.000 0.000 0.000
R2 0.661 0.661 0.661 0.662 0.662 0.662
50 CHAPTER 5 – DISCUSSION
This study was inspired by the recognition that institutional distance is an asymmetrical concept, which challenges the traditional assumption of a symmetrical effect of institutional distance on subsidiary performance, regardless of whether a firm moves to a relatively stronger or weaker institutional environment compared to its home country (Konara and Shirodkar, 2018). Subsidiary performance is not only affected by the magnitude of institutional distance, but the direction in which the firm moves also matters. The implications of the results will be elaborated on below, followed by a discussion of the contributions and the limitations. Based on the limitations, some suggestions for further research are provided.
5.1 – Implications
51
weaker institutional environments often set up research and development (R&D) centres in more developed countries, such as the United States, as stronger institutional environments like these provide better support for the protection of innovative technologies (Gaur and Lu, 2007). These arbitrage benefits may outweigh the adaptation costs that these firms face during their internationalisation process.
Furthermore, the moderating effect of host country experience on the relationship between institutional distance and subsidiary performance was also examined. The results showed a strong and positive effect of host country experience on subsidiary performance. In the upward direction, host country experience was shown to moderate the relationship between institutional distance in a positive manner. However, the main relationship between upward institutional distance and subsidiary performance was hypothesised to be negative, but turned out to be positive. Therefore, the hypothesis of a positive moderating effect could only be partially supported. Nonetheless, the results demonstrate that the positive relationship between upward institutional distance and subsidiary performance becomes stronger, the longer the subsidiary operates in the host country. Moreover, host country experience did not significantly moderate the relationship between downward institutional distance and subsidiary performance. Thus, host country experience moderates the relationship between institutional distance and subsidiary performance more significantly in the upward direction than in the downward direction. Firms moving in an upward direction face positive performance implications as the result of a larger institutional distance between the home and the host country, and this positive effect on performance increases even further with greater host country experience. This moderating effect of host country experience does not apply to firms moving in a downward direction, meaning that their performance might not further increase with greater experience in the host country. In sum, host country experience plays a larger role for firms moving in the upward direction than for firms moving in the downward direction.
