Managing Corporate Social Responsibility in a cross-cultural context : a multiple case study of Dutch multinational enterprises in China

Managing Corporate Social Responsibility in a Cross-Cultural

Context: A Multiple Case Study of Dutch Multinational

Enterprises in China

MSc in Business Administration – Strategy Track

University of Amsterdam, Amsterdam Business School

Supervisor: dr. Arno Kourula

Author: Sytze van den Essenburg

Student number: 10691081

Date: 24-06-2016

Statement of originality

This document is written by student Sytze van den Essenburg 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.

Abstract

Managing corporate social responsibility (CSR) is a complex process for multinational enterprises (MNEs), especially when there are vast national cultural differences between the headquarters and foreign subsidiary. Even though pervious CSR literature has examined the topic from either the home country or host country, the engagement of MNEs in CSR from a cross-country perspective has received little academic attention. The purpose of this study is to investigate how, from the institutional level, national cultural differences between the headquarters and foreign subsidiary affect the management of CSR strategies by MNEs. Moreover, this research examines to which extent the engagement of CSR by foreign subsidiaries can contribute to the creation of external legitimacy and reduce the MNE’s liability of foreignness (LOF). These research gaps are addressed by adopting a qualitative multiple case study research design of three Dutch MNEs with a foreign subsidiary in China. In total, 14 interviews are conducted with both Dutch and Chinese managers responsible for CSR. Following the within- and cross-case analysis, a new multi-level model is developed. Based on the findings, the model contributes to existing literature by outlining the (1) antecedents, (2) implementation and, (3) outcomes of CSR strategies managed by Dutch MNEs in China. This study provides valuable theoretical and managerial insights and discusses avenues for future research.

Keywords: Corporate Social Responsibility, CSR, CSR Strategy, Multinational Enterprise, MNE, National Culture, Liability of Foreignness, Legitimacy, Standard, Local, Transnational

Acknowledgments

First of all, I would like to thank my supervisor Arno Kourula for his support throughout this journey. Your insights and extensive knowledge on CSR have helped me out a lot. Secondly, I owe an enormous amount of gratitude to the gatekeepers and interviewees of the researched companies. Writing this thesis would not have been possible without your help and patience. Last but not least I would like to thank my family and dear friend Chaoran Yang. Your encouragement and moral support have been invaluable.

Table of contents

Statement of originality ... ii

Abstract ... iii

Acknowledgments ... iv

Table of contents ... v

1. Introduction ... 1

1.3 Contributions and Research Question ... 2

1.4 Key Definitions ... 3

2. Literature Review ... 4

2.1 Corporate Social Responsibility ... 5

2.2 National Culture within an Institutional Context ... 6

2.2.1 Power Distance ... 9

2.2.2 Individualism/collectivism ... 10

2.2.3 Masculinity/Femininity ... 10

2.2.4 Uncertainty avoidance ... 11

2.2.5 Long-Term Orientation (Confucius Dynamism) ... 12

2.3 Corporate Social Responsibility Strategies for Multinational Enterprises ... 12

2.3.1 The Standard Corporate Social Responsibility Strategy ... 14

2.3.2 The Local Corporate Social Responsibility Strategy ... 14

2.3.3 The Transnational Corporate Social Responsibility Strategy ... 15

2.4 Corporate Social Responsibility in the Netherlands ... 16

2.5 Corporate Social Responsibility in China ... 16

2.6 Legitimacy and Liability of Foreignness ... 17

2.7 Developing a Preliminary Conceptual Framework ... 19

2.7.1 Antecedent – Cultural Sensitivity of the Headquarters ... 19

2.7.2 Implementation – Managing the CSR Strategy in China ... 19

2.7.3 Outcomes – Creating Legitimacy to Reduce Liability of Foreignness ... 20

3. Research Methodology ... 20

3.1 Research Philosophy –Interpretive Epistemology ... 21

3.2 Research Strategy – Multiple Case Study ... 21

3.3 Data collection ... 23

3.4 Data Analysis ... 25

3.5 Quality of the Research ... 26

3.5.1 Reliability ... 26

3.5.2 Validity ... 27

4. Results ... 29

4.1 Within-case results ... 29

MNE 1 – International Beverages Company (BevCo) ... 29

MNE 2 – International HR Service Company (HRCo) ... 34

5. Discussion ... 45

5.1 Main Findings and the Development of Propositions ... 45

5.1.1 Antecedents Influencing CSR Strategy in China ... 45

5.1.2 Implementation of CSR Strategy in China ... 47

5.1.2.1 The Influence of Local Culture on CSR adaptation ... 49

5.1.3 Outcomes of CSR Strategy ... 51

6. Conclusion ... 53

6.5 Recommendations for Future Research ... 58

References ... 60

Appendices ... 66

Appendix A – Overview of the cases and interviews ... 66

Appendix B – Overview of the consulted secondary data ... 67

Appendix C – Coding Scheme within-case Analysis ... 68

Appendix D – Overview cross-case analysis ... 82

Appendix E – Descriptive results based on cultural sensitivity of the headquarters ... 88

1. Introduction

In 2014, the Fortune Global 500 companies spent over $15 billion on CSR initiatives such as educational programs, partnerships with stakeholders and environmental policies (Smith, 2014). Due to globalization and internationalization of their context, MNEs are faced with the need to satisfy an increasing number of stakeholders with different demands, values and norms (Filatotchev & Stahl, 2015). This requires MNEs to make a well-balanced – and therefore crucial – tradeoff between, on the one hand, having a clear and consistent CSR strategy across all the foreign subsidiaries and, on the other hand, focusing on the numerous interests of their local communities and stakeholders (Carroll, 2015). In other words, tension exists between the overall standardization and local adaptation of CSR (Muller & Kolk, 2010).

1.2 Research Gap

Several empirical studies have found important antecedents and outcomes of CSR (see Aguinis & Glavas, 2012, for a review). However, not only the link between these antecedents and outcomes, but also the actual implementation and execution of CSR initiatives within organizations remains a “black box” (Russell & Griffiths, 2009; Aguinis & Glavas, 2012). At the institutional level, authors have provided frameworks for studying corporate governance structures among different (mainly developed) countries (e.g., Aguilera & Jackson, 2003) and others have examined cross-country differences in CSR practices of domestically operating companies (e.g., Maignan & Ralston, 2000; Matten & Moon, 2008). However, research investigating CSR strategy formation and implementation of MNEs is relatively absent (Yang & Rivers, 2009), especially regarding MNEs that operate in vastly different cross-cultural contexts (Bondy & Starkey, 2014). Hence, this study provides meaningful contributions in terms of combining literature on CSR strategies and international business (IB) (Hah & Freeman, 2013). Furthermore, Bondy and Starkey (2014) call for a better understanding of how MNEs manage CSR, especially with regards to foreign subsidiaries that operate in a different local culture than the company’s home country. To summarize, there is a need to further investigate why

MNEs adopt a certain CSR strategy and how managers of foreign subsidiaries perceive this approach and execute CSR initiatives based on the national culture in their host country.

1.3 Contributions and Research Question

This research will focus on three Dutch MNEs that are headquartered in the Netherlands, and have a foreign subsidiary in China. Studying MNEs that operate in different countries has important implications, especially with regards to cross-cultural research. At the institutional level, the national cultures of the Netherlands and China are vastly different (Hofstede Centre, n.d.). Therefore, this study will investigate how such cultural differences impact decision-making related to CSR by managers working for the same company (Schneider & De Meyer, 1991). Secondly, from a managerial point of view, research found that CSR practices of Chinese companies are still in an early stage of development and are even described as “exotic” by Chinese managers (Wang & Juslin, 2009, p. 435). Therefore, studying how managers of MNEs interpret and implement CSR initiatives in the Chinese market will bring new insights that can build upon the findings of Wang and Juslin (2009). On the other hand, from the theoretical point of view, CSR in China has not received much attention from academic scholars (Kolk, Hong & Van Dolen, 2010). Thirdly, this study does not only respond to the call of Yang and Rivers (2009) for more empirical testing of CSR strategies of MNEs, but also replies to Bondy and Starkey (2014) who emphasize that current literature on MNEs and CSR strategies needs to understand how CSR strategies are managed at local-level management of foreign subsidiaries. Fourthly, engaging in CSR in foreign countries can be a viable source of external legitimacy, which allows MNEs to effectively reduce their liability of foreignness (Castelló & Galang, 2014). This study will examine to what extent Dutch MNEs perceive a liability of foreignness in China and how CSR can contribute to the creation of legitimacy in order to reduce this liability. Lastly, Hah and Freeman (2013) argue that research on CSR by MNEs should adopt a cross-country approach in order to study the complexities faced by these companies. The current study provides fruitful insights to this research stream by conducting interviews with managers responsible for CSR in both the Netherlands and in China.

Consequently, the following research question has been developed:

How do MNEs manage their CSR strategy in a country that is culturally different from the headquarters in order to create legitimacy?

In order to answer this research question, the following sub-questions have been formulated:

- How does the headquarters’ host country knowledge influence the decision to adopt a certain

CSR strategy?

- How is the MNE’s CSR strategy implemented at the local level of the foreign subsidiary? - How does the MNE’s CSR strategy recognize cultural difference between the headquarters

and the foreign subsidiary?

- How does the MNE’s CSR strategy help in creating legitimacy and hence reducing the

company’s liability of foreignness?

1.4 Key Definitions

The purpose of Section 1.4 is to outline the key definitions adopted in this study. Table 1 can be consulted when reading through the literature review (Chapter 2) and provides an overview of the most important constructs used in this research.

Key concept* Definition

Corporate social responsibility (CSR)

Actions of firms that appear to further some social good, beyond the interests of the firm and that which is required by law (McWilliams & Siegel, 2001, p. 117). - Local CSR strategy

- Standard CSR strategy

- Transnational CSR strategy

A CSR strategy that allows managers of foreign subsidiaries to develop and execute CSR policies and practices relatively independent based on their own interpretation of the company’s local environment and culture (Filatotchev & Stahl, 2015; Fisher & Bonn, 2007).

A CSR strategy focusing on consistently implementing the same CSR policies and practices across the different countries in which the organization is operating (Arthaud-Day, 2005; Filatotchev & Stahl, 2015).

A CSR strategy based on a global framework for CSR activities that can guide managerial decision-making and ensures consistency across the organization, but allows managers of foreign subsidiaries to adapt the framework to local needs and circumstances (Filatotchev & Stahl, 2015, p. 123).

Cultural sensitivity The capability to both understand and mange differences between cultures (Paul, Meyskens & Robbins, 2011).

Legitimacy A generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions (Suchman, 1995, p. 547).

Liability of foreignness (LOF) The lack of information about the MNE on the part of the host country environment, the uses of stereotypes and different standards in judging MNEs versus domestic firms and the use of MNEs (especially large and visible MNEs) as targets for attack by host country interest groups (Kostova & Zaheer, 1999, p. 73).

Multinational enterprise (MNE) An organization that operates in two or more countries with multiple subunits linked through shared policies or strategy (Kostova & Zaheer, p. 65).

National culture The assumptions, values, norms and beliefs shared by individuals in in a society (Campbell, Eden and Miller, 2012, p. 87).

- Individualism vs.

collectivism The degree of independence (or interdependence) a society maintains among its members (Hofstede Centre, n.d.). - Long-term orientation The degree to which people are concerned with the future points in time, or whether people are focused on the present and the past (De Mooij & Hofstede, 2010).

- Masculinity vs. femininity The degree to which society is driven by success and competition or whether society is driven by quality of life (i.e., wanting to be the best vs. wanting to do what you like) (Hofstede Centre, n.d.).

- Power distance The degree to which individuals within a society are willing to approve and accept a certain degree of hierarchy and unequal distribution of power (De Mooij & Hofstede, 2010).

- Uncertainty avoidance The degree to which individuals within a society feel threatened by uncertainty and ambiguity and avoid these situations (De Mooij & Hofstede, 2010, p. 90).

Table 1 – Definitions of key concepts

*Key definitions are ordered alphabetically and do not follow the structure of the literature review.

2. Literature Review

The following chapter will provide a literature review on the academic works that create the building blocks of this study. Section 2.1 will consists of a general introduction and discussion on the concept of CSR. Section 2.2 will focus on the cultural aspect of institutional theory, and outlines the notion of cultural sensitivity. Moreover, Section 2.2 explains Hofstede’s work on cultural differences by specifying each dimension of the Hofstede Model and linking these dimensions to CSR. Section 2.3 describes what kind of CSR strategies MNEs can choose at a cross-country level. Section 2.4 and 2.5 will discuss the literature on CSR in the Netherlands and China, respectively. In Section 2.6, the concepts of liability of foreignness and legitimacy are outlined. Finally, Section 2.7 summarizes Chapter 2 by developing a preliminary conceptual framework based on the literature review.

2.1 Corporate Social Responsibility

One of the pioneers of CSR was Bowen (1953) who emphasized the “social responsibilities of businessmen”. To date, CSR has become a widespread practice among companies around the world, such as in Europe (Steurer, 2010), Asia (Chapple & Moon, 2007) and even Africa (Visser, 2006). Because of this accelerated interest in CSR, the concept itself has become confusing and unclear (Angus-Leppan, Metcalf & Benn, 2010; Matten & Moon, 2008). Moreover, many definitions of CSR exist, or as Votaw (1973, p. 11; quoted in Barkemeyer & Figge, 2014, p. 126) put it: “The term [CSR] is a brilliant one; it means something, but not always the same thing, to everybody”. This study will follow the definition by McWilliams and Siegel (2001, p. 117) who define CSR as “actions of firms that appear to further some social good, beyond the interests of the firm and that which is required by law”. Since “actions of firms” and “social good” are rather broad concepts, an overview of Carroll’s (1979; 1991) work is given which can clarify the concept of CSR.

Carroll (1979; 1991) synthesized the field of CSR by developing the ‘Pyramid of CSR’ which consists of four different social responsibilities of organizations: economic responsibilities, legal responsibilities, ethical responsibilities, and philanthropic responsibilities. First of all, in order for firms to exist, it is necessary that they fulfill their economic and legal responsibilities. This reflects the notion of shareholder maximization advocated by Friedman (1970), in which businesses should maximize their corporate profits but are not allowed to conduct illegal activities (Athanasopoulou, 2012; Carroll & Shabana, 2010). Thereafter, firms have their ethical responsibilities. Rather than only focusing on financial gains and obeying the law, firms are expected to conduct social and environmental initiatives that take into account the interest of their stakeholders (Carroll, 1991). Lastly, some firms engage in philanthropic activities. These activities could include the sponsorship of programs, contribution to charity funds, or donation of products for local communities. The main distinction between ethical responsibilities and philanthropic responsibilities is that the latter are more voluntaristic in nature than the former. In other words, whereas a firm’s ethical responsibilities are “expected”, philanthropic responsibilities are “desired” (Carroll & Shabana, 2010, p. 90).

the institutional level, organizational level and individual level. In their extensive review of the CSR literature, Aguinis and Glavas (2012) argue that research has mainly concentrated on either an institutional level or organizational level perspective of CSR, whereas an integration of different perspectives is crucial for the further advancement of CSR in academic literature. Therefore, the focus of this research is to develop a multi-level model of how MNEs engage in CSR (Aguinis & Glavas, 2012; Bansal & Roth, 2000; Basu & Palazzo, 2008). In order to do so, literature on institutional theory, and especially the cultural aspect of this theory, is combined with literature on CSR strategy development and implementation by MNEs. Adopting an institutional perspective can help explain why different CSR practices exist across different countries (Campbell, 2006; Campbell, 2007; Matten & Moon, 2008). This will be discussed in Section 2.2.

2.2 National Culture within an Institutional Context

According to Scott (1995; 2008), institutional theory consists of three different aspects: regulative, normative, and cultural-cognitive elements. Whereas regulative elements stem from formal institutions such as government directives and policies, normative and cultural-cognitive elements are the result of tacit meaning derived from the social context in which organizations and individuals operate and make decisions. Normative elements of institutions are created by educational systems such as schools and universities (Matten & Moon, 2008). Cultural-cognitive elements, on the other hand, consist of values, assumptions and rules that are deeply embedded within society (Scott, 1995; 2008). As briefly discussed, the main emphasis of this research will be on the cultural aspect of institutional theory (Scott 1995; 2008). The decision to focus more specifically on the cultural part of institutional theory will be outlined more thoroughly below.

According to Peng et al. (2014), previous research taking an institutional perspective has mainly investigated the effect of formal institutions (e.g., regulations set out by governments) on CSR, while relatively few studies have explored the role of informal institutions like national cultures. The limited literature that does exist found that national cultures have an important impact on CSR practices and orientations in different countries (e.g., Burton, Farh & Hegarty, 2000; Kim & Kim,

2010; Peng, Dashdeleg & Chih, 2014). However, the aforementioned authors have quantitatively tested this phenomenon and therefore little is known on “how” the relationship between national cultures and CSR engagement affects MNEs (Hah & Freeman, 2013; Bondy & Starkey, 2014).

At the same time, understanding cultural differences between countries is fundamental for MNEs that operate in culturally diverse countries (Robertson & Fadil, 1999). This understanding is also referred to as cultural sensitivity (Shapiro, Ozanne, and Saatcioglu, 2008). Cultural sensitivity is defined as the capability to both understand and manage differences between cultures (Paul et al., 2011). Aguilera-Caracuel, Guerrero-Villegas, Vidal-Salazar and Delago-Máriquez (2014) and Hah and Freeman (2013) recently emphasized that it is crucial for the headquarters of MNEs to understand the local cultural context of their foreign subsidiaries when implementing a CSR strategy. Therefore, this study will investigate how the headquarters’ knowledge of the Chinese culture influences the type of CSR strategy that is chosen.

Schneider and De Meyer (1991) found that national cultures influence the interpretation of important strategic issues, such as the implementation of new CSR activities and practices (Porter & Kramer, 2006). However, Bondy, Moon and Matten (2012) and Yin and Jamali (2016) argue that the headquarters of a MNE frequently neglects the role that the local culture can play when introducing new CSR activities in foreign countries. Implementing a successful CSR strategy across different countries means that the headquarters needs to understand the cultural differences between the national culture of the MNE’s home country and the national culture of a particular host country (Campbell et al., 2012). This difference between cultures has been defined as “cultural distance” (Ghemawat, 2001, p. 3). However, in order to understand what cultural distance means and how cultural sensitivity impacts the CSR strategies of MNEs, it is necessary to first establish a solid definition of a national culture.

According to Campbell et al. (2012, p. 87), a national culture can be defined as “the assumptions, values, norms and beliefs shared by individuals in a society”. One of the founder fathers of research on national cultures is Geert Hofstede (López-Duarte, Vidal-Suárez & González-Díaz, 2015; Peng et al., 2014). Although other cultural researchers such as House and colleagues (2004) of

valuable contributes to the field, Hofstede remains the most cited cultural researcher (López-Duarte et al., 2015). Following his longitudinal research at more than 40 subsidiaries of IBM, Hofstede developed the Hofstede Model that allows one to distinguish between different national cultures (De Mooij & Hofstede, 2010).

However, Hofstede’s work has received a fair degree of criticism throughout the last two decades (Fernandez, Carlson, Stepina & Nicholson, 1997; McSweeney, 2002; Jones, 2007) and this criticism should not be neglected. First of all, McSweeney (2002) criticizes the model developed by Hofstede because it lacks external validity. That is, Hofstede wrongly assumes that the data found at the foreign subsidiaries of one company (IBM) can be generalized to national levels (Jones, 2007). Secondly, Jones (2007) argues that national cultures are not uniform in the first place, since many different subcultures can exist across a nation’s population groups. Moreover, Hofstede retrieved most of his data at IBM through various surveys between 1967 and 1973 (Hofstede & Bond, 1998; McSweeney, 2002). Consequently, authors such as McSweeney (2002) and Jones (2007) have depicted Hofstede’s findings as outdated. Although the Hofstede Model remains one of the most important constructs within the international business field (Taras, Steel & Kirkman, 2010), especially for scholars interested in MNEs and their cross-cultural context (López-Duarte et al., 2015), it is important to keep the criticism on Hofstede’s work in mind.

Each dimension of the Hofstede Model will be discussed more thoroughly below. Moreover, the scores (ranging between 0 and 100) received by the both Netherlands and China will be outlined based on the different dimensions provided by the Hofstede Centre (n.d.). This allows for the illustration of the vast cultural differences between the Netherlands and China. Moreover, each of Hofstede’s cultural dimensions will be linked to CSR by discussing the findings of Peng et al. (2014). These authors analyzed the CSR practices of 1,189 companies listed on the S&P Global Index and 245 companies from the Dow Jones Sustainability Index (DJSI). Together, these companies were incorporated in 29 different countries, therefore representing 29 different national cultures. Lastly, it is important to note that Hofstede added a sixth dimension called “indulgence versus restraint” in 2010. Since Hofstede recently developed this dimension, the researcher only found little empirical research on this new dimension and is therefore not included in this research.

Table 1 (see Section 1.4) provides an overview of each cultural dimension and Figure 1 outlines the scores of both the Netherlands and China.

Figure 1 – Cultural differences between the Netherlands and China (Hofstede Centre, n.d.)

2.2.1 Power Distance

First of all, power distance refers to whether individuals within a culture are willing to approve and accept a certain degree of hierarchy and unequal distribution of power (De Mooij & Hofstede, 2010). In countries with a high score on power distance, hierarchy is taken for granted and therefore accepted by the people. On the other hand, in countries with a low score on power distance, individuals are treated equally regardless of their social position. According to the Hofstede Centre (n.d.), the Netherlands receives a score of 38. This means that power within the Dutch society is dispersed and rather equally distributed. Superiors allow their subordinates to be engaged in decision-making, and the relationship between them is easygoing (e.g., managers are usually called by their first name rather than their last name). In contrast, China scores a high 80 in terms of power distance (Hofstede Centre, n.d.). Consequently, hierarchy based on social ranks is seen as acceptable by the Chinese. Authority drives the relationship between managers and employees and subordinates are expected to follow their leaders under all circumstances (Hofstede Centre, n.d.). This also implies that subordinates put great trust in their supervisor. Research by Peng et al. (2014) found empirical support for the notion that high power distance will have a negative influence on a company’s CSR practices.

2.2.2 Individualism/collectivism

The second dimension of culture is individualism versus collectivism. In individualistic countries, people are oriented towards themselves and their close relatives (De Mooij & Hofstede, 2010). Conversely, collectivistic cultures are focused on the “we”-aspect, putting great emphasis on the larger collective of a certain group. Within this environment of collectivism, the group is more important than a single individual and losing face should be avoided at all costs. According to the Hofstede, Centre (n.d.), the Netherlands is a typical example of an individualistic country, receiving a score of 80. In terms of the dimension individualistic versus collectivistic, China scores significantly lower with a score of just 20. This is not surprising, since China, as a result of its communistic heritage, is a typical example of a collectivistic culture in which in-group members receive superior treatment over members who are not in that group (Ralston, Holt, Terpstra & Kai-Cheng, 1997). According to Peng et al. (2014), managers of individualistic cultures focus more on explicit forms of CSR (see also Matten & Moon, 2008), which means that companies from individualistic cultures are more likely to articulate and embed CSR in their core strategy. As a result, Peng et al. (2014) find a significant positive link between individualism and a firm’s engagement in CSR. That is, companies in an individualistic national culture are more likely to engage in CSR.

2.2.3 Masculinity/Femininity

This dimensions refers to whether cultural values can be seen as masculine or feminine. In masculine countries, success and competition are important (De Mooij & Hofstede, 2010). Because of the great focus on accomplishments, people, particularly men, are driven by materialism and rivalry (Vitell, Nwachukwu & Barnes, 1993). In feminine countries, equality and “quality of life” are important (De Mooij & Hofstede, 2010, p. 89). With regards to this cultural dimension, the Netherlands receives a very low score of 14. Involvement of stakeholders such as employees is important in the decision-making process of companies and the Netherlands is known for its consensus-driven political system (Goodijk, 2000). China has a score of 66 and can therefore be specified as having a masculine national culture (Hofstede Centre, n.d.). For example, it is extremely important for Chinese students (and also their parents) to receive high grades on their exams (Smith & Smith, 1999). Moreover, many Chinese

farmers have left the rural areas hoping to find success and a better life in urban cities along the country’s east coast (Knight & Gunatilaka, 2010). Vitell et al. (1993) suggest that people in masculine cultures are less likely to recognize ethical dilemmas than people in feminine cultures. Moreover, Peng et al. (2014) find empirical proof for a negative relationship between masculinity and a company’s CSR practices. Hence, companies in masculine national cultures are less likely to engage in CSR.

2.2.4 Uncertainty avoidance

According to De Mooij and Hofstede (2010, p. 90), uncertainty avoidance can be defined as “the extent to which people feel threatened by uncertainty and ambiguity and avoid these situations”. This means that people within cultures that score high on uncertainty avoidance are less likely to deviate from existing (social) structures, whereas people within cultures that score low on uncertainty avoidance are more likely to have a positive view towards change. Moreover, in countries scoring high on uncertainty avoidance, it is important for people to find a sense of “truth” in their lives. The results from the Hofstede Centre (n.d.) show that the Netherlands scores a 53 on the dimension uncertainty avoidance, indicating a small inclination towards the avoidance of uncertainty. On the other hand, China has a score of 30 in terms of uncertainty avoidance. Therefore, the results show that Chinese people are more flexible than the Dutch when it comes to the acceptance of ambiguity. According to Peng et al. (2014), companies that operate in a high uncertainty avoidance culture are more likely to engage in CSR. However, even though the Hofstede Centre indicates a score of 30 for China, the uncertainty avoidance dimension is mainly an element of Western cultures (Hofstede & Bond, 1988). Therefore, these authors collaborated with Chinese sociologists to create the so-called Chinese Value Survey (CVS) and discovered that Chinese people are not necessarily worried about seeking a sense of truth or certainty. More interestingly, Hofstede and Bond (1988) discovered a new cultural dimension, which they called “Confucius dynamism”, that especially applies to Eastern cultures such as China. This dimension has also been called “long-term orientation” (Kim & Kim, 2010) and will be outlined more thoroughly in Section 2.2.5.

2.2.5 Long-Term Orientation (Confucius Dynamism)

The last cultural dimension of the Hofstede Model is long-term orientation, and refers to whether people are concerned with the future points in time, or whether people are focused on the present and past (De Mooij & Hofstede, 2010). According to Hofstede and Bond (1988), this dimension closely resembles the ideas of Confucius. When cultures in countries score low on Confucius dynamism, cultural value such as old rituals, balance and perseverance are emphasized. On the other hand, a high country score on Confucius dynamism constitutes a culture that focuses on “thrift, perseverance and having a sense of shame” (Hofstede & Bond, 1988, p. 17). Based on research by the Hofstede Centre (n.d.), the Netherlands scores a 67 and can therefore be seen as quite future-oriented. China scores relatively higher than the Netherlands on Confucius dynamism with a score of 87. Again, these scores have to be interpreted with caution since Confucius dynamism is mainly an element of Eastern cultures (Hofstede & Bond, 1988) and therefore this dimension should be less important in the Dutch culture. Peng et al. (2014) did not test Hofstede’s long-term orientation dimensions but Vitell, Paolillo and Thomas (2003) found a positive link between high Confucian dynamism and the focus on ethics and CSR. Therefore, it is assumed that individuals in countries that score high on this dimension are more likely to care about behaving ethically and maintaining face at all times, hence trying to avoid any unethical conduct.

Now that each cultural dimension of the Hofstede model has been discussed and linked to CSR, Section 2.3 will provide an overview of how MNEs manage CSR and what types of CSR strategies these companies can implement and execute across their foreign subsidiaries.

2.3 Corporate Social Responsibility Strategies for Multinational Enterprises

The accelerated growth of global markets has opened up the gates for MNEs to enter new regions with new opportunities. However, globalization also brings new challenges for MNEs. First of all, globalization enhances the agency-dilemma because it becomes increasingly more difficult for shareholders to closely monitor the decision made by managers of foreign subsidiaries (Filatotchev & Stahl, 2015). Secondly, and even more importantly, when firms start to operate in countries with

different institutional systems, they ultimately have to choose how to adapt to that local environment, especially with regards to the company’s CSR strategy (Arthaud-Day, 2005; Husted & Allen, 2006).

Although the latter challenge proposes a crucial tradeoff for companies, academic research on CSR strategies by MNEs remains insufficiently addressed and understood (Cruz & Boehe, 2010; Hah & Freeman, 2013; Yang & Rivers, 2009). Because of their global orientation, MNEs are constantly called upon to provide social benefits (Jamali, 2010; Kostova & Zaheer, 1999; Matten & Moon, 2008), especially in developing countries (Matten & Crane, 2005; Prahalad & Hart, 2002). However, critics have argued that big corporations use CSR as a means to become more powerful and in fact legitimize that power (Banerjee, 2008). Although this is an interesting and ongoing discussion, the purpose of this study is to go beyond this debate and to focus on developing a model that outlines how MNEs engage in CSR in culturally different countries (Basu & Palazzo, 2008; Hah & Freeman, 2013).

To date, very few studies exist that have tried to empirically examine the complexities faced by MNEs in terms of managing CSR across borders. Jamali (2010) investigated the CSR strategies of 10 MNEs operating in Lebanon. A major limitation is that Jamali (2010) only interviewed managers of local subsidiaries that were not actively involved in developing the MNEs international CSR strategy. Moreover, this study did not investigate whether cultural differences between headquarters and the foreign subsidiary in Lebanon were taken into considerations by the MNEs. Bondy and Starkey (2014), on the other hand, examined the CSR strategies of 37 MNEs headquartered in the United Kingdom. This study interestingly found that most of the CSR strategies of the MNEs did not reflect the local cultural value of their foreign subsidiaries. However, the research by Bondy and Starkey (2014) provided a unilateral side of the story since the authors did not investigate how CSR was managed at the level of the local foreign subsidiaries. Cruz and Boehe (2010) are, to the best knowledge of the researcher, the only authors that have empirically examined CSR strategies of MNEs by taking a cross-country approach. These researchers conducted interviews with managers in France (headquarters) and Brazil (local foreign subsidiary) of two internationally operating retail companies. Consequently, Cruz and Boehe (2010) identified several organizational challenges faced by MNEs when trying to manage a CSR strategy across different countries. Although the authors provide

France and Brazil and whether or not this could impact a MNE’s CSR strategy. Therefore, a significant research gap remains unaddressed.

Researchers agree that MNEs can opt to apply one of three CSR strategies across their national borders (Filatotchev & Stahl, 2015). These approaches and their differences are outlined below.

2.3.1 The Standard Corporate Social Responsibility Strategy

MNEs that apply the standard (also called global) CSR strategy focus on consistently implementing the same CSR policies and practices across the different countries in which their organization is operating (Filatotchev & Stahl, 2015). In other words, these companies always follow similar universal norms and values wherever they do business (Arthaud-Day, 2005; Fisher & Bonn, 2007). According to Donaldson and Dunfee (1999, p. 52) these universal norms and values are called “hypernorms” that apply everywhere around the world and are “acceptable to all cultures and organizations”. When multinational companies follow the standard strategy of CSR, decision-making regarding new policies or practices is centralized (Fisher & Bonn, 2007) and therefore primarily done at the main office (Filatotchev & Stahl, 2015). This makes it easier for the headquarters of MNEs to compare CSR activities and programs across their different host countries. An important disadvantage of this approach is that it neglects that managers of foreign subsidiaries are usually well informed about the company’s local context and are therefore in a better position to make CSR-related decisions (Fisher & Bonn, 2007). Therefore, MNEs that follow the standard strategy of CSR tend to overlook the demands and interests of their foreign stakeholders (Miska et al., 2013).

2.3.2 The Local Corporate Social Responsibility Strategy

The local CSR strategy emphasizes that when firms operate in different countries, they have to understand their local stakeholders (Filatotchev & Stahl, 2015). Based on this strategy, decision-making regarding CSR is decentralized to the foreign subsidiaries (Fisher & Bonn, 2007). Managers of these subsidiaries can develop and execute CSR policies and practices relatively independent based on their own interpretation of the company’s local environment and culture. However, handing over too much authority to the foreign subsidiaries can become problematic. Filatotchev and Stahl (2015)

argue that the headquarters might lose track of the decision-making by local managers, resulting in an information wedge between them. Moreover, when a new local CSR policy is proposed by the foreign subsidiary that is not in line with the company’s overall strategy, it becomes more difficult for the headquarters to manage its worldwide CSR activities (Bondy & Starkey, 2014).

2.3.3 The Transnational Corporate Social Responsibility Strategy

The transnational CSR strategy can be seen as a combined, or hybrid, version of the global and local CSR strategy (Arthaud-Day, 2005). As a result of globalization, MNEs have to strategically address the needs of international and local stakeholders (Bondy & Starkey, 2014). Moreover, by accepting that stakeholders sometimes have conflicting needs, multinational companies that have adopted the transnational CSR strategy allow local managers to develop CSR initiatives that can contribute to context of the foreign subsidiary as well as to the overall company (Filatotchev & Stahl, 2015). In order to achieve this, the transnational strategy proposes that the headquarters should provide a globally consistent CSR framework or template that can be interpreted based on the subsidiary’s local market environment and culture. Authors such as Miska et al. (2013) have advocated the transnational strategy because it acknowledges that local cultural differences exist across the countries in which a MNE is operating. However, Miska et al. (2013) did not empirically test how national cultural differences between countries influence a MNE’s decision to implement the transnational CSR strategy. Moreover, Barkemeyer and Figge (2014) notice an increase in the “headquartering effect”, which hinders the ability of MNEs to pursue a transnational CSR strategy based on both global standardization and local responsiveness. The headquartering effect refers to “a concentration of CSR-related decision making within corporate headquarters”, which makes it more difficult for MNEs to recognize the local needs of foreign subsidiaries (Barkemeyer & Figge, 2014 p. 136). The authors doubt whether the transnational CSR strategy is even practically manageable for MNEs, and question if the strategy is “a realistic ambition or a utopian ideal” (Barkemeyer & Figge, 2014 p. 125).

This study investigates how MNEs manage CSR from two levels (i.e., the headquarters in the Netherlands and foreign subsidiary in China). Therefore, Section 2.4 and 2.5 will outline the literature on CSR in the Netherlands and China, respectively.

2.4 Corporate Social Responsibility in the Netherlands

Starting in the 1960s, CSR has become a popular topic in the Netherlands (Remmé, 2015). Since then, an increasing amount of Dutch organizations started focusing on the “soft side” of their business (i.e., how a company’s people are treated and how a company affects its stakeholders). This is also reflected in the Rhineland model, which has influenced Dutch economic thinking (Goodijk, 2000). The Rhineland model emphasizes that companies need to incorporate various stakeholders, rather than only shareholders. Moreover, the Netherlands is famous for its strong employee councils and various industry-wide codes of conduct (Remmé, 2015).

To date, most empirical research concerning CSR by Dutch companies has mainly focused on small and medium sized enterprises (e.g., Uhlaner, Berent-Braun, Jeurissen & De Wit, 2012; Uhlaner, Van Goor-Balk & Masurel, 2004). On the other hand, Graafland, Van de Ven and Stoffele (2003) made a comparison of CSR engagement of both small (<100 employees) and large (>100 employees) organizations in the Netherlands. The authors argue that large firms are more likely to engage in CSR than small firms because the former companies receive more attention from the Dutch public and media. Moreover, large companies in the Netherlands are in a better position to manage the cost of their engagement in CSR by utilizing economies of scale and scope. However, Graafland et al. (2003) did not specifically investigate the role of Dutch MNEs relative to domestically operating companies.

To the best knowledge of the researcher, there is currently no academic literature on how Dutch MNEs manage CSR on a cross-country level (let alone a cross-cultural level). This seems paradoxical, since the Netherlands has a relatively high amount of MNEs compared to the size of the country (Remmé, 2015). For example, in 2013 there were 24 Dutch MNEs in the Forbes Global 2000, which together generated $1.130 trillion in revenues (Economy Watch, 2013). Hence, by investigating how Dutch MNEs manage CSR in China, this study provides new insights specifically for companies headquartered in the Netherlands.

2.5 Corporate Social Responsibility in China

Research on CSR in China is rather limited (Kolk et al., 2010). Moreover, existing literature shows mixed results. On the one hand, Wang and Juslin (2009) argue that CSR is a Western concept and

therefore does not fit with the Chinese culture. On the other hand, Tang and Li (2009) found that Chinese firms seem to progressively adopt and promote CSR even though the latter authors agree that CSR is a Western concept. There could be a reason why researchers end up with conflicting outcomes. First of all, according to Lo, Egri and Ralston (2008), CSR is a product of the influential Protestant historical culture in many Western countries such as the United States (US). Within this Western culture, businesses themselves are viewed as institutions that need to be socially responsible and ethically correct towards society (Vogel, 1992). Conversely, China still has a communitarian legacy that views the government, rather than companies, as the main driver for social welfare (Lo et al., 2008). Interestingly, Lo et al. (2008) did not find a significant difference between the CSR practices of US and Chinese companies. Moreover, Tang and Li (2009) conclude that, in terms of CSR, Chinese and other global companies do not address their customers, employees, suppliers or shareholders differently. Then why do Chinese companies seem to focus more on CSR? The answer to this question may be attributed, again, to the increased globalization. The sample of both Lo et al. (2008) and Tang and Li (2009) did not consist of Chinese companies that solely operate in China. In contrast, the authors included Chinese MNEs that operate in different countries and therefore have to comply with foreign regulations. This has important implications. First of all, the fact that these Chinese companies focus more on CSR seems to link with their international orientation rather than the context of their home country (Lo et al., 2008; Tang & Li, 2009). Secondly, the results are hard to generalize to a (Dutch) MNE operating in China because the studies are flawed towards explaining CSR through the international context of the companies in their sample, rather than the companies’ local Chinese context. Consequently, additional research is needed that investigates how companies in China engage in CSR (Kolk et al., 2010).

2.6 Legitimacy and Liability of Foreignness

MNEs operate across borders, and therefore face more challenges than domestically operating companies. One of these challenges is the liability of foreignness (LOF) that foreign subsidiaries of MNEs can experience (Campbell et al., 2012; Eden & Miller, 2004; Kostova & Zaheer, 1999). According to Kostova & Zaheer (1999, p. 73), LOF refers to “the lack of information about the MNE

on the part of the host country environment, the uses of stereotypes and different standards in judging MNEs versus domestic firms and the use of MNEs (especially large and visible MNEs) as targets for attack by host country interest groups”. Not only do MNEs have to maintain a good reputation across different countries (Kostova & Zaheer, 1999), they also need an international “license to operate” (Castelló & Lozona, 2014; Yin & Jamali, 2016) in diverse host countries. Consequently, LOF can result in additional costs and a reduced competitive position relative to domestic companies (Zaheer, 1995). In order to reduce their LOF, MNEs should build legitimacy within their external environment (Campbell et al., 2012; Kostova & Zaheer, 1999). Suchman (1995, p. 574) defines legitimacy as “a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions”. The creation of legitimacy relative to the external environment of a company has also been referred to as external legitimacy (Hah & Freeman, 2013).

Several scholars have argued that MNEs’ engagement in CSR activities in host countries can be seen as a way to establish legitimacy, which in turn reduces their LOF (e.g., Campbell et al., 2012; Castelló & Galang, 2014; Hah & Freeman, 2013). For example, MNEs can institute support and a favorable reputation within the local environment of foreign subsidiaries through CSR activities that create social welfare in host countries (Hah & Freeman, 2013). As a result, the foreign subsidiary of a MNE can establish itself as a legitimate entity within its broader social context. According to institutional isomorphism, foreign subsidiaries of MNEs “morph” their practices based on the expectations and pressures within the local environment, and hence create legitimacy (DiMaggio & Powell, 1983; Hah & Freeman, 2013 Kostova & Zaheer, 1993). Although previous literature found that Chinese MNEs often face a high degree of LOF when moving to the Western world (Klossek, Linke & Nippa, 2012), limited academic attention has been given to the LOF of MNEs that operate in China. Therefore, this study will examine to what extent Dutch MNEs perceive a liability of foreignness in China and how engaging in CSR can contribute to the creation of legitimacy and hence this liability.

As previously outlined, this study develops a multi-level model of how Dutch MNEs manage CSR in China. Moreover, it aims to understand how cross-cultural differences between the

Netherlands and China influence this process. In order to so, Section 2.7 will develop a preliminary conceptual framework that connects the subchapters of literature review into an overarching display.

2.7 Developing a Preliminary Conceptual Framework

The preliminary conceptual framework consists of three parts: (1) the researched antecedent that influences which type of CSR strategy is chosen, (2) the actual implementation and management of the CSR strategy in China, and (3) the researched outcome of the CSR strategy in China.

2.7.1 Antecedent – Cultural Sensitivity of the Headquarters

Aguilera-Caracuel et al. (2014) and Hah and Freeman (2013) recently emphasized that it is crucial for the headquarters of MNEs to understand the local cultural context of their foreign subsidiaries when implementing a CSR strategy. Therefore, this study will investigate the role of cultural sensitivity at the headquarters of Dutch MNEs. According to the preliminary conceptual framework, the degree cultural sensitivity of the headquarters will influence which of three CSR strategies (standard, local, or transnational) is chosen and hence will determine how CSR in implemented at the local level in China. The framework proposes that the higher the headquarters’ degree of cultural knowledge of China, the more likely the MNE is going to opt for either a local or transnational rather than a standard CSR strategy. A high degree of Chinese cultural knowledge refers to headquarters’ understanding that cultural differences exist between them and the foreign subsidiary. If this is the case, it is less likely that the headquarters chooses a standard CSR strategy for the Chinese subsidiary because such a strategy would allow for the least amount of local cultural freedom. Having local cultural freedom means that the CSR activities and practices of the foreign subsidiary can be adjusted to the local culture (Hah & Freeman, 2013).

2.7.2 Implementation – Managing the CSR Strategy in China

Both Yang & Rivers (2009) and Bondy and Starkey (2014) have stressed the need for empirical research investigating how MNEs manage CSR strategies at a cross-cultural level. Hence, the study identifies how CSR strategies developed by Dutch MNEs are implemented at the local Chinese level and to which extent the CSR strategies do actually allow for local cultural freedom. Moreover, the

study will investigate what types of CSR activities are implemented at the Chinese level and how these activities are influenced by the Chinese national culture based on the 5 dimensions of the Hofstede Model.

2.7.3 Outcomes – Creating Legitimacy to Reduce Liability of Foreignness

Previous literature has theorized that the engagement in CSR of foreign subsidiaries can provide MNEs with a successful means to create legitimacy in a host country, hence reducing the company’s liability of foreignness (e.g., Campbell et al., 2012; Castelló & Galang, 2014; Hah & Freeman, 2013). Legitimacy is socially constructed (Suchman, 1995), and therefore the perception of a company’s (external) legitimacy is culturally dependent (Kostova & Zaheer, 1999). Because both local and transnational CSR strategies allows for more cultural adaption than the standard CSR strategy, it is assumed that the former two strategies will result in a higher degree of external legitimacy in China. Hence, one can argue that the local and transnational CSR strategies work better for MNEs trying to develop legitimacy and reduce their liability of foreignness.

Figure 2 – Preliminary conceptual framework based on the literature review

3. Research Methodology

The following chapter outlines the research design of this study. Section 3.1 explains the research philosophy that is adopted and describes the assumptions on which this philosophy is build. Thereafter, Section 3.2 focuses on the chosen research strategy and Section 3.3 outlines the data Home country context

(the Netherlands) Low degree of China country cultural knowledge High degree of China country cultural knowledge Antecedent

Implementation of standard CSR strategy

Implementation of local CSR strategy

Implementation of transnational CSR strategy Implementation

Host country (cultural) context China

Low degree of legitimacy

High degree of legitimacy

collection. Section 3.4 provides an in-depth explanation of how the data was analyzed. Lastly, Section 3.5 describes how a high degree of research quality was maintained during this study.

3.1 Research Philosophy –Interpretive Epistemology

The research philosophy adopted in this study influences the research process and is therefore an important first step of the methodology chapter (Saunders & Lewis, 2012). The two most influential research philosophies are ontology and epistemology (Crotty, 1998). Ontology is concerned with what things are within their natural reality (Saunders & Lewis, 2012). Epistemology, on the other hand, focuses on what it means to acquire knowledge within this reality (Crotty, 1998). More specifically, this study adopts the interpretive epistemology philosophy. This philosophy is chosen for two reasons. First, interpretive epistemology assumes that researchers need to understand subjective realities rather than an objective truth (Blumberg, Cooper & Schindler, 2014). Since cultures are subjective (Brannen, 2009), the researcher adopts this approach to investigate how people make sense of national cultures and how cultural differences impact the CSR engagement of MNEs. Second, there is an overrepresentation of quantitative, and therefore rather positivistic, research on CSR (Aguinis & Glavas, 2012). To a large degree these positivistic studies have focused on establishing empirical relationships between CSR and corporate social performance (CSP) but ignored the cultural contexts in which companies operate (Gond & Matten, 2007).

3.2 Research Strategy – Multiple Case Study

Following the interpretive epistemology philosophy, this study adopts a qualitative case study research approach. According to Eisenhardt (1989), case studies can provide crucial exploratory information and evidence for the further development of theory. This type of qualitative research allows one to identify processes across different levels (e.g., institutional, organizational and individual) of analysis (Aguinis & Glavas, 2012) but within their real-life contemporary context (Yin, 2009). Marquis and Raynard (2015) argue that qualitative research provides a rich description of the multiple contexts in which MNEs and their foreign subsidiaries are operating. Rather than trying to establish direct causal

connections between variables, case study research allows one to illustrate and argue how and why certain relationships exist (Siggelkow, 2007).

In order to gain a richer and deeper understanding of how MNEs manage CSR in a cross-cultural context, a multiple-case research approach is selected. Multiple cases are usually better equipped for the building of theory than single-case studies (Eisenhardt & Graebner, 2007; Yin, 2009). Having multiple cases of evidence makes it easier for researchers to generalize and compare the outcomes of a study (Eisenhardt & Graebner, 2007). Although single-case studies usually provide a more complex description of details and how these details are connected to a particular theory, multiple case studies allow for an exploration of the broader picture of the research itself (Eisenhardt & Graebner, 2007).

According to Yin (2009, p. 59), multiple case studies can be either “holistic” or “embedded”. Whereas holistic case studies analyze individual cases with a specific context, embedded case studies investigate units that share a specific context. This research focuses on the latter approach, because for each organization within the sample there are two units of analysis: the headquarters and the foreign subsidiary. This has two implications. First, CSR is investigated across different Dutch MNEs. Second, CSR is investigated across different departments (i.e., the Dutch headquarters that developed the CSR strategy and the Chinese subsidiary that implemented and executed the CSR strategy).

On the other hand, qualitative case study research has its shortcomings. Eisenhardt (1989) argues that scholars conducting case study research can become overwhelmed by the amount of data that is collected. This data is usually rather complex and detailed and therefore researchers can lose track of the broader picture that they are trying to investigate. Furthermore, qualitative case study research does not have as much generalizability as quantitative approaches such as surveys that allow researchers to statistically test hypotheses (Yin, 2009). In other to tackle this weakness, Siggelkow (2007, p. 21) proposes that case study research should provide “motivation, inspiration, and illustration”. This implies that qualitative case study research can help other scholars in motivating and inspiring to search for interesting and unaddressed gaps in the literature. Moreover, case studies are an excellent approach to illustrate how theoretical mechanisms work and how they relate to the context in which they are operating (Siggelkow, 2007).

3.3 Data collection

Based on the multiple case study method, three Dutch MNEs (two of which are part of the Fortune Global 2000) with a foreign subsidiary in China were approached following the personal connections of the researcher within these companies (see Table 2 for a brief description of the MNEs). This can be seen as purposive sampling (Saunders & Lewis, 2012). The Chinese subsidiaries of the MNEs in the sample each had a different size and operated in a different industry, which means that the sample also had characteristics of maximum variety sampling (although this was not the main goal of the sampling method; Saunders & Lewis, 2012).

MNE* Brief description Size of company Years of operation in

China BevCo MNE that produces both alcoholic and

non-alcoholic beverages

76,135 (1,800 in China) Since 1994 (22 years) HRCo MNE that provides human resource

consultancy services

29,750 (600 in China) N/A FoodCo MNE that both produces and processes

raw foods

1,600 (100 in China) Since 2014 (2 years)

Table 2 – Sample of the multiple case study *Names are anonymized.

The researcher collected data through semi-structured interviews with 14 interviewees using a snowball sampling method (Saunders & Lewis, 2012). Contact persons at the MNEs were asked whether they could introduce the researcher to members of their organization responsible for managing CSR. The length of the interviews ranged from approximately 30 minutes to 60 minutes and the majority of the interviews were held in March and April of 2016 (see Appendix A for an overview). The researcher continued to approach new interviewees until similar insights emerged from the data and no additional interviewees were found, therefore reaching saturation (Saunders & Lewis, 2012). The interviews were conducted with managers in the Netherlands and China who were responsible for the development, communication, interpretation, and implementation of the company’s CSR strategy. Due to the limited availability of the CSR Director of BevCo in the Netherlands, one interview was conducted with the CSR Director of BevCo in the United Kingdom, who was also partly responsible for the company’s CSR strategy on an international level. Additionally, the

researcher conducted an interview with the Regional CSR Director of BevCo at the regional headquarters in Singapore. Compared to HRCo and FoodCo, BevCo was the only company with a regional headquarters for the Asia market. The interview with the Regional CSR Director of BevCo in Singapore allowed the researcher to identify what kind of role the regional headquarters of this MNE plays in managing CSR on cross-country level.

Most of the interviews with managers from the Netherlands were conducted in Dutch. Since this study is written in English, these interviews were first transcribed in Dutch and translated into English afterwards. The interviews with managers from China (and also the United Kingdom and Singapore) were conducted in English. When necessary, the researcher contacted the interviewees for follow-up questions after the interview. This is a good method for the clarification of possibly ambiguous answer and ensures the quality of the data (Turner, 2010).

In order build on the preliminary conceptual framework, the semi-structured interview guide was slightly adjusted based on the location of the interviewee. At the level of the headquarters (i.e., the Netherlands and in the case of BevCo also the United Kingdom and Singapore), the interviewees were asked about their cultural knowledge of China. In order to do so, the researcher first introduced each of Hofstede’s cultural dimensions and then asked what score the interviewee would give for China. Due to the limited amount of questions and respondents, it was not possible to run a statistical analysis based on this data. However, the descriptive results are outlined in Appendix E. At the Chinese level, the managers were also introduced to the cultural dimensions of Hofstede. In order to explore the cultural adaptation of each MNE’s CSR strategy, the Chinese interviewees were asked how each of these dimensions influenced CSR at the local level. Although the adjustment of the interview guide based on location did not allow for a strict cross-cultural analysis (e.g., the interviewees in the Netherlands were not asked how the Dutch culture influences CSR strategy formation), it did create the opportunity to explore cultural sensitivity at the level of the headquarters and local cultural responsiveness at the level of the Chinese subsidiary.

3.4 Data Analysis

Using NVivo 11, the transcribed interviews were coded and analyzed following the constant comparative method (Corbin & Strauss, 1990) and the Gioia methodology (Gioia, Corley & Hamilton, 2013) which both find their origin within the inductive grounded theory research approach (Charmaz, 2006; Corley, 2015). The grounded theory approach was chosen because it “consists of systematic, yet flexible guidelines for collecting and analyzing qualitative data to construct theories ‘grounded’ in the data themselves” (Charmaz, 2006 p. 2). This research approach has proven to be relevant for research that addresses theoretically underexplored areas (Lawrence & Tar, 2013), such as the CSR strategies of MNEs in developing countries like China (Yin & Jamali, 2016). Andrade (2009) argues that the combination of case study research and grounded theory is useful for interpretive researchers and can enhance each other in the building of new theory (see also Corley & Gioia (2004) who combined case study research based on a grounded theory approach).

The three cases were first analyzed individually via within-case analysis (Andrade, 2009; Eisenhardt, 1989). During the analysis process, each individual transcribed interview was coded following the Gioia methodology. First, this meant coding a line or several sentences, hence separating the data and creating 1st order concepts (Gioia et al., 2013; similar to open coding). Thereafter, the first-order concepts were combined and grouped together into thematically similar 2nd _{order themes,} by constantly comparing the individual 1st order concepts with each other (Corbin & Strauss, 1990) and the preliminary theoretical framework based on the literature (see Figure 2; Yin & Jamali, 2016). This part of the coding process symbolized axial coding (Gioia et al., 2013). Since the coding process was repeated for each MNE, some 2nd order themes were similar across the cases whereas other 2nd order themes were specific to one case. For example, the 2nd order theme “Developing Transnational CSR Strategy” was created specifically for HRCo to outline how the company was moving from a local CSR strategy (which was still the case in China) to a transnational CSR strategy. Lastly, the 2nd order themes were clustered together into aggregate dimensions (Gioia et al., 2013). Although the preliminary framework specified one specific antecedent and outcomes, the analysis revealed other important antecedents and outcomes. Therefore, the aggregate dimensions consists of “Antecedents of

in China”. This resulted in a coherent “data structure”, which allows for a transparent display of the

evidence in a structured manner (Gioia et al., 2013 p. 20; see Figure 3 and Appendix C).

During the cross-case analysis, the researcher searched for patterns and deviations across the cases, based on 2nd _{order themes and 1}st _{order concepts that emerged from the analysis (Eisenhardt,} 1989). Drawing on the suggestions of Eisenhardt (1989) and Yin (2009) for cross-case analysis, comprehensive tables were made to illustrate the similarities and differences across the cases (see Appendix D).

3.5 Quality of the Research

Ensuring a high degree of research quality is a top priority for academic scholars. Focusing on the reliability and validity of research is a means to achieve this quality (Saunders & Lewis, 2012). These two concepts will be discussed more thoroughly below.

3.5.1 Reliability

Reliability within a research setting refers to the possibility that, when one repeats a specific study based on similar procedures, one would find similar outcomes (Saunders & Lewis, 2012). According to Yin (2009, p. 45), “the goal of reliability is to minimize the errors and biases in a study”. In order to ensure the reliability of a study, it is important to create transparency (Saunders & Lewis, 2012). The researcher aimed to provide a high degree of transparency by outlining both the data collection and data analysis in detail (see Section 3.3 and 3.4), through a step-by-step approach. Saunders and Lewis (2012) argue that it is practically infeasible to replicate semi-structured interviews (given their unstructured aspects), hence lowering the reliability. However, transparency is achieved by providing the interview guide of the study in Appendix F.

Moreover, creating transparency means being open and frank about possible biases that can occur when conducting research (Saunders & Lewis, 2012). First of all, research might suffer from observer bias. If this is the case, the opinion of the researcher could influence the outcomes of a study. In order to tackle this problem, the researcher tried to remain as objective as possible whilst avoiding to direct or steer the answers of the interviewees. Moreover, observer bias was reduced since the researcher did not have a personal relationship with any of the interviewees. Secondly, the quality of