CSR: the effect of a home country’s formal and informal institutional pressures on the relation between internationalization and CSiR

(1)

Master Thesis

CSR: the effect of a home country’s formal and

informal institutional pressures on the relation

between internationalization and CSiR

By

Steven Peper

15 June 2020

University of Groningen

Faculty of Economics and Business

(2)

2

Abstract

Why do multinational corporations exercise irresponsible practices and what are their

motives? Perspectives of different kinds of literature are used to answer this question, such as International Business literature, stakeholder theory, and institutional theory. First, I argue that the higher a firm’s level of internationalization, the more irresponsible practices they will exercise. Stakeholders are quite important in this process since multinationals rely on local stakeholders their legitimacy to thrive in a new business environment. Second, several factors may influence the reason why firms exercise irresponsible practices. The institutional theory argues that the home country’s institutional pressures may be an incentive for a firm to decrease their irresponsible practices. Furthermore, the complex situation of institutional duality plays a major role. Therefore, I argue that the home country's formal and informal institutional pressures have a negative moderating effect on the positive relation between internationalization and corporate social irresponsibility. A sample of 454 observations from 111 European firms is used within this research. The results do not indicate a significant positive and moderating effect between the argued assumptions. However, when formal and informal institutional pressures are both integrated, one factor reduced the irresponsible practices of a firm, where another factor increased it. This research complements the

integrated literature with some theoretical implications and provides managerial implications for practice. Furthermore, convenient limitations may help further research on their journey to find the reason why international firms are irresponsible.

(3)

3

Acknowledgments

(4)

4

1. Introduction

In the last years, social and environmental issues, such as child labor and global warming, have been an important and extensive discussion point in worldwide governments and legislation. Through social media, society enforced itself a large share in this debate. Multinational enterprises (MNE’s) are important factors in diminishing social and environmental issues due to their presence in many countries. However, they are also the executioners of these irresponsible behaviors. For example, Coca-Cola kept its distribution costs low by tapping local water resources (MacDonald, 2018), or Nike who violate the human rights of its employees in developing countries (Connor, 2001). But why should successful corporations, such as Coca-Cola and Nike, be irresponsible?

These irresponsible or unethical behaviors are also described as corporate social irresponsibility (CSiR). Armstrong (1977) defined CSiR as: “A decision to accept an

alternative that is thought by the decision maker to be inferior to another alternative when the effects upon all parties are considered” (185). Corruption, fraud, atrocious working

conditions, price-fixing strategies, environmental issues, and child labor are examples of unethical behaviors (Popa & Salanţă, 2014). Internationalization has become a more and more used strategic tool for firms to develop themselves. Moreover, they create value through gaining access to new resources and knowledge (Hitt, Hoskisson, & Kim, 1997), finding new opportunities to grow (Porter, 1990), and it can increase a firm’s competitive advantage (Nachum & Zaheer, 2005). However, internationalization also entails issues for firms. They face complexities in satisfying home and host country stakeholders (Sanders & Carpenter, 1998), and the well-known phenomenon liability of foreignness (LoF) (Zaheer, 1995;

Kostova & Zaheer, 1999; Eden & Miller, 2004; Johanson & Vahlne, 2009). Thus, the more a firm is internationalized, the more difficulties they encounter while doing business. Within an MNE’s subsidiary network, the focal firm may not meet the minimum expectation of their home and host stakeholder groups, or face an increase in their monitoring and controlling activities, which may result in CSiR (Campbell, 2007; Strike, Gao, & Bansal, 2006).

(6)

6 expectations, and regulatory and social pressures (Keig et al., 2019). A high internationalized firm interacts with a variety of formal and informal institutions in their home and host

countries, where a local firm only is confronted by its home country institutions. Furthermore, the foreign subsidiaries of MNE’s face different sets of pressures, and the challenge of

maintaining legitimacy in the host country and with the MNE. Kostova & Roth (2002) defined this conflict between these pressures as institutional duality. Thus, MNE’s face more complexities in establishing and maintaining legitimacy than local firms (Kostova & Zaheer, 1999). These institutional differences and duality faced by MNE’s became a big incentive for firms to internationalize towards and exploit countries with a lower institutional quality (Surroca, Tribö, & Zahra, 2013; Hitt, Li, & Xu, 2016). For example, the transfer of CSiR towards foreign subsidiaries (Surroca et al., 2013). Within this process, firms move their irresponsible practices from their home country towards their host countries due to home country stakeholder pressure (Surroca et al., 2013). Thus, they exploit the institutional environment of their host countries to maintain legitimacy within their home country. This behavior suggests that the institutional quality of the MNE’s home country is higher, relative to their host countries. However, this is not true in every single case. Nowadays, many international firms do originate from countries that have lower institutional quality and expand their activities worldwide.

Research has shown that firms pursue more social responsibility when they face a “strong state regulation, collective industrial self-regulation, nongovernmental organizations and other independent organizations that monitor them, and a normative institutional

environment that encourages socially responsible behavior” (Campbell, 2007: 962). It is clear that countries with high institutional quality have great leverage, and that MNE’s face

complexities in adhering to the different institutional pressures. Surroca et al.'s (2013) research shows evidence that home country stakeholder pressure on the MNE’s headquarters influences the CSiR activities among subsidiaries of internationalized firms. However, there remains a gap in current studies whether a home country’s formal and informal institutions influence MNE’s to enact in less CSiR. As a firm internationalizes, they may exercise more CSiR practices. They may not meet the expectations of both home and host country

(7)

7 Throughout the above-mentioned theories and reasons, I investigate whether formal and informal pressures from home countries negatively affect the positive relation between internationalization and CSiR of MNE’s with the following research question:

Do the home country’s formal and informal institutional pressures negatively moderate the direct positive relation between internationalization and CSiR practices of MNE’s?

The above-argued issue is examined from the perspective of the stakeholder theory. This theory emphasizes the importance of the firm’s responsibility towards its stakeholders, which goes beyond their financial performance (Guthrie, Petty, & Ricceri, 2006). But are MNE’s doing this? Crilly (2011) found that the demands of MNE’s shareholders are placed above those of host country stakeholders. Stakeholders are defined by Freeman (1984) as: “any group or individual who can affect or is affected by the achievement of the firm’s objectives” (46). Employees, customers, suppliers, shareholders, society, and many more can be categorized as stakeholders (Freeman, 1984). However, shareholders of a firm may not place the demands of their stakeholders above the financial performance of the organization, because, in the end, the business of business is business (Friedman, 1970). Each stakeholder group have their expectations and give different pressures towards MNE’s. Furthermore, as a firm internationalizes, its dependence on home and host country stakeholders will increase (Sanders & Carpenter, 1998). Moreover, complexities in managing the different stakeholder pressures arise (Sanders & Carpenter, 1998). Thus, MNE’s face more stakeholder pressures than local firms, due to their worldwide subsidiary network (Kang, 2013). Firms may exercise CSiR when they do not meet the minimum expectation of their stakeholder groups.

Furthermore, stakeholders even have an impact on the construction and implementation of a firm’s socially responsible strategies (Campbell, 2007).

The question of why firms enact unethical behaviors is becoming more and more a subject within present research. However, this exploration is at its beginning, and many aspects still must be researched. Thus, this research should contribute to the academic

literature of corporate social responsibility (CSR) and CSiR. Internationalization creates new opportunities for firms, whether it is in a positive or negative sense. Besides, all these new environments influence a firm’s behavior. This study does not only provide theoretical

(8)

8 stakeholders in host countries are the missing link for firms in doing business abroad.

Unethical behavior may not only harm the firm, but also the stakeholders in multiple ways. This research paper contains multiple sections that contribute to answering the main research question. First, theories are introduced and reviewed, where the topics of

internationalization, CSR, CSiR, and the institutional environment are discussed.

Furthermore, the hypotheses are developed in this section. Second, the methodology of this research is specified and described. Third, the results of the analysis are interpreted and illustrated. Fourth, a comprehensive discussion combines the theory and results, where at last an overall conclusion is constructed.

2. Literature review and hypothesis development

2.1 Internationalization

Johanson & Vahlne (1977) stated that many studies of international business showed that internationalization is the process of a firm, in which its engagement in international practices progressively increases. They developed a model, nowadays known as the ‘Uppsala model’, which firms can use to expand their businesses abroad. Knowledge about foreign markets and operations are key in this model. Lack of knowledge creates obstacles for firms to

internationalize and can mainly be overcome by operating abroad (Johanson & Vahlne, 1977). In other words, experiential learning is key in a firms’ internationalization process.

Furthermore, through the years the concept of a ‘born global’ firm evolved. Knight & Cavusgil (2004) defined this concept as: “business organizations that, from or near their founding, seek superior international business performance from the application of knowledge-based resources to the sale of outputs in multiple countries” (124).

(9)

9 Eden & Miller (2004) argue that the institutional distance between the home and host country is the main factor that creates LoF for an MNE. North (1990) defined institutions as: “the rules of the game in society or, more formally are the humanly devised constraint that shape human interaction” (3). Institutional distance may lead to difficulties for a firm in obtaining legitimacy in a host country (Kostova & Zaheer, 1999). How larger the distance is between the home and the host country, how more difficult it can be to gain legitimacy, which affects the practices of the firm (Kostova & Zaheer, 1999). Legitimacy is the acceptance of the organization by the environment (Zaheer, 1995; Kostova & Zaheer, 1999; Eden & Miller, 2004; Johanson & Vahlne, 2009). Stakeholders will supply resources more towards firms who are seen as legitimate, which eventually helps to thrive in a new environment (Lounsbury & Glynn, 2001). Furthermore, firms are likely to be unsuccessful without legitimacy from stakeholders (Singh, Tucker, & House, 1986). The establishment and maintenance of legitimacy is a crucial issue faced by MNE’s (Kostova & Zaheer, 1999). Furthermore, as MNE’s operates in two or more countries, the legitimacy complexity of that firm increases. MNE’s can have legitimacy demands from their home country as well as from their host countries, which can differ from each other and increases its complexity. This will become even more complex when an MNE is doing business in multiple countries (Kostova & Zaheer, 1999).

As explained before, high internationalized firms face more issues than local firms. When referring to the stakeholder theory, an MNE faces the complexity of satisfying each of their stakeholders in their home and host countries. Stakeholder expectations may be

dissimilar, and firms are therefore influenced differently (Campbell, 2007). Each country in which a firm operates has its stakeholder groups with different expectations from the focal firm. When a firm faces little or none stakeholder pressures, even with the need of realizing profits, they do not prefer to comply with their stakeholders’ social expectations (Campbell, 2007). Pressures to create profits and compliance may push firms towards adhering to the expectations of stakeholders, but not exceeding these expectations (DiMaggio & Powell, 1983). However, the interests of stakeholders with critical resources, and thus of great

importance for firms, are placed above those from less important stakeholders (Guthrie et al., 2006). Furthermore, the demands of MNE’s shareholders are placed above the demands of host country stakeholders (Crilly, 2011).

(10)

10 2.2 Corporate social responsibility

Corporate social responsibility (CSR) has become the last years more and more important for an MNE’s overall performance (Brammer, Pavelin, & Porter, 2008). Studies show that exercising CSR positively affects firm performance in the long-term (Muller, 2018), and increases the legitimacy of MNE’s in host countries (Chen, Patten, & Roberts, 2008). Firms must adapt and comply with the standards which are expected by their stakeholders. The World Business Council for Sustainable Development (2000) defined CSR as: “the continuing commitment by business to behave ethically and contribute to economic

development while improving the quality of life of the workforce and their families as well as the local community and society at large” (Holme & Watts, 2000: 8). Thus, CSR is exercising social activities beyond the interest of the firm (McWilliams & Siegel, 2001). However, stakeholders may perceive CSR differently. Where one stakeholder can identify an action as socially responsible, others may not (Surroca et al., 2013). High internationalized firms could use CSR to overcome issues such as the LoF or gaining legitimacy in host countries.

However, subsidiaries with more distance between home and host countries are less likely to cooperate in CSR activities (Campbell et al., 2012).

2.3 Corporate social irresponsibility

Within the business environment, firms rely significantly on legitimacy in their home and host countries (Kostova & Zaheer, 1999; Lounsbury & Glynn, 2001; Singh et al., 1986). However, where CSR increases an MNE’s legitimacy (Chen et al., 2008), their reputation may be

harmed when they exercise CSiR, which creates difficulties for them in doing business (Strike et al., 2006; Luo, 2001). This since stakeholders may perceive irresponsible practices as illegal or unethical, which results in loss of reputation and legitimacy when a firm exercises CSiR. Even if CSiR is not illegal in a country, stakeholders may still identify it as

irresponsible (Mazzei, Gangloff, & Shook, 2015). We should not perceive CSR and CSiR as two opposite poles. They are both moving towards the same direction as a firm

internationalizes (Strike et al., 2006). Thus, firms that exercise CSR, can also exercise CSiR (Strike et al., 2006). However, we cannot see both as two isolated perceptions, because “a ‘managers’ decisions/actions are sometimes involuntarily placed between the moral and the immoral” (Popa & Salanţă, 2014: 144).

(11)

11 stakeholder pressures than local firms due to their global subsidiary network. Strike et al.’s (2006) research indicated that firm size positively relates to CSiR, which suggests that large firms may face more issues with handling their social relations than smaller firms.

Furthermore, firms may exercise CSiR when they do not meet the minimum expectation of their stakeholder groups (Campbell, 2007). Thus, an MNE may enact in more CSiR than a local firm, due to their higher level of internationalization. Failing to comply with a

stakeholder’s expectation lead to a loss of legitimacy (Mazzei et al., 2015), which could affect their reputation in the home and host country. However, the variety and high amount of relations between an MNE’s’ headquarter and subsidiaries makes it quite complicated for stakeholders to indicate CSiR activities exercised by the MNE (Strike et al., 2006).

Another factor that affects a firm’s CSR level is its monitoring and controlling capabilities (Strike et al., 2006). Internationalized firms face the challenge of coordinating, exchanging, and integrating resources between their subsidiaries which are spread globally (Kostova & Roth, 2003). The larger the MNE, the more extensive its subsidiary network, and the more subsidiaries must be controlled and monitored (Strike et al., 2006). This increment could outweigh its managing capabilities, resulting in not desired behavior by the firm (Strike et al., 2006). To monitor and control its subsidiaries, an MNE needs to expand its resources and capabilities (Strike et al., 2006). However, this process may be perceived as CSiR by the firms’ stakeholders (Strike et al., 2006). Furthermore, Barney (1991) stated that firms can adapt to their diverse stakeholder demands due to their organizational resources. However, MNE’s face complexities in coordinating, exchanging, and integrating resources in their global subsidiary network (Kostova & Roth, 2003). Thus, these issues could lead to less adherence towards stakeholder demands, which may result in more CSiR exercised by an MNE (Campbell, 2007).

To avoid negative publicity, firms may transfer their irresponsible practices towards foreign subsidiaries to maintain their legitimacy in their home country (Surroca et al., 2013). Moreover, they use CSR policies as a strategy to distract people’s attention from their CSiR (Kotchen & Moon, 2012). Due to all the reasons mentioned in this chapter, I suggest that highly internationalized firms are more involved in CSiR practices than a low

internationalized firm. This relation is graphically demonstrated in figure 1.

(12)

12 2.4 Institutional environment

The institutional environment plays a major role in the CSR practices of a firm. North (1990) identified institutions as: “the rules of the game” (3), and are in each country unique, due to the different regulatory and social pressures and expectations (Keig et al., 2019). Carpenter & Feroz (2001) stated that pressures from a firm’s institutional environment are adopted and implemented in their organizational structures and procedures, and these choices are accepted as socially appropriate by the institutional environment. The institutional environment can be categorized into formal and informal institutions (DiMaggio & Powell, 1983; Keig et al., 2014; Keig et al., 2019; Tolmie et al., 2020). These institutions are important for establishing socially responsible behavior by firms in their environment (Campbell, 2007).

Gaining legitimacy can be a challenge for MNE’s in host countries. Moreover, maintaining legitimacy in their home country is equally, or maybe a bigger issue for many multinationals through several pressures from different institutional environments (Kostova & Zaheer, 1999). The complexity of gaining legitimacy can be overcome through isomorphic behavior by the MNE in their host countries. DiMaggio & Powell (1983) mentioned three isomorphic pressures from a host country which could influence the MNE’s structure:

coercive, mimetic, and normative isomorphism. Coercive isomorphism derives from political pressure and legitimacy, mimetic isomorphism indicates the mimicking of other organizations to overcome uncertainty, and normative isomorphism identifies the compliance of firms to the normative rules of institutions (DiMaggio & Powell, 1983). Adopting these structures also helps firms to maintain legitimacy in their home and host country (DiMaggio & Powell, 1983). Furthermore, monitoring legitimacy continuously and creating strategies to respond towards several pressures support firms to maintain legitimacy in their home and host country (Kostova & Zaheer, 1999).

(13)

13 as institutional duality, where subsidiaries face the complexity to maintain legitimacy in their home and host country. This institutional duality may be an incentive for high

internationalized firms to transfer CSiR towards subsidiaries in host countries to maintain legitimacy in their home country (Surroca et al., 2013). However, does this imply that MNE’s prefer maintaining legitimacy in their home country, instead of in their host countries?

‘Window-dressing’ activities, such as transferring irresponsible practices towards subsidiaries in host countries, show conformity towards home country stakeholders, even while MNE’s enact socially irresponsible practices in their host countries (Surroca et al., 2013). Furthermore, Surroca et al.'s (2013) study confirms that subsidiaries exercise CSiR utilizing country stakeholder pressure on the MNE. The motivation for MNE’s to transfer CSiR may be affected by the formal and informal institutions of the MNE’s home country, when they compromise the expectations of stakeholders through surveillance, enforcement, and sanctions for irresponsible behavior (Surroca et al., 2013).

2.5 Formal institutional pressures

Formal institutions embody all the rules and regulations of a country’s environment and are guided by the public and private sector (DiMaggio & Powell, 1983; Keig et al., 2014; Keig et al., 2019; Tolmie et al., 2020). Regulations may force a firm to act in a specific way to gain or maintain legitimacy, or remain competitive in an environment (DiMaggio & Powell, 1983; Kostova & Zaheer, 1999; Mirchandani & Ikerd, 2008). High regulatory constraints may decrease the amount of CSiR exercised by a firm in an environment (Mazzei et al., 2015). There is more awareness and strictness on irresponsibility in such countries, which influences the behavior of firms (Campbell, 2007). Where in contrast, firms may show a more

opportunistic behavior to exercise CSiR in countries with slack regulations, which may lead to less retribution by stakeholders (Mazzei et al., 2015). Firms may even think that they need to exercise CSiR to remain competitive (Mazzei et al., 2015). In the case of formal

institutional CSR pressures from the host country, firms must decide whether they comply with these pressures due to potential conflicting expectations between the home and host countries (Nyuur, Ofori, & Debrah, 2015). Furthermore, they can even result in conflicting stakeholder pressures between their home and host countries (Keig et al., 2019). Thus, MNE’s find themselves in a conflict between the regulatory forces of multiple countries.

(14)

14 institutional pressures (Kostova & Zaheer, 1999; Mazzei et al., 2015). Furthermore, they despise losing legitimacy in their home country (Kostova & Zaheer, 1999; Mazzei et al., 2015). Internationalized firms face due to institutional duality several formal institutional pressures between their home and host countries (Kostova & Roth, 2002). Thus, the more an MNE is internationalized and therefore conduct business in multiple host countries, the more complex institutional duality becomes. Countries with high institutional quality are more strict on social responsibility (Campbell, 2007), which may result in less CSiR exercised by MNE’s (Surroca et al., 2013). This moderation is graphically demonstrated in figure 1.

H2a: Formal institutional pressures from the home country have a negative moderated effect on the relation between internationalization and CSiR.

2.6 Informal institutional pressures

In a country’s environment, the informal institutions cover all the social and cultural norms and behaviors and these institutions are driven by the view on and perceptions of society on corruptive activities exercised by firms (DiMaggio & Powell, 1983; Keig et al., 2014; Keig et al., 2019; Tolmie et al., 2020). MNE’s may face complexity by using universal values to adjust towards a country’s environment (Matten & Moon, 2008). These social values vary across countries, which create this complex situation for high internationalized firms (Tolmie et al., 2020). Adjusting towards a host countries culture and their social behaviors is even more difficult when their institutions differ largely (Keig et al., 2019). Alike formal institutional pressures, informal frictions (e.g. bribery in a business setting), may result in conflicting stakeholder’s pressures between an MNE’s home and host countries. One stakeholder may perceive bribes as normal in a business setting, where another stakeholder with a different cultural background may consider it as irresponsible behavior. Thus, in a country with high institutional quality, firms may adhere more frequently towards pressures to act socially responsible (Campbell, 2007). Furthermore, not complying with these pressures may result in significant consequences, such as loss of legitimacy, for MNE’s (Surroca et al., 2013). Specifically, if media or various pressures on the government initiate action against irresponsible behavior (Doh & Teegen, 2002; Campbell, 2007). However, firms may exercise CSiR within countries with low institutional quality and low caution on irresponsible

behavior, even with high stakeholder pressures (Surroca et al., 2013).

(15)

15 behavior based on social pressures and expectations from institutions (Campbell, 2007). Not complying with these expectations may result in negative publicity and harm the firm with loss of legitimacy (Mazzei et al., 2015). Furthermore, institutional duality also occurs when informal institutional pressures differ between home and host countries (Kostova & Roth, 2002). Thus, whenever a firm increases its internationalization level, institutional duality will become more complex. Eventually, societies and institutions within a home country with high institutional quality are more aware and strict on irresponsible behaviors, which may result in less CSiR by firms (Campbell, 2007; Surroca et al., 2013). This moderation is also graphically demonstrated in figure 1.

H2b: Informal institutional pressures from the home country have a negative moderated effect on the relation between internationalization and CSiR

2.7 Conceptual model

As explained in the literature review, I suggest that internationalization positively affects the amount of CSiR activities exercised by the MNE. Furthermore, the formal and informal institutional pressures of an MNE’s home country may negatively moderate the direct effect of internationalization on CSiR. This results in a lower amount of CSiR practices exercised by the MNE.

(16)

16

3. Methodology

3.1 Data collection

To test the formulated hypotheses, a quantitative research method is used. The objective is to find evidence that internationalization has a direct positive effect on CSiR practices of a firm, and whether home country formal and informal institutional pressures negatively moderates this direct relation. Therefore, the measures are on the firm and country-level. The required quantitative data for this research is retrieved from four databases: Thomson Reuters ASSET4, Orbis, the World Bank, and Hofstede Insights.

(17)

17 3.2 Sample

Many CSR and CSiR researches, such as those of Strike et al. (2006), did research with only US firms in their sample. Therefore, this research is only containing European multinationals to see whether it makes a difference. Furthermore, a panel dataset is developed by first combining the ASSET4 data with the internationalization data from the Orbis database. Hereafter, the data required for the moderated and control variables from the Orbis, World Bank, and Hofstede Insights databases are combined with the developed dataset. The total sample contains 625 observations from 125 firms. The sample size could have been larger, due to missing organizational and financial information of multiple firms. Furthermore, the sample decreased even more, since data of the control variable return on assets (ROA) was not available for some of the firms. Eventually, my dataset for the stepwise regression contains 454 observations from 111 different European firms. The firms in de final dataset originate from the following seventeen countries: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, Ukraine, and the United Kingdom. I chose to use a timeframe of a total of five years in my sample, ranging from 2014 towards 2018. It takes at least one year to observe any effect from the independent and moderating variables on the dependent variable. Therefore, the independent and moderating variables are lagged.

3.3 Measurements of the different variables 3.3.1 Dependent variable

The dependent variable CSiR is measured through the controversy scores in the ASSET4 database. A percentile score in this database determines whether a firm exercise many, less or none CSiR practices. The lower score a firm has, the more CSiR practices it has exercised (Thomson Reuters, 2018).

3.3.2 Independent variable

A firm’s level of internationalization is measured with the foreign sales ratio. This is the most used measure for internationalization and is measured by dividing a firm’s foreign sales by its total sales (Kang, 2013). This financial data is collected from the Orbis database.

3.3.3 Moderating variables

(18)

18 institutional pressures are measured through the WGI of the World Bank. These are six

indicators based on 31 underlying data sources, which captures the governance perceptions from respondents and expert assessments worldwide (Kaufmann, Kraay, & Mastruzzi, 2011). The six indicators are voice and accountability, political stability and absence of

violence/terrorism, government effectiveness, regulatory quality, rule of law, and control of corruption (Kaufmann et al., 2011). A country’s scores are combined in an average WGI score of each year ranging from 2014-2018. The lower the WGI score, the less stringent the formal institutional pressures from the home country are (Keig et al., 2019). However, where Keig et al. (2019) measured the formal institutional distance, this research is only measuring the home country pressures.

Similar to Keig et al. (2019), culture is used as a measurement of informal institutional pressures. However, this research only contains the home country characteristics of each firm in the sample. Data from Hofstede Insights is used to measure the informal institutional pressures from the home countries (Hofstede Insights, n.d.). The data consists of six dimensions from more than one hundred national cultures (Hofstede & Milosevic, 2011). These six dimensions are power distance, uncertainty avoidance, individualism, masculinity, long-term orientation, and indulgence (Hofstede & Milosevic, 2011). Power distance is the level of inequality in society, organizations, and institutions, where uncertainty avoidance is affiliated to a society’s level of stress for the future (Hofstede & Milosevic, 2011).

Individualism is the level integration of individuals in society, where masculinity indicates whether a society is assertive and competitive, or modest and caring (Hofstede & Milosevic, 2011). Long-term orientation indicates whether society focuses on long- or short-term, where indulgence shows a society’s freedom in enjoying life (Hofstede & Milosevic, 2011). The scores of each national culture represent to what extent that country relates to those dimensions (Hofstede & Milosevic, 2011). Thus, the higher the score, the more a country shows cultural traits that are related to that specific dimension.

3.3.4 Control variables

(19)

19 more stakeholder pressure to act in a certain way (Brammer et al., 2008). Thus, when a firm exercises CSiR, this may be identified earlier by stakeholders than with a small firm. The amount of CSiR practices of firms is affected by stakeholder pressures (Mazzei et al., 2015). The number of employees is the measure for the control variable firm size. This since other studies indicated firms size as a strong influence for CSiR (Strike et al., 2006; Keig et al., 2014). The second firm-level control variable is the logarithm of firm age. Older firms may have grown larger subsidiary networks and therefore face more stakeholder pressure to act in a certain way (Brammer et al., 2008). Furthermore, the firm age affects their participation in social activities (Tang, Qian, Chen, & Shen, 2015). Firm age is measured through subtracting the date of incorporation from the year in the panel data. The date of incorporation is available in the Orbis database. The third firm-level control variable is a firm’s industry. The industry in which the firm enacts may influence levels of social irresponsibility (Strike et al., 2006). Furthermore, the industry type is a significant control variable in international business research (Keig et al., 2014). By adapting the manner of Strike et al. (2006), I make use of the one-digit North American Industry Classification System (SIC). The following industries are included in the sample: Mining, Construction, Manufacturing, Transportation and utilities, Wholesale, Retail, Finance, insurance and real estate, and Services. Eventually, these industries are clustered into two generic industries: manufacturing and service industry. Furthermore, a dummy variable is created, where the manufacturing industry is 0, and the service industry is 1. The fourth firm-level control variable is a firm’s CSR scores. As a firm internationalizes, CSR, and CSiR may move in the same direction (Strike et al., 2006). Thus, firms can exercise CSR at the same time as CSiR. For this reason, the CSR score is implemented as a control variable. The CSR scores of each firm in the sample are collected from the ASSET4 database for the year 2014-2018. The fifth firm-level control variable is the return of assets (ROA). Firms with low profitability have fewer resources to invest in CSR (Campbell, 2007). Therefore, firms with financial strains may act more irresponsibly than firms that are financially healthy (Campbell, 2007). Thus, a firm’s level of CSiR may be influenced by its financial healthiness.

(20)

20 country codes are based on the ISO-3166-3 numeric code and are retrieved from the World Bank. Furthermore, a dummy variable is created, where North en West Europe are 0, and East and South Europe are 1.

3.4 Empirical analysis

To test the hypotheses in this research, I make use of stepwise regression. This stepwise regression is operated in the statistical software STATA. Since the hypotheses are tested with the use of panel data, the regression is exercised through a fixed or a random effect. The Hausman test determines whether a fixed or random effect is used (Park, 2011). With the results, relationships between variables can be examined and results can be interpreted. Moreover, it tests whether the hypotheses can be accepted or not. The fixed and random effect regressions test for multicollinearity and omit the variables which show high correlations (Park, 2011). Furthermore, a variation inflation factor (VIF) tests identify whether multicollinearity occurs between variables. Furthermore, I standardized the moderating variables, which exclude a high correlation between the moderators. Since two dummy variables are used in the dataset, I must be cautious about the dummy variable trap. To prevent this, I drop for each variable one dummy. In the end, a robustness check is exercised to examine if the coefficients are sensitive to changes in the model.

3.5 Descriptive statistics

(21)

21 countries have low institutional quality. The six moderators which show the informal institutional pressure an equal average in the sample. Only the dimension Individualism (mean = 73.06 (16.72)) shows a high level. Since the values of firm size, firm age, and GPD per capita varies quite a lot, I have chosen to create a natural logarithm for each variable. The ESG scores of firms in the sample are on average quite mediate (mean = 56.75 (20,59)). Some firms score quite high, were other performed badly on CSR. On average, firms performed quite good on their ROA (mean = 0.06 (0.48)), where generally a ROA of five percent is seen as good. The sample is distributed over seventeen countries, where the United Kingdom is most represented with 36.80 percent. Furthermore, eight industries are observed in the sample, where the manufacturing industry has a majority of 35.20 percent.

Table 2 indicates the correlations between the dependent, independent, moderating, and control variables. Three variables significantly correlated with the dependent variable, which are: Firm size (log) r (-0.29), p < .05, ESG score r (-0.26), p < .05 and ROA r (0.15), p < .05. Thus, none of the independent and moderating variables significantly correlates with the dependent variable. Furthermore, the independent variable significantly correlates (p < .05) within a total of nine variables in the correlation matrix. These correlations are ranging from -0.13 to 0.24. Moreover, there are no high correlations between the dependent and independent variables with other variables in the dataset. Between the moderating variables, many high correlations can be observed. However, McClelland, Irwin, Disatnik, & Sivan (2017) stated that multicollinearity between moderating variables is irrelevant. Moreover, testing for multicollinearity is only relevant between the independent variables. To be sure of no multicollinearity between variables, VIF tests are exercised on the dataset. The VIF-values indicate that eight variables show multicollinearity. However, in this dataset, it is not a problem, since all these variables are moderating variables.

Table 1 – Descriptive statistics and correlation matrix

Variables Mean SD 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

(22)

22

4. Results

The stepwise fixed effect regression of this research contains a total of seven models, which are presented in table 2. To decide whether I should use a fixed or random effect regression, a Hausman test is performed. The result of the Hausman test was p = 0.0816, which indicates that a random effect regression fits my panel data set (Park, 2011). Furthermore, each model’s F-value is significant at a significance level of 0.01. Model 1 includes only the control variables to indicate if my control variables have significant explanatory power. In model 2, the dependent variable is included to test the first hypotheses. The standardized moderating variables which represent the formal institutional pressures are included in model 3. In model 4, the variables of formal institutional pressures are converted by the standardized moderating variables of the informal institutional pressures. Model 5 is the same as model 3, only interaction moderating variables are included in this model. Model 5 tests whether hypotheses 2a can be accepted or not. Model 6 includes the independent, standardized moderating variables and interaction variables for informal institutions, and the control variables to test if hypotheses 2b can be accepted. In the last model, model 7, all variables are included to test an overall result and to see whether significant changes appear or not. The coefficients and significance levels of each WGI, which are included in the combined WGI score, are described in detail and can be found in Appendix B. Thus, this only applies to models 3,5 and 7, since in these models the combined WGI score is included. Furthermore, the F-value and R-squared are shown in table 3. Hence, the adjusted R-squared and log-likelihood are not included since a random effect regression does not calculate these measures. As stated before, the dependent, moderating (standardized), and control variables are measured in t 2014-2018. The independent variable is measured in t 2013-2017. Thus, this variable is already lagged. Furthermore, the moderating variables and controls are lagged (excluding the natural logarithm control variables).

4.1 Results of regressions

(23)

23 The standardized ‘combined WGI score’ is included in model 3. This variable shows a negative coefficient of 3.36, with a p-value of 0.548. This variable consist of the following WGI’s: ‘Control of Corruption’ (β = 13.37, p = 0.110), ‘Government effectiveness’ (β = 16.69, p = 0.002), ‘Political stability and absence of violence/terrorism’ (β = -1.94, p = 0.526), ‘Regulatory quality’ (β = -8.74, p = 0.086), ‘Rule of Law’ (β = -20.00, p = 0.011, and ‘Voice and accountability’ (β = 6.84, p = 0.168). Where the combined WGI score does not indicate any significance, three WGI factors are showing a significant effect. Model 4 contains the following standardized moderating variables: ‘Power distance’ (β = -1.95, p = 0.656), ‘Individualism’ (β = -1.16, p = 0.812), ‘Masculinity’ (β = -3.27, p = 0.213), ‘Uncertainty avoidance’ (β = 1.46, p = 0.823), ‘Long-term orientation’ (β = 1.57, p = 0.547), and ‘Indulgence’ (β = 1.78, p = 0.635). Thus, none of these variables shows any significant effect. The results of model 5 indicate whether hypothesis 2a can be accepted. Only the interaction moderating variables are described in this section since they are used to test hypothesis 2a. The interaction combined WGI score has a negative coefficient of -19.03, with a p-value of 0.519. However, this negative coefficient is not significant. This variable consist of the following WGI’s: ‘Control of Corruption’ (β = 1.15, p = 0.955), ‘Government effectiveness’ (β = -10.01, p = 0.582), ‘Political stability and absence of violence/terrorism’ (β = -4.98, p = 0.654), ‘Regulatory quality’ (β = -21.23, p = 0.168), ‘Rule of Law’ (β = 50.53, p = 0.043, and ‘Voice and accountability’ (β = 2.12, p = 0.886). Only the interaction moderating variable ‘Rule of Law’ shows a significant effect on the relation between internationalization and CSiR. Thus, hypothesis 2a will be rejected. The results of model 6 show whether hypothesis 2b will be accepted or not. Likewise, only the interaction moderating variables are used to test hypothesis 2b. Model 6 contains the following interaction moderating variables: ‘Power distance’ (β = 20.52, p = 0.308), ‘Individualism’ (β = 10.88, p = 0.537), ‘Masculinity’ (β = -2.30, p = 0.772), ‘Uncertainty avoidance’ (β = -12.59, p = 0.588), ‘Long-term orientation’ (β = -5.63, p = 0.630), and ‘Indulgence’ (β = 0.58, p = 0.960). None of the interaction moderating variables shows any significant effect on the relation between internationalization and CSiR. Thus, hypothesis 2b will be rejected.

(24)

(25)

25 ‘Rule of Law’ (β = 45.11, p = 0.097), and ‘Voice and accountability’ (β = 24.82 p = 0.319). Thus, the interaction moderator ‘Rule of Law’ remains its significant positive effect.

Furthermore, model 7 also contains the following interaction moderating variables: ‘Power distance’ (β = 51.26, p = 0.117), ‘Individualism’ (β = -15.39, p = 0.560), ‘Masculinity’ (β = 17.18, p = 0.205), ‘Uncertainty avoidance’ (β = -47.31, p = 0.197), ‘Long-term orientation’ (β = -6.31, p = 0.663), and ‘Indulgence’ (β = -24.74, p = 0.096). The interaction moderating variable ‘Indulgence’ shows a significant negative effect on the relation between

internationalization and CSiR, where this effect was positive and not significant in model 6. 4.2 Robustness check

To test whether the coefficients in the stepwise regression are sensitive for changes, a mixed effect level regression is performed and can be found in appendix C. Within this multi-level regression, two multi-levels are included: country and firm-multi-level. The models of both the stepwise regression and multi-level regressions are compared with each other in this section. The coefficients and significance levels of each WGI, which are included in the combined WGI score, are described in detail and can be found in Appendix D. The mixed-effect multilevel regression measures the log-likelihood and the Akaike Information Criterion (AIC). This AIC is a measure to indicate which model is better performing (Stylianou, Pickles, & Roberts, 2013). Besides, the log-likelihood is also a measure to argue which model is better performing.

In model 1, both the control variables ‘Firm size (log)’ (β = -3.55, p = 0.001) and ‘ESG score’ (β = -0.27, p = 0.003) have the same sign, size, and significance as those in the random effect regression. The other control variables also remained the same sign, size, and significance. Thus, model 1 shows that the results are robust. Furthermore, the independent variable in model 2 (β = 5.18, p = 0.334) also remains the same sign, size, and significance as in the random effect regression. Thus, model 2 indicates robust results.

(26)

26 The interaction moderating variable ‘Combined WGI score’ (β = -19.08, p = 0.509) shows in model 5 the same sign, size, and significance as in the random effect regression. Furthermore, the WGI factor ‘Rule of Law’ (β = 50.69, p = 0.037) within this combined score prevailed its sign, size, and significance. The other WGI factors also remained the same. Furthermore, the interaction moderating variables in model 6 remained the same sign, size, and significance as in the random effect regression. Thus, the results of model 6 are also robust.

In model 7, the independent variable (β = 26.18, p = 0.493), the interaction moderating variable ‘Combined WGI score’ (β = -16.17, p = 0.591), the other interaction moderating variables, ‘Individualism’ (β = -14.97, p = 0.546), ‘Masculinity’ (β = 16.86, p = 0.187), ‘Uncertainty avoidance’ (β = -46.44, p = 0.174), ‘Long-term orientation’ (β = -6.95, p = 0.610), and ‘Indulgence’ (β = -24.42, p = 0.076) prevailed the same sign, size and significance as in the random effect regression. The WGI factors within the interaction moderating variable ‘Combined WGI score’ also remained the same. However, the interaction moderating variable ‘Power distance’ (β = 51.49, p = 0.095), has the same sign and size, but its p-value has changed. This variable has now a significant positive effect, where it had a positive effect on the random effect regression. Nonetheless, this change is too small to conclude that the results are not robust.

(27)

27

5. Discussion

This study was arranged to help broaden the CSiR studies within the CSR topic. It explores whether the level of internationalization positively influences the amount of CSiR activities a firm exercise. Furthermore, a moderating effect was used to predict if a home country’s formal and informal institutional pressures negatively influence the above-mentioned relation. A panel data sample of 454 observations, consisting out of 111 firms originating from countries in the European Union, is trying to predict these relations. The analyses may help to understand why firms are undertaking CSiR and whether these firms are affected by external factors. In this section, the interpretations of the results are discussed and whether they contribute to the existing literature and practice. Furthermore, the limitations and potential future research are mentioned.

5.1 Interpretation of the results

(28)

28 do not infer a significant positive relation between internationalization and CSiR. Hence, the results did show a positive relation, but not a significant positive relation. Furthermore, the other models in the stepwise regression indicate a positive relation. Yet, these relations are also not significant. Thus, there is not enough evidence in this research to claim that the higher a firm’s internationalization level, the more CSiR it exercises. It may be that the firms in this research did not face any complexities to maintain their legitimacy. Their stakeholder demands may be minimum, or their monitoring and controlling costs did not outweigh their managing capabilities (Strike et al., 2006; Campbell, 2007; Mazzei et al., 2015). Likewise, Strike et al's. (2006) results did also not identify a significant positive relationship between both variables.

With hypotheses 2a and 2b, I argue that a firm’s formal and informal institutional pressures from the home country negatively influences the positive relation between internationalization and CSiR. This infers that different regulatory and social pressures from the MNE’s home country reduce the amount of CSiR activities exercised by that specific firm. These pressures create issues for multinationals in maintaining legitimacy in their home country (Kostova & Zaheer, 1999). For example, adhering towards host country stakeholders may result in loss of legitimacy in the home country (Kostova & Zaheer, 1999). Furthermore, irresponsible practices may be seen as normal in the host country, but irresponsible in the firm’s home country (Strike et al., 2006). This institutional duality may result in a transfer of social irresponsibility towards an MNE’s host country subsidiaries (Surroca et al., 2013). Thus, the following question arises: Do MNE’s prefer their home country legitimacy over host country legitimacy? The level of internationalization plays an important role in this question. The more a firm is internationalized, the more it is relying on its activities or subsidiaries in host countries. However, the formal and informal institutional pressures do also play a significant role.

(29)

29 they do not want to lose their legitimacy. This may result in a lower amount of CSiR practices exercised by MNE’s. However, the results indicate a rejection of hypothesis 2a. The ‘combined WGI score’ does show a negative influence on the relation between internationalization and CSiR. Although, this moderating relation is not significant. The fact that firms may enact in CSiR to remain competitive could explain this outcome (Mazzei et al., 2015). Within the ‘Combined WGI score’, the variable ‘Rule of Law’ does show a significant positive effect. Rule of law can be captured as the acceptance of and believing in the rules of society, such as contract enforcement, the police, and also the likelihood of crime and violence (Kaufmann, Kraay, & Mastruzzi, 2011). It may be that the host countries of the firms in the sample have the same level of formal institutional quality as their home country, or even higher. Besides ‘Rule of Law’, the variables ‘Government effectiveness’, ‘Political stability and absence of violence/terrorism’, and ‘Regulatory quality’ indicates a negative effect. However, these effects are not significant.

(30)

30 MNE which originates from this country exercises less CSiR. A reason for this may be that firms have such good perceptions of a normal and free life by society, that they also want to experience this in the countries where they do business. They may even feel the responsibility to accomplish this in countries where society is suppressed by their government. 5.2 Theoretical implications

Since none of the hypotheses are accepted, there is no absolute explanation of the discussed theory in this research. However, the findings still show some theoretical implications. The stakeholder theory indicates that international firms are greatly dependent on their home and host country stakeholders (Sanders & Carpenter, 1998). Stakeholders will supply resources more towards firms who are seen as legitimate, which helps to compete in a new business environment (Lounsbury & Glynn, 2001). Furthermore, stakeholders have a great influence on the complexities which multinationals face, such as complying with home or host country stakeholder demands (Campbell, 2007; Mazzei et al., 2015). However, when a firm faces little or no stakeholder pressure, they prefer to not satisfy their stakeholders’ social expectations (Campbell, 2007). Different scholars suggest that high internationalized firms may enact more in CSiR than low internationalized firms due to these stakeholder demands (Campbell, 2007; Mazzei et al., 2015). This present research complements this assumption. However, this presumption is not without limits, as this present research does not show this significant positive relation between internationalization and CSiR.

(31)

31 5.3 Managerial implications

Furthermore, the results of this research do give some implications for practices in CSR. The question of whether high internationalized firms do exercise more CSiR is not answered, such as in other researches (Strike et al., 2006). However, this notion can still be a reality. MNE’s still use CSR to gain and maintain legitimacy in home and host countries and both CSR and CSiR move in the same direction as a firm internationalizes (Strike et al., 2006). In the past, the motives of businesses were only to fulfill their shareholder demands (Friedman, 1970). Nowadays, it shifted more towards a stakeholder’s perspective. I believe that in today’s business environment, firms must behave socially and environmentally responsible, since their impact in the world and society has grown, and still increases with the years.

Furthermore, as the results of this research give a diverse answer to the question of whether a home country with high institutional quality reduces a firm’s CSiR, it may still be useful for internationalized firms. A firm’s CSiR reduces when their home country gives society freedom in enjoying life, where acceptance of the rule of society increases a firm’s level of CSiR. Thus, institutional quality can have different effects. Firms and stakeholders should be keen on the differences between home and host countries and decrease the number of complexities to as few as possible. These complexities may be social or environmental externalities, such as poor work condition regulations in a host country, or global warming. A firm that strives to improve or mitigate such externalities increases its legitimacy and firm performance in the long-term (Chen et al., 2008; Muller, 2018). Thus, firms should not only do this for themselves to prevent loss of legitimacy but also create better living circumstances for society. Especially big firms can have a great impact when they pursue social and environmental responsibility. Their influence and voice are far-reaching, and they have the resources to create such development at large scale.

(32)

32 host countries may decrease due to CSiR, which has a major impact on a firm’s financial performance and successes (Singh et al., 1986; Kostova & Zaheer, 1999).

5.4 Limitations and further research

Where this study has made some contributions to the existing literature, several limitations have occurred within this research. Moreover, these flaws may encourage future research to use other approaches. Firstly, this research used a total of four databases to retrieve continuous data. A lot of data was lost within the process of collecting since firm information did not exist or was not complete. Thus, the sample size could have been bigger than in the present. This may have influenced the results of the regressions. Hence, this is an assumption and the results may have remained the same whenever the sample size was larger. However, when future research uses continuous data, the better approach is to collect this data within the same database.

Secondly, CSiR activities exercised by firms are mostly done behind the scenes. Thus, many unethical behaviors may not have been captured, which may imply that scores are different than they truly are. The Environmental Social Governance controversies score in the ASSET4 database is based on the count of negative publicity in public available information sources (Thomson Reuters, 2018). Where the Environmental Social Governance controversies score is a good representative of the unethical behaviors of firms, it may have its flaws in capturing every single CSiR activity. However, this is quite difficult to achieve since most firms try to keep their CSiR hidden from the public. Furthermore, the perspective of ethical behavior is different for everyone. Country’s with great oppression and no freedom of speech and media may not write about a firm’s unethical behavior. Thus, this could influence the Environmental Social Governance controversies score of the ASSET4 database. Since the ASSET4 database uses a numeric score, a subjective sense is missing in the information of these controversies. Future research must keep this in mind and may choose another database that captures this kind of information for unethical behaviors exercised by firms.

Thirdly, within the sample, more than 90 percent of the firms have over 1000 employees. Thus, the results of this research are not generalizable for smaller firms. Future research may also try to indicate the level of CSiR of smaller firms. However, this is quite difficult to capture, since smaller firms are not followed as closely as bigger firms.

(33)

33 may want more distinct results and focus on one instead of both institutions. Furthermore, future research could also use other measures to capture formal and informal institutional quality, which may result in other outcomes.

6. Conclusion

(34)

34

Bibliography

Armstrong, J. S. 1977. Social irresponsibility in management. Journal of Business Research, 5(3): 185–213.

Brammer, S. J., Pavelin, S., & Porter, L. A. 2008. Corporate charitable giving, multinational companies and countries of concern. Journal of Management Studies, 46: 575–596. Bureau van Dijk. 2019. Orbis; The world’s most powerful comparable data resource on

private companies. https://www.bvdinfo.com/en-gb/our-products/data/international/orbis.

Campbell, J. L. 2007. Why would corporations behave in socially responsible ways? An institutional theory of corporate social responsibility. Academy of Management Review, 32(3): 946–967.

Campbell, J. T., Eden, L., & Miller, S. R. 2012. Multinationals and corporate social responsibility in host countries: Does distance matter ? Journal of International Business Studies, (43): 84–106.

Carpenter, V. L., & Feroz, E. H. 2001. Institutional theory and accounting rule choice : an analysis of four US state governments ’ decisions to adopt generally accepted accounting principles. Accounting, Organizations and Society, 26(7–8): 565–596.

Chen, J. C., Patten, D. M., & Roberts, R. W. 2008. Corporate charitable contributions: A corporate social performance or legitimacy strategy? Journal of Business Ethics, 82(1): 131–144.

Connor, T. 2001. Still waiting for Nike to respect the right to organize. Corpwatch. https://corpwatch.org/article/still-waiting-nike-respect-right-organize.

Crilly, D. 2011. Predicting stakeholder orientation in the multinational enterprise: A mid-range theory. Journal of International Business Studies, 42(5): 694–717.

DiMaggio, P. J., & Powell, W. W. 1983. The iron cage revisited institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48: 147–160.

Doh, J. P., & Teegen, H. 2002. Nongovernmental organizations as institutional actors in international business: theory and implications. International Business Review, 11(6): 665–684.

Eden, L., & Miller, S. R. 2004. Distance matters : Liability of foreignness , institutional distance and ownership strategy. Theories of the Multinational Enterprise: Diversity, Complexity and Relevance, 187–221.

Freeman, R. E. 1984. The stakeholder concept and strategic management. Strategic management: A stakeholder approach: 31–51.

Friedman, M. 1970, September 13. The social responsibility of business is to increase its profits. The New York Times Magazine, 122–126.

(35)

35 Guthrie, J., Petty, R., & Ricceri, F. 2006. The voluntary reporting of intellectual capital

Comparing evidence from Hong Kong and. Journal of Intellectual Capital, 7(2): 254– 271.

Hitt, M. A., Hoskisson, R. E., & Kim, H. 1997. International diversification : Effects on innovation and firm Performance in product-diversified firms. Academy of Management Journal, 40(4): 767–798.

Hitt, M. A., Li, D., & Xu, K. 2016. International Strategy : From local to global and beyond. Journal of World Business, 51(1): 58–73.

Hofstede, G., & Milosevic, D. 2011. Dimensionalizing cultures: The Hofstede model in context. Online Readings in Psychology and Culture, 2(1): 1–17.

Hofstede Insights. n.d. Compare countries. https://www.hofstede-insights.com/product/compare-countries/.

Holme, R., & Watts, P. 2000. Corporate social responsbility: making good business sense. World Business Council for Sustainable Development.

Johanson, J., & Vahlne, J.-E. 1977. The internationalization process of the firm - a model of knowledge development and increasing foreign market commitments. Journal of International Business Studies, 8(1): 23–32.

Johanson, J., & Vahlne, J. E. 2009. The Uppsala internationalization process model revisited: From liability of foreignness to liability of outsidership. Journal of International Business Studies, 40(9): 1411–1431.

Kang, J. 2013. The relationship between corporate diversification and corporate social performance. Strategic Management Journal, 34(1): 94–109.

Kaufmann, D., Kraay, A., & Mastruzzi, M. 2011. The Worldwide Governance Indicators: Methodology and analytical issues. Hague Journal on the Rule of Law, 3(2): 220–246. Keig, D. L., Brouthers, L. E., & Marshall, V. B. 2014. Formal and informal corruption

environments and multinational enterprise social irresponsibility. Journal of Management Studies, 52(1): 89–116.

Keig, D. L., Brouthers, L. E., & Marshall, V. B. 2019. The impact of formal and informal institutional distances on MNE corporate social performance. International Business Review, 28(5): 1–10.

Knight, G. A., & Cavusgil, S. T. 2004. Innovation, organizational capabilities, and the born-global firm. Journal of International Business Studies, 35(2): 124–141.

Kostova, T., & Roth, K. 2002. Adoption of an organizational practice by subsidiaries of multinational corporations: Institutional and relational effects. The Academy of Management Journal, 45(1): 215–233.

Kostova, T., & Roth, K. 2003. Social capital in multinational corporations and a micro-macro model of its formation. The Academy of Management Review, 28(2): 297–317.

(36)

36 Kotchen, M. J., & Moon, J. J. 2012. Corporate social responsbility for irresponsibility. The

BE Journal of Economic Analysis & Policy, 12(1): 1–25.

Lounsbury, M., & Glynn, M. A. 2001. Cultural entrepreneurship: Stories, legitimacy, and the acquisition of resources. Strategic Management Journal, 22(6–7): 545–564.

Luo, Y. 2001. Toward a cooperative view of MNC-host government relations: Building blocks and performance implications. Journal of International Business Studies, 32(3): 401–419.

MacDonald, C. 2018. Coke claims to give back as much water as it uses. An investigation shows it isn’t even close. The Verge.

https://www.theverge.com/2018/5/31/17377964/coca-cola-water-sustainability-recycling-controversy-investigation.

Matten, D., & Moon, J. 2008. “Implicit” and “explicit” CSR: A conceptual framework for a comparative understanding of corporate social responsibility. Academy of Management Review, 33(2): 404–424.

Mazzei, M. J., Gangloff, A. K., & Shook, C. L. 2015. Examing multi-level effect on corporate social responsibility and irresponsibility. Management & Marketing, 10(3): 163–184. McClelland, G. H., Irwin, J. R., Disatnik, D., & Sivan, L. 2017. Multicollinearity is a red

herring in the search for moderator variables: A guide to interpreting moderated multiple regression models and a critique of Iacobucci, Schneider, Popovich, and Bakamitsos (2016). Behavior Research Methods, 49(1): 394–402.

McWilliams, A., & Siegel, D. 2001. Corporate social responsibility : A theory of the firm perspective. The Academy of Management Review, 26(1): 117–127.

Mirchandani, D., & Ikerd, J. 2008. Building and maintaining sustainable organizations. Organization Management Journal, 5(1): 40–51.

Muller, A. 2018. When does corporate social performance pay for international firms? Business and Society, 1–35.

Nachum, L., & Zaheer, S. 2005. The persistence of distance? The impact of technology on MNE motivations for foreign investment. Strategic Management Journal, 26(8): 747– 767.

North, D. C. 1990. An introduction to institutions and institutional change. Institutions, Institutional Change and Economic Performance: 3–10.

Nyuur, R. B., Ofori, D. F., & Debrah, Y. A. 2015. The impact of FDI inflow on domestic Firms ’ uptake of CSR activities: The moderating effects of host institutions.

Thunderbird International Business Review, 58(2): 147–159.

Park, H. M. 2011. Practical Guides To Panel Data Modeling : A Step-by-step Analysis Using Stata. Tutorial Working Paper. Graduate School of International Relations, International University of Japan.

Popa, M., & Salanţă, I. 2014. Corporate social responsibility versus corporate social irresponsibility. Management and Marketing, 9(2): 137–146.

(37)

37 Sanders, W. M. G., & Carpenter, M. A. 1998. Internationalization and firm governance : The

roles of CEO compensation , top team composition , and board structure. Academy of Management Journal, 41(2): 158–178.

Singh, J. V, Tucker, D. J., & House, R. J. 1986. Organizational legitimacy and the liability of newness. Administrative Science Quarterly, 31(2): 171–193.

Strike, V. M., Gao, J., & Bansal, P. 2006. Being good while being bad: Social responsibility and the international diversification of US firms. Journal of International Business Studies, 37(6): 850–862.

Stylianou, C., Pickles, A., & Roberts, S. A. 2013. Using Bonferroni, BIC and AIC to assess evidence for alternative biological pathways: Covariate selection for the multilevel Embryo-Uterus model. BMC Medical Research Methodology, 13(1): 1–13.

Surroca, J., Tribö, J. A., & Zahra, S. A. 2013. Stakeholder pressure on MNEs and the transfer of socially irresponsible practices to subsidiaries. Acadamy of Management Journal, 56(2): 549–572.

Tang, Y., Qian, C., Chen, G., & Shen, R. 2015. How CEO hubris affects corporate social (ir)responsiblity. Strategic Management Journal, 36(9): 1338–1357.

Thomson Reuters. 2018. Thomson Reuters ESG scores.

https://www.refinitiv.com/content/dam/gl/en/documents/methodology/esg-scores-methodology.pdf.

Tolmie, C. R., Lehnert, K., & Zhao, H. 2020. Formal and informal institutional pressures on corporate social responsibility: A cross-country analysis. Corporate Social

Responsibility and Environmental Management, 27(2): 786–802.

(38)

38

Appendix A: Sample distribution of country of headquarters, industry, and

year

Country of Headquarters Freq. Percentage Cum.

Austria 5 0.80 0.80 Belgium 25 4.00 4.80 Denmark 5 0.80 5.60 Finland 25 4.00 9.60 France 65 10.40 20.00 Germany 10 1.60 21.60 Greece 35 5.60 27.20 Ireland; Republic of 30 4.80 32.00 Italy 15 2.40 34.40 Luxembourg 5 0.80 35.20 Netherlands 55 8.80 44.00 Norway 10 1.60 45.60 Portugal 5 0.80 46.40 Spain 70 11.20 57.60 Sweden 30 4.80 62.40 Ukraine 5 0.80 63.20 United Kingdom 230 36.80 100.00 Total 625 100.00

Industry Freq. Percentage Cum.

Construction 40 6.40 6.40

Finance, insurance, and real estate 25 4.00 10.40

Manufacturing 220 35.20 45.60

Mining 40 6.40 52.00

Retail 25 4.00 56.00

Services 165 26.40 82.40

Transportation and utilities 65 10.40 92.80

Wholesale 45 7.20 100.00

Total 625 100.00

Year Freq. Percentage Cum.

(39)

39

Appendix B: Additional formal institutional pressure moderating variables

in the stepwise regression

Moderating variables (1) Model 3 Model 5 Model 7

Control of Corruption 13.368 (8.368) 12.750 (15.635) 12.487 (20.294) Government effectiveness 16.685** (5.363) 22.612+ (13.429) 22.912 (15.635) Political stability and absence

of violence/terrorism -1.938 (3.056) 2.610 (8.602) -5.948 (10.871) Regulatory quality -8.739+ (5.089) (11.649) 5.651 (16.853) -2.364 Rule of Law -19.998* (7.842) -52.091** (17.974) -46.658* (19.806)

Voice and accountability 6.843

(4.965) (10.731) 7.498 (17.234) -5.191 Foreign sales ratio *

Control of corruption (20.258) 1.153 (27.463) 6.079

Foreign sales ratio *

Government effectiveness (18.160) -10.006 (22.290) -9.392 Foreign sales ratio *

Political stability and absence of violence/terrorism

-4.979

(11.112) (14.311) 7.030 Foreign sales ratio *

Regulatory quality

-21.233 (15.394)

-8.643 (22.685) Foreign sales ratio *

Rule of Law (24.952) 50.528* (27.188) 45.111+

Foreign sales ratio * Voice and accountability

2.120 (14.789)

(40)

40

CSR: the effect of a home country’s formal and informal institutional pressures on the relation between internationalization and CSiR

Master Thesis