Effects of Foreign Directors’ Nationalities and Director Types on Corporate Philanthropic Behavior: Evidence from Korean Firms

Abstract

Addressing the fact that there are few studies exploring the relationship between board characteristics and corporate social responsibility (CSR) in non-Western contexts, this study examines the relationship in South Korean corporate contexts. We concentrate on foreign directors as a board attribute, which is reported as a remarkable change in Korean corporate boards, and propose that foreign directors have different impacts on CSR investment depending on their nationality (Anglo-Americans vs. non-Anglo-Americans) and director types (insiders vs. outsiders). In detail, the presence of directors from Anglo-American countries (e.g., the United States, the United Kingdom) decreases firms’ CSR involvement, whereas the presence of directors from non-Anglo-American countries (e.g., France, Germany) increases firms’ CSR involvement. Moreover, the effects of Anglo-Americans on CSR are strengthened when they are inside (rather than outside) directors. Empirical analyses using a sample of 1828 Korean firms from 2002 to 2015 provide evidence to support the predictions. This study theoretically contributes to CSR and corporate governance literature in that it sheds light on the CSR in non-Western companies and reveals varied effects of foreign directors contingent upon their individual attributes. It also has practical implications for policymakers and corporate managers by providing insights of the changes generated by foreign members in a boardroom.

1. Introduction

The importance of corporate social responsibility (CSR) has increased in recent years [1,2,3,4]. Companies have begun to recognize their obligations to society and a wide variety of stakeholders, and have engaged in socially responsible activities such as enacting relevant policies, programs, and practices [5,6]. Although companies’ motivations to participate in CSR are not always normative (i.e., to do the right thing), it is widely accepted that CSR is needed for improving social welfare [1]. However, there are variances in the extent of CSR investment among companies [7].

Accordingly, numerous scholars have explored the factors that encourage firms’ CSR activities [7,8,9,10,11,12]. In particular, a corporate board of directors has been suggested as an important determinant of CSR, because the board, as the highest decision-making group, greatly influences the strategic choices of the corporation [13,14,15]. For instance, board characteristics such as board size [16], independent director presence [17,18,19], director ownership [19], director background [20], board diversity [17], and the presence of an audit committee [18] have been examined.

Despite the numerous studies on CSR and corporate boards, there are relatively few studies that explore CSR in non-Western contexts [5]. Addressing this limitation of most prior studies having concentrated heavily on Western companies [5], a few recent studies have begun to pay attention to non-Western companies, especially those in Asia (e.g., [17,18,21,22,23,24]). The reason for the growing attention is that there are unique norms and behaviors in Asian CSR practices owing to institutional and cultural differences, and that the importance of Asian economies has increased continuously [17,23].

In this study, we explore the effects of board characteristics on CSR in South Korean companies. In particular, we focus on foreign directors in a boardroom. The appointment of foreign members is reported as a notable change in Korean corporate boards during the past several decades, because Korean boards have long been ethnically homogeneous, based on a single-ethnic, strong collective culture [21,25]. Korean firms made this change in an effort to improve corporate governance after weak governance systems were identified as an important cause of the Asian Financial Crisis in the late 1990s [26]. However, since the increased foreign board representation is a recent phenomenon, it is still poorly understood [25,27]. Corporations often appoint foreign directors without a deep understanding of the changes made by the foreign members in a boardroom [25]. Moreover, prior relevant studies on corporate governance (e.g., board diversity) are limited to explain this phenomenon, because they usually focus on Western countries, which have different institutions to Korea [17]. For example, foreign directors appointed to a board in a single-ethnic country would experience different conditions than in a multiethnic country, which leads them to behave differently and generates different consequences.

Our research question is how do foreign directors influence CSR decisions in Korea? We extend the question to include the attributes of foreign members in order to understand the effects of foreigners better. To add more detail, we address nationality and director type in the question, how does the influence of foreign directors differ depending on their nationality and director types? We employ national institutional difference literature [5,28] to address the distinct effects of directors’ nationality, and board dynamics literature [29] to address the effects of director type.

The results from the 1828 Korean firms during the 2002–2015 period reveal that the impact of foreign directors on CSR investment varies depending on their nationality (Anglo-Americans vs. non-Anglo-Americans) and director type (inside vs. outside directors). More specifically, the presence of foreign directors from Anglo-American countries (e.g., the United States, the United Kingdom), which operate a shareholder-centered model based on strong investor protection mechanisms and widespread recognition of firms as isolated actors, decreases firms’ CSR involvement. In contrast, the presence of foreign directors from non-Anglo-American countries (e.g., France, Germany), in which the stakeholder-centered model is operated based on weak investor protections and the recognition of firms as associated actors, increases firms’ CSR involvement. Moreover, the impact of Anglo-Americans on CSR is greater when they are inside, rather than outside, directors, owing to decreased constraints faced by foreigners in a minority position.

This study is expected to theoretically contribute to CSR and corporate governance literature in that it is one of a few studies examining the CSR in non-Western contexts and reveals varied effects of foreign directors depending on their individual attributes (i.e., nationalities and director types) above foreignness. It also has practical implications for corporate managers and policymakers who are considering the encouragement of CSR by providing insights of the strategic changes generated by foreign members on corporate boards.

The remainder of this paper is organized as follows: Section 2 reviews the relevant literature to establish theoretical foundations; Section 3 develops the hypotheses proposed; Section 4 documents the data and methods of analysis; Section 5 describes the empirical results; and Section 6 discusses the findings and implications of the study.

2. Theoretical Background

2.1. Corporate Social Responsibility

Although a number of scholars have suggested definitions and frameworks for CSR, the concept of CSR is still ambiguous and inconsistent [3]. For instance, Aguinis defined CSR as “context-specific organizational actions and policies that take into account stakeholders’ expectations and the triple bottom line of economic, social, and environmental performance” [1,30]. Carroll proposed a framework of CSR with four categories: economic, legal, ethical, and discretionary responsibilities [31]. Generally speaking, most definitions of CSR include corporate attention to a wide range of stakeholders beyond shareholders (e.g., employees, customers, communities, minorities, the environment), and contributions to a wide range of society beyond the minimum legally required [3,32].

CSR has become an important subject to companies in recent years. CSR was treated as a trifling joke until the 1970s [6], and most Fortune 500 companies did not even refer to the issue in their annual reports in the late 1970s [33]. However, in recent years, companies have begun to recognize their obligations to a variety of stakeholders and society in general, and actively embraced social and environmental issues in launching relevant policies, programs, and practices [5,6]. Companies’ participation in CSR is explained by normative reasons to do the right thing (i.e., altruistic CSR), or by instrumental reasons to enhance financial performance (i.e., instrumental CSR) [1,3]. The importance of CSR spreads rapidly because the needs of society and companies coincided. Society needs companies to contribute to social welfare, because it is difficult for governments alone to achieve the level of social welfare people desire today, and companies also need to participate in social activities to differentiate themselves in a complex, competitive business environment [4].

2.2. Corporate Boards and Corporate Social Responsibility

Although CSR is expected to grow continuously, there are variances in the extent of CSR involvement among companies [7]. Accordingly, numerous scholars have explored the determinants of CSR involvement from various perspectives. For instance, with respect to external contexts, industry [9], customers [11], and laws and regulations [7] have been examined. With respect to the organization, firm size [8,10], organizational slack [12], and the ethical beliefs of top management teams [34] have been studied.

In this stream of research, the board of directors is suggested as an important organizational factor affecting a firm’s CSR decisions. The corporate board, a top-level decision-making group, has long been considered to influence the strategic choices of a corporation substantially by monitoring and advising managerial decisions [13,15,29]. The board has the responsibility and authority to review, advise, and ratify proposals suggested by managers. It also monitors the management’s execution, which influences the organizational choices made. Given that CSR is an organizational decision, scholars have examined how board characteristics affect CSR. For instance, they have examined board size [16], independent outside directors [17,18,19,20], director ownership [19], director background [20], board diversity [17,20], and the presence of an audit committee [18]. Indeed, 80% of corporate boards were found to be actively engaged in setting the firm’s ethical standards in the late 1990s, a great increase over the prior decades [35].

However, our understanding of the relationship between board characteristics and CSR practices in a non-Western context is relatively poor. Although there are numerous studies on corporate governance and CSR, prior studies have focused heavily on Western contexts such as the United States and Europe [5,18], and thus the theories suggested have limited application to non-Western companies [17]. This is because CSR activities in non-Western countries have distinctive norms and behaviors, as corporate governance and CSR activities are greatly affected by institutions in which companies are embedded [17,22]. Moreover, it is necessary to understand these differences because interdependence among countries has increased rapidly [17] and the importance of non-Western, particularly Asian, companies is increased [23]. Therefore, a few scholars have recently begun to explore non-Western contexts, particularly in Asian countries (e.g., [17,18,21,22,23]), but the number of relevant studies is still small, and our knowledge remains poor.

As part of this line of research, this study examines the effects of board attributes on CSR participation in a non-Western context, particularly in South Korea. South Korea provides an interesting research setting for this study. After weak corporate governance was identified as a critical cause of the Asian Financial Crisis in the late 1990s, the South Korean government introduced regulations to compel companies to reengineer their board structures in ways that improved accountability and transparency [26]. For example, publicly held firms were required by law to have more than 25% (or 51% for firms with $2 billion or more in assets) of their boards be outside independent directors, and these guidelines have been strengthened as the laws have been revised [27,36]. Accordingly, Korean companies have well-structured boards after accepting Western practices and demonstrate a willingness to continuously improve them as well. Furthermore, Korean companies have recognized the importance of CSR and have begun to actively participate in relevant activities [17]. The investigation on the board–CSR relationship is therefore theoretically and practically meaningful.

2.3. Foreign Directors

It is important to understand who are on the boards to understand boards’ effects. This is because board decisions are made by individual members in the boardroom through interactive activities among themselves (e.g., information sharing and information processing), even though board decisions are organizational choices [3,20,29,37]. Moreover, considering that CSR is a strategic choice that pursues long-term performance that could hurt short-term performance, the recognition and value of decision makers have a significant impact [38,39]. Therefore, this study assumes that the attributes of board members are an important board characteristic.

In particular, this study concentrates on the introduction of foreign directors, which was one of the most important changes to Korean boards during the last several decades [25]. Board diversity (e.g., gender, nationality diversity) has been a widely examined characteristic in the literature on boards, because the presence of members distinct from the dominant members in the group is believed to make the board more vigilant and extend the resources utilized, consequently enhancing board quality [37]. However, Korean corporate boards have long been ethnically homogeneous owing to strong unitary nationalism and a collective culture [21]. With the tide of board restructurings since the late 1990s, Korean companies have begun to appoint foreign members to their boardrooms to enhance board composition by benchmarking other firms, or after being forced by institutional investors [21]. The percentage of foreign directors in the top 100 Korean companies reached 1.5% in 2015. This figure is still small, but it represents a radical change to previously excessively homogeneous Korean boards.

However, since the appointment of foreign board members is a recent phenomenon, their effects are not revealed yet. Most Korean companies often recruit foreign members to their boards to externally show that they have effective board composition or to follow the interests of controlling shareholders, without a deep understanding the consequences of having these members [25,40]. Furthermore, despite abundant prior research on board diversity, the presence of national differences makes it difficult to explain this phenomenon [17]. For example, foreign members in a strong, single-ethnic culture like that of Korea experience different conditions than those in multiethnic cultures like those of the United States and Europe, and they exhibit different behaviors, which generate different consequences. Most previous studies on board diversity also have limitations of focusing only on a particular characteristic that categorizes minorities (i.e., foreignness), and not considering other attributes and conditions which could lead the minorities to behave differently [27,41]. However, board members are individuals who have their own attributes and are faced with contextual conditions, it is necessary to address their individual characteristics and board conditions to understand the board effects [15,42].

We examine two critical characteristics—nationality and director type—to reveal the impact of foreign directors. Given that individuals are social actors embedded in social systems [43], the national institutions where they are raised affect their values, beliefs, and cognition. Foreign directors, therefore, have different values and cognitive frames that depend on their nationality, which in turn leads to different behaviors in boardrooms. Director types also act as important contexts. Foreign directors face different conditions depending on their director type, which leads them to behave differently. These two characteristics are meaningful to explain the director effect, in that the former represents their internal dimension and the latter reflects the external dimension of foreign directors.

3. Hypotheses

3.1. The Nationality of Foreign Directors and Corporate Social Responsibility

National institutions historically grown are typically categorized into two types: Anglo-American and non-Anglo-American. Anglo-American countries are mostly those that had been British colonies and follow common law which is English in origin (e.g., the United Kingdom, the United States, Canada), whereas non-Anglo-American countries follow civil law derived from Roman law (e.g., France, Germany) [28]. These two country groups, represented in particular by the United States and Europe, have been documented to have different organizational structures and strategic choices in numerous studies (e.g., [5,44]), and thus, we operationalized countries into one of these two types.

As Anglo-American countries based upon common law institutions generally have strong investor protection mechanisms, shareholders can exercise greater power through their voting rights (Porta et al., 1998). Shareholders present their opinions and demand more to increase their interests. For example, shareholders may assert companies to reduce consideration of other stakeholders in order to ensure that corporate focus on themselves is not diminished. Moreover, since those countries also have liberal market economies which permit individualism and discretion of economic actors, companies are usually regarded as isolated, rather than associated, actors [5]. Therefore, companies in Anglo-American countries tend to concentrate on their own shareholders, adhering to the shareholder-centered model [45].

On the other hand, non-Anglo-American countries generally have weak shareholder rights, and thus shareholders in those countries have relatively little power [28], which lowers corporate interests in shareholders. Furthermore, as those countries have coordinated market economies where collectivism and partnership governance are emphasized, companies are regarded as interlocking, associated actors and have a broad range of relationships with social stakeholders [5]. Accordingly, companies in non-Anglo-American countries take a variety of stakeholders into consideration, which is known as the stakeholder-centered model [45].

Considering these two national institutions as well as the definition of CSR (i.e., corporate attention to broad stakeholders and society) as mentioned above, Anglo-American companies that adopted the shareholder-centered model should be more skeptical of CSR for broad stakeholders, while non-Anglo-American companies that followed the stakeholder-centered model should be more supportive of CSR activities. In a similar vein, individuals from Anglo-American countries are more likely to oppose the argument that companies need to conduct socially responsible activities for a broad segment of society and stakeholders (and not just for their shareholders), whereas individuals from non-Anglo-American countries are more likely to agree with the argument.

Therefore, we argue that the effects of foreign directors on CSR decisions vary depending on the directors’ nationality. When the foreign directors on a board are from Anglo-American countries, they are skeptical of CSR and are less likely to approve CSR activities while on the board, which ultimately decreases the extent of the firm’s CSR participation. In contrast, foreign directors from non-Anglo-American countries are supportive of CSR, and thus their presence will increase the firm’s CSR participation. Based on this logic, we propose the following hypotheses (stated in alternative forms).

Hypothesis 1.

The presence of foreign directors of Anglo-American nationality on a board will decrease the firm’s CSR involvement.

Hypothesis 2.

The presence of foreign directors of non-Anglo-American nationality on a board will increase the firm’s CSR involvement.

3.2. The Director Types of Foreign Directors and Corporate Social Responsibility

Although foreign director representation greatly affects a firm’s CSR decisions, the influence of foreign directors also varies depending on the contextual conditions the directors face. In particular, different director types generate different situations, because foreign directors have distinct knowledge sets and are treated differently contingent upon their directorship. Given that foreign directors are usually minorities on a board, they are often regarded as an out-group and experience unfavorable situations in the boardroom [29], but director type affects the favorability of these situations, either positively or negatively. Directors are typically classified into two groups depending on their degree of independence: inside directors and outside directors [46,47]. Inside directors are those who are current employees (e.g., CEOs, executives), and outside independent directors are those who are not associated with the firm in any way, through either business, employment, or family [46,47].

When foreign directors are insiders, their effect on board CSR decisions is increased. Since inside directors usually serve as full-time executives, foreign inside directors are more likely to possess abundant information and knowledge about the firm and contribute effectively to board discussions [48,49]. Moreover, inside directors are less likely to be excluded from discussions as regarded as out-group members, because the effects of superficial differences (i.e., nationality) decrease over time as they demonstrate other attributes beyond nationality and construct cooperative relationships with their peers [50]. Therefore, foreign directors with inside directorships, who have greater capabilities and face more favorable conditions, are more likely to participate in board discussions effectively, which increases their impact on board CSR decisions.

In contrast, the effects of foreign directors on board CSR decisions decrease when they are outsiders. Although outside directors are qualified people with excellent backgrounds, their knowledge has been accumulated outside of the firm, and they often lack firm-specific information and knowledge [48,49]. Moreover, since outside directors are usually part-time employees, foreign outside directors are more likely to be regarded as an out-group based on salient surficial attributes (i.e., nationality), and then to have fewer opportunities to present their opinions during board discussions. In addition, their independent status and often concurrent positions in other companies further limit their motivation to execute their board roles [48,51]. Accordingly, foreign outside directors, who lack firm-specific knowledge and are placed in unfavorable conditions, are less likely to contribute effectively to board discussions, and consequently decreases their impact on board decisions.

Considering the Korean context makes our predictions more plausible. Most Korean companies belong to business groups called Chaebols, where majority shareholders’ control exceeds their voting rights and companies have long maintained owner-controlled governance structures [26]. Within these governance systems, owners with greater control rights exert a significant influence on the selection of executives, including CEOs, and have close relationships with them [25,26]. Thus, inside directors selected from those executives are likely to have relatively more power, based on their relationship with the owners. On the other hand, outside directors are often only appointed to demonstrate that the company has a well-structured board, so their actual impact on board roles is marginal [26,52]. Indeed, numerous cases and statistics report that outside directors are often rubber stamps. For example, a report from the Korean Development Institute documented that board agendas objected to by at least one outside director amounted to only 0.4% (33 out of 9101 cases) of all agendas, after analyzing 100 Korean firms in the 2010–2012 period. Similarly, CEO Score magazine analyzed 180 companies in 30 Korean business groups and reported that directors on corporate boards in 18 business groups (e.g., SK, LG) agreed 100% on all agendas.

Therefore, we argue that the effect of foreign directors on CSR differ depending on their director type. Their impact on decisions is greater when they are insider directors than when they are outsider directors. Based on this logic, we propose the following hypotheses (stated in alternative forms).

Hypothesis 3.

The negative effect of foreign directors with Anglo-American nationality on a firm’s CSR involvement is strengthened when the foreign directors are insiders rather than outsiders.

Hypothesis 4.

The positive effect of foreign directors with non-Anglo-American nationality on a firm’s CSR involvement is strengthened when the foreign directors are insiders rather than outsiders.

4. Methods

4.1. Data and Sample

We used a panel dataset of South Korean companies from 2002 to 2015. We concentrated on firms listed on the Korea Stock Exchange (KSE) and Korean Securities Dealers Automated Quotations (KOSDAQ), because those firms are pressured to improve board structures and appoint foreign directors. We restricted our sample to manufacturing firms by excluding financial firms (e.g., banks, insurance companies) which followed different accounting practices and regulations. Consequently, our final sample consists of 12,906 firm-year observations on 1828 Korean companies.

We used several databases to collect the data. We utilized the TS2000 to obtain board and director information. TS2000 of the Korea Listed Companies Association provides a wide range of information, including board membership and ownership, of the firms listed on KSE or KOSDAQ [53]. We used KIS-Value and FnGuide in a complementary way for financial and general information of the firms. KIS-Value, operated by National Information and Credit Evaluation (NICE) that is affiliated with Standard & Poor’s (S&P), provides financial statements and company files [54]. This database is regarded equivalent to COMPUSTAT in the U. S. [17], and widely used for financial information. We also used annual reports from FnGuide to cross-check the data gathered to improve the dataset’s quality.

4.2. Measures

The dependent variable was CSR involvement. CSR is often measured using performance, such as CSR rating [55], but this measure has limitations because it captures the results, rather than the intentions and behaviors, of firms in the CSR domain. Addressing these limitations, several scholars have focused on particular corporate choices, such as environmental protection [56], voluntary disclosure of environmental information [57], and engagement in charitable giving activities [58]. Consistent with these studies, we focused on charitable giving engagements, a widely utilized CSR activity in Korea, and measured CSR involvement as the natural log of total donations.

The primary explanatory variables are presence of Anglo-American directors and presence of non-Anglo-American directors. Following prior studies that measured the representation of particular directors on a board as a proportion or number [59,60], we used the same two measures. More specifically, we calculated the ratio of (non-)Anglo-American directors to total directors, as well as the number of (non-)Anglo-American directors on the board.

For the director type variables, we simply used the number of directors belonging to each particular director type. We identified four groups of foreign directors according to their nationality and directorship, and measured the number of directors in each group: Anglo-American inside directors, Anglo-American outside directors, non-Anglo-American inside directors, and non-Anglo-American outside directors.

As this study sought to examine the relationship between foreign directors and CSR involvement, we included several control variables that could affect a firm’s CSR behavior and decisions. Specifically, we controlled for firm size, organizational slack, profitability, and debt ratio, because financial conditions and resource availability could affect firms’ CSR investments [12,39,61]. Firm size was measured by a firm’s total assets, logarithmically transformed to lessen the positive skewness. Organizational slack was calculated as log-transformed operating income to total sales. As a measure of profitability, ROA is net income divided by total assets. Debt ratio is calculated as the ratio of a firm’s total debt to total assets. We also control for firm age, marketing intensity, and media exposure, because a greater degree of firm exposure to, and pressure from, the public makes the firm invest in more social CSR activities [62]. Firm age is calculated as the years of operation after its founding. Marketing intensity is measured by the ratio of marketing costs to total sales, and media exposure is defined as the number of times the firm was reported on in the daily papers. Lastly, year and industry dummy variables are included to exclude unobserved time and industry effects.

4.3. Analysis

To verify our hypotheses, we mainly operationalize fixed effects (FE) panel regressions. The results of a Hausman test demonstrate a significant difference between the random effects and fixed effects models, and the fixed effects estimations are considered more suitable. The fixed effects estimations also control for time-invariant unobserved firm characteristics and serial correlations, which enables us to test our hypotheses properly.

5. Results

5.1. Descriptive Statistics and Correlations

The means, standard deviations, and pairwise correlations for all variables used in this study, except year and industry dummies, are shown in

Table 1. Means, Standard Deviations, and Correlations.

	Mean	S.D.	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
(1) ln(donation)	7.74	4.94	1.00
(2) Anglo director rate	0.00	0.03	0.01	1.00
(3) non-Anglo director rate	0.01	0.06	0.01 *	0.04 *	1.00
(4) Anglo director number	0.03	0.26	0.02 *	0.92 *	0.05 *	1.00
(5) non-Anglo director number	0.10	0.55	0.03 *	0.05 *	0.94 *	0.07 *	1.00
(6) Anglo dir. number: insider	0.03	0.27	0.00	0.77 *	0.05 *	0.75 *	0.05 *	1.00
(7) Anglo dir. number: outside	0.01	0.13	0.06 *	0.46 *	0.05 *	0.62 *	0.08 *	0.18 *	1.00
(8) non-Anglo dir. number: insider	0.08	0.46	0.02 *	0.03 *	0.91 *	0.04 *	0.95 *	0.05 *	0.04 *	1.00
(9) non-Anglo dir. number: outsider	0.02	0.19	0.04 *	0.06 *	0.56 *	0.10 *	0.65 *	0.03 *	0.15 *	0.38 *	1.00
(10) firm size	18.34	1.49	0.53 *	0.04 *	0.08 *	0.07 *	0.10 *	0.03 *	0.13 *	0.08 *	0.12 *
(11) ROA	−3.67	33.45	0.17 *	−0.01	0.02 *	−0.01	0.02 *	−0.00	0.00	0.02 *	0.01
(12) organizational slack	9.55	7.53	0.30 *	−0.02 *	0.02 *	−0.00	0.02 *	−0.01	0.03 *	0.01	0.04 *
(13) debt ratio	1.26	12.26	−0.01	−0.00	−0.01	−0.00	−0.01	−0.00	−0.00	−0.01	−0.00
(14) firm age	25.07	15.50	0.24 *	0.00	0.03 *	0.01	0.05 *	0.01	0.03 *	0.04 *	0.05 *
(15) marketing intensity	0.01	0.02	0.10 *	0.01	−0.01	0.00	−0.02 *	−0.01	0.02 *	−0.03 *	0.01
(16) media exposure	75.72	64.44	0.05 *	−0.01	−0.01	−0.01	−0.02	−0.02 *	−0.00	−0.01	−0.01
			(10)	(11)	(12)	(13)	(14)	(15)	(16)
(10) firm size	-	-	1.00
(11) ROA	-	-	0.16 *	1.00
(12) organizational slack	-	-	0.26 *	0.32 *	1.00
(13) debt ratio	-	-	0.00	−0.02	−0.03	1.00
(14) firm age	-	-	0.44 *	0.02 *	−0.00	0.00	1.00
(15) marketing intensity	-	-	0.01	−0.04	0.00	−0.01	−0.01	1.00
(16) media exposure	-	-	0.22 *	−0.02	0.03 *	−0.00	0.16 *	−0.02	1.00

Note: Pair-wise correlation; * p < 0.05.

. The variances in the extent of donation among firms were large as expected (mean 7.74, standard deviation 4.94). The firms in our sample had a small number of foreign directors (about 1%), and this figure was comparable to the reported numbers (1.5%), albeit slightly lower. Foreign directors were more likely to take inside directors than outside directors (Anglo directors 0.03 vs. 0.01; non-Anglo director 0.08 vs. 0.02).

There were some significant correlations between variables, but none that would cause serious multicollinearity issues. For instance, the variables associated with foreign directors have high correlations with each other (e.g., Anglo director number & Anglo director ratio, non-Anglo inside director number & non-Anglo director number), but these variables are used in different estimations for different objectives. The correlations among other variables did not generate problems either, because their absolute values are less than 0.7 and the variation inflation factor (VIF) values in each estimation is below ten, which is widely used as a cutoff threshold.

5.2. Regression Results

Table 2. Fixed Effects (FE) Regressions: ln(donation).

Variables	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Anglo director rate		−2.57 **
		(1.12)
non-Anglo director rate		1.86 **
		(0.90)
Anglo director number			−0.34 **
			(0.15)
non-Anglo director number			0.18
			(0.11)
Anglo dir. number: insider				−0.46 ***		−0.47 ***
				(0.15)		(0.15)
Anglo dir. number: outside				0.40		0.39
				(0.31)		(0.31)
non-Anglo dir. number: insider					0.23 *	0.23 *
					(0.13)	(0.13)
non-Anglo dir. number: outsider					−0.06	−0.05
					(0.28)	(0.28)
firm size	1.66 ***	1.66 ***	1.66 ***	1.66 ***	1.65 ***	1.66 ***
	(0.07)	(0.07)	(0.07)	(0.07)	(0.07)	(0.07)
ROA	5.92 × 10−4	5.45 × 10−4	5.72 × 10−4	5.97 × 10−4	5.97 × 10−4	6.01 × 10−4
	(1.08 × 10−3)	(1.08 × 10−3)	(1.08 × 10−3)	(1.08 × 10−3)	(1.08 × 10−3)	(1.08 × 10−3)
organizational slack	0.04 ***	0.04 ***	0.04 ***	0.04 ***	0.04 ***	0.04 ***
	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)
debt ratio	−9.15 × 10−4	−9.33 × 10−4	−9.27 × 10−4	−9.23 × 10−4	−9.09 × 10−4	−9.18 × 10−4
	(2.37 × 10−3)	(2.37 × 10−3)	(2.37 × 10−3)	(2.37 × 10−3)	(2.37 × 10−3)	(2.37 × 10−3)
firm age	−0.30 ***	−0.30 ***	−0.30 ***	−0.30 ***	−0.30 ***	−0.30 ***
	(0.06)	(0.06)	(0.06)	(0.06)	(0.06)	(0.06)
marketing intensity	7.34 ***	7.39 ***	7.36 ***	7.20 ***	7.42 ***	7.28 ***
	(2.40)	(2.40)	(2.40)	(2.40)	(2.40)	(2.40)
media exposure	0.01 **	0.01 **	0.01 **	0.01 **	0.01 **	0.01 **
	(2.36 × 10−3)	(2.36 × 10−3))	(2.36 × 10−3)	(2.36 × 10−3)	(2.36 × 10−3)	(2.36 × 10−3)
Constant	−16.14 ***	−16.19 ***	−16.19 ***	−16.21 ***	−16.12 ***	−16.18 ***
	(1.73)	(1.73)	(1.73)	(1.73)	(1.73)	(1.73)
Number of observations	12,906	12,906	12,906	12,906	12,906	12,906
Number of firms	1828	1828	1828	1828	1828	1828
F-value	62.09 ***	52.82 ***	52.70 ***	52.92 ***	52.38 ***	6.88 ***
R-square	0.066	0.067	0.067	0.067	0.066	0.067

Notes: (1) Numbers in parentheses are standard errors; (2) Year and industry dummies are included but not documented; * p < 0.1; ** p < 0.05; *** p < 0.01.

presents the regression results. Model 1, containing the control variables only, is a base model. Models 2 and 3 were utilized to test Hypotheses 1 and 2 by adding the variables related to the foreign directors’ nationalities. Model 4, with variables associated with director types of Anglo directors, was estimated to test Hypothesis 3; Model 5, with non-Anglo directors’ director type variables, was estimated to test Hypothesis 4. Model 6, which contained the directorship variables for both Anglo and non-Anglo directors, was estimated to test Hypotheses 3 and 4 again.

Model 1 documents that firms were likely to invest more in CSR when they were larger (β = 1.66, p < 0.01) and had more organizational slack (β = 0.04, p < 0.01), because companies with greater resources can afford to do more socially responsible activities. Firms were also more likely to conduct CSR activities when their marketing intensity was high (β = 7.44, p < 0.01) and when they received public media attention (β = 0.01, p < 0.05), owing to greater perceived pressure to do the right thing. These effects were consistent with our expectations. However, firm age had a negative impact on CSR involvement (β = −0.30, p < 0.01), inconsistent with our expectations. One explanation may be that younger firms participated in more CSR activities to publicize themselves positively.

The results of Models 2 and 3 provided support for Hypotheses 1 and 2, which predicted that the effects of foreign directors would vary depending on their nationality. In Hypothesis 1, we argued that firms with foreign directors from Anglo-American countries on their boards would be less likely to invest in CSR activities. Hypothesis 2 stated that firms with foreign board members from non-Anglo-American countries would be more likely to invest in CSR. As shown in Model 2, the proportion of Anglo-American directors had a significant and negative impact (β = −2.57, p < 0.05) on CSR involvement, while the proportion of non-Anglo-American directors had a significant and positive impact (β = 1.86, p < 0.05). Similarly, using numbers instead of proportions in Model 3, the number of Anglo-American directors had a negative and significant effect on CSR involvement (β = −0.34, p < 0.05). However, while the effect of the number of non-Anglo-American directors was positive, as expected, it was not significant (β = 0.18, p = 0.11). The results of the Wald test on each model demonstrated that these different impacts were statistically significant (p = 0.00), thus Hypotheses 1 and 2 were supported.

In Hypotheses 3 and 4, we argued that the impact of foreigners would be greater when the directors were insiders rather than outsiders. In other words, the negative impact of Anglo-American directors, and the positive impact of non-Anglo-American directors, on CSR would be greater when they were insiders rather than outsiders. The results of Models 4 through 6 provided evidence for these hypotheses. In Model 4, we found that the inside Anglo-American number had a negative and significant impact (β = −0.46, p < 0.01), whereas the outside Anglo-American number had no significant impact (β = 0.40, p = 0.19). The Wald test verified that these different effects are significant (p = 0.01), thus Hypothesis 3 was supported. In Model 5, the inside non-Anglo-American number had a significant and positive impact (β = 0.23, p < 0.1), while the outside non-Anglo-American number had no significant impact (β = −0.06, p = 0.83). However, the results of the Wald test did not demonstrate that these differences were statistically significant (p = 0.36), thus Hypothesis 4 was not supported. The results of Model 6 were similar to Models 4 and 5. We found different effects between inside Anglo-American directors (significant, with β = −0.47, p < 0.01) and outside ones (insignificant). Although we found that the inside non-Anglo-American number had a positive and significant impact (β = 0.23, p < 0.1) and that the outsider’s impact was not significant, these differences were not statistically significant.

5.3. Additional Analysis

We also conducted additional analyses to verify whether the suggested effects of foreign directors would be rigorously maintained. First, we analyzed incremental changes in CSR involvement, when including the extent of CSR involvement in the prior year.

Table 3. FE Regressions: Increment of ln(donation).

VARIABLES	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Anglo director rate		−2.25 **
		(1.09)
non-Anglo director rate		1.50 *
		(0.88)
Anglo director number			−0.31 **
			(0.15)
non-Anglo director number			0.13
			(0.11)
Anglo dir. number: insider				−0.40 ***		−0.40 ***
				(0.15)		(0.15)
Anglo dir. number: outside				0.29		0.28
				(0.30)		(0.30)
non-Anglo dir. number: insider					0.16	0.17
					(0.13)	(0.13)
non-Anglo dir. number: outsider					−0.03	−0.02
					(0.27)	(0.27)
ln(donation)_last year	0.20 ***	0.19 ***	0.19 ***	0.19 ***	0.19 ***	0.19 ***
	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)
firm size	1.39 ***	1.39 ***	1.39 ***	1.40 ***	1.39 ***	1.40 ***
	(0.07)	(0.07)	(0.07)	(0.07)	(0.07)	(0.07)
ROA	1.52 × 10−3	1.48 × 10−3	1.50 × 10−3	1.52 × 10−3	1.53 × 10−3	1.53 × 10−3
	(1.06 × 10−3)	(1.06 × 10−3)	(1.06 × 10−3)	(1.06 × 10−3)	(1.06 × 10−3)	(1.06 × 10−3)
organizational slack	0.04 ***	0.04 ***	0.04 ***	0.04 ***	0.04 ***	0.04 ***
	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)
debt ratio	−2.00 × 10−3	−2.02 × 10−3	−2.01 × 10−3	−2.01 × 10−3	−2.00 × 10−3	−2.00 × 10−3
	(2.32 × 10−3)	(2.32 × 10−3)	(2.32 × 10−3)	(2.32 × 10−3)	(2.32 × 10−3)	(2.32 × 10−3)
firm age	−0.30 ***	−0.30 ***	−0.30 ***	−0.30 ***	−0.300 ***	−0.30 ***
	(0.06)	(0.06)	(0.06)	(0.06)	(0.06)	(0.06)
marketing intensity	6.55 ***	6.59 ***	6.56 ***	6.43 ***	6.61 ***	6.49 ***
	(2.36)	(2.36)	(2.36)	(2.36)	(2.36)	(2.36)
media exposure	0.01 ***	0.01 ***	0.01 ***	0.01 ***	0.01 ***	0.01 ***
	(2.31 × 10−3)	(2.31 × 10−3)	(2.31 × 10−3)	(2.31 × 10−3)	(2.31 × 10−3)	(2.31 × 10−3)
Constant	−12.88 ***	−12.93 ***	−12.94 ***	−12.95 ***	−12.87 ***	−12.94 ***
	(1.71)	(1.71)	(1.71)	(1.71)	(1.71)	(1.71)
Number of observations	12,904	12,904	12,904	12,904	12,904	12,904
Number of firms	1828	1828	1828	1828	1828	1828
F-value	93.51 ***	80.04 ***	79.96 ***	80.10 ***	79.67 ***	71.29 ***
R-square	0.103	0.104	0.104	0.104	0.103	0.104

Notes: (1) Numbers in parentheses are standard errors; (2) Year and industry dummies are included but not documented; * p < 0.1; ** p < 0.05; *** p < 0.01.

displays the results. Model specifications were similar to those documented in Table 2 above. The regression results were almost identical to those in Table 2. They supported most hypotheses except Hypothesis 4. More specifically, the results of Model 2 showed that the presence of Anglo-American directors, as measured by their proportion on the board, led to a decrease in CSR added (β = −2.25, p < 0.05), while the presence of non-Anglo-American directors led to an increase in CSR added (β = 1.50, p < 0.1). In Model 3, we found that the number of Anglo-Americans decreased the CSR added (β = −0.31, p < 0.05), although the non-Anglo-American number did not have a significant influence. Based on the Wald test (p = 0.00), these different effects of Anglo and non-Anglo directors were significant, which provided support for Hypotheses 1 and 2. Moreover, the results of Models 4 through 6 demonstrated that Anglo-Americans’ impact was greater when they were insiders than when they were outsiders. In Models 4 and 6, the number of inside Anglo-American directors had a negative and significant effect on CSR added (β = −0.40, p < 0.01), but the effect of the number of outsiders was insignificant. Those effects were shown to be significantly different in the Wald test (p = 0.05), thus Hypothesis 3 was supported.

Second, we also used generalized estimating equations (GEE) [63] to account for contemporaneous and time-series correlations, because the results of Woodbridge test reported autocorrelation concerns in our sample (p = 0.00).

Table 4. Generalized Estimating Equations (GEE) Regressions: ln(donation).

VARIABLES	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Anglo director rate		–2.65 **
		(1.04)
non-Anglo director rate		0.86
		(0.73)
Anglo director number			–0.36 ***
			(0.14)
non-Anglo director number			0.06
			(0.09)
Anglo dir. number: insider				–0.41 ***		–0.41 ***
				(0.14)		(0.14)
Anglo dir. number: outside				0.25		0.25
				(0.28)		(0.28)
non-Anglo dir. number: insider					0.15	0.15
					(0.11)	(0.11)
non-Anglo dir. number: outsider					−0.29	−0.29
					(0.24)	(0.24)
firm size	1.70 ***	1.70 ***	1.70 ***	1.70 ***	1.69 ***	1.70 ***
	(0.04)	(0.04)	(0.04)	(0.04)	(0.04)	(0.04)
ROA	2.69 × 10−3 ***	2.66 × 10−3 ***	2.68 × 10−3 ***	2.72 × 10−3 ***	2.70 × 10−3 ***	2.71 × 10−3 ***
	(1.02 × 10−3)	(1.03 × 10−3)	(1.03 × 10−3)	(1.03 × 10−3)	(1.03 × 10−3)	(1.03 × 10−3)
organizational slack	0.07 ***	0.07 ***	0.06 ***	0.06 ***	0.07 ***	0.06 ***
	(4.84 × 10−3)	(4.88 × 10−3)	(4.88 × 10−3)	(4.88 × 10−3)	(4.88 × 10−3)	(4.88 × 10−3)
debt ratio	−5.98 × 10−4	−8.45 × 10−4	−8.43 × 10−4	−8.41 × 10−4	−8.22 × 10−4	−8.29 × 10−4
	(2.31 × 10−3)	(2.33 × 10−3)	(2.33 × 10−3)	(2.33 × 10−3)	(2.33 × 10−3)	(2.33 × 10−3)
firm age	0.01	0.01	0.01	0.01	0.01	0.01
	(4.72 × 10−3)	(4.78 × 10−3)	(4.78 × 10−3)	(4.78 × 10−3)	(4.78 × 10−3)	(4.78 × 10−3)
marketing intensity	14.57 ***	14.63 ***	14.59 ***	14.49 ***	14.70 ***	14.59 ***
	(2.00)	(2.01)	(2.01)	(2.01)	(2.01)	(2.01)
media exposure	0.01 *	0.01 **	0.01 **	0.01 **	0.01 **	0.01 **
	(3.60 × 10−3)	(3.75 × 10−3)	(3.75 × 10−3)	(3.75 × 10−3)	(3.75 × 10−3)	(3.75 × 10−3)
Constant	−24.73 ***	−25.11 ***	−25.14 ***	−25.12 ***	−25.06 ***	−25.11 ***
	(2.97)	(2.99)	(2.99)	(2.99)	(2.99)	(2.99)
Number of observations	13,153	12,906	12,906	12,906	12,906	12,906
Number of firms	1866	1828	1828	1828	1828	1828
Wald-chi2	2773.77 ***	2708.57 ***	2709.17 ***	2710.12 ***	2703.53 ***	2714.10 ***

Notes: (1) Numbers in parentheses are standard errors; (2) Year and industry dummies are included but not documented; * p < 0.1; ** p < 0.05; *** p < 0.01.

reports the results. In Model 2 and 3, we found the negative impact of Anglo-Americans on CSR involvement (β = −2.65, p < 0.05; β = −2.36, p < 0.01), which supported Hypothesis 1. However, the effect of non-Anglo-Americans was insignificant, even though their coefficients were positive as expected. The results of Models 4–6 revealed that the Anglo-Americans’ impact on CSR was greater when they were inside directors than when they were outside directors. In Model 4 and 6, the number of inside Anglo-Americans had a negative and significant effect on CSR (β = −0.41, p < 0.01), while that of outside Anglo-Americans has insignificant impact. These different effects were shown to be significant in the Wald test (p = 0.05), so Hypothesis 3 was supported.

Last, we conducted instrumental variable two-stage least square (IV 2SLS) estimations to address endogeneity concerns, since the decision to recruit foreign talents on a board would be affected by omitted unobservable firm characteristics and thus our explanatory variables would be correlated with the error terms. In more detail, we used two instrumental variables: (1) foreign ownership (measured by the proportion of common stocks held by foreign shareholders) and (2) foreign market development (measured by the ratio of foreign market development costs to sales), following prior relevant literature that suggests foreign ownership and international market operation as critical determinants of foreign director appointments [27,64]. We estimated the endogenous variables related to foreign directors using these two instruments in the first stage, and then we included the predicted values to estimate the CSR involvement in the second stage.

Table 5. Instrumental Variable (IV) Two-Stage Regressions: the Effects of Director Nationalities

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
VARIABLES	1st Stage (DV: Anglo Rate)	1st Stage (DV: Non-Anglo Rate)	2nd Stage (DV: ln(donation))	1st Stage (DV: Anglo num.)	1st Stage (DV: Non-Anglo num.)	2nd Stage (DV: ln(donation))
Anglo director rate			−68.92 **
			(29.47)
non-Anglo director rate			36.45 ***
			(12.27)
Anglo director number						−8.17 **
						(3.35)
non-Anglo director number						4.12 ***
						(1.32)
foreign ownership	7.86 × 10−4 ***	1.94 × 10−3 ***		0.01 ***	0.02 **
	(3.99 × 10−5)	(7.97 × 10−5)		(3.32 × 10−4)	(6.9 × 10−4)
foreign market development	0.47 ***	−0.65 **		4.71 ***	−4.02 *
	(0.13)	(0.26)		(1.00)	(2.09)
firm size	−2.51 × 10−3 ***	−6.46 × 10−3 ***	1.65 ***	−0.01 ***	−0.03 ***	1.62 ***
	(4.2 × 10−4)	(8.38 × 10−4)	(0.06)	(3.63 × 10−3)	(0.01)	(0.07)
ROA	−3.57 × 10−5	−5.2 × 10−5	0.04 ***	−1.09 × 10−4	−6.0 × 10−4	0.03 ***
	(3.39 × 10−5)	(6.76 × 10−5)	(0.01)	(2.45 × 10−4)	(5.11 × 10−4)	(0.01)
organizational slack	1.87 × 10−12	6.56 × 10−12 *	1.60 × 10−10	3.30 × 10−11 **	9.27 × 10−11 ***	2.33 × 10−10
	(1.80 × 10−12)	(3.60 × 10−12)	(3.07 × 10−10)	1.41 × 10−11	(2.94 × 10−11)	(2.99 × 10−10)
debt ratio	2.98 × 10−5	−3.9 × 10−5	−0.01	1.81 × 10−4	2.08 × 10−4	−0.01
	(1.16 × 10−4)	(2.32 × 10−4)	(0.02)	(8.29 × 10−4)	(1.73 × 10−3)	(0.02)
firm age	4.08 × 10−5	−3.1 × 10−5	2.38 × 10−3	4.55 × 10−4	−6.4 × 10−5	2.43 × 10−3
	(3.24 × 10−5)	(6.46 × 10−5)	(0.01)	(2.92 × 10−4)	(6.07 × 10−4)	(0.01)
marketing intensity	−1.22 × 10−12 *	−3.82 × 10−12 ***	−1.10 × 10−10	−1.19 × 10−11 **	−5.56 × 10−11 ***	−6.43 × 10−11
	(6.93 × 10−13)	(1.38 × 10−12)	(1.18 × 10−10)	(5.37 × 10−12)	(1.12 × 10−11)	(1.19 × 10−10)
media exposure	1.61 × 10−5	1.12 × 10−4	0.03 ***	1.48 × 10−4	9.56 × 10−4)	0.03 ***
	(6.38 × 10−5)	(1.27 × 10−4)	(0.01)	(4.46 × 10−4)	(9.28 × 10−4)	(0.01)
Constant	0.01	0.03	−27.83 ***	0.01	−0.18	−26.76 ***
	(0.02)	(0.04)	(3.70)	0.18	(0.37)	(3.69)
Number of observations	4422	4422	4422	4422	4422	4422
Number of firms			1028			1028
Wald-chi2	488 ***	703 ***	1097.68 ***	415 ***	641 ***	911.34 ***
R-square			0.1235			0.1230

Notes: (1) Numbers in parentheses are standard errors; (2) Year and industry dummies are included but not documented; * p < 0.1; ** p < 0.05; *** p <0.01.

and

Table 6. IV Two-Stage Regressions: the Effects of Director Types.

	Model 7	Model 8	Model 9	Model 10	Model 11	Model 12
VARIABLES	1st Stage (DV: Anglo num: Insider)	1st Stage (DV: Anglo num. Outsider)	2nd Stage (DV: ln(donation))	1st Stage (DV: Non-Ang num: Insider)	1st Stage (DV: Non-Ang num. Outsider)	2nd Stage (DV: ln(donation))
Anglo dir. num.: insider			−34.76
			(98.69)
Anglo dir. num.: outside			137.4
			(372.0)
non-Anglo dir. num.: insider						51.81
						(246.5)
non-Anglo dir. num.: outsider						−176.6
						(865.8)
foreign ownership	0.01 ***	1.98 × 10−3 ***		0.01 ***	2.53 × 10−3 ***
	(2.95 × 10−4)	(2.1 × 10−4)		(5.09 × 10−4)	(2.87 × 10−4)
foreign market development	11.55 ***	2.64 **		−1.62	−0.13
	(1.67)	(1.19)		(2.72)	(1.53)
firm size	−0.03 ***	0.01 ***	−1.22	−0.02 ***	0.14 ***	5.19
	(3.08 × 10−3)	(2.19 × 10−3)	(7.79)	(0.01)	(3.51 × 10−3)	(17.14)
ROA	−2.9 × 10−4	−3.1 × 10−4 *	0.07	−2.4 × 10−4	−2.4 × 10−4	−0.01
	(2.61 × 10−4)	(1.86 × 10−4)	(0.09)	(3.2 × 10−4)	(1.81 × 10−4)	(0.15)
organizational slack	−2.56 × 10−11 *	5.46 × 10−11 ***	−8.13 × 10−9	−7.97 × 10−12	8.74 × 10−11 ***	1.63 × 10−8
	(1.35 × 10−11)	(9.60 × 10−12)	(2.29 × 10−8)	(1.94 × 10−11)	(1.09 × 10−11)	(7.76 × 10−8)
debt ratio	5.02 × 10−4	4.06 × 10−4	−0.05	2.53 × 10−4	2.72 × 10−4	0.03
	(9.00 × 10−4)	(6.41 × 10−4)	(0.14)	(1.06 × 10−3)	(5.97 × 10−4)	(0.20)
firm age	6.87 × 10−4 ***	−4.0 × 10−4 **	0.08	1.32 × 10−4	−3.0 × 10−4	−0.06
	(2.35 × 10−4)	(1.67 × 10−4)	(0.22)	(5.55 × 10−4)	(3.13 × 10−4)	(0.30)
marketing intensity	−5.47 × 10−12	−1.36 × 10−11 ***	1.52 × 10−9	−5.88 × 10−12	−6.02 × 10−11 ***	−1.05 × 10−8
	(5.20 × 10−12)	(3.70 × 10−12)	(4.57 × 10−9)	(7.27 × 10−12)	(4.10 × 10−12)	(5.07 × 10−8)
media exposure	−3.4 × 10−4	2.76 × 10−5	0.02	7.02 × 10−4	−2.1 × 10−4	−0.04
	(4.98 × 10−4)	(3.54 × 10−4)	(0.07)	(5.56 × 10−4)	(3.14 × 10−4)	(0.36)
Constant	0.40 **	−0.34 ***	33.27	−0.06	−0.31 *	−79.38
	(0.17)	(0.12)	(168.5)	(0.30)	(0.17)	(254.9)
Number of observations	4422	4422	4422	4422	4422	4422
Number of firms			1028			1028
Wald-chi2	760 ***	260 ***	60.29 ***	404 ***	164 ***	11.98
R-square			0.0071			0.0011

Notes: (1) Numbers in parentheses are standard errors; (2) Year and industry dummies are included but not documented; * p < 0.1; ** p < 0.05; *** p <0.01.

report the results. Table 5 contains the results of nationality effect of foreign directors predicted in Hypotheses 1 and 2, and Table 6 documents those of directorship effects in Hypotheses 3 and 4. In Models 1–3, Model 1 and 2 are the first stage estimations and Model 3 is the second stage estimation. Specifications of other models (Models 4–6, 7–9, 10–12) are similar to those of Models 1–3.

In Table 5, the results of Model 1 and 2 showed that the instruments used are valid, because they were significantly relevant to the endogenous variables (β = 7.86 × 10⁻⁴, p < 0.01; β = 0.47, p < 0.01 in Model 1; β = 1.94 × 10⁻³, p < 0.01; β = −0.65, p < 0.05 in Model 2, respectively), and had no overidentifying restrictions in Sargan test. In Model 3, we found Anglo-Americans have negative and significant impact on CSR (β = −68.92, p < 0.05), while non-Anglo-Americans have positive and significant impact on CSR (β = 36.45, p < 0.01). The results of Models 4–6 were almost identical to these. In Model 6, we found that Anglo directors’ impact is significantly negative (β = −8.17, p < 0.05) and non-Anglo directors’ impact is significantly positive (β = 4.12, p < 0.01). These results provided supports to Hypotheses 1 and 2.

In Table 6, we tested Hypotheses 3 and 4 that argue the different effects of foreign directors depending on their director types. However, the results did not support our hypotheses. In Models 7–9, we found that our instruments are valid to estimate the endogenous variables in the first stage estimations of Model 7 and 8, based on the relevance (β = 0.01, p < 0.01; β = 11.55, p < 0.01 in Model 7; β = 1.98 × 10⁻³, p < 0.01; β = 2.64, p < 0.05 in Model 8, respectively) and overidentifying tests. However, we could not find significant effects of director types of foreign directors, as shown in the second stage regression of Model 9, even though the coefficient signs were consistent with our expectation. Therefore, Hypothesis 3 was not supported. The results of Models 10–12 provide little verification of Hypothesis 4, because the validity of instrumental variables (in Models 10 and 11) and regression model (in Model 12) was low.

6. Discussion and Conclusions

Addressing the fact that the relationship between board attributes and CSR is relatively under-investigated in non-Western contexts, this study examines the impact of foreign directors on CSR involvement in South Korean companies by considering the attributes of the directors in terms of nationality (Anglo-Americans vs. non-Anglo-Americans) and director types (inside vs. outside directors). Grounded upon the national institutions [5,28] and board dynamic literature [29], we argue that foreign board members lead to different consequences, depending on their nationality and directorship type. More specifically, foreign directors from Anglo-American countries (e.g., the United States, the United Kingdom), where a shareholder-centered perspective is widely operationalized, are more skeptical of CSR activities for a wide range of shareholders, and consequently decrease a firm’s CSR participation. On the other hand, foreign directors from non-Anglo-American countries (e.g., France, Germany), where a stakeholder-centered perspective is emphasized, are more supportive of CSR activities, which increases a firm’s CSR involvement. Moreover, the Anglo-American directors’ negative impact on CSR becomes greater when they are inside (rather than outside) directors, because the constraints on minority foreign members are diminished as they spend more time in the firm.

We believe our findings theoretically contribute to the CSR and corporate governance literature. First, this study is one of only a few studies that have investigated CSR of non-Western companies. Despite the widespread recognition that different national institutions can lead to different corporate behavior, previous studies on CSR have mainly focused on Western companies [17]. Only recently have a few scholars begun to pay attention to the CSR phenomenon in non-Western companies (e.g., [17,18,21,22,23]). Thus, our study investigating the board–CSR linkage in South Korean companies has certain implications for this stream of research.

Second, our study is meaningful in that it examines the effects of foreign directors on strategic choices. Although the appointment of foreign members on corporate boards is generally recognized as one of critical changes in Korean corporate boards during the past several decades, the subject of foreign directors has remained underinvestigated [25,27]. Moreover, prior relevant studies on corporate governance have limited implications to explain this phenomenon, because they have heavily focused on Western companies which are embedded in different institutions (e.g., multiethnic culture) from Korean ones (e.g., single-ethnic culture) [17]. A few studies recently have begun to explore the impact of foreign directors in the context of Korean firms (e.g., [27]), but the number is still small and the studies examining their impact on strategic choices (rather than on financial performance) are few. Considering that boards tend to indirectly, rather than directly, affect financial performance by making organizational decisions [13], it is necessary to examine the foreign directors’ effect on strategic choices. In this regard, we believe our study has theoretical implications and proposes many possibilities for future research.

Third, our study is important in that it finds different effects of foreign directors depending on their attributes. Most prior studies on corporate boards have usually approached the effects of minority members through a diversity perspective, based on the assumption that subgroups of minorities are homogeneous by emphasizing a particular characteristic that categorizes them as minorities (e.g., foreigners, women) [12,41]. However, minority subgroups are different in other attributes and conditions, which leads them to behave differently, and consequently generates different outcomes [41]. Several scholars have indeed proposed that it is necessary to consider directors’ individual characteristics and board conditions in more detail, to understand the inner workings of boards (e.g., [15,42]). However, there are few studies that consider individual attributes and conditions of foreign directors to examine their effects. In this regard, our study is meaningful because it addressed other characteristics beyond the mere presence of minorities and proposed theoretical explanations. We specifically employ the national institutional literature to predict the different effects between Anglo-American and non-Anglo-American directors. We also adopt the board dynamics literature to argue that inside and outside directors would influence board decisions differently.

Fourth, our finding that inside foreign directors have a greater impact on board CSR decisions than outside foreign directors in Korean companies is insightful. Since previous studies on corporate boards have usually focused on outside directors in the belief that those directors would be more effective on boards owing to their independent status [65], inside directors have been regarded as relatively unimportant. Our findings imply that inside directorships are more important on Korean boards, and more effective for particular types of directors (e.g., minority foreign members), which advances our understanding of Korean corporate boards and also suggests the possibility of research on inside directors.

This study has practical implications as well. South Korean companies have begun to appoint foreign directors to their boards, but they do not understand the consequences generated by these members [25,40]. Then, how the foreign members change a corporate board and which foreigners are better for the firm? This study provides insights relevant to these questions. Companies can select foreign candidates for organizational objectives better by considering their national backgrounds and entrusting them with roles that are more appropriate. For example, director candidates who are from non-Anglo-American countries and given inside roles are more likely to effectively contribute to firms seeking to enhance their CSR. Policy makers can also use these insights to encourage firms to participate in CSR activities by creating new regulations related to board structures and compositions. For instance, the regulations that provide guidelines for the appointment of foreign board members and require the firms to announce the detailed information of directors (including nationalities and director types) may help facilitate CSR.

Although this study makes both theoretical and practical contributions, it has several limitations that need to be considered in future research. First, we focus on charitable giving engagements and use the amount of donation to measure CSR involvement, but some other measures are sometimes more concrete and preferred. For example, CSR rating is widely used to capture the overall performance of CSR [12,55] and other corporate choices, such as those associated with environmental protection [56,57] and web-site reporting [22], can be used as well. Second, our sample only contains a small percentage of foreign board members (about 1%), since there have not yet been many appointments of foreign directors to Korean companies. If future research examines the board–CSR linkage after foreign board representation increases considerably, it may discover more consequences that foreign directors generate. Third, although we investigate nationality and directorship types as two notable director attributes that could cause changes in their boardroom behaviors and boards’ decisions, there may be other important factors that lead to those changes. For example, the backgrounds of foreign directors (e.g., education, social networks, residence) and those of the others (e.g., studying abroad) on a board may also be important. Fourth, although we have endeavored to address problems contained in our sample and analyses by using alternative methods (e.g., changing controls, using alternative estimations), endogeneity issues may not be perfectly dismissed and still remain.