The Effects of Board Diversity on Korean Companies’ ESG Performance

Abstract

This paper explores the effect of board diversity on environmental, social, and governance (ESG) performance in Korean-listed firms using regression analysis. Our findings reveal that an increased board size significantly correlates with higher ESG scores when combined with other board diversity dimensions. The presence of female directors on boards was found to have a significant effect on environmental and social components of ESG performance. Age diversity exhibits a negative association with ESG scores, emphasizing potential disruptions from intergenerational differences. Foreign directors show no significant impact on ESG performance, suggesting that country-specific contextual factors may limit foreign directors’ influence on boards. The proportion of highly educated directors positively affects the overall ESG performance, aligning with resource dependence and agency theories. Overseas-educated directors play a crucial “bridging” role in adapting sustainable innovations overseas, positively influencing ESG performance. In conclusion, this study provides empirical evidence of the complex relationships between board diversity dimensions and ESG performance in the Korean context. These findings guide stakeholders in shaping inclusive and effective board structures for optimal corporate sustainability.

1. Introduction

Despite its current popularity both in business and politics, sustainability has not become a global trend in a single day. The discussion on the social responsibility of corporations in business management research can be traced back to the early 1960s [1,2]. Environmental movements, scientists, economists, and social activists in the 1960s and 1970s planted the seed of sustainability as we know it today [3].

Van Holt and Whelan [4] point out the abundance of terms and definitions used to describe the responsibility of a firm towards the environment, society, and people. In the management literature and the corporate world, the name “sustainability” acts like an umbrella term to encompass all the other synonymous terms associated with this domain of corporate responsibility.

The earliest term “Corporate Social Responsibility” (CSR) came into use to define a firm’s commitment to philanthropy and maintaining good relations with the community and its employees. CSR is still in use in academia and business but has a broader meaning, in which stakeholders are prioritized over shareholders, and corporate performance is evaluated based on environmental, social, and governance measures. On the other hand, the relatively newer terms like “sustainability” and “ESG” are related to how a firm integrates environmental, social, and governance issues into its business model. They are more precise terms, though their meanings can vary. Sustainability encompasses economic sustainability, whereas ESG concentrates exclusively on non-financial measures. In line with Van Holt and Whelan’s [4] findings in the literature, the terms “sustainability”, “CSR”, and “ESG” are used interchangeably throughout the paper.

Ranked as the 17th largest cumulative CO₂ emitter globally, South Korea initiated a “Green New Deal” to shift towards a green economy. The government aims to improve corporate disclosure rules, beginning with voluntary ‘sustainability reports’ by 2025. Public companies with assets exceeding KRW 2 trillion are obligated to disclose ESG activities from 2025, followed by a mandatory disclosure requirement for all KOSPI-listed firms starting in 2030 [5].

Our research concentrated on South Korea mainly for two reasons. Firstly, relative to Western countries, South Korea has recently adopted ESG practices in business operations, making it a “latecomer” in this aspect. Most Korean firms initiated the integration of sustainable practices into their business processes only recently [5]. Secondly, the effect of corporate governance characteristics on organizations greatly varies across countries owing to the diverging institutional and cultural factors [6]. The effects of CSR have been predominantly explored mostly in Western contexts, with comparatively fewer studies focusing on non-Western and developing markets [7,8]. Moreover, research examining the relationship between board characteristics and CSR performance in Asian countries remains limited [6]. In emerging economies, the priority is frequently on operational efficiency and profits rather than softer values like environmental conservation, fair wealth distribution, and community relations [9]. Considering these contextual differences, we expect to observe a different result in the Korean economy than those results reported in earlier studies.

On the other hand, the cultural context plays a crucial role in shaping the relationship between board diversity and CSR outcomes. Korean cultural norms differ significantly from those of Western societies, being deeply rooted in a strong collectivistic orientation and Confucian teachings [7]. The application of Western theoretical frameworks, such as agency theory and resource dependence theory, which predict a positive association between board diversity and CSR, may not be entirely suitable for the Korean context due to its collectivist nature. As Hofstede et al. [10] emphasize, harmony and consensus are fundamental goals in collectivist cultures, where diversity might be perceived as a potential disruption to these values. As a result, board diversity in Korean firms may not yield the same positive effects on CSR as commonly observed in Western countries.

Kang et al. [8], for example, highlight that board diversity attributes (particularly race) in Korean corporate boards are a relatively under-researched area. This is because Korean boards have historically been ethnically homogeneous, reflecting a strong single-ethnic collective culture. The behavior and the impact of foreign directors appointed to a board in a single-ethnic country differ from those in a multiethnic country, a nuance often overlooked in the predominantly Western-oriented board diversity literature.

Considering Korea’s unique institutional environment (i.e., state-led market economy, Chaebols—family-controlled corporate structures) and cultural context, Chang et al. [7] propose that the relationship between board characteristics and CSR in Korea differs significantly, especially compared to previous studies in Western contexts. In light of these findings, it would be informative to empirically test whether the impact of diverse boards is different in Korean firms compared to what was observed in Western companies.

In light of the unique characteristics of the Korean context, as mentioned in the literature, it becomes crucial to empirically investigate and validate whether the impact of CSR produces outcomes that differ from what has been observed in Western countries. The contribution of our research would be filling the current gap in the literature related to corporate ESG practices in the non-Western context and testing the reported uniqueness of the relationship between board characteristics and CSR in Korea.

Our study distinguishes itself from refs. [11,12], as they measure board diversity by constructing a diversity index from various board diversity attributes. While these studies aim to measure the collective impact of diverse board attributes on ESG performance, they overlook the individual effects of each board diversity attribute. In contrast, our study employs a multiple regression model to not only identify the combined impact of multiple board diversity attributes but also to discern their individual influences on the sustainability performance of firms.

In Section 2, we review the existing literature and relevant theories on board diversity and ESG to build the conceptual framework of the research. Section 3 introduces the proposed hypotheses. The research design and variables used in the study are explained in Section 4. Section 5 reports the empirical results and the main findings of the paper. The last section explores the practical implications of our findings for corporate boards, policymakers, and investors.

2. Literature Review

2.1. Social Responsibility of Businesses

In 1962, Milton Friedman started an academic debate about CSR in a free economy by suggesting that the only social responsibility of a business is “to use its resources and engage in activities designed to increase its profits” [13] (p. 133). This claim was based on the neoclassical economics theory according to which sustainability actions will put a firm in a less competitive position due to their cost [14].

The majority of the early papers from the 1960s to mid-1995s were written as a reactionary response to Friedman’s [13] shareholder approach, addressing questions such as why the firms would engage in sustainable practices and what actions a company should be responsible for [1,4,15]. Also in this period, “ecological economists”, a group of nonconforming economists, challenged the tenets of neoclassical economics in an attempt to reorient economics to be more ethical and eco-friendly [3].

The leading names in this group were systems theorists Donella Meadows, Dennis Meadows, and Jørgen Randers. They argue that the global economic system obsessed with constant growth is running in overshoot mode as the human ecological footprint (the human burden on the natural environment) is above the sustainable level that the physical limits of the planet can support [16]. Backed by the predictions made by computer simulation models based on various scenarios, the book foresees an economic collapse unless the human ecological footprint is deliberately reduced through “dematerialization (less use of energy and materials to obtain the same output); increased equity (redistribution from rich to the poor of the benefits from using energy and materials); and lifestyle changes (lowering demands or shifting consumption toward goods and services that have fever impacts on the physical environment)” [16] (p. 178).

The early 1970s also witnessed a growing role of the United Nations (UN) in coordinating international environmental efforts and guiding member states in pursuing eco-friendly practices, giving the movement an international character [3]. The first milestone was the establishment of the United Nations Environmental Program (UNEP) in 1972. UNEP created the Intergovernmental Panel on Climate Change (IPCC) in 1988 to examine and summarize the latest scientific research on climate change. Its comprehensive assessment reports inform policymakers around the world [17].

The UN-organized summits gave birth to numerous treaties and environmental policies. The first such summit was the Rio Earth Summit of 1992, which produced Agenda 21, a framework for sustainable development, and the legally binding “Framework Convention on Climate Change” (UNFCCC). The decision-making body of the UNFCCC is the so-called “Conferences of Parties” (COPs), which brings together member states to assess the measures taken by the parties. The COPs resulted in two significant agreements: the Kyoto Protocol of 1997 and the Paris Agreement of 2015. The Kyoto Protocol, adopted in COP 3, sets forth the legally binding emission reduction targets for the developed world. The Parties agreed to keep global warming below 2 °C above pre-industrial levels and continue efforts to limit it to 1.5 °C in COP 21 and adopted the Paris Agreement [18]. The latest UN achievement on the sustainability front was the adaptation of the “2030 Agenda for Sustainable Development” in 2015 which sets out 17 Sustainable Development Goals to be achieved by 2030. By the early 2000s, the concept of sustainability had become deeply embedded in academia, the United Nations, various governmental bodies, and businesses [3].

2.2. Stakeholders and Sustainable Development

In the management literature, the most prominent response to Friedman came from Edward Freeman. Freeman’s [19] stakeholder theory posed a major critique of Friedman’s [13] shareholder-oriented approach [20]. The stakeholder approach became the mainstream theory in the sustainability literature after the late 1990s. He considered the existing management theories unable to address the latest changes in the business world of 1980 such as the rising consumer movement, increasing environmental concerns, and changes in communication technology [1]. Freeman’s [19] key argument was that organizations should pay attention to the interests of “any group or individual who can affect or is affected by the achievement of the organization’s objectives” (p. 46). According to this view, a firm has a responsibility not only for shareholders but also for a broader range of stakeholders (such as employees, suppliers, customers, communities, minorities, and the environment) [8].

Through bibliometric and text analyses of 65,000 papers published between 1960 and 2015 on sustainability, Van Holt and Whelan [4] revealed four key milestone periods in the evolution of sustainability research. According to them, sustainability concepts and theories emerged in the mid-1990s, as governments increasingly emphasized corporate environmental and social responsibility issues in the business world.

The concept of “triple bottom line” by John Elkington is considered one of the most important theories developed in the mid-1990s. The main idea behind the “triple bottom line” is that a company should commit to generating profit while respecting the interests of people (society) and the planet (environment), and its success should be measured based on its performance in these three domains [3].

Empirical research in the management literature from 2005 onwards was built on the theoretical concepts developed in the mid-1990s, trying to prove the business benefits of adopting sustainability practices. Only by 2015, research shifted towards exploring how sustainability provides value and how to create the best practices [4].

What was considered a trivial joke until the late 1970s became a significant trend in the corporate world recently [8]. From the mid-1990s, companies began to care about their social impact and address issues like climate change, inclusion, diversity, employee welfare, etc. [14]. For the last two decades, a growing number of companies started to calculate, manage, and disclose their sustainability performance. In the early 2000s, CSR emerged as a powerful movement within the business world [3].

The sustainability hype in the corporate world is due to the pressure from various stakeholders demanding firms to be more responsible for their social impact and the awareness that corporate responsibility practices are valuable for organizational competitiveness [14,21]. According to Borghesi et al. [22], corporate managers initiate sustainability practices for altruism (driven by moral duty), the company’s financial interests (such as attracting highly skilled employees, cultivating loyal customers, avoiding legal and political risk), and professional and personal reputation (agency cost)

In parallel with this heightened attention to sustainability by companies and investors, researchers began to explore what factors might affect a firm’s CSR practices and performance [8]. One of the major themes in this quest revolved around the relationship between the characteristics of firm management (boards and executives) and the ESG performance of a firm [23] (p. 3). Since CSR is an organizational decision, and the board of directors is the top decision-making body of an organization, a large body of the literature examined the link between the board characteristics and the ESG score of a firm [8].

2.3. Board of Directors

The board of directors plays a crucial role in the sound operation of an organization, offering strategic direction and influencing the performance of the firm. According to the resource dependence theory, the key role of the board is to provide effective feedback and efficient advisory services [24]. According to the agency theory, the primary role of boards is to oversee managers. Directors are required to supervise the concerns of stakeholders during the management decision-making process [25]. The theory proposes that diverse boards can mitigate the effect of the agency cost by enabling efficient monitoring of management by boards [12]. In other words, boards with a heterogonous composition are better at monitoring management, compared to those with a homogenous composition.

The corporate board plays a significant role in a firm’s CSR orientation given its strategic leverage in an organization [7]. The board of directors ensures that the interests of management stay aligned with those of the shareholders [26]. Miller and Del Carmen Triana [27] mention the monitoring and controlling role of boards. Boards also play a crucial role by providing strategic advice and bringing information, skills, and legitimacy to the firm. Cucari et al. [28] mention that the main responsibility of the boards is to ensure that the company’s actions and the expectations of all stakeholders are in harmony, directing the company towards the implementation of sustainable development processes.

2.4. Theories on Board of Directors

The resource dependence theory sees the firms as part of interdependent networks and social relationships. Organizations rely on external sources for resources like finance, physical assets, and information obtained from the environment [29]. According to the theory, the main role of the directors is to provide resources by linking the company to the external environment, which creates uncertainty and external dependency [30]. The theory predicts that in a board with a diverse composition, each board member serves as an extra resource in terms of information, expertise, skills, experience, and perspective that will allow a firm to access key constituents and to deal with ESG issues better [7,12]. The stakeholder theory and resource dependency theory hint that a more diverse board is likely to represent a wide range of stakeholders, leading to better ESG performance [28].

On the other hand, the human capital theory posits that an individual’s accumulated education, experience, and skills constitute their stock of human capital, which can be utilized for the benefit of an organization [31]. When applied to corporate boards, boards with a diverse composition have access to highly skilled human capital. Because diverse directors bring varied educational backgrounds, skills, and experiences to the boardroom, they can enhance the overall human capital of the board.

Boards with diverse compositions, in terms of gender, ethnicity, age, and educational backgrounds, are better positioned to address a wide range of business challenges. This aligns with the resource dependence theory, as mentioned earlier, wherein directors’ skills and knowledge acquired through education are regarded as additional resources that contribute to the overall effectiveness of the board and, consequently, the organization. Grounded mainly on these two theories, our research aims to empirically test the predictions outlined by these theories to understand how board diversity affects the ESG performance of a firm.

2.5. Board Diversity

The way a board composition is structured is a main indicator of what a board can do [8,28]. As boards are the principal governing and decision-making representatives in firms, an extensive body of research investigated the relationship between board characteristics such as board diversity and board effectiveness [32].

Prior research proposes that governance characteristics determine organizational practices; as such, governance diversity affects the preferences and decisions of boards [33]. Moreover, diverse boards have more moderated decision-making processes and are less likely to face the problem of “groupthink” [11]. Post et al. [34] suggest that board directors’ life experiences and personal values influence the firms’ strategic orientation. Since directors’ personal beliefs stem from their demographic background, such as age, gender, cultural origins, and educational achievements, it is assumed that boards’ demographic diversity is an important indicator of firms’ performance.

Having a diverse board composition in terms of gender and nationality leads to more informed discussions and well-thought decisions, benefiting from rich human and social capital [27]. Moreover, diverse boards are more likely to have higher expertise, knowledge, and experience in the markets, which results in improved financial and ESG performance [24,26]. Beji et al. [25] provided evidence that the demographic diversity of boards improves the CSR performance of French-listed firms. They measured board diversity based on the number of independent and foreign directors in the board, along with the gender, age, and educational diversity of the board members.

Cormier et al. [33] draw our attention to the “two-sided nature” of board diversity, namely demographic diversity and cognitive diversity. Demographic diversity of boards is differentiation in observable attributes of directors such as culture, gender, age, education experience, and tenure. The latter is the difference in unobservable attributes among directors such as perception, beliefs, and attitudes. They have observed that both demographic and cognitive diversity mediate the relationship between social performance and social disclosure quality. However, cognitive diversity positively mediated the relationship between environmental performance and environmental disclosure quality. This is important research highlighting that diversity goes beyond gender, encompassing a diverse array of demographic and cognitive attributes among board members.

On the other hand, ref. [32] reported that firm performance is negatively affected by culturally diverse boards, due to difficulties originating from coordination and communication in a culturally heterogeneous group. The existing studies on board diversity effects lack a definitive conclusion, necessitating additional research to explore the impact of board diversity.

On the other hand, the cultural context plays a crucial role in shaping the relationship between board diversity and CSR outcomes. Korean cultural norms differ significantly from those of Western societies, being deeply rooted in a strong collectivistic orientation and Confucian teachings [7]. The application of Western theoretical frameworks, such as agency theory and resource dependence theory, which predict a positive association between board diversity and CSR, may not be entirely suitable for the Korean context due to its collectivist nature. As Hofstede et al. [10] emphasize, harmony and consensus are fundamental goals in collectivist cultures, where diversity might be perceived as a potential disruption to these values. As a result, board diversity in Korean firms may not yield the same positive effects on CSR as commonly observed in Western countries. Chang et al. [7] found that the impact of board diversity on CSR in the Asian context (China, Japan, and Korea) is different from the impact observed in the West. In these countries, due to strong collectivist norms, board diversity can also raise the likelihood of violating cultural norms of conformity in group settings.

Using a bibliometric analysis of the published papers on the topic of board diversity, Baker et al. [26] revealed three research clusters in the board diversity literature: the first group investigates the effect of board diversity on corporate governance; the second group researches driving factors behind board diversity, and the last cluster focuses on the impact of board diversity on firms’ CSR policy and firms’ strategy and innovation. This paper will contribute to the studies in the third group by investigating the influence of the different attributes of board diversity on corporate sustainability.

The authors also draw our attention to the lack of research on the other types of demographic diversity (namely age, nationality, race, education, and professional background) in the board diversity literature [26] (pp. 242–243). They urge scholars to examine other dimensions of demographic diversity. Reviewing the most debated questions in the corporate finance literature regarding ESG issues, Gillan et al. [23] also acknowledge little attention toward other dimensions of board diversity and suggest that future research should focus on broader demographic attributes to gain a deeper understanding of the interaction between corporate leadership and firms’ ESG performance. Moreover, a limited number of studies have explored the link between board diversity attributes and ESG performance in Korean context. And those studies focused only on a few aspects of board characteristics to measure board diversity. Kang et al. [8], for example, only examined the impact of the nationality of foreign directors on the firm’s CSR involvement. On the other hand, Chang et al. [7] examined three board characteristics, namely ratio of outside directors, CEO–outside director social ties, and educational diversity in terms of expertise, in a quest to find their impact on the CSR practices of Korean firms. Acknowledging their influence on ESG performance, Kang et al. [6] controlled critical board characteristics such as the size of the board, ratio of outside directors, ratio of female outside directors, and number of outside directors with relevant expertise, while examining how directors’ compensation affects the ESG performance of Korean firms. To best of our knowledge, no study has explored the effect of the age, race, education level, and international educational experience of directors on the ESG performance of Korean firms.

To fill these gaps in the literature, we have evaluated board diversity focusing on the various demographic aspects of board diversity. In line with ref. [33], our research investigated the impact of board diversity in terms of both demographic attributes (board size, age, gender) and cognitive attributes (nationality, education, experience) of directors.

3. Hypothesis Development

3.1. Board Size

Beji et al. [25] found evidence that larger boards are more likely to engage in sustainability practices and show better ESG performance. Having more members on the board brings diverse perspectives from various stakeholders, adding diverse insights on improving CSR activities. Each director contributes distinct attributes to the firm, thus bringing more diversity to the board and providing various resources such as expertise, skill, and information about the external environment [30] enabling a company to handle ESG practices effectively. In contrast, Walls et al. [35] suggest that firms with a larger board have a weaker environmental performance. Following the reasoning of the resource dependence theory and in the light of the previous literature, we develop our first hypothesis:

Hypothesis 1. 

The number of directors on the board is associated with the ESG performance of a firm.

3.2. Female Directors on Boards

Perhaps, gender diversity or female presentation is the most researched topic in the board diversity literature [2,12,27,36]. Research on the topic of women and their presence in board diversity research has dominated the board diversity literature [26,37]. Keyword analysis by Baker et al. [26] shows that gender diversity is the most recurrent theme in this literature.

Studying the CSR ranking of Chinese-listed firms, McGuinness et al. [21] confirmed that the gender of the board members has a significant impact on the CSR engagement of the firms. They also showed that female leadership (CEO, vice-CEO, chair, and vice-chair) increases a firm’s CSR performance. Qureshi et al. [2] concluded that female presence on the board both increases the market value and the ESG disclosure score of European-listed firms. Their finding confirmed previous studies that found that female directors are more effective monitors, and their presence on the board impacts a firm’s reputation, subsequently enhancing both firm value and ESG performance.

Moreover, earlier research showed that female directors are sensitive and show a greater focus on environmental and social practices compared to their male counterparts [38]. Likewise, Boulouta [36] argues that female directors play a significant role in addressing CSR issues and show a greater social awareness to align with gender-based expectations, which has immense pressure on women to adopt a more ‘caring’ approach. Investigating French-listed firms, Kahloul et al. [39] confirmed that woman directors can raise the board’s understanding of sustainability practices and bring a valuable perspective to addressing social responsibility concerns. By improving the decisions taken regarding CSR, gender diversity moderates the relationship between CSR and the financial performance of a firm.

Another way female directors can improve a firm’s commitment to sustainability and ESG performance is that they can bring a different point of view into the decision-making process of the board, and they are more willing to follow regulations and maintain long-term relationships with stakeholders [12].

However, previous research has not arrived at a consensus regarding how gender diversity on boards affects a firm’s sustainability initiatives. Cucari et al. [28] discovered a negative relationship between women on the board and the ESG disclosure of Italian-listed firms, whereas Carter et al. [31] reported no significant relationship between the gender diversity of the board and firm performance. The mixed results in the literature highlight the need for further investigation. Following the reasoning of resource dependence theory and in the light of the previous literature, we hypothesize the following:

Hypothesis 2. 

The presence of female directors on the board is associated with the ESG performance of Korean-listed firms.

3.3. Age Diversity

As Talavera et al. [40] showed, age serves as an indicator of an individual’s life experience and it can have a significant impact on the formation of personal values. Intergenerational differences can be visible in countries that have experienced rapid economic transformation over a short time. Growing up in different times, each generation can develop a unique set of world views. Directors can see themselves as part of a certain generation. Board directors’ life experiences and personal values shape firms’ strategic direction [34]. Having diverse age groups on the boards will bring various perspectives to the table given their unique life experience, affecting the decision-making process on boards [40].

There is no consensus in the literature over the direction of the relationship between the age diversity of the board and the ESG performance of a firm [41]. The age diversity of a board can enhance ESG performance by bringing distinct knowledge, experience, and networks. However, it can also cause low ESG performance due to conflicting views and a lack of group cohesion because group members with similar values and beliefs tend to favor those within the group and may discriminate against those outside the group [41]. Wu et al. [12] similarly documented such a negative relationship between board age diversity and a firm’s ESG performance, suggesting that age difference leads to a conflict of opinions, which in turn damages the team cohesion on the boards. In contrast, [28] could not detect any effect of directors’ age on the firm’s ESG disclosure score.

Previous research failed to establish a consensus on how age diversity affects a firm’s sustainability policies. Therefore, there is a need for further investigation on the topic. In line with the previous literature, we hypothesize the following:

Hypothesis 3. 

Board age diversity is associated with the ESG performance of Korean-listed firms.

3.4. Foreign Directors

Directors’ attitude toward CSR depends much on their country of origin and cultural background, given that the awareness and regulations on sustainability issues vary across countries [34]. Based on the resource dependence theory, we can expect that foreign directors can provide valuable resources to the board with their distinct knowledge, perspectives, and networks.

Investigating the carbon emission performance of French-listed firms, Mardini and Elleuch Lahyani [42] observed that boards with a greater number of foreign directors improve the carbon emission performance of firms by reducing carbon emissions. In other words, the authors confirm that the foreign directors provide valuable resources for the firm to cope with the decarburization and create a sustainable culture within the firm. The evidence suggests that directors from abroad are capable of shaping board decisions toward taking a more proactive stance on addressing climate change.

Iliev and Roth [43] discovered that directors with foreign board experiences transfer corporate governance practices across countries. Boards learn about governance and board characteristics from an international context through the experiences of their directors on international boards. In other words, the foreign experiences of directors play a significant role in shaping the governance and board structures of the firm. This implies that boards can also adopt sustainable practices from a foreign board member.

Miller and Del Carmen Triana [27] highlighted that the racial diversity of boards improves the firm performance (return on investment) through innovation and firm reputation. However, Frijns et al. [32] demonstrated that the national cultural diversity of a board harms the firm performance, whereas Carter et al. [31] found no empirical evidence to suggest that racially diverse boards lead to better financial performance.

Analyzing longitudinal data from the KOSPI200 index, Choi et al. [44] suggest that outside foreign directors (the ones recommended by foreign shareholders) in boards positively affect firm value when there is limited foreign ownership. Their work provides evidence that appointing foreign external directors indicates the firm’s readiness to sever managerial connections with controlling shareholders and enhance both board independence and expertise.

However, analyzing the board composition of the Korean-listed firms, Kang et al. [8] observed that the impact of foreign directors on the firm’s CSR investment differs greatly based on their nationality (Anglo-Americans vs. non-Anglo-Americans) and director types (insiders vs. outsiders). Following the reasoning of resource dependence theory and in line with the previous research, we expect the following:

Hypothesis 4. 

The presence of a foreign director in boards is associated with the ESG performance of Korean-listed firms.

3.5. Educational Diversity

Following the logic of the human capital theory, investment in higher education enhances an individual’s skills, knowledge, and productivity. The theory also predicts that board performance will be influenced by board diversity due to the presence of diverse and unique human capital [31]. Therefore, a board with a higher proportion of directors with an advanced educational degree is expected to bring a greater depth of expertise and awareness to sustainability issues, leading to improved overall ESG performance of a firm. The same thought line can be reached from the resource dependence theory, given that directors’ skills and knowledge, gained through education, are seen as an additional resource for both the board and the organization.

Likewise, Chang et al. [7] detected a U-shaped relationship between a director’s educational diversity and a firm’s CSR score. Their findings show that a board’s educational diversity negatively affects the CSR performance of a firm to a certain level, but the effect turns positive after that point. It also implies that Korean firms can take advantage of board educational diversity in boosting ESG performance when the board is either highly homogenous (little diverse) or highly heterogeneous (highly diverse).

However, in contrast to previous studies, Post’s et al. [34] findings did not support the results of earlier research, which assumed that directors with higher educational achievements are more concerned with environmental issues due to their deep understanding of the issue. The authors found no significant relationship between the proportion of highly educated directors on the board and the environmental performance of a firm. Consistent with the human capital theory and resource dependence theory, and in line with the previous literature, we predict the following:

Hypothesis 5. 

The proportion of highly educated directors (with a master’s degree or above) on the board is associated with the ESG performance of Korean-listed firms.

3.6. Foreign Education

Another attribute of directors that can contribute to the board’s diversity is overseas education. Post’s et al. [34] findings provided insight into the impact of directors educated overseas, particularly in Western Europe, on the environmental CSR performance of European firms. Their research indicates that boards with a greater number of directors who have studied abroad result in better corporate sustainability practices. This research shows that directors’ attitudes toward corporate sustainability might be influenced by their cultural background and educational experience. Indeed, each society has a different level of awareness and sensitivity towards sustainability and environmental issues, together with unique legal systems to regulate them. An international student educated in a society with higher sustainability norms will be more familiar with and informed about sustainability issues. Therefore, the perspective on the sustainability of a director educated abroad will diverge from that of a counterpart who received education in their home country.

Directors who have studied abroad can serve as a bridge that can transfer knowledge and skills related to sustainability from abroad, especially from developed countries where sustainability practices are common. Ullah et al. [45] confirmed this line of thought in the Chinese context, where female directors with foreign experience (study or work experience) boost the environmental and sustainable performance of listed firms. From a resource dependence perspective, the international education experience of a director provides a crucial resource for a firm in terms of knowledge of CSR activities.

Hypothesis 6. 

The proportion of overseas-educated directors on the board is associated with the ESG performance of Korean-listed firms.

4. Research Methodology

4.1. Sample and Data

This paper was based on the data of South Korean firms listed on the KOSPI (Korean Composite Stock Price Index) market in 2021. The final sample included observations on directors from 738 firms.

Table 1. Sample selection.

Category	Obs.
Total number of directors in board data provided by TS2000	5055
Total number of firms with board data provided by TS2000	749
Firms with ESG score data provided by KCGS	765
Firms with financial data provided by DataGuide	811
Firms with missing board, ESG, and financial data	11
Total	738

shows the sample selection criteria.

We have combined the data from three different databases. The data on the demographic characteristics of the directors such as their name, age, gender, ethnicity, and educational level are extracted from the TS2000 database. It is a business information service compiled by the Korea Listed Company Association, providing business reports and other financial information submitted by firms listed on KOSPI and KOSDAQ.

As a proxy for sustainability performance, the ESG scores of Korean firms in 2021 were obtained from the Korea Corporate Governance Service (KCGS). The KCGS is a Korean ESG rating company, and its ESG scores are widely used in sustainability-related research in Korea [8,9,46]. The company also evaluates the ESG scores of Korean firms and provides scores for constituents of the overall ESG performance.

Financial data to construct control variables was obtained from DataGuide, which is FnGuide’s data service. R programming language version 4.2.0 was used to run the statistical analysis and test the hypotheses.

Initially, the raw data from the TS2000 database included individual demographic information about 5055 directors from 749 KOSPI-listed firms in 2021. We created board diversity variables for a total of 749 KOSPI-listed firms by hand-collecting and assembling data about 5055 directors. KCGS provided integrated ESG scores as well as individual environmental, social, and governance scores of 765 firms. We obtained the financial data of 811 KOSPI-listed firms. Finally, we merged the data from the three sources. The merged data contained 749 matched observations. We deleted 11 firms that had missing data either about board directors, ESG, or financial data. Thus, our analysis was based on a final dataset of 738 observations.

4.2. Research Design

To test the hypotheses and to examine the relationship between ESG scores and board diversity, we employed Ordinary Least Squares (OLS) regression. We estimated the ESG score using the following model:

$$ESG={\beta }_0+{\beta }_{1}Br{d}_{size}+{\beta }_{2}Fe{m}_{brd}+{\beta }_{3}Ag{e}_{mean}+{\beta }_{4}Age\_div+{\beta }_{5}High\_edu+{\beta }_{6}For\_dir+{\beta }_{7}For\_edu+{\beta }_{8}ASST+{\beta }_{9}ROA+{\beta }_{10}LEV+{\beta }_{11}GRW+ϵ$$

(1)

In this model, variable ESG is a dependent variable, and it is a measure of a firm’s sustainability performance, as evaluated by the analysts of KCGS. We evaluated the firms’ board diversity based on the following independent variables: board size (Brd_size), gender diversity (Fem_brd), age diversity (Age_div), racial diversity (For_dir), educational diversity (High_edu), and education-level diversity (For_edu).

When measuring the board diversity, consistent with [26,34], we took the multiple dimensions of board diversity into account. We believe that diversity, in general, is a complex concept that has multiple dimensions. Thus, we attempted to measure the combined effect of multiple attributes of board diversity rather than just focusing on one dimension, as it will reflect the complex nature of in-board diversity. Definitions of the variables are provided in

Table 2. Variable definitions.

Variables	Description
Dependent
ESG	ESG scores from KCGS reflect the sustainability and ethical practices of listed firms in the KOSPI index.
Independent
Brd_size	Number of directors on the board.
Fem_brd	Presence of female directors on the board (0 = no, 1 = yes).
Age_div	Age diversity measured by Blau’s index.
For_dir	Presence of foreign directors on the board (0 = no, 1 = yes).
High_edu	Percentage of highly educated directors (with MA or/and PhD degrees, and above)
For_edu	Percentage of board members who have studied overseas.
Control
ASST	Natural logarithmic value of total assets of the firm.
ROA	Return on Assets (ROA): Net income/Total assets.
LEV	Leverage: Liabilities/Total assets.
GRW	Sales growth rate

.

4.3. Variable Measurement

We calculated the variables in the following manner. Board size (Brd_size) was simply the number of directors on the board. Gender diversity (Fem_brd) is a binary variable and following [40] it was measured by the presence of a female director on the board. It was coded 0 for boards with no female director and 1 for boards with at least one female director.

Following [25], we used Blau’s index of heterogeneity to measure the age diversity (Age_div) on the board using the formula $1-\mathsf{\Sigma }{P}_i^2$, where P is the proportion of directors in each category and i indicates the number of categories. There are six age categories: less than 40 years old, 40 to 49, 50 to 59, 60–69, 70–79, and 80 years and older. The formula sums up the squares of proportions of directors in each age group. The higher the number of age categories on the board, the greater the age diversity of the board. Blau’s index is a ratio of heterogeneity in a population and can take any number between 0 and 1. In our case, 0 indicates that all directors belong to the same age group (minimum board diversity), whereas 1 indicates that all directors are from different age groups (maximum board diversity).

Racial diversity (For_dir) is also a binary variable. In line with previous research [8], this variable was coded 0 for the boards with no foreign board member and 1 for the boards with at least one foreigner. Educational diversity (High_edu) was constructed by calculating the percentage ratio of directors with advanced educational degrees (master’s degrees, Ph.D. degrees, and above). Finally, educational-level diversity (For_edu) is a percentage ratio of directors who have studied abroad.

Following [8,46,47], firms’ total assets (ASST), Return On Assets (ROA), financial leverage ratio (LEV), and growth rate (GRW) were controlled (see Table 2) because financial performance and available resources can affect CSR performance [8]. The firm’s total assets are logarithmically transformed to lessen the positive skewness.

5. Results

5.1. Descriptive Statistics

Table 3. Descriptive statistics.

Variables	N	Mean	STD	Min	Q1	Q3	Max
ESG_score	738	3.29	1.3	0.25	2.25	4.45	6
Brd_size	738	6.76	1.95	4	5	8	17
Fem_brd	738	0.35	0.48	0	0	1	1
Age_div	738	0.53	0.17	0	0.46	0.64	0.84
For_dir	738	0.07	0.25	0	0	0	1
High_edu	738	45.3	25.3	0	25	67.7	100
For_edu	738	24.6	21	0	0	40	100
ASST	738	20.8	1.8	16	20	22	27
ROA	738	2.6	7.6	−50	0.63	5.86	48.9
LEV	738	0.49	0.22	0.03	0.32	0.64	1
GRW	738	16.5	25.6	−79.3	3.15	26.2	181

shows the descriptive statistics of the major variables. The mean ESG score of our sample is 3.29. This indicates that the average ESG score of the Korean-listed firms amounts to 3.29. Moreover, the average number of directors on the board is 6.8, indicating the typical size of the boards in the sample Korean firms. Furthermore, 35% of firms in the sample have at least one female director on their boards. The average age diversity, measured by Blau’s index, is 0.53, suggesting a moderate level of age diversity within the boards. On average, approximately 45% of board members in the sample hold MA or PhD degrees and above. The average percentage of board members who have studied overseas is 25%. Finally, on average, only 6.8% of firms in the sample have at least one foreign director on their boards.

Table 4. Correlation matrix.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
(1) ESG	1
(2) Brd_size	0.28 ***	1
(3) Fem_brd	0.35 ***	0.34 ***	1
(4) Age_div	−0.09 *	0.12 **	0.07	1
(5) For_dir	−0.00	0.12 ***	0.03	0.05	1
(6) High_edu	0.42 ***	0.09 *	0.27 ***	−0.06	0.02	1
(7) For_edu	0.42 ***	0.08 *	0.23 ***	−0.05	0.06	0.68 ***	1
(8) ASST	0.69 ***	0.33 ***	0.42 ***	−0.05	0.03	0.39 **	0.42 ***	1
(9) ROA	0.15 ***	0.02	0.01	0.04	0.03	0.08 *	0.09 *	0.19 ***	1
(10) LEV	0.22 ***	0.11 **	0.14 ***	−0.09 *	−0.05	0.09 *	0.09 *	0.37 ***	−0.31 ***	1
(11) GRW	0.05	−0.01	0.00	0.01	0.05	−0.00	0.02	0.07	0.26 ***	−0.01	1

Note: Pearson method with pairwise deletion. * p < 0.10 ** p < 0.05 *** p < 0.01.

displays the Pearson correlation coefficients for the variables under consideration. The outcome variable ESG showed a significant positive correlation with Brd_size (r = 0.28, p < 0.01), Fem_brd (r = 0.35, p < 0.01), High_edu (r = 0.42, p < 0.01), and For_edu (r = 0.42, p < 0.01). This indicates that boards with larger sizes and greater female representation tend to have higher ESG scores. Moreover, it also suggests that as the proportion of highly educated and foreign-educated members increases on the board, companies tend to have better ESG performance. In contrast, ESG has a negligible negative correlation with Age_div (r = −0.09, p < 0.10) and For_div (r = −0.00, p < 0.01). In addition, the ESG score exhibits a statistically significant correlation with the control variables ASST, ROA, and LEV at the 1% significance level, with respective correlation coefficients of 0.69, 0.15, and 0.22. This suggests that a company’s financial indicators affect its ESG performance. Correlations among predictive variables are low enough not to cause concern for multicollinearity, which is also supported by the variation inflation factor (VIF) and tolerance values. None of the VIF values are greater than 2, which is way below the cutoff point of 10. Moreover, all the tolerance values (1/VIF) are way above the threshold level of 0.2, indicating no potential problem of multicollinearity [48].

5.2. Board Diversity and Overall ESG Performance

The regression analysis was conducted using seven different models, adding different aspects of board characteristics as predictors. The full model includes all independent variables simultaneously. The control variables incorporated in each model were the natural logarithm of total assets (ASST), Return on Assets (ROA), leverage (LEV), and sales growth rate (GRW). The results of the regression are reported in

Table 5. The effect of board diversity on integrated ESG score of firms.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Full Model
Predictors
(Intercept)	7.09 *** (0.42)	−6.75 *** (0.45)	−6.83 *** (0.44)	−7.13 *** (0.42)	−6.46 *** (0.42)	−6.34 *** (0.44)	−5.70 *** (0.48)
Brd size	0.04 * (0.02)						0.05 ** (0.02)
Fem brd		0.18 ** (0.08)					0.10 (0.08)
Age div			−0.49 ** (0.21)				−0.50 ** (0.21)
For dir				−0.14 (0.14)			−0.19 (0.14)
High edu					0.01 *** (0.00)		0.01 *** (0.00)
For edu						0.01 *** (0.00)	0.01 ** (0.00)
ASST	0.49 *** (0.02)	0.49 *** (0.02)	0.51 *** (0.02)	0.51 *** (0.02)	0.45 *** (0.02)	0.46 *** (0.02)	0.41 *** (0.03)
ROA	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)	0.00 (0.01)
LEV	−0.24 (0.19)	−0.23 (0.19)	−0.28 (0.19)	−0.27 (0.19)	−0.18 (0.19)	−0.18 (0.19)	−0.18 (0.19)
GRW	−0.00 (0.00)	−0.00 (0.00)	−0.00 (0.00)	−0.00 (0.00)	0.00 (0.00)	−0.00 (0.00)	0.00 (0.00)
Observations	738	738	738	738	738	738	738
R-square	0.481	0.482	0.482	0.479	0.505	0.499	0.518
R-square adjusted	0.477	0.478	0.479	0.476	0.501	0.496	0.511
F statistics	135.61 ***	136.16 ***	136.36 ***	134.69 ***	149.06 ***	146.9 ***	78.02 ***

Note: Numbers in parentheses are standard errors. * p < 0.10 ** p < 0.05 *** p < 0.01.

.

Hypothesis 1 posited a relationship between the number of directors on the board (Brd_size) and ESG performance. The results of Model 1 in Table 5 show that board size alone does not affect a firm’s sustainability level (β = 0.04, p = 0.063). However, the effect of board size becomes significant (β = 0.05, p = 0.014) when the other aspects of board diversity are taken into account in Model 6. This suggests that an increase in the number of directors on the board is associated with higher ESG scores for Korean-listed firms, only when other board characteristics are controlled. As predicted by resource dependency theory, each board member provides valuable expertise, skills, and information to address the firm’s ESG performance.

Hypothesis 2 investigated the association between the presence of female directors and ESG performance. Model 2 in Table 5 shows a positive and statistically significant relationship (β = 0.18, p = 0.028). However, this significance diminishes in the full model (β = 0.10, p = 0.217), where the effects of other dimensions of board diversity are controlled. This hints at the complexity of the relationship between the presence of female directors and ESG scores. In other words, gender diversity of boards is an important determinant of firm ESG performance, but such an effect was not detected when the combined effect of other board diversity factors was taken into account.

Hypothesis 3 examined the association between age diversity on the board and ESG performance as shown in Table 5. The results reveal a negative and statistically significant relationship in both Model 3 (β = −0.49, p = 0.020) and the full model (β = −0.50, p = 0.015). This suggests that higher age diversity is negatively associated with ESG scores. In other words, the higher the proportion of different age groups on boards, the lower the ESG performance of the firm.

Hypothesis 4 explored the impact of the presence of foreign directors on ESG performance. The results do not show a statistically significant association either in Model 4 (β = −0.14, p = 0.303) or the full model (β = −0.19, p = 0.150), indicating that the presence of foreign directors may not be a significant factor in explaining the variance of ESG scores among Korean firms.

Hypothesis 5 examined the relationship between the proportion of highly educated directors and ESG performance. The results indicate a positive and statistically significant association in both Model 6 and the full model (β = 0.01, p < 0.001), suggesting that a higher percentage of highly educated directors is associated with better ESG performance. In other words, a 1-unit increase in the percentage of highly educated directors is estimated to result in a 0.01-unit increase in ESG performance. This effect might seem minute; however, given that the ESG scores range between 0.25 and 6, it can be considered as a moderate effect. In percent ratio, a one-unit increase in the percentage of highly educated directors is estimated to result in approximately a 0.17% increase in ESG score.

Hypothesis 6 investigated the impact of the proportion of overseas-educated directors on ESG performance. The results show a positive and statistically significant relationship in both Model 2 (β = 0.01, p < 0.001) and the full model (β = 0.01, p < 0.017), indicating that firms with a higher percentage of directors educated overseas exhibit better ESG performance.

Regarding the model fit statistics, the R-squared values range from 0.479 to 0.518, and the adjusted R-squared values range from 0.476 to 0.511. These values indicate that our models, especially the full model, explain a substantial portion of the variance in ESG scores. The F-statistics are consistently significant across all models, further supporting the overall model’s goodness of fit.

5.3. Board Diversity and Individual Components of ESG

We also investigated the effect of board diversity on individual constituents of the overall ESG performance: environmental, social, and governance performance. To check the impact of board diversity on firms’ environmental, social, and governance performance, we have regressed the same independent variables on each constituent of ESG separately. As shown in

Table 6. The effect of board diversity on individual components of ESG.

	Environmental		Social		Governance
Predictors	Estimates	p-value	Estimates	p-value	Estimates	p-value
(Intercept)	140.47 *** (9.97)	<0.001	−162.10 *** (9.06)	<0.001	−51.29 *** (5.06)	<0.001
Brd size	0.78 * (0.40)	0.053	0.93 ** (0.36)	0.011	0.76 *** (0.19)	<0.001
Fem brd	4.26 ** (1.70)	0.012	5.78 *** (1.54)	<0.001	1.12 (0.80)	0.164
Age div	−8.73 ** (4.30)	0.043	−15.61 *** (3.91)	<0.001	−5.74 *** (2.05)	0.005
For dir	−2.85 (2.83)	0.313	−6.29 ** (2.57)	0.015	−0.38 (1.33)	0.774
High edu	−0.01 (0.04)	0.769	0.09 ** (0.04)	0.014	0.05 *** (0.02)	0.005
For edu	0.13 *** (0.05)	0.005	0.10 ** (0.04)	0.016	0.03 (0.02)	0.248
ASST	8.12 *** (0.54)	<0.001	9.26 *** (0.49)	<0.001	4.03 *** (0.27)	<0.001
ROA	0.12 (0.11)	0.256	−0.09 (0.10)	0.372	0.04 (0.05)	0.392
LEV	−3.19 (3.88)	0.410	−3.71 (3.53)	0.293	−0.97 (1.86)	0.603
GRW	−0.00 (0.03)	0.945	−0.01 (0.03)	0.714	−0.02 (0.01)	0.163
Observations	738		738		669
R-square	0.459		0.592		0.473
R-square adjusted	0.452		0.586		0.465
F statistics	61.73 ***		105.46 ***		61.38 ***

Note: Numbers in parentheses are standard errors. * p < 0.10 ** p < 0.05 *** p < 0.01.

, only some variables demonstrated notable changes. The following analysis will only focus on the variables that exhibited significant changes across the regression models.

Female presence on boards had no significant impact on overall sustainability performance. However, it was observed that female presence had a significant impact on the environmental performance (β = 4.26, p = 0.012) and social performance of a firm (β = 5.78, p < 0.001).

Earlier, it was detected that foreign directors have a negative but insignificant impact on the overall ESG performance of a firm. It turns out that the social performance of a firm is significantly reduced by the presence of at least one foreign board member (β = −6.29, p < 0.015).

Overall, the ESG performance of a firm is found to be positively and significantly affected by the ratio of highly educated directors. However, educational diversity has a negative but insignificant impact on environmental performance (β = −0.01, p = 0.769).

Initially, a positive and significant effect of foreign-educated directors on the overall ESG performance became non-significant (0.03, p = 0.248) when it was regressed on firms’ governance performance.

5.4. Regression Diagnostics

The reliability of our OLS regression model depends on whether the regression assumptions are met. The assumptions are conditions required for the OLS regression to show unbiased results. Violation of assumption may result in biased parameter estimates, standard errors, and biased p-values, which can negatively affect the reliability of the inference statistics of our model [48,49].

Considering this adverse effect, we have employed diagnostic checking in our model to detect any potential violations of assumption. There are five main assumptions of the OLS regression: linearity, normality, homoscedasticity, multicollinearity, and outliers

5.4.1. Linearity

First and foremost, as the name implies, the linear regression models depend on the linearity of the relationship between dependent and independent variables. Put differently, the response variable should be a linear function of the predictor variable. We have tested the linearity assumption by running the “Rainbow test” for linearity in the lmtest package (version 0.9–40) in the R programming language. The test expects a good linear fit even on a subsample in the middle of the data if the true relationship is indeed linear. In a non-linear model, the central parts of the data would be less curvy than the overall model fit. If the model fit improves on the central part of the fit, we can expect a non-linear relationship. The rainbow test describes whether a model fit improves when data are removed—when we look only at the central part of the data. In other words, the test compares the overall fit to the fit of central parts (subsample) of the data. The null hypothesis states that there is no significant difference between the fits. We have observed no significant deviation from linearity, as we could not reject the null hypothesis (p = 0.169) (see

Table 7. Rainbow test of linearity.

Rain = 1.106

df2 = 358

p-value = 0.169

p-value = 0.169

).

5.4.2. Normality

The assumption of normality states that the errors in the estimation of outcome variables are normally distributed. For the method of least squares to estimate the parameters optimally, it is assumed that the residuals in the population, and thus in the sample, are normally distributed [48].

To check the normality of the residual assumption, we drew a Q-Q plot. Sample quantiles of residuals are aligned with theoretical quantiles of normal distribution. The empirical quantiles align closely with the expected theoretical values, as illustrated in Figure 1. The plot provides visual evidence that the residuals are normally distributed, providing evidence that the model does not violate the assumption of normality.

We also performed a Shapiro–Wilk test to check the normality of the residual assumption. The Shapiro–Wilk test compares the distribution of the residuals in the sample to a normal distribution. The null hypothesis is that the sample distribution is not different from a normal distribution. We found no strong evidence to reject the null hypothesis (p = 0.7311); thus, we can assume that the residuals in the sample are normally distributed, meeting the assumption of normality of residuals (see

Table 8. Shapiro–Wilk test of normality.

W = 0.998

p-value = 0.731

).

Moreover, the skewness is close to 0 and the kurtosis is close to 3 (see

Table 9. Skewness and kurtosis of distribution of residuals.

Skewness = −0.02

Kurtosis = 2.93

), indicating that there is no major skew or kurtosis in the distribution of the residuals [48].

5.4.3. Homoscedasticity

The assumption of homoscedasticity assumes that the variance of the error terms (i.e., spread of the residuals) does not depend on predictor variables. The variance of the residual is constant across every level of the predictor variable [50]. If this assumption is not violated, we would expect homogenous variance of residuals on a graph plotting residuals on fitted values. As we can see from Figure 2, the variance of the residuals is quite homogenous across all fitted values, indicating that the homoscedasticity of the residuals’ assumption is not violated.

To test the assumption of the constant variance of error terms, we employed the Breusch–Pagan Test using the olsrr package (version 0.5.3) in R programing language. It tests whether the variance of errors from a regression is dependent on the values of independent variables. The null hypothesis states that the residuals are distributed with equal variance. As

Table 10. Breusch–Pagan test for heteroscedasticity.

χ2	0.015
p-value	0.903

shows, we fail to reject the null hypothesis (p = 0.903). We have the confidence to assume that the assumption of homoscedasticity is met.

5.4.4. Outliers

Another issue that can bias the regression model is the existence of outliers in the data. An outlier is a case that deviates from the main trend of the date. Outliers not only bias the estimates but also significantly impact the total squared errors. The sum of squared errors is crucial for computing the standard deviation. The standard deviation, in turn, is essential for estimating the standard error, which is then used to determine confidence intervals. In short, if the sum of squared errors is skewed, it affects the standard error and, consequently, the confidence intervals will be biased [48]. Large outliers can make OLS regression results misleading [50].

We can identify outliers by plotting the studentized residuals. The studentized residual is the unstandardized residual divided by an estimate of its standard deviation that varies for each observation. Figure 3 shows the studentized residuals for each observation on a plot. Only one observation is above 3, a threshold deemed as an outlier. No extreme outliers were observed in our data that can cause biased estimation in the mode.

5.4.5. Multicollinearity

Earlier, we saw a correlation between some predictors in our model in Table 4, raising concern for multicollinearity. Multicollinearity is an issue because it becomes impossible to see a direct impact of a particular predictor variable on the outcome variable if that predictor variable is also under the influence of another predictor variable. As the collinearity increases, the estimated coefficients become untrustworthy, the R-square value diminishes, and the importance of predictors becomes interchangeable [48].

Thus, we performed collinearity diagnostics of the variance inflation factor (VIF).

Table 11. Multicollinearity diagnostics.

Variables	Tolerance	VIF
Brd_size	0.81	1.24
Fem_brd	0.75	1.33
Age_div	0.96	1.04
For_dir	0.97	1.03
High_edu	0.51	1.96
For_edu	0.50	1.99
ASST	0.52	1.91
ROA	0.74	1.35
LEV	0.70	1.43
GRW	0.92	1.08

shows the VIF and tolerance values. None of the VIF values are greater than 2 (way below the threshold 10), and the tolerance values (1/VIF) are below 0.2, indicating no potential problem of multicollinearity [48].

6. Discussion and Conclusions

This study investigated the relationship between board diversity attributes and the ESG performance of Korean-listed firms and sheds light on the unique aspects of Korean corporate culture. Our findings provide evidence for the resource dependence theory. Each board member contributes unique information and brings expertise, skills, experience, and perspectives, facilitating access to critical stakeholders and improving the firm’s ability to address ESG issues. This finding is in line with [25], where board size was found to be a significant predictor of the ESG score of a firm. There is a contrast between our results and Walls’s et al. [35], who found that larger boards have a weaker environmental performance. Larger boards, by incorporating a broader range of perspectives and expertise, are better equipped to support the effective implementation of ESG practices.

On the question of whether women on the board improve ESG performance, we have not found enough evidence to support that the presence of women on the board improves the overall sustainability record of a firm. Contrary to theoretical prediction and in contrast to the existing literature, especially in Western economies, we did not find a statistically significant association between the presence of female directors and ESG performance in Korean-listed firms. Our findings differ from the findings of [2,23,36,39]. The results are different from what those observed in Western countries, providing support for [7] who suggested that the board diversity–sustainability relationship depends on the country-specific culture, such as strong collectivism dominant in Asia.

This raises important questions about the role of women and feminism in South Korea’s corporate governance structure. As our sample indicates, South Korean firms have a long way to go in achieving gender equality, with only 35% of listed companies having at least one female director on their boards. Furthermore, the presence of female directors appears to have a limited influence on corporate sustainability policies. In collectivist societies like South Korea, where cultural norms emphasize harmony and conformity, gender diversity within predominantly male boards may be perceived as disruptive, potentially challenging the harmony within the boards.

Moreover, we explored the influence of age diversity on ESG performance, finding a significant negative association. The conflicting views and lack of group cohesion resulting from age differences in the board could be potential reasons for this negative relationship. This result corroborates earlier studies [7,8] that diversity, age diversity in our case, can disrupt the sense of group harmony and conformity through conflicts stemming from differences in the personal values of directors of different ages.

Our results also confirm [40], which states that diverse boards with different age groups are more likely to suffer from miscommunication and personal friction and conflicts in the boardroom. As the authors suggest, such intergenerational differences are common in countries that have experienced rapid economic transformation over a short time, as each generation is a product of the economic and political conditions of their time. South Korea’s rapid economic development might have widened the intergenerational gap, leading to conflicting views about certain topics or a lack of group cohesion within the boards.

Our results showed that age diversity within boards negatively affects the overall ESG performance of a firm due to conflicting views and a lack of group cohesion. This is in line with [11], which states that age diversity might cause polarization in boards due to discrimination of members from the same age group against outsiders. The adverse effect of age diversity on team cohesion within boards was also observed in the Chinese market [14]. However, the negative effect is in contrast with Beji’s et al.’s [27] findings from the French market and with Cucari’s [30] findings from the Italian market.

The presence of foreign directors did not show a statistically significant effect on ESG performance, suggesting that, in the context of Korean-listed firms, the influence of foreign directors may be limited. This contradicts the research conducted in the West, where racial diversity in boards is found to have a significant effect on carbon emission performance [41] and overall CSR performance [27]. This finding also provides strong evidence for Chang et al.’s [7] argument which claims that the impact of board diversity on CSR is context-dependent.

There can be several underlying reasons that prevent foreign board members from contributing to the firm’s ESG performance with their knowledge and perspective. For example, such results may arise from the possibility that factors such as a small number of foreigners on boards or a lack of board independence might impede foreign directors’ contribution to firm value, as suggested by Hwang et al. [45]. They documented that foreign directors have limited value as independent outside directors without board independence.

The lack of effect might be related to the overall small number of foreigners in Korea, which is a mono-ethnic society. According to the monthly immigration report of the Ministry of Justice of the Republic of Korea [51], the number of foreigners residing (working, studying, living) in the country is 2,496,092 as of October 2023. This is just 4.8% of the whole population. This ratio becomes even smaller if we exclude temporary residents such as tourists, visitors, university students, blue-collar workers who are staying in the country temporarily, and a significant number of illegal workers.

Another factor that might hinder foreign directors from influencing the board’s decisions on corporate sustainability might be the lack of board independence in Korean firms, as suggested by Choi et al. [43] and Hwang et al. [45]. They observed that the lack of board independence keeps foreign directors from contributing to firm value. The nationality of the foreign board member is an important predictor of a firm’s commitment to sustainability practices, as observed by Kang et al. [8] in the Korean market. On the other hand, language barriers and cultural differences between foreign and local directors can prevent boards from addressing sustainability issues effectively, as they challenge in-group harmony and conformity [7] and cause friction between directors due to miscommunication and misunderstanding [13].

We also investigated the relationship between the proportion of highly educated directors and ESG performance. This finding provides evidence for predictions of the resource dependence and human capital theories that a larger number of board directors with higher educational degrees can bring a greater depth of expertise and awareness to sustainability issues, leading to improved overall ESG performance of a firm. Highly educated directors are highly likely to be exposed to the latest sustainability trends in business, politics, and academia. More and more universities are integrating sustainability-related subjects in their graduate courses or opening up new departments related to sustainability fields such as environmental and climate sciences and sustainable management.

The overseas education of directors was found to be positively associated with ESG performance, aligning with the notion that they can act as bridges for transferring knowledge and skills related to sustainability. It emphasizes the importance of considering the cultural background and educational experiences of directors in understanding their influence on ESG performance. Consistent with [34], overseas education of directors has a significant impact on environmental CSR performance. Educational and cultural experience affect the director’s stance toward corporate sustainability. A director who has studied in a society with higher sustainability standards will be more informed and sensitive towards sustainability issues, as they will not only learn information but also absorb cultural values. In our data, the majority of the directors who have studied overseas are educated in Europe or North America, where society is more sensitive to climate issues.

This result also confirms the “bridging” role of overseas-educated directors in bringing the latest knowledge and expertise, as highlighted by [45]. Directors can transfer what they have learned about sustainability overseas to boards. As confirmed by [43], boards learn corporate governance practices from the overseas experience of the directors and adopt them in their boards.

When examining the impact of board diversity on individual ESG components, interesting results emerged. The inclusion of women had a considerable influence on both the environmental and social components of a company’s ESG performance. This result supports earlier findings [38,45], which argue that female directors are sensitive and show a greater focus on environmental and social practices compared to their male counterparts. Moreover, women are more likely to voluntarily follow environmental regulations and maintain a long relationship with stakeholders [12]. Gender-based expectations exert group pressure on woman directors to show a more ’caring’ approach toward social issues [36].

Foreign directors were found to negatively and significantly affect the social component of a firm’s overall ESG score. Our result stands in contrast to [25], where the presence of foreign directors on the board was found to be positively and significantly associated with the overall CSR score, along with the environmental and community involvement (social) score of a firm. A foreign director might not be able to grasp the norms and values of the society within which they operate the company.

Despite the fact that highly educated directors positively affected the overall ESG performance, they had a negative but insignificant impact on environmental performance. The same result was observed by Post et al. [34], who found no relationship between directors’ educational achievements and the level of their environmental concern. They may lack the power or independence from the board to exert an influence on the company’s environmental policies.

Initially, we observed a positive effect of foreign-educated directors on overall ESG performance, yet their effect on governance performance was observed to be non-significant. This implies that foreign-educated directors have a significant effect on firms’ environmental and social performance but are incapable of impacting their governance performance.

One explanation for these negative relationships can be related to the adverse effect of cultural diversity. Cultural diversity can bring relationship conflicts with groups due to slower communication, misunderstanding, and a lower level of trust between the members of the group, affecting the group’s decision quality. Frijns et al. [32] observed such a negative effect of culturally diverse boards on the board performance and thus on firm performance in UK firms. They relate this to challenges stemming from difficulty in coordination and communication in culturally heterogeneous boards.

Our research has theoretical and practical implications. It expanded the existing governance literature by exploring the various aspects of board diversity in a non-Western context. This research addresses the gap in understanding the relationship between board characteristics and CSR performance, particularly in the underexplored context of Asian countries. Our results confirmed the prior research that governance characteristics such as board composition are an important indicator in shaping the strategic direction of firms in addressing sustainability issues. We have tested the validity of organizational theories developed in Western contexts within the Korean corporate environment. Our findings reveal that while some aspects align, others diverge significantly, underscoring the need for a localized understanding of corporate governance and sustainability practices.

Our findings also provide guidelines for firms’ leadership (directors and managers) desiring to improve the sustainability profile of their firms. Companies may benefit from diverse and inclusive board composition by gaining the trust of their stakeholders. This will enable firms to build healthier and sustainable relationships with their stakeholders.

Policymakers should consider establishing regulatory frameworks that encourage greater diversity on corporate boards, such as gender quotas or incentives for appointing directors from varied demographic and professional backgrounds.

Korean firms seeking to enhance their ESG performance should consider appointing directors with advanced academic degrees and international educational backgrounds. To maximize the contributions of these directors, it is essential to give them more voice in decisions related to environmental and governance matters. Strengthening board independence—a noted weakness in Korean firms—can be a pivotal step toward achieving this goal.

Furthermore, firms concerned with social and environmental issues should consider increasing the representation of women on their boards, as their presence significantly enhances environmental and social performance. However, leadership in Korean companies must actively address gender diversity and ensure that women have a stronger voice in governance-related decisions. The potential of women directors in governance remains underutilized. Similarly, the expertise of foreign directors is overlooked in Korean boards. Fully tapping into their knowledge can introduce global best practices in ESG to the organization.

Finally, Korean firms should address intergenerational differences among directors by reducing miscommunication and fostering open dialogue. Such efforts can help mitigate conflicts and create a more cohesive board environment, ultimately supporting the firm’s ESG objectives. This paper has some limitations. We conducted our research based on the observations only from a single year. Future research may benefit from using panel data to gain a comprehensive understanding of the relationship between board diversity and ESG performance on a broader time scale. Another limitation that can be addressed in future research is regarding the measurement of the variables. Future research can measure female representation with the percentage ratio of the female directors on the board.

Additionally, moderating and mediating effects are rarely explored in the existing literature. It would be meaningful to further research the underlying mechanisms through which board diversity influences ESG performance. Board independence, suggested by Hwang et al. [46], emerges as a potential moderator/mediator that could uncover the complex dynamics of the board diversity–ESG relationship.

Our study contributes by revealing that board diversity goes beyond simple gender diversity, which has often been used as the sole indicator of board diversity. We emphasize that when implementing board diversity, the multidimensional characteristics of diversity should be taken into account, along with acknowledging the unique impact of the local context and culture.