Effects of CEO Overseas Experience on Corporate Social Responsibility: Evidence from Chinese Manufacturing Listed Companies

Abstract

With increasing economic globalization, CEOs with overseas study or work experience play a crucial role in corporate strategic decision making, especially in emerging economies. Using Chinese manufacturing companies publicly listed on the Shanghai or Shenzhen Stock Exchanges, we explore the influence of CEO overseas experience on corporate social responsibility (CSR), based on the upper echelons theory and behavioral theory of the firm. We find that CEOs with overseas experience have a stronger CSR thinking concept and are more inclined to fulfill CSR than CEOs without overseas experience. We also analyze the effects of the organization’s external environment (market competition intensity) and internal resources (organizational slack resources) on this relationship, and find that all of these factors can strengthen the relationship. We find that the effect of CEO overseas study experience is stronger than that of overseas work experience; CEOs with overseas experience from developed countries such as Europe and the United States have a stronger effect on CSR; CEO overseas experience leads to better performance in state-owned enterprises; but when the CEO is also the chairman of the board, the effect of CEO overseas experience on CSR is weaker; female CEOs’ overseas experience has a stronger positive effect on CSR than male CEOs’; and the effect of CEO overseas experience on CSR is greater when the CEO is highly educated.

1. Introduction

With the rapid development of the global economy and the intensification of market competition, enterprises are increasingly realizing the power of “social responsibility” as well as the value of paying attention to the industry market’s competitive position and integrating their various advantages and resources. Furthermore, enterprises need to continuously enhance their awareness of their social responsibility and actively fulfill corporate social responsibility (CSR). Responsible business activities are also generally considered essential for enterprises to achieve sustainable development [1,2]. Given the importance of CSR to corporate strategic decision making [3], widespread attention is focused on the factors influencing active CSR activities. A range of research findings has emerged, mainly related to external factors such as laws and regulations [4,5], activist groups [6], stakeholder identity orientation [7,8], institutional investment [9], as well as internal factors such as slack resources [10], organizational size [11], and ownership structure [12,13]. Researchers have gradually shifted their attention to the individual level within organizations, for example, considering the influence of top management team (TMT) members on CSR.

The upper echelons theory suggests that TMT members’ background characteristics influence strategic enterprise choices and decision-making behavior [14]. In nearly every organization, the CEO is the most influential decision maker [15,16]. The personality traits and organizational identity of the CEO affect an enterprise’s strategic choices, which play a vital role in the enterprise’s final strategic decision making [17,18]. However, CSR is an essential component of corporate strategic decision making, and CEO background characteristics will inevitably have a significant impact on CSR fulfillment. In this context, existing studies focus on the CEO’s characteristics, identity, and organizational-related attributes: first, the relationship between individual characteristics and CSR, mainly including factors such as the CEO’s moral leadership [19,20,21], transformational leadership [22], gender [23], education level [24,25], ability [26], overconfidence [27], narcissism and hubris [28,29,30,31], responsible leadership styles [32], entrepreneurial orientation [33], greed [34], luck [35]; second, the relationship between individual identity and CSR, including the CEO’s power [36,37,38], family CEO [39], ownership [40], celebrity status [41], political ideologies [42], and the transgenerational transfer of entrepreneur’s political appointments or identity [43]; third, the relationship between organization-related attributes and CSR, including the CEO’s compensation [44,45], tenure [46,47], career horizon problems [48], duality [49,50,51], incentive mechanism [52,53], internal debt [54,55].

Despite the wide range of research fields, few studies have examined CEO overseas experience, i.e., whether the CEO has obtained any work or study experience in a foreign country. In the literature, studies have only explored the impact of returnee senior managers or board members and CSR performance [56,57,58], while ignoring the relationship between the overseas experience of the lead TMT decision maker—the CEO—and CSR. For this reason, we select CEO overseas experience as the influencing factor of CSR, mainly considering the following aspects: First, overseas experience includes overseas work experience or study experience, indicated by exposure to Western living culture or experience of learning styles abroad. Overseas experience is generally regarded as a sign of a good educational background or professional knowledge and skills, reflecting the individual’s vision, knowledge structure, experience, and skills [59,60]. Therefore, compared with CEOs without overseas experience, CEOs with overseas experience are more adept at corporate strategic decision making [36]. Second, since the 1970s, the Chinese economy has undergone rapid development, and is now an indispensable part of the global economy and the most important emerging economy in the world. In this context, China provides an ideal setting for studying the effect of CEO overseas experience on CSR in emerging markets [61]. Finally, the number of Chinese students studying overseas has been increasing year by year. The tremendous development potential and employment opportunities available in China have also promoted China’s transformation from the world’s largest talent outflow country to a talent return country. According to the statistics of the Ministry of Education of the People’s Republic of China, the total number of Chinese students studying abroad in 2018 was 662,000, which was 4.6 times more than the 144,000 in 2007, most of whom were in developed countries such as the United States or European countries. The statistic also shows that the proportion of returnees among the number of students studying abroad in China increased from 30.6% in 2007 to 78.5% in 2018, and is increasing year by year, as more overseas students choose to return to their home countries for employment and entrepreneurship. Therefore, there is no lack of high-end talents to master cutting-edge technologies. After returning to China, they serve as senior managers in enterprises and become the main force of enterprise development.

In this context, based on the upper echelons theory and the behavioral theory of the firm, this paper uses Chinese manufacturing listed companies over the period 2009–2018 as research samples to further explore the effect of CEO overseas experience on CSR performance and the contextual factors that stimulate or constrain this effect.

The remainder of this paper is organized as follows: Section 2 reviews relevant studies and develops our hypothesis. Section 3 introduces the research design and sample distribution. Section 4 provides empirical results and discussion. Section 5 concludes the paper.

2. Theory and Hypothesis

2.1. Analysis of the Relationship between CEO Overseas Experience and CSR

The behavioral theory of the firm was developed by Cyert and March in 1963, based on Simon’s “bounded rationality,” with enterprise decision making as the core [62,63]. The theory takes an organization as the research object, considers that organizational decision makers are constrained by information and technology, and emphasizes that decision making is constrained by bounded rationality. Bounded rationality can be regarded as the cognitive essence of strategic enterprise choice. This kind of bounded rationality has an impact on enterprise decision-making processes and decision-making behavior. Furthermore, CSR is an essential part of enterprise strategic decision making, initiated and implemented by the top decision maker.

CSR can be regarded as a long-term enterprise investment, which requires continuous attention from TMT members [48]. CSR decision making requires enterprise leaders to have foresight and wisdom, foresee future challenges and opportunities, and lead the enterprise. The measurable background characteristics of executives (e.g., education, work, and career experience) are closely related to executives’ cognitive abilities and values [14], which are unique demographic characteristics of TMT members. As the top leader of the corporate executive team, the CEO’s characteristics and life experience may have certain differences from those of other members of the corporate executive team in terms of knowledge structure, vision, cognition, and value judgment [42], which can affect the choice and implementation of CSR decisions. Previous studies have also shown that CEO experience significantly impacts corporate decisions [64,65].

CEO experience can be divided into overseas experience and local experience [58]. Because CEOs who have gained work or study experience in developed countries, such as the United States or European countries, have been exposed to various regional cultures, they may have a more elevated vision, wider horizons, and more advanced ideas. They tend to be better in terms of selection and formulation of corporate decisions, and more likely to grasp future development opportunities. This diversity of experience may make their cognitions and values correspondingly influenced and more adaptable to the rapidly changing internal and external environment. Furthermore, CEOs with overseas experience may gain certain information and resource advantages by relying on the various social network resources they have accumulated, which may be more conducive to implementing enterprise decisions on the ground. CEO overseas experience can significantly improve corporate governance and effectively circumvent management myopia, leading to a greater focus on long-term corporate investment decisions such as being more active in CSR [59].

This paper argues that CEOs with overseas experience are more open-minded in terms of cognition, vision, and experience and will perform better in CSR than those with only local experience. This is because CEOs with overseas experience have a more strongly CSR-oriented mindset and are more aware of its importance. CEOs with overseas experience have a broader strategic vision, global perspective, access to international resources, and understanding of advanced technology. They are better at making long-term corporate decisions, which help companies fulfill their social responsibility and seek long-term investment returns. CEOs with overseas experience have stronger decision-making concepts and higher ethical standards in CSR decision making and implementation [66] and are prone to forming mindsets that value social responsibility and are conducive to CSR activities. Guided by high ethical standards and value concepts, CEOs with overseas experience tend to focus on CSR and are better able to defend their stakeholders’ legitimate rights and interests, thus helping to improve CSR scores [47].

Furthermore, CEOs with overseas experience are more familiar with advanced models and practical experience of fulfilling CSR. Compared with CSR practices in developed countries, the institutional mechanisms for CSR incentives in China are relatively less developed in terms of strategy, business environment, regulatory framework, and supervision, and the legal system and evaluation system. CEOs with overseas experience identify more strongly with the concept of CSR, gain practical experience of advanced CSR in Western countries, and apply it to Chinese companies by combining it with the Chinese contextual culture. Executives who are imbued with CSR culture in Western countries are more familiar with global implementation, which is conducive to guiding CSR activities in their own country, and the CSR score of their firm is thus higher [67]. Lau et al.’s study [66] shows that executive members with work or study experience abroad (especially in developed countries) can directly contact and engage in CSR practices in Western countries, improving and enhancing CSR performance. Based on the above analysis, we propose the following hypothesis:

Hypothesis 1 (H1).

There is a positive relationship between CEO overseas experience and CSR, i.e., companies with CEOs who have overseas experience have higher CSR scores than companies without CEOs who have overseas experience.

2.2. Contextual Mechanism Analysis of the Relationship between CEO Overseas Experience and CSR

Based on the above analysis, CEO overseas experience has a facilitating effect on CSR. This paper further argues that this influencing effect is not invariable because the organization’s characteristics are different, and the external market environment is also dynamically changing. That is to say, the effect of CEO overseas experience on CSR may vary even in the same context, depending on different contextual factors such as the organization’s characteristics and internal and external environment. Based on this, this paper further explores the contextual mechanisms that influence the relationship between CEO overseas experience and CSR, mainly considering the effect of the organization’s external environment (market competition intensity) and internal environment (slack resources) factors.

2.2.1. External Environmental Characteristics: The Moderating Effect of Market Competition Intensity

Market competition intensity is the competitive situation of the industry in which the enterprise is located and has always been considered an important condition for CSR success [68]. The intensity of competition is reflected in fierce price wars, diversified product substitutes, and additional services. It is the level of direct competition faced by enterprises in their business fields and plays a vital role in CSR fulfillment [69]. As an external contextual mechanism, market competition intensity is more likely to facilitate CEO overseas experience. It may enable the CEO to make more favorable strategic decisions, thus contributing to the role of the CEO overseas in fulfilling CSR. Previous related studies are further confirmed. Luo and Zheng [68] found that CSR performance will be enhanced when market competition intensity is high. Soewarno et al. [70] show that competition intensity has a positive direct effect on CSR.

This paper argues that competition intensity in the external market can affect how CEO overseas experience affects CSR performance changes. Specifically, market competition intensity has a positive moderating effect on the relationship between CEO overseas experience and CSR, i.e., the greater market competition intensity, the stronger the positive effect of CEO overseas experience on CSR. This is because, first, CEOs with overseas experience show a higher willingness to engage in CSR, and perform CSR more actively under high competition intensity. In an environment of high market competition, enterprises may implement CSR to achieve their differentiated competitive strategies to meet various stakeholders’ interests, win shareholder support, elicit partner commitment, and achieve consumer loyalty [71]. According to the strategic CSR perspective, as the market competition intensity increases, the CEO may adopt social responsibility behaviors to achieve competitive differentiation and corporate strategic goals, such as green products and public welfare marketing [72].

Second, CEOs who have worked or studied abroad tend to have broader decision-making horizons and can better identify and seize the opportunities embedded in a highly competitive environment to actively develop and implement CSR in a more competitive industry. Compared with lower competition intensity industries, CEOs have a more significant impact on CSR in industries with fierce market competition. Furthermore, the greater the market competition intensity, the more critical corporate reputation becomes [73]. Therefore, compared with that in less competitive markets, CEO overseas experience has a greater impact on CSR in a highly competitive market. Based on the above analysis, we propose the following hypothesis:

Hypothesis 2 (H2).

Market competition intensity will positively moderate the relationship between CEO overseas experience and CSR.

2.2.2. Organization Characteristics: The Moderating Effect of Slack Resources

As a strategic resource, slack resources can reduce the firm’s internal constraints and mitigate the risks that the firm needs to face due to changes in the external environment [74]. According to organization theory, an enterprise’s slack resources are potential resources that can be redirected beyond the needs of the firm to maintain its operations and can effectively help the firm to resist external disturbances, cope with environmental changes, and achieve performance improvement [63,75,76]. Bourgeois [77] believes that slack resources are an actual or potential resource that can serve as a buffer to enable the organization to adapt when faced with internal pressures and adjust under external environmental pressures through policies. Previous related studies have also shown that organizational slack resources are an important contextual factor of CSR and positively contribute to CSR [78,79,80].

This paper argues that slack resources may enhance the positive relationship between CEO overseas experience and CSR, i.e., the more slack resources that a business organization has, the more likely a CEO with overseas experience will use available strategic resources to make and implement CSR decisions. This is because slack resources provide a fundamental guarantee for CEOs with overseas experience to develop and implement CSR decisions. Slack resources represent the firm’s potential resources, and CEOs with overseas experience tend to choose slack resources to cushion the impact of the intensely changing external environment on the firm [81]. The fulfillment of CSR behavior not only requires a strong concept of social responsibility but also sufficient resources to support it.

Furthermore, slack resources are a necessary condition for the CEO to seize market opportunities and sustain the enterprise’s long-term survival and development. CEOs with overseas experience are more likely to take advantage of business development opportunities, use slack resources to develop strategies, and implement strategic changes and innovations to gain and sustain long-term competitive advantages. As the core of TMT, the CEO can make and implement corporate strategic decisions more rationally and efficiently based on the advantages offered by available resources when they have overseas experience. Thus, it is not only necessary for the CEO to have good CSR awareness, but also, more importantly, the company must have slack resources for the CEO to make and implement social responsibility decisions [82]. Based on the above analysis, we propose the following hypothesis:

Hypothesis 3 (H3).

Slack resources will positively moderate the relationship between CEO overseas experience and CSR.

3. Research and Design

3.1. Research Samples and Data Sources

We take Chinese listed manufacturing companies on the Shenzhen Stock Exchange and the Shanghai Stock Exchange from 2009 to 2018 as the research sample. To ensure the rationality of the sample selection, the following selections are made: (1) excluding companies whose corporate nature cannot be determined; (2) excluding S, ST, * ST, S * ST, and PT company sample; and (3) excluding other listed companies with a severe lack of data. Finally, we constructed an unbalanced panel dataset spanning 396 publicly listed companies in the Chinese manufacturing industry from 2009 to 2018, with 3009 observations.

Our dataset was compiled from two major sources: the China Stock Market and Accounting Research (CSMAR) and the Rankins CSR Ratings (RKS) database. Data on CSR came from the RKS database, and other indicator data were mainly from the CSMAR database. RKS is the first independent, authoritative third-party social responsibility rating agency in China.

3.2. Main Research Variables and Definitions

3.2.1. Dependent Variable

We take CSR as the dependent variable to study the effect of independent variables on it. Drawing on the practice of existing research results [15,26,41,80], this paper uses the scoring results of Running Global (RKS) on the social responsibility reports of listed companies to measure CSR performance [58]. RKS is a third-party social responsibility rating agency in China. The CSR evaluation system established by RKS is based on the framework of the Global Reporting Initiative (GRI) and incorporates Chinese-specific CSR elements. Its original evaluation data comes from the enterprises’ publicly released social responsibility reports and the official websites and news media. The evaluation results have reasonable objectivity, authenticity, and reliability. Therefore, this paper selects Runling Global’s CSR score as a measure of the dependent variable. The rating system sets up 15 primary indicators and 63 secondary indicators (excluding the general industry and other manufacturing industries) from four dimensions: Macrocosm(M), Content (C), Technique (T), and Industry (I), to make a comprehensive evaluation of CSR quality. The evaluation system adopts a structured expert scoring method, with a full score of 100. It is shown in Appendix A.

3.2.2. Independent Variable

This paper mainly discusses the relationship between CEO overseas experience and CSR. Because of the late introduction of the CEO concept in Chinese enterprises, this paper regards the positions of “President” or “General Manager” in listed companies as CEO positions. We select CEO overseas experience (COE) as the independent variable, i.e., whether the CEO has overseas experience or not is selected as the determining factor of COE. It should be noted that we exclude CEOs working in foreign branches of Chinese companies or Chinese branches of foreign companies. We do not regard CEOs in these situations as having foreign experience, ensuring that the measure is truly indicative of exposure to a foreign environment.

Drawing on the practice of existing research results [59,60], for the CEO overseas experience variable, we mainly construct the dummy variable COE, which is defined as follows: when the CEO has an overseas study or work experience, it is recorded as 1; otherwise, it is 0.

In further analyses, we distinguish the type of CEO overseas experience into two categories: (1) CEO study experience (COEEDU), a dummy variable is equal to 1 if the CEO has overseas study experience, i.e., COEEDU = 1, and 0 otherwise. (2) CEO work experience (COEJOB), a dummy variable is equal to 1 if the CEO has overseas work experience, i.e., COEJOB = 1, and 0 otherwise.

3.2.3. Moderating Variables

(1)

Market Competition Intensity (MCI)

Market competition intensity is the intensity of competition among competitors in an industry, reflecting the degree of market concentration in the industry. The higher the market concentration, the greater the firm’s market dominance, and its calculation requires the operating income of all firms in the industry [83]. We follow Li et al. [84] to select the Herfindahl index to measure market competition intensity, as follows:

$$HHI={\displaystyle \sum (X_i}/X{)}^2$$

where X is the main business income of all enterprises in the industry, and Xi is the main business income of enterprise i. This is used to measure market competition intensity. The larger the value, the greater the intensity of market competition, and vice versa.

(2)

Slack Resources (SLACK)

Slack resources are considered as resources that exceed the organization’s actual needs and are kept within the organization as a potential and realistically available resource buffer. In the current research literature, slack resources are defined as the difference between a firm’s available working capital and its required capital [48,77,81]. To be consistent with previous studies, based on the practice of Bourgeois’ financial data-based measure of organizational slack resources [77], we measure the organization’s slack resources through the ratio of equity to debt.

3.2.4. Control Variables

Referring to the existing research literature [28,31,36,45,48], combined with enterprises’ actual development status, we control for several firm-level variables that were potentially relevant to CSR.

The firm-level variables include the following indicators: (1) Firm Size (FS): Size is the natural logarithm of the book value of total assets at the end of the period. Larger firms are more likely to implement socially responsible activities because of their visibility, subject to more rigorous public scrutiny. (2) Firm Life (FL): that is, firm age, which is calculated as the number of years from firm establishment to the statistical year, and the natural logarithm is taken. A study by Roberts [85] found that the older a company is, the more concerned it is about its reputation, and thus the more active it is in implementing CSR activities. (3) Capital Structure (CS), that is, debt-to-asset ratio, is calculated as total liabilities divided by total assets. Firms with more liabilities may disclose more information to the public to assure their corporate stakeholders that they will abide by their covenants. (4) Market Share (MS): we measure market share by the ratio of main business income to the industry’s total main business income. Companies with higher market shares tend to have greater social influence and are more willing to perform CSR activities. (5) Profitability (ROE): we measure profitability by the ratio of return on net assets, that is, net profit to the balance of shareholders’ equity. Highly profitable companies are likely to present a better social responsibility to society and often may be more willing to disclose social responsibility information [36]. (6) Firm Growth (FG): we measure firm growth by the ratio of operating income growth to the total operating income of the previous year. Firm growth can reflect a company’s ability to grow sustainably. (7) Ownership Structure (OS): OS represents the equity nature of the firm [86,87]. The system, business environment, and strategic objectives within enterprises with different ownership properties may differ significantly, which have an important impact on CSR fulfillment. This is determined according to whether the enterprise is owned by the state: if it is a state-owned enterprise, EN = 1; for a non-state-owned enterprise, EN = 0.

Furthermore, we control for individual CEO-level variables, mainly including the following indicators: (1) CEO Concurrently (CC): whether the CEO serves as chairman and CEO at the same time, two concurrent posts are 1, otherwise 0. The CEO who is also the chairman of the board has more decision-making power, which has an important impact on CSR [88]. (2) CEO Gender (CG): when the CEO is male, it is recorded as 1; otherwise 0. Previous related studies have shown that female CEOs are more concerned about social responsibility issues and perform better in CSR than male CEOs [23]. (3) CEO Age (CA): the age of the CEO during the sample period. As the age of the CEO increases, their sense of social responsibility will gradually increase, and older CEOs may pay more attention to CSR. (4) CEO Education (CE): CEO education level, which can be recorded as 5, 4, 3, 2, or 1 based on whether the CEO has a doctorate, master’s, or undergraduate degree, junior college, or technical secondary school (including technical secondary school and below). As the education level of CEOs increases, their values are more likely to be influenced by the concept of social responsibility, and they can be more aware of the long-term benefits of actively implementing CSR for their companies. (5) CEO Turnover (CT): when the CEO leaves his position for a specific reason, it is recorded as 1, otherwise 0. Related studies have shown that companies whose CEOs leave their positions due to the corporate governance system may actively engage in CSR disclosure [89].

Finally, we also set up dummy variables for the year (YR) and industry (IT). Industry dummy variables are used to control the possible impact of industry differences on CSR; year dummy variables are used to control the impact of changes in external policies, economic, and other macroeconomic conditions on CSR with annual changes.

3.3. Research Model Construction

To test the hypothesis proposed in this paper, we examine the regression model to be set as follows:

$$CSR={\mathit{\beta }}_0+{\mathit{\beta }}_1\times {\displaystyle \sum CN}+\mathit{\epsilon }$$

(1)

$$CSR={\mathit{\beta }}_0+{\mathit{\beta }}_1\times COE+{\mathit{\beta }}_2\times {\displaystyle \sum CN}+\mathit{\epsilon }$$

(2)

$$CSR={\mathit{\beta }}_0+{\mathit{\beta }}_1\times COE+{\mathit{\beta }}_2\times COE\times MCI+{\mathit{\beta }}_3\times MCI+{\mathit{\beta }}_4\times {\displaystyle \sum CN}+\mathit{\epsilon }$$

(3)

$$CSR={\mathit{\beta }}_0+{\mathit{\beta }}_1\times COE+{\mathit{\beta }}_2\times COE\times \mathit{SLACK}+{\mathit{\beta }}_3\times \mathit{SLACK}+{\mathit{\beta }}_4\times {\displaystyle \sum CN}+\mathit{\epsilon }$$

(4)

CSR is an explained variable that represents corporate social responsibility. The larger the value, the better the CSR performance. COE is an explanatory variable, indicating that the CEO has overseas experience.$\sum CN$ is a control variable at the firm and individual levels. Model (1) is mainly to verify the influence of control variables on CSR. Model (2) is based on model (1) and includes independent variables to test the impact of CEO overseas experience (COE) on CSR. Models (3)–(4) respectively test the moderating effects of market competition intensity (MCI) and slack resources (SLACK) on the relationship between COE and CSR.

3.4. Descriptive Statistics and Correlation Analysis

We perform descriptive statistics and correlation analysis on the main variables and control variables involved, such as CSR, COE, market competition intensity, and slack resources, as shown in

Table 1. Descriptive statistical analysis of main variables.

Variable	Mean	Standard Deviation	Minimum	Max	Median	Observations
CSR	37.940	10.940	19.950	70.940	35.710	3009
COE	0.077	0.267	0	1	0	3009
COEJOB	0.052	0.222	0	1	0	3009
COEEDU	0.036	0.186	0	1	0	3009
FS	22.850	1.324	19.540	27.390	22.700	3009
FL	2.771	0.327	1.609	3.434	2.833	3009
CS	0.463	0.191	0.062	0.868	0.476	3009
MS	0.014	0.053	3.40 × 10−6	0.360	7.73 × 10−5	3009
ROE	0.076	0.110	−0.474	0.387	0.075	3009
FG	0.135	0.266	−0.399	1.250	0.104	3009
OS	0.576	0.494	0	1	1	3009
CC	0.197	0.398	0	1	0	3009
CG	0.939	0.240	0	1	1	3009
CA	49.970	6.179	30	81	50	3009
CE	3.607	0.829	1	5	4	3009
CT	0.084	0.278	0	1	0	3009
MCI	0.217	0.187	0.052	1	0.172	3009
SLACK	1.168	1.008	0.060	5.980	0.903	3009

and

Table 2. Correlation analysis of main variables.

Variable	1	2	3	4	5	6	7	8	9
CSR	1
COE	0.051 ***	1
COEJOB	0.053 ***	0.807 ***	1
COEEDU	0.024	0.666 ***	0.205 ***	1
FS	0.411 ***	0.003	0.001	–0.005	1
FL	0.155 ***	–0.036 **	–0.013	–0.032 *	0.131 ***	1
CS	0.093 ***	0.011	0.022	–0.012	0.517 ***	0.061 ***	1
MS	–0.067 ***	–0.016	–0.027	0.003	0.250 ***	–0.195 ***	0.176 ***	1
ROE	0.023	0.007	0.031 *	–0.006	0.090 ***	–0.064 ***	–0.212 ***	0.113 ***	1
FG	0.007	0.022	0.038 **	–0.003	0.055 ***	–0.070 ***	0.011	0.141 ***	0.301 ***
CC	0.02	0.056 ***	0.072 ***	–0.015	–0.033 *	–0.004	–0.089 ***	–0.039 **	0.061 ***
OS	0.073 ***	–0.098 ***	–0.084 ***	–0.084 ***	0.264 ***	0.083 ***	0.234 ***	0.120 ***	–0.134 ***
CG	–0.073 ***	0.017	0.060 ***	–0.033 *	–0.004	–0.014	0.047 ***	0.033 *	–0.068 ***
CA	0.143 ***	0.015	0.095 ***	–0.078 ***	0.177 ***	0.150 ***	0.051 ***	–0.032 *	0.018
CE	0.135 ***	0.146 ***	0.125 ***	0.124 ***	0.191 ***	–0.036 *	0.080 ***	0.086 ***	–0.011
CT	0.042 **	–0.021	–0.017	–0.013	0.053 ***	0.097 ***	0.031 *	–0.054 ***	–0.109 ***
MCI	0.077 ***	0.043 **	0.022	0.02	–0.043 **	–0.045 **	0.113 ***	0.107 ***	–0.043 **
SLACK	0.057 ***	–0.008	0.007	–0.028	0.429 ***	0.038 **	0.864 ***	0.159 ***	–0.278 ***
Variable	10	11	12	13	14	15	16	17	18
CSR
COE
COEJOB
COEEDU
FS
FL
CS
MS
ROE
FG	1
CC	0.036 **	1
OS	–0.061 ***	–0.236 ***	1
CG	–0.022	–0.009	0.106 ***	1
CA	–0.031 *	0.225 ***	0.156 ***	0.077 ***	1
CE	0.005	0.053 ***	0.185 ***	0.031 *	–0.020	1
CT	0.003	0.015	0.074 ***	0.037 **	0.074 ***	0.045 **	1
MCI	–0.008	–0.007	0.016	–0.009	–0.056 ***	–0.037 **	0.001	1
SLACK	–0.030 *	–0.095 ***	0.229 ***	0.040 **	0.068 ***	0.053 ***	0.025	0.114 ***	1

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Year and industry virtual control variables are not listed.

.

Table 1 reports a descriptive statistical analysis of the main variables. The results show that the maximum value of CSR is 70.94, the minimum value is 19.95, the average value is 37.94, and the standard deviation is 10.94, indicating that there are large differences in CSR performance. While the average value of CEO overseas experience (COE) is 0.077. The standard deviation is 0.267, indicating that CEO overseas experience varies greatly between companies. Moreover, the mean and standards deviation of market competition intensity (MCI) and slack resources (SLACK) also differ significantly. Table 2 reports the correlation analysis between the main variables. The results show that CEOs’ overseas experience and CSR have a significant positive correlation (COE: Beta = 0.051, p < 0.01). However, correlation analysis only reflects the correlation between individual variables and does not control the impact of other factors on CSR. Therefore, to obtain a more accurate and rigorous conclusion, other control factors are added in the follow-up, and further regression analysis is performed.

4. Regression Results and Discussion

Before the empirical analysis, to ensure the consistency and validity of the model estimates, the data are processed as follows: (1) To avoid the influence of outliers, the main continuous variables are tailed at 1% and 99% [90]. (2) We diagnose the variance inflation factor (VIF) of all the explanatory variables and control variables entering the model. The results show that the VIF is about 2.25, and the VIF of each variable is well below the critical value of 10, which avoids multicollinearity. (3) Because this paper’s data are panel data, there may be heteroscedasticity, autocorrelation, and cross-sectional correlation. Using the usual panel data estimation method may underestimate the standard error, leading to biased model estimation results. Driscoll–Kraay(D-K) standard errors were used to ensure that the estimated standard errors are unbiased, consistent, and valid [91]. Therefore, in the follow-up panel data regression and robustness test, we mainly used the D-K standard error estimation method for correction.

4.1. CEO Overseas Experience and CSR

Before the sample regression analysis, it was determined whether a fixed-effects model or a random-effects model should be selected. This paper uses the Hausman test to determine this. The test results show that the deviation between the more effective estimator and the consistent estimator is significant (Prob > chi2 = 0.0000), which indicates that a fixed-effects model should be selected.

Table 3. Regression results of CEO overseas experience and CSR.

Variable			Moderating Effect
	Model (1)	Model (2)	Model (3)	Model (4)	Model (5)
	Benchmark Model	Main Effect	Market Competition	Slack Resources	Full Model
COE		1.113 **	0.278	1.053 **	1.921 ***
		(0.363)	(0.354)	(0.3863)	(0.422)
MCI			2.425 **		2.603 **
			(0.962)		(1.013)
COE × MCI			3.551 ***		3.409 ***
			(0.236)		(0.374)
SLACK				0.328 ***	0.289 **
				(0.070)	(0.096)
COE × SLACK				1.892 ***	1.960 ***
				(0.544)	(0.563)
FS	1.509 ***	1.529 ***	1.453 ***	1.478 ***	1.400 **
	(0.413)	(0.409)	(0.402)	(0.417)	(0.431)
FL	–4.371 ***	–4.644 ***	–4.441 ***	–4.939 ***	–4.768 ***
	(1.199)	(1.086)	(1.109)	(0.929)	(1.003)
CS	–0.350	–0.433	–0.354	−2.660	–2.408
	(1.756)	(1.744)	(1.730)	(1.606)	(1.646)
MS	–21.554 ***	–21.510 ***	–20.956 ***	–21.933 ***	–21.344 ***
	(1.612)	(1.603)	(1.457)	(1.540)	(1.462)
ROE	0.322	0.387	0.414	0.689	0.690
	(0.682)	(0.644)	(0.606)	(0.584)	(0.565)
FG	–0.325	–0.354	–0.431	–0.345	–0.422
	(0.332)	(0.326)	(0.307)	(0.333)	(0.331)
OS	–0.649 *	–0.661 *	–0.738 **	–0.665 **	–0.746 **
	(0.303)	(0.300)	(0.309)	(0.282)	(0.309)
CC	1.718 ***	1.834 ***	1.658 ***	1.948 ***	1.781 ***
	(0.405)	(0.413)	(0.412)	(0.414)	(0.435)
CG	–0.494	–0.480	–0.476	–0.555	–0.544
	(0.799)	(0.806)	(0.796)	(0.808)	(0.840)
CA	0.044 *	0.044 **	0.048 **	0.047 **	0.051 **
	(0.020)	(0.019)	(0.018)	(0.021)	(0.021)
CE	0.493 ***	0.422 **	0.444 **	0.387 **	0.410 **
	(0.147)	(0.155)	(0.162)	(0.142)	(0.157)
CT	–0.491 *	–0.475 *	–0.483 *	–0.456 *	–0.466 *
	(0.252)	(0.250)	(0.250)	(0.239)	(0.252)
Constant term	1.782	2.138	2.362	4.639	4.889
	(8.565)	(8.698)	(8.509)	(8.580)	(8.811)
YR	Yes	Yes	Yes	Yes	Yes
IT	Yes	Yes	Yes	Yes	Yes
N	3009	3009	3009	3009	3009
R2	0.441	0.442	0.443	0.445	0.473

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Driscoll–Kraay standard errors in parentheses; (3) Because of limitations on space, year and industry virtual control variables are included in each model, and the results are not listed.

presents the results of the analysis of the relationship between COE and CSR. Model (1) is the benchmark model, and model (2) uses CEO overseas experience as an explanatory variable for regression. The test results show a significant positive relationship between CEO overseas experience and CSR (COE: Beta = 1.113, p < 0.05). The regression coefficient of COE on CSR is 1.113 and significant at the 5% level. This indicates that companies with CEOs with overseas experience have 11.3% higher CSR scores than companies without CEO overseas experience, and hypothesis 1 is verified. Compared with executives without overseas experience, CEOs with overseas experience may have a broader overseas perspective, a greater reserve of cutting-edge knowledge, and more advanced CSR thinking concepts and ideas, and be more willing to fulfill CSR. Previous studies also have relevant findings. Studies by Wen and Song [56] and Zhang et al. [58] have shown that directors or managers with foreign study or work experience can improve and promote CSR performance. Our study focused on the CEO overseas work or study experience, which is consistent with previous research.

4.2. Analysis of the CEO Overseas Experience and the Moderating Mechanism of CSR

This section further analyzes the moderating effect of market competition intensity and slack resources on COE and CSR, based on the organization’s external environment and internal characteristics.

4.2.1. Characteristics of External Environment: The Moderating Effect of Market Competition Intensity

Model (3) in Table 3 analyzes the moderating effect of market competition intensity on the relationship between COE and CSR. The results show that the cross-term coefficient of market competition intensity (MCI) and CEO overseas experience (COE) is significantly positive (Beta = 3.551, p < 0.01). That is, the regression coefficient of the cross term of market competition intensity (MCI) and CEO overseas experience (COE) is 3.551 and significant at the 1% level, and the result remains robust in the subsequent full model (6). This shows that when market competition intensity is higher, the positive effect of CEO overseas experience on CSR will be enhanced, and hypothesis 2 is verified; the results remain consistent in the subsequent robustness tests. Luo and Zheng [68] explore market competition intensity as a crucial contextual moderator of CSR. Soewarno et al. [70] find that market competition intensity has a positive effect on CSR. Our study also follows previous studies, and the findings are confirmed and supported.

4.2.2. Organization Characteristics: The Moderating Effect of Slack Resources

Model (4) in Table 3 analyzes the moderating effect of slack resources on the relationship between CEO overseas experience and CSR. The results show that the cross-term coefficient of slack resources (SLACK) and CEO overseas experience (COE) is significantly positive (Beta = 1.892, p < 0.01). The regression coefficient of the cross-sectional term of slack resources (SLACK) and CEO overseas experience (COE) is 1.892 and significant at the 1% level, and the results remain robust in the subsequent full model (6). This indicates that the more extensive the slack resources in the organization, the greater the positive effect of CEO overseas experience on CSR. As a result, hypothesis 3 is verified, and the results remain consistent in the subsequent robustness tests. Zhang et al.’s study [80] finds that slack resources have a positive effect on enhancing CSR performance, and its transmission mechanism is to increase the quality of social responsibility disclosure by improving the level of corporate responsibility performance. However, our study also finds that slack resources are an essential contextual moderator of CSR scores, further confirming the related research.

4.3. Robustness Test

To verify the reliability of the research findings, we conduct robustness tests on the following aspects.

4.3.1. Endogeneity Issues

The previous regression results indicate that CEO overseas experience has a positive relationship with CSR. The results may be affected by omitted variables or reciprocal causality, thus generating endogeneity issues that may cause bias in the estimation results. Therefore, we use the instrumental variable method to deal with the endogeneity of the original model. First, we select appropriate instrumental variables to replace endogenous variables, and we then use the appropriate endogeneity treatment to obtain unbiased estimates. In general, the instrumental variables are selected to satisfy both the conditions of being correlated with the endogenous variables and being independent of the random disturbance terms.

According to the common practice in the existing literature [92], we select lagging CEO overseas experience in the period as an instrumental variable to address the original model’s endogeneity issue. Given the possibility of heteroscedasticity or autocorrelation cases of randomly perturbed items in panel data, this paper uses the more efficient two-stage generalized moment estimation (2-stage GMM) method.

Table 4. Endogenous problem processing: two-stage GMM instrumental variable regression.

Variable		Moderating Effect
	Model (1)	Model (2)	Model (3)	Model (4)
	Main Effect	Market Competition	Slack Resources	Full Model
COE	2.896 ***	2.995 *	5.817 ***	5.667 **
	(0.901)	(1.541)	(2.113)	(2.547)
MCI		0.408 *		0.275
		(1.331)		(1.205)
COE × MCI		1.470 ***		1.452 **
		(1.897)		(1.850)
SLACK			0.772 **	0.786 **
			(0.384)	(0.384)
COE × SLACK			2.300 **	1.997 **
			(1.324)	(1.304)
Constant term	–62.185 ***	–62.185 ***	–63.326 ***	–63.367 ***
	(4.373)	(4.429)	(4.391)	(4.452)
CN	Yes	Yes	Yes	Yes
YR	Yes	Yes	Yes	Yes
IT	Yes	Yes	Yes	Yes
N	2417	2417	2417	2417
R2	0.226	0.226	0.226	0.226

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Driscoll–Kraay standard errors in parentheses; (3) Because of limitations on space, the control variables are included in each model, and the results are not listed.

shows the results of the endogeneity treatment. The main and moderating effects are consistent with the results in Table 3 and remain robust in the subsequent full model. Therefore, the findings of this paper remain reliable and robust after controlling for the endogeneity problem.

4.3.2. Other Robustness Tests

First, the sub-sample tests market competition intensity. According to Luo and Zheng’s study [68], on grouping regression according to market competition intensity (divided by the mean value of this variable, greater than the mean value is 1; otherwise, it is 0) to carry out a robustness test, it is found that the main effect coefficient is larger than the mean value of market competition intensity. Second, the proxy variables test slack resources. Drawing on the existing research literature [75], given that organizational slack has a lagged effect on the relationship between CEO overseas experience and CSR, its lagged first order is used for testing. The results also indicate that organizational slack has a positive moderating effect on the relationship. Third, the regression tests different periods. Considering the possible errors from the sample period selection, we also adopt different year sample combinations such as 2012–2018, 2009–2016, and 2009–2017 for testing. The test results show that the main effects and moderating variables are consistent, with no significant differences. Finally, the model was changed for retesting. We select a random effects panel model to test the hypothesis, and the results show that the main effect and moderating effect remain consistent.

In summary, the regression results are not substantially different from the previous findings when retested according to the above method, which indicates that the findings of this paper are reliable and robust.

4.4. Further Analysis

4.4.1. Comparative Analysis of the Effects of Different Overseas Experiences on CSR

Different overseas experiences may expose CEOs to different resources and environments, with different effects on CEO values. Therefore, to further analyze whether there are differences in the effects of different overseas experiences on CSR, CEO overseas experiences are distinguished into CEO overseas work experience (COEJOB) and CEO overseas study experience (COEEDU), and the results of their categorical regressions are compared and analyzed. The regression results are shown in

Table 5. Comparative analysis of the impact of CEO overseas work and study experience on CSR.

Variable	Model (1)	Model (2)	Model (3)
	Overseas Experience	Overseas Work Experience	Overseas Study Experience
COE	1.113 **
	(0.363)
COEJOB		0.942 *
		(0.440)
COEEDU			1.291 **
			(0.441)
Constant term	2.138	2.223	1.164
	(8.698)	(8.820)	(8.394)
CN	Yes	Yes	Yes
YR	Yes	Yes	Yes
IT	Yes	Yes	Yes
N	3009	3009	3009
R2	0.442	0.442	0.442

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Driscoll–Kraay standard errors in parentheses; (3) Because of limitations on space, the control variables are included in each model, and the results are not listed.

.

The regression results show that the regression coefficient and significance model (2) of model (2) (Beta = 0.942, p < 0.1) are significantly smaller than those of model (3) (Beta = 1.291, p < 0.05). This suggests that the positive contribution of CEO overseas study experience to CSR is stronger than CEO overseas work experience. This may be explained by the fact that CEOs will be exposed to more social responsibility concepts, absorb more cutting-edge relevant theoretical knowledge in Western developed countries when studying abroad, and pay more attention to formulating and implementing CSR behaviors. Slater and Dixon-Fowler’s study [47] shows that CEO international assignment experience enhances CSR, but we further find that the positive effect of overseas study experience on CSR is stronger than overseas work experience.

4.4.2. The Effect of Geographical Regions on the Relationship between CEO Overseas Experience and CSR

The degree of social responsibility development in different geographical regions also varies significantly, and the education or experience that CEOs gain in countries with different development levels may also be different. Therefore, to further analyze whether there is a difference in the effect of overseas experience gained by CEOs in different geographical regions on CSR, the full sample is divided into two cases of group regressions according to geographical regions: Europe and the United States countries group (EuropUS) and the non-Europe and United States countries group (Non-EuropUS). This is because, compared with emerging countries, developed countries such as Europe and the United States have a more standardized and mature education, legal and institutional system, as well as business environment and practice methods, etc. The concept of social responsibility is deeply rooted in people’s hearts, and the atmosphere of social responsibility fulfillment is strong.

The regression results are shown in

Table 6. The effect of geographical regions on the relationship between CEO overseas experience and CSR.

Variable	Model (1)	Model (2)	Model (3)
	Full Sample	EuropUS Group	Non-EuropUS Group
COE	1.113 **
	(0.363)
COEEUS		1.248 ***
		(0.355)
COENOEUS			1.273
			(1.104)
Constant term	2.138	1.654	2.127
	(8.698)	(8.537)	(8.767)
CN	Yes	Yes	Yes
YR	Yes	Yes	Yes
IT	Yes	Yes	Yes
N	3009	3009	3009
R2	0.442	0.442	0.442

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Driscoll–Kraay standard errors in parentheses; (3) Because of limitations on space, the control variables are included in each model, and the results are not listed.

. Model (1) is for the EuropUS group, which has a significant positive relationship between CEO overseas experience and CSR (COE: Beta = 1.248, p < 0.01), and the regression coefficient is greater than that of the full sample and significant at the 1% level. In contrast, model (2) shows a positive non-significant relationship between COE and CSR in the Non-EuropUS group (Beta = 1.273, p > 0.1). This suggests that CEOs with overseas experience in developed regions such as Europe and the United States have a stronger effect on CSR. This is consistent with the studies of Wen and Song [56] and Zhang et al. [58] on returnee managers or directors. This may be explained by the fact that developed regions such as Europe and the United States have developed earlier in social responsibility and have initially established a relatively complete social responsibility education system and a comprehensive social responsibility guarantee system. Social responsibility practice has been quite mature and has very rich and advanced experience in social responsibility practice.

4.4.3. The Effect of Different Ownership Structure on the Relationship between CEO Overseas Experience and CSR

Enterprises with different ownership properties have different organizational institutional environments and strategic planning goals, and their CEOs also show differences in CSR decision-making behavior. Therefore, to test the difference in the effect of COE on the formulation and fulfillment of CSR in companies with different ownership properties, this paper distinguishes the full sample into SOE group and non-SOE group, and regression is conducted in groups.

Model (1) in

Table 7. The effect of ownership structure on the relationship between CEO overseas experience and CSR.

Variable	Model (1)	Model (2)	Model (3)
	Full Sample	State-Owned Enterprise Group	Non-State Enterprise Group
COE	1.113 **	1.400 ***	0.955
	(0.363)	(0.268)	(0.572)
Constant term	2.138	–5.999	16.263 **
	(8.698)	(10.590)	(5.447)
CN	Yes	Yes	Yes
YR	Yes	Yes	Yes
IT	Yes	Yes	Yes
N	3009	1732	1277
R2	0.442	0.433	0.455

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Driscoll–Kraay standard errors in parentheses; (3) Because of limitations on space, the control variables are included in each model, and the results are not listed.

is a full sample, while models (2) and (3) are the effects of the relationship between COE and CSR in the state-owned enterprise group and non-state-owned enterprise, respectively. The regression results reported are all fixed effects. The regression results show that COE in the full sample group exerts a positive and significant effect on CSR. Compared with the non-state-owned enterprise group, the CEO overseas experience (COE) has a significant positive effect on CSR in the state-owned enterprise group (Beta = 1.400, p < 0.01).

This suggests that for enterprises with state ownership, CEO overseas experience has a more significant positive impact on CSR. This may be explained by the fact that compared with non-state-owned enterprises, state-owned holding enterprises have clearer medium and long-term strategic planning and development performance goals and are more inclined to focus on the interests of the public, shareholders, and other stakeholders to achieve long-term sustainable development of the enterprise. Related studies by Sufian and Zahan [93] and Kilic et al. [94] show that the ownership structure has a positive effect on CSR. However, we further find differences in firms’ social responsibility performance with different ownership structures, and this positive effect mainly occurs in state-owned enterprises.

4.4.4. The Effect of CEO Concurrent Appointment on the Relationship between CEO Overseas Experience and CSR

As the TMT leader, the CEO sometimes also serves as the chairman of the board of directors. Whether or not the CEO is concurrently appointed may affect the relationship between COE and CSR. Therefore, to further analyze whether there is a difference in the effect of CEO overseas experience on CSR in the case of CEO concurrent appointment, group regressions were conducted by dividing the full sample into the following two cases: CEO concurrent chairmanship and CEO non-concurrent chairmanship.

The regression results are shown in

Table 8. The effect of CEO concurrent appointment on the relationship between CEO overseas experience and CSR.

Variable	Model (1)	Model (2)	Model (3)
	Full Sample	CEO’s Concurrent Chairmanship Group	CEO’s Non-Concurrent Chairmanship Group
COE	1.113 **	2.417	1.285 ***
	(0.363)	(1.395)	(0.289)
Constant term	2.138	–22.246	3.578
	(8.698)	(15.529)	(6.449)
CN	Yes	Yes	Yes
YR	Yes	Yes	Yes
IT	Yes	Yes	Yes
N	3009	594	2415
R2	0.442	0.439	0.440

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Driscoll–Kraay standard errors in parentheses; (3) Because of limitations on space, the control variables are included in each model, and the results are not listed.

. Under Model (2), the CEO concurrently serves as the chairman of the board, and CEO overseas experience has a positive and insignificant relationship with CSR (COE: Beta = 2.417, p > 0.1). Model (3) is the group of CEOs who are not concurrently appointed as chairmen, and CEO overseas experience has a significant and positive relationship with CSR (COE: Beta = 1.285, p < 0.01) that is significant at the 1% level. This suggests that the CEO’s concurrent appointment as chairman has a weakening effect on the positive relationship between COE and CSR. Most of the existing studies [49,50,51] have found that CEO concurrent appointment has a negative effect on CSR performance from the perspective of corporate governance mechanisms or practices, and the findings of this paper have been confirmed and supported.

This may be explained by the fact that according to the principal–agent theory, the chairman and the CEO are part of the principal–agent relationship of supervising and being supervised [95]. If the CEO is also the chairman of the board, it may make it difficult for the board of directors to achieve effective supervision of the CEO, which may negatively affect corporate decision making [96]. In other words, the separation of the CEO and the chairman may ensure the effectiveness and independence of the board of directors’ supervisory function [97]. It does not only achieve effective control over the CEO’s power, but can also reduce the CEO’s risk of only focusing on short-term goals.

4.4.5. The Effect of the Overseas Experience of CEOs of Different Genders on CSR

Executive gender characteristics play an important role in the effective realization of corporate governance and may be a significant influencing factor of CSR. Therefore, to further analyze whether there is a difference in the effect of the overseas experience of CEOs of different genders on CSR, COE is distinguished into the overseas experience of male CEOs and overseas experience of female CEOs according to the gender of the CEOs, and the results of the categorical regression of the two are compared and analyzed. The regression results are shown in

Table 9. The effect of gender on the relationship between CEO overseas experience and CSR.

Variable	Model (1)	Model (2)	Model (3)
	Full Sample	CEO for Male Group	CEO for Women Group
COE	1.113 **	0.987 *	1.856 ***
	(0.363)	(0.466)	(0.483)
Constant term	2.138	2.799	65.597 **
	(8.698)	(9.543)	(28.109)
CN	Yes	Yes	Yes
YR	Yes	Yes	Yes
IT	Yes	Yes	Yes
N	3009	2825	184
R2	0.442	0.320	0.364

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Driscoll–Kraay standard errors in parentheses; (3) Because of limitations on space, the control variables are included in each model, and the results are not listed.

.

The regression results show that the regression coefficient and significance of the model (2) (Beta = 0.987, p < 0.1) are significantly smaller than those of model (3) (Beta = 1.856, p < 0.01). This suggests that the overseas study experience of female CEOs has a stronger positive effect on CSR decision-making than that of male CEOs. This is consistent with the findings of Bernardi et al. [23]. Zhuang’s study [57] suggests that gender difference in the board of directors significantly affects CSR, but does not explore the impact of different genders on CSR in depth.

This may be explained by the fact that, from an ethical perspective, women pay more attention to ethics and morality than men at the attitudinal and behavioral levels [23,98,99]. Women’s care ethics, on the other hand, emphasize the ethical behavioral characteristics of women that are different from those of men, arguing that women are generally more morally and ethically conscious than men, have significantly different ethical event evaluation criteria than men [23,100], and place more emphasis on fulfilling CSR.

4.4.6. The Effect of the Overseas Experience of CEOs with Different Educational Levels on CSR

The education level represents the level of cultural education a person has received [28]. The overseas experience of CEOs with different education levels may directly affect CSR performance. Therefore, to further analyze whether there is a difference in the effect of the overseas experience of CEOs with different education levels on CSR, group regression is conducted according to two scenarios of CEO education level—master’s degree or above and below master’s degree. This is because a master’s degree or above is a higher education level training and focuses more on comprehensive ability, while below master’s degree (i.e., bachelor’s degree and below) focus on developing basic abilities. The regression results are shown in

Table 10. The effect of educational level on the relationship between CEO overseas experience and CSR.

Variable	Model (1)	Model (2)	Model (3)
	Full Sample	Master’s Degree or Above Group	Below Master’s Degree Group
COE	1.113 **	1.334 ***	0.281
	(0.363)	(0.383)	(0.205)
Constant term	2.138	–10.471	–1.728
	(8.698)	(6.198)	(10.753)
CN	Yes	Yes	Yes
YR	Yes	Yes	Yes
IT	Yes	Yes	Yes
N	3009	1770	1239
R2	0.442	0.438	0.403

Note: (1) *** p < 0.01, ** p < 0.05, * p < 0.1; (2) Driscoll–Kraay standard errors in parentheses; (3) Because of limitations on space, the control variables are included in each model, and the results are not listed.

. Model (2) is the master’s degree or above group, and the CEO overseas experience is significantly and positively related to CSR (COE: Beta = 1.334, p < 0.01). The regression coefficient is greater than that for the full sample and is significant at the 1% level.

This suggests that the overseas experience of CEOs with high educational attainment has a stronger impact on CSR. Cacioppe et al.’s study [24] shows that CSR performance receives extra attention from senior managers with higher education levels. Manner’s study [25] also finds that the CEO’s personal educational experience positively affects CSR performance. This may be explained by the fact that highly educated CEOs receive high-level theoretical knowledge and comprehensive ability training, master more cutting-edge professional knowledge, form higher cognitive views and values, have a clearer understanding of enterprise strategy and development ideas, and are more capable of formulating and implementing CSR behaviors.

5. Conclusions and Discussion

We examined the relationship between CEO overseas experience and CSR and analyzed the moderating effects of market competition intensity and slack resources on the relationship between the two. We further explored the effects of different overseas experiences, ownership structure, CEO concurrent appointment, gender, and educational level. Based on the sample data of all listed companies in the Chinese manufacturing sector from 2009 to 2018, the following findings are obtained: (1) There is a significant positive relationship between CEO overseas experience and CSR, i.e., CEO overseas experience can enhance CSR performance. (2) Market competition intensity and organizational slack resources have a strengthening effect on the positive relationship between COE and CSR. (3) CEOs with overseas study experience had a stronger positive contribution to CSR than those with overseas work experience; the effect of CEO overseas experience on CSR is stronger when the CEO gains overseas experience in developed regions such as Europe and the United States; the positive effect of COE on CSR is more significant in the state-owned enterprise group; CEO concurrent appointment has a weakening effect on the positive relationship between COE and CSR; the positive effect of female COE on CSR is stronger; and CEO overseas experience has a stronger effect on CSR when the CEO is highly educated.

This paper’s contributions based on the research conclusions are as follows: First, we contribute to the literature on the impact of CEO characteristics on corporate strategic decision making. In particular, we provide empirical evidence for exploring how CEOs’ past overseas experience can help improve and enhance CSR performance. Second, this paper incorporates CEO overseas experience into the CSR research framework, which is of great significance for further expanding the upper echelons theory. In contrast to previous studies that mainly explored CEO’s characteristics and organization-related attributes, our study examines the role of CEOs’ overseas experiences, especially from developed countries such as Europe and the United States, on corporate decision making and elucidates the drivers of CSR. Third, based on the local cultural context in China, we focus on the contextual mechanism involved in the impact of individual CEO characteristics on CSR in the new Chinese economy. The intrinsic mechanisms of action that influence the relationship between CEO overseas experience and CSR are explored from the external environment’s organizational perspectives (market competition intensity) and the internal organizational environment (slack resources).

Based on this paper’s findings, the following managerial practice implications are obtained: (1) The decision-making efficiency of executives is influenced by the cognitive structure and knowledge of individual members, which are formed by their early life experiences closely related to their past work or study experience. Enterprises should enrich their executives’ knowledge structure to provide new knowledge and cognitive perspectives in making strategic decisions to enhance CSR performance goals. (2) Activities related to the implementation of corporate performance goals should not only consider the impact of the CEO’s overseas experience on CSR but also examine the characteristics of the internal and external environment of the organization. Only by making strategic decisions suitable for the enterprise itself and actively fulfilling CSR according to the internal and external market environment conditions can we ensure that strategic goals are met. (3) Enterprises should make full use of organizational slack resources to actively fulfill CSR, which can help advance their strategic choice and decision making. CEO overseas study experience yields more extensive knowledge reserves and decision-making cognition than overseas work experience. Enterprises should cultivate and introduce highly educated talents, which can help improve CSR performance and positively affect the achievement of CSR performance goals. (4) Our findings have important implications not only for Chinese firms but also for firms in other new economies with low CSR levels. Promoting CEOs with overseas experience can effectively enhance CSR in emerging markets, which has an essential role in attracting stakeholders’ investment.