Can more able CEOs reduce the gap between internal and external CSR disclosures?

ABSTRACT

This study examines the association between chief executive officer (CEO) ability and the gap between internal and external corporate social responsibility (CSR) disclosure. We find that a firm’s CSR disclosure gap decreases when its CEO's ability increases. We also find that this negative association is strengthened when a CEO has political connections and when a CEO is internally promoted (inside CEO). Our results are robust after controlling for firm fixed effect and addressing endogeneity concerns. Overall our findings are consistent with our argument that more able CEOs significantly reduce the CSR disclosure gap and maintain the positive relationship between internal and external stakeholders. These results carry substantial implications for both theory and practice. From a practical standpoint, our findings emphasize the pivotal role of capable CEOs in orchestrating consistent CSR narratives that resonate internally and externally. This study offers valuable insights for corporate leaders striving to enhance their firms' CSR transparency and maintain favourable stakeholder relationships.

¿Pueden los consejeros delegados más capaces reducir la brecha entre la información interna y externa sobre RSE?

RESUMEN

Este estudio examina la relación entre la capacidad del consejero delegado y la brecha entre la divulgación interna y externa de la responsabilidad social corporativa (RSC). Se observa que la brecha de divulgación de la RSE de una empresa disminuye cuando aumenta la capacidad de su consejero delegado. También observamos que esta asociación negativa se refuerza cuando el CEO tiene conexiones políticas y cuando es promovido internamente (inside CEO). Nuestros resultados son robustos tras controlar el efecto fijo de la empresa y abordar los problemas de endogeneidad. En general, nuestros resultados son coherentes con nuestro argumento de que los CEO más capaces reducen significativamente la brecha de divulgación de la RSC y mantienen la relación positiva entre las partes interesadas internas y externas. Estos resultados tienen importantes implicaciones teóricas y prácticas. Desde un punto de vista práctico, nuestras conclusiones subrayan el papel fundamental que desempeñan los directores generales capaces de orquestar narrativas de RSC coherentes que resuenen interna y externamente. Este estudio ofrece valiosas ideas a los directivos de empresas que se esfuerzan por mejorar la transparencia de la RSE de sus empresas y mantener relaciones favorables con las partes interesadas.

1. Introduction

Over the recent years, global companies have heightened their corporate social responsibility (CSR) disclosures in reaction to mounting pressures from various stakeholders (Campbell, 2007; Jin et al., 2023; Wang et al., 2016). These CSR disclosures commonly encompass both internal aspects (like employee policies and diversity) and external facets (such as community involvement, product innovation, and safety and environmental reporting) (Jin et al., 2023; Jin et al., 2022). Lately, scholars have been increasingly attentive to both internal and external CSR disclosures (Al-Shammari et al., 2019; Cruz et al., 2014; Ge & Zhao, 2017). Researchers are now more cognizant of the notable incongruities between internal and external CSR disclosures, with substantial variances among companies: while some prioritize external disclosures, others take the opposite approach (Gosselt et al., 2019; Hawn & Ioannou, 2016). Consequently, this gives rise to a disclosure gap: the disparity between internal and external revelations.

A larger CSR disclosure gap can be punished by the capital market (García-Sánchez et al., 2021), thus affecting the company's market value (Hawn & Ioannou, 2016), the cost of capital (Bastida et al., 2014), and the difficulty in getting finance (García-Sánchez et al., 2021). Therefore, considering the negative impact of the CSR disclosure gap, it is important to study factors that influence the CSR disclosure gap. However, there is scant evidence about when a company’s CSR disclosure gap increases or decreases. Recently, researchers have been exploring the role of top managers' characteristics that impact CSR internal and external disclosures (Al-Shammari et al., 2019; Jin et al., 2023; Jin et al., 2022). Because top managers are considered to influence the companies’ strategic decisions that determine the continuity of companies in the market (Hambrick & Mason, 1984), they may also influence the companies’ CSR disclosure gap (Al-Shammari et al., 2019). This study focuses on one important characteristic of the CEO: CEO ability and asks the following question: Can more able CEOs reduce the gap between internal and external CSR disclosures?

Previous studies have shown that some factors significantly affect CSR internal and external disclosure and the disclosure gap, for example, CEO narcissism (Al-Shammari et al., 2019), independent directors' characteristics such as nationality and political affiliation (Jin et al., 2022) and female independent directors (Jin et al., 2023). We study the role of CEO ability on the CSR disclosure gap. The classic shareholder view shows that a company’s primary goal is to achieve growth and economic benefits; however, the growth of social and environmental performance points to the agency problem of the company (Friedman, 1970). Thus, the CEO has to balance corporate social performance and corporate financial outcomes (García-Sánchez et al., 2019).

In an era where corporate social responsibility (CSR) disclosures wield increasing influence, this study delves into a captivating dimension: the pivotal role of CEO characteristics in shaping the alignment between internal and external CSR disclosures. By exploring how CEO ability, a central but often overlooked factor, influences this alignment, our research addresses a critical gap in the existing literature.

The study builds on the recognition that CSR disclosures serve as a conduit for firms to exhibit their commitment to ethical practices and stakeholder engagement. Yet, despite the burgeoning importance of CSR in modern business landscapes, the intricate relationship between CEO attributes and the consistency of these disclosures remains underexplored. To bridge this gap, we investigate whether CEOs with distinct abilities can facilitate a more cohesive CSR narrative that resonates both internally and externally.

A CEO’s ability is directly related to her career concerns (Yuan et al., 2019). Whereas a manager's decision-making, including investment decisions, largely depends on her career concerns (Moriss, 1998). CEOs with higher managerial ability pay less attention to their careers concern because most of them get a good evaluation in the labour market, and get a good image for effectively operating the company (Ali & Zhang, 2015; Doukas & Zhang, 2021).

The method employed to measure CEO ability in this study is adapted from previous research, particularly following the approach established by Yuan et al. (2019) for Chinese firms. This method builds on the framework introduced by Demerjian et al. (2012), which has been widely adopted in various studies such as Baik et al. (2011), Demerjian et al. (2013), Krishnan et al. (2015), and Koester et al. (2017). The estimation process is openly available online for reference.

The process of estimating CEO ability involves a two-stage procedure, based on the approach proposed by Demerjian et al. (2012). In the initial stage, data envelopment analysis (DEA) is utilized to gauge a firm's efficiency. Subsequently, in the second stage, a regression model is employed to analyze the relationship between the firm's efficiency and several firm-level determinants. The difference between the actual efficiency and the predicted efficiency from the regression model yields the residual, which is considered the CEO's contribution to the firm's overall efficiency.

In the context of this study, this methodology is adapted to calculate CEO ability using data from Chinese public firms. Given the assertion that Corporate Social Responsibility (CSR) disclosure could impact a firm's sales, the CSR score is introduced in the second stage of estimation. By incorporating the firm's CSR score into the second-stage model, the potential influence of CSR disclosure on firm efficiency is accounted for. Following the example set by Yuan et al. (2019), the firm's CSR score is included in the second-stage estimation model, and the resulting residual term is interpreted as the CEO's ability.

In essence, this approach allows for the quantification of CEO ability by considering the CEO's impact on firm efficiency beyond the effects of CSR disclosure. The utilization of a well-established methodology and its adaptation to incorporate relevant factors such as CSR scores ensures a robust measurement of CEO ability in the context of this study.

If a CEO pays close attention to her labour market assessments, he or she is more likely to abandon investing in long-term projects as their returns are uncertain (He & Tian, 2016; Narayanan, 1985). Similar to this argument, Graham et al. (2005) survey 401 top managers and find that most managers agree that they are very careful about their labour market assessment because any failure to meet short-term goals would be their failure in the labour market. To avoid a poor reputation in the labor market, they are ready to cut investment in long-term projects to achieve short-term goals. CSR disclosure is a long-term initiative and affects the firms' financial outcomes (Lee & Choi, 2021). Thus, CEOs with high ability have better career prospects and longer career visions than CEOs with low ability. In addition, they have more incentives to reduce the gap between internal and external CSR disclosures because this can reduce the companies’ financial risk (Lee & Choi, 2021). Furthermore, more able CEOs have more skills to deal with uncertainty related to internal and external disclosures.

We choose China as the research setting because China is the second-largest economy in the world, and CSR disclosure is a new phenomenon in Chinese listed companies. Second, social and governance law is underdeveloped compared to the western world, so it provides us with the opportunity to explore how companies can meet the needs of multiple stakeholders. Therefore, we select Chinese companies listed in the Shanghai and Shenzhen stock exchanges from 2010 to 2020 as our research sample. We use the ordinary least squares regression model as the baseline model to explore the impact of CEO ability on the CSR disclosure gap and find that CEO ability is significantly negatively associated with the CSR disclosure gap. The results show that companies with more able CEOs keep the balance between internal and external CSR disclosures to meet the requirements of multiple stakeholders. In further analysis, we find that this relationship is more pronounced when the CEO has a political connection and when the firm is a state-owned enterprise (SOE).

This study adds to the literature in the following respects. First, this study helps to better understand the causes of the CSR disclosure gap. Consistent with the upper echelons theory, our study finds that more able CEOs are more likely to reduce the gap between internal and external CSR disclosures. Second, our study reveals the heterogeneity of CEO ability in general.

The following section presents the literature review, theoretical backgrounds, and hypotheses. Section III presents the data and the methodology. Section IV reports empirical results and additional tests. Section V concludes the study.

2. Background, theoretical framework, and hypothesis

2.1. Background of the CSR disclosure gap

An increasing body of academic research indicates the misalignment of companies’ internal and external CSR disclosures because companies often engage in internal and external CSR disclosures in many different ways and to different extents (Hawn & Ioannou, 2016). Companies’ internal and external CSR disclosures ideally should be closely aligned with no gaps (García-Sánchez et al., 2020). In reality, companies prioritize one activity above another, focusing on the most important ones (Chen et al., 2021; Luo & Wang, 2021), and the decisions made may result in a gap between internal and external CSR disclosures (Hawn & Ioannou, 2016). Recent research documents that companies need to reduce the gap between internal and external CSR disclosures because such misalignment may result in negative market reactions (García-Sánchez et al., 2020).

In prior literature, terms like green-washing, decoupling, walking the talk, and minding the gap have been used to describe the gap between internal and external CSR disclosures (Wickert et al., 2016). Previous studies find factors that influence the gap between internal and external disclosure. For example, Font et al. (2012) find that larger hotels have more comprehensive policies about CSR disclosure but in general, they have more gaps in the implementation of these policies. Wickert et al. (2016) find that larger companies are more likely to emphasize the disclosure of external CSR to gain organizational legitimacy rather than internal implementation as compared to smaller firms. Cruz et al. (2014) show that a large CSR disclosure gap exists in European-listed family companies and that family companies are more likely to disclose external CSR because of social norms. Ge & Zhao (2017) based on the national survey data set of 1268 Chinese listed companies find that companies with more bureaucratic connections with the state are more likely to focus on external CSR disclosure, while companies with a closer partnership with the state through political connections are more likely to engage in internal CSR disclosure.

Prior studies also suggest that top executives also affect the CSR disclosure gap. For example, Jin et al. (2023) find that gender diversity in terms of more female independent directors on the board is more likely to engage in internal CSR disclosure rather than external CSR disclosure. They also find that politically connected directors in Chinese companies are associated with a wider gap between internal and external CSR disclosures and this gap is narrower when companies have foreign directors on board. Al-Shammari et al. (2019) find that narcissistic CEOs pay more attention to external CSR disclosure than to internal CSR disclosure. Sauerwald & Su (2019) find that overconfident CEOs are more likely to create a gap between the optimistic tone of CSR reports and the actual CSR performance. There is scant evidence on how CEO ability accounts for the misalignment between a firm’s internal and external CSR disclosures.

2.2. Theoretical framework

We use several theories to explain the relationship between CEO ability and the CSR disclosure gap. For instance, a recent study by Yosra and Jihene (2021) based on the legitimacy theory, the impression management theory, and the upper echelons theory explores the relationship between CEO characteristics and the readability of CSR reports. Muttakin et al. (2018) based on multiple theories such as the agency theory and the resource dependency theory examine the association between CEO power and CSR disclosure. In response to the recent trend of using multiple theories to explain corporate governance's effect on CSR (Elmagrhi et al., 2016; Nguyen et al., 2021), we add to the scant literature by exploring the relationship between CEO ability and the CSR disclosure gap. More specifically, we use the upper echelons theory, and the stakeholder theory to explain the above-mentioned relationship.

According to the stakeholder theory, it is a strategic approach to managing trade-offs of multiple stakeholders. Stakeholders often differ in importance, proximity, and visibility to companies, prompting companies to implement a range of CSR approaches in response (Al-Shammari et al., 2019). Internal CSR disclosure refers to a company's management practice for its employees, such as employee training, fair pay, diversity policies, and fair treatment of ethnic minorities (Al-Shammari et al., 2019; Farooq et al., 2017). External disclosure of CSR refers to a company's procedures involving customer interaction, philanthropic work, environmental performance, product safety, and community engagement, which are centred on stewardship toward customers, local communities, or the natural environment (Al-Shammari et al., 2019; Farooq et al., 2017). We believe that the more able CEOs understand the requirements of multiple internal and external stakeholders, and therefore, they are more likely to reduce the gap between internal and external CSR disclosures.

The upper echelon theory suggests that a company’s CEO is an important part of the strategic management process, and the different characteristics of the CEO determine the company's financial and non-financial outcomes (Hambrick, 2007; Hambrick & Mason, 1984). The CEO is regarded as the most important person having the opportunity to choose the optimal and best strategy to allocate a company’s resources to satisfy multiple stakeholders' demands. Prior studies based on the upper echelons theory have documented that the CEO’s tenure, age, experience, and compensation, are key determinants of CSR disclosure (Ali et al., 2020; Chen et al., 2019; Hambrick, 2007; Mahmoudian et al., 2021; Oh et al., 2016). The CEO as the most influential manager in the company has the potential to perceive and adjust the company's strategic goals and objectives through understanding multiple stakeholders' demands. Therefore, we expect that more able CEOs are more likely to reduce the CSR disclosure gap.

2.3. Agency theory

The agency theory focuses on the principal-agent relationship within a corporation, where shareholders (principals) delegate authority to managers (agents) to make decisions on their behalf (Jensen & Meckling, 1976). This delegation creates a potential conflict of interest, as managers might prioritize their interests over those of the shareholders.

Within the context of corporate disclosure, the agency theory provides insights into how CEO characteristics influence the extent and quality of information disclosed to shareholders and other stakeholders. CEOs, as agents, have access to information that shareholders do not, creating information asymmetry. Certain CEO characteristics can affect their incentives and behaviour, which in turn can impact the firm's disclosure decisions.

For instance, CEO ability, as a key CEO characteristic, can significantly influence disclosure practices. CEOs with higher abilities might be more adept at managing the firm's resources, effectively communicating its prospects, and making strategic decisions that enhance firm value. Such CEOs are more likely to be aligned with shareholder interests and motivated to provide accurate and comprehensive information to reduce information asymmetry.

2.4. CEO Ability

We follow the method used by Yuan et al. (2019) to measure the CEO ability of Chinese firms. Their method is derived from Demerjian et al. (2012) which has been used in many studies (e.g., Baik et al., 2011; Demerjian et al., 2013; Krishnan et al., 2015; Koester et al., 2017). The estimation procedure is publicly available online. Demerjian et al. (2012) use a two-stage approach to compute managerial ability. In the first stage, they use data envelopment analysis (DEA) to estimate a firm’s efficiency. In the second stage, they regress a firm’s efficiency on firm-level determinants. The residual from the second stage is considered the manager’s contribution to a firm’s efficiency.

Previous studies show that a company’s performance is a reflection of the CEO's ability (Baik et al., 2011; Doukas & Zhang, 2020; Koester et al., 2017). For instance, Bertrand & Schoar (2003) find that CEOs with MBA degrees have more capital expenditures. Chemmanur & Paeglis (2005) find that organizations with high-ability CEOs perform better in initial public offerings (IPOs), showing that more able CEOs can credibly communicate the intrinsic values of the organization to outsiders. Koester et al. (2017) show CEOs with high abilities can better manage companies' resources to avoid paying taxes.

While previous studies focus on the relationship between CEO ability and company performance, they also find that the greater the CEO's ability is, the higher the firm’s output is (Ali et al., 2015; Doukas & Zhang, 2021). Further, a lower ability can lead to a lower assessment for the CEO in the labor market (Fama, 1980; Gibbons & Murphy, 1992; Weisbach, 1988). For instance, Fee & Hadlock (2003) propose that CEOs who generate higher stock returns are most likely to be hired by other companies with higher compensations. On one hand, Rajgopal et al. (2006) also find that CEOs’ external opportunities are positively related to their managerial expertise. On the other hand, the poor performance of a CEO can lead to dismissal and difficulty in reappointment (Denis & Denis, 1995; Wiersema, 2002). The above studies conclude that, compared with less capable CEOs, CEOs with higher management skills, are better prepared to engage with different stakeholders.

2.5. CEO and disclosure policy

Prior literature recognizes the importance of the role of CEOs in shaping disclosure policies. For example, the study conducted by Bochkay et al. (2019) offers a comprehensive exploration of the dynamic dimensions inherent in CEO disclosure styles. This research not only contributes to the evolution of our understanding of CEO characteristics but also serves to complement and augment the scope of our own investigation into the intricate relationship between CEO attributes and corporate disclosure strategies. By delving into the ever-changing dynamics that underlie how CEOs disseminate information to their stakeholders, this study provides a valuable vantage point from which we can glean further insights into the multifaceted interplay between CEO traits and the pivotal decisions surrounding disclosure practices. Their examination of how CEOs navigate the varying approaches to disclosure not only aligns with our research's focus but also enriches the discourse on the intricate mechanisms that drive the disclosure landscape.

In a similar vein, the study by Li et al. (2018) significantly broadens our understanding by introducing the concept of CEO power as a determinant of environmental, social, and governance (ESG) disclosure and its subsequent implications for firm value. By contemplating the influence of CEO power on disclosure practices, this research presents a compelling facet to the discussion of CEO attributes and their impact on corporate transparency. As we explore the multifaceted interplay between CEO characteristics and disclosure strategies, Li et al.'s examination of how CEO power affects ESG disclosure practices resonates strongly with our inquiry, offering a complementary perspective that unveils a broader canvas of factors shaping disclosure decisions and their outcomes.

The study conducted by Garcia-Sanchez et al. (2021) adds another layer of depth to our understanding by investigating the nexus between CEO power and integrated reporting. This exploration resonates closely with our study's objectives, as we also seek to unravel the intricate threads connecting CEO attributes to disclosure mechanisms. By examining the correlation between CEO power and integrated reporting practices, their work accentuates the multifaceted dimensions of CEO influence on corporate communication strategies. Their findings, when integrated with our research, provide a richer tapestry of insights into the intricate dynamics governing the relationship between CEO characteristics and how corporations convey information to their stakeholders.

2.6. Hypothesis Development

Prior literature documents that a CEO’s ability influences his/her decision-making (Doukas & Zhang, 2020). Prior studies also find that a CEO’s ability is related to his/her career (Ali et al., 2015). Further, Holmstrom (1982) finds that the career concern of managers is one of the reasons for the distortion of accounting numbers. CEOs who are concerned about their careers are less willing to make investments with the uncertainty of long-term repayment (Holmström, 1999).

By focusing on short-term-oriented projects that focus on quick returns, managers can demonstrate their ability to the labour market (Narayanan, 1985). Graham et al. (2005) show that managers believe that not meeting short-term goals is a managerial failure. Bebchuk & Stole (1993) also believe that these short-term-oriented behaviours of managers distort companies’ investment behaviours. Similarly, Porter (1992) finds that managers reduce investments in long-term projects like research and development, employee training, and product promotions, to meet short-term goals. He & Tian (2016) find that managers are facing tremendous pressure to achieve short-term goals.

Based on the above studies' findings, we predict that more able CEOs are more likely to reduce the CSR disclosure gap than less able CEOs. The intuition is that more able CEOs have fewer career concerns and are not much under pressure to just pursue short-term-oriented projects with rapid payoffs. In addition, more capable CEOs are more able to deal with uncertainties, and they are more likely to start projects that uphold the interests of multiple stakeholders. In summary, we believe that the more able CEOs are more likely to reduce the CSR disclosure gap. Therefore, we propose the following hypothesis:

H1: CEO ability is negatively associated with the CSR disclosure gap.

2.7. CEO political connection as a moderating factor

The political connection of Chinese CEOs plays an important role in shaping corporates’ behaviours (Lin et al., 2016). In China, informal political ties offset the drawbacks of institutional environments (Hoskisson et al., 2002; Schweizer et al., 2019). We expect that CEOs' political connection strengthens the relationship between CEO ability and the CSR disclosure gap for the below reasons.

First, CEOs with political ties have important social capital which helps them to get access to governmental support ( Xin & Pearce, 1996; Li et al., 2008), such as getting loans (Dinç, 2005) getting tax reliefs (Chen et al., 2011; Benito et al., 2014; Benito et al., 2019), and obtaining governmental subsidy (Hung et al., 2015). These incentives help more able CEOs to maintain the balance between the interests of multiple stakeholders. Second, in the competitive corporate world, the quality of information plays an important role. CEOs with more political connections have more incentives to disclose information that can satisfy multiple stakeholders. This study argues that CEOs’ political connections help them build deeper and broader relationships with multiple stakeholders. Social capital brought through political connections reduces uncertainty about policy risks, and strengthens the firms’ internal knowledge management. Finally, we argue that more able CEOs having political relationships with the government are more likely to engage with multiple stakeholders when disclosing CSR information. This is because politically connected CEOs under legitimacy, regulatory and normative pressure are more likely to follow ethical guidelines than non-political connected CEOs (Scott & Meyer, 1994). Therefore, politically connected CEOs are likely to follow governmental regulatory policies to reduce the CSR disclosure gap. Based on these arguments. We make the following hypothesis:

H2: CEO political connection strengthens the relationship between CEO ability and the CSR disclosure gap.

2.8. CEO’s firm-specific knowledge (CEO identity: outside CEO versus inside CEO) as a moderating factor

Next, we explore the moderating effect of a CEO’s firm-specific knowledge, using CEO identity as a proxy. CEO identity refers to whether the CEO is promoted from the inside (inside CEO) or hired from the outside (outside CEO). A CEO promoted from the inside has more knowledge than a CEO hired from the outside. An inside CEO is more familiar with a firm’s stakeholders than an outside CEO. Therefore, we argue that CEOs with more firm-specific knowledge can reduce the CSR disclosure gap. We make the following prediction:

H3: The CEO's firm-specific knowledge strengthens the relationship between CEO ability and the CSR disclosure gap.

3. Methodology

3.1. Sample

This study uses a sample of Chinese public companies listed on the Shenzhen and Shanghai stock exchanges for the period 2010-2020. Our data sample starts from 2010 because most firms’ CSR disclosure information is available starting from 2010 in the Chinese Corporate Social Responsibility (CCSR) database, which provides comprehensive information about Chinese firms’ CSR information. We also use the China Stock Market and Accounting Research (CSMAR) database which provides comprehensive details about listed companies' corporate governance and financial variables.

3.2. Dependent variable: CSR Gap

CSR Gap (CSR disclosure gap) is our dependent variable, which measures the gap between internal and external CSR disclosures. Following prior studies (Al-Shammari et al., 2019; Farooq et al., 2017), we derive five areas of CSR disclosure from the CCSR database and further categorize them into internal disclosure (i.e., employee relations and diversity) and external disclosure (i.e., environment, product, and community). In Table 1, we present the benefits and concerns of each disclosure. The CCSR database provides the details of the five areas which include 48 items of binary nature. For example, if a company discloses that it provides ownership to an employee, the “employee stock ownership” indicator is then assigned 1 and 0 otherwise. Similarly, if the company discloses that it has downsized, the “layoffs” are then assigned 1 and 0 otherwise. First, we measure each area’s value using the difference between the advantages summed scores and concerns summed scores (Hull & Rothenberg, 2008; Tang et al., 2012). Second, we sum up the scores of internal categories (employee and diversity) to measure the internal CSR disclosure score (Al-Shammari et al., 2019). Following the same pattern, we sum up the scores of the three external categories (environmental, product, and community) to measure the external CSR disclosure score (Cruz et al., 2014; Farooq et al., 2017). Finally, following prior studies (Farooq et al., 2017; García-Sánchez et al., 2020; Jin et al., 2022), we obtain the value of the CSR disclosure gap by taking the difference between the external and internal CSR disclosure scores (external score-internal score). The larger the misalignment between a company’s internal and external CSR disclosures is, the higher the value of the CSR disclosure gap will be.

Table 1. CSR Disclosure gap

CSR Category	Type	CSR Dimensions
(1) Employees (internal disclosure)	Benefits	Employee stock ownership
		Employee benefits
		Safety management system
		Safety training
		Occupational health and safety
		Job training
		Employee communication
		Other advantages
	Concerns	Security issues
		Layoffs
		Other concerns
(2) Diversity (internal disclosure)	Benefits	Female executives
		Female representation board
		Employment of the disabled and ex-prisoners
		Party members
		Other advantages
	Concerns	Women’s non-representation
		Other concerns
(3) Environment (external disclosure)	Benefits	Environmental-friendly products
		Waste reduction measures
		Clean energy
		Energy saving
		Green office program
		ISO 14001
		Environmental award
		Other advantages
	Concerns	Environmental penalty
		Pollutant emission
		Other concerns
(4) Product (external disclosure)	Benefits	Product quality management
		After-sales service
		Customer satisfaction
		Quality award
		Anti-corruption measures
		Resource-sharing policies
		Business philosophy for integrity
		Other advantages
	Concerns	Product dispute
		Other concerns
(5) Community (external disclosure)	Benefits	Support for education
		Charitable giving
		Volunteer programs
		International aid
		Employment policies
		Local growth promotion
		Other advantages
	Concerns	Financing concerns
		Other concerns

3.3. Independent variable: CEO ability

We follow the method used in Yuan et al. (2019) to measure the CEO ability of Chinese firms. Their method is derived from Demerjian et al. (2012) which has been used in many studies (e.g., Baik et al., 2011; Demerjian et al., 2013; Krishnan et al., 2015; Koester et al., 2017). The estimation procedure is publicly available online.¹ Demerjian et al. (2012) use a two-stage approach to compute managerial ability. In the first stage, they use data envelopment analysis (DEA) to estimate a firm’s efficiency. In the second stage, they regress a firm’s efficiency on firm-level determinants. The residual from the second stage is considered the manager’s contribution to a firm’s efficiency. We follow the above method to compute CEO ability using Chinese public firms’ data. As we argued before, CSR disclosure may affect a firm’s sales. Therefore, we include the CSR score in the second stage to exclude CSR disclosure’s impact on firm efficiency. For instance, following Yuan et al. (2019), we include the firm’s CSR score in the below second-stage estimation model and use the residual term of the estimation as CEO ability.

Firm Efficiency= a + b1*ln(Total Assets) + b2*Market Share + b3*Free Cash Flow Indicator + b4*ln(Age) + b5*Business Segment Concentration + b6*Foreign Currency Indicator + b7*CSR Score + Year + \(\varepsilon\)

3.4. Control variables

We also control for corporate governance-level, board-level, and company-level variables that may potentially impact a company's choice of CSR strategies. For instance, we follow prior studies (García-Sánchez et al., 2020; Yuan et al., 2019) and control for governance-level variables: CEO tenure, CEO age, CEO education (CEO-Edu), CEO duality, and CEO gender (Gender). Also, we control for board-level variables, such as board size and board independence (board-Ind) (Jin et al., 2022). Last, we control for company-level variables, such as company financial performance measured as return on assets (ROA), company leverage (leverage), company’s size (Size), state ownership enterprise (SOE), and firm age. Please refer to the Appendix for variable definitions.

3.5. Empirical Model

To empirically examine the relationship between CEO ability and the CSR disclosure gap (H1), we apply the following regression model, controlling for firm fixed effect and year fixed effect.

\[\begin{equation} \label{eq1} \small \begin{split} \text{CSR Gap} =& \ \text{CEO ability + Control Variables} \\ & \text{+ firm effect + Year effect + e} \ \ \ \ \ \ \ \ \ \ (1) \end{split} \end{equation}\]

To empirically test H2 and H3, we interact CEO political connection and CEO identity with the CEO ability, respectively.

\[\begin{equation} \label{eq2} \small \begin{split} \text{CSR Gap} =& \ \text{CEO ability + CEO political} \\ & \text{+ CEO ability * CEO political + Control Variables} \ \ \ \ \ \ \ \ \ \ (2) \\ & \text{+ Firm effect + Year effect + e} \end{split} \end{equation}\]

Whereas CEO political is a dummy variable that is equal to 1 when the CEO has a political connection, and 0 otherwise.

\[\begin{equation} \label{eq3} \small \begin{split} \text{CSR Gap} =& \ \text{CEO ability + CEO identity} \\ & \text{+ CEO ability * CEO identity + Control Variable} \ \ \ \ \ \ \ \ \ \ (3)\\ & \text{+ Firm effect + Year effect + e} \end{split} \end{equation}\]

Whereas CEO identity is a dummy variable that is equal to 1 when the CEO is promoted from the inside, and 0 otherwise.

4. Empirical results and additional tests

Table 2 shows the descriptive statistics of all variables used in this study. CSR disclosure gap range from a maximum of 36.73 to a minimum of -31.02 with a mean value of -3.78. The mean value of CEO ability is 0.00 which is similar to prior studies (Demerjian et al., 2012; Yuan et al., 2019). CEO tenure range from a maximum of 20.33 years to a minimum of 0.25 years with a mean value of 4.12 years. The average age of the CEO in our sample is 50.5 years. The median value of CEO education is 3 years, which means that most CEOs have a bachelor's degree. The mean value of CEO duality is 18%, which shows that in our sample 18% of CEOs are also the chairperson of the board. The mean value of CEO gender is 91%, which indicates that 91% of sample CEOs are male. The median value of the board size (board) is 9, ranging from a maximum of 21 directors to a minimum of 5 directors on board. Board independence mean value is 37%, which meets the Chinese regulator’s corporate governance requirement that at least one-third of the board should be independent. The mean values of ROA, leverage, and size are 0.04, 0.5, and 22.52 respectively. On average, 57% of the sample firms are SOEs.

Table 2. Summary statistics

	Mean	Sth. Dev.	min	p25	Median	p75	max
CSR Gap	-3.78	12.01	-31.02	-13.25	-6.00	4.85	36.73
CEO ability	0.00	0.15	-0.24	-0.11	-0.02	0.09	0.46
CEO tenure	4.12	3.32	0.25	1.58	3.00	5.50	20.33
CEO age	50.50	5.97	28.00	47.00	51.00	55.00	81.00
CEO edu	3.55	0.79	1.00	3.00	3.00	4.00	5.00
CEO duality	0.18	0.38	0.00	0.00	0.00	0.00	1.00
Gender	0.91	0.28	0.00	1.00	1.00	1.00	1.00
Board	9.42	2.32	5.00	8.00	9.00	11.00	21.00
Board-Ind	0.37	0.06	0.10	0.33	0.36	0.40	0.80
ROA	0.04	0.07	-1.08	0.01	0.04	0.07	0.67
Leverage	0.50	0.22	0.01	0.33	0.51	0.66	3.12
Size	22.52	1.65	17.41	21.37	22.37	23.51	28.72
SOE	0.57	0.49	0.00	0.00	1.00	1.00	1.00
Firm age	2.39	0.76	0.00	2.08	2.64	2.94	3.43

Table 3 shows the correlation matrix among all the variables. We find that CEO ability is negatively associated with the CSR disclosure gap. For the control variables, we find that CEO tenure, age, education, duality, board independence, ROA, and size are negatively correlated with the CSR disclosure gap. We also board size, leverage, SOE, and firm age are positively correlated with the CSR disclosure gap.

Table 3.Pairwise correlations

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)
(1) CSR Gap	1.00
(2) CEO ability	-0.06	1.00
	(0.00)
(3) CEO tenure	-0.02	0.01	1.00
	(0.02)	(0.58)
(4) CEO age	-0.13	-0.01	0.13	1.00
	(0.00)	(0.43)	(0.00)
(5) CEO edu	-0.01	-0.10	0.06	0.04	1.00
	(0.56)	(0.00)	(0.00)	(0.00)
(6) CEO duality	-0.05	-0.02	0.09	0.14	0.00	1.00
	(0.00)	(0.22)	(0.00)	(0.00)	(0.67)
(7) Gender	0.02	0.00	-0.03	-0.02	-0.20	0.05	1.00
	(0.03)	(0.99)	(0.01)	(0.02)	(0.00)	(0.00)
(8) Board	0.05	-0.07	0.00	0.01	0.14	-0.14	0.07	1.00
	(0.00)	(0.00)	(0.99)	(0.17)	(0.00)	(0.00)	(0.00)
(9) Board-Ind	-0.03	-0.01	-0.02	0.08	0.02	0.07	-0.07	-0.34	1.00
	(0.01)	(0.62)	(0.07)	(0.00)	(0.13)	(0.00)	(0.00)	(0.00)
(10) ROA	-0.21	0.22	0.05	-0.02	-0.05	0.10	0.01	-0.01	-0.03	1.00
	(0.00)	(0.00)	(0.00)	(0.02)	(0.00)	(0.00)	(0.22)	(0.25)	(0.01)
(11) Leverage	0.09	-0.02	-0.07	0.08	0.09	-0.13	-0.03	0.16	0.06	-0.42	1.00
	(0.00)	(0.07)	(0.00)	(0.00)	(0.00)	(0.00)	(0.02)	(0.00)	(0.00)	(0.00)
(12) Size	-0.13	0.04	-0.05	0.17	0.14	-0.12	0.02	0.25	0.10	0.00	0.49	1.00
	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.13)	(0.00)	(0.00)	(0.95)	(0.00)
(13) SOE	0.08	0.01	-0.15	0.12	0.10	-0.28	0.03	0.23	0.00	-0.12	0.18	0.25	1.00
	(0.00)	(0.36)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.00)	(0.84)	(0.00)	(0.00)	(0.00)
(14) Firm age	0.00	-0.01	0.06	0.08	-0.04	-0.15	-0.02	0.02	0.00	-0.14	0.12	0.16	0.25	1.00
	(0.98)	(0.38)	(0.00)	(0.00)	(0.00)	(0.00)	(0.04)	(0.05)	(0.97)	(0.00)	(0.00)	(0.00)	(0.00)

In Table 4, we present the findings of testing hypothesis 1. The coefficient of CEO ability is significantly negative (-8.047***), which indicates that more able CEOs can reduce the CSR disclosure gap. Our findings show that more able CEOs pay attention to satisfying multiple stakeholders and reducing the CSR disclosure gap. For the control variables, we find that CEO tenure (0.115***) and age (-0.044**) are significantly negatively associated with the CSR disclosure gap, which indicates that as tenure or age increases, they are more likely to consider multiple stakeholders in their decision making. We also find that a larger board size significantly (-0.198***) reduces the CSR disclosure gap. We find that larger and more profitable companies significantly reduce the CSR disclosure gap because they have more resources to protect multiple stakeholders’ interests. We also find that firm leverage is positively related to the CSR disclosure gap because companies facing financial difficulties are more likely to focus on external stakeholders rather than satisfying internal stakeholders.

Table 4.Main relationship between CEO ability and the CSR disclosure gap (H1)

Variables	CSR Gap
CEO ability	-8.047***
	(-9.316)
CEO tenure	-0.115***
	(-3.256)
CEO age	-0.044**
	(-2.163)
CEO-edu	0.431***
	(2.818)
CEO duality	0.326
	(1.016)
Gender	0.195
	(0.455)
Board	-0.198***
	(-2.877)
Board-Ind	-1.919
	(-0.885)
ROA	-0.576***
	(-4.866)
Leverage	0.007*
	(1.821)
Size	-0.930***
	(-10.364)
SOE	0.112
	(0.390)
Firm age	1.888***
	(10.209)
_cons	28.027***
	(12.272)
Observations	8095
R-squared	0.469
Firm FE	Yes
Year FE	Yes

t-values are in parentheses*** p<.01, ** p<.05, * p<.1
Dependent variable is the CSR disclosure gap.

To test hypothesis 2, we add the interaction of CEO ability and CEO political connection (CEO ability*CEO political). In Table 5, we show that the coefficient of the interaction term is significantly negative (-15.187**), which indicates that a CEO having a political connection strengthens the negative relationship between CEO ability and the CSR disclosure gap. Our findings confirm that CEO with political connections bears more responsibility toward multiple stakeholders.

Table 5. Moderating effect of CEO political connection on the relationship between CEO ability and the CSR disclosure gap (H2)

Variables	CSR Gap
CEO ability	-4.353***
	(-5.374)
CEO political	-0.660*
	(-1.788)
CEO ability*CEO political	-15.187**
	(-1.980)
CEO tenure	-0.111***
	(-3.129)
CEO age	-0.046**
	(-2.241)
CEO-edu	0.300*
	(1.958)
CEO duality	-0.035
	(-0.107)
Gender	-0.084
	(-0.192)
Board	-0.241***
	(-3.601)
Board Ind	-2.647
	(-1.235)
ROA	-0.622***
	(-5.102)
Leverage	0.008**
	(2.203)
Size	-0.628***
	(-7.871)
SOE	-0.028
	(-0.101)
Firm age	2.146***
	(11.947)
_cons	22.732***
	(10.683)
Observations	8095
R-squared	0.465
Firm FE	Yes
Year FE	Yes

t-values are in parentheses*** p<.01, ** p<.05, * p<.1
Dependent variable “CSR Gap” is the CSR disclosure gap.

To test hypothesis 3, we add the interaction of CEO ability and CEO identity (CEO ability*CEO identity) and present the results in Table 6. The coefficient of the interaction term is significant and negative, which indicates that inside CEOs strengthen the negative association between CEO ability and the CSR disclosure gap. This is because CEOs who are promoted from the inside have more firm knowledge than outside CEOs.

Table 6.Moderating effect of a CEO’s firm-specific knowledge (CEO identity) on the relationship between CEO ability and the CSR disclosure gap (H3)

	(1)
	CSR Gap
CEO ability	-4.069***
	(-4.177)
CEO identity	-1.079**
	(-1.962)
CEO ability*CEO identity	-0.836**
	(-2.417)
CEO tenure	-0.046
	(-1.390)
CEO age	-0.039**
	(-2.016)
CEO-edu	0.323**
	(2.223)
CEO duality	0.487
	(1.588)
Gender	-0.007
	(-0.016)
Board	-0.155**
	(-2.353)
Board-Ind	-0.269
	(-0.130)
ROA	-44.190***
	(-22.595)
Leverage	3.232***
	(4.102)
Size	-0.961***
	(-8.993)
Firm age	1.180***
	(2.974)
_cons	30.899***
	(11.593)
Observations	8095
R-squared	0.522
Firm FE	Yes
Year FE	Yes

t-values are in parentheses*** p<.01, ** p<.05, * p<.1

To mitigate the endogeneity concern due to omitted variables that may impact the relationship between CEO ability and the CSR disclosure gap, we apply two different econometric techniques. First, we apply the instrumental variable (IV) approach using the two-stage least square (2SLS). We use CEO compensation as our IV because it is positively associated with CEO ability but not related to the CSR disclosure gap. We present the results of the first stage in column 1 of Table 7, which shows that CEO compensation is positively associated with CEO ability. In column 2 of Table 7, we present the second-stage results, which show that CEO ability is significantly and negatively (-5.330***) associated with the CSR disclosure gap. Second, we apply the generalized method of moments (GMM) technique to mitigate the endogeneity concern. In table 8, we present the findings, which are similar to the main findings. In summary, our results are robust to potential spurious correlations that may arise from endogeneity.

Table 7. Two-Stage least squares

Variables	First stage	Second stage
CEO ability	-	-5.330***
		(-2.785)
CEO tenure	0.001	0.085***
	(1.351)	(2.866)
CEO age	0.000	0.023
	(0.332)	(1.397)
CEO-edu	-0.105***	-0.839***
	(-6.527)	(-3.589)
CEO duality	0.018	-0.071
	(0.543)	(-0.277)
Gender	-0.072	-0.584
	(-1.591)	(-1.561)
Board	-0.029	-0.081
	(-4.076)	(-1.020)
Board-Ind	-0.747	-3.057
	(-3.251)	(-1.402)
ROA	0.051***	0.587***
	(4.032)	(4.392)
Leverage	-0.000	-0.006*
	(-1.062)	(-1.889)
Size	0.043***	0.638***
	(4.313)	(6.934)
SOE	-0.004	0.058
	(-0.144)	(0.253)
Firm age	-0.015	-1.046***
	(-0.813)	(-6.958)
CEO compensation	3.834***	-
	(3.936)
_cons	2.496***	1.018***
	(9.989)	(0.183)
Observations	8095	8095
Pseudo R2	0.036	0.127
Firm-FE	Yes	Yes
Year FE	Yes	Yes

t-values are in parentheses*** p<.01, ** p<.05, * p<.1

Table 8. GMM

Variables	CSR Gap
L.CSR Gap	-0.091***
	(-4.216)
CEO ability	-12.297**
	(-1.500)
CEO tenure	-0.197
	(-1.161)
CEO age	0.027
	(0.175)
CEO-edu	4.029***
	(2.876)
CEO duality	1.175***
	(2.608)
Gender	6.052*
	(1.939)
Board	-1.504*
	(-1.743)
Board-Ind	4.646
	(1.552)
ROA	-4.623***
	(-3.107)
Leverage	0.039
	(1.426)
Size	-0.474
	(-0.708)
SOE	-4.471
	(-1.501)
Firm age	3.120**
	(2.330)
_cons	-37.992*
	(-1.940)
Observations	5145
Pseudo R2	0.148
AR (1) test	-2.705
AR (1) p value	0
AR (2) test	-1.20
AR (2) ρ value	0.232
Sargen test	1.011
Sargen ρ value	0.314

t-values are in parentheses*** p<.01, ** p<.05, * p<.1

We also conducted an additional test to check the robustness of our models. For example, first, we use an alternative measure of CEO ability to check the robustness of our independent variable. By following prior studies, we use the industry average of ROA for the alternative measure of CEO ability and present the results in Table 9 which shows that these findings are consistent with the main findings. Overall, our results are robust to a different measure of CEO ability.

Table 9. Alternativemeasure of CEO ability

	CSR Gap
Average ROA	-43.104***
	(-10.080)
CEO tenure	-0.064**
	(-2.166)
CEO age	-0.039**
	(-2.294)
CEO-edu	0.511***
	(4.061)
CEO duality	0.266
	(0.998)
Gender	0.154
	(0.449)
Board	-0.126**
	(-2.409)
Board-Ind	0.627
	(0.349)
ROA	-0.654***
	(-6.160)
Leverage	0.007**
	(2.064)
Size	-0.914***
	(-12.626)
SOE	0.401*
	(1.727)
Firm age	1.507***
	(10.766)
_cons	29.077***
	(15.577)
Observations	5538
R-squared	0.481
Firm-FE	Yes
Year FE	Yes

t-values are in parentheses*** p<.01, ** p<.05, * p<.1

5. Conclusions

In this study, we ask the question of whether more able CEOs can reduce the gap between internal and external CSR disclosures. Our study is motivated by the fact that more and more firms are disclosing internal and external CSR activities in China. This increasing trend creates an imbalance between internal and external CSR disclosures. However, little is known about the determinants of this imbalance. Following the upper echelons theory, our study finds that more able CEOs are more likely to reduce the gap between internal and external CSR disclosures. We also find that this negative association is strengthened when a CEO has political connections and when a CEO is internally promoted. Our results are robust after controlling for firm fixed effects and addressing endogeneity concerns. Overall our findings are consistent with our argument that more able CEOs significantly reduce the CSR disclosure gap and maintain the positive relationship between internal and external stakeholders.

The moderating role of political connections enhances our understanding of the CEO's effectiveness in driving CSR disclosure consistency. Political connections amplify the positive impact of CEO ability, enabling access to external networks and resources that aid in aligning internal and external CSR perspectives. By linking political connections to CSR disclosure consistency, we extend existing literature on CEO attributes and political influence, shedding light on the interplay between corporate social responsibility and political dynamics. Furthermore, our findings reveal the nuanced influence of internal promotion on the CEO's ability to shape CSR disclosures. CEOs who rise through the ranks possess an intimate familiarity with the organization's culture, enabling them to effectively harmonize internal and external viewpoints on CSR. This insight extends prior research on CEO succession and organizational culture by emphasizing the role of internal promotion in strengthening the coherence between CSR strategies and disclosures.

From a practical standpoint, our study offers actionable insights for firms aiming to enhance their CSR practices and disclosures. Recognizing the pivotal role of CEO ability, organizations can invest in leadership development programs that foster cognitive agility, strategic thinking, and interpersonal skills among executives. Furthermore, firms operating in politically charged environments can strategically leverage political connections to bolster CSR initiatives and communication strategies.

As a direction for future research, our study opens avenues for examining the interplay between CEO ability and other organizational outcomes, such as innovation, employee engagement, and financial performance. Exploring the mechanisms through which CEO ability translates into CSR outcomes could provide a more nuanced understanding of the underlying processes. Additionally, investigating the boundary conditions of CEO ability's impact on CSR disclosure consistency, such as industry-specific factors or firm size, could yield valuable insights.

In conclusion, our study enriches the field of corporate social responsibility by elucidating the role of CEO ability and its interplay with political connections and internal promotion in driving CSR disclosure consistency. By contributing to both theoretical knowledge and practical implications, we hope to inspire further research and facilitate the development of strategies that promote sustainable and transparent business practices. Our study has limitations. First, we are not documenting a causal relationship between CEO ability and the CSR disclosure gap. Second, our findings in the context of China may not generalize to other nations. Future research could focus on other CEO characteristics’ impact on the CSR disclosure gap.

 

---

Appendix

Appendix. Variable definitions

CSR Gap	CSR disclosure gap
CEO ability	CEO ability measure
CEO tenure	We measure CEO tenure as the number of years of a CEO in the office.
CEO age	CEO age in years
CEO edu	A CEO’s years of college education
CEO duality	A dummy variable, 1 If the CEO and chairman of the board as the same person otherwise zero
Gender	a dummy variable that equals 1 if the CEO is male and 0 otherwise
Board	The number of directors serving on the board
Board-Ind	The percentage of independent directors over total directors on the board
ROA	Net profit divided by total assets
Leverage	The ratio of total debt divided by total assets
Size	Natural log of total assets of a firm
SOE	1 if a firm is a state-owned enterprise (SOE),0 otherwise
Firm age	Natural log the number of years when the firm was listed till the observation year
CEO identity	Whereas CEO identity is a dummy variable that is equal to 1 when the CEO is an inside CEO, and 0 otherwise.
CEO political	Whereas CEO political is a dummy variable that is equal to 1 when the CEO has a political connection, and 0 otherwise.

 

---