CEO social capital, board gender diversity and environmental disclosures of public companies in Southeast Asia Open Access

Environmental disclosures, often driven by regulatory expectations and stakeholder demands, reflect the extent to which companies transparently report their environmental impacts and sustainability initiatives. Several studies have highlighted the role of corporate governance and regulatory frameworks in influencing disclosure practices (Cahyono et al., 2024a; Al-Shaer and Zaman, 2018; De Villiers et al., 2011). However, the influence of CEO social capital (CEOSC), which is a strategic human factor, on environmental transparency has not been comprehensively considered (Afzali et al., 2022; Jermias and Gani, 2014; Mundi, 2022; Sauerwald et al., 2016; Wiersema et al., 2018). In this study, we explore whether CEOSC influences corporate environmental disclosure practices. We seek to advance the literature by identifying CEOSC as an important driver of environmental disclosures. Furthermore, we examine how board gender diversity (BGD) serves as a moderating and mediating factor, potentially strengthening CEO influence through monitoring mechanisms and commitments to sustainability.

Various factors influencing environmental disclosures and their impact on reputation and investor appeal have been previously explored (Clarkson et al., 2008; Qian et al., 2021). Such disclosures are often considered to be a signal of sustainability commitment; however, these are usually symbolic or merely a “window dressing”, resulting in inconsistent impacts on long-term performance (Cho et al., 2012; Truong et al., 2021). These diverse views highlight the need for a new understanding. We argue that CEOSC, which includes professional networks, reputation and external relationships, is a crucial determinant of the inconsistent views, because it influences informative and opportunistic incentives for environmental disclosures (El-Khatib et al., 2015; Fracassi et al., 2016; Helfaya and Moussa, 2017).

In social science research, two primary functions of social affiliations are identified: a channel of information and communication and that of power and influence (Coleman, 2009). From an informational perspective, CEOs’ social networks facilitate efficient information dissemination, strengthen credibility and signal firm quality. Engelberg et al. (2012) showed that CEOs’ informal relationships can improve access to loans with better terms. With high social capital, CEOs have the potential to promote environmental transparency more authentically and reduce reliance on symbolic practices (Walls and Berrone, 2017; Afzali et al., 2022). This motivation is related to mitigating creditor and investor risk perceptions, maintaining reputation and reducing debt costs (Martínez-Falcó et al., 2024; Gong et al., 2021).

However, CEOSC does not always translate into transparency. Reputational pressure and market expectations can encourage greenwashing, which is a symbolic disclosure not accompanied by robust action (Long et al., 2024; Schwertner and Sohn, 2024). CEOs with extensive networks are often in the public spotlight; thus, the risk to their reputations increases when dishonest practices are exposed (Griffin et al., 2021; Burke, 2022). Under these circumstances, greenwashing can be a defensive strategy for maintaining reputation without incurring significant environmental costs. Thus, the relationship between CEOSC and the quality of environmental disclosures is not always linear but rather depends on individual cost–benefit calculations.

In addition, CEOs with high social capital can easily gain board support for certain disclosure strategies, even when they deviate from conservative operational policies. They also obtain a form of “insurance” in the labour market, because social relationships can protect against the consequences of incompetency or create new opportunities after their tenure (Faleye et al., 2014; Zou et al., 2019; Cahyono et al., 2024b). This protection can encourage CEOs to take greater risks, including in environmental disclosure strategies. Thus, social relationships can be a paradoxical force, either supporting transparency or increasing opportunities for adaptable practices.

A critical question then arises: does CEOSC reduce symbolic disclosure practices or increase their variety and intensity? When used sincerely, social capital enables CEOs to strengthen stakeholder trust through authentic disclosures (Cao et al., 2015; Zou et al., 2019). Conversely, when focused on personal gain, social capital can support excessive and opportunistic disclosure strategies (Cahyono et al., 2023). We aim to answer this question using cross-country empirical evidence from Southeast Asia.

CEO or board social capital has been primarily related to firm performance (Afzali et al., 2022; Muttakin et al., 2018; Hoi et al., 2019); however, its specific relationship with environmental disclosures remains underexplored. We address this gap by testing a path model involving BGD as a moderator and mediator. Gender diversity has been recognised as influencing governance quality (Ben-Amar et al., 2017; Konadu et al., 2022; Nuber and Velte, 2021). From the perspective of group dynamics theory (Lewin, 1947), the presence of women on boards can broaden discussions and improve oversight; however, it poses coordination challenges. Therefore, gender diversity can strengthen accountability, including in environmental disclosures, although it requires adaptations in strategic decision-making.

This study contributes to the literature in the following way: first, it extends the literature on the determinants of environmental disclosures by emphasising the role of CEOSC. Formal governance structures and individual factors such as social networks and reputation are shown to influence environmental transparency. Second, this study integrates the gender perspective into the analysis of CEOSC. The ambivalent role of gender diversity as a moderator is highlighted to provide new insights into board dynamics in the context of environmental disclosures. Third, this study adds cross-country empirical evidence to Southeast Asia, which is a relatively under-researched context compared with developed markets but highly relevant as it experiences increasing sustainability demands.

This study has several practical implications. Companies should assess the social capital of CEO candidates, such as their social reputation, external engagement and professional networks, during the selection and succession process. Leadership development programmes should also emphasise social capacity development, such as cross-sector collaboration and participation in sustainability forums. Thus, companies can improve the quality of their environmental disclosures while strengthening legitimacy by increasing ESG demand.

The remainder of this paper is organised as follows: Section 2 presents a review of related work and the hypotheses, Section 3 introduces the research methodology, Section 4 presents a discussion on the empirical results, and Section 5 concludes with the study's contributions and policy implications.

Indonesia, Malaysia, Thailand and Singapore experience complex environmental challenges that reflect various policy responses to the global ecological crisis. Southeast Asia experiences air pollution from industrial and transportation emissions, land degradation due to deforestation and urbanisation as well as carbon emissions that are attributed to climate change mitigation commitments (ASEAN, 2021). The water crisis also stands out: Indonesia and Thailand rely on agriculture and manufacturing, which consume large amounts of water; Singapore experiences supply constraints and relies on desalination and NEWater (its high-grade reclaimed water produced from treated wastewater); whereas Malaysia is relatively well-off in terms of water resources and vulnerable to climate change and industrial growth (Lee and Kim, 2023; Subramaniam and Loganathan, 2024). Thus, the environmental contexts of these regions provide an important background for understanding corporate environmental disclosure practices.

The selection of these four countries is driven by a combination of economic factors, policies and market dynamics. Singapore stands out as a financial centre with stringent regulations, such as carbon taxes and mandatory sustainability reporting. Malaysia strengthened its transparency through ESG policies and carbon reporting. Indonesia has implemented PROPER as an environmental performance rating instrument, whereas Thailand is promoting a bio-circular-green strategy. These differences allow for a comparative analysis of how regulatory factors, market demands and institutional conditions influence environmental disclosures (Lee and Kim, 2023).

The issues of CEOSC, gender diversity on boards of directors and environmental reporting practices are becoming increasingly crucial in the ASEAN context. Boards of directors tend to be gender homogeneous, making the examination of how CEOs’ social networks influence environmental decisions in Indonesia and Malaysia relevant (Katmon et al., 2019). Singapore has demonstrated a positive relationship between gender diversity and accountability, whereas Thailand, which actively promotes gender equality, requires empirical evidence of its relationship to environmental disclosure transparency. Understanding the relationship between CEOSC and BGD is crucial in developing countries that experience increasing regulatory pressures and market expectations.

The literature shows that corporate environmental disclosures are influenced by managerial and board characteristics (Riyadh et al., 2019; Arslan et al., 2023; Gerged et al., 2023; Cahyono et al., 2024a). Factors such as gender diversity (Post et al., 2022) and industry experience (Martínez-Falcó et al., 2024; Tekin and Polat, 2025) have been shown to be significant. However, the role of CEOSC remains relatively underexplored, despite the literature confirming that social networks can expand access to information, strengthen legitimacy and promote transparency (García-Sánchez et al., 2021; Li et al., 2023; El-Khatib et al., 2015; Ho et al., 2024).

CEOs with extensive networks have two advantages. First, faster access to best practices, global trends and new regulatory standards leads to more proactive policy adoption consistent with stakeholder demand (Adams et al., 2024; Griffin et al., 2021). Second, interactions with business partners, NGOs and academics strengthen the collective norms that encourage the adoption of transparent practices (Kim and Kim, 2024; Perez-Batres et al., 2012; Sauerwald et al., 2016). Through cross-organisational learning, CEOSC enhances the credibility of disclosures (Harymawan et al., 2024). Thus, CEOSC influence individual decisions and define a governance ecosystem that supports sustainability transparency.

The literature also shows that CEOs with strong social capital tend to encourage boards to adopt progressive sustainability policies (Hao et al., 2021; Wiersema et al., 2018). This is consistent with stakeholder theory (Freeman, 1984), which emphasises corporate responsibilities to shareholders, regulators, investors, communities and environmental organisations. CEOs who can develop productive relationships with stakeholders are more likely to increase their accountability, legitimacy and credible disclosures (Clarkson et al., 2008; de Villiers et al., 2011). Based on this argument, we proposed the following hypothesis:

However, governance mechanisms through BGD may moderate this relationship. Generally, gender-diverse boards broaden perspectives and increase sensitivity to sustainability issues (Nielsen and Huse, 2010). More diverse boards respond better to stakeholder demands (Post et al., 2022). However, gender diversity can also slow the decision-making process because of increased disagreement and coordination complexity (Rao and Tilt, 2016; Nguyen et al., 2021).

Several recent findings have supported this argument. Tejerina-Gaite and Fernández-Temprano (2021) found that although CEOs with high social capital have greater access to resources, the presence of a more gender-diverse board slows down consensus on decisions related to environmental disclosure. This is consistent with group dynamics theory (Lewin, 1947), which emphasises that member heterogeneity broadens discussions but increases coordination complexity (Ben-Amar et al., 2017; Soysa et al., 2024). Delays in consensus can hinder rapid responses to sustainability demands (Ardianto et al., 2024a; Karahanna and Preston, 2013; Qian et al., 2021). Thus, gender diversity can create a paradox: it enhances the quality of discussions but weakens the positive effect of CEOSC on environmental disclosure by hindering internal board dynamics (Ardianto et al., 2024b; Konadu et al., 2022). Therefore, we proposed the following hypothesis:

We focus on non-financial companies in Southeast Asia (Indonesia, Malaysia, Singapore and Thailand) listed in the Carbon Disclosure Project (CDP) database for the 2017–2023 period. Data were obtained from the Bloomberg and Refinitiv databases and annual and sustainability reports. The CDP was selected based on the credibility and standardisation of its data, which has been recognised in the academic literature (Cahyono et al., 2024a; CDP, 2019). The CDP provides extensive, structured and relevant environmental information for comparison across industries and countries, thereby enhancing the legitimacy of the analyses. The CDP’s focus on disclosure also benefits investors and stakeholders when assessing corporate sustainability practices.

These four countries were selected because they are the largest economies in the ASEAN region and experience diverse climatic challenges. The 2017–2023 period was selected because it reflects the pivotal moment after the adoption of the 2015 Paris Agreement and the increasing global focus on sustainable investment. The limited geographic and temporal scope also helps control for confounding factors such as differences in climate or national policies.

The final sample consisted of 973 annual observations. The largest industry distribution was mining (127 observations), whereas the smallest was agriculture, forestry and fisheries (102 observations). In terms of countries, Singapore dominates with 268 annual observations, as presented in Table A1 ^[1].

The CEOSC reflects the network of relationships, influences and reputations held internally and externally of the organisation. This capital enables CEOs to obtain information, mobilise resources and strengthen the company’s legitimacy through interactions with investors, regulators and industry leaders.

This was measured using the following indicators. First, PARTICIPATION is the CEO’s level of involvement in business associations or forums. Second, MENTION refers to the frequency of mentions of the CEO in business publications or mainstream media. Third, AWARDS has a dummy value of one if the CEO received a professional award in the current or previous year. Data were collected from LinkedIn and international business media (e.g. Forbes and Bloomberg) and professional award reports.

We adopt the theory of Freeman (1984) by expanding the measurement to include media exposure and formal recognition, thereby providing a more holistic perspective than methods that only assess formal networks (Iqbal and Rao, 2023; Quan et al., 2024; You et al., 2024; Fang et al., 2024; Cahyono et al., 2024a).

Gender diversity on the board of directors is an indicator of inclusivity and governance and is believed to influence strategic decision-making and environmental disclosures. Measurements included PWOMAN, the percentage of women among total board members (Afzali et al., 2022; Ben-Amar et al., 2017); TWO_WOMAN and THREE_WOMAN, dummy variables based on critical mass theory, which emphasises a significant influence when at least three women are on the board (Nuber and Velte, 2021); and a dummy for the presence of women in strategic positions, such as the chair of the board or on the audit/sustainability committee (Ha et al., 2024; Konadu et al., 2022; Ningsih et al., 2023). Data were obtained from annual reports and capital market databases. These indicators were selected to capture the numerical representation and strategic position of women on the board.

Environmental disclosures were measured using three pillars of the CDP (climate change, water and forests). The CDP assigns scores labelled A–F, which were simplified to percentile rankings for ease of analysis. The conversion was performed by assigning the following numerical weights: A = 8, A = 7, B = 6, B = 5, C = 4, C = 3, D/E = 2 and F = 1. This method refers to the model of Cahyono et al. (2024b) to produce standardised variables reflecting a company’s level of transparency and sustainability practices.

Several control variables were included to reduce estimation bias: BIG4: dummy = 1 if the company is audited by a Big 4 accounting firm. GCAO: dummy = 1 if the company receives a going-concern audit opinion. AUDCOM: dummy = 1 if the company has an audit committee. RMC: dummy = 1 if a risk management committee has been instituted. BODSIZE: total number of board members. SUBS: natural log of the number of subsidiaries. FSIZE: natural log of total assets. LOSS: dummy = 1 if earnings are negative. LEV: debt-to-assets ratio. ROA: net income to total assets. MTB: market value-to-book value ratio. CURRENT: ratio of current assets to liabilities. INVREC: proportion of inventory and receivables to assets. LIT: dummy = 1 if the company is in a high-litigation industry as per the SIC classification. These controls were selected based on the governance, audit and sustainability literature (Al-Shaer and Zaman, 2018; Broadstock et al., 2021; Cahyono et al., 2023; Hoi et al., 2019; Nasih et al., 2025). All definitions and operationalizations are presented in Table A2 ^[1].

Table A3 ^[1] presents descriptive statistics for the variables in the regression model. The mean value for environmental impact (ENV) is 2.029 with a standard deviation, median, minimum and maximum of 2.281, 1.000, 0.000 and 10.000, respectively. This indicates wide variation in environmental reporting practices. The PARTICIPATION variable has a mean of 0.116, with a range of 0.000–3.000, indicating that the majority of companies have low levels of engagement in external forums. The MENTION variable has a mean of 2.555 (SD = 1.577) with a median of 3.000, indicating that CEOs mostly receive limited exposure in the media. The AWARDS variable has a mean value of 0.397, indicating that most CEOs do not receive public awards although some cases to the contrary exist.

Table A4 ^[1] reports the correlation matrix between the variables. The results show that ENV is not significantly correlated with PARTICIPATION (p = 0.786), indicating that the level of external CEO participation does not directly determine the quality of environmental disclosure. Conversely, a significant correlation is observed between BGDUM and PARTICIPATION (p = 0.015), indicating that board composition is related to the level of CEO participation. PWOMAN is significantly positively correlated with BGDUM (p < 0.001), suggesting that increasing female representation on boards is associated with the presence of certain educational backgrounds. Overall, this matrix confirms the existence of an underlying pattern of relationships between the variables, although it cannot describe a causal relationship.

Figure A1 ^[1] presents a path analysis framework used to examine the relationship between CEOSC and environmental disclosures (ENV) and the role of BGD as a moderator and mediator. Furthermore, the contribution of the ENV to firm value (FV) was examined. The path analysis presented in Table 1 reveals several crucial findings. The results show a significant positive effect (ρ = 0.156; p < 0.001), confirming that CEOs with broader social networks encourage environmental disclosure. Gender diversity is shown to influence ENV (ρ = 0.045; p < 0.001), whereas ENV increases the FV (total effect = 0.272; p < 0.001). Thus, BGD contributes to increasing corporate legitimacy through environmental transparency. CEOSC positively influences BGD (ρ = 0.286; p < 0.001), BGD influences ENV (ρ = 0.337; p < 0.001) and ENV increases the FV (ρ = 0.172; p < 0.001). The total mediation effect was 0.412 (p < 0.001). These findings confirm the importance of a diverse approach in which CEOSCs increase the FV through gender-diversity-enhanced ENV.

Table A5 ^[1] presents the effects of BGD moderation on the relationship between CEOSC and ENV. The baseline results (specification 1) indicate that PARTICIPATION has a significantly positive effect on ENV (coefficient = 0.660; p < 0.001). However, when relationships with the board variables are introduced, a more complex pattern emerges. The PARTICIPATION × BGDUM relationship yields a significant negative effect (−0.121; p < 0.001), indicating that a board’s educational background does not always strengthen the influence of CEO participation. Interestingly, the relationship with gender variables (PWOMAN, TWO_WOMAN and THREE_WOMAN) is consistently negative, with the largest effect on THREE_WOMAN (−0.488; p < 0.001). These results imply that in some cases, gender diversity can weaken the positive effect of CEOSCs on ENV. Although CEOs maintain extensive networks, more gender-diverse boards tend to balance their strategic orientation from a broader critical perspective, thereby reducing excessive levels of openness.

A PSM test was conducted as per the study conducted by Caliendo and Kopeinig (2008) to address potential endogeneity. The sample was divided into treatment and control groups (CEOs with high CEOSC) and (CEOs with low CEOSC), respectively. Theoretically, CEOs with high social capital are predicted to be more proactive in environmental disclosure because their network influences and reputation drives sustainability strategies (Cao et al., 2015; Ferris et al., 2017). The CEOSC components (PARTICIPATION, MENTION and AWARDS) were combined using PCA to produce a composite score (PCA_CEOSC), which was used in the PSM model. The results (Table A6 ^[1]) show that PCA_CEOSC has a significant effect on ENV (coefficient = 2.806; p < 0.01). In addition, the CEOSC component variables remain consistently and significantly positive across all models, with coefficients for PARTICIPATION (0.263–0.291), MENTION (0.187–0.206) and AWARDS (0.289–0.331) at p < 0.01. Robustness checks using the average treatment effect method confirmed similar results.

Table A7 ^[1] reports the results of the subsample analysis based on country. The results are consistent with the main model for Indonesia (N = 213), Malaysia (N = 246), Singapore (N = 268) and Thailand (N = 218): CEOSC is significantly positively correlated with ENV across all countries. This suggests that the role of CEOSC in promoting environmental transparency is cross-contextual, despite the differences in regulatory and governance systems across countries.

Additional tests were conducted to assess the moderating roles of CEO characteristics: education (CEOEDU), stock ownership (CEOOWN), tenure (CEOTEN) and family status (FCEO). Table A8 ^[1] shows that the positive effect of the CEOSC on ENV is significant even when these variables are included. This means that individual CEO factors strengthen the effectiveness of social networks but do not alter the course of the primary relationship.

This study highlights the relationship between CEOSC and environmental disclosure and the moderating role of BGD. An analysis of public companies in Indonesia, Malaysia, Singapore and Thailand for the 2017–2023 period reveals that CEOSC has a significant positive effect on environmental transparency. CEOs with extensive social networks tend to increase their disclosures to strengthen their reputations and relationships with stakeholders. BGD weakens this relationship, highlighting the complex dynamics of governance, monitoring and coordination.

These findings have strategic implications for managers and policymakers. The sustainability agenda must consider the synergy between CEOSC and board composition. Board capacity can be strengthened through strategic collaborative training to make more effective environmental decisions. Practically, companies are advised to integrate CEO-board characteristics during recruitment and encourage cross-functional training.

Academically, these results advance the literature on sustainable governance by highlighting that CEO roles and board structures are not always complementary. A limitation of this study is the institutional context of Southeast Asia. In future work, generalisability to countries with established governance systems should be tested through cross-jurisdictional studies.

The supplementary material for this article can be found online.