Revisiting the Corporate Governance-ESG Performance Nexus: Insights from the CEE Companies

From a theoretical standpoint, larger boards bring greater heterogeneity in expertise, knowledge and external links (Adams, Almeida & Ferreira, 2005). Larger boards can draw on a wider pool of skills and resources, which improves the firm's capacity to address complex topics, including ESG issues (Hidalgo, García-Meca & Martínez, 2011). From a stakeholder-theoretic perspective (Freeman, 1984), a larger board can better represent and reconcile the claims of multiple stakeholder groups, enhancing responsiveness to social and environmental concerns. By broadening voice and oversight, board size becomes a governance mechanism that aligns firm strategy with stakeholder welfare.

However, the role of board size in ESG/sustainability performance remains contested. One stream of research argues in favour of smaller boards. The core claim is that smaller boards are more cohesive, better coordinated and communicate more efficiently, which enhances accountability and strengthens directors' commitment to oversight (Faysal, Salehi & Moradi, 2021). Consistent with this position, Ding et al. (2021) suggested that smaller boards tend to be more active in monitoring roles. Smaller boards are more effective in decision-making, have more individual responsibility regarding their monitoring obligation, and fewer conflicts within the Board of Directors (Ahmed, Hossain & Adams, 2006). Other works highlight structural weaknesses of larger boards. Larger boards may suffer from coordination and communication problems, slower decision-making, and lower internal cohesion (Rao, Tilt & Lester, 2012). They may also be more vulnerable to CEO influence and to managerial dominance, increasing agency costs rather than reducing them (Jensen, 1993).

At the same time, arguments in favour of smaller boards are not without limitations. Critics note that small boards have less diversity of expertise and narrower knowledge bases, which may restrict their ability to understand and govern complex sustainability-related issues (Guest, 2008). Smaller boards also face heavier individual workloads, which may reduce their capacity to effectively control, monitor and engage with specialised topics, such as ESG strategy (Hussain, Rigoni & Orij, 2018). From this angle, a board that is ‘too lean’ may become overstretched and ultimately less effective in sustainability governance, even if it is internally cohesive.

Prior evidence on board size and ESG/sustainability performance is mixed but generally suggests capacity benefits that can be offset by coordination costs. Using 313 S&P 500 firms in primary and manufacturing industries, Walls, Berrone and Phan (2012) reported that larger boards are associated with weaker environmental performance, consistent with slower decision-making and diffusion of responsibility. In contrast, multi-country and country-specific studies typically find performance gains from larger boards. In India, Kumari et al. (2022) supported a positive size-environmental performance relation. Almaqtari et al. (2023) showed that board size significantly promotes environmentally friendly production. Broader ESG performance studies confirm this pattern: in Turkish firms, larger boards are positively related to ESG (Aksoy et al., 2020); among 108 listed banks in Europe and the United States, board size predicts higher ESG performance (Birindelli et al., 2018); in a sample of 478 multinationals, board size is positively associated with sustainability performance (Chams & García-Blandón, 2019); and for French listed firms, larger boards are more likely to engage in sustainability practices and display stronger ESG scores (Beji et al., 2021). Studies indicate that bigger boards can marshal diverse expertise and stakeholder access to advance environmental and broader sustainability outcomes, though the early US evidence warns of potential coordination frictions. Whether these mechanisms extend to CEE companies remains unclear. However, given that the vast majority of studies indicate that larger boards promote sustainable practices, the following hypothesis is proposed:

Hypothesis 1 (H1): In CEE, board size is positively associated with corporate ESG performance.

Hollindale et al. (2019) argued that women directors emphasise non-financial outcomes, such as corporate social responsibility (CSR), ethics and reputation, which supports longer-run financial stability by reducing exposure to regulatory, social and reputational risk. Valls Martínez et al. (2022) showed that the presence of women in upper management improves decision-making quality and organisational efficiency, supporting the execution of CSR and sustainability initiatives. Stakeholder theory (Freeman, 1984) argues that firms are accountable to a broad set of stakeholders rather than solely to shareholders. A more diverse board is expected to recognise, interpret and respond to heterogeneous stakeholder claims more effectively, ensuring that employee, community, regulatory, investor and environmental expectations are integrated into strategic decisions.

Empirical evidence increasingly ties board gender diversity to realised ESG/sustainability performance. In the S&P 500, Hafsi and Turgut (2013) associate a larger number of female directors with stronger social outcomes. Similarly, Cordeiro and Tewari (2015), analysing US firms over a 5-year period, concluded that firms with more women on the board achieved higher environmental performance, attributing this to women's greater strategic orientation towards sustainability. Addabbo et al. (2020), using the sample of European firms, showed that the presence of women on boards improves the firm's ability to manage stakeholder expectations, reinforcing the social side of sustainability performance. Slomka-Golebiowska, De Masi and Paci (2023) found that FTSE MIB Italian companies (2007–2018) led by a female board chair perform better on ESG. In Chinese industries with pronounced environmental and social risk, board gender diversity strengthens corporate social performance (Naveed et al., 2021). Broader studies confirm that greater female representation tends to align with stronger sustainability/ESG performance in Australia (Hollindale et al., 2019), among Italian companies (Romano et al., 2020), within banks from the US (Shakil, Tasnia & Mostafiz, 2021), and among 439 publicly listed non-financial firms in 20 emerging economies (Disli, Yilmaz & Mohamed, 2022). An interesting study conducted by Godfrey et al. (2024) on 2,880 US firms from 2007 to 2016 found that boards with greater diversity are significantly more effective than less diverse boards at reducing corporate social irresponsibility incidents once they arise.

On the contrary, the literature also records mixed or adverse findings: Dang et al. (2021) detected no (or negative) links with environmental/ESG outcomes; Husted and de Sousa-Filho (2019) showed that the mere presence of female directors can correspond to lower performance in Latin American companies; and Naciti (2019), studying 362 large industrial firms including members of the Fortune Global 500-likewise questions a universal board gender diversity-performance relationship. Mechanisms and contingencies appear pivotal: evidence of ‘critical mass’ suggests that at least three women can improve environmental decisions in Italy (Muhammad & Migliori, 2023). Amorelli and García‐Sánchez (2021), also in a European setting, argued that a critical mass matters: boards with at least three female directors exhibit stronger social performance, suggesting that token representation is not sufficient. Despite these advances, CEE remains notably underexplored: post-transition governance, concentrated and state/foreign ownership, evolving EU-aligned sustainability enforcement, and heterogeneous gender norms may determine whether female directors can translate board voice into measurable ESG performance rather than symbolic compliance. However, in line with the stream of research underscoring the positive impact of female representation, the following hypothesis is stated:

Hypothesis 2 (H2): In CEE board gender diversity is positively associated with corporate ESG performance.

Board tenure refers to the length of time, typically measured in years, that a director serves on the board of a firm (Sun & Bhuiyan, 2020). At the CEO level, tenure captures the number of years the current CEO has served in that role within the firm in a given year. Stakeholder theory (Freeman, 1984) posits that boards are charged with balancing the interests of multiple constituencies; director tenure can shape how effectively directors recognise, prioritise, and steward those stakeholder claims. Director tenure can influence sustainability-related governance in two opposite ways. On the one hand, longer tenure is associated with accumulated firm-specific knowledge, a deeper understanding of the firm's operations and strategic environment, and a more informed view of managerial capabilities (Ben-Amar et al., 2013; Hafsi & Turgut, 2013). This experience may enhance the board's ability to identify material environmental and social issues, to oversee strategic responses, and to monitor management effectively. Ben-Amar et al. (2013) argued that directors with longer service are better positioned to evaluate managerial practices, make more informed strategic and operational judgements, and exercise monitoring responsibilities with greater competence. In this sense, tenure can strengthen governance quality and thus support improved performance on social and environmental dimensions. On the other hand, longer tenure may generate rigidity and entrenchment. Extended service can lock directors into established routines, foster resistance to new ideas, and increase alignment with management, which weakens independent oversight (Ben-Amar et al., 2013). Walls and Hoffman (2013) argued that boards dominated by long-tenured members tend to conform to habitual practices, display loyalty to existing managerial positions, and converge on managerial values, potentially lowering the board's appetite to surface controversial issues in areas such as environmental risk and social responsibility. Shorter tenure presents the mirror image. Newly appointed directors are less embedded in existing power structures and may be more willing to challenge management. At the same time, limited organisational knowledge can make it difficult for them to propose credible alternatives or to contest managerial narratives in a technically grounded way (Hafsi & Turgut, 2013). Hafsi and Turgut (2013) suggested that newer directors may adopt an overly cautious posture, while more established directors may avoid confrontation; together, this dynamic may produce a board that neither forcefully leads nor openly opposes management on social and environmental issues.

Evidence on director tenure and realised ESG/sustainability performance is mixed across settings and roles. In European samples, tenured directors have been linked to better outcomes: for listed firms in ‘old’ member states of the EU, longer-serving directors enhance environmental performance (Paolone et al., 2024), and broadly across European listed companies, more tenured boards correspond to higher sustainability performance (Collevecchio et al., 2025). In the US, board tenure is associated with stronger social and environmental behaviour and fewer environmental concerns (Harjoto et al., 2015). In contrast, several studies report no robust effect of board tenure on ESG: in Italy (Cucari, Esposito De Falco & Orlando, 2018), in S&P and broader US samples (Hafsi & Turgut, 2013; Walls & Hoffman, 2013), and in a cross-country set of top listed firms from Australia, France, the UK, and the US (Wellalage, Locke & Acharya, 2018). Other findings point to potential downsides of lengthy service: Indonesian evidence links prolonged board tenure to lower CSR activity (Handajani et al., 2014). As a result, the literature hints that accumulated firm- and stakeholder-specific knowledge can facilitate ESG improvements, but benefits may taper or reverse as entrenchment, routine persistence, or reduced openness to innovation set in. Thus, given the context of CEE, the following hypothesis is stated:

Hypothesis 3 (H3): In CEE board tenure has no significant association with corporate ESG performance.

CEO duality arises when the same individual serves simultaneously as chief executive officer and chair of the board, thereby controlling both day-to-day managerial authority and the highest formal oversight position. The performance implications of CEO duality remain one of the most contested questions in governance research and practice (Singh et al., 2018). From an external perspective, CEOs are viewed as ultimately responsible for firms' actions and outcomes because they personify corporate leadership and strategic direction (Ou et al., 2024). Under conditions of intense competition and heightened public and regulatory scrutiny, CEOs are expected to prevent behaviour that could be seen as irresponsible or in conflict with CSR norms (Ou et al., 2024). Senior executives' conduct therefore has material consequences for firms' legitimacy, reputation and performance (Mubeen et al., 2021). Under the lens of stakeholder theory, corporate governance should align managerial discretion with the long-term welfare of diverse constituencies (Freeman, 1984). CEO duality, by concentrating power in a single individual, can decouple decision-making from stakeholder accountability (Fama & Jensen, 1983). This concentration may weaken the board's monitoring capacity and independence, enabling the pursuit of personal or short-term interests at stakeholders' expense and eroding both financial performance and socially responsible behaviour (Singh et al., 2018). In the same vein, unchecked discretion can translate into underinvestment in stakeholder-oriented or sustainability initiatives, thereby impairing ESG performance (Beji et al., 2021). Stakeholder theory does not, however, preclude the possibility that unified leadership might serve stakeholder interests when it enhances timely, coherent actions that reduce deliberation costs and mobilise resources towards salient claims (Prado-Lorenzo & García-Sanchez, 2010). From this perspective, CEO duality could, under conditions of strong stakeholder salience and credible oversight, facilitate decisive deployment of CSR/ESG as instruments for legitimacy and competitive positioning that benefit key stakeholder groups (Mubeen et al., 2021). Thus, stakeholder theory frames CEO duality as a contingent governance arrangement whose net effect on ESG and performance hinges on whether concentrated authority is effectively checked and directed towards creating value for stakeholders.

Reflecting this importance, a considerable body of research examines how CEO duality shapes CSR/ESG performance. Several studies report a negative link between CEO duality and ESG outcomes. In Latin America, Husted and de Sousa-Filho (2019) documented an adverse association across firms in Brazil, Colombia, Chile, and Mexico. In India, Bhaskar et al. (2024) found that CEO duality is associated with weaker CSR practices, which they attribute to weak enforcement and institutional conditions. At a global level, Sepulveda-Nunez, Fong Reynoso and Llamosas-Rosas (2025), using 695 non-financial firms across 59 stock markets in 54 countries spanning Africa, the Americas, Asia, Europe, and Oceania, concluded that combining the CEO and chair roles is associated with poorer ESG performance. Not all settings align, however Nguyen et al. (2021) found no significant relationship in East Asia (China, South Korea, Taiwan), indicating institutional context may shape the effect. Nevertheless, given the fact that the vast majority of research indicates a rather negative impact of CEO Duality, the following hypothesis is formed in the context of CEE:

Hypothesis 4 (H4). In CEE, CEO Duality is negati¬vely associated with corporate ESG performance.

Independent board members are typically not employed by the company and do not maintain significant business ties with it, which positions them to act as external monitors rather than as agents of management. Since they are more concerned with their personal reputation and less aligned with managerial self-interest, they are expected to discourage actions that could damage the firm's legitimacy (Walsh & Seward, 1990). Independent directors are expected to monitor executive directors more rigorously, to protect minority interests and, in some settings, to support stronger firm performance (Ricart, Rodríguez & Sánchez, 2005). From the perspective of stakeholder theory, independent (outside) directors are also expected to be more attentive to the interests of a broad range of stakeholders, including employees, communities, regulators and the natural environment. As a result, they are more likely to push the firm towards socially responsible behaviour, in order to demonstrate accountability to these stakeholders (Johnson & Greening, 1999).

Empirical research generally suggests that greater board independence is associated with stronger sustainability performance. Evidence from large US samples is consistent with this view: examining nearly 15,000 firms, Jo and Harjoto (2011) documented a positive relationship between board independence and sustainability outcomes. A similar pattern is reported in the study of the 500 largest companies (Zhang et al., 2013), where a higher proportion of outside/independent directors is linked to superior sustainability performance. Australia shows similar patterns: among the largest 100 firms, Rao et al., (2012) documented a positive relationship between independence and social/environmental outcomes. In the UK, Liao et al. (2015) studied 329 of the largest British companies and found that boards with more independent directors, especially when an environmental committee exists, are more ecologically proactive and better at balancing stakeholder perspectives around ESG issues. In South Africa, Ntim and Soobaroyen (2013) found that independence improves sustainability performance, and in Turkey, using 165 listed firms, Sahin, Basfirinci and Ozsalih (2011) reported higher sustainability performance where the share of independent directors is larger. India also fits the pattern, with Kumari et al. (2022) reporting a positive association between independence and ESG performance. Whether these mechanisms carry over to CEE countries is unresolved; however, given the prevailing positive influence of board independence, the following hypothesis is formed:

Hypothesis 5 (H5). In CEE, board independence is positively associated with corporate ESG performance.

The empirical analysis is based on an unbalanced panel dataset of 50 publicly-listed companies from CEE countries, identified in accordance with the OECD country classification (Albania, Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia and Slovenia). The observation period covers 2017–2023 years, resulting in a total of 323 firm-year observations. Firms were selected based on data availability for both corporate governance variables and ESG performance indicators. Substantial dropouts occurred because of limited data availability. Consequently, only companies from four CEE countries were examined (Table 1).

Studied companies represented a broad range of industries. The largest two industry groups (Banks and Banking and Investment Services) together represent 26% of the observations, indicating that the sample was not heavily concentrated in any single industry (Table 2).

Using panel data regression over a 7-year period, this study investigates the relationship between the ESG performance (ESGP) and five board characteristics: board size (B_SIZE), board gender diversity (B_DIV), average board tenure (B_TEN), CEO Duality (B_CEO) and the proportion of independent board members (B_IND). To strengthen the robustness of the analysis, the model also incorporates control variables commonly employed in corporate ESG performance research, including cost of capital (COC), presence of CSR Sustainability Committee (CSR_COM), profitability (PROF), leverage (LEV) and company size (SIZE). Based on these variables, the regression model was developed and estimated as follows: ESGPit=β0+β1B_SIZEit+β2B_DIVit+β3B_TEN+β4B_CEOit+β5INDit+β6COCit+β7CSR_COMit++β8PROFit+β9LEVit+β10SIZEit+μi+λt+εit \matrix{{{\rm{ESG}}{{\rm{P}}_{{\rm{it}}}}={\beta_0}+{\beta_1}{\rm{B}}\_{\rm{SIZ}}{{\rm{E}}_{{\rm{it}}}}+{\beta_2}{\rm{B}}\_{\rm{DI}}{{\rm{V}}_{{\rm{it}}}}+{\beta_3}{\rm{B}}\_}\cr{{\rm{TEN}}+{\beta_4}{\rm{B}}\_{\rm{CE}}{{\rm{O}}_{{\rm{it}}}}+{\beta_5}{\rm{IN}}{{\rm{D}}_{{\rm{it}}}}+{\beta_6}{\rm{CO}}{{\rm{C}}_{{\rm{it}}}}+{\beta_7}}\cr{{\rm{CSR}}\_{\rm{CO}}{{\rm{M}}_{{\rm{it}}}}++{\beta_8}{\rm{PRO}}{{\rm{F}}_{{\rm{it}}}}+{\beta_9}{\rm{LE}}{{\rm{V}}_{{\rm{it}}}}+{\beta_{10}}{\rm{SIZ}}{{\rm{E}}_{{\rm{it}}}}}\cr{+{\mu_{\rm{i}}}+{\lambda_{\rm{t}}}+{\varepsilon_{{\rm{it}}}}}\cr}

The dependent variable is Refinitiv's ESG combined score (ESGP). Although Refinitiv Eikon ESG scores are generally regarded as reputable and reliable (Shakil et al., 2020), ESG ratings face documented limitations, including weak agreement across providers (Berg et al., 2022), potential conflicts of interest (Tang et al., 2025), and the methodology changes over time (Berg et al., 2021; Bryl, 2026). Nevertheless, the ESG combined score is widely used in academic research as a standardised measure of firms' ESG performance that draws on both company-reported disclosures and publicly available information. Unlike the standalone ESG score, which reflects only self-reported ESG data, the ESG combined score adjusts for event-driven ESG controversies. Methodologically, it is the average of the ESG score and the ESG controversies score, both reported on a 0–100 scale. The ESG controversies score (23 metrics) captures a firm's exposure to ESG-related controversies reported in global media and is inverse-coded: higher values indicate fewer controversies. A score of 100 indicates no recorded controversies, whereas 0 reflects severe or repeated ESG issues (Refinitiv, 2024). This adjustment is important because ESG scores may not predict controversies; they often reflect potential risk rather than realised misconduct (Stewart, 2024). Therefore, an indicator that incorporates observed ESG controversies (rather than relying solely on company-reported data) can better capture firms' actual ESG performance, which is a focus of this study.

There are five independent variables in the model. The first independent variable, board size (B_SIZE), was estimated by the total number of board members at the end of the fiscal year. Second, board gender diversity (B_DIV) was calculated by the percentage of females on the board. ⁽²⁾ Third, average board tenure (B_TEN) was measured by the average number of years each board member has been on the board. Fourth, the CEO duality is a dummy variable investigating if the CEO simultaneously chair the board or has the chairman of the board been the CEO of the company in the past. Fifth, the proportion of independent board members (B_IND) was estimated by the percentage of independent board members as reported by the company. In addition, to ensure the validity of the model, the following control variables were included for this study: WACC (COC), presence of the CSR_COM, PROF, measured by ROA, LEV, measured as debt-to-equity ratio, and company size (SIZE), measured by natural logarithm of total assets at the end of the fiscal year. Data for independent and control variables were obtained from the Refinitiv database. Table 3 describes the variables included in this study.

Table 4 presents the descriptive statistics of the variables included in the study. The dependent variable, ESG performance (ESGP), has a mean value of 51.64 (SD = 17.05), ranging from 1.80 to 88.43, indicating substantial variation in ESG performance practices across firms. With respect to board characteristics, the average board size (B_SIZE) is 9.33 members (SD = 3.59), with boards ranging from 4 to 26 directors, suggesting heterogeneity in governance structures. Board gender diversity (B_DIV), measured as the percentage of female directors, averages 23.01% (SD = 14.85), with values between 0% and 71.43%, showing that while some boards are fully male-dominated, others achieve relatively high gender representation. Board tenure (B_TEN) averages 5.55 years (SD = 2.78), varying from <1 year (0.72) to almost 17 years, reflecting differences in board continuity and renewal across firms. CEO duality (B_CEO) is relatively rare, with a mean of 0.13, implying that in only 13% of the observations the same person serves as both CEO and board chair. Board independence (B_IND) averages 45.02% (SD = 26.29), with values ranging from 0% to 100%, demonstrating significant differences in the extent of independent oversight among firms. Regarding firm-level control variables, the presence of a CSR_COM is observed in 49% of firm-year cases, reflecting that approximately half of the companies institutionalise sustainability oversight at the board level. The mean COC is 0.06 (SD = 0.04), ranging between 0 and 0.36. PROF exhibits a mean of 0.06 (SD = 0.18), with a wide range from negative values up to 1.63, suggesting strong performance heterogeneity across the sample. LEV averages 1.02 (SD = 6.15), with minimum and maximum values of 0.00 and 223.12, respectively, indicating moderate financial risk. Finally, firm size (SIZE), measured as the natural logarithm of total assets, has a mean of 18.37 (SD = 2.32), with values ranging between 6.51 and 25.57.

To assess potential multicollinearity, a correlation analysis was first performed among the study variables (Table 5).

Although correlations between ESGP and several variables (B_SIZE, B_DIV, CSR_COM, and SIZE) were statistically significant, all remained below the commonly cited threshold of 0.8 (Mason & Perreault, 1991), indicating no evidence of serious multicollinearity concerns (Gujarati, 1995). Furthermore, variance inflation factors (VIFs) were calculated, with mean values reported alongside the regression results (Table 6). The individual VIFs ranged between 1.04 and 1.92, which is well below the threshold of 10 recommended by Chatterjee and Hadi (2013). Overall, both the correlation analysis and VIF diagnostics confirm that multicollinearity does not pose a concern in this study.

The determinants of ESG performance (ESGP) were analysed using generalised least squares (GLS) panel regression. Following Ellili (2022), Models (1)–(5) assessed individual board characteristics, while Model (6) examined their combined effect. Estimation proceeded with pooled OLS, fixed effects, and random effects specifications. Model-selection diagnostics indicated fixed effects for Models (1), (2), and (6): the Breusch-Pagan Lagrange Multiplier test favoured panel estimators over pooled OLS (Prob > χ² < 0.001), and the subsequent Hausman test rejected the random-effects null, suggesting the adoption of fixed effects. For Models (3), (4), and (5), the diagnostics were supported using random-effects panel regression. Subsequently, the Ramsey RESET test was performed on each specification to detect omitted-variable misspecification. Next, although further diagnostic checks did not reveal heteroscedasticity in all models except for Model (6) (Breusch-Pagan test, Prob > χ² > 0.05), the Wooldridge test demonstrated issues with serial correlation (Prob > F = 0.000) in each model. Therefore, to account for this, regressions were estimated with robust standard errors clustered at the firm level. The baseline GLS random effects results are presented in Table 6.

The baseline panel regression results for the board determinants of ESG performance (ESGP) demonstrate that estimates are consistent across models assessing the individual effects of board characteristics (Models 1–5) and their joint impact (Model 6). The within/overall R² values range from 0.19 to 0.48, suggesting that the models account for 19%–48% of the variation in ESG performance within firms over time, what suggests moderate explanatory power of the models. With respect to board characteristics, board size (B_SIZE) exhibits a consistently negative and statistically significant association with ESG performance in Models (1) and (6) (β = −0.979, p < 0.05; β = −0.909, p < 0.05). This indicates that larger boards are associated with lower levels of ESG performance, contrary to the expectations outlined in Hypothesis 1 (H1).

Board gender diversity (B_DIV) shows a positive but statistically insignificant effect in all specifications, suggesting that gender diversity does not exert a meaningful influence on ESG performance. Hence, Hypothesis 2 (H2) is not supported.

Board tenure (B_TEN) is positively related to ESG performance in Models (3) and (6), reaching significance at the 10% level in both cases (β = 0.598, p < 0.10; β = 0.712, p < 0.10). This result indicates that longer average board tenure may enhance ESG performance, thus contradicting Hypothesis 3 (H3).

In contrast, CEO duality (B_CEO) does not demonstrate a statistically significant effect in any model, implying that combining the CEO and chairperson roles does not systematically affect ESG performance. Thus, Hypothesis 4 (H4) is not supported.

Board independence (B_IND), however, shows a positive and statistically significant impact in both Models (5) and (6) (β = 0.096, p < 0.05; β = 0.096, p < 0.1). This indicates that boards with a greater proportion of independent directors are more likely foster corporate ESG performance, supporting Hypothesis 5 (H5). Regarding the control variables, the presence of a CSR_COM is strongly and positively associated with ESG performance across all specifications, with coefficients significant at the 5% and 10% level (ranging from β = 5.573 to β = 7.911). PROF has a negative effect; however, only significant in Model (5). Firm size (SIZE) and COC are positive and statistically significant in some specifications (Models (3), (4), and (5)), indicating that companies with larger assets and with higher WACC are more likely to perform better in terms of ESG, though the effect is not observed in the full model.

To assess the robustness and generalizability of the baseline findings, four additional models were estimated. Model (7) employs an alternative dependent variable; the ESG combined score was replaced with the ESG score. Model (8) introduces an alternative control variable by replacing WACC with COD. Model (9) integrates these alternative specifications by incorporating both the alternative dependent and independent variables within a single model. Finally, to address the problem of reverse causality, Model (10) provides lagged independent variables to address the problem of reverse causality. Hence, a 1-year lag for each board characteristic was employed. The results are presented in Table 7.

Across all three robustness models with alternative variable definitions, the baseline findings hold: board size remains significantly negative for ESG performance, while board tenure and board independence are significantly positive. However, once potential reverse causality is considered, the effect of board tenure loses significance. Consequently, any inference about the tenure-ESG performance relationship should be treated with caution.