According to agency theory, firms need to enhance their CG pillars to decrease agency conflicts among corporate management and stakeholder interests (Jensen & Meckling 1979). The most significant pillars of the CG framework are BGD and AC (Terjesen et al. 2015). In this context, several theories explain the role of corporate BGD and AC on CSP, such as social role theory, agency theory, stakeholder theory, legitimacy theory and critical mass theory (Oussii & Jeriji 2025). In this way, the social role theory argues that female leaders have natural characteristics, such as sensitivity and compassion for environmental issues more than male leaders (Aliani 2023). Drawing on this theory, Liu (2018) suggested that firms with greater female representation on the board are less sued for environmental regulation violations, probably due to ethical performance shown by female directors. The legitimacy theory suggests that firms are required to carry out their business activities in line with the norms prevailing in their societies (Wahyuningrum, Oktavilia & Utami 2022). Hence, the firms must have some accountability towards society. In this way, the firms should act responsibly and maturely to meet stakeholders’ expectations. Drawing on legitimacy theory, Akhter et al. (2023) argue that more BGD can accelerate firms’ reputation by enhancing corporate environmental responsibilities. The stakeholder theory suggests that the appointment of female managers in firms would reinforce the decision-making process and enhance communication among corporate board members, thereby affecting the CSP (Lara et al. 2017). The study’s theoretical framework is based on stakeholder theory, which helps to identify and understand the link between BGD, AC attributes and CSP (Oyewo 2023). According to this theory, the lack of corporate female representation and AC disclosures could be viewed as an adverse sign, as it may lead to a conflict of interest. Thus, female directors and AC can play a critical role in stimulating the CSP (Zhang, Zhu & Ding 2013). In this way, the firms may face pressures from stakeholders to increase the female representation on their boards and enhance corporate transparency to sustain their performance (Caby, Coron & Ziane 2022; Elleuch Lahyani 2022). Thus, BGD can play a positive role in promoting CSP by reducing their consumption of non-renewable resources and adopting more environmentally friendly technologies (Aliani 2023). In addition, the study is based on the critical mass theory to understand the critical mass of BGD on CSP. The essential idea of the critical mass is that the impact of BGD on corporate performance increases when their number reaches a specific threshold (Kanter 1977). Hence, this theory assumes that a critical mass of BGD ensures that their diverse experiences and skills are considered in formulating the firm policies and strategies, which in turn may affect CSP (Glass, Cook & Ingersoll 2016).

In line with stakeholder theory, several empirical papers have found that higher BGD in firms can play a positive role in mitigating the level of carbon emissions. Liu (2018) argues that firms with good female representation face fewer lawsuits for ecological issues and misconduct. Velte (2024) suggests that female managers are more aware about the needs of stakeholders, particularly about environmental issues. Konadu et al. (2022) argue that BGD is an important issue for reinforcing eco-innovation and sustainable practices. In testing the linkage among BGD and carbon emissions, Rjiba and Thavaharan (2022) suggest that BGD drives the reduction of carbon emissions in 43 selected economies. Recent studies conducted by Khatri (2024) in the case of UK firms, Oussii and Jeriji (2025) in the case of French firms and Gull et al. (2025) in the context of 36 selected countries found a positive role of BGD in mitigating the level of carbon emissions. Conversely, Alkurdi et al. (2023) used a sample of 1621 European companies from 2017 to 2021 and found that an increase in BGD increases the level of carbon emissions in these economies. In conclusion, several studies suggest that higher BGD is more effective in making decisions at board meetings, which in turn impacts corporate performance. However, the critical mass theory assumes that a certain threshold of minority group representation in the corporate sector is necessary to reinforce their role in the firm (Pathiranage, Waheduzzaman & Dayarathna 2025). Before reaching this threshold, minority groups are perceived as ‘tokens’ in firms under the control of majority male board members. The thresholds for female on board representation can be captured as numbers such as ‘two or more’ members (Iqbal & Saeed 2025) or as a percentage of total members such as more than ‘10 or 20’% (Burkhardt, Nguyen & Poincelot 2020). Recently, some studies have examined the role of the critical mass of BGD on CSP. Toukabri and Jilani (2023) used a sample of US firms listed from 2011 to 2018 and found that a critical mass of female leaders positively affects carbon emission deductions. Al-Shaer, Zaman and Albitar (2024) used a sample of UK firms from 2011 to 2019 and found that BGD critical mass contributes to CSP by improving ESG performance:

The current empirical studies show that AC attributes can reinforce the reliability of financial reports and align firms’ decisions with stakeholders’ interests, leading to promoting CSP (Alkurdi et al. 2023; Arif et al. 2021; Meqbel et al. 2025; Pozzoli et al. 2022). In this context, Arif et al. (2021) found that a positive role of AC independence in enhancing the ESG ratings. Pozzoli et al. (2022) also support the AC independence reinforces ESG performance. According to Pozzoli et al. (2022), AC-independent members guarantee better monitoring activities in firms, leading to the adoption of more sustainable practices. Alkurdi et al. (2023) suggest that AC independence plays a positive role in mitigating the level of CO₂ emissions among European companies. Meqbel et al. (2025) argue that AC independence has a positive effect on environmental disclosure, underscoring the importance of AC independence in reinforcing the company transparency and accountability regarding ecological performance. These studies suggest that ACI can promote CSP by establishing robust internal control systems within company, thereby mitigating business risk.

Small and large firms may differ in their implementation of governance practices, leading to different impacts on the CSP. In this context, Agnese et al. (2025) suggest that smaller firms frequently encounter obstacles, such as a lack of sustainability expertise or limited financial resources. Meanwhile, larger firms may struggle to balance the needs of different internal and external stakeholders. Chen et al. (2026) highlighted that large firms have more capabilities to promote CSP strategies, such as more capital access, and efficient technological support and teams. These capabilities can help these firms to implement more CSP practices. Empirical studies suggested that firm size affects the effectiveness of AC and BGD. In this way, Kwamboka, Githaiga and Kinuthia (2025) suggested that firm size affects a firm’s internal CG, potentially influencing the scope of auditing. For instance, large firms may require more auditing efforts and processes than smaller firms. Joecks, Pull and Vetter (2013) found that critical mass of about 30% women is positively linked with higher firm performance. The author found this threshold is more observed in larger firms. On the other hand, some studies assessed the impact of firm size on CSP. For example, Ahmad, Mobarek and Raid (2023) used data from 2002 to 2018 in the context of 351 UK companies and found that firm size moderates the linkage of financial crisis and CSP. Al-Sarraf, Al-Swidi and Al-Hakimi (2025) used UK firms spanning 2011–2021 and found that the influence of BGD on corporate performance varies by the company size. Qureshi et al. (2020) argue that larger firms may have the capacity and financial source to invest more in sustainability activities, leading to higher CSP score. Most of these studies ignored the role of firm size on the nexus among critical mass of BGD, ACI and carbon emissions. To the best of the author’s knowledge, the present work is the first in the literature that assesses the effect of firm size on this relationship.

This work aims to explore the effect of BGD and AC characteristics on carbon emissions performance in the context of G7 firms. The empirical analysis of this work is based on a final sample of 2180 non-financial listed firms over the period from 2015 to 2024, drawn from an initial dataset of 13 829 firms. After excluding firms due to missing data, the refined final sample includes 10 592 firm-year observations. We exclude financial firms because their financial structures, processes and requirements differ significantly from non-financial firms. In particular, financial firms operate under different regulatory and disclosure frameworks. Excluding financial firms enhances the clarity and reliability of the study findings. In addition, firms with missing data were dropped from the sample, such as CG, CO₂ emissions and financial information. In addition, we have excluded firms that do not have data in 2022, 2023 and 2024. Table 1 reports the number of observations per country and industry. The study’s data (CO₂, AC characteristics, BGD and firm size) are collected from the London Stock Exchange Group website (LSEG) database. Gross domestic product is collected from the World Bank Database.

To assess the impact of critical mass of BGD and ACI on carbon emissions (direct, indirect and total), the following econometric model is formulated as follows (Equation 1): CO2 it=α0+α1BGDit+α2ACIit   +α3CMBGDit+α4IBMit+α5BSit   +α6FLit+α7CEit+α8MVit+α9FSit   +α10GDPit+α11Year effects   +α12Firm effects+∈it[Eqn 1]

To assess the impact of firm size on the linkage between critical mass of BGD, ACI and carbon emissions, the following econometric model is formulated as (Equations 2 and 3): CO2 it=α0+α1FSit+α2CMBGDit+α3FS*CMBGDit   +α4IBMit+α5BSit+α6FLit+α7CEit+α8MVit   +α9FSit+α10GDPit+α11Year effects   +α12Firm effects+∈it[Eqn 2] CO2 it=α0+α1FSit+α2ACIit+α3FS*ACIit   +α4IBMit+α5BSit+α6FLit+α7CEit   +α8MVit+α9FSit+α10GDPit   +α11Year effects+α12Firm effects+∈it[Eqn 3]

In the above equations, i and t represent firms and the time horizon (year), α means the slope of the selected independent control variables and the constant. CO_{2 total} represents total carbon emissions, CO_{2 direct} are and CO_{2 Indirect} are direct and indirect CO₂ emissions, respectively. Critical mass of Board Gender Diversity (BGD) and Audit Committee Independence (ACI) are the main independent variables of this study. Independent board members (IBM), board size (BS), firm leverage (FL), capital expenditures (CE), market value (MV), firm size (FS), and GDP are control variables in this study. Independent board members and BS are independent board members and board size, respectively. FL, CE and MV represent firm leverage, capital expenditures and market value, respectively. FS and GDP in the above equation represent firm size and country GDP, which is captured by total GDP in USD. The summary of the variables measurements and sources are reported in Appendix 1 Table 1-A1. To examine the study hypotheses, we used the fixed effects method to estimate the coefficients. This approach controls time-invariant unobserved heterogeneity. The article uses Hausman assessment to evaluate whether a fixed effects or random effects method is suitable for testing the models. The findings of this assessment displayed in Table 3 show that p-value of Hausman assessment is significant, suggesting that the fixed effects model is the relevant model (Meqbel et al. 2025). Further, we use robust standard errors clustered at firm level. Besides, we use firm and year fixed effects. The firm fixed effects aim to absorb time-invariant unobserved heterogeneity (Zhang et al. 2026). While year fixed effects control for common macroeconomic conditions, changes regulations and aggregate shocks affecting the firms in a given year. All the focused variables of this work are winsorised at ‘1% and 99%’ levels to delete the effect of outliers and decrease the effect of abnormal values. We performed all tests and regressions in STATA version 16.

This article followed all ethical standards for research without direct contact with human or animal subjects.

The summary of descriptive statistics (Table 2) shows that the mean total carbon emissions value (in log) is 10.816, with a standard deviation of 2.097. Additionally, the outcomes indicate that the means of AC-independent members and AC financial expertise are 50.985 and 53.932, respectively. Table 2 also indicates that mean of BGD in G7 economies is 25.279%, suggesting that these countries have a good representation of female compared to other economies such as OECD (Farhan et al. 2025). Furthermore, Appendix 1 Table 2-A1 reports that correlation coefficients (BGD, ACI, ACE, IBM, BS, FL, CE, MV, FS, CSRS and GDP) with carbon emissions (CO₂) are lower than 0.80, suggesting multicollinearity may not be an issue because the associations between the predictors were below the criterion of 0.80. Multicollinearity between the variables can be observed when the VIF value for each variable is more than 10 (Cheng et al. 2022). According to Appendix 1 Table 2-A1, all the variables are lower than ‘10’, implying the absence of multicollinearity in the tested variables.

Table 3 reports the fixed effects outcomes. In column 1 of Table 3, the regression outcomes report the impact of BGD and selected control variables on carbon emissions (direct, indirect and total emissions). Board gender diversity in column 1 is negative (0.035) and significant at the 1% level, suggesting that a 1% point increase in the percentage of the female on board is associated with a 0.035% reduction of CO₂ emissions. These outcomes are in line with the theory of stakeholder, which suggests that more BGD can positively affect CSP by reinforcing the communication among the corporate board members. Hence, these findings suggest that female leaders are more likely to be involved in green activities and projects that mitigate harmful environmental effects than males. The outcomes support the works of Rjiba and Thavaharan (2022) and Kreuzer and Priberny (2022), which suggest that BGD drives CSP in different selected countries by mitigating the level of CO₂ emissions performance. We estimated nonlinear linkage using a spline model with knots at 10%, 20% and 30%. The spline results (column 1 of Appendix 1 Table 3-A1) show that coefficients of BGD-S (≤ 10%) and BGD-S2 (10% – 20%) segments are insignificant. Meanwhile, the coefficients for the 20% – 30% and > 30% segments are positive and statistically significant. These findings suggest that critical mass of BGD is reached approximately 20% and remains significant beyond 30%. Further, we then re-estimated a threshold model separately (Appendix 1 Table 3-A1 panel II and Table 3). In columns 2–4 of Table 3, we report the impact of critical mass of BGD (captured as a dummy variable, 1 if female representation ≥ 20% and 0 otherwise) on carbon emissions (direct, indirect and total emissions). The outcomes illustrate a significant and negative linkage among critical mass of BGD and carbon emissions, suggesting that firms with at least 20% female representation in the board are associated with carbon emissions reduction. To further examine the critical mass hypothesis, we have tested the impact of female representation in the board below 20% on carbon emissions. The findings show that the effect of female representation below 20% has a limited effect on CSP, supporting that token representation of women (female representation in the board with < 20% representation) limits their effectiveness in promoting CSP. These outcomes are in line with the theory of critical mass, confirming that a critical mass of BGD reinforces monitoring efforts, supports firm value and stimulates innovative practices. These findings also support prior empirical literature, which suggests that there is a threshold for the number of women directors needed to reinforce the CSP (Birindelli, Iannuzzi & Savioli 2019; Nuber & Velte 2021). For example, the findings are in line with Nuber and Velte (2021), who argued that a critical mass of BGD has a more powerful impact on carbon emissions performance in non-financial firms in Europe and Birindelli et al. (2019) who confirmed that a critical mass of BGD is a driver of CSP. On the other hand, the findings from Table 3 show that ACI has a negative and significant effect on corporate carbon emissions in the tested countries (direct, indirect and total emissions). These findings suggest that the more independent AC members promote CSP. More specifically, the coefficient of ACI in column 1 is negative (0.007) and statistically significant at the 1% level, suggesting that a 1% point increase in ACI is associated with a 0.007% reduction in direct and indirect carbon emissions, and these findings suggest that the second hypothesis is supported in context of G7 firms. These outcomes corroborate with the prior empirical literature. For example, Meqbel et al. (2025) and Alkurdi et al. (2023) confirmed the positive role of ACI in promoting CSP in European firms. However, the significant and positive linkage connection among ACI and CSP performance affirms that the more ACI members reinforce monitoring and control activities to meet the stakeholder interests and needs. These outcomes are in line with stakeholder-agency theories, which suggest that firms need to enhance CG mechanisms.

Moving to control variables, we found that both independent board members and board size have an adverse effect on CSP, increasing the level of carbon emissions (direct, indirect and total). In addition, the findings show that firm leverage has an insignificant effect on direct CO₂ emissions performance. Further, we found that increasing capital expenditures and total assets in firms increases the level of carbon emissions. On the other hand, the study shows that an increase in the country’s economic growth has limited effects on the level of corporate carbon emissions. Besides, the study finds that market value has a limited effect on indirect carbon emissions. The findings also find the financial leverage has a limited effect on direct carbon emissions.

Table 5 shows the moderating effect of firm size on the linkage between critical mass of BGD and carbon emissions (direct, indirect and total). The outcomes show that firm size moderates this linkage, suggesting that the third hypothesis of this study is supported in the context of G7 firms. These outcomes may be attributed to the fact that the larger firms face stronger monitoring and scrutiny by the governments and the stakeholders to adopt more governance practices (Githaiga, Muturi Kabete & Caroline Bonareri 2022), which in turn lead to an increase in the share of BGD in their firms, and increases their effectiveness role in decision-making process. Subsequently, it leads to promote CSP. These findings align with Abdullah, Ardiansah and Hamidah (2017), who argue that larger companies have stronger boards, characterised by higher representation of women on their boards. Furthermore, the findings show that the coefficient of FS*ACI is negative and insignificant, highlighting that the moderating effects of firms’ size on the linkage between ACI and CSP is statistically insignificant. These findings suggest that the fourth hypothesis of this study is not supported in the case of G7 firms. These findings suggest that ACI can play a positive role in mitigating the level of carbon emissions, regardless of firm size, underscoring the importance of ACI in stimulating CSP. These findings could be attributed to ACI that plays a significant role in enhancing the transparency and accountability about the sustainable practices in the firms regardless of the firms’ size. Increasing ACI members may mitigate the levels of the conflicts between the companies and their shareholders, thereby enhancing the alignment of management interests with shareholders’ interests, which in turn will sustain corporate performance regardless of the firm size.

We use the S-GMM to offer more robust findings and confirm the main findings of this study. This method is a consistent and efficient approach to address the endogeneity issue, which arises from the correlation among the lagged dependent variable and the error term (Kitulazzi et al. 2025). The study uses the Arellano–Bond (AR) and Hansen assessments to evaluate the instrument variables’ validity in S-GMM. Hansen’s test aims to confirm that the external independent variables in the tested models are uncorrelated with the errors. The AR aims to evaluate and detect any autocorrelation in the tested model (Assfaw & Sharma 2024). The findings from Table 4 report that the p-values of the AR (2) are statistically insignificant, which confirms the absence of autocorrelation. Furthermore, the p-values of Hansen’s tests are insignificant, confirming the instrumental variables are valid. The S-GMM findings are displayed in Table 4. The lag of dependent variables (CO_{2 Total}, CO_{2 Direct} and CO_{2 Indirect}) is significant, suggesting persistence in carbon emissions over time (Assfaw & Sharma 2024; Okoyeuzu et al. 2021). After addressing the endogeneity, the findings show that the coefficients of BGD and ACI remain negative and significant, confirming the positive role of BGD and ACI in driving CSP. To affirm the main findings of this study, we have used several tests. In this way, we have assessed the impact of critical mass of BGD and ACI on carbon emissions scaling by capital expenditures. The findings from column 1 of Table 6 shows that critical mass of BGD and ACI adversely affect the level of carbon emissions scaling by capital expenditures. On the other hand, the share of female on board may show a high degree of homogeneity in terms of board gender. In this way, we use the BLAU index as an alternative measurement of BGD, which takes into account the maximum value when each category share is at a maximum. This proxy ranges from ‘0 to 0.5’. The findings from Table 6 show that the BLAU coefficient is negative, confirming the positive role of BGD in sustaining the CSP in the context of G7 countries. In addition, we have examined the moderator role of CSR strategy on the nexus among critical mass of BGD and carbon emissions. The findings from column 4 of Table 6 show that CSR strategy moderates the nexus among critical mass of BGD and carbon emissions, highlighting the importance of sustainable strategy practices in promoting CSP. In addition, we have examined the impact of AC financial expertise on carbon emissions, and we found that AC with financial expertise mitigates the level of corporate CO₂ emissions. These findings highlight the importance of the AC committee in mitigating the level of carbon emissions performance in the context of G7 firms.