This research seeks to examine the direct impact of audit committee characteristics on carbon emission practices. Furthermore, it attempts to find how corporate social responsibility committee (CSRC) moderates the association between audit committee characteristics on carbon emission practices.
The data for 5,480 firms were retrieved from the Refinitiv Eikon Database for the period from 2017 to 2022. For analyzing the data, the study utilized fixed-effect (FE) and random-effect (RE) estimation methods. Moreover, the study used two-stage least squares (2SLS) estimation and generalized method of moments (GMM) as robust tests to confirm the findings.
Results revealed that the audit committee positively and significantly affects carbon emission practices. Regarding the moderation effect, it is revealed that CSRC positively and significantly moderates the association between audit committee and carbon emission practices.
This research offers a novel contribution to the existing literature by focusing on emerging economies, and it offers a distinct perspective by examining how the presence of a CSRC influences the relationship between audit committee and carbon emissions practices. Further, it provides significant recommendations for policymakers, regulators, board members, auditors, analysts and academics regarding carbon emissions issues. This emphasizes the need to strengthen stakeholder engagement mechanisms involving regulatory bodies, external and internal auditors, shareholders and board oversight to promote ethical and responsible practices.
1. Introduction
In recent years, carbon emissions in the Organization for Economic Cooperation and Development (OECD) countries have become a prominent global issue because of economic development and increased industrialization activities in these countries (Sahu et al., 2025). Researchers argue according to world energy report 2020, developing countries such as South Korea, Saudi Arabia, India, Iran, Japan and China are the major contributors to the increase in carbon emission (Friedlingstein et al., 2020; Lyu et al., 2023). In this context, OECD countries are seeking to adopt innovative technologies, set robust environmental policies and encourage green investment to reduce global carbon emissions (Ganda, 2019). Shen and Wang (2023) believe that on the national level, governments should implement carbon-price schemes and support renewable energy investment to reduce carbon emission. On the corporate level, firms have to be more concerned about climate change in general and carbon emission reduction in particular. For that firms must assign the role of oversight to the audit committee to ensure that there is compliance with environmental regulations and appropriate disclosure on carbon emissions is made. Ika et al. (2021) argue that audit committee independently verifies and oversees firms’ environmental effects. Hence, firms can use the environmental positive effects as a signal to attract more investors (Ammer et al., 2020). Thus, researchers assume that the presence of corporate social responsibility committee positively moderates the influence of audit committee on firms’ performance, which is better illustrated by agency theory.
According to Smith (1937) the relationship between agent and principal is referred to as agency theory. Panda and Leepsa (2017) believe that managers may not run the business according to the owner’s best interests. Hence, audit committees play a crucial role in reducing agency conflicts between owners and managers regarding social issues (Said et al., 2009). Similarly, Mita et al. (2024) believe that corporate social responsibility committee integrates environmental and sustainability goals into corporate governance to reduce carbon emissions. Further, Elmaghrabi (2021) affirms that the existence of CSRC in business is crucial for enhancing corporate social responsibility activates. This suggests that, while CSRC enhances audit committee effectiveness in reducing carbon emissions, further empirical research is required to determine how well CSRC moderates the association between audit committee and carbon emissions.
Building on this discussion, our study seeks to address this gap by exploring how the presence of CSRC influences the relationship between audit committees and carbon emission practices. This study contributes to literature in several ways. First, although various studies have examined the direct relationship between audit committee and carbon emission practices, e.g. (Chariri et al., 2018; Widagdo et al., 2022), little attention has been paid to the interaction effects of CSRC. Second, this study contributes to the evolving literature on corporate governance and sustainability by elucidating the audit committee’s role in influencing carbon emissions practices in OECD Countries. Given the fact that, in 2013, OECD countries accounted for 85% of global GDP and were responsible for 76% of global CO2 emissions (OECD, 2015; Paramati et al., 2021). Third, the findings offer evidence based on present and large datasets of 6,063 enterprises from 34 OECD nations. Further, majority of studies conducted in OCED countries rely on relatively small samples, such as Amin et al. (2021), 225 companies; Ghachem et al. (2022), 192 country-year observations; and Hamad and Cek (2023), 625 companies. Fourth, this study offers significant insights into various aspects of carbon emissions for regulators, policymakers, board members, auditors, analysts, and scholars.
The findings reveal that audit committee has a positive and significant impact on carbon emission practices. Moreover, CSRC positively and significantly moderates the association between audit committee and carbon emission practices. Given these findings, this study offers significant perspectives on many carbon emission-related matters to regulators, policymakers, auditors, board members, analysts, and scholars. First, the authors believe that managers should include an audit committee in their long-term strategic planning to guarantee that the company model incorporates sustainability, especially carbon reduction. Second, long-term investment potential is presented to investors by businesses with robust audits and CSRC, which prioritizes carbon reduction. Third, the authors argue that prioritizing companies with strong sustainability governance can help investors reduce the risks and profits from market trends toward eco-friendly activities. Moreover, lawmakers should encourage audit and CSRC to ensure that businesses consider sustainability.
The remainder of this article is organized as follows. The second section provides a comprehensive overview of the relevant literature by highlighting previous research findings, theoretical frameworks, and key concepts. The third section illustrates research design. The fourth and fifth section demonstrate data analysis and discussion. The last section summarizes the key findings of the study.
2. Literature review
2.1 Theory development
Agency theory is a discipline of economics that studies the interaction between a company’s principals (shareholders) and agents (managers) (Shapiro, 2005). One application of agency theory to carbon emissions is the potential misalignment of interests between shareholders and managers in terms of environmental responsibility (Dalton et al., 2007). Shareholders may emphasize immediate financial returns and profit maximization, while managers may be incentivized to disregard environmental issues to reduce expenses or achieve performance objectives. To diminish carbon emissions and advocate for sustainable practices, corporations can form an audit committee that is essential in overseeing and alleviating these conflicts (Dalton et al., 2007).
Moreover, agency theory provides a framework for understanding the link among CSRC, audit committees, and carbon emissions practices. Audit committees, essential to corporate governance frameworks, are pivotal in mitigating agency problems (Braiotta, 2003) and significantly enhance board efficacy and governance procedures (Brennan and Kirwan, 2015). The attributes of the audit committee can affect corporate social responsibility (CSR) initiatives (Mohammadi et al., 2021), indicating that the effectiveness of an audit committee can determine the level of CSR activities within organizations (Akhtaruddin and Haron, 2010). The frequency of board and audit committee meetings is shaped by agency theory (Greco, 2011), and compliance with audit committee mandates is essential for enhancing governance procedures (Braiotta, 2003). Agency theory offers a theoretical framework for comprehending how governance structures, such as audit committees, influence corporate social responsibility activities, including carbon emission strategies, within organizations.
2.2 Institutional settings
The existence of diverse governance structures, including audit committees, institutional ownership, and corporate governance frameworks, significantly impacts organizational behaviors and outcomes. Institutional investors and the structure of governance bodies, including audit committees, significantly impact companies’ decision-making processes and performance metrics. The importance of the institutional context in governance research is highlighted by Zgarni et al. (2016), who assert that the efficacy of audit committees and the robustness of institutions can either complement or replace one another to bolster investor confidence in reported earnings. Braik and Al-Thuneibat (2023) asserted that institutional investors are crucial for supervising corporate governance, hence enhancing the efficacy of the audit committee. Furthermore, Takain and Hidayah (2024) emphasize that a higher percentage of institutional ownership may lead to improved oversight, thus augmenting the financial performance of businesses. Numerous studies have elucidated the impact of institutional elements on carbon emissions reduction techniques and the significance of governance mechanisms in attaining sustainability objectives. Feng and Xie (2024) underscore the significance of regulatory frameworks and financial mechanisms in advancing sustainable practices, stressing the impact of financial decentralization and institutional policies on fostering environmental enhancements. Moreover, Boateng et al. (2023) assert that the role of institutions in mitigating carbon emissions is vital, as the efficacy of emission reduction efforts hinges on elements such as regulatory quality, impartiality, administrative processes, and licensing limitations. These findings underscore the necessity for robust institutional frameworks to enhance environmental endeavors.
2.3 Audit committee and carbon emission practices
Carbon emission reporting standards in OECD nations are essential for tracking and monitoring advancements in the reduction of greenhouse gas emissions and the fight against climate change (Hannasvik et al., 2023; Aldy and Stavins, 2012). The audit committee is tasked with evaluating business disclosures on carbon emissions and sustainability (Chariri et al., 2018). The audit committee can enhance openness, accountability, and sustainability regarding carbon emissions (Salleh et al., 2022; Annisa Anafiah et al., 2017). Wang et al. (2024) assert that the publication of climate risk information can mitigate carbon emissions. Furthermore, Li et al. (2024) illustrate that government subsidies stimulate the growth of investment and consumption to an equivalent extent of pessimism. Moreover, organizations that regularly have meetings can attain superior carbon emissions disclosure (Widagdo et al., 2022). The committee supervises the incorporation of environmental factors into financial reporting, facilitates ongoing enhancements in reporting procedures, and is crucial in stakeholder communication. The audit committee aids in preserving the integrity of business disclosures, improving transparency, and cultivating trust among stakeholders. Therefore, we propose the subsequent hypothesis:
Audit committee has a positive association with carbon emissions practices.
2.4 The moderation effect of CSR committee
Research highlights the moderating function of CSRC in enhancing environmental performance, indicating its potential effect on organizational sustainability practices (Uyar et al., 2022). Furthermore, the association between Corporate Social Responsibility Committee (CSRC) and corporate social performance has been shown to enhance social sustainability and reduce carbon emissions, underscoring the critical function of CSRC in fostering environmental and social sustainability (Saeed et al., 2021). These findings cumulatively underline the significance of governance frameworks like CSRC in facilitating environmental disclosures, moderating environmental performance, and enhancing social sustainability, hence clarifying the connection between CSRC and carbon emission behaviors. Kılıç and Kuzey (2019) assert that the existence of a CSRC markedly affects carbon emission declarations in organizations. The presence of a CSRC committee is strongly correlated with environmental sustainability disclosure (Helfaya and Moussa, 2017). In summary, the CSR committee serves as a crucial element in mediating the connection between the audit committee and carbon emissions practices by affecting environmental sustainability disclosure, guaranteeing thorough CSR reporting, and executing sustainable business practices to mitigate organizational carbon footprints. In light of the preceding discourse, the subsequent hypothesis was determined:
CSR committee positively moderates the association between audit committee and carbon emission practices.
3. Methodology
3.1 Data collection and sampling
This study examines the impact of audit committee on carbon emission practices in OECD countries, while accounting for factors such as GDP, business size, leverage, and return on assets (ROA). Table 3 shows the study variables, their definations and their support from previous literature. We employed a recent dataset covering 2017 to 2022, based on data availability. The final sample included 5,480 enterprises from 34 OECD nations for a total of 32,880 observations (companies*years). Financial and governance data were extracted from the Refinitiv Eikon Database. At the same time, GDP and inflation were retrieved from the World Bank database. This study includes firms from different sectors, as shown in Table 1.
Tabulation of sector
| TRBC economic sector name | Without CSR committee | With CSR committee | |||||||
|---|---|---|---|---|---|---|---|---|---|
| No obs. | Percent | Cum. | No obs. | Percent | Cum. | No obs. | Percent | Cum. | |
| Technology | 5,052 | 15.36 | 15.36 | 3,143 | 16.8 | 98.57 | 1,909 | 13.47 | 95.96 |
| Healthcare | 4,902 | 14.91 | 30.27 | 3,760 | 20.1 | 61.81 | 1,142 | 8.06 | 58.75 |
| Consumer cyclicals | 4,860 | 14.78 | 45.05 | 2,776 | 14.84 | 21.97 | 2,084 | 14.71 | 27.44 |
| Industrials | 4,854 | 14.76 | 59.81 | 2,482 | 13.27 | 75.08 | 2,372 | 16.74 | 75.49 |
| Financials | 3,252 | 9.89 | 69.7 | 2,008 | 10.73 | 41.71 | 1,244 | 8.78 | 50.7 |
| Basic materials | 3,018 | 9.18 | 78.88 | 1,251 | 6.69 | 7.13 | 1,767 | 12.47 | 12.73 |
| Real estate | 2,244 | 6.82 | 85.7 | 1,253 | 6.7 | 81.77 | 991 | 6.99 | 82.49 |
| Consumer non-cyclicals | 2,034 | 6.19 | 91.89 | 902 | 4.82 | 26.79 | 1,132 | 7.99 | 35.42 |
| Energy | 1,704 | 5.18 | 97.07 | 784 | 4.19 | 30.98 | 920 | 6.49 | 41.92 |
| Utilities | 840 | 2.55 | 99.62 | 267 | 1.43 | 100 | 573 | 4.04 | 100 |
| Academic and educational services | 120 | 0.36 | 100 | 83 | 0.44 | 0.44 | 37 | 0.26 | 0.26 |
| Total | 32,880 | 100 | 18,709 | 100 | 14,171 | 100 | |||
For several reasons, we choose to include sample corporations from OECD nations in our analysis. First, with a total score of 77.8, OECD countries have achieved remarkable SDG achievement, outperforming the worldwide average by 66.7 (Sachs et al., 2022). Second, all G20 countries are included in the OECD, and the majority of OECD members have a population of more than 100 million people (Sachs et al., 2022). Third, the OECD actively assists the UN and its agencies in implementing the SDGs. Table 2 provides comprehensive details of the chosen nations and Table 3 demonstrates variables definition and previous research that support their selection.
Tabulation of country
| Pooled | With CSR committee | Without CSR committee | |||||||
|---|---|---|---|---|---|---|---|---|---|
| No of obs. | Percent | Cum. | No of obs. | Percent | Cum. | No of obs. | Percent | Cum. | |
| United States of America | 15,522 | 47.21% | 47.21% | 4,938 | 34.85% | 34.85% | 10584.00 | 56.57% | 56.57% |
| United Kingdom | 3,360 | 10.22% | 57.43% | 1,690 | 11.93% | 46.77% | 1670.00 | 8.93% | 65.50% |
| Japan | 2,550 | 7.76% | 65.18% | 1,784 | 12.59% | 59.36% | 766.00 | 4.09% | 69.59% |
| Sweden | 2,064 | 6.28% | 71.46% | 1,051 | 7.42% | 66.78% | 1013.00 | 5.41% | 75.01% |
| Canada | 2,052 | 6.24% | 77.70% | 1,156 | 8.16% | 74.93% | 896.00 | 4.79% | 79.80% |
| Australia | 1,824 | 5.55% | 83.25% | 762 | 5.38% | 80.31% | 1062.00 | 5.68% | 85.47% |
| Korea; Republic (S. Korea) | 1,410 | 4.29% | 87.54% | 821 | 5.79% | 86.11% | 589.00 | 3.15% | 88.62% |
| Poland | 780 | 2.37% | 89.91% | 206 | 1.45% | 87.56% | 574.00 | 3.07% | 91.69% |
| Norway | 480 | 1.46% | 91.37% | 178 | 1.26% | 88.82% | 302.00 | 1.61% | 93.30% |
| Chile | 426 | 1.30% | 92.66% | 276 | 1.95% | 90.76% | 150.00 | 0.80% | 94.10% |
| Turkey | 384 | 1.17% | 93.83% | 235 | 1.66% | 92.42% | 149.00 | 0.80% | 94.90% |
| Belgium | 372 | 1.13% | 94.96% | 200 | 1.41% | 93.83% | 172.00 | 0.92% | 95.82% |
| Finland | 324 | 0.99% | 95.95% | 186 | 1.31% | 95.15% | 138.00 | 0.74% | 96.56% |
| Denmark | 312 | 0.95% | 96.90% | 180 | 1.27% | 96.42% | 132.00 | 0.71% | 97.26% |
| Italy | 252 | 0.77% | 97.66% | 124 | 0.88% | 97.29% | 128.00 | 0.68% | 97.95% |
| Austria | 156 | 0.47% | 98.14% | 99 | 0.70% | 97.99% | 57.00 | 0.30% | 98.25% |
| Israel | 156 | 0.47% | 98.61% | 87 | 0.61% | 98.60% | 69.00 | 0.37% | 98.62% |
| Colombia | 132 | 0.40% | 99.01% | 71 | 0.50% | 99.10% | 61.00 | 0.33% | 98.95% |
| New Zealand | 84 | 0.26% | 99.27% | 20 | 0.14% | 99.24% | 64.00 | 0.34% | 99.29% |
| Netherlands | 66 | 0.20% | 99.47% | 29 | 0.20% | 99.45% | 37.00 | 0.20% | 99.49% |
| Czech Republic | 48 | 0.15% | 99.62% | 23 | 0.16% | 99.61% | 25.00 | 0.13% | 99.62% |
| Estonia | 30 | 0.09% | 99.71% | 14 | 0.10% | 99.71% | 16.00 | 0.09% | 99.71% |
| Mexico | 30 | 0.09% | 99.80% | 17 | 0.12% | 99.83% | 13.00 | 0.07% | 99.78% |
| Costa Rica | 18 | 0.05% | 99.85% | 12 | 0.08% | 99.92% | 6.00 | 0.03% | 99.81% |
| Latvia | 18 | 0.05% | 99.91% | 2 | 0.01% | 99.93% | 16.00 | 0.09% | 99.89% |
| Luxembourg | 12 | 0.04% | 99.95% | 7 | 0.05% | 99.98% | 5.00 | 0.03% | 99.92% |
| Spain | 12 | 0.04% | 99.98% | 3 | 0.02% | 100.00% | 9.00 | 0.05% | 99.97% |
| Slovenia | 6 | 0.02% | 100.00% | 0 | 0.00% | 100.00% | 6.00 | 0.03% | 100.00% |
| 32,880 | 100% | 14,171 | 100% | 18,709 | 100.00% | ||||
Variable definition
| Variable name | Symbol | Definition | Previous research |
|---|---|---|---|
| Carbon emission | Carbon Emission | It is a quantitative indicator of how committed and effective a business is at limiting emissions during its operational and manufacturing processes | Aliani (2023), Tanthanongsakkun et al. (2023) |
| Audit committee | Is the independent variable; 1 if the firm has an audit board committee and 0 if it does not have an audit board committee | Simamora and Elviani (2022), Wang and Sun (2022) | |
| CSR committee | Is the moderator variable; 1 if there is a CSR committee and 0 if there is no CSR committee | Uyar et al. (2022), Meqbel et al. (2024) | |
| Interaction term | The interaction term is calculated by centering the independent and moderator variables by subtracting the original values from their means. Then multiply the centered dependent and moderator variables | Kim et al. (2001) | |
| Return on assets | Net income as a percentage of total assets | Farhan (2024), Afrifa and Padachi (2016) | |
| Firm size | Natural logarithm of total assets | Afrifa and Padachi (2016), Tran et al. (2017) | |
| Leverage | Total debt/shareholder’s equity | Afrifa and Padachi (2016), Mehta (2017), Farhan (2024) | |
| Gross domestic product | GDP | The market worth of all the finished products and services produced by a nation in a certain period is expressed in monetary terms as the gross domestic product | Bagadeem et al. (2024), Jakob (2016) |
| Inflation | INF | The rate of change in prices over a specified period | Bagadeem et al. (2024) |
| Board size | BDSIZE | Total number of board members | Nuber and Velte (2021) |
| Board gender diversity | BDGEND | The proportion of the presence of women on board | Nuber and Velte (2021) |
3.2 Statistical tests
This study examines the association between audit committee and carbon emission practices using widely used econometric estimation methodologies, such as Random Effects (RE) and Fixed Effects (FE) approaches. Individual country-level, sector-level, and time-specific impacts from 34 OECD countries were included in the panel data. Tests for specifications were also carried out, including the Hausman and Breusch-Pagan LM tests. The Huber-White and Driscoll-Kraay tests were used in addition to the Two-Stage Least Squares (2SLS) and Generalized Method of Moments (GMM) techniques to verify the validity of the results. The following models are formulated to assess the direct effect of audit committee and the moderation effect of the CSR committee:
4. Analysis and discussion
4.1 Descriptive statistics
Table 4 shows the central tendencies of the dataset across all variables. The table shows that the mean and median carbon emissions values were 0.361 and 0.306, respectively, with a standard deviation of 0.335. The mean and median values for the audit committee were 0.811 and 1, respectively, with a standard deviation of 0.391. In the case of return on assets, firm size, and leverage, the mean values are −0.015, 9.226, and 0.255, respectively. Concerning gross domestic products and inflation, the results in Table 4 show that the mean values are 10.841 and 3.272, respectively. Moreover, all VIF values shown in the descriptive statistics were less than 4. Furthermore, most of the tolerance values (1/VIF) in our data were closer to 1, suggesting the absence of multicollinearity.
Descriptive statistics
| Variable | Obs | Mean | Median | Std. dev. | Min | Max | Swilk | p-value | VIF | 1/VIF |
|---|---|---|---|---|---|---|---|---|---|---|
| EMISSION | 32,880 | 0.361 | 0.306 | 0.335 | 0 | 0.999 | 17.810 | 0.000 | ||
| AUDIT COM | 32,880 | 0.811 | 1 | 0.391 | 0 | 1 | 2.176 | 0.015 | 3.14 | 0.318 |
| ROA | 32,880 | −0.015 | 0.033 | 0.214 | −1.113 | 0.332 | 23.013 | 0.000 | 1.23 | 0.815 |
| FIRM SZ | 32,880 | 9.226 | 9.26 | 0.886 | 7.009 | 11.357 | 8.482 | 0.000 | 1.66 | 0.603 |
| LEVERAGE | 32,880 | 0.255 | 0.229 | 0.216 | 0 | 0.955 | 17.811 | 0.000 | 1.08 | 0.927 |
| GDP | 32,880 | 54340.048 | 59907.754 | 15709.565 | 9661.228 | 84121.93 | 18.026 | 0.000 | ||
| logGDP | 32,880 | 10.841 | 11.001 | 0.399 | 9.176 | 11.34 | 21.851 | 0.000 | 1.22 | 0.819 |
| INF | 32,880 | 3.272 | 1.994 | 5.108 | −2.517 | 96.107 | 24.877 | 0.000 | 1.06 | 0.943 |
| BDSIZE | 32,880 | 9.015 | 8.98 | 0.894 | 1 | 41 | 23.050 | 0.000 | 1.02 | 0.981 |
| BDGEND | 32,880 | 22.90 | 24.09 | 4.803 | 0 | 60 | 21.062 | 0.000 | 1.02 | 0.982 |
| CSR COM | 32,880 | 0.431 | 0 | 0.495 | 0 | 1 | −4.495 | 1.000 | 1.39 | 0.719 |
4.2 Correlation matrix
Table 5 lists the correlation matrices of the variables. The table reveals that audit committee has a positive and significant association with carbon emission practices. This means that organizations with more effective audit committee are more likely to have significant efforts or practices related to managing carbon emissions. This could mean that audit committee plays a role in overseeing or influencing environmental initiatives within organizations, possibly through monitoring, reporting, or guiding strategic decisions related to carbon emission reduction. The reporting of carbon emissions in OECD nations is an essential part of tracking and observing efforts made to combat climate change and reduce greenhouse gas emissions (Hannasvik et al., 2023; Aldy and Stavins, 2012). Similarly, Table 5 shows that return on assets, CSR Committee, firm size, leverage, inflation and board gender diversity have a positive and significant association with carbon emission practices. On the other hand, gross domestic product, board size and dummy variables sector and county, have a negative and significant association with carbon emission practices.
Pairwise correlations pairwise correlations
| Variables | EMISSION | AUDIT_COM | ROA | CSR_COM | FIRM_SZ | Leverage |
|---|---|---|---|---|---|---|
| EMISSION | 1.000 | |||||
| AUDIT_COM | 0.079*** | 1.000 | ||||
| ROA | 0.280*** | 0.068*** | 1.000 | |||
| CSR_COM | 0.646*** | 0.206*** | 0.208*** | 1.000 | ||
| FIRM_SZ | 0.590*** | 0.198*** | 0.407*** | 0.485*** | 1.000 | |
| LEVERAGE | 0.100*** | 0.079*** | 0.003 | 0.094*** | 0.236*** | 1.000 |
| logGDP | −0.234*** | 0.206*** | −0.133*** | −0.131*** | −0.073*** | 0.028*** |
| INF | 0.034*** | 0.102*** | 0.015*** | 0.044*** | −0.046*** | 0.021*** |
| BDSIZE | −0.038*** | 0.003 | −0.002 | −0.023*** | −0.031*** | −0.008* |
| BDGEND | 0.001 | 0.023*** | 0.004 | 0.007 | −0.003 | −0.012** |
| SECTOR | −0.044*** | 0.004 | −0.068*** | −0.062*** | −0.023*** | 0.034*** |
| COUNTRY | −0.166*** | 0.111*** | −0.121*** | −0.161*** | −0.049*** | 0.031*** |
| Variables | logGDP | INF | BDSIZE | BDGEND | BDGEND | SECTOR |
|---|---|---|---|---|---|---|
| logGDP | 1.000 | |||||
| INF | −0.160*** | 1.000 | ||||
| BDSIZE | 0.013** | 0.011** | 1.000 | |||
| BDGEND | 0.023*** | 0.019*** | −0.125*** | 1.000 | ||
| SECTOR | 0.065*** | −0.007 | 0.015*** | 0.004 | 1.000 | |
| COUNTRY | 0.356*** | 0.059*** | 0.025*** | 0.012** | 0.134*** | 1.000 |
Note(s): ***p < 0.01, **p < 0.05, *p < 0.1
4.3 Results estimation
Table 6 presents the findings of two specification tests: the Breusch-Pagan LM test for random effects and the Hausman model specification test. The rejection of the null hypothesis by the Breusch-Pagan LM test indicates that a Random Effects (RE) model might be preferable to Pooled Ordinary Least Squares (OLS). However, upon conducting the Hausman specification test, the null hypothesis, suggesting the suitability of the random-effects model, was rejected. Instead, it was concluded that the Fixed Effects (FE) model was more appropriate for the analysis.
Specification test
| χ2 | p-values | Appropriate model | |
|---|---|---|---|
| Breusch and Pagan Lagrangian multiplier test for random effects | 38862.64 | 0.0000 | RANDOM |
| Hausman | 538.53 | 0.0000 | FIXED |
The results presented in Table 7 are separated into pooled (all samples) columns (1), without CSR committees (column 2), with CSR committees (column 3), and CSR_COM as moderators (column 4). The pooled results demonstrate that the regression model is statistically significant, as evidenced by the F-value of 271.5 at a significance level of 0.000. This indicates a robust relationship between the independent and dependent variables (carbon emission practices). Furthermore, the R-squared value of 0.074 indicates that approximately 0.074 of the variability in carbon emission practices can be explained by independent variables. Notably, the unaccounted variances in the model were ascribed to external factors, which were not included in this study.
Panel regression (fixed effect model)
| Variables | Pooled | Without CSR committee | With CSR committee | CSR committee as moderator |
|---|---|---|---|---|
| (1) | (2) | (3) | (4) | |
| EMISSION_SC | EMISSION_SC | EMISSION_SC | EMISSION_SC | |
| AUDIT_COM | 0.0279*** | 0.0168*** | 0.0228** | −0.0220*** |
| (0.00317) | (0.00299) | (0.0112) | (0.00300) | |
| CSR_COM | 0.157*** | |||
| (0.00787) | ||||
| AUDIT_COM#CSR_COM | 0.0414*** | |||
| (0.00808) | ||||
| ROA | −0.0605*** | −0.0299*** | −0.00288 | −0.0329*** |
| (0.00789) | (0.00693) | (0.0177) | (0.00716) | |
| FIRM_SZ | 0.148*** | 0.0657*** | 0.237*** | 0.105*** |
| (0.00482) | (0.00447) | (0.0116) | (0.00441) | |
| LEVERAGE | 0.0274*** | 0.0247*** | 0.00426 | 0.0188** |
| (0.00841) | (0.00783) | (0.0177) | (0.00762) | |
| logGDP | 0.120*** | 0.0380*** | 0.131*** | 0.0812*** |
| (0.0119) | (0.0144) | (0.0161) | (0.0108) | |
| INF | 0.00285*** | 0.00246*** | 0.00199*** | 0.00181*** |
| (0.000205) | (0.000389) | (0.000219) | (0.000186) | |
| BD_SIZE | −0.00840*** | −0.00828*** | −0.00599** | −0.00680*** |
| (0.00149) | (0.00152) | (0.00244) | (0.00135) | |
| BD_GEND | 0.00234*** | 0.00131*** | 0.000918* | 0.00150*** |
| (0.000378) | (0.000429) | (0.000530) | (0.000342) | |
| SECTOR | Included | Included | Included | Included |
| COUNTRY | Included | Included | Included | Included |
| Constant | −2.313*** | −0.804*** | −3.107*** | −1.536*** |
| (0.134) | (0.161) | (0.200) | (0.122) | |
| Observations | 32,880 | 18,709 | 14,171 | 32,880 |
| R-squared | 0.074 | 0.038 | 0.066 | 0.240 |
| Number of ID | 5,480 | 4,090 | 3,568 | 5,480 |
| F-value | 271.5*** | 71.12*** | 93.09*** | 862*** |
Note(s): Standard errors in parentheses
***p < 0.01, **p < 0.05, *p < 0.1
The first column demonstrates that audit committee has a positive and significant impact on carbon emission practices at a significance level of 0.01, with a coefficient of 0.0279. This suggests that the audit committee’s influence over governance, supervision, and resource allocation encourages businesses to adopt and use environmentally friendly techniques to lower carbon emissions. AUDIT_COM has a positive impact on EMISSION, as shown in Columns (2) and (3). In addition, the impact of AUDIT_COM on EMISSION is higher in the sample with CSR_COM (coefficient = 0.0228) than in the sample without CSR_COM (coefficient = 0.0168). This indicates that the presence of CSR_COM influences AUDIT_COM’s impact on EMISSION. Concerning the moderation variable, Table 7 Column (4) reveals that AUDIT_COM has a negative impact on EMISSION, whereas CSR_COM has a positive impact on EMISSION, with a 0.157 coefficient. This finding confirms that CSR committees have a greater influence on firms’ carbon emission practices. Regarding the moderating effect of CSR committees, CSR committee positively and significantly moderate the association between audit committee and carbon emissions practices at a significance level of 0.01, with a coefficient of 0.00787, indicating that companies are compelled by CSR committees to enhance their environmental effect, which promotes openness and responsibility. Regarding the controlling variables, Table 7 shows that return on assets and board size negatively and significantly impact carbon emission practices at a significance level of 0.01. In contrast, firm size, leverage, GDP, board gender diversity, and INF have a positive and significant impact on carbon emission practices.
4.4 Endogeneity test
To address endogeneity, which could skew the results for firms with high carbon emission practices, two-stage least squares (2SLS) and system Generalized Method of Moments (GMM) methodologies were used, as detailed in Table 8. The authors selected the audit committee, AUDIT _ EXPERT (AUDIT_EXPERT), and board skills (BD_SKILL) based on their roles in overseeing firm activities and mediating between management and stakeholders regarding carbon emission practices (Arif et al., 2021), as well as their impact on the integrity of carbon emission practices (Pozzoli et al., 2022). The validity of these instruments was confirmed by the Sargan test statistic of 2.495, with a p-value of (p = 0.112). In the initial phase, the audit committee was analyzed using these instrumental variables along with other control variables, with the residuals from this phase feeding into the second stage of both the 2SLS and GMM analyses (shown in columns 2 and 3, respectively). The findings show that AUDIT_COM has a positive impact on EMISSION, with coefficients of 0.0346 and 0.0413 in the 2SLS and GMM models, respectively. This shows that with the interference of additional variables, the residuals of such interference do not influence the model. Therefore, these advanced methods are consistent with the original analysis, as presented in Table 7, reinforcing the initial conclusions.
2SLS and GMM
| Variables | (First stage) | (2sls) | (GMM) |
|---|---|---|---|
| AUDIT_COM | EMISSION | EMISSION | |
| AUDIT_COM | 0.0346*** | 0.0345*** | |
| (0.00500) | (0.00472) | ||
| AUDIT_EXPERT | 0.398*** | ||
| (0.00361) | |||
| BD_SKILL | 0.532*** | ||
| (0.00437) | |||
| ROA | −0.0127** | −0.0410*** | −0.0413*** |
| (0.00619) | (0.00774) | (0.00648) | |
| FIRM_SZ | 0.0149*** | 0.219*** | 0.220*** |
| (0.00159) | (0.00201) | (0.00193) | |
| LEVERAGE | 0.00661 | −0.0362*** | −0.0364*** |
| (0.00544) | (0.00680) | (0.00678) | |
| logGDP | −0.0375** | 0.0778*** | 0.0775*** |
| (0.0165) | (0.0207) | (0.0206) | |
| INF | 0.000734** | 0.00272*** | 0.00272*** |
| (0.000298) | (0.000372) | (0.000412) | |
| BD_SIZE | −0.00177 | −0.00417*** | −0.00414*** |
| (0.00121) | (0.00152) | (0.00142) | |
| BD_GEND | 2.13e−05 | 0.000201 | 0.000194 |
| (0.000228) | (0.000285) | (0.000286) | |
| SECTOR | Included | Included | Included |
| COUNTRY | Included | Included | Included |
| Constant | 0.442** | −2.524*** | −2.523*** |
| (0.182) | (0.228) | (0.228) | |
| Observations | 32,880 | 32,880 | 32,880 |
| R-squared | 0.070 | 0.070 | |
| F-test | 616.41*** | 905.7*** | |
| Sargan statistic (Chi-sq) | 2.495 (p = 0.112) | ||
| Hansen J statistic | 2.343 (p = 0.126) |
Note(s): Standard errors in parentheses
***p < 0.01, **p < 0.05, *p < 0.1
4.5 Additional test
Table 9 shows the results of splitting the sample into developed and developing countries. The results for developing countries show that AUDIT_COM has a positive impact on EMISSION with a coefficient of 0.0500. In developed countries (Column 3), AUDIT_COM has a positive impact on EMISSION with a coefficient of 0.0273. This finding indicates that AUDIT_COM has better monitoring mechanisms for companies’ carbon practices in developing countries than in developed ones. Regarding the moderating role of CSR_COM, CSR_COM strengthens the relationship between AUDIT_COM and EMISSION in both developing and developed countries (Columns 2 and 4). The moderating role of CSR_COM is more prominent in developing countries, with a coefficient of 0.0626 compared with 0.0392 in developed countries.
Developed vs developing countries
| Variables | Developing | Developed | ||
|---|---|---|---|---|
| (1) | (2) | (3) | (4) | |
| EMISSION_SC | EMISSION_SC | EMISSION_SC | EMISSION_SC | |
| AUDIT_COM | 0.0500** | −0.0192 | 0.0273*** | −0.0219*** |
| (0.0226) | (0.0247) | (0.00320) | (0.00303) | |
| CSR_COM | 0.0947*** | 0.160*** | ||
| (0.0336) | (0.00811) | |||
| AUDIT_COM#CSR_COM | 0.0626 | 0.0392*** | ||
| (0.0382) | (0.00831) | |||
| ROA | −0.0278 | −0.0493 | −0.0605*** | −0.0328*** |
| (0.157) | (0.148) | (0.00791) | (0.00717) | |
| FIRM_SZ | 0.115*** | 0.0842** | 0.148*** | 0.105*** |
| (0.0401) | (0.0381) | (0.00486) | (0.00445) | |
| LEVERAGE | 0.192* | 0.115 | 0.0260*** | 0.0181** |
| (0.100) | (0.0954) | (0.00845) | (0.00765) | |
| logGDP | 0.243** | 0.148 | 0.123*** | 0.0833*** |
| (0.101) | (0.0958) | (0.0120) | (0.0109) | |
| INF | 0.0114*** | 0.00808*** | 0.00275*** | 0.00174*** |
| (0.00196) | (0.00191) | (0.000207) | (0.000188) | |
| BD_SIZE | −0.00513 | −0.00463 | −0.00833*** | −0.00681*** |
| (0.0135) | (0.0129) | (0.00150) | (0.00136) | |
| BD_GEND | 0.00497 | 0.00224 | 0.00233*** | 0.00150*** |
| (0.00563) | (0.00533) | (0.000379) | (0.000343) | |
| Constant | −3.045*** | −1.782* | −2.355*** | −1.563*** |
| (1.031) | (0.987) | (0.136) | (0.124) | |
| Observations | 990 | 990 | 31,890 | 31,890 |
| R-squared | 0.136 | 0.234 | 0.074 | 0.241 |
| F-test | 9.285*** | 14.36*** | 265.8*** | 849.4*** |
Table 10 shows the results of splitting the sample according to loss and profit firms. The results for loss-making firms show that AUDIT_COM has a positive impact on EMISSION with a coefficient of 0.0188. For profitable firms (Column 3), AUDIT_COM also has a positive impact on EMISSION, with a coefficient of 0.0291. This finding indicates that, in performing firms, AUDIT_COM has a better monitoring mechanism for companies’ carbon practices than loss-making firms. Regarding the moderating role of CSR_COM, CSR_COM strengthened the relationship between AUDIT_COM and EMISSION in profitable firms (Column 4), with a coefficient of 0.0464.
Profit vs loss companies
| Variables | Loss | Profit | ||
|---|---|---|---|---|
| (1) | (2) | (3) | (4) | |
| EMISSION_SC | EMISSION_SC | EMISSION_SC | EMISSION_SC | |
| AUDIT_COM | 0.0188*** | −0.00573 | 0.0291*** | −0.0303*** |
| (0.00451) | (0.00420) | (0.00426) | (0.00413) | |
| CSR_COM | 0.139*** | 0.147*** | ||
| (0.0214) | (0.00895) | |||
| AUDIT_COM#CSR_COM | 0.0369* | 0.0464*** | ||
| (0.0215) | (0.00927) | |||
| ROA | −0.0196** | −0.00752 | 0.0436* | 0.0209 |
| (0.00878) | (0.00801) | (0.0253) | (0.0232) | |
| FIRM_SZ | 0.0485*** | 0.0350*** | 0.322*** | 0.224*** |
| (0.00575) | (0.00526) | (0.00817) | (0.00766) | |
| LEVERAGE | 0.0686*** | 0.0489*** | −0.0424*** | −0.0245* |
| (0.00926) | (0.00845) | (0.0145) | (0.0133) | |
| logGDP | 0.0610*** | 0.0426** | 0.0997*** | 0.0692*** |
| (0.0218) | (0.0199) | (0.0148) | (0.0136) | |
| INF | 0.00315*** | 0.00210*** | 0.00247*** | 0.00166*** |
| (0.000581) | (0.000530) | (0.000227) | (0.000208) | |
| BD_SIZE | −0.00608*** | −0.00563*** | −0.00930*** | −0.00726*** |
| (0.00226) | (0.00206) | (0.00203) | (0.00186) | |
| BD_GEND | 0.00167*** | 0.00162*** | 0.00190*** | 0.00103** |
| (0.000642) | (0.000585) | (0.000468) | (0.000428) | |
| SECTOR | Included | Included | Included | Included |
| COUNTRY | Included | Included | Included | Included |
| Constant | −0.909*** | −0.611*** | −3.679*** | −2.462*** |
| (0.243) | (0.222) | (0.171) | (0.158) | |
| Observations | 9,499 | 9,499 | 23,381 | 23,381 |
| R-squared | 0.049 | 0.212 | 0.117 | 0.261 |
| F-test | 41.31*** | 172.2*** | 306.6*** | 652.3*** |
5. Findings and discussion
Table 7 reveals that audit committee has a positive and significant impact on carbon emission practices at a significance level of 0.01. This result contradicts that of Alkurdi et al. (2023), who argue that the independence of audit committee has a negative relationship with carbon emission performance. However, the results of this study align with those of Chariri et al. (2018) and Patelli and Prencipe (2007), who believe that audit committee positively affect carbon emission practices. Widagdo et al. (2022) argue that the frequency of audit committee meetings has a beneficial impact on the report on greenhouse gas emissions. These results can be explained by the fact that audit committee often advise management on strategic matters including sustainability. Their influence can steer a company towards integrating carbon reduction into its strategic planning, ensuring that environmental sustainability becomes a core component of the business model rather than an afterthought. The audit committee can encourage responsibility, openness, and sustainability of carbon emissions by actively tracking and evaluating the company’s environmental performance (Salleh et al., 2022; Annisa Anafiah et al., 2017). These results could persuade policymakers to create legislation that enhances the function of audit committee in environmental supervision through their beneficial effects on carbon emission practices, which might force audit committee to explicitly address sustainability concerns or mandate that members have environmental competence.
Concerning the moderating effect, it was found that the CSR committee positively and significantly moderated the association between audit committee and carbon emissions practices at a significance level of 0.01. This result suggests that having an active CSR committee strengthens the relationship between audit committee and carbon emissions practices within firms. This could be explained by the fact that having a CSR committee may improve an organization’s environmental accountability. This accountability may encourage the audit committee to examine carbon emissions processes more thoroughly, which might result in improved management and monitoring of these operations. Moreover, the CSR committee may bring specialized expertise and additional resources to the Table, such as environmental consultants or sustainability professionals. This expertise and resources can help the audit committee better understand and address issues related to carbon emissions, leading to more effective governance in this area. Furthermore, a corporation has a competitive advantage if it has a CSR committee. Second, agency issues and systemic risks inside the organization decrease if the firm has an efficient CSR committee (Sadiq et al., 2020). The audit committee may find itself obliged to work with the CSR department to enhance carbon emission procedures as stakeholders place increasing demands on environmental responsibility. Furthermore, stakeholders are less knowledgeable when there is no CSR committee in place (Martínez-Ferrero et al., 2021).
6. Conclusion
This study examines the effect of the audit committee on carbon emission practices. Furthermore, it sought to investigate the moderating effect of CSR committees on the association between audit committees and carbon emissions practices. Fixed-effects (FE) and random-effects (RE) estimation methods were used to achieve the study objectives. Additionally, the Huber-White standard error, Driscoll-Kraay standard error, two-stage least squares (2SLS) estimation, and generalized method of moments (GMM) were employed to guarantee the robustness of our findings. The results reveal that audit committees have a positive and significant impact on carbon emission practices as they advise management on strategic matters, including sustainability issues. Moreover, CSR committee positively and significantly moderates the association between audit committee and carbon emission practices. Given these findings, this study offers significant perspectives on many carbon emission-related matters to regulators, policymakers, auditors, board members, analysts, and scholars. Enhancements should be made to stakeholders’ engagement processes to promote ethical and responsible practices. This mechanism comprises responsible parties, such as regulatory agencies, shareholders, external and internal auditors, and board monitoring.
This study contributes to the literature in various ways. First, unlike previous studies that have examined the direct relationship between audit committee and carbon emissions practices, this study examined the moderating effect of CSR committees on the association between audit committee and carbon emissions. Second, this study contributes to the evolving literature on corporate governance and sustainability by elucidating the role of audit committee in influencing carbon emissions practices in OECD Countries. Third, the findings offer evidence based on present and large datasets of 6,063 enterprises from 34 OECD nations, which may help regulators establish standards for companies to improve their carbon emission practices. Fourth, this study offers several implications for managers, investors, and policymakers regarding the various aspects of carbon emissions. The first implication is for corporate managers who are in strategic positions to check the board of directors’ rules and make the necessary updates that align with the business environment. One of these updates would be to assign audit responsibilities to manage and oversee carbon emission reduction targets. Furthermore, as they are at the strategic management level, they should set carbon emissions targets as part of their corporate strategy. Second, stockholders in the OECD region should consider the existence of audit and corporate social responsibility committees in a business as one of the vital criteria for investment decisions. In addition, stakeholders in the OECD region should review the carbon emission history and ESG performance of businesses before making investment decisions. In other words, investors should prioritize investment options based on these two main factors. The third implication is for regulators and policymakers in the OECD region, who should set mandatory rules and policies regarding their audit and corporate social responsibility practices’ disclosure, their role in sustainability planning, and oversight. At the national level, policymakers should launch programs that incentivize businesses to have an effective corporate sustainability structure. One of these programs could offer subsidies, tax exemptions, or breaks, which would encourage businesses to be more environmentally friendly. Finally, regulators and policymakers should create a long-term plan for mandating audits and assuring ESG reporting.
This research, like many others, has limitations that pave the way for further research. The first limitation is the period from 2017 to 2022. Future research could expand this timeframe to 2024. Second, the independent variables in this study explain only 0.12% of the total variability in carbon emission practices. Hence, researchers should investigate other factors that may affect carbon emission practices. Third, this study relies on secondary data. Therefore, researchers should collect primary data to investigate these results. Fourth, corporate governance variables were not used in the analysis due to data limitations. Hence, future research could use corporate governance variables, such as BIG4, board size, and independent commissioners in the analyzed models.
Consent to publish declaration
The results/data/figures in this submission have not been previously published or under consideration for publication elsewhere.
