Do audit committees impact corporate social responsibility? A review of empirical quantitative research and research opportunities

Abstract

Purpose

The purpose of this study was to analyze whether audit committees (ACs) influence corporate social responsibility (CSR) outputs.

Design/methodology/approach

A structured literature review of 57 archival studies on the influence of ACs on CSR outputs was conducted. According to a stakeholder–agency theoretical framework, the AC variables were structured as follows: presence, composition and resources, incentives and diligence. CSR is mainly divided into CSR performance, CSR reporting and CSR assurance.

Findings

Previous studies have mainly focused on AC composition and CSR reporting. There are indications that AC composition and CSR performance and assurance are positively linked. Moreover, AC resources, incentives and diligence increase CSR reporting.

Research limitations/implications

This study stresses the need for linking AC composition with sustainability, the inclusion of moderator and especially mediator variables and addressing endogeneity concerns via advanced regression models.

Originality/value

This paper reports the first literature review on the interaction between AC and CSR. It presents the main variables that have been included in previous studies, the limitations of these studies and useful recommendations for future research, business practice and regulators.

Citation

Velte, P. (2025), "Do audit committees impact corporate social responsibility? A review of empirical quantitative research and research opportunities", Corporate Governance, Vol. 25 No. 8, pp. 1-20. https://doi.org/10.1108/CG-04-2024-0227

1. Introduction

Many regulators have questioned the quality of the corporate social responsibility (CSR) activities of public interest entities (PIEs) since the 2007–2008 financial crisis. Controversial discussions have related to the influence of boards of directors on CSR outputs (Ali et al., 2023; Al-Shaer et al., 2017). An audit committee (AC) is a subcommittee of a board of directors and a significant corporate governance tool. Supervising financial reports, internal audits and external auditors are the traditional goals of ACs. However, given the increasing number of corporate sustainability regulations, ACs have also become responsible for supervising CSR reports, including performance measures, and the engagements of CSR assurors (Suttipun, 2021; Said et al., 2009). Researchers have assumed that ACs positively influence CSR reporting, performance and assurance (Arif et al., 2021; Uyar et al., 2023). Management systems (e.g. internal control and risk management systems) and CSR reports must be supervised by ACs. While the duties of ACs differ between one- and two-tier systems, the assumption is that they increase corporate governance quality in both systems (Raimo et al., 2021). Moreover, some regimes still do not mandate the formation of an AC for PIEs. However, ACs are recommended by national corporate governance codes and are thus considered a “best practice” (Qaderi et al., 2024).

Many studies have investigated the influence of ACs on CSR (e.g. Pucheta-Martinez et al., 2021; Perez-Cornejo et al., 2019). In comparison with studies that merely examined the impacts of ACs, other studies have addressed their composition, resources, incentives and diligence (e.g. Khemakhem et al., 2023). The major importance of ACs for corporate sustainable transformation has been stressed (e.g. Hasan et al., 2022). Previous research on the link between ACs and CSR has yielded complex and heterogeneous results. Some meta-analyses and literature reviews have been conducted on ACs and their impacts on overall firm outputs without focusing on CSR (e.g. DeZoort et al., 2002; Ghafran and O’Sullivan, 2013; Malik, 2014; Turley and Zaman, 2004; Velte, 2023). In contrast to previous meta-analyses (e.g. Bilal et al., 2018), the aim of this study was not to test the statistical correlations between specific AC proxies. Instead, we emphasized the major tendencies of previous empirical studies, as well as the complexity of the included proxies, and formulated useful research recommendations. To our knowledge, no systematic analysis has focused on AC research or CSR.

We analyzed 57 empirical quantitative (archival) studies on the influence of ACs on CSR. Based on stakeholder–agency theory, the AC variables (independent variables) were as follows: 1) presence, 2) composition and 3) resources, incentives and diligence. The CSR outputs (dependent variables) were as follows: 1) CSR performance, 2) CSR reporting and 3) CSR assurance. We focused on archival studies as the dominant research method to achieve an adequate level of comparability and on the econometric relationships between AC and CSR. This led us to formulate our main research question:

RQ1.

Which AC attributes (presence, composition, resources, incentives and diligence) are correlated with increased CSR performance, reporting and assurance?

In this literature review, we highlight the major gaps in recent AC and CSR research and the need for detailed analysis. Most studies have focused on AC composition and CSR reporting. AC composition, CSR performance and CSR assurance tend to be positively correlated. Moreover, AC resources, incentives and diligence increase CSR reporting. Thus, we emphasize the need for AC composition with explicit environmental and social skills, interrelations between AC and sustainability board committees and the inclusion of endogeneity concerns by advanced regression models.

Our analysis was structured as follows: Section 2 establishes the study’s theoretical foundation, which was based on stakeholder–agency. Section 3 presents the research framework and methodology. Section 4 consists of a content analysis of the included studies and explains our main findings. In Section 5, we divide ACs into three categories (presence, composition and resources, incentives and diligence) and CSR into three proxies (CSR performance, reporting and assurance). In Section 6, we identify the study’s limitations and provide research recommendations, distinguishing between methodological and content-related issues. Section 7 summarizes our analysis.

2. Theoretical foundation

The impact of ACs on CSR can be explained through multiple theories, such as stakeholder theory, legitimacy theory and resource dependency theory. As indicated in Table 2 (see Panel G), most studies on the link between ACs and CSR refer to principal agent theory (Ross, 1973; Jensen and Meckling, 1976). However, management theories are also included to a greater extent. This relates to stakeholder theory (17 studies) and legitimacy theory (14 studies). Many studies also use a mixed-theory approach, combining agency and management theories. In this regard, they mainly stress the conflicting roles of ACs in satisfying the demands of either exclusive shareholders or a broad range of stakeholders. In this study, we used stakeholder–agency theory (Hill and Jones, 1992) as a combination of the principal agent and stakeholder theories (Freeman, 1984) for the following reasons: In line with principal agent theory (Jensen and Meckling, 1976), ACs should decrease agency conflicts (information asymmetries between management and shareholders, and conflicts of interest). The traditional focus of ACs was financial reporting and related management systems, which entailed supervising internal and external auditors and checking the effectiveness of internal control and risk management in accordance with shareholders’ preferences. In addition to these traditional tasks, ACs are responsible for monitoring CSR activities as well as supervising CSR reporting, related CSR performance measures and the engagement of CSR assurors. CSR activities lead to major information asymmetries, as executive directors are better informed about a firm’s real CSR situation than shareholders and other stakeholders. CSR reporting and the included performance measures can decrease as a result of opportunistic management behavior. ACs must fulfill a major monitoring function to guarantee decision-useful CSR reporting and related management processes (Bilal et al., 2018). The effectiveness of ACs should lead to increased CSR reporting, performance and assurance (Al-Shaer and Zaman, 2018).

As CSR reports represent a broad stakeholder tool, they lead to the extension of principal agent theory to aspects related to other stakeholders. While many studies have stressed the conflicts between principal agent theory and stakeholder theory (e.g. Umar et al., 2023b), the aim of this study was to combine them into stakeholder–agency theory (Velte, 2018; Zaman et al., 2021). We assumed that both shareholders and other stakeholders demand that ACs have adequate resources, incentives and diligence as well as a useful composition. An adequate size (resources), meeting frequency (diligence) and compensation (incentives) should lead to increased AC effectiveness in line with stakeholders’ preferences. ACs should include independent members with financial, industry and sustainability expertise as well as high levels of education and experience. Moreover, ACs should recognize diversity factors (e.g. age, internationality and gender) to increase awareness of CSR activities in the firm. In this regard, we do not see a major conflict among stakeholders, as ACs are responsible for financial and CSR communication following the business case argument for CSR.

3. Research framework and methods

We developed an AC research framework to summarize the status quo of the current AC research for this literature review (see Figure 1). This study was mainly structured based on the previous frameworks developed by Malik (2014), Ghafran and O’Sullivan (2013), DeZoort et al. (2002) and Turley and Zaman (2004). Malik (2014) distinguished between AC composition, responsibilities and compensation as input factors and auditors and earnings management/internal control deficiencies as output factors. Ghafran and O’Sullivan (2013) distinguished between AC composition, resources and diligence as AC characteristics and external audit quality, financial reporting quality and internal audit quality as the financial reporting process. Finally, DeZoort et al. (2002) categorized composition, authority, resources and diligence as AC input factors. Based on these distinctions, we defined the major AC categories as follows: 1) AC presence, 2) AC composition and 3) AC resources, incentives and diligence.

AC presence was used as a dummy variable indicating whether firms established an AC in a voluntary regime. Although many countries have requested that PIEs implement ACs, others have included recommendations in national corporate governance codes. According to stakeholder–agency theory, ACs are more effective than boards at monitoring CSR reporting, performance and assurance (Hill and Jones, 1992). Thus, the presence of ACs should positively impact CSR outputs.

AC composition is the second category in the research framework and is rather attractive in archival AC research. This leads to an increased complexity of variables (see Table 1). It guarantees effectiveness and thus should be positively related to CSR outputs (Tumwebaze et al., 2022). Many researchers have relied on effectiveness, independence, expertise, (gender) diversity and experience/education as the most prominent proxies in archival AC research (Uyar et al., 2023). In line with board independence, AC members would be independent of executive directors to conduct proper monitoring of CSR reports, CSR assurors and related processes. AC members’ specific sustainability expertise is important for realizing appropriate monitoring quality (Bilal et al., 2018). In line with financial expertise, industry and sustainability expertise can be classified as a “golden triangle” (Velte, 2018). As CSR reporting is industry specific (e.g. based on environmentally sensitive industries), ACs should include a certain degree of industry expertise. Diversity is another prominent AC dimension. Female AC directors are more independent and consider a broader range of stakeholder concerns in their decision-making (e.g. Bose et al., 2022a). According to stakeholder–agency theory, gender diversity in ACs is associated with critical reflections on CSR reports, performance and assurance procedures, which emphasize increased AC effectiveness (Bose et al., 2022a). This also applies to other types of diversity, such as age and ethnicity. AC members with more experience and higher levels of education should also positively impact CSR outputs.

AC resources (size), incentives (compensation) and diligence (meeting frequency) (Ghafran and O’Sullivan, 2013; Velte, 2023) constitute the third category in this review. They are the foundation of AC effectiveness. We assumed that these elements positively influence CSR outputs (Ghafran and O’Sullivan, 2013). An increased AC size is linked with higher expertise and stakeholder dialog and increased discussions and awareness of CSR (Buallay and Al-Ajmi, 2020). In accordance with stakeholder–agency theory, higher meeting attendance rates should also lead to increased AC monitoring of CSR issues (Dwekat et al., 2020). To align ACs and stakeholders, ACs should have an adequate compensation system (Li et al., 2012). AC members should be rewarded for properly monitoring activities. Therefore, AC members should receive more compensation than other nonexecutive directors (Li et al., 2012).

CSR outputs were the dependent variables in this literature review. CSR reporting refers to the assumption that executives can negatively influence CSR quality as a result of opportunistic management behavior (Ross, 1973; Jensen and Meckling, 1976; Hill and Jones, 1992), including green washing and information overload to attract stakeholder groups. While many regulators have finalized regulations on mandatory CSR reporting, the comparability of CSR figures is currently rather low (Musallam, 2018). Since ACs must supervise CSR reports, specific AC attributes associated with AC effectiveness should lead to higher CSR reporting quality (Pitenoei et al., 2022). We also included similar reporting concepts, such as integrated and intellectual capital reporting.

Whereas CSR performance, the second category of dependent variables, mainly refers to external databases, individual environmental and social scores are still less important (e.g. Yorke et al., 2023). Higher AC effectiveness should promote firm reputation and improve CSR performance scores when stakeholders trust the numbers (Pozzoli et al., 2022).

We included CSR assurance as the third CSR category. It is still voluntary in most regimes, and it should increase stakeholder reputation and, in turn, CSR outputs (Zaman et al., 2021). While external auditors are responsible for CSR assurance services, other external parties (e.g. environmental assurors outside the audit profession) may also be selected. Thus, the choice of assuror is heterogeneous in business practice and crucial for researchers (Zaman et al., 2021). ACs should be motivated to engage external auditors as CSR assurors because of the synergies between financial and CSR reporting. Moreover, CSR assurance quality is measured by analyzing the contents of assurance statements (Zaman et al., 2021). Firms can choose between limited and reasonable assurance levels. Effective ACs should demand reasonable assurance levels for CSR reports, as these represent a higher quality of CSR assurance and better stakeholder reputation.

Figure 1 summarizes the research framework, and Table 1 includes a list of the AC variables and CSR reporting, performance and assurance proxies.

4. Content analysis

Empirical research on the influence of ACs on CSR uses various kinds of data, study designs, theoretical foundations and analytical models (e.g. Bilal et al., 2018), which indicates the need for a literature review. Literature reviews are an important research method for scholars, practitioners and regulators seeking to guide researchers within this field of research (Torraco, 2005; Webster and Watson, 2002). Literature reviews aim to produce new knowledge about specific research topics by summarizing previous studies, which can lead to theory building and recommendations for future research designs. Literature reviews can also improve business practices by stressing organizational developments for future firm strategies and guidance for regulators and implementation (Webster and Watson, 2002).

Accordingly, we stressed the different goals of meta-analyses and structured literature reviews, and our decision to prepare a structured literature review. In corporate governance research, a meta-analysis is an important research method for analyzing the overall statistical significance of specific economic relationships based on the results of several studies (Bilal et al., 2018). Meta-analyses test possible moderator and mediator variables from a statistical perspective. However, our goal in this study was to perform a narrative analysis of the influence of ACs on CSR outputs by including heterogeneous variables. In contrast to meta-analyses on ACs (e.g. Bilal et al., 2018), we addressed the limitations of previous archival studies from content and methodological perspectives. This paper guides future researchers, as it includes explicit recommendations for future research. The variables included in our study were too heterogeneous for a meta-analysis, which is only useful for testing a restricted relationship, such as the impact of AC effectiveness on CSR performance.

In this study, we relied on established processes (Denyer and Tranfield, 2009). We began by developing our research objective. In contrast to previous literature reviews on AC, this review focused on the influence of ACs on CSR performance, reporting and assurance, identifying major research gaps and inconsistencies in previous AC studies. We analyzed the main theories and generated specific terms for the database search. We then searched international databases (EBSCO Business Source Complete and Web of Science). We used asterisks to capture related terms and included relevant keywords (“audit committees” in connection with “ESG performance,” “CSR performance,” “CSR reporting,” “ESG reporting,” “CSR assurance,” “ESG assurance” and related terms) in our search string. Next, we established exclusion and inclusion criteria. We collected an initial sample of 98 studies, focusing on archival research as the most important research method. This strategy led to an increased level of comparability between the included studies. We excluded 24 analytical, experimental, qualitative and conceptual papers. We excluded 15 studies unrelated to CSR performance, reporting or assurance. We did not set a temporal limit. For quality assurance reasons, we referred only to studies published in double-blind reviewed international journals. This resulted in the exclusion of 12 studies and a final sample of 57 studies. Figure 2 illustrates the selection process for the included studies.

We performed a precursory analysis. The titles of the articles were scanned to determine which abstracts to read. We did not further consider articles that matched our exclusion criteria. The theory and method sections of the remaining articles were then scanned.

The included studies were then coded according to the selected AC (sub-)categories of CSR (reporting, performance and assurance) and matched to our research framework. We coded significant findings and their indicators in accordance with the vote-counting technique (Light and Smith, 1971) by recording positive coefficients (+), significant negative coefficients (−) and insignificant results (±). Table 2 provides an overview of the included studies, showing the year of publication (Panel A), region (Panel B), journal (Panel C), dependent variables (Panel D), independent variables (Panel E), regression models (Panel F) and included theories (Panel G). The link between AC and CSR constitutes a rather nascent field of empirical research, as it has only existed since 2009, but it grew significantly in 2022 (2023), with 12 (13) studies (Panel A). Both developed and developing countries have been included in line with cross-country settings (Panel B). Panel C illustrates the diverse use of journal publication outlets in accounting, auditing, corporate governance, management and sustainability journals. The most prominent journal included was Business Strategy and the Environment (5 studies). CSR reporting was the most prominent dependent variable (40 studies; Panel D). Moreover, as illustrated in Panel D, most research studies included AC composition proxies. Panel F presents the regression models used in the included studies. The relationship between ACs and CSR outputs relates to massive endogeneity concerns based on sample selection bias, omitted variable bias and reverse causality. Without properly checking for endogeneity concerns, studies only measure correlations instead of causality. The included studies referred to simple ordinary least squares (OLS) regressions (23 studies), panel regressions (18 studies) and probit/logit models (9 studies). Advanced regression methods were included to a lesser extent: two-stage least squares in connection with instrumental variables (2SLS/IV; 7 studies), the generalized method of moments (GMM; 11 studies), propensity score matching (PSM; 4 studies), the Heckman two-stage approach (4 studies) and structural equation modeling (SEM) or three-stage least squares (3SLS; 4 studies). Thirty-six studies did not include proper endogeneity checks, which restricted the validity of the relationship between AC and CSR. As for theoretical frameworks (Panel G), principal agent theory was the most frequently used (40 studies), followed by selective management theories, such as stakeholder theory (17 studies), legitimacy theory (14 studies) and resource dependency theory (8 studies).

5. Results

5.1 Audit committee presence

Few studies in our literature review analyzed the impacts of AC presence on CSR outputs, and those that did mainly found a positive relationship. In a cross-country sample, Pucheta-Martinez et al. (2021) included CSR reporting and found a positive relationship. This relationship is also related to the Gulf Cooperation Council (Garas and ElMassah, 2018) and Bangladesh (Khan et al., 2013). Jarboui and Moalla (2022) focused on French environmental ACs and found a positive influence on environmental reporting. By contrast, Cancela et al. (2020) included environmental/social expenses in Spanish and Portuguese samples, but their results were inconclusive. Suttipun (2021) documented the negative impacts of ACs on social and governance reporting in Thailand.

5.2 Audit committee composition

5.2.1 Impact on corporate social responsibility reporting.

Although an increasing number of studies have analyzed the influence of AC composition on CSR reporting, the results have been inconclusive. In this study, we distinguished between AC effectiveness, independence, financial expertise, gender diversity and other AC composition variables.

5.2.1.1 Effectiveness.

Based on AC effectiveness scores, positive impacts on CSR reporting were found in Uganda (Tumwebaze et al., 2022), environmental reporting in the UK (Al-Shaer et al., 2017) and integrated reporting in Malaysia (Qaderi et al., 2024) and South Africa (Wang et al., 2020). Habbash (2016) found no significant link between AC effectiveness and CSR reporting in Saudi Arabia.

5.2.1.2 Independence.

A positive impact of AC independence on CSR reporting was found in the Gulf Cooperation Council (Buallay and Al-Ajmi, 2020), Australia (Appuhami and Tashakor, 2017 except for independent chairs; Arif et al., 2021), Malaysia (Said et al., 2009), Iran (Fallah and Mojarrad, 2019) and cross-country settings (Pozzoli et al., 2022; Uyar et al., 2023). This is also related to intellectual capital reporting in Malaysia (Haji, 2015) and integrated reporting in cross-country settings (Raimo et al., 2021). An insignificant impact of AC independence on CSR reporting was found in Jordan (Bataineh et al., 2023), Palestine (Musallam, 2018) and Indonesia (Ridwan and Mayapada, 2022) as well as on integrated reporting in South Africa (Haji and Anifowose, 2016) and intellectual capital reporting in the UK (Li et al., 2012). A negative impact on CSR reporting was found in Pakistan (Hasan et al., 2022) and in cross-country settings (Pucheta-Martinez et al., 2021, 2023) as well as on environmental reporting in China (Wang and Sun, 2022).

5.2.1.3 Financial expertise.

Inconclusive results have also been obtained for financial expertise in ACs. Financial experts were associated with increased CSR reporting in cross-country studies (Pucheta-Martinez et al., 2021; Pozzoli et al., 2022; Uyar et al., 2023) and increased intellectual capital reporting in Malaysia (Haji, 2015). Financial sustainability and combined AC expertise were found to lead to better integrated reporting in Velte’s (2018b) cross-country study. An insignificant impact of financial expertise on CSR reporting was found in Australia (Appuhami and Tashakor, 2017) and Jordan (Bataineh et al., 2023) as well as on environmental reporting in China (Wang and Sun, 2022), integrated reporting in South Africa (Haji and Anifowose, 2016) and in a cross-country setting (Raimo et al., 2021), and intellectual capital reporting in the UK (Li et al., 2012). A negative relationship was also found between financial expertise and CSR reporting in the Gulf Cooperation Council (Buallay and Al-Ajmi, 2020) and Palestine (Musallam, 2018).

5.2.1.4 Gender diversity.

Female AC members tended to have a positive impact on CSR reporting in Spain (Bravo and Reguera-Alvarado, 2019), Australia (Appuhami and Tashakor, 2017), Canada (Khemakhem et al., 2023) and Iran (Pitenoei et al., 2022) as well as on environmental reporting in China (Wang and Sun, 2022).

5.2.1.5 Other audit committee composition variables.

Education (age) was found to increase (decrease) corporate political spending reporting in the USA (Ali et al., 2023). Wang and Sun (2022) found no significant relationship between education, multiple and local directorships and environmental reporting in China. This is also related to chair tenure and CSR reporting (Alodat et al., 2023). Combined audit and risk committees were found to have a negative impact on carbon reporting in Australia (Krishnamurti and Velayutham, 2018).

5.2.2 Impact on corporate social responsibility performance.

The results on the impact of AC composition on CSR performance were more homogeneous than those on CSR reporting and mainly showed positive relationships, which is consistent with stakeholder–agency theory. We distinguished between AC effectiveness, independence, financial expertise, diversity and other variables.

5.2.2.1 Effectiveness.

In a cross-country study, Dwekat et al. (2020) found that AC effectiveness had a positive impact on CSR performance, while in their US-based study, Mohy-ud-Din and Raza (2023) found an insignificant relationship.

5.2.2.2 Independence.

Researchers have found a positive link between AC independence and CSR performance in cross-country settings (Alkurdi et al., 2023; Dwekat et al., 2020; Pozzoli et al., 2022), Spain (Perez-Cornejo et al., 2019), Iran (Mohammadi et al., 2021) and Nigeria (Umar et al., 2023a). However, Alwadani et al. (2024) did not find any significant influence of AC independence on environmental performance in Saudi Arabia.

5.2.2.3 Financial expertise.

Some studies have also found that financial expertise increases CSR performance in cross-country settings (Alkurdi et al., 2023; Dwekat et al., 2020; Pozzoli et al., 2022) and Iran (Mohammadi et al., 2021). This effect was also related to female financial experts in US firms (Yorke et al., 2023).

5.2.2.4 Diversity.

There is evidence that CSR performance is positively influenced by AC members’ gender diversity in the USA (Bose et al., 2022a) and experience diversity in Hong Kong (Cheung and Lai, 2023). However, insignificant results were found for education (diversity) in Spain (Perez-Cornejo et al., 2019) and foreign directors in Nigeria (Umar et al., 2023a).

5.2.2.5 Other audit committee composition variables.

AC tenure was found to either increase environmental performance (Paolone et al., 2022) or decrease CSR performance (Pozzoli et al., 2022) in cross-country settings. Moreover, interconnections between chief executive officers (CEOs) and other firms’ ACs were found to positively impact CSR performance in the USA (Bose et al., 2022b).

5.2.3 Impact on corporate social responsibility assurance.

Few studies have compared the impacts of ACs on CSR assurance with their impacts on CSR reporting and performance. AC dimensions were positively related to CSR assurance, which aligns with stakeholder–agency theory. AC effectiveness was found to positively impact integrated reporting assurance in South Africa (Wang et al., 2020). A positive impact on CSR assurance was found for financial expertise in the UK (Al-Shaer and Zaman, 2018) and cross-country settings (Dwekat et al., 2022; Uyar et al., 2023), industry expertise in New Zealand (Zaman et al., 2021) and AC independence in the UK (Al-Shaer and Zaman, 2018), New Zealand (Zaman et al., 2021) and cross-country settings (Dwekat et al., 2022; Uyar et al., 2023).

5.3 Audit committee resources, incentives and diligence

5.3.1 Impact on corporate social responsibility reporting.

Studies on the impact of AC resources, incentives and diligence on CSR reporting are popular. Positive connections between the variables can often be found, which is in line with stakeholder–agency theory.

5.3.1.1 Size.

Positive impacts of AC size on CSR reporting were found in Saudi Arabia (Alotaibi and Hussainey, 2016; Umar et al., 2023b), Australia (Appuhami and Tashakor, 2017), both Australia and the UK (Kend, 2015), Palestine (Musallam, 2018) and Pakistan (Hasan et al., 2022), as well as on corporate political spending reporting in the USA (Ali et al., 2023) and integrated reporting in a cross-country setting (Raimo et al., 2021), and on intellectual capital reporting in Malaysia (Haji, 2015) and the UK (Li et al., 2012). Some studies found insignificant results in the UK (Albitar et al., 2023), Jordan (Bataineh et al., 2023), the Gulf Cooperation Council (Buallay and Al-Ajmi, 2020), India (Fahad and Rahman, 2020) and Indonesia (Ridwan and Mayapada, 2022; Suyono and Farooque, 2018).

5.3.1.2 Meeting frequency.

In accordance with stakeholder–agency theory, a tendency was found for AC meeting frequency positively influencing CSR reporting in Australia (Appuhami and Tashakor, 2017; Arif et al., 2021), the Gulf Cooperation Council (Buallay and Al-Ajmi, 2020), Palestine (Musallam, 2018) and a cross-country setting (Pucheta-Martinez et al., 2023) as well as corporate political spending reporting in the USA (Ali et al., 2023), intellectual capital reporting in the Malaysia (Haji, 2015) and the UK (Li et al., 2012), and integrated reporting in South Africa (Haji and Anifowose, 2016) and a cross-country setting (Raimo et al., 2021). A few studies found no significant effects on either CSR reporting in Australia and the UK (Kend, 2015) and Saudi Arabia (Umar et al., 2023b) or environmental reporting in the USA (Giannarakis et al., 2020).

5.3.1.3 Other audit committee resources, incentives and diligence proxies.

While AC authority and integrated reporting were positively related in South Africa (Haji and Anifowose, 2016), managerial ownership did not have a significant impact on intellectual capital reporting in the UK (Li et al., 2012).

5.3.2 Impact on corporate social responsibility performance.

Few studies have analyzed the impacts of AC resources, incentives and diligence on CSR performance. Whereas Mohammadi et al. (2021) found that AC size increased CSR performance in Iran, in a cross-country study, Dwekat et al. (2020) found that it decreased CSR performance and Umar et al. (2023a) found no impact in Nigeria. While AC meeting frequency was found to increase CSR performance in a cross-country setting (Dwekat et al., 2020) and Nigeria (Umar et al., 2023a), Perez-Cornejo et al. (2019) found that the variables were unrelated in Spain.

5.3.3 Impact on corporate social responsibility assurance.

Few studies have analyzed the effects of AC resources, incentives and diligence on CSR assurance. AC size was found to have an insignificant impact on CSR assurance in cross-country settings (Dwekat et al., 2022), the UK (Al-Shaer and Zaman, 2018; Kend, 2015) and New Zealand (Zaman et al., 2021). Other tendencies include the positive connection between AC meeting frequency and CSR assurance in cross-country settings (Dwekat et al., 2022), the UK (Al-Shaer and Zaman, 2018; Kend, 2015) and New Zealand (Zaman et al., 2021).

5.4 Summary of the major results

Table 3 summarizes the positive, negative and insignificant relationships between the AC variables (presence; composition and resources, incentives and diligence) and CSR reporting, performance and assurance. Most studies focused on AC composition and CSR reporting. AC resources, incentives, diligence, CSR performance and CSR assurance proxies were less relevant. While research results in other directions have been either too heterogeneous or scarce, those on AC composition, CSR performance and assurance have shown positive significances. Moreover, AC resources, incentives and diligence increased CSR reporting. These positive connections are inconsistent with stakeholder–agency theory, highlighting the monitoring role of ACs in satisfying stakeholder demands. Previous studies have mainly matched traditional AC attributes (e.g. financial expertise, independence and size) with CSR variables.

6. Limitations and future research recommendations

Although many AC and CSR variables have been examined in the literature, we identified major research gaps and developed some recommendations for future research.

6.1 Content-related recommendations

From a content perspective, future research on the presence of ACs will be of less relevance, as this presence is either considered a “best practice” in listed firms related to national corporate governance codes or is already regulated by governments. The use of a dummy variable is associated with the limited validity of empirical research. Analyses of the influence of AC composition on CSR should be related to CSR performance, CSR assurance and the various CSR reporting subtypes. As ACs must monitor CSR reports and related management systems, future studies on the impacts of ACs on biodiversity reporting, climate performance and the circular economy will be more useful. Moreover, researchers should explicitly include the sustainability expertise of AC members (Velte, 2018).

While research on the relationship between CSR committees and CSR has increased, little is known about the sustainability expertise of ACs and their influences on CSR performance, reporting and assurance (Velte, 2018). Future studies should conduct content analyses of AC members’ CVs and evaluate their environmental and social skills. The CVs of AC members include descriptions of the practical and academic backgrounds of the board members (e.g. work experience and education). While the publication of CVs is voluntary, they are normally included in business reports or published separately on firms’ webpages. CVs can be analyzed to determine the sustainability expertise of AC members, including expertise related to the UN Sustainability Development Goals or the environmental goals of the Taxonomy Directive of the European Union (e.g. biodiversity, the circular economy and water management). Both theoretical/academic sustainability skills (e.g. sustainability-related study programs) and practical experience with sustainability practices (e.g. previous jobs in nongovernmental organizations) can be included.

In addition, overlaps between sustainability board committees and ACs should be identified. Sustainability board committees may support both executive and nonexecutive directors in making sustainability decisions in leadership and monitoring. Although these committees are not mandatory in most regimes, many firms have implemented them as a “best practice,” and close cooperation with ACs is crucial to increasing the probability that the firm will become more sustainable.

We should also emphasize that studies on the impacts of chief sustainability officers (CSOs) on CSR outputs are rare. As CSOs are responsible for corporate sustainability strategies, missions and visions, they should work closely with other top management team members, especially chief financial officers (CFOs) and CEOs. While implementation is voluntary in most regimes, many listed firms have implemented a CSO as a signal to their stakeholders. Future research should investigate whether CSOs or the characteristics of CSR committees moderate or mediate the impacts of ACs on CSR. The moderating influence of sustainability board committees on other economic relationships, such as the influence of CSR performance on financial stability, has recently been explored (Orazalin et al., 2024). Since ACs are responsible for monitoring CSR reports, related performance figures and assurance procedures, the strategic function of CSOs in sustainable transformation processes is crucial to ensuring a positive influence of ACs on CSR.

Regarding AC resources, incentives and diligence, there is a need for research on the impact of AC compensation on CSR, especially performance and assurance. In line with previous research on executive compensation, analyses of sustainability incentives in AC compensation contracts and their influence on CSR are useful because research on AC compensation has been neglected. AC members should be motivated to push top managers toward environmental and social targets. Content analyses on compensation, corporate governance and sustainability reports should be conducted. Automatized textual analyses of CVs and AC reports, such as those carried out using Python and algorithms, should be conducted to analyze qualitative information on AC members. These methods are also useful for building individual measures of CSR report quality (e.g. of carbon reports and human rights disclosures).

6.2 Methodological recommendations

Archival studies on the link between ACs and CSR have been affected by significant endogeneity problems, such as omitted variables and reverse causality (e.g. Wintoki et al., 2012). As mentioned earlier, many studies in our literature review measured correlations instead of causality (see Panel F in Table 2). While the included studies focused on the influence of ACs on CSR, inverse, bidirectional or nonlinear relationships are also plausible. As AC variables may not have a linear connection with CSR measures, (inverted) U-shaped links should be included. To improve CSR measures, AC effectiveness may also be related to a critical mass of AC members, including sustainability experts. Advanced regression models, such as 2SLS models and instrumental variables, GMM, PSM and difference-in-difference approaches, should be included as “best practices” to increase the validity of AC research (e.g. Ali et al., 2023; Alkurdi et al., 2023). Studies on voluntary AC implementation should control for sample selection bias using the two-stage Heckman approach and inverse Mills ratio. Only four studies in our review referred to this method. We found no study that used the difference-in-difference approach in line with a solid identification strategy. Quasi-natural experiments, such as the difference-in-difference approach, require a clear identification strategy and the inclusion of exogenous shocks. Owing to the increasing number of CSR and corporate governance regulations, regulatory shocks are helpful.

Cross-country settings should analyze country effects such as the strength of CSR performance within a specific regime, code versus case law regimes and cultural factors. We expect that the impact of ACs on CSR will be more pronounced in countries with a code law regime and that CSR performance will increase with enforcement strength. We summarize the major research recommendations in Table 4.

7. Summary

This literature review summarizes the results of 57 archival studies on the influence of ACs on CSR. This study was based on stakeholder–agency theory (Hill and Jones, 1992), with the assumption that ACs, as corporate governance instruments, increase CSR reporting, performance and assurance (Said et al., 2009). AC effectiveness is associated with fewer information asymmetries and fewer conflicts of interest between executives and stakeholders (Uyar et al., 2023). In addition to monitoring duties related to financial reporting, ACs are responsible for CSR reports, performance measures and assurance (Pozzoli et al., 2022; Pucheta-Martinez et al., 2023). This motivates the theoretical framework of stakeholder–agency. As many regulators have implemented rules on board governance and sustainability reporting in recent years, the relationship between ACs and CSR has been highlighted.

In view of these recent trends and in line with previous AC studies, we separately analyzed 1) presence, 2) composition and 3) resources, incentives and diligence and distinguished between CSR reporting, performance and assurance. The focus on CSR distinguishes our review from previous literature reviews and meta-analyses on ACs (e.g. Bilal et al., 2018). As our focus was on the CSR consequences of ACs and the comparability of the included studies, we included only archival studies. Most of the included studies concentrated on AC composition and CSR reporting. AC composition and CSR performance and assurance were found to be positively correlated. Moreover, AC resources, incentives and diligence increased CSR reporting. Studies on other relationships were either too scarce or produced heterogeneous results. Future research should focus on the sustainability expertise of AC members, as previous studies have focused on financial expertise and largely neglected environmental and social knowledge. As ACs are responsible for financial and CSR reporting, a combination of financial, industry and sustainability knowledge is needed among their members. Moreover, other sustainable board characteristics should be included as moderators or mediators in future research (e.g. CSR committees, CSOs and CSR-related compensation).

Our vote-counting approach limited our research method, as we only recognized the number of significances and did not measure sample or effect sizes (Light and Smith, 1971). Although a quantitative meta-analysis could have overcome these limitations (Bilal et al., 2018), our AC and CSR variables were too heterogeneous. Moreover, the number of studies included was too small regarding specific AC attributes to conduct a meta-analysis. As quantitative meta-analyses have been used in sustainability research, and studies on the link between AC and CSR outputs are expected to increase (Bilal et al., 2018), more meta-analyses will definitely be conducted in the future.

In the following paragraphs, we present the managerial implications of our study. While previous studies have focused on the classical financial reporting duties of ACs, our study emphasizes the managerial implications of ACs for CSR reporting, performance and assurance. Companies should be aware that ACs are crucial to their CSR activities. As stakeholders judge the effectiveness of ACs based on corporate governance reports or CVs, the transparency of the descriptions of AC member profiles and duties should be increased in the future. AC members should explicitly report on their explicit sustainability skills (environmental and social) on a concrete level. They should also report on their practical and theoretical/academic skills related to sustainability. Firms should include a critical mass of environmental and social experts in ACs to increase CSR reporting, performance and assurance practices. Developing environmental skills is a major challenge, as a background in the natural sciences is needed. Second, expertise in a combination of financial, industrial and sustainability knowledge is useful in ACs for the following reasons: ACs are responsible for the monitoring of financial and CSR reports, which leads to an integrated thinking process (Velte, 2018). Sustainability reports and management systems interact with financial reports and related processes. This triangle of financial, industry and sustainability skills in ACs should reduce CSR decoupling and greenwashing (Albitar et al., 2023). Firms should provide AC members with incentives to conduct seminars on sustainability issues in dynamic and complex environments, with a special focus on environmental risks and opportunities. Furthermore, close cooperation between audit and sustainability committees should be realized to promote successful corporate sustainable transformation. If an adequate level of sustainability expertise cannot be achieved in the AC, external environmental experts should be engaged to verify environmental information (e.g. climate neutrality disclosures).

Our study has major theoretical implications. While previous studies have stressed the conflicts between stakeholder theory and principal agent theory, our study aimed to use stakeholder–agency theory (Hill and Jones, 1992) as a combined theoretical framework. By integrating these theories, the complex dynamics between stakeholders and ACs were highlighted (Qaderi et al., 2024). In accordance with stakeholder–agency theory, ACs should facilitate dialog between stakeholders and compromises in terms of different stakeholders’ interests. Contrary to classical agency theory, ACs should not only decrease agency conflicts (information asymmetries and conflicts of interest) between management and shareholders but also be responsible for overseeing an organization’s sustainability practices and ensuring that the company remains answerable to a diverse range of stakeholders. Thus, we stress the urgent need for the integration of stakeholder interests into the theoretical foundation of ACs. ACs are a major corporate governance instrument that should satisfy the needs of shareholders and other stakeholder groups.

Finally, we present the regulatory implications of our study. Over the last few decades, standard setters have increased the mandatory job profiles and composition requirements of ACs. Traditionally, the focus has been on financial reports, related management systems and financial audits. More recently, corporate sustainability regulations have directly affected the CSR duties of ACs. Among other things, in accordance with the EU Corporate Sustainability Reporting Directive, ACs must monitor sustainability reports, management systems and CSR assurance processes. The risks of information overload and greenwashing of CSR actions cannot be reduced without supervision of financial and CSR processes by ACs. Standard setters should increase cooperation between ACs, internal audits and CSR assurors in the future. Moreover, more regulatory incentives should be provided to engage external environmental experts and support corporate environmental audits.

Figures

Figure 1

Research framework

Figure 2

Flow diagram of the sample selection process

Table 1

List of main AC and CSR variables

List of AC variables

List of CSR variables

• Presence • Composition: a) Effectiveness/competencies score of several proxies b) Independence (ratio; chair) c) Expertise (financial; sustainability; industry) d) Diversity (age; foreign; gender) e) Education/experience/tenure (chair as number of years) • Resources, incentives and diligence: a) Size (resources) b) meeting frequency (diligence) c) ownership (incentives)

• CSR reporting: a) Overall CSR reporting (quality, based on content analyses or databases) b) GRI adoption c) Environmental reporting (carbon reporting) d) Political spending reporting e) Intellectual capital reporting f) Integrated reporting • CSR performance: a) Total CSR performance measures (based on databases) b) Environmental performance c) Social performance d) Governance performance e) donations f) Reputation (score) • CSR assurance: a) CSR assurance (presence; provider; level; quality) a) Integrated reporting assurance

Source: Author’s own creation/work

Table 2

Count of cited published papers

Panel A: by publication year
Total: 57	2024: 2 2023: 13 2022: 12 2021: 6 2020: 5 2019: 3 2018: 6 2017: 2 2016: 3 2015: 2 2013: 1 2012: 1 2009: 1
Panel B: by region
Total: 57	Cross-Country: 10 Australia/UK: 1 Australia: 3 Bangladesh: 1 Canada: 1 China: 1 France: 1 Gulf Cooperation Council: 2 Hong Kong: 1 India: 1 Indonesia: 2 Iran: 3 Jordan: 2 Malaysia: 3 New Zealand: 1 Nigeria: 1 Pakistan: 1 Palestine: 1 Saudi Arabia: 4 South Africa: 2 Spain: 2 Spain/Portugal: 1 Thailand: 1 Uganda: 1 UK: 4 USA: 6
Panel C: by journal
Total:	Accounting, Auditing and Corporate Governance Journals (28): Accounting & Finance: 1 Accounting Research Journal: 1 Asian Journal of Accounting Research: 1 Australian Accounting Review: 1 Cogent Economics & Finance: 1 Corporate Governance: 2 European Accounting Review: 1 International Journal of Accounting & Information Management: 3 International Journal of Disclosure and Governance: 4 International Journal of Finance & Economics: 1 Journal of Accounting & Organizational Change: 1 Journal of Applied Accounting Research: 2 Journal of Corporate Accounting & Finance: 1 Journal of Financial Reporting and Accounting: 1 Journal of Islamic Accounting and Business Research: 1 Journal of Sustainable Finance & Investment: 2 Managerial Auditing Journal: 2 Pacific-Basin Finance Journal: 1 The British Accounting Review: 1 Management and sustainability journals (29): Business Strategy and Development: 1 Problems and Perspectives in Management: 1 Business Strategy and the Environment: 5 Business & Society: 1 Academia Revista Latinoamericana de Administracion: 1 International Journal of Productivity and Performance Management: 1 Journal of Cleaner Production: 3 Gender in Management: An International Journal: 1 Benchmarking: An International Journal: 1 Journal of Management and Governance: 1 Corporate Social Responsibility and Environmental Management: 2 European Management Journal: 2 Business Ethics, the Environment & Responsibility: 1 EuroMed Journal of Business: 1 Journal of Business Ethics: 2 Social Responsibility Journal: 3 Critical Perspectives on International Business: 1 Annuals of Operations Research: 1 Sustainability Accounting, Management and Policy Journal: 1
Panel D: by dependent variable
Total: 60*	CSR reporting: 40 CSR performance: 14 CSR assurance: 6
Panel E: by independent variable (AC)
Total: 125*	Presence: 6 Composition: 77 Resources, incentives, & diligence: 42
Panel F: By regression analyses
Total: 79	OLS: 23 panel: 18 GMM: 11 Probit/logit: 9 2 stage least squares (2SLS)//instrumental variable (IV): 7 Propensity score matching (PSM): 3 Heckman 2 stage approach: 4 Structural equation modeling (SEM)/3 stage least squares (3SLS): 4
Panel G: By theoretical framework
Total: 92*	Principal agent theory: 40 Stakeholder theory: 17 Legitimacy theory: 14 Resource dependency theory: 8 Signaling theory: 3 Other theories: 6 N.A.: 4

Note:

*Some studies include more than one dependent variable/regression method/theory

Source: Author’s own creation/work

Table 3

Summary of results of the literature review

Audit committee characteristics		CSR reporting	CSR performance	CSR assurance	Total
Presence	(+) (−) (+/−)	4 1 0	1 0 0	0 0 0	5 1 0
Composition	(+) (−) (+/−)	24 7 16	15 2 3	10 0 0	49 9 19
Resources, incentives and diligence	(+) (−) (+/−)	21 0 14	4 0 2	3 0 4	28 0 20
Total	(+) (−) (+/−)	49 8 30	20 2 5	13 0 4

Source: Author’s own creation/work

Table 4

Summary of research recommendations

Content-related research recommendations

Methodological research recommendations

Focus on CSR performance, CSR assurance and subpillars of CSR reporting (e.g. biodiversity, circular economy, climate) Sustainability expertise of AC members (based on prior practical experiences and/or theoretical/academic background) Connectivity between AC and CSR board committees (interactions; overlaps of members) AC compensation (e.g. equity-based, sustainability-related targets) Automated text analyses of CVs of AC members and business reports to gather innovative AC and CSR variables

Endogeneity checks by quasi-natural experiments (e.g. difference-in-difference-approach and 2SLS/IV) Bidirectional and nonlinear relationships, e.g. critical mass of AC variables Country effects (e.g. code versus case law, CSR performance, cultural dimensions)

Source: Author’s own creation/work