Audit committee effectiveness, audit quality and earnings management: evidence from Ghana

Abstract

The paper investigates the effect of Audit Committee Effectiveness (ACE) on Earnings Management (EM) and consequently examines whether audit quality (Big4) can moderate the link between ACE and EM in Ghana. First, the study uses panel data from 25 non-financial firms in Ghana, and employs the fixed-effect (FE) and Two-stage Least-squares (2SLS) regression methods. Second, the evidence shows that ACE (AC independence, AC size and AC meetings) constrain EM with the effect being stronger for AC independence. Thus, firms with ACE tend to report lower earnings management. Third, further analysis reveals that the combined effects of Big4 and ACE (AC independence and AC meetings) on EM are stronger than ACE alone. This suggests that audit quality negatively moderates the link between ACE and EM. Finally, leverage, profit, cash flow and firm size affect EM.

Keywords:

• Earnings management
• audit committee effectiveness
• audit quality

SUBJECTS:

• Finance
• Economics
• Business
• Management
• Accounting

REVIEWING EDITOR:

• Collins Ntim, University of Southampton, United Kingdom of Great Britain and Northern Ireland

Introduction

Firm efficiency is critical to sustainable economic development (Proença & Soukiazis, 2023). However, firm efficiency cannot be achieved if firms are involved in opportunistic behaviours and compromise on financial reporting quality and integrity. Such behaviour exposes firms to failure. An example of such behaviour, reported as the leading cause of firm failure is earnings management (EM) (Ghazali et al., 2015; Gross et al., 2024; Veganzones et al., 2023). Earnings management is defined as a discretionary financial reporting strategy meant to mislead stakeholders about firms’ actual performance by adjusting accounting transactions or methods to suit such purposes (Sawarni et al., 2023). This paper defines earnings management as discretionary accruals. Discretionary accrual is the extent to which accruals embedded in earnings are opportunistically used by management (Jones, 1991). Earnings management is an acknowledged unethical accounting practice with serious adverse effects on firms. At the extreme, earnings management can collapse a firm (Baskaran et al., 2020; Cai et al., 2020; Tohang et al., 2024). More importantly, however, is that earnings management can reduce a firm’s profitability and productivity (Fariha et al., 2022). Earnings management compromises earnings quality and increases information asymmetry and investment inefficiencies (Le & Nguyen, 2023). At the legal level, management or firms engaged in unethical corporate practices such as earnings management can face legal consequences. This paper responds to the broader ethical and corporate governance accounting literature to make contributions to the current literature by: (i) investigating the direct relationship between audit committee effectiveness (ACE) and earnings management, and (ii) examining the extent to which audit quality can moderate the link between ACE and earnings management, with particular focus in Ghana.

According to the agency theory (Jensen & Meckling, 1976), the separation of ownership in firms increases information asymmetry between managers and owners. Managers take advantage of the information asymmetry to pursue private interests such as hiding their inefficiencies in earnings management. Therefore, an effective audit committee is required to mitigate management opportunistic behaviours (Elghuweel et al., 2017). However, the effectiveness of the audit committee (AC) in monitoring and ensuring effective financial reporting processes depends on its features such as size, independence, gender diversity, frequency of meetings, and financial expertise (Biswas et al., 2023; Chemingui et al., 2023; Elghuweel et al., 2017; Stewart & Munro, 2007). For instance, AC independence is critical to mitigating unethical earnings practices. Independent AC has no affiliation nor tie with the executives which enhances objectivity in their assessment and judgement of financial statements by management. AC independence also ensures that the external auditor is evaluated objectively. Also, due to reputational risk, they strongly scrutinize financial information to avoid managerial opportunism. Moreover, AC gender is another essential feature of effective AC. Studies (e.g. Zalata et al., 2022) posit that females have unique characteristics compared with men. Drawing insights from behavioural, economic, psychology, ethical, moral, development and social role theories, females are less aggressive and competitive which enable them to be more independent, ethical, objective, risk-averse, vigilant, tend to confer greater importance on ‘communal value such as communication, building relationship and trust making them more, caring, credible and responsible in providing effective monitoring and supervisory function over management compared with men (Powel & Ansic, 1997; Kao et al., 2020; Sarin & Wieland, 2016; Zalata et al., 2022). By contrast, Nguyen (2023) find a U-shaped relationship between women on AC and financial reporting quality. Further, the size of AC is critical to its effectiveness. A larger AC size is deemed to have a larger diversity and skill set to achieve effective financial reporting quality and financial monitoring.

As a managerial strategy to meet earnings benchmark, smoothen earnings, or fraudulently misrepresent financial information, management uses obfuscation strategies to make earnings management harder to detect (Li et al., 2024) such that even AC is insufficient and incapacitated to fully identify the earnings discrepancies. Therefore, an external auditor becomes necessary to ensure that the financial statement by management accurately responds to certified accounting and auditing standards, owners, institutions and stakeholders. Audit quality refers to the key elements that create an atmosphere to promote quality audits consistently. Audit quality can be achieved when the team exhibits appropriate values, ethics, and attitudes, is sufficiently knowledgeable, and experienced, has time for the audit work, applies rigorous accounting and auditing methods and procedures within applicable laws and regulations, and interacts with relevant stakeholders (International Auditing & Assurance Standards Board [IAASB], 2014). Also, DeAngelo (1981) defines audit quality ‘as the market assessed joint probability that a given auditor will both discover a material misstatement in the client’s accounting system and subsequently report’. Following the literature (eg Donelson et al., 2020; Rajgopal et al., 2021; Sitanggang et al., 2020), this study measures audit quality with the Big 4. The Big 4 are reputable, sophisticated, technical, resourced, professional and technologically resourced audit institutions which provide effective audit functions, and they include PWC, KPMG, Deloitte, and Ernst and Young. In this case, Jensen and Meckling (1976) suggest that good corporate governance tools such as ACE and audit quality are critical to gain, maintain or repair broken standards in firms because the Big 4 will effectively scrutinise financial reports, and will report any identified earnings discrepancies. We therefore use the agency theory as the central reason underlying how effective AC and audit quality can constrain firms’ earnings management. Strong compliance with CG standards via an effective AC and the appointment of a competent external auditor will increase financial disclosure while reducing earnings practices.

Consequently, the extant literature has explored the associations between audit committee effectiveness and earnings management (e.g. Carrer et al., 2017; Chatterjee & Rakshit, 2023; Fernández-Méndez & Pathan, 2023; Habbash et al., 2013; Hasan et al., 2020; He et al., 2009; Komal et al., 2023; Le & Nguyen, 2023; Mardessi, 2022; Mardessi & Fourati, 2020; Mardjono & Chen, 2020; Mnif & Tahri, 2023; Mohd Saleh et al., 2007; Nikulin et al., 2022; Rahman et al., 2023; Sharma & Iselin, 2012; Sun et al., 2014). However, the current literature appears to have some weaknesses. First, As noted by Ntim (2022, p. 72), scientific studies on accounting and finance which predominantly exist in advanced countries tend to address exclusively the needs of the context of research settings, which impairs the distinctiveness of their findings to be generalizable to emerging economies such as Ghana. Approaches, methods, ethical concerns, resource limitations, regulations, culture, firms’ characteristics and issues of audit distinctively differ between advanced economies and developing African economies. Therefore, studies on emerging economies must factor in and understand the distinctiveness of the emerging country context in which the study is being conducted. Second, current literature on the association between AC effectiveness and earnings management is limited in Ghana (Agyei-Mensah et al., 2019; Agyekum et al., 2014). Particularly, studies in Ghana focus primarily on the performance implications of ACE (e.g. Agyemang, 2020; Awinbugri & Prince, 2019; Glover-Akpey & Azembila, 2016; Ibrahim et al., 2017), and neglect to provide empirical evidence on how it addresses earnings practices. Finally, existing studies that examined the association between ACE and earnings management moderated by audit quality are generally limited (Alia et al., 2020; Imen & Anis, 2021; Iriyadi, 2019; Zgarni et al. 2016) and acutely so in developing countries such as Ghana. Studies in Ghana (e.g. Afenya, Arthur, Kwarteng, & Opoku, 2022; Afenya, Arthur, Kwarteng, & Kyeremeh, 2022; Agyei-Mensah & Yeboah, 2019) investigate the association among ACE, audit quality and EM without investigating their complementarity effects on EM. Arguably, these limitations in the current literature limit understanding of how and why and to what extent ACE might mitigate earnings practices, and whether audit quality moderates the ACE and earnings management nexus.

Again, the study is motivated by its contextual settings to current happenings and shortcomings in audit practices in Ghana. The audit committee function and the role of external auditors in corporate governance is critical to regulators in Ghana following the collapse of some financial institutions in Ghana including the collapse of microfinance institution, DKM in 2015, which cost the country millions of dollars to compensate depositors who lost their deposits. Also, in 2017 and 2018, five banks collapsed, whereas other banks merged as consolidated banks (Afenya, Arthur, Kwarteng, & Kyeremeh, 2022). In any of these situations, the regulators cited corporate governance weaknesses and audit inefficiencies as the source of the crises (Bank of Ghana [BOG], 2017a). In response, the regulator has renewed its commitment to ensure stringent compliance with corporate governance and auditing practices in Ghana. Key legislations include the strengthening of firms’ audit committee functions and external auditor roles in ensuring that firms comply with relevant accounting standards to ensure quality financial reports. The legislation addresses and recommends the right balance of AC members regarding diversity, knowledge, expertise and qualification. In this study, we provide empirical evidence on how effective AC and audit quality complement to address firm inefficiencies such as earnings management.

Thus, this study seeks to contribute to the current literature in the following ways: First, we examine why and how ACE such as AC size, AC gender diversity, AC independence and AC meetings influence firms’ earnings management. Specifically, the paper investigates whether firms with strong (weak) ACE tend to demonstrate lower (higher) earnings management. Second, the paper establishes whether effective AC contributes to less earnings management in the presence of external audit quality. This is distinct from prior studies which only examined the link between ACE and earnings management. The proposition is based on theoretical and empirical evidence which suggests that the selection of a quality external auditor is based on the effectiveness of the selection body (AC) (Azizkhani et al., 2023). The conjecture is influenced by theoretical and empirical evidence that while effective AC is critical to mitigating managerial opportunism, audit quality (Big 4) is better positioned to identify earning discrepancies and align hedging decisions (Alzeban & Sawan, 2015; Azizkhani et al., 2023; Hoitash & Hoitash, 2009; Tai et al., 2020). Thus, the theoretical and empirical implication is that a negative AC – EM relationship can be further strengthened with the Big 4 (Audit quality). However, current literature ignores the possible moderating effect of audit quality on ACE and earnings management nexus. Hence, this paper augments the earlier findings on the direct negative effect of ACE on EM by investigating the moderating effect of audit quality in this link. The findings are two-fold. First, using Ghanaian firm data, the results suggest that ACE (AC size, AC meetings and AC independence) constrain EM with the association being stronger on AC independence. Second, the negative impact of ACE on EM is stronger when combined with the Big4.

2. Background

Good corporate governance (CG) improves firm performance (Bawuah, 2024; Farooq et al., 2022). Effective accountability in the use of public or firm resources is a feature of good CG. Accountability produces greater outcomes in using scarce resources (Mejía Acosta, 2013). Based on these reasons, regulators of firms consistently provide instructions to firms to provide procedural clarity, competency and credible firm outcomes. CG mechanisms such as the ACs and audit quality have been recognized in the aftermath of weak corporate governance-induced corporate failures such as Enron in the US (Alzeban & Sawan, 2015). In response, quality audit practice is recommended as an important component of good corporate governance worldwide (Irwansyah & Zega, 2023; Mangala & Singla, 2023; Nguyen et al., 2024). The audit function provides independent assurance and professional assessment on whether firm resources have been used efficiently to ensure accountability to strengthen public confidence in firms (Appiah et al., 2023; Oppong et al., 2023). Section 5(1) (d) of Ghana’s Public Financial Management (PFM) Act 2016 (Act 921), the Internal Audit Agency Act 2003 (Act 658), the Audit Service Act. (2000 (Act 584), Guidelines for Effective Functioning of Audit Committee 2023, International Professional Practice Framework and International Standards of Supreme Audit Initiations provide regulations for auditors to support government responsibility of oversight, insight and foresight over firms and public institutions (MOF, 2016; Internal Audit Agency (IAA) Act, 2003 (Act 658) [IAA], 2003)

Section 87 of the PFM Act 2016 (Act 921) details the membership of AC of firms: AC shall consist of 5 members; the majority of the AC members must be independent; the Internal Audit Agency (IAA) and the Institute of Chartered Accountants Ghana (ICAG) shall nominate a majority of the AC members from among persons who do not work in the AC’s covered entity; AC chairperson shall be elected from the independent AC members. AC can co-opt a senior management member to join the AC; the AC members and the Chair shall be appointed by the Principal Account Holder (Ministry of Finance (MOF), 2016). The primary objective of the AC is to provide oversight over the financial reporting process, audit process, risk management system, system of internal controls, and compliance with laws and regulations (Ministry of Finance (MOF), 2016). Other supporting roles include providing advice on sound and transparent financial reports, appointment of external auditor, infusing firm-wide commitment to strong internal control, reviewing and approving the internal audit charter, ensuring the independence and objective of the internal audit unit (Ministry of Finance (MOF), 2016; IAA, 2003). Despite these regulatory provisions, the effectiveness of CG and internal audits in Ghana in general have not been effective due to the low level of professional proficiency of internal auditors; lack of managerial support to the internal audit unit; lack of budget support for internal audit activities and weak functioning of audit committee (Adomako et al., 2023; Daud et al., 2020; Musah et al., 2021; Onumah & Yao Krah, 2012). As a result, there have been some weak CG-induced corporate failures in Ghana.

The Bank of Ghana (BOG) in its press release on August 14, 2017, approved the takeover of insolvent UT Bank LTD and Capital Bank LTD by Ghana Commercial Bank (GCB) to protect customers’ deposits and improve the banking sector resilience (BOG, 2017a). Just after a year, on August 1, 2018, BOG consolidated other five indigenous banks (Unibank Ghana LTD, Royal Bank LTD, Beige Bank LTD, Sovereign Bank LTD, and Construction Bank LTD) into a new bank called Consolidated Bank Ghana LTD (CBG) due to same insolvency issues. The regulator cited poor corporate governance as the major cause of the collapse of the banks (Appiah et al., 2023; Boachie, 2023; Osei et al., 2019; Torku & Laryea, 2021). The banks managements were cited to have engaged in personal activities that inured to their private gains rather than the bank’s growth (Appiah et al., 2023; BOG, 2017a). For instance, most of the bank’s loans and the non-performing loans were extended to managers and senior managers of the banks (BOG, 2017b). Further, the board of the banks failed to provide effective oversight and monitoring functions, as well as poor external auditing functions (BOG, 2017b). Also, the internal auditing lacked independence, integrity and transparency in their activities. In summary, poor adherence to corporate governance policies, especially effective internal and external audit functions contributed to the banks collapse (Appiah et al., 2023). In response, the regulator implemented several reforms including the strengthening of firm’s internal and external audit function to avoid the reoccurrence of the crises and restore audit integrity. In this study, we investigate the role of ACE and external auditors in addressing banks’ inefficiencies such as earnings management and provide policy and managerial recommendations.

3. Theoretical framework

This study is based on the agency theory. Agency theory was proposed by Jensen and Meckling (1976). The theory is applied in this study to explain the relationship between ACE, audit quality, and earnings management. The theory states that shareholders are the principal of a firm who appoints the management (agent) to act and maximize their interests (Alkebsee et al., 2022; Lu et al., 2022). The theory further explains that management may act in their interests to the detriment of the principal. Due to the separation of control between shareholders and management, the former is unable to comprehensively observe the actions of management (Jensen & Meckling, 1976). Therefore, agency theory proposes good corporate governance practices to ensure that management does not pursue interests that do not maximize shareholders wealth. Corporate governance controls align the interests of shareholders and management and discourage opportunistic behaviour from one side (Rahman et al., 2023). AC effectiveness (AC independence, AC size, AC gender diversity, AC meetings) and audit quality are examples of corporate governance practices to control management opportunistic behaviours and mitigate agency costs (Afenya, Arthur, Kwarteng, & Opoku, 2022; Carrer et al., 2017). AC meetings ensure the free flow of information between the principal and agents, and reduce information asymmetry (Liu et al., 2023). In general, AC is highly recommended as an effective monitoring mechanism to achieve quality corporate financial reports and discourage earnings management (Alkebsee et al., 2022)

Further, audit quality is an important dimension of corporate governance (Al-Ahdal & Hashim, 2022). Audit quality provides independent and reliable financial reports on a firm’s financial position. As earlier indicated, the separation of ownership and control of a business results in conflicts between stakeholders. Thus, companies are obliged to use corporate governance practices such as audit quality to reduce agency costs and prevent information asymmetry and reputational loss from corrupt management practices (Al-Ahdal & Hashim, 2022; Becker et al., 1998; Chen et al., 2005; Le & Moore, 2023). According to agency theory, upholding quality corporate governance dimensions improves a firm reputation in the industry. However, the mere establishment of an AC does not mean that financial statements will be reliable. However, AC must possess the essential qualities (such as competence, independence, gender-balanced, sufficient size to accommodate diversity, and audit diligence through the frequency of AC meetings) required to achieve effective performance. Generally, audit committee effectiveness and audit quality have negative effects on earnings management (Fernández-Méndez & Pathan, 2023; Mnif & Tahri, 2023), while ACE such as AC size, meetings, and independence have positive effects on audit quality (Habib et al., 2022; Mardessi, 2022; Tai et al., 2020). Agency theory predicts a negative association between ACE and earnings management and further proposes a positive relationship between ACE and audit quality (Tai et al., 2020). In this study, we seek to identify whether the complementarity of ACE and audit quality constrain earnings management as highlighted by the agency theory.

4. Literature review and hypotheses development

4.1. Audit committee effectiveness and earnings management

Agency theory predicts a negative relationship between AC effectiveness such as AC gender diversity, AC independence, AC size, AC meetings and firm inefficiencies such as information asymmetry, managerial opportunism and earnings management (Jensen & Meckling, 1976; Prawitt et al., 2009; Stewart & Munro, 2007; Zalata et al., 2022). Researchers point to agency theory to explain the potential conflicts of interest between managers and owners of a company (Carrer et al., 2017). The conflict arises partly from the information asymmetry problem: management has more information, especially on earnings than investors. Management has higher incentives to capitalize on this information at the expense of investors. To reduce information asymmetry and moral hazards, and maintain financial report integrity, quality internal audits should be pursued vigorously (Christensen, 2022; Kaawaase et al., 2021; Saeed et al., 2024). Agency theory proposes that essential corporate governance characteristics such as effective AC and audit quality can be used to mitigate agency risk and managerial opportunism such as earnings management (Hasan et al., 2020; Ji et al., 2020).

Earnings management can be motivated by both opportunistic and non-opportunistic behaviours (Zalata et al., 2022). For instance, management has a higher incentive to engage in earnings management to meet earnings benchmarks or to avoid reporting lower earnings. As explained by the prospect theory (Levy, 1992), the market or users of financial information are more induced by negative earnings than positive earnings (Zalata et al., 2022). In other words, the market rewards (punishes) firms that meet (miss) earnings benchmarks (Zalata et al., 2022). Therefore, firms are motivated to indulge in earnings management to meet earnings benchmarks. Also, management uses earnings management to smoothen earnings, thereby reducing risk premium and cost of equity (Zalata et al., 2022). Even though such non-opportunistic discretions may dampen fluctuations in earnings, they can be costly and increase restatements which can lead to litigation costs (Albrecht et al., 2018; Zalata & Abdelfattah, 2021). Therefore, the audit committee (responsible for monitoring the firm financial reporting process) are obliged to mitigate both opportunistic and non-opportunistic earnings practices by management. The audit committee mitigates managerial opportunism and protects shareholders’ interests by relying on their effectiveness (size, independence, gender, and meetings) to ensure that credible financial information is reported by management

Evidence shows that effective AC improves financial reporting quality (e.g. Salehi & Shirazi, 2016; Mardessi, 2022; Carrer et al., 2017; Rahman et al., 2023; Mishra & Malhotra, 2016; Zadeh et al., 2023). Firms with a high propensity to engage in financial reporting discrepancies are less likely to maintain effective audit committees (He et al., 2009; Sharma & Iselin, 2012). Also, firms with fraudulent financial reporting intentions rarely insist on AC independence and competency (Albrecht et al., 2018). Firms with effective AC with the right authority and competence are less likely to engage in fraudulent financial reporting or earnings management practices (Xia et al., 2024). Further, extensive evidence shows that ACs with more independent directors have less earning management practices (Chatterjee & Rakshit, 2023), and diverse-gender ACs are less likely to engage in accounting irregularities (Fernández-Méndez & Pathan, 2023; Mnif & Tahri, 2023). Also, Mishra and Malhotra (2016), Xie et al. (2003) and Lin et al. (2006) find a negative and significant relationship between AC size and meetings and earnings management. According to Bedeir (2023) firms that overstate their earnings usually have ineffective AC. Therefore, several studies report that firms with effective AC (independence, gender diversity, larger size, frequent meetings) can curtail earnings management by improving financial reporting quality and reducing accounting irregularities and fraudulent reporting practices (Agyei-Mensah & Yeboah, 2019; Carrer et al., 2017). In Ghana, the PFM Act 2016 recommends AC to constitute five members, have more independent members and possess the right mix of gender, experience, religion, skills and motivation to curtail managerial opportunism. However, few empirical studies exist on the link between ACE and earnings management in Ghana. Notwithstanding that, and based on the above arguments, this paper hypothesizes that:

H1: Audit committee effectiveness has a negative and significant effect on earnings management.

4.2. The role of audit quality

Agency theory predicts that ACE and audit quality provided by external auditors (Big 4) can constrain earnings management (Jensen & Meckling, 1976; Nguyen et al., 2024; Xia et al., 2024). Prior studies demonstrate that effective AC can strengthen external auditors and support them in undertaking activities that improve audit functions (Friedrich & Quick, 2024; Kalia et al., 2023). Effective AC supports the audit function by appointing industry specialist auditors. Also, effective ACs are less likely to have accounting problems or disputes with external auditors (Azizkhani et al., 2023). Effective ACs are more likely to side with an external auditor in a substance-over-form dispute with management (Rainsbury et al., 2009). Therefore, effective AC improves audit quality.

According to Azizkhani et al. (2023), AC chairs with higher business qualifications are more likely to hire Big 4 auditors, pay higher fees and achieve lower discretionary accruals. In contrast, an AC chair who is an executive director is less likely to hire a Big 4 auditor and have higher discretionary accruals. This indicates that high AC independence improves audit quality and mitigates earnings management. Moreover, research suggests that audit quality discourages managerial incentives to engage in fraudulent earnings practices (Alhadab & Clacher, 2018). External auditors, such as the Big 4, are mandated to apply the certified accounting and auditing standards, thus, having the mechanism to identify discrepancies in financial statements. Big 4, from a regulatory standpoint, has the technology to identify earnings management. Also, due to the high litigation risk of the Big 4, they are less likely to compromise with management. Therefore, management, while still having incentives to engage in earnings practices, may be less likely to have room to utilize all the cheating methods, due to robust external audit assessments. AC effectiveness such as AC gender and independence are important in external auditor selection and general efforts to reduce earnings management. For instance, female AC members are generally not members of the ‘old boys club’ and therefore are less likely to approve the appointment of an affiliated auditor. This improves AC independence and competence in achieving audit quality and mitigating earnings practices (Adams & Ferreira, 2009; Azizkhani et al., 2023). Cheung and Chung (2022), however, in Hong Kong, find that audit committee expertise is more likely to increase earnings management. They explain that members of AC, irrespective of their strong accounting, and finance background or experience, encourage managers to switch their earnings management strategies.

A negative relationship between audit quality and earnings management is largely supported (e.g. Becker et al., 1998; Chen et al., 2005; Chowdhury & Eliwa, 2021; Alhadab & Clacher, 2018; Imen & Anis, 2021; 2021; Le & Moore, 2023; Rusmin, 2010), and the positive relationship between effective ACE and audit quality is established (e.g. Afenya, Arthur, Kwarteng, & Opoku, 2022; Habib et al., 2022; Mardessi, 2022; Zgarni et al. 2016). For instance, in Ghana, Afenya, Arthur, Kwarteng, and Opoku (2022) find a negative and significant link between ACE (AC meetings, AC financial expertise, AC size, AC gender) and audit fees. They interpret their results that effective AC solidify the committee to ensure effective monitoring and also lower costs. Evidence on the role of audit quality in the relationship between audit committee effectiveness and earnings management is less explored. This study addresses that. Nonetheless, evidence shows that ACs with more independent members are more likely to appoint the industry’s external best auditor to constrain management fraudulent reporting behaviour (Azizkhani et al., 2023). Also, female AC members, influenced by their cognitively and psychologically distinct characteristics, compared to men, allow them to exhibit more ethical and strong monitoring behaviour, thus making them an effective complement with external auditors to mitigate earnings management (Salem et al., 2023; Kao et al., 2020). A study by Zgarni et al. (2016) in Tunisia find that the complementarity of audit committee effectiveness and audit quality has a negative effect on EM. This suggests that audit quality has a significant moderator effect on the link between ACE and earnings management. Therefore, the study hypothesizes that:

H2: Audit quality has a negative and significant moderating effect on the relationship between audit committee effectiveness and earnings management.

5. Research design

This study uses panel data from 25 firms in Ghana from 2010 to 2021 with a year observation of 275. The paper sampled both listed and non-listed non-financial firms. Financial firms were excluded because they tend to be highly regulated, and heavily leveraged, and therefore need to be analyzed separately (Ntim, 2016). The sample comprises several industrial sectors including consumer services, industrials, technology/telecoms, basic material and oil and gas. The focus was to cover all listed firms and Club 100 firms in Ghana. However, the study restricts it to 25 firms, deemed as purposively sampled, and the selection is based on data availability on all the study constructs. The variables are manually entered from the selected firm’s audited annual reports from their websites and include audit committee effectiveness, measured as AC independence, AC size, AC gender, and AC meetings. Other variables include earning management, audit fees and audit quality. The study uses control variables known to influence earning management, including leverage, firm size, cash flow, and profits.

5.1. Dependent variable

5.1.1. Earnings management

The study employs the modified–Jones (1991) model as used by Dechow et al. (1995) and other studies (e.g. Agustia et al., 2020; Bisogno & Donatella, 2022; Le & Nguyen, 2023; Sam et al., 2022). It estimates the non-discretionary accruals at the event year. It is computed as: $$\mathrm{\text{EM}}={\alpha }_1\mathrm{ }\left(\frac{1}{A_{t-1}}\right)+{\alpha }_2\mathrm{ }\left(\frac{{\Delta \mathit{\text{REV}}}_t-{\Delta \mathit{\text{REC}}}_t}{A_{t-1}}\right)+{\alpha }_3\mathrm{ }\left(\frac{{\mathit{\text{PPE}}}_t}{A_{t-1}}\right)$$

Where EM is earnings management, ${\Delta \mathit{\text{REV}}}_t$ is revenues in year t less revenues in year t-1; ${\mathit{\text{PPE}}}_t$ is property, plant and equipment; $A_{t-1}$ is total assets at the year-end; ${\alpha }_1-{\alpha }_3$ are firm-specific parameters.

5.2. Independent variables

The main independent variable is ACE which includes AC size, AC gender, AC independence and AC meetings.

5.2.1. Audit committee size

This study measures AC size as the number of members serving on the audit committee (Endrawes et al., 2020) as shown in Table 1. Evidence shows that AC size has a positive effect on audit disclosure. Larger AC size has the required diversity regarding AC independence, gender, skills and experience which enhances AC effectiveness in delivering quality audit outcomes (Juwita et al., 2020; Lin et al., 2006; Xie et al., 2003). On the other hand, Nguyen (2022) finds that AC with a small size enhances bank stability. Therefore, we expect AC size to impact negatively on EM.

Table 1. Variable description and measurement.

Variable	Symbol	Variable type	Measurement
Earnings Management	EM	Dependent	The absolute value of discretionary accruals
Audit Committee Independence	AC_IND	Independent	It is the percentage (%) of independent AC members on the AC
Audit Committee Size	AC_Size	Independent	It is the total number of members serving on the AC
Audit Committee Gender Diversity	AC_GENDER	Independent	It is the ratio of female AC members and AC size
Audit committee meetings	AC_MEET	Independent	It is the number of meetings held by the AC in a year
Audit Quality	Big 4	Moderator	The dummy variable equals 1 if the firm is audited by Big4 (KPMG, EY, PWC, Deloitte) and 0 otherwise
Profit	ROE	Control	It is earnings before interest and tax divided by total equity
Leverage	LEV	Control	The ratio of total debts to total assets
Operating Cash Flow	CF	Control	Operating cash flow this year to total assets of the previous year
Firm Size	SIZE	Control	Log 10 of the total assets of the firm

Source: Author’s elaboration from literature.

5.2.2. Audit committee independence

Following the literature, this study measures AC independence as the number of non-executive AC members on the AC (e.g. Alderman & Jollineau, 2020; Afenya, Arthur, Kwarteng, & Opoku, 2022), presented in Table 1. AC independence is generally considered to have a positive impact on audit outcomes (Habib et al., 2022; Mardessi, 2022), and negatively influences earnings management (Carrer et al., 2017; Mardessi, 2022; Nikulin et al., 2022). Generally, independent AC members have unbiased opinions in the committee’s decision and achieve an effective financial monitoring process (Rahman et al., 2023).

5.2.3. Audit committee gender diversity

The study measures AC gender diversity as the number of female members on the AC against the total number of AC members (Chatterjee & Rakshit, 2023). Studies show that the presence of females on AC improves audit functions and enhances quality financial reporting by reducing managerial opportunism (Chatterjee & Rakshit, 2023; Rahman et al., 2023).

5.2.4. Audit committee meetings

The study measures AC meetings as the number of meetings held by the AC in a year (Biswas et al., 2023; Stewart & Munro, 2007), as indicated in Table 1. The mere setting up of AC does not suggest quality managerial monitoring. However, effective ACs are diligent, meet regularly to discuss financial matters and approve important financial decisions. Evidence shows that AC regular meetings restrain earnings practice and enhance general AC efficiency (Xie et al., 2003; Hasan et al., 2020).

5.3. Moderating variable

The moderating variable is audit quality. The study measures audit quality with the Big 4 group (Price Waterhouse Cooper (PWC), Deloitte, Ernst and Young (E&Y) and KPMG) (Donelson et al., 2020; Rajgopal et al., 2021; Sitanggang et al., 2020). A dichotomous variable is used where one (1) is assigned when the firm is audited by the Big 4, and zero (0) otherwise. Big 4 has the resources, competence, skills, international outlook and technology to achieve quality audit outcomes (Habib et al., 2022; Mardessi, 2022).

5.4. Control variables

5.4.1. Cash flow

Cash flow is measured as operating cash flow this year to total assets of the previous year (Ball & Nikolaev, 2022; Q. L. Le & Nguyen, 2023). Empirical studies demonstrate that firms with strong operating cash flow are less likely to use accruals to manage earnings, and vice versa (Liu et al., 2023; Alhadab & Clacher, 2018).

5.4.2. Firm size

Following conventional literature, firm size is measured as the natural logarithm of total assets (Farooq et al., 2022; Ting, 2021), presented in Table 1. Managers of large firms are often politically sensitive and tend to reduce political costs by choosing accounting policies to increase profits during the current period (Veganzones et al., 2023; Madah, 2023). Moreover, large firms may have complex operations, so they have higher incentives and more opportunities to engage in earnings smoothing and overstate earnings (Osianto & Pudjolaksono, 2022). On the other hand, other studies have found that managers of smaller firms can keep information better than larger firms (Nikulin et al., 2022). Information about larger firms is generally more widely available and can be obtained at a lower cost than that of smaller firms (El Diri et al., 2020) because larger firms are more closely scrutinized by investors or regulators. Despite the scrutiny and accessible information from larger firms, they are more likely to report earnings management than smaller firms.

5.4.3. Profitability

Profitability is measured as return on equity (ROE) (earnings before interest and tax divided by total equity) (Bawuah, 2024; Kateb & Belgacem, 2023; Le & Nguyen, 2023). Evidence shows that firms with lower profits are more likely to engage in earnings management to achieve a desired income to issue more shares, and vice versa (Le & Nguyen, 2023).

5.4.4. Leverage

Financial leverage has been measured as the total debt to total assets ratio (Le & Nguyen, 2023). Firms with higher leverage are more likely to use accruals management to achieve desired earnings (Jordan et al., 2008).

5.5. Model specification and estimation strategy

The fixed-effect (FE) method is used in this study. This method is more efficient in dealing with heterogeneity in firms (Aljadba et al., 2022). The FE is more efficient than the ordinary least square (OLS) method. The fixed effect accounts for individual effects by adding separate intercepts for each entity in the regression. Also, it can estimate time-invariant variables not captured by the OLS model. Thus, the FE method provides a more reliable estimate than the OLS because it accounts for individual entity effects and provides reliable and unbiased estimates of the explanatory variables (Garson, 2013; McNeish & Kelley, 2019). However, the study performs an unreported Hausman test. The p-value of the F-test is 0.000, confirming the appropriateness of the FE over the pooled OLS and the random effect (REM) models.

The study further uses the Two-stage least squares (2SLS) regression analysis as a further robustness check and to solve the problem of potential endogeneity in the panel. Before the 2SLS, the assumptions of the model were tested including the endogeneity test, heteroscedasticity test, normality test and autocorrelation test. Conventional panel methods do not check endogeneity problems in explanatory variables. Therefore, outputs from such estimators are often inconsistent and biased (Gujarati & Porter, 2009; Wooldridge, 2015). Given this, the conventional practice is to employ the 2SLS approach—a case of an extraordinary situation of more instrumental variable practice. All the assumptions were met. The 2SLS was however used to verify the results of the FE.

FE assumes that the intercepts vary across units and time. In its simplified form, the FE model is written as: (1) $$Y_{\mathit{\text{it}}}=X_{\mathit{\text{it}}}^{\prime }\hspace{0.17em}\beta +Z_i\alpha +{\epsilon }_{\mathit{\text{it}}}$$(1)

First, the study investigates the effect of AC effectiveness on earnings management, as indicated in Equation 2(2) $$\begin{array}{l}E{M}_{i,t}={\beta }_o+{\beta }_1\mathit{\text{AC}}\_\mathit{\text{IND}}{\hspace{0.17em}}_{i,t}+{\beta }_2\mathit{\text{AC}}\_\mathit{\text{MEE}}{T}_{i,t\hspace{0.17em}}+{\beta }_3\mathit{\text{AC}}\_\mathit{\text{SIZ}}{E}_{i,t\hspace{0.17em}}+{\beta }_4\mathit{\text{AC}}\_\mathit{\text{GENDE}}{R}_{i,t\hspace{0.17em}}+\\ \hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}{\beta }_5\mathit{\text{RO}}{E}_{i,t\hspace{0.17em}}+{\beta }_{6}C{F}_{i,t}+{\beta }_7\mathit{\text{SIZ}}{E}_{i,t}+{\beta }_8\mathit{\text{LE}}{V}_{i,t}+{\mu }_{i\hspace{0.17em}}+e_{i,t}\end{array}$$(2) . (2) $$\begin{array}{l}E{M}_{i,t}={\beta }_o+{\beta }_1\mathit{\text{AC}}\_\mathit{\text{IND}}{\hspace{0.17em}}_{i,t}+{\beta }_2\mathit{\text{AC}}\_\mathit{\text{MEE}}{T}_{i,t\hspace{0.17em}}+{\beta }_3\mathit{\text{AC}}\_\mathit{\text{SIZ}}{E}_{i,t\hspace{0.17em}}+{\beta }_4\mathit{\text{AC}}\_\mathit{\text{GENDE}}{R}_{i,t\hspace{0.17em}}+\\ \hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}\hspace{0.17em}{\beta }_5\mathit{\text{RO}}{E}_{i,t\hspace{0.17em}}+{\beta }_{6}C{F}_{i,t}+{\beta }_7\mathit{\text{SIZ}}{E}_{i,t}+{\beta }_8\mathit{\text{LE}}{V}_{i,t}+{\mu }_{i\hspace{0.17em}}+e_{i,t}\end{array}$$(2)

Second, the study further investigates the moderating role of audit quality between AC effectiveness and earnings management. Audit quality is proxied with Big 4 auditors. Equation 3(3) $${\mathit{\text{EM}}}_{i,t}={\beta }_o+{\beta }_1{B\_\mathit{\text{Big}}4\mathrm{ }}_{i,t}{+{\beta }_2{\mathit{\text{AC}}\_\mathit{\text{IND}}}_{i,t\mathrm{ }}+{\beta }_3{\mathit{\text{AC}}\_\mathit{\text{MEET}}}_{i,t\mathrm{ }}+{\beta }_4{\mathit{\text{AC size}}}_{i,t}+{\beta }_5{\mathit{\text{GENDER}}}_{i,t}+{\beta }_6{(\mathit{\text{Big}}\mathrm{ }4\mathrm{*}\mathit{\text{AC}}\_\mathit{\text{IND}})}_{i,t}+{\beta }_7{(\mathit{\text{Big}}\mathrm{ }4\mathrm{*}\mathit{\text{AC}}\_\mathit{\text{MEET}})}_{i,t}+{{\beta }_8{(\mathit{\text{Big}}\mathrm{ }4\mathrm{*}\mathit{\text{AC}}\_\mathit{\text{SIZE}})}_{i,t}+{\beta }_9{\left(\mathit{\text{Big}}\mathrm{ }4\mathrm{*}\mathit{\text{GENDER}}\right)}_{i,t}+{\beta }_{10}{\mathit{\text{ROE}}}_{i,t\mathrm{ }}+{\beta }_{11}{\mathit{\text{CF}}}_{i,t}+{\beta }_{12}{\mathit{\text{SIZE}}}_{i,t}+{\beta }_{13}{\mathit{\text{LEV}}}_{i,t}+\mu }_{i\mathrm{ }}+e}_{i,t}$$(3) presents the model. (3) $${\mathit{\text{EM}}}_{i,t}={\beta }_o+{\beta }_1{B\_\mathit{\text{Big}}4\mathrm{ }}_{i,t}{+{\beta }_2{\mathit{\text{AC}}\_\mathit{\text{IND}}}_{i,t\mathrm{ }}+{\beta }_3{\mathit{\text{AC}}\_\mathit{\text{MEET}}}_{i,t\mathrm{ }}+{\beta }_4{\mathit{\text{AC size}}}_{i,t}+{\beta }_5{\mathit{\text{GENDER}}}_{i,t}+{\beta }_6{(\mathit{\text{Big}}\mathrm{ }4\mathrm{*}\mathit{\text{AC}}\_\mathit{\text{IND}})}_{i,t}+{\beta }_7{(\mathit{\text{Big}}\mathrm{ }4\mathrm{*}\mathit{\text{AC}}\_\mathit{\text{MEET}})}_{i,t}+{{\beta }_8{(\mathit{\text{Big}}\mathrm{ }4\mathrm{*}\mathit{\text{AC}}\_\mathit{\text{SIZE}})}_{i,t}+{\beta }_9{\left(\mathit{\text{Big}}\mathrm{ }4\mathrm{*}\mathit{\text{GENDER}}\right)}_{i,t}+{\beta }_{10}{\mathit{\text{ROE}}}_{i,t\mathrm{ }}+{\beta }_{11}{\mathit{\text{CF}}}_{i,t}+{\beta }_{12}{\mathit{\text{SIZE}}}_{i,t}+{\beta }_{13}{\mathit{\text{LEV}}}_{i,t}+\mu }_{i\mathrm{ }}+e}_{i,t}$$(3)

Where EM is earnings management, AC_IND is audit committee independence, AC_MEET is audit committee meetings, AC_SIZE is audit commitee size, AC_GENDER is audit committee gender diversity, SIZE is firm size, ROE is profit, LEV is leverage, Fees is auditors’ fees, and CF is cash flow. The moderating variables are generated by interacting audit quality (Big 4) with each of the ACE variables (AC independence, AC gender diversity, AC meetings and AC size)

6. Empirical findings

6.1. Stationarity and co-integration test

The study performs a stationarity test using the augmented Dickey–Fuller (ADF) test. The null hypothesis suggests that the variable is not stationary or has a unit root (p > 5%), while the alternative hypothesis suggests that the variable is stationary or has no unit root (p < 5%). As indicated in Table 2, the variables are stationary at either the level or first difference (of orders 0 and 1).

Table 2. Augmented Dickey-Fuller unit root test results.

Variables	Level	Difference	Conclusion
Earnings Management	−3.64	−8.32**	I (1)
Audit Quality	−4.17	−15.297***	I (1)
AC_Size	−3.06	5.25**	I (1)
AC_Gender	−1.45	−7.83***	I (1)
AC_Independence	−4.26**	−1.53	I (0)
AC_Meetings	−2.74	−4.05**	I (1)
Cash Flow	−4.64	−3.95**	I (1)
SIZE	−3.45**	2.56	I (0)
Profitability	−5.46	−6.66***	I (1)
Leverage	−1.87	−5.73***

Note(s): *, **, *** denote 10%, 5% and 1% significance levels respectively. Source: Author’s estimation.

6.2. Summary statistics

Table 3 presents summary descriptive statistics of the study variables over the period investigated (2010–2021). As indicated, the absolute value of discretionary accruals has a minimal mean value of 2.027. Overall, the results show no outliers since the average values are higher than the standard deviations. 77% of the sample firms use the service of the Big 4 implying high audit quality. The mean of the AC size is 3 members ranging from 3 to 5. AC meetings range between a minimum of 4 and a maximum of 8, with a mean value of 3.96 annual AC meetings. AC gender diversity has a mean of 0.234, indicating that 23.4% of all AC members are females, implying a male-dominated (76.6%) AC of firms in Ghana. The average value of profit is 11.9% indicating poor performance of the sampled firms in Ghana. The firms are highly geared with an average leverage value of 73.3% of their total assets.

Table 3. Descriptive statistics.

Variable	Obs	Mean	SD	Min	Max
EM	275	2.027	1.076	−0.920	4.097
Big4	275	0.770	0.421	0	1
AC_Size	275	3.641	1.132	3.00	5
AC_Gender	275	0.234	0.242	1	3
AC_Independence	275	0.822	0.295	2	4
AC_Meetings	275	3.923	2.028	4	8
Cash flows	275	7.853	1.105	4.130	9.983
Size	275	8.870	1.010	6.160	10.162
ROE	275	0.119	0.134	−1.435	0.412
Leverage	275	0.733	0.239	0.050	1.430

Note: EM is earnings management. Source: Author’s analysis with data from selected firm’s annual reports.

6.3. Correlation

Table 4 presents Pearson correlation results of the study variables to identify possible multicollinearity problems. Generally, the results indicate a weak correlation between the variables, suggesting no serious multicollinearity problem. For instance, Big 4 has a negative and significant (β = −0.407, p < 0.05) correlation with earnings management, consistent with the study’s prediction and other studies (Alhadab & Clacher, 2018; Imen & Anis, 2021). Also, AC size and AC gender have a negative correlation with EM consistent with hypothesis 1 and previous findings (Carrer et al., 2017; Mardessi, 2022; Nikulin et al., 2022). There is a weak negative correlation between firm size and AC gender. Leverage has a positive correlation with earnings management (β = −0.356, p < 0.05). A correlation coefficient of 0.8 or higher (Kennedy, 2008) suggests a problem of multicollinearity. However, as shown in Table 4, the correlation coefficients have correlations lower than 0.8. Therefore, we conclude that the variables suffer from no multicollinearity problems.

Table 4. Correlation.

Var	EM	Big 4	AC_Size	AC_Gender	AC_IND	ACM	CF	SIZE	ROE	LEV
EM	1
Big4	−0.407*	1
AC_Size	0.305*	0.357*	1
AC_Gender	−0.432*	−0.254	0.07	1
AC_IND	0.341*	0.193	0.36	0.23	1
ACM	−0.057	−0.108	0.070	−0.00	−0.11	1
CF	−0.365*	−0.201	−0.41	0.081	−0.04	0.664	1
SIZE	0.418	0.208	−0.034	−0.241*	0.278	−0.157	−0.167	1
ROE	−0.11	−0.15	−0.04	0.209	−0.006	−0.106	0.306*	−0.06	1
LEV	0.356*	−0.35	−0.00	0.022	−0.108	0.342*	0.375*	0.265*	0.321*	1

Note: Significance level * is p < 0.5. Source: Author’s computation with data from selected firms’ annual reports. ACM is AC meetings. All other variables are defined in Table 1.

6.4. Audit committee effectiveness and earnings management

Table 5 (Models 1–4) presents the results of the effect of audit committee effectiveness (AC size, AC meetings, AC independence and AC gender diversity) on EM using FE and 2SLS estimators to provide robust results. Models 1 and 3 estimate only the independent variables on the dependent variable with no control variables. Models 2 and 4 estimate both the independent and the control variables on the dependent variable. The adjusted r-squared indicates that about 35.4%, 51%, 34% and 45% of the variations in the dependent variables are explained by the independent variables in Models 1–4, respectively.

Table 5. Effect of audit committee effectiveness on earnings management.

	FE	FE	2SLS	2SLS
Var	(1)	(2)	(3)	(4)
AC_IND	−0.514*** (0.021)	−0.360*** (0.102)	−0.024*** (0.004)	−0.214*** (0.024)
AC_MEET	−0.478*** (0.040)	−0.106*** (0.056)	0.047 (0.358)	−0.314*** (0.025)
AC_GENDER	0.051 (0.874	1.829* (0.549)	0.024 (0.547)	0.054 (0.045)
AC_SIZE	−0.208*** (0.024)	0.061 (0.051)	−0.351* (0.147)	−0.215** (0.064)
ROE		0.130 (0.888)		0.024 (0.087)
CF		−0.487** (0.147)		−0.458*** (0.061)
LEV		−0.371 (0.510)		−0.054*** (0.008)
SIZE		−1.076*** (0.327)		0.220*** (0.014)
Constant	0.014*** (0.002)	7.203*** (1.149)	0.343 (0.207)	0.354** (0.102)
Year FE	Yes	Yes	No	No
Firm FE	Yes	Yes	No	No
Adj. $R^2$	0.354	0.510	0.340	0.450
F. Stats	12.097	14.097	8.087	11.087
Prob.	0.000	0.000	0.000	0.000

Note: * represents (0.1 sig.); ** represents (0.05 sig.); and *** represents (0.01 sig.). AC_IND is audit committee independence; AC_MEET is audit committee meetings; AC_GENDER is audit committee gender; AC_SIZE is audit committee size; ROE is profit; CF is cash flow; LEV is leverage; and SIZE is firm size. Source: Author’s analysis with data from firms’ annual reports.

Models 1–4 suggest that AC independence has a significant negative (β = −0.514, p < 0.01; β = −0.360, p < 0.01; β = −0.024, p < 0.01; β = −0.214, p < 0.01) association with earnings management, respectively. This is consistent with hypothesis 1 that AC’s effectiveness of independence provide better monitoring of the management financial reporting process and constrain earnings management (Carrer et al., 2017; Mardessi, 2022; Nikulin et al., 2022). In addition, the negative association between AC size and earnings management (β = −0.208, p < 0.01; β = −0.351, p < 0.10; β = −0.215, p < 0.05) in Models 1, 3 and 4 respectively, is consistent with the hypothesis 1 that sufficient AC size brings diverse ideas, knowledge and experience to the committee (Juwita et al., 2020; Lin et al., 2006; Xie et al., 2003) which impact negatively on earnings management (Chatterjee & Rakshit, 2023). Also, AC meetings show a negative and significant (β = −0.478, p < 0.01; β = −0.106, p < 0.01; β = −0.314, p < 0.01) relationship with earnings management in Models 1, 2 and 4, respectively, supporting H1 and other empirical studies (Xie et al., 2003; Alhadab & Clacher, 2018). Frequent AC meetings provide timely and regular information sharing between the management and the committee which enhances monitoring by reducing information symmetry. However, AC gender has no significant effect on earnings management, contrary to H1 and other studies that demonstrate that females have unique abilities that allow them to be ethical, strict, responsible and transparent in audit duties and mitigate earnings practices (Chatterjee & Rakshit, 2023; Zalata et al., 2022). However, it tends to be consistent with Nguyen (2023) who find a U-shaped relationship between women on AC and financial reporting quality. The plausible explanation for the insignificant results is that female AC members in the sample firms are insufficient in number (23.4%) to impact audit functions significantly. Also, despite the benefits of appointing female AC members, the current study contends that a mere appointing of a female will not lead to effective financial reporting monitoring and integrity, but rather female AC members with relevant financial skills will constrain earnings management. Generally, the results support hypothesis 1 that AC effectiveness constrains earnings management, and also in support of the agency theory that predicts a negative association between good corporate governance mechanisms such as effective AC and managerial opportunistic behaviours such as earnings management (Carrer et al., 2017). In summary, firms must strengthen their AC to achieve lower earnings management.

6.5. The moderating role of audit quality in the relationship between AC effectiveness and earnings management

In the previous results (Table 5), the study has established a negative and significant impact of ACE (AC size, AC independence and AC meetings) on earnings management. In Table 6, we investigate whether audit quality (Big4) influences the association between ACE (AC_IND, AC_SIZE, AC_GENDER and AC_MEET) and earnings management. In particular, Models 5 and 7 regress the independent and moderating variables on the dependent variable with no control variables. Models 6 and 8 add the control variables. Noticeably, the coefficients of AC_IND and AC_MEET on earnings management in Table 5 have improved significantly in Table 6. For instance, the interaction coefficient between AC independence and audit quality (AC_IND*Big4) in Models 6-8 are negative and statistically significant. Observably, the negative coefficient of AC independence has not changed. Rather, the absolute coefficient has increased from -0.360 in Table 5 Model 2, to -2.264 (-2.210 - 0.054) in Model 6 of Table 6, and -0.214 in Table 5 Model 4, to -0.882 (-0.450 - 0.432) in Model 8 of Table 6 for FE and 2SLS respectively. Similarly, under the FE results in Model 6, the interaction between AC meetings and audit quality (AC_MEET*Big4) increases the magnitude of the negative impact reported in Table 5. The coefficient of AC meeting is now -1.044 (-0.572 - 0.472) in Model 6, compared with the -0.106 reported in Model 2 of Table 5. Also, under the 2SLS method in Model 8, the coefficient of AC meeting is now -0.558 (-.0354 - 0.204) compared with the -0.314 reported in Model 4 of Table 5. The results indicate that the presence of audit quality (Big4) increases the negative effect of ACE (AC independence and AC meetings) on earnings management. This supports Hypothesis 2. AC independence enhances the implementation of external audit recommendations which constrain earnings management (Alzeban & Sawan, 2015). Also, during the external audit process, frequent meetings with the internal audit units improve financial reporting systems. The committee can remain informed about the audit process and provide assistance to achieve quality audit outcomes. External auditor appointment is crucial in the audit function. When the ACs are involved in the appointment and removal of the external auditor, it improves internal audit empowerment (Alzeban & Sawan, 2015). This enhances collaboration with the AC and external auditor to achieve collective results of efficiency. Suggesting a policy insists that ACE when combined with audit quality, produces a greater effect in constraining earnings management (Zgarni et al. (2016).

Table 6. The role of audit quality in the AC effectiveness and earnings management nexus.

	FE	FE	2SLS	2SLS
Variables	(5)	(6)	(7)	(8)
Big4	−3.024*** (0.024)	−4.029* (1.104)	−0.204 (0.145)	−0.254** (0.058)
AC_IND	-2.147*** (0.254)	-0.054*** (0.007)	−1.354*** (0.103)	-0.432*** (0.042)
AC_MEET	−2.145*** (0.621)	−4.727** (1.340)	0.208 (0.108)	-0.204*** (0.051)
AC_GENDER	0.254 (0.124)	−0.342 (0.298)	−0.210 (0.236)	−0.021** (0.004)
AC_SIZE	0.454*** (0.100)	0.054 (0.060)	0.402*** (0.100)	0.654 (0.432)
Big4*AC_IND	-0.417*** (0.103)	-0.210*** (0.098)	-0.460*** (0.124)	-0.450*** (0.065)
Big4*AC_MEET	0.365 (0.297)	-0.575*** (0.081)	-2.054*** (0.360)	−0.354*** (0.105)
Big4*AC_GENDER	−0.453 (0.609)	0.043 (0.254)	0.034 (0.065)	−0.321 (0.217)
Big4*AC_SIZE	0.056 (0.017)	0.032 (0.109)	0.025 (0.140)	−0.435*** (0.087)
ROE		−9.109*** (3.600)		−0.015 (0.087)
CF		−0.342*** (0.076)		0.147** (0.047)
LEV		−0.231*** (0.031)		0.342 (0.564)
SIZE		−1.269** (0.601)		−0.065*** (0.007)
Constant	−14.564*** (4.907)	−19.975*** (7.920)	0.346*** (0.065)	6.342 (7.875)
Year FE	Yes	Yes	No	No
Firm FE	Yes	Yes	No	No
F. Stats	13.097	16.907	12.907	8.098
Prob.	0.000	0.000	0.000	0.000

Note: * represents (0.1 sig.); ** represents (0.05 sig.); and *** represents (0.01 sig.). Big4 is audit quality; AC_IND is audit committee independence; AC_MEET is audit committee meetings; AC_GENDER is audit committee gender; AC_SIZE is audit committee size; ROE is profit; CF is cash flow; LEV is leverage; and SIZE is firm size. Source: Author’s analysis with data from firms’ annual reports.

Turning to the control variables, profit has a negative and significant (β = −9.109, p < 0.01) effect on EM in Model 6. Profitable firms are less likely to use discretionary accruals to manage earnings, and vice versa. Contrary to expectation, leverage has a negative association with earnings management. The plausible explanation is that firms in Ghana (both successful and struggling) predominantly use debt financing, and therefore the decision to engage in earnings practices is less determined by their gearing ratio, but rather by other matters. As expected, cash flows reduce earnings management. Firms with high cash flow have little incentive to use discretionary accruals to manage earnings (Liu et al., 2023; Alhadab & Clacher, 2018). Also, firm size has a negative association with earnings management consistent with Ruwanti et al. (2019). Generally, from the total assets, most Ghanaian firms are small, relatively, and may have less complex operations and opportunities to vigorously use earnings accruals.

6.6. Sensitivity analysis

Table 7 presents a sensitivity analysis to examine the robustness of the results. Consistent with several empirical studies, audit fees have been used to proxy audit quality (see, Afenya, Arthur, Kwarteng, & Opoku, 2022). Audit fees are the professional service fees for auditors. Audit fee is among the input-based proxies of audit quality including client importance, auditor tenure, client location, city specialist auditor and industry specialist auditor (Rajgopal et al., 2021). The fees are determined by the auditor’s level of experience and expertise (Afenya, Arthur, Kwarteng, & Kyeremeh, 2022). In addition, audit fee is influenced by the length of the process, the complexities and expectations (Afenya, Arthur, Kwarteng, & Kyeremeh, 2022). To avoid firm legal liability, and promote shareholder interests, shareholders, independent and diligent AC members may demand higher audit quality to signal higher assurance (which requires more audit work) (Carcello et al., 2002). Industry specialists with confirmed audit expertise, resources and technology to achieve quality audits charge higher fees. Therefore, a higher audit fee is expected to lead to audit quality. Generally, the results in Table 7 Models 9 and 10 suggest that the results in Table 6 are robust to different measures of audit quality (fees)

Table 7. Robustness analysis using audit fees as a moderator of audit quality on the ECE -EM nexus.

	FE	FE
Variables	(9)	(10)
Fees	−1.024*** (0.214	−2.014*** (0.621)
AC_IND	−2.025*** (0.157)	1.054 (1.984)
AC_MEET	−0.147*** (0.019)	0.564 (0.408)
AC_GENDER	1.341 (1.047)	−0.157*** (0.024)
AC_SIZE	−0.045*** (0.009)	1.324 (1.654)
Fees*AC_IND	−2.235 (0.980)	−2.024*** (0.602)
Fees*AC_MEET	−2.024*** (0.521)	−1.054*** (0.091)
Fees*AC_GENDER	−0.354*** (0.102)	0.243 (0.109)
Fees*AC_SIZE	2.075 (2.000)	−0.432** (0.107)
ROE	−5.087* (2.020)	−0.054 (0.077)
CF	−0.342*** (0.087)	−0.308** (0.043)
LEV	−0.453 (0.509)	−0.426*** (0.031)
SIZE	−4.986** (0.965)	2.087 (1.432)
Constant	−14.086*** (4.097)	−12.674*** (2.075)
Year FE	Yes	No
Firm FE	Yes	No
Adj. $R^2$	0.508)	0.408
F. Stats	10.086	9.097
Prob.	0.000	0.000

Note: * represents (0.1 sig.); ** represents (0.05 sig.); and *** represents (0.01 sig.). Fees (audit fees) is audit quality; AC_IND is audit committee independence; AC_MEET is audit committee meetings; AC_GENDER is audit committee gender; AC_SIZE is audit committee size; ROE is profit; CF is cash flow; LEV is leverage; and SIZE is firm size. Source: Author’s analysis with data from firms’ annual reports.

7. Conclusion

The role of the audit committee in constraining managerial opportunism has been a subject of corporate governance for decades. Extant literature calls for research to identify factors that promote AC effectiveness such as AC independence, AC diligence, AC size, AC gender, and AC financial expertise (Agyei-Mensah et al. 2019; Agyekum et al., 2014; Alzeban & Sawan, 2015; Azizkhani et al., 2023; Hoitash & Hoitash, 2009). AC can be effective if comprises sufficient, independent, highly financially knowledgeable members who have an impartial assessment of financial information from management. Evidence exists on the effect of corporate governance dimensions such as audit quality and AC effectiveness on firm performance and financial reporting quality both in Ghana and elsewhere (see, Agyemang, 2020; Al-Ahdal & Hashim, 2022; Awinbugri & Prince, 2019; Glover-Akpey & Azembila, 2016; Ibrahim et al., 2017). Also, some studies investigate the link between ACE and audit quality and EM (e.g. see, Azizkhani et al., 2023; Hoitash & Hoitash, 2009; Alzeban & Sawan, 2015; Agyei-Mensah & Yeboah, 2019; Agyekum et al., 2014). Despite theoretical and empirical suggestions and predictions that audit quality and ACE constrain earnings practices by management, evidence on this relationship is less explored in Ghanaian literature. In addition, how and why audit quality might moderate the ACE-EM nexus is rare, and the existing studies only investigate the direct relationship (e.g. Afenya, Arthur, Kwarteng, & Opoku, 2022; Afenya, Arthur, Kwarteng, & Kyeremeh, 2022; Agyei-Mensah & Yeboah, 2019). This study investigates the direct effect of ACE on earnings management, and further examines whether audit quality moderates this relationship using 25 firms in Ghana from 2010 to 2021. Applying both the fixed effect and the 2SLS methods, the findings show that AC effectiveness such as AC independence, AC size and AC meetings constrain earnings management with the association being stronger for audit independence followed by AC meeting and then AC size. With sufficient AC size, strong AC independence and regular meetings, the AC is expected to provide effective monitoring of financial and internal control processes to avoid managerial opportunistic behaviours such as earnings management. Further, the evidence shows a negative and significant moderating effect of audit quality on the ACE-EM nexus. This suggests that audit quality strongly conditions ACE to induce a negative effect on EM. The evidence implies that firms with effective AC (AC independence and meetings) and Big 4 report lower earnings management or are less likely to use discretionary accruals to manage earnings. The explanation is that effective AC can strengthen external auditors and support them in undertaking activities that improve audit functions. Also, effective AC supports the audit function by appointing industry specialist auditors. By doing so, the AC is less likely to have accounting problems or disputes with external auditors. Effective ACs are more likely to side with an external auditor in a substance-over-form dispute with management, which contributes to audit quality, and therefore mitigates managerial incentives to use discretionary accruals.

Apart from applying the agency theory to interpret ACE, Big4 and EM, the results make new contributions to current literature in several ways. First, employing a dataset in Ghana, the study offers new evidence indicating that firms with effective AC tend to report lower earnings management through increased disclosure. Noticeably, some of the ACE effectiveness has a stronger effect on EM than others. Thus, this study contributes to one of the important and growing literature, particularly in emerging countries on how and why the CG mechanism can mitigate managerial opportunism and inefficiency-induced corporate failures (Agyei-Mensah & Yeboah, 2019; Carrer et al., 2017; Chatterjee & Rakshit, 2023; Fernández-Méndez & Pathan, 2023; Hasan et al., 2020; He et al., 2009; Mardessi, 2022; Mnif & Tahri, 2023; Nikulin et al., 2022; Rahman et al., 2023; Sharma & Iselin, 2012). This informs effective recruitment of AC members and how they must perform their functions with all deligence. Second, previous evidence on the direct relationship between ACE and earnings management presents mixed findings. This can be due to the lower consideration of how EC functions align with external auditor effectiveness in achieving quality financial control systems. Therefore, the current study makes a distinctive new contribution to the literature by showing that the presence of audit quality (Big4) increases AC oversight over firm financial reporting process, audit process, risk management systems, internal control systems and management compliance with the laws and regulation, and therefore discourages managerial incentives to use earnings management. The theoretical and methodological implication is that the effect of ACE on EM is stronger when combined with the Big4 auditor service. Empirically, the findings provide evidence that ACE effectively mitigates earnings management when Big4 is used.

Third, the results have managerial, practitioner and regulatory implications, particularly for firm management, policymakers and regulators who are interested in achieving firm resilience. In terms of managers, the findings provide them with a strong basis to support AC and allow them to operate independently. While some management with private interests might not do this, shareholders must insist on AC’s independence and effectiveness. Also, AC must operate efficiently and appoint Big 4 auditors and offer the necessary collaboration and support to constrain earnings management. To regulators, governments and policymakers, this evidence shows that insisting on audit compliance such as the use of the Big 4 and providing support for the smaller firms to use the Big 4 or equivalence will promote a sustainable business environment for economic growth.

Finally, whereas the findings are robust, some caveats are applied. First, data was manually collected which is considered very laborious (Ntim, 2016). Second, due to the first limitation, only a few firms were selected. Even though the limited sample still allowed robust estimation and evidence, however, future studies might benefit from larger datasets. Third, the study uses the modified Jones model (1995) to proxy earnings management. Some studies claim that the Jones model (1995) may have its weaknesses to comprehensively capture all elements of earnings management. Therefore, future studies will benefit more from using several proxies of earnings management.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data and materials supporting the results or analyses presented in this paper is available upon reasonable request.

Additional information

Notes on contributors

Isaac Bawuah

Isaac Bawuah holds a post graduate certificate in Finance and an undergraduate certificate in banking and Finance from Kwame Nkrumah University of Science and Technology (KNUST), Ghana. His research interest includes corporate governance, climate finance, institutional quality, sustainable development and assessment in Sub-Saharan Africa.