The relationship between internal audit function quality and earnings quality: the moderating effect of CEO power

Kalembe, Dorcus; Nkundabanyanga, Stephen Korutaro; Kaawaase, Twaha Kigongo; Kayongo, Isaac Newton Article The relationship between internal audit function quality and earnings quality: the moderating effect of CEO power Cogent Business & Management Provided in Cooperation with: Taylor & Francis Group Suggested Citation: Kalembe, Dorcus; Nkundabanyanga, Stephen Korutaro; Kaawaase, Twaha Kigongo; Kayongo, Isaac Newton (2024) : The relationship between internal audit function quality and earnings quality: the moderating effect of CEO power, Cogent Business & Management, ISSN 2331-1975, Taylor & Francis, Abingdon, Vol. 11, Iss. 1, pp. 1-28, https://doi.org/10.1080/23311975.2024.2311159 This Version is available at: https://hdl.handle.net/10419/326032 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Cogent Business & Management ISSN: 2331-1975 (Online) Journal homepage: www.tandfonline.com/journals/oabm20 The relationship between internal audit function quality and earnings quality: the moderating effect of CEO power Dorcus Kalembe, Stephen Korutaro Nkundabanyanga, Twaha Kigongo Kaawaase & Isaac Newton Kayongo To cite this article: Dorcus Kalembe, Stephen Korutaro Nkundabanyanga, Twaha Kigongo Kaawaase & Isaac Newton Kayongo (2024) The relationship between internal audit function quality and earnings quality: the moderating effect of CEO power, Cogent Business & Management, 11:1, 2311159, DOI: 10.1080/23311975.2024.2311159 To link to this article: https://doi.org/10.1080/23311975.2024.2311159 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group Published online: 23 Feb 2024. Submit your article to this journal Article views: 2731 View related articles View Crossmark data Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oabm20 ACCOUNTING, CORPORATE GOVERNANCE & BUSINESS ETHICS | RESEARCH ARTICLE Cogent Business & ManageMent 2024, VoL. 11, no. 1, 2311159 The relationship between internal audit function quality and earnings quality: the moderating effect of CEO power Dorcus Kalembea , Stephen Korutaro Nkundabanyangab , Twaha Kigongo Kaawaasec and Isaac Newton Kayongod aDepartment of accounting and Finance, Makerere university Business school, Kampala, uganda; bDepartment of accounting, Makerere university Business school, Kampala, uganda; cDepartment of auditing and taxation, Makerere university Business school, Kampala, uganda; dDepartment of Finance, Makerere university Business school, Kampala, uganda ABSTRACT This study investigates the relationship between internal audit function quality, Chief Executive Officer (CEO) power and earnings quality and how CEO power moderates the relationship between internal audit function quality and earnings quality. The study is correlational, cross-sectional, and perception-based that obtained 136 usable questionnaires from regulated firms in Uganda. Data was analyzed using Statistical Package for Social Scientists, Smartpls software, and Jose’s Modgraph. The results show that internal audit function quality is positive but not significantly related to earnings quality. The results revealed that internal audit function quality is negatively related to CEO power although the association is not significant. CEO power is significant and negatively related to earnings quality. The interaction effects of internal audit function quality and CEO power on earnings quality are significant, according to which the greater the power of the CEO, the lower the effect of the IAF on earnings quality. The current study adds to literature of internal audit function quality, CEO power and earnings quality and consequently, moderating role of CEO power in the relationship between internal audit and earnings quality. This study is relevant to policy makers and regulators who need to consider these results to identify the needed changes in their regulation of accounting practice and governance in Uganda. 1. Introduction This paper aims to investigate the relationship between internal audit function quality (IAFQ), CEO power, and earnings quality (EQ), as well as the moderating role of CEO power in the relationship between internal audit function quality and earnings quality. In trying to do this, the current study specifically aims to address four research concerns. First, what is the relationship between internal audit function quality and earnings quality? Second, what is the relationship between internal audit function quality and CEO power? Third, what is the relationship between CEO power and earnings quality? Fourth, how does CEO power moderate the relationship between internal audit function quality and earnings quality? Earnings quality are important to stakeholders because they serve as a foundation for decision-making, allow firms to secure funding, permit nations to draw in inexpensive capital, and guarantee effective resource allocation (Aggarwal et al., 2005; Dechow et al., 2010; Demerjian et al., 2013; Francis et al., 2008). Despite the importance of EQ, low-quality profits have continued to be a growing source of worry globally. For instance, Hertz Global Holdings, overstated transactions worth US$ 7.4 billion (PWC Report, 2019), Steinhoff firm overstated its pre-tax income worth US$ 235 million (SEC, 2019), and FTE Networks inflated its revenues by as much as 108% (SEC, 2021) among others. The phenomenon of doubtful © 2024 the author(s). Published by informa uK Limited, trading as taylor & Francis group CONTACT Dorcus Kalembe [email protected] Department of accounting and Finance, Makerere university Business school, Kampala, uganda this article has been corrected with minor changes. these changes do not impact the academic content of the article. https://doi.org/10.1080/23311975.2024.2311159 this is an open access article distributed under the terms of the Creative Commons attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. the terms on which this article has been published allow the posting of the accepted Manuscript in a repository by the author(s) or with their consent. ARTICLE HISTORY Received 20 November 2023 Revised 21 January 2024 Accepted 23 January 2024 KEYWORDS Internal audit function quality; CEO power; Earnings quality; Moderation; Uganda REVIEWING EDITOR Collins Ntim, University of Southampton, United Kingdom of Great Britain and Northern Ireland SUBJECTS Financial Accounting; Auditing; Corporate Governance; Business; Management and Accounting 2 D. KALEMBE ETAL. earnings is also apparent in Uganda. Successive reports have adduced evidence in this direction (PriceWaterhouse Coopers Limited Nyonyi Report, 2016; HFB Audited Annual Reports, 2018; The Ogoola Judicial Commission Report-Government of Uganda, 2000; Mugisha, 2019). When the quality of earnings is doubted, it shines a spot light on the relevance of information contained in financial statements as a basis for decision making, and reduces their financial probity (Jouber & Fakhfakh, 2011). Majority of the studies on earnings quality employed agency theory (Jensen & Meckling, 1976) perspective (see, Ismael & Kamel, 2021; Alzoubi, 2019; Melgarejo, 2019; Egbunike & Odum, 2018; Hashmi et al., 2018; Faricha, 2017; Hessayri & Saihi, 2015; Demerjian et al., 2013; Lyengar et al., 2010). However, these studies produce consistent results only in the context of developed capital markets where the major stakeholder is the shareholder. For instance, Ismael and Kamel (2021) reported that internal audit quality is negatively associated with abnormal accruals. Similarly, Alzoubi (2019) also concluded that internal audit function reduces discretionary accruals. But, in environments such as Uganda, where capital markets are not developed, agency perspective is less likely to be effective. From the stakeholder identification and salience (Mitchell et al., 1997) theory, there are several stakeholders who are interested in earnings quality. In this study, we consider a firm’s CEO to be a key internal stakeholder in less developed markets, where markets for control are weak. In the presence of such a powerful CEO, it is most likely that earnings quality will be low. Kalembe etal. (2023) present empirical evidence that CEO power is negatively associated with earnings quality. Therefore, managerial incentives for poor earnings quality potentially mirror stakeholder attributes such as power to influence the firms’ outcomes (Cai & Mehari, 2015; Mitchell et al., 1997). Existing empirical studies (Rezaee & Safarzadeh, 2022; Nalukenge et al., 2022; Kawaase et al., 2021; Elzahaby, 2021; Melgarejo, 2019; Abbott et al., 2016; Ege, 2015; Prawitt et al., 2009) present positive effects of internal auditing on earnings quality (management). Rezaee and Safarzadeh (2023) reported that internal audit improves EQ. Similarly, Elzahaby (2021) stresses that internal audit function improves firm performance of Egyptian Listed firms. Melgarejo (2019) argues that internal audits provide accounting reports that are more value-relevant, persistent, and conservative. Also, Prawitt et al. (2009) concluded that internal audit quality is negative and significantly related to absolute abnormal accruals. Other studies (Lee et al., 2022; Baker et al., 2019; Jiang etal., 2010) documented a relationship between CEO power and earnings quality. For instance; Baker et al. (2019) revealed that earnings management is greater when a CEO is powerful. Relatedly, Lee etal. (2022) reported that powerful CEOs amplified earnings management in Vietnamese Listed companies. Majority of these studies were carried out in developed countries and, moreover, only tested for direct effects. Studies such as Nalukenge etal. (2022) and Kaawaase et al. (2021) were carried out in Uganda, but their focus was on specific industries such as statutory corporations and financial services firms respectively. But, how CEO power affects a firm’s financial reporting strategies’ outcomes such as earnings quality is rarely discussed in the corporate governance, management or accounting literature. Powerful CEOs tend to create an opaque accounting environment that limits disclosure of accurate earnings. As a result, low EQ is reported. Yet with the presence of a quality internal audit function, it is expected that such tendencies will be mitigated, because internal audit monitors the opportunistic behavior of the CEO. We, therefore, expect powerful CEOs to moderate the link between IAFQ and EQ, if earnings are to be enhanced. Further, despite the numerous studies on corporate governance and its impact on EQ being documented, little is known about the interplay of CEO power and internal audit function quality on earnings quality. This study therefore intends to fill this void in existing literature. Other studies that try to address EQ adopt event study methodology (e.g. Faricha, 2017; Dutillieux etal., 2016) and have been informative but limited in application because the wealth effects of regulatory changes can be difficult to detect (Mackinlay, 1997). In the absence of the event or robust stock market model, other factors explaining earnings quality remain unearthed. Already, Athari and Bahreini (2021, Athari, 2022a) indicate that profitability is shaped by, among others, context, including country-level governance quality (Athari, 2022b), and generally, context offers insights into decisions relating to earnings (Athari et al., 2016). Still, many studies measure EQ using different proxies such as Jones model of discretionary accruals, modified Jones model, and performance-matched discretionary accruals (Hashmi et al., 2018), earnings restatement, earnings persistence, errors in bad debt provision and mapping of accruals into cashflows (Demerjian et al., 2013), abnormal accruals and timely loss recognition (Dutillieux et al., 2016). The problem with the accrual component is its failure to accurately indicate the real driver of COGENT BUSINESS & MANAGEMENT 3 opportunism and failure to address directly perceived EQ that mirrors the behavior of users (Dechow & Dichev, 2002; Healy & Wahlen, 1999). This is contrary to the Framework for Financial Reporting 2018 that addresses quality in terms of qualitative characteristics of financial statements (information). Unlike previous studies (e.g. Rezaee & Safarzadeh, 2022) on earnings quality, which recommend a comprehensive earnings quality model (which includes all different aspects of this construct), we utilize the theoretical framework of financial accounting 2018, specifically the fundamental qualitative characteristics of relevance and faithful representation to measure earnings quality. As such, we aim to capture the perceptions of stakeholders on earnings quality using the above-mentioned measures because perception-based studies provide managerial motivations for improving earnings quality (Belal & Owen, 2007). Consequently, the current research seeks to make the following contributions to existing literature. First, this study adds to existing literature by documenting the relationship between internal audit function quality, CEO power and earnings quality in Uganda. Second, the study articulates the moderation effect of CEO power in the relationship between internal audit function quality and earnings quality. This is because available studies (e.g Rezaee & Safarzadeh, 2022; Lee et al., 2022; Elzahaby, 2021) investigated the direct relation between internal audit and earnings quality, ignoring the role of a third variable-CEO power as a moderator in the Ugandan context. Third, this study uses a multi-theoretical approach to understand earnings quality in a developing country like Uganda. It is evident that previous studies (e.g Ismael & Kamel, 2021; Alzoubi, 2019; Melgarejo, 2019) used a single theory; which may not adequately explain EQ. The findings provide clear policy implications to the regulators and standard setters who may need to ensure that firms comply with international financial reporting standards in order to report high earnings quality. The next section of the paper includes a study setting, theoretical literature review, empirical literature review and hypotheses development, research design, empirical results and discussion, and summary and conclusion. 2. Background Uganda is a developing landlocked country lying on the equator in East Africa. Uganda gained independence in 1962 from the British colonialists but thereafter suffered civil wars and political unrest (Nkundabanyanga etal., 2013). The country is endowed with substantial resources, including fertile soils, regular rainfall, copper, gold, cobalt and the recently discovered oil reserves. The main exports are coffee, fish, flowers, horticultural products, minerals, tobacco and tea, making it an investment hub (Uganda National Investment Policy, 2018). With the Country’s industrial and trade policy in place, there has been a robust flow of investors into the country who have invested in different sectors given its favorable investment policies. This contributes to the economic growth and development of the country. To enhance the sustained growth agenda by the government of Uganda, robust financial reporting by companies is required (Uganda National Investment Policy, 2018). This has led to enactment of sound financial reporting infrastructure consisting of legal enforcement mechanisms. We begin to see the enactment of laws by the government of Uganda during colonial times. Others were enacted immediately after independence, such as the Companies Act of 1961. More regulations came on board around 2000, such as the Financial Institutions Act of 2004. This was later amended in 2016 to include agency banking, bancassurance, special access to the Credit Reference Bureau, and reform Deposit Protection Fund, among others. Since then, other regulations have come on board such as the Companies Act (2012), the Accountants Act (2013), The Micro Finance and Deposit-taking Institutions Act (2003), The Uganda Securities Listings Rules (2003), Insurance Act (2017), National Coffee Act (2021), Uganda Communications Act (2013), Uganda Retirement Benefits Scheme Act (2011), Electricity Act (2022) among others. These regulations provide policies that regulated firms should follow in order to report high earnings quality. For instance, the Companies Act (2012), section 154(1), requires regulated firms to prepare books of accounts that give a true and fair view of the state of affairs of the firm. Similarly, the Accountants Act (2013), section 54(2) necessitates heads of accounts and finance and internal auditors working with public interest firms to be members of the Institute of Certified Public Accountants (ICPAU) of Uganda. Moreover, for one to be a member of ICPAU, he/she must have passed the qualifying examinations conducted by the examinations board or be a full member of any accountancy body recognized 4 D. KALEMBE ETAL. by ICPAU (Accountants Act, Section 5(2a). Further, the international financial reporting standards, as ratified by ICPAU, also guide on the preparation and presentation of accounting information in financial reports. But also, the presence of regulatory authorities who supervise these regulated firms in Uganda offer a fertile ground for transparency, accountability as well as sound and sustainable financial reporting practices. Although regulations are in place, Uganda’s earnings quality is still far from desirable, despite its adoption of international financial reporting standards (IFRS). For instance, the Bank of Uganda Supervision Report (2022) indicates that commercial banks still have financial reporting gaps, corporate governance lapses, threatened independence of the second line of defence by the CEOs, and non-compliance with the requirements of the Financial Institutions Act 2004. Similarly, the Ogoola Judicial Commission report (2000) presented three commercial banks that failed as a result of false revenue statements. In order to escape paying income tax, Mugisha (2019) observed that businesses in Uganda ‘force’ deficits into their books of accounts. Furthermore, errors have been found in financial reports of some Ugandan commercial institutions (see, Housing Finance Bank’s annual financial reports from 2017/2016 and 2018/2017). In addition, there has been evidence of top management usurping board powers in commercial institutions (see, Bank of Uganda Annual Supervision Report, 2020). The Companies Act (2012), Section 14, Table F, mandates firms that are either public or private to adopt the code of corporate governance at the time of registration of its articles because the presence of good corporate practices mitigate fraud and errors resulting in high earnings quality. It is thus reasonable to expect that many parties begin to doubt the accurate depiction of Uganda’s earnings data in the wake of such incidents despite the presence of regulations. Thus, Uganda presents a unique environment for this examination. 3.Theoretical literature review In this study, we use the agency and stakeholder identification and salience theories. The agency theory (Jensen & Meckling, 1976) addresses the paradigm of divergent interests between the agent (managers) and principal (owners) in a contractual relationship. Because of the separation of ownership and control, potential conflicts such as information asymmetry arise between agent(management) and shareholders, especially if management aims to maximise utility. The agent is familiar with the firm’s business operations and financial reporting processes and as a result, he fails to disclose accurate earnings information. This leaves the shareholders unable to evaluate the firm’s true performance. It is worse when, in real practice, the manager shirks in the Job (Choi, 2015), as this leads to reporting low earnings. Moreover, all this is done to mislead stakeholders about the underlying economic performance of the firm (Healy & Wahlen, 1999). To mitigate the self-interest behaviours of the agent, shareholders institute a quality internal audit function to enhance earnings. The ability of IAFQ to mitigate the opportunistic behaviour of the agent largely depends on IA organization status, IA investment, IA competence and IA role performance. For instance, Ismael and Kamel (2021) argue that IAFQ negatively reduces instances of committing earnings management. This suggests that in the presence of IAF, financial information data becomes subject to scrutiny to ensure that it is free from any material misstatements. Accordingly, the CEO, as a key internal stakeholder, has the potential to reconstruct the rules, norms, beliefs, and policies that guide IA’s actions (Kalembe et al., 2023; Cai & Mehari, 2015). The agency theorists emphasize two views. First, that internal auditors are agents of shareholders, reporting to the audit committee. Another view is that internal auditors are mere employees of the CEO and not agents of shareholders. The latter view is appropriate for environments with less developed capital markets and, therefore, the agency’s theoretical position suggesting a connection between internal audit function quality and earnings quality needs to additionally consider the effect of the third variable – CEO power. Yet, the influence of CEO power on earnings quality and its moderating effect on the link between internal audit function quality and earning quality has been omitted in prior studies. The stakeholder identification and salience theory (Mitchell etal. 1997) assumes that stakeholders can be identified if they possess power to influence the firm. A stakeholder is defined as a person or group of persons who can affect or be affected by the achievement or realization of an organization’s purpose (Freeman, Freeman et al., 2010), such as the CEO. Deegan (2014) in his attempt to define financial accounting theories presents two contrasting positions of stakeholder theory, namely; the managerial COGENT BUSINESS & MANAGEMENT 5 position and normative position. This study follows the managerial perspective because of its ability to address issues of stakeholder (in this case, the CEO as a key internal stakeholder) power and how this power affects the ability of the CEO to comply with other stakeholders’ expectations, such as high-quality earnings, known to be one of their primary interests in the firm. In the context of Uganda, salient stakeholders such as the Uganda Revenue Authority, Banks, and investors expect powerful CEOs to report high earnings quality to enable them make valid economic decisions. In a firm setting, the CEO is one of the preparers of the financial reports and can therefore use his power to influence earnings quality. This is possible especially if such a CEO is knowledgeable and has a rich background in accounting. It is thus expected that the internal audit function mitigates any intentions of the CEO to manage earnings. In an environment like Uganda, stakeholders have differing accounting earnings needs. For example, banks are interested in high-earnings quality because this acts as a signal that the firm can finance its cost of borrowing (Waweru et al., 2011). URA’s interest entails the assessment of correct taxes to levy from regulated firms. Moreover, the demands of these stakeholders are usually urgent and may require immediate attention from powerful CEOs (Mitchell etal. 1997). Thus, a powerful CEO, as a preparer of financial statements, assesses the importance of meeting stakeholder demands because their needs’ differentials create a lot of contradictions. This leaves the CEO with the task of fulfilling these multiple objectives. As such, he/she is propelled to choose an accounting policy to report earnings that may suit the needs of different stakeholders, including his/hers. 4. Empirical literature review and hypotheses development 4.1 The concept of earnings quality Dechow and Schrand (2004) define high earnings quality as those that accurately reflect the firm’s current operating performance. Dechow et al. (2010) indicate that high earnings quality provides more information about the features of the firm’s financial performance that are relevant for a specific decision made for a specific decision maker. According to Schipper and Vincent (2003), earnings are of high quality if they faithfully represent the Hicksian income. Earnings quality is a multifaceted concept (Gutiérrez & Rodríguez, 2019). For instance; Dechow et al. (2010) used three broad proxies for earnings quality such as properties of earnings (earnings persistence and accruals; earnings smoothness; asymmetric timeliness and timely loss recognition and target beating), investor responsiveness to earnings (earnings response coefficient), and external indicators of earnings misstatements (restatements and internal control procedure deficiencies). Also, Francis et al. (2008) used accounting-based constructs (accrual quality, persistence, predictability, and smoothness) and market-based constructs (value relevance, timeliness and conservatism). Scholars like Adams etal. (2005) developed a power index comprising CEO founder, tenure, president, compensation, and chair. Likewise, Bebchuk et al. (2011) utilized CEO pay slice. Other stock market studies proxied for earnings quality using panel data in terms of faithful representation and relevance to assess earnings quality (Benkraiem et al., 2021; Brahem etal., 2022). For instance, Benkraiem et al. (2021) measured earnings quality using a composite of accurate depiction (accrual quality) and relevance (persistence, predictability, value relevance, and timeliness). Discretionary accruals and earnings informativeness were used by Brahem etal. (2022) as alternate metrics for faithful representation and relevance, respectively. In Uganda, Nkundabanyanga et al.’s (2013) study on accounting quality focused on comparability, accuracy, reliability, relevance, and understandability of financial information in a government Ministry. However, we fail to generalise these findings because of the differences in firms’ internal structures. Therefore, the current study proxies for earnings quality in terms of the fundamental qualitative characteristics of relevance and faithful representation of accounting information as enshrined in the Conceptual Framework for Financial Reporting, 2018. Earnings information is relevant if it is capable of creating a difference in the decisions made by the user. Such information should either have predictive or confirmatory value. Earnings with predictive value has the ability to predict future events, while earnings with confirmatory value enable users to confirm past or earlier earnings. On the other hand, faithful representation means that the information should be free from any error, unbiased and should be complete. 6 D. KALEMBE ETAL. 4.2. Internal audit function quality and earnings quality The Institute of Internal Auditors (IIA, 2022) defines internal audit (IA) as an independent, objective assurance and consulting activity that is designed to add value and improve an organization’s objectives. Abbott et al. (2016) indicates that an internal audit (IA) function is of quality if it has the ability to prevent or detect material misstatements and its inclination to report such misstatements to the audit committee or the external auditor. Consistent with the agency theory (Jensen & Meckling, 1976), internal audit function is a monitoring mechanism that mitigates low earnings quality. The proponents of the agency theory argue that internal auditors are agents of shareholders, reporting to the audit committee. In this view, it is expected that financial statement data is subjected to adequate disclosure and scrutiny by audit committee members since these are responsible for financial reporting process. Previous studies on IAFQ and EQ have found inconsistent results (Nalukenge et al., 2022; Kaawaase et al., 2021; Ismael & Kamel, 2021; Prasad et al., 2021; Tumwebaze et al., 2018; Abbott et al., 2016; Ege, 2015; Johl et al., 2013; Sierra Garcia et al., 2012; Prawitt et al., 2009; Davidson et al., 2005). For example, Tumwebaze et al. (2018) found that internal audit improves accountability in statutory corporations of Uganda. In addition, Nalukenge et al. (2022) revealed that internal audit quality is positively related to accountability of which earnings quality is part. Also, Kaawaase et al. (2021) documented that internal audit quality improves financial reporting quality among financial services firms in Uganda. Accordingly, Madawaki et al. (2022) finds a positive and significant relationship between internal audit and financial reporting quality among listed firms in Nigeria. Prawitt et al. (2009) found that quality internal audit function is negatively associated with lower levels of earnings quality. Ege (2015) posits that quality internal audit function reduces management misconduct. Also, Sierra García et al. (2012) revealed that internal audit function has a negative relation with earnings management. Contrarily, Davidson et al. (2005) found that internal audit function is not associated with low accruals earnings management. Further, Prasad etal. (2021) reported no significant relationship between internal audit use and earnings quality. Similarly, Ismael and Kamel (2021) documented that internal auditor’s independence does not reduce abnormal accruals among listed firms in UK. Ardianto etal. (2023) revealed that the internal audit function does not efficiently enhance the efficiency of corporate investment decisions in Indonesian public companies. A study by Kabuye et al. (2018) revealed that internal audit activities do not significantly predict fraud management in Uganda’s financial services sector. Kalembe et al. (2023) provides evidence that IAF is a ceremonial function, and its focus is on internal controls, risk management and compliance but not financial statement reviews. This leaves material misstatements go undetected. Similarly, lapses and laxity in corporate governance mechanisms including internal audit, are evident making the internal audit function ineffective in commercial Banks in Uganda (Bank of Uganda supervision report, 2022). This suggests that internal audit may not be a disciplining mechanism in Uganda. We hypothesize that; H1: A relationship exists between IAFQ and EQ. 4.3. Internal audit function quality and CEO power Consistent with the managerial branch of stakeholder theory (Mitchell etal., 1997), internal auditors can be treated by the CEO in the same way as any other employee. As a result, they succumb to social pressure and familiarity from CEOs to perform in their interests (Kabuye et al. 2018), despite their code of ethics. This suggests that the internal auditors may not find it possible to add value to firms through provision of assurance and advisory roles in the presence of a powerful CEO (Roussy & Brivot, 2016). Studies that empirically examine the relationship between internal audit function quality and CEO power report consistent results (for instance, Jiang etal., 2018; Prawitt etal., 2009; Hermalin & Weisbach, 1998). For instance, Jiang et al. (2018) reported that IAFQ is negatively associated with CEO power. Similarly, Hermalin & Weisbach (1998) indicated that IAFQ can be reduced by an opportunistic powerful CEO since he/she has the discretion to bargain with the board to reduce its potential monitoring. Prawitt et al. (2009) documented that a high-quality internal audit function deters earnings manipulation and fraud that would have been committed by a powerful CEO. Literature concurs with the current situation in Uganda. The Bank of Uganda Supervision Report (2022) adduced evidence that internal auditors have COGENT BUSINESS & MANAGEMENT 7 continued to face threatened independence from CEOs in commercial Banks. In addition, it is also indicated that CEOs usurp powers of the board and as a result, they become ineffective in monitoring the internal auditors (Bank of Uganda Supervision Report, 2022). We thus hypothesise that; H2: A negative relationship exists between IAFQ and CEO power. 4.4. CEO power and earnings quality The theory of stakeholder identification and salience (Mitchell et al., 1997) indicates that a CEO is a key internal stakeholder who has power to influence organizational outcomes, hence making him the most powerful figure in a firm (Li et al., 2016). French and Raven (1959) define power as a relationship between two or more people where one person has more influence than the other, translating into psychological change. CEO power is a relational construct, implying that it arises out of social relations and networks (Cormier etal., 2016; Tang etal., 2011). This power can be disastrous to the firm especially when its exercise can be abused for example, through manipulation of earnings (Cormier et al., 2016) and also when such a CEO may not value any independent advice or have his decisions scrutinized (Han et al., 2016). Literature conceptualizes CEO power as structural power, expert power, ownership power, prestige power and formal power (Sheikh, 2019; Lisic et al., 2016; Finkelstein, 1992; Pearce & Zahra, 1991). CEO’s structural power comes from CEO’s legitimate authority, in terms of his relative compensation (CEO Pay slice) (Shiah-Hou, 2021; Sheik, 2019; Lisic et al., 2016; Bebchuk et al., 2011; Finkelstein, 1992). Expert power emanates from CEO’s financial experience, the number of positions (functional positions) served in, and CEO Tenure. CEO ownership is understood in terms of a CEO founder, relative and his shareholdings in a firm (Sheikh, 2019; Finkelstein, 1992). Formal power entails power of the CEO relative to that of the board (Pearce & Zahra, 1991). The majority of these studies (See, Shiah-Hou, 2021; Sheikh, 2019; Lisic et al., 2016; Bebchuk etal., 2011) conceptualised CEO power in terms of structural power, ownership power and expert power, and ignored prestige power, arguing that it is not a proximal measure of CEO power and, that its hard to obtain data using secondary sources (Han et al., 2016; Tang et al., 2011). This gap is closed in this study because it is easier to understand the perceptions of the respondents using a survey instrument. Consistent with stakeholder identification and salience theory (Mitchell et al., 1997), powerful CEOs create an opaque environment which lowers the quality of earnings profitability. This aligns with existing empirical stock market studies (Arif etal., 2023; Kalembe et al., 2023; Shiah-Hou, 2021; Cormier et al., 2016; Friedman, 2014; Bebchuk et al., 2011; Feng et al., 2011; Malmendier & Tate, 2009; Adams et al., 2005) that documented a relationship between CEO power and earnings quality. For instance, Arif etal. (2023) reported that CEOs with political, structural and expert power have significant detrimental effect on earnings quality. Relatedly, Kalembe et al. (2023) reported that CEO power significantly reduces earnings quality in Uganda. Shiah-Hou (2021) revealed that CEO power significantly lowers earnings quality in U.S. Also, Cormier et al. (2016) noted that firms that are accused of financial misreporting exhibit strong CEO power in Canada. Friedman (2014) revealed that powerful CEOs pressure CFOs to bias financial statement results, which consequently lowers the financial reporting quality. According to Feng et al. (2011), accounting manipulations are more likely when CEO power is high. Bebchuk etal. (2011) documented that CEO pay slice is associated with lower accounting profitability. In another study by Malmendier and Tate (2009), it was found that superstar CEOs are more likely to influence reported performance by managing earnings to meet the market’s higher expectations. Further, Adams etal. (2005) reported that CEOs who have more decision-making power experience more variability in firm performance. This may not be any different in Uganda where gaps in financial reporting and non-compliance with implementation of IFRS as per the requirements of Financial Institutions Act 2004 are instigated by powerful CEOs (Bank of Uganda Supervision Report, 2022). It is then likely, that the quality of reported earnings will be low. We note that previous studies that have tested the association of all five dimensions (structural, expert, ownership, formal, and prestige power) of CEO power on earnings quality are few. Moreover, Prawitt et al. (2009) also calls for further studies to explore the association between CEO power and earnings quality (Earnings management) using larger samples in different contexts. This study responds to this call. We therefore hypothesize that; H3: A negative relationship exists between CEO power and EQ. 14 D. KALEMBE ETAL. Table 6. summary of Loading estimates, validity and reliability tests for Ceo Power and path coefficients -the measurement model of Ceo power. item scales α rho_a CR aVe 0.840 0.850 0.868 0.684 Ceo Power expert Power EP04: this firm’s Ceo is a skilled accountant 0.914 0.904 0.917 0.933 0.776 EP05: this firm’s Ceo is a acknowledgeable accountant 0.902 EP06: this firm’s Ceo has certifications in accounting 0.885 EP07: this firm’s Ceo is a finance expert 0.820 Formal Power FP09:this firm’s Ceo makes decisions about future acquisitions 0.852 0.880 0.888 0.913 0.678 FP10: this firm’s Ceo makes decisions about future divestments 0.885 FP11:this firm’s Ceo makes decisions about capital acquisitions 0.785 FP12: this firm’s Ceo approves changes in capital structure 0.866 FP14: this firm’s Ceo plans for top leadership succession 0.719 ownership Power OP01: this firm’s Ceo owns the highest number of shares 0.821 0.823 0.824 0.883 0.654 OP02:this firm’s Ceo turnover is less involuntary 0.803 OP03: this firm’s Ceo is its founder 0.848 OP04: this firm’s Ceo is a relative to the founding family 0.761 Prestige Power PP01:this firm’s Ceo enjoys the support of ruling government 0.772 0.800 0.804 0.868 0.623 PP02:this firm’s Ceo comes from a wealthy family 0.804 PP03: this firm’s Ceo comes from a powerful tribe 0.805 PP06: this firm’s Ceo comes from a powerful family 0.774 structural Power SP01: Comparing with other top executives, this firm’s Ceo enjoys much benefit 0.826 0.853 0.883 0.899 0.690 SP02: Comparing with other top executives, our Ceo is paid much more than necessary 0.869 SP03: Comparing with other top executives, this firm’s Ceo gets larger allowances 0.806 SP04: Comparing with other top executives, this firm’s Ceo takes home (benefits and salary) more than other executives in this company 0.821 Paths: expert Power → Ceo Power (β =.333, t = 4.554, p < .05); Formal Power → Ceo Power (β =.411, t = 4.979, p < .05); ownership Power → Ceo Power (β = .371, t = 9.293, p < .05); Prestige Power → Ceo Power Quality (β =.263, t = 4.801, p < .05); structural Power → Ceo Power (β =.271, t = 3.945, p < .05) Notes: FP = Formal power, PP = Prestige power, eP = expert power, sP = structural power, oP = ownership power. Source: Primary data. Table 7. Descriptive statistics. Variable n Min Max Mean std. Deviation skewness Kurtosis statistic statistic statistic statistic std. error statistic statistic std. error statistic std. error Capital structure 136 1.00 3.00 1.40 0.05 0.518 0.817 0.213 −0.523 0.423 Firm age 136 1.00 4.00 3.07 0.09 0.982 −0.711 0.213 −0.650 0.423 auditor type 136 1.00 3.00 1.53 0.04 0.499 0.066 0.213 −1.463 0.423 ownership 136 1.00 2.00 1.53 0.04 0.493 −0.110 0.213 −1.986 0.423 Board size 136 1.00 2.00 1.53 0.04 0.485 −0.110 0.213 −1.950 0.423 Ceo interlock 136 1.00 2.00 1.26 0.04 0.429 0.084 0.213 −0.752 0.423 employee size 136 1.00 4.00 2.89 0.10 1.156 −0.444 .213 −1.343 0.423 structural Power 136 1.00 3.75 2.15 0.07 0.763 0.025 0.213 −0.879 0.423 Prestige power 136 1.00 3.14 2.09 0.05 0.554 −0.395 0.213 −0.661 0.423 expert power 136 1.00 4.00 2.26 0.08 0.876 0.341 0.213 −0.775 0.423 Formal power 136 1.67 4.00 3.03 0.05 0.618 −0.415 0.213 −0.667 0.423 ownership power 136 1.00 3.50 1.83 0.07 0.745 0.442 0.213 −1.199 0.423 Ceo Power 136 1.13 3.28 2.27 0.04 0.429 0.246 0.213 0.034 0.423 Competence 136 2.17 4.00 3.18 0.04 0.464 −0.23 0.213 −0.291 0.423 investment 136 1.75 4.00 3.02 0.05 0.519 −0.272 0.213 −0.331 0.423 ia org. status 136 2.00 4.00 3.16 0.05 0.512 −0.113 0.213 −0.355 0.423 Role Performance 136 2.00 4.00 3.11 0.04 0.466 −0.426 0.213 0.381 0.423 ia function Quality 136 1.98 3.90 3.12 0.03 0.385 −0.506 0.213 0.492 0.423 Relevance 136 1.67 4.00 2.67 0.04 0.399 0.167 0.213 0.363 0.423 Faithful Representation 136 1.75 3.35 2.63 0.03 0.295 0.001 0.213 0.500 0.423 earnings Quality 136 1.75 3.93 2.92 0.04 0.467 −0.123 0.213 −0.119 0.423 Source: Primary data. COGENT BUSINESS & MANAGEMENT 15 constructs, indicating that the data points are close to the means (Field, 2013), hence representing the observed data. To determine whether firm differences influenced the study variables, a one-way analysis of variance (ANOVA) was used to determine the impact of firm sector on the study variables. The results of the one-way ANOVA in Table 8 show that the p-values of all the study variables are above 0.05. This suggests that the group differences between firms did not significantly influence their responses between the study variables. Also, the overall differences between respondents did not bias the results of the study (see, Table 9). 6.1. Correlation analysis The results in Table 10 revealed that IAFQ is positive but not significantly related to EQ (r = .025, p > 0.01). The results imply that a unit change in IAFQ in terms of competence for example, may lead to a unit change in earnings quality. The correlation between IAFQ and CEO power is negative but not statistically Table 8. anoVa global variables and firm sector. Variable sector NMean std. Deviation df F sig Ceo power Banking institutions 49 2.2478 .59210 3 .404 .751 insurance Companies 27 2.2070 .44861 132 utility Companies 11 2.6555 .59347 135 others 87 2.3364 .48012 total 174 2.3116 .52264 serving two masters Banking institutions 49 1.7918 .76263 3 .317 .813 insurance Companies 27 1.7630 .64220 132 utility Companies 11 1.7273 .55334 135 others 87 2.1103 .87218 total 174 1.9425 .80485 audit committee effectiveness Banking institutions 49 3.1017 .67308 3 .784 .610 insurance Companies 27 3.3706 .55690 132 utility Companies 11 2.5827 .86268 135 others 87 3.1462 .51821 total 174 3.1329 .61421 internal audit function quality Banking institutions 49 3.2046 .63631 3 .668 .773 insurance Companies 27 3.0556 .75855 132 utility Companies 11 3.3977 .43279 135 others 87 3.0440 .49173 total 174 3.1134 .58334 earnings quality Banking institutions 49 2.7245 .70313 3 .283 .838 insurance Companies 27 2.9802 .59888 132 utility Companies 11 2.2636 .89137 135 others 87 2.8686 .56002 total 174 2.8071 .64823 Source: Primary data. Table 9. global variables and position of respondents in a firm. Variable Respondents’ Position NMean std. Deviation df F significance Ceo power CFo 131 2.2527 .49000 3 .404 .798Cia 43 2.4907 .58166 132 total 174 2.3116 .52264 135 serving two masters CFo 131 1.9695 .81768 3 .317 .920Cia 43 1.8605 .76785 132 total 174 1.9425 .80485 135 audit committee effectiveness CFo 131 3.1616 .63645 3 .784 .710Cia 43 3.0454 .53827 132 total 174 3.1329 .61421 135 internal audit function quality CFo 131 3.1179 .58689 3 .668 .823Cia 43 3.0994 .57901 132 total 174 3.1134 .58334 135 earnings Quality CFo 131 2.8621 .61803 3 .283 .838Cia 43 2.6395 .71441 132 total 174 2.8071 .64823 135 Source: Primary data. 16 D. KALEMBE ETAL. Table 10. Zero order matrix. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 CaPstR (1) 1 age (2) .161 1 auD (3) −.048 −.244** 1 oWn (4) .042 −.121 .473** 1 BsiZe (5) −.175*−.287** .303** .209*1 Ceoin (6) .090 .009 −.093 −.080 −.138 1 eMPsiZe (7) .062 .165 −.282** −.169 −.413** .117 1 Comp (8) .021 .013 −.230** −.129 .057 .075 .110 1 inv (9) .049 .129 −.242** −.155 −.038 .118 −.052 .639** 1 orgstat(10) .119 .037 −.012 .045 −.086 .147 .159 .324** .337** 1 Roleperf(11) .026 .034 −.161 −.013 .131 −.024 −.022 .658** .561** .451** 1 iaFQu (12) .070 .070 −.204*−.080 .015 .104 .062 .823** .811** .680** .839** 1 struct (13) −.119 .167 −.066 −.121 −.249** −.060 .030 .025 .118 −.134 .056 .020 1 Prest (14) −.046 .118 .015 .084 −.298** −.041 .255** −.025 −.040 .083 −.025 −.001 .323** 1 exp (15) −.018 .106 −.053 −.057 −.263** −.011 .144 −.021 .036 .032 −.048 .002 .149 .204*1 Form (16) .057 −.170 .347** .385** .031 −.040 −.013 −.137 −.181*.122 −.155 −.108 −.093 .240** .066 1 owners (17) −.132 .055 .255** .261** .028 −.181*−.124 −.040 −.022 −.229** .013 −.092 .321** .180*.233** .404** 1 CeoP (18) −.091 .103 .147 .157 −.254** −.111 .089 −.059 −.014 −.058 −.046 −.056 .585** .588** .614** .484** .720** 1 Rel (19) .048 −.164 −.053 .029 .103 .039 −.148 .104 .094 −.029 −.038 .042 −.072 −.217*−.150 −.066 −.241** −.246** 1 FR (20) .071 −.220*−.054 −.066 .148 .004 .077 .031 −.094 .066 −.022 −.007 −.214*−.168 −.280** .022 −.317** −.338** .328** 1 eQ (21) .071 −.231** −.065 −.015 .150 .029 −.061 .088 .015 .015 −.038 .025 −.163 −.239** −.253** −.034 −.336** −.349** .863** .760** 1 *Correlation is significant at the 0.05 level (2-tailed). **Correlation is significant at the 0.01 level (2-tailed). KEY: CaP = Capital structure, eMPsiZe = employee sizeage = Firm age, auD = auditor type, oWn = ownership, Board = Board size, Ceoint = Ceo interlock, structuralP = structural Power, Prest = Prestige Power, exp = expert Power, Form = Formal Power, owners = ownership Power, CeoP = Ceo power, Comp = Competence, inv = investment, orgstat = internal audit organization status, Role Perf = Role Performance, iaFQu = internal audit Function Quality, Rel = Relevance, FR = Faithful Representation, eQ = earnings Quality. Source (S): Primary Data. COGENT BUSINESS & MANAGEMENT 17 significant (r = –.056, p > 0.01). The finding implies that IAFQ diminishes when the CEO is powerful. The results reveal that all the dimensions of IAFQ are negative but not significantly related to CEO power. Further, the results also indicated that CEO power (CEOP) is negative and significantly related to EQ (r = –.349**, p < 0.01). The finding implies that CEOP causes a reduction in earnings quality. Amongst all the dimensions of CEOP, expert power and prestige power are significant and negatively related to EQ. In terms of control variables, only firm age is negative and statistically significant with EQ. Other control variables such as ownership, capital structure, employee size, CEO interlock, auditor type, and Board size are not significantly related to earnings quality. 6.2. Regression analysis results The regression analysis results are obtained from an estimated PLS-SEM model that was run to establish the extent of the relationships between the independent variable and dependent variable. We used the standardised beta values because they take on real values without a standard measurement and are easier to interpret (Field, 2013). The results in Table 11 indicate that IAFQ is positive but not significantly related with EQ at this level of analysis (β = 0.102, t = 0.594, p > 0.05), hence H1 was not supported. The model returned an NFI of 0.100, which is an indication that there is strong convergent validity. Other model fit indices such as SRMR are assessed and this returns 0.000, a signal that the model fits the data well. Relatedly, the path coefficient between IAFQ and CEOP revealed a negative relation although this association was not significant (β= –0.341, t = 1.070, p > 0.05). Thus, H2was supported. The model returned an NFI = 1.000, chi-square is non-significant at 0.05 level of significance, and SRMR = 0.000. This implies that the model acceptably fits the data well. The results further indicate that CEOP is statistically significant and negatively related to EQ (β = –0.360, t = 4.477, p < 0.05), hence H3 was substantiated. These results suggest that increases in CEOP causes negative variations in EQ. Fit indices for this model are NFI = 1.000 and SRMR = 0.000. According to the results, CEO power explains 11.7% of variations in earnings quality. Table 11. Direct paths and hypothesized relationships. Hypothesized path Hypothesis Coeff t p adj R2 nFi IAFQ -> EQUA H1: Not supported 0.102 0.594 0.553 4.20% 1.000 Control variables auditor size → eQua no Hypothesis −0.143 1.219 0.223 Boardsize → eQua no Hypothesis 0.166 1.492 0.136 CeointerL → eQua no Hypothesis 0.005 0.05 0.96 eMPsize→ eQua no Hypothesis 0.01 0.08 0.936 age → eQua no Hypothesis −0.251* 2.565 0.01 CaPstR → eQua no Hypothesis 0.132 1.406 0.16 ownership → eQua no Hypothesis −0.024 0.209 0.835 IAFQ → CEO power H2: Not supported −0.341 1.07 0.898 2.00% 1.000 Control variables age → Ceo power no Hypothesis 0.067 0.512 0.609 auditortype → Ceo power no Hypothesis 0.222 1.584 0.113 Boardsize → Ceo power no Hypothesis −0.324 1.777 0.076 CaPstR → Ceo power no Hypothesis −0.14 1.113 0.266 Ceointerl → Ceo power no Hypothesis −0.121 1.149 0.25 eMPsize → Ceo power no Hypothesis 0.073 0.683 0.495 ownership → Ceo power no Hypothesis 0.217 1.183 0.237 CEO power → EQUA H3: Supported −0.360 4.477 0.000 11.70% 1.000 Control variables age → eQua no Hypothesis −0.209* 2.553 0.011 auditor type → eQua no Hypothesis −0.089 0.784 0.433 Board size → eQua no Hypothesis 0.058 0.663 0.508 CaPstR → eQua no Hypothesis 0.071 0.857 0.392 Ceointerl → eQua no Hypothesis −0.033 0.365 0.715 eMPsize → eQua no Hypothesis 0.05 0.541 0.588 ownership → eQua no Hypothesis 0.014 0.148 0.882 Notes: CaPstR = Capital structure, Ceointerl = Ceo interlock, eMPsize = employee size, iaFQ = internal audit function Quality, eQua = earnings Quality. Source(S): Primary Data. 18 D. KALEMBE ETAL. Further, results in Table 11 revealed that CEOP is the most significant predictor of EQ in Uganda relative to IAFQ. Overall, the model, which also controls for firm age, employee size, auditor type, board size, CEO interlock, capital structure, and firm ownership, accounts for 12.5% of the variations in earnings quality. However, the position of this study has been that the relationship between IAFQ and EQ is moderated by CEOP, according to which H4 was put forward. We use multivariate regression analysis to test the extent to which CEOP and its interactions with IAFQ significantly predict EQ. Using SmartPLS, we initially tested for the influence of internal audit function and found that it accounts for 4.2% variations in earnings quality (Table 11). We also tested for the influence of CEOP and found that it accounts for 11.7% of significant variances in EQ (See Table 11). Using SPSS (see, Table 12), CEOP is the only significant predictor of EQ, accounting for 10% of variations in EQ. We then add the interaction term (CEOP * IAFQ) to test for H4. With the addition of the interaction term, the variances explained improved to 12.1% (see, Table 12). The interaction term is significant (β = –.494, p<.000). This means that the interaction of IAFQ and CEOP explains more of the variances in overall EQ than the direct influence of IAFQ or CEO power on their own. As earlier indicated, the strength or direction of the relationship between IAFQ and EQ should be affected by CEOP for the moderation effect to be claimed. As can be seen from Table 12, without the interaction term, IAFQ is not a significant predictor of EQ but when the interaction term is included, it becomes a significant predictor (β = 1.213, p<.05). The results show that the interactive-term boosts the main effects (IAFQ and CEOP) to explain the variances in EQ. Since the interaction term is significant it is maintained that H4 is supported. This supports the assertion of the existence of a moderating effect of CEOP in the relationship between IAFQ and EQ in Uganda. As an additional check, we examine the slopes of graphs to determine the moderating effect of the variables using Jose’s (2008) Modgraph. The graph (See, Appendix, Table A4) shows that the lines are not Table 12. the moderation effect of Ceo power in the relationship between internal audit function quality and earnings quality. Coefficientsa Model Unstandardized Coefficients Standardized Coefficients t Sig BStd. Error Beta 1 (Constant) 3.434 .394 8.710 .000 Ceo power_CentReD −.371 .094 −.323 −3.951 .000 internal audit function quality_CentReD .106 .102 .085 1.037 .302 R = 0. .337; R2 = .114; adj.R2 = .100; F = 8.540; sig. = 0.000; e = 0.46952; Durbin-Watson = 1.979 (Constant) −.078 1.767 −.044 .965 Ceo power_CentReD 1.193 .773 1.038 1.543 .125 internal audit function quality_CentReD 1.213 .553 .970 2.195 .030 Ceo power_CentReD * internal audit function quality_CentReD −.494 .242 −1.599 −2.037 .044 R = 0. .375; R2 = .141; Adj.R2 = .121; F = 7.212; Sig. = 0.000; e = 0.46406; Durbin-Watson = 1.961 Model 1 Model2 Model3 (Constant) 3.116** 3.139** 3.198** Finance .103 .099 .062 age −.115** −.117** −.102** ownership −.007 −.005 .033 auditor type −.129 −.116 −.046 Ceo interlock .030 .018 −.021 Board size .120 .109 −.002 internal audit function quality_CentReD .106 .116 Ceo power_CentReD −.336** R.278 .290 .392 R2.077 .084 .153 F Change 1.803 .954 10.390** R2 Change .077 .007 .069 SE .48645 .48654 .46962 df1 6 1 1 df2 129 128 127 Durbin Watson static 1.887 note: **P < 0.01. aDependent Variable: eQu. Source: Primary data. COGENT BUSINESS & MANAGEMENT 19 parallel but do cross each other at some point. This is in line with Aiken and West (1991), who argue that for the interaction to be significant and interpretable and for moderation to have taken place, the graphs should not be parallel or must have different gradients or slopes, implying that the magnitude of an effect is greater at one level of a variable than at another level. We, therefore, conclude that CEOP significantly interacts with IAFQ. The graph (Appendix, Table A4) suggests that the greater the power of the CEO, the lower the effect of IAF on EQ. This proves the inverse relationship between CEOP and EQ even with increasing IAFQ. The results from multivariate analysis further confirm the assertion that the interaction term is causing significant variances in EQ and that the consistency of results even in the face of the control variables is further testimony of the predictive validity of the interaction. So, in closing, the results show that CEOP is negatively related to and a significant predictor of EQ. IAFQ while related to EQ is not a significant predictor of EQ. The interaction effects of IAFQ and CEOP on EQ are significant, according to which when CEOP is high, the effect on IAF and EQ is low. 6.2.1. Additional analyses We further check whether CEOP can mediator in the relationship between IAFQ and EQ. The results reveal that CEOP does not mediate the relationship between IAFQ and EQ (β = 0.087, t = 0.096, p > 0.05). This suggests that in Uganda’s environment, CEOP is a moderator and not a mediator, hence confirming H4. Further, we subject our data to endogeneity bias test using Gaussian Copula beta coefficients (Hult et al., 2018). The results in Table 13 reveal that beta coefficients for all the study variables are not significant (p-values >0.05), indicating the absence of endogeneity issues in our data. 6.3. Discussion The results of the current study indicate that CEOP is statistically significant and negatively related to EQ. As CEOP is a negative element in EQ, Uganda’s earnings are perceived to be low. This factor is likely to be unique to Uganda’s environment and other similar environments. The finding arguments the theory of stakeholder identification and salience (Mitchell et al., 1997) that suggests that powerful CEOs create an opaque financial reporting environment which limits disclosure of accurate financial statement data. This potentially creates problems of accountability leaving stakeholders unable to evaluate the performance of the firm. The finding may suggest that CEOs who are knowledgeable accountants are likely to report values that are not accurate and justifiable. This finding is in line with previous scholars (e.g, Kalembe et al., 2023; Shiah-Hou, 2021; Duong etal., 2020; Ngo & Nguyen, 2022). Kalembe etal. (2023) reported that CEO power is negative and significantly related to EQ. Also, Shiah-Hou (2021) documented a negative and significant relationship between CEOP and EQ. Also, Duong et al. (2020) found out that powerful CEOs magnify the negative impact on EQ especially if the CFO is a board member. Further, Ngo and Nguyen (2022) concluded that firms with CEOs who have financial and accounting experience have more influence and intervention on earnings management which consequently affects financial reporting quality. Similarly, Arif et al. (2023) documented that CEOs with political, structural and expert power have significant detrimental effect on earnings quality. Internal audit function quality is positive and not significantly related to earnings quality. This finding is consistent with the agency theoretical position suggesting a connection between internal audit function quality and earnings quality. This finding is in line with Tumwebaze et al. (2018) who found that IA improves accountability in statutory corporations of Uganda. Similarly, Nalukenge et al. (2022) revealed that internal audit quality is positively related to accountability of which earnings quality is part. Also, Kaawaase et al. (2021) documented that internal audit quality improves financial reporting Table 13. endogeneity test. Models test Beta Coefficients p-values Model 1 gaussian Copula (iaFQ)-> CeoP −1.132 0.094 Model 2 gaussian Copula (CeoP)-> eQu −0.170 0.797 Model 3 gaussian Copula (iaFQ)-> eQu −0.1346 0.622 source (s): Primary Data. Notes: P values > 0.05. 20 D. KALEMBE ETAL. quality among financial services firms in Uganda. Accordingly, Madawaki etal. (2022) revealed a positive and significant relationship between internal audit and financial reporting quality among listed firms in Nigeria. A similar study by Ege (2015) found out that quality internal audit function reduces management misconduct. These previous studies extend the debate that internal audit is a key monitoring mechanism in regulated firms in Uganda. This renders this function a disciplining mechanism that mitigates fraud, errors and financial misreporting. However, contradictory findings have also been found (see, Ismael & Kamel, 2021 and Ardianto et al., 2023; Prasad et al., 2021). Further, internal audit function quality is negative but not related significantly to CEO power. The finding is consistent with agency’s theoretical perspective that argues that the principal’s additional monitoring costs in form of an IA are potentially futile with very powerful CEOs. This finding also contradicts the theory towards stakeholder identification and salience (Mitchell et al., 1997), specifically the managerial perspective (Deegan, 2014), which suggests that the internal auditor is considered as just any other stakeholder to manage by the CEO. This is intuitively sensible because a powerful CEO has the potential to reconstruct the rules, norms, beliefs, and policies that guide his or her action (Cai & Mehari, 2015). The findings are consistent with previous studies such as Hermalin & Weisbach (1998) who argue that IAQ can be reduced by powerful CEOs especially when he/she can bargain with the board to reduce its potential monitoring role. Similarly, Jiang etal. (2018) also reported that IAFQ is negatively associated with CEOP. Finally, the interaction of IAFQ and CEO power explains more variances in overall earnings quality than the direct influence of IAFQ or CEOP on their own. This finding potentially clears the ambivalence existing in literature (see, Davidson et al., 2005; Ege, 2015; Johl et al., 2013; Prawitt et al., 2009) on the role of IAF in enhancing EQ by showing that the greater the power of the CEO, the lower the effect of IAF on EQ. Because of this interplay, ambivalent results on the efficacy of internal audit function quality on earnings quality will continue to surface in studies that do not address the moderating role of CEOP. The current results support the idea that a research design involving at least two independent variables should consider examining the main effects of each of the independent variables (Friedrich, 1982) and also the multiplicative effects. IAFQ will influence earnings quality given a level of the firm’s CEOP and CEOP will influence EQ given a level of IAFQ. These results indicate that CEOP and IAFQ cause a magnitude effect on EQ hence the assumption of non-additivity is met (Aiken & West, 1991; Friedrich, 1982; Jose, 2008). It signifies that the two coexist to influence EQ in Uganda’s regulated firms. In the present case, EQ reduces as CEOP goes up and internal audit function quality is enhanced, suggesting that an interactive effect of CEOP and IAFQ is significant in Uganda’s regulated firms. The finding that CEO power interacts with IAFQ to cause significant variations in earnings quality upholds the framework put forward by agency and stakeholder identification and salience theories suggesting that when a CEO is powerful, he can decide to negatively affect the relationship between internal audit function quality and earnings quality. This supports the view that in environments where market disciplining mechanisms are weak, internal auditors are likely to be mere employees of the CEO and not agents of the principal. 7. Summary and conclusions The aim of the study was to establish the relationship between internal audit function quality, CEO power and earnings quality. The study also examined how CEO power moderates the relationship between internal audit function quality and earnings quality. The findings revealed that internal audit function quality has a positive but not significant relationship with earnings quality. The results also found that internal audit function quality and CEO power are negatively related but the relationship is not significant. Also, CEOP is statistically significant and negatively related to EQ. Similarly, the interaction effects of CEO power in the link between internal audit function quality and earnings quality are significant to the extent that the greater the power of the CEO, the lower the effect of IAFQ on earnings quality. The results of this study offer some important implications for the academic community, practitioners, standard setters, and the regulators. For academics, the results suggest that CEO power is more important for earnings quality than internal audit, and this maybe the reason for poor earnings quality. This is because powerful CEOs have the discretion to influence the quality of reported earnings. COGENT BUSINESS & MANAGEMENT 21 Accordingly, the study unearthed the use of perceptions in understanding earnings quality contrary to previous studies that used panel data. As regards theory, this study has shown that the relationship between internal audit function quality, CEO power and earnings quality and the multiplicative effect of CEO power and internal audit function quality can be investigated following a multi-theoretic approach of agency, stakeholder identification and salience theories. In terms of practice, the results offer a sensitivity of earnings quality (proxied by relevance and faithful representation) to internal audit function quality and CEO power. Importantly, the relationship between internal audit and earnings quality is a function of CEO power when internal audit staff are regarded as employees responsible to the CEO and not agents of the principal (Jensen & Meckling, 1976). Also, the results are of interest to the standard setters-International Accounting Standards Board (IASB) since powerful CEOs appear to be a negative element in their interaction with internal audit function that ensures compliance with accounting policies such as compliance with IFRS. The paper has shown that CEO power is one of the ills vitiating earnings quality. In this regard, it is meaningful to identify the significant multiplicative effects of CEO power and internal audit function quality to uncover what needs to be done to improve earnings quality in Uganda. The regulators such as the Institute of Corporate Governance of Uganda could consider these results to identify the needed changes in their regulation of accounting practice and governance in Uganda. Primarily, the measurement models of earnings quality, CEO power, and internal audit quality reported in this paper allow governments and the institute of internal auditors to evaluate the effectiveness of current firm governance provisions and internal audit status. The salient stakeholders of earnings information such as the Uganda Revenue Authority, Banks, and investors should take note of these findings. As with any study, there are several limitations with the present one. First, the questionnaire was self-administered and we did not undertake follow-up interviews which would have informed me of the reasons why the respondents held certain views. Second, the present study is cross-sectional; the views held by individuals may change over the years. Finally, although there is an attempt at controlling CMV in particular with proactive instrument design, testing the influence of CMV may not have been dealt away completely owing to the failure to find a plausible common marker variable. Well, policymakers of Uganda dealing with public interest firms, academicians, company owners, and even general readers interested in the field of corporate governance and audit, in particular, might find this study useful. Future research may wish to test this study’s model in predicting earnings quality using the framework for financial reporting 2018 in developed economies. Author contribution The authors confirm contribution to the paper as follows: study conception and design: Ms. Dorcus Kalembe, Prof. Stephen K. Nkundabanyanga and Prof. Twaha Kigongo Kaawaase: Data collection: Ms. Dorcus Kalembe: Data analysis and interpretation: Ms. Dorcus Kalembe, Prof. Prof. Stephen K. Nkundabanyanga: Data manuscript and preparation: Ms. Dorcus Kalembe, Prof. Stephen K. Nkundabanyanga, Prof. Twaha Kigongo Kaawaase, Dr. Kayongo Isaac. All authors reviewed the results and approved the final version of manuscript. Disclosure statement No potential conflict of interest was reported by the author(s). Explanation for use of the same dataset and why the study has been separated into different publications This manuscript ‘Submission ID 232677893’ is an extract from my PhD thesis. Already, one paper has so far been published from this thesis, and the same data set is also being used for this submitted manuscript. We thus disclose as authors previous publication of data since these publications are from the same PhD thesis. Second, as a university requirement, I am required to have at least two publications from my PhD thesis. The manuscript submitted to your journal will be the second publication. All these publications are aimed at disseminating Knowledge. 22 D. KALEMBE ETAL. About the authors Dorcus Kalembe is a Lecturer at Makerere University Business School, Department of Accounting and Finance. She is currently persuing her PhD in the area of accounting. Her research interests are in the areas of financial accounting, auditing, corporate governance and sustainability reporting. Stephen Korutaro Nkundabanyanga is a reknowned researcher and a professor of accounting at Makerere University Business School, Department of accounting. He has vast knowledge and research interests in financial accounting, corporate governance, taxation, sustainability reporting among others. Twaha Kigongo Kaawaase is a professor at Makerere University Business School, Department of auditing and taxation. His research interests are in the areas of auditing, financial reporting and accounting, taxation and corporate governance. Isaac Newton Kayongo is a PhD holder and a senior lecturer in the Department of Finance. His research interets are in behavioural finance, accounting and corporate governance. ORCID Kalembe Dorcus http://orcid.org/0000-0002-3466-0635 Stephen Korutaro Nkundabanyanga http://orcid.org/0000-0003-0940-9745 Twaha Kigongo Kaawaase http://orcid.org/0000-0002-4793-6709 Data availability statement I confirm that the data set is available upon request. References Abbott, L. J., Daugherty, B., Parker, S., & Peters, G. F. (2016). Internal audit quality and financial reporting quality; the joint importance of independence and competence. Journal of Accounting Research, 54(1), 1–28. https://doi. org/10.1111/1475-679X.12099 Abdullah, N. S., & Ismail, K. I. N. (2016). Women directors, family ownership and earnings management in Malaysia. Asian Review of Accounting, 24(4), 525–550. https://doi.org/10.1108/ARA-07-2015-0067 Adams, R. B., Almeida, H., & Ferreira, D. (2005). Powerful CEOs and their impact on corporate performance. Review of Financial Studies, 18(4), 1403–1432. https://doi.org/10.1093/rfs/hhi030 Aggarwal, R., Klapper, L., & Wysocki, P. D. (2005). Portfolio preferences of foreign institutional investors. Journal of Banking & Finance, 29(12), 2919–2946. https://doi.org/10.1016/j.jbankfin.2004.09.008 Alzoubi, E. S. S. (2019). Audit committee, internal audit function and earnings management: Evidence from Jordan. Meditari Accountancy Research, 27(1), 72–90. https://doi.org/10.1108/MEDAR-06-2017-0160 Arif, H. M., Mustapha, M. Z., & Abdul Jalil, A. (2023). Do powerful CEOs matter for earnings quality? Evidence from Bangladesh. PloS One, 18(1), e0276935. https://doi.org/10.1371/journal.pone.0276935 Aiken, L. S., & West, S. G. (1991). Testing and interpreting interactions. SAGE. Ardianto, A., Anridho, N., Ngelo, A. A., Ekasari, W. F., & Haider, I. (2023). Internal audit function and investment efficiency: Evidence from public companies in Indonesia. Cogent Business & Management, 10(2), 2242174. https://doi. org/10.1080/23311975.2023.2242174 Athari, S. A. (2022b). Does investor protection affect corporate dividend policy? Evidence from Asian Markets. Bulletin of Economic Research, 74(2), 579–598. https://doi.org/10.1111/boer.12310 Athari, S. A., Adaoglu, C., & Bektas, E. (2016). Investor protection and dividend policy: The case of Islamic and conventional banks. Emerging Markets Review, 27, 100–117. https://doi.org/10.1016/j.ememar.2016.04.001 Athari, S. A. (2022a). Examining the impacts of environmental characteristics on Shariah-Based Bank’s Capital Holdings: Role of country risk and governance quality. The Economics and Finance Letters, 9(1), 99–109. https://doi. org/10.18488/29.v9i1.3043 Athari, S. A., & Bahreini, M. (2021). The impact of external governance and regulatory settings on the profitability of Islamic banks. Evidence from Arab Markets. International Journal of Finance & Economics, 2021, 1–24. Bank of Uganda Supervision Report. (2022). Retrieved from https://www.bou.or.ug/bouwebsite/Supervision/. Bank of Uganda Annual Supervision Report. (2020). Retrieved March 27, 2020, from https://www.bou.or.ug/bou/ bouwebsite/bouwebsitecontent/Supervision/Annual_Supervision_Report/asr/AnnualSupervision-Report-2021.pdf. Bank of Uganda (BOU) website, (2020). https://www.google.com/url?sa=i&url=https%3A%2F%2Fwww.bou.or. ug%2Fbou%2Fbouwebsite%2FRelatedPages%2FPublications%2Farticle-v2%2F. Baker, T. A., Lopez, T. J., Reitenga, A. L., & Ruch, G. W. (2019). The influence of CEO and CFO power on accruals and real earnings management. Review of Quantitative Finance and Accounting, 52(1), 325–345. https://doi.org/10.1007/ s11156-018-0711-z COGENT BUSINESS & MANAGEMENT 23 Bartov, E., Gul, F. A., & Tsui, J. S. L. (2000). Discretionary accruals models and audit qualifications. Journal of Accounting and Economics, 30(3), 421–452. https://doi.org/10.1016/S0165-4101(01)00015-5 Bebchuk, L. A., Cremers, K. M., & Peyer, U. C. (2011). The CEO pay slice. Journal of Financial Economics, 102(1), 199– 221. https://doi.org/10.1016/j.jfineco.2011.05.006 Belal, A. R., & Owen, D. (2007). The views of corporate managers on the current state of, and future prospects for, social reporting in Bangladesh: An engagement-based study. Accounting, Auditing & Accountability Journal, 20(3), 472–494. https://doi.org/10.1108/09513570710748599 Bradburn, M. N., Sudman, S., & Wansik, B. (2004). Asking Questions: The Definitive guide to questionnaire design-for Market Research, Political polls and Social and Health Questionnaires. Revised ed, Jossey-Bass. Benkraiem, R., Ben Saad, I., & Lakhal, F. (2021). New insights into IFRS and earnings quality: What conclusions to draw from the French experience. Journal of Applied Accounting Research, 22(2), 307–333. https://doi.org/10.1108/ JAAR-05-2020-0094 Brahem, E., Depoers, F., & Lakhal, F. (2022). Corporate social responsibility and earnings quality in family firms. Journal of Applied Accounting Research, 23(5), 1114–1134. https://doi.org/10.1108/JAAR-05-2021-0139 Cai, Y., & Mehari, Y. (2015). The use of institutional theory in higher education research. Theory and Method in Higher Education Research (Theory and Method in Higher Education Research, Vol. 1, pp. 1–25), Emerald Group Publishing Limited. Choi, H. (2015). Agency Dilemma: A theory of shirking, embezzlement, and bribery prevention: A theory of out-sourcing and privatization: A theory of cooperation. https://doi.org/10.2139/ssrn.2618155 Cormier, D., Lapointe-Antunes, P., & Magnan, M. (2016). CEO Power and CEO hubris: A prelude to financial misreporting. Management Decision, 54(2), 522–554. https://doi.org/10.1108/MD-04-2015-0122 Davidson, R., Goodwin-Stewart, J., & Kent, P. (2005). Internal governance structures and earnings management. Accounting and Finance, 45(2), 241–267. https://doi.org/10.1111/j.1467-629x.2004.00132.x Dechow, P. M., & Schrand, C. M. (2004). Earnings quality. The Research Foundation of CFA Institute. https://doi. org/10.2469/cp.v2004.n3.3394 Dechow, P., Ge, W., & Schrand, C. (2010). Understanding earnings quality: A review of the proxies, their determinants and their consequences. Journal of Accounting and Economics, 50(2–3), 344–401. https://doi.org/10.1016/j.jacceco.2010.09.001 Dechow, P. M., & Dichev, I. D. (2002). The quality of accruals and earnings: The role of accrual estimation errors. The Accounting Review, 77(s-1), 35–59. https://doi.org/10.2308/accr.2002.77.s-1.35 Deegan, C. (2014). Financial accounting theory (4th Ed.). McGraw Hill. Demerjian, P. R., Lev, B., Lewis, M. F., & McVay, S. E. (2013). Managerial ability and earnings quality. The Accounting Review, 88(2), 463–498. https://doi.org/10.2308/accr-50318 Dillman, D. A. (1991). The design and administration of surveys. Annual Review of Sociology, 17(1), 225–249. https:// doi.org/10.1146/annurev.so.17.080191.001301 Duong, L., Evans, J., & Truong, T. P. (2020). Getting CFO on board – Its impact on firm performance and earnings quality. Accounting Research Journal, 33(2), 435–454. https://doi.org/10.1108/ARJ-10-2018-0185 Dutillieux, W., Francis, R. J., & Willekens, M. (2016). The spillover of SOX on earnings quality in Non-U.S. Jurisdictions. Accounting Horizons, 30(1), 23–39. https://doi.org/10.2308/acch-51241 Ege, M. S. (2015). Does internal audit function quality deter management misconduct? The Accounting Review, 90(2), 495–527. https://doi.org/10.2308/accr-50871 Egbunike, C. F., & Odum, A. N. (2018). Board leadership structure and earnings quality: Evidence from quoted manufacturing firms in Nigeria. Asian Journal of Accounting Research, 3(1), 82–111. https://doi.org/10.1108/AJAR-05-2018-0002 Electricity Act. (2022). Retrieved December 2, 2022, from https://www.parliament.go.ug/cmis/views/55f31964-40644a6a-97bd-adeae2d195cb%253B1.0.pdf. Endaya, K.A., & Hanefah, M.M. (2016). Internal auditor characteristics, internal audit effectiveness, and moderating effect of senior management. Journal of Economic and Administrative Sciences, 32(2), 160–176. https://doi. org/10.1108/JEAS-07-2015-0023. Elzahaby, A. M. (2021). How firm Performance mediates the relationship between corporate governance quality and earnings quality. Journal of Accounting in Emerging Economies, 11(2), 278–311. https://doi.org/10.1108/JAEE-09-2018-0100 Faricha, F. (2017). Relationship of earnings management and earnings quality before and after IFRS Implementation in Indonesia. European Research Studies Journal, XX(4B), 70–81. Feng, M., Ge, W., Luo, S., & Shevlin, T. (2011). Why do CFOs become involved in material accounting manipulations? Journal of Accounting and Economics, 51(1–2), 21–36. https://doi.org/10.1016/j.jacceco.2010.09.005 Field, A. (2013). Discovering statistics (4th ed.). SAGE Publications Ltd. Finkelstein, S. (1992). Power in top management teams: Dimensions, measurement, and validation. Academy of Management Journal, 35(3), 505–538. https://doi.org/10.2307/256485 Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.1177/002224378101800104 Francis, J., Huang, A. H., Rajgopal, S., & Zang, A. Y. (2008). CEO reputation and earnings quality. Contemporary Accounting Research, 25(1), 109–147., https://doi.org/10.1506/car.25.1.4 Francis, J., Olsson, P., & Schipper, K. (2006). Earnings Quality. Foundations and Trends® in Accounting, 1(4), 259–340. https://doi.org/10.1561/1400000004 Freeman, R. (1984). Strategic management: A stakeholder approach. Pitman.