Managerial ownership,audit committees and non-audit services

Abstract

We extend prior research on corporate governance and non-audit services (NAS) by distinguishing between the management entrenchment region of management ownership and regions in which the interests of management and shareholders converge. Management entrenchment and convergence-of-interests regions were estimated for our sample of 5198 Australian firm-years. NAS is negatively (positively) associated with management ownership in the convergence-of-interests (entrenchment) regions. However, using a sub-sample based on firms that have audit committees, we find the association between NAS and management ownership is confined to the entrenchment region, while audit committee strength is negatively associated with NAS fees across all management ownership regions.

JEL Classification: M42, M48

Keywords

Agency theory audit committees corporate governance managerial ownership non-audit service fees

1. Introduction

This article examines the association between non-audit services (NAS), audit committees and managerial ownership. NAS include various forms of consulting and advisory services, such as taxation advice, investment appraisal, compensation advice and information systems design, which contribute to the efficiency and performance of a company. However, there is widespread belief that the provision of NAS by the company’s audit firm undermines auditor independence (Hay et al., 2006b). Arguably, NAS can undermine auditors’ independence by making them reluctant to raise issues concerning their client’s financial statements or management’s questionable accounting policy choices due to the risk of compromising future income from NAS (Hay et al., 2006a; Levitt, 2000; Wu et al., 2016). Prior empirical investigations of the influence of providing NAS on auditor independence, using earnings quality and audit going concern qualified opinions as surrogates for independence, yield mixed results (Wu et al., 2016).¹ The provision of NAS by the company’s audit firm continues to be controversial and has attracted increased regulation (e.g. Sarbanes–Oxley Act in the United States, CLERP 9 in Australia, and Audit Regulation in the European Union (EU)) on the grounds that it may increase audit firms’ financial reliance on their clients and threaten auditor independence (Ratzinger-Sakel and Schönberger, 2015).

The ongoing concerns of regulators and policymakers about the purchase of NAS from the incumbent audit firm, and the potential threat to auditor independence, underpin the importance of identifying factors that drive or reduce demand for NAS (De Fuentes and Porcuna, 2016). In their meta-analysis on the determinants of NAS provided by the company’s audit firm, De Fuentes and Porcuna (2016: 4779) identify several gaps, including the need for further research on the role of corporate governance and ownership structures.

The audit committee is a key corporate governance mechanism (CGM) that can enhance the monitoring of management and protect shareholders by ensuring and maintaining corporate reporting quality and compliance with mandatory disclosure requirements (Davidson et al., 2005; Kent and Stewart, 2008). The audit committee is a particularly relevant CGM in light of the involvement of the audit committee in the NAS purchase decision (Wu et al., 2016).² Research into the relationship between audit committee strength and the purchase of NAS from the incumbent audit firm has produced mixed results.³ Furthermore, as these studies are based in the United States (Abbott et al., 2011; Lee and Mande, 2005) and the United Kingdom (Zaman et al., 2011), their findings are not necessarily generalisable to other contexts suggesting the need for further research in this area. Indeed, corporate governance requirements are characterised by significant variation across jurisdictions (Collier and Zaman, 2005; Khan et al., 2014; KPMG_ACCA, 2014; Monem, 2013; Shan et al., 2019).

Management are also key players in the decision to purchase NAS. Accordingly, a contribution of this study is to extend the literature by investigating how management ownership is associated with the purchase of NAS from the company’s audit firm. Prior studies investigating the role of ownership structure on NAS have focused on institutional ownership (Lee and Mande, 2005; Mitra and Hossain, 2007; Whisenant et al., 2003) or differences in legal form, such as mutual and proprietary insurers (O’Sullivan and Diacon, 2002).

Executive share ownership is a means of aligning the interests of management with those of shareholders, and thus reducing agency costs (Campbell et al., 2012; Jensen and Meckling, 1976). This occurs because managers with larger shareholdings have greater incentives to act in line with shareholders’ interests (Khan et al., 2014; Shuto and Takada, 2010). However, at higher levels of ownership management may become entrenched and have greater scope for pursuing their private interests, and thus increasing agency costs (Berger et al., 1997; Khan et al., 2014; Morck et al., 1988; Shuto and Takada, 2010). The few studies that include management ownership variables in modelling NAS adopt the view that agency costs decrease as management ownership increases (Abbott et al., 2003, 2011; Prakash and Venable, 1993). The differing implications of alignment and entrenchment (as per convergence-of-interests and entrenchment hypotheses) for agency costs provide a framework for a nuanced investigation of the association between management ownership and the purchase of NAS from the company’s audit firm. To the best of our knowledge, no prior studies distinguish between the convergence-of-interest hypothesis and the entrenchment hypothesis in explaining the association between management ownership and NAS.

Australia offers a suitable context in which to investigate the provision of NAS because the Corporate Law Economic Reform Program (Audit Reform and Corporate Disclosure) Act (CLERP 9) permits the provision of NAS by the company’s audit firm, provided they do not compromise auditor independence. In contrast, the Sarbanes–Oxley Act (SOX, 2002) in the United States prohibits the provision of most forms of NAS by the corporation’s audit firm. Similarly, the EU Audit Regulation 537/2014 prohibits certain forms of NAS provided by the incumbent audit firm, while permitting other forms subject to approval of the audit committee and assessing the threats to independence and safeguards applied (Articles 5), and imposes a cap on NAS fees as a function of audit fees (Article 4) (Ratzinger-Sakel and Schönberger, 2015).

The Australian setting is an ideal context for pursuing our investigation for several other reasons. Importantly, the Australian setting is characterised by key differences relative to other countries including the United States and the United Kingdom which are the subject of most extant research in this area. Specifically, there is agreement that agency costs of listed firms are higher in Australia than in other western jurisdictions including the United States and the United Kingdom (Miglani et al., 2015, 2020). For example, while Australia has implemented strong shareholder protection legislation, Australian firms feature greater levels of ownership concentration among non-institutional blockholders who tend to be relatively passive in monitoring management (Khan et al., 2014; La Porta et al., 1999). However, there are relatively lower levels of ownership of institutional investors, particularly in larger listed firms represented in market or sector indices, and also lower levels of corporate debt, at least in part due to Australia’s full dividend imputation taxation system (Hsu and Koh, 2005). In addition, there is limited analyst following, and greater non-coverage of many listed firms in Australia (Henry, 2010). These conditions suggest that external monitoring mechanisms are likely to be weaker in Australia (Henry, 2010). Also, market disciplining of management tends to be weaker in Australia because the economic environment is less competitive and there is a lower level of hostile takeovers than other countries (Dignam and Galanis, 2004). In addition, Australian executives are highly paid which means that alignment might entail greater levels of managerial ownership than in other countries (Khan et al., 2014). The smaller population in Australia also limits the available selection pool of director and managerial labour (Henry, 2010; Monem, 2013). Taken together, these differences mean that traditional external monitoring mechanism is likely to be weaker in Australia (Henry, 2010). These conditions shape the scope of Australian managers’ discretion for making decisions (including NAS purchase decisions), and in turn, agency cost implications in ways that are different to their counterparts’ in other countries (for detailed discussions, see, e.g. Dignam, 2008; Dignam and Galanis, 2004; Khan et al., 2014; Monem, 2013; Shan et al., 2019).

Our investigation is based on firm-year observations for companies listed on the ASX between 2005 and 2015. Using methods introduced by Morck et al. (1988), we identify the entrenchment region of management ownership ranging from 21% to 50% and convergence regions of managerial ownership as below 21% and above 50%.

We find that the association between management ownership and NAS fees is positive within the entrenchment region, and negative within the convergence-of-interests regions of management ownership for the full sample. A sub-sample comprising only those firm years for which the firm has an audit committee is used to examine both management ownership and audit committee strength. While we find a positive association between management ownership and NAS fees in the entrenchment region of management ownership, we find no significant association between management ownership and NAS fees in either of the convergence-of-interests regions. Furthermore, we find that audit committee strength is negatively associated with NAS fees irrespective of the region of management ownership. Our findings suggest that stronger audit committees discourage the purchase of NAS from the incumbent audit firm. Overall, our findings support a regulatory initiative mandating involvement of the audit committee in the decision to purchase NAS.

Although these findings are consistent with those of Abbott et al. (2003), they extend literature by examining the association of audit committee strength and NAS fees under alignment and entrenchment conditions.

This study makes three contributions. First, it enhances understanding of the role of management ownership, a key element in the broader corporate governance mosaic (Cohen et al., 2004; Gedajlovic and Shapiro, 1998), in the decision to purchase NAS from the company’s audit firm. Specifically, we provide evidence of the importance of distinguishing between the convergence-of-interests and entrenchment regions of management ownership to obtain a nuanced understanding of the relationship between management ownership and NAS. Thus, we extend the extant literature on the implications of alignment and entrenchment⁴ to the association between management ownership and NAS.

Second, to the best of our knowledge, our study is the first to specifically examine the association of key CGMs with NAS fees in the Australian setting. This is important because each country uniquely features ‘a particular nexus of constraints’ (Gedajlovic and Shapiro, 1998: 536) on the discretion of management to make key strategic decisions.

Finally, this study appears to be one of the first to estimate specific convergence-of-interests and entrenchment regions of managerial ownership in an Australian setting. We find that the turning points of the entrenchment region of management ownership are higher than those reported in the literature using US data (Morck et al., 1988) and UK data (Short and Keasey, 1999). This demonstrates the inappropriateness of using estimates based on US or UK samples when undertaking research on the convergence-of-interests and entrenchment hypotheses in the Australian context.

This article is organised as follows. The next section reviews the literature and develops the hypotheses, followed by the description of the research design and data collection. The fourth section discusses the empirical results, followed by concluding comments.

2. Literature and hypotheses

The separation of ownership and control gives rise to agency problems between principals (shareholders) and their agents (managers) who exercise control over the resources of the firm (Fama, 1980; Jensen and Meckling, 1976; Jensen, 1986). Management attempt to mitigate the associated agency costs by establishing ‘a series of mechanisms that seek to reconcile the interests of shareholders and managers’ (O’Sullivan, 2000: 399). These mechanisms include external (or statutory) auditing where independent auditors enhance the credibility of a company’s financial statements (Agarwal and Chadha, 2005; Desender et al., 2013, 2016). Other mechanisms pertain to the ways in which companies are directed and controlled (Larcker and Richardson, 2004; O’Sullivan, 2000), including incentive effects of executive share ownership (Jensen and Meckling, 1976), and the review of financial reporting and the engagement of the external auditor by the audit committee (Krishnan, 2005; Krishnan et al., 2011). Together, these mechanisms constitute key instruments for protecting the interests of shareholders (Desender et al., 2013).

Agency theory suggests that executive share ownership enhances the alignment of the managers’ interests with those of shareholders. Alignment means that managers have greater incentives to act in line with shareholders’ interests and fewer incentives to engage in opportunistic behaviours that lower corporate value, as they would then bear a larger share of the costs of such behaviours (Jensen and Meckling, 1976). This situation is described by the convergence-of-interests hypothesis. Accordingly, agency costs decrease as managers’ ownership in the company increases (Chau and Leung, 2006).

Management entrenchment is defined as ‘the extent to which managers fail to experience discipline from the full range of corporate governance and control mechanisms, including monitoring by the board, the threat of dismissal or takeover, and stock- or compensation-based performance incentives’ (Berger et al., 1997: 1411).

Through ownership of a substantial proportion of a company’s equity and associated voting power, and other entrenchment conditions such as status as a founder, or control of the board of directors, management may be able to exercise influence ‘to guarantee . . . employment with the firm at an attractive salary’ (Morck et al., 1988: 294). Entrenched managers may hold sufficient voting power or control of the board to protect their positions such that they become insulated from disciplinary action from the board or markets,⁵ even when their performance is below acceptable levels (Berger et al., 1997). Increased management ownership facilitates deeper entrenchment which can create greater scope for managers to pursue their own private interests at the expense of the shareholders, and increase agency costs as a result (Morck et al., 1988; Shleifer and Vishny, 1997).⁶

Consideration of management alignment and entrenchment is important as it facilitates a nuanced understanding of the implications of management ownership for management incentives and behaviour (Abbott et al., 2003; Fan and Wong, 2002). While the alignment incentives operate throughout the ownership range (Morck et al., 1988; Shuto and Takada, 2010), they could be outweighed by entrenchment implications within a specific region of ownership, where the benefits that managers expect to receive by pursuing private interests are valued more than those expected as a result of engaging in behaviours that maximise firm value.

Extant research has found that the opposing implications of alignment and entrenchment result in a curvilinear relationship between management ownership and firm value (Morck et al., 1988; Navissi and Naiker, 2006; Short and Keasey, 1999). This is reflected in an inverse U-shaped relationship between management shareholding ownership and firm value observed by Morck et al. (1988).⁷ Specifically, management ownership is positively associated with firm value until management become entrenched, and again at higher levels of management ownership, as managers’ interests become aligned with owners’ interests when share ownership is more heavily weighted in the managers’ wealth function. Management shareholding is negatively associated with firm value in the entrenchment region of managerial ownership.

A key explanation of the relationships between management ownership and firm value concerns the differences of underlying agency costs between firms with entrenched managers and those for which management’s interests are aligned with those of shareholders (Morck et al., 1988; Navissi and Naiker, 2006; Short and Keasey, 1999). The differing agency costs associated with alignment and entrenchment may also shape the manner in which management ownership is associated with decisions of companies to purchase NAS from their audit firm.

Managers in management-controlled firms have incentives to capture the benefits of the purchase of NAS from the incumbent audit firm (Prakash and Venable, 1993). However, the purchase of NAS is associated with perceptions of diminished audit independence. Consideration of the trade-off between the benefits of the purchase of NAS and its perceived implications on audit independence is likely to differ between entrenched managers and managers whose interests are aligned with those of shareholders.

Specifically, for managers whose interests are aligned with those of the shareholders’, the costs associated with perceptions of impairment of audit independence (which is associated with purchase of NAS) are likely to be of greater concern than the benefits of NAS. These managers are more likely to make or influence decisions that mitigate these costs. Consequently, they are likely to have less demand for NAS from the incumbent audit firm.

In contrast, entrenched managers have both greater control of the firm and scope for making decisions that enhance their own private interests (Shuto and Takada, 2010). Consequently, entrenched managers are expected to seek benefits by taking advantage of the audit firm’s expertise, and engage them to provide NAS. Implications associated with perceived impairment of audit independence due to NAS purchase decisions are likely to be inconsequential for entrenched managers since they are protected from possible disciplinary action of the board of directors or markets. The purchase of NAS manifests in NAS fees. Thus, we expect management ownership to be negatively associated with NAS fees within the convergence-of-interests regions of ownership but positively associated with NAS fees in the entrenchment region of management ownership. Accordingly, we hypothesise as below:

Hypothesis 1a. Management ownership is negatively associated with NAS fees when management ownership is in the low convergence-of-interests region of ownership.

Hypothesis 1b. Management ownership is positively associated with NAS fees when management ownership is in the entrenchment region of ownership.

Hypothesis 1c. Management ownership is negatively associated with NAS fees when management ownership is in the high convergence-of-interests region of ownership.

A key objective of audit committees is to protect the interests of shareholders by enhancing the quality of financial statements.⁸ Audit committees seek to ensure the objectivity and independence of the external audit process by monitoring and mitigating management pressure on external auditors. To discharge their mandate, a key role of audit committees is to make decisions or advise directors concerning the purchase of NAS (Beasley et al., 2009; Krishnan, 2005; PWC, 2016; Wu et al., 2016). To fulfil this role, an audit committee is likely to scrutinise decisions concerning the purchase of NAS and implications on auditor independence and shareholder interests (Abbott et al., 2003; Gaynor et al., 2006; Zaman et al., 2011). Specifically, an audit committee may be reluctant to support the provision of NAS by the incumbent audit firm in light of perceptions that such action may compromise the auditor’s independence (Abbott et al., 2003; Gaynor et al., 2006; Hay et al., 2006a).

Furthermore, as noted above, in Australia the audit committee is required to advise the board of directors on whether the purchase of NAS would compromise auditor independence (Corporations Act, s.300(11D)(a)). The directors’ statement in the annual report as to whether they are satisfied that the provision of NAS does not compromise auditor independence, and reasons for being so satisfied, must be made in accordance with the advice of the audit committee if the company has an audit committee (Corporations Act, s.300(11B)(b, c),(11D)(a)). The potential exposure of members of the audit committee to litigation and reputation risks may be a further source of reluctance to support the purchase of NAS from a company’s audit firm (Abbott et al., 2003, 2011; Prakash and Venable, 1993).

In Australia, the ASX Corporate Governance Council recommends that all listed companies have an audit committee and provides guidelines for audit committee structure and disclosures (ASXCGC, 2014, Recommendation 4.1(a)). While the Corporate Governance Principles and Recommendations Guidelines require listed companies to comply or to explain non-compliance, the ASX Listing Rules mandate audit committees for all companies included in the Standard & Poor’s (S&P) All Ordinaries Index (ASX Listing Rule 12.7). Furthermore, the audit committees of companies included in the S&P 300 Index must comply with ASXCGC Recommendation 4.1 (a).⁹

The ASX Corporate Governance Council recommends certain audit committee characteristics to provide formal and rigorous processes to safeguard the integrity of financial reporting (ASXCGC, 2014). Recommendation 4.1 (a) of the ASX Corporate Governance Principles and Recommendations states that the audit committee should have at least three members, all of whom are non-executive directors and a majority of whom are independent directors, and that it should be chaired by an independent director, who is not the chair of the board. The following disclosures are also recommended: the audit committee’s charter; the number of meetings of the audit committee held during the year and the attendance of individual members; and the relevant accounting and financial expertise of members of the audit committee (ASXCGC, 2014, Recommendation 4.1 (a)). These recommended characteristics are intended to provide formal and rigorous processes (ASXCGC, 2014), and are thus indicative of the strength of the audit committee.

Companies with stronger audit committees are more likely to discharge their responsibilities and monitor the external audit process more effectively than firms with weaker audit committees (Zaman et al., 2011). Stronger audit committees are thus more likely to use their mandatory advisory role to challenge the purchase of NAS from the incumbent auditor so as to signal higher quality auditing processes, irrespective of the level of management ownership.¹⁰ Accordingly, we hypothesise as below:

Hypothesis 2. Audit committee strength is negatively associated with NAS fees.

3. Research design

3.1. Data collection and sample specification

Corporate governance, accounting, and NAS fees data were obtained from Thomson Reuters DataStream. Hand-collected data include the number of ordinary shares owned by executive directors from the sample firms’ annual reports.

Our initial data set comprised firm-year observations from firms listed on the ASX between 2005 and 2015. Commencing with 21,988 observations, we removed 2959 observations of financial institutions, 3169 observations of foreign companies listed on the ASX,¹¹ 3116 observations of Australian American Depositary Receipts (Australian ADRs) trading on the US Exchanges and on the US Over-the-Counter (OTC) Markets,¹² and 7546 observations due to missing data (see Table 1).¹³ This process resulted in a final sample of 5198 firm-year observations. As shown in Table 1, the sample comprises 2408 firm-year observations from top 300 listed firms which are required to have an audit committee (as per ASX Listing Rule 12.7), 826 firm-year observations from the listed firms that are not in the top 300, but which have voluntarily established an audit committee, and 1964 firm-year observations for firms that are not in the top 300 and have elected not to have an audit committee.

Table 1.

Sample.

	No. of firm years
ASX listed companies (2005–2015)	21,988
Less: Observations for financial institutions	(2959)
Less: Observations for foreign firms listed on ASX	(3169)
Less: Observations for the Australian ADRs trading on the US Exchanges & the US OTC Markets	(3116)
Less: Observations due to missing data	(7546)
Total firm-year observations	5198
Comprising:
ASX top 300 listed firms required to establish an audit committee	2408
Other firms that voluntarily established an audit committee	826
Observations with an audit committee	3234
Observations for which there firm does not have an audit committee	1964
Management ownership breakdown:	No. of firm years
MOS_L	1585
MOS_M	1895
MOS_H	1718
Total firm-year observations	5198
Management ownership breakdown of sample with an audit committee:	No. of firm years
MOS_L	1141
MOS_M	1219
MOS_H	874
Total firm-year observations	3234

To facilitate testing Hypothesis 2, a sub-sample is identified comprising 3234 firm-year observations for listed firms that have an audit committee, irrespective of whether the audit committee was established voluntarily or to comply with ASX Listing Rule 12.7, that is, by virtue of a firm being in the top 300 (see Table 1).

Following Morck et al. (1988) and other related research (e.g. Khan et al., 2014; Lennox, 2005; Lin and Liu, 2013; Short and Keasey, 1999), we estimate a range of segmented and curvilinear regressions and identify turning points of the management ownership entrenchment region for the hypothesised relationships applicable to our sample of Australian listed firms. On this basis, we segment the sample into three subsets consisting of 1585, 1895 and 1718 observations for management ownership in ranges of less than 21% (MOS_L), between 21% and 50% (MOS_M), and greater than 50% (MOS_H), respectively (see Table 1). While these cut-offs differ from earlier estimates (see e.g. Lennox, 2005; Lin and Liu, 2013; Morck et al., 1988),¹⁴ they demonstrate a consistent pattern of turning points in which lower and higher convergence-of-interests regions are separated by an entrenchment region. The different turning points estimated in this study reflect contextual differences in share ownership and demonstrate the importance of distinguishing between jurisdictional settings in this area of corporate governance research.

3.2. Model specification

We use the following segmented regression models to empirically examine the hypotheses

\begin{array}{l} N A S F E E S = α + β_{1} M O S + γ_{1} L O S S + γ_{2} A U D I T O P I N + γ_{3} C O M P L E X \\ + γ_{4} R O A + γ_{5} L E V E R A G E + γ_{6} F I R M S I Z E + γ_{7} F C F O \\ + γ_{8} B I G 4 + \emptyset_{2005 - 2015} Y E A R + φ_{1 - 9} I N D U S T R Y + ε \end{array}

Model 1

We extend Model 1 to include audit committee strength (ACS) as follows

\begin{array}{l} N A S F E E S = α + β_{1} M O S + β_{2} A C S + γ_{1} L O S S + γ_{2} A U D I T O P I N \\ + γ_{3} C O M P L E X + γ_{4} R O A + γ_{5} L E V E R A G E + γ_{6} F I R M S I Z E \\ + γ_{7} F C F O + γ_{8} B I G 4 + \emptyset_{2005 - 2015} Y E A R + φ_{1 - 9} I N D U S T R Y + ε \end{array}

Model 2

After identifying the convergence-of-interests and entrenchment regions of management ownership applicable, following Morck et al. (1988), we regress Model 1 for the three regions to test Hypotheses 1a, 1b and 1c only, followed by Model 2 to examine all hypotheses. ACS is excluded from Model 1 to facilitate testing Hypotheses 1a, 1b, and 1c using the full sample of 5198 firm-year observations (see Section 4.1 and Table 1).

The definitions of variables used in the models are summarised in Table 2. These variables are discussed further in the sub-sections 3.3–3.5.

Table 2.

Definition of variables.

Variables	Description
Dependent variable:
NASFEES	Natural logarithm of 1 + non-audit service fees
Ownership variables:
MOS	Management ownership ratio; the ratio of shares owned by the executive directors to total outstanding shares
MOS_L	Low convergence-of-interests range of management ownership: 0 < MOS_L < 21%
MOS_M	Entrenchment range of management ownership: 21% ⩽ MOS_M ⩽ 50%
MOS_H	High convergence-of-interests range of management ownership: MOS_H > 50%
Audit committee strength variable:
ACS	ACS (audit committee strength) is coded 1 when all of the following conditions are met: (1) the committee has at least three members; (2) all audit committee members are non-executive directors, the majority of whom are independent directors; (3) the audit committee is chaired by an independent director, who is not the chair of the board; (4) audit committee members meet at least four times a year; and (5) at least one audit committee member has accounting and financial expertise; 0, otherwise.
Control variables:
LOSS	Coded 1 if net income is negative in either of the past two years (t – 2 or t – 1); 0 otherwise
AUDITOPIN	Auditor’s opinion, coded 1 if the firm received a qualified audit report, adverse opinion or disclaimer of opinion in the previous year (t – 1); 0, otherwise
COMPLEX	Business complexity, coded 1 if the firm has at least two geographical segments; 0, otherwise
ROA	Return on assets, measured as the ratio of income before extraordinary items to total assets
LEVERAGE	The ratio of total debt to total equity
FIRMSIZE	Measured as the natural logarithm of the book value of total assets
FCFO	The ratio of free cash flow (cash flow from operations – capital expenditure) to current assets
BIG4	Big 4 auditor, coded 1 if the firm is audited by a Big4 auditor; 0, otherwise
YEAR	Year dummy represents dummy variables that reflect the years between 2005 and 2015
INDUSTRY	Industry dummy reflects the two-digit Standard Industry Classification (SIC)

3.3. Dependent variable

Based on prior research on NAS and consulting services (see e.g. Grosse et al., 2020; Hay et al., 2006a; Srinidhi and Gul, 2007; Wu et al., 2016; Zaman et al., 2011), we define the dependent variable, NASFEES, as the natural logarithm of 1 + NAS fees.

3.4. Independent variables

Consistent with prior related research (see e.g. Campbell et al., 2012; Gotti et al., 2012; McKnight and Weir, 2009; Morck et al., 1988), management ownership (MOS) is defined as the proportion of ordinary shares owned by executive directors.

Similar to Zaman et al. (2011), we draw on the applicable corporate governance regulation in classifying audit committees. Accordingly, based on the audit committee requirements of the ASX Corporate Governance Council (ASXCGC, 2014), audit committee strength (ACS) is coded 1 in each firm year when the firm has an audit committee with all of the following characteristics: (1) the committee has at least three members; (2) all audit committee members are non-executive directors, and the majority are independent directors; (3) the audit committee is chaired by an independent director, who is not the chair of the board; (4) the audit committee members meet at least four times a year; and (5) at least one audit committee member has qualifications in accounting or finance.¹⁵ While the ASX Corporate Governance Principles and Recommendations suggest that the members of the audit committee ‘between them should have the accounting and financial expertise . . . to be able to discharge the committee’s mandate effectively’ (ASXCGC, 2014: 22), they do not elaborate on what constitutes sufficient accounting and financial expertise for this purpose. Accordingly, we adopted the SOX (2002) definition of financial and accounting expertise in defining condition (v) of ACS, that is, at least one audit committee member is a financial expert (SOX, 2002) and operationalise this as qualifications or experience in accounting or finance, consistent with other related audit committee research (see e.g. Azim, 2012; Zaman et al., 2011). ACS is coded 0 if the audit committee fails to meet one or more of the five criteria specified above for a given firm year.

3.5. Control variables

The control variables consist of firm’s operating loss (LOSS), auditor’s opinion (AUDITOPIN), business complexity (COMPLEX), firm performance (ROA), firm’s leverage (LEVERAGE), firm size (FIRMSIZE), free cash flow (FCFO), Big 4 auditor (BIG4), year (YEAR) and industry (INDUSTRY).

LOSS is coded 1 if net income is negative in either of the past 2 years (i.e. t – 2 or t – 1), 0 otherwise (Firth, 2002; Zaman et al., 2011). Loss-making firms may have less resources to engage external consultants. Alternatively, it has been suggested that poorly performing firms may be more likely to seek external advice to improve profitability (Zaman et al., 2011). However, loss-making firms are likely to have higher agency costs, which would make them more likely to restrict the purchase of consultancy services from their audit firm in light of perceptions that NAS threaten auditor independence (Firth, 1997, 2002; Prakash and Venable, 1993). Thus, a negative relationship between LOSS and NASFEES is expected.

Conversely, more profitable firms are more likely to purchase NAS because the ensuing lower agency costs may alleviate concerns about threats to auditor independence. Furthermore, more profitable firms have greater capacity to absorb consultancy fees. Profitability is measured as return on assets (ROA), that is, the ratio of income before extraordinary items to total assets (Gul and Tsui, 1997). Thus, a positive relationship between ROA and NASFEES is expected.

AUDITOPIN is coded 1 if the firm received a modified opinion in the previous year (t – 1), 0 otherwise. Modified opinion include qualified opinions, adverse opinions and disclaimers of opinion based on AUASB Standard ASA 705 Modifications to the Opinion in the Independent Auditor’s Reports (refer to paragraph 5(b)) (ASA705, 2015). Auditors’ propensity to issue a qualified report is used in the literature as a surrogate for audit independence (DeFond et al., 2002; Sharma and Sidhu, 2001). A recent history of having received a modified opinion may serve to mitigate concerns about the potential effect of NAS on auditor independence. Accordingly, we expect a positive association between AUDITOPIN and NASFEES.

Adapted from Lennox (2005), COMPLEX represents business complexity and is coded 1 if the firm has at least two geographical segments, 0 otherwise. We expect a positive coefficient for COMPLEX because firms with more complex international operations may have greater demand for professional services, such as taxation, finance, human resources and international business management (Gotti et al., 2012; Hay et al., 2006a; Zaman et al., 2011).

LEVERAGE is measured as the ratio of total debt to total equity. Firms with higher leverage may require additional monitoring to enhance protection from business and financial risk (Zaman et al., 2011) and are thus likely to have more demand for certain NAS, particularly those related to financial management and investment appraisal. Accordingly, we expect a positive association between LEVERAGE and NAS.

FIRMSIZE is measured as the natural logarithm of the book value of total assets (Desender et al., 2016). Size reflects both the magnitude and complexity of the firm’s operations. FIRMSIZE is likely to be positively associated with NAS fees because larger firms are likely to have greater expenditure for NAS (Chan et al., 2012; Hay et al., 2006a; Zaman et al., 2011).

Free cash flow (FCFO) is defined as cash flow from operations minus capital expenditure divided by current assets (Lennox, 2005). Firms with more free cash flow have a greater capacity to engage consultants, including NAS. However, more free cash flow suggests greater managerial discretion and higher agency costs (McKnight and Weir, 2009). Higher agency costs are likely to reduce demand for NAS due to increased sensitivity to the potential threat to auditor independence. Thus, FCFO is expected to be negatively associated with NAS.

BIG4 is coded 1 if the company is audited by a big four auditor, 0 otherwise (Chan et al., 2012; Garg, 2017; Gotti et al., 2012; Zaman et al., 2011). Big four audit firms have greater capacity than smaller audit firms to provide NAS. Thus, BIG4 is expected to be positively associated with NAS fees.

Finally, we control for industry effects and variation over time on the demand for NAS (Chan et al., 2012; Zaman et al., 2011). YEAR is a dummy variable that reflects the year. INDUSTRY is a dummy variable reflecting the two-digit Standard Industry Classification (SIC) code.

4. Results and discussion

4.1. Descriptive statistics

The descriptive statistics for the full sample and the sub-sample with audit committees are presented in Table 3. As shown in Panel A, the means (medians) of NASFEES ($’000) are $116.17 ($106.46), $112.71 ($102.93), $109.80 ($95.00), and $127.36 ($121.01) for the full sample and the MOS regions, namely, 0 < MOS_L < 21%, 21% ⩽ MOS_M ⩽ 50% and MOS_H > 50%, respectively. The means (medians) of MOS are 0.38 (0.36), 0.08 (0.08), 0.36 (0.36) and 0.68 (0.65) for the full sample and each MOS region, respectively.

Table 3.

Descriptive statistics.

Panel A: Full sample
Variables	Full sample							MOS_L (0%–21%)
Variables	Mean	P25	Median	P75	Min	Max	SD	Mean	P25	Median	P75	Min	Max	SD
NASFEES($’000)	116.17	46.58	106.46	283.00	12.50	1184.11	1.22	112.71	42.00	102.93	297.00	12.50	1184.11	1.30
MOS	0.38	0.16	0.36	0.57	0.003	0.88	0.25	0.08	0.01	0.08	0.14	0.003	0.21	0.07
LOSS	0.48	0.00	0.00	1.00	0.00	1.00	0.50	0.46	0.00	0.00	1.00	0.00	1.00	0.50
AUDITOPIN	0.90	1.00	1.00	1.00	0.00	1.00	0.31	0.90	1.00	1.00	1.00	0.00	1.00	0.31
COMPLEX	0.52	0.00	1.00	1.00	0.00	1.00	0.50	0.51	0.00	0.00	1.00	0.00	1.00	0.50
ROA	−0.16	−0.20	0.01	0.07	−1.92	0.19	0.44	−0.16	−0.20	0.01	0.07	−1.92	0.19	0.44
LEVERAGE	0.33	0.00	0.10	0.48	0.00	1.66	0.47	0.35	0.00	0.11	0.51	0.00	1.66	0.48
FIRMSIZE	11.38	9.73	11.25	13.11	6.75	15.00	2.24	11.92	9.98	12.25	13.94	6.75	15.00	2.42
FCFO	−2.42	−2.75	−0.89	−0.21	−14.13	0.21	3.70	−2.55	−2.82	−0.97	−0.22	−14.13	0.21	3.88
BIG4	0.58	0.00	1.00	1.00	0.00	1.00	0.49	0.67	0.00	1.00	1.00	0.00	1.00	0.47
	MOS_M (21–50%)							MOS_H (>50%)
NASFEES($’000)	109.80	44.40	95.00	269.01	12.50	1184.11	1.23	127.36	55.53	121.01	284.99	12.50	1184.11	1.13
MOS	0.36	0.28	0.36	0.43	0.21	0.50	0.08	0.68	0.57	0.65	0.78	0.50	0.88	0.12
LOSS	0.53	0.00	1.00	1.00	0.00	1.00	0.50	0.45	0.00	0.00	1.00	0.00	1.00	0.50
AUDITOPIN	0.90	1.00	1.00	1.00	0.00	1.00	0.30	0.89	1.00	1.00	1.00	0.00	1.00	0.31
COMPLEX	0.54	0.00	1.00	1.00	0.00	1.00	0.50	0.51	0.00	1.00	1.00	0.00	1.00	0.50
ROA	−0.18	−0.24	−0.02	0.07	−1.92	0.19	0.46	−0.14	−0.17	0.01	0.07	−1.92	0.19	0.44
LEVERAGE	0.30	0.00	0.06	0.43	0.00	1.66	0.44	0.34	0.00	0.11	0.50	0.00	1.66	0.48
FIRMSIZE	11.22	9.68	11.08	12.76	6.75	15.00	2.15	11.05	9.63	10.78	12.35	6.75	15.00	2.06
FCFO	−2.27	−2.55	−0.90	−0.23	−14.13	0.21	3.44	−2.46	−2.84	−0.79	−0.16	−14.13	0.21	3.80
BIG4	0.58	0.00	1.00	1.00	0.00	1.00	0.49	0.51	0.00	1.00	1.00	0.00	1.00	0.50
Panel B: Sample with audit committees
	Sample with audit committees							MOS_L (0%–21%)
NASFEES($’000)	121.37	42.00	124.84	341.09	12.50	1184.11	1.41	129.39	42.00	140.17	388.02	12.50	1184.11	1.43
MOS	0.28	0.07	0.25	0.45	0.003	0.88	0.22	0.07	0.01	0.05	0.12	0.003	0.21	0.07
LOSS	0.34	0.00	0.00	1.00	0.00	1.00	0.47	0.32	0.00	0.00	1.00	0.00	1.00	0.47
AUDITOPIN	0.96	1.00	1.00	1.00	0.00	1.00	0.20	0.96	1.00	1.00	1.00	0.00	1.00	0.19
COMPLEX	0.62	0.00	1.00	1.00	0.00	1.00	0.49	0.59	0.00	1.00	1.00	0.00	1.00	0.49
ROA	−0.05	−0.06	0.03	0.08	−1.92	0.19	0.31	−0.06	−0.04	0.03	0.07	−1.92	0.19	0.30
LEVERAGE	0.43	0.01	0.32	0.1	0.00	1.66	0.48	0.45	0.02	0.36	0.63	0.00	1.66	0.49
FIRMSIZE	13.03	11.92	13.48	15.00	6.75	15.00	1.99	13.27	12.28	13.71	15.00	6.75	15.00	1.91
FCFO	−2.48	−2.87	−1.02	−0.19	−14.13	0.21	3.67	−2.69	−3.07	−1.05	−0.21	−14.13	0.21	3.99
BIG4	0.81	1.00	1.00	1.00	0.00	1.00	0.39	0.83	1.00	1.00	1.00	0.00	1.00	0.38
	MOS_M (21%–50%)							MOS_H (>50%)
NASFEES($’000)	107.29	37.00	100.00	272.97	12.50	1184.11	1.39	130.90	51.91	123.43	396.49	12.50	1184.11	1.37
MOS	0.35	0.28	0.35	0.42	0.21	0.50	0.08	0.61	0.54	0.60	0.66	0.50	0.88	0.10
LOSS	0.37	0.00	0.00	1.00	0.00	1.00	0.48	0.31	0.00	0.00	1.00	0.00	1.00	0.46
AUDITOPIN	0.95	1.00	1.00	1.00	0.00	1.00	0.22	0.96	1.00	1.00	1.00	0.00	1.00	0.20
COMPLEX	0.64	0.00	1.00	1.00	0.00	1.00	0.48	0.66	0.00	1.00	1.00	0.00	1.00	0.48
ROA	−0.07	−0.08	0.03	0.08	−1.92	0.19	0.33	−0.02	−0.03	0.04	1.00	−1.78	0.19	0.28
LEVERAGE	0.43	0.01	0.31	0.60	0.00	1.66	0.48	0.41	0.00	0.27	0.61	0.00	1.66	0.47
FIRMSIZE	12.76	11.48	12.97	14.77	6.75	15.00	1.99	12.95	11.76	13.59	15.00	6.80	15.00	2.11
FCFO	−2.08	−2.43	−0.88	−0.17	−14.13	0.21	3.10	−2.71	−3.35	−1.13	−0.15	−14.13	0.21	3.78
BIG4	0.79	1.00	1.00	1.00	0.00	1.00	0.41	0.80	1.00	1.00	1.00	0.00	1.00	0.40

All variables are defined in Table 2.

As shown in Panel B of Table 3, the means (medians) of NASFEES ($’000) are $121.37 ($124.84), $129.39 ($140.17), $107.29 ($100.00) and $130.90 ($123.43) for the sub-sample with audit committees and the MOS regions, namely, 0 < MOS_L < 21%, 21% ⩽ MOS_M ⩽ 50% and MOS_H > 50%, respectively. The means (medians) of MOS are 0.28 (0.25), 0.07 (0.05), 0.35 (0.35) and 0.61 (0.60) for the sub-sample with audit committees and each MOS region, respectively. The control variables show a similar pattern, in which the mean values are not distant from the respective median values. This suggests that our data are normally distributed overall.

4.2. Regression results

To examine regression estimation without bias, we winsorised all continuous variables at the 1st and 99th percentiles to avoid the likely effects of extreme outliers and then conducted multicollinearity diagnostics. The Pearson correlation coefficients are shown in Table 4. All correlation values of independent variables are well below the critical value of 0.8. Following Gujarati (2003), we also conduct the variance inflation factor (VIF) test. The results, which are not presented in the interest of brevity, indicate that the largest VIF value is 1.62, which is well below the critical VIF value of 10. Therefore, we conclude that multicollinearity is not strongly influencing the computational accuracy of our results.

Table 4.

Multicollinearity diagnostics: Pearson’s rank correlation matrix.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
(1) MOS	−
(2) ACS	−0.22***	−
(3) LOSS	0.01	−0.27***	−
(4) AUDITOPIN	−0.01	0.16***	−0.21***	−
(5) COMPLEX	−0.004	0.18***	−0.05***	0.01	−
(6) ROA	−0.02*	0.22***	−0.52***	0.3***	0.05***	−
(7) LEVERAGE	−0.01	0.22***	−0.25***	0.04***	0.07***	0.16***	−
(8) FIRMSIZE	−0.17***	0.57***	−0.56***	0.29***	0.17***	0.54***	0.42***	−
(9) FCFO	−0.01	−0.03*	−0.05***	0.02*	0.08***	0.03**	0.02*	0.001	−
(10) BIG4	−0.15***	0.37***	−0.31***	0.18***	0.13***	0.23***	0.24***	0.53***	0.03**	−

We also conduct variance inflation factor (VIF) test as an alternative method to investigate the likelihood of multicollinearity, and find that the highest VIF value is 1.62, which is well below the critical value of 10 (Gujarati, 2003). All variables are defined in Table 2.

p < .05; **p < .01; ***p < .001. All tests are two-tailed.

The regression estimations for Models 1 and 2 are reported in Table 5.

Table 5.

Regression for testing hypotheses.

Independent variable	Full sample			Sample with audit committees – Model 1			Sample with audit committees – Model 2
Independent variable	MOS_L0%–21%	MOS_M21%–50%	MOS_H>50%	MOS_L0%–21%	MOS_M21%–50%	MOS_H>50%	MOS_L0%–21%	MOS_M21%–50%	MOS_H>50%
	Clmn (1)	Clmn (2)	Clmn (3)	Clmn (4)	Clmn (5)	Clmn (6)	Clmn (7)	Clmn (8)	Clmn (9)
Intercept	10.99*** (9.25)	10.74*** (5.09)	11.15*** (5.28)	8.41*** (10.94)	8.59*** (14.77)	9.42*** (9.06)	8.17*** (19.76)	8.27*** (25.84)	8.30*** (21.19)
Management ownership:
MOS	−1.37** (–3.17)	0.57* (1.93)	−0.56** (–2.83)	−0.89(–1.18)	1.68** (2.77)	−0.36(–0.46)	−0.77(–1.48)	0.78* (2.21)	−0.26(–0.88)
	β₁ < 0	β₁ > 0	β₁ < 0	β₁ < 0	β₁ > 0	β₁ < 0	β₁ < 0	β₁ > 0	β₁ < 0
	[H1a]	[H1b]	[H1c]
Audit committee strength:
ACS							−0.39*** (–4.45)	−0.62*** (–8.90)	−0.58*** (–6.86)
							β₂ < 0	β₂ < 0	β₂ < 0
							[H2]	[H2]	[H2]
LOSS	−0.29*** (–3.9)	−0.25* (–3.82)	−0.41*** (–6.58)	0.10(0.75)	0.18(1.28)	−0.12(–0.60)	0.02(0.17)	−0.03(–0.40)	−0.03(–0.37)
	γ₁ < 0	γ₁ < 0	γ₁ < 0	γ₁ < 0	γ₁ < 0	γ₁ < 0	γ₁ < 0	γ₁ < 0	γ₁ < 0
AUDITOPIN	0.20* (1.98)	0.37*** (4.42)	0.40*** (4.77)	−0.62* (–2.13)	−0.35(–1.40)	−0.46(–1.06)	−0.25^† (–1.81)	−0.03(–0.29)	−0.05(–0.44)
	γ₂ > 0	γ₂ > 0	γ₂ > 0	γ₂ > 0	γ₂ > 0	γ₂ > 0	γ₂ > 0	γ₂ > 0	γ₂ > 0
COMPLEX	0.37*** (6.18)	0.41*** (8.11)	0.48*** (9.93)	0.40*** (4.02)	0.26* (2.36)	0.36* (2.15)	0.29** (4.27)	0.27*** (4.45)	0.30*** (4.41)
	γ₃ > 0	γ₃ > 0	γ₃ > 0	γ₃ > 0	γ₃ > 0	γ₃ > 0	γ₃ > 0	γ₃ > 0	γ₃ > 0
ROA	0.18* (2.20)	0.11^† (1.73)	0.11^† (1.71)	−0.34(–1.58)	−0.22(–1.10)	−0.22(–0.62)	−0.30** (–2.68)	−0.34*** (–3.90)	−0.18^† (–1.82)
	γ₄ > 0	γ₄ > 0	γ₄ > 0	γ₄ > 0	γ₄ > 0	γ₄ > 0	γ₄ > 0	γ₄ > 0	γ₄ > 0
LEVERAGE	0.72*** (11.28)	0.56*** (9.55)	0.45*** (8.92)	0.34** (3.09)	0.20^† (1.68)	−0.19(–1.04)	0.33*** (4.21)	0.20** (2.78)	−0.002(–0.02)
	γ₅ > 0	γ₅ > 0	γ₅ > 0	γ₅ > 0	γ₅ > 0	γ₅ > 0	γ₅ > 0	γ₅ > 0	γ₅ > 0
FIRMSIZE	0.31*** (6.77)	0.31*** (8.36)	0.33*** (11.68)	0.28*** (6.69)	0.26*** (6.79)	0.24*** (3.69)	0.27*** (9.41)	0.27*** (11.55)	0.28*** (10.92)
	γ₆ > 0	γ₆ > 0	γ₆ > 0	γ₆ > 0	γ₆ > 0	γ₆ > 0	γ₆ > 0	γ₆ > 0	γ₆ > 0
FCFO	0.01(1.43)	−0.01(–0.72)	−0.01^† (–.73)	0.01(1.16)	0.01(0.65)	−0.01(–0.63)	0.01(1.16)	0.001(0.11)	−0.004(–0.47)
	γ₇ < 0	γ₇ < 0	γ₇ < 0	γ₇ < 0	γ₇ < 0	γ₇ < 0	γ₇ < 0	γ₇ < 0	γ₇ < 0
BIG4	0.13(1.51)	0.17* (2.11)	0.15(1.60)	0.35* (2.30)	−0.19(–1.30)	−0.19(–0.68)	027** (3.05)	0.20** (2.73)	0.18* (2.28)
	γ₈ > 0	γ₈ > 0	γ₈ > 0	γ₈ > 0	γ₈ > 0	γ₈ > 0	γ₈ > 0	γ₈ > 0	γ₈ > 0
YEAR	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
INDUSTRY	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Adjusted R²	0.2109	0.2661	0.2736	0.2840	0.3030	0.1673	0.3221	0.3441	0.3319
F-Statistic	21.71***	26.44***	24.96***	11.21***	9.78***	3.24***	19.05***	22.30***	15.46***
N	1,585	1,895	1,718	1,141	1,219	874	1,141	1,219	874

We find the low convergence-of-interest effect when management ownership is below 21% (MOS_L), the entrenchment effect when management ownership is between 21% and 50% (MOS_M), and the high convergence-of-interest effect when management ownership is above 50% (MOS_H). The first row (number) represents the estimated coefficient, the second row (number in parentheses) represents the t-value of significance and the third and fourth rows represent expected sign and hypothesis, respectively. All continuous variables are winsorised at the 1st and 99th percentiles. All variables are defined in Table 2.

†

p < .10; *p < .05; **p < .01; ***p < .001. All tests are two-tailed.

4.3. Model 1

The regression estimations for Model 1 for the partitions of the sample based on regions of management ownership, namely MOS_L, MOS_M and MOS_H, are reported in Columns (1)–(3), respectively. The adjusted R²s for Model 1 for MOS_L, MOS_M and MOS_H regions are 0.2109, 0.2661 and 0.2736, respectively. The F-statistics are statistically significant for the three regressions.

Hypothesis 1a and Hypothesis 1c predicted a negative association between management ownership and NAS fees within both the low and high convergence-of-interests regions of ownership, that is, at 0 < MOS_L < 21% (MOS_L) and MOS_H > 50% (MOS_H), respectively. The coefficient of MOS (β₁) is significantly negative (β₁ = –1.37, t = –3.17, p < 0.01), as shown in Column (1) of Table 5. This indicates that management ownership is inversely related to NASFEES in the low convergence-of-interests region of ownership. Thus, Hypothesis 1a is supported. Similarly, the coefficient of MOS (β₁) is significantly negative (β₁ = –0.56, t = –2.83, p < 0.01) in the MOS_H region, shown in Column (3). Thus Hypothesis 1c is also supported. These results provide tentative evidence that when the interests of management and shareholders are aligned through managerial shareholding, management are likely to value the benefits associated with perceptions of auditor independence more than the benefits of purchasing NAS from the incumbent audit firm, thereby reducing purchasing of NAS.

Hypothesis 1b predicts a positive association between NAS and management ownership in the entrenchment region, that is, MOS_M. As shown in Column (2) of Table 5, the MOS coefficient (β1) is significantly positive (β₁ = 0.57, t = 1.93, p < 0.05), indicating that higher levels of management shareholding are associated with higher levels of NASFEES within the entrenchment region. Thus, Hypothesis 1b is supported. This result provides tentative evidence that when management is entrenched, the benefits to management from purchasing NAS from the company’s audit firm are perceived to outweigh the associated costs of perceived threats to auditor independence.

Table 5 also presents the regression results for the control variables in Model 1. As expected, the coefficients of LOSS are significantly negative and the coefficients of AUDITOPIN, COMPLEX, ROA, LEVERAGE and FIRMSIZE are significantly positive in all three management ownership regions. Coefficients of FCFO and BIG4 are not significant in most regions of MOS.

For comparison, we also use Model 1 to regress the sub-sample based on firms with audit committees. The results are presented in Columns (4)–(6) of Table 5. Similar to the regression on the full sample, the coefficient of MOS is negative and significant in the entrenchment region of management ownership. However, the coefficients of MOS are not significant in the regression estimates for the convergence-of-interests regions.

4.4. Model 2

In Model 2, we introduce audit committee strength (ACS) as an independent variable to include examination of Hypothesis 2. As discussed in section 3, the measurement of audit committee strength is based on Corporate Governance Principles and Recommendations (ASXCGC, 2014). Hypothesis 2 is tested using a sub-sample of 3234 firm years in which the firm has an audit committee. The regression estimations for Model 2 are reported in Columns (7)–(9) of Table 5. The adjusted R²s for the three management ownership regions are 0.3221, 0.3441 and 0.3319, respectively. The F-statistics are statistically significant for all regressions.

As expected, there is a negative association between ACS and NASFEES. The ACS coefficients (β₂) are significantly negative (β₂ = –0.39, t = –4.45, p < 0.001; β₂ = –0.62, t = –8.90, p < 0.001; β₂ = –0.58, t = –6.86, p < 0.001) for the management ownership regions of MOS_L, MOS_M and MOS_H, respectively, as shown in Columns (7), (8) and (9) of Table 5. Thus, Hypothesis 2 is supported.

The coefficient of MOS is significant in the expected direction (β₁ = 0.78, t = 2.21, p < 0.05) for the entrenchment region of management ownership (MOS_M), consistent with Hypothesis 1b. However, the coefficients of MOS are not significant in the regression estimates for the convergence-of-interests regions of management ownership (MOS_L and MOS_H), as shown in Columns (7) and (9) of Table 5.

Our findings can be interpreted as evidence that while management ownership within the entrenchment range is associated with higher levels of NAS, strong audit committees constrain the purchase of NAS, irrespective of whether managerial ownership is in a convergence-of-interests or entrenchment region.

Table 5 also provides the regression results for the control variables for Model 2. As expected, the coefficients on COMPLEX, FIRMSIZE and BIG4 are significantly positive in all management ownership regions, while the coefficient on LEVERAGE is significantly positive in two of the three regions. Contrary to expectations, the coefficient on ROA is significantly negative in the three regions of management ownership. The coefficient for AUDITOPIN is insignificant in most regions and there is no evidence to suggest that LOSS and FCFO are associated with NAS.

4.5. Additional tests

We conduct a variety of sensitivity analyses to ensure the robustness of our primary results. First, using trial-and-error method we regress Model 1 to examine the influence of entrenchment on NASFEES by using a number of entrenchment regions as discussed in prior studies. For example, Morck et al. (1988) found that the entrenchment region is at 5%–25% of managerial ownership in their US sample. Using UK samples, Short and Keasey (1999) identify the entrenchment regions to be 12%–40% while Lennox (2005) tested a number of entrenchment regions including those at 10%–30% and 15%–40%. Lin and Liu (2013) identify the entrenchment region to be at 20%–50% in a sample of Hong Kong listed firms. In addition to testing of these regions, we also include other entrenchment regions in our tests such as those at 25%–50% and 5%–60%. Importantly, we also use the entrenchment regions of 21%–50% identified for our Australian sample. The key aim of regressing the hypothesised relationships across these regions is to seek further evidence supporting the robustness of our primary results for Hypotheses 1(a), 1(b) and 1(c).

The range of test results of the identified entrenchment regions and turning points are shown in Columns (1)–(3) of Table 6. Specifically, the identified entrenchment regions with significant turning points (at 5%–50%, 10%–50%, 15%–50%, 20%–50%) prompted further testing of other regions in a trial-and-error fashion. Our tests confirm that the most appropriate entrenchment region in the Australian sample is at 21%–50% with turning points at the significance level of 5%.

Table 6.

Regression for testing alternative entrenchment ranges.

Entrenchment		Full sample			Sample with audit committees
Entrenchment		MOS_L	MOS_M	MOS_H	MOS_L	ACS	MOS_M	ACS	MOS_H	ACS
		Clmn (1)	Clmn (2)	Clmn (3)	Clmn (4)	Clmn (5)	Clmn (6)	Clmn (7)	Clmn (8)	Clmn (9)
	Sign	β₁ < 0	β₁ > 0	β₁ < 0	β₁ < 0	β₂ < 0	β₁ > 0	β₂ < 0	β₁ < 0	β₂ < 0
5%–25%	β	−14.39***	−0.29	0.001	−5.88	−0.47**	−0.28	−0.43***	0.18	−0.58***
	t	(–4.13)	(–0.51)	(0.01)	(–1.39)	(–3.01)	(–0.43)	(–4.61)	(1.38)	(–10.37)
10%–30%		−4.54***	0.14	−0.15	−1.50	−0.54***	−0.39	−0.54***	0.01	−0.54***
		(–3.60)	(0.26)	(–1.42)	(–0.99)	(–4.24)	(–0.62)	(–6.16)	(0.04)	(–9.21)
10%–40%		−4.54***	0.73*	−0.26^†	−1.50	−0.54***	0.64*	−0.52***	−0.07	−0.56***
		(–3.60)	(2.51)	(–1.84)	(–0.99)	(–4.24)	(1.96)	(–7.65)	(–0.33)	(–7.96)
12%–40%		−3.34***	0.76*	−0.26^†	−1.01	−0.51***	0.71*	−0.53***	−0.07	−0.56***
		(–3.53)	(2.33)	(–1.84)	(–0.89)	(–4.36)	(1.94)	(–7.43)	(–0.33)	(–7.96)
15%–40%		−2.19**	0.98*	−0.26^†	−0.95	−0.43***	0.94*	−0.56***	−0.07	−0.56***
		(–3.08)	(2.58)	(–1.84)	(–1.14)	(–4.17)	(2.16)	(–7.53)	(–0.33)	(–7.96)
5%–50%		−14.39***	0.47	−0.56**	−5.88	−0.47**	0.43*	−0.53***	−0.26	−0.58***
		(–4.13)	(1.34)	(–2.83)	(–1.39)	(–3.01)	(2.33)	(–9.24)	(–0.88)	(–6.86)
10%–50%		−4.54***	0.57	−0.56**	−1.50	−0.54***	0.61**	−0.53***	−0.26	−0.58***
		(–3.60)	(1.40)	(–2.83)	(–0.99)	(–4.24)	(2.79)	(–8.96)	(–0.88)	(–6.86)
15%–50%		−2.19**	0.61	−0.56**	−0.95	−0.43***	0.71**	−0.57***	−0.26	−0.58***
		(–3.08)	(1.58)	(–2.83)	(–1.14)	(–4.17)	(2.68)	(–8.84)	(–0.88)	(–6.86)
20%–50%		−1.56**	0.51^†	−0.56**	−1.00^†	−0.37***	0.67*	−0.62***	−0.26	−0.58***
		(–3.34)	(1.85)	(–2.83)	(–1.77)	(–4.14)	(2.01)	(–9.05)	(–0.88)	(–6.86)
22%–50%		−1.44***	0.50^†	−0.56**	−0.93^†	−0.39***	0.71*	−0.64***	−0.26	−0.58***
		(–3.50)	(1.65)	(–2.83)	(–1.87)	(–4.48)	(1.95)	(–9.07)	(–0.88)	(–6.86)
25%–50%		−1.07**	0.41	−0.56**	−0.56	−0.45***	0.85*	−0.60***	−0.26	−0.58***
		(–3.19)	(1.15)	(–2.83)	(–1.41)	(–5.62)	(1.96)	(–8.01)	(–0.88)	(–6.86)
5%–60%		−14.39***	0.53	−0.50^†	−5.88	−0.47**	0.47**	−0.54***	−0.19	−0.59***
		(–4.13)	(1.33)	(–1.67)	(–1.39)	(–3.01)	(3.37)	(–10.20)	(–0.39)	(–5.10)

We examine the entrenchment effect in accordance with prior studies, for example, 5%–25% by Morck et al. (1988); 12%–40% by Short and Keasey (1999); 10%–40%, 10%–50% and 15%–50% by Lennox (2005); 10%–30%, 15%–40% and 20%–50% by Lin and Liu (2013). We also consider other threshold intervals including 25%–50% and 5%–60%. Furthermore, we use trial and error method to test the accuracy of intervals and find that the most appropriate entrenchment interval is 21%–50%. The first row (number) represents the estimated coefficient and the second row (number in parentheses) represents the t-value of significance. All continuous variables are winsorised at the 1st and 99th percentiles. All variables are defined in Table 2.

†

p < .10; *p < .05; **p < .01; ***p < .001. All tests are two-tailed.

We also use Model 2 to test the robustness of results for ACS. The results, presented in Columns (4)–(9) of Table 6, show a significantly negative relationship with NASFEES on all possible intervals, which is consistent with our primary results (in Table 5). This adds further support to our finding that stronger audit committees can constrain the purchase of NAS irrespective of the firm’s management ownership structure.

Second, following prior studies (e.g. Lennox, 2005; Lin and Liu, 2013; Short and Keasey, 1999), we use curvilinear regression models to examine the relationship between management ownership and NAS fees, and the robustness of the turning points for the entrenchment range identified in the primary tests (see Table 5). The curvilinear model for management ownership is developed as follows

\begin{array}{l} N A S F E E S = α + β_{1} M O S + β_{2} M O S^{2} + β_{3} M O S^{3} + γ_{1} L O S S \\ + γ_{2} A U D I T O P I N + γ_{3} C O M P L E X + γ_{4} R O A \\ + γ_{5} L E V E R A G E + γ_{6} F I R M S I Z E + γ_{7} F C F O \\ + γ_{8} B I G 4 + \emptyset_{2005 - 2015} Y E A R + φ_{1 - 9} I N D U S T R Y + ε \end{array}

Model 3

The curvilinear model for management ownership and audit committee strength is developed as follows

\begin{array}{l} N A S F E E S = α + β_{1} M O S + β_{2} M O S^{2} + β_{3} M O S^{3} + λ_{1} A C S + γ_{1} L O S S \\ + γ_{2} A U D I T O P I N + γ_{3} C O M P L E X + γ_{4} R O A + γ_{5} L E V E R A G E \\ + γ_{6} F I R M S I Z E + γ_{7} F C F O + γ_{8} B I G 4 \\ + \emptyset_{2005 - 2015} Y E A R + φ_{1 - 9} I N D U S T R Y + ε \end{array}

Model 4

We expect negative coefficients for β₁ and β₃ for the low and high management ownership regions, and a positive coefficient for β₂. The results are reported in Table 7.

Table 7.

Regression for testing curvilinear.

Independent variable	Curvilinear Model 3	Curvilinear Model 4
	Clmn (1)	Clmn (2)
Intercept	10.70*** (6.90)	11.03*** (7.54)
MOS	−1.33** (–2.89)	−2.83*** (–4.86)
	β₁ < 0	β₁ < 0
MOS²	4.88 × 10⁻²*** (3.75)	8.63 × 10⁻²*** (5.03)
	β₂ > 0	β₂ > 0
MOS³	−3.78 × 10⁻⁶*** (–3.78)	−6.70 × 10⁻⁶*** (–4.89)
	β₃ < 0	β₃ < 0
ACS		−0.19*** (–4.28)
LOSS	−0.22*** (–5.81)	−0.22*** (–.32)
AUDITOPIN	0.17** (3.21)	0.13^† (–1.79)
COMPLEX	0.38*** (12.60)	0.39*** (10.06)
ROA	0.09* (1.47)	0.05(0.86)
LEVERAGE	0.43*** (12.78)	0.45*** (10.23)
FIRMSIZE	0.27*** (10.03)	0.22*** (9.03)
FCFO	−0.01(–1.28)	−0.004(–0.77)
BIG4	0.65*** (9.90)	0.59*** (13.32)
YEAR	Yes	Yes
INDUSTRY	Yes	Yes
Adjusted R²	0.3201	0.2585
F-Statistic	77.43***	37.35***
N	5198	3965

Following prior studies (e.g. Short and Keasey (1999), Lennox (2005) and Lin and Liu (2013)), we examine the curvilinear relationship between management ownership and NAS fees, and robust the entrenchment region. We define the derivative values for the curvilinear models and derive the turning points of 20%–50% as the entrenchment. The first row (number) represents the estimated coefficient, the second row (number in parentheses) represents the t-value of significance, and the third row represents expected sign. All continuous variables are winsorised at the 1st and 99th percentiles. All variables are defined in Table 2.

†

p < .10; *p < .05; **p < .01; ***p < .001. All tests are two-tailed.

As expected, and as shown for Model 3 in Column (1), the coefficient (β₁) of management ownership (MOS) is negative (β₁ = –1.33, t = –2.89, p < 0.01), the coefficient (β₂) of squared management ownership (MOS²) is positive (β₂ = 4.88×10⁻², t = 3.75, p < 0.001), and the coefficient (β₃) of cubed management ownership (MOS³) is negative (β₃ = –3.78×10⁻⁶, t = –3.78, p < 0.001). The results for Model 4 with audit committee strength, presented in Column (2) of Table 7, are similar to those of Model 2.¹⁶ Furthermore, the patterns of the turning points in which lower and higher convergence-of-interests regions are separated by an entrenchment region for Models 3 and 4 are consistent with those of Lennox (2005) and Lin and Liu (2013) (i.e.|β₁| > |β₂| > |β₃|).

We analyse the turning points for the entrenchment region and define the derivative values for the curvilinear model (Model 3) as follows

\begin{array}{l} N A S F E E S = α - 1.33 \times M O S + 4.88 \times 10^{- 2} \times M O S^{2} - 3.78 \times 10^{- 6} \times M O S^{3} \\ + γ_{1} L O S S + γ_{2} A U D I T O P I N + γ_{3} C O M P L E X + γ_{4} R O A \\ + γ_{5} L E V E R A G E + γ_{6} F I R M S I Z E + γ_{7} F C F O + γ_{8} B I G 4 \\ + \emptyset_{2005 - 2015} Y E A R + φ_{1 - 9} I N D U S T R Y + ε \end{array}

Following the methodology used by Lin and Liu (2013), we find that NAS fees decline in the lower region of management ownership in the interval of 0%–20%, increase between 20% and 50%, and gradually decline beyond 50% of management ownership. Thus, the turning points identified using curvilinear model demarcate the entrenchment region to be within 20%–50% which is generally consistent with our primary findings of entrenchment at 21%–50%.

Third, prior studies report that empirical models with ownership variables may suffer from endogeneity which can yield inconsistent estimations (see e.g. Bhagat and Bolton, 2008; Coles et al., 2008; Demsetz and Villalonga, 2001; Hermalin and Weisbach, 1991; Himmelberg et al., 1999; Palia, 2001; Welch, 2003). Following the methodology used by Lennox (2005), we regress the ordinary least squares model with management ownership as the dependent variable, and use the coefficient estimates to obtain the predictable value of management ownership (MOS_HAT). The models are presented as follows

{\hat{y}}_{2} = π_{0} + π_{1} z_{1} + π_{2} z_{2} + π_{3} z_{3} + π_{4} z_{4} + v

y_{1} = β_{0} + β_{1} y_{2} + β_{2} z_{1} + β_{3} z_{2} + δ_{1} v + ε

\begin{array}{l} N A S F E E S = α + β_{1} M O S + δ_{1} M O S_H A T + β_{2} A C S + γ_{1} L O S S + γ_{2} A U D I T O P I N \\ + γ_{3} C O M P L E X + γ_{4} R O A + γ_{5} L E V E R A G E + γ_{6} F I R M S I Z E \\ + γ_{7} F C F O + γ_{8} B I G 4 + \emptyset_{2005 - 2015} Y E A R + φ_{1 - 9} I N D U S T R Y + ε \end{array}

Model 5

We then conclude using the following hypothesis indicating whether management ownership (MOS) is exogenous or endogenous:

H₀: δ₁ = 0 → Hypothesis that y₂ (MOS) is exogenous

H₁: δ₁ ≠ 0 → Hypothesis that y₂ (MOS) is endogenous

The results (not tabulated here) indicate that the coefficients of MOS_HAT are statistically significant for the sub-sample in the regions of MOS_L, MOS_M and MOS_H. Thus, H₁ is rejected, which means that MOS_HAT is endogenous. We therefore conclude that the management ownership (MOS) is exogenous.

We also use propensity score matching (PSM) to further account for the possibility of endogeneity in Model 2. PSM requires the sample to be divided into six groups, that is, three treatment groups of MOS_L, MOS_M and MOS_H with ACS, and three control groups of MOS_L, MOS_M and MOS_H without ACS. The variables are matched including LOSS, ROA, FCFO, YEAR and INDUSTRY.¹⁷ The results based on PSM samples (not tabulated in the paper) are consistent with the findings in Columns (4)–(6) of Table 5 and Columns (4)–(9) of Table 6. These results also suggest that sample selection bias is not a concern.

Fourth, Leenaerts and Van Bokhoven (1998) suggest that the piecewise regression model allows many linear models to be fitted to the data for different segments of an independent explanatory variable. Accordingly, we employ the piecewise regression models as an additional method to check the robustness of the results presented in Table 5 and the results for alternative entrenchment ranges shown in Table 6. We do this because the key independent variable, management ownership, is partitioned into interval regions (MOS_L, MOS_M and MOS_H)

\begin{array}{l} N A S F E E S = α + β_{1} M O S_L + β_{2} M O S_M + β_{3} M O S_H + γ_{1} L O S S + γ_{2} A U D I T O P I N \\ + γ_{3} C O M P L E X + γ_{4} R O A + γ_{5} L E V E R A G E + γ_{6} F I R M S I Z E \\ + γ_{7} F C F O + γ_{8} B I G 4 + \emptyset_{2005 - 2015} Y E A R + φ_{1 - 9} I N D U S T R Y + ε \end{array}

Model 6

Following Model 2, we extend Model 6 to include ACS as follows:

\begin{array}{l} N A S F E E S = α + β_{1} M O S_L + β_{2} M O S_M + β_{3} M O S_H + λ_{1} A C S + γ_{1} L O S \\ + γ_{2} A U D I T O P I N + γ_{3} C O M P L E X + γ_{4} R O A + γ_{5} L E V E R A G E \\ + γ_{6} F I R M S I Z E + γ_{7} F C F O + γ_{8} B I G 4 + \emptyset_{2005 - 2015} Y E A R \\ + φ_{1 - 9} I N D U S T R Y + ε \end{array}

Model 7

The results of the piecewise regression models (not tabulated in the paper) confirm the entrenchment range of 21%–50% of management ownership and are consistent with results reported in Table 5 for Models 1 and 2.

Fifth, we control for additional corporate governance variables in Model 1 to check the robustness of the findings reported in Table 5. These variables include audit committee size, nomination committee size, compensation committee size, board size, the proportion of independent directors, the numbers of board meeting, CEO duality, and block shareholding ownership. The results (not tabulated in the paper) are consistent with the findings in Table 5.

Finally, in a further robustness test, we exclude 55 observations where NAS fees were equal to zero. Our results (not tabulated in the article) are consistent with the primary results in Table 5.¹⁸

5. Conclusion

This study extends existing wider research in the domain of corporate governance and NAS. Drawing on the earlier works of Morck et al. (1988), Lennox (2005) and Lin and Liu (2013) that distinguish between the convergence-of-interests and entrenchment regions of management ownership, this study investigates the associations between management ownership and audit committees and NAS fees. Both management ownership and audit committee are important pieces in the corporate governance mosaic (Cohen et al., 2004), particularly in the context of NAS purchase decisions which involve both management and the audit committee.

By focusing on companies listed in Australia, the study responds to growing calls by academics and practitioners to consider individual jurisdictions as a way of better understanding the specific nuances of various corporate governance measures and regulatory reforms (Collier and Zaman, 2005; Doidge et al., 2007; Houghton et al., 2013; Porter, 1992; Zaman et al., 2011). The Australian setting provides a useful setting to examine corporate governance mechanisms and NAS because the purchase of NAS from the incumbent audit firm is permitted, subject to certain measures aimed at mitigating the associated perceived threat to auditor independence.

We find that in Australia, the entrenchment region is between 21% and 50% of managerial ownership while regions of management ownership below 21% and above 50% constitute convergence-of-interests regions. While these cut-offs differ from those suggested in extant empirical research (see for example Lennox, 2005; Lin and Liu, 2013; Morck et al., 1988), they suggest that in the broader range of management ownership, convergence-of-interests and entrenchment regions feature similar patterns of distribution, both confirming theory expectations but also highlighting jurisdictional variation influences (Campbell et al., 2012).

Based on a sample of 5198 firm-year observations of ASX listed companies over the period of 2005–2015, we find that management ownership is positively associated with NAS fees within the entrenchment region of ownership. In contrast, management ownership is negatively associated with NAS fees within the convergence-of-interests regions of ownership. However, using a sub-sample confined to firms that have audit committees, we find no association between management ownership and NAS within the convergence-of-interests regions while continuing to observe a positive association between management ownership and NAS within the entrenchment region.

The finding constitutes a departure from what extant literature had established in terms of the positive association between management ownership and NAS fees. It also underscores the importance of considering both the convergence of interests and management entrenchment for a nuanced analysis of management ownership in relation to NAS fees.

Furthermore, we provide evidence that NAS fees decrease in firms with audit committees that simultaneously meet the following criteria: (1) have at least three members; (2) all audit committee members are non-executive directors, the majority of whom are independent directors; (3) are chaired by an independent director, who is not the chair of the board; (4) conduct at least four meetings a year; and (5) have at least committee member with accounting and financial expertise. Within the entrenchment region of management ownership, NAS fees are increasing with the proportion of shares held by management, but negatively associated with the presence of a strong audit committee.

The reduced purchase of NAS by a company from the incumbent audit firm when audit committees are more able to impose more effective monitoring and control over management, is interpreted as reflecting concerns about the perceived threat to auditor independence. Conversely, the increasing purchase of NAS from a company’s incumbent audit firm as management becomes entrenched is interpreted as evidence that entrenched managers may perceive the benefits of the audit firm’s expertise to be more valuable than the costs associated impairment to auditor independence.

Similar to other studies, this study also suffers from inherent limitations. In particular, this study does not show nor explain the manner in which managers and audit committee members interact as decisions about NAS purchases are made (for example, in audit committee meetings or other types of meetings). Further research could adopt survey, observation or interviews for additional insight to better understand these interactions (Beasley et al., 2009; Krishnan, 2005; PWC, 2016; Wu et al., 2016). Second, while desired statistical power effects have been satisfied in this study, generalisation beyond the Australian setting could be an avenue of further research to confirm (or refute) our findings in other jurisdictions.

Footnotes

Acknowledgements

We are grateful to Mandy Cheng (Deputy Editor) and Mark Wilson (Associate Editor) at AJM, and anonymous reviewers for their constructive comments and suggestions in the review process. We thank the anonymous AFAANZ 2016, 2017 Conference Reviewers, Paul Coram (University of Adelaide), Karla Johnstone (University of Wisconsin-Madison), David Woodliff (University of Western Australia), Douglas DeJong (University of Iowa), Millicent Chang (University of Wollongong) and Henry Lahr (Open University) for their helpful comments. We also acknowledge the constructive comments on earlier versions of this article received from seminar participants at the University of Adelaide (Australia), Australian National University (Australia), University of Newcastle (Australia), Curtin University (Australia), Murdoch University (Australia), Renmin University of China (China), Shanghai University of Finance and Economics (China), Tsinghua University (China), University of Rome Tor Vergata (Italy), and Open University (UK).

Final transcript accepted 5 June 2020 by Mark Wilson (AE Accounting and Auditing).

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Yuan George Shan

Notes

References

Abbott

Parker

Peters

(2011) Does mandated disclosure induce a structural change in the determinants of nonaudit service purchases? Auditing: A Journal of Practice and Theory 30: 51–76.

Abbott

Parker

Raghunandan

(2003) An empirical investigation of audit fees, non-audit fees, and audit committees. Contemporary Accounting Research 20: 215–234.

Agarwal

Chadha

(2005) Corporate governance and accounting scandals. Journal of Law and Economics 48: 371–406.

Arcot

Bruno

Faure-Grimaud

(2010) Corporate governance in the UK: Is the comply or explain approach working. International Review of Law and Economics 30: 193–201.

ASA705 (2015) Modifications to the opinion in the independent auditor’s report.

Ashbaugh

Lafond

Mayhew

(2003) Do non-audit services compromise auditor independence? The Accounting Review 78: 611–639.

ASXCGC (2014) ASX Corporate Governance Principles and Recommendations. Sydney: ASX Corporate Governance Council. Available at: http://www.asx.com.au/documents/asx-compliance/cgc-principles-and-recommendations-3rd-edn.pdf (accessed 31 August 2019).

Azim

(2012) Corporate governance mechanisms and their impact on company performance: A structural equation model analysis. Australian Journal of Management 37: 481–505.

Beasley

Carcello

Hermanson

, et al. (2009) The audit committee oversight process. Contemporary Accounting Research 26: 65–122.

10.

Berger

Ofek

Yermack

(1997) Managerial entrenchment and capital structure decisions. Journal of Finance 52: 1411–1438.

11.

Bhagat

Bolton

(2008) Corporate governance and firm performance. Journal of Corporate Finance 14: 257–273.

12.

Campbell

Sirmon

, et al. (2012) Shareholder influence over director nomination via proxy access: Implications for agency conflict and stakeholder value. Strategic Management Journal 33: 1431–1451.

13.

Chan

Chen

Janakiraman

, et al. (2012) Reexamining the relationship between audit and nonaudit fees: Dealing with weak instruments in two-stage least squares estimation. Journal of Accounting, Auditing & Finance 27: 299–324.

14.

Chau

Leung

(2006) The impact of board composition and family ownership on audit committee formation: Evidence from Hong Kong. Journal of International Accounting, Auditing & Taxation 15: 1–15.

15.

Cohen

Krishnamoorthy

Wright

(2004) The corporate governance mosaic and financial reporting quality. Journal of Accounting Literature 23: 87–154.

16.

Coles

Daniel

Naveen

(2008) Boards: Does one size fit all? Journal of Financial Economics 87: 329–356.

17.

Collier

Zaman

(2005) Convergence in European corporate governance: The audit committee concept. Corporate Governance: An International Review 13: 753–768.

18.

Davidson

Goodwin-Stewart

Kent

(2005) Internal governance structures and earnings management. Accounting & Finance 45: 241–267.

19.

De Fuentes

Porcuna

(2016) Main drivers of consultancy services: A meta-analytic approach. Journal of Business Research 69: 4775–4780.

20.

DeFond

Raghunandan

Subramanyam

(2002) Do non-audit service fees impair auditor independence? Evidence from going concerns audit opinions. Journal of Accounting Research 40: 1247–1274.

21.

Demsetz

Villalonga

(2001) Ownership structure and corporate performance. Journal of Corporate Finance 7: 209–233.

22.

Desender

Aguilera

Crespi

, et al. (2013) When does ownership matter? Board characteristics and behavior. Strategic Management Journal 34: 823–842.

23.

Desender

Aguilera

Lópezpuertas-Lamy

, et al. (2016) A clash of governance logics: Foreign ownership and board monitoring. Strategic Management Journal 37: 349–369.

24.

Dignam

(2008) The globalisation of General Principle 7: Transforming the market for corporate control in Australia and Europe? Legal Studies 28: 96–118.

25.

Dignam

Galanis

(2004) Australia inside-out: The corporate governance system of the Australian listed market. Melbourne University Law Review 28: 623–653.

26.

Doidge

Karolyi

Rene

(2007) Why do countries matter so much for corporate governance? Journal of Financial Economics 86: 1–39.

27.

Fama

(1980) Agency problems and the theory of the firm. Journal of Political Economy 88: 88–107.

28.

Fan

Wong

(2002) Corporate ownership structure and the informativeness of accounting earnings in East Asia. Journal of Accounting and Economics 33: 401–425.

29.

Farinha

(2003) Dividend policy, corporate governance and the managerial entrenchment hypothesis: An empirical analysis. Journal of Business Finance & Accounting 30: 1173–1209.

30.

Firth

(1997) The provision of non-audit services and the pricing of audit fees. Journal of Business Finance & Accounting 23: 511–525.

31.

Firth

(2002) Auditor-provided consultancy services and their association with audit fees and audit opinions. Journal of Business Finance & Accounting 29: 661–693.

32.

Frankel

Johnson

Nelson

(2002) The relationship between auditors’ fees for nonaudit services and earnings management. The Accounting Review 77: 71–105.

33.

FRC (2016) The UK Corporate Governance Code. London: Financial Reporting Council (FRC). Available at: https://www.frc.org.uk/Our-Work/Publications/Corporate-Governance/UK-Corporate-Governance-Code-April-2016.pdf (accessed 31 August 2019).

34.

Garg

(2017) Value relevance of voluntary internal control certification: An information asymmetry perspective. Australian Journal of Management 42: 527–559.

35.

Gaynor

McDaniel

Neal

(2006) The effects of joint provision and disclosure of non-audit services on audit committee members’ decisions and investors’ preferences. The Accounting Review 81: 873–896.

36.

Gedajlovic

Shapiro

(1998) Management and ownership effects: Evidence from five countries. Strategic Management Journal 19: 533–553.

37.

Gotti

Han

Higgs

, et al. (2012) Managerial stock ownership, analyst coverage, and audit fees. Journal of Accounting, Auditing & Finance 27: 412–437.

38.

Grosse

Scott

(2020) Evidence on compensation consultant fees and CEO pay. Australian Journal of Management 45: 15–44.

39.

Gujarati

(2003) Basic Econometrics. New York: McGraw-Hill.

40.

Gul

Tsui

JSL

(1997) A test of the free cash flow and debt monitoring hypothesis: Evidence from audit pricing. Journal of Accounting and Economics 24: 219–237.

41.

Hay

Knechel

(2006a) Non-audit services and auditor independence: New Zealand evidence. Journal of Business Finance & Accounting 33: 715–734.

42.

Hay

Knechel

Ling

(2008) Evidence on the impact of internal control and corporate governance on audit fees. International Journal of Auditing 12: 9–24.

43.

Hay

Knechel

Wong

(2006b) Audit fees: A meta-analysis of the effect of supply and demand attributes. Contemporary Accounting Research 23: 141–191.

44.

Henry

(2010) Agency costs, ownership structure and corporate governance compliance: A private contracting perspective. Pacific-Basin Finance Journal 18: 24–46.

45.

Hermalin

Weisbach

(1991) The effects of board composition and direct incentives on firm performance. Financial Management 20: 101–121.

46.

Himmelberg

Hubbard

Palia

(1999) Understanding the determinants of managerial ownership and the link between ownership and performance. Journal of Financial Economics 53: 353–384.

47.

Houghton

Kend

Jubb

(2013) The CLERP 9 audit reforms: Benefits and costs through the eyes of regulators, standard setters and audit service suppliers. Abacus 49: 139–160.

48.

Hsu

Koh

(2005) Does the presence of institutional investors influence accruals management? Evidence from Australia. Corporate Governance: An International Review 13: 809–823.

49.

Jensen

Meckling

(1976) Theory of the firm: Management behavior, agency costs and ownership structure. Journal of Financial Economics 3: 305–360.

50.

Jensen

(1986) Agency costs of free cash flow, corporate finance, and takeovers. American Economic Review 76: 323–339.

51.

Kent

Stewart

(2008) Corporate governance and disclosure on the transition to International Financial Reporting Standards. Accounting & Finance 48: 649–671.

52.

Khan

Mather

Balachandran

(2014) Managerial share ownership and operating performance: Do independent and executive directors have different incentives. Australian Journal of Management 39: 47–71.

53.

Khlif

Ahmed

Souissi

(2017) Ownership structure and voluntary disclosure: A synthesis of empirical studies. Australian Journal of Management 42: 376–403.

54.

Kole

(1995) Measuring managerial equity ownership: A comparison of sources of ownership data. Journal of Corporate Finance 1: 413–435.

55.

KPMG_ACCA (2014) Balancing Rules and Flexibility for Growth. A Study of Corporate Governance Requirements across 25 Markets. Singapore: KMPG and ACCA. Available at: https://home.kpmg.com/mu/en/home/insights/2017/06/balancing-rules-and-flexibility-for-growth.html (accessed 31 August 2019).

56.

Krishnan

(2005) Audit committee quality and internal control: An empirical analysis. The Accounting Review 80: 649–675.

57.

Krishnan

Wen

Zhao

(2011) Legal expertise on corporate audit committees and financial reporting quality. The Accounting Review 86: 2099–2130.

58.

La Porta

Lopez-de-Silanes

Shleifer

(1999) Corporate ownership around the world. Journal of Finance 54: 471–517.

59.

Larcker

Richardson

(2004) Fees paid to audit firms, accrual choices, and corporate governance. Journal of Accounting Research 42: 625–658.

60.

Lee

Mande

(2005) The relationship of audit committee characteristics with endogenously determined audit and non-audit fees. Quarterly Journal of Business and Economics 44: 93–112.

61.

Leenaerts

DMW

Van Bokhoven

WMG

(1998) Piecewise Linear Modeling and Analysis. New York: Springer.

62.

Lennox

(2005) Management ownership and audit firm size. Contemporary Accounting Research 22: 205–227.

63.

Levitt

(2000) Renewing the Covenant with Investors. Speech at New York University Center for Law and Business, May 10. New York: New York University. Available at: http://www.sec.gov/news/speech/spch370.htm (accessed 31 August 2019).

64.

Lin

Liu

(2013) The effects of managerial shareholding on audit fees: Evidence from Hong Kong. International Journal of Auditing 17: 227–245.

65.

McConnell

Servaes

(1990) Additional evidence on equity ownership and corporate value. Journal of Financial Economics 27: 595–612.

66.

McKnight

Weir

(2009) Agency costs, corporate governance mechanisms and ownership structure in large UK publicly quoted companies: A panel data analysis. Quarterly Review of Economics and Finance 49: 139–158.

67.

Miglani

Ahmed

Henry

(2015) Voluntary corporate governance structure and financial distress: Evidence from Australia. Journal of Contemporary Accounting & Economics 11: 18–30.

68.

Miglani

Ahmed

Henry

(2020) Corporate governance and turnaround: Evidence from Australia. Australian Journal of Management. Epub ahead of print 14 February. DOI: 10.1177/0312896220902225.

69.

Mitra

Hossain

(2007) Ownership composition and non-audit fees. Journal of Business Research 60: 339–347.

70.

Monem

(2013) Determinants of board structure: Evidence from Australia. Journal of Contemporary Accounting & Economics 9: 33–49.

71.

Morck

Shleifer

Vishny

(1988) Management ownership and market valuation: An empirical analysis. Journal of Financial Economics 20: 293–315.

72.

Navissi

Naiker

(2006) Institutional ownership and corporate value. Managerial Finance 32: 247–256.

73.

O’Sullivan

(2000) The impact of board composition and ownership on audit quality: Evidence from large UK companies. The British Accounting Review 32: 397–414.

74.

O’Sullivan

Diacon

(2002) The impact of ownership, governance and non-audit services and non-audit fees: Evidence from the insurance industry. International Journal of Auditing 6: 93–107.

75.

Palia

(2001) The endogeneity of managerial compensation in firm valuation: A solution. Review of Financial Studies 14: 735–764.

76.

Porter

(1992) Capital disadvantage: America’s failing capital investment system. Harvard Business Review 70: 65–82.

77.

Prakash

Venable

(1993) Auditee incentives for auditor independence: The case of nonaudit services. The Accounting Review 68: 113–134.

78.

PWC (2016) Audit Committee Guide: How Can Audit Committee Members Add Value? Sydney: PwC Australia. Available: http://www.pwc.com.au/assurance/assets/audit-committee-guide/ac-guide-booklet-dec11.pdf (accessed 31 August 2019).

79.

Ratzinger-Sakel

NVS

Schönberger

(2015) Restricting non-audit services in Europe – The potential (lack of) impact of a blacklist and a fee cap on auditor independence and audit quality. Accounting in Europe 12: 61–86.

80.

Ruddock

Taylor

(2006) Nonaudit services and earnings conservatism: Is auditor independence impaired? Contemporary Accounting Research 23: 701–746.

81.

Schooley

Barney

(1994) Using dividend policy and managerial ownership to reduce agency costs. Journal of Financial Research 17: 363–373.

82.

SEC (2003) SECURITIES AND EXCHANGE COMMISSION 17 CFR PARTS 228, 229, 240, 249 and 274 [RELEASE NOS. 33-8220; 34-47654; IC-26001; File No. S7-02-03] RIN 3235-AI75 STANDARDS RELATING TO LISTED COMPANY AUDIT COMMITTEES (effective date April 25, 2003), US SEC. Available at: https://www.sec.gov/rules/final/33-8220.htm (accessed 31 August 2019).

83.

Shan

Troshani

Tarca

(2019) Managerial ownership, audit firm size, and audit fees: Australian evidence. Journal of International Accounting, Auditing & Taxation. https://doi.org/10.1016/j.intaccaudtax.2019.05.002:1-19

84.

Sharma

Sidhu

(2001) Professionalism vs commercialism: The association between Non-Audit Services (NAS) and audit independence. Journal of Business Finance & Accounting 28: 563–594.

85.

Shleifer

Vishny

(1997) A survey of corporate governance. Journal of Finance 52: 737–783.

86.

Short

Keasey

(1999) Managerial ownership and the performance of firms: Evidence from the UK. Journal of Corporate Finance 5: 79–101.

87.

Shuto

Takada

(2010) Managerial ownership and accounting conservatism in Japan: A test of management entrenchment effect. Journal of Business Finance & Accounting 37: 815–840.

88.

SOX (2002) Sarbanes-Oxley Act of 2002. Public Law 107–204. 116 STAT. 745. US Congress. Available at: https://www.sec.gov/about/laws/soa2002.pdf (accessed 31 August 2019).

89.

Srinidhi

Gul

(2007) The differential effects of auditors’ nonaudit and audit fees on accrual quality. Contemporary Accounting Research 24: 595–629.

90.

Teshima

Shuto

(2008) Managerial ownership and earnings management: Theory and empirical evidence from Japan. Journal of International Financial Management and Accounting 19: 107–132.

91.

Welch

(2003) The relationship between ownership structure and performance in listed Australian companies. Australian Journal of Management 28: 287–305.

92.

Whisenant

Sankaraguruswamy

Raghunandan

(2003) Evidence of joint determination of audit and non-audit fees. Journal of Accounting Research 41: 721–744.

93.

CYH

Hsu

Haslam

(2016) Audit committees, non-audit services, and auditor reporting decisions prior to failure. British Accounting Review 48: 240–256.

94.

Zaman

Hudaib

Haniffa

(2011) Corporate governance quality, audit fees and non-audit services fees. Journal of Business Finance and Accounting 38: 165–197.