Manager Attributes, Firm Characteristics, and Audit Committee Independence: Evidence from the Canadian Context

To improve the information quality and defend investors’ interests, the current challenge is no longer only to set up an audit committee, but also to ensure its independence. Nevertheless, this independence is not always guaranteed and it depends on several factors. The study aims to identify the determinants of the audit committee independence. We identify factors linked to the manager attributes and factors linked to the firm characteristics. The empirical study is drawn on a sample of Canadian firms over a period of five years. The results show that the independence of the audit committee is negatively related to the size of the board of directors and to the presence of the manager within the remuneration committee. Furthermore, independence of the audit committee seems to be positively linked to the independence of the board of directors and the existence of intangible assets.


I. INTRODUCTION
Over the past decades, there has been considerable interest in setting up an audit committee due to the important role it plays in corporate governance (Stewart and Munro, 2007). Indeed, investors give more importance to companies with an effective governance system. Thus, to protect and restore the confidence of these investors in the published financial information, professional and regulatory bodies have tried to encourage companies to set up specialized committees. The institutionalization of these committees corresponds to a desire to improve the functioning of the board of directors on the one hand (Pochet and Yeo, 2004), and to help ensure a balance in the directors' and shareholders' powers on the other hand (Ebondo et al., 2014). However, the current challenge to improve the quality of the information published is no longer just to set up such a body, but also to ensure its independence since the objective of defending the interests of investors has been attributed mainly to an independent audit committee. Indeed, it has been proven that an independent audit committee has the potential to influence investment decision-making (Al-Hadrami et al. 2020). This independence criterion has several consequences such as improving the efficiency of the audit committee (Report of the National Commission for Fraudulent Financial Reports, 1987), the internal control and the external audit quality (Krishnan, 2005;Carcello and Neal, 2003), and the limitation of the earnings management practice (Wan Mohammad and Wasiuzzaman, 2020, Koh et al., 2007, Bradbury et al., 2006 and restatements (Pucheta-Martínez and De Fuentes, 2007). Consequently, by controlling the audit process and reducing the possibility of information manipulation, the quality of published information improves. Indeed, Abbott et al. (2000) claim that the existence of an independent committee reduces errors in financial statements. They support the idea that companies with an audit committee composed entirely of independent members are less sanctioned by the SEC for publishing false information in their financial statements. This idea was also supported by Beasley et al. (2000) who show that firms publishing erroneous information are those which have fewer independent audit committees. Likewise, Abbott et al. (2004) provide that the corrections of errors in the previous years' financial statements are low in companies with independent audit committees. This idea has been confirmed by Persons (2005) who hypothesizes that the level of fraud is low when the audit committee is independent. This evidence on the independence of the audit committee makes it a corporate governance system that guarantees the quality of the financial information published (Bédard and Géndron, 2010; DeFond and Zhang, 2014). The vast amount of literature devoted to the audit committee gives evidence on the importance of this body over time. Current efforts are directed not only towards the establishment of an audit committee but also to ensure its independence from management. A set of reports has been issued on this subject. For example, the AICPA Public Oversigh Board (1993) argues that to be effective, an audit committee must be composed entirely of independent directors. Furthermore, the National Commission on Fraudulent Financial Reports (1987) reveals that the mere presence of an audit committee does not mean that it is effective, but that it must be independent. Likewise, Ebondo et al. (2014) argue that the independence criterion can be used when assessing the effectiveness of the audit committee. However, the independence of the audit committee is not always obvious and its presence depends on several factors. From this perspective, this study seeks to identify the determinants that can influence the independence of the audit committee in the Canadian context. The study of this context is interesting, firstly, because following the bankruptcy of several Canadian listed companies; the Toronto Stock Exchange had formed a working group in 1993 to assess the corporate governance of Canadian limited companies. This group produced the Dey Report (1994) setting out a series of guidelines for improving corporate governance and recommending the creation of independent audit committees. Secondly, the set of reports 1 published by the United States (the Bleu Ribbon Committee report, 1999, the Sarbanes-Oxley law, 2002 and the rules set by the Securities Exchange Commission) aimed at improving the information transparency through the creation of independent audit committees had several consequences on the governance system of Canadian companies (Niu, 2006). Indeed, in response to the publication of the Sarbanes-Oxley Act (2002), Canadian regulators set in June 2003 several governance rules similar to those set by that act and by the SEC. Some of these rules are related to the audit committee. They require Canadian companies to create audit committees composed of independent directors with experience in finance and to disclose this information to help investors assess the expertise of audit committee members (Niu, 2006). Finally, The Canadian Corporate Statutes require that public corporations establish an audit committee composed of at least three directors, a majority of whom are not inside directors. Canadian securities law requirements significantly exceed the Canadian corporate law requirements for audit committees. Under securities law, public corporations must have at least three directors on their audit committees, all of whom must be independent. The rest of the article is organized as follows: the theoretical background is presented in section 2. The literature review and research hypotheses are provided in section 3. Section 4 presents the research methodology and section 5 highlights results and discussions. Finally, we conclude in section 6.

II. THEORETICAL BACKGROUND
The audit committee derives its theoretical background from the agency theory (Jensen and Meckling, 1976) and the institutional theory (Meyer and Rowan, 1977). According to the agency theory, the probability of establishing an audit committee increases with increasing agency costs (Pincus et 1 The list of reports requiring the creation of an audit committee is presented in appendices. al., 1989). Indeed, in managerial companies characterized by a separation between management and ownership, managers do not often act in the interests of shareholders, hence the need to establish a control system responsible for reducing conflicts of interests between managers and shareholders. Fama and Jensen (1983) reveal that in managerial companies, shareholders delegate their rights of control to the board of directors. The reduction in agency costs resulting from these conflicts can be ensured by the audit committee (Pincus et al. 1989). In addition, due to the weak explanatory power of agency theory, some researchers suggest using institutional theory to better understand the audit committee's functioning. According to this paradigm, the creation of specialized committees represents a compliance phenomenon with accepted practices. Based on these theories, many studies both continental (Saada, 1998;Pochet and Yeo, 2004) and Anglo-Saxon (Klein, 2002b;Krishnan, 2005) try to approach the theme of the audit committee in different ways. Some of these studies show the important role played by an independent audit committee in protecting the external auditor and in improving the quality of disclosed financial information. In fact, in carrying out its missions, the audit committee may make decisions contrary to management. It is, therefore, necessary to increase the number of independent directors within the audit committee, since they are expected to have the capacity to act in the best interests of the shareholders.

III. LITERATURE REVIEW AND HYPOTHESES
DEVELOPMENT The literature gives insights into several factors that may influence the independence of the audit committee. Based on the mixed results shown by earlier studies investigating, we have grouped these factors into two categories: factors linked to the manager attributes, and factors linked to the firms' internal characteristics.

A. Factors related to the manager attributes
The control mechanisms represent an obstacle to the opportunistic behaviors of the managers. Hence, they always try to limit these features. Thus, Paquerot (1996) confirms that the second phase of the entrenchment strategy consists in reducing the means of control. In this phase, the managers try to reduce the effectiveness of the control mechanisms by increasing the number of inside directors both on the board of directors and on the audit committee. We have identified four factors related to manager attributes that may influence the independence of the audit committee: the presence of the manager on the remuneration committee, his presence on the nomination committee, and managerial propriety.

Presence of the manager within the remuneration committee
The remuneration committee is a body attached to the board of directors. Its functions are to determine the remuneration principles of directors and corporate officers, to monitor the application of these principles, and to ensure their consistency with the annual performance assessment (Ebondo et al., 2014). The remuneration committee must first be independent of management to carry out its missions effectively. Hence, the presence of the manager within the remuneration committee is an indicator of his entrenchment. Klein (2002b) asserts that the manager is usually a member of the compensation committee when he has good relations with the members of the board of directors. By occupying this position, he can exert an influence on the control mechanisms. Hence, our hypothesis: Hypothesis 1: The presence of the manager within the remuneration committee negatively influences the independence of the audit committee.

Presence of the manager within the nomination committee
The role of the nomination committee is to nominate people likely to occupy key positions in the company and to study potential succession candidates (Ebondo et al., 2014). It also plays an important role in achieving a balance of power between managers and shareholders. Ebondo et al. (2014) show that specialized commissions balance power relations. Likewise, Sarkar et al. (2008) argue that the main objective of governance mechanisms is to ensure the alignment of interests between shareholders and managers. According to Saada (1998), audit committees appear mainly in managerial firms to reduce agency costs. It is recommended that this committee be independent of management. The presence of the manager in this committee can affect its effectiveness since he will intervene in the appointment of directors and will try to increase the number of internal directors. However, Shivdasani and Yermack (2002) show that the number of directors outside the board of directors is lower in companies where the manager is a member of the nomination committee. Hypothesis 2: The probability of having an independent audit committee is low in companies whom directors are members of the nomination committee.

Managerial ownership
Previous research has not yielded a consensus on the relationship between managerial ownership and board independence. Indeed, Weisbach (1988) shows the existence of a negative linear relationship. This finding was not confirmed by Peasnell et al. (2003) who find the existence of a non-linear relationship. Their study shows a negative link between managerial ownership and the percentage of outside directors on the board when managerial ownership is low. This relationship becomes positive when managerial ownership increases. This can be explained by the fact that when managerial ownership is low an alignment of interests can be observed. There is therefore no need to increase the number of outside directors. Whereas, when managerial ownership increases this leads to an increase in conflicts of interest, hence the need to increase the number of outside directors to resolve these conflicts. Furthermore, Julie and Mark (2003) show a negative and linear relationship between managerial ownership and board independence. This observation is supported by Deli and Gillan (2000). While, for Pochet and Yeo (2004), the establishment of specialized committees decreases when managerial ownership is high. Hypothesis 3: The relationship between the independence of the audit committee and the managerial ownership is nonlinear.

B. Factors related to the firms' internal characteristics
The literature shows that the audit committee independence is influenced by some internal characteristics of the firm. Indeed, Peasnell et al. (2003) state that the audit committee cannot directly influence the audit process. There must be consensus and coordination with the members of the board of directors for the audit committee to perform its task properly. We study the board of directors' independence and size, the firm size, the firm performance, and the existence of intangible assets.

The board of directors' independence
The board of directors is responsible for protecting shareholders against manager opportunism. Jensen (1993) asserts that the board has disciplinary power over leaders. Likewise, Weisbach (1988) finds that leadership change when achieving low performance is more likely in companies with a board of directors composed largely of outside directors. Based on this observation, we assume that an independent board of directors plays an important role in the control of the leader. The importance of the board of directors' composition is also identified through its influence on performance. Charreaux (2000) argues that there is a relationship between firm performance and the board of directors' size and composition. In addition, Menon and Williams (1994), Klein (1998aKlein ( , 2002b, and Julie and Mark (2003) find a positive relationship between the board of directors' independence and the audit committee independence. This result contradicts Saada (1998), who shows through a study of a sample of French companies, that the board of directors' composition is not an explanatory factor for the decision to set up an audit committee. This idea was confirmed by Pochet and Yeo (2004).

Hypothesis 4:
The board of directors' independence positively influences the audit committee independence.

The board of directors' size
The board of directors' size and the audit committee vary by country and by the firm. According to the corporate governance principles of the American Law Institute (1994), audit committees should have at least three members; there is no optimal size and it all depends on the needs and situation of each firm. According to Yermack (1996), the small board of directors' size is a good indicator of its effectiveness and has a significant impact on the firm performance. Similarly, a relationship was found by Omri (2003) between the size of the board of directors and the firm performance. In addition, Klein (2002b) finds a positive relationship between the audit committee independence and the board size. Thus, Saada (1998) and Bradbury (1990) reveal a link between the board of directors' size and the probability of setting up an audit committee. Hypothesis 5: The audit committee independence is negatively related to the board of directors' size.

The firm size
The increase in the firm size can cause an increase in conflicts of interest between managers and shareholders (Fama and Jensen, 1983), which requires the establishment of a significant control system responsible for resolving these conflicts. Moreover, according to Pochet and Yeo (2004), the need for control mechanisms increases with the increase in agency costs. Klein (1998a;2002a) and Deli and Gillan (2000) show a positive relationship between the firm size and the audit committee independence. Similarly, a positive relationship is observed by Pincus et al. (1989) between the firm size and the probability of creating an audit committee.

Hypothesis 6:
The audit committee independence is more likely in large companies.

The firm performance
The studies attempting to determine the relationship between firm performance and certain governance mechanisms lead to contradictory results. Some of these studies confirm the existence of a relationship (Alodat et al., 2021), and some others infirm it (Al-ahdal and Hashim (2021). Brown and Caylor (2004) find that the independence of different governance mechanisms (board of directors, nomination committee, and remuneration committee) is associated with good firm performance. While Klein (1998b) shows that there is no link between the firm performance and the independence of the specialized committees (the remuneration committee and the audit committee). Hypothesis 7: The audit committee independence is positively linked to the firm performance.

The existence of intangible assets
The intangible asset provides measurement problems (Whitwell et al., 2007) due to the determination of the entry cost, the determination of the depreciation period, and the method of recording value permanent and substantial depreciation. Solutions to these difficulties are not obvious due to the specific characteristics of these assets. They lack physical quality and therefore the values of these assets are difficult to estimate (Klapper and Love, 2004). In such a situation, the manager can take advantage of these shortcomings to manage earnings in the direction that suits him. To limit these behaviors, it is necessary to improve the governance system quality (in terms of independence and skills of the directors) to ensure effective control of drawing up financial statements process and to reduce the possibility of accounting manipulation through intangible assets. According to Klapper and Love (2004), the higher the percentage of intangible assets, the greater the need for an effective governance system. Since the audit committee is a part of the control system, it is responsible for improving the quality of the information disclosed. It is, therefore, necessary to improve its efficiency by increasing the number of independent and competent directors sitting within it, so that it can reduce the possibility of accounting information manipulation. Klein (2002a) confirms that the audit committee has the responsibility of controlling the process of preparing financial reports. Consequently, it organizes regular meetings with the external auditor and managers. Hence, the governance system is effective in firms with a large fraction of intangible assets. In this study, we assume that: Hypothesis 8: The audit committee independence is positively linked to a significant portion of intangible assets.

IV. RESEARCH METHODOLOGY
We seek to identify the determinants of the audit committee independence in the Canadian context. Indeed, the Canadian economy has been affected by several financial scandals (Bre-X Minerals Ltd; Livent Inc; YBM Magnex international Ltd; Ginar Corp and Visual abs Inc), which has forced Canada to improve the government system of their companies by improving their audit committees effectiveness. The study is drawn on a sample including 62 Canadian firms over five years. The data used was taken from the Canadian Electronic Regulatory Database (SEDAR). Two models are specified and estimated. The first model is used to determine the manager attributes on the audit committee independence, while the second is used to study the relationship between the firms' internal characteristics and the audit committee independence.

A. Models specification
Two models are specified in the study. The first model tests the influence of entrenched managers on the audit committee independence. The second model is used to assess the relationship between the firms' internal characteristics and the audit committee independence.

Model 1.
ACINDik=β0 + β1REMCik + β2NOMCik + β3SENIOik + β4MOWNik + 5LINEAik + ζik  (Klein, 1998a); and (2) the absence of the CEO within the audit committee (Julie and Mark, 2003). In this research, we use the first measure (percentage of independent directors which is equal to the ratio between the number of external directors and the total number of directors on the audit committee) since Canadian firms do not allow their managing directors to be members of the audit committee.

Independent variables
 Variables of Model 1 REMC: the presence of the manager within the remuneration committee: a dummy variable equals 1 if the manager is a member of the remuneration committee and 0 if not. The presence of the manager within the remuneration committee is a good indicator of his entrenchment. NOMC: the presence of the manager within the nomination committee: a dummy variable equals 1 if the manager is a member of the nomination committee and 0 if not. SENIO: the manager seniority: a control variable measured by the logarithm of the number of years the manager has spent with the firm as the manager of the firm (Omri, 2003). MOWN: managerial ownership: equals the percentage of capital held by the manager. LINEA: the linearity of the relationship between the audit committee independence and the managerial ownership: equals to the square of the share of capital held by the manager.
 Variables of Model 2 BDIND: the board of directors' independence equals the ratio between the number of directors outside the board and the total of directors. This measure was identified and used by Klein (2002b) and Julie and Mark (2003). BDSIZE: board of directors' size, equals the natural logarithm of total board members (Yermak, 1996; Klein, 2002b andOmri, 2003). FSIZE: the firm size. In the literature, several measures are used for the size of the firm: the sum of the market value of common stocks and the book value of current liabilities, long-term debts and preferred stocks (Pincus et al. 1989), and the natural logarithm of total assets (Klein, 2002b). In our study, we will use this last measure. PERF: the firm performance. Several measures were used to measure performance (Tobin's Q, Market-to-book, return on assets, return on equity, turnover growth rate...). We use the turnover growth rate as a measure of performance. INTEN: intangible assets, a dummy variable equals 1 if there is a significant fraction of intangible assets within the firm and 0 if not.

Model 1: Manager attributes and audit committee independence
The results of model 1 are presented in Table 2. The table shows the existence of a negative and statistically significant relationship at a confidence level of 1% between the variables ACIND and REMC suggesting that the presence of the manager on the remuneration committee negatively influences the audit committee independence. This result is consistent with Klein (2002b) and is in line with our expectations. It seems logical to us that the director's membership of the remuneration committee is a good indicator of his entrenchment and power. So, once the manager is a member of the compensation committee, it means that he exercises power to influence the control mechanisms. Surprisingly, we found that the variable NOMC presents a non-significant sign with the dependent variable ACIND, which means that the presence of the manager within the nomination committee has no impact on the audit committee independence. This means that the audit committee independence cannot be affected when the manager is a member of the nomination committee. Likewise, we found that the variable SENIO is not statistically significant suggesting that the manager seniority does not influence the audit committee independence. This result contradicts that of Klein (1998aKlein ( , 2002b. Finally, we observed a significant correlation at a level of 1% confidence between the variables MOWN and LINEA and the dependent variable ACIND. The relationship between these two variables is nonlinear. Indeed, when the percentage of capital held by the manager is low, a negative relationship has been observed because of the existence of an alignment of interests between the managers and the shareholders. But when managerial ownership increases, a positive relationship was observed (increase in the number of external directors within the audit committee) and this is "Manager Attributes, Firm Characteristics, and Audit Committee Independence: Evidence from the Canadian Context" because of the increased conflicts of interest. This result confirms our hypothesis. .005** **Significant at 1% *Significant at 5%

Model 2: Firm characteristics and audit committee independence
The results of model 2 are provided in Table 3. We notice that the variable BDIND has a positive and significant coefficient at a confidence level of 1% on ACIND suggesting that the board of directors' independence has a positive impact on the audit committee independence. Indeed, the greater the number of outside directors on the board of directors, the more possible it is to have an independent committee. This result is consistent with Menon and Williams (1994), Bradbury (1990), Vafeas (1999), and Julie and Mark (2003). The findings show that the variable BDSIZE has a negative and significant coefficient at a confidence level of 5% with ACIND, which means that the board size has a negative impact on the audit committee independence. This result contradicts that obtained by Klein (2002b). In our opinion, this result can be attributed to the diligence factor not studied in this research. Indeed, the effectiveness of the board of directors can also be perceived through the number of meetings, its authority, and the competence of its members (DeZoort et al., 2002). Table 3 doesn't show a significant relationship between the variables FSIZE and ACIND. This means that the audit committee independence cannot be affected by the firm size. The structure of the audit committee is therefore the same in both large and small companies. This result contradicts that found by Deli and Gillan (2000). Likewise, no significant relationship was found between the variables PERF and ACIND suggesting that the firm performance has no impact on the audit committee independence. This result is consistent with Klein (1998a) but does not confirm our expectations. These implications can be interrelated and linked to the reporting obligations that require Canadian companies to set up audit committees regardless of the size and performance of the firm.
Finally, we found a positive and significant relationship at a confidence level of 1% between the variables INTEN and ACIND. This result seems logical because the existence of a significant portion of intangible assets within the firm can pose problems of evaluation of their value (Whitwell et al., 2007), hence the need to set up an independent audit committee responsible for monitoring the process of accounting for financial information to prevent any attempt to manipulation.

CONCLUSIONS
The study aimed to identify the determinants of the audit committee independence in the Canadian context. We have grouped these factors into two categories: factors related to the manager attributes and factors related to the firms' internal characteristics. The empirical study shows that there is a significant relationship between audit committee independence and manager entrenchment. This is especially evident in the relationship between the presence of the manager within the remuneration committee and the audit committee independence. The study also provides a nonlinear relationship between audit committee independence and managerial ownership. Indeed, when managerial ownership is low, it seems unnecessary to increase the number of outside directors because of the existence of an alignment of interests. As managerial ownership increases, it becomes useful to increase the number of outside directors to mitigate conflicts of interest between the manager and the shareholders. In addition, the study shows the existence of a negative relationship between the presence of the manager on the remuneration committee and the audit committee independence. Indeed, the presence of the manager within the remuneration committee is an indicator of his power, which he can use to neutralize the control mechanisms (including the audit committee). Moreover, codes of conduct indicate that the presence of the manager on the remuneration committee can affect his credibility. It is therefore necessary to prohibit the manager from being a member of this committee.
The audit committee independence is also affected by the board of directors' size. This result is supported by the finding of Yermak (1996), asserting that the effectiveness of the board improves with the reduction in its size. This can be explained by the fact that increasing the size of the board can make communication between its members difficult, which can negatively affect the quality of the control it exercises. In the same way, the study shows the existence of a positive relationship between the board of directors' independence and the composition of the audit committee. Indeed, since the audit committee is made up of directors who are members of the board, therefore, its probability of being independent is higher when the board of directors is independent.
Finally, the study also shows that the audit committee independence increases with the existence of intangible assets. Indeed, the manipulation of financial information becomes easier when the firm owns a significant portion of these types of assets, hence the need to improve the efficiency of the control system (including the audit committee) by ensuring its independence.