Getting inside the black box: a field study of practices in "effective" audit committees.
Most regulators and practitioners view the audit committee (AC) as an important corporate governance mechanism deemed to improve the quality of financial reporting (e.g., Dey 1994; KPMG 1999). However, in contrast to the significance typically conferred to the AC, knowledge is scant about what AC members actually do in meetings (DeZoort et al. 2002; McMullen 1996). Prior studies dealing with the effectiveness of ACs indeed mostly examine the relationship between the AC's inputs (e.g., independence and knowledge of members) and outputs (e.g., publicly reported deficiencies in financial statements). Process issues have been widely neglected. Process issues matter, however, since ACs made up of independent and knowledgeable members are unlikely to be effective if members remain passive in meetings.
For example, although the AC of Enron was made up exclusively of outside directors, included at least one financial expert, and met four times or more a year, the committee did not "probe the [external auditor] independence issue, nor did it initiate the type of communications with Andersen personnel that would have led to its discovering Andersen concerns with Enron accounting practices" (U.S. Senate 2002, 57).
This paper provides insights into practices that AC members carry out in meetings. Specifically, we seek to answer the following questions:
* What matters do AC members emphasize in meetings?
* How do AC members evaluate these matters?
* How do AC members assess responses and comments made by managers and auditors during meetings?
Given the difficulty in assessing the effectiveness of ACs from an ex ante point of view (notably before corporate scandals occur), a better understanding of practices carried out in AC meetings constitutes a valuable knowledge base to assess and perhaps improve the effectiveness of ACs. We conducted the investigation using the field-study method at three large public corporations listed on the Toronto Stock Exchange (TSX). To our knowledge, our study is the first to gather interview data in several public corporations from the key groups of individuals who attend AC meetings, namely, board members, managers, and auditors.
In addition to complying to a large extent with regulatory guidelines of the TSX (Dey 1994) and of the Blue Ribbon Committee (BRC 1999), the three participating ACs are generally viewed as effective by the individuals who attend their meetings. Our paper therefore provides information on practices carried out in normatively compliant and perceptively effective ACs. The practices that we document may provide a benchmark for attendees in other corporations to evaluate their AC. Our findings may also allow regulators to assess the extent to which practices in normatively compliant ACs meet their expectations.
We did not seek to assess the extent to which the practices detailed in the paper are common among other corporations, nor whether these practices are actually effective in preventing financial reporting deficiencies. Our study should rather be conceived as an initial step in the undertaking of research on AC practices. In particular, although skill in questioning is often seen as a prime quality of AC members, we found that a key aspect of the work carried out by audit committee members consists of asking challenging questions and assessing responses provided by managers and auditors. Whether this finding can be generalized to other normatively compliant ACs is an open question.
The paper is organized as follows. The next section locates our research within the literature on the effectiveness of ACs. The third section describes the research method. Field evidence is presented in the fourth section. The last section emphasizes the paper's main findings and suggests areas for future research.
PRIOR RESEARCH ON THE EFFECTIVENESS OF AUDIT COMMITTEES
A growing number of studies have investigated AC effectiveness in the last 15 years. Findings from large-sample archival studies generally suggest that the quality of financial reporting in corporations that have an AC is higher than that of corporations that do not have an AC (Beasley 1999). For example, corporations with an AC have been shown to be less likely to have financial reporting problems such as quarterly earnings restatements or investigative actions undertaken by regulatory agencies (Beasley et al. 2000). Gibbins et al.'s (2001) survey, however, casts doubt on the effectiveness of ACs. They found that ACs are often not involved in negotiations between the external auditor and the auditee concerning significant financial reporting issues.
Researchers have also sought to identify factors that contribute to the effectiveness of the AC (DeZoort et al. 2002). In general, large-sample archival studies suggest that the quality of financial reporting is positively associated with: independence of AC members (e.g., Beasley et al. 2000), members' knowledge and experience in financial reporting (e.g., McMullen and Raghunandan 1996), and frequency of meetings (e.g., Abbott et al. 2000). Further, Carcello and Neal (2003) found that ACs with greater independence and governance expertise are more likely to shield external auditors from dismissal after the issuance of new going-concern reports. Several surveys also show that member independence and knowledge are positively correlated with measures of effectiveness, such as the extent to which members support the auditor in auditor-corporate management conflict situations (e.g., DeZoort and Salterio 2001; Knapp 1987), or the extent to which members oversee the internal auditing function (Raghunandan et al. 2001). Kalbers and Fogarty's (1998) survey, however, indicates that member independence is not correlated with CFOs' and auditors' perceptions of AC effectiveness.
Although a few discordant studies are noted, the stream of research described above generally suggests that ACs are positively related to several measures of effectiveness, and points to member independence and knowledge as significant determinants of effectiveness. These results, however, have to be interpreted cautiously. First, the effectiveness of the AC is, in and of itself, an elusive and multidimensional notion that is difficult to measure (DeZoort 1998). For example, archival studies generally measure effectiveness by taking into account only those deficiencies in financial statements that are reported, or by relying on proxies such as discretionary accruals. Second, studies have relied on a "black box" approach to investigate effectiveness. That is, archival studies' analysis is restricted to the examination of the relationships between externally observable features of ACs and indicators of effectiveness (Kalbers and Fogarty 1998). Archival studies therefore do not provide any meaningful information on activities performed by AC members. Surveys only provide limited insight into AC activities since respondents answer structured questions and complete fictitious cases individually, in a way that prevents the effects of group dynamics from occurring (DeZoort and Salterio 2001).
A few field studies, however, provide insight about AC activities and the extent to which they are deemed effective. For example, Canadian auditors interviewed by Gibbins et al. (2001) highlight wide differences in the effectiveness of ACs. U.S. auditors interviewed by Cohen et al. (2002) describe AC meetings as being generally characterized by members" passivity, and the number of questions asked by members being relatively small. Their interviewees also perceive ACs as not powerful enough to resolve contentious matters with top managers. Spira's (1999) interviews with U.K. informants (AC members, CFOs, auditors) having a relationship with various ACs suggest that corporate managers exert much more influence on the AC than auditors. Managers exert influence notably through the control that they have on various aspects of AC meetings, such as the selection of information to send to members before meetings. Moreover, several of the managers and auditors that Spira (1999) interviewed underlined that questions asked by AC members are unlikely to be sufficiently penetrating to challenge managers, and that members usually are satisfied with vague answers such as "We're going to sort it." In sum, field studies generally view skill in questioning as a prime quality of AC members, but their results cast doubt on members' ability to be skillful questioners. In contrast, our fieldwork indicates that AC effectiveness is a strongly held perception among interviewees in each of our participating corporations. (1)
Research methods typically are categorized as focusing either on scope or depth. While large-sample studies provide a generalizable set of findings pertaining to a few pre-determined constructs, in-depth studies such as field research produce much more detailed information but about a limited number of people and cases (Patton 1990). Selecting a research method therefore involves a tradeoff between generalizability and understandability. We decided to focus on the latter, given the scarcity of knowledge about concrete practices of ACs. The field study method is also particularly relevant when the researcher seeks to better understand phenomena in their naturally occurring settings (Patton 1990).
Selection of the Participating Corporations
We approached 17 public corporations listed on the TSX to participate in the research and sent a summary of the research project to all of them. Ten of these corporations did not answer our numerous attempts to contact them. Four others answered negatively. The other three agreed to participate. These corporations are herein referred to as Corporations "A," "B," and "C." (2)
We mainly used interviews to gather data on AC activities. (3) Specifically, we relied on the "interview guide approach" (Patton 1990, 288) in order to make data collection somewhat systematic for each respondent, while keeping the interview fairly conversational and situational. Prior to fieldwork, we developed an interview instrument that specified certain themes to discuss while being flexible enough to explore emerging paths during the interview. The instrument was developed based on our research objective, literature review, and our experience and intuition. We customized the instrument for each category of interviewees. (4) Interviewees were asked about the process of a typical AC meeting, the extent to which the company's AC fulfills its mandate, the concerns that AC members manifest during meetings, and the type of questions that they ask. (5)
We took several steps to increase the reliability of the field study. First, we began each interview by describing the objective of the research and by introducing an informed consent form, which both the interviewer and interviewee needed to sign before going through the interview questions. We then asked interviewees for permission to tape the interview, while emphasizing that complete anonymity would be provided to them and to their respective organization, and that no other organization member would examine the interview transcript. (6) Finally, we told participants that they would have the opportunity to subsequently verify the accuracy of their transcript and add changes that they felt might be needed to make them comfortable with what they said during the interview. (7) Second, in each corporation we gathered evidence from multiple interviewees. Third, we regularly asked interviewees to provide concrete examples to substantiate their thoughts. Finally, it is worth noting that five interviewees referred, on their own initiative, to other ACs in which they are involved and whose activities are problematic in terms of effectiveness. A number of interviewees therefore appeared not to be biased toward overstating the effectiveness of ACs.
In all, we interviewed 22 individuals (7 in Corporation A; 8 in Corporation B; 7 in Corporation C). Most interviews were conducted at the interviewee's office. In each corporation, we had a meeting with the following persons: CEO, CFO, chief internal auditor, partner in charge of the external audit engagement, chairperson of the AC, and two AC members. (8) Importantly, we had access to the majority of AC members in each corporation. The length of interviews varied from 45 to 75 minutes. All interviews were tape-recorded and transcribed, and were conducted by one of the authors.
We analyzed the interview transcripts via grounded theory procedures (Gibbins et al. 1990; Strauss and Corbin 1990). Significant words or phrases were first highlighted in a few transcripts to develop a coding scheme. We then used this scheme to code the other transcripts--modifying the scheme when new themes were emerging from the data. After coding was completed, we developed a conceptual matrix for each corporation to summarize the main themes discussed by the interviewees and to examine the relationships that might exist among these themes. To further ensure the validity of the interviews, we also made sure that interviewees' answers within the same corporation were converging (Yin 1989). The three conceptual matrices were then used to prepare an outline for the paper, which we discussed in a face-to-face meeting. (9)
General Overview of the Participating Corporations
Table 1 provides descriptive information for each participating corporation. In particular, two of them (A and C) are medium-sized financial institutions. Also, while in Corporations A and C audit fees account for most of the fees charged by the external auditor, in Corporation B the annual proportion of audit fees to total fees has in the last few years been about 50 percent.
Table 2 compares along corporate governance features (at the time of the interviews) the three participating corporations to the 14 that we contacted but did not participate. It also provides comparative information for a random sample of 260 firms listed on the TSX (for year 2000). On average, the three participating corporations comply to a large extent with the TSX governance guidelines. Compared to the sample of 260 firms and the 14 nonparticipating entities, (10) Corporations A, B, and C rank in the top 3 deciles (8, 10, and 9 respectively) for their level of compliance with the guidelines. (11) Also, as for 64 percent of the 260 firms and 50 percent of the 14 nonparticipating entities, the participating corporations do not have the same person acting as both chairperson of the board and CEO. (12)
Information on ACs is categorized in Table 2 using the four components that DeZoort et al. (2002) highlight as contributing to AC effectiveness, namely, composition, authority, resources, and diligence. Compared to the ACs of the 260 firms, the participating ACs rank very well in terms of independence and expertise. (13) All three meet the TSX guideline that states that the AC should be comprised of only outside directors. In Corporations A and C the AC also meets the BRC recommendation that specifies that the AC should be composed only of outside, nonrelated directors. In terms of expertise, all three ACs meet the recommendations of the BRC (1999) report and the Joint Committee on Corporate Governance's (2001) final report--both of which propose that at least one AC member should have "accounting or related financial expertise."
Table 3 provides further information on members' expertise. In Corporation A, every member either is a professional accountant or holds a M.B.A. Two members also have experience as top managers of financial institutions (one as CEO, the other as CFO). Another member is experienced in financial institution auditing. In Corporation B, one member is a CA and another holds an M.B.A. In addition, the majority of the members are, or have been, CEO/CFO in large corporations. In Corporation C, two members occupy, or have occupied, the highest management position in large organizations. Therefore, AC members in the corporations under study have a financial and accounting background that goes beyond what the BRC recommends.
Practices of Sampled Companies' Audit Committees
Table 4 provides information on documentary and logistical features of AC meetings. In particular, members in every participating corporation are provided with the opportunity to read information prior to the meetings. Private meetings take place in every corporation--though there are significant differences in their frequency and participants. Also, in contrast with Spira's (1999) study that indicates that meetings are attended by up to 20 people, our data shows that about ten individuals participate in the meetings. In addition to AC members, attendees typically include the internal and the external auditors, the CEO, CFO, and corporate secretary.
Our analysis of the interviews shows that practices carried out in AC meetings generally aim to make members comfortable with regard to several matters (Pentland 1993). Table 5 highlights these matters and the corresponding evaluation criteria that members use in meetings to assess written and verbal information submitted by managers and auditors.
Members in every participating AC are concerned about the accuracy of financial statements. Several of them stressed that they are very worried about the ways in which readers could misinterpret the company's financial statements, and about the eventuality that the company would have to restate statements already approved by the AC. Interestingly, one member metaphorically compared the process by which members evaluate the accuracy of financial statements to a "smell test":
I find that generally applying, if you will, common business sense to what I see as the results. And generally the issues that I raise aren't because of my formal training as a CA, but rather things that don't pass the smell test. That you look at and say this doesn't really represent what I understood to be happening in the [corporation], what I understood to be the progress of the [corporation]. So it is that sort of test you are applying. (AC member, Corporation A)
Our interviews provide insights into the criteria used by members to evaluate financial information in accordance with the "smell-test" process. Coherence is one of these criteria. Members reportedly seek to ensure that submitted figures are coherent with regard to figures that may be expected, given trends in data and members' understanding of the impact that contextual events have on the company's accounts. As such, the judgmental process by which members assess coherence bears a significant resemblance to the external auditors' analytical process, as described by Hirst and Koonce (1996), in which auditors develop expectations from contextual information, compare them to financial statement figures, and then seek explanations for material differences. In contrast with the auditors' analytical process, however, the way in which AC members assess coherence is less structured than in audit firms. It also takes place in an arena where members, managers, and auditors collectively debate the validity of any incoherence claim.
Another criterion used by members in Corporations A and B to evaluate accuracy is the extent to which the balances of "softer" accounts are justifiable. For example, in Corporation A members especially seek to become comfortable with loan loss provisions by examining the mathematical formulas upon which they are predicated. Members pay careful attention to the logic that underlies the formulas--to make sure that they will be recognized as rational in the event that people (in court or elsewhere) cast doubt on the amounts involved. Justifiability concerns were not mentioned by Corporation C's interviewees. AC practices therefore appear to vary in accordance with the corporate context.
In Corporations A and C (but not in B), members also evaluate accuracy by considering the extent to which the financial statements are understandable. For example, members in Corporation A reportedly seek to avoid expensing restructuring charges--these charges being perceived as nebulous and difficult to grasp by readers.
Corporations also differ in the extent to which AC members are concerned about the appropriateness of the wording used in the narrative part of financial reports. Members showed wording concerns in Corporations A (regarding the possibility of potentially misleading disclosures) and C (regarding understandability). The following quote describes how members in Corporation A seek to make sure that the message to shareholders found at the beginning of financial reports cannot be misinterpreted by readers:
We review on a line-by-line basis the message to shareholders to make sure there is nothing misleading in there whatsoever. Generally speaking this takes a little bit of time because there is always some point. Maybe we have made an acquisition and now we have additional shares outstanding or something like that and we want to make sure that the reader understands that the earnings per share may be affected by the number of outstanding shares or things of that nature. Likewise we want to make sure that there is nothing that can be construed as overly optimistic or a promise of future results. (AC member, Corporation A)
This quote suggests that members examine the content of the message to shareholders by conducting an intellectual simulation (Schon 1983) or a simulation heuristic (Kahneman and Tversky 1982), in which they seek to predict how readers are likely to interpret every sentence. Such a simulation may increase the ability of the AC member to consider whether a sentence is phrased in an overly optimistic way (Galinsky and Moskowitz 2000). Like any other heuristic, however, it may be "subject to characteristic errors and biases" (Kahneman and Tversky 1982, 207).
In contrast with Corporations A and C, members' evaluations of financial reports in Corporation B are less focused on the qualitative narration than on quantitative information. (14) AC members in each corporation therefore use different interpretive schemes to make sense of financial reports.
In all three corporations, members had an interest in the extent to which internal control is effective--not least because internal control "underlies" the credibility of financial reports. To assess control effectiveness, members reportedly ensure that control systems cover the company's key risks, notably the financial ones. As highlighted by one member:
[We] would be probing to be satisfied that the perceived high-risk areas are getting the proper attention, [such as] areas that had a problem before for internal control, where there might be major new systems being integrated into the company's operation ... So there will be questions along the lines that would satisfy the Board that none of the high-risk areas have been overlooked. (AC member, Corporation B)
Members especially rely on the work of the internal auditor to develop their own appreciation of the controls' effectiveness. As highlighted by one AC member in Corporation C: "Internal auditing is almost like an insurance policy, a type of surveillance device that prevents a company from making mistakes. In the financial industry in particular, it is quite easy to get involved in a situation where suddenly you wake up and it is too late. So internal auditing is useful." A recurrent theme in our interviews is that members become comfortable with internal control by assessing the extent to which managers adopt appropriate measures to solve deficiencies highlighted in internal audit reports. Deficiencies indeed generate many questions:
[For unsatisfactory items], number one we [i.e., AC members] want to know what it is. We then get the total internal audit report for that audit, so we can read it. Let us say one of the unsatisfactory things, and this is just an example, is they are just six months behind on reconciling the bank accounts. Well, we would say: "When is it going to be done? And we want you, the internal auditor, when it is done, to go back in and check and see that it is done and tell us." Because we always have a follow up ... We also want management's response ... We want management's response to see if they agree or disagree, and why they disagree if they disagree. And in most cases they say: "We agree and we will have it fixed by such and such a date." So there is always a follow up to ensure it is done. (AC member, Corporation B)
Also, members in all three ACs are aware that to fulfill their responsibilities, they are dependent on the quality of the work performed by the company's internal and external auditors. For example, one member metaphorically referred to auditors as the "eyes and ears" of the AC in the organization. Members perceive auditor performance in terms of competence and independence. To assess auditor competence, members notably make sure that audit plans reflect their understanding of the impact that contextual events have on audit risks, and examine the profile of audit team members. Auditors, then, are judged on the responses they provide "to the extent that they are knowledgeable, that they are candid with us" (AC member, Corporation A).
In Corporations B and C, members' evaluation of auditor performance also takes independence into account. In Corporation C, members consider it important that their internal auditor regularly expresses points of view that conflict with those of management. In Corporation B, members devote significant energy during meetings to assess the independence of the external auditor. One member mentioned that by constantly "probing and pushing" the external auditor on the matter, s/he hoped that the auditor becomes convinced that it is the AC that ultimately "drives" external auditing--not the CEO. Members' expectations are that raising and discussing independence openly with both managers and auditors may avoid members later being confronted with unacceptable situations. Corporation B's external auditor reportedly has responded to independence challenges raised by members by highlighting the risk of the firm's reputation being tarnished and the nature of the firm's partner-compensation scheme. However, one AC member in Corporation B mentioned that, due to regulators' recent concerns about independence (e.g., Brown 1999), s/he was considering changing her/his point of view about the appropriateness of the current level of nonaudit fees earned by the external auditor. Thus, a few months before Enron's debacle, external auditors appeared to walk on thin ice when attempting to persuade AC members that they are independent from managers no matter how much nonaudit fees they earn in addition to audit fees (see also Frankel et al. 2002).
In contrast with Corporation B, members in Corporations A and C are much less concerned about external auditor independence and members do not discuss it extensively during meetings. (15) Recall that in Corporations A and C nonaudit fees represent a relatively small proportion of the total fees earned by the audit firm, while this proportion is significantly higher in Corporation B. It appears that AC members in these two corporations adapt their practices to the organizational context and keep their attention focused on other matters.
Table 6 provides details about matters emphasized and the corresponding evaluation criteria used by AC members during the private part of the meetings. Panel A deals with the private meeting with auditors while Panel B refers to that with management. Panel A indicates that, in each corporation, private meetings with auditors deal with the quality of the relationship that they have with corporate management. In this respect, auditors are questioned about whether difficult issues emerged during the course of their audits, and whether they had unrestricted access to information. (16) AC members in corporations A and B also use private meetings with auditors to inquire about the competence of management, asking questions such as: "Are there enough accounting people to do the job? What is the staffing like? What is the morale like?" (AC member, Corporation B)
In summary, our fieldwork shows that deriving comfort from reviews and challenging questions is a pivotal aspect of the work carried out by AC members. One member indeed highlighted that AC members are involved in an "ask the right question" game. In fact, our data suggests that the ability to ask "right" questions is key for members to become more comfortable in their role as AC members, and in constructing members' legitimacy in the eyes of other attendees. For example, when describing the effectiveness of B's AC, the external auditor highlighted that the company's AC members ask "relevant" questions--not the "generic" ones found in checklists distributed by her/his own accounting firm or questions about the "basics of financial statements and internal controls." Interestingly, managers in the three corporations mentioned that they appreciate being confronted with challenging questions, which implies that members' ability to ask such questions is a key dimension of AC effectiveness in the eyes of managers.
Finally, our fieldwork provides a better understanding of the criteria used by members to evaluate answers provided by managers and auditors to the questions asked during meetings. For example:
So, I think our audit committee, particularly our Chair, likes to question people. Then [s/he] will get to a point, and so will the committee, where they are comfortable with the answers you have been able to give them over time. Sometimes, you may see them focus on something that you might not think important, but really they are not focusing on that particular thing but on the person presenting it, to see if that person has enough knowledge and understanding of all of the implications and ramifications of what they are talking about. That is a way for them to get comfort that the person is capable and knowledgeable. And they would always look for consistency among all the various people that they hear from; that we are all singing from the same song sheet. (CFO, Corporation A)
This excerpt suggests that members assess whether individuals having financial and control responsibilities are capable of experimenting intellectually with the diverse possible outcomes of a given situation (Schon 1983). The quote also indicates that members assess the extent to which the diverse parties they question present a consistent picture of the situation. Furthermore, not only are members sensitive to consistency across respondents at a specific point of time, but also they take into account the respondents' record concerning the extent to which their past statements, claims, and points of view proved to be true. Indeed:
Over time a committee develops a level of confidence and trust and you just gain that by the dialogue and interaction. And time has a way of confirming what representations have been made to you previously. And through that you develop the trust and confidence. (AC member, Corporation A)
This quote (and several others) implies that AC meetings are not mere rituals devoid of interest to managers and auditors. Managers and auditors have their reputation (in providing acceptable answers to members' questions) at stake during meetings. Given members' interest in the consistency of answers provided by managers and auditors over time, it seems crucial to the effectiveness of the AC that members sit on the committee for more than a few years--in order for them to develop a thorough understanding as to whether answers provided in past meetings by managers and auditors are reliable. As indicated in Table 2, a majority of the members in the three corporations that we studied have sat on the AC for at least three years.
This paper sought to better understand practices that AC members carry out in meetings. Our fieldwork shows that these practices aim to make members comfortable with regard to matters such as the accuracy of financial statements and the quality of the work performed by auditors. Thus, not only does the notion of comfort play a key role when auditors conduct audit engagements (Pentland 1993), but also it is a fundamental aspect of the work performed by AC members. Deriving comfort from reviews and questions seems to be a key aspect of the way in which various parties confer trust to corporate disclosures. Our analysis also brings to light that meetings in the three participating corporations are much more than formal rituals for managers and auditors, who actually need to demonstrate to AC members that they are trustworthy. Members especially assess trustworthiness through the degree of consistency across responses provided by the diverse parties that they question, and the credibility that these parties develop over time with regard to the validity of their previous answers. Meetings in "effective" ACs may therefore be conceived of as arenas where attendees establish and secure (or not) their reputation of trustworthiness.
Our paper's findings also suggest that one opportunity for future research consists of studying the judgmental processes of AC members (e.g., McDaniel et al. 2002). The trend analyses made by members when evaluating information indeed imply a wide spectrum of cognitive activities, such as forming mental representations as well as generating and testing hypotheses (Solomon and Shields 1995).
Of course, the paper is characterized by limitations. First, our interviews took place before Enron's failure and the impact that this event has had in the business community may have motivated members in our participating ACs to change their practices. Second, our data is unlikely to be as comprehensive as it could have been if, for example, we would have been able to observe meetings. Nonetheless, the in-depth interviews that we conducted provided us with a rich database, and with significant patterns across interviewees' transcripts. Third, the ACs that we studied did not include committees whose perceived effectiveness and degree of regulatory compliance are moderate or low, thereby preventing us from comparing practices between more "effective" and less "effective" ACs. Other researchers may use our findings as a basis for undertaking this comparison. Fourth, our paper did not focus on the examination of antecedents and consequences of members' ability to ask challenging questions. For example, whether the ability to ask challenging questions corresponds to members' background characteristics as described in corporate proxies is an open question. Further, in spite of the general perception held by attendees about the effectiveness of their ACs, it remains to be seen whether the committees would actually make a difference in resolving serious conflicts between managers and auditors--in a way that is consistent with shareholders' interests. Although interviewees often claimed that their AC would be able to deal appropriately if confronted with a significant dispute between managers and auditors, these claims were intentional in nature since interviewees were unanimous in highlighting that their respective ACs had never been confronted with such disputes. Further research is needed, first, to examine how AC members react when a serious and contentious matter comes to their attention and, second, to validate the low frequency of conflicting situations to which ACs are confronted--given that prior studies tend to assume a much higher frequency (e.g., DeZoort and Salterio 2001; Knapp 1987).
In conclusion, this paper demonstrates that researchers are able to have access in public corporations to data on AC activity through interviews with attendees of AC meetings, thereby allowing academics to get into the "black box" and investigate process issues. Although the potential of field research to generate knowledge is highlighted in auditing literature (e.g., Solomon and Shields 1995), very few field-based studies have been carried out on AC processes. Obstacles to field research on corporate governance processes, such as practitioners' time pressures and confidentiality concerns, can be overcome.
TABLE 1 Descriptive Features of the Participating Corporations Feature Corporation A Industry Financial services Financial characteristics (a) Book value of total Within the range of $1 assets (Canadian dollars) billion to $5 billion Return on common Consistently between shareholders' equity 10% and 15% within the last 5 years Dispersion of ownership (b) Higher: no shareholder possesses more than 10% of the company's voting shares (d) External auditor (c) Number of years in More than 10 years which the current audit firm has been in charge of the external audit engagement Annual proportion (in About 80% the last few years) of audit fees to total fees charged by the external auditor Feature Corporation B Industry Service industry (nonfinancial) Financial characteristics (a) Book value of total Within the range of $5 assets (Canadian dollars) billion to $10 billion Return on common Consistently between shareholders' equity 10% and 15% within the last 5 years Dispersion of ownership (b) Lower: the company has one significant corporate block-holder, which however does not have control of the company's outstanding voting shares External auditor (c) Number of years in More than 10 years which the current audit firm has been in charge of the external audit engagement Annual proportion (in About 50% the last few years) of audit fees to total fees charged by the external auditor Feature Corporation C Industry Financial services Financial characteristics (a) Book value of total Within the range of $10 assets (Canadian dollars) billion to $15 billion Return on common Consistently between shareholders' equity 10% and 15% within the last 5 years Dispersion of ownership (b) Higher: no shareholder possesses more than 10% of the company's voting shares (d) External auditor (c) Number of years in More than 10 years which the current audit firm has been in charge of the external audit engagement Annual proportion (in Above 80% the last few years) of audit fees to total fees charged by the external auditor (a) As indicated in the corporation's latest annual report (at the time of the interviews). (b) As indicated in the corporation's latest management proxy circular (al the time of the interviews). (c) Information provided by interviewees. None of the participating corporations voluntarily discloses in financial reports the amount of audit and nonaudit fees charged by the external auditor. (d) As required by financial institution regulation in Canada. TABLE 2 Comparison of Participating and Nonparticipating Corporations to a Random Sample of 260 Corporations Listed on the TSX Random Sample of 260 Corporations Listed on Nonparticipating the TSX Corporations Dimension (year 2000) (n = 14) Board Score on TSX guidelines [max 14] (a) 8.7 9.7 (Decile) (NA) (NA) CEO is also chairperson of the 36% 50% board Audit Committee (AC): Composition Independence AC comprises solely outside 73% 93% directors (b) AC comprises solely outside, 43% 71% unrelated directors (c) Expertise At least one outside director 36% 50% on the AC is a financial expert (d) At least one outside director NA NA on the AC has experience serving on other ACs as an outisde director Proportion of members who sat 61% 77% on the AC during the last 3 years Audit Committee: Authority Formal oversight responsibilities NA NA (FR: financial reporting; FA: financial auditing; IC: internal control) Audit Committee: Resources Number of individuals serving 3.4 3.8 on the audit committee Direct access to auditors NA NA Access to legal advice NA NA Audit Committee: Diligence (Process) Number of meetings per year 3.8 (e) 4.6 (f) Corporation Dimension A B C Board Score on TSX guidelines [max 14] (a) 10.8 12.8 12.4 (Decile) (8.) (10.) (9.) CEO is also chairperson of the No No No board Audit Committee (AC): Composition Independence AC comprises solely outside Yes Yes Yes directors (b) AC comprises solely outside, Yes No Yes unrelated directors (c) Expertise At least one outside director Yes Yes Yes on the AC is a financial expert (d) At least one outside director Yes Yes Yes on the AC has experience serving on other ACs as an outisde director Proportion of members who sat 100% 60% 75% on the AC during the last 3 years Audit Committee: Authority Formal oversight responsibilities FR; FR; FR; (FR: financial reporting; FA; IC FA; IC FA; IC FA: financial auditing; IC: internal control) Audit Committee: Resources Number of individuals serving 4 5 4 on the audit committee Direct access to auditors Yes Yes Yes Access to legal advice Yes Yes Yes Audit Committee: Diligence (Process) Number of meetings per year 5 4 6 (a) The TSX Corporate Governance Guidelines, as stated in Sec. 474 (13) of the Toronto Stock Exchange Company Manual (at the time of the interviews), consists of 14 guidelines on the role, composition, and powers of the board and its subcommittees, such as the nominating committee and the audit committee. The TSX requires listed corporations to disclose on an annual basis the extent to which they follow the 14 guidelines--either in their proxy statement or annual report. We read the compliance disclosure of every one of the 260 firms included in the sample. A code of 1 was given each time a firm complied with one of the 14 guidelines. We calculated for each firm a global score of compliance by adding up the firm's individual codes. The maximum global score that a firm could receive is 14. (b) An outside director is a director who is not "an officer or employee of the corporation or of any of its affiliates" (Dey 1994, 26). (c) An unrelated director "is a director who is independent of management free from any interest and any business or other relationship which could, or could reasonably be perceived to, materially interfere with the director's ability to act with a view to the best interests of the of the corporation, other than interests and relationships arising from shareholding" (Dey 1994, 4). (d) A member is classified as a financial expert when s/he holds a professional certification in accounting (CA/CPA) or in financial analysis (CFA), or when s/he has experience in finance or accounting. (e) Only 65 corporations provided that information in their proxy statement or annual report. (f) Only 7 corporations provided that information in their proxy statement or annual report. TABLE 3 Additional Information on the Background of Audit Committee Members in the Participating Corporations Whether Member Holds a Professional Whether Accounting Member Was Designation Member Interviewed (CA or CPA) Corporation A Member 1 Yes No (chairperson) Member 2 Yes Yes Member 3 Yes Yes Member 4 No Yes Corporation B Member 5 Yes Yes (chairperson) Member 6 Yes No Member 7 Yes No Member 8 No No Member 9 No No Corporation C Member 10 Yes No (chairperson) Member 11 Yes No Member 12 Yes No Member 13 No No Whether Member Is Member Other Qualifications Unrelated (a) Corporation A Member 1 M.B.A.; former employee Unrelated (chairperson) and top manager in two financial institutions; former director of two financial institutions; currently CEO of a private holding and investment company Member 2 Retired partner of a Big 5 Unrelated firm; when active, was in charge of the audit of a large financial institution Member 3 Currently CFO of a large Unrelated financial institution Member 4 Currently CEO of a private Unrelated holding and investment company Corporation B Member 5 Former CEO of a large Unrelated (chairperson) public company; currently CEO of a private holding and investment company Member 6 M.B.A.; former CFO of a Unrelated large private company (service industry); currently CEO of a private holding and investment company Member 7 University courses in Related, but business and economics; outside director currently CEO of a public company (service industry) Member 8 Currently partner Related, but in a law firm outside director Member 9 Top manager at a corporate Related, but block-holder of corporation outside director B's common stocks Corporation C Member 10 Former CEO of a large Unrelated (chairperson) public sector institution; currently CEO of another large public sector institution Member 11 Degree in law; M.B.A.; Unrelated former CEO of two companies (one being a state-owned corporation, the other a subsidiary of a foreign public corporation) Member 12 Lawyer; former top manager of a financial institution; currently business consultant Member 13 Degree in business; former comptroller of a medium-sized Unrelated businessl currently president of a private holding and investment company (a) As indicated in the corporation's latest annual report or management proxy circular (at the time of the interview's). An unrelated director "is a director who is independent of management and is free from any interest and any business or other relationship which could, or could reasonably be perceived to, materially interfere with the director's ability to act with a view to the best interests of the corporation, other than interests and relationships arising from shareholding" (Dey 1994, 4). TABLE 4 Documentary and Logistical Features of Audit Committee Meetings (a) Feature Corporation A Corporation B Documentation Associated with Quarterly Meetings Agenda Predefined in Predefined in accordance with a accordance with a schedule of routine schedule of routine activities; adapted to activities; adapted to reflect nomecurrent reflect nonrecuurent items items Information sent in advance to members Content Draft of financial Draft of financial reports and proxies; reports and proxies; press releases; press releases; reports from internal reports from internal and external auditors; and external auditors; information on riskier environmental assets; minutes of reports ... last meeting ... Thickness About one inch About one inch Range of time to From 1 1/2 to 3 hours From 3 to 5 hours read Logistical Features of Quarterly Meetings Duration of meeting About 3 hours From 2 to 4 hours (including the private meeting(s)) Frequency and At every regular At every regular duration of private meeting, members meet meeting, members meet meeting(s) privately with the privately with the external and the external auditor and internal auditor then with the internal together. The private auditor. Subsequently, meeting last normally members have a meeting between 5 and 20 just among themselves. minutes. The three meetings altogether last normally between 10 and 15 minutes. Also, once a year members meet privately with the CFO--for about 15 minutes. People present Audit committee Audit committee during the members; internal and members; internal and non-private part external auditors; external auditors; CEO of the meeting CEO; CFO; chairperson (but the interim CEO of another board's at the head of the committee (who is not corporation in 2000 member of the AC) did not always attend); CFO; corporate secretary Feature Corporation C Documentation Associated with Quarterly Meetings Agenda Prepared from prior agendas and minutes; adapted to reflect nonrecurrent items Information sent in advance to members Content Draft of financial reports and proxies; press releases; reports from internal and external auditors; balanced scorecard information; listing of complaints and legal proceedings against the company; minutes of last meeting ... Thickness About one inch Range of time to From 2 to 6 hours read Logistical Features of Quarterly Meetings Duration of meeting From 1 1/2 to 3 hours (including the private meeting(s)) Frequency and Members meet privately duration of private with the external meeting(s) auditor once a year. Members also meet privately once a year with corporate managers and the internal auditor together. Each of these meetings lasts about 15 minutes. People present Audit committee during the members; internal non-private part auditor; external of the meeting auditor (attends the majority of meetings); CEO; CFO; chief accountant; corporate secretary (a) As reported by interviewees. TABLE 5 Matters Emphasized and Evaluation Criteria Used by Audit Committee Members during Meetings (a) Evaluation Criteria Matter Corporation A Corporation B Financial Statements Are financial Coherence with regard Coherence with regard statements to trends or to trends or accurate? contextual events contextual events Justifiability of Justifiability of softer accounts (e.g., softer accounts (e.g., loan loss provisions) restructuring charges) Understandability (a) Is the wording in No potentially financial reports misleading disclosures appropriate? Internal Control Is internal Risk coverage Risk coverage control effective? Appropriateness of Appropriateness of corrective actions corrective actions taken by management taken by management on control weaknesses on control weaknesses highlighted by highlighted by internal auditor internal auditor (a) (a) Auditing Do auditors Competence Competence perform appropriately? (b) NA NA (a) Independence of the external auditor Evaluation Criteria Matter Corporation C Financial Statements Are financial Coherence with regard statements to trends or accurate? contextual events (a) Understandability Is the wording in Understandability financial reports appropriate? Internal Control Is internal Risk coverage control effective? Appropriateness of corrective actions taken by management on control weaknesses highlighted by internal auditor Justifiability of management's position when it refuses to implement internal auditor's recommendations Auditing Do auditors Competence perform appropriately? (b) Independence of the internal auditor (a) (a) Gray shaded cells indicate either that interviewees did not raise and discuss the criterion, or that interviewees' discussion suggests that the criterion only plays a minor role in members' evaluations. (b) Due to time constraints, interviews in Corporations A and B did not deal with the independence of the internal auditor. In Corporation C interviewees brought it up during interview without prompting. Note: Gray shaded cells indicate either that interviewees did not raise and discuss the criterion, or that interviewees' discussion suggests that the criterion only plays a minor role in members' evaluations indicated with (a). TABLE 6 Matters Emphasized and Evaluation Criteria Used by Audit Committee Members during the Private Part of Meetings Panel A: Private Meeting with Auditor(s) (a) Evaluation Criteria Matter Corporation A Corporation B Does the external (a) Independence of the auditor perform external auditor appropriately? Is the relationship Occurrence of Occurrence of between external/ difficult issues difficult issues internal auditors and corporate Unrestricted access Unrestricted access management healthy? to information to information Does management Competence Competence perform appropriately? Panel B: Private Meeting with Management (b) Evaluation Criteria Matter Corporation A Corporation B Does the external (a) Degree of management's auditor perform satisfaction with appropriately? regard to the competence of the external auditor (a) Independence of the external auditor Evaluation Criteria Matter Corporation C Does the external (a) auditor perform appropriately? Is the relationship Occurrence of between external/ difficult issues internal auditors and corporate Unrestricted access management healthy? to information Does management (a) perform appropriately? Panel B: Private Meeting with Management (b) Matter Corporation C Does the external Degree of management's auditor perform satisfaction with appropriately? regard to the competence of the external auditor (a) (a) Gray shaded cells in Panel A indicate either that interviewees did not raise and discuss the criterion, or that interviewees' discussion suggests that the criterion only plays a minor role in members' evaluations. (b) Gray shaded cells in Panel B indicate either the absence of a formal private meeting between AC members and management (Corporation A), or that interviewees did not raise and discuss the criterion (Corporation C). Note: Gray shaded cells in Panel A indicate either that interviewees did not raise and discuss the criterion, or that interviewees' discussion suggests that the criterion only plays a minor role in members' evaluations indicated with (a).
We acknowledge the comments provided on earlier versions of the paper by Joane Martel, Michael Power, Steven Salterio, Laura Spira, Arnie Wright, workshop participants at the Ecole des Hantes Etudes Commerciales de Montreal and at Universite Laval, and participants at the 2002 Congress of the European Accounting Association in Copenhagen. We also thank the contact persons and participants that made it possible for us to gather field data in their respective organizations. We finally acknowledge the financial support of the Social Sciences and Humanities Research Council of Canada.
(1) Many comments made by interviewees point to the perceived effectiveness of the three participating ACs. When asked about the handicaps to the accomplishment of the role of their corporation's AC, 13 interviewees (out of the 20 who were asked this question) answered by underlining that their AC works very well or that there is no area needing improvement. Several interviewees also highlighted that their company's AC complies with the latest best practices.
(2) In the "Results" section, we compare along corporate governance features the 14 corporations that did not participate with the three that participated and a random sample of 260 corporations listed on the TSX.
(3) In each corporation we also examined the charter of the AC as well as the information disclosed about the committee in the latest management proxy circular (at the time of the interviews).
(4) The instrument is available from the first author upon request.
(5) Each interviewer had the latitude to ask the questions in his own words, to sequence them differently, and to ask probes and follow-up questions when necessary.
(6) It was key to stress the latter since interviewees were aware that several of their colleagues were to be interviewed.
(7) Only minor modifications resulted from interviewees' revision of their transcript.
(8) We also interviewed B's corporate secretary, who is in charge of producing the minutes of AC meetings.
(9) An earlier version of the paper was sent to the chairperson of every participating AC, the CFO of Corporation A, and the chief internal auditor of Corporation B. These individuals did not report any significant concern regarding the validity of the data and the appropriateness of the conclusions.
(10) Individual scores for the 14 nonparticipating corporations range from 4.2 to 14. Six of these corporations have an individual score higher than 10, thereby indicating a relatively high degree of compliance with the TSX guidelines.
(11) In particular, AC policies in each participating corporation fully comply with guideline #13 pertaining to the AC, which states that the AC should be composed only of outside directors, that the responsibilities of the committee should be defined, that the AC should have direct communication channels with the internal and external auditors, and that the AC's duties should include oversight responsibility for management reporting on internal control.
(12) We also ensured that none of the participating corporations appears in the statements of allegations issued by the Ontario Securities Commission. Statements of allegations from 1997 onward were consulted on the Ontario Securities Commission's website (as of April 8, 2002).
(13) Although regulated corporations in Canada, on average, have lower-quality ACs based on measures of member experience and knowledge (Beasley and Salterio 2001), two of our participating organizations (Corporations A and C) are regulated corporations having a higher-quality AC.
(14) Whether the lower proportion of unrelated directors on B's AC played any role in this between-firm difference is, however, an open question.
(15) It is worth noting that AC members in Corporations A and C are aware of regulators' concerns about accounting firms providing significant management advisory services to auditees. Unawareness therefore does not account for members' not discussing external auditor independence extensively during meetings.
(16) Although no significant issue was ever brought to the attention of AC members that would have led them to doubt the quality of the relationship between auditors and management, members unanimously saw regular private meetings with auditors as useful, allowing members and auditors to interact more freely while avoiding suspicion being raised by managers if these meetings only took place if need be.
Abbott, L. J., S. Parker, and G. F. Peters. 2000. The effectiveness of Blue Ribbon Committee recommendations in mitigating financial misstatements: An empirical study. Working paper, University of Memphis.
Beasley, M. S. 1999. Examining corporate governance research: Any consistencies with calls for reform? The Auditor's Report 23 (1): 6-8.
--, J. V. Carcello, D. R. Hermanson, and P. D. Lapides. 2000. Fraudulent financial reporting: Consideration of industry traits and corporate governance mechanisms. Accounting Horizons 14 (4): 441-454.
--, and S. E. Salterio. 2001. The relationship between board characteristics and voluntary improvements in audit committee composition and experience. Contemporary Accounting Research 18 (4): 539-570.
Blue Ribbon Committee on Improving the Effectiveness of Corporate Audit Committees (BRC report). 1999. Report and Recommendations of the Blue Ribbon Committee on Improving the Effectiveness of Corporate Audit Committees. New York: New York Stock Exchange/National Association of Securities Dealers.
Brown, D. A. 1999. Public accounting at a crossroads. Remarks by the Chair of the Ontario Securities Commission to the Business Leaders' Luncheon, sponsored by the Institute of Chartered Accountants of Ontario. June 8. Available at: http://www.osc.gov.on.ca/en/About/News/Speeches.
Carcello, J. V., and T. L. Neal. 2003. Audit committee characteristics and auditor dismissals following "new" going-concern reports. The Accounting Review 78 (1): 95-117.
Cohen, J., G. Krishnamoorthy, and A. M. Wright. 2002. Corporate governance and the audit process. Contemporary Accounting Research 19 (4): 573-594.
Dey, P. 1994. Where Were the Directors? The Toronto Stock Exchange Committee on Corporate Governance in Canada. Toronto, Canada: Toronto Stock Exchange.
DeZoort, F. T. 1998. An analysis of experience effects on audit committee members' oversight judgments. Accounting, Organizations and Society 23 (1): 1-21.
--, and S. Salterio. 2001. The effects of corporate governance experience and financial-reporting and audit knowledge on audit committee members' judgments. Auditing: A Journal of Practice & Theory 20 (2): 31-47.
--, D. R. Hermanson, D. S. Archambeault, and S. A. Reed. 2002. Audit committee effectiveness: A synthesis of the empirical audit committee literature. Journal of Accounting Literature 21: 21-38.
Frankel, R. M., M. F. Johnson, and K. K. Nelson. 2002. The relation between auditors' fees for nonaudit services and earnings management. The Accounting Review 77 (Supplement): 71-105.
Galinsky, A. D., and G. B. Moskowitz. 2000. Counterfactuals as behavioral primes: Priming the simulation heuristic and consideration of alternatives. Journal of Experimental Social Psychology 36 (4): 384-409.
Gibbins, M., A. Richardson, and J. Waterhouse. 1990. The management of corporate financial disclosure: Opportunism, ritualism, policies, and processes. Journal of Accounting Research 28 (1): 121-143.
---, S. Salterio, and A. Webb. 2001. Evidence about auditor-client management negotiation concerning client's financial reporting. Journal of Accounting Research 39 (3): 535-563.
Hirst, D. E., and L. Koonce. 1996. Audit analytical procedures: A field investigation. Contemporary Accounting Research 13 (2): 457-486.
Joint Committee on Corporate Governance. 2001. Beyond Compliance: Building a Governance Culture. Final report of the Committee, sponsored by the Toronto Stock Exchange, the Canadian Venture Exchange and the Canadian Institute of Chartered Accountants, Toronto, Ontario: Joint Committee on Corporate Governance.
Kahneman, D., and A. Tversky. 1982. The simulation heuristic. In Judgment under Uncertainty: Heuristics and Biases, edited by D. Kahneman, P. Slovic, and A. Tversky, 201-208. New York, NY: Cambridge University Press.
Kalbers, L. P., and T. J. Fogarty. 1998. Organization and economic explanations of audit committee oversight. Journal of Managerial Issues 10 (2): 129-150.
Knapp, M. C. 1987. An empirical study of audit committee support for auditors involved in technical disputes with client management. The Accounting Review 62 (3): 578-588.
KPMG. 1999. Shaping the Audit Committee Agenda. New York, NY: KPMG LLP.
McDaniel, L., R. D. Martin, and L. A. Maines. 2002. Evaluating financial reporting quality: The effects of financial expertise vs. financial literacy. The Accounting Review 77 (Supplement): 139-167.
McMullen, D. A. 1996. Audit comrmttee performance: An investigation of the consequences associated with audit committees. Auditing: A Journal of Practice & Theory 15 (1): 87-103.
--, and K. Raghunandan. 1996. Enhancing audit committee effectiveness. Journal of Accountancy 182 (2): 79-81.
Patton, M. Q. 1990. Qualitative Evaluation and Research Methods. Second edition. Newbury Park, CA: Sage Publications.
Pentland, B. T. 1993. Getting comfortable with the numbers: Auditing and the micro-production of macro-order. Accounting, Organizations and Society 18 (7/8): 605-620.
Raghunandan, K., W. J. Read, and D. V. Rama. 2001. Audit committee composition, "gray directors," and interaction with internal auditing. Accounting Horizons 15 (2): 105-118.
Schon, D. A. 1983. The Reflective Practitioner. New York, NY: Basic Books.
Solomon, I., and M. D. Shields. 1995. Judgment and decision-making research in auditing. In Judgment and Decision-Making Research in Accounting and Auditing, edited by R. H. Ashton, and A. H. Ashton, 137-175. New York, NY: Cambridge University Press.
Spira, L. F. 1999. Ceremonies of governance: Perspectives on the role of the audit committee. Journal of Management and Governance 3:231-260.
Strauss, A. L., and J. Corbin. 1990. Basics of Qualitative Research: Grounded Theory Procedures and Techniques. Newbury Park, CA: Sage Publications.
U.S. Senate. 2002. The Role of the Board of Directors in Enron's Collapse. Report 107-70, Permanent Subcommittee on Investigations of the Committee on Governmental Affairs United States Senate. Washington, D.C.: Government Printing Office.
Yin, R. K. 1989. Case Study Research: Design and Methods. Revised edition. Newbury Park, CA: Sage Publications.
Yves Gendron is an Assistant Professor at the University of Alberta, and Jean Bedard and Maurice Gosselin are Professors, both at Universite Laval.
Submitted: May 2002
Accepted: November 2003
|Printer friendly Cite/link Email Feedback|
|Author:||Gendron, Yves; Bedard, Jean; Gosselin, Maurice|
|Publication:||Auditing: A Journal of Practice & Theory|
|Date:||Mar 1, 2004|
|Previous Article:||Comparing audit team effectiveness via alternative modes of computer-mediated communication.|