Comments of the Standards Committee of the Auditing Section of the American Accounting Association on the SEC's Concept Release No. 33-9862; 34-75344 File No. S7-13-15, Possible Revisions to Audit Committee Disclosures: Participating Committee Members: Free

Dear Mr. Fields:

The Auditing Standards Committee of the Auditing Section of the American Accounting Association is pleased to provide comments on the Release No. 33-9862; 34-75344 File No. S7-13-15, Possible Revisions to Audit Committee Disclosures.

The views expressed in this letter are those of the members of the Auditing Standards Committee and do not reflect an official position of the American Accounting Association. In addition, the comments reflect the overall consensus view of the Committee, not necessarily the views of every individual member.

We hope that our attached comments and suggestions are helpful and will assist the Commission. Please feel free to contact the subcommittee chair should the Commission have any questions about our comments and suggestions.

• Respectfully submitted,

• Auditing Standards Committee

• Auditing Section, American Accounting Association

The Committee commends the Commission for considering requiring additional disclosures related to the work of audit committees of publicly held companies. The following presents a number of specific comments or suggestions on questions posed by the Commission in the related Concept Release.

The document asks if current Audit Committee (AC) reporting is relevant to investors (Q1), whether additional information would be relevant (Q3), and what potential challenges block the ability of preparers and the AC to disclose meaningfully to shareholders. Our response is that auditing is a credibility-enhancing service (Causholli and Knechel 2012), even though the public is not able to observe the quality of audits. Therefore, provision of credible signals of audit quality is crucial to the proper functioning of an audit market. In the absence of such signals, auditing is at risk of becoming a market for lemons (Akerlof 1970). A recent working paper, Reid, Carcello, Li, and Neal (2015), shows that additional disclosure from ACs and auditors provide useful information to investors in the U.K. In this response we outline the types of information required for a well-functioning audit market. Admittedly, there are potential costs associated with a greater level of disclosure, and there is a risk that additional disclosures may create more “disclosure overload.” However, academic research suggests that the benefits of the increased disclosure outweigh the associated costs. Furthermore, investors have become increasingly more interested in enhanced AC disclosure.¹ For example, in 2013, the Council of Institutional Investors (CII) expanded its corporate governance policies to promote greater information to investors about how the AC carries out its responsibilities. Our key concern is that ACs that do not want to provide meaningful disclosure will provide boilerplate disclosures, thus defeating the purpose of mandating disclosure. To forestall such boilerplate, we propose focusing on the disclosure of AC policies and scope, as well as objective information (e.g., number of times the AC meets with the auditor in private) with calls for substantive comments in only very few instances.

There is a long list of questions in the SEC document. To simplify our message we have organized our response into two categories identified by DeAngelo (1981) as being crucial for audit quality, namely:

• (1) 

  Auditor competence so the auditor can detect any breaches in financial reporting, and

• (2) 

  Auditor independence so the auditor reports the findings of his audit.

Generally, independence is considered to be the most important attribute of an auditor, and so we address independence first.

There are some scope and policy issues with respect to the AC that could be of interest to investors. They include:

• (1) 

  The AC's policy with respect to auditor provision of nonaudit services. This issue has been of considerable interest to regulators and large research literature (Ashbaugh, LaFond, and Mayhew 2003; Kinney, Palmrose, and Scholz 2004).

• (2) 

  The scope of the AC's work (Q6) and whether it is limited to financial reporting only, or whether it also accompanies other compliance and risk management functions such as ERM, reporting, and other such issues (Jamal and Sunder 2011).

• (3) 

  Whether the AC sought proposals or has a policy in that regard (Q30) (see Fiolleau, Hoang, Jamal, and Sunder 2013).

There are also several items that could be helpful to assessing independence:

• (4) 

  Auditor tenure (number of years).²

• (5) 

  Significant issues discussed with the auditor (Q11) and how they were resolved. While there is some possibility that such a disclosure could be provided by the auditor, it would be informative for investors to know what issues the AC found to be most important. This has been found to be relevant to investors in the U.K. (Reid et al. 2015).

• (6) 

  Emerging literature suggests that investors highly value auditor independence (Christensen, Glover, Omer, and Shelley 2015). Disclosure of the number of meetings held between the auditor and the AC outside the presence of management would provide assurance about the independence of the auditor and be useful information to investors. While the specific topics discussed need not be disclosed, certainly, the number of these meetings can be useful to investors. Additionally, the Commission should consider disclosure about the AC's oversight of the internal audit function. Specifically, meetings with the chief internal auditor, and its involvement and oversight of the internal audit function would provide relevant information to investors given the important role the internal audit function plays in the financial reporting process.

In a recent report, the Center for Audit Quality (CAQ 2015) noted that investors are seeking greater disclosure about how the AC oversees the external auditor. Academic research has generally supported the view that ACs that meet more frequently are more effective (Abbott, Parker, and Peters 2004). Greater disclosure about the number of meetings with the auditor and the types of issues discussed at those meetings would provide investors with relevant information about how the AC oversees the external audit. Additional disclosures that specify the nature and substance of discussions between the auditor and AC could be helpful, but would also carry some downsides. For investors, a greater understanding of the nature of this relationship can offer further assurances of the reporting oversight process and the thoroughness of an audit. But, for such disclosures to be informative, they would likely have to be detailed, which could be problematic for a few reasons. However, one member of our Committee notes that some of these conversations between the AC and the auditor may be private and not intended for public consumption. For instance, suppose the auditors made certain suggestions or pointed out some immaterial mistakes to the AC. If this communication were made public, investors may make improper conclusions using this otherwise innocuous information. In other words, even though this communication was natural and proper, its release may lead some to believe that there is an unaddressed issue with the firm or that an improper audit was done. This may lead to an improper belief among investors that the AC is not properly monitoring the reporting process. Thus, releasing new information without proper context of the nature of communications between the auditor and AC could lead to problematic interpretations. The frequency of conversations between the auditor and AC is a number that could be hard to interpret. On the one hand, more conversations could suggest that there is a major problem that needs a lot of dialogue between the two parties. But on the other hand, frequent conversations could reflect a conscientious AC requesting regular feedback from the auditor. Again, without proper context about the “normal” number of conversations, such a disclosure is up for interpretation, thus making its release less helpful. However, if a checklist form were to be developed, that could be much more helpful. For instance, if a form containing items most likely to be discussed by auditors and the AC were required, then it could be much more informative. Then, the company could release this form and check off which items its AC discussed with the auditors. The investing public, then, could see that certain topics were discussed, which could give them greater insights into the AC's oversight process. At the same time, companies could offer a brief explanation as to why certain items on that checklist were not discussed. A checklist disclosure would give the public some new information about the auditor-AC relationship, but still keep the nature of their conversations private. Thus, this type of disclosure would limit the downside of broad disclosures mentioned above.

Also, as a result, dialogue between auditors and the AC can become overly formalized. Auditors, out of fear of conveying information that could ultimately become public, could become overly guarded in their communications with the AC, which would in turn limit the amount of information the AC gets about the audit process. The result could be that the AC is in worse shape because of the chilled communications with the auditor, as it would have one less channel for feedback about the reporting process.

The most significant items are:

• (7) 

  Audit firm industry experience (number of years of experience of the partner and manager and the percent of time these two individuals spend on industry specific audits (Q45). This has been found to have a major effect on audit fees (Ferguson, Francis, and Stokes 2006). Further, DeFond, Hann, and Hu (2005) provide evidence that investors respond favorably to the appointment of accounting financial expertise to the AC. This suggests that information about the AC is useful to investors. Current disclosure requirements provide useful information to investors but, as noted in the request for comment, the current disclosure requirements have not been updated since before the adoption of Sarbanes-Oxley (SOX). Therefore, we believe that enhancing the AC disclosure requirements would better inform investors.

• (8) 

  The use of specialists, and especially forensic experts, on the audit (Q35) (see Boritz, Kochetova-Kozloski, and Robinson 2015).

• (9) 

  The firm's internal review and PCAOB inspection reports. Prior research (e.g., Gunny and Zhang 2013; Abbott, Gunny, and Zhang 2013) suggest that PCAOB inspection findings are negatively associated with audit quality. The AC should acknowledge its knowledge that a PCAOB inspection occurred on its audit, and comment on its findings (Q20)

With respect to Q26 on the process the AC undertook during the auditor selection, we would like to note that if the disclosure requirements are too specific, ACs might be reluctant to publicly reveal such private information. Furthermore, given that such information is only known by the AC members, it is unlikely that there can be an appropriate enforcement mechanism for nondisclosures. But since very little is known about why firms select auditors, any new information would be helpful to investors and the public. To that end, any disclosure requirement must consider the fact that ACs may be unwilling to reveal the details of their decision making.

We believe the following information is not too specific, yet informative about the auditor selection process:

• 1. 

  How many auditors were considered?

• 2. 

  Did the selected auditor have an office near your headquarters? How big of a factor was geographic proximity?

• 3. 

  Did the auditor's industry experience play a factor in selection?

• 4. 

  Did the audit partner's experience play a factor in selection?

• 5. 

  Did the audit firm's national or regional reputation play a factor in selection?

Furthermore, enhancing auditor independence has been one of the hallmarks of the efforts by the SEC, PCAOB, and the accounting profession in the last several years. Although the Sarbanes-Oxley Act is unequivocal on the party (i.e., the AC) that has the authority to select, retain, and compensate external auditors, research results point to a more complex and a varying set of practices among issuers. For instance, Cohen, Krishnamoorthy, and Wright (2010), in a study involving audit managers and partners, find that management is still perceived to have a significant influence in the auditor selection, retention, and compensation decisions. While AC's consultations with the management when making auditor appointment and dismissal decisions should be expected and could in fact be functional, given the close working relationship between management and auditors, undue influence by management has the potential to undermine the AC's authority and can negatively impact auditor independence. In fact, during casual conversations, some auditors still refer to management as the “client,” potentially reflecting a mindset that could consciously, or subconsciously, influence audit decisions when dealing with management. While current AC disclosures in issuer regulatory filings (e.g., proxy statements) usually affirm the AC's authority with respect to auditor selection, retention, and compensation decisions, they tend to be more “boilerplate,” and management's role in the process is seldom discussed. Additional disclosures about the processes that the AC pursued in auditor selection and compensation decisions, and role of the company management in such decisions, where applicable, may help bolster investor confidence with respect to auditor independence, and consequently increase the value of the assurance provided by audit reports.

We feel it is quite appropriate to disclose these items by the AC and to include hyperlinks to the AC charter and other AC relevant material (Q66).

Finally, we believe it is important to name the audit engagement partner, and the main audit manager (Q34). There are two key reasons supporting this recommendation. First, it enables individual auditors an opportunity to develop a reputation for providing high-quality audits (see Aobdia, Lin, and Petacchi [2015] and Chi, Lisic, Myers, and Pevzner [2015] for evidence from Taiwan). Signing the audit report can also induce increased accountability on the part of the audit partner. However, psychology results suggest such accountability effects arise when individuals sign a document before they do the work, rather than at the end (as current auditors do—see Shu, Mazar, Gino, Ariely, and Bazerman [2012]). One possibility would be to sign an “audit contract” at the start of fieldwork and then both can also sign at the end of the audit. Since the bulk of audit fieldwork is usually overseen directly by the audit manager (rather than the partner—see Hoang, Jamal, and Tan [2015]) it may be important to include the key audit manager in the accountability process. We also note that the Committee provides its detailed views on the audit partner name disclosure proposal by the PCAOB in two separate “Comment Letter to PCAOB” (see, Jones et al. 2012a; Anderson, Gaynor, Hackenbrack, Lisic, and Wu 2014).

Question 15, in Part 1, asks whether additional AC disclosures would be used by institutional and retail investors, investment advisors, and proxy advisory firms. We believe that enhanced AC disclosure could be quickly picked up by proxy advisors, such as Institutional Shareholder Services. Anecdotal evidence suggests that proxy advisors have in the past considered some audit quality indicators in their considerations of firms' governance practices, and therefore it is likely that they would consider at least more informative and insightful AC disclosures (for example, with respect to process related to resolving disagreement with managers, as it is one of the key indicators of auditor turnover [DeFond and Jiambalvo 1993]). Moreover, institutional investors are sensitive to voluntary disclosures provided by firms, and therefore would be likely to pick up any disclosure that they believe could provide information about future firm performance. For example, detailed information on how audit fees are set may be useful in predicting future crash risk (Hackenbrack, Jenkins, and Pevzner 2014), and therefore may be useful to institutional investors. However, we believe that any disclosure provided by ACs should be sufficiently clear to be interpretable and useful in making predictions. Therefore, while conceptually it would be very useful to learn how ACs evaluate auditors' professional skepticism and objectivity (Q25), we find it somewhat difficult to think of a clear way to disclose such an evaluation process, as there is no unified agreed-upon clear measure of professional skepticism that we are aware of.³

Question 55 asks whether additional AC disclosures would promote audit quality. We believe so. For example, a large research literature in auditing suggests that various auditor characteristics (e.g., tenure, industry expertise, history of association with past restatements) are all indicative of audit quality (or lack thereof).⁴ Thus, understanding how the AC considers various audit quality indicators, and how it operates in general, would be useful to investors. We echo the concern of Question 60 on whether the disclosures could result in overload of boilerplate information that is not particularly useful to investors. To avoid it, the proposed rules should encourage more specific disclosure, as investors tend toward value-specific⁵ and more readable⁶ disclosures. We also believe that as long as disclosures are specific and readable, they will result in reduced information asymmetry between investors and the firm because we know from prior research that ACs affect audit quality, and audit quality in turn affects cost of equity and debt capital.⁷

With respect to Question 56, which deals with specific issuer, industry, AC member, or auditor characteristics that should be considered in establishing new disclosure requirements, we note that there has been considerable focus from regulators and from academic researchers on the characteristics of AC members that make them effective in discharging their responsibilities with respect to their oversight of the financial reporting process. For instance, there is a considerable body of literature that documents the positive association between accounting financial expertise of the AC and financial reporting quality. Regulations currently require issuers to disclose whether their AC consists of at least one member who is a “financial expert” in accordance with the definition set forth in SEC regulations. However, there has been little discussion about the importance of disclosing industry expertise of AC members. The value of auditor industry expertise has been well documented by a number of academic studies, and a recent study by Cohen, Hoitash, Krishnamoorthy, and Wright (2014) finds that AC industry expertise contributes positively to an AC's effectiveness in monitoring the financial reporting process. However, it is an open question as to whether more detailed or targeted disclosures about the industry expertise of AC members are value relevant to investors. More broadly, however, this raises the question of whether there should be more mandated or voluntary disclosures about the characteristics of AC members, especially those that are value relevant to investors. Current biographical information about board members (including AC members) disclosed in regulatory filings tends to be brief sketches, with no coherent or consistent information that may be value relevant to investors. While interested investors may be able to seek and gather biographical information about AC members from a variety of public sources, an important issue to consider is whether more information, such as the industry expertise of AC members, should be disclosed by the issuer in regulatory filings with the SEC.

With respect to Questions 12 and 73, which ask whether the AC should disclose how it deals with disagreements between auditors and management and how the AC interacts with the company's management, respectively, we believe that specific and clear disclosure of such interactions could be highly helpful in assessing overall governance quality and ensuring auditor independence. For example, if the AC becomes informed of existing disagreements between the auditor and management over accounting policies, it would be highly helpful to understand how these disagreements are resolved. For example, over time investors may benefit from knowing whether a particular AC tends to side with management over these issues, which could suggest a potential threat to auditor independence. However, the current picture of this is not necessarily very clear.

More specifically, currently there appears to be variability in practice on how ACs deal with disagreements between company management and the auditor. Prior research (e.g., Cohen et al. 2010; Gendron and Bédard 2006) has examined the AC's role in helping resolve disagreements between company management and the auditor, and evidence seems to suggest a lack of consistency across issuers, with some ACs being more proactive and others taking the position that the auditor and management resolve contentious issues before they are brought to the attention of the AC (Cohen et al. 2010). However, it is unclear as to which one of the many postures that the AC could pursue, with respect to dealing with disagreements between company management and the auditor, is optimal or if the variability in practices among issuers should be expected given the complexities of business, industry, and the working relationship between the management and the auditor. Hence, what may be helpful to investors is if ACs disclosed, wherever feasible, their policy or posture with respect to dealing with disagreements between management and the auditor. The potential downside is that such disclosures could induce audit behavior that may be suboptimal for the issuer. For instance, if investors generally view more proactive ACs as being more effective, ACs may be induced to pursue a more proactive strategy in dealing with disagreements between management and auditors, even when such a strategy may be suboptimal or ineffective for the specific issuer. In summary, while there are potential benefits to disclosures in this area, further research is warranted before specific disclosures are recommended or mandated.