Comments by the Auditing Standards Committee of the Auditing Section of the American Accounting Association on PCAOB Rulemaking Docket Matter No. 042: PCAOB Release No. 2017-005, Proposed Amendments Relating to the Supervision of Audits Involving Other Auditors and Proposed Auditing Standard— Dividing Responsibility for the Audit with Another Accounting Firm Free

On April 12, 2016, the Public Company Accounting Oversight Board (PCAOB) issued a proposal to strengthen the requirements related to audits involving accounting firms and individual accountants other than the accounting firm that issues the audit report. The 2016 proposal is “intended to increase the involvement by the lead auditor in the work of other auditors, enhance the ability of the lead auditor to prevent or detect deficiencies in the work of other auditors, and facilitate improvements in the quality of the work of other auditors” (PCAOB 2017). The current supplemental request for comment presents revisions to the initial proposal based on several comment letters and discussions through the PCAOB's Standing Advisory Group (SAG).

Many of the questions posed in the supplemental request for comment focus on the clarity and appropriateness of the proposed changes. In general, the Auditing Standards Committee of the Auditing Section of the American Accounting Association (the committee) is supportive of the proposal and revisions; however, the committee did make suggestions to improve clarity and raised potential concerns. The responses of the committee included in the comment letter are as follows.

• Question 1:

  Is the revised requirement for determining the sufficiency of participation to serve as lead auditor, based on risk and importance of the locations, appropriate and clear?

The word “importance” in paragraph .B2.b creates some ambiguity. Scoping of audit work at locations is based on “significance” of the location, which is usually linked to materiality. The use of the word “importance” of locations in this case does not link to either concept—significance or materiality. It appears that auditors should interpret this word in a similar manner to the way they would interpret based on “materiality.” We encourage the Board to further clarify how auditors should interpret this word choice. For example, should “importance” imply a user perspective in the same way that “materiality” would?

The word “importance” may also be construed as less precise than the word “materiality.” While there is relatively little research on the effects of the auditing standard precision, several studies find that accounting standard precision influences the way jurors evaluate auditor negligence. Namely, jurors experience more difficulty determining the extent to which an accounting treatment conforms to imprecise standards, as compared to precise standards; as a result, jurors are relatively less able to use compliance with imprecise accounting standards to evaluate audit quality (e.g., Gimbar, Hansen, and Ozlanski 2016; Kadous and Mercer 2016). We, therefore, encourage the Board to consider whether using relatively less precise wording (i.e., “importance” versus “materiality”) may have unintended effects on the way stakeholders evaluate audit quality. In addition to improving the clarity of .B2.b, more precise wording here would also be more consistent with wording used in other standards.

• Question 2:

  Is the additional sufficiency threshold for divided responsibility engagements clear? Should this be a bright-line requirement, or does this threshold need to allow for exceptional situations? Are there any other implications of this threshold that the Board should consider, such as investor protection implications or auditing challenges related to the revised requirement?

The threshold requirement proposed to be added to paragraph .B2 is an improvement. As noted in the discussion of this modification (AICPA 2017, 10, 11), while the “50 percent” threshold has been a standard historically applied in practice, qualitative considerations should also influence the determination of “sufficiency of participation.”

To aid in the auditor's evaluation of “sufficiency,” the Board should consider including examples where it might be appropriate for an auditor to serve as lead auditor even when auditing less than 50 percent of an issuer (e.g., significant late-year acquisitions or other unanticipated events or conditions that increase the portion of assets or revenue audited by other auditors beyond the 50 percent threshold [AICPA 2017, 11]), as well as when it might not be appropriate for a lead auditor to rely on other auditors even though a component represents less than 50 percent of an issuer's assets and revenues (e.g., a component representing 49 percent of assets or revenues, and more than 50 percent of cash flows). Given the potential for acceptable situations outside of the 50 percent threshold, it seems appropriate to allow for exceptional situations rather than implementing a bright-line requirement.

As it appears the Board's intention is to require justification for reliance on other auditors in cases where a component's assets or revenues exceed 50 percent of a company's consolidated assets and revenues (AICPA 2017, 10), it would be helpful to state that requirement in paragraph .B2.

• Question 3:

  Are the revised requirements relating to the other auditors' compliance with the independence and ethics requirements appropriate? Are there any practical challenges associated with the revised amendments? If so, what are they, and how could the proposed requirements be revised to address the challenges?

The revised requirements relating to the independence of other auditors seem clear. However, such requirements could be enhanced by providing practical examples. In particular, paragraph .B4.a could be amended to include the following wording from the release: “For example, the lead auditor may obtain a written description of the other auditor's process and results of the process, or may obtain this understanding through inquiry, and perform follow-up procedures as necessary to address gaps in the process or indications of potential noncompliance” (AICPA 2017, 13).

Because the “auditor's processes for determining compliance with the SEC independence requirements and PCAOB independence and ethics requirements” (paragraph .B4.a) are part of each audit firm's quality control (QC) program (Bedard, Deis, Curtis, and Jenkins 2008; Church, Jenkins, McCracken, Roush, and Stanley 2015), and firms' QC programs are regularly inspected by the PCAOB, information about the independence process at the other auditor's firm should be readily available to the lead auditor.

For some smaller auditors that may currently lack formalized and/or documented procedures for assessing and addressing independence, the importance of independence and its influence on audit quality (Tepalagul and Lin 2014) merits the possible imposition of additional costs associated with the proposed modifications. However, as noted on page 14 of the Release (AICPA 2017), although the Board has decided not to allow “reliance” on a network in determining the other auditor's compliance with independence ethics requirements, research has shown that membership in networks has been associated with higher levels of audit quality for smaller firms (Bills, Cunningham, and Myers 2016). Therefore, the Board may consider allowing membership in a network to be a factor that reduces the risk associated with independence for firms that may not have a robust independence and ethics process of their own. In other words, membership in a network would not substitute for the lead auditor obtaining an understanding of the other firm's independence and ethics processes, but a strong independence process at the network level that is adhered to by the local-level firm could be one factor that helps address a gap in or lack of documented processes at the local firm level.

• Question 4:

  Are the proposed amendments relating to the knowledge, skill, and ability of the other auditor, revised by this release, appropriate? Are there any practical challenges associated with the revised amendments? If so, what are they, and how could the proposed requirements be modified to address the challenges?

A lead auditor's assessment of the knowledge, skill, and ability of auditors assigned to component audits is important, and the proposed modifications seem clear. The knowledge level of each individual assigned to the engagement team is difficult to assess, so a focus on the firm-level processes for disseminating knowledge throughout the firm is appropriate, as firms have multiple knowledge sharing avenues to ensure that auditors can access the resources they need to conduct the audit (Bedard et al 2008; Vera-Muñoz, Ho, and Chow 2006; Carson 2009). The standard as amended (paragraphs .B6.a and .B6.b) focuses heavily on ensuring that the lead auditor has assessed whether engagement team members in the other auditor's firm have received proper training and that the other audit firm itself has the proper industry expertise, but there is no mention of other resources that can enhance an individual auditor's knowledge, such as the ability to consult with other experts in the firm about matters that are highly complex. As noted in research on quality control and firm networks (Bedard et al. 2008; Bills et al. 2016), the presence of consultation units and electronic decision aids are two such resources that can help enhance audit quality within firms and engagements by giving engagement auditors direct access to firm-level expertise. It may therefore be appropriate to include some mention of the lead auditor's assessment of the presence of these resources in the standard in addition to assessment of training. Also, in the Release on page 15, the Board notes that some commenters suggested relying on the network's system of quality control when the other auditor and lead auditor are in a common network (AICPA 2017, 15). Although the lead auditor must still assess the training and industry expertise of the other auditor within the network, knowledge of the network's processes for consultation and provision of other knowledge-sharing resources could aid the lead auditor in assessing the knowledge, skill, and ability of the other auditor.

Additionally, some examples may be helpful in guiding the lead auditor in understanding this section of the standard. For example, paragraph .B6.b could be enhanced by adding the following wording from the Release: “Possible sources of information that are relevant to the lead auditor's understanding of the knowledge, skill, and ability of relevant personnel include the lead auditor's own experience working with them, the other auditor's policies regarding the nature, scope, and timeliness of relevant training for them, information about internal inspection results regarding them, and publicly disclosed disciplinary action by regulators against them” (AICPA 2017, 16).

• Question 5a:

  Are the proposed new additions to AS 1015 (Due Professional Care) and revision to AS 1201 (Supervision) relating to the other auditors' responsibility appropriate and clear?

The proposed modifications seem clear and address commenters' requests for an explicit reference to other auditors' responsibilities (AS 1015) and guidance regarding what should be in the other auditors' written report (AS 1201).

• Question 5b:

  Is it clear that AS 1015, Due Professional Care, already applies to referred-to auditors that perform audits under PCAOB standards?

Yes, it is clear that AS 1015 applies to referred-to auditors that perform audits under PCAOB standards.

• Question 6:

  Are the proposed new additions to AS 2101.B2 appropriate and clear? Also, is it clear that the necessary level of detail of the other auditor's audit documentation that the lead auditor obtains and the necessary extent of the lead auditor's review according to requirements in proposed Appendix B of AS 1201 are scalable based on the factors in the existing standard regarding the necessary extent of supervision?

The proposed new additions to the standard seem appropriate and clear, and the scalable nature of the documentation to be provided for review is clearly stated in the standard. As written, the standard indicates that the other auditor would prepare documentation for review, and the lead auditor would review the documentation and draw conclusions based on this review. If the documentation indicates that sufficient appropriate evidence has not been obtained, the standard recommends that the lead auditor consider whether additional evidence should be obtained. However, it seems that waiting until the final summary memo is prepared to make this determination may result in an inefficient audit. Research has demonstrated that one challenge to group audits is lack of clear communication between the lead auditor and other auditors (Downey and Bedard 2016), and knowledge sharing within geographically dispersed teams can be negatively affected by cultural differences, particularly in how teams challenge the lead auditor, or even ask questions of the lead auditor (Vera-Muñoz et al. 2006; Downey and Bedard 2016). One solution proposed by Downey and Bedard (2016) is to ensure frequent communication between the group auditor (lead auditor) and component auditor (other auditor) throughout the engagement. The proposed standard should acknowledge the complex and iterative nature of this process and provide additional guidance in terms of frequency of communication and the iterative nature of the documentation.

Additionally, although on page 20 of the Release it states that the proposed amendments to the standard would cover “Obtaining and reviewing the other auditor's description of the nature, timing and extent of its audit procedures,” (AICPA 2017, 20) the words “nature, timing, and extent” are struck from the actual standard paragraph .B2b. It is unclear why these words are struck from the standard given the goal stated in the Release.

• Question 7:

  Are the revised proposed requirements for situations in which the lead auditor directs another auditor to perform supervisory procedures with respect to a second other auditor on behalf of the lead auditor clear? If not, how should the revised proposed requirements be revised?

The proposed requirements with respect to multi-tiered engagement team structures in the standard (paragraph .B3) seem clear in terms of specifying that the lead auditor can delegate certain tasks outlined in paragraph .B2. Research has not specifically addressed the multi-tiered nature of group audit engagements, but Downey and Bedard (2016) do note that efforts taken to reduce the complexity of these engagements can have positive impacts on how group audits are conducted. Therefore, it seems that in certain circumstances, having the first other auditor oversee the second other auditor has the potential to improve communication and oversight, and ultimately, as a result, improve the outcome of the engagement.

Since the lead auditor is ultimately responsible for reviewing and retaining all documentation required by AS 1215.19, it is unclear why only documentation required under paragraph .B2a and .B2.d is mentioned in paragraph .B3 and documentation under .B2.b and .B2.c is excluded from the paragraph.

• Question 8:

  Is the revision to the proposed standard relating to the division of responsibility when the company and its business unit use different reporting frameworks appropriate and clear?

This revision relating to the division of responsibility seems clear. However, similar to other circumstances where the lead auditor refers to other auditors, if the lead auditor indicates that the referred-to auditor audited conversion adjustments, then this reference may be construed as a disclaimer of responsibility by certain various groups (e.g., investors, attorneys, jurors). Contemporaneous research around Critical Auditing Matter (CAM) disclosures finds that these disclosures reduce users' assessments of auditor responsibility for misstatements in the same area as the CAM (e.g., Kachelmeier, Schmidt, and Valentine 2017). This suggests that the proposed requirement for the lead auditor to indicate that the referred-to auditor audited the conversion adjustments may, perhaps unintentionally, mitigate the lead auditor's legal liability. We encourage the Board to consider clarifying the proposed standard to avoid this potential unintended effect.

Relatedly, research demonstrates that certain investor groups have difficulty weighing information about the audit in their valuation judgments (e.g., Vera-Muñoz, Gaynor, McDaniel, and Kinney 2015; Kachelmeier et al. 2017), but visual cues in financial reports (such as those provided in “Circle-ups” that accompany letters to underwriters) can facilitate users' weighting of this information (Dennis, Griffin, and Johnstone 2017). We, therefore, encourage the Board to consider mechanisms (such as “Circle-ups”) that may aid in facilitating users' weighting of such information by linking audit report information about referred-to auditors with the related amounts and disclosures in the financial statement.

• Question 9:

  The committee did not provide feedback on Question 9.

• Question 10:

  Comment is requested on the matters discussed in this section. Would any revisions the Board is considering for adoption affect the scalability of PCAOB standards in this area? Would any have a significant effect on the competitiveness of smaller audit firms? Would the revisions significantly change the costs and benefits associated with the proposed changes discussed in the 2016 Proposal? Are there any unintended consequences that the Board should consider? Are there any other matters not addressed in this release the Board should consider in its economic analysis?

Although existing rules require any auditor who plays a substantial role in the preparation or furnishing of an audit report to register with the PCAOB, the modification proposed to paragraph B5 may be perceived as lowering the threshold requirement for registration. If so, extant research suggests some smaller auditors may exit the issuer-audit market (e.g., Abbott, Gunny, and Zhang 2012; Daugherty, Dickins, and Tervo 2011). That said, it is likely that the number of impacted auditors would be small.

Additional documentation requirements proposed to be mandated (e.g., paragraph B4) will likely add to the cost of conducting an audit, and these costs will likely be passed along to issuers in the form of higher audit fees.