The Board is reproposing today amendments to its standards to require firms to disclose in the audit report the name of the engagement partner and to provide certain information regarding other persons and independent public accounting firms that took part in the audit.
While I support gathering additional feedback on our possible courses of action, I do have strong reservations about today's proposal and take exception to a number of generalizations in the release about what the Board believes. I cannot say today that I would support the ultimate adoption of the reproposal as currently drafted.
As is the case with all of our regulatory activities, there is a balance we must achieve. We must weigh the potential benefits of our actions, determining the degree to which they have a positive effect on investor protection, and we must consider the anticipated costs to those we regulate and the capital markets at large. We strive to base this consideration on the best evidence available. In some cases, the scale tips easily and the Board unanimously supports a new standard. In other cases, it is a much closer call, and this is one of those situations.
Many investors have told us that they would find useful the information that we are proposing to require, including the identity of the engagement partner and other participants in the audit. Some investors believe that knowing the firms involved and the engagement partner's identity will help them evaluate the audit's quality, while others believe the identifications will enhance accountability of those leading and participating in the audit.
I support investors in their desire for relevant and meaningful information. If investors want to know who is responsible for the audits of the companies in which they invest, they should have access to that information. If they believe that it is important to understand what other firms or individuals participate in the audit, then we should provide that information, as long as it does not impose any unreasonable costs, uncertainties or other adverse consequences.
What troubles me about today's reproposal, however, is the mechanism for the disclosures. In the 2011 proposal, we raised questions about whether auditors should disclose the relevant information in the auditor's report or in the firms' annual filings with the Board (known as Form 2) or both. Today, we are issuing a reproposal to require all of the disclosures in the audit report. In my view, requiring these disclosures in the audit report — as opposed to on our website in a firm's annual filing on Form 2 or another filing — involves substantial uncertainties and potentially unnecessary risks. I believe that the evidence cited in the release for the potential benefits of the disclosures is weak. And certainly the incremental benefit, if any, from including the disclosure in the audit report rather than in another filing is minimal, at best.
Nevertheless, because I support the concept of increased transparency overall and believe in the process of seeking further input, I agree to the issuance of the reproposal. I am hopeful that we receive robust and thoughtful comments to help guide us as we further consider the best approach to responding to investor calls for more relevant information about audits.
Identification of Other Firms and Individuals
With regard to the disclosure of other firms and individuals, the Board is reproposing amendments to PCAOB standards that would require firms to disclose in the audit report participation in the audit by other independent public accounting firms and individuals not employed by the auditor, as well as their locations and certain information about the work performed by those other firms or accountants.
As I noted in my statement in support of the original proposal, the disclosures are particularly important in the context of audits of companies with multi-national operations. In such audits, the auditor issuing the audit report frequently refers certain audit procedures to firms or accountants located in other countries. In most situations, the auditor performs certain procedures to review the work of the other firms or accountants and assumes full responsibility for that work. Currently, investors have little or no information about whether other firms or accountants participate in such audits or about the identity or role of such other firms or accountants in conducting the audit procedures that ultimately support the opinion of the firm issuing the report. In some cases involving multi-national audits, the other audit firm is unrelated to the firm issuing the opinion. In other cases, the firm issuing the audit report and the other firm are members of an international network, which may subject both firms to similar governance and quality control procedures. Currently, investors have little or no insight into this issue.
Moreover, even where the other firm is a member of the international network of the firm issuing the audit report, the network affiliate firm may be subject to different, and potentially conflicting, legal and regulatory requirements that investors may want to consider in evaluating the overall audit. For example, in light of the obstacles that have prevented PCAOB inspections of firms located in certain jurisdictions, investors may be particularly interested in knowing when a firm that participated in the audit is located in one of these jurisdictions, and if so, the significance of that firm's role in the audit.
Identifying in the audit report that other firms or individuals participated in the audit also will permit investors to consider other relevant information. This may include basic information such as the size of the firm or the expertise and experience of the firm or the individual. Investors may want to consider whether the firm is registered with the PCAOB, whether the firm is subject to PCAOB inspection, and, if so, whether the firm has been inspected, and its inspection results — information that is available on the PCAOB website.
While the reproposal requires disclosures very similar to those previously proposed, there are several changes worth pointing out.
First, the Board is no longer proposing to require that names of individuals participating in the audit, but not employed by the auditor, be disclosed. Rather, the reproposal requires only the disclosure of names of public accounting firms participating in the audit (if they meet a certain threshold), while individuals who participated may be identified as "persons not employed by our firm."
Second, while the reproposal retains the original requirement for the firm issuing the audit report to disclose the percentage of the total audit hours attributable to the respective firm or individual, the thresholds for when firms or individuals must be individually listed, as opposed to aggregated, has changed. Under the original proposal, signing firms had to disclose each firm or individual who performed work amounting to more than 3 percent of the total audit hours. In the reproposal, that number is 5 percent. In other words, if the contribution of another firm or a single individual exceeds 5 percent of the total audit hours, that firm or individual must be listed separately in the audit report (with the other firm identified by name or the individual identified as a person not employed by the signing firm), along with the percentage of their participation. Any firms or individuals whose individual contribution is 5 percent or less may be disclosed as a group with an aggregated percentage of audit hours.
Third, to make the disclosures less burdensome, the reproposal permits for the percentage of total audit hours to be disclosed in ranges, based on estimates, rather than requiring specific numbers.
Finally, while the original proposal did not require the disclosure of the use of individuals with specialized skill or knowledge in a field other than auditing or accounting, the reproposal would require firms to disclose the use of such specialists in the same way and to the same degree as any other individual persons not employed by the auditor.
In general, I believe information about other participants in the audit is meaningful information that we should provide to investors. While the percentage of audit hours may not fully capture the actual importance of the other firm's or other accountant's role in the audit, it may provide more insight to investors, and I believe it comes at a reasonable cost to firms who generally track audit hours and should be able to provide estimates and ranges as required in the proposal without too much difficulty.
Nevertheless, I am particularly interested in comments about certain aspects of this proposal. First, is 5 percent the right threshold for disclosing individual firms or persons? The increase from 3 to 5 percent is a response to comments that 3 percent was too low, but 5 percent is not a magic number and is not based on evidence that the work of another auditor becomes particularly "material" at a given point. As explained in the release, a 5 percent threshold results in the disclosure of several "other auditors" in many large audits, but the importance of the work of each of those auditors may vary with the nature of the work more than with the number of hours. Our rule for what constitutes "playing a substantial role" in an audit for other purposes, such as registration, is 20 percent (based on audit hours, or several alternative metrics). The reproposal assumes that the requirement for the disclosure of other firms participating in audits should be broader than the threshold for firm registration, but is 5 percent the right number? Would 10 percent or another level be more appropriate? One reason to consider this threshold carefully is that audit firms who are individually named in the audit report will have to provide their consent to be named as an expert pursuant to Section 7 of the Securities Act of 1933 ("Securities Act"). Conceivably, the logistical challenge of obtaining relevant consents from foreign firms could be significant. We have not fully explored this issue, so I hope that we will receive meaningful input on these points.
In addition, we have posed questions in the release about the potential disclosure of specialists. As written, disclosing a specialist will reveal the same information as disclosing individual accountants participating in the audit. In other words the specialist will not be named or identified as a specialist, but the audit report will include him or her as a person participating in the audit but not employed by the signing firm, along with the respective audit hours. But is the percentage of audit hours meaningful for this subset of participants, especially since their contribution of specialized skills could be critical to the audit while the hours are few? Should the nature of each specialist's work be disclosed, or do the hours suffice? What other effects could this disclosure have that the Board should consider?
The circumstance of individuals or firms participating in an audit in the context of so-called "off-shoring arrangements" also presents a difficult question. Currently, the reproposal takes a bright line approach: If the "off-shored" entity is part of the firm signing the audit report — for example an office or service center of the firm located in another country — no disclosures would be required. But if the service center is created by a signing firm as a separate legal entity — for legal or tax reasons, for example — then disclosures about its involvement in the audit are required. While this approach may provide better clarity about when disclosures are required, does it provide useful information to investors? Are entities that are wholly owned and controlled by the signing firm, but legally separate entities, really any different in terms of the work performed? This is an issue that I believe would benefit from more comment and consideration.
Finally, I am concerned about the potential liability implications for other firms that would be named in the audit report under the reproposal, particularly with respect to liability under the Securities Act. Because firms who contribute more than 5 percent of the audit hours would be named in the audit report, such firms would be required to provide consents as experts under Section 7 of the Securities Act and would be subject to liability under Section 11 for material misstatements or omissions in the report. This liability currently does not exist for audit firms that participate (but are not named) in the audit report, and I believe this is a potentially significant cost that may weigh in favor of making these disclosures elsewhere. Because these liability concerns overlap with my liability concerns about the engagement partner disclosure, let me first say a few words about that aspect of today's reproposal.
Engagement Partner Identification
As the staff has explained, the Board is also reproposing today a requirement for firms to disclose the identity of the engagement partner in the audit report. My concerns about this aspect of the reproposal are significant, because, based on what we know, I believe the potential costs associated with requiring this disclosure in the audit report may far outweigh the more limited benefits.
Investors have told us that they believe the identity of the engagement partner is meaningful and relevant information. Some investors have told us that they believe that identifying audit partners will enhance the accountability of those partners and therefore enhance audit quality. I am skeptical of this, and the Board has found no evidence that such an effect is likely. Indeed, the release cites several studies suggesting that disclosure of the engagement partner's name produces limited or no observable improvement in audit quality and offers no studies to the contrary. (One study we cite suggests that audit partner signature — as opposed to disclosure — may enhance audit quality, while also noting that audit fees increase by 13.2 percent after implementation of a signature requirement.) The work of public company auditors is governed by performance standards and is reviewed by engagement quality reviewers and internal firm inspectors. They also have PCAOB inspectors judging the most difficult aspects of their work. They understand that with every issuance of an audit report, their firm's reputation is on the line. Ultimately, those individuals who stray from their obligations despite all of these incentives may face the specter of regulatory enforcement proceedings and the subsequent loss of their ability to practice in their profession, as well as civil litigation. Accountability for audit engagement partners, in my experience, is already built into the system.
Other commenters have told us they believe that knowing the identity of the engagement partner will permit them to better evaluate the quality of the audit, particularly after some period of time has passed, allowing for the accumulation of information about engagement partners and their audit experience and history. I do not question the desire by investors to have that information and believe they should have access to it, to use as they see fit. I do believe, however, that in order to justify regulatory action, the benefits referenced in today's release need further explanation and support, and I would like to understand better how information about the engagement partner is used today by investors in jurisdictions where that information is currently provided.
I also believe that the engagement partner's identity, alone, will provide only limited insight into the quality of a given audit. While engagement partners certainly have significant responsibilities under applicable auditing standards, in practice, they are not solely responsible for the quality of an audit. Many firms have complex quality control systems. Large audit engagements often include multiple partners involved in supervising and reviewing the work of the firm's staff, as well as partners in foreign affiliates. Engagement partners frequently consult national office experts to help resolve technical questions, and many firms have policies requiring engagement partners to follow national office determinations on such technical questions. Thus, while the engagement partner plays an important role in the audit, he or she is, by no means, the only key factor in audit quality.
Nevertheless, while the benefits of learning the engagement partner's identity may be limited, I understand investors' desire for more information about the audit, including the engagement partner's name. I am not opposed to providing that information, in some form, but I have significant reservations about whether the disclosure of the information should be made in the audit report, rather than through other means.
When we issued our proposal in 2011, we asked a series of questions about the potential increase in liability to engagement partners that would arise out of the disclosure of their names in the audit report. My concerns at that time included not only the potential that some limited number of engagement partners could face significant personal liability, but also that the imposition of such liability could make the auditing profession less attractive. This could provide a disincentive for the best and the brightest young accountants to becoming auditors, at a time when financial statements are more complicated than ever. In addition, such potential liability could cause "defensive auditing," resulting in unnecessary audit procedures and increasing audit costs but not necessarily enhanced quality.
Based on the limited feedback we received in response to the original proposal, we know little more about this topic today. What appears clear is that engagement partners, as well as other firms proposed to be named in the audit report, would be subject to liability under Section 11 of the Securities Act for any material misstatement or omission in the audit report because they were "named as having prepared or certified any report or valuation which is used in connection with the registration statement.…" The extent to which this would spark additional litigation against engagement partners or other named firms, or would affect their ultimate liability, remains unclear. Potential engagement partner liability for fraud under Section 10(b) of the Exchange Act and SEC Rule 10b-5, likewise, is uncertain. While the release attempts to explain the import of the Supreme Court's Janus decision and suggests that the proposed engagement partner disclosure "is unlikely to change the status quo regarding private liability for fraud," the release also concedes that it is unknown how courts will interpret Janus in the context of audit engagement partners.
The release goes on to state that "[u]ncertainty . . . does not, in the Board's view, justify depriving investors of the benefits of the additional information that would be provided pursuant to the reproposed amendments." That, however, is not the dilemma we are facing. The question is not whether we should "deprive investors" of the relevant information, but about how best to provide that information while doing the least harm. In my opinion, requiring the relevant disclosure in the audit report, as opposed to finding another disclosure mechanism, is the wrong approach.
On the other hand, I would support the required disclosure of the name of the engagement partner and other participants in the audit through one or more of the alternatives described in the release. Personally, I believe the best place for this information would be in the audit committee report, where the audit committee could expand on the information to provide additional context to its selection of the engagement partners and its oversight over the audit firm. We do not have the authority to require audit committee disclosures, but I believe this would be an appropriate topic for consideration by the SEC, while respecting its already very full agenda and understanding that any formal action would take time. In the meantime, I am encouraged by a recent "Call to Action" by the Audit Committee Collaboration encouraging audit committees to enhance their disclosures, including by providing context about their selection of the lead audit engagement partner. I would like to see the disclosures we are considering today added to their action call, in order to provide relevant information to investors more quickly than any standard-setting project by the PCAOB or SEC could accomplish.
Barring SEC action, I also would support a PCAOB standard requiring the relevant disclosures in a form to be filed with the Board, such as the firms' annual report (Form 2) or another, special filing made closer in time to the issuance of the audit report. The release discounts these two alternative approaches because they would "require investors ... to search two different places, at two different regulators … to see both the auditor's report and the disclosures about the participants in the audit." I find this rationale unconvincing. There is no value to obtaining only the engagement partner's name. In order to make use of that information to evaluate the potential quality of a given audit, investors would still have to look at other regulatory filings, including searching for disciplinary records on the PCAOB website, and do other background research in order to find out relevant information about the partner's experience and history conducting audits.
I also believe that other objections to this approach — such as those relating to the ease of use and accessibility of the information on our website — could be easily solved. Indeed, I believe the PCAOB generally should begin to move away from its "form based" reporting and static disclosure approach and, over time, establish web-based systems that are more easily searchable and feature more timely data, to make all of our information more usable for investors, audit committees and others. Establishing a mechanism to allow a user to search for an issuer, its audit firm, and the firm's most recent engagement partner or other participants in its most recent audit would be modest first steps in that direction.
In closing, I believe that the Board has more work to do. I believe investors will gain some benefits from the disclosures that we are proposing today, but I do not believe that the disclosures must be made, at all costs, in the auditor's report. As noted in one qualitative analysis on engagement partner identification that is cited in the release:
[R]elevant academic findings do not provide the basis for making strong inferences about this issue because the data are scarce and inconclusive. However, given the research . . . , the PCAOB's implicit assumption about unconditional benefits of increased transparency and accountability is inconsistent with the empirical evidence. Both increased transparency and increased accountability have a positive impact only in a certain set of conditions, while different circumstances might lead to adverse outcomes.
I believe we need to do a more thorough job considering the potential adverse outcomes of putting these disclosures in the audit report. Ultimately, I believe, that work may lead us to conclude that the disclosures should be made through another mechanism. Short of that, I doubt that I would be able to support final adoption of the standards as currently reproposed.
* * *
Substantive disagreements aside, I would like to add my appreciation to that already expressed by my fellow Board members for the work of the staff of the Office of the Chief Auditor, the Office of General Counsel, and the Office of Research and Analysis. This is a difficult project, and, as our statements demonstrate, Board members have strong views. Despite the occasional tensions this creates, it is my hope that you feel appreciated as the professionals that you are, and I thank you for your hard work and dedication.
I would also like to thank the staff of the SEC for their input, as well as all of the individuals and organizations that provided comments on our initial proposal. I encourage all of you to continue to participate in this important dialog.