Rethinking the Relevance, Credibility and Transparency of Audits

DATE: June 2, 2011 
SPEAKER(S): James R. Doty, Chairman 
EVENT: SEC and Financial Reporting Institute 30th Annual Conference 
LOCATION: Pasadena, CA 

I am honored to be with you.

I want to discuss what I see as important issues in auditing today and for the future. These are issues that occupy the PCAOB today. Chairman Schapiro, the SEC's Commissioners, and Chief Accountant Kroeker, who spoke earlier today, also take a deep interest in the PCAOB's work. I am grateful for their support.

I am also grateful for the work of the Leventhal School of Accounting. For 30 years, you have fostered important thought and thought-leaders in the profession. I know the PCAOB has benefited from your experience and opinions.

And now, about my opinions, I must say that the ideas I express today are my own and should not be attributed to the PCAOB as a whole or any other members or staff.

If you will, first permit me a biographical note.

One night early in my career as a corporate lawyer, I found myself late at the printer's on a registration statement. A young auditor was working across the table and "footing" some numbers. He saw that I had paused in my laborious parsing of the text of, let's say, the attached merger agreement, and that I was watching him do his thing. Having a few years on me, he smiled and said, "You know, Jim, you can think of this as an airplane: what you are doing is polishing the fuselage. I'm tuning the engine!"

In the intervening decades, I have looked back with respect at the self-confidence and sense of purpose of that young auditor, who retired several years ago with distinction as senior partner of his firm and whom I still count as a friend.

I.  The Cultural Importance of Independence and Skepticism

Auditors confirm the flight-worthiness of the engine of reliable financial data that drives our economy. They are an integral part of the basic checks and balances in the system.

To perform their role properly — to assure that reported financial and economic successes are not illusory — auditors must approach their jobs with independence and skepticism. They cannot allow themselves to be caught up in their audit clients' business goals. How do we instill those necessary traits in auditors? This may be the most important auditing question of our time.

For several hundred years, historians have looked to the "tulip scandal" and the "South Sea bubble" to illustrate a cultural malady. The financial crisis of 2007-2009 surpasses them all for an entire business culture suspending skepticism.

Former Fed Chairman Alan Greenspan was shocked to find that senior managers of the major financial institutions did not behave the way the laws of economics should have dictated: how could they all have abandoned the caution necessary to the survival of their institutions and (by the way) to the preservation of their considerable stakes in those entities?

Some regulators too lost their skepticism. The Fed's former Vice Chairman Donald Kohn is reported to have said last month in a confirmation hearing for appointment to the Bank of England's new financial policy committee that "I believe I will not make the same mistake twice"![1]

The financial crisis was not, we know, caused by auditors. But it demonstrates the broader cultural challenge of the audit. The audit must be founded on independence and skepticism. We need to focus on and preserve the underpinnings of that audit culture.

To refer to the language of an auditing standard, AU 316, on "The Importance of Exercising Professional Skepticism":

The auditor should conduct the engagement with a mindset that . . . fraud could be present, regardless of any past experience with the entity and regardless of the auditor's belief about management's honesty and integrity.

Over my career, I have developed pockets of skepticism that influence my view of the audit culture. Do auditors read the literature? Do young auditors learn to hold onto skepticism and look for fraud regardless of long association with management? These are questions I leave with you, as I turn to the decisions facing the PCAOB regarding audit culture.

II.  Cultural Challenges Still Impede Auditor Independence and Skepticism

It is the rare case in which an auditor knowingly compromises his or her integrity. As you know, however, well-intentioned auditors, as with other people, fail to recognize and guard against their own unconscious biases.

Auditors are, after all, paid by the clients they are charged with policing. As in other professions, auditors want to advance in their chosen profession which often means keeping the client happy and growing their business.

Auditor independence requirements serve as counterweights to those forces. One example of those counterweights may be found in the SEC rule that says an accountant will not be considered to have the necessary independence from its audit client if an audit partner earns or receives compensation based on selling non-audit services to the audit client.[2] The purpose of this rule is to keep auditors singularly focused on the quality of their audits and not on nurturing a relationship that will make management more receptive to cross-selling efforts.

Despite those requirements, PCAOB inspection reviews of partner evaluation and compensation processes find examples of seemingly unrestrained enthusiasm — in partners' self-evaluations, in their supervisors' evaluations of their performance, and in agreed performance goals — for selling services to audit clients.

For example, one large firm audit partner set a 2010 performance metric to "achieve service line integration of 15% on SEC clients," which he reportedly met by obtaining multiple transaction consulting services "wins" at several of his clients, by increasing the level of tax services provided at all of his public clients, and by winning additional tax work at two in particular.

Another audit partner's self-assessment claimed that he "overcame long-standing barriers against non-audit services at [two audit clients] with a series of well-planned meetings and supporting presentations with the Audit Committee Chair, the full Audit Committee, the CEO and the CFO at both companies."

In response, his reviewing partner noted that he was —

highly alert to cross service line opportunities and has successfully penetrated both of his accounts where few services had been provided in the past. The results of these efforts were a number of proposals and wins but the efforts will likely impact FY 11 in [a] more significant way.

We don't see these problems in all the files we look at, but we have seen them in sufficient number to raise troubling questions, not the least of which are whether these audit partners are unaware of, or simply unconcerned about, the independence rule that should make such considerations irrelevant to their compensation, and why a firm would allow such unawareness or unconcern to continue unabated.

We can and do, of course, address quality control issues and violations of independence rules through our inspection and enforcement processes. More broadly, though, these examples force us to confront the possibility that despite existing independence rules, auditors still too often approach the audit with an inappropriate mindset.

III.  Fostering a Broad Public Policy Debate to Enhance the Relevance, Credibility and Transparency of the Audit

We need a holistic approach to addressing the cultural challenges inherent in auditing, based on a deep, fact-based analysis of the problem. By a "holistic approach," I mean addressing relevance, credibility, and transparency of the audit by all available and effective means.

I expect the PCAOB to play an important role in supporting a full and informed debate about how to counteract the conflicts auditors face. With this role in mind, the PCAOB expects to issue several policy documents in the near-term. The objective is to foster broad debate and research about ways to enhance both the relevance and credibility of audits, and to provide the investing public a better understanding of what an audit is through enhanced transparency.

We plan to issue these documents over the next two months. We plan on long comment periods to allow for evidence-gathering and research, and to hold public roundtables to explore issues through discussion. We expect interested parties on all sides to do their homework and come back to us in comment letters and public dialogues with facts and evidence.

Please let us know your feedback and suggestions. Please also inform investors that you interact with about these projects and encourage their input. We need constructive insights to help us design effective solutions.

A.  The Auditor's Reporting Model

The PCAOB has already commenced the debate in one project — whether changes to the standard auditor's report can better address investor needs. We've heard from many investors that the current auditor's report, which provides only a pass-fail opinion, is of limited relevance. Given the effort involved in an audit of a large company, and the complexity of many financial statements, investors want deeper insight from the auditor, especially in the aftermath of the financial crisis.

In March, the Board held an open meeting to hear from its Chief Auditor on the staff's outreach about audit reports. The staff presented views they had received over several months from numerous in-depth meetings with dozens of people experienced in using or preparing audit reports, including investors, auditors, preparers, audit committee members, researchers, and others.

The Board's outreach effort, especially at such an early stage in the project, was unprecedented. In addition, the PCAOB's open meeting to discuss the range of opinion received was the first of its kind.

Based on this outreach, and the discussion at the Board's public meeting, the PCAOB is developing a concept release on several alternatives for changing the auditor's reporting model, which we expect to issue this month.

The concept release will lay out a broad range of options, including more insight into procedures already included in the audit: i.e., areas of disclosure the auditors already review. Some of these options are already in use in other systems. Whether they make sense here will be part of our analysis. Marty Baumann, the PCAOB's Chief Auditor, will talk about this project later today.

B.  Auditor Independence

Enhancing the relevance of the auditor's report will do no good if we don't at the same time provide investors a sound basis for confidence in the audit. An audit has value to the public only to the extent that it is performed by a third party who is viewed as having no financial stake in the outcome.

As I've described, though, auditors face real pressures to please their clients. Too often, PCAOB inspectors find that auditors have failed to exercise the required skepticism and have accepted evidence that is less than persuasive. Indeed, the examples are galling in their simplicity.

For example, PCAOB inspectors found at one large firm that an engagement team was aware that a significant contract was not signed until the early hours of the fourth quarter. Nevertheless, the audit partner allowed the company to book the transaction in the third quarter, which allowed the company to meet its earnings target. Although the firm discussed the timing of the transaction with the customer, it failed to obtain persuasive evidence of an arrangement for revenue recognition purposes in the third quarter. The company had been an audit client of the firm for close to 50 years.

Similar concerns have surfaced in the Board's enforcement program. In one case, an engagement partner helped his client avoid the possibility that an earnings announcement would have to be retracted after an overstatement was discovered shortly before the company intended to file. As the company looked for last-minute accounting adjustments that might offset the errors, the audit response to the situation included the partner initiating a 50-percent increase in the firm's planned tolerance for misstatements.[3] The company had been a client of the firm for 98 years.

To be sure, conduct such as this would be a problem no matter how long or short the auditor's tenure. In the early years of a relationship. the auditor might be trying to build a long-term relationship by pleasing the client. In later years, however, the incentive is to avoid being the engagement partner that lost the client. It's worth exploring how we can mitigate these incentives, and the answer may not be the same for both.

The PCAOB's efforts to address these problems through inspections and enforcement are ongoing. But considering the disturbing lack of skepticism we continue to see, and because of the fundamental importance of independence to the performance of quality audit work, the Board is prepared to consider all possible methods of addressing the problem of audit quality — including whether mandatory audit firm rotation would help address the inherent conflict created because the auditor is paid by the client.

The idea of a regulatory limit on auditor tenure is not new. Over the years, it has been considered by a variety of commentators and organizations. Through this public debate, the basic arguments both for and against mandatory term limits have been fairly well described. [4]

I won't revisit all the history now. But most recently, in 2002, Congress considered requiring firm term limits during the debates that led to the Sarbanes-Oxley Act. It ultimately decided that the idea required more study and directed the GAO to prepare a report. That report, issued in 2003, noted that the SEC and the Board would need several years to evaluate whether the Sarbanes-Oxley reforms — including audit partner rotation — were sufficient, or whether further independence measures are necessary to protect investors.

The PCAOB has now conducted annual inspections of the largest audit firms for eight years. Our inspectors have reviewed more than 2,800 engagements of such firms and discovered and analyzed hundreds of cases involving what they determined to be audit failures. We have conducted more than 1,500 inspections of smaller domestic firms and of non-U.S. firms. These include multiple inspections of hundreds of those firms. And our inspectors have identified hundreds more cases involving what they determined to be audit failures.

Based on this work, I believe it is incumbent on the PCAOB to take up the debate about firm tenure and examine it, with rigorous analysis and the weight of evidence in support and against. I don't have a predetermined idea as to whether the PCAOB ultimately should adopt term limits. My only predilection is that the PCAOB deepen the analysis of how we can better insulate auditors from client pressure and shift their mindset to protecting the investing public.

As such, the Board plans to issue another concept release to explore whether there are other approaches we could take that could more systematically insulate auditors from the forces that pull them away from the necessary mindset.

We expect to issue this concept release around the same time that we issue the concept release on the auditor's reporting model, in order that they can be considered together in a holistic manner.

C.  Providing More Context for Audit Committees

We are also looking for ways to enhance audit committees' understanding of our inspection process. Like other regulatory examination processes in the world of financial and securities regulation, the details and results of any particular PCAOB inspection are, by law, largely nonpublic. That prevents us from disclosing to an audit committee the concerns we might have about its auditor, and possibly its audit — including in some cases our concern that the company's financial statements appear to be materially misstated in ways not identified by the auditor.

To a great extent, that information void can be filled by an auditor's candid discussion with an audit committee about an inspection. But we hear that auditors are often less than forthcoming with audit committees that try to elicit information about inspection results.

I recognize that firms may approach such audit committee discussions with one eye on taking care not to waive any privilege the firm might have, in a different context, against compelled disclosure of inspection information. That caution, however, does not explain other more troubling assertions by firms — such as that a particular audit deficiency cited by our inspectors is based on nothing more than incomplete documentation; or that it reflects merely a difference of professional judgment within a range of reasonable judgments.

An audit committee armed with a proper understanding of our process would recognize that those kinds of assertions are seriously suspect. Those assertions are, without exception, directly at odds with the considered collective conclusion of a group of very experienced auditors on the inspection staff.

Such a conclusion means that, in a concrete, identifiable respect that is not reducible to a mere difference in professional judgment, the inspections staff has determined that the firm failed to perform an audit that provides what the audit committee contracted for and what investors deserve – reasonable assurance about whether the financial statements are free of material misstatement.

In addition, in response to deficiencies cited by inspectors, firms often represent that they have complied with applicable standards governing an auditor's conduct with regard to post-opinion indications of possible deficiencies.[5] Typically, the representation is that the firm has taken necessary follow-up steps and determined that it can support its previously issued opinion. Too often, however, that representation seems at odds with reality as it appears to the inspection staff.

The PCAOB can neither open its inspection files to audit committees nor compel auditors to disclose inspection information to them. But we can and should help audit committees be better informed about our processes and better equipped to engage with their auditors about inspections without settling for responses that distort the significance of inspection results. You will hear more from us along these lines in the near future.

D.  Audit Transparency

Finally, I expect the Board to consider certain initiatives relating to audit transparency. In July 2009, the Board issued a concept release seeking comment on whether the Board should require engagement partners to sign audit reports.

The Board received thoughtful comment letters. Fair arguments were made both for and against the concept. The next step will reflect due consideration of those arguments and articulate a compromise approach that would retain the substance of the benefits discussed while addressing legitimate concerns.

I expect this proposal to take up another disclosure issue as well. Specifically, for many large, multi-national companies, a significant portion of the audit may be conducted abroad. Many of those non-U.S. firms, including the affiliates of the largest audit firms, are independently registered with the PCAOB, but are based in jurisdictions that bar our inspections.

Nevertheless, investors are left with the impression that the signing firm performed all the procedures described in the audit report. That is generally not the case. (Nor is it always the case that the PCAOB has inspected affiliates that participated.) Enhancing transparency about how multi-national audits are conducted should help investors and audit committees gain a better understanding of how an audit was conducted and make more informed decisions about how to use the audit report.

*   *   *

This is an ambitious agenda. Rethinking the structural foundation for auditing is not only in the best interest of investors, but in the best interest of auditors and preparers.

We will, of course, work closely with our colleagues at the SEC as we move forward. We plan to hear all the arguments, gather all available data, and immerse ourselves in the details (however devilish), fairly and dispassionately. Whether you are an investor, auditor, preparer, audit committee member, academic scholar or other interested party, we will need your involvement.

Thank you.


[1]  Chris Giles, Kohn Apologises for Fed Role in Crisis, Fin. Times, May 18, 2010, at 4.

[2]  Rule 2-01(c)(8) of Regulation S-X.

[3]  Thomas J. Linden , PCAOB Rel. No. 105-2009-004 (Aug. 11, 2009).

[4]  To give a bit of this history, in 1977, in the wake of the Equity Funding and other scandals, Senator Lee Metcalf published a wide-ranging study of the American accounting establishment. The Metcalf Report noted that "long association between a corporation and an accounting firm may lead to such a close identification of the accounting firm with the interests of its client's management that truly independent action by the accounting firm becomes difficult." The Accounting Establishment: A Staff Study, Subcommittee on Reports, Accounting and Management of the Committee on Government Operations, United States Senate 21 (1977). In recommending reforms, he proposed that "one alternative is mandatory change of accountants after a given period of years, or after any finding by the SEC that the accounting firm failed to exercise independent action to protect investors and the public." Id.

A year later, the AICPA's Cohen Commission considered a variety of proposals to increase auditors' ability to resist management pressure, including audit firm rotation. The Cohen Commission identified two potential benefits of such a requirement. First, "[s]ince the tenure of the independent auditor would be limited, the auditor's incentive for resisting pressure from management would be increased." Second, "a new independent auditor would bring a fresh viewpoint." The Commission on Auditor"s Responsibilities, Report, Conclusions, and Recommendations 108 (1978).

Ultimately, though, the Cohen Commission recommended against requiring rotation, because it believed that "the cost of mandatory rotation would be high and the benefits that financial statement users might gain would be offset by the loss of benefits that result from a continuing relationship." Id. at 109.

[5]  See AU 390, "Consideration of Omitted Procedures After the Report Date," and AU 561, "Subsequent Discovery of Facts Existing at the Date of the Auditor’s Report."