I support the issuance of this concept release and I welcome the robust and thoughtful discussion that it will surely provoke. Simple human experience tells us that any arrangement in which one party pays for the services of another is likely to lead to a certain, albeit perhaps limited, commonality of interest between the parties. That commonality of interest raises the question whether the party being paid can ever truly be independent of the payer as our accounting and auditing standards and securities laws require. To use a humble analogy, most creatures do not bite the hand that feeds them. In the case of the auditing profession, a panoply of independence rules and professional standards are designed to counteract the potential for conflict that seems inherent in the current auditor payment model and judicial decisions have at times imposed heavy penalties on auditors who fail to meet those standards. Indeed, the Board itself has been the source of some of those rules. Out of the questions for consideration here is whether these constraints have achieved their purpose.
Despite the numerous rules, regulations and judicial decisions intended to promote auditor independence, the Board's inspection program, now having included the detailed examination of several thousand independent audits, continues to reveal a troubling number of audit deficiencies many of which appear to be attributable to a lack of appropriate skepticism on the part of the auditor. This approach absence of skepticism may be attributable to many factors, but one may be a subtle restraint on the auditor's willingness to challenge management judgments for fear of losing a client and a stream of revenue. Experience also leads at least this Board member to suspect that while auditors may want to do the right thing, no auditor wants to have a dissatisfied client or to be the reason his or her firm lost a long standing audit engagement, particularly if the sources of disagreement do not have answers that are crystal clear.
The questions being raised in this release have been addressed by knowledgeable people periodically over many years, usually without definitive resolution. So why raise them again at this point? There are a number of answers. Perhaps most important, the world's economic and business environments are rapidly evolving. Accounting standards have been evolving equally rapidly in response to new economic realities and financial statements now increasingly include estimated amounts for highly complex assets and liabilities that must be reported at fair value rather than historical cost. The market value of these assets are often not easily ascertainable. As management's judgments about these items as reported in financial statements grow ever more important and ever more complex, the job of the independent auditor grows ever more difficult. Indeed, because there is often a high degree of uncertainty in these judgments the auditor may frequently be put in the uncomfortable position of questioning strongly held views that management believes it has reached carefully and in good faith. In this environment it is reasonable to ask again whether the current combination of the auditor payment model and independence rules, even coupled with individual audit partner rotation requirements, is adequate to deal with the potential conflict of interest that appear to be inherent in the auditor payment model.
Alternatives to the current model that have periodically been discussed are mandatory audit firm rotation, mandatory periodic audit tendering requirements, alterations to the audit payment and auditor selection model, joint audits and others discussed in the release. Each of these approaches has costs and limitations. Mandatory audit firm rotation inevitably increases costs as a new auditor becomes familiar with a new client, and may pose greater risks of audit failure, particularly in the early years, where the new auditor fails to find a problem simply from lack of familiarity with the new client's business. Indeed, what data does exist on the subject of auditor rotation (most of which is inconclusive) suggests that audit failures tend to be higher in the first years of an auditor's tenure with a new client. Mandatory tendering would impose costs both in terms of time and money on firms making proposals for new audits and on audit committees that would be required to evaluate the proposals. Joint audits run the risk of either duplication of work or areas of appropriate audit inquiry dropping between the cracks.
The question remains whether the current auditor payment model in fact poses an inherent conflict of interest for an auditor, how serious such a conflict is, whether existing rules and standards or additional rules and standards can sufficiently offset that conflict so that audits can truly be said to be independent or whether alternatives, taking fully into account the costs of change, would yield a net improvement in auditor independence and audit quality.
Finally, I would like to stress that in this, as in all other instances where we consider regulatory change, we take seriously the Hippocratic maxim, that has application to anyone attempting to ameliorate anything, of "first, do no harm". We look forward to a vigorous debate.
I would also like to thank the Office of the Chief Auditor, the Office of General Counsel, and the Division of Registration and Inspections for their time and effort in developing this release.