Comments from ACCA to the Public Company Accounting Oversight Board (PCAOB), 2011.
ACCA is pleased to provide comments relating to Rulemaking Docket Matter No. 37. Since the outbreak of the global financial crisis, we have provided comments on independence issues and audit firm rotation to standard setters, regulators and lawmakers, most recently the European Commission (EC). While this submission is consistent with our earlier views, we have extended our thinking in the light of indications from the EC as to the proposed direction of European legislation and developments in thinking on reports on audited financial statements and audit quality more generally.
ACCA is not in favour of mandatory rotation of audit firms. We believe that the provisions in the International Ethics Standards Board for Accountants Code of Ethics for Professional Accountants concerning the rotation of key audit partners provide a more appropriate safeguard to the familiarity threat to independence.
In this section of our response, we address the general issues highlighted in section D General Questions, of the Concept Release.
Should the Board focus on enhancing auditor independence, objectivity and professional scepticism? How significant are the problems in those areas relative to problems in other areas on which the Board might focus? Should the Board simply defer consideration of any proposals to enhance auditor independence, objectivity and professional scepticism?
The Board will be aware of developments in the European Union flowing from the 2010 European Commission's green paper Audit Policy: Lessons from the Crisis. While a formal announcement of legislative proposals was made in late November 2011 (too late for this submission to consider them), draft proposals have been widely available. They include measures to address the above and, should the proposals be confirmed, the Board in its deliberations could consider them immediately.
Auditor independence, objectivity and professional scepticism are all relevant to audit quality. The wider aspects of audit quality are under active consideration by the International Auditing and Assurance Standards Board (IAASB). The IAASB and the Board are also considering possible revisions to reports on audited financial statements, which may enhance user perceptions regarding factors relevant to audit quality.
ACCA believes that audit quality is deserving of holistic consideration by the Board and that possible piecemeal changes directed at independence, objectivity and professional scepticism might validly be delayed.
Would audit firm rotation enhance auditor independence, objectivity and professional scepticism?
The theoretical position was enumerated by the Cohen Commission in 1978 and is we believe still widely accepted - that firm rotation increases (at least the perception of) auditor independence. Its actual impacts on objectivity and professional scepticism are, however, dependent on individual circumstances and it is therefore difficult to generalise about their magnitudes.
What are the advantages and disadvantages of mandatory audit firm rotation? If there are potential disadvantages or unintended consequences, are there ways a rotation requirement could be structured to avoid or minimize them?
The Concept Release fully enumerates the advantages and disadvantages of mandatory rotation. The disadvantages can be attributed to (1) forcing an audited entity to replace its optimum auditor with one that is potentially inferior and (2) the incoming auditor has to accumulate client knowledge.
We see no regulatory mechanism to minimise the first disadvantage but the second could be addressed by enabling and requiring close cooperation between incoming and outgoing auditors. There would remain nevertheless, for all parties, the additional costs of familiarisation and the costs of the tendering process.
Because there appears to be little or no relevant empirical data directly on mandatory rotation available, should the Board conduct a pilot program so that mandatory rotation of registered public accounting firms could be further studied before the Board determines whether to consider developing a more permanent requirement? How could such a program be structured?
Conducting a pilot program would be an innovative approach and would pose many practical challenges. Irrespective of the likelihood of extracting valid empirical data, we doubt whether the time necessary for such a pilot is available given the pace of change in capital markets. As a minimum, such a pilot would have to consider the impact of a full cycle of rotation after implementation. Overall, we would not expect fully analysed results to be available any earlier than a decade after announcing the pilot.
According to the 2003 GAO Report, large firms estimated that a rotation requirement would increase initial year audit costs by more than 20 percent. What effect would a rotation requirement have on audit costs? Are there other costs the Board should consider, such as the potential time and disruption impact on company financial reporting staff as a result of a change in auditors? Are there implementation steps that could be taken to mitigate costs? The Board is particularly interested in any relevant empirical data commenters can provide in this area.
We see no reason to doubt the estimate, quoted above, that initial year audit costs would be increased by more than 20 percent. Depending on the length of tenure of the auditors, the annualise cost would be commensurately smaller but the value of the 'extra spend' has to be judged against the perceived benefits and indeed whether the money could be spent more efficiently on other ways to deliver such benefits.
On a change of auditor, there are undoubtedly significant calls on the company financial reporting staff and the Board should consider these. In contrast to staff of auditors, company financial reporting staff (and indeed other operational staff affected) do not usually record and account for their time, which makes it less easy to quantify such costs. If the Board does undertake a pilot study (as suggested in one of the Board's questions), this aspect could be investigated more fully.
As we said in our answer to an earlier questions regarding mitigation of costs, enabling and requiring close cooperation between incoming and outgoing auditors could reasonably be expected to reduce the costs of rotation by improving knowledge transfer.
A 2003 report by the Conference Board Commission on Public Trust and Private Enterprise recommended that audit committees consider rotation when, among other factors, "the audit firm has been employed by the company for a substantial period of time eg, over 10 years. To what extent have audit committees considered implementing a policy of audit firm rotation? If audit committees have not considered implementing such a policy, why not? What have been the experiences of any audit committees that have implemented a policy of rotation?
As the Board will be aware, in December 2010 the UK Financial Reporting Council updated its Guidance on Audit Committees (formerly known as the Smith Guidance), which was first published in 2003. Under that guidance, the audit committee should assess annually the qualification, expertise and resources, and independence of the external auditors and the effectiveness of the audit process. The committee should explain in the annual report to shareholders how it reached its recommendation to the board on the appointment, reappointment or removal of the external auditors. This explanation should normally include supporting information on tendering frequency and the tenure of the incumbent auditor.
Thus while it has been normal for companies listed in the UK to consider each year the choice of external auditors, none have opted to implement a system of fixed term mandatory rotation.
We believe this to be the prevalent position in major capital markets and one that demonstrates no support amongst audit committees of mandatory auditor rotation.
Are there alternatives to mandatory rotation that the Board should consider that would meaningfully enhance auditor independence, objectivity and professional scepticism? For example, should broader alternatives be considered that relate to a company's requirement to obtain an audit, such as joint audits or a requirement for the audit committee to solicit bids on the audit after a certain number of years with the same auditor? Could audit committee oversight of the engagement be otherwise enhanced in a way that meaningfully improves auditor independence?
Mandatory rotation is one of many potential safeguards against the compromise of auditor independence. It is relatively attractive as a mechanism as it is a very visible indication of independence. Its popularity as a 'solution' has risen in recent years due in part to its perceived value in addressing audit market concentration.
It must not be forgotten, however, that mandatory rotation is primarily a safeguard of the appearance of independence. It does not diminish the need for there to be a host of other safeguards as enumerated in the independence provisions of, for example, the International Ethics Standards Board for Accountants Code of Ethics for Professional Accountants (the IESBA Code). Indeed, an expectation gap may develop if investors believe that under mandatory rotation auditors are automatically more independent.
The alternative mechanisms mentioned in the question (joint audit, mandatory tendering) also suffer from the drawback that they apparently achieve enhanced auditor independence, when in reality many other factors must be taken into account.
Responses to the EC Green Paper from ACCA and others provide views inter alia on the relative merits of measures to enhance auditor independence. We believe that it is better to consider how these work individually and in combination, rather than attempt to impose solutions, such as mandatory rotation. There is a complex interplay between the effect of measures on perceived and actual independence and their attendant costs and their positive or negative effects on audit quality. We believe that an holistic approach to audit quality is necessary, in which independence plays but one part.
Should the Board continue to seek to address its concerns about independence, objectivity and professional scepticism through its current inspection program? Is there some enhanced or improved form of inspection that could better address the Board's concerns? If mandatory rotation were in place, could an enhanced inspection, perhaps focused particularly on professional scepticism, serve as a substitute in cases in which it would be unusually costly, disruptive or otherwise impracticable to rotate auditors?
Inspection by the Board is an important promoter of audit quality so long as auditors are held to account against criteria that are appropriate drivers of such quality. While there exist standards for the performance of audit engagements, quality control and ethics, there are as yet no overall accepted standards for audit quality. The Board will be aware of the current project of the IAASB examining audit quality and will no doubt wish to continue its involvement in that as well as its own relevant considerations.
Professional judgements made by auditors can reasonably be expected to be influenced by independence, objectivity and professional scepticism. It is important that the inspection regime pays particular attention to the acceptability of professional judgements in the particular circumstances and, where they are open to challenge, establishes the likely cause of any deficiencies.
As we have noted above, the primary role of mandatory auditor rotation is to provide a visible indication that certain threats to independence have been addressed by the safeguard of restricting an auditor's length of tenure. We do not believe that enhanced inspection can serve as a substitute for this primary characteristic. Enhanced inspection could, nevertheless, have a broader focus in circumstances where the risk to audit quality is higher, for example where mandatory rotation has been foregone because it is judged to be unusually costly, disruptive or otherwise impracticable to rotate auditors.
In this section of our response, we answer certain of the numbered questions in the Concept Release. Where we do not answer a specific question, it is because we have nothing significant to add to the information and references provided in the Concept Release. For completeness, we note that this is the case for the following questions: 3, 9, 10, 11 and 12.
Question 1
If the Board determined to move forward with development of a rotation proposal, what would be an appropriate term length?
On the assumption that different term lengths give rise to different costs and benefits, there may theoretically be one length of term that could be accepted as being generally more appropriate. Discussions in relation to the rotation of audit engagement partners have resulted in the adoption of rotation terms for individuals of the order of 5 to 7 years, although it is recognised that there may be circumstances, such as a change of ownership of the company, that merit an extension of the ordinary period.
ACCA has no particular view on the length of a general term for auditor rotation, as we would look to the outcome of any relevant research on this matter. We caution that provision should made for there to be a response to circumstances in which a mandatory rotation would impose an unreasonable burden on the company concerned.
Question 2
Should different term lengths for different kinds of engagements be considered? If so, what characteristics, such as client size or industry, should this differentiation be based on?
As we said in answer to question one, there is an underlying assumption that different term lengths give rise to different costs and benefits, and that there may theoretically be one length of term that could be accepted as being generally more appropriate. Some of the factors considered in making that determination may relate to characteristics of the engagement such as those referred to above.
It may be the case that there would be agreement that the appearance of independence is more important when the client is of a significant size, or involved in an industry such as banking, and is therefore of high public interest. We expect there to be a simplistic view that a shorter rotation period equates to a higher degree of independence. This may not actually be the case and indeed, if the focus of the investigation is the quality of the audit, that overall quality may diminish. Nevertheless, if for certain companies in which there is high public interest the optimum term of auditor engagement is more than a year different to the norm, there would be an argument to say that there should be some stratification in the application of the rotation model. It is not just high public interest companies that should influence such thinking as smaller companies may justifiably claim that their investor interests are best served by a longer continuity of auditor.
Thus, any consideration of the length of tenure of an auditor under mandatory rotation should we suggest take into account issues of public interest and not reject the possibility of stratification of the requirement.
Question 3
Does audit effectiveness vary over an auditor's tenure on a particular engagement? For example, are auditors either more or less effective at the beginning of a new client relationship? If there is a "learning curve" before auditors can become effective, generally how long is it, and does it vary significantly by client type?
The Concept Release summarises relevant studies and views and in the absence of ACCA research on these particular issues, we do not respond further to this question.
Question 4
Some have also suggested that, in addition to being less effective at the beginning of an engagement, an auditor may be less diligent toward the end of the allowable term. On the other hand, others have suggested that auditors would be more diligent towards the end of the allowable term out of concern about what the replacement auditor might find. Would auditors become more or less diligent towards the end of their term?
Does the answer depend on the length of the term?
We are not convinced that either of the causes mentioned above would have a significant impact on auditor diligence. Theoretically, the longer the term of tenure, the less impact effects at the end of it would have on the overall quality of the appointment over the years.
Question 5
How much time should be required before a rotated firm could return to an engagement?
Mandatory rotation is aimed at removing the threat to auditor independence created by a possible income stream for many years into the future, and at breaking the familiarity between the audit firm and the audited entity. In the absence of collusion between the firm and audited entity, mandatory rotation removes the guarantee of future income immediately a firm is rotated off an audit. The length of prohibition on retendering does not affect this factor. In the case of familiarity, it could be argued that a break of less time than the ordinary period before mandatory rotation would not be sufficient to have an impact, particularly if the firm provides non-audit services to the client.
Question 6
Should the Board consider requiring rotation for all issuer audits or just for some subset, such as audits of large issuers? Should the Board consider applying a rotation rule to some other subset of issuer audits? For example, are there reasons for applying a rotation requirement only to audits of companies in certain industries?
As we indicated in our response to the general questions in the Concept Release, there are theoretical reasons why different rotation requirements might be appropriate. Accordingly, the Board may choose to implement rotation requirements for subsets of issuers as indicated above.
Question 7
To what extent would a rotation requirement limit a company's choice of an auditor? Are there specific industries or regions in which a rotation requirement would present particular difficulties in identifying an auditor with the necessary skills and expertise? Is it likely that some smaller audit firms might decide to leave the public company audit market due to the level of uncertainty regarding their ongoing client portfolios?
We agree with the Board that the purpose of it adopting any rotation requirement would be to enhance auditor independence, objectivity and professional scepticism. We do not believe that there is sufficient evidence linking a rotation requirement with increasing competition in the audit market to consider that as a further reason for requiring rotation. Indeed, there are contrary indications from recent experience in Belgium, where the introduction of compulsory tendering has, anecdotally, resulted in firms losing audit work to the Big Four, which audit committees view as a 'safe option'.
There are undoubtedly practical difficulties in implementing a rotation requirement where there is a restricted choice of auditor. This may be because of the capacity and expertise required to service a particular type of client, for example a large multinational bank; or it may be that, in a particular jurisdiction, there is insufficient capacity generally. The current restrictions on the provision of non-audit services may also rule out potential audit firms as they provide non-audit services to companies that otherwise might wish them to tender for an audit. The practice also poses practical difficulties for global corporations wishing to preserve the arrangement of having a single audit firm network for the group.
Question 8
If rotation would limit the choice of auditors, are there steps that could be taken to allow a company sufficient time to transition out of non-audit service arrangements with firms that could be engaged to perform the audit? Are there other steps that could be taken to address any limitation on auditor choice?
It is not just during the initial implementation of a system of mandatory rotation that non-audit service arrangements would be problematic. Similar considerations would affect subsequent rotations and there may be significant cost implications of changing the supplier of a non-audit service. One could envisage a tendering process involving several firms that would be unable to carry out non-audit engagements so that, taken together with the prohibition on the existing auditor, it may make it very difficult for, particularly the larger, public corporations to find suitable providers of certain services.
Question 9
If rotation were required, would audit firms have the capacity to assign appropriately qualified personnel to new engagements? If they do not currently have that capacity, could firms develop it in order to be able to compete for new clients, and would they do so?
This question is best answered by the firms concerned.
Question 10
Would rotation create unique challenges for audits of multinational companies? For voluntary rotations that have taken place, what have been the implementation and cost issues and how have they been managed?
This question is best answered by the firms concerned.
Question 11
Would increased frequency of auditor changes disrupt audit firms' operations or interfere with their ability to focus on performing high quality audits? How would any such disruption vary by firm size? For example, would a rotation requirement pose fewer or more implementation issues for small firms than for large ones?
This question is best answered by the firms concerned. However, if mandatory rotation is introduced for a subset of companies it may be that an audit firm having relatively few such audits might be more significantly affected than one with a large number of appointments.
Question 12
Would audit firms respond to a rotation requirement by devoting fewer resources to improving the quality of their audits? Would firms focus more on non-audit services than on audit services?
We see no particular reason why this should occur, but the question is best answered by the firms concerned.
Question 13
Would rotation have any effect on the market for non-audit services? Would any such effect be harmful or beneficial to investors?
On the assumption that provision of a particular non-audit service would create an unacceptable threat to independence for an existing auditor, a prospective auditor would be deterred from providing that service in advance of a known possible appointment. This would have the effect of narrowing the choice of provider for the service, which may be harmful for investors as it prevents a company from using the optimum supplier.
Question 14
Some have expressed concern that rotation would lead to "opinion shopping," or that in competing for new engagements firms would offer favourable treatment. Others have suggested that rotation could be an antidote to opinion shopping because companies would know that they could not stick with a firm promising favourable treatment forever. Would opinion shopping be more or less likely if rotation were required? If rotation limits auditor choice, could it at the same time increase opinion shopping?
We do not believe that 'opinion shopping' is a significant driver of auditor change and hence its impact when rotation is mandatory would be minimal. Indeed if mandatory rotation after a set period of years is in place, auditor change in the interim would likely be viewed as an unusual matter deserving of full explanation, which would deter opinion shopping over a longer period.
Question 15
What effect would a rotation requirement have on competition for audit engagements? If competition would be increased, how might that affect audit quality?
While increased competition might be thought to be a driver of an increase in the number of audit firms available to undertake audit engagements it is more directly linked to intensification of price competition. The Bocconi report noted that, 'In the case of auditing which is generally considered as a "public interest" activity, this could be considered as inappropriate.'
As set out in our earlier answer to one of the general questions, '...mandatory rotation is primarily a safeguard of the appearance of independence.' Its impact, if any, on competition should be regarded as a consequence rather than a reason to introduce mandatory rotation.
Question 16
Are there any requirements the Board should consider to mitigate any risks posed by rotation? For example, are there enhancements to firms' quality control systems that might address such risks?
We address the risks to audit quality in our response to question 17. The responses to such risks inevitably add costs to audits and if fees are constrained then that introduces further risks to quality.
Question 17
If the early years of an auditor-client relationship pose higher audit risks than later years, should the Board require firms to provide additional audit supervision and oversight in the first year or two of a new engagement? Should the Board impose such a requirement for auditor changes even if it does not further consider requiring audit firm rotation? If firms are accepting new clients but are unable to perform quality audits for them until several years have passed, should the Board require enhanced client acceptance procedures? What impact would additional requirements of this type have on audit costs?
The matters at issue go much deeper than the question as to whether mandatory rotation should be implemented. Mandatory rotation would be expected to increase the number of occasions of auditor change and magnify the effect of audit deficiencies that are acknowledged to be more prevalent immediately after a change of auditor.
Audit failure is a natural consequence when audit is constrained by the limits of what it is theoretically possible to achieve and the economic and time limitations on audit procedures. Failures are more likely to occur where audits are more difficult, whether that is due to client characteristics (such as having a complex business model) or to general factors, such as whether the audit is 'first time through'. The introduction of mandatory rotation may indeed result in an increase in the number of audit failures unless correcting mechanisms are put in place. These could include strengthening of auditing and quality control standards and more effective inspection. Inevitably, the extra costs of such measures generally would be added to the specific costs incurred by auditors and companies implementing auditor rotation.
The overall rate of audit failure and the profile of those failures (for example, how many arise on the first time through) have to be judged for acceptability in the light of societal need (often interpreted by regulators and standard setters). It may be that the status quo is sufficient in that any worsening in audit failures brought about by mandatory rotation would be compensated by the benefits of increased independence overall.
Question 18
If mandatory rotation were required, are existing standards relating to communications between predecessor and successor auditors sufficient? Should additional communications be required? For example, should the outgoing auditor provide the incoming auditor with a written report outlining audit risks and other important information about the company?
Increased communication from outgoing to incoming auditors would be expected to assist in the building up of client knowledge. This should be the case in all changes of auditor, not just on mandatory rotation. The additional cost to the company would be expected to be small because extra costs for the outgoing auditor are balanced by cost savings for the incoming auditor. There may be circumstances where communication is precluded (for example in a dispute) but generally, subject to appropriate provisions being in place to mitigate liability, greater cooperation between auditors would make a positive contribution of audit quality.
Question 19
Are there other audit procedures that should be required to mitigate any risks posed by rotation?
Auditing standards are not predicated on the audit being 'first time through' and it would be worthwhile to consider whether, for a first time through audit, standards ought to prompts a greater degree of professional scepticism, restrict options and drive further procedures in specific areas.
In the absence of agreement on the desired impact on the level of reasonable assurance it is impossible to suggest specific other audit procedures that should be required. The approach in International Standards on Auditing (ISA) 240 The Auditor's Responsibilities Relating to Fraud in an Audit of Financial Statements might serve as a model, as a way of directing and amplifying the application of general requirements in the circumstances of a first time through audit.
Question 20
If the Board moved forward with development of a rotation proposal, should consideration be given to the recommendation for a cause restriction on the company's ability to remove an auditor before the end of a fixed term? Would such a provision be useful? Would there be unintended consequences of such a requirement? Should the Board work with the SEC on implementation of this recommendation? Are there other matters on which the Board should coordinate with the SEC?
We see no reason why auditors should not be appointed for a term longer than one year but we also recognise that the needs of a company may change. If adequate disclosure were to be required of the reasons for change we believe that the market would act as an appropriate mechanism to deter inappropriate auditor change.
Question 21
What other transition issues might arise in the first year of a rotation requirement? How should the Board address these issues?
On the assumption that most auditors will have been in office for longer than the rotation interval it is likely that transitional measures will be necessary to avoid almost all audits rotating in the first affected year. Such circumstances would further increase the costs to all parties and may challenge the capacities of audit firms and companies to achieve change.