I am making this submission in a purely personal capacity. Over the last 25 years, I have been Chairman or a director of ten different companies, from small privately owned to FTSE100, participating in some 200 audit committee meetings. I have never worked for an auditing firm, big or small. I am now Chairman of Flybe Group plc, a non-executive director/Chairman of the Audit Committee at Watkins Jones plc and Chairman of the Audit Committee of Dentsu Aegis Network.
The CMA needs to define what a high-quality audit looks like.
The CMA notes ‘widespread public concerns’, but concedes that part of the problem may be an ‘expectation gap’, where commentators do not understand what an audit is intended to do. However, the CMA doesn’t define what it thinks an audit should achieve and what a high-quality audit should look like, nor is there any description of a ‘poor’ audit. The latter needs to distinguish between concerns that the auditor signed off; a going concern (but the company subsequently went bust); the ‘wrong’ number; a number biased towards management; or where it failed to detect fraud.
The CMA should define clearly its objectives in enhancing audit quality and trace through remedies to show how they would address the specific issues noted.
The CMA should consult with Audit Committees to understand what they think a high- quality audit looks like. They would find that it differs significantly from the quality benchmark applied by the AQR. The latter essentially audits an audit, looking at technical process and documentation. The Audit Committee looks for an audit to identify issues, challenge management’s assumptions and identified risks, propose improvements to controls, and work with management to finalise accounts that satisfy all regulation in a timely manner.
The CMA seems to imply that a poor-quality audit is one where the auditors are lax in agreeing whatever number management wants to declare. However, this is a very simplistic interpretation of quality. The CMA would do better to understand a quality audit in terms of the process of applying technical standards, speediness of response, identification of control and risk issues, collaborative working and good technical judgement. Crucially, only management and Audit Committees are able to judge most of these facets of quality, and they also have the strongest incentives to employ an auditor who can deliver on them.
Most of the ‘public concerns’ stem from a few well-publicised ‘failures’. These are important, but need to be out in the context of many thousands of audits that are completed successfully. If the CMA wishes to address these ‘failures’, it must study them to understand what actually happened. Then it should test its possible remedies to show that they would have stopped, or at least reduced the risk of, these events happening. The CMAneeds to be careful not to impose wide-reaching remedies just to address relatively rare individual failures.
The CMA needs to be realistic about the failures of previous regulatory interventions and learn from these when proposing new ones
The CMA takes satisfaction that, despite ‘only three years in force, the Competition Commission (CC) remedies have generated some positive change in the operation of the audit market, with increases in both tendering of audit contracts and switching.’ Since retendering and switching became compulsory, it’s not too surprising to see it happening. However, the CMA provides no evidence that this has made a positive change to the market, as opposed simply to increasing churn.
The CMA argues that ‘Alongside this, the FRC has reported broad increases in quality, albeit sampling means that we should be cautious in interpreting a trend over time.’ Indeed, the CMA should be cautious, since the AQR reviewed only 6% of audits within its scope. This included 24 – only 4% of those in scope – at much-criticised KPMG.
The AQR admits; ‘Our report focuses on the key areas requiring action by the firm to safeguard and enhance audit quality. It does not seek to provide a balanced scorecard of the quality of the firm’s audit work.’ In fact, the AQR targets particular ‘problematic’ sectors for review, underlining that this is not random sample.
The CMA admits that the objective of reducing the dominance of the Big Four has signally failed. 97% of FTSE350 audits are now performed by the Big Four, up from 95% at the time of the last review.
The CMA should make a balanced appraisal of the benefits of the previous CC remedies. The 2013 remedies imposed significant extra cost on companies and audit firms, and the regulator should feel accountable for those costs being borne by the market to deliver whatever market benefits have been achieved. The CMA admits that the dominance of the Big Four has increased and public confidence is reducing. How can this be reconciled with the CMA’s conclusion of ‘some positive change’?
With a new review and remedies being proposed, the CMA should be fully confident, with real evidence to support, that any new remedies will indeed be effective and proportionate, and will not have counter-productive unintended consequences.
The CMA needs to understand its own limitations in this market
The CMA doesn’t tackle the inherent contradiction between regulation and competition. Since auditors are themselves regulators of financial reporting, is greater competition between regulators going to produce better regulation? After all, there Is only one CMA. The CMA must acknowledge that greater competition (even assuming the CMA could deliver this) may not be the solution to improving audit quality. There may be a better solution in improving the regulatory oversight of audits.
Without an understanding of what audit quality is, it is not clear how the CMA will assess the effectiveness of the AQR process (3.21). The consultation says that the CMA will ‘explore what quality means’, but appears to prejudge this by saying it will be ‘building on the CC’s report that quality involves scepticism, objectivity, integrity and independence’ (3.34 d). It then says that “We do not expect to focus our work on theme 1 (Scope and purpose of audit” (3.46). The consultation is confused and contradictory.
The CMA is correct that there is a lack of choice in selecting an auditor
The question whether there is sufficient choice of auditor is a pertinent one. The CMA will need to address whether this is a UK domestic or international problem. There may be little point in designing a solution for the former, if the problem is the latter, or if addressing the former makes the latter more problematic.
The CMA should also be aware that some companies now feel that they have reduced choice as some audit firms are actually competing with them (3.25). For example, the booming area of data analytics, increasingly being offered by the Big 4, is also a key business area for many media, IT and consultancy companies. Can you be regulated by another party that competes with your business, and do you want to open up your business to a competitor?
The CMA analysis of auditor selection and perverse incentives is naive and not backed by evidence
The demand for independent external audit arises from the social or stakeholder need for reliable financial information, not just shareholder’s needs (para 3.4 as then noted in para 3.5). This sets up a false analysis of incentives. To say that ‘the auditor is selected and paid by the company’ (3.8, error repeated in 3.18) is to ignore corporate governance that forms the heart of UK listed company regulation. The auditor is selected by independent non- executive directors, not the company.
In any case, the proposition that management only wants low prices and shareholders only high quality is naïve and not backed by any evidence. In my experience, management wants a quality audit above everything else. The consequences of having to deal with poor quality auditors during the highly time-pressured Results process are significant. The dominance of the Big Four in winning tenders reinforces that companies put a high value on their perception of quality, as smaller audit firms usually charge less and may be in a weaker position faced with management pressure. Management has to work with auditors. It is unlikely that they would seek out poor quality audit.
The scope and purpose of audit may be determined by international rules, but the CMA should try to understand what value companies expect to get from an audit. In the ten or so audit tenders, in which I have participated in, audit firms place little emphasis on how accurately they apply requisite standards. This is rightly taken as read, with the emphasis on a quality service to the company. Furthermore, the general attitude of audit committees in a tender is to select the best audit firm and then negotiate the price, not vice versa.
The scope is aimed at large listed companies, but the market is much bigger than this
The scope will be large companies. Why is the scope limited to large companies (3.46)?
The concerns about agency are much reduced in private companies, especially where management may be very close to shareholders, but again there may not be independent non-executives running an audit committee. However, the vast bulk of the consultation is devoted to large listed companies, so how will the CMA fully understand the position in private companies?
The CMA should avoid designing remedies just for the FTSE 100 that then get applied inappropriately to the whole market for audit services.
The potential outcomes may be counter-productive
The list of potential remedies emphasises the difficulty of applying competition solutions to a regulatory issue. There is undoubtedly a competition issue in that dominance of the Big Four reduces effective choice for companies. However earlier, well-intended regulatory reforms have actually made this worse. The CMA’s forerunner, the Monopolies and Merger Commission waived through the merger of Coopers & Lybrand and Price Waterhouse in 1998, reducing the Big Six to Big Five, despite much protest from industry. The restrictions on non-audit work by auditors has actually reduced choice and mandatory tendering has in fact strengthened the oligopoly of the Big Four.
Several of the remedies suggested are likely to reduce audit quality, for example; by shrinking audit firms to audit only; forcing some companies to take non-Big Four audits (the consequence of a market cap on the Big Four) and forcing inefficient joint audits. Other ideas, such as a regulator appointing auditors have no coherent connection with increasing choice.
The CMA should;
Define what it means by a quality audit, after understanding what audit committees and management want and value in an audit.
Be clear about the difference between the benefits of competition and higher quality regulation, and not try to achieve the latter by the former.
Recognise the limitations of competition policy and not propose measures in order to be seen to do something under political pressure.
Address the specific causes of public disquiet about audits and test any remedies against whether they would have avoided well-publicised company ‘failures’.
Develop an evidence-based case for any further competition measures that takes explicit account of costs generated and fully understands the possible unintended consequences.