Steve Clapham, founder of research and training firm Behind the Balance Sheet, sets out four simple steps that could be taken to dramatically improve audit quality.
A series of high-profile corporate failures have highlighted the unacceptable quality of auditing in the UK. The collapse of former FTSE 350 companies, such as Carillion, Debenhams and Thomas Cook – to name but a few – have quite correctly led investors to question whether they are receiving the protection they have a right to expect.
The reputation of the auditing profession is in tatters. Government and regulators are squaring up to act (see ‘Auditing the auditors’ box below). Earlier this year, Andrew Tyrie, chairman of the Competition and Markets Authority (CMA), warned that “addressing the deep-seated problems in the audit market is now long overdue”, pledging his organisation “will persist until the problems are addressed”.
Meanwhile, an independent review into the industry’s regulator, the Financial Reporting Council (FRC), concluded that the organisation needs to be replaced with a body capable of commanding respect and fear among auditors. The government is acting on the recommendations.
Whatever the FRC’s flaws, the watchdog itself has been highly critical of the audit industry, too. It targets 90 per cent of FTSE 350 audits to be classified as good or requiring only limited improvements – an official classification of 1 or 2A. However, its July 2019 report indicated that audit quality was significantly below its threshold, with a quarter of FTSE 350 audits falling below the acceptable standard. For Grant Thornton, which received a special mention for its poor performance, half of audits were below standard.
Stephen Haddrill, chief executive of the FRC, lamented: “At a time when the future of the audit sector is under the microscope, the latest audit quality results are not acceptable. Audit firms must identify the causes of their audit shortcomings and take rapid and appropriate action to improve quality. Our latest results suggest that they have failed to achieve this in recent years.”
FRC data shows these problems are nothing new.
What can be done
Four simple initiatives could go a long way towards solving the problem. These could help the Audit, Reporting and Governance Authority (ARGA), the new replacement for the FRC, to have a much higher success record than the current system.
The initiatives are:
1. FRC/ARGA letters to companies should be published openly.
2. Auditors should audit non-GAAP measures and publish an ‘acceptable’ EPS range.
3. Companies should disclose any conflict between audit committees and audit firm.
4. Audit committees should receive a formal report from the company’s broker on short interest in the stock with explanations.
Any one of these recommendations alone would generate a significant uplift in audit quality.
We’ve used the examples of three recent high-profile stock market disasters – Carillion, Debenhams and Thomas Cook – to illustrate how some of these initiatives (EPS ranges and short positions) could have offered clear warning signs for investors and auditors. We’ve also collected data on potentially valuable information that investors do not currently get ready access to (FRC letters and audit committee conflicts).
FRC letters
The FRC Conduct Committee issued 220 letters to companies in 2017 (2018 data have not yet been announced). These letters are treated as a confidential exchange with company management. This makes no sense.
In the US, the Securities and Exchange Commission (SEC) publishes comment letters when it writes to quoted companies. Notably, one such letter preceded Citron Research’s short-selling attack on Valeant Pharmaceuticals, which led to an $80bn reduction in its market capitalisation. Valeant’s practice of channel stuffing using ‘subsidiary’ Philidor was first raised by the SEC. There is no reason why the FRC or its successor body should not publish its letters. This could happen the day following (or two days after) their receipt by the company, to allow management time to frame its response to investors.
This simple change, which requires no additional work or cost, would create a climate of fear and respect among company managements – the last thing a finance director would want would be for the FRC (or ARGA) to publish criticism of their accounts.
A range of acceptable EPS
The non-GAAP (generally accepted accounting principles) earnings measures reported by the majority of companies are unaudited. Introducing a requirement to audit these critical data, relied on by investors, would introduce discipline into the process.
And instead of simply stating that earnings are true and fair, auditors should calibrate that opinion. They should provide a range of acceptable earnings per share. This will entail additional work and a lot of argument as to how to frame it, but that debate would be worthwhile, as it would force the profession to introduce greater precision in this critical parameter.
Of course, for some companies the range would be very wide, but that is surely useful information for investors. For all companies, it would be clear if the Finance Director was adopting a conservative or aggressive approach. Imagine the range at Carillion and where the reported number fell within that range – it would be a stark warning to investors.
In 2012, academics at Emory and Duke Universities published a study into earnings quality based on a survey of 169 finance directors of public companies and 12 in-depth interviews. They found that about a fifth of companies manipulate earnings to misrepresent economic performance, and for this cohort about a tenth of EPS was typically “managed”.
The researchers found that among the most reliable indicators for good earnings quality identified by finance directors were strong cash conversion and an absence of ‘one-time’ adjustments. We’re looking at these measures as proxy indicators of whether our three stooges – Carillion, Debenhams and Thomas Cook – may have been pushing earnings to the limit of (or beyond) an acceptable range in the five years prior to their demises. The cash-conversion charts are based on rolling five-year free-cash conversion, which has been used to smooth out the annual fluctuations in free cash flow that can be caused by large investment projects. As a rule of thumb, 80 per cent free-cash conversion can be considered decent. This level is shown as a dotted line on the cash conversion charts.
So much of Thomas Cook’s headline profits were accounted for by making “one-off” adjustments, that it has not been possible to construct a cash conversion chart. This is because the measure of cash conversion we use is based on rolling-five-year statutory profits, whereas Thomas Cook had losses. The company’s lavish adjustments were actually something its auditor EY did comment on in its final full-year audit report, stating: “We communicated to the Audit Committee that we had strongly recommended to management that they strengthen the process over the identification and approval of separately disclosed items.” With some prescience, the auditor also commented: “Covenants in the last two quarters of FY2019 remain tight.” That said, with hindsight the auditor was overly optimistic in taking assurance from the fact the covenants looked alright “under management’s severe but plausible scenarios.”
An acceptable EPS range is likely to have provided a much more valuable pointer for Thomas Cook shareholders than the measured warnings buried in the auditor’s 7,350-word report, which began by declaring the accounts “true and fair”.
Audit committee conflicts
Where the chair or another member of the audit committee used to be a partner of the company’s audit firm, the independence of the committee is called into question. Investors have a right to know such information.
In March, press reports highlighted criticism from shareholder advisory group Pensions & Investment Research Consultants (PIRC) of eight UK companies where there were potentially inappropriate connections between directors and the auditors. One example was BHP, the mining group listed in the UK and Australia, where Lindsay Maxsted, the chair of the firm’s audit committee, was until 2007 CEO of KPMG Australia; yet KPMG has been its auditor since 2003. PIRC considers such a relationship “inappropriate” and has recommend voting against the re-election of the director and reappointment of the auditor in previous years. This year, BHP proposes to replace KPMG with EY – a process set in train two years ago and expected to be voted through on 17 October.
Making it easier for investors to scrutinise potential audit committee conflicts, could prove a highly effective means to encourage good practice.
Audit committee information
A May 2019 paper, Audit Quality, by Eric Tracey – a former Deloitte Partner turned trouble-shooting finance director at Amey and Wembley – suggested that an independent body, for example the company’s broker, should report to the audit committee if there is a large short interest in the company’s shares and the reasons. This would ensure that the committee was aware of the concerns prevailing in the market. It’s a simple idea and would be cheap and quick to implement.
The chart on the right shows this would have certainly helped at Carillion where the stock was heavily shorted for a number of years before its collapse. Market practitioners were apparently more aware of the true situation than some of the board. There were also warning signs at the time of Debenhams’ final September year-end with shorts at over 15 per cent. This was seven months before the company went into administration. The major spike in Thomas Cook shorts, meanwhile, came only shortly before the finale. But with shorts running at over 5 per cent at the time of two previous year ends, more scrutiny of its bold profit adjustments could have been prompted.
Conclusion
Some of these measures would have a cost, notably the auditor framing an EPS range. Some would be free – the publication of FRC letters, for example. This is too important an issue to leave and it’s worth spending some time and money correcting it. Some of our proposals could be implemented tomorrow. Change is overdue.
