Problem: The big four accountancy firms are facing reform calls over perceived conflicts of interest, since they are paid by the companies they audit and sell consultancy services to the same clients.
Solutions: Complete separation of the audit and consultancy wings of the big firms or financial separation of audit and consultancy profits, together with some specific changes to the UK Corporate Governance Code on how auditors are appointed.
In April, the Financial Reporting Council’s CEO Stephen Haddrill announced it was time to look at how the audit could be improved to give a better alignment of interest between shareholders and auditors.
Of the various ideas in the press, the best is a complete separation of the audit and consultancy wings of the big firms and some specific changes to the Corporate Governance Code on how auditors are appointed.
While audits are meant to verify a company’s soundness for the benefit of investors, a string of failures including Enron and Lehman show it doesn’t always work. This has led people to question whether auditors should really be paid by the companies they police, and whether they should be allowed to sell consultancy services at the same time.
As a backdrop, most large companies are audited by the same four firms, which have near-identical partnership structures with mechanisms to share profits from consultancy and audit work. Meanwhile, fear of litigation is said to have made audits blander and less useful to shareholders, while the concentration of audit firms has left regulators terrified that another Enron will reduce the big four to a big three.
Ideas for reform include splitting the audit and consultancy wings of firms and finding more subtle ways to make auditors serve shareholders better. Unsurprisingly, the accountancy profession regards these as unnecessary and impractical. The necessity argument rings pretty hollow in the wake of recent audit failures: PwC’s failure to enforce the separation of client and bank monies at JPMorgan cost shareholders GBP 33m in fines, while Ernst & Young’s approval of the infamous Repo 105 at Lehman’s came shortly before the bank went bust. But practicality will be vital for any reform to be introduced.
One attractive idea, proposed in March by the Evening Standard’s Anthony Hilton, is to let the FRC allocate and pay for audits and recover the cost from the company or an industry levy. “The conflict of interest is obvious... it is time for auditors to be paid differently so there is no possibility of conflict of interest when they are paid by the people they are supposed to be keeping a check on,” he wrote.
His scheme would make it easier for medium-sized firms to grow and compete with the big four, but might need limiting so the FRC does not have to get involved with every single small company.
The FRC itself seemed bemused by the idea, and a spokesperson said it was not a serious industry proposal. Not yet, at least – Vince Cable called for a competition review of the big four when he was in opposition. If he follows through on this, perhaps Hilton’s FRC idea could make a come-back.
Meanwhile, Hilton writes today in a different article about a proposal from a former PwC audit partner to replace mandatory audits with voluntary ones. This would replace litigation-minded box-ticking with a cooperative environment in which auditors could work closely with the board that hired them and agree to provide challenging yet constructive feedback. On the surface, it might improve matters by switching the focus from box-checking to offering valuable advice. This would be great for well-behaved companies and auditors with good ideas to offer, but it would be terrible for avoiding the next Enron.
A better idea is to make a series of adjustments to the Corporate Governance Code to put shareholders at the centre of the auditor appointment process, instead of executives.
Here’s how the system works at the moment. Shareholders vote annually on a resolution to appoint an auditor. The resolutions are put forward at each AGM by the board, based on a recommendation from the independent non-executive directors on its audit committee. It sounds lovely in principle, but in practise the sticky fingerprints of executives can be seen across the whole process, and these are the very people the audit is meant to scrutinise.
While the AGM vote gives an impression of shareholder control, auditor resolutions are seldom voted down. One of the most active shareholder representatives in the market, Hermes, last year voted on auditor resolutions at just one of the nearly 900 UK AGMs it attended. Most shareholders wouldn’t even manage that as they don’t have enough information to bring a challenge.
The make-up of the audit committee gives an impression of independence, but the board often appoints and decides on the fees of members of the audit committee, and is always free to ignore their recommendations as long as it explains why. In other words, the executives on the board pull all the strings.
According to the FRC’s Guidance on Audit Committees, the audit committee must include at least two non-execs, one of whom should represent shareholders’ interests (but does not have to be a shareholder or a shareholder’s nominee). Members are appointed directly by the board or the nomination committee, which also happens to appoint the executives. The audit committee’s duties include scrutiny of any consultancy work that might compromise the audit, but committee members rely for their judgement on information provided by management. The directors who sit on the audit committee are paid extra for their efforts. Guess who sets the level of the fees? The board or the remuneration committee, which is itself appointed by the board. Can you see the pattern?
This arrangement could be vastly improved. First, the composition of audit committees needs to be put out of reach of the executives. Audit committees should include an independent representative of the biggest shareholder or shareholders. Other members could include, say, representatives of the company’s customers and/or regulators and be chosen by the nomination committee, which should itself comprise a majority of shareholder representatives instead of board nominees, as is the case in Scandinavia. This would mean that the people who choose the auditors are those that benefit from the audit, rather than those being policed by it.
To make audit committees even safer, their fees should be set at a flat rate, perhaps determined by the FRC, and their recommendations made binding on the board, subject to AGM vote so that minority shareholders still have a voice. This would remove all executive influence from auditor selection and encourage auditors to be tougher and more critical. The most successful auditors would be those who help shareholders the most in their role as stewards, providing ammunition for them to hold management to account.
This would solve some of the independence problem, but still leaves the temptation for auditors to suck up to management – regardless of who appoints them – to win consultancy work for their firm. As long as audit partners share consultancy profits, this will always be the case.
One idea is simply to ban audit firms from engaging in consultancy work at the same time as an audit, so firms must choose whether to be auditor or consultant for each client at any given time.
This reform would face less opposition from the industry and be easier to implement than a full separation. It would allow firms to hold on to the staffing and operational synergies that they claim justify their multi-disciplinary structure. On the downside, it could be highly distortive: firms might raise audit fees to compensate for lost consultancy work, and might resign from audits so consultant colleagues can sell solutions to problems "uncovered" during the audit. Big consultancy projects might face disruption when audit tenders come up, while companies that use a lot of consultants might struggle to find an auditor at all, unless they turn to smaller firms (which would at least improve competition).
A compromise reform might be to insist that big firms separate profits from their different activities, so that audit partners do not gain financially from consultancy wins. To some extent this happens already with more profitable partners being allocated more profit units. Going a step further and segregating the profit pools completely would be relatively simple and would still allow firms to enjoy cross-referrals and other benefits of size. It might weaken the business case for multi-disciplinary firms, leading firms to break themselves up voluntarily. But as a behavioural remedy, it would require supervision, which is always worse than a structural fix.
A full legal split of the profession, combined with reforms to the Corporate Governance Code, would be the best way to tackle conflicts of interest in audit.