At a London lunchtime meeting hosted by Hambros Bank Ltd on 23 November, Paul Grout and Geoffrey Whittington presented research on the UK auditing profession that confirms fears of `cosy relationships' between auditors and firms. These relationships can compromise the independence of auditing firms, and the profession is facing considerable criticism as a result, particularly in its perceived failure to provide adequate protection against corporate failure. But the speakers cautioned against a tightening of the regulatory structure or an increase in auditors' legal liability. Such a cure may prove worse than the disease itself, they argued: instead, reforms should aim to encourage auditors to exercise their professional judgement. Grout is Professor of Economics at the University of Bristol and Whittington is Price Waterhouse Professor of Accounting at the University of Cambridge. Their talk was based on an ESRC-funded research project on `Regulation in Accounting', and in particular their article in Economic Policy 19, `The Professional Judgement of Auditors: Regulation and Legal Liability', written with Ian Jewitt (University of Bristol and CEPR) and Chris Pong (University of Manchester).

Grout and Whittington's research reveals strong evidence of `cosy relationships'. On average, an auditing firm can expect to receive as much income from consulting work for the companies it audits as it earns in audit fees. In addition, investigation of approximately 3,500 audits of UK companies over the period 1980–88 indicates that only 2% of them were subject to genuine terminations. It follows that the expected life of an audit was in excess of 40 years. Most of the audits in question were large and relatively complex, and over 70% were conducted by the `Big 8' firms (now the `Big 6'). The research also provides evidence of `low balling' of audit fees, in which auditors bid for work at below cost. This means that audits are only profitable if they are maintained for many years: an audit's longevity, combined with associated consulting income, is expected to offset the initial losses necessary to win the business.

It is often argued that `cosy relationships' compromise the position of auditors and that steps should be taken to ensure their independence. But Grout and Whittington do not advocate a dramatic increase in regulation. While broadly in favour of present measures being adopted by the Accounting Standards Board and the Auditing Practices Board, they argue against moving too far towards restrictive and prescriptive regulation, and see merit in some limitation of liability. They believe that it is essential for auditors to be able to exercise professional judgement: limiting that scope may actually reduce the value of the information they provide to shareholders and the market. A balance needs to be struck between two undesirable extremes: in a world free of regulation and penalties, auditors may lose all credibility, as a result of which accounts would have limited value. But in an overly regulated world, where there are severe penalties and auditors face unlimited liability to a broad class of users of company accounts, they could become dominated by the rule book, afraid to exercise their professional judgement in seeking to report a `true and fair' view of the firms they audit.

Assessing whether a business is a going concern is an example of an auditing issue &nbspin which professional judgement plays a key role. Auditors will often have information that, by its very nature, is impossible to verify and communicate formally, but which may still provide valuable insights into a firm's future prospects. Favourable information of this type may encourage the auditors to take a positive view of those prospects. But if there are enormous penalties facing auditors, they may err too strongly on the side of caution, playing down positive information and presenting all negative information in the strongest light lest they become liable should the firm fail. On the other hand, there must be sufficient regulatory pressure to ensure that auditors are not cavalier in their presentation and can resist corporate attempts at influencing their reports. The penalties need to be sufficiently strong that auditors will not gain from succumbing to influence when an audited firm is unlikely to remain a going concern, but not so strong that auditors feel they must protect themselves in other cases by curbing the exercise of their professional judgement. It is beneficial for the transmission of information to the markets that auditors have an incentive for their clients to remain in business, provided that regulation is strong enough to make it disadvantageous to misrepresent the position of the weakest firms.

Grout and Whittington conclude that draconian regulation is not necessary and that there may even be a case for limiting auditors' liability. So how should auditors be regulated? They argue that the degree of latitude should depend on the nature of the financial markets which use the information provided by the profession. For example, in markets like the UK and the US where there are many small shareholders who do not exercise direct control over companies, very detailed audit information is likely to prove pointless or even counterproductive, discouraging auditors from exercising their professional judgement. Small investors may be prepared to forego accurate, detailed but closely circumscribed information for the vaguer information conveyed in `true and fair' company assessments. In contrast, owners of companies in a country like Germany where the stock market does not play such an important role will be much more interested in information of a kind useful to management. When banks hold seats on the board of a company, for example, their position and financial requirements are closely aligned with those of management itself. Thus, given the nature of the capital market in Germany, tough and closely prescribed accounting standards are probably more appropriate, while in the capital markets of the UK and the US, relatively flexible accounting standards are more suitable. This suggests possible dangers in too rapid EU harmonization of accounting practices, and Grout and Whittington argue that it is essential that harmonization should not be rushed nor take the form of averaging across existing legislation of the EU members.