Bias in Accounting Estimates - A Grey Area for Auditors
Author:
Ronan Nolan
Auditing financial statements has never been a question of a black and white, rules-based approach. In particular, dealing with the potential for bias in accounting estimates takes the auditor into a grey area where the exercise of professional judgement is of critical importance.
It is quite possible for an auditor to conclude that whereas no single accounting estimate falls outside a reasonable range of outcomes, the overall picture presented by the financial statements has been distorted. This can occur as a result of a biased approach by management in arriving at the totality of accounting estimates. In these circumstances even though no individual estimate constitutes a misstatement, there is an overall misstatement, which if material, will require resolution.
Background
The fact that many elements of financial statements are the result of accounting estimates is, of course, not new. One factor which changed preparers and auditors' perspective on this has been the relative downgrading of the traditional concept of prudence, and the overriding importance given to reliability in FRS 18, Accounting Policies, issued in December 2000. FRS 18 notes that: "there can also be tension between two aspects of reliability - neutrality and prudence. Whilst neutrality involves freedom from deliberate or systematic bias, prudence is a potentially biased concept that seeks to ensure that, under conditions of uncertainty, gains and assets are not overstated, and losses and liabilities are not understated. This tension exists only where there is uncertainty, because it is only then that prudence needs to be exercised. In the selection of accounting policies, the competing demands of neutrality and prudence are reconciled by finding a balance that ensures that the deliberate and systematic understatement of assets and gains and overstatement of liabilities and losses do not occur."
Similar language had been included in the ASB Statement of Principles for Financial Reporting issued in December 1999, which went on to state that: "When there is uncertainty, the competing demands of neutrality and prudence are reconciled by finding a balance that ensures that the deliberate and systematic understatement of gains and assets and overstatement of losses and liabilities do not occur."
Neutrality is an important objective, as underlined in the ASB Statement of Principles as follows: "Financial information is not neutral if it has been selected or presented in such a way as to influence the making of a decision or judgement in order to achieve a predetermined result or outcome."
The overall objective therefore, is to strike a balance, and not allow excessive prudence (or indeed the excessive lack of same!), whether evidenced in unnecessary provisions or reserves, or in the understatement of asset values, to override the presentation of a fair and neutral position.
Some Financial Directors, brought up on a strict diet of prudence, still find this difficult to deal with, and are shocked when auditors do not automatically accept a 'prudent' view as valid in all cases!
Guidance for Auditors
SAS 470, Overall Review of Financial Statements, issued in March 1995, provides some guidance for auditors in dealing with bias; auditors are required to review the financial statements as a whole for consistency and reasonableness, and in carrying this review, should have regard, amongst other factors, to the following:
"the potential impact on the financial statements of the aggregate of uncorrected misstatements (including those arising from bias in making accounting estimates) identified during the course of the audit and the preceding period's audit, if any."
An important point identified here is the need to have regard to bias not only in the current year's numbers, but also in the preceding year's estimates. A shift in the bias within the acceptable range of outcomes from one year to the next can give rise to a misstatement.
The APB Consultation Paper on Aggressive Earnings Management, issued in June 2001, also sets out some relevant thinking on this matter. It may be noted that this paper was as much concerned with the potential for bias through a lack of prudence as from an excess of that attribute!
Firstly, the need for estimation is clarified:
4 Valuations and estimates, inevitably requiring judgment, are needed in relation to many elements of an entity's financial statements, particularly in respect of transactions that span the year-end or several years; and
4 Businesses are often highly complex, operate in many countries and undertake transactions for which the accounting is also complex (e.g. the use of derivatives)."
Then the auditor's role is outlined: "When there is a difference between the auditors' estimate of the amount best supported by the available audit evidence and the amount included in the financial statements, auditors consider whether the estimate falls within a range of acceptable results or requires adjustment."
In concluding, the paper underlines the point made above in respect of the aggregate effect of individually reasonable estimates:
"When evaluating the results of the audit, the auditors consider whether judgments and decisions made by the directors and management, including any resistance to adjusting misstatements identified by the auditors, could be part of a pattern of bias, even though individually they may appear reasonable, to avoid the financial statements reflecting the underlying reality."
Future Developments
It is evident from the above that to date guidance for auditors in dealing with this rather complex area has not been very detailed. This has been recognised by IAASB which in December 2004 issued an exposure draft with the rather imposing title ISA 540 (revised) - Auditing Accounting Estimates and Related Disclosures (Other than those arising from Fair Value Measurements and Disclosures).1
The draft standard provides significant additional guidance which should be of assistance to auditors.
The proposed standard introduces requirements for "greater rigour and scepticism into the audit of accounting estimates, including the auditor's consideration of indicators of possible management bias".
Guidance for auditors is provided in the context of risk assessment procedures, and "estimation uncertainty" is defined as the "susceptibility of a financial statement item to a lack of precision in its measurement because the outcome of future events is not known".
The significance of this is that because of the sensitivity of some estimates to changes in assumptions, where there are several reasonable sets of assumptions, differing results can emerge.
The draft standard therefore requires auditors to focus "not only on estimates that have a risk of material misstatement, but in particular on those that have high estimation uncertainty".
The draft standard suggests substantive procedures to respond to estimation uncertainty, in particular requiring the auditor to evaluate:
4 Whether the significant assumptions made by management provide a reasonable basis for the estimate; and
4 Whether and how management have considered alternative assumptions or outcomes, and why they have rejected them.
If management have not dealt with these matters adequately in support of its point estimate, the auditor is required to develop a reasonable range of outcomes, if practicable, and detailed practical guidance on such evaluation is provided. This range is defined as "not having high and low outcome values whose likelihood of occurrence is judged, by the auditor, to be remote, or otherwise less likely to occur than not". The purpose of this language is to define a range which is narrow enough to be useful, without using the terms 'probable' and 'possible'.
The standard makes it clear that the prime responsibility for estimation, including sensitivity analysis, rests with management, and it is only in the absence of adequate work by management that the auditor should seek to develop such a range of out comes.
The draft standard highlights the need to be alert for indicators of possible management bias, and links this consideration in respect of estimates to the review required by proposed standard ISA 320 (revised)2 to determine whether the financial statements as a whole are free from material misstatement.
An interesting matter raised in the draft standard is the position where management decides to change the location of an accounting estimate within a reasonable range of outcomes from one period to the next. It concludes that a misstatement will arise if such a change occurs without a good reason being demonstrated. The amount of the misstatement will be the difference between the estimate decided on by management and the result which would have been the outcome had a consistent basis been applied.
The onus for demonstrating good reason to the auditor rests with management. An example given in the draft standard is that of a change of management, where the new management takes a different commercial view, and therefore evaluate risks differently, than its predecessor. This could constitute 'good reason', as the different view is a bona fide basis for arriving at a different point estimate.
One difficulty which could arise, particularly in the Irish and UK context, would be where there has been a change of auditor. The former auditor would not be expected to make his file available to the new appointee (a requirement in many other jurisdictions), giving rise to a potential problem in carrying out the look back process envisaged in the draft standard.
Conclusion
Auditors should welcome the additional guidance which will be provided by the new ISA in dealing with an area where a high degree of professional judgement must be exercised. ISA 540, together with its companion draft standard ISA 320 require careful attention from all practitioners as they will have a pervasive impact on audits in the future.
Notes
1 The period for comment is up to April 30th, 2005.
2 'Materiality in the Identification and Evaluation of Misstatements' - exposure period ending April 30th, 2005.
Ronan Nolan is Chairman of the ICAI's Auditing Standards Committee, and is an audit partner in the Financial Services group of Deloitte, based in Dublin.