What could we mean by ‘audit quality’ and why is it difficult to measure?

Download 40.54 Kb.
Size40.54 Kb.
Unfortunately for the researcher, it is not possible to directly observe audit quality unless litigation related to auditor negligence test the audit work in the courts. Discuss this statement, with reference to attempts at directly testing audit quality.

  1. What could we mean by ‘audit quality’ and why is it difficult to measure?

  2. Discuss the advantages and disadvantages of bankruptcy studies and event studies, as measures of audit quality in term of the information content of audit opinions.

  3. How useful is litigation in ensuring audit quality?

Auditing is considered as the ‘quality control’ of financial accounts. Much importance is often placed on the audit opinions given as all sorts of users of financial statements are seen to rely upon these to make financial decisions. Despite many differing opinions on the purpose of an audit, the objective is laid out in APB Statements of Auditing Standards (SAS) 100;

‘The objective of an audit of financial statements is to enable auditors to give an opinion on those financial statements taken as a whole and thereby to provide reasonable assurance that the financial statements give a true and fair view’
Auditors do not guarantee that the statements are correct and as with any procedure is unlikely to be 100% accurate, there are many risks both inherent and otherwise involved in auditing accounts.
For many years research has surrounded the level of audit quality offered by auditors to both companies and those relying on the accounts and audit opinions such as shareholders. However, it is not always easy to test audit quality.
Firstly, it is often difficult to decide what exactly is meant by ‘audit quality’. Arguably a good quality audit is one which reaches it’s objective. However when we look at the given objective of an audit it is clear that auditors are expected to give an opinion and a reasonable assurance that the accounts give a true and fair view of the accounts.
The first issue comes with the word opinion. An auditor must form his opinion by gathering evidence, this evidence can be hard evidence that is observable and verifiable or soft evidence which is observable but not verifiable. Benito Arrunada, 2000 argued that in order to produce an audit of maximum value (which he considers to be the same measure as quality), an auditor must draw on the soft information available and be free to use his professional judgement.
Mautz and Sharaf, (1961) recognised nine types of evidence that an auditor is likely to draw on, however SAS 400 recognises only five:

  1. Inspection

  2. Observation

  3. Inquiry

  4. Computation

  5. Analytical procedures

It is clear to see that in gathering evidence from these five sources there is likely to be a wide range of knowledge gathered both hard and soft that an auditor could use to produce a quality report.

Mautz and Sharaf, encouraged the concept of the prudent auditor as a way of measuring the ‘due audit care’ that they should display when gathering evidence and reporting on the evidence gathered.
Arrunada stated that he believed audit quality could be clearly split into two factors:

  1. The auditor’s ability to complete the audit (technical competence) and

  2. The auditor’s ability to offer an objective opinion (independence).

Arrunada also argued that quality can be enforced in one of two ways, by explicit or implicit terms. Explicit contract terms are clearly defined and breach is likely to result in litigation. Implicit contract terms are less obvious and are similar to unspoken expectations, in this instance breach is more likely to result in loss of confidence, although harder to regulate Arrunada feels that implicit terms result in a more onerous requirement on auditors and is more natural and automatic in scope.

Furthermore, Arrunada recognizes that as a group of professionals, auditors should be expected to have a minimum level of enforceable quality. This is achieved through barriers to entry. In order to practice as an auditor a certain level of qualification is required. Although it is recognized that this increases the chance of monopolistic behavior, it is generally agreed that this minimum level is required for the more indirect mechanism to be truly effective.
L.E. De Angelo, (1981) also questioned the true meaning of ‘audit quality’ and how researchers can attempt to measure such an entity. She suggested that auditor quality is the probability that an auditor will both;

  1. Find any breach in the statements and

  2. Report that breach if found.

There is a direct parallel with the opinions of Arrunada in that quality depends on technical competence (point (a) above) and independence (point b).

However De Angelo recognizes that as procedures are not directly observable it is very costly to get a true measure of quality. It is also true to state that the consumers of audits (e.g. shareholders, general public etc.) rarely know the true motivations of the company auditors.
De Angelo therefor proposes that an easier to observe variable could be used instead of quality. She suggests that the size of an auditor is directly related to the quality of an auditor. She suggests this for several reasons:

  • Larger firms have less cheating incentive as an auditor has a larger incentive to cheat as the level of the ‘quasi-rent’ of the client. (Quasi-rent being the expected stream of incomes that are likely to come from that client). This suggests that the more an auditing company is relying on future business from a given client the more likely they are to produce an inaccurate lower-quality audit in order to please and therefor retain their client.

  • Generally the larger the firm the less important every individual client is, reducing the incentive to cheat to retain clients.

  • Also a larger firm tends to have more partners so that each individual partner has less personally to gain from cheating and thus further reducing the incentive.

Clive Lennox, (1999) also produces evidence that auditor size is directly related to audit quality. He notes that the stock market generally reacts favorably to a company that switches to use a larger audit firm thus implying a greater degree of confidence in the larger firms.

Lennox, suggests that there are two main reasons for this;

  1. The depth of the pocket of the auditors. Larger firms are generally more wealthy. Consumers of the audits often see auditors as an insurance policy that they can get financial compensation from. The larger the firm the more likely they are to get a larger compensation if there should be a detrimental error later discovered.

  2. Larger firms are generally more concerned with their reputation as when an error is discovered this can have an adverse effect on the fees they can charge the remaining clients to compensate for the potential loss of value in the opinions that they are giving.

De Angelo, (1981) believes that the loss of reputation is the key factor in linking auditor size with quality.

Ronald Dye, (1993) also did research on the auditor wealth in terms of auditor quality. He notes that the larger firms are generally more active in the standard setting procedure and are therefor often better versed in what is required to produce a ‘quality’ audit.
Dye, does note one major problem in using auditor size as a measure of quality. The question of quality is rarely fully investigated and does not become public knowledge unless the client is facing financial difficulty and is suing for financial compensation. Larger firms are known for having ‘deep pockets’ and are therefor more prone to litigation. He suggests that we cannot therefor use litigation levels as a way of measuring quality.
Dye recognises that auditor wealth is likely to be a direct link with the quality that they produce. He suggests that the auditor wealth is in effect his bond that ensures performance of an audit of acceptable quality.
Stephen Hillegeist (1999), placed a different emphasis on measuring audit quality. He suggests that the important issue that is of concern to audit users is the level of audit failure not the level of audit quality per-se.

Event studies are often used by researchers as a means of measuring audit quality. An event study is simply a study that looks at the effect of any given event on the stock market share price of a given company. In terms of audit quality researchers often look at the way the stock market reacts to the announcement of a modified audit report.

The principle that governs such research is that if a modified audit creates a change in the share price this implies that an audit does provide valuable information for investors. However this is not entirely reliable, audit reports are normally announced at the same time as the financial statements. It is therefor very difficult to determine which event has caused the share price change.
The problem of using event studies as a means of measuring audit quality is evident in the disparity of results that researchers have found. At one end of the spectrum, Firth 1978 along with Banks and Kinney 1982, both found that share prices were seen to fall after the a qualified audit report. This implies that investors do see the audit as useful in terms of information offered. Dopuch et al,.1986 also found that share prices fell after a qualified report, however he studied cases where the report had been announced through the news which is likely to provide a more obvious reaction due to the nature of the disclosure. (People tend to pay more attention to announcements heard on the news). However, other researchers, such as Elliott, 1982 have found that qualified audit reports have little or no effect on share prices.
A different perspective was taken by Choi and Jeter 1992, who found that following a qualified audit report investors tended to be less receptive to future earning announcements. This implies that greater uncertainty is felt in the message held by greater dividends.
There are several problems inherent in using event studies as a way of measuring audit quality:

  • Identifying the exact effect of the audit due to other concurrent events

  • The effect of a qualified report can only be used of a relatively short period of time, and

  • There are relatively few companies that receive a qualified audit effecting the possible sample that can be examined.

However, as previously discussed, there is no direct way of measuring audit quality therefor researchers are required to find an alternative variable. Returns on shares and abnormal returns that could be associated with an audit report are relatively easy and quantifiable to measure (once the event date has been established). This makes using event studies a viable option as an alternative to directly observing g audit quality.

As an attempt to look at audit content and the value of the information content other researchers have looked at bankruptcy studies. This is a way of determining whether audits successfully identify failing companies.
Evidence suggests that audits do not identify failing companies;

  • Auditors disclosed a going-concern qualification on 20% of failing companies’ in the UK. (Carcello et. Al 1995)

  • In private companies this figure was only 5%.(Barnes and Hooi 1987)

  • Out of 40 companies to receive modified reports only 10 then went on to fail. (Taffler and Tseung 1984)

Much research has been done to attempt to determine the link if any between audit reports and the level of bankruptcy. Hopwood et al. (1989) found that audit opinions did have a large effect on bankruptcy levels however many other variables were omitted such as company size and the industry sector.

Other researchers such as Lennox 1(999) included these variables and found that as result there was no significant link between audit opinions and the level of bankruptcy. He felt that this was due to two main factors:

  • Audit reporting does not reflect a lot of information that is available to the public through other means. This information includes industry specific crises such as foot and mouth in the farming industry and general economic conditions such as recession.

  • There is also the important factor of the previous audit opinion. If the previous audit report was qualified then the strength of the current audit report being qualified is likely to be considerably smaller than in the case where a company is getting a qualified opinion for the first time.

A major disadvantage of using bankruptcy is that it is often a highly unpredictable event and therefor an auditor not detection this is not necessarily a measure of poor quality.

Bankruptcy as a means of measuring audit quality suffers the same limitation as event studies in that there is a limited number of companies that can be observed.
Bankruptcy studies also suffer a large disadvantage with what has been named the self-fulfilling prophecy hypothesis. It was suggested by Matsumara et al. (1997) that by declaring that a company is suffering from going concern issues this may in itself cause a company to fail as creditors may refuse credit etc.
Despite this suggestion Citron and Taffler (1992) have found no empirical evidence to support the hypothesis. They found that 17% of modified companies went bankrupt before the next set of accounts in comparison to 8% of non-modified companies going bankrupt this figure was not significant at the 5% level.
These problems suggest that no direct correlation can be seen between audit reports and the level of bankruptcy, mainly due to the limits in the possible samples, the self-fulfilling hypothesis and the unpredictability of bankruptcy.
Many other researchers have suggested that inference can be drawn on the quality of audits by the level of litigation. However, this may not be accurate due to the fact that auditors are required to hold professional indemnity insurance and are therefor seen as having deep pockets. This makes the auditor a favored target in the case of shareholders losing in the event of a failing company.
Zoe Palmorse (1988), suggested that low (high) litigation levels could be seen as saying that audits are of high (low) quality. However, she does recognize that this cannot be treated as a direct correlation due to the fact that auditors deep pockets mean that people will often litigate auditors even if the quality was high simply as a means of recouping losses that they would not be able to recoup elsewhere.
It should also be noted that Lennox (1999) showed evidence to support the deep pockets hypothesis by showing that larger auditors were more prone to litigation despite providing higher quality. He also showed that auditors who faced litigation did not generally suffer obvious created client loss. This research implies that litigation does not provide useful information on audit quality.
Currently, in the UK, auditors share ‘joint and several’ liability with the directors of the company on the contents of the financial statement. This is often regarded as unjust as auditors potentially have to pay for the entire damages should the directors be unable to pay themselves. A likely scenario if the company has gone bankrupt. An added problem to the auditor is the fact that most audit firms are partnerships which therefor means that the partners’ personal funds could be at risk.
There have also been many cases arguing whether the auditor was in fact negligent. In order for an auditor to be found negligent it must be established that he owed a ‘duty of care’ to the claimant. A contract (namely between the company and the auditor) would provide this duty. More controversial however, is the fact that 3rd parties can sue under the tort of negligence if they can prove that they were entitled to a duty of care, that this duty was breached and that financial losses have resulted. English law currently recognizes that an auditor has a legal duty of care to 3rd parties when 3 conditions are met:

  1. The auditor must be aware of the nature of the transaction that the 3rd party is considering,

  2. He must know that the report is going to be made available to the 3rd party

  3. And that the 3rd party is likely to rely on his report.

It was held in Caparo Industries v Dickman (1990) that an auditor did not owe a duty of care to shareholders using the auditors’ report as a means of making financial investment choices, whether they were potential or current shareholders. This position has been upheld in the courts and it is currently very difficult for anyone other than the current shareholders to successfully recover any losses.

Increased levels of litigation during the 1980’s has lead firms to look at ways of minimizing the risk of litigation. These include incorporating a company in order to make use of limited liability. This however has not proved popular with only KPMG out of the big five making use of this option.
There are propositions by the Department of Trade and Industry that suggest a shift from joint and several liability to proportional liability. It has been suggested by the International Federation of Accountants that this may increase auditor quality. This position is justified by considering auditor risk as two separate issues. There are controllable risks (those that the auditor can effect by increasing effort and thus quality) and uncontrollable risks (those that the auditor cannot affect). It is argued that simply limiting liability limits both types of risk. However with proportional liability greater importance is placed on the controllable risk therefor encouraging auditors to put in a greater level of effort.
It should be noted however that according to research by Balachandran and Nagarajan (1991), audit quality could be linked directly to the level of damages imposed. This suggests that courts are already shifting towards the idea of proportional liability despite no actual reforms being approved yet.
In terms of measuring audit quality, I suggest that the current litigation system does not provide any form of accurate measure mainly due to the ‘deep pocket hypothesis’. I do however feel that a shift towards proportional liability will see a greater importance being placed on controllable risk and will therefor encourage greater auditor quality. It will also mean that 3rd parties will not simply be able to sue auditors as they are the last option but will sue auditors when the losses are DIRECTLY related to the audit quality.
It is fair to conclude that directly observing auditor quality is impossible and therefor many attempts have been made by researchers to provide a suitable variable. Some variables such as size of audit firm appear to be relatively successful whereas others such as levels of litigation provide little useful audit quality evidence.
Some of the above mentioned variables are proving to be relatively consistent with auditor quality, in that they give users of audits a reasonable reflection of the quality of the audit they are receiving. Despite this, the only true and accurate reflection of an individual audit quality is when it is assessed by the courts in a litigation case.

Download 40.54 Kb.

Share with your friends:

The database is protected by copyright ©sckool.org 2022
send message

    Main page