Auditors’ Responsibilities with Respect to Fraud: A Possible Shift?

By Nicholas Apostolou and D. Larry Crumbley

E-mail Story Print Story

Fraud Is Pervasive

Corporate fraud continues to be a pervasive problem. The Association of Certified Fraud Examiners’ (ACFE) 2006 “Report to the Nation on Occupational Fraud and Abuse” estimated that a typical organization loses 5% percent of its annual revenues to fraud, or about $4,500 per employee each year. Pricewaterhouse-Coopers’ 2005 “Global Economic Survey” disclosed a dramatic increase in the number of companies reporting fraud as compared with its 2001 and 2003 results.

The restatement of financial statements almost doubled from 2004 (616) to 2005 (1,195), representing almost 8.5% of U.S. publicly traded companies. In 2006, restatements totaled 1,420, representing one out of every 10 public companies. The 2006 total is 12 times higher than in 1997, according to statistics compiled by Glass, Lewis & Co. (2003–2006) and Huron Consulting (1997–2002). The top three areas responsible for restatements were equity, revenue recognition, and misclassification.

Although the number of shareholder class-action lawsuits has trended downward in the past several years, the settlements have become much larger. At the peak of Fannie Mae’s restatement efforts, it was spending $50 million a month on outside accounting services. It had 2,000 contractors working on restatement-related matters, along with 400 employees from its controllers’ organization (Roy Harris, “Say Again,” CFO, April 2007).

Most frauds involve a lack of adequate internal controls (opportunity), the need to maintain an expensive lifestyle or pressure to meet goals (incentive), and the perpetrators’ lack of awareness that their actions are wrong (self-rationalization) or simple lack of integrity (see Grace Duffield and Peter Grabosky, “The Psychology of Fraud,” Australian Institute of Criminology, no. 199, March 2001). Some fraudsters, however, wish to make fools of their victims. They take delight in the fraud act itself. Three Duke Energy employees illustrate how this incentive factor can work. These employees were charged in April 2004 for allegedly creating phony electricity and material-gas trades to boost trading volumes, thus inflating profits in a trading book that was the basis of their annual bonuses. This trading scheme inflated their bonuses by at least $7 million between March 2001 and May 2002 (Rebecca Smith, “Former Employees of Duke Charged Over Wash Trades,” Wall Street Journal, April 22, 2004, p. A-15).

The increased awareness of both fraud and the importance of transparent financial reporting has spurred the concern of regulatory bodies, as well as the accounting profession. Below is a discussion of two recent reports regarding the audit function and, more specifically, the responsibility of the auditor with respect to fraud.

The PCAOB Report

On January 22, 2007, the PCAOB issued a report titled Observations on Auditors’ Implementation of PCAOB Standards Relating to Auditors’ Responsibilities with Respect to Fraud (PCAOB Release No. 2007-001). The auditor’s responsibility is principally defined in AU section 316 (Consideration of Fraud in a Financial Statement Audit). In Release 2007-001, the PCAOB emphasized that it was not changing or proposing to change any existing standard. Based upon observations made during its inspection of audit work, the PCAOB believes that a further explication of the standards that relate to fraud would be constructive. The report can be divided into five sections:

• Auditor’s overall approach to the detection of financial fraud;
• Required brainstorming sessions and fraud-related inquiries;
• Auditor’s response to fraud risk factors;
• Financial statement misstatements; and
• Fraud associated with management override of controls.

Auditor’s Overall Approach to the Detection of Financial Fraud

AU section 316.52 discusses changing the nature, timing, and extent of auditing procedures needed to address identified risks of material misstatement due to fraud. These changes are described by the PCAOB as follows:

• The nature of auditing procedures may require obtaining evidence that is more reliable or verifiable. For example, computer-assisted audit techniques might provide corroborative evidence about significant accounts.
• The timing of substantive tests might need to be adjusted. For example, substantive testing might be of particular value in assessing the risk of material misstatement due to fraud on or near the reporting date.
• The extent of the procedures employed should reflect the assessment of the risk of material misstatement due to fraud. For example, increasing the sample size might be appropriate.

The root of the PCAOB’s concern, as expressed in Release 2007-001, is that its inspection teams have observed that auditors often obtain reasonable assurance of risk mitigation by checking off items on standard audit programs and checklists. One author of this article has heard of an external auditor giving the walkthrough checklist to an employee of the company and asking the employee to fill it out. Release 2007-001 states that PCAOB standards require additional documentation to confirm the performance of required audit procedures. Furthermore, the lack of sufficient documentation makes it difficult for senior members of audit engagement teams to properly review the procedures performed by other members. Finally, the PCAOB inspection teams expressed concern about audit procedures being applied mechanically, rather than the audit plan being modified in accordance with risk detection.

Brainstorming Sessions and Fraud-Related Inquiries

AU section 316.14–17 discusses how an effective audit team should conduct an audit. AU section 316.14 discusses auditors’ planning, including a consideration by the audit team of the potential for material misstatement due to fraud. The discussion should include an exchange of ideas or “brainstorming” among team members about the susceptibility of the financial statements to material misstatement due to fraud, and the role of management in perpetrating and concealing fraudulent financial reporting. This discussion should alert the audit team to how fraud might be perpetrated and concealed based upon both the general and the client-specific knowledge of members of the team.

AU section 316.15-–16 emphasizes how important it is that audit team members have a questioning mind. The auditors should be aware of the management’s incentives to commit fraud as well as the opportunity for fraud to be perpetrated. This evaluation requires an awareness of the culture and environment of the firm that might enable management to
rationalize committing fraud. Furthermore, audit team members should be diligent in obtaining appropriate evidence to support their judgments.

AU section 316.17 discusses the involvement of audit team members in an engagement. The PCAOB makes clear that key members of the audit team must be involved in discussions and brainstorming sessions that evaluate the potential for material misstatement due to fraud. AU section 316.17 also specifies some factors that will influence the extent of the audit team’s discussions and how they should occur. For audits involving more than one location, the PCOAB suggests multiple discussions with team members in differing locations. Specialists assigned to the team should be included in brainstorming sessions that continue throughout the audit.

The planning steps outlined in paragraphs AU section 316.15–17 are critical to an effective audit. PCAOB inspection teams have, however, noted instances of failures to comply with the standards. For example, PCAOB inspectors have “(1) identified audits in which the audit team was unable to demonstrate that brainstorming sessions were held; (2) identified audits in which the audit teams’ brainstorming sessions occurred after planning and after substantive fieldwork had begun; and (3) identified audits in which key members of the audit team did not attend the brainstorming sessions” (PCAOB Release 2007-001, p. 5).

Auditor’s Response to Fraud Risk Factors

AU section 316.48 describes an auditor’s response to risks of material misstatement due to fraud in the following ways:

• A response that has an overall effect on how the audit is conducted—that is, a response involving more general considerations apart from the specific procedures otherwise planned.
• A response to identified risks involving the nature, timing, and extent of the auditing procedures to be performed.
• A response involving the performance of certain procedures to further address the risk of material misstatement due to fraud involving management override of controls, given the unpredictable ways in which such an override could occur.

The overall response to risk assessment involves procedures that are not predictable, such as modifying the assignment of personnel, the degree of supervision, and selecting auditing procedures. The auditor can respond to specifically identifiable risks of material misstatement such as significant related-party transactions (e.g., shifting debt to special purpose entities, as done by Enron) by changing the nature, timing, and extent of auditing procedures.

The PCAOB observed instances of auditors “failing to respond appropriately to identified fraud risk factors” (Release 001, p. 6). The PCAOB also noticed instances in which there was no evidence that auditors had considered any associated fraud risk factors for transactions that the PCAOB considered questionable.

Financial Statement Misstatements

When the auditor’s procedures detect misstatements in the financial statements, the auditor should document the nature and effect of the misstatements (AU section 316.52) and evaluate whether the misstatements might be indicative of fraud (AU section 316.75). PCAOB inspectors noted the improper calculation of the threshold for posting proposed audit adjustments to a summary schedule (Release 2007-001, p. 7). Because the summary schedule was incomplete, the PCAOB inspectors noted that certain uncorrected misstatements were not properly evaluated. Consequently, the auditors failed to determine whether these departures from GAAP were indicative of fraud.

Fraud Associated with Management Override of Controls

AU section 316.08 recognizes that management has a unique ability to perpetrate fraud as a result of being in a position to directly or indirectly manipulate accounting records and present fraudulent financial information. It notes: “Fraudulent financial reporting often involves management override of controls that otherwise may appear to be operating effectively.” To address the risk of management override of controls, AU section 316 requires auditors to perform certain procedures, such as checking journal entries and other adjustments and reviewing accounting estimates for possible biases that could result in material misstatement due to fraud. PCAOB inspection teams noted cases in which it did not appear that the auditor had properly addressed the risk of management override of controls with respect to “top drawer” journal entries and accounting estimates.

For example, Rite Aid’s management directed its accounting staff to make improper adjusting entries to reduce cost of goods sold and accounts payable in every quarter from the first quarter of fiscal year 1997 through the first quarter of fiscal year 2000. These entries had no substantiation, and were intended purely to manipulate Rite Aid’s reported earnings. As a result of these entries alone, Rite Aid overstated pretax income by $100 million in the second quarter of fiscal year 1999 (www.sec.gov/news/press/2002-92.htm).

In testing specific entries and other adjustments, and the support for them, auditors should consider several important issues:

• Fraud risk factors that might help identify specific classes of journal entries for testing, such as entries made by unauthorized personnel or personnel who do not ordinarily wake journal entries; or entries that lack detailed explanations or other supporting documentation;
• The characteristics of fraudulent entries, including entries made at unusual times, such as nights, weekends, or holidays, and entries made to intercompany or suspense accounts; and
• Nonstandard journal entries that might not be subjected to the same level of internal control as recurring journal entries—for instance, entries at the close of quarterly and annual reporting periods and those that are part of the post-closing process. (Release 2007-001, p. 8).

Most of the original entries in the WorldCom fraud were initially correct, but later topside entries moved the expenses into asset accounts. In the HealthSouth fraud, where there were at least 2,000 ledgers, most of the fraudulent entries occurred in the intercompany entries. The major entry was a debit to the suspense account with a credit to revenues. The suspense account was later closed, and asset accounts were established with a credit from the suspense account. The external auditors did not catch on to this financial engineering.

After a fraud, the forensic work can be expensive. During 2005, 2004, and 2003, professional fees associated with the reconstruction of HealthSouth’s financial records and restatement of 2001 and 2002 approximated $206.2 million and $70.6 million, respectively.

Fraudulent financial reporting often is accomplished by intentionally biasing assumptions and judgments used to estimate account balances (AU section 316.63). For example, if the effect of each individual estimate, although reasonable, were to increase income, the auditor should further evaluate the estimates taken as a whole. Furthermore, management can manipulate income by distorting accounting estimates, such as not recognizing losses due to the asset impairment, or overestimating estimates in one period so that the estimates can be reversed in later periods to manage earnings (e.g., cookie-jar accounting). Examples of such estimates include the allowance for bad debts, pension estimates, and restructuring reserves.

PCAOB inspections revealed that “some auditors have failed to test, or failed to document their testing of, management’s assumptions and other aspects of issuers’ accounting estimates. The inspection teams also noted that some auditors failed to assess, or failed to include in their audit documentation evidence that they had assessed whether the overstatement or understatement of accounting estimates indicated a bias in management’s estimates that could result in material misstatements due to fraud” (Release 2007-001, p. 10).

Global Capital Markets and the Global Economy

Published in November 2006, Global Capital Markets and the Global Economy: A Vision From the CEOs of the International Audit Networks represents the views of the CEOs of the six leading international audit firms: Pricewaterhouse-Coopers, Grant Thornton, Deloitte, KPMG, BDO, and Ernst & Young. The Global Capital Markets report represents a comprehensive discussion of how public company auditing procedures must adapt to better serve capital markets around the world. The report identifies six vital elements:

• Investor needs for information are well defined and met;
• The roles of the various stakeholders in those markets (preparers, regulators, investors, standards setters, and auditors) are aligned and supported by effective forums for continuous dialogue;
• The auditing profession is vibrant, sustainable, and provides sufficient choice for all stakeholders in these markets;
• A new business-reporting model is developed to deliver relevant and reliable information in a timely way;
• Large, collusive frauds are more and more rare; and
• Information is reported and audited pursuant to globally consistent standards.

The report identifies no single issue “as the subject of more confusion, yet is more important, than the nature of the obligation of auditors to detect fraud—or intentional material misstatement of financial information by public companies.” The report emphasizes how essential it is that all parties engaged in business reporting adopt appropriate procedures and policies to prevent and detect fraud. However, it claims that there is an “expectations gap” between what various stakeholders believe should be done to detect fraud and what auditors are actually doing because of the fees that companies are willing to pay for audits. The expectations gap occurs because many investors and policy makers expect auditors to detect all fraud, and if they do not, the auditors are “presumed to be at fault.” The report suggests “a constructive dialogue among investors, other company stakeholders, policy makers and our own professionals about what should be done to close or at least narrow the ‘expectations gap’ relating to fraud.”

As the report states, SAS 99 (the source of AU 316) and its international counterpart (IAS 240) provide similar directions to auditors with respect to fraud. Both require auditors to approach their task with a healthy degree of skepticism. Furthermore, both present specific requirements for auditors to follow, which can be summarized as follows:

• Considering a company’s internal controls and procedures, and how these are actually implemented, when planning the audit;
• Designing and conducting audit procedures to respond to the risk that management could override internal controls and procedures;
• Identifying specific risks where fraud may occur;
• Considering whether any misstatement uncovered during the audit may be indicative of fraud;
• Obtaining fraud-related written representations from management; and
• Communicating with appropriate managers and the board if the auditor finds an indication that fraud may have occurred.

For example, Symbol Technologies, Inc., engaged in a number of channel-stuffing techniques. The company arranged transactions to make it appear that Symbol was selling products, but it would simultaneously eliminate the resellers’ obligation to pay for the products. These resellers typically did not need and often could not afford to pay for the products they ordered, but Symbol negated any risk to the resellers by granting them contingent payment terms and unconditional return rights. Furthermore, the resellers did not have to pay Symbol unless and until they resold the product and received payment from an end user. The resellers also had the right to return any unsold product to Symbol at no cost. These special terms did not appear anywhere in the purchase orders or resulting invoices, which simply recited Symbol’s standard “net 45 day” payment terms. Side agreements also superseded the stock rotation terms that Symbol normally granted to channel partners in its standard contracts, which did not permit unlimited returns and in many circumstances charged a restocking fee.

The reports emphasize that even if auditors follow all these guidelines, a sound audit may not uncover fraud, especially if that fraud is perpetrated by senior management. Both financial statement fraud and misappropriation of assets involve intent and deception. Financial statement fraud is difficult to catch because it is often perpetrated by highly motivated, clever teams of knowledgeable managers with the capacity to persuade or intimidate both their own employees and their auditors. Financial statement audits involve extensive complexity spread across time, people, geography, and economic settings (see R.J. Nieschewietz, “Empirical Research on External Auditors’ Detection of Financial Statement Fraud,” Journal of Accounting Literature, vol. 19, 2000, p. 236). At least 40 individuals knew of or were involved in the massive WorldCom fraud.

Given the challenge of detecting fraud, the Global Capital Markets report offers additional ideas for enhancing fraud detection. Without embracing any of the ideas, the authors of the report believe these ideas have sufficient merit and should be debated by stakeholders:

• Subject all public companies to a forensic audit on a regular basis. A forensic audit is the most aggressive and costly way to detect fraud. A forensic audit is a much more detailed audit involving the evaluation of all company records and the questioning of all company employees. Public companies would be required to undergo a forensic audit on a regular basis (e.g., every three to five years).
• Subject all public companies to a forensic audit on a random basis. A less costly version of the preceding proposal would be to subject a sample of public companies to a forensic audit on a random basis. Although this recommendation might uncover fewer frauds, the possibility of being scrutinized in such detail could have a significant deterrent effect.
• Other choice-based options. Another possibility would be to enlist more participants in selecting the intensity of the audit for fraud. As an example, the report suggests the possibility of investors deciding on the type of fraud detection effort they want the auditors to perform. The shareholders could base this decision on information presented to them regarding the costs of the different types of audits as well as the historical experience of the company with respect to fraud. Alternatively, the board of directors or the audit committee could decide on the intensity level of fraud detection.

A Forensic Audit

A forensic audit may be significantly more expensive and time-consuming than regular audit work. Whereas a financial audit is a sampling activity that does not look at every transaction, a forensic audit focuses on a specific aspect of the books and may examine each item. According to Jake Poinier (“Fraud Finder,” Future Magazine, Fall 2004), “[W]hile the average accountant is trying to make everything add up, a forensic accountant is performing a detailed financial analysis to find out why everything doesn’t or shouldn’t add up.”

Lorraine Horton of the forensic accounting firm L. Horton & Associates in Kingston, R.I., says that in investigative accounting you are “looking for one transaction that will be the key. The one transaction that is a little different, no matter how small the difference, and that will open the door” (H.W. Wolosky, “Forensic Accounting to the Forefront,” Practical Accountant, February 2004.)

Robert J. DiPasquale, a partner with the Business Investigations Group in the Parsippany, N.J., office of J.H. Cohn, says that forensic accounting “is a very competitive field. What is interesting is that you may be a good accountant, but not a good forensic accountant. The training and the way you look at transactions are different” (Wolosky, 2004). Horton suggests that “forensic accounting is very different from auditing in that there is no template to use. There are no set rules. You don’t know when you go into a job how it is going to be” (Wolosky, 2004).

When an audit turns into a forensic investigation, auditors must comply with the litigation services standards. Although auditors should use professional skepticism, a forensic accountant is often trying to establish scienter; that is, trying to prove “intent” on part of the fraudster. Horton cautions that, unlike auditing, lower-level staff often cannot be used for a forensic engagement. “They normally will not spot anything out of the ordinary, and an experienced person should be the one testifying as well as doing the investigative work.”

Fighting Fraud

Fraud continues to be a prominent issue commanding the attention of regulators as well as of the accounting profession. The auditor’s concern is that the financial statements of a company be stated fairly in all material respects. Because auditors cannot evaluate every transaction of a company, they have to make judgments and decisions dictated by a risk assessment and cost-benefit analysis. Both regulators and stakeholders are strengthening the role of auditors in the deterrence and detection of fraud. The discussion above represents regulators’ current view of auditors’ performance in detecting fraud as well as a discussion by our most prominent auditors of their role in fraud deterrence and detection.

©2009 The New York State Society of CPAs. Legal Notices