An Overview of Research
on Auditors’ Detection
of Financial Statement Fraud

Mark F. Zimbelman, Brigham Young University

“No problem confronting the (auditing) profession is as demanding, or as difficult to resolve as the problem of management fraud and its detection by auditors.…(T)he profession must accept responsibility for the detection of fraud…(and) develop means of increasing significantly the likelihood of detecting fraud.” [Public Oversight Board, 1993, p. 42]

Introduction
The purpose of this article is to provide an overview of academic accounting research on financial statement auditors’ detection of fraudulent financial reporting (hereafter fraud).¹ Research on auditors’ detection of fraud is extremely motivated for at least three reasons. First, policy makers, academics, government officials, and practitioners have expressed concern over auditors’ performance in detecting fraud for several decades (e.g., see Accounting Series Release No. 19, 1940, p. 4 and Mautz and Sharaf, 1961, p. 113). Importantly, fraud detection continues to be a priority in the profession as indicated by the recent issuance of SAS No. 82 (AICPA 1997) and the fraud research projects funded by the Auditing Standards Board.² Second, significant audit policies have been adopted to influence auditors’ behavior with respect to fraud during the past two decades. Thus, significant opportunity exists for research to inform audit policy with respect to detecting fraud. Third, auditors’ detection of fraud is extremely important to both the profession and society. One indication of the validity of this claim is the Securities and Exchange Commission’s active involvement in policy making to address audit failures involving fraud.³

The remainder of this article is organized as follows: the next section describes the challenging setting that auditors face in attempting to detect fraud; next, a few characteristics of this setting are highlighted and extant and future research issues related to these characteristics are considered; and, finally, concluding comments regarding the future of research on auditors’ detection of fraud are offered.

The Setting
Detecting fraud is an extremely challenging task for the auditor. First, when fraud exists, auditee management has likely undertaken complex efforts to deceive the auditor. Clever teams of highly motivated, knowledgeable managers who often have considerable political capital that can be used to influence their employees and auditors, develop fraud schemes. Furthermore, audits involve extensive complexity spread across time, people, geography, and economic conditions. Typically, an auditor’s first consideration of fraud occurs in the client acquisition/decision (Shelton, Whittington, and Landsittel 1999), which often involves auditors with considerable expertise who assess numerous interrelated risks (e.g., business, engagement, fraud and audit risk). Once the client acceptance decision is made, less experienced auditors strive to conduct the audit efficiently and effectively. Conducting the audit involves (re)assessing inherent risk, control risk, and fraud risk, developing a detailed audit program, gathering evidence of varying reliability, communicating with one another, and reaching an opinion based on a large body of interrelated evidence. Auditors typically perform these tasks using various decision tools. In reality, few auditors will encounter fraud and those that do may not become aware of it until years later, if at all. Additionally, these auditors must cultivate positive client relations (especially with client personnel most likely to perpetrate fraud) and their mixed accountability may lead them to make judgments that are less skeptical than one might expect. Thus, there is little wonder that the typical fraud goes undetected for more than three years. Finally, some management fraud may cure itself in times of economic prosperity, thus inducing greater uncertainty.

Given this setting, numerous questions arise. Some major issues include: How do auditors and auditees respond to and anticipate one another’s behavior (i.e. strategic interaction)? How should and do auditors assess fraud risk and how do these assessments affect audit plans? What audit methods can be used to optimize audit decisions for fraud detection?

Strategic Interaction
A key characteristic of fraud and its detection is that both parties must anticipate their opponent’s response to be successful. Auditees must hide the fraud in a way that the auditor will fail to detect it and auditors must look for fraud where the auditee has hidden it. Failure to consider that management may be acting strategically, e.g., falsifying audit evidence, may result in ineffective levels of skepticism and audit failures. Most experimental research on fraud has not directly examined auditors’ and auditees’ abilities to anticipate and respond to one another’s behavior. A few recent studies (Bloomfield 1997; Zimbelman and Waller 1999) present research rooted in behavioral game theory (Camerer 1990). These studies are motivated by the assumption that fraud introduces behavior that cannot be understood using theories and experiments framed in a decision theoretic context (Kinney 1975). Future experimental research should attempt to discover how boundedly rational auditors’ decisions can be optimized within a game theoretic context. For example, how can skepticism be elevated at a reasonable cost? What proportion of audit procedures should be randomized? Should rotation for whole audit teams or firms occur periodically? How will auditees and auditors respond if all audits require forensic audit procedures as proposed by the Panel on Audit Effectiveness?⁴

Auditor Performance in Fraud Risk Assessment and Audit Decisions
Humans have difficulty achieving accuracy when making complex decisions. As described above, auditors are confronted with numerous stimuli and many seemingly conflicting objectives. Research suggests that this environment may lead to detrimental effects with respect to fraud detection (Braun 2000). Prior research suggests at least two ways that fraud risk assessment and detection may be improved to deal with this complexity. Both involve dealing with the notion of procedural knowledge, which develops over many trials (e.g., Bonner and Walker 1994; Herz and Schultz 1999). Procedural knowledge frees greater cognitive resources to consider knowledge germane to the immediate task. Studies show that requiring auditors to explicitly assess fraud risk results in changes to the extent but not the nature of audit plans (Zimbelman 1997; Glover et. al. 2000) and that a specific instruction that the fraud risk task is primary affects auditors’ decisions in interpreting audit evidence (Knapp and Knapp 2000). More research is needed to discover additional effects of focusing auditors’ attention on the tasks of fraud risk assessment and detection.

A second response to the complexity inherent in fraud- related decision contexts is to bring greater expertise to the decision task. While research in this area is sparse, some findings suggest that auditors with general experience perform better in fraud-related judgments (Knapp and Knapp 2000). More research is needed to determine the effect on fraud detection of experience across varied settings (e.g., audit structure) and under the demands of strategic interaction. As a substitute for experience, auditors may be able to provide effective training to improve fraud detection. For example, unaided audit decision making may be improved by training auditors to avoid the pitfalls of bounded rationality (Johnson, Jamal, and Berryman 1991). Additionally, new audit methodologies might be used to alter auditors’ cognitive representations to improve their fraud detection.

Audit Methodologies and Decision Aids
Recent revisions in audit methodology may affect auditors’ detection of fraud (Bell et al. 1997; Lemon, Tatum, and Turley 2000). These new approaches require that auditors gain a deeper understanding of the economic and regulatory forces affecting their clients’ risks and opportunities. In theory, this knowledge helps auditors understand the economics underlying clients’ transactions, which may be crucial for detecting fraudulent transactions (Erickson, Mayhew, and Felix 2000). More research is needed to determine the effects of these new audit methodologies on fraud detection.

As for decision aids, fraud risk models offer the potential for allocating audit effort to detect fraud in an efficient manner. Empirically derived models generally classify fraud cases more accurately than unaided auditors (Bell and Carcello 2000) but classify nonfraud engagements as slightly more likely to have fraud than unaided auditors (Nieschweitz, Schultz, and Zimbelman 2000). Hansen et al. (1996) assigned different costs to these two misclassifications and derived an “optimal” model suggesting that audit firms might implement a model using cost-benefit criteria. In this area of research, new sources of data are needed as many of the extant studies used the same data set. A few recent studies tested the predictive ability of new sources of fraud cues (Beasley 1996; Summers and Sweeney 1998). Additionally, research should determine how reliance on these models can be accomplished because research outside of auditing suggests auditors may be unlikely to rely on these decision aids for fraud detection (Arkes, Dawes, and Christensen 1986). Research could fruitfully explore ways to increase auditors’ propensity to rely on these models. Extensions are needed for work like that of Eining, Jones, and Loebbecke (1997) who found that auditors achieved better results using expert systems vs. other models. Research is also needed to determine potential litigation consequences that may accompany the use of fraud decision aids and how these consequences may be mitigated.

Concluding Comments
In sum, solving the challenges of detecting fraud will likely remain a priority for the profession for many years. Thus, research will be highly valued by both academics and professionals as long as academics carefully design studies that consider prior research findings and the setting that auditors face to investigate issues that will assist policy makers in their efforts to reduce the costs of fraud Even so, one challenge remains for academics interested in conducting fraud research. Because of fraud’s high profile in litigation against auditors, some audit firms are very reluctant to participate in fraud research. For example, even after committing to support the Auditing Standards Board on SAS No. 82 research projects, the firms were reluctant to participate. Once academics design meaningful studies, cooperation from auditing firms may be the key predictor of meaningful progress in this area.

Footnotes:

1. While the term “fraud” is often defined to include employee theft or defalcation, its use in this article is limited to acts resulting in management fraud or fraudulent financial reporting. A stream of related, but neglected, research exists in the earnings management literature (e.g. Phillips 1999). This vast body of research is beyond the scope of this article (see Healy and Wahlen [1999] for a review). In excluding earnings management research, I acknowledge the difficulty in distinguishing these topics and encourage efforts to meld these bodies of research.
2. See www.aicpa.org/news/p032299a.htm for the AICPA’s press release on these projects.
3. For a more comprehensive review of empirical research on auditors’ detection of fraud see Nieschwietz, Schultz, and Zimbelman (2000).
4. See www.pobauditpanel.org.

References

Accounting Series Release No. 19. 1940. In the matter of McKesson & Robbins, Inc., Securities and Exchange Commission.

American Institute of Certified Public Accountants (AICPA): Auditing Standards Board. 1997. Consideration of Fraud in a Financial Statement Audit. Statement on Auditing Standards No. 82. New York, NY: AICPA.

Arkes, H. R, R. M. Dawes, and C. Christensen. 1986. Factors influencing the use of a decision rule in a probabilistic task. Organizational Behavior and Human Decision Processes 37: 93–110.

Beasley, M. S. 1996. An empirical analysis of the relation between the board of director composition and financial statement fraud. The Accounting Review 71: 443–465.

Bell, T. B., and J. V. Carcello. 2000. A decision aid for accessing the likelihood of fraudulent financial reporting. Auditing: A Journal of Practice & Theory, (Spring): 169–184.

Bell, T., F. Marrs, I. Solomon, and H. Thomas. 1997. Auditing Organizations through a Strategic-Systems Lens: The KPMG Business Measurement Process. New Jersey: KPMG Peat Marwick LLP.

Bloomfield, R. J. 1997. Strategic dependence and the assessment of fraud risk: A laboratory study. The Accounting Review 72: 517–538.

Bonner, S., and P. Walker. 1994. The effects of instruction and experience on the acquisition of auditing knowledge. The Accounting Review 69: 157–178.

Braun, R. L. 2000. The effect of time pressure on auditor attention to qualitative aspects of misstatements indicative of potential fraudulent financial reporting. Accounting, Organizations and Society 25 (3): 243–259.

Camerer, C. 1990. Behavioral game theory. In Insights in Decision Making: A Tribute to Hillel J. Einhorn, edited by R. M. Hogarth. Chicago, IL: University of Chicago Press.

Eining, M. M., D. R. Jones, and J. K. Loebbecke. 1997. Reliance on decision aids: An examination of auditors’ assessment of management fraud. Auditing: A Journal of Practice & Theory 16 (Fall): 1–19.

Erickson, M., B. W. Mayhew, and W. L. Felix. 2000. Why do audits fail? Evidence from Lincoln Savings and Loan. Journal of Accounting Research (Spring): 165–194.

Glover, S. M., D. F. Prawitt, J. J. Schultz, Jr., and M. F. Zimbelman. 2000. A comparison of audit planning decisions in response to increased fraud risk: Before and after SAS No. 82. Working paper, Brigham Young University.

Hansen, J. V., J. B. McDonald, W. F. Messier, Jr., and T. B. Bell. 1996. A generalized qualitative–response model and the analysis of management fraud. Management Science 42: 1022–1032.

Healy, P., and J. Wahlen. 1999. A review of the earnings management literature and its implications for standard setting. Accounting Horizons (December): 365–383.

Herz, P. J., and J. J., Schultz, Jr. 1999. The role of procedural knowledge in accounting judgment. Behavioral Research in Accounting: 1–26.

Johnson, P. E., K. Jamal, and R. G. Berryman. 1991. Effects of framing on auditor decisions. Organizational Behavior and Human Decision Processes 50: 75–105.

Kinney, W. 1975. A decision theory approach to the sampling problem in auditing. Journal of Accounting Research (Spring): 117–132.

Knapp, C. A., and M. C. Knapp. 2000. The effects of experience and explicit fraud risk assessment in detecting fraud with analytical procedures. Accounting, Organizations and Society (forthcoming).

Lemon, W. M., K. W. Tatum, and W. S. Turley. 2000. Developments in the audit methodologies of large accounting firms. Mautz, R., and H. Sharaf. 1961. The Philosophy of Auditing. American Accounting Association, Monograph No. 6.

Nieschwietz, R. J., J. J. Schultz, Jr., and M. F. Zimbelman. 2000. Empirical research on external auditors’ detection of financial statement fraud. Journal of Accounting Literature (forthcoming).

Phillips, F. 1999. Auditor attention to and judgments of aggressive financial reporting. Journal of Accounting Research (Spring): 167–189.

Public Oversight Board. 1993. In the Public Interest: A Special Report by the Public Oversight Board of the SEC Practice Section, AICPA. Stamford, CT: Public Oversight Board.

Shelton, S., R. Whittington, and D. Landsittel. 1999. Fraud risk assessment: An analysis of practices of auditing firms. Working paper, DePaul University.

Summers, S. L., and J. T. Sweeney. 1998. Fraudulent misstated financial statements and insider trading: An empirical analysis. The Accounting Review (January): 131–146.

Zimbelman, M. F. 1997. The effects of SAS No. 82 on auditors’ attention to fraud risk factors and audit planning decisions. Journal of Accounting Research (Supplement): 75–97.

______, and W. S. Waller. 1999. An experimental investigation of auditor–auditee interaction under ambiguity. Journal of Accounting Research (Supplement): 135–155.

Back to Contents Page