An Historical Perspective on Fraud Detection: From Bankruptcy Models to Most Effective Indicators of Fraud in Recent Incidents

Why are auditors unable to detect fraudulent financial reporting in some audits? Should auditors be able to anticipate when fraud is likely to occur? Various studies have discussed not only the need for experience and supervision of auditors, but also the effect of decision models, and even expert systems, on fraud detection and the study of internal controls. This paper presents a discussion of regulation regarding fraud and the evolution of statistical models in the accounting and auditing literature designed to measure financial difficulties and fraudulent financial reporting. We test selected models using company data from two different time periods representing recent incidents of fraudulent financial reporting. Our results show high accuracy rates of fraud detection on both datasets, supporting the fact that a combination of accounting rules, regulation and government enforcement has contributed to the development of successful decision models to detect fraud. * The authors are, respectively, Associate Professor of Accounting at Meredith College and Professor of Accounting at Kent State University.