Authors: Stefano Grazioli (U. of Virginia), Karim Jamal (U. of Alberta) & Paul E. Johnson (U. of Minnesota)
Publication: Working Paper Series
Year: 2006
Focus Area: Detection
Relevance: While we all possess some measure of deception detection abilities, analyzing those who are professionals at detecting fraud (in this case financial auditors) highlights both valuable tools and common errors in our detection strategies.
Summary: This article focuses on “financial statements fraud,” in which statements are manipulated or falsified to reflect inaccurate figures for financial gain (such as phony investment fraud). By combining accounting and fraud detection knowledge, auditors must decipher and diagnose fraudulent cases.
The authors examine the relative success rates of auditors, compare professional auditors’ success rate with that of a computer model, and propose an explanatory theory of fraud detection performance.
- The authors constructed a computer model designed to diagnose cases of fraud (85% success rate).
- In contrast, auditors demonstrated a high rate of errors and failure to diagnose fraudulent cases (45% success rate).
- Auditors may be hesitant to diagnose fraud given the negative emotional and financial consequences of misdiagnosing a case as fraudulent.
- The computer model, on the other hand, did not consider potential repercussions.
- While auditors consistently identified discrepancies, they often let these “cues” of fraud pass. Experienced auditors, having previously seen legitimate explanations for such a discrepancy, hypothesized a similarly innocent explanation for the cue at hand. Their extensive exposure to “clean” cases, and their tendency to generalize from these, made auditors less prepared to identify the relatively rare instances of fraud. Computer models are less likely to apply previous “innocent” explanations to evaluate new situations.
- Those few auditors who demonstrated abnormally high success rates (90%+) were more likely to hypothesize fraud when they located errors.
While there are ongoing developments in fraud detection by statistical analysis and modeling, the authors hold that fraud detection professionals remain the “most powerful means to detect strategic misrepresentations of financial information” (p. 26). To increase effectiveness, detection professionals must be trained in both their specific field (e.g. accounting) and in deception and detection methods (training which the majority of the auditors studied largely lacked).
Author Abstract: Fraud detection researchers have spent a great deal of effort looking for information cues (often termed ‘red flags’) that signal the presence of fraud (Albrecht and Romney 1986). This research has been motivated by the desire to improve on auditors’ accuracy at detecting fraud, in particular financial statements fraud. Despite the intuitive appeal of the red flag approach, studies have shown that this search for cues has not been entirely successful. Red flags have frequently been found to be ineffective, sometimes even hindering the ability of an auditor to detect fraud (e.g., Pinkus 1989; Johnson, Jamal and Berryman 1989).
consolidates information
