A decision aid for assessing the likelihood of fraudulent financial reporting

The auditor's responsibility for detecting fraudulent financial reporting i s of continuing importance to both the profession and society. The Auditing Standards Board has recently issued SAS No. 82, Consideration of Fraud in a Financial Statement Audit, which makes the auditor's responsibility for t he detection of material fraud more explicit without increasing the level o f responsibility. Using a sample of 77 fraud engagements and 305 nonfraud engagements, we dev elop and test a logistic regression model that estimates the likelihood of fraudulent financial reporting for an audit client, conditioned on the pres ence or absence of several fraud-risk factors. The significant risk factors included in the final model are: weak internal control environment, rapid company growth, inadequate or inconsistent relative profitability, manageme nt places undue emphasis on meeting earnings projections, management lied t o the auditors or was overly evasive, the ownership status (public vs. priv ate) of the entity, and an interaction term between a weak control environm ent and an aggressive management attitude toward financial reporting. The l ogistic model was significantly more accurate than practicing auditors in a ssessing risk for the 77 fraud observations. There was not a significant di fference between model assessments and those of practicing auditors for the sample of nonfraud cases. These findings suggest that a relatively simple decision aid performs quite well in differentiating between fraud and nonfraud observations. Practitio ners might consider using this model, or one developed using a similar proc edure, in fulfilling the SAS No. 82 requirement to "assess the risk of mate rial misstatement of the financial statements due to fraud."