Detecting Financial Statement Fraud: An Alternative Evaluation of Automated Tools Using Portfolio Performance

This research combines the fields of accounting fraud detection and investment, the intersection of which is largely unexplored in research, to evaluate financial statement fraud detection models in an investment context and thus complement traditional evaluation. Financial statement fraud involves the intentional publication of false or misleading information in financial statements. Large-scale fraud cases such as Enron, WorldCom, Theranos, and Satyam (India) have made news headlines around the world. Far from being victimless crimes, such frauds leave behind real financial losses borne by individuals. Not only is there a negative effect on the company involved (Dyck et al., 2021), but also on stakeholders such as employees and investors, and further on the whole economy through the misallocation of resources (Kedia et al., 2009).
Despite efforts to measure the cost of fraud, its true cost is difficult to determine because much of it goes undetected. However, a staggering value of over $4.7 trillion USD was projected as the total global fraud loss in 2021 (ACFE, 2022).
This ACFE study focused on three major categories: asset misappropriation, corruption, and financial statement fraud. Out of the three, financial statement fraud has the lowest frequency, but the highest severity per incident.