Objective: This study examines the effectiveness of the Beneish M-Score model in detecting financial statement manipulation in companies charged with fraud by the U.S. Securities and Exchange Commission (SEC), aligning with Sustainable Development Goal (SDG) 16: Peace, Justice, and Strong Institutions, which promotes financial transparency and accountability. Theoretical Framework: Earnings management is a major concern in financial reporting, as it affects the reliability of financial statements. The Beneish M-Score model is a mathematical tool developed to detect earnings manipulation. Additionally, artificial intelligence (AI) is emerging as a complementary method to enhance fraud detection. Method: The research applies the Beneish M-Score model to three SEC-charged companies: DXC Technology Co., GTT Communications, Inc., and Luckin Coffee Inc. Financial data is collected from the SEC’s EDGAR database and company reports to compute the M-Score over multiple fiscal years. Results and Discussion: The model successfully flagged Luckin Coffee as a likely manipulator during its peak fraudulent period. DXC Technology Co. showed signs of manipulation in 2018, aligning with the SEC’s findings of misleading non-GAAP disclosures. GTT Communications, Inc. had a more stable M-Score, with no strong indications of manipulation. AI techniques, such as machine learning and neural networks, have the potential to improve fraud detection by analyzing financial patterns. Research Implications: This study supports the Beneish M-Score as a useful tool for detecting earnings manipulation and suggests integrating AI-driven methods to enhance accuracy in fraud detection. Originality/Value: The study contributes to forensic accounting research by validating the Beneish M-Score in real-world fraud cases and highlighting AI’s role in financial fraud detection, supporting SDG 16’s objectives of promoting transparency and strong institutions.
Read full abstract