Financial Shenanigans in Listed Manufacturing Firms in Nigeria: Fraud Diamond Indicators and Beneish-flagged Earnings Manipulation
Reuben Osita Nwosu *
Department of Auditing and Forensic Accounting, College of Public Sector Accounting, ANAN University, Kwall, Plateau State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
This study examines associations between five observable Fraud Diamond indicators and Beneish M-Score-flagged earnings-manipulation risk among listed manufacturing companies in Nigeria. The indicators are asset change, board monitoring, managerial ownership, auditor change and chief executive officer (CEO) financial expertise. The analysis uses 340 firm-year observations for 40 companies over 2014-2022 and estimates population-averaged logistic models through generalised estimating equations (GEE), incorporating year effects and Mancl-DeRouen bias-reduced robust covariance to account for within-firm dependence and the modest number of firm clusters. In the contemporaneous model, a one-standard-deviation increase in asset growth is associated with 1.52 times the odds of a Beneish flag (p = .047), while auditor change is associated with 2.53 times the odds (p = .014). Board monitoring is negatively associated at the 10% level (odds ratio = 0.82, p = .073); managerial ownership and CEO financial expertise are not statistically significant. These relationships do not persist in the one-year-lagged GEE or conditional firm fixed-effects logit models. The main model has moderate in-sample discrimination (AUC = 0.703). The evidence therefore supports interpreting asset growth and auditor change as contemporaneous red flags, not established causes or reliable one-year-ahead predictors. A Beneish flag is a screening signal and should not be read as proof of financial statement fraud.
Keywords: Beneish M-Score, earnings manipulation, fraud diamond theory, auditor change, board independence, nigeria, generalised estimating equations