Financial statement fraud and earnings manipulation have attracted high-profile concerns recently. There have been countless cases by businesses of what occurs to be financial statement fraud, which have been undetected by the auditors. Fraudulent misrepresentation of financial statements has debilitating consequences not only on the perpetrators but the other related parties that were not involved in the fraud. Detection and prevention of fraud is therefore crucial to parties related to any company. The main purpose of this study is to carry out a pairwise multiple cases study investigation of financial statement fraud of fraud and non-fraud companies to ascertain whether the Beneish model can detect financial statement fraud before the prediction of the company's annual report by the media. Financial statements of five selected companies that are known to have committed financial statement fraud and five companies with no reported incident of fraudulent misrepresentation over the same period. Companies known to have committed fraud are selected based on news articles reported on these companies as well as the reasons of suspension of these companies from listing on the Johannesburg Stock Exchange (JSE). Uneven composition of the board of directors in companies has resulted in numerous issues of misappropriation in South Africa and globally in the recent past. The study applies both quantitative and qualitative techniques to achieve its stated objectives. The study investigates board composition, management holdings, CEO tenure, and CEO duality of both fraudulent and non-fraudulent companies. The study further applies the Beneish model to determine whether a company can be flagged for fraudulent misrepresentation one year before the publication of the fraudulent event in the media and suspension from the JSE. 

Keywords:   financial statement fraud, earnings management, fraud and non-fraud companies, non-executive directors, executive directors.