Earnings Manipulation and Excess Returns A study on the Swedish stock exchange
Abstract: This study aims to evaluate the efficacy of one of the most well-known quantitative methods for discovering earnings manipulation, the M-score model (Beneish 1999), as a tool for investors on the Swedish stock exchange by testing the model's predictive power on future stock returns. The data used in the study covers 126 firms during the years of 2005-2017. The M-score's predictive power is tested by regressing the firm's assigned M-score against the stock's excess returns during the period, also testing for potential lagged effects. In accordance with previous research on the American stock exchange, the hypotheses of this study are that M-score higher has explanatory power on stock returns, and that a higher M-score should yield lower stock returns. Results conclude however that the M-score does not have significant explanatory power on future stock returns. Companies with a high M-score were not found to have been performing worse than companies with a low M-score on the stock exchange during the sample period, contrary to the hypothesis of this study. A likely cause of these results stems from the ongoing evolution of more complicated financial manipulation schemes, and knowledge of the model becoming more common amongst manipulating firms. With the hypotheses being rejected, M-score is not deemed to be a reliable tool to investors on the Swedish stock exchange and will likely be even less so in the future.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)