FINANCIAL RATIOS FOR DETECTION OF COMPANY’S INSOLVENCY AND BANKRUPTCY FRAUD: SIMILARITIES AND DIFFERENCES

This research is aimed to identify the similarities and differences of financial ratios used in the international methods for detecting insolvency (bankruptcy) and bankruptcy fraud. This is the first step in the process of developing a Latvia model of bankruptcy fraud detection. The methods of research: literature review, analysis and synthesis, comparative and correlation analysis. The selection of the analysed financial ratios is based on data from 28 bankruptcy forecasting models tested in Latvia and Lithuania, developed on the basis of multiple discriminant and regression analysis. In addition, the ratios of 3 analogical models for detecting fraud are also analysed. 58 indicators for bankruptcy forecasting were included in the study, selecting 19 indicators, which are also used to reveal fraud in financial statements (FFS) and to detect bankruptcy fraud in experts’ and auditors’ practice. The forensic methods to detect bankruptcy fraud, the international guidelines to reveal FFS, legislative acts and scientific research in this area are reviewed with the purpose to identify and analyse the similarities and differences in detecting insolvency (bankruptcy) and bankruptcy fraud. This type of complex analysis has not been presented in the Baltic countries yet. The article presents the testing results of fraud’s revealing score models (M-score, F-score) using a sample of data obtained manually from financial statements of 114 small and medium companies in Latvia. It has been concluded that it is necessary to implement the system of financial ratios and assess the possibility to use fraud’s revealing models in accounting examination. The authors stated the impossibility of using the M-score and F-score models without calibration for Latvia companies. The logistic regression model for revealing fraud (Lithuania) one year before the fact of bankruptcy forecasts the possibility of the bankruptcy fraud with an accuracy of 61.5%. The results of the study can be used both to create a model for detecting bankruptcy fraud and to develop an accounting expertise procedure to reveal fraud.