The rate of mortality among newly founded firms is very high. Failure statistics universally show that over 50% of newly founded firms will fail during their first five years. The economic, financial, and social losses resulting from these failures are significant. Thus, it is valuable to try to develop methods to predict and to avoid suchfailures. However, there are only very few studies dealing with failure prediction methods for newly founded firms. The aim of this study was to develop such methods that would also have practical value. Failure was defined as cash insolvency. The aim was to develop afailure prediction model based on financial statement datafrom newly founded firms. The ratios based on these data are hard measures and easy to calculate also for a person forced to only use publicly available information. A prediction model based on these ratios could, then, be a practical method for financial analysts and other interest groups in evaluating the failureprobability ofa new firm. This model does not require information about management, products, or markets because it rather deals with the symptoms of failure than the causes. The study is based on a supposition that failure process in a newly founded firm is characterized by a too high initial indebtedness and by too low revenue financing as compared with the budget. This supposition was supported by the empirical results based on the comparison of the financial ratios in first four years between 20 failed and 20 nonfailed small, newly founded, entrepreneurial, industrial firms. These results may give an explanation as to why some newly founded firms will fail and some will survive. This explanation is, of course, based on the development of financial variables in the very first years. The results showed that it is possible, to some degree, to predict thefailure of a newly founded firm already in the first year after foundation. The prediction accuracy will increase, when the data of failure is approached. The best univariate predictors proved to be stockholders capital to total capital ratio (indebtedness), cash flow to net sales ratio (revenue financing), and cash flow to total debt ratio (sufficiencv of revenue financing to pay financial obligations). The prediction accuracy can be increased by adjusting the critical value of the ratio stepwisely to the year of operation. Thus, the optimal critical value will change year by year afterfoundation. The prediction accuracy can slightlv be improved by a multivariate analysis. This kind of multivariate model also included the logarithmic size (net sales) of the firm as a variable so that the probability to fail increased when the size increased. The results also showed that the same multivariate model estimated for older small firms, can be used as a failure prediction model for newly founded firms. However, this requires that quite different critical values are selected for both types of firms. To summarize, the results show that the failure of a newly founded firm is, to some degree, predictable by univariate or multivariate analyses. Thefailure risk is increased with high indebtedness, insufficient revenue financing, and large size in the first years. Hence, the risk to fail can be reduced by using less debt as initial financing and by paying special attention to the generation of a sufficient amount of revenues in initial stages. Furthermore, it may be less risky to start business operations with a smaller, rather than larger initial size.
