Predicting the default probability of companies in USA and EU during the financial crisis, A study based on the KMV model

Abstract: The purpose of this study is to determine whether it is easier to predict the default probability in EU than in the USA or vice versa by analyzing the time period 2006-2009 which is characterized by the financial crisis. We will also establish whether there are any differences in the prediction of default probabilities between non-financial and financial firms. To be able to perform the study, we have collected data from 28 firms, 14 from each economy. Further, we have used the KMV model in order to calculate the default probabilities of the companies. The model is owned and used by Moody’s, one of the major rating agencies. When calculating the default probabilities, we assume normal distribution while Moody’s instead uses a large data base of historical default information. From the study we can conclude that the model to a large extent is able to predict the default probability of a company. Although, since this is a qualitative study, the conclusions are only an indication of the reality. The results imply that the model is able to predict a default of a non-financial firm approximately 1,5 years before the default actually occurs. Since the financial crisis has affected financial companies especially bad, leading to a severe increase in their already high leverage, the default probabilities of these companies are extremely high. The high probability values as well as similarities between the companies make it difficult for us to draw any conclusions about this industry.