Many bankruptcy models have been created by a variety of authors from different countries. A problem arises with the use of these models in an environment other than that in which the model was created. Several studies have pointed out that the prediction accuracy of bankruptcy models (their ability to differentiate correctly between a company threatened by bankruptcy and a prospering company) falls markedly when they are applied to a different branch, period or economic environment than the original environment. Another study showed there is a link between the manager’s decisions and financial results and macroeconomic factors, as successful companies often analyze foreign and domestic macroeconomic conditions and corresponding changes in the corporate sectors.
From that viewpoint, incorporating macroeconomic factors into bankruptcy prediction models is very legitimate. However, the aforementioned scientific problem is not yet solved as these studies conclude that the macroeconomic factors are specific for given environments. In other words further research is needed to understand the problem.
Our research sample consists of 20,308 manufacturing small-and-medium-sized enterprises (SMEs) from 28 EU countries. We used a balanced sample which consists of 10,154 defaulted companies that went bankrupt during 2014 and 2016 and 10,154 non-defaulted companies. In course of the research, five bankruptcy prediction models were analyzed. The relationship between macroeconomic factors and the accuracy of the models was explored using correlation analysis and linear regression models with fixed effects.
We identify macroeconomic factors which influence the financial health of the company or rather enhance the probability of its bankruptcy. During downturns in the economy, the number of companies experiencing solvency problems is growing, which is evident in the average values of financial ratios of these companies. This directly effects the prior probabilities of bankruptcy and will be incorporated into the model.