Macroeconomic factors and credit risk.

Summary The objective of this thesis is to study the impact of macroeconomic factors on credit risk. In this thesis, we use two types of models that allow us to exploit a large number of series. The first model used refers to the GVAR model (Global Vector AutoRegressive model) developed by Pesaran et al. (2004). We consider a fictitious portfolio of 83 firms covering 16 developed countries. We find that default probabilities increase significantly during recessions but do not decrease during growth periods. We also confirm the fact that companies with a good rating are less sensitive to changes in economic conditions than those with a poor rating. The second model studied is the dynamic factor model of the FAVAR type (Factor Augmented Vector Autoregressive Model), proposed by Stock and Watson (2005). We propose two empirical applications, one focused on the United States, the other on the Eurozone. We find that common factors explain relatively little of the default rates of firms. This result demonstrates the importance of risk diversification. We also find that the most explanatory factor of the default rate is the factor related to real activity, such as production and employment. Another important explanatory factor is the one related to stock market indices. Finally, we find that monetary policy is far from being the dominant factor in determining changes in the default rate. Changes in the FeD's policy rate do not therefore appear to be one of the causes of the subprime crisis.