The importance of credit risk assessment and monitoring has increased since the recent financial turmoil. This paper presents a toolkit for credit risk modeling that follows the top-down approach proposed by Wilson (1997). The analysis is conducted separately for the household and corporate sector, by means of panel techniques and seemingly unrelated equations, using default aggregated data at county or business sector level. The results indicate that the determinants of default on bank loans for the household sector are unemployment, exchange rate, industrial production, indebtedness and interest rate spreads, while for the corporate sector the output gap, indebtedness and exchange rate are the main factors. Comparing the two models, it arises that default events from the corporate sector occur sooner than for the household sector in case of adverse macroeconomic developments. There are two possible explanations: i) there is no personal bankruptcy law for individuals in Romania, and ii) public administration appears to adjusts slower during recessions, an important part of the work force being part of this system. Furthermore, stress-testing analysis is performed on arbitrarily built portfolios by considering the impact of macroeconomic shocks on the probabilities of default over a one year time horizon.
