Assessing Portfolio Credit risk ChAnges in A sAmPle of eU lArge And ComPlex BAnking groUPs in reACtion to mACroeConomiC shoCks

In terms of regulatory and economic capital, credit risk is the most signi…cant risk faced by banks. We implement a credit risk model –based on publicly available information – with the aim of developing a tool to monitor credit risk in a sample of large and complex banking groups (LCBGs) in the EU. The results indicate varying credit risk pro…les across these LCBGs and over time. Furthermore, the results show that large negative shocks to real GDP have the largest impact on the credit risk pro…les of banks in the sample. Notwithstanding some caveats, the results demonstrate the potential value of this approach for monitoring …nancial stability. Key words: Portfolio credit risk measurement; stress testing; macroeconomic shock measurement JEL classi…cation: C02, C19, C52, C61, E32 4 ECB Working Paper Series No 1002 February 2009 Non-technical summary This paper attempts to address the issue of measuring credit risk in the European banking sector using an approach based on publicly available data. For the ECB this is particularly relevant because although it does not have supervisory responsibility for individual institutions and consequently access to supervisory data, it is responsible for contributing to the stability of the European …nancial system. By linking publicly available bank exposure data to information received from a global marcoeconomic model we are able to simulate the e¤ects of di¤erent macroeconomic shocks on corporate sector credit quality/default probabilities. This enables a model based assessment of credit risk in European bank portfolios under di¤erent macroeconomic scenarios and provides a tool for …nancial stability scenario analysis. To run the credit risk model without any scenario analysis we use bank exposure data, probabilities of default (PD) for each exposure and information on loss given default (LGD). All of them can be related to the following sectors: corporates, …nancial institution, households and the public sector. Using these inputs, the credit Value at Risk (VaR) for each single bank is calculated. The results show that credit risk varies across banks depending on the business lines pursued as well as their geographical and sector exposures. The remainder of the paper elaborates on the e¤ects of macroeconomic shocks on Value at Risk (VaR) which are calculated in two steps. First, the impulse responses of the GVAR (Global Vector Autoregression) model to a …ve standard standard deviation shocks in di¤erent macroeconomic variables are calculated. These shocks include real Gross Domestic Product (GDP), real stock prices, in‡ation, short-term and long-term interest rates and the euro-dollar exchange rate. In the second step, the results of these macroeconomic shocks are regressed on the sector speci…c PD values. The main results from the macro shocks show that the e¤ect on credit risk of European banks depends on the source of the macroeconomic shock as well as the banks loan exposures. Overall, shocks to real GDP increase the median VaR of banks in the sample the most from the shocks considered in this paper. 5 ECB Working Paper Series No 1002 February 2009