The new Basel II regulation contains a number of new regulatory features. Most importantly, internal ratings will be given a central role in the evaluation of the riskiness of bank loans. Another novelty is that retail credit and loans to small and medium-sized enterprises will receive a special treatment in recognition of the fact that the riskiness of such exposure derives to a greater extent from idiosyncratic risk and much less from common factor risk. Much of the work done on the differences between the risk properties of retail, SME and corporate credit has been based on parameterized models of credit risk. In this paper we present new quantitative evidence on the implied credit loss distributions for two Swedish banks using a non-parametric Monte Carlo re-sampling method following Carey . Our results are based on a panel data set containing both loan and internal rating data from the banks complete business loan portfolios over the period 1997-2000. We compute the credit loss distributions that each rating system implies and compare the required economic capital implied by these loss distributions with the regulatory capital under Basel II. By exploiting the fact that a subset of all businesses in the sample is rated by both banks, we can generate loss distributions for SME, retail and corporate credit portfolios with a constant risk profile. Our findings suggest that a special treatment for retail credit and SME loans may not be justified. We also investigate if any alternative definition of SMEs and retail credit would warrant different risk weight functions for these types of exposure. Our results indicate that it may be difficult to find a simple risk weight function that can account for the differences in portfolio risk properties between banks and asset types.
Internal ratings Credit risk Value-at-Risk Banks Basel II Retail credit SME risk weights