Banks use credit-rating models in underwriting and pricing loans, managing portfolios, and estimating loan loss reserves and capital, among other purposes. While such models are common in both corporate and retail lending, there are differences in how they are developed and used. The advent of Basel II will bring about some convergence in how credit risk is evaluated in retail and corporate lending.
Corporate credit-rating models are used to quantify credit risk exposures at the obligor level. Each rating grade maps to a specific range of either overlapping or mutually exclusive expected default likelihood or expected losses. Retail credit-scoring models are also typically applied to individual credit applicants, but are used to generate a rank-ordering of applicants by their credit risk, measured as the expected likelihood of delinquency or default.
Now, with the imminent advent of Basel II, convergence in the use of the output from the two approaches can be expected. The Basel II framework for assessing capital requirements for credit risk is based on corporate approach, which means that the output from retail methodologies will need to be modified to align more closely with the corporate approach. In particular, it is likely that the methodologies now used in retail lending will be enhanced to place more emphasis on assigning specific credit grades or ratings that relate to specific default likelihood rather than simply rank-ordering risk. This is not expected to lead to an overhaul of the retail methodologies; rather, information from existing methodologies will be expressed in ways that are more comparable to corporate models in terms of enhanced transparency and new model governance requirements. The model governance requirements are going to be particularly challenging due to the presence of numerous scoring models in the various retail lending environments of the bank, unlike the single corporate model usually utilized by the bank. |
