Ebook Credit Risk versus Capital Requirements under Basel II: Are SME Loans and Retail Credit Really Different?
Although non-financial corporate debt (bond issues and privately issued debt) has become more common in the past 10-20 years, bank loans are still the prime source of business finance, especially for small and medium-sized enterprises (SMEs). As a consequence, banks’ ex-ante assessment of the riskiness of loan applicants and the resulting decision to grant credit (or not) at some risk-adjusted interest rate, is of great importance for businesses. Bank regulators increasingly lean on the risk assessments made by banks: in the Basel Committee’s proposal for new capital adequacy rules, the so called Basel II Accord, internal risk ratings produced by banks have been given a prominent role.
Unlike previous regulation, the rules of Basel II will make the size of the required buffer capital contingent on a bank’s appraisal of ex-ante individual counterpart risk. It will be up to each bank to characterize the riskiness of the counterparts and loans in its portfolio by means of a relatively small number of risk categories or ”rating classes”. A special feature of the new regulation is that retail credit and loans to SMEs will receive a different treatment than corporate loans and will require less regulatory capital for given default probabilities. The main reason for this differential treatment is the supposedly low correlation between small business loans. Their risk is generally thought to be largely of an idiosyncratic nature.
The Basel proposal has been criticized extensively because of its implications. Altman and Saunders found that relying on traditional agency ratings may produce cyclically lagging rather than leading capital requirements and that the risk based bucketing proposal lacks a sufficient degree of granularity. Among other things, they advise to use a risk weighting system that more closely resembles the actual loss experience on loans. Criticism like this has spurred subsequent research by authors like Carling, Jacobson, Lindé and Roszbach, Dietsch and Petey, Estrella, Calem and LaCour-Little, and Hamerle, Liebig, and Rösch. Their work employs credit risk models for the ultimate goal of calculating capital requirements under a variety of alternative systems. It makes clear how the proposed internal ratings based (IRB) approach relates to general Value-at-Risk (VaR) models of credit risk and state-of-the-art risk rating and how the technical specification of the final IRB design will affect banks’ policies.
To what extent a different treatment of retail credit and SME loans is justified will depend on at least two factors: the ability of banks’ internal risk rating systems to adequately capture the differences between different loans and different types of assets, and the methods used to calculate the relevant risk measure. Several authors have studied the ability of internal ratings (IR) systems to handle differences between (portfolios of) assets and the implications for credit risk measurement and the eventual functioning of Basel II. Gordy shows that IRB bucket models of credit can be reconciled with the general class of credit Value-at-Risk (VaR) models.
Carey concludes that the success of the IRB approach will depend on the extent to which it will take into account differences in assets and portfolio characteristics, such as granularity, risk properties and remaining maturities. Jacobson, Lindé and Roszbach find that IRB parameters such as the target forecasting horizon, the method to estimate average probabilities of default (PDs) and banks´ business cycle sensitivity will also affect the way in which the IRB system can function. Carey and Hrycay study the effect of internal risk rating systems on estimated portfolio credit risk and find that some of the commonly used methods to estimate average probabilities of default (PDs) by rating class are potentially subject to bias, instability and gaming. Jacobson, Lindé and Roszbach investigate the consistency of internal ratings at two major Swedish banks. They find that loan size and portfolio size are very important determinants of the shape of credit loss distributions and that the banks differ significantly in their perception of an identical loan portfolio’s riskiness.
Posted in :
