Bank Lending Policy, Credit Scoring and Value at Risk
by Tor Jacobson of Sveriges Riksbank, and
Abstract: In this paper we apply a bivariate probit model to investigate the implications of bank lending policy. In the first equation we model the bank's decision to grant a loan, in the second the probability of default. We confirm that banks provide loans in a way that is not consistent with default risk minimization. The lending policy must thus either be inefficient or be the result of some other type of optimizing behavior than expected profit maximization. Value at Risk, being a value weighted sum of individual risks, provides a more adequate measure of monetary losses on a portfolio of loans than default risk. We derive a Value at Risk measure for the sample portfolio of loans and show how analyzing this can enable financial institutions to evaluate alternative lending policies on the basis of their implied credit risk and loss rate, and make lending rates consistent with the implied Value at Risk.
Keywords: Banks, lending policy, credit scoring, Value at Risk, bivariate probit.
Published in: Journal of Banking & Finance, Vol. 27, No. 4, (April 2003), pp. 615-633.