Economics of credit scoring management

Credit scoring models constitute an inevitable element of modern risk and profitability management in retail financial lending institutions. Quality, or separation power of a credit scoring model is usually assessed with the Gini coefficient. Generally, the higher Gini coefficient the better, as in this way a bank can increase number of good customers and/or reject more bad applicants. In the paper a simple simulation framework for analysis of microeconomics of credit scoring management is presented. The model takes into account competition among banks (there are 10 competing banks in the model), risk-based pricing (the banks differentiate prices based on their credit scoring models), “loan-shopping” practices by credit applicants (each applicant checks the price offered by three randomly selected banks). Such a setup enables us to perform a simulation where one of the banks improves the credit scoring model used and benefits from it. As the simulation shows, even small changes in Gini coefficient may lead to substantial improvement of bank’s standing measured by its profitability and market share.