|M.Sc Student||Mashiah Rinat|
|Subject||Identifying High Risk Borrowers in Markets with Asymmetric|
|Department||Department of Industrial Engineering and Management||Supervisor||Dr. Benjamin Bental|
This study deals with the problem of identifying high-risk borrowers in the credit market. In this market, lenders face potential borrowers under asymmetric information. This problem prevents the market from achieving a standard competitive equilibrium.
The question of the study is whether borrowers have information that may help in forecasting any borrower’s likelihood to fail. If so, the information may be used to formulate a credit strategy, based on a mixture of criteria that the lender can observe, in order to identify high risk borrowers.
The empirical part of the study is based on the credit history of a sample of small business borrowers of a bank branch in Israel. In the sample there are "good borrowers" who have not failed, and "bad borrowers" who have failed. Using a logistic regression, a failure probability function is estimated, based on variables taken from the criteria for credit approval. The thesis then shows that the predicted failure probability helps predict the amount of credit granted to borrowers. Accordingly, the thesis concludes that the lender uses relevant information when deciding the credit policy towards any individual borrower. However, the moderate proportion of variance (about 30%) for which the model accounts indicates that there are many others considerations in the decision of granting credit.