A Credit Scoring Model for Commercial Loans : Comment

Ogler's article observes that while analytical models have long been developed to screen new loan applications, much less attention has been devoted to the loan review function. While it is of primary importance for a loan officer to ascertain credit quality prior to granting a loan, nonetheless, many apparently qualified borrowers become poor credit risks for reasons that develop subsequent to granting of the loan. In such cases it is important to identify weak loans as early as possible so that appropriate corrective action can be taken to avoid or minimize losses. In this regard, it is Ogler's stated objective "to develop a general credit-scoring model for evaluating existing commercial loans."As is customary in statistical studies of credit quality, Ogler defends the dependent variable in qualitative terms and then tests several sets of independent variables to find the set that best best discriminates between the groups defined by the dependent variable. While Orgler's statistical methodology appears sound, the definition and justification of his dependent variable is less satisfactory. The dependent variable measures loan quality as being either good or bad based on the judgment of bank examiners. It is unfortunate that Orgler's data dictated his research objective and model because they also limited the generality of his conclusions. Nonetheless, Orgler has considered an important question deserving of further research. If possible, such research should give careful attention to the objective measurement of actual loan quality and not personal judgments about quality, no matter how refined such judgments may be