Evaluating statistical models of mortgage lending discrimination: A bank-specific analysis

We present our efforts to develop bank-specific models to test for the pres ence of mortgage lending discrimination. We discuss the potential for selec tion and simultaneity biases and delineate the conditions under which a sin gle-equation model is appropriate. The results from three national banks de monstrate that, by incorporating the specific underwriting guidelines of ea ch bank, our alternative approach significantly improves the ability of the model to explain the outcomes of the mortgage lending decision process whe n compared to a single generic specification applied across all banks. Our results also demonstrate the difficulties encountered in attempting to inco rporate the specifics of a bank's underwriting criteria and the remaining p otential for omitted-variables problems.