ABSTRACT

In a 1992 study on lending discrimination for the Federal Reserve Bank of Boston, Munnell and colleagues found that race played an independent role in the mortgage lending decision. Three years later, these authors refined and extended their results (Munnell et al. 1996). The earlier study, however, was subjected to several criticisms, ranging, on a conceptual level, from concerns about the effects of possible omitted variable bias to, on a more mundane level, questions about the specification of the model and the integrity of the data. Whereas any of these criticisms could potentially be important in a study of this sort, this chapter demonstrates that, in fact, none are.