Abstract
The purpose of this research note is to document the importance of empirical specification in the analysis of the influence of race on home mortgage lending decisions. Using a variety of testing procedures and detailed lending decision data for a large midwestern lending institution, the authors show that dramatically different results may be obtained from slight changes in empirical specification. The case study presented here can serve as a prototype for the large number of such case studies needed to obtain convincing evidence on the question of where and how discrimination may, or may not, emerge as a marketwide phenomenon.
Get full access to this article
View all access options for this article.
