Racial Discrimination in Home Purchase Mortgage Lending Among Large National Banks

• Activity Number: 4
• Type: Invited
• Date/Time: Sunday, August 3, 2003 : 2:00 PM to 3:50 PM
• Sponsor: Business & Economics Statistics Section
• Author(s): Todd Vermilyea*+ and McKinley Blackburn
• Companies: Federal Reserve Bank of Philadelphia and University of South Carolina
• Keywords: Discrimination ; Disparate Impact ; Disparate Treatment

Abstract

Statistical models of mortgage lending underwriting have been used by bank regulators to assist in enforcing fair lending laws for the past several years. In general, these bank specific models have not produced evidence of discrimination, a finding that appears to be at odds with the well-known conclusion of the Boston Fed study. In this paper, we combine data from several bank-specific models constructed by the Office of the Comptroller of the Currency to obtain a single measure of disparate treatment discrimination in the banking sector. Our estimates suggest that blacks (and, less conclusively, Hispanics) are less likely to receive loans, given stated criteria used by individual banks. Our estimated impact of race is smaller than that suggested in the Boston Fed Study, which is explained largely by their imposition of a single decision-making framework across all lending institutions. We show that earlier studies have likely overstated--by a significant magnitude--the degree of discrimination, as these studies utilized pooled rather than lender-specific models. Their common specification approach may reflect disparate impact discrimination, although the estimated effect of race in this kind of model can also reflect nondiscriminatory influences.