Racial Patterns in Mortgage Lending Outcomes During and After the Subprime Boom

Abstract

ABSTRACTThe racial contours of the United States’ subprime lending boom, foreclosure crisis, and subsequent recovered market illustrate that the benefits and risks of homeownership were unevenly distributed. Existing studies reveal general lending patterns in these periods but fail to scrutinize racially homogeneous neighborhoods where outcomes often diverge from aggregate trends. Using 2005 and 2015 U.S. Home Mortgage Disclosure Act mortgage application data, I construct a series of logistic regression models estimating lending outcomes in neighborhoods grouped by racial composition to reveal new patterns of differences in borrower outcomes. In both time periods the gap between white borrowers’ comparatively high application approval rates and minority borrowers’ lower approval rates grows as the proportion of white residents in a target neighborhood increases. In 2005 the likelihood of a borrower of any race receiving subprime terms generally increased as the proportion of white residents in a target neighborhood decreased. However, Asian borrowers experience comparatively favorable outcomes in homogeneously nonwhite neighborhoods. Additionally, applicants in homogeneously white neighborhoods experience less favorable results than those in slightly more diverse neighborhoods. Collectively, these findings suggest that outcomes in homogeneously white and nonwhite neighborhoods do not uniformly align with previously identified trends and warrant closer inspection.