The Rainbow of Credit: Same-sex Mortgage Discrimination and Two-sided Spillover Effect

Using the Home Mortgage Disclosure Act (HMDA) national data from 1990 to 2015, we propose a method to identify potentially homosexual borrowers without requiring any self-disclosure on sexual orientation. Augmenting the HMDA data with the classical 1990 Boston Fed data and Fannie Mae Loan Performance data, we test whether their perceived homosexual status affects the approval, cost, and performance of the mortgages for which they apply. The results reveal that, in contrast with otherwise comparable loan applicants, the average approval rate for potentially homosexual applicants is about 3% to 8% lower. Furthermore, conditional on being approved, their financing cost is about 0.02% to 0.2% higher. This is equivalent to an annual total of $8.6 million to $86 million in additional interest/fees paid by same-sex borrowers nationwide. Meanwhile, we find no evidence that homosexual status is associated with higher default risk. Finally, we document evidence of a twosided spillover effect. That is, holding other factors constant, when a neighborhood’s same-sex population increases, although same-sex applicants still experience lower approval rate and higher financing cost, the treatment discrepancy between same-sex and straight applicants narrows from both sides.