This research assesses the relationship between subprime lending rates among applicants to traditional and fintech mortgage lenders and metropolitan-level racial and ethnic segregation in the United States. Fintech—short for financial technology—mortgage lenders underwrite loans using all-online applications and proprietary machine learning underwriting algorithms that process unprecedented amounts of applicant data. While traditional lenders have long been associated with high rates of subprime lending in segregated metropolitan areas, it is unknown whether fintech lenders also exhibit this relationship. Using Home Mortgage Disclosure Act data from the nation’s 200 largest metropolitan areas in 2015–2017 and a series of binomial logistic regressions, I find the probability of an applicant receiving a subprime loan at both traditional and fintech lenders is positively associated with metropolitan area Black and Hispanic segregation. However, fintech lending is associated with significantly lower rates of subprime lending, relative to traditional lending, in metropolitan areas with high levels of Black segregation. This relationship holds true when analyzing both Black-white dissimilarity and Black isolation. Results related to white-Hispanic segregation are mixed. Fintech lenders are more likely than traditional lenders to originate subprime loans in metropolitan areas with high levels of white-Hispanic dissimilarity, but less likely as a metropolitan area’s Hispanic isolation increases. Findings suggest the structural forces connecting subprime lending to metropolitan segregation—especially Black segregation—have a weaker association with the fintech lending market than the traditional market, but still play a significant structural role in shaping fintech lending outcomes.

本研究评估了传统和金融科技抵押贷款机构申请人之间的次级贷款利率与美国大都市种族和民族隔离之间的关系。Fintech（金融技术的缩写）抵押贷款人使用全在线应用程序和专有的机器学习承销算法来承销贷款，这些算法可以处理前所未有的大量申请人数据。虽然传统贷方长期以来一直与隔离大都市地区的高利率次级贷款有关，但尚不清楚金融科技贷方是否也表现出这种关系。使用 2015-2017 年全国 200 个最大都市地区的《房屋抵押贷款披露法》数据和一系列二项式逻辑回归，我发现申请人在传统和金融科技贷方获得次级贷款的可能性与大都市区黑人和西班牙裔隔离呈正相关。然而，与传统贷款相比，在黑人隔离程度高的大都市地区，金融科技贷款与次级贷款利率显着降低有关。在分析黑白差异和黑色隔离时，这种关系成立。与西班牙裔白人隔离相关的结果好坏参半。与传统贷款人相比，金融科技贷款人更有可能在白人与西班牙裔差异程度较高的大都市地区发放次级贷款，但随着大都市地区西班牙裔隔离程度的增加，这种可能性较小。