New Technology, Old Patterns: Fintech Lending, Metropolitan Segregation, and Subprime Credit

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 app...

Ausführliche Beschreibung

Gespeichert in:  
Bibliographische Detailangaben
1. VerfasserIn: Haupert, Tyler (VerfasserIn)
Medienart: Elektronisch Aufsatz
Sprache:Englisch
Veröffentlicht: 2022
In: Race and social problems
Jahr: 2022, Band: 14, Heft: 4, Seiten: 293-307
Online-Zugang: Volltext (lizenzpflichtig)
Journals Online & Print:
Lade...
Verfügbarkeit prüfen: HBZ Gateway
Schlagwörter:
Beschreibung
Zusammenfassung: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.
ISSN:1867-1756
DOI:10.1007/s12552-021-09353-0