Description
TitleFHA and the dual mortgage delivery system in Philadelphia
Date Created2022
Other Date2022-05 (degree)
Extent292 pages : illustrations
DescriptionThis dissertation explores how a mortgage product is related to racial equity in Philadelphia. Locally and nationally, Black and Latino/a homebuyers are much more likely to rely on mortgages that are insured by the Federal Housing Administration (FHA). Yet, FHA-insured mortgages are generally understood to be an expensive loan of last resort for borrowers unable to access conventional financing. I use a combination of methodologies to understand how an apparently racially neutral mortgage insurance program operates in a racially segregated and unequal city. My approach combines: 1) exploratory spatial data analysis of fourteen years of Home Mortgage Disclosure Act (HMDA) data, 2) fieldwork with home purchase counselors, lenders and realtors, and 3) regression analysis using logit and linear approaches. In this dissertation, I show there is a dual mortgage delivery system operating in Philadelphia, as of 2019. Using spatial cluster procedures, I find a geographic divide in the kind of housing finance used, with low-down payment conventional home purchase loans densely clustered in the city’s affluent and rapidly appreciating neighborhoods close to the downtown area, and FHA-insured loans are heavily clustered in Philly’s moderate- and middle-income neighborhoods some distance removed from downtown. There are no areas of overlap in the spatial patterns of these two financing streams. The conventional clusters all have growing non-Hispanic White populations, and shrinking Black populations. The FHA clusters have growing Black, Latino/a and/or immigrant populations, and shrinking NH White populations. Philly’s lenders are related to this dual system. One group of lenders, including the largest banks and nonbanks, is serving primarily conventional mortgages to predominantly White non-Hispanic borrowers in affluent and rapidly appreciating neighborhoods. Another group of lenders, including smaller community banks and nonbanks recently entering the local market, is serving significant FHA-insured mortgages to larger shares Black, Latino/a, and lower-income borrowers, in the city’s diverse moderate and especially middle-income neighborhoods, that are geographically removed from downtown. As of 2019, it’s the largest banks who offer the best low-credit low-down payment mortgages, with reduced or eliminated mortgage insurance. Yet these same large banks have significantly higher denial rates for Black and Latino/a applicants and have limited loan footprints outside of the greater downtown area. My research points to the ongoing importance of place-based scholarship to fair credit access. I find FHA-insured buyers are disadvantaged in appreciating areas with competitive markets because of realtors’ preference for conventional bids. At the same time, the long-term concentration of FHA-insured mortgages in Philly’s middle-income neighborhoods, especially in the city’s Black middle-income neighborhoods, is troubling because it raises questions about equitable access to conventional mortgage products. A more recent concentration of higher-cost FHA-insured mortgages may threaten neighborhood stability because they are more likely to default, with the potential to concentrate foreclosure and blight. Finally, this dissertation complicates our traditional understanding of FHA-insured mortgages as loans of last resort. Many Black and Latino/a borrowers are funneled into FHA because of the way socioeconomic inequality intersects with the racial logics of pricing, especially for private mortgage insurance. Risk-based pricing translates lower income, credit scores, savings, and more debt into higher-priced mortgages. These higher costs can push even people who qualify for conventional loans into FHA-insured products.
NotePh.D.
NoteIncludes bibliographical references
Genretheses
LanguageEnglish
CollectionSchool of Graduate Studies Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Statistical Profile