Reverse Redlining and the Fintech Marketplace: Evidence from US Zip Code

The rise of financial technologies (fintech) in household finance raises an urgent need to understand how it shapes the already uneven banking landscape in the U.S. Given the historical origins of unequal financial access in racist economic systems such as redlining and housing segregation, it is critical to examine the impact of fintech on the state and trajectory of financial access from the lenses of racialized and segregated financial markets. A growing body of research indicates that use of alternative financial services (AFS) is associated with poor financial and health outcomes; such negative consequences are disproportionally borne by the poor, ethnic/racial minority communities. Moreover, the question of how fintech affects the banking gap and whether it will alleviate the negative impact of AFS is unknown. Using a uniquely merged dataset, this study investigates community-level fintech and financial access with a focus on the intersection of race and poverty.

Data were merged from five sources including 2015 Esri Business Analyst Market Potential, 2014 Federal Deposit of Insurance Corporation’s Summary of Deposit, 2014 National Credit Union Association, 2015 InfoGroup USA, and 2010-14 US Census Bureau’s American Community Survey (ACS). All data were collected at the zip code level and merged based on US Census Bureau Zip Code Tabulation Areas, which were used as a proxy for communities (N ≈ 30,000). Financial access is the focal dependent variable and is measured by the density of mainstream financial service providers including bank and credit union branches as well as alternative financial providers within zip code area. Fintech use is the focal independent variable and is measured by zip codes’ estimated rate of online banking and mobile banking. Race variables were calculated using the percentage of the populations within zip codes who identified as different racial groups, indicating the percentages of Black, Latinx, Asian, American Indian/Alaska Native residents. Poverty, another focal variable that assesses poverty status within zip codes, was constructed by calculating the percentage of the population living at or below the federal poverty line. Several zip code-level demographics and socioeconomic indicators were included as control variables.

Ordinary Least Square regressions were used to estimate the relationship between fintech and financial access in the full sample and high-poverty subsample. Interaction terms of race and fintech were added to assess the added effect of race on the relationship between fintech and AFS density. Preliminary findings show a positive association between fintech and access to mainstream banking services as well as AFS, the association between fintech and traditional banking services were 6 to 10 times stronger than the association between fintech and AFS density. In the high-poverty subsample, the relationship between fintech and mainstream banking services further strengthened, the association between fintech and AFS remained the same. Findings revealed that the interaction between race and fintech explained 1.1% of the variance in AFS density These results suggest that fintech might have a moderate impact on AFS density, and likely have a smaller impact among communities that have higher percentage of black and brown populations.