Association for Public Policy Analysis & Management

Given the ubiquity and necessity of financial services in today’s economy (Davis 2009), the clear patterns of racial disparities in United States financial services are alarming. While financial exclusion and exploitation via racial segregation are not new phenomena (Baradaran 2017; Caplovitz 1968; Rothstein 2017), contemporary manifestations of this “Ghetto Tax” (or “Poverty Tax”) are concerning given the concurrent trends of increasing income inequality (Piketty 2014), volatility (Morduch and Schneider 2017), and segregation (Bischoff and Reardon 2014). Therefore, it is crucial to understand how sociospatial variation in “day to day” interactions with financial institutions (e.g. traditional banking and financial services) shape the cost of participating in today’s economy (Campbell, Martínez-Jerez, and Tufano 2011; Carruthers and Ariovich 2010). To address these concerns, this study leverages novel survey data from a stratified random sample of 1,364 banks located in neighborhoods (i.e., census tracts) across the United States to investigate racial variation in the costs and fees associated with entry-level checking accounts.

Using linear and logistic regressions, we analyze banks' entry-level checking account costs and fees that may prevent consumers from opening accounts and heighten the possibility of their closing accounts. We combine survey responses with neighborhood data (i.e., census tracts and places) from the 2011-2015 5-year sample of the American Community Survey (ACS), measuring racial makeup and demographic characteristics such as educational attainment and poverty rate. We also incorporate data from Federal Deposit Insurance Corporation (FDIC), National Credit Union Administration (NCUA), and InfoGroupUSA to calculate the geographic density of mainstream and alternative service providers and to measure local market competition.

Our evidence indicates that it is more expensive to open and maintain a checking account in neighborhoods and cities with larger populations of color—especially black and Latinx populations. We find associated increases of $0.32 in the required opening balance, $0.02 in the maintenance fee (p < .10), and $0.03 in the overdraft fee amounts for every one-unit increase in a neighborhood's black population, all else being equal. The corresponding associated increases for a one-unit increase in the neighborhood's Latinx population include $0.77 in the required opening balance, $0.04 in the maintenance fee, and $6.24 in the minimum balance amounts. These amounts are higher when considering how segregation shapes the cost of mainstream banking for different races (Reardon, 2002). For instance, the estimated minimum balance varies substantively across race: $647.75 for whites, $821.73 for blacks, $644.83 for Asians, and $878.96 for Latinx. Moreover, by interacting neighborhood racial makeup with survey respondents' job title, we find evidence that bank employees wield discretionary power in shaping the racialized costs of banking. As compared to customer service representatives, tellers in places with small white populations report significantly higher overdraft fees than tellers in places with large white populations.