No Slack: The Financial Lives of Low-Income Americans

And Banking for All?

54 The use of alternative financial services—such as check cashers, pawnshops, and payday lenders—among low- and moderate-income (LMI) households presents challenges to policymakers seeking to improve financial outcomes among those with few economic resources. That LMI households choose these alternative financial services, in spite of their monetary cost, suggests that they value these services and would find regulations banning them outright harmful. Policies such as “lifeline” banking have aimed to make traditional bank accounts more available, but banks have generally seen these programs as unattractive and have not marketed them widely; as a result, the programs have had low participation rates and have been too limited in scope to increase the number of households with bank accounts (Doyle, Lopez, and Saidenberg 1998; Prescott and Tatar 1999; Washington 2006). Among the demand-side explanations for low bank-account take-up, one possibility is that users of alternative financial services may not be especially sensitive to relative prices, on average, making policy interventions that solely lower the monetary costs of banking services yield low take-up rates.1 Another possibility is that other factors influencing the demand for financial services, such as income or nonmonetary factors (for example, inertia, trust, convenience), are more important or salient to LMI households. Using “holistic” data on household balance And Banking for All? michael s. barr, jane k. dokko, and benjamin j. keys 3 1. An alternative, supply-side explanation for low participation rates may be that banks avoid low-cost products and low-revenue customers. and banking for all? 55 sheets, financial services decisions, and expenditures on financial services from the DAHFS survey, we show that, though annual outlays on financial services are low on average, LMI households incur substantial nonpecuniary costs to obtain and use banking services, suggesting that policies focused solely on lowering the costs of bank accounts might do little to increase bank-account ownership or discourage the use of alternative financial services. Instead, improving account functionality (including convenience) may be more likely to increase demand for banking services. Broadly speaking, our household-level survey data on financial services behavior enable us to fully measure the portfolio of financial services used by LMI households to transact, save, and borrow, which previous work could not do. We obtain financial service usage patterns, estimates of annual financial services outlays, and household demographics, socioeconomic characteristics, and attitudes from this in-person survey. Our estimates of outlays are based on selfreported use of financial services and the self-reported fees paid for these services rather than posted fees alone, which distinguishes this study from previous work that extrapolates the financial burden of financial services based on posted fees. The LMI households in our sample reported spending, on average, about 1 percent of annual income on all financial services, which suggests that many LMI households are able to avoid regular use of the most expensive financial services options. The top spenders, however, take up a disproportionate share of spending . Moreover, the economic burden of financial services for LMI households is not well measured solely by their outlays, particularly if households curtail their use of welfare-enhancing financial services in response to the high posted fees they face. When we compare annual reported outlays on financial services between observably similar “banked” and “unbanked” households, we find that annual outlays for transactional and credit products are higher for banked households than for the unbanked. As we show, banked households are more “economically active” than unbanked households, on average, which corresponds to higher financial service use and total outlays. That is, the banked have higher annual incomes, are more likely to be employed, and have stronger labor-force attachment . Accordingly, they make more transactions and have greater access to credit (for example, through credit cards). Surprisingly, most of the outlays for the median banked household are for alternative financial services. That the banked have higher annual outlays suggests that diminished economic activity of the unbanked may contribute to their lower outlays. In addition, higher spending on financial services may not necessarily be welfare reducing for banked households if it is accompanied by the benefits of financial services associated with greater economic activity. Furthermore, we find that LMI households incur substantial nonpecuniary costs to obtain financial services—such as waiting in line to pay bills, lacking 56 michael s. barr, jane k. dokko, and benjamin j. keys ready mechanisms to save, and burdening friends and family...