The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
EPR
FEDERAL RESERVE BANK OF NEW YORK
ECONOMIC POLICY REVIEW
Financial Inclusion
in the United States:
Measurement,
Determinants, and
Recent Developments
Volume 31, Number 3
September 2025
Matteo Crosignani, Jonathan Kivell,
Daniel Mangrum, Donald Morgan,
Ambika Nair, Joelle Scally,
and Wilbert van der Klaauw
Financial Inclusion
in the United States:
Measurement,
Determinants, and
Recent Developments
Matteo Crosignani, Jonathan Kivell, Daniel Mangrum, Donald Morgan,
Ambika Nair, Joelle Scally, and Wilbert van der Klaauw
OVERVIEW
• New financial technologies
and initiatives aimed at
mitigating hardship during
the pandemic helped expand
access to banking, credit, and
payment services to more consumers and small businesses.
• Despite this progress, many
people and small firms still lack
access to financial services
that would help them better
manage day-to-day finances,
absorb economic shocks, and
build wealth.
• This study takes a measure
of the current state and
evolving landscape of financial
inclusion in the United States,
drawing on survey evidence
and a growing number of academic studies.
• The authors highlight how
differences in survey questions
and sampling methodologies
can cause large differences
in many of these measures
across surveys. They also
explore evidence on the relationship between a person’s
access to financial services
and their ability to weather
financial difficulties.
Federal Reserve Bank of New York
I
n recent years technological change and pandemic-related
policies have helped expand access to banking, credit,
and payment services to more consumers. Digital financial
services in particular have opened the financial system to
more households and small and medium-sized businesses.
Despite this progress, many obstacles remain, especially for
people with low financial security who continue to lack
access to the financial tools needed to manage their day-today finances, absorb economic shocks, and build financial
wealth. In this article, we review the current state and evolving landscape of financial inclusion in the United States. In
doing so, we draw on survey evidence and a growing
number of academic studies. Our review identifies key
opportunities for improving our understanding of the
causes of financial exclusion and of the effectiveness of
alternative efforts to expand financial inclusion. We also
emphasize the need for new measures of financial inclusion
Matteo Crosignani and Donald Morgan are financial research advisors, Jonathan Kivell
is director of community investments, Daniel Mangrum a research economist, Ambika
Nair a community development outreach analyst, Joelle Scally an economic policy
advisor, and Wilbert van der Klaauw an economic research advisor at the Federal
Reserve Bank of New York. Email: matteo.crosignani@ny.frb.org; jonathan.kivell@
ny.frb.org; daniel.mangrum@ny.frb.org; don.morgan@ny.frb.org; ambika.nair@ny.frb.
org; joelle.scally@ny.frb.org; wilbert.vanderklaauw@ny.frb.org.
The views expressed in this article are those of the authors and do not necessarily reflect
the position of the Federal Reserve Bank of New York or the Federal Reserve System. To
view the authors’ disclosure statements, visit https://www.newyorkfed.org/research/
epr/2025/epr_2025_financial-inclusion_crosignani.
https://doi.org/10.59576/epr.31.3.1-49
Economic Policy Review 31, No. 3, September 2025
1
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
that go beyond access and use of specific financial services to how effectively these services
are used to improve financial resiliency and well-being.
The article begins with a definition of financial inclusion and a discussion of its importance.
We review several studies highlighting the consequences of financial exclusion. While exclusion
from the financial system may, in part, be driven by preferences, we highlight several frictions
that cause involuntary and inefficient exclusion, opening the door for government interventions
to improve welfare by expanding financial inclusion.
Next, Section 2 discusses the measurement of financial inclusion across several individual
and household surveys. In addition to bank account ownership, it covers several other measures of financial inclusion, among them access to credit and the use of other bank and
nonbank financial services. We highlight that differences in survey questions and sampling
methodologies can cause large differences in many of these measures across surveys. We then
explore the relationship between whether a person has access to a bank account and their use
of nonbank financial services and their ability to weather financial shocks.
Section 3 examines differences in these financial inclusion measures across the population
by demographics. We also investigate how the unbanked rate varies with education and income
after conditioning on race and ethnicity and describe regional differences in the unbanked rate.
We then explore credit exclusion using the New York Fed Consumer Credit Panel (CCP),
which is a 5 percent representative sample of credit reports from Equifax. Further, we look at
two measures of credit inclusion—the share of the population with a credit report and the
share with a credit card—over time and across neighborhood income. Last, we consider heterogeneity in the use of nonbank financial services across race and ethnicity, age, and income.
In Section 4, we discuss several of the driving factors behind why people are financially
excluded. We start by presenting survey data from the Federal Deposit Insurance Corporation
(FDIC) biennial National Survey of Unbanked and Underbanked Households suggesting that
some households decide to not open a checking account because of high minimum balance
requirements, mistrust of banks, and high and unpredictable fees—with this latter motive
being particularly important for Black and Hispanic households according to data from the
Board of Governors of the Federal Reserve System’s Survey of Consumer Finances (SCF).
Survey data from the New York Fed’s SCE Credit Access Survey also indicate significantly
higher rates of rejection of credit applications for Black and Hispanic applicants relative to
white applicants, with these racial gaps narrowing sharply after controlling for credit scores.
Finally, we contrast government programs, such as the Community Reinvestment Act (CRA),
designed expressly to improve credit access with others, such as usury limits, that can unintentionally limit access, particularly when credit may be most clearly needed, for example, after
weather-related disasters.
Section 5 discusses several recent private, public, and social sector initiatives to expand
financial inclusion, focusing on efforts to make banking more accessible and affordable and on
programs that offer financial education, literacy, and counseling. We review the literature on
their effectiveness.
Section 6 reviews recent market developments, including a tightening of underwriting standards for mortgages and the increased adoption of alternative data and advanced technology
in the underwriting of fintech lending, with a particular focus on unsecured consumer loans.
We discuss evidence from the literature on the impact of this development on financial
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
2
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
inclusion and risks. We also review new evidence from the New York Fed’s Survey of Consumer Expectations (SCE) on the use of “buy now, pay later” loans. Finally, we highlight the
trade-offs in nonbank financial institution lending more broadly between increased financial
inclusion, consumer protection, and financial stability.
In Section 7, we briefly discuss opportunities and areas of further research that will contribute to the improvement of financial inclusion and to the broader goal of building an inclusive
and well-functioning financial system that works for everyone in the United States.
Section 8 concludes the article.
1. What Is Financial Inclusion and Why Does It Matter?
Financial inclusion broadly refers to the availability of opportunities to access and participate
in the financial system, but specific definitions vary somewhat across studies. The World Bank
defines financial inclusion as the condition whereby “individuals and businesses have access
to useful and affordable financial products and services that meet their needs—transactions,
payments, savings, credit, and insurance—delivered in a responsible and sustainable way.”1
Alternatively, financial inclusion could be defined as having access to basic payment and
credit services that are offered at transparent and fair prices commensurate with the cost and
risks of providing the services and that do not aim to exploit consumer naivete and behavioral biases. Similarly, financial exclusion generally refers to individuals and businesses being
unable to access or being prevented from accessing financial services and products that meet
the needs of the customer and are offered under conditions that reflect a fair level of risk.
Participation in the financial system facilitates making payments for goods and services,
securely transferring money, safely storing funds, borrowing money at reasonable rates, and
establishing a credit history. A credit history showing a successful record of repayment, in
turn, implies greater access to a wider and more affordable range of loans. Meanwhile, its
absence could negatively affect an individual’s ability to find work or affordable housing, obtain
approval for small business loans or private student loans, and access affordable insurance. Importantly, access to banking and credit services improves one’s ability to build wealth
through saving and investment (for example, through homeownership and debt-financed
investments in education) and to weather economic difficulties (Ampudia and Ehrmann 2017;
Celerier and Matray 2019; Stein and Yannelis 2020).2
Empirical evidence shows that households with at least some savings are better able to deal
with income volatility and less likely to be evicted or miss mortgage or utility payments
(McKernan, Ratcliffe, and Vinopal 2009; McKernan et al. 2016; Mills and Amick 2010).
Enabling individuals and businesses to pursue economic gain, invest and manage risks, build
wealth, and achieve economic mobility, access to the financial system is important for
U.S. economic growth and resiliency. Furthermore, financial inclusion can help lift people out
of poverty and reduce economic disparities (Beck et al. 2007). The inclusivity of the financial
system also relates to the delivery of tax refunds, government cash transfers, and relief funds
to households and small businesses, such as the Small Business Administration’s Paycheck
Protection Program (PPP) loans and economic impact payments during the COVID-19 pandemic.3 Furthermore, since the direct impact of changes in policy interest rates will generally
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
3
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
be smaller for those who are financially excluded, the extent of financial inclusion can affect
the effectiveness and optimal design of monetary policy (Haughwout, Koşar, and Pinkovskiy
forthcoming; Mehrotra and Yetman 2014).4
The extent of an individual’s participation is determined by demand and supply factors. A
key focus in the study of financial inclusion is the extent to which financial exclusion is due to
frictions in the financial system that prevent banking and financial services from operating
efficiently (Boel and Zimmerman 2022). These inefficiencies reduce overall welfare while contributing to and perpetuating existing inequalities in wealth and opportunities, because the
individuals most likely affected are those with lower incomes and wealth. These frictions in the
financial system could be associated with a lack of market competition, criminal behavior, discrimination, behavioral biases and misconceptions, and information asymmetries. Their
presence provides an argument for federal and state government intervention, similar to the
government’s role in addressing inefficiencies and inequalities in the provision of education
and health care. Such intervention could involve efforts to reduce these frictions and improve
consumer protection. As discussed in Section 4, there is also a role for the Federal Reserve and
other federal banking regulators in addressing financial inclusion through the CRA, by
encouraging federally insured banks to help meet the credit needs of the communities in
which they do business, especially low- and moderate-income (LMI) communities, consistent
with safe and sound operations.5
2. Measuring Financial Inclusion
In this section we discuss the measurement of financial inclusion while drawing on four different national surveys, including the FDIC’s National Survey of Unbanked and Underbanked
Households, the Federal Reserve Board’s SCF, FINRA’s National Financial Capability Study
(NFCS), and the Federal Reserve Bank of Atlanta’s Survey of Consumer Payment Choice
(SCPC), as well as credit report data from the New York Fed’s CCP.
An important aspect of the literature on financial inclusion concerns the multidimensional
nature of its measurement. Rather than a simple binary indicator for whether someone is
financially excluded or not, it is common to use multiple measures to capture access to a range
of financial products. In our description of these different measures, we do not take a position
on their relative importance. Rather we see these measures as complementary, together providing a more holistic characterization of the extent and nature of financial inclusion. For
example, while access to a bank account and its associated saving and transfer options may
affect households’ day-to-day living, access and use of various types of credit could affect one’s
ability to borrow and handle larger purchases of durable goods.
Accordingly, we focus on several dimensions of financial inclusion that have been proposed
in the literature, including bank account ownership, access and use of bank credit, use of bank
and nonbank payment and transfer services, and nonbank credit and insurance, as well as
measures of financial resiliency.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
4
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
2.1 Access and Use of Bank Accounts, Credit, and Payment Services
Owning a bank account provides a way to deposit earnings securely, to pay bills easily, to make
purchases safely, and to save for the future. The FDIC defines an individual as “unbanked” if no
one in the household has a checking or savings account with a bank or credit union. Other
measures of bank account ownership, such as those based on the SCPC, are at the individual,
rather than the household level.
Among those with a bank account, the FDIC further defines as “underbanked” those individuals who are banked but underserved by existing saving, credit, and financial products. The
latter is measured by use in the past twelve months of at least one alternative high-cost
nonbank transaction or a credit product or service disproportionally used by the unbanked to
meet their transaction and credit needs, such as money orders, check cashing, international
remittances, rent-to-own services, pawnshop or payday loans, tax refund anticipation loans,
and auto title loans. Underbanked individuals usually pay high fees for accessing their money
and for transactions while having few opportunities to build savings and assets.
According to the latest 2021 FDIC National Survey of Unbanked and Underbanked Households, there are about 5.9 million households (15.6 million adults) that are unbanked, while
18.7 million households (51.1 million adults) are underbanked. As shown in Table 1, those who
are unbanked accounted for 4.5 percent of U.S. households in 2021, with a further 14.1 percent of
households being underbanked. Similar rates for the unbanked are found in three other surveys:
5.5 percent of households in the Board of Governors’ 2019 SCF, 4.7 percent of respondents in
FINRA’s 2021 NFCS, and 6.5 percent of adults in the 2020 Atlanta Fed’s SCPC.6
Table 1
Measures of Financial Inclusion
Measure (in percent)
Have bank account
FDIC
Household
SCF
Household
95.5
94.5
NFCS Household/
Individual
SCPC
Individual
95.3
93.5
Have checking account
93.0
93.0
92.7
Have savings account
55.9
74.5
75.2
74.5
79.6
79.1a
Underbanked
14.1
Have credit card
71.5
Sources: FDIC, National Survey of Unbanked and Underbanked Households; Board of Governors of the
Federal Reserve System, Survey of Consumer Finances (SCF); FINRA, National Financial Capability Study
(NFCS); Federal Reserve Bank of Atlanta, Survey of Consumer Payment Choice (SCPC).
Notes: The SCF and FDIC measures are at the household level (anyone in the household), while the SCPC
measures are at the individual level. All measures in the NFCS are at the individual level, except bank account
ownership, which is at the household level. SCF data are from the 2019 wave, SCPC data are from the 2020
wave, FDIC data are from the 2021 wave, and NFCS data are from the 2021 wave. Although we cluster
variables together across surveys, estimates may vary between surveys due to other differences in sample
construction and variable definitions.
a
Share with credit or charge card.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
5
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 1
Shares of Households That Are Unbanked and Underbanked
Unbanked
Underbanked
Percent
25
20
15
10
5
0
2009
2011
2013
2015
2017
2019
2021
Source: FDIC National Survey of Unbanked and Underbanked Households, 2009-21.
The rate of unbanked households in the 2021 FDIC survey was the lowest since the survey
began in 2009. Between 2019 and 2021, the unbanked rate dropped by 0.9 percentage point,
contributing to an overall decline of 3.7 percentage points since 2011 (see Chart 1). Improvements in household finances and advances in online and mobile account opening technologies
and bank offerings in recent years likely contributed to the decline in the share of households
who are unbanked. Further, a growing number of banks and credit unions have joined a
national effort to expand banking access since 2006 by offering safe and low-cost Bank On
accounts that meet the Cities for Financial Empowerment Fund’s national account standards.
About one-third of new account openings since 2019 were associated with referrals by the
federal government to Bank On certified accounts (as an alternative to spending cards) for
enabling transfers of economic impact payments, advance child tax credits, and unemployment benefit payments during the pandemic.7 In fact, among the 77.9 percent of recently
banked households that received a government benefit payment, almost half (44.8 percent, or
about 1.9 million households) said that the payment contributed to opening an account.
While the pandemic created unique opportunities and challenges for promoting financial
inclusion, there is some prospect that these changes may be longer lasting. For example, there are
several ongoing efforts at the state level to swap out pay cards for unemployment insurance payments in favor of direct deposit into bank or credit union accounts.8 More broadly, the provision
of safe and affordable bank accounts to consumers receiving income and other government payments appears to be an effective financial inclusion strategy that is likely to further increase bank
account ownership.9
In parallel with the decline in the rate of unbanked households, the share of underbanked
households has declined from 20.0 percent in 2011 to 14.1 percent in 2021 (see Chart 1). There
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
6
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
was also an increase in the share of the adult population owning a credit card, another
common measure of credit access in the United States. Recent surveys yield a range of estimates for the rate of credit card ownership, varying from 70.7 percent to 79.1 percent,
reflecting differences in survey timings and definitions. As shown in Table 1, 71.5 percent of
households in the 2021 FDIC survey and 74.5 percent of households in the 2019 SCF have at
least one member who owns a credit card. Further, 79.1 percent of individuals in the 2020
SCPC own a credit or charge card, and 79.6 percent of individuals in the 2021 NFCS report
owning a credit card. Similarly, the New York Fed CCP shows that 72.3 percent of adults
between ages 21 and 79 held a credit card account in 2021. We can further draw on the CCP to
examine the trend over time: credit card participation fell from 64.9 percent in 2011 to
63.3 percent in 2012. By 2013, only 62.6 percent of Americans held credit card accounts. But
between 2013 and 2021, the share of Americans with credit card accounts grew by nearly
1 percentage point per year. By 2021, 72.3 percent of Americans held credit card accounts.
Besides the increase in bank account and credit card ownership, the past decade has also
seen a sharp decline in the use of some nonbank financial services, such as check cashing.
Despite these improvements in connecting households to the banking sector and declining
rates of unbanked and underbanked households, a significant number of Americans remain
unbanked or underbanked, especially in some segments of the population.
2.2 Having a Credit History and Credit Score
In addition to the ownership of bank accounts and credit cards and the use of banking services, large persistent gaps in credit access and use remain, as captured by an individual’s credit
history. Brevoort, Grimm, and Kambara (2015) distinguishes between two measures of credit
exclusion: being “credit invisible” and “credit unscorable.”10 Credit invisible consumers are
those who have no credit history with one of the major credit bureaus. Those who are credit
unscorable are individuals with a credit history that is too limited to produce credit scores
using traditional credit models, even in cases where the person has a long history of bill payments not traditionally reported to credit bureaus, such as for rent and utilities. Traditional
credit scores are based on borrowing and repayment experiences on most loan products,
including home-based loans, such as mortgages and home equity lines of credit, auto loans and
student loans, and credit cards, which are the most common loan types reported on
credit reports.
Brevoort, Grimm, and Kambara (2015) found that in December 2010 some 11 percent of
adults, or about 26 million Americans, were credit invisible. This CFPB report found that
credit invisibles were also more likely to be unbanked or underbanked. An additional
19 million consumers (8 percent) have unscorable credit files: their files are either thin with
insufficient credit history (9.9 million) or stale with no recent credit history (9.6 million). So,
in total there are about 45 million consumers (almost 20 percent of the U.S. adult population)
who could be denied access to credit because they lack a traditional credit score.
To obtain a more recent perspective on credit inclusion, we draw again on the CCP, which
constitutes a representative sample of anonymized Equifax credit reports. Here, we compute the
share of the adult population without a credit file or credit score in 2011 and in 2021.11 For the
United States overall we find that, in 2021, 10.2 percent of individuals ages 21 to 79 were credit
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
7
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
invisible or unscorable, that is, without a credit score. This is slightly higher than the 9.2 percent
of adults who were in those categories in 2011; the share remained relatively stable during the
intervening years. Note that we estimate a lower rate of credit exclusion compared to the CFPB
report (which estimated this rate to be 19 percent in 2010) for a few key reasons. First, the CFPB’s
computation uses foresight and discards credit files that do not appear four years after the reference year (such that a credit record must appear in 2014 to be counted in 2010). This is a rather
restrictive requirement, which likely drops a significant number of legitimate credit files; we
instead use a two-year lookback window. Second, the minimum scoring criteria for a specific
credit score are highly determinative, and these criteria vary between scores. For example, FICO
scores will be generated with at least a six-month credit history on an open account, whereas VantageScore can be generated with less than a six-month history.12 Our estimates use the Equifax
Risk Score 3.0, which is likely able to generate more scores on credit files with less substantial borrowing histories and thus is more likely to score newer credit users or those with fewer accounts.
Third, we draw our sample for ages 21 to 79, while the CFPB analysis uses ages 20 to 74. While
these differences may lead to a lower estimate of credit exclusion than that of the CFPB, differences in sampling methodology (specifically the treatment of those without a Social Security
number) may cause our estimates to overstate the extent of credit exclusion relative to the CFPB.13
Due to these differences, our estimates and those from the CFPB are not directly comparable on
levels, but the dynamics should move in the same direction.
2.3 Access to Nonbank Services and Financial Resiliency
As discussed earlier, the use of nonbank services is often considered an indication of exclusion
from more traditional bank services. Of the households in the 2021 FDIC survey, 14.1 percent
report using some nonbank financial service (for example, check cashing or a payday loan).
This rate has declined over the past decade, driven in part by increased participation in the
banking system, reduced demand for those services, and the increasing supply of new online
and mobile nonbank products and services, such as prepaid cards and online payment services
such as PayPal, Venmo, and Cash App. While banked households also use nonbank services,
unbanked households may use such services as substitutes for banking services. Using data
from the 2021 FDIC survey, we show in Chart 2 the share of the banked and unbanked populations that have used common nonbank products and services in the last twelve months.
Those who are unbanked are more likely to have used each of the nonbank financial services
than those with a bank account. Money orders and check cashing are the most-used nonbank
financial services by both unbanked and banked individuals; however, unbanked individuals
are roughly four times as likely to have used money orders in the past year and ten times as
likely to have used check cashing services. Since these products and services typically incur
higher service fees than those associated with a checking account, these unbanked individuals
often pay more for the use of these services than they would with a bank account.
Additional insight into access and use of financial services by the unbanked is presented in
Table 2. In addition to ownership of different payment cards and ownership of a checking account
in the past, the table presents statistics on where the unbanked and the banked obtain cash, as
well as several measures of financial resiliency. Note that the unbanked have much lower ownership rates of credit and store cards than the banked and are instead more likely to own prepaid
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
8
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 2
Use of Nonbank Financial Services
Unbanked
Banked
Percent
30
25
20
15
10
5
0
Money
order
Check
cash
Payday
loan
Auto
title loan
Tax
refund
loan
Rentto-own
Pawn
loan
Source: 2021 FDIC National Survey of Unbanked and Underbanked Households.
Note: The chart reports the percentage using nonbank financial services in the last twelve months.
cards and government benefit rewards. Also, a little over half of those currently unbanked had a
checking account in the past, pointing to the dynamic nature of the unbanked status.
Where do those without a bank account get cash most often, and how does that differ for those
with a bank account? Table 2 shows a reduced reliance on ATMs and bank tellers and an increased
dependence on cash wage payments, check cashing stores, and family and friends for getting cash.
Another measure of financial inclusion concerns an individual’s ability to weather financial
shocks. Table 2 shows that 65 percent of adults report being able to obtain financial assistance
of $3,000 from family or friends in case of an emergency. This rate drops to 30 percent for
those who are unbanked, emphasizing their broader vulnerability to financial shocks. Compared with those who are banked, a larger proportion of the unbanked would rely on
borrowing money (from family, friends, a credit card, a car title lender, a pawn shop, a title
loan, or a payday lender), postponing payments, and working more hours or getting a second
job, while fewer would rely on their savings and investments. The unbanked are also considerably less likely to have or live with someone who has life insurance.
3. Who Is Financially Excluded?
Access to and use of financial services differ dramatically across education, income, and demographic groups in the United States. To study differences in access to and use of financial
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
9
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Table 2
Use of Financial Services and Financial Resiliency of the Banked and Unbanked
Shares (in percent)
Unbanked
Banked
All
Credit, cash, and payment methods
Have a credit card
12.2
78.2
74.5
Have a store card
6.0
47.5
45.2
Have a prepaid card
43.1
7.2
9.2
Have a government benefit card
54.5
9.2
11.7
Ever had a checking account
56.2
99.3
96.9
ATM
33.9
58.6
57.0
Bank teller
8.5
19.1
18.4
Check cashing store
5.7
0.4
0.7
Cash back at retail store
8.4
12.5
12.2
Some/All of wage in cash
13.6
2.1
2.9
Family or friend
24.4
5.1
6.3
Other
5.5
2.2
2.5
30.2
67.3
65.2
Borrow money from others
15.0
10.8
11.0
Use savings/investments
12.3
42.1
40.5
Postpone payments
8.0
4.2
4.4
When get cash, where get it most often?
Financial resiliency
Could get an emergency $3,000 from
friends?
How deal with financial emergency?
Cut back
7.3
5.8
5.9
Work more/get an extra job
32.7
24.7
25.2
Any household member has life
insurance?
22.6
61.6
59.4
Source: Survey of Consumer Finances (2019), except for “When get cash, where get it most often” section,
which is from the 2020 Survey of Consumer Payment Choice.
Notes: The “How deal with financial emergency” question was asked only of those who did not run an
income deficit (that is, their spending exceeded their income). The question was worded: “If tomorrow you
experienced a financial emergency that left you unable to pay all of your bills, how would you deal with it?”
The remaining answer category in each column for this question is “ran income deficit.” The wording of the
emergency $3,000 question was as follows: “In an emergency could you (or your {husband/wife/partner}) get
financial assistance of $3,000 or more from any friends or relatives who do not live with you?”
services, we rely on the National Financial Capability Study, which surveys more than 25,000
respondents in each wave, provides a state representative survey sample, and asks survey
respondents a range of questions regarding their ownership of certain financial products and
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
10
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
their use of various financial services. According to the 2021 wave of the NFCS, Asian and
white non-Hispanic respondents were the least likely to be unbanked at 2 percent and
4 percent, respectively, while Black non-Hispanic and Hispanic respondents were the most
likely to be unbanked at 8.7 percent and 5.4 percent, respectively (Chart 3). The rate at which
the share of the unbanked has fallen since 2018 also differed dramatically by race/ethnicity.
Between the 2018 and 2021 waves, the unbanked rate fell by 0.7 percentage point for white
non-Hispanic respondents (from 4.7 percent to 4 percent), 1.6 percentage points for Black
non-Hispanic respondents (from 10.3 percent to 8.7 percent), 3.1 percentage points for Hispanic respondents (from 8.5 percent to 5.4 percent), and 1.7 percentage points for Asian
non-Hispanic respondents (from 3.7 percent to 2 percent).
Chart 3
Share Without a Bank Account in 2018 and 2021 by Race/Ethnicity
2018
2021
Percent
12
10
8
6
4
2
0
White
non-Hispanic
Black
non-Hispanic
Hispanic
any race
Asian
non-Hispanic
Other
non-Hispanic
Source: FINRA, National Financial Capability Study 2018 and 2021.
Those with lower incomes are also less likely to have a bank account. As shown in Chart 4,
the unbanked rate for those with annual household income below $15,000 is more than
15 percent, and the rate decreases dramatically over the income spectrum such that fewer than
2 percent of those with household income above $75,000 are without a bank account. While
race is correlated with income, there are stark differences across racial and ethnic categories
within income bands. For instance, for those earning less than $15,000 per year, each demographic category had an unbanked rate greater than 15 percent except for Asian non-Hispanic
respondents, who were half as likely to be unbanked at a similar income. On the other end of
the income spectrum, the overall unbanked rate for those with incomes of more than $75,000
was 1.3 percent, but Black non-Hispanics in this income range were more than twice as likely
to be unbanked, at 2.9 percent.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
11
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 4
Share Without a Bank Account by Income
All
White
Black
Hispanic
Asian
Other
Percent
20
15
10
5
0
Less than
$15,000
$15,000 to
$50,000
$50,000 to
$75,000
Over $75,000
Source: FINRA, National Financial Capability Study 2021.
A similar trend emerges for educational levels (Chart 5). Those without a high school
diploma are most likely to not have a bank account (nearly 20 percent), but the overall rate
masks significant heterogeneity across demographic groups. Across almost every educational
level, Asian non-Hispanics are the least likely to be unbanked, while Black non-Hispanics are
the most likely.
The unbanked rate also varies across geography in the United States, consistent with the
demographic, income, and educational trends shown above. As shown in Exhibit 1, the rate is
highest in the Southern census region, which is home to the states with the six highest
unbanked rates (Mississippi, Oklahoma, Arkansas, Kentucky, Louisiana, and Tennessee).
Using individual-level credit report data from the CCP, we can create estimates of credit inclusion by various demographics. Here, we examine the share of the population with a credit
score and with a credit card account, charted against the age (Chart 6) and median zip code
income (Chart 7) of the borrower. We find credit score and credit card ownership to be
strongly increasing in age, especially at younger ages. In 2021, by age 30 the share with a credit
score increased to about 90 percent and the share owning a credit card climbed to over
70 percent. Comparing 2021 levels with those in 2011, we find noticeable increases in the share
with a credit score among those in their early 20s and more broad-based increases in credit
card ownership rates among those under age 60.
Similarly, we find that both measures increase with neighborhood median income: in the
higher-income areas, more than 80 percent have a credit card and nearly everyone has a credit
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
12
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 5
Share Without a Bank Account by Education
All
White
Black
Hispanic
Asian
Other
Percent
20
15
10
5
0
No high school
High school
Some college
College degree
Source: FINRA, National Financial Capability Study 2021.
Note: Shares are not reported for cells with too few respondents to produce reliable estimates.
Exhibit 1
Share Without a Bank Account by State
12
8
6
Share unbanked
10
4
Source: FINRA, National Financial Capability Study 2021.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
13
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
score. By contrast, in areas with median income under $30,000, fewer than 50 percent of adults
had access to a credit card and less than 80 percent had a credit score in 2021. Comparing the
2021 and 2011 coverage rates, we find the largest increase in credit card ownership among
those in lower-income areas. This increase could reflect a general increase in risk tolerance and
easing of credit standards by banks and credit card companies, for example, as suggested by
the lower credit card application rejection rates reported during June 2019 to June 2020 in the
SCE Credit Access Survey.14 Yet it may also reflect an improvement in credit scores among
lower-income individuals owing to student loan and other loan forbearance programs during
the pandemic (Fulford, Rush, and Wilson 2021).
Charts 8, 9, and 10 show characteristics of households using specific nonbank financial services associated with the underbanked—services that typically are more costly than services
provided to those owning a bank account. The charts show considerably greater use of money
orders and check cashing services by Black and Hispanic households, as well as by lower-income
and younger individuals, especially those under age 25.15 Further, in Table 3 we jointly control for
fixed effects for each of these groups to test for statistically significant differences in the use of
nonbank financial services. The patterns largely remain when controlling for other demographic
characteristics. Across each of the nonbank financial services, income above $75,000 is the largest
and most precisely estimated negative predictor of use. However, age and race/ethnicity vary in
their explanatory pattern across different services. Age is a (negative) predictor for money order
use and check cashing after conditioning on the other factors but it does not perform well
Chart 6
Share of the 2011 and 2021 Adult Population with a Credit Score or a Credit Card, by Age
With a credit score
2011
2021
With a credit card
2011
2021
Percent
100
80
60
40
20
20
30
40
Age
50
60
Source: New York Fed Consumer Credit Panel/Equifax; American Community Survey.
Note: Credit score is Equifax Risk Score 3.0.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
14
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 7
Share of the 2011 and 2021 Adult Population with a Credit Score or a Credit Card, by
Median Zip Code Income
With a credit score
2011
2021
With a credit card
2011
2021
Percent
100
80
60
40
$20,000
$40,000
$60,000
$80,000
$100,000 $120,000
Median household income (zip code)
$140,000
Source: New York Fed Consumer Credit Panel/Equifax; American Community Survey.
Note: Credit score is Equifax Risk Score 3.0.
Chart 8
Use of Nonbank Financial Services by Race/Ethnicity
White non-Hispanic
Black non-Hispanic
Hispanic
Asian non-Hispanic
Percent
25
20
15
10
5
0
Money order
Check
cashing
Payday loan
Auto title
loan
Pawn shop
loan
Source: 2021 FDIC National Survey of Unbanked and Underbanked Households.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
15
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 9
Use of Nonbank Financial Services by Income
Under $15,000
$15,000-$30,000
$50,000-$75,000
Over $75,000
$30,000-$50,000
Percent
20
18
16
14
12
10
8
6
4
2
0
Money order
Check
cashing
Payday loan
Auto title
loan
Pawn shop
loan
Source: 2021 FDIC National Survey of Unbanked and Underbanked Households.
explaining the other nonbank financial services. Even after controlling for age and income
factors, Black and Hispanic individuals were still more likely to use most of these services.
4. Why Are Households Financially Excluded?
This section explores the reasons households are unbanked or credit constrained/excluded.
The services we are discussing, and their prices, and quantities, are ultimately (or in large
measure) market determined so we naturally frame the discussion in market terms: demand
and supply. In some cases, the distinction is fairly clear. For example, a sizable fraction of
unbanked households cite distrust of banks or a desire for privacy as important reasons why
they go without (that is, do not demand) a bank account. In other cases, the distinction is less
clear. For example, individuals may be discouraged from opening a bank account or applying
for a loan because they expect to be rejected. Similarly, the limited presence of nearby bank
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
16
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 10
Use of Nonbank Financial Services by Age
15-24
25-34
35-44
45-54
55-64
>65
Percent
16
14
12
10
8
6
4
2
0
Money order
Check cashing
Payday loan
Auto title loan
Pawn shop
loan
Source: 2021 FDIC National Survey of Unbanked and Underbanked Households.
branches may be due to a lack of sufficient local demand. The supply forces we consider
include the providers’ decisions (or algorithms) and government policies and programs such as
the Community Reinvestment Act and, lately, state usury limits.
4.1 Demand- and Supply-Side Factors
The FDIC’s biennial survey of unbanked and underbanked households provides a glimpse into
the reasons why some households operate without a checking account. Chart 11 reports the
share of respondents who listed each reason and the share who listed a given reason as their
main one. High minimum balance requirements rank at the top, followed by a desire for
privacy and a mistrust of banks. The latter two are strictly demand-side, while the former is
partly supply-side as well (since banks set minimum balance requirements and the maintenance fees incurred when balances fall below the minimum). High and unpredictable fees also
rank near the top. This reason may reflect overdraft fees, although, surprisingly, the FDIC
report never mentions overdrafts. Bank location ranks relatively low as a main reason for being
unbanked, although some 15 percent of respondents listed it as a factor.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
17
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Table 3
Multivariate Regressions on the Household Use of Nonbank Financial Services
Variable
25 to 34 years old
35 to 44 years old
45 to 54 years old
55 to 64 years old
65 years or more
Black
Hispanic
Asian
Other
$15,000 to $30,000
$30,000 to $50,000
Money Order
Check Cashing
Payday Loan Auto Title Loan Pawn Shop Loan
-1.486
-1.124
-0.262
-0.397
0.113
(1.431)
(0.911)
(0.571)
(0.494)
(0.489)
-1.318
-0.219
-0.228
-0.718
0.088
(1.410)
(0.913)
(0.560)
(0.478)
(0.475)
-2.704*
-0.815
-0.369
-0.502
0.456
(1.395)
(0.900)
(0.553)
(0.484)
(0.499)
-2.448*
-1.963**
-0.836
-0.732
-0.110
(1.380)
(0.875)
(0.530)
(0.471)
(0.474)
-6.133***
-2.967***
-1.170**
-1.046**
-0.696
(1.332)
(0.848)
(0.520)
(0.460)
(0.457)
13.308***
3.166***
1.384***
-0.023
0.218
(0.896)
(0.515)
(0.345)
(0.209)
(0.253)
9.129***
2.860***
0.726***
0.325
0.640**
(0.746)
(0.448)
(0.269)
(0.218)
(0.256)
0.975
-0.218
-0.554***
-0.116
-0.318
(0.790)
(0.445)
(0.146)
(0.238)
(0.237)
6.789***
2.777***
0.486
1.050
2.112***
(1.561)
(1.004)
(0.519)
(0.705)
(0.781)
-2.455**
-0.502
-0.091
0.078
0.043
(1.103)
(0.739)
(0.364)
(0.250)
(0.416)
-6.009***
-2.789***
-0.120
0.581**
-0.672*
(1.021)
(0.655)
(0.357)
(0.267)
(0.370)
-8.930***
-4.115***
-0.166
0.096
-1.466***
(0.989)
(0.627)
(0.354)
(0.239)
(0.341)
-11.617***
-4.973***
-0.839***
-0.016
-1.824***
(0.922)
(0.597)
(0.306)
(0.215)
(0.326)
Observations
30,434
30,434
30,434
30,434
30,434
R-squared
0.062
0.023
0.006
0.002
0.009
Baseline mean
9.720
3.181
1.091
0.897
0.957
$50,000 to $75,000
At least $75,000
Source: Authors’ calculations, using data from the 2021 FDIC National Survey of Unbanked and Underbanked
Households.
Notes: Each column in the table above presents the regression results of the dependent variable listed in the
column header on a set of fixed effects for age groups, race/ethnicity, and annual household income. The
dependent variable is measured as zero or 100 for ease of interpreting the coefficients in percentage point
terms. The omitted groups for each category are ages 18 to 24 years, white (non-Hispanic), and income less
than $15,000, respectively. The estimates presented are the average differences in predicted use of nonbank
financial services from the baseline group conditional on all included regressors. Observations are weighted using
household sample weights. Robust standard errors are presented in parentheses below each estimate. .
* denotes p<0.10, ** denotes p<0.05, and *** denotes p<0.001.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
18
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 11
Unbanked Households’ Reasons for Not Having a Bank Account, 2021 (in Percent)
Cited
Main
Don’t have enough money to meet
minimum balance requirements
Avoiding a bank gives more privacy
Bank locations are inconvenient
Other reason
Did not select a reason
29.5
6.0
27.3
1.5
19.2
2.4
15.4
4.4
Problems with past banking or credit history
Don’t have personal identification
required to open an account
33.0
13.2
Bank account fees are too high
Banks do not offer needed products and services
34.1
8.4
Don’t trust banks
Bank account fees are too unpredictable
40.1
21.7
5.3
2.7
13.6
11.6
17.7
21.5
16.8
16.8
Source: Figure ES.3 in the 2021 FDIC National Survey of Unbanked and Underbanked Households.
Reproduced with permission.
“Problems with past banking or credit history” also ranks low. This is surprising given evidence that unpaid overdrafts are a primary reason for involuntary account closures (Campbell,
Martínez-Jerez, and Tufano 2012). A history of unpaid overdrafts and bounced checks can lead
to a low ChexSystems Consumer Score that prevents households from opening an account at
any depository institution.16 Table 4, from the Federal Reserve’s SCF, breaks out the main
reasons households do not have a checking account by race of survey respondent. The first two
reasons are roughly equal across racial groups, but high service charges rank higher for Black
and Hispanic households than for others. Inconvenient location or hours ranks relatively low,
consistent with the FDIC survey.
A study by Rhine and Greene (2013) identifies “shocks” that lead households to lose or
close bank accounts. They find that families are significantly more likely to become unbanked
following a decline in family income, a job loss, or the loss of health insurance. Given those
factors, race and ethnicity, education, and marital or housing status also predict whether families are banked or not. Campbell, Martínez-Jerez, and Tufano (2012) use data from
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
19
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Table 4
Most Important Reason for Not Having a Checking Account, 2019
Share of Households Ranking Each Reason as Most Important
Overall
White
Black
Hispanic
Don’t write enough checks to make it
worthwhile
28.1
27.7
27.6
31.8
Do not like dealing with banks
22.9
21
25.7
23.8
Service charges are too high
12.6
9
16.5
14.1
Not enough money
7.9
5.5
8
10.6
Don’t need/want a checking account
7.1
9.2
4.7
5.9
Minimum balance is too high
4.5
3.3
5.4
4.7
Credit problems; Bankruptcy; Don’t meet
depositor qualifications for checking
account
3.6
4.9
3.3
0
Can’t manage/balance a checking account
3.2
4.1
2.2
2.1
Inconvenient hours or location
2.2
3.8
1.1
1.5
Someone else writes checks for me
2.1
4.3
1.1
0
Concern about overdraft fees
2.1
2.8
0.8
1.6
Checkbook has been/could be lost/stolen
1.6
1.8
1.8
1.1
1
1.6
0.9
0
0.6
0.8
0.8
0
Haven’t gotten around to it
Have other account with checking privileges
Source: Survey of Consumer Finances (2019).
ChexSystems, the debit bureau used by banks to track depositor behavior, to study involuntary
bank account closures due to excessive overdrafts. They find that involuntary closure rates per
capita are more frequent in U.S. counties with a larger fraction of single mothers, lower educational levels, lower wealth, and higher unemployment. Closures are higher in communities
with high rates of property crime and low electoral participation. Financial structure also
matters: counties having more competitive banking markets and more multi-market banks
have higher account closure rates as do counties that allow payday lending.17
Another dimension of financial inclusion is credit access. Household credit supply has
expanded dramatically in the last several decades, both in quantity and variety, yet concern
remains that some households are unable to borrow at terms that accurately reflect their risk
profile. Table 5 summarizes responses to this question from the SCF in 2019: “Have you been
turned down for credit in the past five years?” The first three columns show that while white, Black,
and Hispanic respondents were about equally likely to have applied for credit in the past five years,
Blacks and Hispanics were about twice as likely to report being rejected.18 Black respondents were
also more likely to report being granted less credit than requested. The fifth and sixth columns
compare responses for households in the lowest income quintile with those of higher-income
groups. Low-income households were considerably less likely to apply for credit and more likely to
be turned down. These disparities go only so far in identifying credit exclusion, however, because
average credit scores also vary by race and income (Bhutta, Hizmo, and Ringo 2022).
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
20
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Table 5
Credit Rejection Rates by Race and Income
Reponses to “Have You Been Turned Down for Credit in the Past 5 Years?” (in Percent)
Race/Ethnicity
Reason
Low-Income
White
Black
Hispanic
Other
No
Yes
NA, did not apply
47.5
46.6
46.6
40.3
43.4
60.9
Yes, turned down
7.6
14.5
15.2
9.5
8.7
12.1
Yes, did not receive
as much credit as
requested
1.4
3.1
0.9
1.6
1.6
1.5
No
43.5
35.9
37.3
48.6
46.4
25.5
Total
100.0
100.0
100.0
100.0
100.0
100.0
Source: Survey of Consumer Finances (2019).
To investigate credit exclusion further, we draw on data from the SCE Credit Access Survey.
Since 2014, this survey has been fielded every four months and collects data on the credit
experiences and expectations of households. In addition to asking about rejection of applications (experienced for any type of loan), the survey also asks respondents whether they did not
apply for credit because they expected their application to be rejected (so-called “discouraged
borrowers”). The latter question provides a measure of latent credit demand.
We conducted a series of regression analyses relating the rejection rates among applicants and
discouraged borrowing rates among all adults to an applicant’s race and household income. The
analysis finds economically and statistically significant higher rates of rejection of credit applications for Black and Hispanic applicants relative to white applicants (See top left panel of Chart 12)
and a monotonically declining relationship with household income (bottom left panel of Chart 12).
Adding controls for employment status, homeownership, and education (gold bars) only slightly
weakens these relationships. In contrast, controlling for credit scores (gray bars) causes the racial
gaps to narrow sharply: only the higher rejection rate for Hispanics of 3 percentage points remains
statistically significant, as indicated by the confidence intervals in the chart. While controlling for
credit scores similarly reduces household income gaps (relative to those with incomes under
$20,000) in discouraged borrower rates, a strong negative income effect on discouraged borrowing
remains (bottom right panel of Chart 12). These findings indicate that much, but not all, of the race
and income gaps in credit access can be accounted for by differences in credit scores.19 While the
literature on financial inclusion has mostly focused on the role of household demand, a growing
body of research highlights the role of supply-side factors. Next, we discuss some forces that drive
credit supply away from underserved communities, focusing on the role played by discrimination
and drawing on the recent evidence on the allocation of pandemic relief funds.
The U.S. Fair Housing Act and the Equal Credit Opportunity Act prohibit lenders from
making credit determinations that disparately affect minority borrowers if those determinations are based on characteristics unrelated to creditworthiness. In a recent analysis of Home
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
21
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Chart 12
Race and Income Gaps in Loan Application Rejection and Discouraged Borrowing Rates (in
Percent)
No controls
Controls for employment status, homeownership, and education
Same controls plus credit scores
Race/Ethnicity Gaps
Rejection Rate
Discouraged Borrowing Rate
Percent
Percent
10
10
8
8
6
6
4
4
2
2
0
0
-2
-2
-4
Black
Asian
Hispanic
-4
Black
Asian
Hispanic
Household Income Gaps
Rejection Rate
Discouraged Borrowing Rate
Percent
15
0
1
ov 502
e
20 00
0
et
c.
00
-1
Ab
10
0-
75
30
20
-6
50
30
Thousands of dollars
0
-45
-7
5
-40
-45
60
-35
-40
0
-35
-6
-30
-5
-25
-30
50
-20
-25
-4
-20
40
-15
-3
0
-10
-15
0
60
-7
5
75
-1
00
10
015
Ab 15 0
ov 02
e
20 0 0
0
et
c.
-10
-4
0
40
-5
0
0
-5
20
-3
0
0
-5
0
Percent
Thousands of dollars
Source: Authors’ calculations based on data from the SCE Credit Access Survey, 2014-23.
Notes: Each bar represents the estimated rejection rate or discouraged borrowing rate for each racial/ethnic
group relative to white respondents, which are the reference group, after controlling for different sets of
borrower characteristics. Also shown are 95 percent confidence intervals around the estimated gaps.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
22
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Mortgage Disclosure Act data, Bhutta, Hizmo, and Ringo (2022) find little evidence of discrimination in largely automated mortgage approval decisions. However, Bartlett et al. (2022) show
that risk-equivalent Hispanic and Black borrowers pay significantly higher interest rates than
other borrowers on loans securitized by government-sponsored enterprises (GSE) and loans
insured by the Federal Housing Administration (FHA), particularly in high-minority-share
neighborhoods. Using a methodology that considers that borrowers are offered potentially different “menus” of interest rate and upfront fee options, Willen and Zhang (2023) find
discrimination by race in conforming mortgage lending in the United States.20 These findings
are particularly worrying for underrepresented groups, given the ongoing redistribution of
mortgage credit from small and mid-size loans to large loans that started in 2011 (D’Acunto
and Rossi 2022). Gerardi, Willen, and Zhang (2020) examine the extent to which the racial
interest rate gap on outstanding mortgages can be explained by a lower rate of refinancing or
moving during periods of falling interest rates. They find that Black and Hispanic borrowers
exploit such opportunities at a much lower rate than white borrowers, in part because they face
much higher obstacles to refinancing, even conditional on credit scores and income.
Lenders might be reluctant to lend to underserved communities because of barriers
unrelated to the risk of these borrowers. For example, Blattner and Nelson (2021) document
that underserved groups have statistically noisier (that is, less informative) credit scores,
primarily because of thin credit files, which affects their access to credit. In a counterfactual
exercise, the authors show that equalizing the precision of credit scores can shrink disparities in loan approval rates for underserved groups by around 50 percent. Ambrose, Conklin,
and Lopez (2021) find that racial discrimination plays an important role in the context of
mortgage lending. They document that minorities pay between 3 and 5 percent more in fees
than similarly qualified white borrowers when obtaining a loan through the same white
broker, with this premium changing based on the race of the broker. Bhutta, Fuster, and
Hizmo (2020) use survey data to demonstrate the important roles of financial knowledge
and shopping and negotiation behavior in explaining variation in mortgage rates across
consumers, with many homebuyers overpaying for their mortgage. They find that overpayment trends decrease when market rates rise, consistent with increased shopping when
rates are high.
Analysis of the allocation of PPP funds has provided more insights into supply-side
factors of financial inclusion. Wang and Zhang (2020) document that minority communities
have less access to programs (in this case small business loans) that rely on the financial
sector to supply the funds. Those living in banking deserts lacking large lenders with an
established relationship with the Small Business Administration are especially disadvantaged. The authors estimate that a 10 percentage point increase in the Black share of the
population in a zip code is correlated with a 1 percent lower likelihood of that zip code
having any PPP-enrolled lender and a 4.1 percent decrease in the number of branches with
lenders enrolled in the PPP program (conditional on having at least one such lender). This
reduced access to enrolled lenders in turn was shown to lead to significantly lower take-up
of PPP loans in heavily Black neighborhoods, particularly in more rural areas where this
disparity is most salient.21
Finally, expanding the research beyond the demand for and supply of bank accounts and
credit, to financial instruments that facilitate saving, Karlan, Ratan, and Zinman (2014) in a
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
23
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
cross-country study document important barriers to savings experienced disproportionately
by the poor. They find that market frictions, including transaction costs, lack of trust, and regulatory barriers, often hinder the supply of savings products. In addition, social constraints,
information and knowledge gaps, and/or behavioral biases may lead to suboptimal saving even
among those with access. The resulting undersaving can have important welfare consequences:
more variable consumption, low resilience to shocks, and forgone profitable investments in
human and business capital.22 We return to the importance of financial literacy and behavioral
factors in Section 5 below.
4.2 Government Policy
Usury limits
Usury limits—government caps on loan interest rates—can also limit credit access, particularly
for riskier borrowers. Lenders are known to charge riskier borrowers high interest rates to
compensate for expected default costs. If that “breakeven” rate exceeds the usury limit, lenders
will deny (“ration”) loans to riskier borrowers. This is not just an academic concern, since
usury limits are increasingly common in the United States, as shown in Exhibit 2. All but seven
states now cap interest rates, with limits ranging from 17 percent in Arkansas to 59 percent in
Mississippi. While these double-digit limits may seem high in the context of secured loans
such as mortgages, they are below the rates that payday and installment lenders charge for
their unsecured loans.
Evidence confirms the credit-constraining effect of usury limits. Bolen, Elliehausen, and
Miller (2023) find that after Illinois imposed a 36 percent APR cap in 2021, the number of
online installment loans made to subprime borrowers declined 44 percent relative to Missouri,
an adjacent state without any usury limit. Relaxing usury limits has been found to increase
credit access. Using data from Prosper.com (an online lender), Rigbi (2013) finds more online
installment lending in states with higher limits and increased lending when limits were raised.
Dlugosz, Melzer, and Morgan (2021) find spillovers from credit access to deposit inclusion.
They find that when caps on overdraft fees were lifted, banks raised fees but also allowed
(covered) more overdrafts and lowered minimum balance requirements on checking accounts.
The result was more credit and more low-income households with checking accounts.
While credit rationing under usury limits is borne out by theory and evidence, proponents
may see reduced access to credit as precisely the point: that is, to protect borrowers from
high-cost credit that may do more harm than good. For example, borrowers with behavioral
biases may overborrow or overpay for credit. The evidence on whether high-cost loans benefit
users is mixed. While we can say with some confidence that usury limits hinder credit inclusion,
we cannot necessarily say that the excluded borrowers are worse off. By limiting access to credit,
usury limits may also leave already vulnerable higher-risk households less able to cope with
weather disasters (see Box 1).
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
24
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Exhibit 2
Usury (APR) Limit on a Two-Year, $2,000 Installment Loan
Caps APR between 17% and 36% (32 states and D.C.)
Caps APR between 37% and 60% (11 states)
Caps APR at more than 60% (no states)
No cap other than unconscionability (no cap*) (5 states)
No cap (2 states)
29%
WA
36%
36%
ND
MT
36%
OR
NOCAP*
31%
MN
NOCAP*
WI
SD
WY
NV
25%
NO CAP*
UT
CO
CA
41%
KS
54%
36%
AZ
30%
36%
IL
38%
OK
NM
36%
TX
AK
24%
PA
31%
40%
OH
IN
NO CAP
33%
VA
KY
59%
NO
CAP*
AL
36%
NH
MA 24%
RI 24%
30%
NJ
CT 36%
DE NO CAP
MD 30%
DC 24%
35%
43% TN
NO
CAP*
17%
AR
50%
WV
42%
MO
MS
31%
NY
MI
IA
NE
31%
25%
36%
30%
40%
ME
VT
36%
ID
30%
21%
33%
32%
NC
SC
GA
38%
LA
37%
31%
FL
HI
Source: National Consumer Loan Center, “Predatory Installment Lending in the States: How Well Do the
States Protect Consumers Against High-Cost Installment Loans?” (2024). Reproduced with permission.
Box 1
Weathering the Storm: Natural Disasters and Financial Inclusion
Research shows that access to credit, even high-cost credit, helps households cope with weather
disasters. A study by Morse (2011) finds that households in California with better access to payday
loans are less likely to face foreclosure after weather disasters. Tiurina (2022) shows that living in
“credit deserts” worsens the financial impact of weather disasters. She looks at Arkansas, where
interest rates on consumer loans are capped at 17 percent, the lowest usury limit in the United
States. Because of the low limit, payday lenders and consumer finance companies do not operate
in Arkansas, though they operate in bordering states. She finds that Arkansas borrowers living
near border states are less likely to fall behind on mortgage debt and have a lower drop in their
credit scores after weather disasters compared with borrowers in the center of Arkansas. A third
study by Dobridge (2018) finds that access to payday loans mitigates declines in spending on food,
mortgage payments, and home repairs following periods of temporary financial distress owing to
extreme weather events. These authors all point out that their findings are agnostic on whether
high-cost credit is good for households in “normal” times, but collectively, they make a strong case
that it can help households cope with increasingly extreme weather events.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
25
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
CRA and Other Regulations
Several studies have examined the impact of government regulations and restrictions on
lower-income borrowers’ access to financial services. Celerier and Matray (2019) show that the
Riegle-Neal Act, which deregulated interstate banking, led to an increase in bank branches,
which, in turn, increased the rate of bank account ownership among low-income households.
They found that having a bank account leads to higher accumulation of interest-bearing assets,
more investment in durable assets, greater access to credit, and a lower likelihood of facing
financial difficulties.
A large body of research analyzes the effect of the Community Reinvestment Act, a law
enacted in the 1970s aimed at improving credit access for disadvantaged communities. The
CRA requires depository institutions to serve the needs of the communities in which they
operate by assigning a CRA-compliance grade largely based on loans extended to low- to
moderate-income (LMI) census tracts. Evidence on the impact of the CRA on lending, as
reviewed by Conway, Glaser, and Plosser (2023), has been somewhat mixed, and overall, the
evidence of a positive impact of CRA eligibility on credit access in LMI communities appears
weak. Conway, Glaser, and Plosser (2023) employ three different identification strategies to
evaluate the impact of the CRA on consumers’ access to credit since 1999. These methods are
generally based on a comparison of outcomes for applicants in census tracts close to the
CRA-eligibility threshold (with the threshold defined in terms of a census tract’s median
household income). They conclude that the CRA has had no meaningful effect on consumer
borrowing in several debt categories or on credit outcomes such as credit scores, bankruptcy,
and delinquency. They show that for mortgage debt, the overall null effect masks a substitution
of mortgages in CRA-eligible areas from nonbanks to banks, with banks purchasing mortgages
at a higher rate in CRA areas, which they then sell on to GSEs. The latter finding is also consistent with recent evidence by Brevoort (2022), who similarly finds evidence of banks complying
with the CRA by purchasing loans originated by other lenders, rather than originating loans
themselves. Thus, the CRA induces some degree of substitution in origination and purchases
while not affecting overall originations and lending.
In contrast, using a different identification strategy that allows estimating the effect in areas
more broadly, including areas further from the CRA-eligibility threshold, Ringo (2023) finds
that the CRA was effective in increasing the availability of mortgage loans to LMI borrowers in
targeted census tracts. It increased mortgage lending to LMI borrowers in those tracts by 2 to
4 percent, with the effect being even stronger among the lowest-income borrowers. He also
found that there was no effect on LMI lending by nonbanks (which are not subject to the CRA).
The CRA recognizes and supports activities that promote financial inclusion of low-income
people and communities through partnerships and loans and investments in Community
Development Financial Institutions (CDFIs) and Minority Depository Institutions (MDIs). To
help inform the overall impact of the CRA more research is needed to examine the effectiveness of different types of CDFIs and MDIs in promoting community development, economic
empowerment, and financial inclusion in underserved communities.
Fuster, Plosser, and Vickery (2021) examine how regulatory oversight by the CFPB, which
is dedicated to overseeing and enforcing consumer financial protection laws, affects the supply
of mortgage credit and other aspects of bank behavior. They find that CFPB oversight leads to
a reduction in FHA lending, a market where mortgage borrowers are typically lower income
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
26
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
and often first-time homebuyers. At the same time, oversight appears to have reduced mortgage foreclosures.
Agarwal et al. (2015) examine the impact of the 2009 Credit Card Accountability Responsibility and Disclosure (CARD) Act in the United States, which imposed regulations on fees
(over-limit fees and late fees) on credit cards. They find that regulatory limits on credit card
fees reduced overall borrowing costs for consumers by an annualized 1.7 percent of average
daily balances, with a decline of more than 5.5 percent for consumers with FICO scores below
660. While the CARD Act led to a substantial reduction in borrowing costs, it did so without
an offsetting increase in interest charges or a reduction in overall access to credit. In investigating the distributional impacts, Nelson (forthcoming) found that a subset of borrowers faced
higher prices due to the act and exited from borrowing; this was especially true for subprime
borrowers.
Debbaut, Ghent, and Kudlyak (2016) study a different aspect of the CARD Act: restrictions
on credit to individuals under age 21. They find that following the passage of the law, individuals under age 21 were 8 percentage points (15 percent) less likely to have a credit card, to have
fewer cards, and, conditional on having a card at all, were 35 percent more likely to have a
co-signed card. They found young individuals from higher-income neighborhoods, those
whose parents had higher credit scores, and those whose parents did not have a serious default
to be most affected.
More recently, the CFPB has introduced new regulation that further limits credit card fees.
The impact of this regulation on credit supply to subprime borrowers and on financial inclusion is an important topic for future research.
5. What Is Being Done to Increase Financial Inclusion?
In this section we first discuss recent efforts by the private, public, and social sectors to further
financial inclusion. This will be followed, in Section 6, by an assessment of how financial inclusion
has been affected by recent market developments, with a particular focus on fintech lending.
5.1 Initiatives to Make Banking More Accessible and Affordable
There have been several significant efforts to increase access to affordable bank accounts to
unbanked and underbanked populations. Those efforts include allowing opening deposits under
$25, providing access to free online and mobile services, and limiting overdraft fees. Access to a
basic and affordable transaction account at an insured institution is seen as a key first step to
financial inclusion, providing a safe place to save, conduct basic financial transactions, build a
credit history, access credit on favorable terms, and achieve financial security. For example, to
benefit from instant payment services through FedNow requires access to a bank account.23
An important ongoing effort is the Cities for Financial Empowerment Fund’s Bank On
accounts. Since 2008, the Bank On National Account Standards (2023-24) have helped financial
institutions connect consumers to safe and affordable bank accounts, centered on safety, cost, and
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
27
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
transactional ability.24 According to the National Consumer Law Center, there are currently more
than 375 Bank On nationally certified accounts. Those accounts are offered by banks and credit
unions representing over 60 percent of the U.S. consumer deposit market, according to the FDIC.
More than half (53 percent) of all U.S. bank branches offer a Bank On certified account.
The latest Bank On National Data Hub report by the Federal Reserve Bank of St. Louis finds
that, to date, more than 14.1 million Bank On–certified accounts have been opened by consumers in 85 percent of U.S. zip codes.25 Further, in 2021 almost 80 percent of the accounts
were opened by customers who were new to the financial institution, indicating that Bank On
accounts are bringing new customers into financial institutions. Working through coalitions,
Bank On continues to push more banks to adopt its 2023-24 national account standards and
grant programs. In addition, Bank On pilot programs are focusing on helping high schoolers
connect to the financial mainstream and helping unbanked individuals open bank accounts
when they start a new job.
As previously discussed in Section 2, federal and state governments have recently relied on
the Bank On certification to refer households to safe bank accounts as an alternative to spending cards for pandemic-related stimulus, advances on child tax credits, and unemployment
insurance benefit payments.26 Such newly opened Bank On accounts represent 34.9 percent of
all new account openings since March 2020. In fact, among the 77.9 percent of recently banked
households that received a government benefit payment, 44.8 percent (about 1.9 million
households) reported that the government payment contributed to their opening of an
account. This experience illustrates the potential of further integrating account opening into
large-scale government payment programs.27
Over the years an increasing number of initiatives similar to Bank On have been implemented to provide safe and affordable bank accounts. In 2011, in a one-year pilot study, the
FDIC introduced Model Safe Accounts, designed to evaluate the feasibility of financial institutions offering safe, low-cost transactional and savings accounts that are responsive to the needs
of underserved consumers. Nine financial institutions participated in the pilot by offering electronic deposit accounts that followed the FDIC Model Safe Accounts Template.28 The findings
of the one-year pilot showed that such accounts offered to LMI consumers performed equally
well as or better than other transaction and savings deposit accounts offered by the pilot banks.
At the end of the pilot, 81 percent of transaction accounts and 95 percent of savings accounts
remained open. Furthermore, most of the pilot institutions reported that the cost of offering
Safe Accounts was roughly the same if not lower than other types because the pilot accounts
did not involve costs related to paper checks. While the pilot program showed that opportunities exist, many participating financial institutions experienced challenges in marketing and
advertising the Safe Accounts and in establishing a presence in new markets where providers
of nonbank financial services compete for customers.
Credit unions, owned and controlled by their members, have also traditionally played an
important role in offering more affordable financial products and services. There are a number
of additional local and national initiatives led by financial institutions that focus on expanding
access to banking services, including Wells Fargo’s ten-year Banking Inclusion Initiative,29
JPMorgan Chase’s $30 billion Racial Equity Commitment, and the BankBlackUSA movement,
which is focused on helping underserved communities and the general public connect to
Black-owned financial institutions in their local community.30
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
28
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Another important recent development is the increased access to online bank accounts provided by digital banks (usually the online-only division of a traditional bank or credit union),
neobanks (nontraditional banks that operate online without physical locations), and other
fintech firms providing mobile or online banking services, such as Aspiration, Chime, Current,
SoFi, and Varo.31 We will discuss the role of fintechs in increasing financial inclusion through
the provision of banking and credit services more broadly in the next section.
Data on Bank On and neobank account openings suggest a general increase in bank
account ownership, consistent with the latest FDIC survey indicating a meaningful reduction
in the shares of households that are unbanked and underbanked. However, relatively little is
known about how long, how frequently, and for what purpose these accounts are used, and
how access to these accounts has affected overall financial inclusion and the well-being of
those previously unbanked.
5.2 Financial Education, Literacy, and Counseling
The research literature on financial literacy and financial education is extensive. Reviews of this
work point to the importance of financial literacy for consumer financial behavior and individual welfare (Lusardi and Mitchell 2014). For example, evidence from the New York Fed’s
Survey of Consumer Expectations shows financial literacy and numeracy to be positively
related to homeownership, household income, and credit scores, and negatively related to loan
delinquency and loan application rejection rates (see Table 6).32
Even conditional on demographic characteristics, including education, income, and credit
scores, those with higher financial literacy and numeracy experience higher rates of loan application approval and homeownership and lower rates of loan delinquency.
While not necessarily causal, the relationship between financial knowledge and financial
behavior and outcomes suggests a potential role for public support for financial education.
However, there is at best conflicting evidence that financial education leads to improved economic outcomes either through increasing financial literacy directly or otherwise (Hastings,
Madrian, and Skimmyhorn 2013). Some studies find financial education to be successful in
increasing financial knowledge and, to a smaller extent, financial behavior, especially for school
students (Kaiser and Menkhoff 2020; Consumer Financial Protection Bureau 2019; Brown et al.
2016; Urban et al. 2020; Kaiser et al. 2022; Mangrum 2022). However, others more broadly consider the evidence on the effectiveness of financial education to be weak, especially in light of
the significant financial resources and time they require, and, in the case of school courses, the
opportunity cost of other coursework (Fernandes, Lynch, and Netemeyer 2014; Willis 2011).
For those who see education as largely ineffective, the literature suggests several alternative approaches to improve financial outcomes, including regulations and interventions
aimed at reducing biases, product information complexity, and decision-making costs,
and providing incentives and nudges to steer consumers toward improved financial decision making (Thaler and Sunstein 2008). For example, Madrian and Shea (2001) and
Beshears et al. (2008) show evidence of the positive impacts of default and opt-out rules
on retirement savings outcomes.
In recent years, we have witnessed a proliferation of private initiatives to help improve financial well-being through financial counseling and financial education, often through online
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
29
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Table 6
Financial Literacy/Numeracy and Financial Outcomes
Low Financial Literacy/
Numeracy
High Financial Literacy/
Numeracy
Homeownership rate
62%
77%
Household income > $75,000
20%
48%
Credit score > 720
39%
68%
Credit score > 760
25%
48%
Late payment (30+ days past due)
11%
5%
Credit application rejected in past year
11%
7%
Source: New York Fed SCE Credit Access Survey, 2014-23.
Notes: Financial literacy/numeracy is based on the number of correct answers to five questions gauging
respondents’ literacy and numeracy skills. Respondents are partitioned into high (four or five correct) or low
literacy/numeracy (three or fewer correct answers). In the total sample, 71.8 percent of all respondents were
categorized as having high numeracy. The credit application rejection data are for those with at least one
credit application. Calculations are based on a total of 34,680 observations.
education tools, to learn about budgeting, online banking, and saving to build credit. Examples
include Operation HOPE; Prosperity Now; America Saves; Ready, Set, Bank; World of Money;
and MoneySmart.33 The Cities for Financial Empowerment (CFE) Fund helps local governments
embed financial empowerment strategies into their work, which includes the provision of professional financial counseling as a free public service to their residents.34 The Financial Health
Network unites entities across business, government, technology, and academia in a shared
mission to improve financial health. Some organizations are also doing important work in the
financial inclusion area.35 For example, companies are starting to collect better data on financial
inclusion to help strengthen organizational capacity, build awareness, and improve offerings.36
Overall, little is known about the effectiveness of these interventions, since few have been
subjected to rigorous scientific evaluation through a randomized controlled trial or
quasi-experimental design.37 Additional research would be valuable for determining what
types of programs are most beneficial and for whom.
6. Impact of Recent Market Developments
The market for consumer lending has seen a number of significant changes in recent years.
Since the financial crisis, there has been a tightening of underwriting standards for mortgages,
partly in response to regulatory changes. This trend has translated into a reduction in mortgage lending to borrowers with low credit scores, leading to an almost complete disappearance
of subprime mortgage lending.38 This reduction has disproportionately affected Black and Hispanic borrowers (Bhutta and Ringo 2016), with the largest banks particularly reducing their
lending to LMI households (Bhutta, Laufer, and Ringo 2017).
Another significant change has been the adoption of more advanced technology in consumer
lending and the rise of fintech companies.39 Before assessing the impact of this market development
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
30
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
on financial inclusion, we first discuss what fintechs do and specifically their use of alternative data
and new underwriting methods. While fintech activities span consumer, small business, and real
estate loans, we focus here on unsecured consumer loans, given their greater relevance for LMI
communities. By providing greater access to such loans, which are frequently used by borrowers to
pay off their credit card balances or to consolidate their debt, fintech companies have the potential
for increasing financial inclusion.40 On the one hand, the growth of fintech companies sustains the
aggregate supply of financial services by providing new access to all households, including marginalized ones. On the other hand, marginalized borrowers might lack the financial sophistication
needed to take advantage of these new offerings, limiting the reach of fintech companies.
6.1 The Current Market for Fintech Unsecured Personal Loans
Baker (2017) developed a framework that categorizes the existing market for a fintech products ecosystem that is most relevant to LMI individuals. These categories are:
• Digital Credit Access/Cost Improvement Lenders: Companies that lend money to consumers for short-term needs but seek to do so in a less costly/better structured manner
than the traditional short-term, small-dollar credit (STSDC) system and/or lend at a
lower cost to people with thin-file credit profiles.
• Digital Credit Builders: Companies that primarily seek to help consumers improve
their credit score to be able to access lower-cost credit in the future, by lending credit
to borrowers short term and/or helping lenders identify creditworthy consumers using
alternative data and underwriting.
• Digital Cash Flow Management Applications: Companies that provide online services
to guide consumers toward financially healthy behaviors and outcomes.
• Alternative Digital Banks: Companies that, although technically not banks, create
mobile transaction deposit solutions for consumers through bank partners.
• Earned Wage Access/Expense Variability Management: Companies that provide
employers with liquidity solutions, often called payroll loans, to help employees manage
the variability of expenses through early income access or other means.
• Digital Savings: Companies providing mobile apps that facilitate consumer savings for
liquidity management and other purposes.
6.2 Alternative Data and Underwriting for Fintech Unsecured
Personal Loans
A distinctive feature of most fintechs is their use of alternative data for underwriting loans.
This use is motivated by a view that traditional underwriting methods based on credit scores
have not always captured the full picture of a borrower’s ability to repay. Alternative data and
underwriting methods can provide additional insight into a borrower’s creditworthiness,
enabling lenders to offer loans to borrowers who may have been rejected based on traditional
underwriting methods. They may also improve default rates on loan products and allow
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
31
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
lenders to precisely target creditworthy borrowers while widening their consumer base more
precisely to those borrowers who have a no-file or thin-file credit history.
Alternative data are defined as any data that can be used to enhance consumer lending decisions and are not traditionally included in the credit databases of the national credit reporting
agencies. Examples of alternative data and underwriting methods include:
1. Income and employment information;
2. Consumer banking information;
3. Utilities and telecom payment history;
4. Rental payments;
5. Inquiries and payment records to specialty lenders (such as payday lenders);
6. Peer-to-peer (P2P) lending history; and
7. Social media profiles and personal information.
These additional information sources can open the door to lower-interest credit options for
those who are low and moderate income as well as those who have been underserved by traditional credit options. However, it is important to consider the ways in which alternative data
are being used in lending decisions, given that LMI individuals are more likely to face acute
financial shocks that may be more pronounced in alternative data such as utilities, telecom,
and rental payment histories and cash flow information. Some uses of alternative data may
widen existing socioeconomic disparities in access to credit, by potentially amplifying the
effects of the financial shocks that underserved groups are vulnerable to and making it more
difficult to improve their credit standing. This potential issue raises the question of whether
alternative data and underwriting in practice are used in a way that promotes inclusion for the
underserved or instead widens existing gaps in access to credit.
6.3 Underwriting Based on Cash Flow and Utility, Telecom, and
Rental Payments
Cash flow information focuses on borrowers’ cash flow, looking at income and expenses to
identify creditworthy applicants. Cash-flow–based underwriting provides a more detailed and
timelier picture of how applicants manage their finances than traditional credit reports.41 This
approach to underwriting is especially useful for borrowers who may have a low credit score but
a steady income stream. As discussed earlier, some 95 percent of American households have
bank or prepaid accounts, and account records are increasingly easy to access electronically.
Utility, telecom, and rental (UTR) payment history provides direct information about
whether consumers have the financial capacity to take on additional expenses or, in the case of
rent, substitute expenses. Payment history also reflects consumers’ general propensity to meet
continuing obligations. These records are particularly relevant to LMI individuals because
payment histories allow lenders to expand credit to renters, who are about seven times more
likely than homeowners to lack sufficient credit history. This population includes disproportionate numbers of Black and Hispanic consumers, low-income households, and young adults,
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
32
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
as does the general population of renters. Studies suggest that a sizable number of no-file and
thin-file consumers (the credit invisible and unscorable) would be able to satisfy many lenders’
minimum credit score thresholds if their UTR payments were considered.
In terms of UTR data’s predictive ability, research is limited at this point in time, but stakeholder interviews and studies show that UTR used in combination with other data can improve
predictive accuracy and expand access. However, UTR data are not widely available in credit
bureau files today and are often left unaccounted for in scoring. Newer FICO scoring models
(including FICO Score 10) are testing the implementation of small samples of UTR data.42
Another feature of fintech lending besides the use of alternative data is its reliance on
greater automation of underwriting. The net effect of lenders’ adoption of technology on financial inclusion is unclear. On the one hand, automation reduces processing costs, thus allowing
banks to provide smaller loans and reach a population of borrowers located in a larger geographical area. The increasing prevalence of automation in lending decisions might also reduce
the footprint of human bias in lending decisions. On the other hand, automation and the
adoption of technology might increase the existing disparities in access to credit. For example,
automation may be associated with a reduction in help provided to applicants in completing
their applications. If such help was particularly valuable for minority applicants, then more use
of automation may reduce lending to minority households and businesses. Some evidence to
date points to an overall net positive effect of automation and technology adoption on financial
inclusion for business owners (Howell et al. 2024).
6.4 Buy Now, Pay Later (BNPL)
Since the mid-2010s, there has been a steep increase in the use of buy now, pay later loans, a
type of short-term financing that allows consumers to make purchases and pay for them over
time, usually with no interest. BNPL loans can be a low-cost substitute for other credit products, imposing significantly lower direct financial costs on consumers than legacy credit
products. While this type of lending has the potential to expand credit access through the use
of alternative underwriting methods, the evidence to date indicates that BNPL loans go almost
entirely to those with a bank account and usually a credit score.
BNPL lenders typically require borrowers to pay a share of the purchase price (usually
25 percent) as a down payment at checkout, usually requiring borrowers to use their existing
debit or credit card for the down payment, which can also be used for autopayment of subsequent installments. This provides instantaneous verification that the applicant has sufficient
checking account funds (debit card) or available credit (credit card) to cover the down
payment. As part of the underwriting of new and returning applicants, several lenders also pull
the applicant’s credit report and review the prior repayment history with that lender.
New evidence from the June 2023 Survey of Consumer Expectations shows that about
64 percent of respondents have ever been offered a BNPL payment option, while 19 percent
have used it as a payment method in the past year (29 percent of those who were ever offered
the option).43 Among users, 77 percent made installment repayments using a debit card, bank
account, or bank check; 10 percent used a credit card; 6 percent used a prepaid card; and
8 percent used a payment service such as Venmo. Thus, indeed, BNPL loan users tend to be
banked and usually have a credit history.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
33
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
We find the use of BNPL loans in the population to be somewhat higher for females,
renters, and individuals without a college degree, and to be monotonically decreasing in
income (see Chart 13).44 Overall, although 19 percent of individuals used BNPL, its usage is
noticeably higher for those with credit scores below 620 (35 percent) and those who were
thirty days or more delinquent at some point during the past year (33 percent). Those who
applied for some other type of credit over the past year (an indicator of higher credit demand)
were generally also more likely to report using BNPL in the past year, compared with those
who did not apply for credit. Among those who applied for credit, BNPL use was particularly
high for those who reported a credit application rejection over the past year (42 percent).45
Despite usage being fairly broad-based, with significant take-up also among more highly
educated and higher-income respondents, we find that those with lower credit scores and
greater unmet credit needs make up a disproportionate share of all BNPL users. Indeed,
32.7 percent of BNPL users had either a credit score below 620 or a credit application rejected
or were delinquent on a loan over the past year; however, this group makes up just 16.6 percent
of the sample. Furthermore, we find that BNPL users overall are more financially fragile, as
measured by the average likelihood of being able to come up with $2,000 in the next month in
case of an emergency. This likelihood is 66 percent across all respondents and for respondents
ever offered the BNPL option, but it is only 52 percent among those who reported using BNPL
over the past year. BNPL users are also less likely to rely on savings when facing a financial
shock. While 68 percent of individuals overall would rely on savings to come up with $2,000,
only 42 percent of BNPL users would. Instead, they report that they are more likely to rely on
borrowing from friends or family or through a bank or credit card. The disproportionate share
of the already financially fragile among BNPL users raises questions about the resilience of
BNPL lending and its performance following an adverse economic shock.
BNPL usage may reflect supply as well as demand factors. For example, those with lower
incomes and credit scores may find interest-free BNPL financing more attractive and affordable but may have greater access to such loans due to where (at what retailers) and to whom
BNPL is offered. BNPL usage could also capture individuals with a lower credit score and past
loan delinquency that were the result of, rather than the reason for, BNPL use (debt overextension), although most BNPL loans are not reported to credit bureaus. To further examine the
extent to which these patterns reflect demand, rather than “targeting,” we related a respondent’s background and circumstances to his or her expected year-ahead BNPL use. We again
found a much higher reported average probability of using BNPL over the next year among
those with lower credit scores and those who had a credit application rejected over the
past year.46 These results, therefore, suggest that while BNPL may not benefit those who are
unbanked or those who are credit invisible or unscorable, BNPL loans appear particularly
attractive to those with unmet credit needs and limited access to credit. Because they represent
an additional attractive source of credit to help borrowers smooth consumption and manage
their debt payments at lower cost, BNPL loans may thus expand financial inclusion, especially
to those with low credit scores.
At the same time, our evidence substantiates to some extent a concern expressed by some
BNPL critics that BNPL may attract consumers who already have financial difficulties and are
struggling to pay their existing bills and debt. One concern in this regard is that the increase in
retail revenue generated by adding a BNPL payment option, the higher average amount per
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
34
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
0
5
10
15
20
25
30
Note: BNPL is buy now, pay later.
Source: Survey of Consumer Expectations, June 2023, October 2023, and February 2024.
30+ day delinquency in past year
No
Yes
Credit applications in past year
Accepted
Rejected
Did not apply
Credit score
Below 620
620-719
720-760
Above 760
Home ownership
Own
Rent
Census region
Midwest
Northeast
South
West
Household income
<$30,000
$30,000-$50,000
$50,000-$75,000
$75,000-$150,000
>$150,000
Education
High school
Some college
College
Age
Age < 40
40 ≤ Age < 60
Age ≥ 60
Gender
Female
Male
Share Using BNPL over the Past Year by Respondent Characteristics
Chart 13
35
40
45
50
55
60
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
35
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
transaction spent by a BNPL lender’s shoppers, and a higher level of repeat usage suggest that
many BNPL consumers may not be simply shifting their existing purchases to a new payment
platform but may instead be spending (and borrowing) more than they otherwise would.
Berg et al. (2024) find causal evidence for this, showing that when BNPL is available, customers spend 20 percent more, with low-creditworthiness customers being most responsive to the
availability of BNPL.
The CFPB has raised two important potential risks associated with current BNPL lending
practices.47 The first concerns the risk of overextension, whereby frequent BNPL usage over
time may lead to excessive debt accumulation and affect a consumer’s ability to meet
non-BNPL obligations. Second, there is a risk that borrowers may take out several loans within
a short time frame with different BNPL lenders. Because most BNPL lenders do not currently
furnish data to the major credit reporting agencies, both BNPL and other lenders are unaware
of the borrower’s current liabilities when making a decision to originate new loans. In both
cases we would expect BNPL users to exhibit greater signs of financial stress, such as higher
indebtedness, higher delinquencies, and lower credit scores. We found evidence of this exact
pattern above, but more data and analysis are necessary to investigate the extent to which the
greater financial stress and need for credit among current BNPL users are the result of previous
BNPL borrowing.
Thus, while our findings indicate that BNPL services enjoy considerable broad-based interest, appear to fill a gap in the credit market, and expand credit access and financial inclusion,
as noted above, more data and analysis are needed to determine the overall effects of BNPL
borrowing on financial well-being, especially over the course of the business cycle.
6.5 Existing Research on the Impacts of Fintech Products on
Financial Inclusion
While much remains to be learned about the short- and longer-run impacts of fintech consumer lending on credit access, financial inclusion, loan pricing, and performance, several
recent studies have begun to analyze fintech consumer lending practices. FinRegLab, a nonprofit organization focused on promoting innovation in financial services, has conducted
research showing that cash flow–based underwriting can significantly improve the accuracy of
credit risk assessment, especially for borrowers with limited credit histories.48
Jagtiani and Lemieux (2017) and Jagtiani and Lemieux (2018a, 2018b) using account-level
data from LendingClub found that its lending served areas that may be underserved by traditional banks, such as in highly concentrated markets and areas that have fewer bank branches
per capita. They also found that the portion of LendingClub loans increased in areas where the
local economy is not performing well. They further explored the key roles of additional information in expanding credit access to creditworthy borrowers whom banks may not be serving
and demonstrated that this credit seems to be “appropriately” risk-priced. Using alternative
data appears to have allowed some borrowers who would be classified as subprime by traditional criteria to be slotted into better loan grades and to access lower-priced credit. These
findings point to the potential advantages that financial technology can provide consumers.49
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
36
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Dolson and Jagtiani (2021) used a rich data set of credit offers from Mintel, in conjunction
with credit information from TransUnion and other consumer credit data from the New York
Fed Consumer Credit Panel, to compare similar credit offers that were made by banks, fintech
firms, and other nonbank lenders. The authors found that fintech firms were more likely than
banks to offer mortgage credit to consumers with lower income and lower credit scores and
those who have been denied credit in the recent past. Fintechs were also more likely than
banks to offer personal loans to consumers who had filed for bankruptcy, who, as a result, were
also more likely to receive credit card offers. For both personal loans and mortgage loans,
fintech firms were found to be more likely than other lenders to reach out and offer credit to
nonprime consumers. At the same time, they found that consumers with higher balances and
more credit cards were also more likely to get fintech offers.
Di Maggio, Ratnadiwakara, and Carmichael (2022) also examined whether cash-flow-based
underwriting with alternative data results in broader credit access using anonymized administrative data from Upstart. They found that using cash flow data resulted in a lower probability
of rejection and lower interest rates for those who were approved while leading to better loan
performance and lower default rates. They attribute these results to cash flow data providing a
more comprehensive view of a borrower’s financial situation, enabling lenders to identify borrowers who are likely to repay their loans.
While this research appears to indicate that using alternative data for assessing borrowers’
creditworthiness results in broader credit access for those who have trouble accessing credit
and lower interest rates for nonprime consumers, other evidence suggests that automation and
technology adoption might maintain or even increase the existing disparities in access
to credit.
Fuster et al. (2022) show that the adoption of statistical technology for credit screening
creates “losers” and “winners,” with the former being Black and Hispanic borrowers and the
winners being non-Hispanic White and Asian borrowers. Statistical methods efficiently
capture relationships between observable borrower characteristics (for example, income and
credit score) and default outcomes, but they also manage to uncover characteristics such as
race and gender that are not usually part of a credit application, effectively de-anonymizing
borrowers and potentially further penalizing underrepresented groups. This insight is consistent with evidence of racial discrimination in algorithms (Arnold, Dobbie, and Hull 2021).
Finally, in the context of GSE-securitized mortgage loans, Bartlett et al. (2022) find that the
higher interest rate (unrelated to risk) paid by minority borrowers and their higher rejection
rates were similar for fintech lenders and non-fintech lenders. One potential explanation for
this similarity is that by using historical patterns within the data to make predictions about the
future, machine learning systems risk repeating biases that were present in past lending data.
Given the growing market share of fintech lenders, these forces are set to become even more
important in the short to medium term. For a review of fintech lending, see Berg, Fuster, and
Puri (2022).
Fintech and other nonbank financial institutions generally are subject to less stringent regulation and supervision and, as evidenced in some of the studies cited above, may pose risks to
more vulnerable financial consumers. The resulting trade-offs between increased financial
inclusion, consumer financial protection, and financial stability are an important area for
further study.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
37
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
7. Where Are the Opportunities to Address
Financial Inclusion?
Over the past decade we have seen significant improvements in bank account ownership and
in access to various forms of credit, including credit cards. These gains not only reflect the
effectiveness of various government interventions but also capture market developments
through which lending to underserved communities has increased. This increased engagement
by private institutions in equitable lending and in improved outreach in underserved communities is, in part, driven by their clients and stakeholders, who increasingly value these
initiatives. For example, Crosignani and Le (2023) document that banks differ in their propensity to lend to minorities based on their stakeholders’ aversion to inequality. Despite these
improvements, a significant number of Americans remain unbanked or underbanked, especially in some segments of the population, such as lower-income and minority households.
Our review of the literature suggests a number of areas that offer opportunities to help improve
understanding of financial inclusion and ways to address it.
An important aspect of the current state of financial inclusion in the United States is the
rapidly evolving nature of consumer lending, driven by changing technology and alternative
credit products, as well as changes in consumers’ need for financial services. Our review of the
literature points to a general need for better data and analysis to uncover the underlying
reasons for financial exclusion and to determine the effectiveness of alternative solutions. We
identify five broad areas of opportunity.
First, there is a need for more qualitative research based on in-depth interviews to understand the current-day challenges faced by households in accessing financial services.50 Much of
what is known about such challenges is based on research done at a time that preceded the rise
in online banking and the digital lending revolution.
Second, with respect to the more traditional measure of bank account ownership, more
information and analysis are needed to determine the extent to which gains in bank account
ownership through Bank On and other initiatives are sustained over time in terms of subsequent use and general bank account ownership. Furthermore, there is a need to develop
broader measures of financial inclusion that go beyond being banked or having a credit score
or credit card to include having access to nonbank financial services and social support networks and insurance, as well as digital access, skills, and literacy. Such measures would provide
a broader assessment of the economic fragility and resiliency of households and of their financial health. Moreover, we acknowledge an ambition to move beyond expanding financial
inclusion per se to optimizing the utility of financial inclusion, that is, the benefits and outcomes that financial inclusion creates for its beneficiaries (Sirtaine 2023). Of particular
importance is the critical role financial services can play in helping vulnerable populations
navigate the risks and shocks associated with natural disasters.
Third, despite important new academic research on this topic, important gaps remain in
our understanding of the overall impact of the use of alternative data and technology, including the role of BNPL lenders and fintech lending more broadly, on the financial well-being of
households. Efforts to close these gaps should include an assessment of the risks associated
with the use of such new products and with the way lenders use alternative information about
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
38
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
households. Inconsistent reporting of loans to credit bureaus and inadequate oversight of nontraditional lenders represent potential future risks to financial stability and household
well-being.
Fourth, despite a large number of federal and private initiatives to improve access to bank
accounts and to improve financial literacy through financial education and financial counseling, little is known about the effectiveness of these interventions, since few of them have been
subjected to rigorous scientific evaluation through a randomized controlled trial or
quasi-experimental design. Additional research would be valuable for determining what types
of programs are most beneficial and for whom.
Finally, there is a need for further research on the potential effectiveness of educational and
financial literacy and counseling initiatives and efforts to improve the quantity and quality of
credit report data in reducing inequality in access and costs due to gaps in credit scores, especially for those with a limited credit history or limited knowledge of their credit history (see,
for example, Homonoff, O’Brien, and Sussman [2019]; Blattner and Nelson [2021]).
8. Conclusion
Financial inclusion matters for wealth building, for households’ ability to weather income volatility and economic difficulties, consumption, employment, and for economic mobility.
Broader access to and participation in the financial system is therefore important for U.S. economic growth and resiliency. Access and participation also affect the effectiveness and optimal
design of monetary policy. So central banks need to understand and monitor financial inclusion. In this article, we highlighted several opportunities for advancing our understanding
through more qualitative research, the development of broader measures of financial inclusion,
additional research into the overall impact of the use of alternative data and technology on the
financial well-being of households, and more rigorous assessments of the effectiveness of
public and private initiatives to expand access to credit and to improve financial literacy
through financial education and financial counseling. This knowledge in turn will improve our
understanding of the potential trade-offs between increased financial inclusion, consumer
financial protection, and financial stability.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
39
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Notes
Acknowledgments: The authors benefited from helpful comments from Felix Aidala, Karla Brom, Erica Bucchieri, Raji
Chakrabarti, Julian di Giovanni, Jack Gutt, Daphne Ha, Otho Kerr, Krista Schmidt, and two anonymous referees. They
would also like to thank Eileen Goodman, Madison Lloyd, and Jessica Orsini for their support. They are grateful to Cathie
Mahon, Inclusiv; Lisa Servon, University of Pennsylvania; Jonathan Mintz, Cities for Financial Empowerment Fund; Wole
Coaxum, CEO, MoCaFi; and Nichole Davis and Haidee Cabusora, New York City Office of Financial Empowerment, for
sharing their topical expertise.
1
https://www.worldbank.org/en/topic/financialinclusion/overview.
2
An improved ability to generate and transfer wealth also enables greater financing of small businesses and
entrepreneurs by friends and family.
3
describe the importance of a preexisting banking relationship for the successful receipt
of PPP funds by small businesses, while Holtzblatt and Karpman (2020) document delays and higher nonreceipt of
economic impact payments for those without a bank account and without internet access,
https://libertystreeteconomics.newyorkfed.org/2020/05/where-have-the-paycheck-protection-loans-gone-so-far/.
4
A smaller effect through the interest rate transmission channel does not necessarily imply that those financially
excluded will be less affected by monetary policy, after accounting for indirect effects through the labor market,
inflation, and other macroeconomic conditions.
5
Federal Reserve Board - Community Reinvestment Act (CRA), https://www.federalreserve.gov/
consumerscommunities/cra_about.htm.
6
There are some differences across surveys in the measurements of the share of unbanked and underbanked.
For example, while the FDIC and SCF surveys measure access to bank accounts at the household level, the NFCS
and SCPC measures are at the individual respondent level. More specifically, the NFCS measures the share of
individuals where none of the adults in the household have either a checking or savings/money market account.
It then defines as underbanked all those in a banked household who personally used alternative financial
services more than two times in the past five years. The SCF measures the share of respondents who do not have a
checking or savings/money market account, and, of those who do, it measures the share that used at least one of
five financial services over the past twelve months.
7
2021 FDIC National Survey of Unbanked and Underbanked Households. https://www.fdic.gov/analysis/householdsurvey/index.html.
8
See, for example, what the State of Maryland has done in this regard. Illinois and Connecticut are moving to
implement the same types of payments. https://www.dllr.state.md.us/whatsnews/uiddtransition.shtml.
9
A related issue concerns the potential effectiveness of strategies to capture remittances through bank accounts for
improving financial inclusion in emerging markets (see Naceur, Chami, and Trabelsi [2020]).
10
See “Data Point: Credit Invisibles,” a CFPB report by Brevoort, Grimm, and Kambara (2015).
11
As with the CFPB analysis, we divide the number of individuals with credit reports, scores, and specific accounts by
the total adult population in the zip code from the 2019 American Community Survey five-year estimates.
12
Information on minimum scoring criteria for FICO and VantageScore can be found here: https://assets.equifax.
com/marketing/US/assets/vantagescore-40-product-sheet.pdf, https://www.myfico.com/credit-education/faq/
scores/fico-score-requirements.
13
The New York Fed CCP draws the sample primarily using Social Security numbers (SSN); thus, those without
an SSN do not enter our sample. However, the CFPB draws its samples from the universe of credit reports and
thus can contain individuals without SSNs.
14
SCE Credit Access Survey, https://www.newyorkfed.org/microeconomics/sce/credit-access#/experiences-creditapplications9.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
40
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Notes (Continued)
15
These findings are largely consistent with those of Rhine, Greene, and Toussaint-Comeau (2006) based on a 2000
survey on the use of check-cashing establishments in the Chicago metropolitan area.
16
ChexSystems is a verification service and consumer reporting agency used by a large majority of commercial
banks and credit unions to screen applicants for checking and savings accounts.
17
As explanations for these findings, Campbell, Martínez-Jerez, and Tufano (2012) posit that: more competitive
banking markets might be ones in which banks reach deeper for new customers, leading to less financially
secure customers and more closure; that local banks may have better customer knowledge or may be more likely to
show forbearance than a multi-market bank; and that payday lending may imply greater financial difficulties of
customers due to payday loans representing a debt trap.
18
Crook (2001) reports similar findings showing comparable application rates, but higher rejection rates
among Black respondents during the 1980s and early 1990s.
19
Controlling for additional loan and borrower characteristics may further reduce the race and income
gaps. The importance of a borrower’s credit score in denials of applications to extract home equity was
similarly demonstrated by Conklin, Gerardi, and Lambie-Hanson (2022).
20
There is also evidence of discrimination in lending outside the United States. For example, Dobbie et al. (2021)
find significant bias against both immigrant and older loan applicants in the United Kingdom.
21
More specifically, they find that doubling the number of PPP-enrolled lenders in a zip code would increase
take-up by between 11.8 and 12.6 percentage points.
22
Barr (2012) discusses the importance of financial slack for lower-income households in weathering shocks, in the
context of the Detroit area during the 2007-09 financial crisis. The analysis considers the failure of regulators to
protect consumers and investors and how households and firms took on risks they did not fully understand.
The author calls for a policy of “behaviorally informed regulation” as an approach for fostering resilience.
23
FedNow is a new service for instant payments launched by the Federal Reserve in July 2023 to help make everyday
payments fast and convenient for American households. With FedNow, households can make last-minute or
emergency payments helping them avoid late and overdraft fees and better manage their credit score. Similarly,
those employed as gig workers, such as rideshare drivers, can get paid immediately for the work they perform,
reducing their need for short-term and costly financing. See https://explore.fednow.org/.
24
https://joinbankon.org; Bank-On-National-Account-Standards-2023-2024.pdf.
25
Bank On National Data Hub, available at https://www.stlouisfed.org/community-development/bank-on-nationaldata-hub/bank-on-report-2021.
26
For making unemployment insurance payments, the State of Maryland has replaced pay cards with direct
deposit into bank or credit union accounts, emphasizing Bank On certification status as a trusted referral (https://
www.dllr.state.md.us/whatsnews/uiddtransition.shtml) Other states, including Illinois and Connecticut, are moving
to do the same.
27
Recently, the Summer Jobs Connect program does the same with Summer Youth Employment Programs (SYEPs)
across the country. More information is available at https://cfefund.org/project/summer-jobs-connect/.
28
https://www.fdic.gov/consumers/template/template.pdf.
29
https://www.wellsfargo.com/jump/enterprise/banking-inclusion-initiative/.
30
https://www.jpmorganchase.com/news-stories/jpmorgan-chase-releases-rec-audit-report; https://bankblackusa.
org/.
31
https://www.aspiration.com/; https://www.chime.com/; https://current.com/; https://www.sofi.com/; https://www.
varomoney.com/.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
41
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
Notes (Continued)
32
SCE Credit Access Survey questionnaire, https://www.newyorkfed.org/medialibrary/Interactives/sce/sce/
downloads/glossary/FRBNY-SCE-ChartGlossary.pdf.
33
operationhope.org; prosperitynow.org; americasaves.org; readysetbank.org; worldofmoney.org; www.fdic.gov/
resources/consumers/money-smart/index.html.
34
cfefund.org.
35
finhealthnetwork.org.
36
Courtney Davis. “Accelerating Toward Greater Financial Inclusion,” Deloitte Insights. September 28, 2021. https://
www2.deloitte.com/us/en/insights/industry/financial-services/alternative-data-innovation-financial-inclusion.html.
37
A notable exception is the study by Theodos, Plerhoples Stacy, and Daniels (2018), which presented evidence based
on randomized controlled trials in New York City and Miami showing the effectiveness of financial coaching
in increasing savings and credit scores and reducing aggregate debt and debt delinquency.
38
New York Fed, Quarterly Report on Household Debt and Credit, 2023:Q1. https://www.newyorkfed.org/
medialibrary/interactives/householdcredit/data/pdf/HHDC_2023Q1.
39
Buchak et al. (2018) find that regulatory constraints on traditional banks have contributed to the increase in the
shadow bank and fintech market shares in residential mortgage origination, with technology also contributing
to this shift. They find that fintech lenders charge a premium of 14 to 16 basis points and appear to provide
convenience rather than cost savings to borrowers.
40
In this section, the discussion of the market for unsecured personal loans, the role of fintechs, and the use of
such loans by LMI individuals, is based in part on Nair and Beiseitov (2023).
41
See FinRegLab (2019), https://finreglab.org/wp-content/uploads/2023/12/FinRegLab_2019-07-25_ResearchReport_The-Use-of-Cash-Flow-Data-in-Underwriting-Credit_Empirical-Research-Findings.pdf.
42
See Thompson, Cochran, and Stegman (2022) for further discussion of the use of UTR data in underwriting credit.
43
Felix Aidala, Daniel Mangrum, and Wilbert van der Klaauw, “Who Uses ‘Buy Now, Pay Later’?,” Federal Reserve
Bank of New York Liberty Street Economics, September 26, 2023. https://libertystreeteconomics.newyorkfed.
org/2023/09/who-uses-buy-now-pay-later/.
44
These demographic patterns in BNPL use are largely consistent with those reported in “Consumer Use of Buy
Now, Pay Later: Insights from the CFPB Making Ends Meet Survey” (Shupe, Li, and Fulford 2023). https://files.
consumerfinance.gov/f/documents/cfpb_consumer-use-of-buy-now-pay-later_2023-03.pdf.
45
When we control for all covariates jointly in a multivariate regression, the higher rates for those with low credit
scores or a recent credit application rejection remain highly statistically significant.
46
Controlling for all covariates jointly in a regression does not alter these findings.
47
Consumer Financial Protection Bureau. “Buy Now, Pay Later: Market Trends and Consumer Impacts,” September
2022. https://s3.amazonaws.com/files.consumerfinance.gov/f/documents/cfpb_buy-now-pay-later-market-trendsconsumer-impacts_report_2022-09.pdf.
48
FinRegLab (2019), https://finreglab.org/research/the-use-of-cash-flow-data-in-underwriting-credit-empiricalresearch-findings/.
49
Similarly, Danisewicz and Elard (2023) find that the availability of marketplace fintech lending of personal
loans reduces personal bankruptcy filings, which they argue points to the importance of these loans as a source of
external funding in the face of income shocks.
50
For an example of such an approach for studying how low- and middle-income Americans earn, spend, borrow,
and save, see the “U.S. Financial Diaries” discussed by Morduch and Schneider (2017).
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
42
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
References
Agarwal, S., S. Chomsisengphet, N. Mahoney, and J. Stroebel. 2015. “Regulating Consumer Financial
Products: Evidence from Credit Cards.” Quarterly Journal of Economics 130, no. 1
(February): 111-64.
Ambrose, B. W., J. N. Conklin, and L. A. Lopez. 2021. “Does Borrower and Broker Race Affect the Cost of
Mortgage Credit?” Review of Financial Studies 34, no. 2 (February): 790-826.
Ampudia, M., and M. Ehrmann. 2017. “Financial Inclusion: What’s It Worth?” European Central Bank
Working Paper no. 1990, January. https://doi.org/10.2866/979715.
Arnold, D., W. Dobbie, and P. Hull. 2021. “Measuring Racial Discrimination in Algorithms.” AEA Papers
and Proceedings 111: 49-54.
Baker, T. H. 2017. “FinTech Alternatives to Short-Term Small-Dollar Credit: Helping Low-Income
Working Families Escape the High-Cost Lending Trap.” Harvard Kennedy School M-RCBG Associate
Working Paper no. 75. https://www.hks.harvard.edu/centers/mrcbg/publications/awp/awp75.
Barr, M. S. 2012. No Slack: The Financial Lives of Low-Income Americans. Washington, D.C.:
Brookings Institution Press. https://muse.jhu.edu/book/29038.
Bartlett, R., A. Morse, R. Stanton, and N. Wallace. 2022. “Consumer-Lending Discrimination in the
FinTech Era.” Journal of Financial Economics 143, no. 1 (January): 30-56.
Beck, T., A. Demirgüç-Kunt, and R. Levine. 2007. “Finance, Inequality, and the Poor.” Journal of
Economic Growth 12, no. 1 (March): 27-49.
Berg, T., A. Fuster, and M. Puri. 2022. “FinTech Lending.” Annual Review of Financial
Economics 14: 187-207.
Berg, T., V. Burg, J. Keil, and M. Puri. 2024. “The Economics of ‘Buy Now, Pay Later’: A Merchant’s
Perspective.” NBER Working Paper no. 33152, November. w33152.pdf.
Beshears, J., J. J. Choi, D. Laibson, and B. C. Madrian. 2008. “The Importance of Default Options for
Retirement Saving Outcomes: Evidence from the United States.” In S. J. Kay and T. Sinha, eds.,
Lessons from Pension Reform in the Americas, 59-87. Oxford: Oxford University Press.
Bhutta, N., and D. Ringo. 2016. “Credit Availability and the Decline in Mortgage Lending to Minorities
after the Housing Boom.” Board of Governors of the Federal Reserve System FEDS Notes,
September 29. http://dx.doi.org/10.17016/2380-7172.1842.
Bhutta, N., A. Fuster, and A. Hizmo. 2020. “Paying Too Much? Price Dispersion in the U.S. Mortgage
Market.” Board of Governors of the Federal Reserve System Finance and Economics Discussion
Series, no. 2020-062, August. http://dx.doi.org/10.17016/FEDS.2020.062.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
43
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
References (Continued)
Bhutta, N., A. Hizmo, and D. Ringo. 2022. “How Much Does Racial Bias Affect Mortgage Lending?
Evidence from Human and Algorithmic Credit Decisions.” Board of Governors of the Federal
Reserve System Finance and Economics Discussion Series, no. 2022-067, October. https://doi.
org/10.17016/FEDS.2022.067.
Bhutta, N., S. Laufer, and D. R. Ringo. 2017. “The Decline in Lending to Lower-Income Borrowers by
the Biggest Banks.” Board of Governors of the Federal Reserve System FEDS Notes, September 28.
https://doi.org/10.17016/2380-7172.2077.
Blattner, L., and S. Nelson. 2021. “How Costly Is Noise? Data and Disparities in Consumer Credit.”
Stanford Graduate School of Business Working Paper no. 3978, May.
Boel, P., and P. Zimmerman. 2022. “Why Worry about Financial Exclusion?” Federal Reserve Bank of
Cleveland Economic Commentary, no. 2022-09, August 3. https://doi.org/10.26509/frbc-ec-202209.
Bolen, J. B., G. Elliehausen, and T. Miller. 2023. “Credit for Me but Not for Thee: The Effects of the Illinois
Rate Cap.” Public Choice 197: 397-420. https://doi.org/10.1007/s11127-023-01087-4.
Brevoort, K. P. 2022. “Does Giving CRA Credit for Loan Purchases Increase Mortgage Credit in Low-toModerate Income Communities?” Board of Governors of the Federal Reserve System Finance and
Economics Discussion Series, no. 2022-047, July. https://doi.org/10.17016/FEDS.2022.047.
Brevoort, K., P. Grimm, and M. Kambara. 2015. “Data Point: Credit Invisibles.” Consumer Financial
Protection Bureau, Office of Research report, May 5. https://www.consumerfinance.gov/
data-research/research-reports/data-point-credit-invisibles/.
Brown, M., J. Grigsby, W. van der Klaauw, J. Wen, and B. Zafar. 2016. “Financial Education and the Debt
Behavior of the Young.” Review of Financial Studies 29, no. 9 (September): 2490–2522. https://
doi.org/10.1093/rfs/hhw006.
Buchak G., G. Matvos, T. Piskorski, and A. Seru. 2018. “Fintech, Regulatory Arbitrage, and the Rise of
Shadow Banks.” Journal of Financial Economics 130, no. 3 (December): 453-83.
Campbell, D., F. Asís Martínez-Jerez, and P. Tufano. 2012. “Bouncing Out of the Banking System: An
Empirical Analysis of Involuntary Bank Account Closures.” Journal of Banking and Finance 36,
no. 4 (April): 1224-12.
Celerier, C., and A. Matray. 2019. “Bank-Branch Supply, Financial Inclusion, and Wealth
Accumulation.” Review of Financial Studies 32, no. 12 (December): 4767-4809. https://doi.
org/10.1093/rfs/hhz046.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
44
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
References (Continued)
Conklin, J. N., K. Gerardi, and L. Lambie-Hanson. 2022. “Can Everyone Tap into the Housing Piggy
Bank? Racial Disparities in Access to Home Equity.” Federal Reserve Bank of Atlanta Working Paper
no. 2022-17a, November, revised June 2024.
Consumer Financial Protection Bureau. 2019. “A Review of Youth Financial Education: Effects and
Evidence,” April. https://files.consumerfinance.gov/f/documents/cfpb_youth-financial-education_
lit-review.pdf.
Conway, J., J. Glaser, and M. C. Plosser. 2023. “Does the Community Reinvestment Act Improve
Consumers’ Access to Credit?” Federal Reserve Bank of New York Staff Reports, no. 1048, January.
Crook, J. 2001. “The Demand for Household Debt in the USA: Evidence from the 1995 Survey of
Consumer Finance.” Applied Financial Economics 11: 83-91.
Crosignani, M., and H. Le. 2023. “Stakeholders’ Aversion to Inequality and Bank Lending to Minorities.”
Federal Reserve Bank of New York Staff Reports, no. 1079, November.
D’Acunto, F., and A. G. Rossi. 2022. “Regressive Mortgage Credit Redistribution in the Post-Crisis Era.”
Review of Financial Studies 35, no. 1 (January): 482-525.
Danisewicz, P., and I. Elard. 2023. “The Real Effects of Financial Technology: Marketplace Lending and
Personal Bankruptcy.” Journal of Banking and Finance 155: 106986.
Debbaut, P., A. Ghent, and M. Kudlyak. 2016. “The CARD Act and Young Borrowers: The Effects and the
Affected.” Journal of Money, Credit, and Banking 48, no. 7 (October): 1495-1513.
Di Maggio, M., D. Ratnadiwakara, and D. Carmichael. 2022. “Invisible Primes: Fintech Lending with
Alternative Data.” NBER Working Paper no. 29840, March.
Dlugosz, J. L., B. T. Melzer, and D. P. Morgan. 2021. “Who Pays the Price? Overdraft Ceilings and the
Unbanked?” Federal Reserve Bank of New York Staff Reports, no. 973, June, revised July 2023.
Dobbie, W., A. Liberman, D. Paravisini, and V. Pathania. 2021. “Measuring Bias in Consumer Lending.”
Review of Economic Studies 88, no. 6 (August): 2799-2832.
Dobridge, C. L. 2018. “High-Cost Credit and Consumption Smoothing.” Journal of Money, Credit,
and Banking 50, no. 2–3: 407-33.
Dolson, E., and J. Jagtiani. 2021. “Which Lenders Are More Likely to Reach Out to Underserved
Consumers: Banks versus Fintechs versus Other Nonbanks?” Federal Reserve Bank of Philadelphia
Working Paper no. 21-17, April. http://dx.doi.org/10.21799/frbp.wp.2021.17.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
45
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
References (Continued)
Federal Deposit Insurance Corporation (FDIC). 2022. “The 2021 FDIC National Survey of Unbanked and
Underbanked Households,” October.
Fernandes, D., J. G. Lynch Jr., and R. G. Netemeyer. 2014. “Financial Literacy, Financial Education, and
Downstream Financial Behaviors.” Management Science 60, no. 8 (August): 1861-83.
FinRegLab. 2019. “The Use of Cash-Flow Data in Underwriting Credit: Empirical Research Findings,”
July. https://finreglab.org/wp-content/uploads/2023/12/FinRegLab_2019-07-25_Research-Report_
The-Use-of-Cash-Flow-Data-in-Underwriting-Credit_Empirical-Research-Findings.pdf.
Fulford, S., M. Rush, and E. Wilson. 2021. “Changes in Consumer Financial Status During the Early
Months of the Pandemic: Evidence from the Second Wave of the Making Ends Meet Survey.”
Consumer Financial Protection Bureau Data Point no. 2021-2, April.
Fuster, A., P. Goldsmith-Pinkham, T. Ramadorai, and W. Ansgar. 2022. “Predictably Unequal? The Effects
of Machine Learning on Credit Markets.” Journal of Finance 77, no. 1 (February): 5-47.
Fuster, A., M. Plosser, and J. Vickery. 2021. “Does CFPB Oversight Crimp Credit?” Federal Reserve Bank
of Philadelphia Working Paper no. 21-08, February.
Gerardi, K. S., P. Willen, and D. H. Zhang. 2020. “Mortgage Prepayment, Race, and Monetary
Policy.” Federal Reserve Bank of Atlanta Working Paper no. 2020-22, December. http://dx.doi.
org/10.2139/ssrn.3829923.
Hastings, J. S., B. C. Madrian, and W. L. Skimmyhorn. 2013. “Financial Literacy, Financial Education, and
Economic Outcomes.” Annual Review of Economics 5: 347-73.
Holtzblatt, J., and M. Karpman. 2020. “Who Did Not Get the Economic Impact Payments by Mid-to-Late
May, and Why? Findings from the May 14–27 Coronavirus Tracking Survey.” Urban Institute, July.
Homonoff, T. A., R. O’Brien, and A. B. Sussman. 2019. “Does Knowing Your FICO Score Change Financial
Behavior? Evidence from a Field Experiment with Student Loan Borrowers.” NBER Working Paper
no. 26048, July.
Howell, S. T., T. Kuchler, D. Snitkof, J. Stroebel, and J. Wong. 2024. “Lender Automation and Racial
Disparities in Credit Access.” Journal of Finance 79, no. 2 (April): 1457-1512.
Jagtiani, J. A., and C. Lemieux. 2017. “Fintech Lending: Financial Inclusion, Risk Pricing, and Alternative
Information.” Federal Reserve Bank of Philadelphia Working Paper no. 17-17, July.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
46
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
References (Continued)
Jagtiani, J. A., and C. Lemieux. 2018a. “Do Fintech Lenders Penetrate Areas that Are Underserved by
Traditional Banks?” Journal of Economics and Business 100: 43-54.
Jagtiani, J. A., and C. Lemieux. 2018b. “The Roles of Alternative Data and Machine Learning in Fintech
Lending: Evidence from the LendingClub Consumer Platform.” Federal Reserve Bank of Philadelphia
Working Paper no. 18-15, April, revised January 2019. http://dx.doi.org/10.21799/frbp.wp.2018.15.
Kaiser, T., and L. Menkhoff. 2020. “Financial Education in Schools: A Meta-Analysis of Experimental
Studies.” Economics of Education Review 78: 101930.
Kaiser, T., A. Lusardi, L. Menkhoff, and C. Urban. 2022. “Financial Education Affects Financial
Knowledge and Downstream Behaviors.” Journal of Financial Economics 145, no. 2, Part A
(August): 255-72.
Karlan, D., A. L. Ratan, and J. Zinman. 2014. “Savings by and for the Poor: A Research Review and
Agenda.” Review of Income and Wealth 60, no. 1 (March): 36-78.
Liu, H., and D. Volker. 2020. “Where Have the Paycheck Protection Loans Gone So Far?” Federal Reserve
Bank of New York Liberty Street Economics, May 6.
Lusardi, A., and O. S. Mitchell. 2014. “The Economic Importance of Financial Literacy: Theory and
Evidence.” Journal of Economic Literature 52, no. 1 (March): 5-44.
Madrian, B., and D. F. Shea. 2001. “The Power of Suggestion: Inertia in 401(k) Participation and Savings
Behavior.” Quarterly Journal of Economics 116, no. 4: 1149-87.
Mangrum, D. 2022. “Personal Finance Education Mandates and Student Loan Repayment.” Journal of
Financial Economics 146, no. 1 (October): 1-26.
McKernan, S-M., C. Ratcliffe, and K. Vinopal. 2009. “Do Assets Help Families Cope with Adverse
Events?” Urban Institute Perspectives on Low-Income Working Families Brief 10, November.
McKernan, S-M., C. Ratcliffe, B. Braga, and E. Kalish. 2016. “Thriving Residents, Thriving Cities: Family
Financial Security Matters for Cities.” Urban Institute, April.
Mehrotra, A., and J. Yetman. 2014. “Financial Inclusion and Optimal Monetary Policy.” BIS Working
Papers no. 476, December.
Mills, G., and J. Amick. 2010. “Can Savings Help Overcome Income Instability?” Urban Institute
Perspectives on Low-Income Working Families Brief 18, December. https://www.urban.org/sites/
default/files/publication/32771/412290-Can-Savings-Help-Overcome-Income-Instability-.PDF.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
47
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
References (Continued)
Morduch, J., and R. Schneider. 2017. The Financial Diaries: How American Families Cope in a
World of Uncertainty. Princeton, N.J.: Princeton University Press.
Morse, A. 2011. “Payday Lenders: Heroes or Villains?” Journal of Financial Economics 102, no. 1
(October): 28-44.
Naceur, S. B., R. Chami, and M. Trabelsi. 2020. “Do Remittances Enhance Financial Inclusion in LMICs
and in Fragile States?” IMF Working Paper no. 20/66, May.
Nair, A., and E. Beiseitov. 2023. “The Role of Fintech in Unsecured Consumer Lending to Low- and
Moderate-Income Individuals.” Federal Reserve Bank of New York, November. https://www.
newyorkfed.org/medialibrary/media/outreach-and-education/household-financial-well-being/
the-role-of-fintech-in-unsecured-consumer-lending-to-low-and-moderate-income-individuals.
National Consumer Law Center. 2024. “Predatory Installment Lending in the States: How Well Do the
States Protect Consumers Against High-Cost Installment Loans?” https://www.nclc.org/resources/
predatory-installment-lending-in-the-states-how-well-do-the-states-protect-consumers-against-highcost-installment-loans-2024/.
Nelson, S. Forthcoming. “Private Information and Price Regulation in the U.S. Credit Card
Market.” Econometrica.
Rhine, S. L. W., and W. H. Greene. 2013. “Factors That Contribute to Becoming Unbanked.” Journal of
Consumer Affairs 47, no. 1: 27-45. https://doi.org/10.1111/j.1745-6606.2012.01244.x.
Rhine, S. L. W., W. H. Greene, and M. Toussaint-Comeau. 2006. “The Importance of Check-Cashing
Businesses to the Unbanked: Racial/Ethnic Differences.” Review of Economics and Statistics 88,
no. 1: 146-57. https://doi.org/10.1162/rest.2006.88.1.146.
Rigbi, O. 2013. “The Effects of Usury Laws: Evidence from the Online Loan Market.” Review of
Economics and Statistics 95, no. 4: 1238-48.
Ringo, D. 2023. “Mortgage Lending, Default, and the Community Reinvestment Act.” Journal of
Money, Credit, and Banking 55, no. 1 (February): 77-102.
Shupe, C., G. Li, and S. Fulford. 2023. “Consumer Use of Buy Now, Pay Later: Insights from the CFPB
Making Ends Meet Survey”. Consumer Financial Protection Bureau Office of Research Publication
No. 2023-1, March.
Sirtaine, S. 2023. “The Future of Financial Inclusion.” CGAP Leadership Essay Series, September 13.
https://www.cgap.org/blog/future-of-financial-inclusion.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
48
Financial Inclusion in the United States: Measurement, Determinants, and Recent Developments
References (Continued)
Stein, L. C. D., and C. Yannelis. 2020. “Financial Inclusion, Human Capital, and Wealth Accumulation:
Evidence from the Freedman’s Savings Bank.” Review of Financial Studies 33, no. 11 (November):
5333-77. https://doi.org/10.1093/rfs/hhaa013.
Thaler, R. H., and C. R. Sunstein. 2008. Nudge: Improving Decisions About Health, Wealth, and
Happiness. New Haven, Conn.: Yale University Press.
Theodos, B., C. Plerhoples Stacy, and R. Daniels. 2018. “Client Led Coaching: A Random Assignment
Evaluation of the Impacts of Financial Coaching Programs.” Journal of Economic Behavior and
Organization 155: 140-58. https://doi.org/10.1016/j.jebo.2018.08.019.
Thompson Cochran, K., and M. Stegman. 2022. “Utility, Telecommunications, and Rental Data in
Underwriting Credit.” Urban Institute and FinRegLab Research Report.
Tiurina, M. 2022. “Tornado in Credit Desert: Role of Consumer Credit Access in Disaster Recovery.”
Master’s thesis, MIT. Tiurina-mtiu-SMMR-Management-2022-thesis.pdf.
Urban, C., M. Schmeiser, J. M. Collins, and A. Brown. 2020. “The Effects of High School Personal
Financial Education Policies on Financial Behavior.” Economics of Education Review 78: 101786.
Wang, J., and D. H. Zhang. 2020. “The Cost of Banking Deserts: Racial Disparities in Access to
PPP Lenders and Their Implications.” Harvard Business School Working Paper, December,
revised April 2021.
Willen, P., and D. Zhang. 2023. “Testing for Discrimination in Menus.” Harvard University Working
Paper, February.
Willis, L. E. 2011. “The Financial Education Fallacy.” American Economic Review 101,
no. 3 (May): 429-34.
Federal Reserve Bank of New York
Economic Policy Review 31, No. 3, September 2025
49
FEDERAL RESERVE BANK OF NEW YORK
ECONOMIC POLICY REVIEW
The Economic Policy Review is published by the Research and Statistics Group of the Federal
Reserve Bank of New York. The views expressed are those of the individual authors and do
not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal
Reserve System.
Series Editor: Julian di Giovanni. Editorial Staff: Robert Powell, Trevor Delaney,
Maureen Egan, Anna Snider, and Peter Stevens. Design Staff: Theresa Izzillo, Alyona
Botogoeva, and Laura Gharrity.
Economic Policy Review articles may be reproduced for educational or training purposes,
provided they are reprinted in full; include credit to the author(s), the publication, and the
Bank; and include the publication’s disclaimer.
©2025 The Federal Reserve Bank of New York
www.newyorkfed.org/research
Follow us on X (Twitter): @NYFedResearch