New York State
Department of Financial Services
Report on Inquiry into Redlining in Buffalo, New York
February 4, 2021
1
I. Introduction
The New York State Department of Financial Services (the “Department” or “DFS”) is
the State’s banking and insurance regulator. The Department regulates the activities of
approximately 1,500 banks and other licensed financial institutions with assets totaling more
than $2.6 trillion, including State-chartered banks, foreign bank branches, State-chartered credit
unions, credit reporting agencies, and mortgage loan originators and servicers. The Department
also regulates the activities of nearly 1,800 insurance companies with assets totaling more than
$4.7 trillion, including life insurance, property/casualty insurance companies, and health insurers
and managed care organizations. The Department also works to ensure transparency, fair
business practices, and public responsibility in these areas and to educate and protect consumers
of financial products and services.
The Department’s work includes enforcement of federal and state fair lending laws,
including the Fair Housing Act (“FHA”), the Equal Credit Opportunity Act (“ECOA”), and the
New York State Human Rights Law, N.Y. Executive Law § 296-a (“Exec. Law § 296-a”), as
well as administering New York State Banking Law § 28-b, the State’s Community
Reinvestment Act (“CRA”). Among the purposes of these laws is to eliminate housing
discrimination, including redlining.
Redlining is a form of housing discrimination that includes such illegal practices as
refusing to do business in a neighborhood based on the racial or ethnic composition of a
neighborhood’s population, or imposing more onerous terms on home loans in a particular
neighborhood in a discriminatory manner. The term “redlining” derives from the practice of
lenders or real estate brokers drawing lines, whether actual or metaphorical, around
neighborhoods they will not serve at all or around areas in which they will not serve minorities.
2
Such practice is both wrong and illegal — ECOA and Exec. Law § 296-a bar the use of race,
color, or national origin in making any credit determination, and the FHA outlaws reliance on
these characteristics in the making of home mortgage loans. Meanwhile, the CRA requires,
among other things, that each bank specifically define its service area (called an “assessment
area”) for evaluation of how well it provides banking services to its entire community rather than
excluding particular neighborhoods or groups. The CRA applies to banks, but does not currently
apply to non-depository lenders, such as non-depository mortgage companies.
Redlining (and housing discrimination more generally) has substantial negative societal
impacts. Homeownership is a key source of wealth building and financial stability, and
discrimination in lending therefore inhibits the economic opportunity of affected groups. Not
surprisingly, housing segregation is correlated with greater social vulnerability, including
decreased economic and educational opportunity and negative health outcomes. The COVID-19
pandemic crisis is an important and timely lesson: a September 2020 study from the National
Community Reinvestment Coalition points out that although COVID-19 has pushed racial health
inequities to the fore, “[r]acial residential segregation has systematically shaped characteristics of
the built environment that may increase susceptibility to not only COVID-19, but other
underlying health conditions,” including “through poor housing conditions, disparity in
educational and employment opportunities, inadequate transportation infrastructure, access to
healthcare and economic instability.”
1
The study further draws connections between policies of
residential segregation dating back more than 80 years (discussed in more detail in Section II
1
Jason Richardson, et al., Redlining and Neighborhood Health, National Community Reinvestment Coalition,
September 10, 2020. See also, Heather Long, et al., The Covid-19 Recession is the Most Unequal in Modern U.S.
History, The Washington Post, September 30, 2020.
3
below) and housing patterns today, finding many neighborhoods that were redlined in the 1930s
are suffering the most severe effects of the COVID-19 pandemic now.
The City of Buffalo has, unfortunately, long been one of the most racially segregated
cities in the United States.
2
The Department has recently conducted an inquiry into mortgage
lending patterns in the Buffalo metropolitan statistical area (“Buffalo MSA”), which consists of
Erie, Niagara, and Cattaraugus counties, in essence encompassing the city of Buffalo and its
surrounding towns.
3
The Department, using Home Mortgage Disclosure Act (“HMDA”) data to
map out and analyze patterns of mortgage lending in the Buffalo area, identified a distinct lack of
lending by mortgage lenders, in particular several non-depository lenders, in neighborhoods with
majority-minority populations and to minority homebuyers in general.
This report recounts the findings and outcomes of the Department’s inquiry and includes:
a summary of the history of housing segregation in Buffalo and its negative effects; an analysis
of current mortgage lending in the Buffalo metropolitan area; a description of the Department’s
investigations targeting several lenders that have underserved minority populations
4
; and
proposed legislative changes aimed to help remediate this tenacious problem.
2
See, e.g., Joe Cortright, America’s Least (and Most) Segregated Cities, City Observatory, August 17, 2020,
(showing Buffalo to be the fourth most segregated urban area in the U.S.); Erica Brecher, A Legacy of Racism: How
Past Practices Affect Segregation in Buffalo Today, August 10, 2020, WIVB News 4 Buffalo; William H. Frye,
Black-White Segregation Edges Downward Since 2000, Census Shows, Brookings Institution, December 17, 2018,
(Buffalo ranked sixth most segregated); Anna Blatto, A City Divided: A Brief History of Segregation in Buffalo,
April 2018.
3
An MSA is “an area containing a large population nucleus and adjacent communities that have a high degree of
integration with that nucleus,” and the purpose of MSAs is to “provide[] a nationally consistent set of delineations
for collecting, tabulating, and publishing Federal statistics for geographic areas.” They are used particularly with
respect to census data. https://www.census.gov/programs-surveys/metro-micro/about.html.
4
According to 2019 Census Bureau data, the estimated racial and ethnic makeup of the city of Buffalo is 47.4%
white non-Hispanic or Latino, 36.7% Black, 11.6% Hispanic or Latino, 5.6% Asian, and 0.5% American Indian or
Alaska Native, with 4.0% reporting more than one race/ethnicity.
4
II. Redlining in the 20th Century in Buffalo
The standard home mortgage that most Americans are familiar with today dates to the
1930s. Until that time, a mortgage typically required a 50% down payment and included interest-
only payments before the principal would be repaid, typically within five to seven years. In 1933,
as Americans struggled to weather the Great Depression, the U.S. government took action to
prevent mass foreclosures, establishing the Home Owners’ Loan Corporation, commonly known
as HOLC, which was authorized to purchase existing mortgages subject to imminent foreclosure
and to offer revised repayment terms to the borrowers, usually consisting of longer repayment
schedules and lower interest rates.
HOLC also offered mortgages that allowed payment of
principal along with interest over extended periods, bringing home ownership into reach for
working- and middle-class Americans.
In its first two years, HOLC granted over one million mortgages; within its first three
years, it had refinanced about 10% of the country’s non-farm mortgages. HOLC has been
credited with “protecting and expanding home ownership,” “standardizing lending practices,”
and “encouraging residential and commercial real estate investment in a flagging economy,” as
well as “work[ing] with public and private sector partners to create millions of jobs and help
millions of Americans buy or keep their homes.”
5
Supported by the work of the Federal Housing
Administration, created in 1934, the rate of American homeownership grew from 30% in 1930 to
60% in 1960.
6
5
Robert K. Nelson, LaDale Winling, Richard Marciano, Nathan Connolly, et al., Mapping Inequality: Redlining in
New Deal America,
“Mapping Inequality,” American Panorama, ed. Robert K. Nelson and Edward L. Ayers,
accessed November 29, 2020 https://dsl.richmond.edu/panorama/redlining/#loc=5/39.1/-94.58&text=intro; Bruce
Mitchell and Juan Franco, HOLC “Redlining” Maps: The Persistent Structure Of Segregation And Economic
Inequality, National Community Reinvestment Coalition, March 20, 2018.
6
Thomas Sugrue, Sweet Land of Liberty: The Forgotten Struggle for Civil Rights in the North (2008).
5
HOLC’s benefits, however, were not equally distributed through the country’s
population, and elements of HOLC’s work likely produced harms that echo even in today’s
housing market. Perhaps the most notorious of HOLC’s programs was its creation of its color-
coded “security maps,” which graded neighborhoods purportedly on the basis of their mortgage
security risk level, from “safe” (grade A, which was colored green) to “hazardous,” (grade D,
colored red). HOLC advised that, “conservative, responsible lenders . . . would ‘refuse to make
loans in [hazardous] areas [or] only on a conservative basis.”
7
These maps included narrative
descriptions accompanying the grades, including, “the neighborhood’s quality of housing, the
recent history of sale and rent values, and, crucially, the racial and ethnic identity and class of
residents that served as the basis of the neighborhood’s grade.”
8
Scholars have not fully settled to
what extent these maps were used by lenders outside the government,
9
and there is no doubt that
an array of other factors, including private restrictive covenants, segregation in public housing,
and practices such as steering and blockbusting contributed to and perpetuated segregation. That
said, the HOLC security maps are widely understood to have laid the groundwork for the
segregation we still see today. The maps helped “set the rules for nearly a century of real estate
practice” and offer a useful source for comparison of conditions today to conditions nearly a
century ago.
10
The racism expressed in the HOLC security maps was not subtle, and HOLC explicitly
expressed the perverse logic that segregation was good for home values. HOLC security map
7
Id.
8
Id.
9
Robert K. Nelson, LaDale Winling, Richard Marciano, Nathan Connolly, et al., Mapping Inequality: Redlining in
New Deal America, “Mapping Inequality,” American Panorama, ed. Robert K. Nelson and Edward L. Ayers;
Thomas Sugrue, Sweet Land of Liberty: the Forgotten Struggle for Civil Rights in the North (2008); Keeanga-
Yamahtta Taylor, Race for Profit: How Banks and the Real Estate Industry Undermined Black Homeownership
(2019) (National Association of Real Estate Boards opposed the Fair Housing Act).
10
Id.
6
documents include, as positive attributes of Buffalo-area neighborhoods, that the white, non-
immigrant population was “[w]ell protected by restrictions” and “sufficiently large to resist
outside influences.”
11
Meanwhile, other Buffalo security map area descriptions identified the
supposed “detrimental influences” on “[a] small old area occupied almost entirely by railroad
porters
12
and their families” with a population of 95% Black residents and 5% Italian immigrants
as, “[a]ge and condition of buildings as well as type of occupant”
13
; another described a
neighborhood made up of about 75% Italian immigrants and 25% African Americans as, “[a]n
extremely old section which has been taken over by Italians and Negros of a poor type [sic]”
(emphasis added).
14
The language clearly conveys the belief that the value of the homes of
majority white neighborhoods was to be preserved through segregation and excluding non-white
and foreign-born people.
15
Buffalo, like many cities, is still reckoning with the effects of the abhorrent racism
represented in the HOLC descriptions. Redlining by lenders has continued to take place long
after it was banned, and many areas deemed “hazardous” by HOLC in the 1930s today remain
segregated, with “much higher proportions of low- and moderate-income (LMI) families (74%)
11
1937 HOLC Area Description of the Township of Amherst, NY, Robert K. Nelson, LaDale Winling, Richard
Marciano, Nathan Connolly, et al., “Mapping Inequality,” American Panorama, ed. Robert K. Nelson and Edward
L. Ayers, accessed November 29, 2020; 1937 HOLC Area Description of Buffalo, NY, area A-5, Central Park
neighborhood.
12
At the time, the occupation of “railroad porter” would likely have been intended by the writer and understood by
the reader, even without the other statistics in the Area Description document, to indicate an African-American
population. See, e.g., R.R. Com. of Tex. v. Pullman Co., 312 U.S. 496, 497, 61 S. Ct. 643, 644 (1941) (“As is well
known, porters on Pullmans are colored.”) (Cited in Martha A. Sandweiss, Passing Strange: A Gilded Age Tale of
Love and Deception Across the Color Line, at 140 (2009).)
13
1937 HOLC Area Description of Buffalo, NY, Area D-1, Cold Springs neighborhood. Robert K. Nelson, LaDale
Winling, Richard Marciano, Nathan Connolly, et al., “Mapping Inequality,” American Panorama, ed. Robert K.
Nelson and Edward L. Ayers, accessed November 29, 2020.
14
1937 HOLC Area Description of Buffalo, NY, Area D-3, covering parts of Willert Park, Emslie, and Broadway-
Fillmore neighborhoods. Id..
15
See Keeanga-Yamahtta Taylor, Race for Profit: How Banks and the Real Estate Industry Undermined Black
Homeownership (2019).
7
and people of color (64%) than the higher-graded neighborhoods” across the U.S.
16
According to
a 2018 report, in Buffalo, approximately 85% of people who identify as Black live in
neighborhoods to the east of Main Street, which is also where many of HOLC’s 1930s redlined
areas were located.
17
These populations also continue to experience economic disadvantage, lack
of access to quality financial services, environmental hazards, lower life expectancy, and worse
health outcomes than the overall population.
18
The homeownership rate for the Black population
in Buffalo is also much lower than for the white population.
19
As recently as 2015, a Buffalo-
based bank, Evans Bank, entered into a settlement with the New York State Attorney General to
resolve charges that it engaged in redlining majority African-American areas of Buffalo, denying
access to mortgages to those communities based on the race of their population.
20
The results of the Department’s own analysis in this matter are consistent with the
conclusion that the echoes of the HOLC maps are evident today. In the figure that follows, the
HOLC “declining” and “hazardous” areas are outlined in yellow and red, respectively, and red
dots show the location of every mortgage issued during a recent three-year period by all of the
major mortgage originators. The shaded areas are census tracts in which minorities constitute a
majority of the population. This map demonstrates that even today there are far fewer mortgages
issued in areas that have a large proportion of minority residents.
16
Id.
17
Anna Blatto, A City Divided: A Brief History of Segregation in Buffalo, at 2 (2018).
18
Id. at 14; http://www.buffalo.edu/ubnow/stories/2020/10/cheri-update.html; Marian Hetherly, Can Buffalo
Overcome Its Racial Inequities? The Answer Isn’t Black and White, WBFO Buffalo-Toronto Public Media, June 19,
2020.
19
Marian Hetherly, Can Buffalo Overcome Its Racial Inequities? The Answer Isn’t Black and White, WBFO
Buffalo-Toronto Public Media, June 19, 2020.
20
https://ag.ny.gov/press-release/2015/ag-schneiderman-secures-agreement-evans-bank-ending-discriminatory-
mortgage
8
III. The Department’s Inquiry
In light of the long-standing problems with housing discrimination in Buffalo, the
Department conducted an inquiry into the mortgage market in the Buffalo MSA. Among other
things, the Department analyzed HMDA data to develop a broad picture of home mortgage
lending activity in the area; conducted in-depth investigations of several lenders that the HMDA
data showed performed poorly in serving minority communities; and met with lenders who
City of Buffalo
2016–2019 Mortgage Originations


HOLC Residential Security Map Areas
(circa 1938)

9
performed well to examine best practices to promote lending to minorities and in majority-
minority areas.
A. HMDA Lending Patterns Data
HMDA, a federal law enacted in 1975 to address lack of lending to minority buyers and
in majority-minority neighborhoods, requires residential mortgage lenders to collect and report
information about the applications they receive and their underwriting determinations on those
applications. The data includes the race, ethnicity, gender, and income of the applicant, the type
of loan applied for, the location of the property, and whether the application was approved or
denied. As with all credit transactions, the lenders are not permitted to rely on race, ethnicity, or
gender of an applicant in their lending determinations — this data is collected and reported to be
used for, among other things, the evaluation of the extent to which lenders are serving the credit
needs of their communities and the identification by supervisory and enforcement agencies of
potentially discriminatory practices of specific lenders as well as broad patterns of
discrimination.
21
HMDA data is also publicly available and may be used for research and
analysis by non-governmental entities and individuals.
The following charts are based on HMDA data and show lending patterns from 2016
through 2019 in the Buffalo MSA by lenders originating 100 or more mortgages annually, as
well as aggregate data for all lenders in the MSA during that time period for comparison. The
first chart shows lenders’ performance lending in census tracts in which a majority of the
residents are minorities; that chart is sorted by the percentage of a lender’s loans in the majority-
minority census tracts as a percentage of that lender’s total loans in the MSA as a whole, with the
best-performing lenders at the top. The second chart shows lenders’ performance lending to
21
HMDA is a notable exception to the fair lending statutory and regulatory prohibitions on creditors from inquiring
into the race, ethnicity, and sex of applicants to prevent discrimination on those bases.
10
minority borrowers throughout the MSA; that chart is sorted by the lender’s percentage of loans
made to minorities, once again with the best performers at the top.
13
One statistic that leaps from the data is that, market-wide, loans made to minorities in the
Buffalo MSA comprise only 9.74% of the total loans made in Buffalo — less than half of what
would be expected given that minorities make up roughly 20% of the MSA’s population. This is
stark confirmation that minority participation in the mortgage market (and therefore their
opportunities for home ownership, one of the biggest drivers of long-term wealth for American
families) lags far behind that of non-minorities. Though this statistic by itself does not shed light
on why this opportunity gap persists, or by itself suggest that there is intentional redlining
occurring, at the very least it demonstrates that this substantial societal problem remains to be
solved.
The data also shows that the mortgage market is comprised of a wide variety of different
kinds of lenders. Banks of all sizes are active participants in the market, from very large national
(indeed, international) banks such as Bank of America and Wells Fargo, to large regional banks
like M&T Bank and KeyBank, to small local and community banks. Some of the institutions
heavily focus on online lending, Quicken Loans being the most prominent example, while others
rely on a more traditional brick-and-mortar business with physical locations within the
community. And many of the participants are not banks at all but are rather non-depository
mortgage lenders.
Nonbank mortgage lenders typically focus exclusively on mortgage lending and do not
offer other services and products such as savings and checking accounts. They often rely on
outside financing to fund the origination of mortgages and then sell the mortgages on the
secondary market. The prominence of non-depository lenders has recently dramatically
increased, especially since the 2008 financial crisis. Whereas in 2013 banks originated 70% of
new mortgages, nonbanks are now responsible for a majority of mortgage originations
14
nationally.
22
That number is somewhat lower in the Buffalo MSA, but is still substantial: from
2016 to 2019, nonbank mortgage lenders were responsible for originating about 37% of
mortgages in the Buffalo MSA.
The Department’s analysis showed that non-depository institutions in the Buffalo MSA
lent at a lower rate in majority-minority census tracts than depository institutions did, making
3.70% of their mortgage originations for properties in these areas whereas depository institutions
made 5.05%. Researchers have found that nationally, nonbanks generally tend to serve more
first-time, low-and moderate-income, and minority homebuyers than depository institutions, and
their borrowers, on average, have lower FICO scores,
23
and the Department’s analysis of HMDA
data showed that within New York State, on average from 2016 to 2019, non-depository
institutions lent at a higher rate in majority-minority census tracts than depository institutions,
originating 20.75% of their loans for homes in these areas versus depository institutions’
15.07%. However, a study based on 2017 HMDA data and analyzing individual lenders rather
than the market overall, found that among banks, more made a greater share of their home loans
to low- and moderate-income borrowers and for homes in low- and moderate-income census
tracts than non-depository lenders did.
24
Thus, it is not clear that non-depository lenders
22
See, e.g., Consumer Financial Protection Bureau, FFIEC Announces Availability of 2019 Data on Mortgage
Lending, June 24, 2020, available at (nonbanks were responsible for 56.4% of home-purchase mortgages in 2019,
down from 57.2 percent in 2018); Bank and Nonbank Lending Over the Past 70 Years, 13 FDIC Quarterly 4 (2019)
at 34.
23
Id., Karan Kaul and Laurie Goodman, Should Nonbank Mortgage Companies Be Permitted to Become Federal
Home Loan Bank Members? Urban Institute Housing Finance Policy Center Comment Ltr. at 3, June 2020; Neil
Bhutta, et al., The Decline in Lending to Lower-Income Borrowers by the Biggest Banks, FEDS Notes, September
28, 2017.
24
Jason Richardson and Josh Silver, Home Lending To LMI Borrowers And Communities By Banks Compared To
Non-Banks, National Community Reinvestment Coalition, Apr. 18, 2019. (Nearly 14% of all banks issued more than
40 percent of their loans to low- and moderate-income borrowers and for homes in low- and moderate-income
census tracts, whereas 5.3% of non-bank lenders did so. Of banks that made government-backed home purchase
mortgages, 51.9% made more than 40% of them to low- and moderate-income borrowers and for homes in low- and
moderate-income census tracts, whereas 32.1% of non-depository institutions did so.)
15
necessarily are better at reaching minority borrowers or making mortgage loans in majority-
minority neighborhoods.
The data from the Buffalo MSA reflected in the charts above also show that there is a
wide discrepancy as to how successful different lenders are in providing mortgage services to
minorities and in majority-minority census tracts. Both Evans Bank and Five Star Bank,
25
for
example, originated more than 12% of their loans in majority-minority census tracts, far
exceeding the market average of roughly 4.5%. By contrast, six lenders in the Buffalo MSA
originated less than 3% of their loans in majority-minority tracts: Cornerstone Community
Federal, Fairport Savings Bank, Caliber Home Loans, Inc., Hunt Mortgage Corporation, First
Priority Mortgage Inc., and USAA Federal Savings Bank. With respect to providing loans to
minorities, four institutions, Five Star Bank, Bank of America, Evans Bank, and Homestead
Funding Corp. made more than 15% of their loans in the Buffalo MSA to minorities, far
exceeding the market average of 9.74%. Six entities, meanwhile, made less than 6% of their
loans to minorities: Cornerstone Community Federal, Fairport Savings Bank, Alden State Bank,
Community Bank, N.A., Northwest Bank and Caliber Home Loans, Inc. Although poor
performance by a lender in these measures does not necessarily mean that there is discrimination
or a fair lending violation, consistent poor performance suggests that scrutiny is warranted by
regulators and law enforcement agencies.
25
It is worth emphasizing that the success of Evans Bank and Five Star is influenced by each of those banks having
been required to remediate fair lending violations in connection with enforcement investigations. Irrespective of the
reason, the Evans Bank example demonstrates that efforts to promote lending in minority communities can indeed
be effective.
16
B. The Department’s Enforcement Investigations
Informed by the HMDA data, the Department has undertaken fair lending investigations
into several of the DFS-regulated
26
entities that performed poorly on one or both of the measures
reflected in the charts above. The Department generally has found that these companies, which
accounted for approximately 15% of the Buffalo MSA market, were lending overwhelmingly to
white borrowers in majority-white neighborhoods and served minority borrowers and minority-
majority areas at a substantially lower rate than other institutions in the Buffalo metropolitan
area. Moreover, in some cases, even looking at the mortgages these institutions did originate for
homes in majority-minority neighborhoods, most of those loans were to white borrowers.
Notably, these institutions’ lack of mortgage originations for properties in majority-
minority census tracts and the lack of lending to individuals identifying as members of a
minority group did not appear to be explained by excessive denials of loan applicants based on
race or ethnicity. In fact, the Department found that the companies generally have high approval
rates for their applications, even to minority borrowers, and that the pricing and rates for the
mortgages were typically generated by standard underwriting software and not subject to
substantial discretion from loan officers.
Instead, the Department’s investigation revealed that these companies had little or no
engagement with minorities, and generally made scant effort to do so. There was little marketing
directed to minorities; no real effort to serve majority-minority neighborhoods within the Buffalo
MSA; and little or no effort to track how well the lenders were serving minority communities.
26
Many of the institutions listed in the charts are federally chartered banks or credit unions. Those institutions are
regulated by their respective federal regulators, and pursuant to federal law the Department has no authority to
investigate or take any enforcement action against them. The Department encourages the federal regulators to
investigate whether fair lending violations are occurring at those institutions that are performing poorly on these
statistical measures. The Department’s own jurisdiction in this area is limited to state-chartered banks and credit
unions and DFS-licensed non-depository mortgage lenders.
17
The unsurprising result of this has been that few minority customers or individuals seeking
homes in majority-minority neighborhoods have made loan applications to these lenders in the
first instance. Moreover, this lack of minority clients is not convincingly explained by lack of
opportunities for minority business. Although the data shows, as noted above, that minorities are
significantly underrepresented in the mortgage market on a per capita basis, there are a
substantial number of minority borrowers (nearly 10% of the market) in Buffalo, and other
lenders are consistently serving minority communities successfully.
To date, the Department has concluded and resolved its investigation into Hunt Mortgage
Corporation (“Hunt Mortgage”). As reflected in the charts above, HMDA data for the period
from 2016 through 2019 revealed a significant lack of lending to minorities and in majority-
minority neighborhoods in Buffalo, and additional analysis showed similar issues in other cities,
such as Syracuse. The Department obtained documents from Hunt Mortgage and took testimony
from its officials. Although the Department found no evidence of intentional discrimination on
the part of Hunt Mortgage and made no finding of any violation of fair lending laws, the
Department concluded that weaknesses in Hunt Mortgage’s fair lending and compliance
programs — and a lack of sufficient attention to fair lending issues generally — contributed to
Hunt Mortgage’s poor performance in lending to minorities and in majority-minority
neighborhoods. Hunt Mortgage was cooperative throughout the investigation, and, in a good-
faith attempt to increase lending in majority-minority neighborhoods and to minority borrowers,
Hunt Mortgage entered into an agreement
27
with the Department pursuant to which Hunt
Mortgage will take significant steps to improve its service to the entire community. These steps
include:
27
The Written Agreement between Hunt Mortgage and the Department is available on the Department’s website.
18
Increasing marketing to minorities and within majority-minority neighborhoods;
Developing a special financing program which will provide at least $150,000 in
discounted or subsidized financing on loans for properties located in majority-minority
neighborhoods;
Providing annual fair lending training to Hunt Mortgage employees and agents with
significant involvement in lending; and
Conducting an annual fair lending compliance audit.
Investigations into multiple other lenders remain ongoing, and the Department will
announce findings as those cases are resolved. In the meantime, however, it bears noting in this
report that the Department has found that many of the companies suffer from the same basic
failing: a general lack of attention to whether they are serving the entire Buffalo community,
including minorities and majority-minority neighborhoods. By and large, the companies make
little or no effort to obtain business in minority areas, do not have adequate fair lending
compliance programs, and do not track whether or how well they are serving minority
populations. Indeed, some of the companies did not know that they were performing worse than
the market as a whole until the Department began its investigation. This is all particularly true
for non-depository lenders, which, although they are required to abide by fair lending laws, are
not required to define an assessment area and track performance as required for banks under the
CRA. As discussed more fully below, this gap in the law between the treatment of banks and
non-depository lenders should be addressed.
C. Best Practices
The Department surveyed several New York banks to learn more about current best
practices among financial institutions. More specifically, the Department reached out to western
New York banks with robust fair lending programs to learn of their successful strategies. Among
those, Five Star Bank (“Five Star”) entered into a 2015 settlement with the New York State
19
Attorney General to resolve redlining claims in Rochester. The bank viewed the Attorney
General’s review of its operations as a guide to reform. Today, Five Star serves minority
communities, including those in Buffalo, in an exemplary manner, as reflected in the charts
above. Likewise, M&T Bank credits a prior consent order with federal regulators as turning point
in improving its fair lending efforts. The 2015 consent order, which resolved ECOA and FHA
violations, was entered into by Hudson City Bank prior to its acquisition by M&T Bank. The
Department additionally reached out to Evans Bank, referenced above, which since its settlement
with the New York State Attorney General in 2015 has also performed well in serving minority
borrowers and making loans in majority-minority neighborhoods. Further, the Department noted
in its evaluation of lending data that Homestead Funding Corp., a nonbank mortgage lender, also
appeared to be among the better performers in the market and spoke with a member of its
compliance team.
The fair lending programs of Five Star, M&T, Evans, and Homestead presently
incorporate the following elements:
Robust fair lending compliance policies and practices, including:
o Meaningful engagement of senior management in fair lending programs
and policy development and implementation;
o Periodic fair lending training for staff;
o A responsive and engaged compliance committee;
o Regular engagement of staff on Community Reinvestment Act-related
lending with attention to improving performance; and
o Regularly scheduled internal and third-party analysis of institutional
lending data with prompt responses to findings.
Specialized product and service offerings, such as:
20
o Products specifically designed for low- and moderate-income customers
that may integrate subsidies and tailored underwriting standards; and
o Lending programs focused on community revitalization and development.
Community outreach and engagement efforts, for example:
o Partnering with community organizations to understand community needs
in the bank’s service areas, and to connect with customers they previously
did not;
o Enhanced attention to marketing content and reach including marketing
targeted at underserved communities;
o Hiring loan officers specializing in community development and working
with brokers who serve and have connections to communities the
institution had not been reaching before; and
o Providing educational programming to the community relating to home
buying.
Significantly, the banks report that these efforts have increased profits in addition to
reversing fair lending issues.
D. Referral to New York State Department of State
The Department also found in the course of its inquiry that the non-depository mortgage
lenders relied heavily on client referrals from real estate agents employed by affiliated real estate
companies and independent companies for their business. The New York State Department of
State (“DOS”) licenses real estate brokers and salespeople and enforces related laws and
regulations, including those barring housing discrimination. DFS has informed DOS of the
findings of this inquiry so that DOS may investigate, as it deems appropriate, whether the real
estate brokers and salespeople referring business to the mortgage lenders DFS reviewed are
engaged in any prohibited discriminatory behavior that could be affecting the mortgage lenders’
lending patterns.
21
IV. Proposal for Legislative Reform
As noted above, the prominence of non-depository mortgage lenders in the marketplace
has increased substantially over time. As a result, a substantial portion of the mortgage lending
market is exempt from CRA requirements, which are limited to banks. Applying the CRA to
non-depository lenders would be an important step in addressing fair lending abuses in the New
York residential loan market. Accordingly, the Department recommends New York State
Banking Law § 28-b, the State’s Community Reinvestment Act, be amended to apply to non-
depository mortgage lenders.
The federal CRA, 12 U.S.C. § 2901, et seq., enacted by Congress in 1977, recognizes that
banks have a “continuing and affirmative obligation to help meet the credit needs of the local
communities in which they are chartered” and directs federal banking regulators to assess the
performance of all banks insured by the Federal Deposit Insurance Corporation, including state-
chartered banks, in meeting the credit needs of those entire communities. 12 U.S.C. §§ 2901(a),
2902(2), 2903(a).
New York State is one of few states with a state-enacted version of CRA, Banking Law §
28-b. The law, enacted in 1978, largely mirrors the federal CRA and empowers the Department
to conduct CRA examinations of New York state-chartered banks. Both the federal and New
York CRA statutes sought to address redlining concerns, and the Department and the federal
banking regulators have promulgated regulations setting forth a detailed framework for
evaluating banks’ CRA performance. Each bank must define its “assessment area,” the area each
bank is obligated to serve, based on the locations of bank branches and loan issuances.
Regulators evaluate each bank’s residential mortgage and small business lending to determine
22
whether a bank sufficiently lends both to low- and moderate-income borrowers, and in low- and
moderate-income neighborhoods.
In conformity with CRA requirements, regulators also review loans and investments of
larger banks in support of community development, as well as the extent of retail services in
CRA assessment areas. Regulators issue one of four ratings to banks (Outstanding, Satisfactory,
Needs to Improve, and Substantial Noncompliance), and make public CRA evaluation results. In
addition to the reputational harm to the bank of a negative assessment from its regulator, a poorly
performing institution may be prevented by the Superintendent or federal regulators from
engaging in corporate transactions and opening new branches.
Although the CRA does not directly give the Department (or other regulators) any
additional authority to enforce fair lending statutes, the steps that lenders must take to comply
with CRA requirements are extremely helpful in investigating and prosecuting fair lending
violations for those banks that violate the law. Because the CRA requires an institution to
identify the areas it serves and track its success in serving all communities within its service
areas, fair lending problems are more readily identified and can be remedied, either by the lender
itself or by regulators or enforcement agencies.
The Department submits that requiring non-depository mortgage lenders to comply with
CRA requirements would lead to these institutions to focus more on serving the entirety of the
communities in which they operate, resulting in an increase in lending to minorities and low-
income borrowers. The non-depository institutions’ increasing share of the home mortgage
market, and the variability in their performance as demonstrated by the case of Buffalo, weigh in
favor of this measure even if broadly speaking, non-depository lenders serve minority and low-
and moderate-income communities at higher rates than banks. It is not clear that non-depository
23
lenders consistently outperform banks in lending to historically underserved populations. As
noted above, at least one study has concluded that, at an institution level, more banks do better at
serving their whole communities than nonbanks do.
28
Further, in response to the Department’s
inquiry regarding its fair lending policies, Homestead Funding Corp., the non-depository lender
noted above with a good record of reaching minority populations in the Buffalo area explained
that one reason for its success is that it operates in another state that happens to impose CRA
requirements on non-depository institutions and therefore, analyzing lending data to ensure the
lender is reaching all parts of its markets is a regular practice. Especially with respect to the
portion of non-bank mortgage lenders that are not reaching low- and moderate-income
borrowers, application of the CRA could improve their performance as it has for banks and help
ensure continued good performance for those institutions already serving their whole
communities, thereby increasing the availability of mortgage lending in communities that have
heretofore experienced unequal access.
28
Jason Richardson and Josh Silver, Home Lending To LMI Borrowers And Communities By Banks Compared To
Non-Banks, National Community Reinvestment Coalition, Apr. 18, 2019.