Letter to the Hon. Dave Uejio, Acting Director of the Consumer Financial Protection Bureau - Booker, Warren and Colleagues Request Information from CFPB on Steps They Are Taking to Prevent Tenant Screening Technologies from Hurting Renters of Color

Letter

Date: March 3, 2021
Location: Washington, DC

Dear Acting Director Uejio,
We write to request information about the Consumer Financial Protection Bureau (CFPB)'s
oversight authority and regulatory history with regard to tenant screening technology companies.
The COVID-19 pandemic has led to a significant rise in economic and housing insecurity; in
January 2021, more than 20% of renters were behind on rent payments and millions were at risk
of eviction.1 Despite eviction bans put in place as public health measures, evictions have
continued throughout the pandemic, and even if unsuccessful or illegal, these evictions often still
appear on reports generated by tenant screening companies.2 Landlords are already relying on
tenant screening technologies that are known to collect faulty, publicly available data from court
records,3
and a rise in eviction filings that may violate local, state or, federal bans4
could mean
that millions of Americans may find themselves blacklisted in future housing searches due to
these faulty screening technologies. We are particularly concerned that communities of color
that have been disproportionately affected by job losses in the pandemic, and have historically
been precluded from homeownership, will be the hardest hit, and we write to request more
information on the regulation of these tenant screening software companies.
Uses of Tenant Screening Technologies
Tenant screening technologies are used by landlords to review potential tenants. Spurred on by a
rise in the number of Americans renting, historic declines in rental vacancy rates, and the
growing data economy, the tenant screening industry has seen rapid expansion in recent years and is now valued at $1 billion.
5 These technologies are widely used--survey results indicate
that approximately 90% of landlords are relying on them to make decisions about whether to rent
to a potential tenant.
6 Landlords report that they rely on these technologies to review income and
employment history, eviction history, and criminal background of prospective tenants.7 But these
data sources are often plagued by inaccuracies.
Many tenant screening companies employ search algorithms and use web scrapers to search for
and collect data on publicly available court websites.8 Companies then compile this data into
reports containing either a numerical recommendation score or a "thumbs-up or thumbs-down"
recommendation of whether a landlord should rent to the potential tenant.9 Easy access to
criminal databases and publicly available court data has led to a proliferation of background
screening companies, with the CFPB identifying 1,954 background screening companies in
2019.10 Because tenant screening companies are not required to register with any government
agency, the exact number of these companies is unknown.11
Tenant Screening Companies' Practices Have a Disproportionately Harmful Effect on
Tenants of Color
Employees at some background screening companies have stated in lawsuits against the
companies that they "err on the side of including any possible match, rather than excluding
possible errors."12 This is reflected in the design of some companies' search algorithms which
employ an error-prone tool, termed a "wild-card," that requires only a partial match of the letters
in a tenant's name during a search.13 This practice may have disproportionately negative
consequences for potential tenants with common surnames, which can include many potential tenants of color. The U.S. Census Bureau found that "twenty-six surnames cover a quarter of the
Hispanic population and 16 percent of Hispanic people reported one of the top 10 Hispanic names," and this pattern is similar for Asian and Black Americans.14 This means that, because an
individual with the same or a similar name has a criminal record or court history, the potential
tenant is unable to obtain housing if a landlord requires a tenant screening record.
This common surname problem is not the only concern. Because many screening services only
search for whether the potential tenant has been named in an eviction case -- and don't include
the reasons for the landlord filing the eviction, how long ago it occurred, or the outcome of the
case including whether it was dismissed -- the potential tenant can find themselves blacklisted as
a result of incomplete data from the screening software.15
This problem of incomplete and incorrect data on eviction filings poses a risk to renters every
year, but may create a heightened threat in 2021. Due to the economic and housing instability
wrought by the COVID-19 pandemic, an estimated 30-40 million Americans remain at risk of
eviction;16 in January 2021, more than 20% of renters reported being behind on rent.17
High Eviction Rates Due to the COVID-19 Pandemic
According to data from Princeton University's Eviction Lab, more than 20,000 evictions were
filed between the start of the CDC moratorium through October 17, 2020 in the 27 cities the
project tracks, bringing the total number of evictions since the start of the pandemic in those
cities to at least 240,000.18
These evictions pose a severe threat to families and to public health,19 and people of color are
disproportionately at risk of the harms of eviction. Communities of color and women have been
hardest hit by job losses during the pandemic--with women of color at particular risk.20 As a direct result of our nation's legacy of racist, harmful policies, Black families are less likely to
own homes, less likely to have generational wealth from past homeownership to afford a down
payment, and more likely to be renters21
-- so Black families are at greater risk in the absence of
renter protections.22
The threat of eviction particularly harms Black women.23 In Massachusetts, Black women renters
were three times more likely than white renters to have an eviction filed against them that
ultimately resulted in dismissal, making Black women "more likely to be denied housing due to
prior eviction filings, even when they won" their cases.24
While the CDC federal eviction moratorium has been a critical public health intervention, the
Trump administration undermined its intent and effectiveness both in the original order and in
subsequent guidance issued in October 2020.25 Eligible renters do not automatically receive
eviction protection under the order, yet the CDC did not conduct public outreach to inform
renters they must take action to receive protection. Additionally, the CDC did not work with
other agencies to ensure enforcement of the moratorium, even as local jurisdictions defied the
order26 and landlords and courts evicted covered renters.27 The October 2020 guidance further
weakened the order, stating it only prevents "the actual eviction of a covered person," but allows
landlords to begin eviction proceedings. The guidance also states that "landlords are not required
to make their tenants aware of the Order and Declaration." This allows for landlords to evict a
tenant who should be protected by the moratorium but is unaware they must take action -- a
problem compounded by the CDC's failure to conduct any renter outreach.
Federal Regulation of Tenant Screening Technologies
Under the Fair Credit Reporting Act (FCRA), tenant screening companies are considered
specialized consumer reporting agencies (CRAs) and must "follow reasonable procedures to
assure maximum possible accuracy of the information concerning the individual about whom
the report relates."28 However, numerous lawsuits have resulted in fines for CRAs -- including tenant screening companies -- for violating this requirement.29 In 2018, the Federal Trade
Commission (FTC) fined the company RealPage $3 million for its use of error prone wild-card
searches; the same company "brings in about $48 million annually" from producing these
inexpensive tenant screening reports.
30
Landlords are required to inform tenants if they were turned down because of a negative report
and the name of the company used; screening companies then have 30 days to respond if a tenant
requests a correction to their report but, "[i]n responses to lawsuits, tenant screening companies
say renters dispute fewer than 1 percent of reports."31 In a deposition for a federal lawsuit, an
employee from one company, CoreLogic, said that independently verifying reports before
sending them to landlords "would be an overwhelming task."32 This assessment may be at odds
with CRAs' requirements under FCRA, and suggests these companies are particularly ill-suited
to navigate their requirements with the surge in eviction filings in 2020.
Despite lawsuits at the state and federal level, tenant screening companies continue to produce
flawed products that are highly consequential for millions of renters, some of whom have faced
homelessness as a result of errors in these tenant reports.33 The CFPB and the FTC jointly
enforce compliance with the FCRA. Effective oversight of these companies requires proactively
investigating and auditing their effects on protected classes, auditing their methods for
generating these reports to ensure accuracy and compliance in the use of data, and the company's
provision of guidance for landlords on the accuracy in utilizing these technologies.
In the absence of federal regulation, some cities and states have recognized the harm tenant
screening companies can cause and have enacted laws to seal or expunge eviction records in
some circumstances, or regulated the content of tenant screening reports, or the ability of
landlords to use publicly reported eviction data in choosing tenants.34 It is vital, with the recent
wave of evictions nationwide caused by the pandemic, that these companies do not further
disadvantage Americans looking to find safe and reliable housing. In order to ensure that a
handful of companies using questionable methods to screen renters are not unfairly preventing
millions of Americans from finding future housing, we ask that you provide answers to the
following questions no later than March 14, 2021:
1) Describe the CFPB's jurisdiction over tenant screening companies and how the Bureau
coordinates with the FTC to examine and enforce compliance with the FCRA How does the CFPB prioritize and evaluate risk when determining which tenant
screening companies to examine for compliance with the FCRA?
a. Does the CFPB evaluate the following when making such a decision, and if so,
how?
i. The way in which these companies obtain data
ii. The techniques used in matching data records to potential tenants
iii. The use of algorithmic models
3) The FCRA requires CRAs to have "reasonable procedures in place to ensure maximum
possible accuracy." Describe how the CFPB evaluates the effectiveness of tenant
screening companies' procedures and how the Bureau evaluates whether such procedures
are ensuring the maximum possible accuracy.
4) Algorithmic design decisions, like the use of wild-cards and the inclusion of whether an
eviction has been filed without including data on the outcome of the eviction, have
contributed to inaccuracy plaguing many tenant screening reports.
a. What additional factors contribute to inaccuracy in reports?
b. What internal processes does CFPB recommend to ensure accuracy among
screening report algorithms?
c. Does the CFPB have specialized procedures in place to evaluate tenant screening
companies' algorithms to ensure these algorithms are accurate?
5) Does the CFPB examine tenant screening companies' algorithms and data collection
methods for compliance with fair lending laws?
a. Is there evidence that current tenant screening software algorithms and data
collection methods discriminate against communities of color?
b. Is there evidence that current tenant screening software algorithms and data
collection methods discriminate against women?
6) What steps will the CFPB take to ensure companies using eviction data from 2020 and
2021 are not reporting data despite a federal, state, or local moratorium at the time of
eviction filing?


Source
arrow_upward