GAO 23 105410 Emergency Rental Assistance Risk Assessment

Summary

A U.S. Government Accountability Office report to congressional committees, GAO-23-105410, titled Emergency Rental Assistance: Treasury's Oversight Is Limited by Incomplete Data and Risk Assessment, dated December 20, 2022. The report examines the Department of the Treasury's Emergency Rental Assistance (ERA) program, which provides nearly $46.55 billion to tribal, state, territorial and local governments. It reports that by June 2021 about one-quarter of grantees had not made any payments, and that Treasury reallocated about $3.1 billion in ERA1 funds by the end of October 2022. It finds that Treasury has not collected or reported complete data as required and that its improper payment risk assessment did not address potentially duplicative payments. GAO makes three recommendations to Treasury, which agreed with them.

Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used

Full text

United States Government Accountability Office

Report to Congressional Committees

December 2022

EMERGENCY
RENTAL
ASSISTANCE
Treasury’s Oversight
Is Limited by
Incomplete Data and
Risk Assessment

GAO-23-105410

December 2022

EMERGENCY RENTAL ASSISTANCE
Treasury’s Oversight Is Limited by Incomplete Data
and Risk Assessment
Highlights of GAO-23-105410, a report to
congressional committees

Why GAO Did This Study

What GAO Found

Congress twice authorized funding for
the ERA program in response to
financial and housing instability
resulting from the COVID-19
pandemic. Because of the emergency
nature and expedited implementation
of the program, Treasury had to
develop program guidance and
oversight procedures as grantees were
beginning to make payments. GAO
previously reported on the program’s
early implementation challenges and
the need for effective oversight of
grantees and payments.

The Department of the Treasury administers the Emergency Rental Assistance
(ERA) program, which provides nearly $46.55 billion to tribal, state, territorial,
and local governments (grantees) to help low-income households affected by the
COVID-19 pandemic pay rent and utilities. By June 2021—5 months into the
program—about one-quarter of grantees had not made any payments. Multiple
factors contributed to payment delays, based on GAO’s interviews with grantees
and review of their documentation. These factors included limited grantee staff
and technology resources, gathering information for tenant eligibility
determinations, and unclear program guidance. Grantees had spent less than
half of the available funding by Treasury’s deadline to begin reallocating excess
funds (September 30, 2021).

The CARES Act includes a provision
for GAO to monitor federal efforts to
respond to the COVID-19 pandemic.
This report examines (1) factors that
affected the timeliness of ERA
payments, (2) reallocation of excess
ERA funds, (3) ERA recipient and
grantee spending characteristics, and
(4) Treasury’s ERA data collection and
oversight efforts.
GAO analyzed ERA payment and
demographic data, as well as
reallocation documentation; reviewed
relevant federal laws and agency
documentation; and interviewed
officials from Treasury and 21 grantees
selected based on expenditures,
government type, and geography.

Treasury’s reallocation of excess funds did not consistently align with renters’
needs. Rebalancing the distribution of funds was critical to maximizing their use
because renters’ needs and grantees’ capacity to deliver assistance varied
across states. Treasury reallocated about $3.1 billion in ERA1 funds by the end
of October 2022, but its prioritization of transfers between grantees in the same
state limited its ability to address large funding differences across states and
better align payments to the needs of renters and grantee capacity.
Available data suggest that ERA payments served low-income renters and varied
to meet local needs. According to limited Treasury data, 85 percent of
households served by the first ERA appropriation had incomes below 50 percent
of the area median, consistent with a legal requirement to prioritize such renters.
Data also suggest grantees used funds to address local needs. Households
received larger average payments in urban areas ($7,200) than in rural areas
($5,200) and in counties with higher rents and more rent-burdened households.
Emergency Rental Assistance (ERA1) Distribution by Income Level, Quarter 4, 2021

What GAO Recommends
GAO is making three
recommendations to Treasury to
complete a detailed assessment of
improper payment risks and improve
ERA data collection and reporting.
Treasury agreed with GAO’s
recommendations.

View GAO-23-105410. For more information,
contact Daniel Garcia-Diaz at (202) 512-8678
or garciadiazd@gao.gov.

Oversight of the ERA program would benefit from improved data collection and
assessment of improper payment risks. Although Treasury has taken recent
steps to improve data collection and reporting, including issuing detailed final
reporting requirements, to date the agency has not collected or reported
complete data as required by the authorizing statute. Without better data
collection and reporting, Congress and Treasury will lack information on the
program’s outcomes. And although Treasury completed a required assessment
of improper payment risks for the program, the assessment focused on
allocations to grantees and did not address missing payment data and potentially
duplicative payments to households. Without a more detailed assessment of
improper payment risks, Treasury’s awareness of such risks and oversight of the
ERA program will be limited.
United States Government Accountability Office

Contents

Letter

1
Background
Limited Grantee Capacity and Other Implementation Challenges
Slowed Payments to Renters
Treasury’s Reallocation Process Did Not Consistently Benefit
States with Greater Renter Needs
Available Data Suggest the ERA Program Served Low-Income
Renters, Including Some Facing Eviction
Data Are Incomplete and Treasury Has Not Fully Assessed
Improper Payment Risks
Conclusions
Recommendations for Executive Action
Agency Comments

36
45
46
46

Appendix I

Objectives, Scope, and Methodology

Appendix II

Comments from the Department of the Treasury

Appendix III

GAO Contact and Staff Acknowledgments

10
16
26

Tables
Table 1: Emergency Rental Assistance (ERA1) Allocations per
Low-Income Renter in Highest and Lowest States, Before
and After Reallocation
Table 2: Percentage of Missing Data and Nonreporting Grantees
for Emergency Rental Assistance Data, by Reporting
Period

23
37

Figures
Figure 1: Emergency Rental Assistance Program Expenditure
Ratio, by Month and Appropriation, January 2021–June
2022
Figure 2: Emergency Rental Assistance (ERA1) Allocations per
Low-Income Renter by State
Figure 3: Emergency Rental Assistance Funds Reallocated by
Payment Type, as of October 2022

Page i

11
18
20

GAO-23-105410 Emergency Rental Assistance

Figure 4: Percentage Change in Emergency Rental Assistance
(ERA1) Allocations by State, as of October 2022
Figure 5: Quartiles of Low-Income Renter Households That
Received Emergency Rental Assistance (ERA1 and
ERA2), by County, 2021
Figure 6: Percentage of Households That Received Emergency
Rental Assistance (ERA1), by Race, Fourth Quarter 2021
Figure 7: Emergency Rental Assistance (ERA1 and ERA2)
Payments to Households by Eligible Use, 2021
Figure 8: Quartiles of Emergency Rental Assistance (ERA1 and
ERA2) Payments to Households for Rent or Utility
Arrears, by County, 2021

22
27
29
33
34

Abbreviations
ERA
HUD
OMB

Emergency Rental Assistance
Department of Housing and Urban Development
Office of Management and Budget

This is a work of the U.S. government and is not subject to copyright protection in the
United States. The published product may be reproduced and distributed in its entirety
without further permission from GAO. However, because this work may contain
copyrighted images or other material, permission from the copyright holder may be
necessary if you wish to reproduce this material separately.

Page ii

GAO-23-105410 Emergency Rental Assistance

Letter

441 G St. N.W.
Washington, DC 20548

December 20, 2022
Congressional Committees
Congress appropriated $46.55 billion for the Emergency Rental
Assistance (ERA) program to assist low-income renters and their
landlords during a period of increased financial and housing instability
caused by the COVID-19 pandemic. 1 The Department of the Treasury
was required to disburse the funds quickly to over 700 tribal, state,
territorial, and local governments (grantees) to provide assistance
payments for renters. Because of the emergency nature and expedited
implementation of the program, Treasury had to develop program
guidance and oversight procedures as grantees were beginning to make
payments.
We previously reported on issues with administration and oversight in the
ERA program. As grantees were beginning to make assistance payments
in March 2021, we reported on the need for Treasury to provide clear and
timely guidance and appropriate oversight of grantees to manage the risk
of improper payments. 2 We subsequently reported on administrative
flexibilities that increased improper payment risks in the ERA program
and related oversight challenges. 3
The CARES Act includes a provision for us to monitor and oversee
federal efforts to prepare for, respond to, and recover from the COVID-19

1In December 2020, the Consolidated Appropriations Act, 2021 authorized Treasury to

disburse about $25 billion to ERA grantees, and in March 2021 the American Rescue Plan
Act of 2021 authorized Treasury to disburse about $21.55 billion. We refer to the two
appropriations (ERA1 and ERA2) as the ERA program for convenience. Treasury uses the
same approach but considers each appropriation to be a separate program managed by
the same office.

2See GAO, COVID-19 Housing Protections: Moratoriums Have Helped Limit Evictions, but

Further Outreach Is Needed, GAO-21-370 (Washington, D.C.: Mar. 15, 2021); and
COVID-19: Additional Actions Needed to Improve Accountability and Program
Effectiveness of Federal Response, GAO-22-105051 (Washington, D.C.: Oct. 27, 2021).

3GAO, Emergency Rental Assistance: Additional Grantee Monitoring Needed to Manage

Known Risks, GAO-22-105490 (Washington, D.C.: Feb. 10, 2022); and COVID-19:
Significant Improvements Are Needed for Overseeing Relief Funds and Leading
Responses to Public Health Emergencies, GAO-22-105291 (Washington, D.C.: Jan. 27,
2022).

Page 1

GAO-23-105410 Emergency Rental Assistance

pandemic. 4 We were also asked to review how Treasury has
administered and overseen the ERA program. This report continues our
reporting on ERA and examines (1) factors that affected the timeliness of
ERA payments and grantee spending, (2) Treasury’s reallocation of
excess funds and its effect on grantees, (3) characteristics of recipient
households and grantee spending, and (4) Treasury’s oversight of
grantee data reporting and improper payment risks.
To address our first objective, we reviewed grantee spending trends and
identified spending delays using Treasury’s monthly compliance report
data. 5 We interviewed a nongeneralizable sample of 21 state, local, and
territorial grantees, which we selected to represent a mix of grantees
based on government type, spending rate, and location. We also
conducted on-site visits with five of the selected grantees. We collected
and reviewed 127 program improvement plans (of about 140 in total) from
Treasury, and we categorized the spending obstacles grantees identified
in those plans.
To address our second objective, we used Treasury’s reallocation reports
through October 2022 and data from the Department of Housing and
Urban Development (HUD) to determine how allocations varied before
and after reallocation relative to low-income renter populations in each
state. 6 We also interviewed Treasury officials and reviewed agency data
and documentation to determine how Treasury reallocated excess funds.
We used this information in conjunction with grantees’ requests for
reallocated funds and Treasury documentation (decision memorandums)
to review how reallocation payments aligned with renter needs and
grantee capacity.

4Pub. L. No. 116-136, div. B, § 19010(b), 134 Stat. 281, 579-81 (2020). All of GAO’s

reports related to the COVID-19 pandemic are available on GAO’s website at
https://www.gao.gov/coronavirus.

5We assessed the reliability of Treasury’s monthly data by reviewing technical

documentation, interviewing knowledgeable agency officials, and conducting electronic
testing for outliers and errors. We found the data to be sufficiently reliable for reporting
general spending trends and grantee expenditures.

6To assess the reliability of HUD’s data, we reviewed technical documentation and

determined the data were sufficiently reliable for reporting community characteristics and
population sizes at the county level.

Page 2

GAO-23-105410 Emergency Rental Assistance

To address our third objective, we analyzed Treasury’s ERA payment and
demographic data for calendar year 2021. 7 We analyzed the proportion of
payments by payment type and recipient type, as well as geographic
spending trends. To describe differences in ERA spending by county, we
identified recipients’ counties and matched them to county-level
characteristics from the Census Bureau’s American Community Survey 5year estimates for 2016–2020. 8 We also reviewed payments relative to
low-income population using HUD data, as well as payments by urban
and rural area using data from the Department of Agriculture’s Economic
Research Service. 9 We reviewed the relationship between county-level
characteristics and total ERA payments by county to determine whether
payments were associated with certain characteristics. Furthermore, we
analyzed demographic data to describe the characteristics of households
served by the program.
To address our fourth objective, we reviewed payment and demographic
data grantees reported for calendar year 2021, as well as Treasury’s
public reporting of data through 2022. We assessed the completeness
and reliability of the data by reviewing summary statistics for key
variables, reviewing technical documentation, and interviewing Treasury
officials responsible for the data. 10 We identified potentially duplicative
payments, which we first cleaned using the U.S. Postal Service’s Address
Management System. We compared the completeness and reliability of
Treasury’s data and its reporting of the data against statutory
requirements for data collection and reporting published in the
Consolidated Appropriations Act, 2021. We also interviewed Treasury
7As discussed for our fourth objective, we assessed the reliability of Treasury’s payment

and demographic data by reviewing summary statistics for key variables, reviewing
technical documentation, and interviewing Treasury officials responsible for the data.
While we include analysis in this report based on these data, significant portions are
missing or erroneous. We accounted for and note these limitations in our analysis.

8To assess the reliability of the Census data, we reviewed technical information and

determined the data were sufficiently reliable for reporting community characteristics and
population sizes at the county level. Findings from each survey are subject to sampling
errors.

9To assess the reliability of the Department of Agriculture data, we reviewed technical

documentation and determined the data were sufficiently reliable for categorizing
population density by county.

10After we provided Treasury with a draft of this report for comment on November 1, 2022,

the agency released updated data on required performance measures and demographics.
We updated our analysis for this objective based on these data; however, given the timing
of the data release, we did not update our analysis of grantee spending and recipient
characteristics in other sections of this report.

Page 3

GAO-23-105410 Emergency Rental Assistance

officials and reviewed agency documentation on activities related to
managing improper payment risk, and we compared these efforts to
payment integrity requirements for federal agencies and federal internal
control standards for risk assessment. See appendix I for more
information on our scope and methodology.
We conducted this performance audit from September 2021 to December
2022 in accordance with generally accepted government auditing
standards. Those standards require that we plan and perform the audit to
obtain sufficient, appropriate evidence to provide a reasonable basis for
our findings and conclusions based on our audit objectives. We believe
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.

Background

Appropriations,
Allocations,
Disbursements, and
Grantee Expenditures

Congress twice appropriated funding for the ERA program. In December
2020, the Consolidated Appropriations Act, 2021 authorized Treasury to
disburse about $25 billion to remain available to grantees until September
30, 2022 (referred to as ERA1 by Treasury). 11 In March 2021, the
American Rescue Plan Act of 2021 authorized Treasury to disburse about
an additional $21.55 billion to remain available to grantees until
September 30, 2025 (referred to as ERA2 by Treasury). 12 Treasury’s
Office of Recovery Programs administers the ERA program and shares

11Pub. L. No. 116-260, div. N, tit. V, § 501, 134 Stat. 1182, 2069-78 (2020) (to be codified
at 15 U.S.C. § 9058a). The statute reserved $15 million of the ERA1 appropriation for
administrative expenses of the Secretary of the Treasury. The American Rescue Plan Act
of 2021 extended the availability of ERA1 funds from December 30, 2021, to September
30, 2022. Pub. L. No. 117-2, tit. III, § 3201(h), 135 Stat. 4, 58. Grantees may request an
extension to continue obligating ERA1 funds received through reallocation until December
29, 2022.
12American Rescue Plan Act of 2021 §§ 3201(a),(g) (to be codified at 15 U.S.C. §§
9058c(a),(g)). The law reserved $30 million of the ERA2 appropriation for administrative
expenses and technical assistance made by the Secretary of the Treasury.

Page 4

GAO-23-105410 Emergency Rental Assistance

some monitoring and oversight authority with Treasury’s Office of
Inspector General. 13
The ERA1 authorizing statute required Treasury to allocate funds to each
state proportionate to its share of the U.S. population. 14 The law also
required that no state receive less than $200 million. In addition, city and
county governments with more than 200,000 residents were eligible to
receive a portion of their state’s allocation based on their share of the
population. 15
The ERA2 authorizing statute required Treasury to use an allocation
formula similar to that for ERA1, but it did not include funding for tribes or
tribally designated housing entities. 16 Unlike ERA1, the ERA2 statute
required Treasury to initially disburse not less than 40 percent of each
grantee’s allocation for ERA2. 17 Grantees could request the remainder of
their ERA2 allocation in two payments after obligating at least 75 percent
of the amount already disbursed. 18 Additionally, $2.5 billion of the ERA2
funding was set aside for payments to high-need grantees based on
rental market costs, change in employment, and the number of very-low13Consolidated Appropriations Act, 2021 § 501(i) (to be codified at 15 U.S.C. § 9058a(i)).
The Consolidated Appropriations Act, 2021 provided the Office of Inspector General with
$6.5 million to monitor and oversee the disbursement, receipt, and use of ERA1 funds and
to recoup certain funds spent in violation of eligible uses. For ERA2 funds, the American
Rescue Plan Act of 2021 provided the Office of Inspector General with $3 million for
oversight but did not provide monitoring or recoupment authority. American Rescue Plan
Act of 2021 § 3201(a)(2)(C) (to be codified at 15 U.S.C. § 9058c(a)(2)(C)).
14Consolidated Appropriations Act, 2021 § 501(b) (to be codified at 15 U.S.C. § 9058a(b)).
15Allocations to territories were based on their share of the total territorial population,
although the sum of the amounts allocated for territories other than Puerto Rico could not
be less than $75 million. Allocations for tribes and tribally designated housing entities were
based on the amounts they were eligible to receive for fiscal year 2020 from the Indian
Housing Block Grant program. The Department of Hawaiian Home Lands was directly
allocated $2.4 million in the statute. Consolidated Appropriations Act, 2021 § 501(b)(2)-(3)
(to be codified at 15 U.S.C. § 9058a(b)(2)-(3)).
16American Rescue Plan Act of 2021 § 3201(b) (to be codified at 15 U.S.C. § 9058c(b)).
17American Rescue Plan Act of 2021 § 3201(c)(1) (to be codified at 15 U.S.C. §
9058c(c)(1)).
18Treasury provided information on the drawdowns on its web site. See “Instructions on

Drawing Down Additional Tranche Payments for ERA2 Awards,” accessed at
https://home.treasury.gov/policy-issues/coronavirus/assistance-for-state-local-and-tribalgovernments/emergency-rental-assistance-program/allocations-and-payments.

Page 5

GAO-23-105410 Emergency Rental Assistance

income renter households paying more than 50 percent of their income
on rent or living in substandard or overcrowded conditions. 19
Treasury began disbursing ERA1 and ERA2 allocations to grantees in
January and May 2021, respectively, based on Treasury payment data.
As of August 2022, Treasury had disbursed about $42 billion of the nearly
$46.55 billion in total ERA funding available for grantees, including all of
the ERA1 appropriation and about $17.1 billion of ERA2 (about 79
percent). Grantees had expended (spent) about 69 percent ($32 billion) of
the ERA1 and ERA2 appropriations, based on Treasury data through
June 2022.

Grantee Responsibilities
and Program
Requirements

Grantees are responsible for establishing policies and procedures to
accept applications and screen households for eligibility; making
payments to landlords and utility providers on behalf of eligible renter
households; and designing controls to deter the misuse of funds.
Grantees are to administer their ERA programs in accordance with
requirements of the two authorizing statutes and Treasury’s guidance. In
addition, grantees must report certain data to Treasury on a monthly and
quarterly basis, including detailed information on their ERA program
activities, expenditures, recipient demographics, and compliance issues. 20
Grantees were to primarily use their allocations to provide financial
assistance payments to landlords and utility providers on behalf of eligible
renter households for past due (arrears) or current and future
(prospective) rent and utilities, as allowed by law. 21 The ERA1 and ERA2
statutes required grantees to establish a preference system that
prioritized assistance to households with incomes not exceeding 50

19American Rescue Plan Act of 2021 § 3201(b)(3) (to be codified at 15 U.S.C. §
9058c(b)(3)).
20Department of the Treasury, Reporting Guidance: Emergency Rental Assistance
Program (Washington, D.C.: Mar. 29, 2022). Consolidated Appropriations Act, 2021 §
501(g) (to be codified at 15 U.S.C. § 9058a(g)).
21Households that were eligible to receive ERA assistance were generally those that (1)
experienced a financial hardship directly or indirectly due to or during the COVID-19
pandemic or qualified for unemployment benefits, (2) demonstrated a risk of housing
instability or homelessness, and (3) were low income, defined as having household
income not exceeding 80 percent of their area median or as established by the
Department of Housing and Urban Development.

Page 6

GAO-23-105410 Emergency Rental Assistance

percent of their area median or with one or more members who were
unemployed for at least 90 days. 22
Under ERA1, eligible households could receive up to 12 months of
assistance plus an additional 3 months to ensure housing stability, if
necessary. When combined with financial assistance under ERA2,
eligible households cannot receive more than a total of 18 months of
assistance. In addition, grantees can use 10 percent of their ERA funds to
provide housing stability services, such as eviction diversion programs
and housing counseling. For ERA1, grantees also were allowed to use up
to 10 percent of their funds (up to 15 percent for ERA2) for administrative
expenses.

Treasury Guidance and
Reallocation

Treasury provides program guidance and requirements to grantees
through a series of frequently asked questions. 23 Treasury published the
guidance in January 2021 and has periodically revised and supplemented
it since, as recently as July 2022. The guidance communicated
requirements in the ERA statutes, as well as additional requirements and
flexibilities created and implemented by Treasury. For example, the
guidance specified outreach requirements before grantees could make
payments directly to renters, and it allowed grantees to accept written
attestations from renters that their households met eligibility
requirements. Treasury also separately published guidance on reporting
requirements in June 2021, which the agency also updated periodically.
Furthermore, the agency provided grantees with additional resources,
including promising practices, sample forms, and guidelines for program
and service design.
In addition, Treasury published guidance on reallocating excess ERA1
and ERA2 funds. 24 For ERA1, Treasury was required to recapture excess
unobligated funds beginning on September 30, 2021, and reallocate them
22Consolidated Appropriations Act, 2021 § 501(c)(4) (to be codified at 15 U.S.C. §
9058a(c)(4)). American Rescue Plan Act of 2021 § 3201(d)(2) (to be codified at 15 U.S.C.
§ 9058c(d)(2)).
23Department of the Treasury, Emergency Rental Assistance Frequently Asked Questions
(Washington, D.C.: July 27, 2022).
24Department of the Treasury, Emergency Rental Assistance under the Consolidated
Appropriations Act, 2021: Reallocation Guidance (Washington, D.C.: Sept. 6, 2022); and
Emergency Rental Assistance under the American Rescue Plan Act of 2021 (ERA2):
Reallocation Guidance (Washington, D.C.: June 28, 2022).

Page 7

GAO-23-105410 Emergency Rental Assistance

to grantees that obligated at least 65 percent of their allocations. 25 The
ERA1 statute provided Treasury with discretion on how to determine
excess funds and reallocate them to grantees that met the obligations
benchmark.
For ERA2, Treasury was required to begin reallocating excess funds
beginning on March 31, 2022, to grantees that had obligated at least 50
percent of their allocations. For ERA2, Treasury can only reallocate funds
that were allocated to grantees but not yet disbursed to them. 26 In
general, Treasury defined excess funds as the difference between a
grantee’s expenditures and the amount needed to reach a certain
spending target. For example, grantees were required to have spent 30
percent of their initial ERA1 allocation (not including the portion allocated
for administrative expenses) by September 30, 2021, to avoid having
excess funds subject to recapture. 27

Our Prior Work and
Recommendations

We published four reports on and made three recommendations related
to administration and oversight challenges in the ERA program. In March
2021, we reported on Treasury’s early efforts to implement the ERA
program and update its guidance to address concerns with its
comprehensiveness and clarity. 28 We also highlighted the need to
balance efforts to expedite payments with appropriate controls and
monitoring to help limit improper payment risks.
In October 2021, we reported on grantee spending challenges and
Treasury’s efforts to help improve payment rates, including updating
guidance to allow grantees to adopt administrative flexibilities that
introduce improper payment risks, such as self-attestation of eligibility and
shortening the required landlord outreach period before grantees can
directly pay households. 29 We continued to emphasize the need for
25We refer to these as excess funds throughout this report. Consolidated Appropriations
Act, 2021 § 501(d) (to be codified at 15 U.S.C. § 9058a(d)). For ERA1 and ERA2,
Treasury’s reallocation guidance clarifies that the agency will consider ERA funds to be
obligated if the grantee spent the funds on financial assistance and housing stability
services or committed the funds under certain circumstances.
26American Rescue Plan Act of 2021 § 3201(e) (to be codified at 15 U.S.C. § 9058c(e)).
27Department of the Treasury, Emergency Rental Assistance under the Consolidated
Appropriations Act, 2021: Reallocation Guidance (Washington, D.C.: Sept. 6, 2022).
28GAO-21-370.
29GAO-22-105051.

Page 8

GAO-23-105410 Emergency Rental Assistance

grantees and Treasury to implement appropriate payment controls and
oversight.
In January 2022, we reported that Treasury had not developed processes
to identify and recover overpayments made by grantees and
recommended that Treasury implement such processes. 30 Treasury
agreed with this recommendation and stated that it was working to
establish post-payment reviews and recovery audit activities within the
schedule prescribed in Treasury’s Implementation Guide for Office of
Management and Budget (OMB) Circular A-123, Appendix C:
Requirements for Payment Integrity. We reiterated the importance of this
recommendation when we issued our list of priority open
recommendations to Treasury in May 2022. 31 As of September 2022,
Treasury had not implemented this recommendation. 32
In our January 2022 report, we also recommended that OMB, in
consultation with Treasury, issue single audit guidance to help auditors
identify deficiencies in grantees’ programs. OMB and Treasury
implemented this recommendation by issuing an update to the 2022
Compliance Supplement in May 2022. 33
In February 2022, we reported that Treasury had not developed
procedures to monitor and evaluate the controls grantees are required to
implement when relying on self-attestations. 34 Treasury required grantees
to implement reasonable validation and fraud-prevention procedures
when relying on self-attestations, but it had not taken steps to monitor
how grantees had implemented the procedures and whether they were
effectively managing improper payment risks. We recommended that
Treasury develop and implement procedures to monitor and evaluate
grantees’ controls and include information on the minimum internal
30GAO-22-105291.
31GAO, Priority Open Recommendations: Department of the Treasury, GAO-22-105633

(Washington, D.C.: May 10, 2022).

32In October 2022, Treasury provided GAO with a completed risk assessment and cost-

effectiveness analysis pursuant to OMB Circular A-123, Appendix C as well as examples
of the agency’s implementation of its compliance testing procedures. We will continue to
monitor and evaluate Treasury’s efforts to address this recommendation.

33Office of Management and Budget, 2 CFR Part 200, Appendix XI, Compliance
Supplement (Washington, D.C.: April 2022).
34GAO-22-105490.

Page 9

GAO-23-105410 Emergency Rental Assistance

control systems expected for ERA grantees that employ self-attestation.
Treasury said it planned to update its monitoring procedures, but as of
September 2022 had not implemented this recommendation. 35

Limited Grantee
Capacity and Other
Implementation
Challenges Slowed
Payments to Renters

Several challenges slowed spending during the early stages of ERA’s
implementation, based on our analysis of data and documentation from
Treasury, and interviews with grantees. 36 ERA was a new program, and
grantees had to design programs, assess and build capacity, and conduct
outreach before they were able to screen applicants and make
payments. 37 Grantees also experienced other challenges after programs
began operating, which stemmed from untimely and unclear guidance,
difficulty documenting tenant eligibility and securing landlord participation,
and ineffective coordination with other grantees to avoid duplication.
Although there was significant renter and landlord need, nearly onequarter of ERA1 grantees (22 percent) had not made any payments at the
end of May 2021 (5 months into the program), and the overall expenditure
ratio among all grantees was 7 percent (see fig. 1). 38 The overall
35In response to this recommendation, Treasury has developed some grantee monitoring
procedures, including a Single Audit Act Compliance Supplement for ERA. We will
continue to monitor and evaluate Treasury efforts to address this recommendation.
36We selected a nongeneralizable sample of 21 grantees to interview to represent a mix of
grantees based on whether they met Treasury’s 30-percent spending target, type (state,
territorial, or local grantees), and Census division and region. We also reviewed program
improvement plans, which grantees were required to submit if they did not meet
Treasury’s spending target by September 30, 2021. We categorized the key spending
obstacles that grantees had to identify in the plans. See app. I for more information on our
methodology.
37In October 2021, we reported that most grantees could not deploy funds immediately
because they needed additional time to enhance existing rental assistance programs or
develop new programs to accommodate significant increases in scales of operations that
ERA1 funding made possible. Grantees had to develop new policies and procedures, hire
additional staff, and develop electronic application and data collection systems. Some
grantees also experienced overwhelming demand for ERA1 funds and lacked sufficient
staff capacity to address applications in a timely manner. Payments were also delayed in
some areas while state and local legislatures deliberated how to administer the new funds.
See GAO-22-105051.
38Treasury’s ERA1 reallocation guidance defines a grantee’s expenditure ratio as its ratio
of expenses for financial assistance to its total allocation (less the portion available for
administrative costs). Treasury also subtracts any amounts previously recaptured or
transferred from the initial allocation when calculating the expenditure ratio. The ERA2
reallocation guidance uses a similar calculation to determine the expenditure ratio but
accounts for differences in how ERA2 grantees may use funds for administrative
expenses and housing stability services.

Page 10

GAO-23-105410 Emergency Rental Assistance

expenditure ratio subsequently increased but was less than 50 percent at
the end of September 2021 when Treasury was required to begin
reallocating excess funds. For ERA2 funds, the overall expenditure ratio
at that time was 4 percent. 39
Figure 1: Emergency Rental Assistance Program Expenditure Ratio, by Month and Appropriation, January 2021–June 2022

Note: Expenditure ratios generally include grantees’ expenses for financial assistance payments as a
percentage of their initial allocation (less the portion of the allocation available for administrative costs
or recaptured). Treasury collected and reported grantee expenditures for financial assistance
payments on a quarterly basis from January to March 2021 and from April to June 2021 for the first
round (ERA1) and second round (ERA2) of funding, respectively. We divided these values by three to
calculate monthly expenditure ratios. We excluded tribes and tribally designated housing entities
because of data limitations.

Based on interviews we conducted with 21 grantees and our review of
127 grantee program improvement plans, we identified seven factors that
affected the timeliness of grantees’ early payments: (1) limited staff and
technology resources, (2) difficulty collecting documentation to determine
eligibility, (3) unclear program guidance and limited technical assistance,
39According to Treasury, grantees generally prioritized spending ERA1 funds before ERA2
funds because of the program’s shorter period of availability.

Page 11

GAO-23-105410 Emergency Rental Assistance

(4) local legislative involvement, (5) limited landlord participation, (6)
overlapping grantee jurisdictions, and (7) funding allocations that
exceeded local need.
Staff and technology resources. More than half of the state and local
grantees we interviewed cited a lack of sufficient staff and technology
resources to administer an ERA program as a challenge to making timely
payments. Among the 21 grantees we interviewed, 13 identified staffing
and capacity and eight identified technology issues as key factors that
limited their ability to make timely payments. The initial volume of
applications overwhelmed some grantees, requiring them to hire
additional staff. Most grantees (17 of 21) also contracted with third-party
vendors to administer or design aspects of their program, such as policies
and procedures, application portals, application screening and approval,
and payment processing. These contracting processes added to their
implementation timelines.
Eligibility determinations. Grantees most commonly cited the collection
of required information for eligibility determinations, including efforts to
obtain complete documentation from applicants, as an obstacle to making
timely payments in the program improvement plans we reviewed.
Specifically, 46 percent of the plans cited documentation requirements as
an obstacle, and 24 percent citied duplication of benefits reviews. In
addition, 13 percent of plans cited applicant responsiveness to requests
for required information as an obstacle. Some grantees similarly noted
that determining eligibility was time consuming, in part because it required
frequent contact with renters, landlords, and other grantees to collect
documentation and confirm that payments would not be duplicative.
Treasury guidance and technical assistance. Eight of the 21 grantees
we interviewed cited issues with the timeliness, clarity, and frequent
updating of Treasury’s guidance and reporting requirements. In addition,
20 percent of the program improvement plans we reviewed cited
Treasury’s guidance as an obstacle to making timely payments. Treasury
released ERA program guidance in January 2021, significantly revised it
after the presidential transition that month, and updated it another five
times (through August 2021). Notable revisions included ongoing
clarification of administrative flexibilities around self-attestation of
eligibility and modifications to reporting requirements.
Some grantees told us program implementation was delayed because
Treasury was slow to develop guidance. For example, one grantee told
us that the early frequently asked questions lacked enough parameters

Page 12

GAO-23-105410 Emergency Rental Assistance

for it to feel comfortable designing a program, and Treasury’s changing
reporting requirements required it to reassign staff from processing
applications to retroactively compiling data to meet updated reporting
requirements.
In addition, grantees generally told us the technical assistance they
received from Treasury was limited. Among the 20 grantees we
interviewed that requested technical assistance, 11 said that Treasury’s
responses were not timely or did not meet their needs, or both. For
example, several grantees that requested technical assistance on
interpretations of the guidance told us that Treasury responded by
directing them to its frequently asked questions.
Local legislative involvement. Some grantees experienced delays
stemming from required involvement of local legislatures, such as state
legislatures or city councils, in approving the use of ERA funds,
contracting decisions, and other requirements. Specifically, 8 of the 21
grantees told us they experienced implementation delays while waiting for
legislative approval. For example, one grantee’s program was delayed for
about 6 months waiting for its local legislature to resolve a disagreement
on approval of a third-party vendor.
Landlord participation. About half of the grantees we interviewed (10 of
21) said they initially experienced issues with landlords’ participation,
which made it challenging to establish tenant eligibility and make timely
payments. In addition, 22 percent of the program improvement plans we
reviewed cited limited landlord participation as an obstacle. Eighteen
percent cited a lack of awareness of ERA programs among landlords and
renters. Landlord participation may have been limited for several reasons,
such as hesitancy to accept government assistance or preferring to
replace the tenant through eviction or expiration of lease.
Grantees varied in the use of flexibilities and promising practices
Treasury introduced to help grantees assist tenants when landlord
participation was limited. Treasury updated its guidance to shorten the
required outreach period before grantees could make payments directly

Page 13

GAO-23-105410 Emergency Rental Assistance

to tenants when landlords would not accept payment. 40 However, fewer
than half of the grantees we interviewed (10 of 21) allowed such
payments, and rarely cited them as significantly increasing payment
rates. According to payment data grantees submitted to Treasury for
calendar year 2021, grantees made the vast majority of rent payments to
landlords (86 percent), rather than renter households directly (13
percent). 41
Treasury—in an effort to help grantees increase payment rates—updated
its guidance to allow renter self-attestations in lieu of landlord-provided
information to establish eligibility, payment amount, and nonduplication of
benefits. All but one grantee we interviewed adopted self-attestation. In
addition, Treasury encouraged grantees through its promising practices to
engage landlords to increase their awareness of and willingness to
participate in the program. Almost all the grantees we interviewed (19 of
21) pursued such intentional engagement with landlords.
Overlapping grantee jurisdictions. About one-quarter of the program
improvement plans we reviewed and six of the 21 grantees we
interviewed cited overlapping jurisdictions among grantees as an obstacle
to making timely payments. Treasury allocated ERA funds to state
grantees, as well as to city and county grantees with populations over
200,000, and some grantees chose to serve the same geographic area.
Local grantees that cited overlapping jurisdictions as an obstacle
commonly discussed a sense of competition with state grantees that
limited the use of their program and required additional time to review

40ERA1 grantees were required to make reasonable efforts to obtain the cooperation of

landlords and utility providers to accept payments before making payments to renters.
Treasury shortened the required waiting period from 21 days to 7 days if conducted in
writing, or 5 days if conducted electronically. ERA2 grantees are not required to seek the
cooperation of the landlord or utility provider before providing assistance directly to the
tenant, but Treasury strongly encourages them to apply the same ERA1 outreach
requirements if they do.

41Payee type was missing for 1 percent of rent payments. In addition, while 43 percent of

grantees reported providing at least one payment to a renter household, the majority of
such payments were made by two grantees—the State of Texas and State of Florida—
which issued more than 60 percent of all direct-to-tenant payments in 2021. We discuss
these payment data in more detail later in this report.

Page 14

GAO-23-105410 Emergency Rental Assistance

applications to prevent duplicative payments. 42 For example, several local
grantees in the same state discussed how awareness of and participation
in their program was limited because their state’s grantee chose to serve
applicants who resided in their service area.
Funding allocations that exceeded local need. The Consolidated
Appropriations Act, 2021, required that Treasury allocate ERA1 funds to
grantees based on their populations but required that no state receive
less than $200 million (minimum allocation). This effectively provided
more funding per household to states with lower populations. One-third of
states and the District of Columbia received the minimum allocation of
$200 million, but most would have received significantly less based on
their populations. For example, Wyoming received $200 million but would
have received $42 million based on its share of the U.S. population.
Because some grantees received allocations considerably greater than
their need, they struggled to meet spending targets. Four of the 21
grantees we interviewed told us their supply of funds exceeded the needs
of eligible renters and identified the oversupply as a spending obstacle. In
addition, 14 of the 18 state grantees in states that received the minimum
allocation did not have a sufficient expenditure ratio (30 percent) by the
end of September 2021 to meet Treasury’s spending target and avoid
recapture of excess funds.
According to some grantees, allocations exceed the needs of renters for
various reasons. For example, one state grantee that received the
minimum allocation told us that its allocation was too large relative to its
population of eligible renters because, compared to other states, it had
fewer renters and businesses did not shut down to the same extent
during the pandemic. Another state grantee that also received the
minimum allocation told us that it analyzed potential demand for ERA
based on cost burden, renter population, occupancy, and eviction risks
and found that it would have had excess funds even with a relatively high
degree of utilization among potential recipients. We discuss the alignment

42The ERA1 statute requires that, to the extent feasible, grantees ensure that rental
assistance provided not be duplicative of other federally-funded rental assistance, and
Treasury encourages grantees to minimize duplicative assistance using ERA2 funds.
Treasury also encourages grantees with overlapping or contiguous jurisdictions to
collaborate on joint administrative and oversight efforts to avoid making duplicative
payments.

Page 15

GAO-23-105410 Emergency Rental Assistance

of the ERA1 allocation formula with renter needs in more detail in the next
section.

Treasury’s
Reallocation Process
Did Not Consistently
Benefit States with
Greater Renter
Needs

ERA1 allocations did not always align with the needs of states’ lowincome renters. Additionally, Treasury prioritized reallocating excess
funds within states. As a result, Treasury’s ability to target excess funds
to states and grantees based on renter needs and grantee capacity was
limited.

Population-Based ERA1
Funding Formula Was Not
Consistent with Renter
Needs

ERA1 funding did not always align with each state’s need for emergency
rental assistance. 43 As discussed previously, the ERA1 statute required
that Treasury disburse ERA1 allocations to grantees largely based on
their populations, but no state could receive less than $200 million. In
addition, the ERA1 allocation formula did not take into account the
following factors related to the needs of low-income renters:
•

Low-income renter population. The allocations did not directly
account for each state’s share of potential recipients (low-income
renters). The share of low-income renters in several states varied
from their share of the total population. For example, about 6 percent
of the total U.S. population resides in New York, while about 8 percent
of the low-income renter population does, according to data from
Census and HUD. 44

43The ERA2 allocation formula accounted for differences in need across states and
localities. The American Rescue Plan Act of 2021 sets aside $2.5 billion for eligible
grantees with a high need for ERA2 assistance. The act requires Treasury to allocate the
funds based on the number of very-low-income renter households paying more than 50
percent of income on rent or living in substandard or overcrowded conditions, rental
market costs, and change in employment since February 2020. For example, California
received an additional $521 million on top of its population-based allocation ($1.5 billion)
to bolster the allocations of 37 of its 49 eligible local grantees. See American Rescue Plan
Act of 2021 §§ 3201(a)(2)(D),(b)(3) (to be codified at 15 U.S.C. §§ 9058c(a)(2)(D),(b)(3)).
44Total population is based on Census’s 2020 Vintage Population data, and low-income
renter population includes renter households with income at or below 80 percent of their
area median income and is based on HUD’s 2014–2018 Comprehensive Housing
Affordability Strategy data.

Page 16

GAO-23-105410 Emergency Rental Assistance

•

Housing costs and burden. The allocations also did not account for
housing costs and rent burden. For example, the average monthly
rent for a two-bedroom apartment across counties in California in
2022 ($1,659) was more than twice that for Alabama ($814). 45
Differences in the number of cost-burdened households (those paying
more than 30 percent of their income on housing) also existed across
states.

Because the ERA1 allocation formula did not account for these factors,
the total amount of funding available to low-income renters varied
significantly across states. Specifically, the allocation per low-income
renter ranged from $602 in New York to $4,588 in Wyoming. The
minimum allocation also greatly contributed to these differences. All 15
states that received $1,300 or more per low-income renter received the
minimum allocation (see fig. 2).

45To determine average rents, we averaged the 50th Percentile Rent Estimates for 2022
(developed by HUD) across all counties in each state.

Page 17

GAO-23-105410 Emergency Rental Assistance

Figure 2: Emergency Rental Assistance (ERA1) Allocations per Low-Income Renter by State

Note: Low-income renters include renter households with income at or below 80 percent of their area
median income, based on the Department of Housing and Urban Development’s Comprehensive
Housing Affordability Strategy data for 2014–2018.

Treasury Reallocated Most
Funds within States

Congress provided Treasury with an opportunity to rebalance the
distribution of ERA funds, which was critical to maximizing their use
because renters’ needs and grantees’ capacity to deliver assistance

Page 18

GAO-23-105410 Emergency Rental Assistance

varied significantly across states. But Treasury primarily reallocated
ERA1 funds within the same states from which they were recaptured.
Treasury completed the final round of ERA1 reallocation in October 2022
and, in total, facilitated the reallocation of about $3.1 billion through
voluntary transfers between grantees in the same state and reallocations
(generally across states), based on our analysis of Treasury data. 46
Treasury’s reallocation procedures prioritized reallocating excess funds to
grantees in the same state where they were recaptured and to grantees
that were likely to expend their ERA allocations promptly. To implement
these priorities, Treasury reallocated excess funds within states before
transferring them to grantees in other states and applied benchmarks to
prioritize payments to higher-performing grantees. Specifically,
•

Voluntary transfers. Treasury first fulfilled approved transfers
between grantees in the same state. Grantees could transfer some or
all of their allocation to another eligible grantee in their state that had
obligated at least 65 percent of their initial allocation at the time of the
transfer.

•

State pools. After facilitating voluntary transfers within states,
Treasury allocated the remaining excess funds in each state into their
own state-level pools. The agency used state pools to fulfill as much
of the amounts requested by eligible grantees in the same state as
possible, adjusted to reflect prior spending trends if needed. 47

•

National pool. Treasury moved any remaining funds from each state
pool into a national pool that it used to fulfill requests from grantees in
any state. In the first round of reallocation, Treasury also prioritized
payment to grantees from the national pool that spent 95 percent or

46Treasury’s ERA1 reallocation procedures included four potential rounds of reallocation
based on spending through September 2021, November 2021, January 2022, and March
2022, as well as a separate round for tribal grantees. According to an internal decision
document, Treasury skipped the third round because it would have been operationally
challenging and confusing to grantees (implementation delays would have caused the
third round to overlap with the final round). We excluded the round of ERA1 reallocation
among tribal grantees from our analysis.
47Treasury reviewed and limited each request, if necessary, to 110 percent of the largest
single month of expenditures through October 2021 reflected in the most recent reporting
data, multiplied by four. (For the final round of ERA1 reallocation, this amount was
multiplied by two to account for the limited time available for grantees to expend their
funds.) Agency officials said they used this benchmark to acknowledge that expenditures
among all grantees had increased quickly, and some grantees might need to spend more
than in previous months.

Page 19

GAO-23-105410 Emergency Rental Assistance

more of their initial ERA1 allocation by October 31, 2021, and all
requests from tribal grantees. 48
Because of these procedures, Treasury reallocated the majority of excess
ERA1 funds through voluntary transfers among grantees in the same
state to local grantees. According to Treasury’s reports, about $1.6 billion
of the total amount reallocated (52 percent) was voluntarily transferred
among grantees within the same state. In comparison, Treasury
reallocated about $1.5 billion, generally across states, using funds that it
involuntary recaptured (about $875 million) or that grantees voluntarily
transferred to Treasury (about $610 million) (see fig. 3).
Figure 3: Emergency Rental Assistance Funds Reallocated by Payment Type, as of
October 2022

Note: Treasury reallocations do not equal voluntary transfers to the general fund and involuntary
recapture because Treasury held over some funds from the round of tribal reallocation, and some
grantees declined or did not claim awards.

Local grantees received the majority (about 61 percent or $1.9 billion) of
the total amount reallocated—comparable to about 31 percent of their
initial allocations. States received about $1.2 billion, and tribal grantees
received about $54 million—comparable to 7 percent of their initial
48In the second round of ERA1 reallocation, Treasury prioritized grantees with combined

ERA1 and ERA2 expenditures equal to or exceeding their ERA1 expenditure ratio
denominator (generally 90 percent of their award).

Page 20

GAO-23-105410 Emergency Rental Assistance

allocations. About 94 percent of the excess funds voluntarily transferred
were to local grantees.

Reallocation Payments
Did Not Consistently Align
with Renters’ Needs and
Grantees’ Capacity to
Deliver Assistance in
Some States

Treasury’s priority—keeping funds within states—limited its ability to
address the initial misalignment between their ERA1 allocations and lowincome renters’ needs. The total allocation of ERA1 funds generally did
not change substantially (5 percent or more) in most states because of
the design of the reallocation procedures (see fig. 4). Specifically, the
total ERA1 allocation did not change substantially after reallocation in 28
of the 50 states and the District of Columbia, including several smaller
states that received the minimum allocation ($200 million). In other states,
the total allocation decreased by as much as 82 percent (Wyoming) and
increased by as much as 24 percent (New Jersey). Among the 14 states
that had a substantial decrease in total allocation, 10 had received the
minimum allocation.

Page 21

GAO-23-105410 Emergency Rental Assistance

Figure 4: Percentage Change in Emergency Rental Assistance (ERA1) Allocations by State, as of October 2022

As a result, large differences in the amount of funding per low-income
renter in each state remained after reallocation, especially in states that
received relatively low allocations. For example, the total allocation per
low-income renter in Alaska was about $4,100 after reallocation, as

Page 22

GAO-23-105410 Emergency Rental Assistance

compared to about $770 in Massachusetts. However, some states
experienced a significant reduction, such as Wyoming, where the
allocation per low-income renter decreased from about $4,600 to $810
(see table 1).
Table 1: Emergency Rental Assistance (ERA1) Allocations per Low-Income Renter
in Highest and Lowest States, Before and After Reallocation
Allocation per lowincome renter before
reallocation

Allocation per lowincome renter after
reallocation

Percent
change

$4,588

$809

-82%

$4,094

$3,456

-16%

$4,069

$4,094

$3,259

$2,054

-37%

$2,807

$654

-77%

$2,734

$691

-75%

$2,432

$2,834

17%

$2,293

$807

-65%

$2,110

$1,898

-10%

$1,967

$822

$843

$816

$762

-7%

$802

$839

$802

$806

$774

$775

$774

$819

$771

$891

16%

$762

$771

$759

$748

-1%

$602

$741

23%

States
(highest)

States (lowest)

Source: GAO analysis of Department of the Treasury and Department of Housing and Urban Development data. | GAO-23-105410

Prioritizing payments within states resulted in some inconsistences in the
extent to which payments aligned with renters’ needs and capacity to
deliver assistance. Treasury relied primarily on grantees’ requests to
determine the amount of excess funds it reallocated to each eligible

Page 23

GAO-23-105410 Emergency Rental Assistance

grantee. When sufficient funds were available, Treasury generally funded
the full amount requested by each grantee, adjusted to reflect prior
spending trends if needed. When the amount of funds requested
exceeded the amount of excess funds available, Treasury reduced
grantees’ requests and funded each proportionately to its share of the
total amount requested by all grantees. 49 Because of these procedures,
grantees that requested reallocated funds in states with large state pools
had an opportunity to have their requests fulfilled preferentially and
potentially receive larger payments than those that did not have large
state pools available, including some grantees that did not meet
Treasury’s spending targets. Examples include the following:
•

Approval not aligned with grantee capacity or renter need. In the
first two rounds of reallocation, a county grantee with an expenditure
ratio of 47 percent (as of October 2021) received about $8.8 million,
and a city grantee with an expenditure ratio of 90 percent did not
receive any reallocated funds. 50 According to HUD’s Comprehensive
Housing Affordability Strategy data for 2014–2018, the city had about
twice as many low-income renters as the county. However, after
reallocation, the county’s new allocation was about $13 million greater
than the city’s. Consistent with Treasury’s procedures, the county
received a large award because its state pool included about $64
million recaptured from the state grantee. In comparison, the city was
eligible to receive funds only from the national pool because its state
pool did not include any excess funds; however, Treasury did not
prioritize it for a payment because it had not meet the 95-percent
benchmark.

•

Award reduced because few funds from state pool. A city grantee
received 100 percent of the amount it requested ($6.6 million) in the
second round of reallocation because its state pool included $42
million recaptured from the state grantee. In comparison, a state
grantee in another state received about 7 percent of the amount it
requested ($7.4 million), because its state pool only included about
$21,000 recaptured from a county grantee. The remainder of its
award was provided from the national pool, but because grantees
requested more funds than were available, Treasury’s procedures

49For the purposes of calculating the proportional share in the first round of ERA1

reallocation, Treasury also multiplied by three requests from grantees that had spent any
ERA2 funds, as well as requests from tribal grantees.

50The city received about $1.4 million in the last two rounds of ERA1 reallocation.

Page 24

GAO-23-105410 Emergency Rental Assistance

required it to significantly reduce each grantee’s award and
recalculate it as a proportion of all other grantees’ requests.
•

Funds awarded to grantee that did not meet spending target. A
state voluntarily transferred about $229 million in excess funds to local
grantees within its state, including a $9 million transfer to a county that
exceeded its initial allocation. But the county had spent 23 percent of
its initial allocation through September 2021 and had been required to
submit a program improvement plan because it did not meet
Treasury’s spending target. As of June 2022, the county had spent 45
percent of its allocation.

Treasury officials acknowledged that some stakeholders had raised
concerns about prioritizing reallocation payments within states when the
initial allocations did not match the needs of renters in all states.
However, officials told us their goals in prioritizing payments within states
were to encourage voluntary reallocation, limit the amount of funds
subject to involuntary recapture, and avoid taking money away from lowincome renters in one state to benefit those in another state—especially
in light of spending delays in some jurisdictions. In their written response
to a draft of this report, Treasury also asserted that they believed
reallocating funds within states was in alignment with Congress’s initial
state-by-state allocation formula. As noted earlier, Congress provided
Treasury with discretion on how to determine excess funds and reallocate
them to grantees that met statutory obligation benchmarks.
Treasury was still reallocating excess ERA2 funds at the time of our
review. In October 2022, Treasury completed its first quarterly
assessment of excess ERA2 funds by reallocating about $519 million
(including $231 million voluntarily transferred between grantees and $288
million involuntarily recaptured and reallocated by Treasury). 51

51The ERA2 statute requires Treasury to reallocate excess ERA2 funds to grantees

beginning on March 31, 2022, and its procedures include four quarterly assessment
periods that conclude with the reallocation of funds that grantees had not drawn from their
allocations by December 31, 2022. Treasury issued guidance in September 2022 that
clarified that grantees could re-characterize ERA2 expenditures as ERA1 expenditures to
minimize the amount of unobligated funds subject to expiration, to the extent allowed by
applicable requirements set out by law and program guidance. Department of the
Treasury, Emergency Rental Assistance Under the Consolidated Appropriations Act,
2021: Notice Regarding Unobligated ERA1 Funds (Washington, D.C.: Sept. 13, 2022).

Page 25

GAO-23-105410 Emergency Rental Assistance

Available Data
Suggest the ERA
Program Served LowIncome Renters,
Including Some
Facing Eviction
Available Data Indicate
ERA Reached Very-LowIncome and RentBurdened Households

ERA payment and demographic data are limited and have missing and
erroneous values, which we discuss later in this report. Although limited,
the available data indicate that ERA funds served low-income and rentburdened households and that receipt of funds across demographic
groups varied. 52 Based on our comparison of available household
payment data to the estimated number of low-income renters, about 7
percent of low-income renter households (those with incomes at or below
80 percent of the area median) in the United States received an ERA
payment in 2021 (assuming that all payments were made to low-income
renter households, as required). 53 In counties we reviewed with available
data, ERA funds served from 0.4 percent to 15.6 percent (5th and 95th
percentiles) of the estimated low-income renter households. Figure 5
shows the percentage of low-income renters in each county that received
ERA1 assistance in 2021.

52We used data on the demographics of households served using ERA1 in the fourth
quarter of 2021. Data were missing from 34 percent of grantees (or 19 percent of
households served) in the quarter. We excluded data from the first three quarters of 2021
because the share of missing data for these quarters ranged from 44 to 55 percent.
Treasury’s public reporting on household demographics for calendar year 2021 was
generally consistent with that from the fourth quarter.
53We used data from HUD’s Comprehensive Housing Affordability Strategy to measure

low-income renter population, which are a custom tabulation of 2014–2018 Census data
on low-income renter populations by county and state. We defined low-income as not
exceeding 80 percent of the area median income, which matched the statutory limit for
ERA1 (the ERA2 statute sets a similar limit) and assumes all payments were made to
eligible low-income renters. For more information on our methodology, see app. I.

Page 26

GAO-23-105410 Emergency Rental Assistance

Figure 5: Quartiles of Low-Income Renter Households That Received Emergency Rental Assistance (ERA1 and ERA2), by
County, 2021

Among those served in the fourth quarter of 2021, 85 percent had verylow incomes (below 50 percent of the area median), which is consistent
with the ERA requirement that grantees prioritize payments to such

Page 27

GAO-23-105410 Emergency Rental Assistance

households. 54 Specifically, 63 percent of households had incomes at or
below 30 percent of the area median and 22 percent had incomes
ranging from 31 to 50 percent of the area median. The remaining 15
percent for which data were reported had incomes ranging from 51 to 80
percent of the area median.
Counties with less-affordable rent received larger ERA payments, based
on our analysis of Treasury and Census Bureau data. Specifically,
counties that have greater median gross rents and greater shares of rentburdened households (those paying at least 30 percent of their income on
rent) were associated with a greater number of ERA payments, greater
total ERA funding, and higher average ERA payments received per
household. 55
Black households received the largest share of ERA1 assistance of any
racial group, according to Treasury’s household demographic data.
Among households for which grantees reported data, Black households
received 44 percent of ERA1 payments in the fourth quarter of 2021 (see
fig. 6). 56 White households received 37 percent of the assistance,
households that identified as multiracial received 6 percent, American
Indian or Alaska Native households received 2 percent, and Asian
households received 2 percent. Among ethnic groups represented in
54The ERA1 and ERA2 statutes require grantees to prioritize assistance for households

with incomes less than 50 percent of their area median or households with one or more
individuals who are unemployed as of the date of the application for assistance and have
not been employed for the 90-day period preceding the date of application. Consolidated
Appropriations Act, 2021 § 501(c)(4) (to be codified at 15 U.S.C. § 9058a(c)(4)) ;
American Rescue Plan Act of 2021, § 3201(d)(2) (to be codified at 15 U.S.C. §
9058c(d)(2)).

55There was a positive correlation between the median gross rents and the total number
of ERA payments to a county, total amount of ERA funding in a county, and average
payment received per household served. There was also a positive correlation between
the share of rent-burdened households and these three measures. We calculated median
gross rent and share of rent-burdened households by county using Census Bureau’s
2016–2020 American Community Survey data.
56Race data were reported in the following categories based on the primary applicant:
American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or
Pacific Islander, White, Mixed, declined to answer, or data not collected. We use “Black”
to refer to the Black or African American racial category and “multiracial” to refer to the
Mixed racial category. Because tribal grantees were not required to report demographic
data during the reporting period, the percentage of assistance to American Indian or
Alaskan Native and Native Hawaiian or Pacific Islander underrepresents the total amount
of assistance to such households. Data were missing for 19 percent of households that
received ERA1 payments in the fourth quarter of 2021.

Page 28

GAO-23-105410 Emergency Rental Assistance

Treasury’s data, Hispanic households received 18 percent of ERA
assistance (compared to 72 percent among non-Hispanic households). 57
Figure 6: Percentage of Households That Received Emergency Rental Assistance
(ERA1), by Race, Fourth Quarter 2021

Note: Data were missing for 19 percent of households that received ERA1 payments in the fourth
quarter of 2021.

Black households tended to receive more ERA1 assistance than their
proportion of the low-income renter population. According to HUD’s
Comprehensive Housing Affordability Strategy data, Black households
make up 23 percent of the national low-income renter population. Among
the 51 states and territories for which demographic and low-income renter
population data are available, the proportion of Black recipients served by
ERA met or exceeded the share of Black households among the low-

57Ethnicity data were reported in the following categories: Hispanic or Latino, Not Hispanic
or Latino, declined to answer, or data not collected. We use “Hispanic” to refer to the
Hispanic or Latino ethnic category. Nine percent of households for whom data were
provided did not report their ethnicity. The rate of Hispanic households served was
calculated separately; households that identified as Hispanic could separately identify with
any race.

Page 29

GAO-23-105410 Emergency Rental Assistance

income renter population in 47 states. 58 The high proportion of Black
households served by the program could be an indication of greater need
during the pandemic. For example, according to the Harvard Joint Center
for Housing Studies, Black and Hispanic households faced a greater risk
of eviction during the pandemic. 59
Based on available demographic and low-income renter population data
for 51 states and territories, Hispanic households received a lower share
of the assistance provided in about two-thirds of states and territories (35)
than their share of the low-income renter population. 60 According to
HUD’s data, Hispanic households also make up 23 percent of the national
low-income renter population. Potential reasons that grantees made
proportionately fewer payments to Hispanic households could include
language and technology barriers, as well as eligibility limitations based
on immigration status. The availability of ERA applications in non-English
languages varied by grantee, based on interviews we conducted with a
selection of grantees. 61 In addition, research by the Department of

58We calculated the share of the low-income rental population by race for each state using
HUD’s Comprehensive Housing Affordability Strategy for 2014–2018. For more
information, see app. I.
59Sophia Wedeen, Black and Hispanic Renters Face Greatest Threat of Eviction in
Pandemic (Joint Center for Housing Studies: Jan. 11, 2021), accessed at
https://www.jchs.harvard.edu/blog/black-and-hispanic-renters-face-greatest-threateviction-pandemic. The author used Census Bureau’ Household Pulse Survey data to
analyze the share of households behind on rent by race or ethnicity.
60Measures of Hispanic identity differ between the ERA data and HUD’s Comprehensive
Housing Affordability Strategy. While ERA data measures Hispanic identity as an ethnicity
separate from race, HUD’s data collapses ethnicity and race into a single measure, and
indication of Hispanic ethnicity supersedes an individual’s indication of race in the data.
We do not believe this difference limits our analysis. The General Services
Administration’s Office of Evaluation Sciences (using a different methodology) also found
that Hispanic households were underrepresented among those with incomes between 30
and 80 percent of their area median but were overrepresented among those with income
below 30 percent of the area median. See General Services Administration, Office of
Evaluation Sciences, Equity in the Distribution of the Emergency Rental Assistance
Program, accessed October 26, 2022, https://oes.gsa.gov/projects/era-equity/.
61Eighteen of the 21 grantees we interviewed provided applications in Spanish and three
did not.

Page 30

GAO-23-105410 Emergency Rental Assistance

Commerce found that Hispanic households historically have had the
lowest levels of internet use among racial or ethnic groups. 62
Counties with a greater share of households without internet access were
less likely to receive ERA payments (most ERA grantees we interviewed
had internet-based applications). 63 Also, although the ERA law does not
impose restrictions based on immigration status, some grantees limited
payments to U.S. citizens or legal residents. However, we found a
positive relationship between immigrant populations and ERA payments.
Counties with larger shares of immigrants were associated with larger
average ERA payments received per household. 64
Based on available demographic data, female-led households also
received a significant share of ERA payments. Specifically, 69 percent of
ERA1 recipients were female, according to Treasury’s demographic data
from the fourth quarter of 2021, which exceeded the national share of
very-low-income households led by women. 65 While a narrower measure
than the eligible ERA population, about 61 percent of rental households
with income not exceeding the poverty line were led by women with no
spouse present, according to Census’ American Community Survey
estimates. 66 The somewhat greater prevalence of female-led households
that received ERA assistance also could be explained by disproportionate
financial impacts during the pandemic. For example, female workers were
disproportionately affected by job losses in the pandemic, and fewer
62Department of Commerce, National Telecommunications and Information
Administration, Language and Citizenship May Contribute to Lower Internet Use Among
Hispanics (Nov. 17, 2015), https://www.ntia.gov/blog/2015/language-and-citizenship-maycontribute-low-internet-use-among-hispanics.
63There was a negative correlation between the share of households without internet
access in a county and (1) the total number of ERA payments in the county and (2) the
total amount of ERA funding in a county. There was also a slight negative correlation
between the share of households without internet access in a county and the percentage
of households served by the ERA program in a county.
64There was a positive correlation between the share of immigrants in a county and both
the total number of ERA payments in the county and the total amount of ERA funding in a
county. This could be partially attributed to the distribution of immigrants between urban
and rural areas. Urban counties received larger average ERA payments per household,
and urban counties also have larger shares of immigrants.
65Gender data were reported in the following categories: male, female, nonbinary,
declined to answer, and did not collect.
66The breakdown of renter households at or below 80 percent of the area median income
by gender was not available using HUD’s data.

Page 31

GAO-23-105410 Emergency Rental Assistance

women than men have returned to work, according to National Women
Law Center’s analysis of Bureau of Labor Statistics data. 67

Grantees Used ERA to
Address Local Needs and
Priorities

Differences in ERA payments suggest that grantees used ERA to address
local needs and priorities. We observed differences in the eligible uses
and amount of payments across grantees and geography. These
differences may reflect policy decisions, such as how grantees prioritized
applications. For example, differences in eligible uses (arrears versus
prospective payments) could partially reflect whether grantees adopted
policies to prioritize payments based on applicants’ arrearages or risk of
eviction. Differences we observed also could reflect socioeconomic
conditions within a state, such as the extent of housing cost burden or
unemployment.
Across all grantees, payments for arrears and prospective rent and
utilities generally were evenly distributed (see fig. 7). 68 Based on available
payment data from Treasury, about 48 percent of payments to
households were for prospective rent and utilities (39 and 9 percent,
respectively) and about 50 percent of payments were for rental and utility
arrears (34 and 16 percent, respectively). 69 One percent of payments
were for other housing expenses, such as security deposits or internet
service.

67National Women’s Law Center, Men Have Now Recouped Their Pandemic-Related

Labor Force Losses While Women Lag Behind (February 2022), accessed September 9,
2022, https://nwlc.org/resource/men-recouped-losses-women-lag-behind/.
68Eligible payment types included rental arrears, utility arrears, prospective rent (which
includes payments for current rental obligations), prospective utilities, and other housing
expenses. Data are missing for 26 percent of the total payment amount in 2021. Because
of data reliability concerns, we excluded negative payment values and payments made by
two grantees from our analysis. For more information, see app. I.
69Data reported by grantees for 2021 cover $12.6 billion of the $17.1 billion (74 percent) in

payments provided to households for ERA1 and ERA2 through December 31, 2021.
Payment use was missing in 0.4 percent of the data.

Page 32

GAO-23-105410 Emergency Rental Assistance

Figure 7: Emergency Rental Assistance (ERA1 and ERA2) Payments to Households
by Eligible Use, 2021

Note: Data are missing for 26 percent of the total payments in 2021, and payment use is missing for
0.4 percent of reported payments. We excluded negative payment values and payments made by two
grantees from our analysis because of data reliability concerns. Available data do not include the
number of payments made to households for housing stability services. The percentages in the figure
do not add to 100 because of rounding.

In addition to payments to households, grantees also spent 4.8 percent of
their ERA1 allocations on administrative costs and 0.9 percent on
household stability services (such as eviction diversion programs).
The extent to which grantees made prospective versus arrears payments
was clustered and varied by state (see fig. 8). For example, over 62
percent of payments made in most counties in Minnesota, Arkansas, and
Mississippi were for arrears payments. In comparison, over 59 percent of
payments made in most counties in Louisiana and Florida were for
prospective payments.

Page 33

GAO-23-105410 Emergency Rental Assistance

Figure 8: Quartiles of Emergency Rental Assistance (ERA1 and ERA2) Payments to Households for Rent or Utility Arrears, by
County, 2021

Page 34

GAO-23-105410 Emergency Rental Assistance

Grantees made larger average payments to households in urban
counties, which are generally associated with higher costs of living. 70 The
median ERA payment to households served by the program in 2021 was
$6,450, but households in urban counties received about $7,200, as
compared to about $5,200 in rural areas. Average payment size was
generally greater in counties near larger cities and across higher-cost
states, such as California.

ERA Payments, Combined
with Other Factors,
Contributed to Lower
Eviction Rates

Eviction data and third-party studies suggest that ERA payments in
conjunction with other factors helped limit evictions. However, data are
not available to assess the full impact of the program on eviction rates,
and ERA alone cannot explain lower eviction rates and improved
perceptions of housing stability. Changes in outcomes during the
pandemic are explained by a variety of factors in addition to emergency
rental assistance. Such factors may include eviction moratoriums, lower
unemployment, declining business closures and disruptions, court
closures that limited eviction hearings, and a greater focus on eviction
diversion.
•

Eviction data. National data to assess eviction rates across all
grantees do not exist. Our analysis of data collected by the Federal
Reserve Bank of Cleveland from 49 jurisdictions (representing 10
percent of U.S. renter households) highlight that eviction rates
remained below pre-pandemic levels even after the national
moratorium ended. 71 Specifically, evictions in the jurisdictions without
any eviction prohibitions, including local bans on eviction filings,
declined as grantees began spending ERA funds in early 2021 and
remained below 2019 levels until February 2022.

70For the purposes of this report, rural counties are nonmetropolitan counties and urban
counties are metropolitan counties, as designated by the Department of Agriculture in
2017. More specifically, the department’s Economic Research Service designated
metropolitan areas to include all counties with urban areas containing 50,000 people or
more. Metropolitan areas also include outlying counties that are economically tied to the
central counties, as measured by the share of workers commuting on a daily basis to the
central counties. Nonmetropolitan counties are outside the boundaries of metropolitan
areas and have no cities with 50,000 residents or more.
71Federal Reserve Bank of Cleveland, Data Updates: Measuring Evictions during the
COVID-19 Crisis (accessed Oct. 27, 2022). See
https://www.clevelandfed.org/publications/cd-reports/2020/db-20200902-data-updatesmeasuring-evictions-during-the-covid-19-crisis. Because of weekly fluctuations in data
reporting, we analyzed the 4-week moving average. The Centers for Disease Control and
Prevention’s eviction order prohibited evictions of covered renter households for the
nonpayment of rent from September 4, 2020, through July 31, 2021.

Page 35

GAO-23-105410 Emergency Rental Assistance

•

Third-party studies. Analysis conducted by Harvard’s Joint Center
for Housing Studies using Census Household Pulse Survey data
identified increased perceived housing stability among ERA
recipients. 72 Specifically, the odds that ERA recipients who had been
behind on rent thought eviction was at least somewhat likely in the
next 2 months was an estimated 59 percent lower than for those who
had applied for but not received ERA assistance. An estimated 25
percent of ERA recipients were behind on rent at the time they were
surveyed, compared to an estimated 65 percent of those who had
applied for but not received ERA assistance.
Consistent with our findings, one academic study also concluded that
ERA contributed to reducing evictions but that the program’s impact
was limited by a slow rollout. 73 Dynamic modeling conducted by the
researchers suggested that the federal eviction moratorium combined
with ERA assistance resulted in lower eviction rates than the expected
eviction rates in the absence of both programs. According to the
study, faster initial distribution of ERA funds could have minimized
evictions further.

Data Are Incomplete
and Treasury Has Not
Fully Assessed
Improper Payment
Risks
Treasury Has Not
Collected or Reported
Complete Data on ERA
Payments and Recipients

Treasury has not collected and reported complete data on ERA payments
and recipients as required under the ERA1 authorizing statute. The
Consolidated Appropriations Act, 2021, requires Treasury to collect and
report six performance measures disaggregated by demographics (race,
72Whitney Airgood-Obrycki, The Short-Term Benefits of Emergency Rental Assistance

(Joint Center for Housing Studies: June 2022), accessed September 8, 2022,
https://www.jchs.harvard.edu/research-areas/working-papers/short-term-benefitsemergency-rental-assistance. The study cited some limitations, including that survey
responses may reflect a household’s receipt of ERA funds explicitly or other state and
local rental assistance funds.

73Katherine Marcal, Patrick J. Fowler, and Peter S. Hovmand, Feedback Dynamics of the
Low-Income Rental Housing Market: Exploring Policy Responses to COVID-19 (June 28,
2022), https://doi.org/10.48550/arXiv.2206.12647.

Page 36

GAO-23-105410 Emergency Rental Assistance

ethnicity, and gender) from all grantees on at least a quarterly basis. 74
Additionally, Treasury’s ERA award terms require each grantee to agree
to comply with program reporting obligations and its reporting guidance
emphasizes the importance of collecting complete and accurate data.
As of November 1, 2022, when we provided a draft of this report to
Treasury for comment, the agency had collected and reported
significantly incomplete grantee data. 75 Specifically, the data Treasury
reported publicly for the first three quarters of 2021 that were
disaggregated by demographics were missing for 44–55 percent of
households served. The high proportion of missing data was largely
driven by grantee nonreporting in those quarters (47–65 percent of
grantees did not report any demographic information for households they
served). Reporting levels improved for the fourth quarter of 2021—data
were missing for 19 percent of households served for that period. In
addition, data were missing for 26 percent of payments in 2021 (see table
2).
Table 2: Percentage of Missing Data and Nonreporting Grantees for Emergency
Rental Assistance Data, by Reporting Period
Proportion of
missing data

Proportion of
nonreporting
grantees

Payment (ERA1 and ERA2) 2021

26%

20%

Household demographics
(ERA1)

Quarter 1, 2021

55%

65%

Quarter 2, 2021

44%

51%

Quarter 3, 2021

52%

47%

Quarter 4, 2021

19%

34%

Type of data

Reporting period

Source: GAO analysis of Department of the Treasury data. | GAO-23-105410

74Consolidated Appropriations Act, 2021 § 501(g) (to be codified at 15 U.S.C. § 9058a(g)).
The performance measures are the number of households served, type of assistance
provided, average household payment amount, household income level, applicationfunding rate, and the average payment length. The statute requires each of these
performance measures to be disaggregated by race, ethnicity, and gender of the primary
applicant for each household.
75Treasury collected the data used to calculate performance measures from grantees

across separate data files. Grantees are required to report payment information (including
recipient address, amount of payment, and type of payment) and household demographic
information (including the number of households served by race, ethnicity, and gender)
separately for each quarter.

Page 37

GAO-23-105410 Emergency Rental Assistance

Also as of November 1, 2022, Treasury had not publicly reported
demographic data for any quarters in 2022 or for the remaining two
required performance measures (average payment length or average
application-funding rate) for any quarters. These two performance
measures are limited by missing and irregular data values. Treasury’s
payment data contain fields for the start and end date for each payment,
which Treasury could use to calculate average payment length. However,
data were missing for 26 percent of the total payment amount issued in
calendar year 2021, and the payment end date is missing for about 60
percent of reported payments. 76
Additionally, data available as of November 1, 2022, indicate that the
application-funding rate may be unreliable. Among the 79 percent of
grantees that reported both the number of complete applications
submitted and number of households served with ERA1 funds by quarter
in 2021, 5 percent reported a greater number of households served than
the total number of applications submitted. While the median applicationfunding rate by grantee was 38 percent, the rates ranged from 0 percent
to 441 percent. 77
Grantees we interviewed also cited Treasury’s delayed implementation of
reporting guidance and technical limitations as contributing to incomplete
and inaccurate data reporting. Although Treasury released reporting
guidance in June 2021, it revised certain reporting requirements in
September 2021—1 month before grantees were required to submit
quarterly reports for the first three quarters of 2021. Some grantees told
us they had not collected certain data elements that became required by
Treasury after it revised the reporting guidance. As a result, the grantees
had to collect new or update existing data retroactively, which challenged
their ability to report complete and accurate data to Treasury in a timely
manner. In addition, some grantees told us they experienced technical
challenges uploading and submitting required information to Treasury’s
data portal, including making corrections to their data once submitted.

76ERA program guidance instructed grantees to leave the payment end date blank in

certain cases. Specifically, in situations when the start and end date of payments was not
known, Treasury’s guidance advised grantees to provide a start date, but leave the end
date blank. Six percent of payments are missing a start date.
77Some outlier funding rates could be due to jointly administered programs in which

applications were ultimately funded by a different grantee than the grantee to which the
household applied.

Page 38

GAO-23-105410 Emergency Rental Assistance

In September 2022, Treasury officials stated that grantee capacity
limitations were the primary cause of missing data. Treasury officials
stated previously that missing data also stemmed from a lack of
awareness among some grantees that their monthly and quarterly
reporting requirements continued after their programs ended or if they
otherwise did not serve any households in the reporting period.
Treasury has taken some steps to improve data completeness and
accuracy. In September 2022, Treasury officials told us they were testing
and monitoring grantees’ data submissions and following up with
grantees for clarification and potential updates when the agency identified
incomplete and erroneous reporting. Officials also told us the agency was
implementing technical updates to its reporting portal to address prior
reporting challenges, such as allowing grantees to correct certain fields in
prior data submissions.
Furthermore, in October 2022, Treasury officials told us they were
working with OMB to finalize closeout reporting requirements that could
address some data quality concerns. 78 According to a framework
published by the agency in September 2022, Treasury planned to require
grantees to submit a closeout report with cumulative (aggregate) financial
and performance information, including payment and recipient
demographic data. On November 4, 2022, after we provided a draft of this
report to Treasury for comment, OMB approved the agency’s ERA1
closeout reporting requirements. As part of the closeout reporting
process, grantees are required to input missing data for and correct
inaccurate data reported in prior quarters, as applicable. The closeout
reporting period provides Treasury with the opportunity to address the
data collection and quality issues we identified in this report. However,
given the previous challenges with collecting and submitting data through
Treasury’s portal, Treasury will need to continue to be proactive in
working with grantees to collect complete and reliable data.
Public reporting of these data is also needed to meet the requirements of
the Consolidated Appropriations Act, 2021. As previously discussed,
Treasury published significantly incomplete data for its reported
78OMB’s Uniform Administrative Requirements, Cost Principles, and Audit Requirements
for Federal Awards require recipients of federal grants to report closeout data. Grantees
must report these data within 120 days after their period of performance ends. For
grantees that did not receive reallocated funds, the period of performance ended on
September 30, 2022. For grantees that received reallocated funds, the period of
performance for these funds ends on December 29, 2022.

Page 39

GAO-23-105410 Emergency Rental Assistance

performance measures and had not reported any data for two required
performance measures as of November 1, 2022. After we provided a draft
of this report to Treasury for comment on November 1, 2022, the agency
published an additional data report on the required performance
measures through the second quarter of 2022 that included the two
previously missing performance measures (average payment length and
average application-funding rate). However, these performance metrics
do not fully meet the reporting requirements of the statute, which may be
a result of data collection challenges. For example, these two
performance measures were not disaggregated by demographics, as
required by statute. According to the data report, fewer than two-thirds of
non-tribal grantees (62 percent) reported disaggregated demographic
data. In addition, Treasury noted in the report that it was working with
grantees to update certain data elements due to data validation issues. It
will be important for Treasury to fully report on the required performance
measures disaggregated by demographics through the end of grantees’
performance period and include information needed to determine their
quality.
Without complete and accurate performance data, Congress will continue
to lack key information it mandated in the authorizing statute. In addition,
in the absence of quality data, Treasury cannot reliably analyze or
publicly report recipient demographics and grantees’ use of funds. For
instance, incomplete data could bias analysis or underrepresent certain
populations served by the program. 79 A lack of complete and accurate
data also hinders Treasury and its Office of Inspector General from
effectively evaluating the program for compliance with relevant laws and
guidance. For example, the office would not be able to comprehensively
evaluate whether grantees exceeded ERA1 payment limitations and, if
necessary, recoup overpayments, as required by law. 80

79Across ERA datasets, nearly all nonreporting grantees were local governments and
dispersed among states. For demographic data, 94 percent of nonreporting grantees were
local. Data were entirely missing for three of the 56 states and territories and partially
missing for 33 states. For payment data, 83 percent of nonreporting grantees were local
and data was entirely missing for four states and five territories and partially missing for 26
states.
80Consolidated Appropriations Act, 2021 § 501(i) (to be codified at 15 U.S.C. § 9058a(i)).

Page 40

GAO-23-105410 Emergency Rental Assistance

Potential Duplicate
Payments and Other Data
Anomalies Highlight the
Importance of Completing
a More Detailed Risk
Assessment

Our review of Treasury data highlights improper payment risks in the ERA
program, but Treasury has not conducted a detailed risk assessment or
implemented our prior recommendation on payment recovery audits. We
reviewed ERA1 payments made to households in 2021 and found that
about 2 percent of the households assisted (or about 43,000 households)
received payments from more than one grantee. 81 These payments
accounted for about 6 percent of all payments reported by grantees in
2021. In some states, a significant proportion of payments may have
been duplicative. For example, one city grantee and its county grantee
each may have made duplicative payments in about 20 percent of their
total payments in 2021. These findings indicate that grantees may be
making duplicative payments despite a statutory requirement that

81We reviewed the addresses of eligible households that received ERA1 assistance as
reported to Treasury by grantees in their household payment data file. Because we
reviewed households’ addresses, our analysis does not count the addresses of landlords,
who could have received multiple payments from more than one grantee to cover different
households. In addition, we included unit numbers in our analysis to the extent available
data allowed, which accounts for payments made to different households that reside at the
same address (for example, within a multifamily apartment building).

Page 41

GAO-23-105410 Emergency Rental Assistance

grantees avoid making such payments to the extent feasible. 82 Treasury
guidance further encourages grantees to avoid making such payments. 83
Other data anomalies we observed further highlight improper payment
risks. About 2 percent of ERA1 payments reported by grantees in 2021
with complete data on payment dates exceeded the statutory limit of 15
months of assistance. Similarly, about 2 percent of households that
received payments from more than one grantee received more than 15
months of ERA1 assistance. However, this rate is likely higher given that
data needed to calculate the length of assistance for each payment were
missing for about 60 percent of payments reported by grantees in 2021. 84
Furthermore, a significant number of grantees may have exceeded
statutory limitations on the use of ERA1 funds for administrative
expenses. Based on Treasury data through April 2022, 42 of the 406

82Consolidated Appropriations Act, 2021 § 501(k)(3)(B) (to be codified at 15 U.S.C. §

9058a(k)(3)(B)).

83Department of the Treasury, Emergency Rental Assistance Frequently Asked
Questions. Treasury officials told us in May 2022 that they were aware that grantees in
some overlapping jurisdictions had made, and in some cases recovered, duplicative
payments. However, households may have legitimate reasons for receiving payments
from multiple grantees. According to Treasury officials, grantees in overlapping
jurisdictions may divide payment responsibilities for rent and utilities, households may
apply to a second program after the first program exhausts its allocation, tenants in
multifamily properties may omit unit numbers from their applications, or both a prior and
current tenant of the same unit may have received payments. Treasury officials also noted
that eligible applicants may rent rooms or spaces in single-family homes and as a result
may share an address with other renters. Treasury updated its guidance in July 2022 to
address instances in which grantees retroactively identify that an ERA payment duplicated
a payment made by another grantee. The guidance allows grantees that identify such
payments to decline to recover the overpayment and modify the intended period covered
by the payment, if the grantee documents the expenses covered by the payment and the
additional months of assistance do not exceed the 15-month limit. Treasury officials told
us they developed the guidance following requests for technical assistance from grantees
about options for addressing duplicative payments. Officials from the Office of Inspector
General told us that Treasury officials responsible for the program’s administration did not
consult with them in developing and publishing the updated guidance. The office is
required to monitor and oversee the use of ERA1 funds and recover funds spent in
violation of the eligible uses. Consolidated Appropriations Act, 2021 § 501(i) (to be
codified at 15 U.S.C. § 9058a(i)).
84We also found that 5 percent of grantees reported a greater number of households
served than the total number of applications submitted.

Page 42

GAO-23-105410 Emergency Rental Assistance

ERA1 grantees reported administrative expenses that exceeded the
statutory limit of 10 percent of their allocation. 85
Treasury has not determined the estimated amount or rate of improper
payments to households in the ERA program. According to federal
internal control standards, to respond to risks, agencies should first
analyze identified risks to estimate their significance. 86 Treasury officials
told us they provided data on potentially duplicative payments to their
Office of Inspector General in March 2022 but were in the early phases of
analyzing payment data to determine the extent of improper payments as
of September 2022. In addition, Treasury completed an improper
payment risk assessment for the ERA program in September 2022 that
focused on the risk of improper payments in Treasury’s disbursement of
ERA allocations to grantees. This assessment concluded that the
program was not susceptible to significant improper payments. 87
However, the risk assessment consisted of a qualitative questionnaire
that did not account for missing data or duplicative payments at the
household level.
Given the data concerns and other challenges we have highlighted, a
more detailed assessment of improper payment risks is warranted. 88
OMB requires that agencies develop risk assessment methodologies that
are appropriate to ensure that the result of the risk assessment
reasonably supports whether the program is or is not susceptible to
85Consolidated Appropriations Act, 2021 § 501(c)(5) (to be codified at 15 U.S.C. §
9058a(c)(5)). The number of grantees that have potentially exceeded the ERA1
administrative expense limit is based on Treasury data as of April 30, 2021, which reflects
the first two rounds of ERA1 reallocation. The administrative expenses of some grantees
relative to the limit may have changed after the third round of reallocation.
86GAO, Standards for Internal Control in the Federal Government, GAO-14-704G

(Washington, D.C: Sep. 10, 2014).

87Treasury is directed by OMB guidance to assess the ERA program’s susceptibility to

significant improper payments and, if determined to be susceptible, develop and report a
reliable estimate of improper payments, identify root causes, and develop a corrective
action plan to reduce them. For newly established programs, an improper payments risk
assessment should be completed after the first 12 months of the program. Office of
Management and Budget, Appendix C to OMB Circular A-123, Requirements for Payment
Integrity Improvement, M-21-19 (Washington, D.C.: Mar. 5, 2021). The circular
implements requirements in the Payment Integrity Information Act of 2019.

88For example, we have noted the hurried implementation of the ERA program that
resulted in payments being made as guidance was being developed. We also have noted
overlapping service areas and challenges grantees cited in limiting duplicative payments.

Page 43

GAO-23-105410 Emergency Rental Assistance

significant improper payments. As we noted, data were missing for about
26 percent of the $17.1 billion in payments grantees reported in 2021,
which potentially places the ERA program well above the statutory
definition for significant improper payments (1.5 percent of program
outlays). 89 Without a more detailed assessment of improper payment
risks in the ERA program, such as a quantitative analysis that
incorporates grantee payment data, Treasury will continue to lack a
complete understanding of the program’s susceptibility to improper
payments and the need for further efforts to reduce them.
We also reiterate the importance of our prior recommendations to
manage improper payments in the ERA program in light of the risks we
highlighted about Treasury’s payment data. In January 2022, we reported
that Treasury lacked processes to identify and recover overpayments
made by grantees and recommended that Treasury implement such
processes, for example, through post-payment reviews and recovery
audits. 90 Treasury agreed with this recommendation and stated that it was
working to establish post-payment reviews and recovery audit activities.
However, Treasury had not implemented this recommendation as of
October 2022.
Treasury’s Office of Inspector General has taken some steps to review
grantee data and help identify and recover improper payments. In April
2022, the office implemented desk review procedures that include steps
to evaluate payments made by grantees that demonstrated data reporting
and compliance challenges. 91 Officials from the Office of Inspector
General told us they have been developing an ERA risk model to identify
89A program is considered to be susceptible to significant improper payments if, in the

preceding fiscal year, the sum of the program’s improper payments and payments whose
propriety cannot be determined due to lacking or insufficient documentation (unknown
payments) may have exceeded either (1) 1.5 percent of program outlays and $10 million
or (2) $100 million (regardless of the improper payment rate). See 31 U.S.C. § 3352.
OMB’s guidance clarifies that payments include transfers of federal funds by a federal
grantee to any nonfederal person or entity, such as a landlord or tenant. Unknown
payments may not necessarily be improper payments but are to be included in the risk
assessment for the purpose of determining the program’s susceptibility to improper
payments.
90GAO-22-105291. We also previously recommended that Treasury implement monitoring

procedures to evaluate the effectiveness of the controls it requires grantees to implement
to prevent improper payments when using self-attestation to determine eligibility. Treasury
also had not implemented this recommendation as of October 2022. GAO-22-105490.

91Department of the Treasury, Office of Inspector General, Emergency Rental Assistance

Government Grantee Quarterly Reporting Desk Review Procedures, OIG-CA-22-013
(Washington, D.C.: Apr. 22, 2022).

Page 44

GAO-23-105410 Emergency Rental Assistance

potentially problematic payments through these audits and their ongoing
oversight work. As of August 2022, the office had initiated two audits of
ERA grantees based on its desk review procedures.
But Treasury’s Office of Recovery Programs, which maintains
responsibility for the ongoing administration and oversight of the ERA
program, plans to reduce its monitoring capacity because it lacks
sufficient administrative funds for the ERA program. According to a letter
sent by the Deputy Secretary of the Treasury to grantees in September
2022, Treasury was appropriated funding to administer multiple pandemic
relief programs, but for some programs like ERA, the available funding is
insufficient to allow Treasury to maintain current levels of administrative
support. Without the authority to repurpose administrative funds from
other programs, the letter states that Treasury is preparing to reduce its
recipient reporting and monitoring, institute a hiring freeze, and suspend
some grantee technical support functions. Given potential improper
payments and the nascent nature of Treasury’s oversight efforts, reducing
its monitoring capabilities would leave the ERA program in a precarious
position that requires immediate resolution to avoid significant risks to its
ongoing integrity. 92

Conclusions

Treasury’s efforts during the pandemic to administer the ERA program,
assist grantees in developing local programs, and introduce
administrative flexibilities to expedite payments have helped provide
temporary financial and housing stability to millions of renters and
landlords. As Treasury transitions into the next phase of the ERA
program, it is vital that the agency consider the challenges and risks we
have highlighted in the past 2 years to improve its administration and
oversight of the program.
Effective oversight of ERA payments and grantees is critical because of
the size and hurried implementation of the program. However, Treasury
has yet to collect complete and accurate information on ERA payments
and recipients or publicly report fully disaggregated quarterly performance
measures required under statute. Treasury has taken some steps to
improve data collection, including developing final reporting requirements.
However, given the difficulties many grantees have faced in collecting
92Treasury officials told us they have sought a legislative solution to provide the agency

with additional flexibility in the use of administrative funds for pandemic relief programs.
Legislation was introduced in 2021 that would provide for such flexibility. State, Local,
Tribal, and Territorial Fiscal Recovery, Infrastructure, and Disaster Relief Flexibility Act, S.
3011, 117th Cong. (2021).

Page 45

GAO-23-105410 Emergency Rental Assistance

and reporting data, Treasury will need to continue to work with grantees
to ensure they submit complete and reliable data. Collecting these data
will better situate Treasury and its Office of Inspector General to oversee
the ERA program. In addition, reporting these data will provide Congress
and taxpayers with a better understanding of the program’s outcomes,
including how grantees used ERA funds and which populations they
served.
Treasury also has yet to complete a detailed assessment of the ERA
program’s susceptibility to improper grantee payments to households.
While Treasury has expressed concerns about its ability to appropriately
support and monitor grantees with existing administrative resources,
these functions are critical in light of the improper payment risks we
identified. Completing a detailed analysis that utilizes grantee payment
data is a necessary first step to better understanding and addressing
improper payment risks. Implementing our prior recommendation on
payment recovery audits would bolster this effort.

Recommendations for
Executive Action

We are making the following three recommendations to Treasury:
The Chief of the Office of Recovery Programs should expediently collect
complete and accurate data, including quarterly payment data and
performance measures required by the Consolidated Appropriations Act,
2021. (Recommendation 1)
The Chief of the Office of Recovery Programs should expediently publish
complete ERA program data, including all required disaggregated
performance measures required by the Consolidated Appropriations Act,
2021, for all applicable quarters from program inception through the end
of the award performance period. Such reporting should include
information necessary for determining data quality, such as the rate of
missing or erroneous data for key data elements. (Recommendation 2)
The Chief of the Office of Recovery Programs should complete a detailed
assessment of the ERA program’s susceptibility to improper payments,
such as a quantitative analysis that incorporates grantee payment data
and other relevant data sources. (Recommendation 3)

Agency Comments

We provided a draft of this report to Treasury for review and comment. In
its written comments, reproduced in appendix II, Treasury agreed with our
recommendations. Treasury also provided technical comments, which we
incorporated as appropriate.

Page 46

GAO-23-105410 Emergency Rental Assistance

We are sending copies of this report to the appropriate congressional
committees, the Secretary of the Treasury, and other interested parties.
In addition, the report is available at no charge on the GAO website at
http://www.gao.gov.
If you or your staff have any questions about this report, please contact
me at (202) 512-8678 or GarciaDiazD@gao.gov. Contact points for our
Offices of Congressional Relations and Public Affairs may be found on
the last page of this report.

Daniel Garcia-Diaz
Managing Director, Financial Markets and Community Investment

Page 47

GAO-23-105410 Emergency Rental Assistance

List of Committees
The Honorable Patrick Leahy
Chairman
The Honorable Richard Shelby
Vice Chairman
Committee on Appropriations
United States Senate
The Honorable Sherrod Brown
Chairman
The Honorable Pat Toomey
Ranking Member
Committee on Banking, Housing, and Urban Affairs
United States Senate
The Honorable Ron Wyden
Chairman
The Honorable Mike Crapo
Ranking Member
Committee on Finance
United States Senate
The Honorable Patty Murray
Chair
The Honorable Richard Burr
Ranking Member
Committee on Health, Education, Labor, and Pensions
United States Senate
The Honorable Gary C. Peters
Chairman
The Honorable Rob Portman
Ranking Member
Committee on Homeland Security and Governmental Affairs
United States Senate
The Honorable Rosa L. DeLauro
Chair
The Honorable Kay Granger
Ranking Member
Committee on Appropriations
House of Representatives

Page 48

GAO-23-105410 Emergency Rental Assistance

The Honorable Frank Pallone, Jr.
Chairman
The Honorable Cathy McMorris Rodgers
Republican Leader
Committee on Energy and Commerce
House of Representatives
The Honorable Maxine Waters
Chairwoman
The Honorable Patrick McHenry
Ranking Member
Committee on Financial Services
House of Representatives
The Honorable Bennie G. Thompson
Chairman
The Honorable John Katko
Ranking Member
Committee on Homeland Security
House of Representatives
The Honorable Carolyn B. Maloney
Chairwoman
The Honorable James Comer
Ranking Member
Committee on Oversight and Reform
House of Representatives
The Honorable Richard Neal
Chair
The Honorable Kevin Brady
Republican Leader
Committee on Ways and Means
House of Representatives

Page 49

GAO-23-105410 Emergency Rental Assistance

Appendix I: Objectives, Scope, and
Methodology
Appendix I: Objectives, Scope, and
Methodology

This report examines (1) factors that affected the timeliness of
Emergency Rental Assistance (ERA) program payments and grantee
spending, (2) the Department of the Treasury’s reallocation of excess
funds and its effect on grantees, (3) characteristics of recipient
households and grantee spending, and (4) Treasury’s oversight of
grantee data reporting and improper payment risks.

Factors That Affected
Timeliness of Payments
and Spending

To address our first objective, we calculated and reviewed expenditures
ratios from Treasury’s ERA monthly compliance report for January 2021–
June 2022 to identify initial spending delays and expenditure ratios by
month. 1 We then interviewed ERA grantees and reviewed program
improvement plans to identify factors that enhanced and hindered ERA
program implementation. We interviewed a nongeneralizable sample of
21 ERA1 grantees, which we selected to represent a mix of grantees
based on expenditure ratio, type (state, territorial, or local grantees), and
Census Bureau division and region. 2 We selected four state grantees, 16
city and county (local) grantees, and one territorial grantee to ensure the
sample was generally representative of the types of grantees in the
overall population. We did not select tribal grantees because of
programmatic differences and data limitations.
For state and local grantees, we randomly selected an equal number of
lower- and higher-performing grantees from each type. Specifically, we
selected grantees that had expenditure ratios below 30 percent (lower
performers) or above 65 percent (higher performers) through September

1We assessed the reliability of Treasury’s monthly data by reviewing technical

documentation, interviewing knowledgeable agency officials, and conducting electric
testing for outliers and errors. We found the data to be sufficiently reliable for reporting
general spending trends and grantee expenditures.

2Congress twice appropriated funding for the ERA program (in December 2020 and in

March 2021). We refer to the two appropriations as ERA1 and ERA2. We selected and
interviewed the following 21 grantees: Brown County (Wisconsin), City of Philadelphia
(Pennsylvania), City of Santa Clarita (California), City of Oklahoma City (Oklahoma),
Clermont County (Ohio), Gaston County (North Carolina), Gloucester County (New
Jersey), Henry County (Georgia), Marion County (Oregon), Onondaga County (New
York), Pima County (Arizona), Polk County (Iowa), Puerto Rico, Ramsey County
(Minnesota), Rockingham County (New Hampshire), State of New York, State of North
Carolina, State of Nevada, State of North Dakota, Tuscaloosa County (Alabama), and
Yuma County (Arizona).

Page 50

GAO-23-105410 Emergency Rental Assistance

Appendix I: Objectives, Scope, and
Methodology

2021, based on Treasury’s November 2021 data. 3 We also selected
these grantees equally from each Census division and region to ensure
geographic representation. For territorial grantees, we selected the
territorial grantee that received the largest ERA1 allocation.
We conducted interviews with the 21 grantees and on-site visits with five
of the grantees to collect information on their administrative operations,
policies and procedures, and factors they identified that enhanced or
hindered implementation. We developed interview questions based on a
background review of existing studies on ERA grantee performance, as
well as interviews we conducted with the National Council of State
Housing Agencies and National Low-Income Housing Coalition. We pretested the questions with two state grantees we selected based on
convenience. We then interviewed the 21 grantees and collected
responses using structured interview questions and responses, which we
summarized to identify common themes.
We also reviewed and summarized the spending obstacles grantees cited
in the program improvement plans they submitted to Treasury. We
collected and reviewed 127 program improvement plans (of 140 in total)
that Treasury received from grantees by December 15, 2021. The plan
document includes three narrative fields in which grantees were asked to
identify the three key obstacles to increasing their ERA1 expenditures
and increasing the number of households served. We reviewed those
fields and categorized the contents of each into common themes. We
then compared the common themes we identified across the interviews
and program improvement plans and synthesized a series of
implementation challenges.

Reallocation of Excess
Funds

To address our second objective, we first reviewed how ERA1 allocations
in each state varied before and after reallocation relative to their lowincome renter populations.
•

We determined the low-income renter population in each state by
state using the Department of Housing and Urban Development’s
(HUD) Comprehensive Housing Affordability Strategy to measure lowincome renter population. The data are a custom tabulation of 2014–

3We used these expenditure ratios to capture grantees that did not meet Treasury’s first

spending target to avoid reallocation, as well as those that met the ERA1 statute’s
spending requirement to receive reallocated funds. Although the ERA1 statute requires
Treasury to reallocate funds based on a 65-percent obligations rate, we used the
expenditure ratio because obligations data were not available.

Page 51

GAO-23-105410 Emergency Rental Assistance

Appendix I: Objectives, Scope, and
Methodology

2018 Census data on low-income renter populations by county and
state, which apply the same definition for low-income renters as the
ERA law (not exceeding 80 percent of the area median income).
•

We used Treasury’s reallocation reports to identify amounts
recaptured from and reallocated to each state and we calculated the
allocation in each state after reallocation.

•

We calculated the allocation per low-income renter household in each
state before and after reallocation to identify how much the supply of
funds varied and changed, especially in states that received the
minimum allocation ($200 million).

•

We then calculated the percentage change in allocation in each state
to identify those that had a substantial change (5 percent or more),
which could indicate a potential misalignment between allocations in
each state and their renter needs and grantee capacity.

We also interviewed Treasury officials and reviewed agency reallocation
guidance, procedures, and data to determine how Treasury reallocated
excess ERA1 funds. We reviewed and summarized ERA1 reallocation
payments using Treasury’s reallocation reports through October 2022,
including how funds were recaptured and reallocated based on payment
and grantee type and geography.
In addition, we reviewed how reallocation payments aligned with renter
needs and grantee capacity. We collected and reviewed requests for
reallocated funds and Treasury’s internal decision documentation to
identify which grantees requested reallocated funds and which grantees
had their requests prioritized and approved. We then reviewed the
amounts grantees received, if any, and the extent to which payments
aligned with grantees’ requests and prior obligations and expenditures—
an indication of renter need and grantee capacity.

Grantee Use of Funds and
Characteristics of
Recipient Households

To address our third objective, we analyzed Treasury’s ERA payment and
demographic data for calendar year 2021 to examine how grantees used
ERA funds and the characteristics of recipients. We determined the
spending trends of grantees by analyzing the proportion of payments by
payment type and recipient type, as well as differences in spending
trends based on geography. To describe differences in ERA spending by
county, we geocoded each reported address to its county and matched
each county to county-level characteristics from the Census’ American

Page 52

GAO-23-105410 Emergency Rental Assistance

Appendix I: Objectives, Scope, and
Methodology

Community Survey 5-year estimates for 2016–2020. 4 These
characteristics included median gross rent, percentage of rent-burdened
households (those paying at least 30 percent of their income on rent), and
percentage of households with internet access. We calculated the
percentage of immigrants in a county. We calculated the number of lowincome rental units and the percentage of low-income rental units within a
county using HUD’s Comprehensive Housing Affordability Strategy for
2014–2018.
We also determined the urban or rural status of counties by using data
from the Department of Agriculture’s Economic Research Service to
classify counties. 5 Counties were coded as urban if they had one or more
high-density urban area with 50,000 or more residents or were outlying
counties that were economically tied to such central counties, as
measured by the share of workers commuting to them on a daily basis.
Counties were coded as rural if they were outside the boundaries of
urban areas and had no cities with 50,000 or more residents. In addition,
we reviewed the linear relationship (Pearson’s correlation coefficient)
between county-level characteristics and total ERA payments by county
to determine whether higher or lower ERA payments were associated
with certain characteristics. The correlations we report are statistically
significant at the 95 percent confidence interval.
We analyzed demographic data by income level, race, ethnicity, and
gender to describe the characteristics of households served by the
program. Because a high proportion of the demographic data had missing
observations, we omitted the first three quarters of 2021 reporting from
our analysis. We also omitted duplicate observations. To calculate the
percentage of low-income renters served in 2021, we compared payment
data by household to data from HUD’s Comprehensive Housing
Affordability Strategy. We compared the race and ethnicity reported by
4To assess the reliability of the Census data, we reviewed technical information and

determined the data were sufficiently reliable for reporting community characteristics and
population sizes on the county level. Findings from each survey are subject to sampling
errors. The American Community Survey uses a series of monthly samples to produce
annually updated estimates for different geographic units, including counties, across the
United States. The survey collects data on the economic, social, housing, and
demographic characteristics of communities at various geographic levels, including
metropolitan areas, states, and counties.

5To assess the reliability of the Department of Agriculture data, we reviewed technical

documentation and determined the data were sufficiently reliable for categorizing
population density by county.

Page 53

GAO-23-105410 Emergency Rental Assistance

Appendix I: Objectives, Scope, and
Methodology

applicants to HUD’s data to identify variation in the percentage of
households served by race or ethnicity to the percentage of low-income
renters by race or ethnicity in each state. We also compared the
percentage of households served by gender to data on low-income
renters by gender from Census.

Treasury Oversight of
Data Reporting and
Improper Payments

To address our fourth objective, we first reviewed payment and
demographic data grantees reported for calendar year 2021. We
assessed the completeness and reliability of the data by reviewing
summary statistics for key variables, reviewing technical documentation,
and interviewing Treasury officials responsible for maintaining the data. 6
For each reporting period, we determined the percentage of grantees that
reported data for each dataset and calculated the percentage of required
payment and household data reported by grantees. We assessed the
reliability of the data by comparing reported payment lengths and grantee
administrative expense ratios to statutory limits. We also assessed the
application-funding rate by comparing each grantee’s reported number of
households served to the number of complete applications submitted in
2021. We compared the completeness and reliability of Treasury’s data
and its reporting of performance measures to the ERA1 statute’s
requirements for data collection and reporting.
We also analyzed Treasury data to identify potentially duplicative
payments. We reviewed the number and proportion of payments for
calendar year 2021 that were made by more than one grantee to the
same household. We took steps to clean payment and demographic data
by standardizing formatting and removing duplicative and anomalous
observations, among other steps. To facilitate household-level analysis,
we standardized reported household recipient addresses using the U.S.
Postal Service’s Address Management System and generated an
identifier for each unique address. Among reported household addresses,
we dropped observations for which the address was not disclosed, those
for which reported payment amounts were negative, and those reported
by Cabarrus County (North Carolina) because of known issues with
erroneous payment reporting. We also reviewed the data to identify other

6After we provided Treasury with a draft of this report for comment on November 1, 2022,

Page 54

GAO-23-105410 Emergency Rental Assistance

Appendix I: Objectives, Scope, and
Methodology

potential compliance challenges, including payments that exceeded the
allowable length of assistance and cap on administrative expenses.
Lastly, we interviewed Treasury officials and reviewed agency guidance
to determine how Treasury was collecting and reporting required data,
including efforts to improve data quality. We also interviewed Treasury
officials and requested documentation on the status of efforts to manage
improper payment risk, including implementing our prior
recommendations. We compared the status of Treasury’s efforts against
requirements in Appendix C of the Office of Management and Budget’s
Circular A-123.
We conducted this performance audit from September 2021 to December
2022 in accordance with generally accepted government auditing
standards. Those standards require that we plan and perform the audit to
obtain sufficient, appropriate evidence to provide a reasonable basis for
our findings and conclusions based on our audit objectives. We believe
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.

Page 55

GAO-23-105410 Emergency Rental Assistance

Appendix II: Comments from the Department
of the Treasury
Appendix II: Comments from the Department
of the Treasury

Page 56

GAO-23-105410 Emergency Rental Assistance

Appendix II: Comments from the Department
of the Treasury

Page 57

GAO-23-105410 Emergency Rental Assistance

Appendix II: Comments from the Department
of the Treasury

Page 58

GAO-23-105410 Emergency Rental Assistance

Appendix III: GAO Contact and Staff
Acknowledgments
Appendix III: GAO Contact and Staff
Acknowledgments

GAO Contact

Dan Garcia-Diaz, (202) 512-8678 or GarciaDiazD@gao.gov

Staff
Acknowledgments

In addition to the contact named above, Cory Marzullo (Assistant
Director), Brandon Kruse (Analyst in Charge), Chelsea Carter, Efrain
Magallan, Alicia Martinez Melton, John Mingus, Marc Molino, Loren
Lipsey, Ying Long, Julia Robertson, Barbara Roesmann, Farrah Stone,
Juliann Vadera, and Sean Worobec made key contributions to this report.

(105410)

Page 59

GAO-23-105410 Emergency Rental Assistance

GAO’s Mission

The Government Accountability Office, the audit, evaluation, and investigative
arm of Congress, exists to support Congress in meeting its constitutional
responsibilities and to help improve the performance and accountability of the
federal government for the American people. GAO examines the use of public
funds; evaluates federal programs and policies; and provides analyses,
recommendations, and other assistance to help Congress make informed
oversight, policy, and funding decisions. GAO’s commitment to good government
is reflected in its core values of accountability, integrity, and reliability.

Obtaining Copies of
GAO Reports and
Testimony

The fastest and easiest way to obtain copies of GAO documents at no cost is
through our website. Each weekday afternoon, GAO posts on its website newly
released reports, testimony, and correspondence. You can also subscribe to
GAO’s email updates to receive notification of newly posted products.

Order by Phone

The price of each GAO publication reflects GAO’s actual cost of production and
distribution and depends on the number of pages in the publication and whether
the publication is printed in color or black and white. Pricing and ordering
information is posted on GAO’s website, https://www.gao.gov/ordering.htm.
Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537.
Orders may be paid for using American Express, Discover Card, MasterCard,
Visa, check, or money order. Call for additional information.

Connect with GAO

Connect with GAO on Facebook, Flickr, Twitter, and YouTube.
Subscribe to our RSS Feeds or Email Updates. Listen to our Podcasts.
Visit GAO on the web at https://www.gao.gov.

To Report Fraud,
Waste, and Abuse in
Federal Programs

Contact FraudNet:
Website: https://www.gao.gov/about/what-gao-does/fraudnet
Automated answering system: (800) 424-5454 or (202) 512-7700

Congressional
Relations

A. Nicole Clowers, Managing Director, ClowersA@gao.gov, (202) 512-4400, U.S.
Government Accountability Office, 441 G Street NW, Room 7125, Washington,
DC 20548

Public Affairs

Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, DC 20548

Strategic Planning and
External Liaison

Stephen J. Sanford, Managing Director, spel@gao.gov, (202) 512-4707
U.S. Government Accountability Office, 441 G Street NW, Room 7814,
Washington, DC 20548

Please Print on Recycled Paper.

File and source

File: gao-23-105410-emergency-rental-assistance-risk-assessment.pdf
Size: 4,748,575 bytes
SHA-256: 51a6808010cf638f4057fcb6452c7e0e5e022c1678e47dde06d501effede6199
Our copy: gao-23-105410-emergency-rental-assistance-risk-assessment.pdf
Original: www.gao.gov

GAO 23 105410 Emergency Rental Assistance Risk Assessment

Summary

Full text

File and source

Related records