COVID-19: Additional Actions Needed to Improve Accountability and Program Effectiveness of Federal Response (GAO-22-105051, October 2021)

Summary

GAO-22-105051, an October 2021 report to congressional committees, examines the federal government's continued response to the COVID-19 pandemic. As of September 23, 2021, the U.S. had about 43 million reported cases and about 699,000 deaths, according to CDC, with new daily cases generally above 100,000 from late July because of the Delta variant, and about 64 percent of the eligible population fully vaccinated. Six relief laws had been enacted by August 31, 2021, appropriating $4.8 trillion, of which the government had obligated $3.9 trillion and expended $3.4 trillion. GAO says it had made 209 recommendations in COVID-19 reports since June 2020, of which agencies had addressed 33 by September 30, 2021, and makes 16 new recommendations.

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

Cite this document

“COVID-19: Additional Actions Needed to Improve Accountability and Program Effectiveness of Federal Response (GAO-22-105051, October 2021).” Original at www.gao.gov. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/appendix-8-https-files-gao-gov-reports-gao-22-105051-index-html-append-704f27eab1/, SHA-256 f867cf0af0604c839ce485cfe782e542f69fa352ed71dda3717738ebffec0240.

Full text

United States Government Accountability Oﬃce

Report to Congressional Committees

October 2021

COVID-19
Additional Actions Needed to
Improve Accountability and
Program Eectiveness of Federal
Response

GAO-22-105051 www.gao.gov
October 2021

COVID-19
Additional Actions Needed to Improve Accountability
and Program Effectiveness of Federal Response
Highlights of GAO-22-105051, a report to
congressional committees

Why GAO Did This Study What GAO Found
As of September 23, 2021, the U.S. As the nation continues to respond to, and recover from, the COVID-19 pandemic,
had about 43 million reported cases of increases in COVID-19 cases in July, August, and September 2021, primarily due to
COVID-19 and about 699,000 reported the Delta variant of the virus, have hampered these efforts. From the end of July
deaths, according to CDC. The country 2021 to September 23, 2021, the number of new cases reported each day generally
also continues to experience economic exceeded 100,000, according to Centers for Disease Control and Prevention (CDC)
repercussions from the pandemic. data. This was a daily case count not seen since February 2021 (see figure).
Six relief laws, including the CARES Reported COVID-19 Cases per Day in the U.S., Mar. 1, 2020–Sept. 23, 2021
Act, had been enacted as of August
31, 2021, to address the public health
and economic threats posed by
COVID-19. As of that same date (the
most recent for which government-
wide data was available), the federal
government had obligated a total of
$3.9 trillion and expended $3.4 trillion
of the $4.8 trillion in COVID-19 relief
funds that had been appropriated by
these six laws, as reported by federal
agencies.
The CARES Act includes a provision
for GAO to report on its ongoing
monitoring and oversight efforts related
to the COVID-19 pandemic. This report
examines the federal government’s
continued efforts to respond to, and
recover from, the COVID-19 pandemic.
Meanwhile, COVID-19 vaccination efforts continue. As of September 23, 2021, about
GAO reviewed data, documents, and 64 percent of the U.S. population eligible for vaccination (those 12 years and older),
guidance from federal agencies about or almost 183 million individuals, had been fully vaccinated, according to CDC.
their activities. GAO also interviewed
federal and state officials, stakeholders The government must remain vigilant and agile to address the evolving COVID-19
from organizations for localities, and pandemic and its cascading impacts. Furthermore, as the administration implements
other stakeholders. the provisions in the COVID-19 relief laws, the size and scope of these efforts—from
distributing funding to implementing new programs—demand strong accountability
What GAO Recommends and oversight. In that vein, GAO has made 209 recommendations across its body of
COVID-19 reports issued since June 2020. As of September 30, 2021, agencies had
GAO is making 16 new
addressed 33 of these recommendations, resulting in improvements including
recommendations for agencies that are
increased oversight of relief payments to individuals and improved transparency of
detailed in this Highlights and in the
decision-making for emergency use authorizations for vaccines and therapeutics.
report.
Agencies partially addressed another 48 recommendations. GAO also raised four
matters for congressional consideration, three of which remain open.
In this report, GAO is making 16 new recommendations, including recommendations
related to fiscal relief funds for health care providers, recovery funds for states and
localities, worker safety and health, and assessing fraud risks to unemployment
insurance programs. GAO’s recommendations, if swiftly and effectively implemented,
can help improve the government’s ongoing response and recovery efforts as well as
help it to prepare for future public health emergencies. GAO’s new findings and
View GAO-22-105051. For more information,
contact Jessica Farb, (202) 512-7114 or
recommendations, where applicable, are discussed below.
farbj@gao.gov.

United States Government Accountability Office
Highlights of GAO-22-105051 (Continued)

Relief for Health Care Providers
A total of $178 billion has been appropriated to the Provider Relief Fund (PRF) to reimburse eligible providers for health
care–related expenses or lost revenues attributable to COVID-19. As of August 31, 2021, the Department of Health and
Human Services (HHS) had allocated and disbursed about $132.5 billion of this amount and had allocated but not yet
disbursed about $21.5 billion; the remaining $24.1 billion was unallocated and undisbursed. On September 10, 2021, HHS
announced that $17 billion of the previously unallocated $24.1 billion would be allocated for a general distribution to a
broad range of providers who could document COVID-related revenue loss and expenses. HHS expected to begin
disbursing the funds in December 2021.
As of September 2021, HHS’s Health Resources and Services Administration (HRSA) had not established time frames for
implementing and completing post payment reviews for all PRF payments. In addition, the agency had not finalized
procedures for recovery of overpayments or recovered the bulk of the overpayments that it had already identified.
Without post-payment oversight to help ensure that relief payments are made only to eligible providers in correct amounts
and to identify unused payments or payments not properly used, HHS cannot fully address stated payment integrity risks
for the PRF and seek to recover overpayments, unused payments, or payments not properly used. GAO recommends
that HRSA take steps to finalize and implement post-payment oversight. Specifically, HRSA should establish time
frames for completing post-payment reviews to promptly address identified risks and identify overpayments
made from the PRF, such as payments made in incorrect amounts or payments to ineligible providers; and it
should finalize procedures and implement post-payment recovery of any PRF overpayments, unused payments,
or payments not properly used. HHS—which includes HRSA—partially agreed with these recommendations.
Coronavirus State and Local Fiscal Recovery Funds
In March 2021, the American Rescue Plan Act of 2021 (ARPA) appropriated $350 billion to the Department of the
Treasury (Treasury) to provide payments from the Coronavirus State and Local Fiscal Recovery Funds (CSLFRF). The
CSLFRF allocates funds to states, the District of Columbia, localities, tribal governments, and U.S. territories to cover a
broad range of costs stemming from the COVID-19 pandemic’s fiscal effects. According to Treasury data, it had
distributed approximately $240 billion from the CSLFRF to recipients as of August 31, 2021 (see figure).
Coronavirus State and Local Fiscal Recovery Funds Allocations and Treasury Distributions as of Aug. 31, 2021, by Recipient Type

Note: For more details, see the Coronavirus State and Local Fiscal Recovery Funds enclosure in appendix I.
a
Non-entitlement units of local government are local governments typically serving populations of less than 50,000.

As of July 2021, some of the 48 states that responded to GAO’s survey reported that they had somewhat less than or
much less than sufficient capacity to report on their use of CSLFRF allocation consistent with federal requirements (17 of
48 states), capacity to disburse the funds (13 of 48 states), and apply appropriate internal controls and respond to
inquiries about requirements (10 of 48 states). In addition, most states (44 of 48) reported that they had taken or planned
to take additional steps—such as hiring new staff or reassigning existing staff—to help them manage their CSLFRF
allocations.
Highlights of GAO-22-105051 (Continued)

As of August 2021, Treasury was developing—but had not finalized or documented—key internal processes and control
activities to monitor recipients’ use of their CSLFRF allocations for allowable purposes and to respond to internal control
and compliance findings. According to officials, these internal processes and control activities were in the development
stage, partly because of the short time frame since ARPA’s enactment and because Treasury’s Office of Recovery
Programs, established in April 2021, continues to work to recruit and onboard key team members.
Until Treasury properly designs and documents policies and procedures to guide CSLFRF program officials and other
responsible oversight parties in the Office of Recovery Programs, there is a risk that key control activities needed to help
ensure program management fulfills its recipient monitoring and oversight responsibilities may not be established or
applied effectively and consistently. This risk may be particularly acute with respect to monitoring state and local
recipients that face capacity challenges in managing their CSLFRF allocations in accordance with federal requirements,
as some survey respondents noted. GAO recommends that Treasury design and document timely and sufficient
policies and procedures for monitoring CSLFRF recipients to provide assurance that recipients are managing
their allocations in compliance with laws, regulations, agency guidance, and award terms and conditions.
Treasury agreed with the recommendation.
Unemployment Insurance Fraud Risk Management
GAO continues to have concerns about potential fraud in the unemployment insurance (UI) program, including concerns
about Department of Labor (DOL) efforts to assess and manage program fraud risks. During the pandemic, fraudulent and
potentially fraudulent activity has increased substantially and new types of fraud have emerged, according to DOL
officials. For example, in June 2021, DOL’s Office of Inspector General reported that it had identified nearly $8 billion in
potentially fraudulent UI benefits paid from March 2020 through October 2020. Improper payments have also been a long-
standing concern in the regular unemployment insurance program, suggesting that the program may be vulnerable to
fraud. While DOL continues to identify and implement strategies to address potential fraud and has some ongoing
program integrity activities, it has not comprehensively assessed fraud risks in alignment with leading practices identified
in GAO’s Fraud Risk Framework, which by law must be incorporated in guidelines established by the Office of
Management and Budget for agencies.
DOL has not clearly assigned defined responsibilities to a dedicated entity for designing and overseeing fraud risk
management activities. Without a dedicated entity with defined responsibilities to lead antifraud initiatives, including the
process of assessing fraud risks to UI programs, DOL may not be strategically managing UI fraud risks. GAO
recommends that DOL designate a dedicated entity and document its responsibilities for managing the process
of assessing fraud risks to the unemployment insurance program, consistent with leading practices as provided
in GAO’s Fraud Risk Framework. This entity should have, among other things, clearly defined and documented
responsibilities and authority for managing fraud risk assessments and for facilitating communication among
stakeholders regarding fraud-related issues. DOL neither agreed nor disagreed with this recommendation.
DOL also has not comprehensively assessed UI fraud risks in alignment with leading practices identified in GAO’s Fraud
Risk Framework. These leading practices call for federal managers to plan regular fraud risk assessments and determine
their fraud risk profile, among other things. Such assessments would provide reasonable assurance that DOL has
identified the most significant fraud risks for the regular UI program that will exist after the pandemic. For example, some
fraud risks identified in the CARES Act UI programs may continue to exist in the regular UI program after the temporary UI
programs expire. GAO recommends that DOL (1) identify inherent fraud risks facing the unemployment insurance
program, (2) assess the likelihood and impact of inherent fraud risks facing the program, (3) determine fraud risk
tolerance for the program, (4) examine the suitability of existing fraud controls in the program and prioritize
residual fraud risks, and (5) document the fraud risk profile for the program. DOL neither agreed nor disagreed with
these recommendations.
FEMA’s Disaster Relief Fund and Assistance to State, Local, Tribal, and Territorial Governments
The Federal Emergency Management Agency (FEMA) has used the Disaster Relief Fund to respond to the COVID-19
pandemic—the first time the fund has been used during a nationwide public health emergency. For example, from
September 1, 2020 to August 31, 2021, FEMA obligated a total of approximately $26.8 billion through one type of disaster
assistance, Public Assistance, for emergency protective measures, such as eligible medical care, the purchase and
distribution of food, and distribution of personal protective equipment.
GAO found that FEMA inconsistently interpreted and applied its policies for expenses eligible for COVID-19 Public
Assistance within and across its 10 regions. For example, officials in one state said that FEMA at one point had deemed
the provision of personal protective equipment at correctional facilities as ineligible for reimbursement in their region but
that states in other regions had received reimbursement for the same expense. These inconsistencies were due to,
among other things, changes in policies as FEMA used the Public Assistance program for the first time to respond to a
nationwide emergency. FEMA officials stated that it was difficult to ensure consistency in policies as different states and
regions are not experiencing the same things at the same time.
Highlights of GAO-22-105051 (Continued)

FEMA is likely to receive applications for reimbursement for a larger number of projects than it estimated earlier in 2021,
given the surge in COVID-19 cases this summer. To improve the consistency of the agency’s interpretation and
application of the COVID-19 Public Assistance policy, GAO recommends that FEMA further clarify and communicate
eligibility requirements nationwide. GAO also recommends that FEMA require the agency’s Public Assistance
employees in the regions and at its Consolidated Resource Centers to attend training on changes to COVID-19
Public Assistance policy. The Department of Homeland Security—which includes FEMA— agreed with both of these
recommendations.
Loans for Aviation and Other Eligible Businesses
Treasury has executed 35 loan agreements with certain aviation businesses and other businesses deemed critical to
maintaining national security. These loans have totaled about $22 billion of the $46 billion authorized by the CARES Act
for loans and loan guarantees to such businesses. As directed by the CARES Act, Treasury required certain loan
recipients to provide financial assets, such as warrants that give the federal government an option to buy shares of stock
at a predetermined price before a specified date, to protect taxpayer interests.
According to Treasury officials, it is likely that, if the airline industry continues to recover and borrowers do not default, the
warrants could have higher values than the predetermined price Treasury would have to pay to act on them. Treasury has
not exercised any of the warrants for stock it received from nine businesses, nor has it developed policies and procedures
for determining when to act on the warrants to benefit the taxpayer. GAO recommends that Treasury develop policies
and procedures to determine when to act on warrants obtained as part of the loan program for aviation and other
eligible businesses to benefit the taxpayers. Treasury agreed with this recommendation.
Payroll Support Assistance to Aviation Businesses
As of September 2021, Treasury had made payments totaling $59 billion of $63 billion provided for the Payroll Support
Programs to support aviation business. These payments were to be used exclusively for the continuation of wages,
salaries, and benefits.
Similar to Treasury’s requirement for loans for aviation and other eligible businesses, Treasury required certain Payroll
Support Program recipients to provide warrants, as allowed by the CARES Act. As of September 2021, 14 recipients had
provided a total of 58 million warrants.
As Treasury continues to hold these warrants for stock purchases, the warrants may increase in value as the airline
industry recovers. Treasury has not exercised any of the warrants for stock it holds in the 14 businesses, nor has it
documented policies and procedures to guide when to act on the warrants to fulfill the statutory purpose to provide
appropriate compensation to the federal government. GAO recommends that Treasury develop policies and
procedures to determine when to act on warrants obtained as part of the Payroll Support Program to provide
appropriate compensation to the federal government. Treasury agreed with this recommendation.
COVID-19 Testing
Use is increasing for antigen tests, one of two types of COVID-19 diagnostic and screening tests for which HHS’s Food
and Drug Administration has issued emergency use authorizations. These “rapid” antigen tests typically have a
turnaround time of about 30 minutes or less for results, compared with 1 to 3 days for molecular tests, the second type of
test HHS authorized. Antigen tests can be conducted at doctors’ offices or in homes or other settings; some antigen tests
can be conducted without a prescription.
Since June 2020, HHS has worked to encourage and improve the reporting of antigen testing data to local, state, and
federal health officials. However, HHS officials told GAO reporting of antigen test results is incomplete, which prevents
HHS from using antigen testing data for COVID-19 surveillance. HHS is taking additional steps aimed at improving
reporting of antigen test data. For example, officials told GAO that HHS will continue to make enhancements to data
reporting by building reporting methods into the testing process, such as for testing in schools and workplaces.
HHS is also considering surveillance approaches to supplement or enhance current surveillance efforts. For example,
HHS is exploring wastewater surveillance approaches, which provide data that can complement and confirm other forms
of surveillance for COVID-19 and an efficient pooled community sample that is particularly useful in areas where timely
COVID-19 clinical testing is underutilized or unavailable, according to HHS officials.
Worker Safety and Health
The Occupational Safety and Health Administration (OSHA) faced challenges in enforcing workplace safety and health
standards during the COVID-19 pandemic, but the agency has not assessed lessons learned or promising practices.
According to inspectors from area offices, they faced challenges related to resources and to communication and
guidance, such as a lack of timely guidance from OSHA headquarters. GAO recommends that OSHA assess—as soon
as feasible and, as appropriate, periodically thereafter—various challenges related to resources and to
communication and guidance that the agency has faced in its response to the COVID-19 pandemic and take
Highlights of GAO-22-105051 (Continued)

related actions as warranted. The Department of Labor—which includes OSHA—partially agreed with this
recommendation.
Advance Child Tax Credit Payments
ARPA temporarily expanded eligibility for the child tax credit (CTC) to additional qualified individuals by eliminating a
requirement that individuals must earn a minimum amount annually to be eligible. ARPA also temporarily increased the
maximum amount of the CTC from $2,000 per qualifying child to $3,000 or $3,600, depending on the child’s age. As
required by ARPA, the Internal Revenue Service (IRS) and Treasury are responsible for issuing half of the CTC through
periodic advance payments, known as advance CTC payments.
IRS reported disbursing more than 106 million advance payments totaling over $45.5 billion as of September 25, 2021
(see figure).
Dollar Amount and Count of Advance Child Tax Credit Payments, by Month, as of Sept. 25, 2021

IRS is conducting and planning several outreach efforts to increase the public’s awareness of advance CTC payments.
However, IRS and Treasury have not developed a comprehensive estimate of individuals who are potentially eligible for
advance CTC payments and the agencies have not set a participation goal. Such an estimate would enable Treasury and
IRS to measure the tax credit’s participation rate, providing greater clarity regarding populations at risk of not receiving the
payments. GAO recommends that Treasury, in coordination with IRS, estimate the number of individuals,
including nonfilers, who are eligible for advance CTC payments, measure the 2021 participation rate based on
that estimate, and use that estimate to develop targeted outreach and communications efforts for the 2022 filing
season; the participation rate could include individuals who opt in and out of the advance payments. Treasury
neither agreed nor disagreed with this recommendation.
Child Nutrition
Child nutrition programs administered by the Department of Agriculture’s Food and Nutrition Service (FNS) supply cash
reimbursements to schools or other programs for meals and snacks provided to eligible children nationwide. In fiscal year
2019, before the pandemic, the four largest programs—the National School Lunch Program, School Breakfast Program,
Summer Food Service Program, and Child and Adult Care Food Program—along with other child nutrition programs,
received $23.1 billion in federal funds. During a typical year, two of these programs—the National School Lunch Program
and the School Breakfast Program—subsidize meals for nearly 30 million children in approximately 95,000 elementary
and secondary schools nationwide.
As of July 2021, FNS officials were unable to provide a plan showing how FNS intends to comprehensively analyze
lessons learned during the pandemic, such as from operational and financial challenges. Further, according to FNS
officials, while the School Meals Operations study—launched in spring 2021—is surveying school districts and state
agencies that administer the federal child nutrition programs, the study is not gathering local perspectives directly from
child care centers and day care homes or other local program sponsors that are not school districts. As a result, FNS may
miss opportunities to identify lessons learned and will lack comprehensive information to aid its future planning. GAO
recommends that the Department of Agriculture document its plan to analyze lessons learned from operating
child nutrition programs during the COVID-19 pandemic. This plan should include a description of how the
department will gather perspectives of key stakeholders, such as Child and Adult Care Food Program institutions
and nonschool Summer Food Service Program sponsors. The Department of Agriculture—which includes FNS—
agreed with this recommendation.
Contents

Recommendations for Executive Action................................................................................ 1

Introduction................................................................................................................................... 3

Background.................................................................................................................................... 8
Public Health and Economic Eﬀects.................................................................................................. 8
Federal COVID-19 Funding and Spending........................................................................................ 12

Executive Summary................................................................................................................... 19
Overview..................................................................................................................................... 19
Relief for Health Care Providers..................................................................................................... 19
Coronavirus State and Local Fiscal Recovery Funds........................................................................... 20
Unemployment Insurance Fraud Risk Management.......................................................................... 21
FEMA’s Disaster Relief Fund and Assistance to State, Local, Tribal, and Territorial Governments............... 23
Loans for Aviation and Other Eligible Businesses............................................................................. 24
Payroll Support Assistance to Aviation Businesses............................................................................ 25
COVID-19 Testing......................................................................................................................... 25
Worker Safety and Health............................................................................................................. 26
Advance Child Tax Credit Payments................................................................................................ 27
Child Nutrition............................................................................................................................. 28

Conclusions.................................................................................................................................. 30

Closing........................................................................................................................................... 31

Congressional Addressees....................................................................................................... 32

Appendixes and Enclosures..................................................................................................... 34
Appendix I: Enclosures.................................................................................................................. 34
Economic Indicators..................................................................................................................... 34
COVID-19 Testing......................................................................................................................... 44
FDA Oversight of COVID-19 Vaccine Manufacturing Quality................................................................ 55
FDA Inspections of Biologic Manufacturing during the COVID-19 Pandemic.......................................... 64
Health Insurance Loss.................................................................................................................. 72
HHS COVID-19 Funding................................................................................................................. 81
Relief for Health Care Providers..................................................................................................... 90
Veterans Health Care.................................................................................................................. 101
HHS Cybersecurity..................................................................................................................... 110
Worker Safety and Health........................................................................................................... 115
Child Care................................................................................................................................. 139
K-12 Education.......................................................................................................................... 148

Page i GAO-22-105051
Child Nutrition........................................................................................................................... 155
Unemployment Insurance Programs............................................................................................. 169
Unemployment Insurance Fraud Risk Management........................................................................ 185
Emergency Rental Assistance....................................................................................................... 200
Advance Child Tax Credit and Economic Impact Payments............................................................... 209
Agricultural Debt Relief............................................................................................................... 222
Shuttered Venue Operators Grant................................................................................................ 226
Restaurant Revitalization Fund..................................................................................................... 233
Assistance for Fishery Participants............................................................................................... 240
Food Safety Inspections.............................................................................................................. 247
Airport Grants........................................................................................................................... 252
Payroll Support Assistance to Aviation Businesses.......................................................................... 262
Loans for Aviation and Other Eligible Businesses........................................................................... 271
Aviation Manufacturing Jobs Protection Program............................................................................ 277
Transit Industry.......................................................................................................................... 282
Coronavirus Economic Relief for Transportation Services (CERTS)..................................................... 288
Paycheck Protection Program...................................................................................................... 293
Coronavirus State and Local Fiscal Recovery Funds......................................................................... 300
Federal Contracts and Agreements for COVID-19............................................................................ 316
Vaccines Provided Abroad........................................................................................................... 327
International Trade..................................................................................................................... 338
Federal Fraud-Related Cases........................................................................................................ 345
FEMA’s Disaster Relief Fund and Assistance to State, Local, Tribal, and Territorial Governments............. 353
COVID-19 Impact on IRS Enforcement........................................................................................... 369
Postal Service............................................................................................................................ 378
Appendix II: Highlights Pages from Recently Issued GAO COVID-19 Products...................................... 384
Contracting Flexibilities............................................................................................................... 384
VA CLC Infection Prevention........................................................................................................ 386
Contractor Paid Leave Reimbursements........................................................................................ 387
Child Welfare............................................................................................................................. 388
Implementation of Paycheck Protection Program........................................................................... 389
BOP Response to COVID-19......................................................................................................... 390
Assessing Vendors and Contracting Lessons Learned...................................................................... 391
Oversight of COVID-19 Assistance................................................................................................ 392
Economic Injury Disaster Loans................................................................................................... 393
Biodefense Preparedness............................................................................................................ 395
Hospital Capacity Data Collection................................................................................................. 396
Federal Debt Management.......................................................................................................... 397
Immigration Courts.................................................................................................................... 398
Medicaid HCBS Waivers.............................................................................................................. 399
Food Box Program..................................................................................................................... 400
Exposure Notiﬁcation................................................................................................................. 401
Characteristics of Paycheck Protection Program Borrowers.............................................................. 403

Page ii GAO-22-105051
DOD Pandemic Response........................................................................................................... 404
Dedicated User Fees.................................................................................................................. 405
Telework Cybersecurity............................................................................................................... 406
HUD Oversight.......................................................................................................................... 407
IHS COVID-19 Contracting........................................................................................................... 408
Federal Reserve Lending Programs............................................................................................... 409
IRS IT Investments..................................................................................................................... 410
Appendix III: List of Ongoing GAO Work Related to COVID-19 as of September 30, 2021....................... 411
Appendix IV: Comments from the Department of Health and Human Services.................................... 414
Appendix V: Comments from the Department of Homeland Security................................................ 417
Appendix VI: Comments from the Internal Revenue Service............................................................. 421
Appendix VII: Comments from the Department of Labor, Employment and Training Administration....... 423
Appendix VIII: Comments from the Department of Labor, Occupational Safety and Health
Administration........................................................................................................................... 426
Appendix IX: Comments from the Department of the Treasury......................................................... 427
Appendix X: Comments from the United States Postal Service......................................................... 430
Appendix XI: Comments from the Department of Veterans Aﬀairs.................................................... 432

Contacts..................................................................................................................................... 433

Tables
Table 1: COVID-19 Relief Appropriations, Obligations, and Expenditures, as of Aug. 31, 2021................... 14
Table 2: Appropriations, Obligations, and Expenditures for Federal Programs and Funds Receiving $10 Billion
or More in COVID-19-Related Aid for States, the District of Columbia, Localities, U.S. Territories, and Tribes, as
of Aug. 31, 2021.......................................................................................................................... 16
Number of FDA Emergency Use Authorizations for COVID-19 Antigen Tests by Setting, as of October 14,
2021........................................................................................................................................... 47
Prior GAO Recommendations Related to COVID-19 Testing................................................................ 53
Selected Studies of Employer-Sponsored Insurance (ESI) Loss during the COVID-19-Associated Economic
Downturn.................................................................................................................................... 74
HHS-Reported COVID-19 Relief Appropriations, Obligations, and Expenditures, by Relief Law, as of August 31,
2021........................................................................................................................................... 83
HHS-Reported Allocations, Obligations, and Expenditures of COVID-19 Relief Funding, by Agency or Key
Fund, as of August 31, 2021.......................................................................................................... 84
HHS-Reported Allocations, Obligations, and Expenditures by Selected COVID-19 Response Activity, as of
August 31, 2021........................................................................................................................... 86
Prior GAO Recommendation Related to Department of Health and Human Services (HHS) COVID-19
Funding...................................................................................................................................... 89
Provider Relief Fund: Summary of Allocations and Disbursements, as of August 31, 2021....................... 92
Prior GAO Recommendation Related to Department of Health and Human Services (HHS) Relief to
Providers.................................................................................................................................. 100

Page iii GAO-22-105051
Number of GAO-Identiﬁed Information Security Program and System Control Deﬁciencies at the U.S.
National Institutes of Health and Associated Recommendations by Core Security Function................... 112
Prior GAO Recommendation Related to Cybersecurity.................................................................... 114
COVID-19-Related Reports to OSHA and OSHA Enforcement Actions, March 2020 through August 2021.. 119
COVID-19-Related Violations and Penalties, for Most Frequently Cited Standards and the General Duty
Clause, from February 2020 through August 2021.......................................................................... 121
Prior GAO Recommendations Related to Worker Safety and Health.................................................. 136
Coronavirus Supplemental Appropriations to the Child Care and Development Fund (CCDF)................. 140
State Uses or Planned Uses of CARES Act and Consolidated Appropriations Act, 2021, Division M (CAA)
Funds....................................................................................................................................... 144
Examples of Vulnerable Student Groups and Barriers to Staying Engaged in School During the COVID-19
Pandemic, According to Oﬃcials from Selected States, School Year 2020-2021.................................... 150
Examples of Reported Eﬀorts to Engage K-12 Students Disconnected by COVID-19 Pandemic during the
2020-2021 School Year, in Selected States..................................................................................... 152
Prior GAO Recommendations Related to Unemployment Insurance (UI) Programs.............................. 183
Summary of Emergency Rental Assistance (ERA) Laws and Amounts Appropriated, Disbursed, and Spent
(dollars in billions) through Aug. 2021........................................................................................... 202
GAO’s Prior Recommendations Related to Economic Impact Payments.............................................. 219
SBA’s Implementation of Suggested Actions in the OIG Shuttered Venue Operators Grant Management Alert,
as of Aug. 2021......................................................................................................................... 231
Disbursement of CARES Act Funding, as of September 15 , 2021...................................................... 243
Prior GAO Recommendation Related to Assistance for Fisheries Participants...................................... 246
FAA Obligations and Expenditures for CARES Act Airport Grants, as of September 27, 2021................... 253
FAA Obligations and Expenditures for the Consolidated Appropriations Act, 2021 Airport Grants, as of
September 27, 2021................................................................................................................... 255
FAA Obligations and Expenditures for the American Rescue Plan Act of 2021 Airport Grants, as of September
27, 2021................................................................................................................................... 256
Information about Notes and Warrants Provided to the Department of the Treasury for the Payroll Support
Program (PSP1), Payroll Support Program Extension (PSP2), and Second Payroll Support Program Extension
(PSP3), as of September 2021...................................................................................................... 266
Prior GAO Recommendation Related to Payroll Support Assistance for Aviation Businesses.................. 270
Executed and Outstanding Loans for the CARES Act Loan Program for Aviation and Other Eligible Businesses
as of October 1, 2021................................................................................................................. 272
FTA Allocations and Transit Agency Obligations and Expenditures of COVID-19 Relief Funds for Formula
Programs, as of Aug. 31, 2021..................................................................................................... 284
Prior GAO Recommendations Related to the Paycheck Protection Program........................................ 298
Prior GAO Recommendations Related to Single Audits.................................................................... 314
Prior GAO Recommendations Related to Federal Contracts and Agreements for COVID-19................... 324
COVAX Key Lines of Eﬀort and Leadership..................................................................................... 330
Other Key Challenges Aﬀecting COVAX Eﬀorts to Deliver Vaccines and Steps Taken to Address These
Challenges, as Reported by COVAX Partners.................................................................................. 335
Prior GAO Recommendations Related to COVID-19......................................................................... 367

Page iv GAO-22-105051
Enforcement Caseload Trends for IRS Examination, Automated Underreporter, and Collection Activities for
June 2019, 2020, and 2021.......................................................................................................... 371
Enforcement Revenue in June 2019, 2020, and 2021 for Selected Major Enforcement Activities.............. 373
Amounts Requested by USPS under the CARES Act, as Amended, for COVID-19-Related Operating Expenses
Incurred Mar. 2020–May 2021..................................................................................................... 381

Figures
Reported COVID-19 Cases per Day in the U.S., Mar. 1, 2020–Sept. 23, 2021...........................................H 1
Coronavirus State and Local Fiscal Recovery Funds Allocations and Treasury Distributions as of Aug. 31,
2021, by Recipient Type.................................................................................................................H 2
Dollar Amount and Count of Advance Child Tax Credit Payments, by Month, as of Sept. 25, 2021..............H5
Figure 1: Report Enclosures by Topic Area......................................................................................... 5
Figure 2: Status of Prior GAO Recommendations from COVID-19-Related Work, by Federal Department or
Agency, as of Sept. 30, 2021............................................................................................................ 7
Figure 3: Reported COVID-19 Cases per Day in the U.S., Mar. 1, 2020–Sept. 23, 2021................................ 9
Figure 4: Higher-Than-Expected Weekly Mortality in the U.S., Jan. 2020 to Sept. 2021............................... 9
Figure 5: Daily Count of COVID-19 Vaccine Doses Administered in the U.S. and Reported to CDC, Dec. 14,
2020–Sept. 23, 2021..................................................................................................................... 11
Figure 6: Employment-to-Population Ratio, Jan. 2019–Sept. 2021........................................................ 12
Figure 7: Percentage of COVID-19 Relief Appropriations Obligated and Expended, July 31, 2020–Aug. 31,
2021........................................................................................................................................... 13
Indicators for Areas of the Economy Supported by the Federal COVID-19 Pandemic Response, June–Sept.
2021, Cumulative Changes since Feb. 2020..................................................................................... 35
Employment-to-Population Ratio, Jan. 2019–Sept. 2021..................................................................... 36
Serious Delinquency Rates on Single-Family Residential Mortgages, Jan. 2019–Aug. 2021....................... 37
Indicators of Inﬂation, Feb.–Aug. 2021, and Average Inﬂation Rates, 2000–2019.................................... 39
Percentage Change in Inﬂation Indicators over the Previous Month, Jan. 2019-Aug. 2021........................ 40
Public Health Reporting Requirements or Guidance by Antigen Test Authorized Settings........................ 48
Establishments Manufacturing the COVID-19 Vaccines, by the Date of the Most Recent Historical FDA
Inspection Prior to Initial Emergency Use Authorization (EUA) Request and Manufacturing Type.............. 58
Establishments Manufacturing the COVID-19 Vaccines Not Authorized or Licensed, by Date of Their Most
Recent FDA Inspections and Manufacturing Type, as of May 2021...................................................... 61
FDA Biologic Inspections by Year and Type of Inspection, Fiscal Years 2012–2021.................................. 66
Number of FDA Biologic Inspections by Month, Fiscal Years 2019–2021............................................... 67
Number of FDA Inspections and Percentage of Foreign and Domestic Inspections by Biologic Product Type,
Fiscal Year 2019........................................................................................................................... 69
Plan Selections during Open Enrollment Periods in Federal and State-Based Exchanges, 2017-2021......... 77
Federal Exchange Special Enrollment by Qualifying Event, 2017-2020.................................................. 78
Medicaid Enrollment from January to December 2020...................................................................... 79
HHS-Reported COVID-19 Relief Appropriations, Obligations, and Expenditures from COVID-19 Relief Laws, as
of August 31, 2021....................................................................................................................... 81
Timeline of OSHA’s Key Actions to Respond to the COVID-19 Pandemic, from January 2020 through
September 2021........................................................................................................................ 117

Page v GAO-22-105051
COVID-19-Related Reports to OSHA among Industries Targeted by OSHA’s National Emphasis Program,
February 2020 through August 2021............................................................................................ 127
OSHA Enforcement Activities Based on Report Type, for Selected Periods from March 2019 through August
2021......................................................................................................................................... 130
Percentage of Open Child Care Providers Increased during 2020...................................................... 141
Financial Problems for Child Care Providers Topped States’ List of Greatest Child Care Challenges in
2020......................................................................................................................................... 142
Federal Child Care Flexibilities States Most Commonly Used During COVID-19.................................... 143
Small Child Care Providers Received More Paycheck Protection Program Loans than Larger Providers.... 146
Total Meals Served in Key Child Nutrition Programs in 2019, 2020, and 2021, by Month, as of Sept. 10,
2021......................................................................................................................................... 159
Weekly Continued Claims Submitted Nationwide for Regular Unemployment Insurance, Pandemic
Emergency Unemployment Compensation, and Extended Beneﬁts, Mar. 1, 2020–Sept. 4, 2021............. 174
Timeliness of First Payments of Regular Unemployment Insurance (UI) Beneﬁts, Jan. 2020–Aug. 2021..... 176
First-Round Emergency Rental Assistance (ERA1) Spending and Unique Renter Households Served, by
Reporting Period (Jan.–Aug. 2021)................................................................................................ 203
Percentage of First-Round Emergency Rental Assistance (ERA1) Allocations Spent by Grantees in U.S. States
and Territories through Aug. 2021................................................................................................ 204
Income Qualiﬁcations for the Expanded Child Tax Credit, Based on Tax Year 2021............................... 210
Advance Child Tax Credit Disbursals, by Month, as of Sept. 25, 2021................................................. 211
Shuttered Venue Operators Grant (SVOG) Initial Grant Recipients, by Grant Amount, as of Sept. 20,
2021......................................................................................................................................... 227
Timeline for the Shuttered Venue Operators Grant Program, as of Sept. 20, 2021............................... 229
Timeline for the Restaurant Revitalization Fund Program, as of July 2021........................................... 235
Restaurant Revitalization Fund Recipients, as a Percentage of the Total Number of Restaurants with 500 or
Fewer Employees in Each State, as of June 30, 2021....................................................................... 237
Restaurant Revitalization Fund (RRF) Recipients, by Assistance Amount, as of June 30, 2021.................. 238
Payment Amount and Number of Agreements Signed for the Payroll Support Program (PSP1), Payroll
Support Program Extension (PSP2), and Second Payroll Support Program Extension (PSP3), as of September
2021......................................................................................................................................... 264
Coronavirus State and Local Fiscal Recovery Funds Allocations and Treasury Distributions as of Aug. 31,
2021, by Recipient Type.............................................................................................................. 301
Treasury’s Reporting Requirements for the Coronavirus State and Local Fiscal Recovery Funds, by Recipient
Type, as of October 7, 2021........................................................................................................ 303
Number of States and Localities That Reported Having a Proposed or Adopted Plan for Spending Their
CSLFRF Allocations, as of July 2021............................................................................................... 304
States’ and Localities’ Reported Capacity for Managing Their CSLFRF Allocations, as of July 2021............ 306
States’ and Localities’ Views on Clarity of Treasury’s CSLFRF Guidance on Allowable Uses of Funds and
Reporting Requirements, as of July 2021....................................................................................... 307
Contract Obligations in Response to COVID-19 by Federal Agency, as of September 30, 2021................ 318
Government-wide COVID-19-Related Contract Obligations and Conﬁrmed COVID-19 Cases by Month, Feb.
2020–Sept. 2021........................................................................................................................ 319

Page vi GAO-22-105051
Contract Obligation Amounts for Top Five Goods and Services Procured in Response to COVID-19 by month,
Feb. 2020–Sept. 2021................................................................................................................. 320
COVAX Progress toward Goal of Delivering 2 Billion Vaccine Doses to Participating Countries in 2021, as of
Sept. 15, 2021........................................................................................................................... 332
Monthly U.S. Imports of COVID-19-Related Products, by Product Type, Jan. 2018–June 2021.................. 339
U.S. Trade of COVID-19 Vaccines with the European Union and United Kingdom, Jan.–May 2021............ 341
Number of Individuals or Entities Who Have Pleaded Guilty to or Faced Federal Charges for Consumer Fraud,
as of July 31, 2021..................................................................................................................... 349
FEMA’s Disaster Relief Fund Balance, by Month, Feb. 2020– Aug. 2021............................................... 354
Number of Projects and Amounts Obligated for Public Assistance for COVID-19, by FEMA Region, Sept. 1,
2020–Aug. 31, 2021.................................................................................................................... 356
FEMA’s Disaster Relief Fund Obligations for COVID-19 by Program and Activity through August 2021...... 357
Internal Revenue Service (IRS) Staﬀ Work Status, Mar. 2020–June 2021.............................................. 370
IRS Pandemic-Related Enforcement Challenges and Actions through June 2021.................................. 374
Change in Mail Volume for Selected Mail Types between Third Quarters in Fiscal Year 2020 and 2021 (in
thousands of piece of mail)......................................................................................................... 379

Page vii GAO-22-105051
Abbreviations

ARPA American Rescue Plan Act of 2021
CDC Centers for Disease Control and Prevention
COVID-19 Coronavirus Disease 2019
CSLFRF Coronavirus State and Local Fiscal Recovery Funds
CTC child tax credit
DOL Department of Labor
FDA Food and Drug Administration
FEMA Federal Emergency Management Agency
FNS Food and Nutrition Service
HHS Department of Health and Human Services
HRSA Health Resources and Services Administration
IRS Internal Revenue Service
OSHA Occupational Safety and Health Administration
PRF Provider Relief Fund
UI unemployment insurance
USDA Department of Agriculture

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.

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 viii GAO-22-105051
Recommendations for Executive Action
Recommendations for Executive Action

We are making a total of 16 recommendations to federal agencies:

• The Administrator of the Health Resources and Services Administration should establish
time frames for completing post-payment reviews to promptly address identiﬁed risks
and identify overpayments made from the Provider Relief Fund, such as payments made
in incorrect amounts or payments to ineligible providers. See the Relief for Health Care
Providers enclosure. (Recommendation 1)
• The Administrator of the Health Resources and Services Administration should
ﬁnalize procedures and implement post-payment recovery of any Provider Relief Fund
overpayments, unused payments, or payments not properly used. See the Relief for Health
Care Providers enclosure. (Recommendation 2)
• The Secretary of the Treasury should design and document timely and suﬃcient policies
and procedures for monitoring recipients of Coronavirus State and Local Fiscal Recovery
Funds to provide assurance that recipients are managing their allocations in compliance
with laws, regulations, agency guidance, and award terms and conditions, including
ensuring that expenditures are made for allowable purposes. See the Coronavirus State
and Local Fiscal Recovery Funds enclosure. (Recommendation 3)
• The Secretary of Labor should designate a dedicated entity and document its
responsibilities for managing the process of assessing fraud risks to the unemployment
insurance program, consistent with leading practices as provided in our Fraud
Risk Framework. This entity should have, among other things, clearly deﬁned and
documented responsibilities and authority for managing fraud risk assessments and for
facilitating communication among stakeholders regarding fraud-related issues. See the
Unemployment Insurance Fraud Risk Management enclosure. (Recommendation 4)
• The Secretary of Labor should identify inherent fraud risks facing the unemployment
insurance program. See the Unemployment Insurance Fraud Risk Management enclosure.
(Recommendation 5)
• The Secretary of Labor should assess the likelihood and impact of inherent fraud risks
facing the unemployment insurance program. See the Unemployment Insurance Fraud
Risk Management enclosure. (Recommendation 6)
• The Secretary of Labor should determine fraud risk tolerance for the unemployment
insurance program. See the Unemployment Insurance Fraud Risk Management enclosure.
(Recommendation 7)
• The Secretary of Labor should examine the suitability of existing fraud controls in
the unemployment insurance program and prioritize residual fraud risks. See the
Unemployment Insurance Fraud Risk Management enclosure. (Recommendation 8)
• The Secretary of Labor should document the fraud risk proﬁle for the unemployment
insurance program. See the Unemployment Insurance Fraud Risk Management enclosure.
(Recommendation 9)
• The Federal Emergency Management Agency Administrator should improve the
consistency of the agency’s interpretation and application of the COVID-19 Public
Assistance policy within and across regions by further clarifying and communicating
eligibility requirements nationwide. See the FEMA’s Disaster Relief Fund and Assistance to
State, Local, Tribal, and Territorial Governments enclosure. (Recommendation 10)

Page 1 GAO-22-105051
• The Federal Emergency Management Agency Administrator should require the agency’s
Public Assistance Program employees in the regions and at its Consolidated Resource
Centers to attend training on changes to COVID-19 Public Assistance policy to help
ensure it is interpreted and applied consistently nationwide. See the FEMA’s Disaster
Relief Fund and Assistance to State, Local, Tribal, and Territorial Governments enclosure.
(Recommendation 11)
• The Secretary of the Treasury should develop policies and procedures to determine
when to act on warrants obtained as part of the loan program for aviation and other
eligible businesses to beneﬁt the taxpayers. See the Loans for Aviation and Other Eligible
Businesses enclosure. (Recommendation 12)
• The Secretary of the Treasury should develop policies and procedures to determine when
to act on warrants obtained as part of the Payroll Support Program to provide appropriate
compensation to the federal government. See the Payroll Support Assistance to Aviation
Businesses enclosure. (Recommendation 13)
• The Assistant Secretary of Labor for Occupational Safety and Health should assess—as
soon as feasible and, as appropriate, periodically thereafter—various challenges
related to resources and to communication and guidance that the Occupational Safety
and Health Administration has faced in its response to the COVID-19 pandemic and
should take related actions as warranted. See the Worker Safety and Health enclosure.
(Recommendation 14)
• The Secretary of the Treasury, in coordination with the Commissioner of Internal Revenue,
should estimate the number of individuals, including nonﬁlers, who are eligible for
advance child tax credit payments, measure the 2021 participation rate based on that
estimate, and use that estimate to develop targeted outreach and communications eﬀorts
for the 2022 ﬁling season; the participation rate could include individuals who opt in and
out of the advance payments. See the Advance Child Tax Credit and Economic Impact
Payments enclosure. (Recommendation 15)
• The Secretary of Agriculture should document the Department of Agriculture’s plan to
analyze lessons learned from operating child nutrition programs during the COVID-19
pandemic. This plan should include a description of how the department will gather
perspectives of key stakeholders, such as Child and Adult Care Food Program institutions
and nonschool Summer Food Service Program sponsors. See the Child Nutrition
enclosure. (Recommendation 16)

Page 2 GAO-22-105051
Introduction
October 27, 2021

Congressional Committees

As the nation continues to respond to the Coronavirus Disease 2019 (COVID-19) pandemic,
response and recovery eﬀorts have been hampered by increases in COVID-19 cases, due primarily
to the Delta variant of the virus.1 Although the daily number of new cases had begun to decline
earlier in the summer, the number of new cases reported each day from the end of July 2021, to
September 23, 2021, generally exceeded 100,000, according to the Centers for Disease Control
and Prevention (CDC)—a daily case count not seen since February 2021 and substantially higher
than the approximately 8,000 new cases reported per day in mid-June. As a result of the rise in
cases, CDC, state and local governments, and private businesses revised their mask guidance or
requirements.2

While vaccination eﬀorts continue, vaccination rates across the U.S. vary. As of September
23, 2021, about 64 percent of the U.S population eligible for vaccination (those 12 years and
older)—about 183 million individuals—had been fully vaccinated, according to CDC.

Hospitals reported an average of more than 9,000 individuals hospitalized daily for the 7-day
period from September 17 to September 23, 2021, a decrease from more than 12,000 individuals
hospitalized daily during a 7-day period in August 2021.3 According to CDC, at the end of August
2021, new admissions of patients with conﬁrmed COVID-19 were at their highest levels since the
beginning of the pandemic for all age groups under 50 years old.4 As of the end of September
2021, CDC reported that weekly hospitalization rates for children aged 11 and younger due to
COVID-19 were at their highest since the beginning of the pandemic, although hospitalizations
due to COVID-19 are lower in children than they are in adults. As the pandemic continues, the U.S.
and the world may continue to see ﬂuctuating increases in new cases, making an agile federal
response to the pandemic even more important.

1As of September 2021, the Centers for Disease Control and Prevention (CDC) listed the Delta variant as the only variant
of concern in the U.S. and accounting for close to 100 percent of COVID-19 cases nationally. CDC reports that the Delta
variant is nearly twice as contagious as the original COVID strain. The U.S. had previously characterized three other
variants (Alpha, Beta, and Gamma) as of concern but downgraded them. According to CDC, the prevalence of these
variants is less than 0.1 percent.
2Data from a sample of state and local health departments showed that, in August 2021, unvaccinated individuals
accounted for the majority of new COVID-19 cases in those states, according to CDC. No vaccine is 100 percent eﬀective.
CDC expects that, as the number of vaccinated individuals increases, so will the number of so-called “breakthrough”
cases. The agency notes, however, that the risks of infection, hospitalization, and death are much lower in vaccinated
individuals compared to unvaccinated individuals. In addition, vaccinated individuals are less likely to experience severe
illness if they do become infected with COVID-19 after vaccination. As of May 1, 2021, CDC shifted from providing data
on all breakthrough cases to providing data on those cases that result in hospitalization or death.
3Data on new daily hospital admissions are pulled from a snapshot of the Department of Health and Human Services
(HHS) Uniﬁed Hospital Timeseries Dataset. See CDC, “COVID Data Tracker: New Admissions of Patients with Conﬁrmed
COVID-19, United States,” accessed September 30, 2021, https://covid.cdc.gov/covid-data-tracker/#new-hospital-
admissions.
4According to CDC, hospitals began consistently reporting admissions data August 1, 2020.

Page 3 GAO-22-105051
Ongoing demand for medical supplies for the COVID-19 response, including testing materials and
personal protective equipment, has resulted in ﬂuctuating shortages. For example, on September
2, 2021, CDC announced a temporary shortage of point-of-care and over-the-counter COVID-19
testing supplies. In addition, the federal government continues to provide personal protective
equipment—N95 respirators, surgical masks, surgical and isolation gowns, and nitrile and other
gloves—to states, with gloves accounting for the largest number of shipments. For example,
during the 7-day period from September 18 to September 24, 2021, the federal government
and its commercial partners shipped close to 700 million units of gloves, over 44 million surgical
masks, over 13 million surgical gowns, and close to 5 million N95 respirators to all 50 states, the
District of Columbia, and Puerto Rico.

To help prevent medical supply shortages for future public health emergencies, the Department of
Health and Human Services (HHS) released its pandemic supply chain resilience strategy, as called
for in Executive Order 14001, in September 2021.5 The strategy outlines the goals and objectives
for a resilient public health supply chain and the “path for implementation” of the strategy.6

Since March 2020, Congress has provided about $4.8 trillion through the CARES Act and other laws
that were enacted to fund eﬀorts to help the nation respond to and recover from the COVID-19
pandemic (COVID-19 relief laws).7

Ongoing implementation of the provisions in the COVID-19 relief laws and the size and scope of
these eﬀorts—from distributing funding to implementing new programs—continue to demand
strong accountability and oversight. Furthermore, the government must remain vigilant and
agile to address the evolving COVID-19 pandemic well into its second year. The current annual
hurricane and ﬂu seasons could place further burdens on the already overtaxed health care,
medical supply, and emergency management sectors.8

The CARES Act includes a provision for us to report regularly on the federal response to the
pandemic. Speciﬁcally, the act requires us to monitor and oversee the federal government’s eﬀorts

5A Sustainable Public Health Supply Chain, Exec. Order No. 14001, § 4, 86 Fed. Reg. 7,219, 7220-21 (Jan. 21, 2021). The
executive order directed the Department of Defense, HHS, and the Department of Homeland Security, among others, to
develop a pandemic supply chain resilience strategy.
6Department of Health and Human Services, National Strategy for a Resilient Public Health Supply Chain (Washington, D.C.:
July 2021). We plan to report on the strategy and its implementation in a future quarterly CARES report.
7For the purposes of our review, we consider COVID-19 relief laws to include the six laws providing comprehensive
relief across federal agencies and programs. These six laws are the American Rescue Plan Act of 2021(ARPA), Pub. L.
No. 117-2, 135 Stat. 4; Consolidated Appropriations Act, 2021, Pub. L. No. 116-260, 134 Stat. 1182 (2020); Paycheck
Protection Program and Health Care Enhancement Act, Pub. L. No. 116-139, 134 Stat. 620 (2020); CARES Act, Pub. L. No.
116-136, 134 Stat. 281 (2020); Families First Coronavirus Response Act, Pub. L. No. 116-127, 134 Stat. 178 (2020); and the
Coronavirus Preparedness and Response Supplemental Appropriations Act, 2020, Pub. L. No. 116-123, 134 Stat. 146.
8To prepare for a hurricane, hospitals in a potentially aﬀected region would typically evacuate critical patients to nearby
hospitals in unaﬀected areas. However, hospitals in most regions in Louisiana were already at or near capacity when
Hurricane Ida made landfall in the state on August 29, 2021. Aﬀected hospitals were unable to evacuate patients
because most hospitals in other regions of Louisiana and surrounding states, such as Alabama and Mississippi, were
also at or near capacity because of COVID-19.

Page 4 GAO-22-105051
to prepare for, respond to, and recover from the COVID-19 pandemic.9 To date, we have issued
seven recurring oversight reports in response to this provision.10

This report examines the federal government’s continued eﬀorts to respond to and recover from
the COVID-19 pandemic. We are making 16 new recommendations to federal agencies in areas
including ﬁscal relief funds for health care providers, worker safety and health, assessing fraud
risks to unemployment insurance programs, and state and local recovery funds.

This report also includes 37 enclosures addressing a range of federal programs and activities
across the government concerning public health and the economy (see app. I). Figure 1 lists these
enclosures by topic area and highlights those with new recommendations.

Figure 1: Report Enclosures by Topic Area

9Pub. L. No. 116-136, § 19010, 134 Stat. at 579–81.
10Our recurring oversight reports are GAO, COVID-19: Continued Attention Needed to Enhance Federal Preparedness,
Response, Service Delivery, and Program Integrity, GAO-21-551 (Washington, D.C.: July 19, 2021); COVID-19: Sustained
Federal Action Is Crucial as Pandemic Enters Its Second Year, GAO-21-387 (Washington, D.C.: Mar. 31, 2021); COVID-19:
Critical Vaccine Distribution, Supply Chain, Program Integrity, and Other Challenges Require Focused Federal Attention,
GAO-21-265 (Washington, D.C.: Jan. 28, 2021); COVID-19: Urgent Actions Needed to Better Ensure an Eﬀective Federal
Response, GAO-21-191 (Washington, D.C.: Nov. 30, 2020); COVID-19: Federal Eﬀorts Could Be Strengthened by Timely and
Concerted Actions, GAO-20-701 (Washington, D.C.: Sept. 21, 2020); COVID-19: Brief Update on Initial Federal Response to the
Pandemic, GAO-20-708 (Washington, D.C.: Aug. 31, 2020); and COVID-19: Opportunities to Improve Federal Response and
Recovery Eﬀorts, GAO-20-625 (Washington, D.C.: June 25, 2020).

Page 5 GAO-22-105051
In addition to the seven recurring oversight reports, we have issued over 100 targeted COVID-19-
related reports, testimonies, and science and technology spotlights in areas such as housing
protections, Medicare and Medicaid program ﬂexibilities, and digital vaccine credentials. We also
have reviews ongoing in other areas. See appendix II for highlights pages from our recently issued
work on COVID-19 and appendix III for a list of our ongoing work related to COVID-19.

Across our body of COVID-19-related reports, we have made 209 recommendations to federal
agencies and have raised four matters for congressional consideration to improve the federal
government’s response eﬀorts. As of September 30, 2021, agencies had addressed 33 of these
recommendations and partially addressed 48.11

See ﬁgure 2 for an overview of the status of our COVID-19-related recommendations by
department. For a complete list of our COVID-related products, see https://www.gao.gov/
coronavirus.

11We consider a recommendation to be addressed when the target agency has completed the implementation of the
recommendation, and we consider a recommendation to be partially addressed when the agency is in the process of
developing an action, has started but not yet completed or has partially implemented an action, or has taken steps
toward implementation.

Page 6 GAO-22-105051
Figure 2: Status of Prior GAO Recommendations from COVID-19-Related Work, by Federal Department or Agency,
as of Sept. 30, 2021

Note: For this ﬁgure, recommendations made to the Internal Revenue Service are counted toward the total of
recommendations made to the Department of the Treasury.

Given the government-wide scope of this report, we undertook a variety of methodologies to
complete our work, including examining a wide range of data sources and conducting interviews
with federal and state oﬃcials and stakeholders, such as those from four antihunger organizations
and organizations that represent landlords and lower-income households. We also examined
federal laws, agency documents, and guidance, among other things. In each enclosure, we include
a summary of the methodology speciﬁc to the work conducted.

We conducted this performance audit from March 2021 to October 2021 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 7 GAO-22-105051
Background

Public Health and Economic Eects
The COVID-19 pandemic continues to have devastating eﬀects on public health and the economy.
As of September 23, 2021, the U.S. had about 43 million reported cases of COVID-19, according to
CDC.12 As of the week ending September 25, 2021, the U.S. had about 699,000 reported deaths
attributed to COVID-19.13 In addition, the country continues to experience high unemployment.
As of September 2021, about 7.7 million individuals were unemployed, compared with nearly 5.8
million at the beginning of 2020.14

The number of newly reported COVID-19 cases began increasing at the end of July 2021, following
a decrease in daily cases since the January 2021 peak. Between September 10 and September
23, 2021, new reported COVID-19 cases averaged about 138,000 per day—close to 60 percent of
the peak that occurred during January 2021.15 See ﬁgure 3 for 7-day case averages. During this
same period, reported new COVID-19 cases per day, on average, increased in 14 jurisdictions, held
steady in 20 jurisdictions, and decreased in 18 jurisdictions.16

12Data on COVID-19 cases in the U.S. are based on aggregate case reporting to CDC and include probable and conﬁrmed
cases as reported by states and jurisdictions. CDC COVID-19 counts are subject to change due to delays or updates in
reported data from states and territories. According to CDC, the actual number of COVID-19 cases is unknown for a
variety of reasons, including that people who have been infected may have not been tested or may have not sought
medical care. See, CDC, “COVID Data Tracker: Trends in Number of COVID-19 Cases and Deaths reported to CDC, by
State/Territory,” accessed September 30, 2021, https://covid.cdc.gov/covid-data-tracker/#trends_totalcases.
13CDC’s National Center for Health Statistics COVID-19 death counts in the U.S. are based on provisional counts
from death certiﬁcate data, which do not distinguish between laboratory-conﬁrmed and probable COVID-19 deaths.
Provisional counts are incomplete due to an average delay of 2 weeks (a range of 1–8 weeks or longer) for death
certiﬁcate processing. See CDC, National Center for Health Statistics, “Provisional Death Counts for Coronavirus Disease
2019 (COVID-19),” accessed October 6, 2021, https://www.cdc.gov/nchs/nvss/vsrr/covid19/index.htm.
14Bureau of Labor Statistics, “Unemployment Level (UNEMPLOY)," retrieved from FRED, Federal Reserve Bank of St.
Louis, accessed October 8, 2021, https://fred.stlouisfed.org/series/UNEMPLOY.
15CDC COVID-19 case counts are subject to change based on any delays or updates in reported data from states
and territories. We compared the relative diﬀerence between the average of new cases between September 10 and
September 23, 2021, and the average of new cases in a 14-day window around the peak in the winter of 2021.
16The 52 states and jurisdictions include all 50 states; Washington, D.C.; and New York, N.Y. COVID-19 case counts for
New York, N.Y., are reported separately from the state of New York. We deﬁned states as holding steady if they had less
than a 1 percent increase or decrease in average daily new cases over the time frame. The average percentage change
in daily new cases was calculated as the average of the daily rates of change of the 7-day moving average between
September 10 and September 23, 2021. CDC, “United States COVID-19 Cases and Deaths by State Over Time,” accessed
on October 4, 2021, https://data.cdc.gov/Case-Surveillance/United-States-COVID-19-Cases-and-Deaths-by-State-o/9mfq-
cb36. These COVID-19 case counts may change as new or updated data are reported by states.

Page 8 GAO-22-105051
Figure 3: Reported COVID-19 Cases per Day in the U.S., Mar. 1, 2020–Sept. 23, 2021

Note: Reported COVID-19 cases include conﬁrmed and probable cases. Beginning April 14, 2020, states could include probable
as well as conﬁrmed COVID-19 cases in their reports to CDC. Previously, counts included only conﬁrmed cases. According to
CDC, the actual number of cases is unknown for a variety of reasons, including that people who have been infected may not
have been tested or may not have sought medical care. See CDC, “COVID Data Tracker: Trends in Number of COVID-19 Cases
and Deaths in the U.S. Reported to CDC, by State/Territory,” accessed September 30, 2021, https://covid.cdc.gov/covid-data-
tracker/#trends_dailytrendscases.

According to data from CDC’s National Center for Health Statistics, the number of deaths in the
U.S. has been higher during the pandemic than the expected number of deaths based on previous
years’ data. For example, from January 1, 2020, through September 4, 2021, about 687,000 more
deaths occurred from COVID-19 and other causes than would be normally expected (see ﬁg. 4).

Figure 4: Higher-Than-Expected Weekly Mortality in the U.S., Jan. 2020 to Sept. 2021

Note: The data shown represent the number of deaths from all causes reported in the U.S.in a given week through September
4, 2021, that exceeded the upper-bound threshold of expected deaths calculated by CDC’s NCHS on the basis of variation in
mortality in prior years. For further details of CDC’s methodology for estimating this upper-bound threshold, see CDC, National
Center for Health Statistics, “Excess Deaths Associated with COVID-19,” accessed October 4, 2021, https://www.cdc.gov/nchs/
nvss/vsrr/covid19/excess_deaths.htm. The number of deaths in recent weeks should be interpreted cautiously, as this ﬁgure
relies on provisional data that are generally less complete.

Page 9 GAO-22-105051
Providing the public with safe and eﬀective vaccines to protect people from getting critically ill
with COVID-19 is crucial to mitigating the public health and economic impacts of the virus and
ending the pandemic. Two COVID-19 vaccines requiring two doses were authorized by the Food
and Drug Administration (FDA) for emergency use in December 2020 and a third vaccine, requiring
one dose, was authorized in February 2021.17 On August 23, 2021, FDA approved Pﬁzer’s biologics
license application for its two-dose vaccine for individuals aged 16 years and older.18

On August 18, 2021, the administration recommended that individuals who received the two-
dose vaccines should get a third “booster” shot 8 months after the second dose, pending FDA
authorization and a recommendation from CDC’s immunization advisory committee.19 On
September 22, 2021, FDA amended the authorization for the Pﬁzer vaccine to allow for a booster
shot to be administered to individuals aged 65 years and older, individuals aged 18 to 64 years
who are at high risk of developing severe illness from COVID-19, and individuals aged 18 to 64
years whose frequent institutional or occupational exposure to COVID-19 puts them at high risk of
serious complications from COVID-19, including severe illness. Boosters for these individuals are to
be administered at least 6 months after completion of the ﬁrst series of shots.20 In mid-October,
FDA’s vaccine advisory panel recommended boosters of the Moderna and Johnson & Johnson
vaccines.21

As of September 23, 2021, almost 390 million doses of COVID-19 vaccine had been administered,
according to CDC. Since the vaccination peak in early April 2021, the number of doses of COVID-19
vaccine administered each day have generally declined. As of September 23, 2021, the number of
daily administered doses was less than one-ﬁfth of those administered in the April peak (see ﬁg.
5).

17Pﬁzer’s two-dose COVID-19 vaccine was authorized for emergency use on December 11, 2020, and Moderna’s
two-dose COVID-19 vaccine on December 18, 2020. Janssen’s (Johnson & Johnson) one-dose COVID-19 vaccine was
authorized for emergency use on February 27, 2021.
18Pﬁzer’s two-dose COVID-19 vaccine continues to be authorized for emergency use in individuals aged 12 to 15.
19FDA had previously authorized, and CDC’s immunization advisory committee recommended, third booster shots of
Pﬁzer’s and Moderna’s vaccines for certain immunocompromised individuals.
20On September 24, 2021, CDC issued recommendations that certain populations—individuals aged 65 years and older,
residents in long-term care settings, and individuals aged 50 to 64 years with underlying medical conditions—should
receive a booster shot of the Pﬁzer vaccine. CDC also noted that individuals aged 18 to 64 years with underlying medical
conditions or at increased risk for COVID-19 exposure and transmission because of their occupational or institutional
setting may receive a booster shot, based on their individual beneﬁts and risks.
21The FDA vaccine advisory panel recommended a half-dose of the Moderna vaccine at least 6 months after completion
of the ﬁrst series of shots for the booster for the same groups of individuals for which it authorized boosters of the
Pﬁzer vaccine. The panel did not restrict its recommendations for the Johnson & Johnson booster to speciﬁc groups
of individuals beyond those who had already received the one-dose vaccine. FDA is not required to implement the
vaccine advisory panel’s recommendations. As of October 18, 2021, FDA had not yet amended the emergency use
authorizations for either the Moderna or the Johnson & Johnson vaccine to allow for boosters.

Page 10 GAO-22-105051
Figure 5: Daily Count of COVID-19 Vaccine Doses Administered in the U.S. and Reported to CDC, Dec. 14, 2020–
Sept. 23, 2021

Notes: The data shown reﬂect COVID-19 vaccine doses administered in the U.S. as reported to CDC by state, territorial, and
local public health agencies and by federal entities since the national vaccine program began on December 14, 2020. The data
include doses administered through all vaccine partners, including jurisdictional partner clinics, retail pharmacies, long-term
care facilities, Federal Emergency Management Agency and Health Resources and Services Administration partner sites, and
federal entity facilities. See CDC, “COVID Data Tracker: COVID-19 Vaccinations in the United States,” accessed on September 30,
2021, https://covid.cdc.gov/covid-data-tracker/#vaccinations.
As of September 30, 2021, one COVID-19 vaccine had been licensed by the Food and Drug Administration for individuals
aged 16 years and older and was authorized for emergency use for individuals aged 12 to 15 years. Two additional COVID-19
vaccines were authorized for emergency use for individuals aged 18 years and older. The approved vaccine and one of the
vaccines authorized for emergency use are two-dose regimens; the other vaccine with emergency authorization requires
one dose. The number of doses administered on a given day may be aﬀected by several factors, such as weekend days,
holidays, weather, and vaccine availability. The most recent days of reporting may be more impacted by reporting delays, and
all reported numbers may change over time as historical data are reported to CDC.

In addition to the impact on public health, the pandemic continues to present economic
challenges, particularly for the labor market, though the economy has improved in recent months.
According to data from the Department of Labor, labor market conditions improved in June,
July, August, and September 2021 but remained worse relative to the prepandemic period. For
example, although initial unemployment insurance claims generally declined through September
2021, initial claims remain high compared to the prepandemic period.

Moreover, in September 2021, the employment-to-population ratio, which measures the share of
the population employed, was 58.7 percent—a slight increase from the previous month. However,
this ratio was 2.4 percentage points lower than in the prepandemic period, indicating that labor
market conditions remain worse than in the prepandemic period (see ﬁg. 6).22 See the Economic
Indicators enclosure in appendix I for more information.

22The employment-to-population ratio represents the number of employed people as a percentage of the civilian
noninstitutional population 16 years and older. The ratio is subject to misclassiﬁcation errors with respect to consistently
identifying workers as employed and absent from work or unemployed on temporary layoﬀ.

Page 11 GAO-22-105051
Figure 6: Employment-to-Population Ratio, Jan. 2019–Sept. 2021

Federal COVID-19 Funding and Spending
As of August 31, 2021, the most recent date for which government-wide information was available
at the time of our analysis, the federal government had obligated a total of $3.9 trillion and
expended $3.4 trillion of the $4.8 trillion in appropriated COVID-19 relief funds as reported
by federal agencies to the Department of the Treasury’s Governmentwide Treasury Account
Symbol Adjusted Trial Balance System.23 Obligations and expenditures relative to the amounts
appropriated through COVID-19 relief laws have varied over time, as new relief laws have
appropriated additional relief funds and as the federal government has obligated and expended
those funds (see ﬁg. 7).

23An appropriation provides legal authority for federal agencies to incur obligations and make payments out of
the U.S. Treasury for speciﬁed purposes. An obligation is a deﬁnite commitment that creates a legal liability of the
U.S. government for the payment of goods and services ordered or received, or a legal duty on the part of the U.S.
government that could mature into a legal liability by virtue of actions on the part of another party that are beyond the
control of the U.S. government. An expenditure is the actual spending of money, or an outlay. Expenditures include
some estimates, such as estimated subsidy costs for direct loans and loan guarantees. Increased spending in Medicaid
and Medicare is not accounted for in the appropriations provided by the COVID-19 relief laws. Federal agencies use the
Governmentwide Treasury Account Symbol Adjusted Trial Balance System to report proprietary ﬁnancial reporting and
budgetary execution information to Treasury.

Page 12 GAO-22-105051
Figure 7: Percentage of COVID-19 Relief Appropriations Obligated and Expended, July 31, 2020–Aug. 31, 2021

Notes: The percentages shown represent the portions of appropriated funds available as of each date shown that had been
obligated and expended. An appropriation provides legal authority for federal agencies to incur obligations and make payments
out of the U.S. Treasury for speciﬁed purposes. Appropriation amounts are based on appropriation warrant information
provided by the Department of the Treasury as of July 31, 2020; September 30, 2020; November 30, 2020; January 31, 2021; May
31, 2021; June 30, 2021; July 31, 2021; and August 31, 2021, for the six COVID-19 relief laws, four of which were enacted before
July 2020. These amounts have increased over time and could increase in the future for programs with indeﬁnite appropriations
(i.e., appropriations that, at the time of enactment, are for an unspeciﬁed amount).
An obligation is a deﬁnite commitment that creates a legal liability of the U.S. government for the payment of goods and
services ordered or received, or a legal duty on the part of the U.S. government that could mature into a legal liability by
virtue of actions on the part of another party that are beyond the control of the U.S. government. An expenditure is the actual
spending of money, or an outlay. Expenditures reﬂected in the percentages shown include some estimates, such as estimated
subsidy costs for direct loans and loan guarantees. Increased spending in Medicaid and Medicare is not accounted for in the
appropriations provided by the COVID-19 relief laws. Under Oﬃce of Management and Budget guidance, federal agencies were
not directed to report COVID-19 related obligations and expenditures until July 2020.

The nine major spending areas shown in table 1 represent $3.9 trillion, or 81 percent, of the total
amounts appropriated. For these nine spending areas, agencies reported obligations totaling $3.3
trillion and expenditures totaling $3.0 trillion as of August 31, 2021. Table 1 provides additional
details on appropriations, obligations, and expenditures of government-wide COVID-19 relief
funds, including the nine major spending areas as of August 31, 2021.

Page 13 GAO-22-105051
Table 1: COVID-19 Relief Appropriations, Obligations, and Expenditures, as of Aug. 31, 2021
Total Total Total
b c c
appropriations obligations expenditures
a
Major spending area ($ in billions) ($ in billions) ($ in billions)

Unemployment Insurance 858.6 660.3 650.2
(Department of Labor)

Economic Impact Payments 855.3 841.6 841.6
(Department of the Treasury)

Business Loan Programs 838.0 829.2 827.6
d
(Small Business Administration)

Public Health and Social Services Emergency Fund 350.1 240.0 172.1
(Department of Health and Human Services)

Coronavirus State and Local Fiscal Recovery Funds 350.0 239.8 239.8
(Department of the Treasury)

Education Stabilization Fund 278.6 257.0 51.7
(Department of Education)

Coronavirus Relief Fund 150.0 149.9 149.9
(Department of the Treasury)

Disaster Relief Fund 97.0 63.8 9.9
e
(Department of Homeland Security)

Supplemental Nutrition Assistance Programs 91.7 66.1 64.6
(Department of Agriculture)

Other areas
f 881.6 532.4 391.9

Total
g 4,750.9 3,880.1 3,399.3

Source: GAO analysis of data from the Department of the Treasury and applicable agencies. | GAO-22-105051

a
Major spending areas shown are based on federal accounts in Treasury’s Governmentwide Treasury Account Symbol Adjusted
Trial Balance System. Each spending area may include multiple programs.
b
COVID-19 relief appropriations shown reﬂect amounts appropriated under the American Rescue Plan Act of 2021 (ARPA), Pub.
L. No. 117-2, 135 Stat. 4; Consolidated Appropriations Act, 2021, Pub. L. No. 116-260, 134 Stat. 1182 (2020); Paycheck Protection
Program and Health Care Enhancement Act, Pub. L. No. 116-139, 134 Stat. 620 (2020); CARES Act, Pub. L. No. 116-136, 134 Stat.
281 (2020); Families First Coronavirus Response Act, Pub. L. No. 116-127, 134 Stat. 178 (2020); and Coronavirus Preparedness
and Response Supplemental Appropriations Act, 2020, Pub. L. No. 116-123, 134 Stat. 146. These amounts are based on
appropriation warrant information provided by Treasury as of August 31, 2021. These amounts have increased over time
and could increase in the future for programs with indeﬁnite appropriations, which are appropriations that, at the time of
enactment, are for an unspeciﬁed amount. The amounts shown do not include transfers of funds that federal agencies may
make between appropriation accounts or transfers of funds they may make to other agencies.
c
Obligation and expenditure data shown are based on data reported by applicable agencies. An obligation is a deﬁnite
commitment that creates a legal liability of the U.S. government for the payment of goods and services ordered or received,
or a legal duty on the part of the U.S. government that could mature into a legal liability by virtue of actions on the part of
another party that are beyond the control of the U.S. government. An expenditure is the actual spending of money, or an
outlay. Expenditures shown include some estimates, such as estimated subsidy costs for direct loans and loan guarantees.

Page 14 GAO-22-105051
d
The Small Business Administration’s Business Loan Program account includes activity for the Paycheck Protection Program
loan guarantees and certain other loan subsidies. These expenditures relate mostly to the loan subsidy costs (i.e., the loan’s
estimated long-term costs to the U.S. government).
e
Appropriations to the Disaster Relief Fund are generally not speciﬁc to individual disasters. Therefore, Treasury’s methodology
for determining COVID-19-related obligations and expenditures does not capture obligations and expenditures for the
COVID-19 response based on appropriations other than those in the COVID-19 relief laws. Further, Treasury’s methodology
includes all obligations and expenditures based on appropriations in the COVID-19 relief laws, including those for other
disasters. In its Disaster Relief Fund Monthly Report dated September 9, 2021, the Department of Homeland Security reported
COVID-19-related obligations totaling $80.0 billion and expenditures totaling $60.6 billion as of August 31, 2021.
f
Several provisions in the Families First Coronavirus Response Act and ARPA authorized increases in Medicaid payments to
states and U.S. territories. The Congressional Budget Oﬃce estimated that federal expenditures from these provisions would be
approximately $76.9 billion through ﬁscal year 2030. The largest increase to federal Medicaid spending is based on a temporary
formula change rather than a speciﬁc appropriated amount. Some of the estimated costs in this total are for the Children’s
Health Insurance Program, permanent changes to Medicaid, and changes not speciﬁcally related to COVID-19. This increased
spending is not accounted for in the appropriations provided by the COVID-19 relief laws and therefore not included in this
table.
g
Because of rounding, amounts shown in columns may not sum to the totals.

The COVID-19 relief laws provided more than $1 trillion to federal agencies to provide assistance
related to the COVID-19 pandemic to states, the District of Columbia, localities, U.S. territories, and
tribes through existing and newly created programs and funds.24 Table 2 lists programs and funds
that each received $10 billion or more—exclusively or primarily for states, the District of Columbia,
localities, U.S. territories, and tribes—in at least one of the six laws. It also provides obligations and
expenditures for these programs and funds as of August 31, 2021.

24This total is based on (1) an analysis of the appropriated amounts in ARPA, Divisions M and N of the Consolidated
Appropriations Act, 2021, the Paycheck Protection Program and Health Care Enhancement Act, the CARES Act, the
Families First Coronavirus Response Act, and the Coronavirus Preparedness and Response Supplemental Appropriations
Act, 2020 that are available to agencies for assistance to states, the District of Columbia, localities, U.S. territories, and
tribes, and (2) the Congressional Budget Oﬃce’s estimated outlays for Medicaid resulting from authorized increases in
payments to states and U.S. territories under those laws.

Page 15 GAO-22-105051
Table 2: Appropriations, Obligations, and Expenditures for Federal Programs and Funds Receiving $10
Billion or More in COVID-19-Related Aid for States, the District of Columbia, Localities, U.S. Territories,
and Tribes, as of Aug. 31, 2021
Appropriations Obligations Expenditures
Program fund/description ($ in billions) ($ in billions) ($ in billions)

Coronavirus State and Local Fiscal Recovery Funds 350 239.8 239.8
Administered by the Department of the Treasury,
these funds provide payments to states, the District
of Columbia (D.C.), U.S. territories, tribal governments,
and localities to mitigate the ﬁscal eﬀects stemming
from the COVID-19 pandemic, among other things.

Elementary and Secondary School Emergency 190.3 172.3 17.3
Relief Fund
Administered by the Department of Education, this
fund generally provides formula grants to states
(including D.C. and Puerto Rico) for education-
related needs to address the impact of the COVID-19
pandemic.

Coronavirus Relief Fund 150 149.9 149.9
Administered by Treasury, this fund provides payments
to states, D.C., localities, U.S. territories, and tribal
governments to help oﬀset costs of their response to
the COVID-19 pandemic.

Disaster Relief Fund 95
a
31.3
b
19.6
b
Administered by the Federal Emergency Management
Agency, this fund provides federal disaster recovery
assistance for state, local, tribal, and territorial
governments when a major disaster occurs.

Medicaid 76.9
c
50.9
d
50.9
d
Administered by states and U.S. territories according
to plans approved by the Centers for Medicare
& Medicaid Services, which oversees Medicaid at
the federal level. This program ﬁnances health
care for certain low-income and medically needy
individuals through federal matching of states’ and U.S.
territories’ health care expenditures. The Families First
Coronavirus Response Act and American Rescue Plan
Act of 2021 temporarily increased federal Medicaid
matching rates under speciﬁed circumstances, among
other changes.

Transit grants 69.5 37.0 22.8
Administered by the Federal Transit Administration,
these funds are distributed through existing grant
programs to provide assistance to states, localities,
U.S. territories, and tribes to prevent, prepare for, and
respond to the COVID-19 pandemic.

Child Care and Development Fund 52.5 52.4 7.0
Administered by the Department of Health and Human
Services (HHS), this program provides funds to states,

Page 16 GAO-22-105051
Appropriations Obligations Expenditures
Program fund/description ($ in billions) ($ in billions) ($ in billions)
D.C., territories, and tribes to subsidize the cost of
child care for low-income families. COVID relief funds
have supported assistance to health care and other
essential workers without regard to income eligibility
requirements. Additional child care stabilization
funding was provided for subgrants to eligible child
care providers to support the stability of the child care
e
sector during and after the COVID-19 pandemic.

Emergency Rental Assistance 46.6 33.2 33.2
f
Administered by Treasury, this program provides
grants to states, D.C., U.S. territories, localities, and
tribes to provide assistance to eligible households for
rent and utility payments.

Public Health and Social Services Emergency Fund 33.4 30.3 7.7
Administered by HHS, this fund provides for
grants to states, U.S. territories, localities, and
tribal governments to support COVID-19 testing,
surveillance, and contact tracing, among other uses.

Airport grants 20 15.8
h
7.7
h
Administered by the Federal Aviation Administration,
these grants provide funds for eligible airports to
prevent, prepare for, and respond to the eﬀects of the
g
COVID-19 pandemic.

Highway infrastructure programs 10 3.9
h
1.5
h
Administered by the Federal Highway Administration,
these programs provide funds to states, D.C., U.S.
territories, and tribes for highway construction and
authorize the use of these funds for maintenance,
personnel, and other purposes to prevent, prepare for,
and respond to the COVID-19 pandemic.

Coronavirus Capital Projects Fund 10 0 0
Administered by Treasury, this fund provides payments
to states, D.C., U.S. territories, and tribal governments
for critical capital projects that directly enable work,
education, and health monitoring in response to the
i
COVID-19 pandemic.

State Small Business Credit Initiative 10 0 0
Administered by Treasury, this program provides funds
to states, D.C., U.S. territories, tribal governments, and
eligible localities to fund small business credit support
j
and investment programs.

Source: GAO analysis of federal laws, data from the Congressional Budget Oﬃce, and obligations and expenditures data from Treasury and applicable agencies. | GAO-22-105051

Notes: The COVID-19 relief laws providing the appropriations shown are the American Rescue Plan Act of 2021 (ARPA), Pub.
L. No. 117-2, 135 Stat. 4 (2021), the Consolidated Appropriations Act, 2021, Pub. L. No. 116-260, div. M and N, 134 Stat. 1182
(2020), the Paycheck Protection Program and Health Care Enhancement Act, Pub. L. No. 116-139, 134 Stat. 620 (2020), the

Page 17 GAO-22-105051
CARES Act, Pub. L. No. 116-136, 134 Stat. 281 (2020), and the Families First Coronavirus Response Act, Pub. L. No. 116-127, 134
Stat. 178 (2020). The Coronavirus Preparedness and Response Supplemental Appropriations Act, 2020 did not provide any
speciﬁed amounts for these programs or funds for states, D.C., localities, territories, or tribes. The amounts shown are the
cumulative amounts for each program or fund under the other ﬁve laws. Some appropriation amounts include an amount
available for administration expenses or for the relevant inspectors general. Numbers are rounded to the nearest hundred
million.
We did not independently verify obligations and expenditures amounts.
a
Appropriations for the Disaster Relief Fund generally are not speciﬁc to individual disasters and may be used for various
disaster assistance programs, including the Public Assistance program, which provides assistance to state, local, territorial, and
tribal governments.
b
The obligations and expenditures listed in the table are for the Public Assistance program for the COVID-19 response.
c
Several provisions in the Families First Coronavirus Response Act and ARPA authorized increases in Medicaid payments to
states and U.S. territories. The Congressional Budget Oﬃce estimated that federal expenditures from these provisions would be
approximately $76.9 billion through ﬁscal year 2030. The largest increase to federal Medicaid spending is based on a temporary
funding formula change rather than a speciﬁc appropriated amount. Some of the estimated costs in this total are for the
Children’s Health Insurance Program, permanent changes to Medicaid, and changes not speciﬁcally related to COVID-19.
d
Medicaid obligations and expenditures are as of June 30, 2021. COVID-19 related obligation and expenditure amounts
for Medicaid only reﬂect provisions in the Families First Coronavirus Response Act. Obligation and expenditure amounts
for COVID-19 related Medicaid provisions in the American Rescue Plan Act are not currently available from the Centers for
Medicare & Medicaid Services.
e
The Child Care and Development Fund is made up of two funding streams: mandatory and matching funding authorized
under section 418 of the Social Security Act, and discretionary funding authorized under the Child Care and Development Block
Grant Act of 1990, as amended. See 42 U.S.C. §§ 618 and 9858m.
f
Expenditures represent funding disbursed to grantees by Treasury for distribution to renters, landlords, and utility providers.
As of August 31, 2021, grantees had spent about $7.7 billion of these amounts. For additional information on grantee spending,
see the enclosure on the Emergency Rental Assistance program in appendix I.
g
Funds are available to eligible sponsors of airports. Nearly all of these airports are under city, state, county, or public-authority
ownership.
h
Obligations and expenditures for these funds are as of August 30, 2021.
i
Treasury issued implementing guidance in September 2021 that provides that the application deadline for requesting
allocations of the Coronavirus Capital Projects Fund from Treasury is (1) December 27, 2021, for states, D.C., and U.S. territories;
and (2) June 1, 2022, for tribal governments.
j
States, the District of Columbia, territories, and tribal governments must initiate applications for the State Small Business Credit
Initiative program with Treasury by December 11, 2021. Eligible jurisdictions must submit completed applications by February
11, 2022.

Page 18 GAO-22-105051
Executive Summary

Overview
As the nation continues to respond to the pandemic and signiﬁcant increases in COVID-19 cases
from the Delta variant, this report provides key updates on the government’s pandemic response
and makes 16 new recommendations aimed at improving the accountability and program
eﬀectiveness of the federal response.

In our prior CARES Act reports and other targeted COVID-19-related reports, we have made a
total of 209 recommendations to federal agencies.25 As of September 30, 2021, agencies had fully
addressed 33 of these recommendations, resulting in improvements including increased oversight
of relief payments to individuals and improved transparency of decision making for emergency
use authorizations for vaccines and therapeutics. Agencies have also partially addressed an
additional 48 recommendations. Fully addressing our previous recommendations as well as the
new recommendations we are making will enhance the transparency and accountability of the
federal government’s response to and recovery from the COVID-19 pandemic.

Relief for Health Care Providers
To respond to the pandemic, $178 billion has been appropriated to the Provider Relief Fund (PRF)
to reimburse eligible providers for health care-related expenses or lost revenues attributable
to COVID-19. As of August 31, 2021, HHS had allocated about $153.9 billion. Of the $153.9
billion allocated, HHS had disbursed about $132.5 billion and about $21.5 billion remained to be
disbursed. Approximately $24.1 billion of PRF funds remained unallocated and undisbursed as
of August 31, 2021. On September 10, 2021, HHS announced that $17 billion of the previously
unallocated $24.1 billion would be allocated for a general distribution to a broad range of
providers who could document COVID-related revenue loss and expenses. HHS expected to begin
disbursing these funds in December 2021.

The Health Resources and Services Administration (HRSA) has taken some oversight actions
regarding post-payment reviews of PRF payments and recovery of identiﬁed overpayments;
however, it has not established key next steps. While the agency has conducted post-payment
reviews for certain priority types of provider payments, it has not established time frames for
implementing and completing all remaining post-payment reviews or set review schedules beyond
the ﬁrst quarter of calendar year 2022. In regards to recovery of identiﬁed overpayments, the
agency has yet to recover most of the overpayments that had been identiﬁed as of September
2021. HRSA oﬃcials stated they had plans for recovering overpayments, but had not ﬁnalized
procedures for doing so.

Without timely post-payment oversight that includes time frames for conducting reviews to help
ensure that relief payments are made only to eligible providers in correct amounts and to identify

25This number includes recommendations from our June 2020, September 2020, November 2020, January 2021, March
2021, and July 2021 CARES Act reports as well as other targeted COVID-19-related reports in areas such as international
humanitarian assistance and Indian Health Services response to COVID-19. For a complete list of our COVID-related
products, see https://www.gao.gov/coronavirus.

Page 19 GAO-22-105051
unused payments or payments not properly used, HHS cannot fully address its stated payment
integrity risks for the PRF and seek to recover overpayments, unused payments, or payments not
properly used. Moreover, setting time frames for completion of these oversight eﬀorts can help
the agency achieve its objectives and increase the likelihood of recovering funds.

We are recommending that the Administrator of the Health Resources and Services
Administration take several steps to ﬁnalize and implement post-payment oversight. Speciﬁcally,
the Administrator should establish time frames for completing post-payment reviews to promptly
address identiﬁed risks and identify overpayments made from the Provider Relief Fund, such
as payments made in incorrect amounts or payments to ineligible providers. The Administrator
should also ﬁnalize procedures and implement post-payment recovery of any Provider Relief Fund
overpayments, unused payments, or payments not properly used. HHS, which includes HRSA,
partially agreed with both recommendations. HRSA stated that it has a schedule for reviewing
the payment types it initially prioritized, and that reviews for the remaining types and payment
recovery eﬀorts will occur in the future. We maintain that establishing time frames for completing
reviews and ﬁnalizing procedures and implementing recovery eﬀorts expeditiously will help the
agency succeed in recovering overpayments.

See the Relief for Health Care Providers enclosure in appendix I for more information.

Coronavirus State and Local Fiscal Recovery Funds
In March 2021, the American Rescue Plan Act of 2021 (ARPA) appropriated $350 billion to Treasury
for the Coronavirus State and Local Fiscal Recovery Funds (CSLFRF).26 The CSLFRF allocates funds
to states, the District of Columbia, localities, tribal governments, and U.S. territories to cover a
broad range of costs stemming from the ﬁscal eﬀects of the COVID-19 pandemic.27 According to
Treasury data, it had distributed approximately $240 billion in CSLFRF funds to recipients as of
August 31, 2021.

As of July 2021, some of the 48 states that responded to a GAO survey reported that they had
somewhat less than or much less than suﬃcient capacity to report on use of CSLFRF allocation
consistent with federal requirements (17 of 48), to disburse the funds (13 of 48), and to apply
appropriate internal controls and respond to inquiries about requirements (10 of 48). In addition,
most states (44 of 48) reported that they had taken or planned to take additional steps—such as
hiring new staﬀ or reassigning existing staﬀ—to help them manage their CSLFRF allocations.

As of August 2021, Treasury was developing its key internal processes and control activities for
the timely monitoring of recipients’ use of their CSLFRF allocations for allowable purposes and for
responding, as appropriate, to internal control and compliance ﬁndings. According to Treasury
oﬃcials, the key internal processes and control activities had not been ﬁnalized or documented.
The oﬃcials noted that program development has occurred within a short time frame since the
enactment of ARPA in March 2021, and that ﬁnalizing and documenting internal processes and

26Pub. L. No. 117-2, § 9901, 135 Stat. at 223. This section of ARPA appropriated $350 billion for two funds—the
Coronavirus State Fiscal Recovery Fund and the Coronavirus Local Fiscal Recovery Fund, which we discuss as one for the
purposes of this report. These funds are codiﬁed, respectively, at 42 U.S.C. §§ 802, 803.
27Pub. L. No. 117-2, § 9901, 135 Stat. at 223 (codiﬁed at 42 U.S.C. §§ 802, 803).

Page 20 GAO-22-105051
control activities for this new program requires time and resources. Further, vacancies in top-
level leadership positions in the Oﬃce of Recovery Programs, which Treasury established in April
2021, have contributed to uncertainty about how the ﬁnal program policies and procedures will be
implemented.

Until Treasury properly designs and documents policies and procedures to guide CSLFRF program
oﬃcials and other responsible oversight parties in the Oﬃce of Recovery Programs, there is a
risk that key control activities needed to help ensure program management fulﬁlls its recipient
monitoring and oversight responsibilities may not be established or applied eﬀectively and
consistently. This risk may be particularly acute with monitoring state and local recipients that face
capacity challenges in managing their CSLFRF allocations in accordance with federal requirements,
as some noted in our survey.

We are recommending that the Secretary of the Treasury design and document timely and
suﬃcient policies and procedures for monitoring CSLFRF recipients to provide assurance that
recipients are managing their allocations in compliance with laws, regulations, agency guidance,
and award terms and conditions, including ensuring that expenditures are made for allowable
purposes. Treasury agreed with the recommendation and stated that it is in the process of
designing, documenting, and implementing a risk-based compliance program to monitor recipient
use of CSLFRF program funds.

See the Coronavirus State and Local Fiscal Recovery Funds enclosure in appendix I for more
information.

Unemployment Insurance Fraud Risk Management
Federal and state entities continue to investigate and report on high levels of fraud, potential
fraud, and fraud risks in the unemployment insurance (UI) programs overseen at the federal level
by the Department of Labor (DOL). For example, in June 2021, DOL’s Oﬃce of Inspector General
reported that it had identiﬁed nearly $8 billion in potentially fraudulent UI beneﬁts paid from
March 2020 through October 2020. In addition, from March 2020 through July 2021, 71 individuals
pleaded guilty to federal charges of defrauding UI programs, and federal charges were pending
against 192 individuals.

In addition to a substantial increase in fraudulent and potentially fraudulent activity in UI
programs, DOL oﬃcials stated that the types of fraud observed during the pandemic diﬀered
from historical UI fraud risks and schemes observed before the pandemic. While DOL continues
to identify and implement strategies to address potential unemployment insurance fraud and has
ongoing program integrity activities to identify risks, it has not comprehensively assessed fraud
risks in alignment with leading practices identiﬁed in our Fraud Risk Framework, which by law
must be incorporated into guidelines established by the Oﬃce of Management and Budget for
agencies.

First, DOL has not clearly assigned deﬁned responsibilities to a dedicated entity for designing and
overseeing fraud risk management activities such as managing the fraud risk assessment process.
Without a dedicated entity with deﬁned responsibilities to lead antifraud initiatives, including the
process of assessing fraud risks to UI programs, DOL may not be strategically managing UI fraud

Page 21 GAO-22-105051
risks. For example, a dedicated antifraud entity could, among other activities, manage the fraud
risk assessment process and coordinate antifraud initiatives across an agency’s various programs
to assure that agency activities called for by the Fraud Risk Framework are conducted.

We are recommending that the Secretary of Labor designate a dedicated entity and document its
responsibilities for managing the process of assessing fraud risks to the unemployment insurance
program, consistent with leading practices as provided in our Fraud Risk Framework. This entity
should have, among other things, clearly deﬁned and documented responsibilities and authority
for managing fraud risk assessments and for facilitating communication among stakeholders
regarding fraud-related issues. DOL neither agreed nor disagreed with this recommendation.
DOL stated that the department’s Chief Financial Oﬃcer and the Employment and Training
Administration’s Assistant Secretary are the designated senior executive oﬃcials responsible for
risk assessment and management of the UI program. While this approach may incorporate the
roles and responsibilities of a dedicated antifraud entity, it is important that, consistent with our
Fraud Risk Framework, DOL clearly document this designation and these senior staﬀ members’
antifraud responsibilities.

Second, DOL has not comprehensively assessed UI fraud risks in alignment with leading practices
or documented a prioritized approach to managing fraud risks. Our Fraud Risk Framework calls
for federal managers to plan regular fraud risk assessments and determine a fraud risk proﬁle.
Speciﬁcally, the fraud risk assessment should be tailored to the program and conducted at regular
intervals as well as when there are changes to the program or operating environment, such as for
program operations and expansions during emergencies.

Without comprehensively assessing UI fraud risks, DOL lacks reasonable assurance that it has
identiﬁed the most signiﬁcant fraud risks for the regular UI program that will exist after the
pandemic. For example, some fraud risks identiﬁed in the CARES Act UI programs may continue
to exist in the regular UI program after the temporary UI programs expire. An analysis of fraud
risks across all UI programs would also help DOL determine whether additional fraud controls are
needed for the regular UI program and could position DOL to deal more eﬀectively with any future
emergency UI programs.

We are also recommending that the Secretary of Labor (1) identify inherent fraud risks facing the
unemployment insurance program; (2) assess the likelihood and impact of inherent fraud risks
facing the program; (3) determine fraud risk tolerance for the program; (4) examine the suitability
of existing fraud controls in the program and prioritize residual fraud risks; and (5) document the
fraud risk proﬁle for the program. DOL neither agreed nor disagreed with this recommendation.
DOL said its current process allows it to identify, evaluate, and manage risks. However, DOL also
said it will incorporate the recommended practices and approaches moving forward.

See the Unemployment Insurance Fraud Risk Management enclosure in appendix I for more
information.

Page 22 GAO-22-105051
FEMA’s Disaster Relief Fund and Assistance to State, Local,
Tribal, and Territorial Governments
The Federal Emergency Management Agency (FEMA) is using the Disaster Relief Fund to respond
to the COVID-19 pandemic, which is the ﬁrst time the fund has been used during a nationwide
public health emergency.28 For example, FEMA’s Public Assistance Program helps state, local,
tribal, and territorial governments, and certain types of private nonproﬁt organizations respond
to and recover from major disasters or emergencies. From September 1, 2020, to August 31, 2021,
FEMA obligated a total of approximately $26.8 billion to Public Assistance projects for emergency
protective measures, such as eligible medical care, the purchase and distribution of food, and
distribution of personal protective equipment.

We found that FEMA inconsistently interpreted and applied its policies for expenses eligible
for COVID-19 Public Assistance within and across its 10 regions. For example, oﬃcials in one
state said that, at one point, FEMA had deemed the provision of personal protective equipment
at correctional facilities as ineligible for reimbursement in their region but that states in other
regions had received reimbursement for the same expense.

We identiﬁed four key areas that contributed to the inconsistent interpretation and application of
COVID-19 policies for Public Assistance based on our discussions with FEMA headquarters oﬃcials
and state emergency managers. These four areas are (1) changes in policy that were interpreted
and applied diﬀerently by FEMA personnel as FEMA used the Public Assistance Program for the
ﬁrst time to respond to a nationwide emergency; (2) delegation of authority to FEMA regions for
making ﬁnal application eligibility determinations; (3) lack of required training on COVID-19 policies
for staﬀ handling Public Assistance applications; and (4) variation in the experience level of staﬀ
making eligibility determinations for applications. FEMA oﬃcials stated that it has been diﬃcult to
ensure consistency in policies as diﬀerent states and regions are not experiencing the same things
at the same time.

FEMA oﬃcials have acknowledged that in spite of eﬀorts to ensure consistency in interpretation
and application of its Public Assistance COVID-19 policy, inconsistent interpretation and
application of its policy continue to occur within and across regions. Given the current rise in the
COVID-19 Delta variant across the nation, FEMA is likely to receive applications for reimbursement
for a larger number of projects than it estimated earlier in 2021.

We are recommending that the Federal Emergency Management Agency Administrator improve
the consistency of the agency’s interpretation and application of the COVID-19 Public Assistance
policy within and across regions by further clarifying and communicating eligibility requirements
nationwide.

28The Disaster Relief Fund receives an annual appropriation and has routinely received supplemental appropriations.
In March 2020, the CARES Act appropriated $45 billion for the Disaster Relief Fund. Pub. L. No. 116-136, div. B, tit. VI, 134
Stat. at 543. The Consolidated Appropriations Act, 2021, appropriated $17 billion to the Disaster Relief Fund for major
disasters and an additional $2 billion to provide assistance for COVID-19-related funeral expenses. Pub. L. No. 116-260,
div. F, tit. III, div. M, tit. II,134 Stat. at 1462, 1910. In March 2021, ARPA appropriated $50 billion to the Disaster Relief
Fund. Pub. L. No. 117-2, § 4005, 135 Stat. at 79.

Page 23 GAO-22-105051
We are also recommending that the Federal Emergency Management Agency Administrator
require the agency’s Public Assistance program employees in the regions and at its Consolidated
Resource Centers to attend training on changes to COVID-19 Public Assistance policy to help
ensure it is interpreted and applied consistently nationwide.

The Department of Homeland Security agreed with both recommendations and outlined actions
it has taken to improve the consistency of its interpretation and application of COVID-19 Public
Assistance policy and to train employees in the regions and at its Consolidated Resource Centers.

See the FEMA’s Disaster Relief Fund and Assistance to State, Local, Tribal, and Territorial
Governments enclosure in appendix I for more information.

Loans for Aviation and Other Eligible Businesses
Treasury has executed 35 loan agreements with certain aviation businesses and other businesses
deemed critical to maintaining national security (national security businesses).29 These loans
have totaled about $22 billion of the $46 billion authorized by the CARES Act for loans and loan
guarantees. Of these 35 loans, as of October 1, 2021, 10 loans had been fully repaid and the total
value of outstanding loans was about $1.1 billion.

As directed by the CARES Act, Treasury required certain loan recipients to provide ﬁnancial assets,
such as warrants—an option to buy shares of stock at a predetermined price before a speciﬁed
date—which give the federal government the ability to protect taxpayer interests. In addition,
the CARES Act provided that for the primary beneﬁt of taxpayers Treasury may sell, exercise, or
surrender ﬁnancial instruments it obtained. Treasury received warrants from nine businesses
equal to 10 percent of the total loan amount drawn. Treasury has not exercised any of the
warrants for stock it holds in these nine businesses.

According to Treasury oﬃcials, it is likely that—if the airline industry continues to recover and
borrowers do not default—the warrants could have higher values than the predetermined price
Treasury would have to pay to act on them. For example, based on the stock price at market close
on October 1, 2021, its warrants from one borrower would be valued at 159 percent above the
initial value at which Treasury received them. However, Treasury has not developed policies and
procedures to guide when to act on the warrants to beneﬁt the taxpayer.

We are recommending that the Secretary of the Treasury develop policies and procedures to
determine when to act on warrants obtained as part of the loan program for aviation and other
eligible businesses to beneﬁt the taxpayers. Treasury agreed with our recommendation and said it
is in the process of creating a policy that will allow it to evaluate when and how to act to dispose of
the warrants obtained as part of the loan program.

See the Loans for Aviation and Other Eligible Business enclosure in appendix I for more
information.

29CARES Act, Pub. L. No. 116-136, § 4003, 134 Stat. at 470 (codiﬁed at 15 U.S.C. § 9042).

Page 24 GAO-22-105051
Payroll Support Assistance to Aviation Businesses
As of September 2021, Treasury had made $59 billion in payments out of $63 billion provided to
the Payroll Support Program to support aviation business.30 These payments, made to air carriers
and aviation contractors, were to be used exclusively for the continuation of wages, salaries, and
beneﬁts.

Similar to Treasury’s loan program for aviation and other businesses described above, the CARES
Act allowed the department to receive ﬁnancial instruments from these businesses to provide
appropriate compensation to the federal government for providing the ﬁnancial assistance, and
Treasury required 14 recipients to provide warrants. These 14 recipients provided a total of 58
million warrants.

As Treasury continues to hold these warrants for stock purchases—and as the airline industry
recovers—these warrants may increase in value. Treasury has not exercised any of the warrants
for stock it holds in the 14 businesses, nor has the agency documented policies and procedures
to guide when to act on the warrants to provide appropriate compensation to the federal
government.

We are recommending that the Secretary of the Treasury develop policies and procedures to
determine when to act on warrants obtained as part of the Payroll Support Program to provide
appropriate compensation to the federal government. Treasury agreed with our recommendation
and said it is in the process of creating a policy that will allow it to evaluate when and how to act to
dispose of the warrants obtained as part of the Payroll Support Program.

See the Payroll Support Assistance to Aviation Businesses enclosure in appendix I for more
information.

COVID-19 Testing
Antigen tests are one of two types of COVID-19 diagnostic and screening tests for which FDA
has issued emergency use authorizations. These “rapid” tests typically have a turnaround time
of about 30 minutes or less for results. Antigen tests can be conducted in doctors’ oﬃces,
pharmacies, and other health care settings, as well as in homes or other non-health care settings;
some antigen tests can be conducted without a prescription. The second type of COVID-19 tests,
molecular tests—which are considered the “gold standard” for diagnostic testing—typically have
a 1–3 day turnaround period, mainly due to the time needed to send a sample to the laboratory,
according to FDA oﬃcials.

30In March 2020, the CARES Act established the Payroll Support Program, which included up to $32 billion in ﬁnancial
assistance for passenger air carriers, cargo air carriers, and aviation contractors. CARES Act, Pub. L. No. 116-136, § 4112,
134 Stat. at 498 (codiﬁed at 15 U.S.C. § 9072). The Consolidated Appropriations Act, 2021 established a Payroll Support
Program Extension in December 2020, and ARPA created a new round of the program in March 2021. These second two
rounds of the program provided up to $16 billion and up to $15 billion, respectively, in ﬁnancial assistance for passenger
air carriers and aviation contractors. Pub. L. No. 116-260, div. N, tit. IV, § 402, 134 Stat. at 2052–61. Pub. L. No. 117-2, §
7301, 135 Stat. at 104–07.

Page 25 GAO-22-105051
The use of antigen testing is increasing. According to HHS data, the number of reported antigen
tests per month increased from about 50,000 in June 2020 to nearly 12 million in August 2021. As
a percentage of total tests reported, antigen tests increased from less than 1 percent in June 2020
to over 20 percent of all tests reported in July and August 2021. In addition, on September 9, 2021,
the administration announced the “Path Out of the Pandemic” plan, which should further increase
the use of antigen tests for COVID-19.

Since June 2020, HHS and its component agencies and testing-related working groups have
worked to encourage and improve the reporting of antigen testing data to local, state, and
federal health oﬃcials. However, HHS oﬃcials told us that limited reporting of antigen test results
prevents HHS from using antigen testing data for COVID-19 surveillance.

HHS is taking additional steps aimed at improving reporting of antigen test data and exploring
additional approaches for eﬀective COVID-19 surveillance. For example, oﬃcials told us that
HHS will continue to work with test manufacturers and make enhancements to data reporting
by building reporting methods into the testing process and emphasizing reporting from speciﬁc
settings, such as schools. HHS is also considering surveillance approaches to supplement or
enhance current surveillance eﬀorts. For example, HHS is exploring wastewater surveillance
approaches, which provide data that can complement and conﬁrm other forms of surveillance
for COVID-19, and an eﬃcient pooled community sample that is particularly useful in areas where
timely COVID-19 clinical testing is underutilized or unavailable, according to HHS oﬃcials.

See the COVID-19 Testing enclosure in appendix I for more information.

Worker Safety and Health
The Occupational Safety and Health Administration (OSHA) is responsible for setting and enforcing
workplace safety and health standards for the private sector in 29 states, the District of Columbia,
and four territories. The other 21 states and Puerto Rico set and enforce their own workplace
safety and health standards for private sector and state and local government employers under
state plans approved by OSHA.

During the ﬁrst 15 months of the pandemic, OSHA primarily relied on existing workplace safety
and health standards and voluntary employer guidance for its enforcement. However, until June
2021, OSHA standards did not contain provisions speciﬁcally targeted at the COVID-19 hazard. As a
result, OSHA inspectors faced challenges in applying OSHA requirements to COVID-19 cases.

OSHA took steps to help protect employees in high-risk industries from the hazard of COVID-19 by
initiating a 1-year COVID-19 National Emphasis Program in March 2021 and issuing an emergency
temporary standard in June 2021.31 Although the emergency temporary standard covers only
employers in the health care industry, in its other policies, OSHA has acknowledged the potential

31In January 2021, President Biden signed an executive order that, among other things, directed OSHA to initiate the
COVID-19 National Emphasis Program and directed the Secretary of Labor to consider whether a COVID-19 emergency
temporary standard was necessary. This standard went into eﬀect on June 21, 2021, with employer compliance with
certain provisions required by July 6, 2021, and others by July 21, 2021. It applies to workplaces where employees
provide health care services or health care support services, with some exclusions.

Page 26 GAO-22-105051
for high risk of workplace COVID-19 exposure in industries beyond health care. The agency
is engaged in rulemaking on two standards: the June 2021 COVID-19 health-care emergency
temporary standard and a separate infectious disease standard.

OSHA area oﬃces faced challenges in enforcing workplace safety and health standards during
the COVID-19 pandemic, but the agency has not assessed lessons learned or promising practices.
According to inspectors from area oﬃces, resource challenges included managing a high volume
of incoming reports and working in a telework environment. Communication and guidance
challenges for inspectors included a lack of timely guidance from OSHA headquarters and
diﬃculty ﬁnding and using the most up-to-date guidance.

We are recommending that the Assistant Secretary of Labor for Occupational Safety and Health
assess—as soon as feasible and, as appropriate, periodically thereafter—various challenges
related to resources and to communication and guidance that the Occupational Safety and Health
Administration has faced in its response to the COVID-19 pandemic and take related actions as
warranted.

DOL partially agreed with our recommendation. DOL stated that it agrees that it is important to
assess lessons learned and best practices for OSHA’s operational response to COVID-19. However,
DOL oﬃcials said they believe that while the pandemic is ongoing, the agency’s resources are best
used to help employers and workers mitigate exposures to COVID-19. Because it is unclear when
the COVID-19 pandemic will end, we maintain that assessing—as soon as feasible—the challenges
that OSHA faced in responding to the pandemic, and taking related actions, would enable the
agency to improve its enforcement eﬀorts during this pandemic and help it prepare for operations
during any future pandemic.

See the Worker Safety and Health enclosure in appendix I for more information.

Advance Child Tax Credit Payments
ARPA made several temporary changes to the child tax credit (CTC). First, it temporarily expanded
eligibility to additional qualiﬁed individuals by eliminating the earned income requirement to
receive the CTC.32 Second, it temporarily increased the maximum amount of the CTC from $2,000
per qualifying child to $3,000 or $3,600, depending on the child’s age.33 As required by ARPA, the
Internal Revenue Service (IRS) and Treasury are responsible for issuing half of the CTC through
periodic advance payments (advance CTC).34 IRS reported that as of September 25, 2021, it had
disbursed more than 106 million advance payments totaling over $45.5 billion.

IRS is conducting several outreach eﬀorts to increase the public’s awareness of advance CTC
payments. For example, IRS continues to coordinate with community organizations to raise
awareness of the advance CTC payments. IRS is also planning to include advance CTC messaging

32Pub. L. No. 117-2, § 9611(a), 135 Stat. at 144–145.
33Pub. L. No. 117-2, § 9611(a), 135 Stat. at 145. In 2017, the maximum CTC amount was temporarily raised from $1,000
to $2,000 for tax years 2018 through 2025. 26 U.S.C. § 24(h)(2).
34Pub. L. No. 117-2, § 9611(b)(1), 135 Stat. at 145–148.

Page 27 GAO-22-105051
for the 2022 tax ﬁling season in its annual Get Ready campaign, which IRS oﬃcials said typically
begins in November.

However, IRS and Treasury have not developed a comprehensive estimate of individuals who
are potentially eligible for advance CTC payments and have not set a participation goal. An
eligibility estimate and participation rate, including individuals who have opted in and out of the
advance CTC payments, would provide greater clarity about which populations may be at risk
of not receiving the payments. These populations would beneﬁt from targeted outreach and
communications to learn more about the payments and how to claim the advance CTC during
the 2022 ﬁling season. Moreover, this information could inform IRS’s administration of other
refundable tax credits as well as any future changes to the CTC that Congress is considering.

We are recommending that the Secretary of the Treasury, in coordination with the Commissioner
of Internal Revenue, estimate the number of individuals, including nonﬁlers, who are eligible for
advance child tax credit payments, measure the 2021 participation rate based on that estimate,
and use that estimate to develop targeted outreach and communications eﬀorts for the 2022
ﬁling season; the participation rate could include individuals who opt in and out of the advance
payments. Treasury neither agreed nor disagreed with this recommendation. Treasury stated that
it supports the goal of the recommendation but has not estimated the eligible population for the
advance CTC. Treasury also stated that it and IRS continue to undertake advance CTC outreach,
education, and media campaign eﬀorts. We maintain that without a comprehensive estimate of
eligibility and a participation rate, which includes more nonﬁlers, Treasury and IRS are missing
an opportunity to assess the eﬀectiveness of their outreach eﬀorts in reaching nonﬁlers who are
more likely experiencing poverty or hardship and may be more in need of the CTC payments.

See the Advance Child Tax Credit and Economic Impact Payments enclosure in appendix I for more
information.

Child Nutrition
Child nutrition programs administered by the Department of Agriculture’s (USDA) Food and
Nutrition Service (FNS) supply cash reimbursements to schools or other programs for meals and
snacks provided to eligible children nationwide. In ﬁscal year 2019, before the pandemic, the
four largest programs—the National School Lunch Program, School Breakfast Program, Summer
Food Service Program, and Child and Adult Care Food Program—along with other child nutrition
programs, received $23.1 billion in federal funds. During a typical year, two of these programs, the
National School Lunch Program and the School Breakfast Program, subsidize meals for nearly 30
million children in approximately 95,000 elementary and secondary schools nationwide.

Various COVID-19 relief laws have provided funding or authority to USDA to support child nutrition
programs during the pandemic. For example, the Families First Coronavirus Response Act granted
FNS authority to issue nationwide waivers in certain programs.35 These waivers are intended to
support access to nutritious meals, reduce the administrative burden associated with eligibility
determinations, and minimize potential exposure to COVID-19.

35Pub. L. No. 116-127, § 2202(a), 134 Stat. at 185.

Page 28 GAO-22-105051
Agencies can leverage lessons learned from an event to inform future eﬀorts and limit the chance
of recurring challenges. According to FNS oﬃcials, FNS is primarily using the existing FNS School
Meals Operations study to gather information about lessons learned during the pandemic for
child nutrition programs. The study, launched in spring 2021, will collect administrative and survey
data on each of the four child nutrition programs from state agencies and will collect survey data
from school district nutrition programs. However, as of July 2021, FNS was unable to provide
us with a plan showing how FNS intends to comprehensively analyze lessons learned from the
pandemic for child nutrition programs, such as from operational and ﬁnancial challenges.

Although FNS is collecting some information on these topics from states and school districts,
without documenting its plan for analyzing lessons learned from the pandemic, FNS may miss
opportunities to comprehensively identify lessons learned. Further, according to FNS oﬃcials,
while the School Meals Operations study will survey state agencies that administer the federal
child nutrition programs, the study will not gather local perspectives directly from child care
centers and day care homes or from other local program sponsors that are not school districts.
Without gathering perspectives from a full range of meal program operators—rather than only
from state agencies and school districts––FNS will lack comprehensive information to aid its future
planning.

We are recommending that the Secretary of Agriculture document the department’s plan to
analyze lessons learned from operating child nutrition programs during the COVID-19 pandemic.
This plan should include a description of how the department will gather perspectives of key
stakeholders, such as Child and Adult Care Food Program institutions and nonschool Summer
Food Service Program sponsors. The Department of Agriculture generally agreed with this
recommendation.

See the Child Nutrition enclosure in appendix I for more information.

Page 29 GAO-22-105051
Conclusions

The federal government’s eﬀorts to respond to and recover from COVID-19 continue. The
spread of the Delta variant in the U.S. this summer—and the subsequent rise in cases and
hospitalizations—illustrates the challenges to the nation’s response and recovery eﬀorts and the
work that remains. We are pleased that agencies have fully addressed 33 and partially addressed
48 of our 209 recommendations. Fully implementing our recommendations, including the new
recommendations we are making in this report, can help improve the federal response and
recovery eﬀorts.

Page 30 GAO-22-105051
Closing

We are sending copies of this report to the appropriate congressional committees, the Oﬃce of
Management and Budget, and other relevant agencies. In addition, the report is available at no
charge on the GAO website at https://www.gao.gov.

If you or your staﬀ have any questions about this report, please contact me at (202) 512-5500
or dodarog@gao.gov. Questions can also be directed to Orice Williams Brown, Chief Operating
Oﬃcer, at (202) 512-5600; Jessica Farb, Managing Director, Health Care, at (202) 512-7114 or
farbj@gao.gov; or A. Nicole Clowers, Managing Director, Congressional Relations, at (202) 512-4400
or clowersa@gao.gov. Contact points for our Oﬃces of Congressional Relations and Public Aﬀairs
may be found on the last page of this report.

Gene L. Dodaro

Comptroller General of the United States

Page 31 GAO-22-105051
Congressional Addressees
The Honorable Patrick Leahy
Chairman
The Honorable Richard Shelby
Vice Chairman
Committee on Appropriations
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

The Honorable Frank Pallone, Jr.
Chairman
The Honorable Cathy McMorris Rodgers
Republican Leader
Committee on Energy and Commerce
House of Representatives

The Honorable Bennie G. Thompson
Chairman
The Honorable John Katko
Ranking Member
Committee on Homeland Security

Page 32 GAO-22-105051
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
Chairman
The Honorable Kevin Brady
Republican Leader
Committee on Ways and Means
House of Representatives

Page 33 GAO-22-105051
Appendixes and Enclosures
Appendix I: Enclosures

Economic Indicators
Based on data available in early October 2021, the national economy has continued to
recover from the COVID-19 pandemic, and areas of the economy we are monitoring saw some
improvement in recent months. Indicators for labor markets, household ﬁnances, and small
business credit conditions improved in June and July 2021, with notable gains in leisure and
hospitality and state and local government employment, while the labor market recovery slowed
in August 2021 and state and local government employment fell in September 2021 (see table).36

36In previous work, we identiﬁed a number of economic indicators to facilitate ongoing and consistent monitoring of
areas of the economy supported by the federal pandemic response, namely labor markets, household ﬁnances, small
business credit and ﬁnancial conditions, corporate credit market conditions, and state and local government ﬁnances. To
the extent that federal pandemic responses are eﬀective, we would expect to see improvements in outcomes related to
these indicators. However, while trends in these indicators may be suggestive of the eﬀect of provisions of the COVID-19
relief laws over time, those trends will not on their own provide deﬁnitive evidence of eﬀectiveness. Beginning with
this report, we have removed economic indicators related to corporate and municipal bond markets because of the
signiﬁcant and consistent recovery in these markets. Going forward, we intend to focus our reporting on labor markets,
household ﬁnances, and small business credit and ﬁnancial conditions.

Page 34 GAO-22-105051
Indicators for Areas of the Economy Supported by the Federal COVID-19 Pandemic Response, June–Sept. 2021,
Cumulative Changes since Feb. 2020

a
The employment-to-population ratio represents the number of employed people as a percentage of the civilian
noninstitutional population 16 years and over. The ratio is subject to a misclassiﬁcation error with respect to identifying workers
as employed and absent from work who are likely unemployed on temporary layoﬀ.
b
State and local government and leisure and hospitality employment data from August and September 2021 are preliminary.
c
Higher levels in the Consumer Credit Default Composite Index rate indicate more defaults on consumer loans, including auto
loans, bank cards, and mortgages. The Consumer Credit Default Composite Index could be subject to seasonal variation but is
not seasonally adjusted.
d
Seriously delinquent loans are 3 months or more past due or in foreclosure, based on mortgages insured by the Federal
Housing Administration (FHA). Increases in serious delinquency rates on FHA loans could, to some extent, reﬂect borrowers
taking advantage of mortgage forbearance provisions of the CARES Act, but may also indicate ﬁnancial challenges facing the
minority and low- to moderate-income households that disproportionately take out mortgages insured by FHA.
e
Lower levels of the small business credit card delinquency index indicate more delayed payments on credit. The small business
credit card delinquency index is published under license and with permission from Dun & Bradstreet, and no commercial use
can be made of these data.

Gross domestic product (GDP) grew at a 6.7 percent annual rate in the second quarter of 2021,
and for the ﬁrst time now exceeds its prepandemic level from the fourth quarter of 2019. The
recovery from the pandemic has also been associated with a notable increase in inﬂation which,
should it persist, could cause ﬁnancial challenges that would be felt most acutely by low-income
households. The strength of the economic recovery will continue to depend on the success of
public health measures against the COVID-19 pandemic.

Key trends in economic indicators. Federal debt held by the public rose to $22.3 trillion in
September 2021 from $22.0 trillion in March 2021, after falling slightly as a share of GDP, from 99.8

Page 35 GAO-22-105051
percent in the ﬁrst quarter of 2021 to 98.3 percent of GDP in the second quarter 2021. Interest
rates on 3-month Treasury securities were relatively stable, rising to 0.04 percent in September
2021 from 0.02 percent in May 2021. Interest rates on 10-year Treasury securities, in contrast, fell
from 1.62 percent to 1.37 percent over the same period. The long-term ﬁscal challenges facing the
U.S. have been exacerbated by the pandemic and will require attention as the economy continues
to recover and public health goals are attained, as we reported in March 2021.

Based on monthly and weekly data from the Department of Labor, the labor market showed
improvement in June, July, August, and September 2021 but remained worse relative to the
prepandemic period. Although weekly initial unemployment insurance claims generally declined
through September 2021, initial claims remain high compared to the prepandemic period (see
the Unemployment Insurance Programs enclosure in app. I). The employment-to-population ratio
in September 2021 was 58.7 percent, which was 0.2 percentage points higher than the previous
month but 2.4 percentage points lower than the prepandemic period (see ﬁgure).

Employment-to-Population Ratio, Jan. 2019–Sept. 2021

Changes in employment across sectors continue to reﬂect the diﬀerential impact of the pandemic
on various sectors of the economy. For example, some industries that experienced strong gains
in employment in the ﬁrst half of 2021, including leisure and hospitality, experienced slower job
growth in August and September 2021 as the Delta variant drove a resurgence in cases in the U.S.
Employment in the leisure and hospitality sector is still 9.4 percent lower than it was in February
2020. State and local government employment decreased in September 2021 following months
of increases, and employment in these sectors remains 4.3 percent lower than the prepandemic
period.

Serious delinquency rates—loans that are 90 or more days past due or in foreclosure—for single-
family mortgage loans insured by the Federal Housing Administration (FHA) decreased from
March through August 2021, to 8.64 percent of loans, but still remained much higher than rates
prior to the pandemic (see ﬁgure). FHA loans disproportionately serve minority and low- to

Page 36 GAO-22-105051
moderate-income borrowers, and therefore falling delinquencies may indicate some improvement
in the ﬁnances of those households in recent months, as well as fewer borrowers relying on
mortgage forbearance provisions of the CARES Act.37 Trends in rent payments among low-
income households suggest that these households remain under ﬁnancial stress as well (see the
Emergency Rental Assistance enclosure in app. I).

Serious Delinquency Rates on Single-Family Residential Mortgages, Jan. 2019–Aug. 2021

Note: Seriously delinquent single-family loans are 3 months or more past due or in the foreclosure process. We excluded
February 2021 data from the ﬁgure because the delinquency rates for February 2021 are likely understated due to late
reporting by a large servicer, according to FHA.

Key trends in inﬂation. Inﬂation has increased notably in recent months, while measures of
underlying inﬂation pressure and longer-term inﬂation expectations have been more stable.

Inﬂation is the increase in the price of goods and services over time, and is typically measured as
the percentage change in those prices over a set period, often 1 year.38 As the prices of goods and
services rise, inﬂation decreases the purchasing power of consumers. That is, inﬂation decreases
the value of currency or other highly liquid assets, like checking accounts; as the prices of goods

37In ﬁscal year 2020, 34.2 percent of all FHA purchase and reﬁnance borrowers were minorities, 50.4 percent of
FHA forward mortgage borrowers were of low- to moderate-income, and 83.1 percent of home purchasers under
the FHA forward mortgage insurance program were ﬁrst-time homebuyers. See Department of Housing and Urban
Development, FHA Annual Management Report Fiscal Year 2020. The CARES Act provided temporary protections for
millions of households against foreclosure and eviction, as well as temporary forbearance, suspending mortgage
payments for up to 360 days. In addition, FHA allows mortgage servicers to initiate new forbearance through September
30, 2021, and it allows borrowers who requested an initial forbearance on or before June 30, 2020, to request up to 6
months of forbearance extensions. Moreover, on July 23, 2021, FHA introduced additional COVID-19 recovery options
to help borrowers transitioning out of forbearance to permanent sustainable payments. For example, FHA will require
mortgage servicers to oﬀer a no-cost option to eligible homeowners and enhance servicers’ ability to provide all eligible
borrowers that cannot resume their monthly mortgage with a 25 percent monthly principal and interest reduction.
38For example, an inﬂation rate of 2 percent would mean that the prices of goods and services, on average, increased 2
percent over the last year.

Page 37 GAO-22-105051
and services rise, each dollar will buy less. Some level of inﬂation on average can help promote
stable economic conditions, but persistently high levels of inﬂation can cause ﬁnancial challenges
that are experienced more acutely by certain households. The Federal Reserve System’s Federal
Open Market Committee (FOMC) aims for inﬂation of 2 percent on average over time, and aims to
achieve rates of inﬂation that are above 2 percent for some time after periods in which inﬂation is
persistently below 2 percent.39

Higher levels of inﬂation over short periods—described as transitory—are not unusual and are
less cause for concern. The prices of goods and services regularly shift in response to economic
changes, and any impact on household ﬁnances is more limited because prices increase more
rapidly for only a short period of time. In addition, transitory inﬂation during an economic
expansion can also be associated with improving labor market opportunities, including for
some low-income and minority workers who could ﬁnd more stable employment during a long
economic expansion.

In contrast, high levels of inﬂation that persist for long periods are more cause for concern, and
can reduce the pace of economic growth.40 Higher inﬂation that persists for a longer period can
also inﬂuence consumers’ and businesses’ expectation of future inﬂation, which can inﬂuence
their current behavior. For example, consumers and businesses may make large purchases
sooner, increasing current demand and making it more likely that those expectations of future
inﬂation are realized.41

We identiﬁed a number of indicators of inﬂation to facilitate ongoing and consistent monitoring
of the inﬂation experience of consumers to help assess the extent to which higher inﬂation may
be transitory or persistent. We included both the Personal Consumption Expenditures (PCE)
price index and the Consumer Price Index (CPI) as broad measures of the prices consumers pay
for goods and services.42 We also included the median and trimmed mean CPIs, which are less
volatile because they omit both small and large price changes and may provide clearer signals of
underlying inﬂation.43

39See the FOMC’s 2020 Statement on Longer-Run Goals and Monetary Policy Strategy.
40High inﬂation could, for example, reduce the eﬃciency of ﬁnancial services, thereby reducing investment and
economic growth. For related empirical work, see Tolga Omay and Elif Öznur Kan, “Re-examining the Threshold Eﬀects
in the Inﬂation–Growth Nexus with Cross-Sectionally Dependent Non-linear Panel: Evidence from Six Industrialized
Economies,” Economic Modelling, vol. 27 (2010): pp. 996–1005, or S. Kremer, A. Bick, and D. Nautz, “Inﬂation and Growth:
New Evidence from a Dynamic Panel Threshold Analysis,” Empirical Economics, vol. 44 (2013): pp. 861–878.
41Regarding consumers, see Lena Dräger and Giang Nghiem, “Are Consumers' Spending Decisions in Line With an
Euler Equation?” The Review of Economics and Statistics, vol. 103, no. 3 (2021): pp. 1–17. Regarding ﬁrms, see Olivier
Coibion, Yuriy Gorodnichenko, and Tiziano Ropele, "Inﬂation Expectations and Firm Decisions: New Causal Evidence," The
Quarterly Journal of Economics (2020): pp. 165–219.
42The major diﬀerence between the two indexes is that they measure diﬀerent baskets of goods and services. The CPI
includes out-of-pocket expenditures on goods and services purchased but excludes other expenditures that are not paid
for directly, like medical care paid for by employer-provided insurance, Medicare, and Medicaid. However, these indirect
expenditures are included in the PCE. The Federal Reserve System’s FOMC states its longer-run inﬂation goal in terms of
PCE inﬂation.
43The median CPI is the 1-month inﬂation rate of the component whose expenditure weight is in the 50th percentile
of price changes. The trimmed mean CPI is a weighted average of 1-month inﬂation rates of components whose
expenditure weights fall below the 92nd percentile and above the 8th percentile of price changes.

Page 38 GAO-22-105051
We include two diﬀerent time periods for each of these inﬂation measures: the percentage change
over the last 12 months and the percentage change from the previous month. Measuring inﬂation
over the last 12 months provides a longer and potentially less volatile perspective on inﬂation,
while the percentage change from the previous month is more useful for assessing whether recent
price pressures are waning or intensifying.44

Finally, we included two measures of expected future inﬂation, as inﬂation expectations can
inﬂuence current economic behavior and indicate whether recent inﬂation experiences are
changing views about future inﬂation (see table).

Indicators of Inﬂation, Feb.–Aug. 2021, and Average Inﬂation Rates, 2000–2019

Note: Underlined, red text indicates a higher rate of inﬂation than the previous month while black text indicates a lower rate
of inﬂation than the previous month but with prices still rising overall. Deﬂation, or falling prices, would be indicated with a
negative sign.
a
PCE is based on the PCE price index, which reﬂects changes in the prices of goods and services purchased by or on behalf of
consumers in the U.S. The Federal Open Market Committee states its longer-run inﬂation goal in terms of PCE inﬂation and

44Twelve month measures are subject to “base eﬀects,” where the level of prices during the ﬁrst month has a signiﬁcant
inﬂuence on measured inﬂation over the year that might obscure more recent changes in trends.

Page 39 GAO-22-105051
typically aims for inﬂation of 2 percent on average over time, including by aiming to achieve inﬂation rates above 2 percent for
some time after periods in which inﬂation is persistently below 2 percent.
b
CPI is based on data from the Consumer Price Index for all urban consumers (CPI-U).
c
Median CPI is based on the 1-month inﬂation rate of the component whose expenditure weight is in the 50th percentile of
price changes. By omitting outliers (small and large price changes) and focusing on the interior of the distribution of price
changes, the median CPI may provide a better signal of the underlying inﬂation trend than the all-items CPI.
d
The 16 percent trimmed-mean CPI is based on a weighted average of 1-month inﬂation rates of components whose
expenditure weights fall below the 92nd percentile and above the 8th percentile of price changes. By omitting outliers (small
and large price changes) and focusing on the interior of the distribution of price changes, the 16 percent trimmed-mean CPI
may provide a better signal of the underlying inﬂation trend than the all-items CPI.
e
The 10-year expected inﬂation rate comes from a model that decomposes the TIPS to nominal Treasury spread into three
components: inﬂation expectations, the inﬂation risk premium, and a third component that may capture the TIPS liquidity
premium or other factors that inﬂuence the relative demand for TIPS. See S. D'Amico, D. H. Kim, and M. Wei, "Tips from TIPS:
The Informational Content of Treasury Inﬂation-Protected Security Prices," Journal of Financial and Quantitative Analysis, vol.
53, no. 1 (2018): pp. 395–436.
f
Professional forecast of inﬂation is an average forecast of expected CPI inﬂation in 2022–2023 (annualized) from forecasts
collected by Bloomberg. Absent data for January 2021, we have used black text for February 2021 data based on the data for
10-year expected CPI inﬂation from TIPS.

Based on data available in early October 2021 covering price trends from February through
August 2021, inﬂation has generally increased over the past several months. Indicators of inﬂation
increased substantially relative to a year ago. While inﬂation remains higher than averages in
recent decades, indicators of more recent price pressures (measured relative to the previous
month) have recently been moderating somewhat (see ﬁgure). The median and trimmed mean
CPI measures, as well as professional forecasts and investor expectations of future inﬂation,
are generally below the broader inﬂation measures, although there has been some meaningful
increase in expectations covering the next two years. As we note above, the FOMC aims to achieve
rates of inﬂation that are above 2 percent for some time after periods in which inﬂation has been
persistently below 2 percent, as it was leading up to and during the early months of the pandemic.

Percentage Change in Inﬂation Indicators over the Previous Month, Jan. 2019-Aug. 2021

Page 40 GAO-22-105051
The eﬀects of inﬂation on diﬀerent populations. Diﬀerent populations can experience
higher or lower levels of inﬂation than the national average (e.g., the Consumer Price Index for
All Urban Consumers, known as CPI-U, which is the broadest and most comprehensive CPI).
A number of studies examining distributional consequences of inﬂation have found that low-
income households and the elderly in particular have historically experienced higher levels
of inﬂation than other groups.45 Regarding low-income households, these studies found that
inﬂation experienced in recent years by the lowest income quartile was notably higher than
inﬂation experienced by the highest income quartile.46 This trend has also been evident during
recessions. For example, one study found that during the ﬁrst several months of the COVID-19
pandemic, although inﬂation was low as demand for many goods and services fell, those in the
lowest income quintile experienced higher inﬂation than others.47 One potential explanation for
these diﬀerences is that low-income populations spend a higher share of their income on speciﬁc
goods and services with prices that have increased at a faster rate. For example, during COVID-19
the price of food purchased for consumption at home increased faster than other purchases, and
those with the lowest incomes typically spent a higher share of their income in this category.

The elderly have also tended to experience higher rates of inﬂation than the national average.
The Bureau of Labor Statistics produces an experimental inﬂation index based on a set of goods
and services that reﬂects the spending patterns of elderly consumers (individuals 62 years or
older), the CPI-E. Since 2011, annual growth in the CPI-E has been roughly 0.1 percentage point
higher than the CPI-U. The rate of increase in the CPI-E has generally been higher than the CPI-U,
primarily because the elderly generally spend a higher share of their income on health care and
shelter, and, historically, health care and shelter prices have risen at a faster rate than most other
goods and services.

How households experience a given level of inﬂation depends in large part on their sources of
income and their assets.

• Labor income. As inﬂation rises, some workers are able to bargain for higher wages, and as a
result, the eﬀects of inﬂation on their purchasing power could be—at least partially—oﬀset
by higher wages. However, wages may fail to keep up with inﬂation, and if they do, it may take
years for wages to adjust.48

45For example, see Xavier Jaravel, "Inﬂation Inequality: Measurement, Causes, and Policy Implications," Annual Review
of Economics, vol. 13 (2021), and Joshua Klick and Anya Stockburger, “Experimental CPI for Lower and Higher Income
Households” (Bureau of Labor Statistics working paper 537, 2021).
46For example, see Xavier Jaravel, "The Unequal Gains from Product Innovations: Evidence from the U.S. Retail Sector,"
The Quarterly Journal of Economics (2019): pp. 715–783, and David Argente and Munseob Lee, "Cost of Living Inequality
during the Great Recession," Journal of the European Economic Association, vol. 19, no. 2 (2021): 913–952.
47Alberto Cavallo, “Inﬂation with Covid Consumption Baskets” (NBER working paper 27352, 2020).
48A Congressional Research Service report found that over the period from 1979 to 2019, real wages rose for
higher-wage workers but rose at lower rates or fell for middle- and lower-wage workers. A number of factors could
explain these wage trends, including the strength and structure of labor unions and employment practices that
aﬀect workers’ ability to bargain over compensation. For example, a movement toward greater use of contractors
and subcontractors in some industries may have reduced the bargaining power of lower-paid workers in service
occupations and put downward pressure on their wages. See Congressional Research Service, Real Wage Trends,
1979 to 2019, R45090 (Washington, D.C.: Dec. 28, 2020). In addition, minimum-wage earners may see declines in
real wages to the extent that the minimum wage is not increased or increases do not keep pace with inﬂation. The

Page 41 GAO-22-105051
• Asset holdings. During periods of high inﬂation, the value of currency depreciates more rapidly,
while other sources of wealth, such as investments in the stock market or real estate, will
typically increase in value to reﬂect higher rates of inﬂation, and therefore would mitigate or
hedge against the impact of inﬂation on household savings. Households with a higher share
of their savings in currency or in low-interest savings vehicles would have fewer opportunities
to hedge against inﬂation. Because low-income households typically have a larger share of
their ﬁnancial assets in cash or transaction accounts, higher levels of inﬂation would be more
likely to negatively aﬀect these households. At higher levels of inﬂation, individuals may reduce
holdings of convenient, liquid assets to better hedge against inﬂation.

• Government beneﬁts and indexation. The impact of government transfers and large social
programs on the distributional eﬀects of inﬂation varies depending on the extent to which
programs are adjusted (or indexed) for inﬂation. Public beneﬁts typically make up a larger
share of household income for low-income households and retirees who rely predominantly
on Social Security retirement beneﬁts or other government pensions. Therefore, if public
beneﬁts are indexed to inﬂation, low-income households and some retirees will be shielded
from some of the eﬀects of inﬂation. In the U.S., most large government transfer programs are
adjusted for inﬂation and, as such, families who receive a considerable share of their income
from public beneﬁts are likely to be shielded somewhat from the eﬀects of inﬂation. However,
adjustments to public beneﬁts are typically made annually, so during times of high inﬂation,
low-income households and other program beneﬁciaries would continue to feel some eﬀects
of inﬂation in between periodic adjustments.

Methodology
To identify indicators for monitoring areas of the economy supported by the federal response to
the COVID-19 pandemic, in particular by the six COVID-19 relief laws, we reviewed a number of
sources. Speciﬁcally, we used prior GAO work, data from federal statistical agencies, information
from the Board of Governors of the Federal Reserve System (Federal Reserve) and relevant federal
agencies responsible for the pandemic response and oversight of the health care system, data
available on the Bloomberg Terminal, and input from internal GAO experts. We reviewed the most
recent data from these sources as of August and September 2021, depending on availability.

To identify indicators for monitoring inﬂation, we reviewed data from federal statistical agencies,
academic and other research literature, information from the Federal Reserve, the Federal Open
Market Committee, written responses to our questions provided by the Bureau of Labor Statistics
and the Federal Reserve, data available on the Bloomberg Terminal, and input from internal
GAO experts. We also reviewed selected academic research and government reports to better
understand the distributional eﬀects of inﬂation and how higher levels of inﬂation could aﬀect
certain households in diﬀerent ways.

We assessed the reliability of the economic indicators we used through a number of steps,
including reviewing relevant documentation, reviewing prior GAO work, and interviewing data
providers. Collectively, we determined the indicators were suﬃciently reliable to provide a general

federal minimum wage, for example, was not increased from 2009 through 2021, falling in real value for more than
a decade.

Page 42 GAO-22-105051
sense of (1) how the areas of the economy supported by the federal pandemic response were
performing and (2) trends in the inﬂation experience of consumers.

Agency Comments
We provided the Department of Housing and Urban Development (HUD), the Department of
Labor (DOL), the Department of the Treasury (Treasury), the Federal Reserve, and the Oﬃce of
Management and Budget (OMB) with a draft of this enclosure. Treasury and the Federal Reserve
provided technical comments, which we incorporated as appropriate. HUD, DOL, and OMB did not
provide comments on this enclosure.

GAO’s Ongoing Work
We plan to monitor and report on changes in economic indicators, including developments in
inﬂation, in future quarterly reports.

Related GAO Product
The Nation’s Fiscal Health: After Pandemic Recovery, Focus Needed on Achieving Long-Term Fiscal
Sustainability. GAO-21-275SP. Washington, D.C.: March 23, 2021.

Contact information: Lawrance L. Evans, Jr., (202) 512-8678, evansl@gao.gov

Page 43 GAO-22-105051
COVID-19 Testing
Antigen testing, also known as “rapid testing,” is on the rise. Limited reporting of test results has
thus far prevented the Department of Health and Human Services from using antigen testing data
to monitor COVID-19. The Department is taking further steps aimed at improving reporting and
exploring additional approaches for eﬀective COVID-19 surveillance.

Entities involved: The Department of Health and Human Services, including the Assistant
Secretary for Preparedness and Response, the Centers for Disease Control and Prevention,
the Food and Drug Administration, and COVID-19 testing-related working groups (Testing and
Diagnostics Working Group and Data Strategy and Execution Workgroup) under the Department.

Background
Following a downward trajectory in reported COVID-19 cases nationally from about mid-April to
mid-June, the Delta variant is driving a resurgence in cases in the U.S. This resurgence highlights
the continued importance of diagnostic and screening testing, including antigen testing, in the
COVID-19 response.49

COVID-19 testing types. Antigen tests are one of two types of COVID-19 diagnostic and
screening tests for which the Food and Drug Administration (FDA) has issued emergency use
authorizations:50

• Molecular tests are considered the “gold standard” for detecting an active SARS-CoV-2 infection,
the coronavirus that causes COVID-19, but may require specialized laboratory equipment and
often have a 1-3 day test turnaround time, mainly due to the time needed to send a sample to
the laboratory, according to FDA oﬃcials.

• Antigen tests (sometimes known as “rapid tests”) have a faster turnaround time than molecular
tests in most cases—about 30 minutes or less. Most antigen tests may be conducted at point-
of-care or at-home settings.51 Some point-of-care settings are health-care related—such
as doctors’ oﬃces, pharmacies, and nursing homes, while others are not related to health
care, such as workplaces. There are two types of tests that are authorized for use at home:
those requiring a prescription and those not requiring a prescription, also known as over-the-
counter tests.

49See Centers for Disease Control and Prevention (CDC) guidance, Testing Strategies for SARS-CoV-2 (Atlanta, Ga.: updated
August 13, 2021). According to the guidance, diagnostic testing is intended to identify current infection in anyone with
symptoms consistent with COVID-19, or is to be used for testing vaccinated and unvaccinated people who were exposed
to someone with a conﬁrmed or suspected case of COVID-19. Screening testing is intended to identify infection in
unvaccinated persons who are asymptomatic and have no known or suspected exposure. Some, but not all, antigen and
molecular tests can be used for screening.
50Generally, medical devices must be cleared or approved by the FDA to be marketed in the U.S.; however, the Secretary
of Health and Human Services may declare that circumstances, prescribed by statute, exist justifying the authorization of
temporary emergency use of unapproved medical products, including devices. See 21 U.S.C. § 360bbb-3.
51Some molecular tests may be conducted at point-of-care settings. According to FDA oﬃcials, these tests have
turnaround times closer to that of point-of-care antigen tests because they do not have to be shipped to a
laboratory for analysis.

Page 44 GAO-22-105051
According to FDA, antigen tests generally perform better in detecting active infections when
there are high concentrations of virus found in upper respiratory specimens. Further, the
probability of detecting an active infection is enhanced when testing is repeated more than
once over a few days (referred to as “serial testing”). However, antigen tests generally have a
lower sensitivity—or ability to correctly identify people with COVID-19—than molecular tests.
As a result, the Centers for Disease Control and Prevention (CDC) recommends conﬁrming
some negative antigen tests for people with symptoms with molecular tests.

In general, antigen testing for COVID-19 provides two public health beneﬁts:

1. It helps reduce disease spread because antigen tests typically have a faster turnaround time
and therefore individuals can more quickly identify whether they are infected and then self-
isolate and take other precautions, accordingly; and

2. It can help inform public health disease surveillance and response eﬀorts to the extent that
test results are reported to public health authorities.52

The Department of Health and Human Services (HHS) has emphasized the importance of data
from reported tests, including antigen tests, for public health disease surveillance purposes. The
department noted that access to clear and accurate testing data is essential to the public and
community leaders as they use data to make response and reopening decisions.53

HHS entities and stakeholders involved in testing and data reporting. Test result data are
reported to HHS through jurisdictional health authorities. In general, laboratories and point-of-
care settings (such as doctors’ oﬃces) report test results to local, state, territorial, and tribal public
health authorities. These jurisdictional health authorities then report the data to CDC and HHS.

FDA plays a role in testing by issuing emergency use authorizations or approvals of test products.
In addition, housed within HHS are two testing-related working groups (referred to as HHS working
groups in this enclosure). The ﬁrst is the Testing and Diagnostics Working Group, whose purpose
is to accelerate and support U.S. testing capacity. This working group is under the Oﬃce of the
Assistant Secretary for Preparedness and Response (ASPR), while members come from CDC, FDA
and other HHS and non-HHS agencies and departments.54 The second is the Data Strategy and
Execution Workgroup, which is a multidisciplinary U.S. Government interagency team that was

52Public health disease surveillance is the ongoing, systematic collection, analysis, and interpretation of health-
related data essential to planning, implementation, and evaluation of public health practice. Response eﬀorts
include activities such as contact tracing to identify persons potentially exposed to the infected individual. In
collaboration with state, local, territorial, academic, commercial, and other partners, the Centers for Disease
Control and Prevention is using multiple disease surveillance systems to monitor COVID-19 that draw upon a
combination of data sources and systems, including laboratory data (i.e., testing data), case and death reporting,
variant surveillance, and other sources.
53See Department of Health and Human Services guidance: COVID-19 Pandemic Response, Laboratory Data Reporting:
CARES Act Section 18115. June 4, 2020 (Updated January 8, 2021).
54Members of the Testing and Diagnostics Working Group come from agencies within HHS, including ASPR, CDC, FDA,
the National Institutes of Health, and the Centers for Medicare and Medicaid Services. Members outside HHS include the
Department of Defense.

Page 45 GAO-22-105051
created in June 2020 under direction of the White House COVID-19 Task Force. ASPR and CDC
serve as co-leads of the Data Strategy and Execution Workgroup.55

Overview of Key Issues
Antigen testing is on the rise. Over the course of the pandemic, the number of authorized
antigen tests has increased. FDA authorized the ﬁrst antigen tests for use at point-of-care settings
in May 2020 and the ﬁrst antigen tests for at-home testing in December 2020. As of October
14, 2021, FDA has issued 36 emergency use authorizations for antigen tests, including tests
authorized for use in laboratory, point-of-care, and at-home settings (see table below). 56

55Members of the Data Strategy and Evaluation Workgroup come from agencies within HHS, including CDC, ASPR, FDA,
the Centers for Medicare and Medicaid Services, and Indian Health Service. Members outside HHS include the United
States Digital Service, Federal Emergency Management Agency, Department of Education, Department of Defense, and
Department of State.
56Laboratories refers to those certiﬁed under the Clinical Laboratory Improvement Amendments of 1988, 42 U.S.C.
§263a, that meet requirements to perform high and moderate complexity tests.

Page 46 GAO-22-105051
Number of FDA Emergency Use Authorizations for COVID-19 Antigen Tests by Setting, as of October 14, 2021
Authorized settings Number of tests

Laboratory only 6

Laboratory and point of care 19

Laboratory, point of care, and at home with a
prescription 3

Laboratory, point of care, and at home without a
prescription (over-the-counter) 8

Total 36

Source: GAO analysis of Food and Drug Administration (FDA) information. I GAO-22-105051

HHS working group oﬃcials told us that antigen testing has comprised a larger share of testing
over time, which will likely continue to increase. Our analysis of CDC’s monthly counts of reported
testing data also suggests that antigen testing has increased as a whole and as a percent of
total tests reported (antigen and molecular). According to the CDC data, the number of reported
antigen tests per month increased from about 50,000 in June 2020 to nearly 12 million in August
2021, and as a percentage of total tests, reported antigen tests increased from less than 1 percent
in June 2020 to more than 20 percent of all tests reported in July and August 2021.57

In addition, on September 9, 2021, the Administration announced the “Path out of the Pandemic”
plan, which is expected to further increase the availability of antigen tests for COVID-19.58
According to the plan, the Administration will exercise authorities in the Defense Production Act to
support sustained manufacturing capacity and will spend nearly $2 billion to procure 280 million
antigen tests, including rapid point-of-care and over-the-counter COVID-19 tests. As part of the
plan, the Administration announced that certain major retailers will sell over-the-counter tests
at cost, resulting in discounts of up to 35 percent. The administration also plans to make antigen
tests free for Medicaid beneﬁciaries and provide 25 million tests to 1,400 community health
centers and hundreds of food banks.

HHS entities have taken actions in an eﬀort to improve reporting of antigen test results. In
June 2020, HHS issued guidance prescribing reporting requirements for laboratories and point-
of-care settings conducting COVID-19 testing, which includes antigen tests.59 However, federal

57The testing database is housed and maintained by CDC. Data may not include results from all testing sites within a
jurisdiction, such as all point-of-care testing sites, according to CDC.
58The White House, Path out of the Pandemic: President Biden’s COVID-19 Action Plan (Washington, D.C.: September 9,
2021), https://www.whitehouse.gov/covidplan/, accessed September 16, 2021.
59Starting in March 2020, all laboratories and point-of-care settings that perform or analyze COVID-19 diagnostic tests
must report the results to public health authorities. CARES Act, Pub. L. No. 116-136, div. B, tit. VIII § 18115, 134 Stat.
281, 574 (3-27-2020). For the duration of the COVID-19 emergency declaration, every laboratory that performs or
analyzes a test that is intended to detect SARS-Cov-2 or to diagnose a possible case of COVID-19 is required to report
the results to HHS “in such form and manner, and at such timing and frequency as the Secretary may prescribe”. In
addition, the Secretary may make prescriptions under this section by guidance without regard to formal rulemaking

Page 47 GAO-22-105051
reporting requirements do not pertain to individuals using at-home over-the-counter antigen
tests. See ﬁgure below for information on reporting requirements by setting.

Public Health Reporting Requirements or Guidance by Antigen Test Authorized Settings

Note: All antigen tests are authorized for laboratory settings, and some are also authorized for point-of-care and at-home
settings.

Since HHS published guidance in June 2020, the department and CDC have tried to improve
the reporting of testing data—including antigen testing data—to local, state, and federal health
oﬃcials, including by taking the following actions:

The department has updated its guidance to clarify test reporting requirements, including those
for antigen testing, for laboratories and point-of-care settings. In January 2021, HHS updated its
guidance on reporting requirements in an eﬀort to further facilitate complete and comprehensive

procedures, including which laboratories must submit reports. See, e.g. Department of Health and Human Services,
COVID-19 Pandemic Response, Laboratory Data Reporting: CARES Act Section 18115 (Jun. 4, 2020; updated January 8, 2021).
Points of care that wish to perform COVID-19 testing themselves, including non-health-care entities, must have a Clinical
Laboratories Improvement Amendments Certiﬁcate of Waiver from the Centers for Medicare & Medicaid Services before
performing COVID-19 testing and are required to report all test results.

Page 48 GAO-22-105051
laboratory testing data reporting.60 For example, the January 2021 guidance clariﬁed the various
methods for submitting test results. This guidance applies to antigen testing conducted in
laboratory and point-of-care settings—including non-health-care settings such as workplaces and
schools.

CDC issued guidance to encourage reporting of antigen test results from individuals and non-health-care
settings. Although federal reporting requirements do not pertain to individuals using at-home over-
the-counter antigen tests, CDC issued guidance for over-the-counter antigen testing in April 2021,
intended in part, to encourage reporting. The guidance noted that users should report test results
to their health care providers or to local or state health departments if they do not have a health
care provider.61

CDC also issued guidance to employers about the importance of reporting antigen test results. It
noted that employers operating a testing program as a point-of-care setting are required to report
results to public health oﬃcials in accordance with the CARES Act. It also encouraged employers
who collect test results from employees who self-test using over-the-counter tests to report these
test results as well—stating that “sharing results with local public health authorities supports
contact tracing eﬀorts to slow the spread of COVID-19.”62

FDA and HHS working groups collaborated with test manufacturers to encourage and facilitate user
reporting of results. FDA does not have the authority to require individuals to report over-the-
counter test results. However, FDA oﬃcials told us they worked with manufacturers during the
emergency use authorization application process, as well as during town halls they hosted with
manufacturers, to identify approaches that encourage and facilitate such reporting by individuals.
For example, one approach is a smart phone application that communicates tests results to the
user and automatically transmits results to public health oﬃcials. In addition, HHS working group
oﬃcials told us they have held collaborative discussions with test manufacturers to identify ways
to ensure high-quality diagnostic data are captured from antigen tests used both in at-home and
in point-of-care settings, such as schools.

Limited antigen test reporting prevents HHS and CDC from using these data for COVID-19
surveillance. HHS working group and CDC oﬃcials told us that limited antigen test reporting to
jurisdictional health departments and from jurisdictions to CDC, prevents them from using these
data for COVID-19 surveillance. Instead, these oﬃcials told us that they use antigen test data to
track trends in tests distributed, test supply availability, and limited programmatic metrics.

HHS working group and CDC oﬃcials noted that the number of reported antigen test results
is much lower than the expected number of administered tests. CDC oﬃcials also told us they
do not presently have an approach to capture the number of antigen tests administered. This
total number is needed to calculate the proportion of antigen tests in a given population that are
positive for infection (also known as, the test positivity rate)—a key public health indicator for

60Department of Health and Human Services, COVID-19 Pandemic Response, Laboratory Data Reporting: CARES Act Section
18115 (Jun. 4, 2020; updated January 8, 2021).
61Centers for Disease Control and Prevention, Self-Testing, (May 14, 2021; updated August 2, 2021).
62Centers for Disease Control and Prevention, Antigen Testing in Non-Healthcare Settings: A Tool to Prevent the Spread of
COVID-19, (May 4, 2021).

Page 49 GAO-22-105051
surveillance of COVID-19, as we reported in January 2021. As a result, test positivity rates have
generally included only molecular tests.63 As antigen test availability increases, especially over-the-
counter tests, test positivity rates based only on molecular tests, or based on limited antigen test
data, could be less useful as indicators of trends in COVID-19 spread.

At the state and local levels, representatives from four national stakeholder organizations told us
that the limited reporting of antigen test data can hinder the use of these data for surveillance.64
For example, representatives from one organization told us that some local health department
oﬃcials are limited in their ability to use antigen testing data to understand the spread of
COVID-19 in their communities, including helping to identify local outbreaks, perform contact
tracing, and identify individuals where sequencing may need to be performed to monitor the
spread of variants.65

HHS and CDC are taking steps aimed at improving reporting and exploring additional
approaches for eﬀective COVID-19 surveillance. HHS working group and CDC oﬃcials told
us they are considering approaches to further improve antigen test data reporting, as they
believe these data could be valuable for surveillance if reporting were more complete. For
example, these approaches include continuing to work with test manufacturers and making
enhancements to data reporting by building reporting methods into the testing process, such
as for testing processes used in schools and workplaces. More complete data and reporting
from such approaches could provide an earlier indication and fuller picture of where community
transmission is occurring, especially among individuals who are asymptomatic or experiencing less
severe symptoms, according to oﬃcials.

Representatives from one stakeholder group told us that targeted antigen testing in speciﬁc areas
can have a role in surveillance. Additionally, representatives from another stakeholder group
told us that not every antigen test result is expected or needs to be reported in order to use the
data for surveillance purposes. Stakeholder groups also commented on the reporting challenge
speciﬁc to over-the-counter test results, stating that there is likely no realistic way to mandate that
individuals report the results from these types of tests.

HHS working group and CDC oﬃcials stated that while they are trying to improve reporting of
antigen testing data, they are also considering surveillance approaches to supplement or enhance
current surveillance eﬀorts. For example, CDC is exploring wastewater surveillance approaches.
According to CDC oﬃcials, wastewater surveillance provides data that can complement and
conﬁrm other forms of clinical case-based surveillance for COVID-19, and it can provide an
eﬃcient pooled community sample that is particularly useful in areas where timely COVID-19
clinical testing is underutilized or unavailable.

63Molecular tests are more reliably reported than are antigen tests, as the former are mainly conducted through health
care providers and laboratories.
64These organizations collectively represent local, state, and territorial public health agencies and laboratories, and state
epidemiologists.
65States vary in whether they report information about antigen test results on their COVID-19 public websites or
dashboards—a monitoring tool the public can access and that states and territories have been using throughout the
pandemic.

Page 50 GAO-22-105051
If antigen testing continues to expand as expected, especially with the increasing availability of
over-the-counter tests, the ongoing limited reporting of antigen test results could reduce the
ability of public health oﬃcials to more comprehensively monitor and eﬀectively respond to
COVID-19. These issues highlight the importance of HHS working group and CDC eﬀorts aimed at
improving reporting and exploring additional approaches for eﬀective surveillance.

Methodology
To understand federal eﬀorts to collect antigen test data and the role these data play in
monitoring COVID-19, we reviewed relevant CDC, FDA, and HHS documentation, such as HHS
and CDC guidance related to the reporting of antigen test results. We reviewed publicly available
information from FDA’s website on its emergency use authorizations for COVID-19 antigen
tests, including the conditions of the authorizations, as well as to determine the number of tests
authorized and the authorized settings for these tests. In addition, we collected written responses
to questions we submitted to CDC, FDA, and HHS working groups.

To gather perspectives from public health stakeholder groups involved in testing and surveillance,
we interviewed representatives from four national organizations that collectively represent
local, state, and territorial public health agencies and laboratories, and state epidemiologists:
the Association of State and Territorial Health Oﬃcials, the Council of State and Territorial
Epidemiologists, the National Association of County and City Health Oﬃcials, and the Association
of Public Health Laboratories.

To describe public reporting of antigen test data, we assessed general trends in the numbers of
monthly reported antigen tests relative to molecular tests as reported to HHS. We assessed the
reliability of the monthly testing data by reviewing HHS and CDC information about the data and
limitations. We determined that the data were suﬃciently reliable for the purpose of assessing
high level, general trends in antigen test data.

Agency Comments
We provided a draft of this enclosure to HHS and the Oﬃce of Management and Budget for review
and comment. HHS provided technical comments, which we incorporated as appropriate. The
Oﬃce of Management and Budget did not provide comments on this enclosure.

GAO’s Ongoing Work
We are continuing to review federal eﬀorts related to COVID-19 testing and surveillance. Our
ongoing work will further explore the evolving testing landscape and related data collection,
as well as surveillance approaches CDC is using, or plans to use, to monitor COVID-19 in the
continued response and into recovery.

Page 51 GAO-22-105051
GAO’s Prior Recommendations
The table below presents our recommendations on testing from prior bimonthly and quarterly
CARES Act reports.

Page 52 GAO-22-105051
Prior GAO Recommendations Related to COVID-19 Testing
Recommendation Status

The Director of the Centers for Disease Control and Prevention Open—not addressed. As of September 2021, we
should work with appropriate stakeholders—including public are awaiting updates from the agency.
health and private laboratories—to develop a plan to enhance
laboratory surge testing capacity. This plan should include
timelines, deﬁne agency and stakeholder roles and responsibilities,
and address any identiﬁed gaps from preparedness exercises (July
2021 report).

The Director of the Centers for Disease Control and Prevention Open—not addressed. As of September 2021, we
should assess the agency's needs for goods and services for are awaiting updates from the agency.
the manufacturing and deployment of diagnostic test kits in
public health emergencies. This assessment should evaluate
how establishing contracts in advance of an emergency could
help the Centers for Disease Control and Prevention quickly and
cost-eﬀectively acquire these capabilities when responding to
future public health emergencies, including those caused by novel
pathogens, and should incorporate lessons learned from the
COVID-19 emergency (July 2021 report).

The Secretary of Health and Human Services (HHS) should develop Open—not addressed. HHS partially agreed
and make publicly available a comprehensive national COVID-19 with our recommendation. In January 2021, HHS
testing strategy that incorporates all six characteristics of an agreed that the department should take steps to
eﬀective national strategy. Such a strategy could build upon more directly incorporate some of the elements
existing strategy documents that HHS has produced for the public of an eﬀective national strategy, but expressed
and Congress to allow for a more coordinated pandemic testing concern that producing such a strategy at this
approach (January 2021 report). time could be overly burdensome on the federal,
state, and local entities that are responding to the
pandemic, and that a plan would be outdated by
the time it was ﬁnalized or potentially rendered
obsolete by the rate of technological advancement.
In May 2021, HHS told us that the White House and
HHS plan to execute a National Testing Strategy
that will act upon the administration's testing
goals. According to HHS, a ﬁnalized document is
forthcoming that includes speciﬁc actions as well
as timelines to achieve these goals. HHS said the
National Testing Strategy will speak to the country's
short-term COVID-19 needs as well as the long-
term needs associated with the country's broader
biopreparedness. We will continue to monitor the
implementation of this recommendation. As of
September 2021, we are awaiting updates from the
agency.

To improve the federal government’s response to COVID-19 and Open—not addressed. HHS partially agreed
preparedness for future pandemics, the Secretary of Health with our recommendation. As of July 2021, HHS
and Human Services should immediately establish an expert stated that it plans to consider ways to establish
committee or use an existing one to systematically review and more permanent work groups to incorporate
inform the alignment of ongoing data collection and reporting best practices for ongoing interagency data
standards for key health indicators. This committee should include needs and to scale up as necessary during future
a broad representation of knowledgeable health care professionals public health emergencies. HHS also stated that
from the public and private sectors, academia, and nonproﬁts the Data Strategy and Execution Workgroup,
(January 2021 report). established as part of the HHS COVID-19 response,

Page 53 GAO-22-105051
Recommendation Status
has helped address the need for a federal
interagency coordination process to align ongoing
COVID-19 data collection and reporting eﬀorts.
We maintain that immediately establishing
an expert committee—not limited to federal
agency oﬃcials—that includes knowledgeable
health care professionals from the public and
private sectors, academia, and nonproﬁts is an
important and worthwhile eﬀort to help improve
the federal government's response to COVID-19
and its preparedness for future pandemics. As of
September 2021, we are awaiting updates from the
agency.

The Secretary of Health and Human Services should ensure that Open—partially addressed. HHS agreed with our
the Director of the Centers for Disease Control and Prevention recommendation and has begun to implement it.
(CDC) clearly discloses the scientiﬁc rationale for any change to For example, on February 16, 2021, CDC issued
testing guidelines at the time the change is made (November 2020 Interim Guidance on Testing Healthcare Personnel
report). that stated asymptomatic health care personnel
who have recovered from COVID-19 may not need
to undergo repeat testing or quarantine in the
case of another exposure within 3 months of their
initial diagnosis. To support this guidance, CDC's
website provided links to studies that explained the
scientiﬁc rationale. Additionally, CDC told us that
it continues to consult with scientiﬁc stakeholders
when issuing or updating guidance documents,
and outlined a series of steps the agency plans to
take to strengthen its testing guidance. However,
as of September 2021, CDC had not fully addressed
the recommendation. We will monitor the
implementation of this recommendation to ensure
that these eﬀorts continue.

Source: GAO. I GAO-22-105051

Contact information: SaraAnn Moessbauer, 202-512-4943, MoessbauerS@gao.gov, and Mary
Denigan-Macauley, 202-512-8552, DeniganMacauleyM@gao.gov

Page 54 GAO-22-105051
FDA Oversight of COVID-19 Vaccine Manufacturing Quality
The Food and Drug Administration took a variety of steps to help ensure the manufacturing
quality of the COVID-19 vaccines authorized for emergency use.

Entity involved: Food and Drug Administration, within the Department of Health and Human
Services

Background
The Food and Drug Administration’s (FDA) oversight of manufacturing establishments is a critical
tool in how the agency helps ensure the manufacturing quality of vaccines, including those used to
prevent COVID-19.

FDA generally licenses vaccines for marketing in the U.S. through approval of a biologics license
application (BLA). BLAs contain data intended to support the application, including data from
non-clinical studies and clinical studies on the safety and eﬀectiveness of the product, as well as
manufacturing data and information. According to FDA, as part of each BLA review, it assesses
manufacturing processes, establishments involved in manufacturing, and the quality and
consistency of the product.

FDA also inspects the establishments involved in manufacturing vaccines as part of the BLA
review process or after the product is licensed for the U.S. market. These inspections are oﬃcial
examinations of establishments to determine compliance with the law and applicable regulations.
Inspections may result in written observations provided to each manufacturer and an inspection
classiﬁcation, which is an assessment of the seriousness of the observations from the inspection.
According to FDA, pre-license and pre-approval inspections are needed in about 20 percent of
instances, usually in cases in which an establishment has a history of compliance issues or when
FDA has not previously inspected the establishment.66

Under certain circumstances, such as the COVID-19 pandemic, FDA may temporarily allow the
use of an unlicensed product through an emergency use authorization (EUA), provided certain
statutory criteria are met.67 For example, FDA may issue an EUA if it is reasonable to believe that
the product may be eﬀective and the known and potential beneﬁts of the product outweigh
the known and potential risks. FDA guidance indicates that companies requesting EUAs should
provide information about manufacturing processes and controls for establishments involved in
manufacturing. FDA may conduct inspections of manufacturing establishments prior to issuing
an EUA, but the statutory criteria for EUAs do not require FDA to conduct such inspections prior
to issuance. FDA guidance for COVID-19 vaccines states that any vaccine company that initially

66Depending on the circumstances, a sponsor may be required to seek and obtain FDA approval of certain changes to
an existing BLA to ensure the safety and eﬀectiveness of the biologic has not been adversely aﬀected. This may include
changes to the product, production process, quality controls, equipment, facilities, or responsible personnel. See 21
C.F.R. § 601.12 (2020).
67See 21 U.S.C. § 360bbb-3. In November 2020, GAO reported on FDA’s use of EUAs for COVID-19 therapeutics and
recommended that the agency identify ways to uniformly disclose to the public the information from its scientiﬁc review
of safety and eﬀectiveness data when issuing EUAs for COVID-19 therapeutics and vaccines. FDA has since implemented
this recommendation.

Page 55 GAO-22-105051
receives an EUA is expected to conduct further research and work towards submission of a BLA as
soon as possible.

Overview of Key Issues
To help ensure manufacturing quality for the COVID-19 vaccines authorized for emergency
use, FDA reviewed documentation and conducted on-site reviews. As of August 2021, FDA
had authorized three COVID-19 vaccines (Pﬁzer, Moderna, and Janssen) for emergency use, and
one of these vaccines (Pﬁzer) had also been licensed.68 To assess each of the 18 establishments
that manufactured these COVID-19 vaccines, FDA took a variety of steps, including reviewing
documentation provided by the vaccine companies for their EUA requests and reviewing prior
inspections reports.69

Documentation from EUA requests. According to FDA, the agency reviewed documentation and
data that each of the COVID-19 vaccine companies included in their EUA requests about the
establishments used in manufacturing the vaccines. Examples of key manufacturing information
that FDA reviewed included:

• Quality systems and controls, the adequacy of the building design and equipment, and the
container storage and closure conditions to ensure the sterility of the product in the ﬁnal
container.

• Cross-contamination controls to ensure they are suitable to mitigate risk of cross
contamination.

• The adequacy of multiple product manufacturing areas used to manufacture the vaccine,
including cleaning and changeover procedures.

68
These three vaccine companies participated in the HHS-DOD COVID-19 Countermeasures Acceleration Group, formerly
known as Operation Warp Speed. This partnership was formed in 2020 to accelerate the availability of a vaccine to
prevent COVID-19 by awarding federal contracts and other transaction agreements to six vaccine companies for
diﬀerent types of activities, including clinical development and manufacturing activities or the purchase of COVID-19
vaccine doses. AstraZeneca, Novavax, and Sanoﬁ/GSK also participated in the partnership’s eﬀorts. In February 2021,
GAO reported on the status of the partnership’s accelerated vaccine development eﬀorts as well as participating vaccine
companies’ technology readiness levels.
Pﬁzer’s COVID-19 vaccine was ﬁrst authorized for emergency use on December 11, 2020 for those 16 years of age
and older. Moderna’s vaccine was ﬁrst authorized on December 18, 2020, and Janssen’s vaccine was ﬁrst authorized
for emergency use on February 27, 2021, both for those 18 years of age and older. FDA subsequently amended the
EUAs for the Pﬁzer and Moderna vaccines to expand their use, such as for diﬀerent age groups or patients with certain
conditions. On August 23, 2021, Pﬁzer’s COVID-19 vaccine was licensed for individuals 16 years of age and older. Pﬁzer’s
vaccine continues to be available under an EUA for individuals 12 to 15 years of age. Pﬁzer developed its COVID-19
vaccine in collaboration with BioNTech. Janssen Pharmaceutical Companies are a part of Johnson & Johnson.
69These 18 establishments were identiﬁed by the vaccine companies as manufacturing the COVID-19 vaccines for the
U.S. market as of June 2021. Of the 18 establishments, 11 were located in the U.S. and seven in Europe. These include
establishments to manufacture the drug substance—bulk amounts of the unformulated active substance, and for ﬁll-
ﬁnish—the transfer of the vaccine into sterile containers.

Page 56 GAO-22-105051
• The qualiﬁcation of critical equipment for manufacturing the drug substance and drug
product.70

Documentation of prior FDA inspections. For the purpose of the EUA review process, FDA was
not required to and did not conduct inspections speciﬁcally examining COVID-19 vaccine
manufacturing for any of the establishments prior to authorizing the three vaccines for emergency
use.71 However, FDA did review the establishments’ inspection histories.

Our review found that 14 of the 18 establishments that manufacture COVID-19 vaccines had
been previously inspected at least once by FDA from October 2011 through the date the initial
EUA request for the COVID-19 vaccine was submitted to the agency.72 This amounted to 90 total
inspections. For all but one of these 14 establishments, FDA’s most recent inspection was within
the last 4 years. (See ﬁgure.)

70FDA also reviewed information on the manufacturing process qualiﬁcation of the drug substance, certiﬁcates of
analyses, and comparability assessments for batches of drug substance, among other things.
71The statutory criteria for EUAs do not require FDA to have conducted an inspection prior to issuance. See 21 U.S.C. §
360bbb-3.
72Establishments may manufacture both drugs and biologics, and thus be eligible to receive both drug and biologic
inspections from FDA. We included both biologic and drug manufacturing inspections in our analysis.

Page 57 GAO-22-105051
Establishments Manufacturing the COVID-19 Vaccines, by the Date of the Most Recent Historical FDA Inspection
Prior to Initial Emergency Use Authorization (EUA) Request and Manufacturing Type

Notes: Each icon or group of icons is an establishment identiﬁed by the vaccine companies as manufacturing the COVID-19
vaccines for the U.S. market as of June 30, 2021. We excluded establishments that were not being used in the manufacturing
process as of June 2021, as well as those used for packaging, storage, and laboratory work.
Data are for biologic establishment inspections and drug establishment inspections conducted from October 2011 through
the date on which each vaccine company initially requested an EUA. FDA subsequently amended the EUAs for the Pﬁzer and
Moderna vaccines to expand their use, such as for diﬀerent age groups or patients with certain conditions.
Drug substance refers to the bulk amounts of the unformulated active substance and ﬁll-ﬁnish refers to the transfer of the
vaccine into sterile containers. Lipid nanoparticles are added to the drug substance to stabilize the mRNA.
a
This establishment manufactures both the Moderna and Janssen vaccines.

FDA classiﬁes inspections based on the seriousness of the deﬁciencies identiﬁed during the
inspections. FDA classiﬁed the 90 prior inspections of these establishments, which were conducted
prior to the initial EUA requests, as follows:

• In 22 of 90 inspections, FDA found no objectionable conditions or practices or the conditions
found did not justify further regulatory action (known as no action indicated).

• In 64 of 90 inspections, FDA found objectionable conditions or practices that were not deemed
serious deﬁciencies and did not recommend any administrative or regulatory actions (known
as voluntary action indicated).

Page 58 GAO-22-105051
• In two of the 90 inspections, FDA found serious deﬁciencies requiring regulatory and/or
administrative actions (known as oﬃcial action indicated).73
• One of the establishments was inspected in 2013. FDA reinspected this establishment
later in 2013 and did not ﬁnd serious deﬁciencies during that inspection or during
subsequent inspections in 2014, 2015, and 2018.

• The second establishment was inspected in 2020. FDA later reinspected this
establishment in 2021 and identiﬁed serious deﬁciencies.74 In response, the
establishment submitted a corrective action plan and worked with FDA to correct the
deﬁciencies.

Other information sources. As needed, in addition to reviewing EUA requests and past inspection
documentation, FDA also used a variety of other information sources including on-site reviews,
record requests, reports from foreign regulators, and inspections to further collect information
about and assess the establishments manufacturing the COVID-19 vaccines. FDA utilized some of
these sources prior to issuing the initial EUAs, as well as after.

• On-site reviews. FDA conducted on-site reviews—investigations or site visits—for some of
the establishments manufacturing the COVID-19 vaccines. During the COVID-19 pandemic,
FDA conducted EUA investigations to assess the current operational status of establishments
manufacturing the COVID-19 vaccines and provide feedback on manufacturing earlier in the
process. Site visits are a longstanding tool used by FDA to learn and observe establishment
operations to improve understanding and open a dialogue between industry and the agency.
Investigations and site visits, unlike inspections, do not lead to written observations or
inspection classiﬁcations.75

As of August 2021, FDA conducted ﬁve investigations and three site visits to assess
manufacturing quality for the COVID-19 vaccines at three establishments. For example,
FDA conducted a site visit prior to issuing an EUA at one establishment that had not been
inspected by FDA within the past 9 years.

• Record requests. FDA may request that establishments send records and other information
in advance of or in lieu of certain types of inspections.76 Such records include those that FDA
commonly reviews during an on-site inspection, such as reports on product quality, lists of
all products manufactured at a facility, and summaries of any discrepancies identiﬁed during
manufacturing and testing and any corresponding investigations. FDA uses record requests to
inform inspection planning or decisions to adjust the intervals in between inspections, or as

73As of May 2021, FDA had not ﬁnalized the classiﬁcation for one inspection, and one inspection was classiﬁed as
administrative closure, which indicates that a ﬁnal classiﬁcation had not been indicated at the time of the original
inspection and FDA has no regulatory concern about the establishment’s compliance status.
74This establishment manufactured both a COVID-19 vaccine authorized for emergency use and a COVID-19
vaccine that has not yet been authorized in the U.S.
75FDA oﬃcials said that if serious deﬁciencies are identiﬁed during an investigation, the agency could change the
review to an inspection.
76See 21 U.S.C. § 374(a)(4).

Page 59 GAO-22-105051
substitutes for certain pre-license or pre-approval inspections as determined by the agency.77
For the authorized COVID-19 vaccines, FDA requested records for three establishments prior
to conducting inspections, as of August 2021.

• Foreign regulator reports. FDA also received reports from foreign regulators about COVID-19
vaccine manufacturing for all seven of the establishments that were located in foreign
countries as of August 2021. Two of these establishments had not been inspected by FDA in
almost 10 years, while another had not been inspected since 2013.

• Inspections. While FDA did not conduct any new inspections of these establishments prior to
issuing the initial EUAs, FDA conducted several inspections after. Speciﬁcally, FDA subsequently
inspected six establishments manufacturing the COVID-19 vaccines from April through July
2021. Three of the six inspections were for establishments manufacturing Pﬁzer’s COVID-19
vaccine, including one in Europe, and occurred after the company began submitting its BLA for
review.
One of these six inspections was for a Janssen contractor establishment (Emergent) in April
2021. This establishment had previously been inspected in April 2020 and FDA indicated
that it found serious deﬁciencies. The April 2021 inspection occurred following a report of
contamination with the drug substance for another COVID-19 vaccine not currently authorized
in the U.S. (AstraZeneca).78 This inspection identiﬁed multiple serious deﬁciencies related to
manufacturing quality control systems, building design, equipment, and personnel training.
Following the April 2021 inspection, manufacturing of new batches of the drug substance for
the Janssen vaccine at the establishment and distribution of existing batches were halted.
Manufacturing of the AstraZeneca vaccine was also removed. Emergent subsequently
submitted a corrective action plan and worked with FDA to correct the deﬁciencies. In July
2021, FDA informed Emergent that it did not object to resuming manufacturing of the Janssen
COVID-19 vaccine at the Emergent establishment, according to FDA oﬃcials.

FDA examined the manufacturing quality of the three other COVID-19 vaccines that
have not been authorized or licensed. As of August 2021, three other vaccine companies
(AstraZeneca, Novavax, and Sanoﬁ/GSK) that also participated in the HHS-DOD COVID-19
Countermeasures Acceleration Group (previously known as Operation Warp Speed) had not
requested EUAs or submitted BLAs.79 FDA oﬃcials told us that they have been examining
manufacturing quality for the establishments manufacturing these vaccines as part of the clinical
trial process.

77FDA will only substitute the review of records and other information for pre-license or pre-approval inspections in
certain cases. For example, FDA may choose to do this if the establishment has an acceptable inspection history for
related manufacturing operations. Establishment records alone cannot be used as a substitute for FDA surveillance
inspections.
78Emergent is one of three of HHS’ Biomedical Advanced Research and Development Authority’s Centers for
Innovation in Advanced Development and Manufacturing that are intended to develop and manufacture medical
countermeasures for the federal government.
79As of June 15, 2021, AstraZeneca and Novavax had both announced ﬁndings from phase 3 clinical trials for their
COVID-19 vaccine candidates, while Sanoﬁ/GSK announced on May 27, 2021, that it had started enrollment in a global
phase 3 clinical trial for its vaccine candidate.

Page 60 GAO-22-105051
FDA used information from past FDA inspections and supplemented with information from on-site
reviews and inspections as needed for the 13 establishments manufacturing these vaccines.80

Inspection history. In examining the establishments’ FDA inspection histories, we found that from
October 2011 through May 2021, FDA had previously inspected 11 of the 13 establishments
manufacturing these COVID-19 vaccines.81 This resulted in a total of 73 inspections. All of the
inspected establishments were inspected in the last 3 years. (See ﬁgure.)

Establishments Manufacturing the COVID-19 Vaccines Not Authorized or Licensed, by Date of Their Most Recent
FDA Inspections and Manufacturing Type, as of May 2021

Notes: Each icon or group of icons is an establishment identiﬁed by the vaccine companies as manufacturing the COVID-19
vaccines for the U.S. market as of June 30, 2021. We excluded establishments that were not currently being used in the
manufacturing process as of June 2021, as well as those used for packaging, storage, and laboratory work. Data are for biologic
establishment inspections conducted from October 2011 through May 26, 2021, and for drug establishment inspections
conducted October 2011 through April 29, 2021.
As of August 2021, AstraZeneca, Novavax, and Sanoﬁ/GSK had not requested emergency use authorizations or submitted
biologics license applications for their COVID-19 vaccines. FDA oﬃcials told us that they have been examining manufacturing
quality for the establishments manufacturing these vaccines as part of the clinical trial process.
Drug substance refers to the bulk amounts of the unformulated active substance and ﬁll-ﬁnish refers to the transfer of the
vaccine into sterile containers. An adjuvant is an ingredient used in some vaccines that helps create a stronger immune
response in people receiving the vaccine.
a
This establishment manufactures both the Novavax and Sanoﬁ/GSK vaccines.

80
These 13 establishments were identiﬁed by the vaccine companies as manufacturing the COVID-19 vaccines for the
U.S. market as of June 2021. Of the 13 establishments, 11 were located in the U.S. and two in Europe. These include
establishments to manufacture the drug substance and ﬁll-ﬁnish.
FDA also requested records from four of the 13 establishments, but these records were unrelated to the COVID-19
vaccines.
81These prior inspections may have been related to either the manufacture of a drug or biologic product.

Page 61 GAO-22-105051
FDA classiﬁed these 73 inspections as follows

• In 16 of 73 inspections, FDA classiﬁed the inspection as no action indicated.

• In 49 of the 73 inspections, FDA classiﬁed the inspection as voluntary action indicated.

• In 3 of the 73 inspections, FDA classiﬁed the inspections as oﬃcial action indicated.82
• One of these inspections was conducted in 2013. FDA later reinspected this
establishment twice in 2014, and again in 2015, 2016, 2017, 2018, 2019, and 2021 and
did not identify serious deﬁciencies.

• The second was conducted in 2019. FDA later reinspected this establishment in 2020
and did not identify any serious deﬁciencies.

• The third inspection was conducted in 2020. FDA later reinspected this establishment
in 2021 and identiﬁed serious deﬁciencies.83 In response, manufacturing of the
COVID-19 vaccine at this establishment was removed.

Onsite reviews. As of August 2021, FDA had conducted seven EUA investigations and two site
visits at six establishments manufacturing these vaccines. For example, FDA conducted an EUA
investigation related to a COVID-19 vaccine in March 2021 for an establishment that had not been
inspected by FDA in almost 10 years. This establishment is a contract manufacturer for Texas
A&M University, which serves as one of three of the HHS Biomedical Advanced Research and
Development Authority’s Centers for Innovation in Advanced Development and Manufacturing
intended to develop and manufacture medical countermeasures for the federal government.

Inspections. FDA also reported that it conducted inspections speciﬁcally for the COVID-19 vaccines
at two establishments in April and July 2021.

Methodology
To conduct this work, we obtained information from the six COVID-19 vaccine companies
participating in the HHS-DOD COVID-19 Countermeasures Acceleration Group on the
establishments involved in manufacturing their vaccines for the U.S. market as of June 2021.
We used this information in our examination of FDA’s inspections from the agency’s Field
Accomplishments and Compliance Tracking System. Speciﬁcally, we reviewed FDA data on vaccine
and other biologic inspections from October 1, 2011 through May 26, 2021 and data on drug
inspections from October 1, 2011 through April 29, 2021. We examined data from October 1,
2011, to provide a wide enough range to capture inspection activities over time. We also reviewed
agency guidance and documents, as well as interviews and written responses from FDA oﬃcials
related to the agency’s vaccine oversight activities during the COVID-19 pandemic.

82FDA’s ﬁnal classiﬁcations were not yet available for 5 of the 73 inspections.
83This establishment manufactured both a COVID-19 vaccine authorized for emergency use and a COVID-19
vaccine that has not yet been authorized in the U.S.

Page 62 GAO-22-105051
To assess the reliability of the data, we conducted electronic data testing for missing data and
outliers, reviewed relevant documentation, and obtained information from knowledgeable agency
oﬃcials. We found the data suﬃciently reliable for our purposes.

Agency Comments
We provided the Department of Health and Human Services (HHS) and the Oﬃce of Management
and Budget (OMB) with a draft of this enclosure. HHS provided technical comments, which we
incorporated as appropriate. OMB did not provide comments on this enclosure.

GAO’s Ongoing Work
We will continue to report on the federal eﬀorts surrounding the development, manufacturing,
and distribution of the COVID-19 vaccines, including the use of the HHS-DOD COVID-19
Countermeasures Acceleration Group (formerly known as Operation Warp Speed).

Related GAO Products
Operation Warp Speed: Accelerated COVID-19 Vaccine Development Status and Eﬀorts to Address
Manufacturing Challenges. GAO-21-319. Washington, D.C.: February 11, 2021.

COVID-19: Federal Eﬀorts Accelerate Vaccine and Therapeutic Development, but More Transparency
Needed on Emergency Use Authorizations. GAO-21-207. Washington, D.C.: November 17, 2020.

Contact information: Mary Denigan-Macauley, 202-512-7114, deniganmacauleym@gao.gov

Page 63 GAO-22-105051
FDA Inspections of Biologic Manufacturing during the
COVID-19 Pandemic
The Food and Drug Administration signiﬁcantly reduced inspections of biologic manufacturing
establishments due to the COVID-19 pandemic but identiﬁed alternative tools it could use to help
oversee manufacturing quality during this emergency.

Entity involved: Food and Drug Administration, within the Department of Health and Human
Services

Background
The Food and Drug Administration’s (FDA) inspections of manufacturing establishments are a
critical tool in how the agency oversees the manufacturing quality of biologics used to prevent,
treat, and cure diseases and medical conditions, such as cancers and infectious diseases.

Biologics are a diverse category of products that include vaccines and allergenics, blood and
blood components, and cells, certain human tissues, and gene therapy products.84 Biologics
are generally derived from living material, such as the human body or a microorganism, and
are generally more complex than other, chemically synthesized drugs. Biologics tend to be heat
sensitive and susceptible to microbial contamination, making the manufacturing process for
biologics typically more complex than it is for other drugs.

Unlike chemically synthesized drugs, biologics marketed in the U.S. are mostly manufactured
domestically, according to FDA. However, there is variation by product type. For example, blood
and blood components are overwhelmingly manufactured at establishments located in the U.S.
while vaccine and allergenic manufacturing establishments are mostly located both in Europe and
the U.S.

Most biologics are regulated by FDA’s Center for Biologics Evaluation and Research.85 FDA
generally licenses biologics, such as vaccines, for marketing in the U.S. through approval of
a biologics license application (BLA). BLAs contain data intended to support the application,
including data from non-clinical studies and clinical studies on the safety and eﬀectiveness of the
product, as well as manufacturing data and information. According to FDA, as part of each BLA
review, it assesses manufacturing processes, establishments involved in manufacturing, and the
quality and consistency of the biologic.

FDA inspects the establishments involved in manufacturing biologics as part of the BLA review
process or after the product is licensed for the U.S. market. These inspections are oﬃcial
examinations of establishments to determine compliance with the law and applicable regulations.
Inspections may result in written observations provided to each manufacturer and an inspection

84See 42 U.S.C. § 262(i)(1). Biologics are subject to licensure under the Public Health Service Act, and they are also
considered drugs under the Federal Food, Drug, and Cosmetic Act. Allergenics include patch tests to diagnose the
causes of allergies as well as extracts to diagnose and treat allergies.
85This report focuses on biologics regulated by the Center for Biologics Evaluation and Research. FDA’s Center for Drug
Evaluation and Research also regulates some drugs that may be considered biologics, such as monoclonal antibodies.
The two centers collaborate in regulating and overseeing biologics.

Page 64 GAO-22-105051
classiﬁcation, which is an assessment of the seriousness of the observations from the inspection.
FDA conducts several types of inspections:

• Pre-license and pre-approval inspections. FDA may conduct pre-license and pre-approval
inspections in response to new BLAs or manufacturing changes to existing BLAs.86 According
to FDA, pre-license and pre-approval inspections are needed in about 20 percent of instances,
usually when an establishment has a history of compliance issues or if FDA has not previously
inspected the establishment.

• Surveillance inspections. FDA conducts surveillance inspections after a product is
marketed to determine an establishment’s ongoing compliance with current good
manufacturing practice (CGMP) regulations.87 FDA uses a risk-based approach to select
biologic establishments for surveillance inspections.88 According to FDA, the agency typically
aims to inspect the highest risk establishments every 2 years and inspect the remaining
establishments every 2 to 4 years.

• For-cause inspections. FDA conducts for-cause inspections after a product is marketed to
investigate speciﬁc issues or follow up on a previous regulatory action, such as when serious
deﬁciencies were identiﬁed during a prior inspection.

Overview of Key Issues
FDA inspections of biologic manufacturers had generally been declining since 2012
and mostly stopped during the COVID-19 pandemic. Prior to the COVID-19 pandemic, FDA
inspections of biologic manufacturing establishments had declined slightly since 2012—from
a high of 1,922 in ﬁscal year 2013 to 1,668 in ﬁscal year 2019 (see ﬁgure below). Surveillance
inspections make up the majority of biologic inspections and particularly decreased in 2019. FDA
oﬃcials attributed this decrease in inspections in ﬁscal year 2019 to a lapse in appropriations and
to vacancies among the staﬀ who conduct inspections.89

86Depending on the circumstances, a sponsor may be required to seek and obtain FDA approval of certain changes
to an existing BLA to ensure the safety and eﬀectiveness of the biologic has not been adversely aﬀected. This
may include changes to the product, production process, quality controls, equipment, facilities, or responsible
personnel. See 21 C.F.R. § 601.12 (2020).
87CGMP regulations contain minimum requirements for the methods, facilities, and controls used in
manufacturing, processing, and packing of a product. See 21 C.F.R. Parts 210 and 211 (2020).
88See 21 U.S.C. § 360(h)(3). Historically, FDA regulations required the agency to conduct surveillance inspections of
biologic establishments at least once every 2 years. See 21 C.F.R. § 600.21 (2018). FDA issued a ﬁnal rule to remove
this requirement, which became eﬀective in May 2019. See 84 Fed. Reg. 12,505 (Apr. 2, 2019).
89FDA activities for ﬁscal year 2019 were funded under a continuing resolution through December 21, 2018. See Pub. L.
No. 115-298, 132 Stat. 4382 (2018). FDA funding lapsed until the enactment of another continuing resolution on January
25, 2019. See Pub. L. No. 116-5, 133 Stat. 10 (2019). In March 2021, we also reported on vacancies among staﬀ who
conduct foreign drug inspections.

Page 65 GAO-22-105051
FDA Biologic Inspections by Year and Type of Inspection, Fiscal Years 2012–2021

Notes: FDA conducts pre-approval and pre-license inspections for new biologic license applications or manufacturing changes
to existing licenses. FDA conducts surveillance inspections after a product is marketed to ensure ongoing compliance with
current good manufacturing practice regulations. For-cause inspections are conducted after a product is marketed to
investigate speciﬁc issues or follow up on a previous regulatory action, such as when serious deﬁciencies were identiﬁed during
a prior inspection.
Fiscal year 2021 data are as of May 26, 2021.

Similar to what we reported in March 2021 for FDA’s drug inspections, due to the pandemic,
beginning on March 17, 2020, FDA paused most biologic inspections and transitioned to focus
solely on those deemed mission-critical, a designation which it determines on a case-by-
case basis. During the COVID-19 pandemic, FDA gave higher priority to establishments that
manufactured products related to the COVID-19 response or products used to treat serious
diseases or medical conditions for which there is no substitute. Mission-critical inspections may
include both domestic and foreign establishments.

In addition to mission-critical inspections, beginning in July 2020, FDA reported that it resumed
certain other high priority domestic surveillance and for-cause inspections, such as following
up on serious deﬁciencies identiﬁed in previous inspections, in areas where it is safe to do so.
Citing concern for the safety of its employees, these prioritized domestic inspections have all
been preannounced; whereas, prior to the pandemic, FDA generally did not announce domestic
surveillance inspections.90 As of August 2021, FDA oﬃcials said the agency had not set a date for

90In March 2021, we reported that prior to the COVID-19 pandemic, almost all domestic drug inspections were
unannounced, whereas foreign inspections were generally preannounced. As we previously reported, FDA’s practice of
generally preannouncing inspections up to 12 weeks in advance may have given establishments the opportunity to ﬁx
problems before the investigator arrives.

Page 66 GAO-22-105051
when it planned to resume unannounced domestic inspections. FDA’s inspection priorities during
the COVID-19 pandemic are generally the same for all medical products, including drugs, which we
reported on in March 2021.

As a result of the pause and limited restart, inspections further declined signiﬁcantly in ﬁscal years
2020 and 2021. In the year following the decision to pause inspections, from April 2020 through
March 2021, FDA conducted 58 biologic inspections, compared to more than 1,500 inspections
conducted during the same period in the year prior to the pause in inspections (see ﬁgure below).
These 58 biologic inspections were all of domestic establishments.

Number of FDA Biologic Inspections by Month, Fiscal Years 2019–2021

Note: Fiscal year 2021 data are as of May 26, 2021.

Due to limited in-person inspections, FDA identiﬁed alternative tools it could use to help
oversee the quality of biologic manufacturing establishments during the pandemic. These
alternative tools include:91

Records requests. FDA may request that establishments send records in advance of or in lieu of
certain types of inspections.92 Such records include those that FDA commonly reviews during
an on-site inspection, such as reports on product quality, lists of all drugs manufactured at a
facility, and summaries of any discrepancies identiﬁed during manufacturing and testing and
any corresponding investigations. FDA uses record requests to inform inspection planning or
decisions to adjust the intervals in between inspections, or as substitutes for certain pre-approval

91These alternative inspection tools are generally the same as those used for drug inspections. In the case of drug
inspections, FDA also has mutual recognition agreements with European regulators, which allow inspections conducted
by these regulators to substitute for FDA inspections. These mutual recognition agreements currently only apply to
drugs and certain biologics regulated by the Center for Biologics Evaluation and Research (e.g., allergenics). FDA is
considering expanding these agreements to include vaccines in ﬁscal year 2022.
92See 21 U.S.C. § 374(a)(4).

Page 67 GAO-22-105051
inspections as determined by the agency.93 FDA reported issuing record requests to 179 biologic
establishments in ﬁscal year 2020 and 601 biologic establishments in ﬁscal year 2021 (as of May
2021). Prior to the pandemic, FDA did not use record requests as an alternative inspection tool for
those biologics regulated by the Center for Biologics Evaluation and Research, according to FDA
responses.

Investigations and site visits. During the COVID-19 pandemic, FDA conducted emergency use
authorization (EUA) investigations to assess the current operational status of establishments
manufacturing the COVID-19 vaccines and to provide feedback on manufacturing earlier in the
process. Site visits are a longstanding tool used by FDA to learn and observe establishment
operations for the purpose of improving understanding and opening a dialogue between industry
and the agency. Unlike inspections, investigations and site visits do not lead to an inspection
classiﬁcation.94 FDA utilized these tools for several of the establishments manufacturing the
COVID-19 vaccines. For more information about the steps FDA used to help ensure manufacturing
quality for the COVID-19 vaccines, see the FDA Oversight of COVID-19 Vaccine Manufacturing
Quality enclosure.

Remote interactive evaluations. These evaluations include remote live-streaming video of
operations, teleconferences, or screen sharing. In April 2021, FDA issued guidance noting
that remote interactive evaluations, in which establishments have to agree to participate,
do not constitute inspections. Instead, FDA would use these evaluations to determine the
scope, depth, and timing of potential future inspections. At the conclusion of such a remote
interactive evaluation, FDA provides establishments with written observations, as it would for an
inspection. However, unlike what happens following an inspection, FDA does not issue inspection
classiﬁcations for remote interactive evaluations.

Prior to the pandemic, FDA did not use remote interactive evaluations and, as of August 2021,
FDA oﬃcials said the agency had not yet conducted any remote interactive evaluations of biologic
establishments. Agency oﬃcials said that was because the agency is reviewing internal processes
to help ensure implementation of remote interactive evaluations adequately provides the
information needed to assess manufacturing quality.

Representatives of almost all biologic manufacturing associations we spoke to generally supported
FDA’s use of alternative inspection tools both during the COVID-19 pandemic and in the future,
but noted some challenges with the use of these tools. For example, representatives from three
associations noted a lack of communication from FDA following record requests. According to
FDA, the agency plans to revise its record request procedures in ﬁscal year 2022 to increase
communication with manufacturers.

Representatives from three associations also noted a lack of clarity on whether the agency could
use these alternative tools as substitutes for in-person inspections, such as to resolve ﬁndings

93FDA will only substitute the review of records and other information for pre-approval inspections in certain cases. For
example, FDA may choose to do this if the establishment has an acceptable inspection history for related manufacturing
operations. Establishment records alone cannot be used as a substitute for FDA surveillance inspections.
94FDA oﬃcials said that if serious deﬁciencies are identiﬁed during an investigation, the agency could change the review
to an inspection.

Page 68 GAO-22-105051
from prior FDA inspections. FDA’s guidance does not state whether these alternative inspection
tools may be used to resolve prior inspection ﬁndings. However, in a May 2021 FDA presentation
to industry oﬃcials, the agency stated that they could be used in this way. FDA oﬃcials later told
us that alternative inspection tools are not intended to replace inspections to resolve inspection
ﬁndings and are reluctant to use these tools in lieu of an inspection.

FDA may face challenges resuming routine surveillance inspections of vaccine and
allergenic manufacturing establishments, but said vaccine inspections are a priority.
While inspections of vaccine and allergenic establishments represent a small number of total
biologic inspections each year—2 percent in ﬁscal year 2019 (the last ﬁscal year prior to the
pandemic)—inspections of these establishments are largely in foreign countries, such as the
United Kingdom, Germany, and France. In contrast, inspections of other biologic product types,
which comprise 98 percent of total biologic inspections, are mostly domestic (see ﬁgure below).

Number of FDA Inspections and Percentage of Foreign and Domestic Inspections by Biologic Product Type, Fiscal
Year 2019

Note: Inspections may be double counted as they may be categorized as more than one biologic type.

In July 2021, FDA announced it resumed routine domestic surveillance inspections, thereby putting
the agency in a position to begin to return to its prepandemic inspection rates for blood and blood
components and tissue and gene therapy products manufacturing establishments. However, in
August 2021, FDA oﬃcials stated that the agency was uncertain whether the ongoing pandemic
will impose additional disruptions to the domestic inspection operations in the future.

In addition, FDA oﬃcials said the agency does not have a time frame for resuming routine
foreign inspections, as of August 2021, which may aﬀect its ability to conduct routine surveillance
inspections for vaccine and allergenics manufacturing establishments. Oﬃcials said that the
additional challenges with resuming foreign inspections, including travel restrictions and the
length of time needed for trip planning, make it diﬃcult to estimate when foreign surveillance
inspections will resume.

According to FDA oﬃcials, vaccine inspections are a high priority for the agency. They said FDA
is continuously examining which establishments to prioritize for surveillance inspections as

Page 69 GAO-22-105051
determined by the agency’s risk-based approach. FDA oﬃcials said they are shifting resources to
address the highest inspection priorities, including vaccine oversight. Further, if FDA determines
that a foreign manufacturing establishment becomes a high enough risk—for instance, the
establishment has not been inspected in the typical period of time—then the agency may
designate that inspection as mission critical and conduct an inspection, according to FDA oﬃcials.

Methodology
To conduct this work, we reviewed FDA data on biologic inspections from October 1, 2011 through
May 26, 2021 (the most recent data available), from the agency’s Field Accomplishments and
Compliance Tracking System. We examined this time frame to provide a wide enough range to
capture inspection activities over time. We also reviewed agency guidance and documents, as well
as interviews and written responses from FDA oﬃcials related to the agency’s biologic oversight
activities during the COVID-19 pandemic.

To assess the reliability of the inspection data, we conducted electronic data testing for
missing data and outliers, reviewed relevant documentation, and obtained information from
knowledgeable agency oﬃcials. We found the data suﬃciently reliable for our purposes.

Additionally, we interviewed AABB (formerly known as the American Association of Blood Banks),
Alliance for Regenerative Medicine, Association for Accessible Medicines, Biotechnology Innovation
Organization, Pharmaceutical Research and Manufacturers of America, Pharma & Biopharma
Outsourcing Association, and Plasma Protein Therapeutics Association, which represent blood
centers and manufacturers of tissue and cell products, generic drugs and biosimilars, drugs and
biologics, brand-name drugs and biologics, contract manufacturers, and plasma manufacturers,
respectively, on the eﬀects of the temporary postponement of inspections and FDA’s use of
alternative tools.

GAO’s Ongoing Work
As FDA’s inspections of manufacturing establishments remain a critical tool to oversee the
manufacturing quality of biologics and other drugs, even during the pandemic, we will continue to
monitor FDA’s inspection program.

Page 70 GAO-22-105051
Related GAO Product
Drug Safety: FDA’s Future Inspection Plans Need to Address Issues Presented by COVID-19 Backlog.
GAO-21-409T. Washington, D.C.: March 4, 2021.

Contact information: Mary Denigan-Macauley, 202-512-7114, deniganmacauleym@gao.gov

Page 71 GAO-22-105051
Health Insurance Loss
Estimates of employer-sponsored insurance suggest more than 3.1 million non-elderly adults lost
their insurance during the COVID-19 pandemic; some losing this insurance were able to obtain an
alternative source of coverage, though complete data are not yet available.

Entity involved: The Centers for Medicare & Medicaid Services, within the Department of Health
and Human Services

Background
Many Americans receive health insurance through their employer, which is known as employer-
sponsored insurance (ESI).95 COVID-19 and the associated economic downturn likely caused
disruptions in ESI for millions of Americans, although estimates of the magnitude of ESI loss vary.
For those who lost ESI, there were a number of health coverage alternatives available, including
the following options:

• coverage through a federal or state health insurance exchange established under the
Patient Protection and Aﬀordable Care Act. Through federal and state health insurance
exchanges, individuals can compare and select among plans that meet certain federal
standards oﬀered by participating private insurers.96

• Medicaid. Medicaid is a joint federal-state health ﬁnancing program for certain low-income
and medically needy individuals.

• beneﬁts under the Consolidated Omnibus Budget Reconciliation Act of 1985 (COBRA)
and COBRA-like provisions for state and local employees. COBRA provides certain
individuals who lose their employer-sponsored health coverage with temporary access to
continue it for limited periods of time under certain circumstances.

• short-term, limited duration insurance. Short-term, limited duration insurance is a type of
health insurance coverage that was primarily designed to ﬁll temporary gaps in coverage and
is generally exempt from federal health insurance requirements.

• other sources of ﬁnancial support for medical expenses. Other forms of ﬁnancial
support for medical expenses include health care sharing ministries, which are faith-based
organizations that share resources for medical needs among their members but do not have
to comply with federal health insurance requirements.

Congress included several provisions in COVID-19 relief laws to support access to health coverage,
including for those who have lost ESI. For example, the American Rescue Plan Act of 2021
temporarily increases and expands eligibility for tax subsidies for individuals enrolled in coverage
through a health insurance exchange.

95According to data from the U.S. Census Bureau’s Current Population Survey Annual Social and Economic Supplement
ﬁelded in 2019, 55.4 percent of the civilian, noninstitutionalized population surveyed had ESI at the time of interview.
96As of September 2021, 36 states used a federally facilitated exchange and 14 states and the District of Columbia
operated their own state-based exchanges.

Page 72 GAO-22-105051
Overview of Key Issues
ESI loss during the COVID-19-associated economic downturn. Though estimates vary widely,
studies suggest millions likely lost their ESI during the COVID-19-associated economic downturn.
Estimates of ESI loss range from 3.1 to 3.3 million non-elderly adults, or between 12 and 14.6
million workers and their dependents, depending on the methods used to develop the estimate
and the time period and population studied (see table below).97

97We conducted a literature search to identify studies of ESI loss during the COVID-19 pandemic. The table contains a
selection of these studies and does not include all studies of ESI loss identiﬁed.

Page 73 GAO-22-105051
Selected Studies of Employer-Sponsored Insurance (ESI) Loss during the COVID-19-Associated Economic
Downturn
Source and date
published (month Time period
a
and year) Title studied ESI loss estimate Methods

Urban Institute ACA Oﬀers Protection March/April to 3.1 million non-elderly The authors used a nationally
and Robert Wood as the COVID-19 September 2020 adults lost ESI representative Internet-based
Johnson Foundation Pandemic Erodes survey designed to assess
(November 2020) Employer Health how the COVID-19 pandemic
Insurance Coverage is aﬀecting nonelderly adults
and their families to examine
the number of adults who lost
ESI between March/April and
September 2020.

Employee Beneﬁt How Many Americans The start of 7.7 million workers The authors merged health
Research Institute and Have Lost Jobs with the COVID-19 and 6.9 million insurance coverage data
the Commonwealth Employer Health pandemic dependents, or 14.6 with data on unemployment
Fund (October 2020) Coverage During the (approximately million people total, beneﬁt recipients to estimate
Pandemic? February 2020) lost ESI the number of jobs with ESI
to June 2020 coverage that were lost as well
as the number of dependents of
these workers who potentially
lost coverage.

Congressional Federal Subsidies All of calendar 3.9 million people The authors used CBO’s health
Budget Oﬃce (CBO) for Health Insurance year 2020 under 65 would lose insurance simulation model. The
(September 2020) Coverage for People a job with ESI in the model includes the most recent
Under 65: 2020 to 2030 year 2020 (projection) administrative and survey data
on enrollment and premiums;
recently enacted legislation,
judicial decisions, or changes
in regulations; and CBO’s most
recent macroeconomic forecast.

Urban Institute As the COVID-19 April/May to mid- 3.3 million non-elderly The authors used the U.S.
and Robert Wood Recession Extended July 2020 adults lost ESI Census Bureau’s Household
Johnson Foundation Into the Summer of Pulse Survey to assess how
(September 2020) 2020, More Than 3 health coverage changed among
Million Adults Lost adults ages 18 to 64 between
Employer-Sponsored April/May and mid-July 2020.
Health Insurance
Coverage and 2 Million
Became Uninsured

The Economic Policy Health Insurance and February to July 6.2 million workers The authors merged data on
Institute (August the COVID-19 Shock 2020 and 12 million people net employment changes from
2020) lost ESI the Bureau of Labor Statistics
with estimates of ESI coverage
by industry from the U.S. Census
Bureau to generate estimates of
access to ESI, based on industry
employment changes.

Page 74 GAO-22-105051
Source and date
published (month Time period
a
and year) Title studied ESI loss estimate Methods

Urban Institute and Changes in Health April 2020 to 10.1 million people The authors used their health
Robert Wood Johnson Insurance Coverage December 2020 would lose ESI insurance microsimulation
Foundation (July Due to the COVID-19 (projection) model, which incorporated
2020) Recession: Preliminary data on employment losses
Estimates Using by industry and other
Microsimulation characteristics published by the
U.S. Department of Labor to
project loss of ESI following loss
of employment in the last three
quarters of 2020, from April
through December 2020.

Source: GAO summary of selected studies. | GAO-22-105051

Note: We reviewed a number of studies that provided a numerical estimate of ESI loss during the COVID-19-associated time
period and selected to report several here that represent a range of data sources and methods for developing an ESI loss
estimate, including surveys and simulation models.
a
These studies vary in population studied. “Non-elderly adults” refers to adults age 18 through 64; “workers” refers to employed
people who receive ESI through their job; “dependents” refers to those who receive ESI through another person, such as a
spouse or parent.

The Congressional Budget Oﬃce estimated that 14.3 million people would lose their jobs in 2020,
but not all job losses would result in a loss of ESI.

ESI loss was likely less than originally expected:

• Many people who lost their jobs during the COVID-19 pandemic had never been enrolled in ESI
through their jobs. The highest COVID-19-related job loss has been seen in small companies
and lower-wage industries less likely to oﬀer health insurance, such as retail and food service.

• Many individuals were able to retain their ESI while furloughed as the pandemic continued
through 2020. According to a Bureau of Labor Statistics survey, 42 percent of establishments
surveyed paid a portion of health insurance premiums for some or all furloughed employees
while they were not working in calendar year 2020.98

• Some job loss and any associated ESI loss may have been temporary. According to a Bureau of
Labor Statistics survey, of the 13.5 million who reported not working in June 2020, 10.6 million
(or 79 percent) expected to be recalled to work at some point.99

It is not yet clear how the COVID-19 pandemic has more recently aﬀected ESI in 2021. A September
2021 study used data from U.S. Census Bureau surveys to estimate that nearly 3 million people
lost ESI in 2020. However, based on its review of health care administrative data, the study

98See U.S. Department of Labor, Bureau of Labor Statistics, Business Response Survey to the Coronavirus Pandemic,
(Washington, D.C.: Dec. 7, 2020).
99See U.S. Department of Labor, Bureau of Labor Statistics, The Employment Situation—June 2020, (Washington, D.C.:
July 2, 2020).

Page 75 GAO-22-105051
estimated that many of the people who lost ESI and became uninsured during the spring and
summer of 2020 may have eventually found an alternative source of coverage, particularly
Medicaid, later in 2020 or in early 2021.100

Another recent study, a Commonwealth Fund survey ﬁelded March through June 2021, found
that 6 percent of working-age adults reported that they lost ESI because of a job loss during the
COVID-19-associated time period; of those, 67 percent of those who reported losing ESI reported
gaining other coverage.101 Of those who lost ESI and gained other coverage, 20 percent reported
they became insured through another ESI plan, 20 percent reported that they elected COBRA, 16
percent reported they gained Medicaid coverage, and 9 percent reported they gained coverage
through an exchange plan.102 Additional and updated estimates of the eﬀect of the COVID-19-
associated economic downturn on ESI may become available as new data from household surveys
are released in late 2021 and 2022.

Health coverage alternatives for those losing ESI during the pandemic. Some of those losing
ESI during the COVID-19 pandemic were able to obtain coverage through a health insurance
exchange or Medicaid. Enrollment in other options, such as COBRA, may have also increased, but
comprehensive data are not available.

Exchange coverage. Enrollment in exchange coverage increased during the COVID-19 pandemic,
both during annual open enrollment periods available to all consumers and special enrollment
periods available under certain circumstances.103

Plan selections during open enrollment in federal and state exchanges was at its highest level
since 2017 during the 2021 open enrollment period, which occurred during the COVID-19
pandemic. (See ﬁgure.) The Centers for Medicare & Medicaid Services (CMS) data show that there
were about 600,000 (about 5.2 percent) more plan selections during the 2021 open enrollment
period than during the 2020 open enrollment period, which occurred before the COVID-19
pandemic.

100 M. Kate Bundorf, Sumedha Gupta, Christine Kim, Trends in US Health Insurance Coverage During the COVID-19 Pandemic
(September 2021).
101 Sara R. Collins, Gabriella N. Aboulaﬁa, and Munira Z. Gunja, As the Pandemic Eases, What is the State of Health Care
Coverage and Aﬀordability in the U.S.? Findings from the Commonwealth Fund Health Care Coverage and Covid-19 Survey,
March-June 2021, (July 2021).
102 The remaining respondents reported gaining coverage through Medicare or an unspeciﬁed source.
103Open enrollment periods for the federally-facilitated exchanges are generally held between November 1 and
December 15 the year prior to the coverage year. State-based exchanges may use diﬀerent dates.

Page 76 GAO-22-105051
Plan Selections during Open Enrollment Periods in Federal and State-Based Exchanges, 2017-2021

Note: Open enrollment periods for the federally facilitated exchanges are generally held between November 1 and December
15 the year prior to the coverage year. State-based exchanges may use diﬀerent dates.

Special enrollment through the federally facilitated exchange was also higher in 2020, during the
COVID-19 pandemic, than in any prior year data were collected.104 Generally, special enrollment
periods allow an individual to apply for health coverage after experiencing a qualifying event, such
as losing minimum essential health coverage or getting married.

Compared to 2019, more than 300,000 (about 23 percent) more consumers obtained coverage
through a special enrollment period in 2020.105 Most of this increase resulted from consumers
who qualiﬁed for the special enrollment period because of a loss of health insurance. Speciﬁcally,
in 2020, there was a 293,563 (43 percent) increase in enrollments using a special enrollment
period with loss of health coverage as the qualifying event compared to a 25,136 (3.7 percent)
increase in enrollments among those using a special enrollment period who qualiﬁed for another
reason. (See ﬁgure.)

104Only data from the federally facilitated exchange are available for coverage years 2017 through 2020.
105 From April 2020 to July 2020, CMS paused certain requirements for consumers to submit documents verifying
their eligibility for a special enrollment period. Oﬃcials told us this was intended to reduce consumer burden and ease
enrollment in the exchanges during the COVID-19 pandemic.

Page 77 GAO-22-105051
Federal Exchange Special Enrollment by Qualifying Event, 2017-2020

Note: Generally, a special enrollment period is a period during which an individual who experiences certain qualifying events,
such as losing of health coverage or having a child, may enroll in exchange coverage outside of the annual open enrollment
period.

Furthermore, on January 28, 2021, the administration announced a new special enrollment period
for the federally facilitated exchange in response to the COVID-19 public health emergency, which
was available from February 15, 2021, through August 15, 2021. This special enrollment period
was open to all individuals and no qualifying event was required to obtain coverage.106 According
to CMS, the new federal special enrollment period in response to COVID-19 was accompanied by
an outreach campaign to raise awareness among the uninsured about it and about the availability
of ﬁnancial assistance to pay for premiums for those who qualiﬁed.

CMS reported that 2.1 million people obtained coverage through the federal exchange during
this special enrollment period, exceeding total special enrollment in each prior coverage year.107
Reasons for this increase in enrollment may include the availability of the 2021 special enrollment
period open to all consumers and its associated outreach campaign, and provisions supporting
exchange coverage in COVID-19 relief laws. For example, the American Rescue Plan Act of 2021
expanded eligibility for exchange subsidies to those making above 400 percent of the poverty level

106All of the states that operated their own exchanges also opened similar special enrollment periods in response to
COVID-19, though timeframes varied.
107 CMS reported that an additional 738,000 consumers obtained state-based exchange coverage through the end of
each state’s respective reporting period. Several states have extended their special enrollment periods beyond August
15, 2021. For these states, the number of new plan selections through the special enrollment period reﬂects data
reported as of August 31, 2021.

Page 78 GAO-22-105051
and also increased subsidies for those making between 100 and 400 percent of the poverty level
for 2021 and 2022.

Medicaid. Medicaid enrollment increased 8.7 million, or 13.5 percent, between January 2020 and
December 2020, according to data available from CMS. 108 (See ﬁgure below)

Medicaid Enrollment from January to December 2020

Note: Enrollment counts presented in this ﬁgure generally represent the total unduplicated number of individuals enrolled in
comprehensive beneﬁts as of the last day of the reporting period.

Part of the increase in Medicaid enrollment in 2020 may be due to more applications for Medicaid
coverage as well as requirements in the Families First Coronavirus Response Act. The federal
government matches states’ spending for Medicaid services according to a statutory formula
known as the Federal Medical Assistance Percentage (FMAP). The Families First Coronavirus
Response Act provided a temporary 6.2 percentage point increase in the Medicaid FMAP funding
states receive if they meet certain conditions, including providing continuous coverage to Medicaid
beneﬁciaries who were enrolled in Medicaid on or after March 18, 2020, regardless of any changes
in circumstances or redeterminations at scheduled renewals that otherwise would result in
termination, through the end of the month in which the public health emergency ends.109

COBRA. There are no comprehensive data for the time period associated with COVID-19 on the
take-up of COBRA. Generally, the employee must elect COBRA coverage within a 60-day election

108States and territories administer their Medicaid programs within broad federal rules and according to state plans
approved by CMS. States are responsible for determining applicants’ eligibility for Medicaid, including redetermining
eligibility at regular intervals and disenrolling individuals who are no longer eligible. In assessing eligibility for Medicaid,
states must determine whether applicants meet eligibility criteria, such as ﬁnancial and citizenship requirements.
109States may terminate coverage for individuals who request a voluntary termination of eligibility, or who are no longer
considered to be residents of the state.

Page 79 GAO-22-105051
period and must pay the full premium plus an administrative fee, which may be prohibitively
expensive. In May 2020, the administration eﬀectively extended the election period within which
individuals must elect COBRA until 60 days after the end of the COVID-19 public health emergency.
Additionally, for certain qualifying individuals, the American Rescue Plan Act of 2021 included a
100 percent subsidy for COBRA premiums from April through September 2021.

Short-term, limited duration insurance. There are limited data available for the time period
associated with COVID-19 on sales of short-term, limited duration insurance plans sold to
individuals. While it is prohibited in several states and in some states no insurers choose to oﬀer it,
most states had issuers oﬀering short-term, limited duration insurance during 2020.

Other sources of ﬁnancial support for medical expenses. While other sources of ﬁnancial support for
medical expenses may be available, such as health sharing ministries, there are no comprehensive
federal data on general use of these arrangements during the time period associated with
COVID-19, or on their use by those who lost ESI.

Methodology
For this work, we conducted a literature search to identify studies of ESI loss during the time
period associated with COVID-19. We reviewed a number of studies that provided a numerical
estimate of ESI loss during the COVID-19-associated time period and selected to report several
that represent a range of data sources and methods for developing an ESI loss estimate, including
surveys and simulation models. We reviewed CMS data and reports on enrollment in exchanges
from 2017 to 2021 and on enrollment in Medicaid from January 2020 to December 2020. We
assessed the reliability of these data using manual checks and discussions with CMS oﬃcials
and determined they were suﬃciently reliable for our purposes. We also reviewed reports about
other sources of health coverage, such as COBRA, from government agencies and research
organizations for descriptive information about health coverage options.

GAO’s Ongoing Work
GAO will continue to assess the eﬀect of the COVID-19-associated economic downturn on ESI and
examine health coverage options.

Contact information: John Dicken, (202) 512-7114, dickenj@gao.gov

Page 80 GAO-22-105051
HHS COVID-19 Funding
The Department of Health and Human Services was appropriated approximately $484 billion
in COVID-19 relief funds. The Department reported that it had obligated about $351 billion and
expended about $196 billion of this amount—about 72 percent and 40 percent, respectively—as of
August 31, 2021.

Entity involved: The Department of Health and Human Services.

Background
The Department of Health and Human Services (HHS) received approximately $484 billion in
COVID-19 relief appropriations from six COVID-19 relief laws enacted as of August 31, 2021. HHS
COVID-19 relief funds may be used for a range of purposes, such as assistance to health care or
child care providers, testing, therapeutic, or vaccine-related activities, or procurement of critical
supplies. Many HHS COVID-19 relief funds are available for a multiyear period or are available until
expended.

Overview of Key Issues
As of August 31, 2021, of the approximately $484 billion in COVID-19 relief funds appropriated,
HHS reported that it had obligated about $351 billion and expended about $196 billion—about 72
percent and 40 percent, respectively (see ﬁgure below).

HHS-Reported COVID-19 Relief Appropriations, Obligations, and Expenditures from COVID-19 Relief Laws, as of
August 31, 2021

a
These amounts reﬂect appropriations provided in Divisions M and N of the Consolidated Appropriations Act, 2021 that are
speciﬁcally designated for COVID-19 relief.

Page 81 GAO-22-105051
The table below shows HHS appropriations, obligations, and expenditures by COVID-19 relief law
that HHS reported as of August 31, 2021.

Page 82 GAO-22-105051
HHS-Reported COVID-19 Relief Appropriations, Obligations, and Expenditures, by Relief Law, as of August 31,
2021
Obligations Expenditures
Date of Appropriations ($ in millions (% ($ in millions (%
Legislation enactment ($ in millions) obligated)) expended))

Coronavirus Preparedness March 6, 2020 6,497 5,785 (89) 3,655 (56)
and Response Supplemental
Appropriations Act, 2020
(Pub. L. No. 116-123)

Families First Coronavirus Response March 18, 2020 1,314 1,307 (99) 1,261 (96)
Act
(Pub. L. No. 116-127)

CARES Act March 27, 2020 142,833 136,091 (95) 119,656 (84)
a
(Pub. L. No. 116-136)

Paycheck Protection Program and April 24, 2020 100,000 58,387 (58) 50,721 (51)
Health Care Enhancement Act
(Pub. L. No. 116-139)

Consolidated Appropriations Act, December 27, 73,175 57,970 (79) 11,943 (16)
2021 2020
b
(Pub. L. No. 116-260)

American Rescue Plan Act of 2021 March 11, 2021 160,494 91,210 (57) 8,796 (5)
(Pub. L. No. 117-2)

Total 484,313 350,750 (72) 196,032 (40)

Source: Department of Health and Human Services (HHS) data. | GAO-22-105051

Note: The Department of Health and Human Services (HHS) reported that, of its total appropriations for COVID-19 relief, the
agency transferred $289 million to the Department of Homeland Security that is not included in the reported obligations or
expenditures, and that $300 million in appropriations are not available until HHS has taken certain actions.
a
HHS reported that it transferred $289 million from CARES Act appropriations to the Department of Homeland Security and this
amount is not included in HHS’s reported obligations or expenditures.
b
This amount reﬂects appropriations provided in Divisions M and N of the Consolidated Appropriations Act, 2021 that are
speciﬁcally designated for COVID-19 relief. An additional $638 million in COVID-19 relief funds were appropriated under
Division H to the Administration for Children and Families, an agency within HHS, to prevent, prepare for, and respond to the
coronavirus, for necessary expenses for grants to carry out a low-income household drinking water and wastewater emergency
assistance program. However, these funds were not included in the HHS-reported data on HHS COVID-19 relief appropriations,
obligations, and expenditures, as HHS noted that the funds were not considered COVID-19 relief funding for USAspending.gov
reporting purposes.

The table below shows allocations, obligations and expenditures of COVID-19 relief appropriations
made to HHS under the six relief laws by HHS agency or fund as of August 31, 2021.

Page 83 GAO-22-105051
HHS-Reported Allocations, Obligations, and Expenditures of COVID-19 Relief Funding, by Agency or Key Fund, as
of August 31, 2021
Agency or key fund Allocations Obligations Expenditures
($ millions) ($ millions) ($ millions)

Administration for Children and Families 65,054.0 64,253.2 10,094.8

Administration for Community Living 3,200.0 2,990.6 1,031.5

Agency for Toxic Substances and Disease Registry 12.5 12.5 9.5

Centers for Disease Control and Prevention 27,770.0 18,692.3 4,332.7

a
Centers for Medicare & Medicaid Services 935.0 138.0 80.3

Enhanced Use of Defense Production Act 10,000.0 2,043.2 33.0

Food and Drug Administration 718.0 145.2 50.1

Health Resources and Services Administration 11,729.8 9,014.3 2,595.3

Indian Health Service 7,980.0 4,425.9 4,208.1

National Institutes of Health 3,977.4 2,259.7 1,230.0

Oﬃce of Inspector General 17.0 4.7 3.9

b
Public Health and Social Services Emergency Fund (PHSSEF) 344,684.7 240,005.1 172,124.6
c
Oﬃce of the Assistant Secretary for Health 7,206.0 5,011.7 3,888.0
c
Oﬃce of the Assistant Secretary for Preparedness and Response 28,013.1 14,375.5 8,358.9
c
Biomedical Advanced Research and Development Authority 38,246.8 30,694.0 13,335.9
c, d
Provider Relief Fund 178,000.0 135,652.2 132,469.9

c
Other PHSSEF 93,218.8 54,271.7 14,071.9

Substance Abuse and Mental Health Services Administration 8,235.0 6,764.9 238.1

Grand Total 484,313.4 350,749.6 196,031.9

Source: Department of Health and Human Services (HHS) data. | GAO-22-105051

Note: For the purpose of this table, the term allocation includes both direct appropriations and transfers between HHS
agencies. For example, according to HHS, the agency transferred $860 million to the Administration for Children and
Families’ Unaccompanied Children Program from National Institutes of Health appropriations provided in the Consolidated
Appropriations Act, 2021, Pub. L. No. 116-260, 134 Stat. 1182, 1913 (2020), citing the Secretary’s authorities under that act.
HHS reported that of its total appropriation for COVID-19 relief, the agency transferred $289 million to the Department of
Homeland Security that is not included in the reported obligations or expenditures, and that $300 million in appropriations are
not available until HHS takes certain actions. With respect to the Consolidated Appropriations Act, 2021, the amounts reﬂect
appropriations speciﬁcally designated for COVID-19 in Divisions M and N of the act.
a
These amounts do not reﬂect Medicaid and Medicare expenditures that resulted from statutory changes to these programs
under the COVID-19 relief laws.

Page 84 GAO-22-105051
b
PHSSEF is an account through which funding is provided to certain HHS oﬃces, such as the Oﬃce of the Assistant Secretary
for Preparedness and Response. Amounts have been appropriated to this fund for the COVID-19 response to support certain
HHS agencies and response activities. Amounts appropriated to the PHSSEF and transferred to agencies within HHS listed in the
table are included in the allocation amounts for the speciﬁed receiving agencies. For example, the National Institutes of Health
(NIH) received about $1.8 billion in transfers from the PHSSEF and this amount is included in the NIH allocation listed above,
and not in the PHSSEF total.
c
The italicized amounts are subtotals of the PHSSEF and are already reﬂected in amounts listed for the PHSSEF.
d
The Provider Relief Fund reimburses eligible health care providers for health care-related expenses or lost revenues that are
attributable to COVID-19. Provider Relief Fund expenditures also may be referred to as disbursements.

HHS reported allocations, obligations, and expenditures of appropriations from the six COVID-19
relief laws for a variety of COVID-19 response activity categories (see table). When response
activities had spending related to multiple categories, they were only assigned to one. For
example, certain funds for testing and vaccine distribution were included in the response activity
category for support to states, localities, territories, and tribal organizations rather than in the
testing or vaccine activity categories. HHS oﬃcials noted that allocations for COVID-19 response
activities are determined by appropriations made by Congress in combination with approved
spend plan decisions. The timing of obligations and expenditures of allocations for response
activities can vary due to a variety of factors, including the timing of the appropriations, and the
planned uses of funds. For example, some research programs are planned in phases, which
aﬀects the timing of the release of the funds.

Page 85 GAO-22-105051
HHS-Reported Allocations, Obligations, and Expenditures by Selected COVID-19 Response Activity, as of August
31, 2021
Allocations Obligations Expenditures
COVID-19 response ($ in millions) ($ in millions) ($ in
activity Description millions)

Provider Relief Fund
a Includes reimbursements to eligible health care 178,000.0 135,652.2 132,469.9
providers for health care-related expenses or lost
revenues that are attributable to COVID-19.

Testing Includes procurement and distribution of testing 61,416.3 27,678.2 9,359.5
supplies, community-based testing programs,
testing in high-risk and underserved populations
and Indian Health Services’ programs, screening
in schools, Centers for Disease Control and
Prevention (CDC) testing-related activities such as
technical assistance, and other activities.

Child Care and Includes funding for states and other 52,465.0 52,441.5 6,985.2
Development Fund
b governments for child care subsidies for eligible
families and quality improvement activities, sub-
grants to child care providers to stabilize the
child care market, and payments for child care
assistance.

Vaccines Includes Biomedical Advanced Research and 40,039.9 31,857.1 9,462.7
Development Authority (BARDA) funding for
vaccine development and procurement; National
Institutes of Health (NIH) research activities; and
CDC vaccine distribution, administration, and
technical assistance related activities.

Support to state, Includes funding for states and other 40,084.3 39,467.6 9,679.9
local, territorial, and governments to support testing, contact tracing,
tribal organizations’ and surveillance; vaccine distribution; and other
preparedness activities.

Strategic National Includes funds for acquiring, storing, and 13,919.9 10,439.7 6,987.3
Stockpile maintaining ventilators, testing supplies, and
personal protective equipment (PPE) and
increasing manufacturing capacity for certain
PPE.

Drugs and therapeutics Includes BARDA funding for development and 11,459.2 7,068.8 4,861.4
procurement of therapeutics and NIH research
activities.

Health centers Includes support for COVID-19-related activities, 9,620.0 8,533.7 2,358.2
such as testing, at health centers, which provide
health care services to individuals regardless of
their ability to pay.

Rural Provider Includes assistance for rural providers and 8,500.0 0.0 0.0
Payments suppliers that will be administered using the
same mechanism as the Provider Relief Fund,
with disbursement of funds anticipated to begin

Page 86 GAO-22-105051
Allocations Obligations Expenditures
COVID-19 response ($ in millions) ($ in millions) ($ in
activity Description millions)
in approximately the fourth week of November
2021, according to HHS oﬃcials.

Mental health and Includes substance abuse prevention and 8,315.0 6,777.7 238.1
substance use–related treatment, community-based mental health
services services, and other activities.

Diagnostics research Includes BARDA diagnostic development 3,382.1 1,828.1 953.7
and development programs and NIH projects, such as the Rapid
Acceleration of Diagnostics Initiative.

Head Start Includes grants to local programs for high- 2,000.0 1,966.9 603.3
quality learning experiences and to respond to
other immediate and ongoing consequences of
COVID-19.

Testing for uninsured
c Includes reimbursements to eligible providers 2,000.0 1,998.3 1,973.2
for COVID-19 testing for individuals who are
uninsured.

Global disease Includes support to governments and other 1,550.0 609.4 195.7
detection and organizations to rapidly diagnose cases and to
emergency response ensure readiness to implement vaccines and
therapeutics.

Telehealth Includes eﬀorts to support safety-net health care 301.7 152.0 128.9
providers transitioning to telehealth, telehealth
access—especially for vulnerable maternal
and child health populations—and a telehealth
website.

Other response Includes additional activities such as activities 51,260.0 24,278.4 9,774.9
activities
d conducted by the Administration for Community
Living, certain CDC-wide activities and program
support, and activities conducted by the Food
and Drug Administration.

Total 484,313.4 350,749.6 196,031.9

Source: Department of Health and Human Services (HHS) data, written HHS responses, and GAO analysis of HHS spend plans. | GAO-22-105051

Notes: The selected response activities represent examples of certain targeted activities that fall within particular HHS agencies,
such as funding for health centers or Head Start, as well as broader categories of response activities that may span HHS
agencies, such as testing-, vaccine-, and therapeutics-related response activities.
HHS reported allocations, obligations, and expenditures for these activities based on the primary programmatic recipient
organization of the funds, although some activities apply to multiple categories. For example, certain funds in the “support to
state, local, territorial, and tribal organizations for preparedness” category were provided for testing but are not reﬂected in the
“testing” category. However, HHS also noted that testing-related funding awarded to states or localities that was appropriated
under the American Rescue Plan Act of 2021 (ARPA) was included in the “testing” category. HHS oﬃcials explained that the
activity names align with how funds were appropriated under diﬀerent COVID-19 relief laws.
According to HHS oﬃcials, the allocations reported for the key activities above are based on amounts appropriated for these
activities in the COVID-19 relief laws, HHS transfers of funds, and approved spend plan decisions made by HHS in coordination
with the Oﬃce of Management and Budget. According to HHS, the agency used about $1.7 billion in appropriations provided
under ARPA, including $1.2 billion appropriated for COVID-19 testing, contact tracing, and mitigation activities, for the

Page 87 GAO-22-105051
Administration for Children and Families’ Unaccompanied Children Program, citing the Secretary’s authorities under the Public
Health Service Act and the Consolidated Appropriations Act, 2021. See Pub. L. No. 116-260, div. H, tit. II, § 204, 134 Stat. 1182,
1589 (2020); 42 U.S.C. 238j(a). With respect to the Consolidated Appropriations Act, 2021, the amounts reﬂect appropriations
speciﬁcally designated for COVID-19 in Divisions M and N of the act. HHS reported that, of its total appropriation for COVID-19
relief, the agency transferred $289 million to the Department of Homeland Security that is not included in the reported
obligations or expenditures.
a
For additional information about Provider Relief Fund allocations and disbursements, see the Relief for Health Care Providers
enclosure.
b
The Child Care and Development Fund is made up of two funding streams: mandatory and matching funding authorized
under section 418 of the Social Security Act, and discretionary funding authorized under the Child Care and Development Block
Grant Act of 1990, as amended. See 42 U.S.C. §§ 618 and 9858m.
c
According to HHS oﬃcials, HHS has allocated an additional $4.8 billion to the testing for the uninsured program from section
2401 of ARPA, which HHS included in the “testing” response activity category.
d
According to HHS oﬃcials, the agency transferred $860 million from NIH appropriations for research and clinical trials related
to long-term studies of COVID-19 and $850 million from the Public Health and Social Services Emergency Fund, Strategic
National Stockpile, to the Administration for Children and Families’ Unaccompanied Children Program citing authority provided
in section 304 of the Consolidated Appropriations Act, 2021 for both transfers. See Pub. L. No. 116-260, div. M, tit. III, § 304, 134
Stat. at 1913, 1916, 1923 (2020).

Methodology
We requested, and HHS provided, data on appropriations, allocations, obligations, and
expenditures of COVID-19 relief funds by HHS agency and by selected response activity, as
of August 31, 2021. We also reviewed appropriation warrant information provided by the
Department of the Treasury as of August 31, 2021. To assess the reliability of the data reported
by HHS, we reviewed HHS documentation; Department of the Treasury appropriation warrant
information; and other available information on HHS’s use of COVID-19 relief funds. We did not
independently validate the data provided by HHS. We determined that the HHS-reported data
were suﬃciently reliable for the purposes of our reporting objective. We also reviewed the six
COVID-19 relief laws to assist the response to COVID-19.

Agency Comments
We provided HHS and the Oﬃce of Management and Budget (OMB) with a draft of this enclosure.
HHS provided technical comments, which we incorporated as appropriate. OMB did not provide
comments on this enclosure.

GAO’s Ongoing Work
We will continue to examine HHS’s use of COVID-19 relief appropriations contained in COVID-19
relief laws.

GAO’s Prior Recommendation
The table below presents our recommendation related to HHS COVID-19 funding from a prior
quarterly CARES Act report.

Page 88 GAO-22-105051
Prior GAO Recommendation Related to Department of Health and Human Services (HHS) COVID-19
Funding
Recommendation Status

To communicate information about and facilitate oversight Open—not addressed. HHS partially concurred
of the agency’s use of COVID-19 relief funds, the Secretary of with the recommendation and stated that the
Health and Human Services should provide projected time department would aim to incorporate some
frames for the planned spending of COVID-19 relief funds in time frames on planned spending where that
the Department of Health and Human Services’ spend plans information may be available such as time
submitted to Congress. (July 2021 report). frames for select grants to states. However, HHS
stated that the department would not be able to
provide speciﬁc time frames for all relief funds
since the evolving environment requires the
department to remain ﬂexible in responding to
incoming requests for response activities. As of
September 2021, we are awaiting updates from
the agency.

Source: GAO. I GAO-22-105051

Contact information: Carolyn L. Yocom, (202) 512-7114, yocomc@gao.gov

Page 89 GAO-22-105051
Relief for Health Care Providers
As of August 31, 2021, the Department of Health and Human Services had disbursed payments
to providers totaling about $132.5 billion (about 74 percent) of the $178 billion appropriated by
COVID-19 relief laws to the Provider Relief Fund. Health and Human Services has begun payment
integrity eﬀorts but lacks time frames to help ensure that post-payment oversight keeps pace with
the distribution of Provider Relief Funds and the agency can expeditiously determine whether
payments were appropriately made, used, and recovered as warranted.

Entities involved: Department of Health and Human Services, including its Health Resources and
Services Administration

Recommendation for Executive Action
The Administrator of the Health Resources and Services Administration should establish time
frames for completing post-payment reviews to promptly address identiﬁed risks and identify
overpayments made from the Provider Relief Fund, such as payments made in incorrect amounts
or payments to ineligible providers.

The Administrator of the Health Resources and Services Administration should ﬁnalize procedures
and implement post-payment recovery of any Provider Relief Fund overpayments, unused
payments, or payments not properly used.

HHS partially concurred with both recommendations. HRSA stated that it has a schedule for
reviewing the payment discrepancy types it initially prioritized, and that reviews for the remaining
discrepancy types and payment recovery eﬀorts will occur in the future. We maintain that time
frames are still needed for implementing recovery eﬀorts.

Background
Relief funds to health care providers have been allocated and disbursed by Health and Human
Services (HHS) through the following programs.

Provider Relief Fund. To respond to the pandemic, three of the six COVID-19 relief laws
appropriated a total of $178 billion to the Provider Relief Fund (PRF) to reimburse eligible
providers for health care-related expenses or lost revenues attributable to COVID-19. Speciﬁcally,

• the CARES Act appropriated $100 billion;

• the Paycheck Protection Program and Health Care Enhancement Act appropriated $75 billion;
and

• the Consolidated Appropriations Act, 2021, appropriated $3 billion for this purpose.110

110Pub. L. No. 116-260, div. M, tit. III, 134 Stat. 1182, 1920 (2020); Pub. L. No. 116-139, div. B, tit. I, 134 Stat. 620, 622
(2020); Pub. L. No. 116-136, div. B, tit. VIII, 134 Stat. 281, 563 (2020).

Page 90 GAO-22-105051
HHS’s Health Resources and Services Administration (HRSA) administers payments from the PRF,
including allocations to the COVID-19 Uninsured Program and the COVID-19 Coverage Assistance
Fund. In addition to the PRF, the American Rescue Plan Act of 2021 appropriated $8.5 billion for
payments to eligible rural health care providers for health care related expenses and lost revenues
that are attributable to COVID-19.111

HRSA’s planned oversight for the PRF includes post-payment (1) analysis and reviews to determine
whether HRSA made PRF payments to eligible providers in the correct amounts, (2) audits to
determine whether PRF funds were used by providers in accordance with laws and agency
guidance, and (3) recovery of overpayments, unused payments, and payments not properly used.

Accelerated and Advance Payments. HHS’s Centers for Medicare & Medicaid Services (CMS)
Accelerated and Advance Payments Program provides loans to active Medicare providers
and suppliers. Section 3719 of the CARES Act authorized, due to the COVID-19 pandemic, the
expansion of this program, though no new loans have been made since January 2021 as they
relate to the COVID-19 pandemic. Of the $107.3 billion in COVID-19 related loans disbursed under
the program as of September 7, 2021, $36.3 billion in repayments have been made by providers
and suppliers and the current outstanding loan balance for the program is $71.0 billion. The
remainder of this enclosure addresses the PRF and other distributions.

Overview of Key Issues
Provider Relief Fund allocations, disbursements, and returned funds. As of August 31, 2021,
HHS had allocated $153.9 billion of the $178 billion appropriated to the PRF. Of that $153.9
billion allocated, HHS had disbursed about $132.5 billion, and about $21.5 billion remained to be
disbursed. HHS allocated PRF funds, in phases, for general relief to health care providers, relief
for seven targeted areas, and “other distributions,” including funding for treatment, testing, and
vaccine administration, as well as administration of the program. Speciﬁcally, the PRF allocations
included $72.4 billion for general distributions to health care providers; about $55.8 billion for
targeted distributions to certain types of providers and facilities; and $25.8 billion for “other
distributions.”

Approximately $24.1 billion of additional PRF funds remained unallocated and undisbursed, as
of August 31, 2021. On September 10, 2021, HHS announced that $17 billion of the previously
unallocated $24.1 billion would be allocated as part of Phase IV general distributions to a broad
range of providers who could document COVID-related revenue loss and expenses between July 1,
2020 and March 31, 2021. HRSA opened up the application period for these funds on September
29, 2021, and expects to begin disbursing these funds in December 2021. According to HRSA
oﬃcials, the remaining unallocated funds are reserved for future contingencies and emerging
needs for the Uninsured Program.

See table below for a summary of PRF allocations and disbursements as of August 31, 2021.

111Pub. L. No. 117-2, § 9911, 135 Stat. 4, 236-38. According to HHS oﬃcials, these funds will be administered using the
same mechanism as the PRF.

Page 91 GAO-22-105051
Provider Relief Fund: Summary of Allocations and Disbursements, as of August 31, 2021
Disbursement
Allocation ($ in billions)
a
Description ($ in billions) Date of initial disbursement

General distributions

Phase I: Medicare 42.816 April 10, 2020 42.282

Phase II: Medicaid and 3.678 July 3, 2020 3.309
Children’s Health Insurance
Program (CHIP) providers

Phase II: dental providers 1.002 July 28, 2020 0.997

Phase II: assisted living 0.405 September 25, 2020 0.380
b
facilities

Phase III: general distribution 24.500 November 14, 2020 17.362

Subtotal of general distributions 72.401 64.330

Targeted distributions

Rural health care facilities 10.990 May 6, 2020 10.963

High-impact hospitals
c 20.685 May 7, 2020 20.668

Skilled nursing facilities 4.785 May 22, 2020 4.781

Indian health care providers 0.520 May 29, 2020 0.510

Safety net hospitals 13.074 June 12, 2020 12.907

Children’s hospitals 1.063 August 20, 2020 1.062

Nursing home infection control, 4.650 August 27, 2020 4.496
quality, and performance

Subtotal of targeted distributions 55.767 55.387

Subtotal of general and targeted 128.168 119.717
distributions

Other distributions

Treatment, testing, and 10.000 May 15, 2020 6.193
vaccine administration for the
d
uninsured and underinsured

Vaccine and therapeutic 14.801 November 25, 2020 6.484
development and procurement
activities

Page 92 GAO-22-105051
Disbursement
Allocation ($ in billions)
a
Description ($ in billions) Date of initial disbursement

Administration 0.980 n/a 0.076

Subtotal other distributions 25.781 12.753

Unallocated funds
e 24.051 n/a 0.000

Total Provider Relief Fund 178.000 132.470

Legend: n/a = not applicable
Source: Summary of Department of Health and Human Services funding data. | GAO-22-105051

a
Provider Relief Fund disbursements may also be referred to as expenditures.
b
In March 2021, we reported that assisted living facilities were disbursed funds as part of Phase III. In May 2021, Health
Resources and Services Administration (HRSA) oﬃcials told us that these funds were disbursed as part of phase II.
c
High-impact hospitals are hospitals that have a high number of conﬁrmed COVID-19 inpatient admissions.
d
HRSA covers treatment, testing, and administering the vaccine for the uninsured through its COVID-19 Uninsured Program. In
May 2021, HRSA announced that it would cover the cost of administering the vaccine for the underinsured through its COVID-19
Coverage Assistance Fund.
e
HRSA oﬃcials told us that the amount of unallocated funds are available for other Provider Relief Fund allocations. On
September 10, 2021, HHS announced that $17 billion of the previously unallocated $24.1 billion would be allocated as part of
Phase IV general distributions to a broad range of providers who could document COVID-related revenue loss and expenses.
According to HRSA oﬃcials, the remaining unallocated funds are reserved for future contingencies and emerging needs. HHS
also refers to unallocated funds as reserved funds.

Fund disbursements and returned funds. According to our analysis of information provided by HRSA,
as of August 31, 2021, HHS had disbursed about $64.3 billion from general distribution allocations,
about $55.4 billion from the targeted allocations, and $12.8 billion for other distributions.112 As of
August 31, 2021, about 431,163 providers have received 674,549 payments made from the PRF.

Examples of disbursements from the PRF illustrate some of the variation in amounts and
purposes for which the funds were disbursed:

• Providers enrolled in Medicare—some of which were also enrolled in Medicaid or the
Children’s Health Insurance Program (CHIP)—received, on average, approximately $150,000
in relief funds under Phase I of the general distributions. The average COVID-19 relief
disbursement for providers solely enrolled in Medicaid and CHIP was approximately $58,000,
distributed through Phase II of the general distributions, beginning July 3, 2020. HRSA oﬃcials
noted that providers solely enrolled in Medicaid and CHIP tended to be smaller entities with
lower revenue than providers also enrolled in Medicare.

• As of August 31, 2021, approximately $6.2 billion from the PRF had been disbursed for
COVID-19 treatment, testing, and vaccine administration of uninsured and underinsured

112The disbursement of $64.3 billion represents about 89 percent of allocations of the current general distributions, and
$55.4 billion represents about 99 percent of allocations from the targeted distributions.

Page 93 GAO-22-105051
individuals.113 In addition to the allocation from the Provider Relief Fund for treatment,
testing, and vaccine administration for uninsured individuals, the Families First Coronavirus
Response Act and the Paycheck Protection Program and Health Care Enhancement Act each
appropriated $1 billion to reimburse providers for the testing of uninsured individuals.114

Providers return funds on a regular basis. HRSA oﬃcials explained that providers may return funds
if they believe

• the original payment calculation sent to them was too high or they expect a reissuance of a
diﬀerent amount from HRSA—possibly a corrected lower amount or a reissuance to a diﬀerent
entity in their health care system—or

• the funds were not needed, and they have no intention of receiving a new amount through a
reissued payment.

According to HRSA oﬃcials, their system does not have a ﬁeld for providers to indicate their
reasons for returning funds.115 Further, providers can return funds at any time.

According to HRSA, providers had returned about $8.8 billion from previous disbursements as
of August 31, 2021, with about 75 percent ($6.6 billion of $8.8 billion) from general distributions,
and about 25 percent ($2.2 billion of $8.8 billion) from targeted distributions.116 In general, no
funds were returned from providers from the “other distributions.”117 When funds are returned,
the disbursement totals reported are calculated after deducting the returned funds. Oﬃcials
explained that the returned funds are not included in the disbursement totals shown in the above
table, and are available for subsequent allocations.118 For example,

113In May 2021, HHS announced that HRSA would also use PRF funds to cover the costs of administering COVID-19
vaccines to underinsured patients through its COVID-19 Coverage Assistance Fund.
114Pub. L. No. 116-127, div. A, tit. V, 134 Stat. 178, 182 (2020); Pub. L. No. 116-139, div. B., tit. I, 134 Stat.at 626
(2020). These funds were also administered by HRSA as part of the COVID-19 Uninsured Program. According
to HRSA oﬃcials, these funds for claims reimbursement have been depleted; therefore, HRSA will continue to
reimburse COVID-19 testing claims through the COVID-19 Uninsured Program using other funding sources.
115In our March 2021 enclosure, we reported that one large health system headquartered in California returned $771
million in funds from the Medicare, high-impact, skilled nursing, and nursing home infection control distributions. The
system’s press release noted it was able to return the majority of funds due to actions taken to manage expenses.
116According to HRSA, most of the returned funds were linked to payments automatically issued to providers based on
HRSA’s determination of provider eligibility and payment calculation, rather than on providers’ applying for the funds.
117Unlike the general and targeted distributions, “other distributions” were not categorized as disbursements to
providers. According to HRSA, the funds in “other distributions” were used for Vaccine and therapeutic development
and procurement activities; administration; and Treatment, testing, and vaccine administration for the uninsured
and underinsured programs. Providers submitted claims—payment requests—to HRSA for the costs of covering the
uninsured and underinsured. HRSA oﬃcials told us that some recoveries have been made for the Uninsured Program,
which includes returned payments from claims submitted for treatment, testing, and administering the vaccine for the
uninsured.
118The Consolidated Appropriations Act, 2021, provided that not less than 85 percent of PRF funds unobligated as of
the date of enactment and funds recovered from providers after the date of enactment shall be for any successor to the
Phase 3 General Distribution to reimburse health care providers based on applications that consider ﬁnancial losses and

Page 94 GAO-22-105051
• nearly three quarters (about 69 percent, $6.1 billion of $8.8 billion) of all returned funds
were returned to HRSA based on Phase I Medicare payments, which were initially disbursed
beginning on April 10, 2020; and

• about 14 percent or $1.2 billion of returned funds was returned after being disbursed as part
of the targeted allocation for high-impact hospitals—those with a high number of conﬁrmed
COVID-19 inpatient admissions—which were initially disbursed beginning on May 7, 2020.

Provider Relief Fund payment integrity. While HRSA has taken actions to initiate PRF oversight,
the agency has not established time frames to help ensure that its oversight keeps pace with
the distribution of PRF funds and that HRSA expeditiously completes post-payment analysis and
reviews, and recovery eﬀorts to ensure that

• relief payments made by HRSA only went to eligible providers in the correct amounts (post-
payment analysis and reviews), and

• any overpayments, unused payments, or payments not properly used are recovered as soon
as possible (recovery).

Federal internal control standards state that management should design control activities to
respond to identiﬁed risks and achieve objectives. As part of these standards, management
designs speciﬁc actions to respond to the program’s risks, including the potential for fraud,
on a timely basis.119 These standards also state that management should deﬁne objectives,
including clearly deﬁning what is to be achieved, and the time frames for achievement. HRSA’s risk
assessment plan for the PRF, dated September 30, 2020, identiﬁes speciﬁc risks associated with
disbursing funds to providers for use.

Post-payment review and analysis of relief payments made by HRSA. As of September 2021, HRSA
was implementing a post-payment analysis and review process to identify overpayments from
the nearly $120 billion in PRF general and targeted distributions—payments that were made in
incorrect amounts or to ineligible providers. However, we found that agency documents did not
specify time frames for implementing and completing all the remaining post-payment analysis and
reviews. Further, in September 2021, agency oﬃcials provided documentation of time frames set
through the ﬁrst quarter of calendar year 2022, but oﬃcials told us that schedules beyond this
date have not been set.

HRSA has developed a post-payment manual that includes procedures for post-payment analysis
and reviews, as well as a post-payment matrix for scheduling and tracking the reviews. Oﬃcials
told us in September 2021 that the draft was ﬁnalized and implemented in December 2020, and

changes in operating expenses attributable to COVID-19 occurring in the third and fourth quarters of 2020 and the ﬁrst
quarter of 2021.
119Additionally, the Oﬃce of Management and Budget (OMB) Circular No. A-123, Management’s Responsibility for
Enterprise Risk Management and Internal Control, requires executive agencies to evaluate the risks to accomplishing
their strategic, operations, reporting, and compliance objectives and provide an annual Statement of Assurance that
represents the agency head’s informed judgment as to the overall adequacy and eﬀectiveness of the agency’s internal
controls.

Page 95 GAO-22-105051
noted that the manual is evolving, with 11 versions documented between December 2020 and
August 2021. In the matrix, HRSA has identiﬁed 54 types of payment discrepancies for review,
and oﬃcials told us that they began pilot reviews in October 2020. HRSA oﬃcials told us that they
deﬁne payment discrepancy types by reviewing provider data and PRF payment calculations to
identify potential overpayments.120

In September 2021, HRSA provided us information that it was continuing to work on reviews and
that it had closed post-payment reviews for six of the 54 payment discrepancy types identiﬁed.
Oﬃcials told us that HRSA started with reviews of payment discrepancy types identiﬁed as a
priority. Furthermore, they told us that some of the discrepancy types may not require review, and
others will only be reviewed after the prioritized discrepancy types are closed.

According to HRSA oﬃcials, they have begun, but not completed, setting time frames for the
remaining reviews. As of September 15, 2021, of the 48 payment discrepancy types remaining (54
total minus six closed), reviews for 19 are currently either underway or scheduled through the
ﬁrst quarter of 2022.121 According to HRSA and agency documents, time frames for implementing
reviews for the other 29 payment discrepancy types have not been speciﬁed. HRSA oﬃcials stated
that due to the interdependencies of the payment discrepancy types, schedules beyond the ﬁrst
quarter of 2022 have not been made to date. However, establishing time frames for all payment
discrepancy types, regardless of their interdependencies, will assist the agency in tracking the pace
of its completion of reviews and assessing its progress in oversight and recovery eﬀorts.

In addition to reviewing payment discrepancies, HRSA also reported taking action to address
circumstances the agency identiﬁed in its risk assessment plan, which is currently being updated.
In particular, HRSA reported that it was implementing reviews to address certain circumstances
the agency identiﬁed in its risk assessment plan as having a high risk of occurring, such as:122

• Erroneous payments to providers with multiple taxpayer identiﬁcation numbers. HRSA
guidance allows provider organizations with multiple taxpayer identiﬁcation numbers (TINs) to
apply for, receive, and transfer PRF payments among both parent and subsidiary organizations
for up to 1,200 subsidiary TINs. Such a high volume of TINs could make it diﬃcult for HRSA to
calculate and distribute PRF funds appropriately based on TINs.

• Overpayments to providers with a change in ownership. HRSA guidance to providers
states that for those providers with a change in ownership, the original recipient must use
funds for eligible expenses and lost revenues and return unused funds to HHS; the PRF
funds do not transfer to the new owner. HRSA guidance states that providers that have not
received payments under the PRF due to issues related to change of ownership will be eligible
to apply for future PRF payments. However, ensuring the appropriate distribution of funds

120HRSA designed payment discrepancy types to identify computation errors and ineligible providers, among other
things. For example, discrepancy types include providers with multiple submissions, providers with more than 20 TINs,
and providers on the HHS Oﬃce of the Inspector General sanctions list.
121The post-payment matrix speciﬁes time frames for the 19 payment discrepancy types—two were for fourth quarter
2020, seven for ﬁrst quarter 2021, one for second quarter 2021, seven for fourth quarter 2021, one for ﬁrst quarter
2022, and one is ongoing.
122HRSA developed an A-123 risk assessment plan to identify and assess PRF risks and developed internal control
activities in response to such risks, as of September 30, 2020.

Page 96 GAO-22-105051
requires actions by HRSA to ensure that funds unused by the original recipient are tracked and
returned.

Audit of providers’ use of payments. HRSA oﬃcials provided us an audit strategy manual dated
September 30, 2021, that it says will guide the agency’s audits to determine whether providers
complied with the requirements for the use of PRF funds, speciﬁcally that they used payments
to cover only COVID-19 eligible expenses or related revenue losses not reimbursed from other
sources in accordance with laws and HRSA guidance.123

HRSA oﬃcials noted that they would schedule their audits to coincide with the receipt of the ﬁrst
round of provider reports, which were due by September 30, 2021.124 However, on September 10,
2021, HRSA announced that a 60-day grace period would be applied to the ﬁrst reporting deadline,
potentially delaying the submission of provider reports to as late as November 30, 2021. HRSA
will not be able to use provider reports to determine whether funds were used appropriately until
provider reports are received.

Recovery of overpayments, unused payments, and payments not properly used. HRSA oﬃcials told
us that as of August 2021, they were in the process of planning to recover funds where they
identify payment discrepancies. At that time, the oﬃcials reported that they had completed
reviews for two of the 54 payment discrepancy types (representing 125 TINs) identiﬁed and
recovered about $2.9 million (about 20 percent) of $15.1 million in potential overpayments
identiﬁed for recovery.125 For another eight payment discrepancy types (representing 3,048 TINs),
agency documentation also identiﬁed an additional $356.4 million for recovery, but action to
recover those funds had not been taken as of August 2021.126 For the remaining 44 payment
discrepancy types, overpayments have not yet been identiﬁed for recovery. In addition, HRSA has
not yet identiﬁed unused payments or payments not properly used for recovery. In September
2021, HRSA oﬃcials told us that to more eﬃciently recover overpayments, they are planning to
oﬀset or reduce future PRF payments to be made at the end of 2021 by identiﬁed overpayment
amounts, rather than incurring administrative costs associated with recovering funds directly
from providers. However, some providers that received overpayments may not apply for future
PRF payments, which could delay recovery of funds. Documenting overpayment amounts
and beginning recovery eﬀorts as soon as possible will increase the likelihood of recovering
overpayments.

123Providers may have received other assistance from several sources, including the Department of the Treasury, the
Small Business Administration, the Federal Emergency Management Agency, the Paycheck Protection Program, and
local, state, and tribal government assistance sources.
124For the ﬁrst PRF payments disbursed from April 10 to June 30, 2020, providers must use the funds by June 30, 2021,
and report on the use of these funds no later than September 30, 2021. According to HRSA oﬃcials, the 60-day grace
period will be applied to this ﬁrst reporting deadline, and no compliance action will be initiated against providers who
submit reports by November 30, 2021.
125According to agency oﬃcials, the ﬁrst payment discrepancy type to be analyzed—renal dialysis providers—was the
pilot for the reviews and recovery.
126In August 2021, agency oﬃcials told us that these are being re-analyzed to account for additional payments made
and will go to recovery once the analysis is complete.

Page 97 GAO-22-105051
Without timely post-payment oversight to help ensure that relief payments are made only to
eligible providers in correct amounts, HHS cannot fully address its stated payment integrity risks
for the PRF and seek to recover overpayments. Similarly, unused payments or payments not
properly used, if not identiﬁed through post-payment oversight, are at risk of not being recovered.
Setting time frames for completion of these oversight eﬀorts can help the agency achieve its
objectives and increase the likelihood of recovering funds.

Methodology
To conduct our work, we examined publicly released HHS information, federal laws and agency
guidance, and obtained information from CMS and HRSA in the form of written responses to
questions, documents (including payment integrity oversight materials), and a dataset. Our review
of the data sources provides reasonable assurance of the data’s reliability. The Provider Relief
Fund dataset, which includes disbursements as of August 31, 2021, came from HRSA, which is the
only available source for the disbursement data. The allocation amounts and categories that were
provided by HRSA match publicly available information. CMS provided data on the current status
for loans and repayments under the COVID-19 Accelerated and Advance Payments Program, as of
September 7, 2021.

Agency Comments
We provided HHS and the Oﬃce of Management and Budget (OMB) with a draft of this enclosure.
HHS provided written comments, reproduced in Appendix IV and technical comments on this
enclosure, which we incorporated as appropriate. OMB did not provide comments on this
enclosure.

HHS partially concurred with both recommendations. HRSA stated that it has a schedule for
reviewing the payment discrepancy types it initially prioritized, and reviews for the remaining
discrepancy types will occur after HRSA completes review of the prioritized discrepancy types. In
addition, HRSA stated that recovery of payments not properly used cannot begin until after the
provider reporting grace period ends on November 30, 2021, and recovery of unused payments
cannot begin until January 1, 2022—another 30 days after the grace period ends. However, we
believe that review of reports and recovery could start earlier, since HRSA oﬃcials told us in
September 2021 that they were already receiving provider reports. Regardless, establishing time
frames for completing reviews of all payment discrepancy types and implementing recovery
eﬀorts expeditiously will help the agency succeed in recovering overpayments.

GAO’s Ongoing Work
As HHS works to distribute funds for COVID-19 relief activities and to eligible providers, it will
continue to be important that HHS oﬃcials ensure funds are appropriately distributed and used.
We plan to conduct additional work examining HHS’s COVID-19 relief funds, including payment
oversight and funds returned by providers.

Page 98 GAO-22-105051
GAO’s Prior Recommendation
The table below presents our recommendation related to a post-payment review process for the
Uninsured Program, funded from the PRF, from a prior quarterly report.

Page 99 GAO-22-105051
Prior GAO Recommendation Related to Department of Health and Human Services (HHS) Relief to
Providers
Recommendation Status

The Secretary of Health and Open—partially addressed. HHS agreed with our recommendation to ﬁnalize
Human Services should and implement a post-payment review process. In July 2021, HHS stated it is
ﬁnalize and implement a post- currently developing the post-payment review audit strategy for the Uninsured
payment review process to Program, which includes detailed protocol and procedures for the assessments
validate COVID-19 Uninsured of the Uninsured Program to be executed by audit contractors. While the
Program claims and to help Uninsured Program post-payment review strategy is being developed, HHS
ensure timely identiﬁcation of has also developed an interim process with standard operating procedures.
improper payments, including HHS oﬃcials added that all claims determined to have been paid to ineligible
those resulting from potential providers or providers that in any way did not comply with the program
fraudulent activity, and recovery terms and conditions, will be required to return the funds. We will continue
of overpayments. (March 2021 to monitor the implementation of this recommendation to ensure that these
report) eﬀorts continue.

Source GAO. I GAO-22-105051

Related GAO Products
Standards for Internal Control in the Federal Government, GAO-14-704G (Washington, D.C.:
September 2014)

Contact information: Carolyn L. Yocom, (202) 512-7114, yocomc@gao.gov

Page 100 GAO-22-105051
Veterans Health Care
Health disparities among the nation’s veteran population have been well documented during the
pandemic and before. Although addressing these disparities has been a goal of the Department of
Veterans Aﬀairs’ Veterans Health Administration for almost a decade, the agency continues to lack
performance measures to evaluate its eﬀorts, which we previously recommended it develop and
with which the Department of Veterans Aﬀairs agreed.

Entity involved: The Veterans Health Administration within the Department of Veterans Aﬀairs

Background
The Veterans Health Administration provides care to a diverse veteran population. The
Department of Veterans Aﬀairs’ (VA) Veterans Health Administration (VHA) operates the nation’s
largest integrated health care system with 171 VA medical centers and 1,112 community-based
outpatient clinics across the country. VHA provides health care nationwide to a diverse population
of enrolled veterans, including women, Black, American Indian or Alaska Native, Asian, Native
Hawaiian or Other Paciﬁc Islander, and Hispanic or Latino people. According to ﬁscal year 2017
data, the latest available data, females comprised about 9 percent of the 18.3 million total veteran
population. Additionally, Black, American Indian or Alaska Native, Asian, Native Hawaiian or Other
Paciﬁc Islander, and Hispanic or Latino veterans comprised about 23 percent of the total veteran
population and 35 percent of the total female veteran population.127

As the veteran population becomes increasingly more diverse, VHA has recognized the importance
of ensuring health equity. According to VHA, “health equity is the attainment of the highest level of
health for all people, and achieving health equity requires valuing everyone equally with focused
and ongoing societal eﬀorts to address avoidable inequalities, historical and contemporary
injustices and the elimination of health and health care inequities.” However, VHA has identiﬁed
racial and ethnic disparities in its health care outcomes, mirroring trends seen across the U.S. in
public and private health care systems.128 For example:

• In 2020, VHA reported sex-based disparities in some areas, such as immunization rates where
women veterans older than 65 had a 10 percent lower pneumococcal immunization rate than
men.129

• In 2019, VHA reported evidence of disparities in health care outcomes within VA medical
centers in the form of lower survival rates for Black veterans with cancer and cardiovascular-

127Less is known about the number of lesbian, gay, bisexual, and transgender (LGBT) veterans. For example, population
estimates for LGBT veterans are more than a decade old in some cases and based on non-VA data, such as a 2004
estimate using U.S. Census data that roughly one million veterans identiﬁed as lesbian or gay and a 2014 estimate that
more than 130,000 veterans identiﬁed as transgender.
128According to VHA, a health disparity is a particular type of health diﬀerence that is closely linked with social or
economic disadvantage. Health disparities adversely aﬀect groups of people who have systematically experienced
greater social and/or economic obstacles to health and/or a clean environment based on racial or ethnic groups; gender;
age; geographic location; religion; socio-economic status; sexual orientation; mental health; military era; or cognitive,
sensory, or physical disability.
129Veterans Health Administration, VHA Clinical Performance Measures Gender Disparity Update – FY 20 thru 2nd
Quarter (Washington, D.C.: 2020).

Page 101 GAO-22-105051
related illnesses compared with veterans from other racial and ethnic groups and White
veterans.130

• In 2021, VHA reported that from 2013-2018, female lesbian, gay, and bisexual veterans faced
depressive and anxiety symptoms at double the rate of heterosexual female veterans.131

• In 2011, VHA found outcomes for controlling blood pressure, blood glucose, and cholesterol
levels were signiﬁcantly worse for Black veterans than they were for White veterans.132

In 2012, VHA established its Oﬃce of Health Equity (OHE) to identify and address health
care outcome disparities and to develop an action plan to help achieve health equity. OHE is
responsible for several eﬀorts, including providing education, training, research, communications
and information; promoting common awareness about health care disparities and working to
improve health care outcomes; and serving as a liaison to support other governmental and non-
governmental organizations working to achieve health equity.

Overview of Key Issues
VHA data show health disparities among minority veterans for some but not all COVID-19
indicators, including health care services provided virtually during the pandemic. Although
there are issues with the completeness of data on race and ethnicity, VHA has found that health
disparities exist.133 For example, interviews with VHA oﬃcials and a review of agency studies based
on varying time frames throughout the pandemic, show that minority veterans, such as Blacks and
Hispanic or Latinos, experienced health disparities among COVID-19 cases, hospitalizations, and
death rates earlier in the pandemic, but there were improvements seen over time in cases and
death rates.134 According to VHA oﬃcials, COVID-19 pandemic conditions are constantly evolving
and VHA observes the data regularly to identify any patterns of concern.

130See M. S. Wong et al., “Racial/Ethnic Disparities in Mortality Across the Veterans Health Administration.” Health
Equity 3, no. 1 (2019): 99, 103, 104. Overall, the study found fewer disparities in the VHA health system than in the
general population.
131See J. McGirr, K. Jones, and E. Moy, Chartbook on the Health of Lesbian, Gay, & Bisexual Veterans, 2021, Oﬃce of
Health Equity, Veterans Health Administration, U.S. Department of Veterans Aﬀairs (Washington, D.C.).
132See A. Trivedi et al., “Despite Improved Quality of Care in the Veterans Aﬀairs Health System, Racial Disparity
Persists for Important Clinical Outcomes,” Health Aﬀairs 30, no. 4 (2011): 707, 712, 713.
133We have previously identiﬁed data quality issues regarding the completeness and accuracy of race and ethnicity data
in VHA’s electronic health record, which is similarly seen in COVID-19 data for the general population reported to the
Department of Health and Human Services.
134VHA studies and internal analyses included for the purposes of this analysis reﬂect data beginning in February 2020
and through February 2021. For the purposes of this review, minority refers to those who are non-White, such as Blacks
and Hispanics or Latinos. Study authors acknowledge limitations in interpreting outcomes for minority groups and sex-
based analyses due to the small numbers of these groups represented in the study populations. In addition, inclusion
in the cited studies of only those veterans who were tested for COVID-19 by a VHA facility may omit minority veterans
who were tested or treated outside VHA. While VHA’s analyses on health disparities during the COVID-19 pandemic were
primarily reported by race, ethnicity, and sex, VHA has completed studies examining other groups, such as veterans
with post-traumatic stress disorder (PTSD) and homeless veterans. For example, according to a study published by VHA
researchers in 2020, veterans suﬀering from PTSD show a greater use of COVID-19 testing but lower rates of positivity.

Page 102 GAO-22-105051
Testing. According to a 2020 study conducted by VHA researchers based on data from February
through July 2020, Black veterans were more likely to be tested than Hispanic or Latino veterans
and White veterans.135 A VHA oﬃcial said there were no observed diﬀerences in the levels of
COVID-19 testing by sex.

Cases. According to a study completed by VHA researchers in 2021 based on 2020 data, after
adjusting for age, sex, and prior diagnoses of COVID-risk factors, veteran groups such as Blacks,
Hispanics or Latinos, and American Indian/Alaska Natives had higher odds of testing positive for
COVID-19 when compared to White veterans, though those gaps had closed somewhat by the
fall of 2020.136 For example, the odds of Black veterans and Hispanic or Latino veterans testing
positive for COVID-19 were 2.32 and 2.24 times higher, respectively, when compared to White
veterans during the spring and summer of 2020. By the fall of 2020, rates had declined for all
groups, with only American Indian/Alaska Native veterans having substantially higher odds (1.33)
of testing positive for COVID-19 than White veterans, suggesting an improvement in this indicator
for some minority veterans.137

Hospitalizations. VHA examined COVID-19 related hospitalizations among veterans seeking VHA
health care in March through May 2020.138 After adjusting for racial and ethnic diﬀerences in age
and sex,

• Black veterans were more likely to be hospitalized than White veterans (38 and 26 percent
adjusted hospitalization rates, respectively);

• Hispanic or Latino veterans were also more likely to be hospitalized than White veterans (34
and 26 percent adjusted hospitalization rates, respectively).139

Conversely, female veterans tended to fare better than their male counterparts. Based on an
internal VHA analysis of hospitalization among COVID-19 positive veterans using VHA health care
based on 2020 data, after adjusting for sex and diﬀerences in age and underlying comorbidities,
male veterans had higher odds of hospitalization (about 22 percent) compared to female veterans
(about 10 percent). Among COVID-19-positive female veterans, there were no disparities in

Elevated COVID-19 testing rates among veterans with PTSD may reﬂect increased COVID-19 health concerns and/or
hypervigilance. Lower rates of COVID-19 test positivity among veterans with PTSD may reﬂect increased social isolation,
or overrepresentation in the tested population due to higher overall use of VHA services.
135C.T. Rentsch et al., “Patterns of COVID-19 Testing and Mortality by Race and Ethnicity among United States Veterans: A
Nationwide Cohort Study,” PLoS Med 17(9): e1003379 (2020).
136M.S.Wong et al. “Time Trends in Racial/Ethnic Diﬀerences in COVID-19 Infection and Mortality,” Int. J. Environ. Res.
Public Health, 2021, 18, 4848.
137According to VHA, the act of receiving a test represents a form of health care access. In addition, the likelihood of
testing positive may be more indicative of exposure risk, rather than an indicator of either health care access or equity.
138VHA cited the following study as the data source for this analysis: G.N. Ioannou et al. “Risk Factors for Hospitalization,
Mechanical Ventilation, or Death Among 10 131 US Veterans With SARS-CoV-2 Infection,” JAMA Network Open. 2020;
3(9):e2022310.
139According to VHA, this analysis was updated with data through August 5, 2020. The results showed lower
hospitalization rates for all groups. Speciﬁcally, the age- and sex-adjusted hospitalization rates were 26, 20, and 18
percent for Blacks, Hispanics or Latinos, and White veterans, respectively.

Page 103 GAO-22-105051
hospitalization rates for Black veterans or for Hispanic or Latino veterans when compared to White
veterans.

Deaths. According to a 2021 study by VHA researchers, after adjusting for age, sex, and prior
diagnoses of COVID-risk factors, in the spring of 2020, Black veterans had a higher adjusted case
fatality rate (12 percent) and other racial and ethnic veteran groups had similar case fatality rates
(10–13 percent) when compared to White veterans (10.2 percent).140 However, adjusted case
fatality rates declined from spring to summer of 2020 and were lower for American Indian/Alaska
Native veterans (10.7 percent), Asian veterans (10.5 percent) and Hispanic or Latino veterans (7.2
percent), although these groups still had greater case fatality than White veterans (5.7 percent).

The adjusted case fatality rate declined further for all groups from summer to fall of 2020 when
compared to White veterans in the same time period. The overall spring-to-fall 2020 decline was
greatest for Black veterans; in the fall, Black veterans had a lower adjusted case fatality rate (1.9
percent) compared to White veterans (2.5 percent), whereas other groups (2.4–3.3 percent) were
similar to White veterans.

Female veterans experienced lower rates of mortality due to COVID-19 than male veterans. For
example, based on an internal VHA analysis of mortality among COVID-19 positive veterans using
VHA health care based on 2020 data, after adjusting for sex, diﬀerences in age and underlying
comorbidities, compared to COVID-19 positive male veterans, female veterans had lower rates of
mortality (about 7 and 1 percent, respectively).

Vaccinations received within VHA. Certain minority veterans aged 65 and older were more likely to
receive COVID-19 vaccinations than their White counterparts.141 Speciﬁcally, according to VHA,
based on an internal analysis of veterans aged 65 or older who received COVID-19 vaccinations
through VHA from December 2020 through February 2021, Black, Hispanic or Latino, and Asian
veterans were 5, 19, and 39 percent more likely (respectively) than White veterans to have
received a COVID-19 vaccination during that time frame.142 The analysis also found that American
Indian/Alaska Native veterans were 18 percent less likely than White veterans to have received a
COVID-19 vaccination, but according to VHA, this ﬁnding was limited to American Indian/Alaska
Native veterans who resided in areas where they could potentially access vaccinations from Indian
Health Service facilities. According to VHA, as of September 1, 2021, female veterans were slightly
more likely (about 3 percent) than male veterans to receive a COVID-19 vaccination.

Virtual health care. According to a study by VHA researchers, by June 2020, 58 percent of VHA
health care was provided virtually (by phone only or video) compared to 14 percent prior to the

140M.S. Wong, “Racial/Ethnic Diﬀerences in COVID-19 Infection and Mortality,” Int. J. Environ. Res. Public Health, 18, 4848.
According to the Centers for Disease Control and Prevention, the case fatality rate is the proportion of persons with a
particular condition who die from that condition.
141According to VHA, capturing vaccination rates depends on those who were seeking care through VHA facilities. VHA
oﬃcials do not know system-wide who received vaccinations outside of VHA facilities.
142According to VHA oﬃcials, neither OHE nor the Quality Enhancement Research Initiative Partnered Evaluation Center
have conducted COVID-19 vaccination studies of other age groups. The Oﬃce of Analytics and Performance Integration
in the VHA Oﬃce of Quality and Patient Safety, along with Population Health and the Oﬃce of Information Technology/
Business Intelligence Service Line, track and report, both internally and publicly, on vaccinations provided by VHA. These
reports include gender, race, ethnicity, and age stratiﬁcations.

Page 104 GAO-22-105051
pandemic.143 Veteran patients with lower incomes, higher levels of service-connected disability,
and more chronic conditions were more likely to receive virtual care during the pandemic.

VHA actions to address and prevent disparities. According to VHA oﬃcials and relevant agency
documents, VHA has used research on the COVID-19 pandemic, as well as other pandemics to
inform the various actions it has taken to address and prevent disparities during the COVID-19
pandemic among veterans from diﬀerent racial and ethnic groups, including the following:144

VHA held a focus group to identify accessibility gaps related to the COVID-19 pandemic for veterans of
various races and ethnicities.145 The focus group provided suggestions for VHA such as,

• instilling conﬁdence in getting a COVID-19 vaccine by having veterans and leaders representing
diﬀerent racial and ethnic groups share their stories of receiving the COVID-19 vaccine;

• ensuring information communicated by VHA was factual and dispelled COVID-19
misinformation;

• addressing language barriers by translating communications; and

• communicating information about VHA and non-VHA COVID-19 resources through diverse
media outlets.146

VHA developed an equity dashboard. Through the dashboard, created in May 2020, VHA generated
and shared weekly reports to the Veterans Integrated Service Networks (VISN), which manage
and oversee VA medical facilities within a deﬁned geographic area. These reports allowed VISN
staﬀ to track and map positive COVID-19 test rates by demographic category; identify new
community case rates to help direct outreach eﬀorts; and track veteran vaccination rates by sex,
race, ethnicity, and rurality.

VHA developed and communicated COVID-19 information to veterans. According to VHA oﬃcials,
VHA used a number of ways to provide veterans with information about COVID-19 such as texts,
weekly newsletters, virtual events, blog posts, videos, podcasts, and through resources found on
its website.

143J.M. Ferguson et al.,“Virtual Care Expansion in the Veterans Health Administration during the COVID-19 Pandemic:
Clinical Services and Patient Characteristics Associated with Utilization,” Journal of the American Medical Informatics
Association, 28(3), 2021, 453–462. According to VHA, prior to the COVID-19 pandemic, Black and Hispanic or Latino
veterans lagged behind White veterans in terms of virtual health care use, suggesting that those gaps have been
mitigated during the COVID-19 pandemic.
144Health disparities for persons from various racial and ethnic groups were known prior to the pandemic; however,
some of the health disparity gaps became more apparent as the COVID-19 pandemic continued. VHA actions to address
and prevent disparities in COVID-19 indicators among veterans from diﬀerent racial and ethnic groups began in May
2020.
145The focus group was conducted in September 2020 and included 10 participants, both male and female, from
communities of color such as Black, Hispanic or Latino, American Indian, and Asian American/Paciﬁc Islander.
146According to VHA oﬃcials, creating trust around the COVID-19 vaccine has proven challenging for VHA,
particularly for racial and ethnic groups that have experienced unethical experimentation and behavior from the
medical community in the past.

Page 105 GAO-22-105051
Through these means, VHA addressed topics including concerns about the safety of the COVID-19
vaccine, dispelled COVID-19 vaccine misinformation, and detailed how to access COVID-19 testing
and services during the pandemic. For example, to provide information to women and Black
veterans, who had concerns about receiving the COVID-19 vaccine, VHA posted blog entries
written by Black VHA leaders and disseminated vaccine information though a podcast for women
veterans.

VHA also translated many of its communications such as VHA webpages and brochures on
COVID-19 vaccine information into Spanish to ensure information was accessible to Hispanic or
Latino veterans and their families.

VHA developed a social risks screening tool to be used when veterans make appointments. In May
2020, VA medical center staﬀ began using the tool. According to VHA, the screening tool may help
identify the risks of catching COVID-19 that Black and Hispanic or Latino people may experience
disproportionately; for example, research shows Black and Hispanic or Latino Americans are more
likely to hold jobs that are not amenable to social distancing that would put them at a higher risk
for contracting the COVID-19 virus.

To prioritize individuals for COVID-19 testing, screening questions on the tool inquire about a
veteran’s social risks, such as use of public transportation and living in overcrowded housing,
which veterans from some racial and ethnic groups may more likely experience. The screener
tool also asks questions on topics such as shopping during the pandemic, COVID-19 prevention
strategies, and social interactions. These questions may also be used separately to assess an
individual’s need for counselling or assistance to minimize COVID-19 exposure.

VHA developed resources for veterans with chronic medical conditions. In March 2021, VHA provided
educational material and information on how to access services during the pandemic through
brochures posted to its website. For example, a brochure for veterans exposed to airborne
hazards and open burn pits during deployment, who are at higher risk for conditions like cancer,
chronic kidney disease, and chronic obstructive pulmonary disease, provides information on how
to reﬁll prescriptions online and resources for setting up virtual health care appointments.147

VHA is expanding virtual care. To better target the expansion of virtual care to veterans from
diﬀerent racial and ethnic groups, VHA oﬃcials told us that they held four focus groups and one
follow-up interview in June through August of 2021. These focus groups included women veterans,
Black veterans, disabled veterans, and LGBT veterans. According to VHA, the focus groups were
a way for VHA to learn more about some of the barriers and assistance veterans experienced in
seeking or engaging in virtual care and to better understand priorities regarding virtual health
care services. VHA oﬃcials also told us that they are working with the Navajo Health Foundation
to complete a Memorandum of Agreement that will allow American Indian veterans to access VHA
virtual care services from Sage Memorial Hospital.148

147Additional VHA brochures include resources for Black, Hispanic or Latino, and American Indian or Alaska Native
veterans; veterans who are 65 years and older; and pregnant veterans, all of which are persons that may be at greater
risk for serious illness from COVID-19.
148The Navajo Health Foundation – Sage Memorial Hospital, Inc. is a private, nonproﬁt corporation. It is the ﬁrst
Native-managed private comprehensive health care system in the country and has been managed since 1978 by an
independent, entirely Navajo Board of Directors.

Page 106 GAO-22-105051
According to VHA oﬃcials, VHA has funded evaluation projects in ﬁscal years 2020 and 2021 that
address health equity issues related to virtual care and expect results from these evaluations
in ﬁscal year 2022. These studies broadly examine telehealth use by VHA during the pandemic
and range from assessing health disparities in telehealth use, broadband access and telehealth
utilization, quality of care, and access to resources bridging the digital divide.

VHA continues to lack performance measures to evaluate its actions to address health
disparities, including for the COVID-19 response. OHE lacks performance measures to
assess the eﬀectiveness of its various COVID-19 response eﬀorts to address and prevent health
disparities among veterans from diﬀerent racial and ethnic groups. OHE oﬃcials told us that
they measure the eﬀectiveness of their actions by tracking and analyzing data on a variety of key
indicators during the COVID-19 pandemic, such as cases and vaccination rates. VHA oﬃcials noted
that OHE uses data on key indicators as a way to assess the eﬀectiveness of its various COVID-19
response eﬀorts to address and prevent health disparities among veterans from diﬀerent racial
and ethnic groups. According to OHE, when staﬀ observe changes in the data indicating that
disparities are decreasing, this indicates that their actions are working.

Speciﬁc and measurable performance indicators for agency actions could help the oﬃce more
accurately determine if any changes in data related to health disparities can be attributed to
agency actions or other external inﬂuences. For the COVID-19 pandemic, changes in disparities
data during the pandemic could be due to VA eﬀorts or could also be due to factors, such as
national, state, local, or other non-VHA response eﬀorts. VHA oﬃcials told us that unlike national
COVID-19 vaccination rates, within VHA, Black and Hispanic or Latino veterans are vaccinated at
higher rates than White veterans. However, as previously noted, Black and Hispanic or Latino
veterans, experienced health disparities among COVID-19 cases, hospitalizations, and death rates,
and those disparities are also generally observed in the national population.

Our prior work on eﬀectively managing performance shows that performance measures should
assess how well the organization is meeting its goals and should be linked directly to oﬃces that
have responsibility for the program or activity. For example, as part of its goal to promote better
health outcomes for racial and ethnic populations, VHA’s OHE is responsible for spreading the use
of its COVID-19 equity dashboard among VHA’s 18 VISNs. However, OHE has no way of measuring
how eﬀective this eﬀort has been in promoting better health outcomes for racial and ethnic
populations.149

Additionally, OHE’s communication plan to promote awareness about health disparities states that
it should regularly update OHE website with new information briefs, newsletters, Cyberseminar
announcements, and other relevant publications and updates. While OHE has made COVID-19
speciﬁc updates to its website and has communicated information on COVID-19 though
newsletters and other media outlets, it has no way of determining how successful these actions
have been in raising awareness about health disparities during the pandemic.

In December 2019, we found that OHE’s Health Equity Action Plan—VHA’s action plan to
address health equity across the agency—lacked performance measures to assess progress

149According to VHA, measuring the eﬀectiveness of a single eﬀort may not be possible when many VHA eﬀorts work
together to promote better health outcomes for racial and ethnic populations.

Page 107 GAO-22-105051
and we recommended that its action plan include such measures. VHA concurred with our
recommendation and told us they plan to add performance measures to its Health Equity Action
Plan. However, as of the date of this report, the recommendation remains unimplemented. It
is important that VHA implement performance measures concerning its actions to address and
prevent health dipartites among veterans. Until VHA implements these performance measures,
VHA runs the risk of not knowing the eﬀectiveness of its eﬀorts for the COVID-19 pandemic or
pandemics or public health emergencies that may occur in the future.

Methodology
To conduct this work, we reviewed VHA studies published in peer-reviewed journals and other
internal analyses identiﬁed by VHA that examined the presence of disparities in COVID-19
indicators among veterans from diﬀerent racial and ethnic groups.150 We also reviewed VHA
guidance and documents used to address and respond to health disparities among veterans.
In addition, we interviewed and reviewed written responses from oﬃcials in VHA’s OHE and its
Oﬃce of Women’s Health about identiﬁed disparities in COVID-19 indictors among veterans from
diﬀerent racial and ethnic groups, actions to address these disparities, and ways VHA is measuring
the eﬀectiveness of its actions.

Agency Comments
We provided VA and the Oﬃce of Management and Budget with a draft of this enclosure. VA
concurred with our ﬁndings and provided technical comments, which we incorporated as
appropriate. The Oﬃce of Management and Budget did not provide comments on this enclosure.

GAO’s Ongoing Work
We will continue to monitor racial and ethnic health disparities during the COVID-19 pandemic,
including as they relate to the provision of equitable access to health care.

Related GAO Products
VA Health Care: Better Data Needed to Assess the Health Outcomes of Lesbian, Gay, Bisexual, and
Transgender Veterans, GAO-21-69 (Washington, D.C.: October 19, 2020).

VA Health Care: Opportunities Exist for VA to Better Identify and Address Racial and Ethnic Disparities,
GAO-20-83 (Washington, D.C.: December 11, 2019).

150We have previously identiﬁed data quality issues regarding the completeness and accuracy of race and ethnicity data
in VHA’s electronic health record, which is similarly seen in COVID-19 data for the general population reported to the
Department of Health and Human Services. However, for the purposes of this review, we determined that the studies we
cite that use the incomplete data are suﬃciently reliable for the purpose of discussing the observed health disparities,
while acknowledging the incomplete data.

Page 108 GAO-22-105051
VA Health Care: Improved Monitoring Needed for Eﬀective Oversight of Care for Women Veterans,
GAO-17-52 (Washington, D.C.: December 2, 2016).

Contact information: Sharon M. Silas, (202) 512-7114, silass@gao.gov; Alyssa M. Hundrup,
hundrupa@gao.gov, (202) 512-7114

Page 109 GAO-22-105051
HHS Cybersecurity
The U.S. National Institutes of Health has not consistently implemented security controls in its
information security program or selected information technology systems that receive, process,
and maintain sensitive information, putting conﬁdentiality, integrity, and availability of information
at risk. GAO has made numerous recommendations to Department of Health and Human
Services component agencies to improve information security. Those component agencies have
implemented, or are in the process of implementing, the recommendations.

Entity involved: U.S. National Institutes of Health, within the Department of Health and Human
Services

Background
The U.S. National Institutes of Health’s (NIH) responsibilities include conducting research on the
prevention of infectious diseases (such as COVID-19), administering over $30 billion annually in
medical research grants, and supporting research on pathogens, including those that have the
potential to pose a severe threat to public health and safety. In carrying out its mission, NIH relies
extensively on information technology systems to receive, process, and maintain sensitive data.
Accordingly, eﬀective information security controls are essential to ensure the conﬁdentiality,
integrity, and availability of the information on the agency’s systems.

Overview of Key Issues
During the course of a prior audit we conducted from January 2019 to June 2021, we found that
NIH implemented numerous security controls within its information security program and over
the 11 systems we reviewed across four NIH entities. These controls included, among other things,
taking steps to develop security plans, ensuring that the majority of personnel had basic security
awareness training, and developing remedial action plans for correcting deﬁciencies.

However, the agency had not always eﬀectively implemented other controls—both within its
information security program and for the selected systems—to protect the conﬁdentiality,
integrity, and availability of these systems and the information maintained on them. Deﬁciencies
existed in some of the controls intended to identify risks, protect systems from threats and
vulnerabilities, detect and respond to cybersecurity events, and recover system operations in
cases of system disruptions. As a result, NIH was at increased risk that sensitive research and
health-related information could be disclosed or disrupted.

In June 2021, we issued a report with limited distribution because of the sensitive information it
contained. In that report, we made 219 recommendations to NIH, including:

• 66 to improve NIH’s information security program by, among other things, assessing risks,
as needed; documenting complete and accurate security controls; assessing controls more
comprehensively; and remediating deﬁciencies in a timely manner; and

• 153 to resolve system control deﬁciencies by implementing stronger access controls,
encrypting sensitive data, conﬁguring devices securely, applying patches in a timely manner,

Page 110 GAO-22-105051
strengthening ﬁrewall rules, improving incident response, and implementing monitoring
controls more eﬀectively, among other things.

NIH concurred with the recommendations to improve its information security program.
Additionally, NIH agreed to implement 148 of the 153 recommendations to resolve system
control deﬁciencies, and disagreed with the remaining ﬁve recommendations for various
reasons. However, we believe these ﬁve recommendations are warranted in order to further
improve the security over NIH’s systems. The table below shows the number of deﬁciencies and
recommendations for NIH’s information security program and system control deﬁciencies across
the core security functions of identify, protect, detect, respond, and recover.

Page 111 GAO-22-105051
Number of GAO-Identiﬁed Information Security Program and System Control Deﬁciencies at the U.S. National
Institutes of Health and Associated Recommendations by Core Security Function
Number of
Number of Number of information Number of selected selected system
Core security information security security program system control control deﬁciency
function program deﬁciencies recommendations deﬁciencies recommendations

Identify 12 26 0 0

Protect 4 6 78 141

Detect 5 11 5 11

Respond 7 16 1 1

Recover 4 7 0 0

Total 32 66 84 153

Source: GAO. | GAO-22-105051

In commenting on our June 2021 report, NIH stated that it had taken corrective actions to address
many of the deﬁciencies identiﬁed. We plan to issue a public report that provides the results of our
determination of the status of the agency’s corrective actions in addressing our recommendations
later in 2021 or early in 2022. Until these recommendations are addressed, NIH’s systems, and the
information maintained in those systems, are at increased risk of unauthorized access.

Methodology
To conduct this work, we selected four entities for review from the agency’s 28 institutes, centers,
and the director’s oﬃce. Our selection focused on entities that provide information technology
and security for the agency and are essential to NIH’s mission. From the four entities, we selected
11 systems for review that, for example, (1) collect, process, and maintain private or potentially
sensitive proprietary business, medical, and personally identiﬁable information; (2) are essential to
NIH’s mission; (3) could have a catastrophic or severe impact on operations if compromised; and/
or (4) share some common infrastructure.

To assess NIH’s information security program, we examined security policies, procedures, and
other documents; compared selected systems’ risk assessments, security plans, security control
assessments, remedial action plans, and contingency plans to federal guidance; and interviewed
personnel at NIH entities. To assess controls over the 11 selected systems, we reviewed the
agency’s network infrastructure and assessed the extent to which controls associated with system
access, encryption, conﬁguration management, and monitoring met federal guidance and industry
best practices.

Page 112 GAO-22-105051
Agency Comments
We provided NIH, the Department of Health and Human Services, and the Oﬃce of Management
and Budget with a draft of this enclosure. NIH provided technical comments, which we
incorporated as appropriate. The Oﬃce of Management and Budget did not provide comments on
this enclosure.

GAO’s Ongoing Work
We plan to issue a public report that describes the ﬁndings discussed in our June 2021 report,
with references to sensitive information removed. In addition, we will report on the status of
NIH’s actions to implement our recommendations related to improving its security program and
resolving system control deﬁciencies.

GAO’s Prior Recommendations
The table below presents our recommendation on cybersecurity from the September 2020 CARES
Act report.

Page 113 GAO-22-105051
Prior GAO Recommendation Related to Cybersecurity
Recommendation Status

To ensure Health and Human Services (HHS) component agencies Open-partially addressed. HHS concurred with
involved in supporting the critical health care infrastructure this recommendation. The Food and Drug
and systems responding to COVID-19 are protected from Administration, Centers for Medicare and Medicaid
cybersecurity threats, we recommended that HHS expedite Services, and Centers for Disease Control and
the implementation of our prior recommendations to address Prevention have implemented an additional
cybersecurity weaknesses at its component agencies. (September 71 cybersecurity recommendations since the
2020) September 2020 CARES Act report. This brings
the total number of implemented cybersecurity
recommendations to 421 (of 434) from September
2020 through July 2021—a 16 percent increase in
the number of corrective actions taken to bolster
cybersecurity at these agencies.

Source: GAO. I GAO-22-105051

Contact information: Jennifer R. Franks, (404) 679-1831, franksj@gao.gov

Page 114 GAO-22-105051
Worker Safety and Health
The Occupational Safety and Health Administration should conduct an analysis of the challenges it
has faced to ensuring worker safety during the pandemic to improve its response to this pandemic
and prepare for a future one.

Entity involved: Occupational Safety and Health Administration, within the Department of Labor

Recommendation for Executive Action
The Assistant Secretary of Labor for Occupational Safety and Health should assess—as soon as
feasible and, as appropriate, periodically thereafter—various challenges related to resources and
to communication and guidance that the Occupational Safety and Health Administration has faced
in its response to the COVID-19 pandemic and should take related actions as warranted.

The Department of Labor partially agreed with our recommendation. The Department of Labor
stated that it agrees that it is important to assess lessons learned and best practices for the
Occupational Safety and Health Administration’s operational response to COVID-19. However,
Department of Labor oﬃcials said they believe that while the pandemic is ongoing, the agency’s
resources are best used to help employers and workers mitigate exposures to COVID-19. Because
it is unclear when the COVID-19 pandemic will end, we maintain that assessing—as soon as
feasible—the challenges that the Occupational Safety and Health Administration faced in
responding to the pandemic, and taking related actions, would enable the agency to improve its
enforcement eﬀorts during this pandemic and help it prepare for operations during any future
pandemic.

Background
The Department of Labor’s (DOL) Occupational Safety and Health Administration (OSHA) helps
ensure safe and healthful conditions for workers by setting mandatory workplace safety and
health standards; conducting inspections; investigating complaints and reports of injuries,
illnesses, and fatalities at workplaces; and oﬀering training, guidance, and outreach; among other
eﬀorts.151

OSHA has 10 regional oﬃces and 89 area oﬃces that implement and oversee enforcement in the
ﬁeld.152 OSHA is responsible for setting and enforcing workplace safety and health standards for
the private sector in 29 states, the District of Columbia, and four territories.153 Twenty-one states

151OSHA carries out these activities under the Occupational Safety and Health Act of 1970 (OSH Act), Pub. L. No. 91-596,
84 Stat.1590 (codiﬁed as amended at 29 U.S.C. §§ 553, 651-678).
152OSHA also has four district oﬃces that are subordinate to an area oﬃce.
153In ﬁve of these states and the U.S. Virgin Islands, the state or territory is responsible for setting and enforcing
standards for state and local government employers, under a state plan approved by OSHA.

Page 115 GAO-22-105051
and Puerto Rico set and enforce their own workplace safety and health standards for private
sector and state and local government employers under state plans approved by OSHA.154

OSHA has almost 1,900 employees, and its appropriation for ﬁscal year 2021 was approximately
$592 million. OSHA received $105.8 million in additional funding under the CARES Act and the
American Rescue Plan Act of 2021.155

Overview of Key Issues
OSHA conducted enforcement using existing worker safety standards during the pandemic.
During the ﬁrst 15 months of the pandemic, OSHA primarily relied on existing standards and
voluntary employer guidance to conduct its enforcement. In March 2021, OSHA started a
COVID-19 National Emphasis Program (NEP) to target its inspections on both health-care and
non-health-care industries with high risk of worker exposure to COVID-19. In June 2021, OSHA
issued an emergency temporary standard (ETS) related to COVID-19 exposure for the health-care
industry. See ﬁgure for a summary of key OSHA actions during the COVID-19 pandemic.

154State standards and their enforcement must be at least as eﬀective as the federal standards in protecting workers
and in preventing work-related injuries, illnesses, and fatalities. Federal agencies are generally responsible for
maintaining their own occupational safety and health programs, consistent with OSHA’s regulations.
155The CARES Act appropriated $15 million to DOL for “Departmental Management,” to remain available through
September 30, 2022, to prevent, prepare for, and respond to the COVID-19 pandemic, including to enforce worker
protection laws and regulations, among other things. Speciﬁcally, the CARES Act authorized the Secretary of Labor
to transfer the amounts provided under this heading as necessary to OSHA and certain other DOL components, to
prevent, prepare for, and respond to COVID-19, including for enforcement, oversight, and coordination activities in those
accounts. Pub. L. No. 116-136, div. B, tit. VIII, 134 Stat. 281, 553-554 (2020). DOL oﬃcials said the department transferred
$5.5 million of this amount to OSHA. As of September 30, 2021, $4.2 million has been obligated and, of that, $3.1 million
has been expended, according to OSHA oﬃcials. The American Rescue Plan Act of 2021 (ARPA) appropriated $200
million to DOL to remain available until September 30, 2023. Pub. L. No. 117-2, tit. II, sub. B, 135 Stat. 4, 30 (2021). OSHA
oﬃcials said the department provided $100.3 million of this amount to OSHA. As of September 30, 2021, according
to OSHA oﬃcials, $35.5 million has been obligated for COVID-19-related activity (including $12.8 million for federal
enforcement), of which $15.5 million has been expended (including $11.2 million for federal enforcement).

Page 116 GAO-22-105051
Timeline of OSHA’s Key Actions to Respond to the COVID-19 Pandemic, from January 2020 through September
2021

a
OSHA published updated versions of its Interim Enforcement Response Plan for COVID-19 on May 19, 2020, March 12, 2021,
and July 7, 2021.
b
OSHA published updates to this employer guidance on June 10, 2021 and August 13, 2021.
c
OSHA published an update to its COVID-19 National Emphasis Program on July 7, 2021.

OSHA has recorded data related to COVID-19 in the workplace since February 2020. From February
2020 through August 2021, related to COVID-19, OSHA received 16,667 complaints and referrals,
1,678 employer reports of severe injuries or illnesses, 1,225 reports of fatalities, and 3 reports of
catastrophes.156 During the same time period, OSHA conducted 16,820 informal inquiries, 1,621

156Complaints refer to reports notifying OSHA of alleged workplace safety or health hazards. Complaints can be made
by employees, their representatives, or others. Referrals and employer reports: OSHA uses the term “referrals” to
encompass two diﬀerent report types, (1) reports of work-related severe injuries and illnesses, which employers are

Page 117 GAO-22-105051
on-site inspections, and 1,190 remote inspections related to COVID-19.157 As a result of these
inspections, OSHA cited 917 violations and issued about $6.4 million in penalties.158 (See table.)

required to submit to OSHA (which OSHA calls employer-reported referrals); and (2) reports of potential workplace
hazards from selected other entities, such as local government agencies or media outlets. In this report, we use
“referrals” to describe those reports from selected non-employer sources, and “employer reports” to describe those
reports from employers. Employers are required to report all work-related in-patient hospitalizations, amputations, and
losses of an eye within 24 hours. 29 C.F.R. § 1904.39(a)(2). Fatalities: Employers are required to report the work-related
death of an employee to OSHA within 8 hours. 29 C.F.R. § 1904.39(a)(1). According to OSHA oﬃcials, most reports of
fatalities come from employers. However, oﬃcials noted that they do receive reports of fatalities from other sources,
such as the media or emergency medical personnel. In this report we refer to all reported fatalities as “reports of
fatalities.” Catastrophes: OSHA’s Field Operations Manual deﬁnes a catastrophe as the hospitalization of three or more
employees resulting from a work-related incident or exposure. Data throughout this report include enforcement activity
performed by OSHA only, and not by state agencies that operate under OSHA-approved state plans.
157An informal inquiry is a process conducted in response to a complaint, referral, or employer report of severe
injury or illness that does not meet OSHA’s criteria for conducting an inspection. According to OSHA oﬃcials, informal
inquiries conducted in response to an employer-reported severe injury or illness, such as an in-patient hospitalization
or amputation, are called rapid response investigations, and informal inquiries conducted in response to complaints
from employees or referrals from entities other than employers are called phone/fax investigations. According to OSHA’s
Field Operations Manual, if Area Directors consider employers’ responses to these informal inquiries to be inadequate,
they may decide to initiate a related inspection. During the COVID-19 pandemic, OSHA’s pandemic-related enforcement
policies have allowed area oﬃces to conduct some inspections remotely, instead of being physically at the workplace.
According to OSHA policy, data on remote inspections include only those inspections that were conducted completely
oﬀ-site.
158OSHA assesses ﬁnancial penalties for violations based on various factors outlined in statute and OSHA policy.
For example, after January 15, 2021, violations determined to be serious were subject to penalties of up to $13,653
per violation, and violations determined to be willful or repeated were subject to penalties of up to $136,532 per
violation. See 29 C.F.R. § 1903.15(d). Some of these cases are still open and may have been contested or appealed by
the employers, which could ultimately result in changes to the violations cited or penalties issued. In addition, under the
OSH Act, OSHA has 6 months from the occurrence of a violation to issue a citation and any related penalties, so totals
for the number of violations cited and penalties issued from March 2021 through August 2021 may not reﬂect the total
that will eventually be cited or issued related to inspections initiated during those months. These data are current as of
September 7, 2021.

Page 118 GAO-22-105051
COVID-19-Related Reports to OSHA and OSHA Enforcement Actions, March 2020 through August 2021

June- June-
March-May August September- December 2020- March-May August
a a a
2020 2020 November 2020 February 2021 2021 2021 Total

Reports to
OSHA

Complaints 4,843 3,757 2,579 2,708 1,213 750 15,850

Referrals 148 134 77 348 59 37 803

Employer 507 387 222 273 66 223 1,678
reports of
severe injury
b
or illness

Reports of 369 325 117 237 82 95 1,225
b
fatalities

Reports of 2 0 0 1 0 0 3
b
catastrophes

OSHA
enforcement
actions

On-site 114 219 150 210 500 427 1,620
inspections

Remote 338 369 139 264 61 19 1,190
inspections

Violations 225 272 105 210 95 10 917
c
cited

Penalties ($ 1,616 1,740 731 1,634 634 41 6,396
c
thousands)

Source: GAO analysis of Occupational Safety and Health Administration (OSHA) Information System data. | GAO-22-105051

Notes: Complaints refer to reports notifying OSHA of alleged workplace safety or health hazards. Complaints can be made by
employees, their representatives, or others.
Referrals and employer reports: OSHA uses the term “referrals” to encompass two diﬀerent report types, (1) reports of work-
related severe injuries and illnesses, which employers are required to submit to OSHA (which OSHA calls employer-reported
referrals); and (2) reports of potential workplace hazards from selected other entities, such as local government agencies or
media outlets. In this report, we use “referrals” to describe those reports from selected non-employer sources, and “employer
reports” to describe those reports from employers. Employers are required to report all work-related in-patient hospitalizations,
amputations, and losses of an eye within 24 hours. 29 C.F.R. § 1904.39(a)(2).
Fatalities: Employers are required to report the work-related death of an employee to OSHA within 8 hours. 29 C.F.R. §
1904.39(a)(1). According to OSHA oﬃcials, most reports of fatalities come from employers. However, oﬃcials noted that they do
receive reports of fatalities from other sources, such as the media or emergency medical personnel. In this report we refer to all
reported fatalities as “reports of fatalities.”
Catastrophes: OSHA’s Field Operations Manual deﬁnes a catastrophe as the hospitalization of three or more employees
resulting from a work-related incident or exposure.

Page 119 GAO-22-105051
During the COVID-19 pandemic, OSHA’s pandemic-related enforcement policies have allowed area oﬃces to conduct some
inspections remotely, instead of being physically at the workplace. According to OSHA policy, data on remote inspections
include only those inspections that were conducted completely oﬀ-site. Data in this table include enforcement activity
performed by OSHA only, and not by state agencies that operate under OSHA-approved state plans.
a
Since OSHA has 6 months from the occurrence of a violation to issue a citation and any related penalties, totals for the
number of violations cited and penalties issued from March 2021 through August 2021 may not reﬂect the total that will
eventually be cited or issued related to inspections initiated during those months. These data are current as of September 7,
2021.
b
Data reliability issues regarding COVID-19-related employer reports, speciﬁcally reports of hospitalizations, were discussed in
an enclosure to our January 2021 CARES Act report and are summarized elsewhere in the current enclosure.
c
Some of these cases are still open and may have been contested or appealed by the employers, which could ultimately result
in changes to the violations cited or penalties issued.

From February 2020 through May 2021, without COVID-19-speciﬁc standards in place, OSHA
enforced existing applicable standards, such as those related to respiratory protection, and
issued general and industry-speciﬁc voluntary guidance for employers on COVID-19-related
precautions.159 OSHA also occasionally used the “general duty clause” for enforcement. According
to OSHA’s Field Operations Manual, if hazards not covered by an OSHA standard are discovered
during an inspection, a general duty clause violation may be cited, if certain criteria are met.160
From February 2020 through August 2021, OSHA cited 20 COVID-19-related general duty clause
violations (see table).

159Beginning in March 2020, OSHA issued a variety of COVID-19 voluntary guidance and safety tips for employers,
including Protecting Workers: Guidance on Preparing Workplaces for COVID-19 in March 2020, supplemented with Guidance
on Returning to Work in June 2020. During 2020, OSHA and the Centers for Disease Control and Prevention issued
joint voluntary employer guidance on protecting workers in speciﬁc industries, such as farmworkers, manufacturing,
meat and poultry processing, and seafood processing. OSHA published Guidance on Mitigating and Preventing the
Spread of COVID-19 in the Workplace on January 29, 2021, as directed by the President’s Executive Order on Protecting
Worker Health and Safety. According to OSHA oﬃcials, this guidance updated the earlier employer guidance, based
on knowledge of the current state of the pandemic, and included input from multiple stakeholders on COVID-19
prevention measures and their feasibility. Although the new guidance did not provide new required standards for
employers to follow, it reaﬃrmed that employers have an obligation to protect workers under the OSH Act and that a
“general duty clause” violation could otherwise be cited. The new guidance also provided example abatement measures
for fulﬁlling this obligation. OSHA updated this guidance on June 10, 2021, to focus on protecting unvaccinated or
otherwise at-risk workers in the workplace. The updated guidance states that most employers no longer need to take
steps to protect their fully vaccinated workers who are not otherwise at risk from COVID-19 exposure. The update also
recommended steps to encourage workers to get vaccinated, including paid time oﬀ for employees to receive their
COVID-19 vaccination. OSHA further updated the voluntary employer guidance on August 13, 2021, including to reﬂect
the July 27, 2021 Centers for Disease Control and Prevention mask and testing recommendations for fully vaccinated
individuals.
160The “general duty clause” requires employers to provide a workplace free from recognized hazards that are causing
or are likely to cause death or serious physical harm to their employees. 29 U.S.C. § 654(a)(1). A general duty clause
violation may only be cited if all four of the following required elements are identiﬁed: (1) the employer failed to keep
the workplace free of a hazard to which employees of that employer were exposed; (2) the hazard was recognized; (3)
the hazard was causing or was likely to cause death or serious physical harm; and, (4) there was a feasible and useful
method to correct the hazard. See OSHA Field Operations Manual, CPL 02-00-164, Ch. 4, Sec. III, A. OSHA, under 29 U.S.C.
§ 658(c), must issue a citation for a violation, whether it be a violation of the general duty clause or a violation of speciﬁc
rules or standards, within 6 months of the date the violation occurred.

Page 120 GAO-22-105051
COVID-19-Related Violations and Penalties, for Most Frequently Cited Standards and the General Duty Clause,
from February 2020 through August 2021
Internal Reporting
employer to OSHA Provision of
recordkeeping fatalities requested
General for fatalities, and severe Personal records to
duty Respiratory injuries, and injuries and protective government Hazard
b c d e f g
clause
a protection illnesses illnesses equipmentrepresentatives
communication

Violations 20 576 128 101 30 15 15
h
cited

Penalties 471 4,900 214 681 70 16 9
($
h
thousands)

Source: Occupational Safety and Health Administration (OSHA) Information System data. | GAO-22-105051

a
The general duty clause requires employers to provide a workplace free from recognized hazards that are causing or are likely
to cause death or serious physical harm to their employees. The general duty clause is a part of the Occupational Safety and
Health Act of 1970, as amended, and is distinct from standards, which OSHA promulgates under the OSH Act. The general duty
clause is used when no standard applies to a particular hazard. See 29 U.S.C. § 654(a)(1).
b
29 C.F.R. § 1910.134 generally requires employers to provide respiratory protection to employees when necessary to protect
employee health.
c
29 C.F.R. § 1904.4 generally requires employers to keep an internal record of all work-related fatalities, injuries, and illnesses.
d
29 C.F.R. § 1904.39 generally requires employers to report to OSHA all work-related in-patient hospitalizations, amputations,
and losses of an eye within 24 hours, and all work-related fatalities within 8 hours.
e
29 C.F.R. § 1910.132 generally requires employers to provide personal protective equipment to employees when necessary,
such as for eyes, face, and head.
f
29 C.F.R. § 1904.40 generally requires employers to provide records to government representatives within 4 business hours of
a request.
g
29 C.F.R. § 1910.1200 generally requires employers to provide employees information concerning chemical hazards.
h
Some of these cases are still open and may have been contested or appealed by the employers, which could ultimately result
in changes to the violations cited or penalties issued. Since OSHA has 6 months from the occurrence of a violation to issue a
citation and any related penalties, totals for the number of violations cited and penalties issued may not reﬂect the total that
will eventually be cited or issued. These data are current as of September 7, 2021.

OSHA inspectors faced some challenges applying OSHA requirements to COVID-19 cases.
OSHA standards existing prior to OSHA’s June 2021 emergency temporary standard (ETS) for
certain health-care employers do not contain provisions speciﬁcally targeted at the COVID-19
hazard. As a result, it has been diﬃcult for employers and employees to determine what particular
COVID-19 safety measures are required, or how existing standards are expected to work when
applied to COVID-19. Moreover, according to the preamble to OSHA’s health-care ETS, OSHA’s
eﬀorts to enforce existing standards to address the COVID-19 hazard have been hindered by the
absence of any speciﬁc requirements in these standards related to some of the most important
COVID-19-mitigation measures. OSHA inspectors or managers from three of ﬁve area oﬃces
we spoke with said that it was diﬃcult to apply existing OSHA standards to COVID-19 cases, for
example, because existing standards did not cover certain COVID-19 hazard mitigations, such as
wearing a face covering.

Page 121 GAO-22-105051
In addition, although not unique to COVID-19 inspections, violations of the general duty clause
were challenging to cite since such violations require a large amount of documentation to
demonstrate that all four elements required to use the clause are present.161 According to the
preamble to OSHA’s health-care ETS, the general duty clause does not provide employers with
speciﬁc requirements to follow or a road map for implementing appropriate COVID-19 abatement
measures. In addition, OSHA’s burden of proof to establish a general duty clause violation is
heavier than a standard violation.

Inspectors from one area oﬃce told us that they did not have enough knowledge to determine
what should be considered a dangerous level of COVID-19 exposure or risk in order to cite a
related violation. Inspectors or managers from four of ﬁve area oﬃces we spoke with said it was
diﬃcult to apply the general duty clause to COVID-19-related hazards, for example, because it
would likely only be cited if an employer was making no eﬀort to use any COVID-19 mitigation
strategies.

According to the preamble to OSHA’s health-care ETS, in many cases during the pandemic,
inspectors found that employers were following some minimal COVID-19 mitigation strategy, while
ignoring other crucial components of employee protection.162 The preamble further notes that,
in such instances, because the employer had taken some steps to protect workers, successfully
proving a general duty clause citation would have required OSHA to show that additional missing
measures would have further materially reduced the COVID-19 hazard.

OSHA’s COVID-19 National Emphasis Program (NEP) is focused on high-risk industries and
uses data that are incomplete. In March 2021, OSHA initiated a 1-year COVID-19 NEP that aims
to ensure that employees in high-risk industries are protected from the hazard of COVID-19.163
The NEP includes a speciﬁc focus to ensure that workers are protected from retaliation.164

161We reported in April 2012 that violations of the general duty clause are challenging for OSHA to cite. According to
OSHA oﬃcials, using the general duty clause requires signiﬁcant agency resources so it is not always a viable option, for
example when OSHA cannot prove an employer knows the hazard exists or when a hazard is just emerging.
162The health-care ETS did not provide a speciﬁc example of such a minimal mitigation strategy compared to other
crucial components of employee protection.
163OSHA’s National Emphasis Programs provide for programmed inspections of establishments in industries with the
potential for high injury or illness rates, and are intended to focus outreach eﬀorts and inspections on speciﬁc workplace
hazards, according to the agency’s Field Operations Manual. In January 2021, President Biden signed Executive Order
13999 on Protecting Worker Health and Safety which, among other things, directed OSHA to initiate the COVID-19
National Emphasis Program (NEP) which identiﬁed high-risk industries.
164OSHA’s National Emphasis Program’s anti-retaliation focus includes, among other things, distributing anti-retaliation
information during inspections and promptly referring allegations of retaliation to OSHA’s Whistleblower Protection
Program, under which OSHA enforces protections for employees who ﬁle a complaint or otherwise engage in protected
activity under the federal laws over which OSHA has jurisdiction. See, e.g., 29 U.S.C. § 660(c)(1). According to Department
of Labor (DOL) data on whistleblower reports to OSHA—not including state agencies that operate under OSHA-
approved state plans—for the 20-month period from February 2020 through September 2021, the program received
19,800 whistleblower complaints, of which 5,816 were related to COVID-19. The total number of complaints was
29 percent higher than during the prepandemic 20-month period from June 2018 through January 2020, indicating
a signiﬁcant increase in whistleblower complaints during the pandemic. In August 2020, DOL’s Oﬃce of Inspector
General (OIG) recommended that OSHA (1) ﬁll ﬁve open whistleblower investigator positions; (2) monitor, evaluate,
and consider expanding to all regions a process to expedite screening whistleblower complaints; and (3) develop a
caseload management plan to more equitably distribute whistleblower complaints among investigators. According to a
February 2021 memorandum from DOL OIG to OSHA, as of February 2021, OSHA (1) had ﬁlled the ﬁve open investigator

Page 122 GAO-22-105051
Prior to the NEP, all COVID-19-related inspections were unprogrammed—conducted in response
to incoming reports of hazardous working conditions at a speciﬁc workplace. In addition to
continuing unprogrammed inspections, the NEP includes plans for OSHA to conduct programmed
COVID-19 inspections—not in response to any speciﬁc report of a hazard—at workplaces where
workers have higher risks of exposure to COVID-19. From February through August 2021, OSHA
has performed a total of 960 COVID-19 NEP inspections, including 366 programmed and 594
unprogrammed.165

To implement the NEP, OSHA created two targeting lists of establishments within the local
jurisdiction of each area oﬃce from which to select for programmed inspections—the ﬁrst based
solely on industries with a higher risk of exposure to COVID-19 and the second based on those
higher-risk industries plus establishment-speciﬁc respiratory and other illness rates in 2020.

The ﬁrst NEP targeting list includes establishments in industries where OSHA conducted the
highest number of COVID-19-related enforcement activities from February 2020 through mid-June
2021. This list aims to identify where the highest number of workers are expected to perform tasks
associated with exposure to COVID-19.166 This includes establishments in health-care industries,
such as hospitals and assisted living facilities, and non-health-care industries, such as poultry
processing and grocery stores.167

The second NEP targeting list includes establishments with higher-than-average per capita
respiratory and other illness rates, based on employer-reported Form 300A summary injury

positions, (2) expected to complete monitoring and evaluation of a pilot program to expedite complaints by March
2021, and (3) planned to develop a long-term solution to minimize complaint backlogs by March 2021. As of October
2021, OSHA had not yet implemented the second and third recommendations and did not have a new estimated date
for their implementation; OSHA oﬃcials said they would work with DOL OIG when they are ready to close out the
recommendations. For DOL OIG’s report, see DOL, OIG, COVID-19: OSHA Needs to Improve Its Handling of Whistleblower
Complaints During the Pandemic, Report No. 19-20-010-10-105 (Washington, D.C., August 14, 2020).
165Under the NEP, OSHA’s goal is to perform approximately 1,600 COVID-19 inspections—or 5 percent of all
inspections—during the NEP’s 1-year duration. OSHA oﬃcials told us that area oﬃces can meet this goal with
unprogrammed enforcement activities based on COVID-19-related complaints, employer reports, referrals, fatalities, or
catastrophes, or through programmed inspections based on establishment-targeting lists created for the NEP. OSHA
oﬃcials told us that the balance of programmed versus unprogrammed NEP inspections will vary by area oﬃce; this
is because area oﬃces in diﬀerent parts of the country have received diﬀerent levels of complaints, employer reports,
and referrals related to COVID-19—which can result in unprogrammed enforcement activities—and the NEP allows area
oﬃces to use their discretion to perform COVID-19 programmed inspections as resources allow.
166To create this list, OSHA included establishments from the 11 health-care industries and 11 non-health-care
industries with the highest numbers (with no per capita adjustment) of COVID-19-related complaints, fatalities/
catastrophes, employer reports, referrals, inspections, violations, and Hazard Alert Letters (which are to be used when
a workplace hazard is serious enough that employers should be notiﬁed, though the case does not meet all necessary
criteria for a general duty clause violation, and no speciﬁc OSHA standard applies). OSHA summed the number of these
enforcement activities it performed within industries in the same North American Industry Classiﬁcation System 6-digit
industry code. A July 2021 revision to the NEP eliminated a secondary group of targeted industries—those in critical
infrastructure sectors that, at the height of the pandemic, were considered to have elevated risks of exposure, such
as critical manufacturing, and transportation and logistics. According to the revised NEP, with the changing COVID-19
environment and the increase in vaccination rate in the general population, the need for a very broad range of targeted
industries—to include critical infrastructure sectors—diminished.
167See the enclosure on Food Safety Inspections for more information on our review of the U.S. Department of
Agriculture’s eﬀorts to conduct meat and poultry plant inspections and track inspectors’ exposure to COVID-19.

Page 123 GAO-22-105051
and illness data (or “300A data”).168 However, as we found in a January 2021 report on summary
injury and illness data reporting, these 300A data are incomplete and thus may be of limited
eﬀectiveness in developing targeting lists that identify workplaces most in need of COVID-19
inspections. Speciﬁcally, we estimated that employers did not submit 300A data for more than half
of required establishments for calendar years 2016 through 2018. Further, not all employers with
the potential for worker exposure to COVID-19 are required to submit 300A data.169

OSHA oﬃcials acknowledged that some establishments included in the ﬁrst targeting list are not
required to report 300A data, and thus will not show up in the second targeting list. According
to the NEP Directive, area oﬃces may add establishments to the NEP targeting lists, based on
their local knowledge, among other things, so they are not limited to establishments selected
based on incomplete data. However, because more than half of required establishments may not
be reporting 300A data, OSHA’s second targeting list may be missing many establishments with
higher rates of respiratory and other illness.170

As we found in our January 2021 report, because OSHA is less likely to conduct certain inspections
on employers who do not report 300A data, employers have an incentive to avoid reporting these
data.171 Such employers that avoid reporting 300A data may still merit a COVID-19 NEP inspection.
In our January 2021 report on summary injury and illness data reporting, we recommended
that OSHA evaluate the agency’s procedures for ensuring that employers report their 300A
data and make needed changes, and the agency generally agreed with our recommendation.
As of September 2021, DOL had not addressed this recommendation.172 By implementing

168OSHA requires employers to record work-related injuries and illnesses on an ongoing basis using Form 300. At
the end of the year, employers must submit a summary of the Form 300 data they recorded using Form 300A. See 29
C.F.R. §§ 1904.4-.29, .41. The second NEP targeting list includes establishments from the ﬁrst list that, according to the
300A data they reported to OSHA, had elevated per capita respiratory and other illness rates in 2020 compared to their
industry. While 300A data do not include speciﬁc data on COVID-19 infections, since COVID-19 is a respiratory illness with
additional possible symptoms, according to OSHA oﬃcials, it is likely that high rates of respiratory and other illness in
these targeted establishments are due to COVID-19 illness.
169OSHA regulations require two groups of employers to report 300A data: those with establishments in certain
industries, such as manufacturing and nursing care facilities, that had between 20 and 249 employees at any point
during the previous calendar year; and those with establishments in any industry that had 250 or more employees at
any point during the previous calendar year and were required to maintain injury and illness records.
170In August 2021, OSHA oﬃcials estimated that employers submitted 300A data for 47 percent of required
establishments in 2020. They further noted that, although the 300A data do not include more than half of required
establishments, since some data were available, it was prudent to use the data to the extent possible. Oﬃcials said
that they are conﬁdent that the selected establishments had substantial challenges with COVID-19 illness rates, based
on their submitted 300A data. However, many other establishments that did not report 300A data, and could not
be selected for the second targeting list because of this, may have had similar or more substantial challenges with
COVID-19. By implementing our January 2021 recommendation, the 300A data could be signiﬁcantly improved and
OSHA would be able to select establishments to devote resources to those most in need of COVID-19 inspections.
171In addition to NEP targeting, OSHA uses 300A data to select establishments for Site-Speciﬁc Targeting inspections,
which are comprehensive inspections that are to examine all potentially hazardous areas of an establishment. Although
employers may avoid reporting 300A data to avoid these inspections, employers did not know about the March 2021
NEP when collecting data throughout 2020 to submit as 300A data.
172According to OSHA oﬃcials, the agency conducted a pilot program in two regions in July and August 2021 to test
the feasibility of matching workplaces with newly opened inspections against a list of workplaces that potentially did
not report required 300A data, in order to encourage and enforce 300A data compliance at more establishments. As

Page 124 GAO-22-105051
this recommendation, OSHA could base its targeting on more complete workplace illness
data—potentially including twice as many establishments—and thus, better target its COVID-19
NEP inspections.

OSHA issued a COVID-19 emergency temporary standard (ETS) for the health-care industry;
a separate infectious disease standard and a vaccination ETS are in process. In June 2021,
OSHA issued a COVID-19 ETS for certain health-care employers that treat suspected or conﬁrmed
COVID-19 patients, such as hospitals and long-term care facilities.173 The health-care ETS may be in
eﬀect for up to 6 months after publication and may be superseded by a permanent standard. This
ETS does not cover employers in other industries, some of which OSHA has identiﬁed as at high
risk of COVID-19 exposure in its other policies.174 The health-care ETS requires covered employers
to comply with several provisions to protect workers from COVID-19 hazards, including

• developing and implementing a COVID-19 plan and related policies and procedures, and
providing related training;

• screening and managing patients and visitors, including contractors, for COVID-19;

• implementing various COVID-19 mitigation measures, such as use of personal protective
equipment for employees, physical distancing, physical barriers, cleaning and disinfection, and
ventilation;

• providing time and paid leave for COVID-19 vaccination;

• ensuring anti-retaliation principles are upheld related to employee rights under the ETS;

of September 2021, OSHA was analyzing the results of the pilot. It is not yet clear if OSHA’s eﬀorts will substantially
improve required reporting.
173Under 29 U.S.C. § 655(c), OSHA has the authority to issue an “emergency temporary standard” (ETS) without going
through the normal rulemaking process if it determines that “employees are exposed to grave danger from exposure
to substances or agents determined to be toxic or physically harmful or from new hazards,” and that an ETS “is
necessary to protect employees from such danger.” In our September 2020 enclosure, we reported that the agency had
determined that an ETS was not necessary. However, in January 2021, President Biden signed Executive Order 13999 on
Protecting Worker Health and Safety, which, among other things, directed the Secretary of Labor to consider whether a
COVID-19 ETS was necessary. The COVID-19 health-care ETS went into eﬀect on June 21, 2021, with employer compliance
with certain provisions required by July 6, 2021 and others by July 21, 2021. The health-care ETS applies to workplaces
where employees provide health-care services or health-care support services, and exempts some health-care facilities,
such as non-hospital ambulatory care settings where all non-employees are screened prior to entry and people with
suspected or conﬁrmed COVID-19 are not permitted to enter, and well-deﬁned hospital ambulatory care settings
where all employees are fully vaccinated, all non-employees are screened prior to entry, and people with suspected or
conﬁrmed COVID-19 are not permitted to enter.
174Because a determination of employees being exposed to a “grave danger” is a requirement under 29 U.S.C. § 655(c)
(1)(A) for issuing an ETS, OSHA devoted substantial discussion in the health-care ETS to the “grave danger” to health-care
employees, citing the “severe health consequences of COVID-19, the high risk to employees of developing the disease
as a result of transmission of SARS-CoV-2 in the workplace, and that [health-care] workplace settings provide direct care
to known or suspected COVID-19 cases” as the basis for the determination. See 86 Fed. Reg. 32376, 32381-32384 (June
21, 2021). According to OSHA oﬃcials, the agency focused the scope of the health-care ETS on the areas of greatest
COVID-19 exposure and did not make any legal ﬁndings about workers outside healthcare settings because those were
not necessary to justify the requirements in the health-care ETS.

Page 125 GAO-22-105051
• screening and managing employees for COVID-19, including, for example, daily screening and
requiring employees to notify the employer of COVID-19 positive tests and symptoms;

• keeping a log of all employee COVID-19 cases, regardless of whether they are work-related;
and,

• reporting work-related COVID-19 fatalities and hospitalizations to OSHA, regardless of the
amount of time between the exposure to COVID-19 and the fatality or hospitalization.

The health-care ETS allows OSHA to obtain some worker safety data related to COVID-19 from
covered health-care employers—which they would not otherwise be required to provide.
Speciﬁcally, an enclosure to our January 2021 CARES Act report stated that OSHA did not receive
employer reports of all work-related hospitalizations related to COVID-19 because disease
symptoms do not appear within 24 hours, the required reporting time frame for work-related
hospitalizations.175 We recommended that OSHA determine what additional data may be needed
from employers or other sources.

The new reporting requirement in the health-care ETS addresses this challenge for covered health-
care employers by requiring them to report to OSHA all work-related COVID-19 fatalities and
hospitalizations, regardless of the amount of time between worker exposure to COVID-19 and the
death or hospitalization, and takes into consideration our prior recommendation. Thus, we are
closing this recommendation. OSHA is not requiring COVID-19-related hospitalization or fatality
reports under the adjusted time frames for workplaces where the health-care ETS does not apply.

The health-care ETS also adjusts the existing recordkeeping requirement, to require covered
health-care employers to keep an internal log of all COVID-19 cases among their employees,
regardless of whether they are determined to be work-related.176 When conducting inspections,
OSHA oﬃcials should be able to obtain these logs to better assess workers’ COVID-19 exposure
and risk, and identify any COVID-19-related violations.

Although the health-care ETS covers only employers in the health-care industry, in its other policies,
OSHA has acknowledged the widespread impact of COVID-19 on industries beyond health care. In
its March and July 2021 updates to its pandemic-related enforcement policy, OSHA expanded
its higher risk designation for COVID-19 exposure from applying only to health-care settings
to applying to any workplace that can be crowded or involve a high level of interaction with
people, providing poultry processing and correctional facilities, two non-health-care industries,
as examples of higher-risk workplaces. Also, as previously discussed, OSHA’s COVID-19 NEP is
designed to focus on industries where the agency has determined that workers face increased

175The existing reporting requirement in 29 C.F.R. § 1904.39 generally requires employers to report all work-related
in-patient hospitalizations, amputations, and losses of an eye within 24 hours, and all work-related fatalities within 8
hours. Under 29 C.F.R. § 1904.39, employers are only to report such hospitalizations if they occur within 24 hours of the
workplace exposure to COVID-19; however, symptoms of COVID-19 may appear within 2 to 14 days after exposure, far
beyond the reporting time frame.
176The existing recording requirement in 29 C.F.R. § 1904.4 generally requires employers to keep an internal record of all
work-related fatalities, injuries, and illnesses. Determining work-relatedness for those related to COVID-19 is challenging
because of the virus’s incubation period and the diﬃculties in tracking the source of exposure.

Page 126 GAO-22-105051
potential for exposure to COVID-19, and targets both health-care and non-health-care industries
(see ﬁg.).

COVID-19-Related Reports to OSHA among Industries Targeted by OSHA’s National Emphasis Program, February
2020 through August 2021

Notes: Reports to OSHA include the total number of complaints, referrals, employer reports of severe injury or illness, and
reports of fatalities or catastrophes.
Complaints refer to reports notifying OSHA of alleged workplace safety or health hazards. Complaints can be made by
employees, their representatives, or others.
Referrals and employer reports: OSHA uses the term “referrals” to encompass two diﬀerent report types, (1) reports of work-
related severe injuries and illnesses, which employers are required to submit to OSHA (which OSHA calls employer-reported
referrals); and (2) reports of potential workplace hazards from selected other entities, such as local government agencies or
media outlets. In this report, we use “referrals” to describe those reports from selected non-employer sources, and “employer
reports” to describe those reports from employers. Employers are required to report all work-related in-patient hospitalizations,
amputations, and losses of an eye within 24 hours. 29 C.F.R. § 1904.39.
Fatalities: Employers are required to report the work-related death of an employee to OSHA within 8 hours. 29 C.F.R. § 1904.39.
According to OSHA oﬃcials, most reports of fatalities come from employers. However, oﬃcials noted that they do receive
reports of fatalities from other sources, such as the media or emergency medical personnel. In this report we refer to all
reported fatalities as “reports of fatalities.”
Catastrophes: OSHA’s Field Operations Manual deﬁnes a catastrophe as the hospitalization of three or more employees
resulting from a work-related incident or exposure.

Page 127 GAO-22-105051
OSHA is engaged in rulemaking for the health-care ETS and a separate infectious disease standard,
and is developing a vaccination ETS. An ETS may serve as a proposal for a permanent standard,
and OSHA must generally take ﬁnal action on the proposal within 6 months of publication, in the
case of the health-care ETS, by December 2021.177 In September 2021, OSHA oﬃcials said they
were reviewing the comments they received on the health-care ETS during the public comment
period, which ended on August 20, 2021. They further noted that they were reviewing the ongoing
need for this ETS every 30 days and had not yet determined whether to extend this ETS beyond 6
months. Oﬃcials said they would continue to monitor trends in COVID-19 infections and deaths
and would update the health-care ETS, as appropriate, if and when “new information indicates a
change in measures necessary to address the grave danger [from the virus].” On June 24, 2021, the
AFL-CIO and United Food and Commercial Workers unions petitioned the U.S. Court of Appeals
for the D.C. Circuit to review OSHA’s decision not to issue an ETS applicable to employees outside
the health-care industry who face occupational exposure to COVID-19, including but not limited to
employees in the meatpacking and food processing industries.178

OSHA is also working on separate rulemaking for an infectious disease standard to protect
workers in high-risk environments from long-standing and emerging infectious diseases, with a
notice of proposed rulemaking currently projected to be published in December 2021. According
to the White House Oﬃce of Information and Regulatory Aﬀairs’ Spring 2021 regulatory agenda,
the rulemaking considers targeting health-care workers and others who are exposed in high-
risk environments, potentially covering workplaces such as hospitals, correctional facilities, some
laboratories, and other occupational settings where workers can be at increased risk of exposure
to infectious people.

On September 9, 2021, the White House announced a plan to increase COVID-19 vaccination.
Speciﬁcally, OSHA is to develop an ETS that would require all employers with 100 or more
employees to implement COVID-19 vaccination and testing requirements for employees and to
provide vaccine-related paid time oﬀ. In September 2021, OSHA oﬃcials said the agency was
working expeditiously to develop the vaccination ETS; oﬃcials did not provide an estimate for
when it would be ﬁnalized.

OSHA’s updated COVID-19 enforcement policy separates health-care and non-health-care
industries and reduces the use of some adapted methods used during the pandemic. In
June 2021, soon after issuing the health-care ETS, OSHA published an ETS enforcement policy for
inspectors designed to ensure uniform enforcement among ETS-covered health-care employers.
For key requirements in the health-care ETS, the ETS enforcement policy provides detailed
inspection guidance for what to include in an inspection, for example, determining whether
the employer has a designated eating and drinking area with suﬃcient space to accommodate
physical distancing. It also provides citation guidance for key requirements in the health-care ETS,
such as speciﬁc examples for when a violation may be cited, as well as guidance for inspector
safety, among other things.

17729 U.S.C. § 655(c)(3).
178The case is United Food and Commercial Workers Int’l Union v. OSHA, No. 21-1143 (D.C. Cir. ﬁled June 24, 2021).
On September 15, 2021, the court granted a joint request from the petitioners and OSHA that case proceedings be
temporarily suspended because of the September 9, 2021 White House announcement that OSHA would issue a new
COVID-19-related ETS, which may aﬀect the claims at issue in the case.

Page 128 GAO-22-105051
In July 2021, OSHA released its updated pandemic-related enforcement policy, which generally
covers non-health-care employers (as health-care employers are covered under the ETS
enforcement policy).179 The updated pandemic-related enforcement policy for non-health-care
employers

• focuses enforcement on protections for workers who are unvaccinated or not yet fully
vaccinated;

• revises its assessment of workplace risk, as discussed above, to include in its higher risk
category any workplace that can be crowded or involve a high level of interaction with people;
and

• provides updated guidance to protect OSHA inspectors.

Reﬂecting a progression toward a return to normal enforcement operations, the July 2021 policy
continues to allow use of two of OSHA’s COVID-19-adapted enforcement methods that we
described in our January 2021 enclosure—(1) remote inspections and (2) informal inquiries in
place of inspections—but generally removes citation discretion.180 The updated pandemic-related
enforcement policy states that OSHA will perform on-site COVID-19 inspections, instead of remote
inspections, in most cases.181 According to the policy, citation discretion was intended to be time
limited, applied on a case-by-case basis, and related to supply shortages, such as shortages in
N95 masks. OSHA is generally ending use of citation discretion in this context based on updated
Centers for Disease Control and Prevention and Food and Drug Administration guidance regarding
supply availability, according to the policy.

Inspectors or managers from all ﬁve area oﬃces we spoke with told us that the use of these
adapted enforcement methods during the pandemic varied by factors such as industry, risk
of worksite COVID-19 exposure, and severity of reported incidents. For example, according to
inspectors at one area oﬃce, inspectors for the construction industry transitioned from remote

179The July 2021 pandemic-related enforcement policy replaces previous iterations from April 2020, May 2020, and
March 2021.
180OSHA adapted its enforcement methods in response to the COVID-19 pandemic to include: (1) remote inspections
or a combination of remote and on-site inspections, in place of on-site inspections of workplaces; (2) use of informal
inquiries in place of inspections; OSHA refers to informal inquiries as phone/fax investigations or rapid response
investigations, as inspectors obtain information about an incident from employers by phone, fax, or email; (3) citation
discretion (which OSHA refers to as enforcement discretion) to take into account employers’ good faith eﬀorts to
comply with certain requirements—when they could not do so due to the pandemic—when determining whether to cite
violations. This third adapted method generally ended with the July 2021 pandemic-related enforcement policy.
181OSHA oﬃcials told us that, beginning with the March 2021 policy, on-site inspections are the default, instead of
remote inspections being the default as they were at the beginning of the pandemic. The policy states that area oﬃces
still have discretion to conduct remote inspections and to conduct informal inquiries in place of inspections to ensure
inspector health and safety. The NEP Directive states that remote-only COVID-19 inspections are reserved for limited
circumstances and are subject to the Area Director’s approval. The July 2021 pandemic-related enforcement policy
states that, in most cases, OSHA will perform on-site workplace inspections, while minimizing in-person meetings
with employers and employees. For example, OSHA will, when appropriate, use phone and video conferencing, in lieu
of face-to-face employee interviews, to reduce potential exposures to inspectors. OSHA will also minimize in-person
meetings with employers if necessary, and encourage employers to provide documents and other data electronically to
inspectors.

Page 129 GAO-22-105051
inspections back to onsite inspections earlier than for other industries due to the nature of
construction work and the lower risk of COVID-19 at construction sites. According to OSHA
guidance, some construction work tasks are classiﬁed as low risk activities, as they may be
outdoors and may allow for social distancing.

Data from the OSHA Information System (OIS), which the agency uses to track its enforcement
activities, indicate that over the course of the pandemic, some COVID-19-adapted enforcement
methods have generally declined. In addition, OIS data show that OSHA enforcement activities
shifted substantially from inspections to informal inquiries at the start of the pandemic, and that
then the proportion of informal inquiries generally declined through August 2021 (see ﬁg.).182

OSHA Enforcement Activities Based on Report Type, for Selected Periods from March 2019 through August 2021

Notes: Complaints refer to reports notifying OSHA of alleged workplace safety or health hazards. Complaints can be made by
employees, their representatives, or others.
Referrals and employer reports: OSHA uses the term “referrals” to encompass two diﬀerent report types, (1) reports of work-
related severe injuries and illnesses, which employers are required to submit to OSHA (which OSHA calls employer-reported
referrals); and (2) reports of potential workplace hazards from selected other entities, such as local government agencies or
media outlets. In this report, we use “referrals” to describe those reports from selected non-employer sources, and “employer
reports” to describe those reports from employers. Employers are required to report all work-related in-patient hospitalizations,
amputations, and losses of an eye within 24 hours. 29 C.F.R. § 1904.39.
An informal inquiry is a process conducted in response to a complaint, referral, or employer report of severe injury or illness
that does not meet OSHA’s criteria for conducting an inspection. According to OSHA oﬃcials, informal inquiries conducted
in response to an employer-reported severe injury or illness are called rapid response investigations, and informal inquiries
conducted in response to complaints from employees or referrals from entities other than employers are called phone/fax
investigations.
According to OSHA’s Field Operations Manual, if Area Directors consider employers’ responses to these informal inquiries to be
inadequate, they may decide to initiate a related inspection.
a
COVID-19-related enforcement activities are a subset of all enforcement activities from March 2020 through August 2021.
The related bars represent the percentages of all COVID-19-related enforcement activities that were informal inquiries or
inspections.

182OIS documents when informal inquiries were used, but does not identify when they were used in place of
inspections. That is, OIS does not specify when COVID-19-related constraints caused an area oﬃce to use an informal
inquiry, in place of an inspection, to address the complaint, referral, or employer report. Therefore, the precise extent to
which OSHA’s adapted enforcement methods aﬀected the shifts between inspections and informal inquiries is unclear.
An enclosure to our January 2021 CARES Act report recommended that OSHA ensure that OIS include comprehensive
information on use of the agency’s COVID-19-adapted enforcement methods. This recommendation remains open and
an update is provided below.

Page 130 GAO-22-105051
Area oﬃces faced enforcement challenges during the COVID-19 pandemic, but OSHA has
not assessed lessons learned or promising practices. Oﬃcials we interviewed in OSHA area
oﬃces reported facing operational challenges in enforcement throughout the COVID-19 pandemic.
However, OSHA has not yet assessed these challenges to improve its response during the current
pandemic and prepare for any future pandemic.183 In particular, area oﬃce inspectors and
managers identiﬁed challenges related to resources and to communication and guidance.

Resource challenge: staﬃng and volume of incoming reports. Inspectors from all ﬁve area oﬃces we
spoke with described challenges with processing a large volume of complaints and other reports
of workplace hazards at the beginning of the pandemic. Inspectors from one area oﬃce said the
number of incoming complaints was overwhelming, while inspectors from another area oﬃce
described not being able to give cases the full attention they would have received, prior to the
pandemic.

This workload was exacerbated by staﬃng challenges throughout the agency. Inspectors or
managers from three of ﬁve area oﬃces we spoke with described experiencing staﬃng challenges
during the pandemic, including high turnover and high numbers of inexperienced staﬀ, who could
not conduct inspections on their own.

The volume of COVID-19-related work varied across the country because the pandemic’s impact
diﬀered from area to area. Staﬃng shortages varied among oﬃces, and certain area oﬃces
developed useful strategies for addressing these challenges. For example, oﬃcials from one
area oﬃce described a helpful practice of directing phone calls from the public, or other work, to
another area oﬃce that had fewer COVID-19-related complaints.

Workload concerns also made citing COVID-19-related violations challenging because of the
substantial time commitment needed for inspectors to collect the evidence and documentation
necessary to support a citation. Inspectors or managers from three of ﬁve area oﬃces we spoke
with described challenges during the COVID-19 pandemic with meeting the requirement that
citations must be issued within 6 months of the violation. Particular challenges included the large
amount of paperwork required for a COVID-19 citation, and inspectors sometimes learning of a
COVID-19-related fatality several months after it occurred—when much of the 6-month window
had already expired.184

Resource challenge: telework and technology. OSHA headquarters oﬃcials said the agency was very
well prepared to transition operations from in-person inspections and oﬃce work to telework in
response to the pandemic, aside from occasional minor issues with technology. However, based
on our interviews, comfort with telework and advance preparation for telework varied across area
oﬃces and individual OSHA staﬀ.

183Such assessments would be distinct from OSHA’s oversight plans for COVID-19-adapted enforcement methods about
which we made recommendations in our January 2021 enclosure. These oversight plans relate to individual enforcement
activities, whereas an assessment of OSHA’s operational challenges throughout the pandemic would be a broader
analysis of the agency’s operations and related challenges.
184Under the OSH Act’s statute of limitations, OSHA may not issue a citation to an employer for violating the act or any
OSHA regulations after the expiration of 6 months following the occurrence of the violation. 29 U.S.C. § 658(c). For more
information on this 6-month statute of limitations, see our January 2021 report on reporting of summary injury and
illness data.

Page 131 GAO-22-105051
Some inspectors from all ﬁve area oﬃces we spoke with described challenges with technology,
including scanning and printing enforcement-related documents, while other inspectors from
two of these ﬁve oﬃces described positive experiences working remotely. According to OSHA
oﬃcials, many regions have electronic case ﬁle pilot programs in place, and the agency is working
to implement electronic case ﬁles nationwide, though the standard OSHA case ﬁle remains paper-
based. Oﬃcials noted that, even once electronic case ﬁles are implemented nationwide, OSHA will
continue accepting paper materials from employers and workers, which would then need to be
scanned into the electronic case ﬁles, and some documents that are generated electronically will
still need to be printed and mailed to employers.

Communication and guidance challenge: Guidance and tools for inspectors provided later than needed.
Inspectors from all ﬁve area oﬃces we spoke with described challenges in performing their roles
due to a lack of timely guidance from OSHA headquarters, or frequent guidance changes. An
inspector from one area oﬃce stated that OSHA’s COVID-19-related guidance often conﬂicted with
guidance and recommendations from other government agencies. OSHA headquarters oﬃcials
said they provided information to inspectors regularly during the pandemic, including interim
enforcement memoranda, via OSHA’s internal webpage, by email, and through meetings with
regional oﬃces. However, inspectors in the area oﬃces said the lack of timely guidance aﬀected
their operations in the following ways:

• Inspectors from one area oﬃce described “scrambling” early in the pandemic to ﬁgure out how
to apply existing standards to COVID-19 hazards and said they did not know how to advise
employers and others who asked for guidance.

• According to inspectors from another area oﬃce, the lack of adequate guidance from the
Solicitor of Labor made citing COVID-19-related violations time-consuming and diﬃcult.
Inspectors said they did not know, in advance of the Solicitor’s review of COVID-19-related
violation cases, what supporting documentation would be required.185 According to
headquarters oﬃcials, as they conducted more COVID-19 violation reviews, OSHA issued
guidance on common citation language to reﬂect lessons learned and guide inspectors going
forward. Headquarters oﬃcials also said the agency posted templates for preparing COVID-19-
related citations in November 2020—8 months into the pandemic.

• A tool to calculate the probability of exposure to COVID-19 was provided in April 2021, leaving
the oﬃce without this type of resource for too long, according to inspectors from one area
oﬃce.

Communication and guidance challenge: diﬃculty ﬁnding and using the most up-to-date guidance.
OSHA headquarters oﬃcials told us that a COVID-19 team scans Centers for Disease Control
and Prevention, National Institute for Occupational Safety and Health, and Food and Drug
Administration websites for updates to COVID-19 employer guidance on a daily basis, and emails
policy updates to regional oﬃces. The regional oﬃces would then send the updates to area oﬃce
leadership, who would send them to inspectors.

185Oﬃcials from the Solicitor of Labor’s oﬃce contested that area oﬃces did not know what documentation was
required for COVID-19-related violation cases because the Solicitor’s reviews were conducted using the established
signiﬁcant case review process that was laid out in a memorandum to the ﬁeld.

Page 132 GAO-22-105051
This method resulted in communication inconsistencies and gaps. According to inspectors
from three of ﬁve area oﬃces we spoke with, the frequent changes in guidance were diﬃcult
to keep track of; this led to inspector confusion about which guidance was in eﬀect, according
to inspectors from one area oﬃce. For example, inspectors from another area oﬃce we spoke
with said that they were unsure how to handle changing N-95 mask-ﬁt testing requirements,
particularly for employers like nursing homes that were unfamiliar with this requirement.
Speciﬁcally, they noted it was challenging to determine whether to cite these workplaces for
violation of a standard they were not familiar with due to the frequently changing guidance.

Communication and guidance challenge: managers unaware of inspectors using COVID-19-related
adapted enforcement methods. In some selected area oﬃces, managers seemed unaware of
inspectors using enforcement methods that had been adapted due to the pandemic, limiting their
ability to oversee and evaluate those methods. For instance, managers from one area oﬃce we
spoke with said their inspectors had conducted only one inspection remotely, while inspectors
from that oﬃce said they had regularly used remote inspections during the pandemic. OIS data
indicate that inspectors have used adapted enforcement methods during the pandemic, including
conducting inspections remotely. As shown in the above ﬁgure on enforcement activities based on
report type, OSHA enforcement also shifted substantially from inspections to informal inquiries
during the pandemic months.

Additionally, managers from the same area oﬃce said they had not used citation discretion (i.e.,
not citing a violation for an identiﬁed hazard, due to extenuating circumstances or employer good
faith eﬀorts), while inspectors from that oﬃce said they used this discretion.

Communication and guidance challenge: OSHA oﬃcials had diﬀerent understandings of the OIS
code developed to track inspectors’ use of citation discretion, and inspectors may have used it
inconsistently throughout the pandemic.186 From February 2020 through August 2021, the OIS code
for citation discretion was used 4 times, indicating limited use of this discretion. While some OSHA
headquarters oﬃcials said the OIS code should only apply to citation discretion involving certain
requirements, other OSHA headquarters oﬃcials said that all instances of citation discretion
should be tracked using this code.187 In addition, inspectors from all ﬁve area oﬃces we spoke
with described using this citation discretion, but not consistently using the designated OIS code to
track it.

Communication and guidance challenge: lack of process for sharing of promising practices among
area oﬃces. OSHA may have missed opportunities to share knowledge and tools among area

186OIS allows users to tag inspections or investigations with various identifying codes, such as inspection type, related
emphasis programs and, during the COVID-19 pandemic, certain COVID-19-related codes, including a code indicating
that hazard abatement was deferred because of citation discretion.
187OSHA’s April and May 2020 pandemic-related enforcement policies allowed citation discretion related to (1)
challenges in meeting certain recurring requirements, such as annual trainings; (2) shortages of personal protective
equipment; or (3) challenges in meeting recordkeeping requirements. However, only the policy related to citation
discretion for the ﬁrst area—recurring requirements—included an OIS code to track those instances of citation
discretion. Thus, some OSHA oﬃcials said that, according to the policy, the OIS code should only apply to citation
discretion involving the recurring requirements; other oﬃcials said that any of the three types of citation discretion
should be tracked using the code. As discussed earlier in this enclosure, these citation discretion policies are generally
no longer in eﬀect. However, related challenges while they were in eﬀect may be useful in assessing lessons learned for
the future.

Page 133 GAO-22-105051
oﬃces throughout the pandemic. While some area oﬃces and regional oﬃces developed unique
practices and tools to enhance their eﬀorts during the pandemic, OSHA did not have a process
in place for area oﬃces to systematically share promising practices with each other during the
pandemic. Inspectors or managers from all ﬁve area oﬃces we spoke with described a number
of unique practices that area and regional oﬃces had implemented during the pandemic; for
example, creating web-based document management sites to share templates, guidance, and
other documents, developing systems to help area oﬃces protect OSHA staﬀ, and coordinating
solutions to technology challenges. Managers from two of ﬁve area oﬃces we spoke with said it
would be good if OSHA were to compile promising practices and lessons learned to better prepare
for another emerging issue.

In August 2021, OSHA headquarters oﬃcials said they had not yet conducted a formal evaluation
or formally collected lessons learned or helpful practices for operating during a pandemic from
area oﬃces. Federal internal control standards state that agencies should evaluate issues and
remediate deﬁciencies. OSHA oﬃcials cited the ongoing pandemic as the reason they had not
performed such an assessment and said they intended to do so when feasible. However, even
while the COVID-19 pandemic is ongoing, analyzing OSHA’s response—including its response
to challenges it has faced, such as those related to resources and to communication and
guidance—and taking related actions as warranted would enable OSHA to make improvements
to better support ongoing enforcement eﬀorts during the COVID-19 pandemic and prepare for
operations during any future pandemic.

Methodology
To conduct this work, we reviewed OSHA guidance and enforcement policy, relevant federal
laws and regulations, and the most recent OSHA data through August 2021.188 To assess the
reliability of OSHA’s data, we reviewed technical documentation and interviewed OSHA oﬃcials.
We determined that OSHA’s data were suﬃciently reliable for the purposes of our reporting
objectives. We also interviewed OSHA headquarters oﬃcials, and managers and inspectors from 5
of OSHA’s 89 area oﬃces, selected to represent areas with industries aﬀected by COVID-19 and a
higher than average number of COVID-19-related complaints, employer reports, and referrals from
February through September 2020, among other things.

Agency Comments
We provided DOL and the Oﬃce of Management and Budget (OMB) with a draft of this enclosure.
OMB did not provide comments on this enclosure. DOL provided written comments, reproduced
in appendix VIII, and technical comments, which we incorporated as appropriate. In its comments,
DOL partially agreed with our recommendation to assess—as soon as feasible and, as appropriate,
periodically thereafter—various challenges related to resources and to communication and
guidance that OSHA has faced in its response to the COVID-19 pandemic and take related actions
as warranted.

188These data are current as of September 7, 2021.

Page 134 GAO-22-105051
DOL stated that the agency agrees that it is important to assess lessons learned and best practices
for OSHA’s operational response to COVID-19. However, DOL oﬃcials said they believe that while
the pandemic is ongoing, the agency’s resources are best used to help employers and workers
mitigate exposures to COVID-19. DOL stated that the agency intends to conduct a review of
OSHA’s response to the COVID-19 pandemic after operations return to normal.

However, it is unclear when the COVID-19 pandemic will end, and OSHA analyzing its response
and taking related actions, as warranted, even as the pandemic is ongoing, would enable
the agency to improve its enforcement eﬀorts during this pandemic, in addition to helping it
prepare for operations during any future pandemic. We continue to believe that assessing—as
soon as feasible and, as appropriate, periodically thereafter—various challenges that OSHA
faced in responding to the pandemic, and taking related actions, would enhance transparency
and accountability in the federal government’s response to, and recovery from, the COVID-19
pandemic.

GAO’s Ongoing Work
Our review of worker safety and health during the COVID-19 pandemic is shifting to focus on
the safety and health of workers at meat and poultry processing plants during the pandemic.
We will continue to examine OSHA’s eﬀorts to protect workers in these industries, and monitor
developments in overall worker safety and health during the ongoing pandemic.

GAO’s Prior Recommendations
The table below presents our recommendations on Worker Safety and Health from prior
bimonthly and quarterly CARES Act reports.

Page 135 GAO-22-105051
Prior GAO Recommendations Related to Worker Safety and Health
Recommendation Status

The Assistant Secretary of Labor Open—partially addressed. The Department of Labor (DOL) neither agreed
for Occupational Safety and Health nor disagreed with our recommendation. The Occupational Safety and Health
should develop a plan, with time Administration’s (OSHA) COVID-19-adapted enforcement methods continue to lack
frames, to implement the agency’s ﬁrm oversight plans. We reported in January 2021 that OSHA did not have speciﬁc
oversight processes for COVID-19- plans or time frames for how and when to conduct the oversight of its COVID-19-
adapted enforcement methods, adapted enforcement methods that was outlined in the agency’s pandemic-
as described in its pandemic related enforcement policy—to ensure the methods are eﬀective.
enforcement policies. (January 2021 • For oversight of remote inspections, in May 2021, OSHA oﬃcials said that
report)
the agency was no longer planning to conduct the oversight outlined in its
May 2020 pandemic-related enforcement policy, which provided guidance to
inspectors for COVID-19-related enforcement until the March 2021 pandemic-
related enforcement policy was issued. Instead, oﬃcials said that follow-up
for some, but not all, remotely-conducted inspections would be performed
according to area oﬃces’ discretion as part of OSHA’s COVID-19 National
Emphasis Program (NEP), as resources permit area oﬃces to focus more on
programmed inspections.

• For oversight of informal inquiries conducted in place of inspections, in
February 2021, OSHA oﬃcials said that they planned to conduct follow-
up inspections for a random sample of cases where COVID-19-related
informal inquiries were conducted. However, this plan would target all
informal inquiries, and not just those that were conducted in place of
inspections because of the pandemic, as originally planned in OSHA’s May
2020 enforcement policy. This change in sampling technique could make
it less likely that the cases meriting further scrutiny would be identiﬁed for
follow-up. In August 2021, OSHA oﬃcials told us they would consider this
issue when they make further plans for this oversight.

• For oversight of citation discretion, in February 2021, OSHA oﬃcials said
that they would conduct a follow-up inspection for each case coded in
the OSHA Information System (OIS) as having used discretion to not cite
violations. OSHA’s COVID-19 NEP includes instructions for conducting these
follow-up inspections. However, it is unclear whether all instances where
citation discretion was used can be identiﬁed in order to conduct follow-up
inspections, as discussed below.

In September 2021, OSHA oﬃcials said that the agency intends to conduct a
“comprehensive lookback” on OSHA’s response to the COVID-19 pandemic
after the pandemic ends. While such a review is the subject of our October
2021 recommendation, it is unclear whether it would meet the intent of this
recommendation to develop a plan for the speciﬁc oversight processes for
COVID-19-adapted enforcement methods that OSHA described in its pandemic
enforcement policies. The oversight processes described in this recommendation
are speciﬁcally intended to evaluate the eﬀectiveness of adapted enforcement
methods used throughout the pandemic. We therefore continue to recommend
that OSHA complete speciﬁc plans for its oversight of informal inquiries conducted
in place of inspections and its citation discretion or clearly state that these speciﬁc
oversight plans have changed.

The Assistant Secretary of Labor Open—partially addressed. DOL neither agreed nor disagreed with our
for Occupational Safety and Health recommendation. OSHA continues to be unable to reliably track some of its
should ensure that the Occupational COVID-19-adapted enforcement methods. We reported in January 2021 that OSHA
Safety and Health Administration could not reliably track some types of adapted enforcement methods in OIS, which
Information System includes may hinder its ability to conduct its planned oversight. For example, although OIS

Page 136 GAO-22-105051
Recommendation Status
comprehensive information on use documents when informal inquiries are used, in general, it does not identify when
of the agency’s COVID-19-adapted the informal inquiry constituted an adapted enforcement method—that is, when
enforcement methods suﬃcient COVID-19-related constraints caused an area oﬃce to use an informal inquiry,
to inform its oversight processes in place of an inspection, to address a complaint, referral, or employer report.
for these methods. (January 2021 Therefore, OSHA does not know the extent to which this adapted enforcement
report) method is being used. As a result, OSHA will not be able to target its oversight for
this adapted enforcement method, once it makes plans to do so.
In addition, OSHA oﬃcials had diﬀerent understandings of the OIS code developed
to track inspectors’ use of citation discretion, and inspectors may have used it
inconsistently throughout the pandemic. From February 2020 through August
2021, the OIS code was used 4 times, indicating limited use of this citation
discretion. This may be due to confusion concerning how OSHA is tracking citation
discretion.
• Some OSHA headquarters oﬃcials told us that the OIS code should only apply
to citation discretion involving certain requirements.

• Other OSHA headquarters oﬃcials told us that all instances of citation
discretion would be tracked using this code.

• OSHA does not have a method to ensure that inspectors are consistently
using the designated code to identify cases where inspectors observed
violations, but did not cite them, according to OSHA oﬃcials.

• Finally, inspectors from all ﬁve area oﬃces we spoke with described using this
citation discretion, but did not consistently use the OIS code to track it.

As a result of this confusion, follow-up inspections that OSHA has planned to
monitor this COVID-19-adapted enforcement method may not cover all instances
when it was used. As of August 2021, OSHA oﬃcials said they had conducted an
informal review to identify inspections with the OIS code for citation discretion,
and that they had worked with regional oﬃces to identify any instances when the
code had been recorded in error. In August 2021, oﬃcials said that OSHA was
conducting a ﬁnal review, to determine the number of inspections that had been
accurately coded as using citation discretion, and the results of any follow-up. Until
OSHA completes this ﬁnal review, we will be unable to assess whether it addressed
all of our above concerns.
In September 2021, OSHA oﬃcials said that OIS allows for suﬃcient coding. They
also stated that relevant citation discretion is no longer in use. However, OSHA has
not responded to our concerns about tracking informal inquiries used in place of
inspections. OSHA also has not provided more information on the agency’s ﬁnal
review of its use of the citation discretion OIS code, described above. We therefore
continue to recommend that OSHA ensure that OIS includes comprehensive
information on the use of the agency’s COVID-19-adapted enforcement methods
suﬃcient to inform its oversight processes for these methods.

The Assistant Secretary of Labor Closed—addressed. In February 2021, OSHA said that, in response to our
for Occupational Safety and recommendation, it had determined that it did not need additional information
Health should determine what from employers to identify where pandemic-related enforcement should be
additional data may be needed targeted. However, OSHA’s June 2021 health-care emergency temporary standard
from employers or other sources to (ETS) speciﬁcally addressed the data gap that we identiﬁed in January 2021 related
better target the agency’s COVID-19 to employer reporting of COVID-19-related hospitalizations for certain health-
enforcement eﬀorts. (January 2021 care employers whose employees, OSHA determined, face “grave danger”. OSHA
report) therefore did determine that it needed additional data from certain employers for
its enforcement eﬀorts, in accordance with our recommendation.

Page 137 GAO-22-105051
Source: GAO I GAO-22-105051

Related GAO Products
Workplace Safety and Health: Actions Needed to Improve Reporting of Summary Injury and Illness Data.
GAO-21-122. Washington, D.C.: January 27, 2021.

Standards for Internal Control in the Federal Government. GAO-14-704G. Washington, D.C.:
September 10, 2014.

Workplace Safety and Health: Multiple Challenges Lengthen OSHA’s Standard Setting. GAO-12-330.
Washington, D.C.: April 2, 2012.

Contact information: Thomas M. Costa, 202-512-7215, costat@gao.gov

Page 138 GAO-22-105051
Child Care
The Oﬃce of Child Care is taking initial steps to ensure accountability over COVID-19 supplemental
funds, and these funds—along with federal child care ﬂexibilities—have been critical to states to
help mitigate the impacts of the COVID-19 pandemic on child care, according to our 2021 national
survey of state child care administrators.

Entity involved: Oﬃce of Child Care, Administration for Children and Families, within the
Department of Health and Human Services

Background
The federal child care subsidy program known as the Child Care and Development Fund (CCDF)
assisted, on average, about 1.3 million eligible children from low-income families per month in
ﬁscal year 2018, the most recent year for which ﬁnal data are available.189 CCDF was appropriated
nearly $62 billion in federal funds since the March 2020 declaration of COVID-19 as a national
emergency—including more than $52 billion in CARES Act and other COVID-19 supplemental
funds, in addition to annual appropriations, for CCDF to help states prevent, prepare for, and
respond to the COVID-19 pandemic (see table).190 CCDF is administered as a block grant to the
states by the Oﬃce of Child Care (OCC), an oﬃce within the Department of Health and Human
Services’ (HHS) Administration for Children and Families (ACF).191

189According to preliminary data for ﬁscal year 2019, about 1.4 million eligible children received subsidies per month.
190The Child Care and Development Fund is made up of two funding streams: mandatory and matching funding
authorized under section 418 of the Social Security Act, and discretionary funding authorized under the Child Care
and Development Block Grant Act of 1990, as amended. See 42 U.S.C. §§ 618 and 9858m. In ﬁscal year 2019, prior to
COVID-19, total CCDF federal funding was $8.1 billion.
191For reporting purposes, in this enclosure we use “states” to also refer to the District of Columbia and U.S. territories,
unless otherwise indicated. Additionally, we include CCDF funding to tribes in the overall total but did not speak to tribal
governments about their planned uses of funds.

Page 139 GAO-22-105051
Coronavirus Supplemental Appropriations to the Child Care and Development Fund (CCDF)
Act Appropriations to CCDF

CARES Act
a $3.5 billion

Consolidated Appropriations Act, 2021, Division M
b $10.0 billion

American Rescue Plan Act of 2021
c $39.0 billion

Total $52.5 billion

Source: Pub. L. No. 116-136, div. B, tit. VIII, 134 Stat. 281, 557 (2020); Pub. L. No. 116-260, div. M, 134 Stat. 1182, 1914; and Pub. L. No. 117-2, §§ 2201, 2202, 135 Stat. 4, 31. I
GAO-22-105051

a
States have until September 30, 2022, to obligate the CARES Act funds and until September 30, 2023, to spend them.
b
Division M of the Consolidated Appropriations Act, 2021 is the Coronavirus Response and Relief Supplemental Appropriations
Act, 2021. States have until September 30, 2022, to obligate the Coronavirus Response and Relief Supplemental Appropriations
Act, 2021 funds and until September 30, 2023, to spend them.
c
The American Rescue Plan Act of 2021 includes $24 billion in child care stabilization funds and $15 billion in supplemental
CCDF funds. States have until September 30, 2022, to obligate the stabilization funds and until September 30, 2023, to spend
them. States have until September 30, 2023, to obligate the supplemental CCDF funds and until September 30, 2024, to spend
them.

In addition to COVID-19 supplemental funds for CCDF, the Small Business Administration’s (SBA)
Paycheck Protection Program (PPP) provided a large potential source of funding to child care
providers who applied and met program eligibility requirements, primarily to help keep workers
employed.192

Overview of Key Issues
Availability of child care. COVID-19 notably impacted the availability of child care for families
early in the pandemic, with the percentage of all open child care providers increasing throughout
the year, according to our 2021 national survey of state CCDF administrators (see ﬁgure). States
reported that, in March 2020, 59 percent of child care centers and 82 percent of home-based
providers were open for business.193 About one-half of states also reported that the availability
of child care to meet the needs of essential and non-essential workers was greatly challenging in
March 2020 (23 and 25 states, respectively).194 However, ﬁve or fewer states reported child care
availability for essential and non-essential workers as greatly challenging in December 2020.

192The Paycheck Protection Program was created as part of the CARES Act, Pub. L. No. 116-136, div. A, tit. I, § 1102, 134
Stat. 281, 286 (2020). The Paycheck Protection Program Extension Act of 2021 extended the application period from
March 31, 2021, to May 31, 2021, and allowed SBA until June 30, 2021, to process those applications. On May 4, 2021,
SBA stopped accepting applications from new lenders except those processed by a community ﬁnancial institution
lender. Borrowers may qualify for full loan forgiveness of their PPP loans if certain conditions are met.
193Our 2021 survey asked states to report on four points in time: March 31, June 30, September 30, and December 31,
2020. Our analysis was limited to states that reported data on open providers at each point in time. Data for child care
centers were missing from up to 13 states. Data for home-based/family child care providers were missing from up to 15
states.
194We use greatly challenging to refer to challenges that states reported as very or extremely challenging in our survey.

Page 140 GAO-22-105051
Percentage of Open Child Care Providers Increased during 2020

Note: Our 2021 survey asked states to report on four points in time: March 31, June 30, September 30, and December 31,
2020. Our analysis was limited to states that reported data on open providers at each point in time. Data for child care centers
were missing from up to 13 states. Data for home-based/family child care providers were missing from up to 15 states. The
percentage of open home-based/family child care providers states reported in September 2020 was the same—93 percent—as
in December 2020.

The availability of child care exceeded demand at times during the pandemic, according to some
state CCDF administrators we interviewed, which may have mitigated the impact that closures
had on families.195 For example, administrators from two of the eight states we interviewed
cited insuﬃcient demand for their child care programs targeted to essential workers, such as
hospital workers. Overall, according to state oﬃcials, the need for child care may have decreased
as a result of changes to parental preferences due to concerns for health and safety, and more
parents working from home. State oﬃcials added that these types of changes make it diﬃcult for
providers and states to plan for future child care needs.

Child care challenges. Alongside closures, child care providers faced many other challenges due
to the pandemic—most frequently ﬁnancial—according to our national survey. In March 2020,
39 of 50 states that responded to our survey rated as greatly challenging ﬁnancial problems for
providers due to (1) decreased child care enrollment and (2) temporary closures. By December,
fewer states did so (35 and 26, respectively). Closely behind, 37 states reported that providers
being able to obtain personal protective equipment or cleaning supplies was greatly challenging
in March 2020. Overall, each of the top challenges states reported in our survey showed
improvement between March and December 2020, as illustrated in the ﬁgure below.

195We interviewed state CCDF administrators in eight states: Arizona, Florida, Illinois, Michigan, New York, Rhode Island,
Tennessee, and Washington.

Page 141 GAO-22-105051
Financial Problems for Child Care Providers Topped States’ List of Greatest Child Care Challenges in 2020

Note: The challenges shown were those that states most frequently rated as “very” or “extremely challenging” in March 2020.

Restrictions on the capacity of child care during the pandemic, including whether child care
was considered safe to open, added ﬁnancial strain to providers, according to some CCDF
administrators we interviewed. One state administrator told us some providers were unable to
maintain operations due to state-mandated reductions in numbers of children, group sizes, and
ratios of children to staﬀ. Oﬃcials in several states also noted that the pandemic exacerbated
pre-existing challenges with staﬀ recruitment and retention, due to increased health and safety
concerns. State administrators told us that they struggled to appropriately balance the ﬁnancial
struggles of child care providers while also ensuring health, safety, and oversight of public funds.

Federal child care ﬂexibilities. To mitigate COVID-19-related challenges for child care providers
and families, states implemented various federal child care ﬂexibilities, according to our national
survey.196 Of these ﬂexibilities, states most often preserved (1) pay for child care providers,
by paying them based on more generous absence day policies; and (2) subsidies for eligible
children, by increasing the time before their next eligibility redetermination.197 As shown in the
ﬁgure below, however, states sometimes implemented ﬂexibilities only on a short-term basis; by
December 31, 2020, about one-half of states or more no longer used some of them.

196OCC has issued a number of guidance documents to provide information to states on the statutory and regulatory
ﬂexibilities available to them to help respond to challenges caused by the COVID-19 pandemic. Thirty-nine states (80
percent) reported in our survey that they took state action (i.e., statutory or regulatory changes) to take advantage of
available child care ﬂexibilities.
197States are responsible for administering funding provided under CCDF, and, within broad federal requirements,
states generally have discretion to determine (1) child care subsidy eligibility; (2) family copayment contribution
requirements; and (3) payment rates and practices for eligible providers (e.g., paying providers based on children’s
enrollment rather than attendance, which may ﬂuctuate because of absences due to COVID-19).

Page 142 GAO-22-105051
Federal Child Care Flexibilities States Most Commonly Used During COVID-19

Note: This ﬁgure shows federal child care ﬂexibilities implemented due to COVID-19 by at least 50 percent of states.

State CCDF administrators we interviewed said the use of federal ﬂexibilities was vital to mitigate
impacts of the pandemic. One administrator called the ﬂexibility to change how states could pay
providers (e.g., based on enrollment instead of attendance) the most useful. Others noted how
their states’ use of certain ﬂexibilities changed over time. For example, oﬃcials in one state said
that from March through June 2020, the state paid providers based on enrollment, but since July
2020, it has continued to pay based on enrollment only if families attend child care at least 50
percent of the time.

Funding to respond to COVID-19. In our national survey, states most frequently reported
using or planning to use CARES Act and Consolidated Appropriations Act, 2021, Division M (CAA)
funds to help address their greatest COVID-19 related ﬁnancial challenges.198 Speciﬁcally, most
states reported that they used or planned to use these funds to provide assistance to child care
providers experiencing temporary closures or decreased enrollment, or to provide assistance
to child care providers not receiving CCDF as of March 1, 2020 (see table). One administrator we
interviewed said, for example, that being able to provide CARES Act funds to providers—whether
they had previously received CCDF funds or not—helped to ensure as many providers as possible
remained open during the pandemic.

198The American Rescue Plan Act of 2021 (ARPA) was enacted in March 2021, after our survey had been sent to state
child care administrators.

Page 143 GAO-22-105051
State Uses or Planned Uses of CARES Act and Consolidated Appropriations Act, 2021, Division M (CAA) Funds
Number of states that
reported using CARES Number of states that
Uses of CARES Act or CAA funds Act funds planned to use CAA funds

Provide assistance to child care providers experiencing 46 46
temporary closures or decreased enrollment due to
COVID-19

Provide assistance to child care providers not receiving 42 42
Child Care and Development Fund funding as of March 1,
2020

Provide child care assistance to essential workers 29 15
regardless of income

Pay two child care providers for the same child for the 24 11
a
same time period

Support child care resource and referral agencies 14 19

Support family child care network(s) as a means to increase 8 13
supply of home-based child care providers

Source: GAO Survey of State Child Care and Development Fund Administrators, 2021. | GAO-22-105051

Note: The CAA was enacted in December 2020, a few weeks before our survey was deployed, and the American Rescue Plan Act
of 2021 was enacted in March 2021, after our survey had been sent to state child care administrators.
a
OCC guidance states that CARES and CAA funds can be used to pay two providers for the same child should one of the
providers be temporarily closed due to COVID-19.

OCC provides guidance to states regarding ﬂexibility to spend their annually appropriated CCDF
funds during times of national or state emergency, and 29 states reported in our national survey
that they also used CCDF funds intended to improve the quality of child care services to respond
to the pandemic.199 States reported using these funds in various ways, including to help providers
obtain critical supplies and personal protective equipment, complete health and safety training,
and obtain ﬁnancial assistance, through grants or direct payments, as a result of decreased
enrollment or temporary closures. In deciding when to use CARES Act or CCDF funds, state
administrators we interviewed said they considered various factors, such as when funding was
available, allowable uses, and the period in which funds must be spent.200

Paycheck Protection Program. The Paycheck Protection Program (PPP) was also a large source of
federal ﬁnancial assistance for the child care market. According to our analysis of SBA data, child

199Oﬃce of Child Care, Information Memorandum (IM), Flexibility in Spending CCDF Funds in Response to Federal or State
Declared Emergency Situations, CCDF-ACF-IM-2017-02 (Washington, D.C.: Dec. 6, 2017). CCDF quality funds are typically
used by states to improve the quality of child care services and increase parental child care options provided in the state.
See 42 U.S.C. § 9858e.
200States have up to September 30, 2022, to obligate CARES Act funds appropriated for the Child Care and Development
Block Grant program and must spend these funds by September 30, 2023.

Page 144 GAO-22-105051
care providers received more than $5.5 billion in PPP loans. In total, 97,965 child care providers
received 133,100 PPP loans, according to our analysis.201

Small child care providers received more PPP loans than larger providers, but a smaller share of
the overall amount loaned, according to our analysis of SBA data. Providers that reported having
zero to two employees received about 60 percent of the loans whereas larger providers with six
or more reported employees received about 33 percent of the loans (see ﬁg.).202 In total, however,
larger providers received about $4.4 billion of the approximately $5.5 billion loaned to child care
providers through PPP, likely because the amount of the loan a borrower was eligible to receive
was based on the size of their payroll. The median loan amount providers received ranged from
about $10,200 for providers with zero or one employee to about $91,600 for providers with 11 or
more employees.

201Eligible businesses could have received a second PPP loan, or “second draw.” Of the 97,965 providers that received a
PPP loan, about a third, or 35,135, received a second draw. Canceled loans were excluded from our analysis. SBA oﬃcials
said child care providers would have been eligible for PPP loans if they met all program eligibility requirements. For
instance, entities eligible for PPP loans included small businesses that met applicable SBA small business size standards,
independent contractors, eligible self-employed individuals, sole proprietors, and businesses with no more than 500
employees. The entities must have been in operation and had paid employees or have been eligible self-employed
individuals, independent contractors or sole proprietorships with no employees on February 15, 2020. PPP borrowers
were eligible for second draw loans if they met several conditions, including having no more than 300 employees and
experiencing reductions in revenue in 2020 relative to 2019. We did not identify how many providers applied for, but did
not receive, PPP loans because SBA oﬃcials said they could not provide these data as they reside with individual lenders.
However, ACF’s Oﬃce of Planning, Research and Evaluation (OPRE) is collecting nationally representative data from child
care providers about whether they applied for and received (or did not receive) PPP loans. According to OPRE oﬃcials,
results from their ﬁrst wave of data collection should be publicly available in fall 2021.
202Borrowers were asked to report the number of employees they had as part of their PPP loan application and may
have reported zero if they were, for example, sole proprietors, independent contractors, or self-employed individuals,
although SBA application instructions stated that these borrowers should enter one for number of employees.

Page 145 GAO-22-105051
Small Child Care Providers Received More Paycheck Protection Program Loans than Larger Providers

Note: Canceled loans were excluded from our analysis. Borrowers may have reported having zero employees if they were, for
example, sole proprietors, independent contractors, or self-employed individuals, although SBA application instructions stated
that these borrowers should enter one for number of employees.

We also found that areas with high minority populations or high poverty received more PPP loans
relative to their representation within the overall U.S. population.203 Speciﬁcally, providers in areas
with high minority populations, which include about 58 percent of the overall U.S. population,
received about 67 percent of the PPP loans. High poverty areas, which include about 14 percent of
the overall U.S. population, received about 20 percent of the PPP loans.204

OCC oversight. OCC has adapted or plans to adapt existing oversight practices to help ensure
the accountability of COVID-19 supplemental funding. For instance, OCC added a column to its
quarterly ﬁnancial reporting form to capture information on state CARES Act expenditures, and
oﬃcials said they will do the same for state CAA and American Rescue Plan Act of 2021 (ARPA)
expenditures.205 OCC also modiﬁed annual administrative reporting requirements and asked
states to estimate, among other things, how many families and children were served whose
subsidy was fully or partially paid using CARES Act funds.

203We use “areas” to refer to U.S. census tracts in this enclosure. Census tracts are statistical subdivisions of counties
whose boundaries follow geographic features, such as streams, highways, railroads, and legal boundaries, and that
generally contain between 1,200 and 8,000 people. For our analysis, we deﬁned high-minority areas as census tracts
with a minority population of at least 26 percent and high-poverty areas as census tracts where 20 percent or more of
the population lived below the poverty line.
204Our analysis did not control for the extent to which child care providers may be concentrated in areas with high or
low poverty or minority populations.
205In addition, nonfederal entities (states, U.S. territory and tribal governments, local governments, or
nonproﬁtorganizations) that expend $750,000 or more in federal awards in a ﬁscal year are required to undergo a single
audit—that is, an audit of the entity’s ﬁnancial statements and federal awards, or a program-speciﬁc audit, for the ﬁscal
year.

Page 146 GAO-22-105051
OCC oﬃcials said they plan to also collect other information speciﬁc to CAA and ARPA funds, as
appropriate. For instance, OCC oﬃcials said they will collect information about the number and
characteristics of providers that receive an ARPA child care stabilization grant, and are seeking
public comment.206 OCC is also seeking public comment to collect information on state uses of
CCDF quality funds, including funds designated speciﬁcally for quality infant and toddler care
and ARPA child care stabilization grants. Additionally, OCC oﬃcials said they plan to discuss state
uses of COVID-19 supplemental assistance during upcoming on-site or virtual monitoring visits
beginning October 2021.207

Methodology
To conduct this work, we reviewed relevant federal laws and agency guidance and interviewed OCC
and SBA oﬃcials, and CCDF administrators in eight states. We selected these states based on the
prevalence of COVID-19 among adults and children and for geographic diversity. Additionally, we
surveyed state CCDF administrators in 50 states and the District of Columbia between January and
March 2021, and received responses from all but one state. We also analyzed SBA PPP loan data
and used U.S. Census Bureau, American Community Survey data (2015 through 2019, the most
recent data available) to identify census tracts with certain demographics. We determined that
these data were suﬃciently reliable by interviewing federal oﬃcials and performing data checks to
identify any missing data, outliers, or errors.

Agency Comments
We provided HHS, SBA, and the Oﬃce of Management and Budget (OMB) with a draft of this
enclosure. HHS provided technical comments, which we incorporated as appropriate. SBA and
OMB did not provide comments on this enclosure.

GAO’s Ongoing Work
We will continue to review states' implementation of the various coronavirus relief and recovery
packages to identify long-term strategies for improving the child care industry and supporting
child care businesses, including the use of grants and/or contracts, improving payment practices,
and strategies to recruit and retain the workforce.

Contact Information: Kathryn A. Larin, (202) 512-7215, larink@gao.gov

206ARPA included approximately $24 billion in funding for child care stabilization grants. According to OCC, states must
provide at least 90 percent of these funds to qualiﬁed child care providers to support the stability of the child care sector
during and after the COVID-19 public health emergency.
207OCC has a 3-year monitoring cycle and divides states (including the District of Columbia and Puerto Rico) into three
cohorts with about 17 or 18 states per cohort. One cohort is visited each year, either on-site or virtually.

Page 147 GAO-22-105051
K-12 Education
Disruptions due to the COVID-19 pandemic led hundreds of thousands of students
nationwide—primarily vulnerable students, such as those who are English learners or low-
income—to miss days or weeks of virtual instruction or not show up for school at all during the
2020-2021 school year.

Several states conducted outreach to locate and re-engage these disconnected students. As
education is primarily a state and local responsibility, these eﬀorts undertaken by selected states
may provide insights for educators nationwide attempting to reach these students, which is
especially important as the pandemic continues to aﬀect schools during the 2021-2022 school
year.

Entity involved: The Oﬃce of Elementary and Secondary Education within the U.S. Department of
Education (Education).

Background
The COVID-19 pandemic signiﬁcantly disrupted the lives of students, families, and teachers
nationwide, and the consequences may be felt for some time. For example, when school buildings
closed during the 2020-2021 school year, many students, especially low-income students,
did not have the devices and internet access they needed for virtual (online) learning. Even
though many schools have provided students with computers and internet access to participate
in virtual instruction, many students faced diﬃculties staying engaged in school. As a result,
hundreds of thousands of students—possibly as many as three million students, according to
one estimate—missed days or weeks of instruction or disappeared from school altogether.208
Research and media reports have identiﬁed many reasons why students disconnect from school,
such as having parents who are frontline workers and cannot stay home to help them with virtual
learning, or the student having responsibility to assist younger siblings.

To help address the impact of the pandemic, Education has distributed to states COVID-19 relief
funds that can be used for a broad range of needs, including eﬀorts to address learning loss
and engage students in virtual instruction.209 In addition, Education has established the Student
Engagement and Attendance Center, which supports states and school districts in a variety of

208Bellwether Education Partners, Missing in the Margins: Estimating the Scale of the COVID-19 Attendance Crisis, accessed
August 9, 2021, https://bellwethereducation.org/publication/missing-margins-estimating-scale-covid-19-attendance-
crisis.
209As we previously reported, Education has been tracking how states and territories are spending COVID-19 relief
funds and according to the department as of August 31, 2021, states (including D.C. and Puerto Rico) had spent about
$17 billion of the $197 billion in funding for the Elementary and Secondary School Emergency Relief (ESSER) Fund and
Governor’s Emergency Education Relief (GEER) Fund appropriated under three COVID-19 relief laws. Federal spending
data alone provide an incomplete picture of the status of funds, as there are several factors that inﬂuence spending
rates. For example, when school districts use available funds, school district oﬃcials said they have to budget for and
obligate these funds before they are permitted to request payment from the state, which is when Education recognizes
the funds as spent. This process can result in a signiﬁcant lag between the rate at which the funds are being obligated,
or used, and when a state reports it has spent the funds. According to Education oﬃcials, Education plans to modify its
annual report on state and school district spending data to include obligations data in the subsequent annual reporting
cycle.

Page 148 GAO-22-105051
activities, such as identifying strategies to reengage with students and families to facilitate learning
recovery.

Overview of Key Issues
The pandemic caused or exacerbated many inequalities faced by students, making it more likely
for some to disengage from school, particularly those who were already vulnerable. Public schools
nationwide experienced signiﬁcant declines in student enrollment during the 2020-2021 school
year, according to the National Center for Education Statistics, with some students shifting to
homeschooling.210 States used a variety of eﬀorts to locate unaccounted for students and re-
engage them, but incomplete contact information for some students made this more diﬃcult.

Vulnerable students were more likely to disengage from school. Oﬃcials in our four selected
states said that certain groups of vulnerable students—such as students with disabilities, English
learners, and those from low-income families—more often disengaged from school due to
pandemic related barriers. (See table.)

210In June 2021, the National Center for Education Statistics reported that preliminary data show that public schools
across 49 states (excluding Illinois) and the District of Columbia experienced a 3 percent decrease in student enrollment
in the 2020-2021 school year.

Page 149 GAO-22-105051
Examples of Vulnerable Student Groups and Barriers to Staying Engaged in School During the COVID-19
Pandemic, According to Oﬃcials from Selected States, School Year 2020-2021
Student group Examples of challenges oﬀered by state oﬃcials

Disabilities It was more diﬃcult during the pandemic for some students to receive special
education services, such as physical therapy and speech therapy, which are typically
provided in person. These services were oﬀered online but are more diﬃcult to
a
provide in an online setting.

English Learners Some parents of students for whom English is not their ﬁrst language had additional
diﬃculty working with schools and school staﬀ to support their students. The
language barrier meant parents did not always know how to ask for help to address
b
problems with learning at home.

Foster Care Some students in foster care did not want to participate in virtual classes or turn on
their device’s camera because their foster facilities were noisy, distracting, or messy,
and potentially embarrassing.

Homeless Some students experiencing homelessness were very mobile during the pandemic,
including students who moved around in families, shelters, and potentially out of
state. These students often lacked access to or had spotty connections to the internet,
devices, and a quiet place to learn and complete assignments.

Low-income Many students in low-income families disengaged from school during the pandemic
to take jobs and handle other responsibilities to support their families. In some cases,
parents were in poor health or lost their jobs, meaning students had to work or care
for siblings.

Native American Some Native American students did not have reliable internet connectivity or access
c
to direct educational support they normally received at school.

Source: GAO analysis of information from state oﬃcials in California, Mississippi, New Mexico, and South Carolina. | GAO-22-105051

a
We found that a variety of factors complicated the delivery of special education services to students during the COVID-19
pandemic. These factors include the wide range of needs of students with disabilities served under the Individuals with
Disabilities Education Act (IDEA); the services speciﬁed in their individualized education programs; and the capacity of parents
or caregivers to assist teachers and service providers in delivering general education, specialized instruction, and related
services to their children. See GAO, Distance Learning: Challenges Providing Services to K-12 English Learners and Students with
Disabilities during COVID-19, GAO-21-43 (Washington, D.C.; Nov. 19, 2020).
b
We found that some English learners and their families had diﬃculty fully participating in distance learning during spring
2020 due to a lack of necessary technology, language barriers, and the demands of meeting basic family needs. Also, limited
English comprehension aﬀected the ability of families to assist students with the curriculum, according to representatives of
professional associations and a technical assistance center. See GAO-21-43.
c
Bureau of Indian Education-funded schools faced additional pandemic diﬃculties related to distance learning and internet
connected devices during the 2020-2021 school year. See GAO, Indian Education: Schools Need More Assistance to Provide Distance
Learning, GAO-21-492T (Washington, D.C.: Apr. 28, 2021).

States reported outreach eﬀorts to locate and re-engage disengaged students. Oﬃcials in
our selected states said they took a variety of approaches to locate and re-engage disengaged
students (see table). State oﬃcials stressed the importance of relationships between students
and teachers as a key factor in re-engaging students who were not participating in virtual learning
during the pandemic. In addition, representatives from a non-proﬁt organization that works with
disengaged students said state leadership in developing and implementing outreach eﬀorts

Page 150 GAO-22-105051
is important, as many school districts do not have the resources. However, incorrect contact
information for families that was outdated due to the pandemic made it more diﬃcult to locate or
work with students, according to state oﬃcials and the non-proﬁt organization.211

211States with data systems across all school districts made obtaining and providing contact information easier than it
was for states with data kept by individual districts, according to representatives from a non-proﬁt organization.

Page 151 GAO-22-105051
Examples of Reported Eﬀorts to Engage K-12 Students Disconnected by COVID-19 Pandemic during the 2020-2021
School Year, in Selected States
California Mississippi New Mexico and South Carolina

Eﬀort Learning Continuity and School Attendance Academic Coaches
Attendance Plans Oﬃcers

Description California asked school Mississippi employs New Mexico and South Carolina separately contracted
districts to create school attendance with a non-proﬁt organization to conduct outreach
plans detailing their oﬃcers who attempt to and re-engagement work with disconnected students.
engagement eﬀorts locate and re-engage This program provides students with a personal
during the pandemic, disconnected students academic coach to help them overcome social,
including eﬀorts to and were asked to emotional, and academic barriers as well as answer
engage disconnected focus on these students questions about technology and curriculum, and
students. during the pandemic.
connect students to community support.
This information included Attendance oﬃcers take
Academic coaches can connect with students
how the school district a team approach and
by phone, email, and social media and interview
was going to accelerate work with schools and
students about the barriers they face to learning.
learning, close the student families as part
technology gap, provide of the process. Oﬃcials in both states said they initially targeted
virtual learning, support students who had the highest need, and they
teachers and provide reported that as they saw the eﬀort beneﬁting
social and emotional students, they expanded access to the program.
support to students, and
engage and re-engage
students.

Source: GAO summary of information from state oﬃcials in California, Mississippi, New Mexico, and South Carolina. | GAO-22-105051

Note: States may also refer to disconnected students as chronically absent and academically at risk.

State oﬃcials also shared a number of school-district-level eﬀorts to address unaccounted for and
disengaged students, including virtual and in-person tutoring oﬀered at various times, providing
additional social-emotional support, and partnering with expanded service providers to make
home visits and work with students, such as the Boys and Girls Clubs of America. Some school
districts are providing targeted summer enrichment activities to students, according to state
oﬃcials.

States have seen signiﬁcant declines in public school enrollment, especially in kindergarten.
Public school enrollment in our selected states was down by tens of thousands of students in
the 2020-2021 school year, and oﬃcials in our selected states said that this decline was mostly
due to the pandemic. Oﬃcials in California told us that over 160,000 fewer students (including
over 60,000 fewer kindergarteners) enrolled in the 2020-2021 school year than in the previous
school year. Although enrollments in California’s public schools had been decreasing annually,
oﬃcials said the decline during the pandemic was far greater than in prior years. In all four states,
the largest declines occurred in elementary school grades, in part because many families chose
not to enroll their children in kindergarten, according to the some of the oﬃcials we interviewed.
They said these students will be less prepared for ﬁrst grade, and this condition would place more
pressure on ﬁrst grade teachers to cover missed material. Or, alternatively, the students could
enter as older kindergarteners, which will put additional strain on kindergarten teachers and
resources.

Page 152 GAO-22-105051
Students who left traditional public schools enrolled in private schools or public charter schools,
were homeschooled, or reported no engagement in educational activities, according to oﬃcials
we interviewed and data they provided. For example, oﬃcials in California shared information that
showed parent registrations for private school and homeschool in school year 2020-2021 more
than doubled over the previous year. In New Mexico, homeschool was the most common choice
for families who did not re-enroll in public schools, according to information provided by state
oﬃcials.212

Several thousand students were still unaccounted for toward the end of 2020-2021 school
year. Even with these states’ outreach eﬀorts, oﬃcials in our selected states told us that some
students remained unaccounted for towards the end of the 2020-2021 school year.213 For
example, oﬃcials in Mississippi said that even though, as of December 2020, they had over 2,700
unaccounted for students, that was a signiﬁcant improvement from about 23,000 unaccounted
for students they had at the beginning of the school year. Similarly, New Mexico had 2,010
unaccounted for students as of July 2021 compared to the 12,000 unaccounted for students at the
beginning of the school year.

Methodology
To review how states addressed the engagement of students in virtual learning as a result of the
pandemic, we interviewed state education oﬃcials in California, Mississippi, New Mexico, and
South Carolina. We selected these states based on their enrollment and attendance data related
to disconnected students for the 2020-2021 school year, information about their eﬀorts to engage
disconnected students, and demographic information, including proportion of students from low-
income families, in diﬀerent racial and ethnic groups, and states’ diverse geographic locations.214
We also reviewed related documentation provided by these states. We interviewed representatives
of a non-proﬁt organization that works with state educational agencies to address the needs of
disconnected students, including in New Mexico and South Carolina.

Agency Comments
We provided Education and the Oﬃce of Management and Budget with a draft of this enclosure.
Education provided technical comments, which we incorporated as appropriate. The Oﬃce of
Management and Budget did not provide comments on this enclosure.

212New Mexico oﬃcials said they polled parents about their plans for the 2021-2022 school year and the most common
answer was they were unsure if they would return to public school. Others said they planned to return to public schools,
would continue in private school, or homeschool.
213Oﬃcials in California said they would not have updated numbers of unaccounted for students until the end of the
2021 calendar year. Oﬃcials in South Carolina said that during the end of the 2019-2020 school year they suspended
tracking student attendance because of the pandemic and as a result do not have updated information on the number
of students that remain unaccounted for.
214For low-income families we analyzed data on students eligible for free or reduced price lunch from Education’s
National Center for Education Statistics Common Core of Data 2018-2019 as well as Census Bureau data on the percent
of families below poverty by state in 2019. For racial composition we analyzed data from the Common Core of Data
2018-2019.

Page 153 GAO-22-105051
GAO’s Ongoing Work
We will continue to monitor Education’s eﬀorts to help schools recover from the pandemic, how
states and school districts are using the COVID-19 relief funds, and the challenges of pandemic-
related learning loss and the approaches educators are ﬁnding to eﬀectively address it.

Related GAO Products
Indian Education: Schools Need More Assistance to Provide Distance Learning. GAO-21-492T.
Washington, D.C.: Apr. 28, 2021.

Distance Learning: Challenges Providing Services to K-12 English Learners and Students with Disabilities
during COVID-19. GAO-21-43. Washington, D.C.: Nov. 19, 2020.

Contact information: Jacqueline M. Nowicki, (617) 788-0580, nowickij@gao.gov

Page 154 GAO-22-105051
Child Nutrition
The Food and Nutrition Service and states have used a variety of approaches to oversee child
nutrition programs during the COVID-19 pandemic. However, states identiﬁed ongoing challenges
with overseeing these programs, and the Food and Nutrition Service may be missing opportunities
to fully leverage lessons learned from the pandemic to improve the management of child nutrition
programs.

Entity involved: Food and Nutrition Service, within the Department of Agriculture

Recommendation for Executive Action
The Secretary of Agriculture should document the Department of Agriculture’s plan to analyze
lessons learned from operating child nutrition programs during the COVID-19 pandemic. This plan
should include a description of how the department will gather perspectives of key stakeholders,
such as Child and Adult Care Food Program institutions and nonschool Summer Food Service
Program sponsors. The Department of Agriculture generally concurred with this recommendation.

Background
Child nutrition programs administered by the Department of Agriculture’s (USDA) Food and
Nutrition Service (FNS) supply cash reimbursements to schools or other programs for meals and
snacks they provide to eligible children. In ﬁscal year 2019, before the pandemic, the four largest
programs—the National School Lunch Program (NSLP), School Breakfast Program (SBP), Summer
Food Service Program (SFSP), and Child and Adult Care Food Program (CACFP)—along with other
child nutrition programs, received $23.1 billion in federal funds.215

During a typical year, NSLP and SBP subsidize meals for nearly 30 million children in
approximately 95,000 elementary and secondary schools nationwide. These two programs are
the largest of the child nutrition programs and typically serve children at school during the school
year. In addition, SFSP and the Seamless Summer Option (SSO) typically provide meals for school-
age children during the summer months.216 Finally, CACFP provides meals to younger children
enrolled for care at participating child care centers and day care homes and to school-age children
participating in CACFP At-Risk Afterschool programs.217

215This review includes NSLP, SBP, SFSP, and CACFP. The Special Supplemental Nutrition Program for Women, Infants,
and Children is not included in this review.
216State agencies may approve public or private nonproﬁt school district nutrition programs or organizations to
participate in SFSP. SSO allows school districts to operate a modiﬁed version of NSLP and SBP in the summer or during
unanticipated school closures.
217State agencies enter into agreements with CACFP institutions, which are independent centers or sponsoring
organizations of day care homes or child care centers that assume responsibility for CACFP operations. CACFP also
provides reimbursement for meals served to children who are residing in participating emergency shelters and to adults
older than 60 years and functionally impaired adults who are enrolled in day care facilities.

Page 155 GAO-22-105051
The Families First Coronavirus Response Act (FFCRA) granted FNS authority to issue nationwide
waivers in certain programs for speciﬁc purposes.218 As we reported in July 2021, FNS extended
several nationwide waivers in April 2021 for the 2021–22 school year.219 FNS also issued a pair
of waivers to allow schools to operate SSO during the school year and to claim SSO meals at the
higher SFSP reimbursement rate. These waivers are intended to support access to nutritious
meals, reduce the administrative burden associated with eligibility determinations, and minimize
potential exposure to COVID-19.220

FNS’s National Oﬃce is responsible for providing regulatory guidance, policy materials, and
monitoring tools to its seven regional oﬃces, which have the primary responsibility for oversight
of state agencies administering child nutrition programs. Typically, regional oﬃces monitor
program compliance through management evaluations. In the course of management evaluations
during typical years, regional oﬃcials review program areas through a combination of oﬀ-site and
on-site monitoring activities.

At the state level, state agencies—generally education or agriculture agencies—administer the
programs and issue guidance to school district nutrition programs and other local program
operators.221 The state agencies responsible for child nutrition oversee school meal programs,
which includes conducting administrative reviews of local operators’ administration of such
programs. These reviews must include the accuracy of meal counting and claiming, nutritional
quality, resource management, and other focus areas. State agencies also conduct regular reviews
of local program operators’ administration of CACFP and SFSP. School district nutrition programs
and other local program operators are responsible for certifying students as eligible for free or
reduced-price meals and for counting and claiming eligible meals for federal reimbursement,
among other monitoring activities.

FNS has issued several nationwide waivers related to program monitoring activities during the
pandemic to facilitate state and local monitoring of the child nutrition programs while allowing
for social distancing for staﬀ. Most recently, FNS issued three waivers allowing state agencies and
local operators to conduct monitoring entirely oﬀ-site, rather than both oﬀ-site and on-site, for
the school meal programs (NSLP, SBP, and SSO) and CACFP until 30 days after the end of the public
health emergency.

Various COVID-19 relief laws have provided funding or authority to USDA to support child nutrition
programs during the pandemic. For example:

218Pub. L. No. 116-127, § 2202(a), 134 Stat. 178, 185 (2020).
219According to FNS, although most of these waivers are available through June 30, 2022, FNS expects that the non-
congregate feeding, meal time ﬂexibility, and parent or guardian pick-up waivers will be used only for the duration and
extent needed, as schools and child care providers work to safely and successfully reopen.
220For school year 2020–21, schools were allowed to operate either SFSP or SSO rather than the traditional school
year programs. For school year 2021–22, schools will be able to operate SSO but not SFSP. According to FNS, SSO has
stricter nutrition standards than SFSP and is typically reimbursed at the same rate as NSLP, while SFSP has a higher
reimbursement rate. During school year 2021–22, schools operating SSO will be reimbursed at the SFSP rate.
221Throughout this enclosure, “school district nutrition program” refers to a school food authority—that is, the local
authority responsible for operating school meal programs. Local program operators are school district nutrition
programs, sponsors, or institutions that operate NSLP, SBP, SSO, SFSP, or CACFP.

Page 156 GAO-22-105051
• The Families First Coronavirus Response Act (FFCRA), enacted in March 2020, authorized and
provided an indeﬁnite appropriation for a new program, Pandemic Electronic Beneﬁts Transfer
(Pandemic EBT), which provides beneﬁts to purchase food to households with children who
would have received free or reduced-price school meals if not for school closures due to
COVID-19.222 The program also provides these beneﬁts to households with eligible children in
child care. According to FNS, during the summer of 2021, Pandemic EBT beneﬁts were oﬀered
to all eligible children who resided in states with approved Summer Pandemic EBT plans.223 As
of August 31, 2021, FNS had obligated $34.432 billion for Pandemic EBT.

• The CARES Act, enacted in March 2020, provided $8.8 billion in supplemental funds.224 As
of August 31, 2021, FNS had obligated nearly all of this funding for child nutrition programs.
According to FNS, it provided nearly all of this funding to states and other meal program
operators and used the majority of the funds—$8.615 billion—to reimburse operators for the
cost of meals served during the pandemic.225

• The Continuing Appropriations Act, 2021 and Other Extensions Act, enacted in October 2020,
extended certain waiver authority granted in the FFCRA through September 2021 and provided
an indeﬁnite appropriation to cover the costs incurred as a result of the waiver extensions.226
As of August 31, 2021, FNS had obligated $1.470 billion of this funding for child nutrition
programs.

• The Consolidated Appropriations Act, 2021, enacted in December 2020, provided an indeﬁnite
appropriation to support CACFP institutions and school district nutrition programs that
replaced some of the decline in reimbursement funding in spring 2020.227

Overview of Key Issues
School districts and other meal program operators served fewer meals during the ﬁrst year
of the pandemic than in the previous year, but the number of meals served in spring 2021
approached prepandemic levels. According to the most recent available data from FNS, during

222Pub. L. No. 116-127, § 1101, 134 Stat. 178, 179-180 (2020).
223According to FNS, Summer Pandemic EBT was oﬀered to (1) income-eligible school children who were enrolled in
a school that participated in NSLP in school year 2020–21 and (2) enrolled SNAP recipients who were younger than
6 years or who were otherwise enrolled in an eligible child care institution, as deﬁned by the FFCRA.
224Pub. L. No. 116-136, div. B, tit. I, 134 Stat. 281, 507 (2020).
225FNS used the remainder of the funds, $185 million, to operate Emergency Meals-to-You, a new partnership that
delivered meals to address pandemic-related nutrition needs among children from low-income households in rural
areas throughout spring and summer 2020. See our July 2021 report for more information about this program.
226Pub. L. No. 116-159, § 4602(a), (d), 134 Stat. 709, 745 (2020).
227Pub. L. No. 116-260, div. N, tit. VII, § 722, 134 Stat. 1182, 2097 (2020). This law provided an indeﬁnite
appropriation of funds, based on a formula that generally takes into account the diﬀerence between
reimbursements paid from March through June 2019 and those paid from March through June 2020. According
to FNS, because the Consolidated Appropriations Act, 2021, extends the authority under the Continuing
Appropriations Act, 2021 and Other Extensions Act, obligations and expenditures under the Consolidated
Appropriations Act, 2021, are accounted for under the Continuing Appropriations Act, 2021 and Other Extensions
Act.

Page 157 GAO-22-105051
the ﬁrst year of the pandemic (March 2020 through February 2021), school districts and other
meal program operators operating NSLP, SBP, SFSP, and CACFP served 2.8 billion fewer meals
than in the prior year—an overall drop of 30 percent.228 The decline in meals served during the
pandemic’s ﬁrst year was not uniform among meal types. Compared with the number of meals
served in the prior year, the number of lunches served through NSLP, SFSP, and CACFP dropped
by 40 percent while the number of breakfasts served through SBP, SFSP, and CACFP dropped by
only 14 percent. FNS oﬃcials attributed the smaller decline in breakfasts served to the packaging
of multiple meals at a time for grab-and-go service.229

The numbers of meals served during March and April 2021, as the pandemic entered its second
year, were closer to the numbers of meals served during March and April 2019, a year before the
pandemic began (see ﬁgure). Speciﬁcally, in April 2021, as vaccines became widely available for
American adults and more schools oﬀered in-person learning, the number of meals served was
only 9 percent lower than in April 2019.230

228As we reported in July 2021, the drop in meals served was not uniform among the child nutrition programs. While
meals served under NSLP, SBP, and CACFP dropped, the total number of meals served under SFSP increased during the
pandemic.
229In the year before the pandemic began, the number of lunches served was nearly twice the number of breakfasts
served, according to FNS data.
230Data for meals served in April 2021 are the most recent data available as of September 10, 2021 that are suﬃciently
reliable for our purposes. According to FNS, state agencies submit monthly meal-claim reports to FNS; initial monthly
tabulations reported 30 days after the end of the claim month include estimated data based on the previous year.
However, the uncertainty of meal service during the COVID-19 pandemic has made it diﬃcult for states to use historical
data to report estimates of meals served, according to FNS. Data reported by states to FNS 90 days after the end of the
claim month are based on actual meal claims rather than on estimates. We determined that 90-day data are suﬃciently
reliable for our purposes. According to FNS, these data are subject to revision.

Page 158 GAO-22-105051
Total Meals Served in Key Child Nutrition Programs in 2019, 2020, and 2021, by Month, as of Sept. 10, 2021

Notes: The monthly totals include four child nutrition programs: National School Lunch Program, School Breakfast Program,
Summer Food Service Program, and Child and Adult Care Food Program (child meals only). Totals also include meals served
through the Seamless Summer Option, a program that allows school districts operating the National School Lunch Program
and School Breakfast Program to continue using the same meal service rules and claiming procedures as in the regular school
year throughout the summer and during unanticipated school closures. According to the Food and Nutrition Service (FNS), the
number of meals reported for any given month is subject to marginal revisions over time for a variety of reasons, including late
claims and changes resulting from routine monitoring activity.
The totals shown are for meals served each month before and during the pandemic. Pandemic-related disruptions to school
meal programs began in March 2020. As we reported in July 2021, meals served during June and July 2020 were higher than
in the same months in 2019. In 2020, some districts and other providers were able to provide meals throughout the summer
without congregate feeding (i.e., by using a grab-and-go model), which allowed some areas to increase the overall number of
meals served in the summer months in 2020 compared with prior years.
Data for meals served in April 2021 are the most recent data available as of September 10, 2021 that are suﬃciently reliable for
our purposes. According to FNS, state agencies submit monthly meal-claim reports to FNS; initial monthly tabulations reported
30 days after the end of the claim month include estimated data based on the previous year. However, the uncertainty of meal
service during the COVID-19 pandemic has made it diﬃcult for states to use historical data to report estimates of meals served,
according to FNS. Data reported by states to FNS 90 days after the end of the claim month are based on actual meal claims
rather than on estimates. We determined that 90-day data are suﬃciently reliable for our purposes. According to FNS, these
data are subject to revision.

FNS oﬃcials told us that the numbers of meals served approached prepandemic levels during
March and April of the 2020–21 school year because more schools were feeding more children

Page 159 GAO-22-105051
who were attending school in person rather than virtually. In August 2021, oﬃcials from two of the
four states where we conducted interviews told us that their states had mandated that all schools
provide in-person learning during school year 2021–22, while oﬃcials from two other states told
us that schools could decide whether to oﬀer a virtual instruction option. FNS oﬃcials anticipated
that the increase in students attending traditional, in-person schooling in 2021–22 would result
in more meals served than during school year 2020–21. Although the most recent data show the
numbers of meals served during March and April 2021 were closer to prepandemic levels, the
extent to which meals were served to low-income students is not known because of the expanded
eligibility for free meals during the pandemic.231

FNS and state agencies used a variety of monitoring approaches while balancing competing
priorities, and state agencies identiﬁed ongoing challenges to maintaining program
integrity. Recognizing that child nutrition programs may operate diﬀerently during the pandemic
given the numerous ﬂexibilities and waivers provided—including nationwide waivers that allow all
monitoring to be conducted oﬀ-site—FNS allowed states to waive traditional program monitoring
requirements. This ﬂexibility was provided to state nutrition oﬃces that provided a waiver request
to FNS that included an alternative oversight plan that ensured continued program integrity.
According to FNS oﬃcials, FNS had approved more than 60 of these oversight plans as of June
2021.232 FNS provided state nutrition oﬃces with a framework and template for these plans, and
oﬃcials from two of the four FNS regional oﬃces where we conducted interviews reported helping
state nutrition oﬃces in their region develop their plans, either directly or by facilitating best
practice–sharing sessions with other states. According to FNS oﬃcials, much of the state oversight
during the pandemic has taken the form of technical assistance.

FNS oversight and monitoring. To ensure that state nutrition oﬃces were following their FNS-
approved plans and implementing waivers correctly, the FNS National Oﬃce asked FNS regional
oﬃces to monitor, in real time, the state oﬃces’ implementation of the plans. Regional oﬃcials
conducted real-time monitoring, known as touchpoints, for each state nutrition oﬃce by
participating in three oversight activities per child nutrition program (NSLP, SFSP, and CACFP) and
then providing a brief report of the activities to the FNS National Oﬃce. For example, if a state’s
oversight plan said the state would oﬀer webinars to program operators operating CACFP under
the nationwide waivers, a regional oﬃcial would attend the webinar and, if any of the information
conveyed in the webinar was incorrect, provide technical assistance to the state nutrition oﬃce
to correct the information. According to FNS oﬃcials, this monitoring was important because it
allowed the regional oﬃcials to provide real-time technical assistance as the pandemic evolved.

According to FNS oﬃcials, in addition to conducting the real-time monitoring, FNS regional oﬃces
continued to conduct traditional, retrospective management evaluations during the pandemic.
However, owing to capacity constraints and pandemic complications, they reported conducting
fewer management evaluations than they would in a typical year, targeting areas that were still
applicable during the pandemic and focusing on speciﬁc areas of concern. To determine which

231In spring 2020, FNS began allowing schools and other meal operators to operate under summer meal programs. In
addition, FNS waived the requirement that to provide free meals to all children, summer meal sites must be located in
areas in which at least half the children are from low-income households.
232There are more than 60 state plans because in some states, multiple state agencies operate child nutrition programs.
For example, in some states NSLP and CACFP are operated by diﬀerent agencies. According to FNS oﬃcials, at least one
state agency from most states submitted an oversight plan.

Page 160 GAO-22-105051
states and programs to review, FNS oﬃcials used a risk-based assessment tool to identify those for
which the evaluations were most critical.

According to FNS oﬃcials, all management evaluations were conducted virtually; as a result, some
portions could not be completed. For example, according to oﬃcials from one FNS regional oﬃce,
warehouse reviews—an optional component of management evaluations for NSLP—have been
put on hold during the pandemic to allow for social distancing. Oﬃcials from each state nutrition
oﬃce we interviewed told us they had taken part in virtual management evaluations during
the pandemic; in general, both state and FNS regional oﬃcials said these evaluations went well
despite challenges. Because these reviews are retrospective, FNS oﬃcials from the National Oﬃce
said they were aided by the regional oﬃce touchpoint reviews that provide timelier monitoring
and technical assistance.

State agency oversight and monitoring. In a typical year, state nutrition oﬃces conduct
administrative reviews of a portion of their school district nutrition programs operating NSLP.233
As a result of a ﬂexibility allowing school districts to operate summer meal programs (through SSO
and SFSP) rather than operating traditional school meal programs (through NSLP and SBP) during
the pandemic, states have conducted fewer administrative reviews than is typical, according to
state nutrition oﬃcials we interviewed. Speciﬁcally, as of June 2021:

• Nutrition oﬃcials in two of the four states said they had conducted no administrative reviews
during the previous school year. In one of the two states, nutrition oﬃcial attributed this to the
fact that none of the state’s school districts had operated NSLP. In the other state, where very
few school districts chose to operate NSLP, state nutrition oﬃcials said they had focused their
eﬀorts on providing technical assistance rather than conducting administrative reviews.

• Nutrition oﬃcials in the two states that conducted administrative reviews of the school
districts that chose to operate NSLP reported conducting more of the review components oﬀ-
site than is typical. According to FNS guidance, strategies for conducting virtual monitoring
include, for example, reviewing and verifying records by observing photographs or videos
in situations where direct observation would normally occur and conducting interviews via
telephone or video conference.

Some state nutrition oﬃcials we interviewed said that, in addition to conducting traditional
monitoring such as administrative reviews, they used various strategies for conducting real-time
oversight during the pandemic as outlined in their oversight plans. For example, oﬃcials from
two state nutrition oﬃces reported providing targeted technical assistance to local operators with
identiﬁed risk factors, such as signiﬁcant recent staﬀ turnover. Speciﬁcally:

• Staﬀ from one state nutrition oﬃce began in spring 2021 to make in-person coordinated
support visits to high-risk school district nutrition programs. During these visits, they
conducted components of administrative reviews with the local program operators to ensure
the operators understood program requirements, including requirements concerning

233State nutrition oﬃces are required to conduct administrative reviews of each school district at least once in a 3-year
review cycle. However, FNS has approved longer review cycles for some states, allowing them, for example, to review
each district every 4 or 5 years.

Page 161 GAO-22-105051
monitoring. Although the state nutrition oﬃce staﬀ conducted most of these review
components virtually before the visits, they conducted a limited portion of the reviews on-site
by observing a meal service.

• Staﬀ from the other state nutrition oﬃce provided one-on-one technical assistance to school
districts where they identiﬁed potential risks—for example, if they noticed meals being served
to numbers outside the expected range or if the school nutrition program had new leadership.
As of June 2021, this state nutrition oﬃce had conducted all of its pandemic monitoring and
technical assistance virtually, using a variety of technology platforms.

According to state nutrition oﬃcials, these forms of technical assistance were intended to
minimize confusion and assist operators in meeting program requirements. Some state nutrition
oﬃcials noted that it can be diﬃcult for local program operators to keep track of changes in
program requirements that resulted from waivers.

Eﬀects of competing priorities on monitoring. Federal and state nutrition oﬃcials we interviewed
said that competing priorities, such as implementing waiver ﬂexibilities and providing technical
assistance, had created challenges to monitoring throughout the pandemic, particularly early
on. For example, oﬃcials from FNS regional oﬃces said that they delayed program monitoring
at the start of the pandemic to help states interpret and implement the various FNS waivers. In
general, FNS regional oﬃcials we interviewed said that they had not seen new types of program
integrity concerns during the pandemic. However, oﬃcials from one regional oﬃce noted that
state nutrition oﬃces are aﬀected by constrained resources, including time.

Additionally, because of a ﬂexibility that allowed districts to operate summer meal programs
during the regular school year, a wider variety of child nutrition programs were operating at the
same time during the pandemic.234 State nutrition oﬃcials from two states said that this had
made monitoring and oversight more diﬃcult because they were not accustomed to so many
diﬀerent child nutrition programs’ operating concurrently.

Challenges and beneﬁts of oﬀ-site monitoring. Nutrition oﬃcials we interviewed at the federal and
state levels reported encountering both challenges and beneﬁts in virtual monitoring. According
to the oﬃcials, insuﬃcient technology made oﬀ-site monitoring diﬃcult, particularly at the start of
the pandemic, when many staﬀ were adjusting to remote work and implementing new technology.
For example, oﬃcials from one FNS regional oﬃce noted that some state nutrition oﬃces were
not set up to telework at the start of the pandemic and did not have systems to forward their
oﬃce phones, which hindered communication. Similarly, virtual desk-audits can require operators
to scan large quantities of documents, which can be time consuming; also, in some instances,
operators did not have access to the technology. Another challenge aﬀecting virtual monitoring
was the lack of physical presence and face-to-face interaction, according to some state and FNS
oﬃcials.

Despite these challenges, in addition to health and safety beneﬁts, in general, nutrition oﬃcials at
the state and federal level said that oﬀ-site monitoring had oﬀered beneﬁts, such as cost savings

234During a typical school year, school districts operate NSLP and SBP. However, because of the pandemic and FNS
ﬂexibilities, school districts in many states have instead operated SFSP and SSO during the school year.

Page 162 GAO-22-105051
and limited travel. One state nutrition oﬃcial we interviewed said that oﬀsite monitoring had
facilitated innovation because it caused the state nutrition oﬃce to reassess and streamline its
monitoring process.

Considerations and ongoing challenges for program monitoring in school year 2021–22. To facilitate
administrative reviews in the 2021–22 school year, FNS issued a waiver in May 2021 that allows
states to conduct administrative reviews of school district nutrition programs operating only SSO
(in addition to those operating NSLP).235 Speciﬁcally, FNS waived certain administrative review
requirements for programs operating only SSO during the school year and allowed the reviews to
count toward state monitoring requirements. According to FNS oﬃcials, because administrative
reviews are conducted on a multiyear cycle, this will help state nutrition oﬃces to fulﬁll state
monitoring requirements, given the April 2021 waiver allowing school districts to operate SSO
during the school year.

According to nutrition oﬃcials in one state, they requested a monitoring waiver to extend their
review cycle time by 1 year because they recognized that their state had conducted few, if any,
administrative reviews in the prior school year and that conducting the necessary number of
reviews in school year 2021–22 might be diﬃcult. Nutrition oﬃcials in a second state said they
were considering requesting a monitoring waiver for the same reasons. One state nutrition oﬃcial
expressed concern that after programs return to normal operations, the high degree of staﬀ
turnover in child nutrition programs and the length of time that programs will have operated
with waivers could result in program integrity issues, such as lack of adherence to program
requirements. According to FNS, allowing administrative reviews to continue for school districts
operating only SSO should help ease the transition back to traditional program monitoring after
the pandemic.

FNS is taking some steps to identify lessons learned for child nutrition programs from the
pandemic, but it may be missing additional opportunities. FNS oﬃcials told us that they are
primarily using an existing FNS School Meals Operations (SMO) study to gather information about
lessons learned during the pandemic for child nutrition programs.236 According to a notice in the
Federal Register, this study will help FNS obtain (1) general descriptive data on characteristics of
the child nutrition programs to inform the budget process and answer questions about topics
of current policy interest; (2) program operations data to identify potential topics for training
and technical assistance for state and school district nutrition programs; (3) administrative data
to identify program trends and predictors; and (4) information on the use and eﬀectiveness of
the child nutrition waivers, which will be used to satisfy states’ reporting requirements on those
waivers under FFCRA.237

235Traditionally, state agencies conduct administrative reviews for school district nutrition programs operating NSLP
and SBP. For additional information about this waiver, see U.S. Department of Agriculture, Food and Nutrition Service,
“Nationwide Waiver to Provide Flexibility for School Meal Programs Administrative Reviews of SFAs Operating Only the
SSO in SY 2021–22,” accessed Aug. 18, 2021, https://www.fns.usda.gov/cn/covid-19-child-nutrition-response-97.
236FNS oﬃcials told us that, in addition to conducting the SMO study, they are assessing pandemic-related challenges
and lessons learned through regularly scheduled meetings with regions, stakeholders, and department oﬃcials.
According to the oﬃcials, they respond to questions and address challenges through webinars, question-and-answer
sessions, and policy guidance, as well as conducting listening sessions with state agencies.
23786 Fed. Reg. 20,654 (April 21, 2021). FNS plans to use the study to fulﬁll states’ reporting requirements pursuant to
section 2202 of FFCRA. FFCRA requires each state that receives a waiver under that section of the law to (1) report on the

Page 163 GAO-22-105051
Launched in spring 2021, the study will collect administrative and survey data on each of the
four child nutrition programs from state agencies and will collect survey data from school district
nutrition programs.238 For example, as part of the state survey, FNS is collecting, as part of the
state survey, perspectives from state agencies regarding state and local operational and ﬁnancial
challenges during the pandemic. FNS stated that because it recognized that the pandemic
changed the way school meal programs operated, it expanded the SMO study’s data collection
eﬀorts—initially planned prior to the pandemic—to include gathering survey and administrative
data from the state agencies that oversee the CACFP and SFSP.239

Agencies can leverage lessons learned from an event to inform future eﬀorts and limit the chance
of recurring challenges. The experience of providing meals to children during the pandemic
presents an opportunity for FNS to assess potential lessons learned for managing child nutrition
programs.240 Although FNS oﬃcials told us that the SMO study will be used to gather information
about lessons learned from the pandemic, as of July 2021, FNS was unable to provide us with a
plan for how it intends to comprehensively analyze lessons learned from the pandemic for child
nutrition programs.241 Further, the Federal Register notice mentioned above does not indicate
whether FNS will analyze lessons learned to address operational and ﬁnancial challenges.

Although FNS is collecting some information on these topics from states, FNS may miss
opportunities to comprehensively identify lessons learned during the pandemic unless it
documents a plan for analyzing them. Further, according to FNS oﬃcials, while the SMO study
will survey state agencies that administer the federal child nutrition programs, the study will not
gather local perspectives directly from CACFP institutions (e.g., child care centers and day care
homes) or SFSP sponsors that are not school districts.242 Without gathering perspectives from a
full range of meal program operators—including CACFP institutions and SFSP sponsors (discussed
below)—rather than only from state agencies and school districts, FNS will lack comprehensive
information to aid its future planning.

District and state child nutrition oﬃcials identiﬁed challenges as well as opportunities
during the pandemic. Nutrition oﬃcials we interviewed from districts, states, and organizations
identiﬁed several challenges and opportunities related to operating child nutrition programs

use of such waiver by the state and eligible service providers and (2) describe whether such waiver resulted in improved
services to children. Pub. L. No. 116-127, § 2202(d), 134 Stat. 178, 185.
238Federal law authorizes FNS to conduct an annual national performance assessment of the school meal programs,
which FNS plans to do through the SMO study in school year 2021–22. According to FNS, the SMO study is divided into
three separate eﬀorts, with data collection beginning in spring 2021 and ending in spring 2023.
239In addition, the SMO study will collect from state agencies, disaggregated data for the Child Nutrition programs.
According to FNS, the data will be used to assess service levels and meal service reach during the waiver periods.
240The use of lessons learned is a principal component of an organizational culture committed to continuous
improvement. Leading practices of a lessons learned process that we and others have identiﬁed include collecting,
analyzing, validating, saving or archiving, and disseminating and sharing information and knowledge gained from
positive and negative experiences.
241According to standards for internal control in the federal government, documentation is a necessary part of an
eﬀective internal control system.
242FNS oﬃcials told us that, rather than contact CACFP and SFSP sites directly, they plan to collect CACFP and SFSP site-
level data from states and to gather perspectives of state agency oﬃcials and school district nutrition programs. FNS
oﬃcials said that many school district nutrition programs operated CACFP and SFSP programs.

Page 164 GAO-22-105051
during the pandemic. Speciﬁcally, the oﬃcials identiﬁed challenges and, in some cases,
opportunities with respect to information technology (IT) systems, food supply and storage, loss
of revenue for local meal operators, waiver rollout and extensions, and the possibility of making
some ﬂexibilities permanent.

IT systems. State oﬃcials we interviewed identiﬁed ﬁnancial and resource burdens and potential
challenges to maintaining program integrity related to their child nutrition IT systems during the
pandemic. For example:

• One state oﬃcial explained that its state child nutrition information system was developed to
be compatible with federal child nutrition programs under normal operations. The oﬃcial said
it took staﬀ time and ﬁnancial resources to update the state’s system to accommodate each
new waiver that FNS announced.

• Oﬃcials in a second state said that IT changes they made to accommodate waivers meant
that built-in program integrity checks no longer functioned correctly. Instead, staﬀ had to
spend additional time conducting manual reviews to check for claims errors and help ensure
program integrity.

• Oﬃcials from the second state also said that institutions at the local level that operate CACFP
are particularly prone to technological challenges. FNS oﬃcials in one regional oﬃce noted
that day care homes that previously used libraries to submit claims electronically often had to
mail hard-copy claims during the pandemic because of extended library closures.

• Oﬃcials from three state agencies said that the pandemic highlighted opportunities for the
federal government to limit the ﬁnancial and resource burdens on their states, such as by
investing in software, establishing a standalone fund to assist states with IT maintenance
costs, and providing communication and technical assistance to states’ IT software vendors.243

Food supply and storage. Oﬃcials from 11 of the 12 districts and three of the four states where we
conducted interviews, and others, identiﬁed food supply or food storage challenges to operating
child nutrition programs during the pandemic. These oﬃcials identiﬁed the following supply
challenges, among others:

• Oﬃcials from two state nutrition programs said that districts experienced cancelations in their
USDA Foods in Schools commodity orders because certain foods were no longer available.244

• Oﬃcials from 11 district nutrition programs told us it was diﬃcult to acquire items such
as individually wrapped foods or certain types of foods, such as milk, breads, and proteins
because of food shortages, competition, or both.

243FNS oﬃcials told us in April 2021 that, with the pandemic still ongoing, they had not considered ways to better
capture data during emergencies but that FNS had provided technology innovation grants to states for system
upgrades. However, nutrition oﬃcials in two states expressed some concerns related to these grants, including
concerns that the grants are short term and add to states’ administrative burden.
244The USDA Foods in Schools program provides USDA-purchased domestic agricultural products to schools and
institutions participating in NSLP, SFSP, and CACFP.

Page 165 GAO-22-105051
• Oﬃcials from the National CACFP Sponsors Association told us that small, rural child care
centers and day care homes may have found that required milk was not available on the day
that they bought food for the week.

In addition, oﬃcials from half of the district nutrition programs in which we conducted interviews
said that limits on storage space had prevented them from ordering food in large quantities or
leveraging their USDA Foods in Schools commodities to oﬀset pandemic-related costs. Oﬃcials
from the Boys and Girls Clubs of America noted that storage space was also a challenge for local
clubs providing food under SFSP and CACFP At-Risk Afterschool programs, especially if they were
not colocated with schools, because they did not have access to additional storage located at
schools.

State nutrition oﬃcials we interviewed suggested opportunities for a federal role in facilitating the
use of USDA Foods in Schools commodities in a national emergency. For example, they suggested
that USDA evaluate the food and menu items that would be needed in an emergency situation,
identify vendors oﬀering those items, and allow increased ﬂexibility in timing of orders.245

Loss of revenue for local meal operators. In March 2021, we reported that a drop in revenue from
meals served and an increase in program costs had caused ﬁnancial challenges for local meal
operators. In April 2021, FNS oﬃcials noted that the decline in meals served had been especially
pronounced in spring 2020, when many institutions that provide meals—schools, child care
centers, and day care homes—were closed. Helping oﬀset these ﬁnancial losses, in December
2020 the Consolidated Appropriations Act, 2021 was enacted, which provided additional funding
for reimbursements for meals served in spring 2020.246 Oﬃcials from three of the four state child
nutrition programs cited challenges related to the process for receiving funding approval from
FNS, noting that the information required was diﬃcult or time consuming to compile.247

One state nutrition oﬃcial we interviewed noted that CACFP institutions, such as child care
centers, were especially in need of this funding because, unlike schools, they were less likely to
have other sources of funding to compensate in the short term. Oﬃcials from the National CACFP
Sponsors Association noted that some child care centers and day care homes, especially smaller
providers, had to close before the funding became available.

Waiver rollout and extensions. As we reported in March 2021, nutrition oﬃcials in nearly all the
districts where we conducted interviews found the waivers helpful in providing needed ﬂexibilities

245FNS oﬃcials told us that they collaborated with stakeholders to cancel orders no longer needed because of menu
changes, freeing up funds for other items. FNS oﬃcials said they continue to provide technical assistance to help states
eﬃciently use food from the USDA Foods Program.
246Pub. L. No. 116-260, div. N, tit. VII, § 722, 134 Stat. 1182, 2097. This law provided an indeﬁnite appropriation of funds
based on a formula that generally takes into account any diﬀerences in reimbursements paid from March through June
2019 and reimbursements paid from March through June 2020.
247FNS established two temporary programs, one for schools and one for CACFP programs, to provide the
reimbursement funds. To participate in these programs, state agencies were required to submit an application to FNS by
February 25, 2021, and an implementation plan by April 26, 2021, showing how the state would calculate and disburse
payments. In June 2021, state nutrition oﬃcials for the four states we interviewed said that they had not received these
funds from FNS, so they had not been able to disburse funds to local program operators. However, FNS reported that as
of August 11, 2021, it had disbursed funds to 52 state agencies.

Page 166 GAO-22-105051
to serve meals during the pandemic. However, they, along with oﬃcials from state child nutrition
agencies and anti-hunger organizations in all four states, identiﬁed challenges related to the
rollout of waivers. These challenges included the timing of waiver announcements and extensions,
interpreting the waivers, delays in waiver-related guidance documents, and keeping up with the
number of waivers announced.

State and local oﬃcials said that dealing with such challenges exhausted their ﬁnancial and staﬀ
resources and made it diﬃcult to plan for meal service and communicate accurate information to
families. Oﬃcials from the Boys and Girls Clubs of America said that implementation of waivers
was confusing for many of their clubs that operate under both SFSP and CACFP At-Risk because
waivers were not always consistent across programs.

FNS oﬃcials told us they are examining potential policy changes, including identifying authority or
legislative action that FNS may need to help prepare for future national emergencies. Additionally,
oﬃcials in three of the four state child nutrition agencies and two of the FNS regional oﬃces
suggested that establishing a single emergency child nutrition feeding program or plan could
help streamline the waiver implementation process and reduce the burdens on states and school
district nutrition programs.

Potential to make some ﬂexibilities permanent. Oﬃcials of some districts and states told us they
believed that some of the child nutrition ﬂexibilities provided during the pandemic should be
made permanent. For example, oﬃcials from two state nutrition oﬃces suggested that FNS should
assess whether there are opportunities to expand oﬀ-site monitoring because of the cost and
time savings. In June 2021, FNS oﬃcials told us that FNS is gathering information through the SMO
study with the intent of assessing whether any ﬂexibilities could be made permanent.

Methodology
To conduct our work, we reviewed relevant federal laws and agency guidance and documents. We
also reviewed the most recent data available from FNS on meals served through child nutrition
programs. To assess the reliability of these data, we reviewed existing information about the data
and FNS’s reporting processes, interviewed agency oﬃcials, and conducted manual testing of the
data. We determined the data were suﬃciently reliable for our purposes.

We also interviewed oﬃcials from FNS’s National Oﬃce and four of its regional oﬃces.
Additionally, we interviewed state nutrition directors and oﬃcials representing anti-hunger
organizations from four states—Georgia, Maine, Texas, and Washington, which we selected in
part on the basis of variation in geographic location and school operating policies at the time of
selection—as well as district nutrition oﬃcials from three school districts in each state. Further,
we interviewed oﬃcials from the Boys and Girls Clubs of America, the National CACFP Association,
and the School Nutrition Association to gain additional perspectives on feeding children in and
outside school settings. The information gathered from these interviews is intended to provide
examples of experiences of meal operators, states, FNS oﬃcials, and nutrition organizations
during the COVID-19 pandemic and is not intended to be representative.

Page 167 GAO-22-105051
Agency Comments
We provided FNS and the Oﬃce of Management and Budget with a draft of this enclosure.
FNS provided technical comments, which we incorporated as appropriate. Although it did not
provide a formal letter, FNS stated in it technical comments that it generally concurred with
our recommendation. FNS also said that the COVID-19 pandemic may continue to aﬀect Child
Nutrition program operations.

FNS stated that it is developing a plan to interview every state agency to determine how they used
waivers to operate and administer all Child Nutrition programs during the COVID-19 pandemic.
According to FNS, it will evaluate the information it collects to capture crucial information, lessons
learned, and best practices to inform future policy making. FNS also stated that in ﬁscal year
2022 it will gather information from stakeholders at conferences to obtain perspectives of key
stakeholders such as CACFP institutions and nonschool SFSP sponsors.

We are encouraged by FNS’s plans to gather local perspectives directly from such stakeholders.
Given the various eﬀorts FNS is planning, we believe that now is an appropriate time to document
the agency’s plans to analyze lessons learned during the pandemic.

GAO’s Ongoing Work
Our work on FNS’s response to COVID-19 through its nutrition assistance programs, including child
nutrition, is ongoing. We will continue to examine FNS’s use of COVID-19 relief funds, its eﬀorts to
ensure program integrity, and its eﬀorts to help vulnerable populations access the programs.

Related GAO Products
Federal Real Property Security: Interagency Security Committee Should Implement A Lessons-Learned
Process. GAO-12-901. Washington, D.C.: September 10, 2012.

Grants Management: OMB Should Collect and Share Lessons Learned from Use of COVID-19-Related
Grant Flexibilities. GAO-21-318. Washington, D.C.: March 31, 2021.

Standards for Internal Control in the Federal Government. GAO-14-704G. Washington, D.C.:
September 2014.

Summer Meals: Actions Needed to Improve Participation Estimates and Address Program Challenges.
GAO-18-369. Washington, D.C.: May 31, 2018.

Contact Information: Kathryn A. Larin, (202) 512-7215 or larink@gao.gov

Page 168 GAO-22-105051
Unemployment Insurance Programs
Although fewer weekly claims for regular and CARES Act unemployment insurance beneﬁts were
submitted during summer 2021 than earlier in the pandemic, millions of unemployed workers
continued to claim the CARES Act beneﬁts until those programs ended in early September. The
historic numbers of claims during the pandemic contributed to challenges—such as delayed
beneﬁt payments and increased amounts of overpayments—that have implications for future
crises.

Entity involved: Department of Labor

Background
The unemployment insurance (UI) system is a federal–state partnership that provides temporary
ﬁnancial assistance to eligible workers who become unemployed through no fault of their
own. States design and administer their own UI programs within federal parameters, and the
Department of Labor (DOL) oversees states’ compliance with federal requirements, such as by
ensuring that states pay beneﬁts when they are due. Regular UI beneﬁts—those provided under
the state UI programs in place before the CARES Act was enacted—are funded primarily through
state taxes levied on employers and are intended to typically be lower than a claimant’s previous
employment earnings, according to DOL.248

The CARES Act created three federally funded temporary UI programs that expanded beneﬁt
eligibility and enhanced beneﬁts, which were amended by the Consolidated Appropriations Act,
2021 and the American Rescue Plan Act of 2021.249 These programs expired on September 6,
2021, although some states ended their participation before that date.

1. Pandemic Unemployment Assistance (PUA) authorized UI beneﬁts for individuals not
otherwise eligible for UI beneﬁts, such as self-employed and certain gig economy workers,
who were unable to work as a result of speciﬁed COVID-19-related reasons.250

248To be eligible for regular UI beneﬁts, applicants generally must be able and available to work and actively seeking
work. 42 U.S.C. § 503(a)(12). Administration of the regular UI program is ﬁnanced by a federal tax on employers,
according to DOL.
249The CARES Act also addressed other aspects of the UI system, such as authorizing certain ﬂexibilities for states to
hire additional staﬀ and funding for Short-Time Compensation programs. In addition to the CARES Act, the Families
First Coronavirus Response Act provided up to $1 billion in emergency grant funding to states in ﬁscal year 2020 for UI
administrative purposes. In addition, on August 8, 2020, the President signed a memorandum directing the Department
of Homeland Security’s Federal Emergency Management Agency (FEMA) to provide up to $44 billion in lost wages
assistance. Pursuant to the presidential memorandum, upon receiving a FEMA grant, states and territories could provide
eligible claimants $300 or $400 per week—which included a $300 federal contribution—in addition to their UI beneﬁts.
The White House, Memorandum on Authorizing the Other Needs Assistance Program for Major Disaster Declarations Related
to Coronavirus Disease 2019 (Aug. 8, 2020). FEMA approved 53 states and territories to provide lost wages assistance to
eligible claimants for a maximum of 6 weeks of unemployment experienced from the week ending on August 1, 2020,
through the week ending on September 5, 2020. These 53 states and territories included the District of Columbia. For
the purpose of this enclosure, we categorize the District of Columbia as a state.
250At the time of the program’s expiration, PUA generally authorized up to 79 weeks of beneﬁts. Pub. L. No. 117-2, §
9011(a), (b), 135 Stat. 4, 118; Pub. L. No. 116-260, div. N, tit. II, § 201(a), (b), 134 Stat. 1182, 1950-1951 (2020); Pub. L.
No. 116-136, § 2102, 134 Stat. 281, 313 (2020).

Page 169 GAO-22-105051
2. Federal Pandemic Unemployment Compensation (FPUC) generally authorized an additional
weekly beneﬁt for individuals who were eligible for weekly beneﬁts under the regular UI and
CARES Act UI programs.251

3. Pandemic Emergency Unemployment Compensation (PEUC) generally authorized additional
weeks of UI beneﬁts for those who had exhausted their regular UI beneﬁts.252

In addition, the Consolidated Appropriations Act, 2021 created the Mixed Earner Unemployment
Compensation (MEUC) program, which was extended by the American Rescue Plan Act of 2021.253
According to DOL, the MEUC program was intended to cover regular UI claimants whose beneﬁts
do not account for signiﬁcant self-employment income and who thus may have received a lower
regular UI beneﬁt than they would have received had they been eligible for PUA.254

During the pandemic, regular UI claimants who exhausted their regular UI and PEUC beneﬁts
in certain states also had access to the Extended Beneﬁts program. The program, which existed
before the pandemic and provides up to an additional 13 or 20 weeks of beneﬁts, is activated in
states during periods of high unemployment, according to DOL.255

Overview of Key Issues
About half of the states stopped participating in at least one CARES Act UI program before
the programs expired in September 2021. As of August 17, 2021, 26 states had announced their

251FPUC generally authorized an additional $600 beneﬁt through July 2020 as well as an additional $300 beneﬁt for
weeks beginning after December 26, 2020, through the end of the program. Pub. L. No. 117-2, § 9013, 135 Stat. 4,
119; Pub. L. No. 116-260, div. N, tit. II, § 203, 134 Stat. 1182, 1953; Pub. L. No. 116-136, § 2104, 134 Stat. at 318.
252At the time of the program’s expiration, PEUC generally authorized an additional 53 weeks of beneﬁts for
claimants who were fully unemployed. Pub. L. No. 117-2, § 9016(a), (b), 135 Stat. 4, 119-120; Pub. L. No. 116-260, div.
N, tit. II, § 206(a), (b), 134 Stat. 1182, 1954; Pub. L. No. 116-136, § 2107, 134 Stat. at 323.
253The MEUC program, which was voluntary for states, authorized an additional $100 weekly beneﬁt for certain UI
claimants who received at least $5,000 of self-employment income in the most recent tax year prior to their application
for UI beneﬁts. Pub. L. No. 117-2, § 9013(a), 135 Stat. 4, 119; Pub. L. No. 116-260, div. N, tit. II, § 261(a)(1), 134 Stat. 1182,
1961. The $100 weekly beneﬁt was in addition to other UI beneﬁts received by claimants; however, individuals receiving
PUA beneﬁts could not receive MEUC payments.
254According to DOL, 51 states and territories elected to participate in the MEUC program, and Idaho and South
Dakota opted not to participate. However, 23 states terminated their participation in June or July 2021, before the
program expiration date, and one intended to terminate participation but did not because of litigation at the state level,
according to DOL. The remaining 27 states and territories continued participating in the MEUC program until it expired
in September 2021. According to DOL, as of August 30, 2021, 1 week before the MEUC program expired, the following 18
states and territories had not begun paying beneﬁts: Colorado, Georgia, Illinois, Iowa, Kansas, Louisiana, Nevada, New
Jersey, Ohio, Oklahoma, Oregon, Rhode Island, Texas, the U.S. Virgin Islands, Vermont, Washington, West Virginia, and
Wyoming. Before paying any beneﬁts, Wyoming submitted notice to DOL that it planned to return all MEUC funds and
not operate the program, according to DOL.
255DOL reported that as of September 26, 2021, the Extended Beneﬁts program was activated in four states and
territories because of high levels of unemployment. The Extended Beneﬁts program was activated in all states except
South Dakota at some point during the pandemic, according to DOL. If unemployment was not high enough to activate
the Extended Beneﬁts program in a state, or if regular UI claimants exhausted their PEUC and Extended Beneﬁts, they
may have been eligible for PUA beneﬁts if they also met PUA eligibility requirements.

Page 170 GAO-22-105051
intention to terminate participation in at least one of these programs.256 For example, according
to DOL, all 26 of these states notiﬁed DOL that they intended to end participation in the FPUC
program, and 22 notiﬁed DOL that they also intended to end participation in the PEUC and PUA
programs.257 Of these 26 states, 24 proceeded to terminate their participation in these programs
between mid-June and late July, according to DOL.258 For the week ending on June 12, 2021,
DOL reported that 1.8 million continued claims for PUA and PEUC beneﬁts were submitted in
states that proceeded to terminate these programs over the next several weeks. This suggests
that demand for these beneﬁts likely would have continued if these states had not ended their
participation.259

In public announcements, states generally cited labor shortages among their reasons for
withdrawing from the CARES Act UI programs. However, preliminary data from DOL’s Bureau of
Labor Statistics (BLS) do not show an association between termination of participation in these
programs and reductions in states’ unemployment rates.260 Also, initial results from our review of
the economic literature suggest that expanded UI beneﬁts during the pandemic generally did not
discourage unemployed workers from returning to work; however, this conclusion could shift over
time with changes in economic conditions.261

256States participated in these UI programs under agreements with the Secretary of Labor. According to DOL, states
were permitted to terminate, on providing 30 days’ written notice to DOL, their participation in the FPUC, PEUC, PUA,
and MEUC programs before the programs expired.
257The 22 states that submitted notice of their intent to withdraw from FPUC, PEUC, and PUA are Alabama, Arkansas,
Georgia, Idaho, Indiana, Iowa, Louisiana, Maryland, Mississippi, Missouri, Montana, Nebraska, New Hampshire, North
Dakota, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, West Virginia, and Wyoming. The four states
that submitted notice of their intent to withdraw only from FPUC are Alaska, Arizona, Florida, and Ohio. According to
DOL, 24 of these 26 states also submitted notice of their intent to withdraw from the MEUC program; the other two
states (Idaho and South Dakota) had not been participating in the MEUC program.
258According to DOL, Indiana and Maryland announced their intent to terminate participation in the FPUC, PEUC, and
PUA programs but, because of litigation at the state level, did not end their participation before the programs expired.
259According to DOL, after ﬁling an initial claim to establish eligibility for UI beneﬁts, individuals then generally ﬁle
a continued claim on a weekly basis to claim beneﬁts for the preceding week of unemployment. According to DOL,
individuals did not submit claims for FPUC, and states added FPUC payments to existing regular UI, PUA, or PEUC claims.
Nineteen of the 20 states that terminated the PUA and PEUC programs between mid-June and late July 2021 reported
PUA and PEUC continued claims for the week ending on June 12, 2021. These 19 states include Georgia, which reported
PUA but not PEUC continued claims data. The remaining state, Mississippi, did not report PUA or PEUC claims data.
The number of continued claims provides some indication of the number of claimants potentially aﬀected by these
terminations; however, the number of continued claims is not an approximation of the number of unique individuals
ﬁling these claims, because individuals sometimes submitted multiple continued claims in a single period for multiple
prior weeks of unemployment.
260According to BLS data, only six of the 24 states that terminated participation in the CARES Act UI programs in June
had statistically signiﬁcant decreases in their unemployment rates from June to July 2021. Two additional states that
terminated participation in these programs in July also had statistically signiﬁcant decreases in their unemployment
rates from June to July 2021. However, given the short time frame, an association between program termination in July
and a decrease in the unemployment rate from June to July is unlikely. Furthermore, BLS data show that 10 of the 29
states and territories that did not terminate participation in the CARES Act UI programs also had statistically signiﬁcant
decreases in their unemployment rates from June to July 2021. These BLS data include states’ unemployment rates from
July 2021, 1 month after states began terminating participation in the CARES Act UI programs, and BLS did not take into
account potential explanatory factors. Data are preliminary and may change in future months. See Bureau of Labor
Statistics, “News Release: State Employment and Unemployment—July 2021,” Aug. 20, 2021.
261For example, one study found that over half of all unemployed individuals receiving the $600 FPUC supplement
returned to work before the supplement expired at the end of July 2020. See Peter Ganong et al., Spending and Job

Page 171 GAO-22-105051
In July 2021, DOL issued guidance to states regarding their responsibilities after they stopped
participating in the CARES Act UI programs or after the programs expired, whichever came ﬁrst.262
For example, states must process and pay PUA, FPUC, PEUC, and MEUC beneﬁts to eligible
claimants for all weeks of unemployment before the programs ended. In addition, for 30 days
after the PUA program ended, states were required to continue accepting new PUA applications
for weeks of unemployment before the program ended. States were also required to accept new
PEUC and MEUC applications for weeks of unemployment before those programs ended, if state
law allowed claims to be backdated.

In August 2021, the Secretary of the Treasury and the Secretary of Labor issued a joint letter to
Congress aﬃrming that states can use American Rescue Plan Act of 2021 funding to provide
additional weeks of UI beneﬁts to workers whose beneﬁts expired in September and to workers
who are not covered by the regular UI program.

Although claims for UI beneﬁts were generally lower during summer 2021, overall demand
for beneﬁts remained high through early September. During the week ending on September
25, 2021, DOL reported that 298,255 initial claims for regular UI beneﬁts were submitted
nationwide, which was close to the lowest number since the surge of initial claims at the beginning
of the pandemic.263 DOL reported that more than 100,000 initial claims for PUA beneﬁts were
submitted nationwide during each week in August 2021; the PUA program expired on September
6.264

Despite the overall decline in initial claims since the beginning of the pandemic, initial claims
throughout summer 2021 remained at a level that indicated more Americans were continuing
to experience job losses than was typical in the year before the pandemic.265 For example, the
number of initial claims for regular UI beneﬁts submitted each week from late June through late
September ranged from about 55 percent to more than 100 percent higher than the number of
initial claims submitted during the corresponding weeks in 2019.

Search Impacts of Expanded Unemployment Beneﬁts: Evidence from Administrative Micro Data, Working Paper No. 2021-19
(Chicago: Becker Friedman Institute for Economics, University of Chicago, February 2021). An update to this study
concluded that unemployment supplements were not the key driver of the job-ﬁnding rate through April 2021. See
Peter Ganong et al., Micro and Macro Disincentive Eﬀects of Expanded Unemployment Beneﬁts (July 29, 2021). In addition,
a Federal Reserve Board survey conducted in November 2020 found that unemployed workers most commonly cited
health limitations, an inability to ﬁnd work, and child care or family obligations as reasons for not returning to work. The
survey question did not include expanded UI beneﬁts as a potential response. See Board of Governors of the Federal
Reserve System, Economic Well-Being of U.S. Households in 2020 (May 2021).
262Department of Labor, State Responsibilities after the Temporary Unemployment Beneﬁt Programs under the Coronavirus
Aid, Relief, and Economic Security (CARES) Act, as Amended, End Due to State Termination of Administration or When the
Programs Expire, UIPL 14-21, Change 1 (Washington, DC: July 12, 2021).
263An initial claim is the ﬁrst claim ﬁled by an individual to determine eligibility for UI beneﬁts after separating from an
employer. Initial claims counts presented are not seasonally adjusted, and counts for the week ending on September 25,
2021, reﬂect advance initial claims, which are preliminary and subject to revision. In some cases, advance initial claims
represent estimates submitted by states that are later revised.
264During the week ending on September 4, 2021, DOL reported that 94,638 initial PUA claims were submitted
nationwide. For 30 days after the PUA program ended, states were required to continue accepting new PUA applications
for weeks of unemployment before the program ended. During the week ending on September 25, 2021, DOL reported
that 16,752 initial PUA claims were submitted nationwide.
265See the enclosure on Economic Indicators in appendix I for more information about economic conditions during
summer 2021.

Page 172 GAO-22-105051
According to DOL oﬃcials, the number of continued claims may be a better measure of continuing
demand for beneﬁts than the number of initial claims. For example, states and territories reported
that during the week ending on September 11, 2021, about 2.5 million continued claims were
submitted for regular UI beneﬁts and 431,340 continued claims were submitted for Extended
Beneﬁts.266

The number of regular UI continued claims submitted each week declined overall after the peak in
late April and early May 2020 through September 2021 (see ﬁgure). Although some of this decline
was due to claimants’ ﬁnding employment, the decline was also likely due to other factors, such
as claimants’ exhausting regular UI beneﬁts and beginning to claim PEUC or other beneﬁts.267
For example, the persistently high numbers of PEUC continued claims from fall 2020 through the
program’s expiration in early September 2021 suggest that many individuals were experiencing
long-term unemployment and had likely exhausted their regular UI beneﬁts.268

266States also reported that about 44,000 continued claims were submitted in other programs, including those for
federal employees and newly discharged veterans, state additional beneﬁt programs, and Short-Time Compensation or
work-sharing, during the week ending on September 11, 2021. In addition, states reported continued claims submitted
that week for the PUA and PEUC programs, which had expired on September 6, 2021. These continued claims were likely
for weeks of unemployment before the programs expired.
267After exhausting regular UI beneﬁts—generally available for up to 26 weeks in most states, according to
DOL—eligible individuals were generally able to apply for (1) PEUC; then (2) the Extended Beneﬁts program, if activated
in a state; and ﬁnally, in certain circumstances, (3) PUA beneﬁts. According to DOL, individuals did not submit claims for
FPUC, and states added FPUC payments onto existing regular UI, PUA, or PEUC claims.
268In its Employment Situation news releases, BLS deﬁnes the long-term unemployed population as those who are
jobless for 27 weeks or more.

Page 173 GAO-22-105051
Weekly Continued Claims Submitted Nationwide for Regular Unemployment Insurance, Pandemic Emergency
Unemployment Compensation, and Extended Beneﬁts, Mar. 1, 2020–Sept. 4, 2021

Note: After exhausting regular UI beneﬁts—generally available for up to 26 weeks in most states, according to the Department
of Labor (DOL)—eligible individuals were generally able to apply for (1) PEUC and then (2) the Extended Beneﬁts program, if
activated in a state. The weekly counts of continued claims shown are not seasonally adjusted. Counts are from DOL data that
include any adjustments submitted by states as of September 30, 2021. All 53 states and territories reported regular UI claims
in each week shown. The number of states and territories reporting PEUC claims varied by week; for example, fewer than half
of the states and territories reported data before mid-May 2020 and at least 50 states and territories reported data each week
from mid-July 2020 through mid-June 2021, when certain states began terminating their PEUC programs. The number of states
reporting Extended Beneﬁts claims each week varied, partly on the basis of the number of states with the program activated
each week. The Extended Beneﬁts program, which existed before the pandemic, is activated in states during periods of high
unemployment, according to DOL.

During the week ending on September 4, 2021, before the pandemic UI programs expired on
September 6, states and territories reported about 8.5 million continued claims submitted in
the PUA and PEUC programs. This large number of claims, in addition to the approximately
2.4 million regular UI continued claims submitted that same week, suggests that demand for
unemployment beneﬁts remained high as the CARES Act UI programs expired. During the week
ending on September 4, 2021, regular UI continued claims were about 81 percent lower than in
the corresponding week of 2020 but were about 62 percent higher than in the corresponding
week of 2019, before the pandemic.

As we have previously reported, because of backlogs in processing historic numbers of claims in
many states, among other data issues, the number of continued claims did not approximate the
number of individuals claiming beneﬁts during the pandemic. For example, backlogs in claims

Page 174 GAO-22-105051
processing led to individuals submitting claims for multiple weeks of beneﬁts in a single reporting
period, which states counted as multiple claims for that reporting period, particularly in the PUA
program. As a result, reliable conclusions about trends in the number of individuals claiming
beneﬁts cannot be drawn from data on continued claims.

Although the timeliness of regular UI ﬁrst payments improved through early 2021, it has
generally declined since April 2021, and some claimants still face substantial delays in
receiving beneﬁts. The timeliness of ﬁrst payments of regular UI beneﬁts declined substantially
early in the pandemic, as states faced extensive claims-processing backlogs resulting from
historically high numbers of claims.269 As we have previously reported, ﬁrst-payment timeliness
nationwide improved from fall 2020 through January 2021. Subsequently, regular UI ﬁrst-payment
timeliness nationwide ﬂuctuated from February through April 2021, then generally declined
through August 2021.270 First-payment timeliness was about 23 percentage points lower in August
than in January 2021 (see ﬁgure).

269DOL monitors timeliness of beneﬁt payments in the regular UI program. One of DOL’s core performance measures
is the percentage of all regular UI ﬁrst payments made within either 14 or 21 days of the ﬁrst week of beneﬁts for
which claimants are eligible; DOL considers 87 percent to be an acceptable level of performance. DOL uses 14 days
as the timeliness goal for states with a waiting week requirement and uses 21 days for states without a waiting week
requirement. According to DOL, some states require individuals who are otherwise eligible for beneﬁts to serve a
waiting period—generally 1 week—before receiving beneﬁts. In its guidance released at the start of the pandemic, DOL
recommended that states consider temporarily waiving their waiting week requirements. Thus, we focus on payments
made within 21 days. We analyzed ﬁrst-payment timeliness data that states had reported to DOL as of September 27,
2021. At that point, 52 of the 53 states and territories had reported data for August 2021 and all 53 had reported data
for July 2021 and prior months.
270The extension of the CARES Act UI programs at the end of 2020, according to DOL oﬃcials, may have aﬀected
payment timeliness because states had to reassign staﬀ and focus resources on implementing new program
requirements.

Page 175 GAO-22-105051
Timeliness of First Payments of Regular Unemployment Insurance (UI) Beneﬁts, Jan. 2020–Aug. 2021

Note: We analyzed UI ﬁrst-payment timeliness data that states had reported to the Department of Labor (DOL) as of September
27, 2021. At that point, 52 of the 53 states and territories had reported data for August 2021 and all 53 had reported data
for July 2021 and prior months. One of DOL’s core performance measures is the percentage of all regular UI ﬁrst payments
made within either 14 or 21 days of the ﬁrst week of beneﬁts for which claimants are eligible, depending on whether the
state requires that individuals who are otherwise eligible for beneﬁts serve a waiting period—generally 1 week—before
receiving beneﬁts. We focus on payments made within 21 days because in guidance released at the start of the pandemic, DOL
recommended that states consider temporarily waiving their waiting week requirements. According to DOL, states must pay at
least 87 percent of regular UI claims within 14 or 21 days to reach an acceptable level of performance.

According to DOL oﬃcials, when they asked oﬃcials in some states about the reasons for the
decrease in timeliness in May and June 2021, they cited the additional time and eﬀort needed to
process backlogs of claims needing adjudication and appeals, decreased numbers of staﬀ, and
enhanced fraud prevention eﬀorts that have resulted in more adjudication issues for states to
resolve. In some states, many regular UI claimants continue to face delays before receiving their
ﬁrst payments. For example, in 16 states, at least half of regular UI claimants who received their
ﬁrst beneﬁts in August 2021 had been waiting longer than 3 weeks. In addition, nationwide, about
21.7 percent of regular UI claimants who received their ﬁrst beneﬁts in August 2021 had been
waiting longer than 10 weeks. By comparison, of the regular UI claimants nationwide who received
their ﬁrst beneﬁts in March 2020, less than 3 percent had been waiting longer than 3 weeks and
less than 1 percent longer than 10 weeks.

The number of states holding federal loans to pay UI beneﬁts, and the total amount of
these loans, decreased slightly in late summer 2021.271 Because of persistently high numbers

271While the CARES Act UI programs are federally funded, regular UI is funded primarily through state and federal
taxes on employers. When a state exhausts the funds available for regular UI beneﬁts, it may borrow from the federal
government. According to DOL data, even before the pandemic, many states were not collecting enough UI tax revenue
to satisfy the solvency standard speciﬁed in DOL regulations providing for interest-free loans to states. See 20 C.F.R. §
606.32 (2019).

Page 176 GAO-22-105051
of UI claims during the pandemic, some states have held substantial federal loans to pay UI
beneﬁts. As of September 24, 2021, 12 states and territories held federal loans totaling about
$45.3 billion. This total loan balance is approximately $8.2 billion less than we reported in July
2021, and seven states that previously held these loans have repaid them.272

As we have previously reported, some states have used their Coronavirus Relief Fund payments,
under guidance from the Department of the Treasury, to pay for UI beneﬁts in order to reduce or
prevent loan balances and avoid possible future increases in employer tax rates.273 Generally, if a
federal loan balance to pay UI beneﬁts is held by a state for 2 or more years, the rate of the federal
tax on employers that is used to fund the UI program will increase.274 States may continue to use
these Coronavirus Relief Fund payments for expenses through the end of 2021.275 In addition,
the American Rescue Plan Act of 2021 provided funds to states, local governments, territories,
and tribal governments. States and territories may use these funds, under an interim ﬁnal rule
from the Department of the Treasury, to restore their UI trust funds or to repay federal loan
balances.276

States have continued to identify overpayments in the regular UI and CARES Act UI
programs, and 30 states are reporting data to DOL on recovered PUA overpayments. DOL
reported that as of September 27, 2021, states and territories had identiﬁed approximately
$18.3 billion in overpayments made in UI programs during the ﬁrst 5 quarters of the pandemic

272At the end of 2010, after the 2007–2009 recession and early recovery, 30 states and territories held approximately
$40.2 billion in federal loans to pay UI beneﬁts. We did not adjust this 2010 loan balance amount for inﬂation. According
to the National Bureau of Economic Research, the 2007–2009 recession began in December 2007 and ended in June
2009. The total loan balance held by states at the end of 2010 represented the highest year-end balance from the
2007–2009 recession until 2020. Total loan balances ﬂuctuate throughout any given year.
273The CARES Act established the $150 billion Coronavirus Relief Fund to provide payments to state, local, territorial,
and tribal governments to cover the costs of necessary expenditures incurred because of the COVID-19 pandemic. Pub.
L. No. 116-136, § 5001, 134 Stat. 281, 501-504. In its guidance on the Coronavirus Relief Fund, the Department of the
Treasury reported that states may use this funding to make payments to their state UI trust funds to prevent expenses
related to the COVID-19 public health emergency from causing these UI trust funds to become insolvent. On January
15, 2021, the Department of the Treasury republished in a ﬁnal form the guidance it had previously made available
on its website regarding the Coronavirus Relief Fund. Prior to publication in the Federal Register, the last version of its
guidance and frequently asked questions documents were dated September 2, 2020, and October 19, 2020, respectively.
86 Fed. Reg. 4,182 (Jan. 15, 2021).
274The regular UI program is primarily funded through state and federal taxes on employers. Under the Federal
Unemployment Tax Act, employers are generally required to pay a federal unemployment tax at a rate of 6.0 percent on
the ﬁrst $7,000 of wages paid to an employee each year, which funds administrative costs associated with the regular
UI program and the federal share of beneﬁts paid under the Extended Beneﬁts program, among other things. The
Federal Unemployment Tax Act provides a credit of up to 5.4 percent against federal tax liability to employers who
pay state taxes timely under an approved state UI program. If a state has outstanding loan balances on January 1 for
2 consecutive years and does not repay the full amount of its loans by November 10 of the second year, the Federal
Unemployment Tax Act credit rate for employers in that state will be reduced. Thus, the federal taxes paid by employers
will increase, all else being equal.
275The Department of the Treasury reported that as of December 31, 2020, it had obligated all of the $150 billion from
the Coronavirus Relief Fund to state, local, territorial, and tribal governments. The Consolidated Appropriations Act, 2021
extended the period for states and other entities to use these funds through December 31, 2021. Pub. L. No. 116-260,
div. N, tit. X, § 1001, 134 Stat. 1182, 2145.
276Coronavirus State and Local Fiscal Recovery Funds, 86 Fed. Reg. 26,786, 26,822 (May 17, 2021). Funds may be used to
restore a state’s unemployment trust fund to its balance on January 27, 2020, or to pay back advances received for the
payment of beneﬁts between January 27, 2020, and May 17, 2021. 31 C.F.R. § 35.6(b)(5).

Page 177 GAO-22-105051
combined (April 2020 through June 2021).277 These reported overpayments are not necessarily a
result of fraud, though some may be.278 This $18.3 billion in reported overpayments includes

• $6.3 billion in FPUC overpayments,279

• $6.0 billion in PUA overpayments,280

• $5.4 billion in regular UI and Extended Beneﬁts overpayments, and

• $0.7 billion in PEUC overpayments.281

States and territories may waive and not recover overpayments in certain circumstances.282 States
and territories reported waiving about $0.2 billion of regular UI, Extended Beneﬁts, PEUC, and

277As we have previously reported, DOL data show that the dollar amount of state-reported overpayments in the
regular UI program increased substantially during the pandemic, coinciding with historically high numbers of UI claims.
States have also reported large amounts of overpayments in the CARES Act UI programs. Regarding underpayments,
states do not report the actual amount of underpayments to DOL. However, states estimate underpayments based on
representative samples of paid and denied regular UI claims and report these estimates to DOL. In calendar year 2020,
states estimated about $696 million in regular UI underpayments, although this estimate does not include the second
quarter (April 1, 2020, through June 30, 2020). According to DOL oﬃcials, DOL provided operational ﬂexibilities to states
in response to the pandemic by temporarily suspending this sampling process during the second quarter of 2020 to
allow states to implement the CARES Act UI programs.
278While overpayments may be caused by unintentional error, fraud involves obtaining something of value through
willful misrepresentation. Whether an act is fraudulent is determined through the judicial or other adjudicative systems.
279FPUC beneﬁts are paid in addition to other UI beneﬁts. About 93 percent of reported FPUC overpayment
amounts were paid on regular UI or PUA claims.
280States and territories report PUA overpayments data to DOL on a monthly basis, and the total amount shown
includes overpayments related to identity theft. We accessed the PUA overpayments data on September 27,
2021; these data are subject to change as more states report data and as states revise previously reported data.
For consistency with the regular UI overpayment data, which states and territories report on a quarterly basis,
the PUA overpayment amount shown is for April 2020 through June 2021. As of September 27, 2021, 42 states
and territories had reported an additional approximately $1.4 billion of PUA overpayments identiﬁed in July and
August 2021. The number of states and territories that have reported PUA overpayments data varies by month; for
example, 18 reported overpayment amounts in April 2020, 41 reported overpayment amounts in July 2021, and 20
reported overpayment amounts in August 2021.
281Because overpayment amounts by program are rounded to billions of dollars, the sum of these rounded
amounts diﬀers from the total overpayment amount. States and territories report regular UI, Extended Beneﬁts,
PEUC, and FPUC overpayments data to DOL on a quarterly basis. We accessed the data on September 27, 2021. At
that point, not all states and territories had reported overpayment amounts for all programs in all quarters. States
and territories may revise the amount of overpayments they have identiﬁed for 3 years after the reporting quarter,
according to DOL. For this report, we excluded overpayments related to emergency unemployment compensation
programs prior to the pandemic. Thus, the totals we calculate diﬀer slightly from those available on DOL’s UI
overpayment recovery website, which, according to DOL oﬃcials, include nonpandemic emergency programs, such
as those enacted during the 2007–2009 recession.
282According to DOL, states generally may waive a nonfraud overpayment, in accordance with state law, if the
overpayment was not the fault of the claimant and if requiring repayment would be against equity and good conscience
or would otherwise defeat the purpose of the UI law. States were authorized to waive PUA overpayments under the
Consolidated Appropriations Act, 2021. According to this act, if an individual receives PUA beneﬁts they were not entitled
to, the state must generally require such individuals to repay the amount; however, the state can waive that requirement
if the individual was without fault and repayment would be contrary to equity and good conscience. Pub. L. No. 116-260,

Page 178 GAO-22-105051
FPUC overpayments during the ﬁrst 5 quarters of the pandemic combined (April 2020 through
June 2021).283 In response to a recommendation in our March 2021 report, DOL updated state
reporting requirements for the PUA program in September 2021 to include the collection of data
on waived PUA overpayments.284

In late May 2021, DOL’s Oﬃce of Inspector General (OIG) reported, among other things, that some
states did not perform required overpayment recovery activities.285 Speciﬁcally, DOL’s OIG found
that 19 states (38 percent) did not perform the required overpayment recovery activities, such
as beneﬁt oﬀsets, for the recipients to repay the UI overpayments.286 The OIG further reported
that once states identify overpayments, it is essential that they complete recovery activities to
mitigate the risk of ﬁnancial loss as a result of overpaid claims.287 The OIG recommended that
DOL assist states with reporting of claims, overpayments, and fraud to create clear and accurate
information and then use the overpayment and fraud reporting to prioritize and assist states with
fraud detection and recovery. DOL agreed with the OIG’s recommendation and said it would take
steps to implement the recommendation.

When states and territories recover overpayments, they report the recovered amount in the
period when the recovery occurs. For example, states and territories reported recovering about
$0.4 billion in the PEUC and FPUC programs combined from April 2020 through June 2021 (i.e.,
during the ﬁrst 5 quarters those programs existed).288 In response to a recommendation in our
January 2021 report, DOL updated its state reporting requirements for the PUA program to include
the collection of data on recovered PUA overpayments. As of September 27, 2021, 30 states had

div. N, tit. II, § 201(d), 134 Stat. 1182, 1952. According to DOL, states are able to retroactively waive PUA overpayments
from the beginning of the program onward.
283We accessed the waived overpayments data on September 27, 2021; these data are subject to change as more states
and territories report data and as states and territories revise previously reported data.
284Department of Labor, Pandemic Unemployment Assistance (PUA) Program: Updated Operating Instructions and Reporting
Changes, UIPL 16-20, Change 6 (Washington, DC: Sept. 3, 2021).
285Department of Labor, Oﬃce of Inspector General, COVID-19: States Struggled to Implement CARES Act Unemployment
Insurance Programs, Report No. 19-21-004-03-315 (Washington, DC: May 28, 2021).
286In this case, beneﬁt oﬀsets are beneﬁts withheld by the state agency to satisfy the requirement for the recipient to
repay an overpayment.
287DOL has issued various guidance during the pandemic related to overpayments, including the following: Department
of Labor, Program Integrity for the Unemployment Insurance (UI) Program and the UI Programs Authorized by the Coronavirus
Aid, Relief, and Economic Security (CARES) Act of 2020—Federal Pandemic Unemployment Compensation (FPUC), Pandemic
Unemployment Assistance (PUA), and Pandemic Emergency Unemployment Compensation (PEUC) Programs, UIPL 23-20
(Washington, DC: May 11, 2020); Addressing Fraud in the Unemployment Insurance (UI) System and Providing States with
Funding to Assist with Eﬀorts to Prevent and Detect Fraud and Identity Theft and Recover Fraud Overpayments in the Pandemic
Unemployment Assistance (PUA) and Pandemic Emergency Unemployment Compensation (PEUC) Programs, UIPL 28-20
(Washington, DC: Aug. 31, 2020); Beneﬁts Held by Banks and Financial Institutions as a Result of Suspicious and/or Potentially
Fraudulent Activity and the Proportional Distribution Methodology Required for Recovering/Returning Federally Funded
Unemployment Compensation (UC) Program Funds, UIPL 19-21 (Washington, DC: May 4, 2021); and Grant Opportunity to
Support States with Fraud Detection and Prevention, Including Identity Veriﬁcation and Overpayment Recovery Activities, in All
Unemployment Compensation (UC) Programs, UIPL 22-21 (Washington, DC: Aug. 11, 2021).
288We accessed the recovered overpayments data on September 27, 2021; these data are subject to change as more
states and territories report data and as states and territories revise previously reported data.

Page 179 GAO-22-105051
reported some data on recovered PUA overpayments, reporting a combined total of about $0.3
billion recovered from April 2020 through June 2021.289

Because of the limited number of states and territories that had reported data on recovered
and waived PUA overpayments to DOL as of September 27, 2021, our related recommendations
remained open. Sustained reporting by more states is needed to help inform DOL, policymakers,
and the public about the amount of PUA overpayments that states have waived and recovered
and about the amount that remains outstanding.

States and territories also report the amounts of fraud overpayments—a subset of the total
overpayment amounts.290 During the ﬁrst 5 quarters of the pandemic combined (April 2020
through June 2021), states and territories reported about $1.5 billion in overpayments they had
identiﬁed as fraud across the UI programs.291 However, according to DOL, states do not report
these overpayments as fraud until investigations are complete and fraud has been conﬁrmed,
which may take a long time. As a result, states’ and territories’ ongoing investigations into
whether overpayments were due to fraud could contribute to increasing amounts of fraud
overpayments reported in the coming months. For example, four of the 47 states and territories
that had reported PUA overpayments as of September 27, 2021, had not yet reported any fraud
overpayments.

In addition to reporting overpayments they identify under program requirements, states conduct
independent assessments of representative samples of paid and denied claims of permanent UI
programs to determine the accuracy of UI beneﬁt payments and estimate the amount and rate of
improper payments.292 According to DOL, by conducting these assessments, states can develop
and implement corrective actions if the assessments identify improper payments, including
potentially fraudulent payments.293 For ﬁscal year 2020, DOL allowed states to suspend these
assessments for 3 months to enable the states to reassign staﬀ to address increased claims

289As of September 27, 2021, states and territories had also reported recovering about $1.6 billion in the regular UI
and Extended Beneﬁts programs during the ﬁrst 5 quarters of the pandemic combined (April 2020 through June 2021).
However, the amounts recovered for any quarter may be from overpayments established in many previous periods.
Thus, the total amount does not measure the extent to which overpayments made during the pandemic have been
recovered.
290According to DOL guidance, an overpayment is established when a formal notice of determination has been issued.
Whether an act is fraudulent is determined through the judicial or other adjudicative systems. According to DOL,
because states may use diﬀerent deﬁnitions for categorizing an overpayment as fraudulent, an overpayment that is
classiﬁed as fraudulent in one state might not be classiﬁed as fraudulent in another state.
291We accessed the fraud overpayments data on September 27, 2021; these data are subject to change as more
states and territories report data and as states and territories revise previously reported data. Of the $1.5 billion in
overpayments identiﬁed as fraud, states and territories reported about $624 million from FPUC, $587 million from PUA,
$293 million from the regular UI and Extended Beneﬁts programs, and $33 million from PEUC.
292An improper payment is deﬁned as any payment that should not have been made or that was made in an incorrect
amount (including overpayments and underpayments) under statutory, contractual, administrative, or other legally
applicable requirements. 31 U.S.C. § 3351(4). For the purpose of producing an improper payment estimate, when the
executive agency cannot determine, due to lacking or insuﬃcient documentation, whether a payment is proper or not,
the payment shall be treated as an improper payment. 31 U.S.C. § 3352(c)(2).
293The state assessments of improper payments include potential fraud; however, states’ deﬁnitions for categorizing an
overpayment as fraudulent may vary.

Page 180 GAO-22-105051
volume.294 According to DOL oﬃcials, the methodology for ﬁscal year 2020, including the 3-month
suspension, was approved by the Oﬃce of Management and Budget (OMB) and complied with
OMB guidance.295

Additionally, as we reported in November 2020, DOL made the decision not to include claims
ﬁled exclusively under the CARES Act UI programs in its existing program for estimating
improper payments.296 According to DOL oﬃcials, because PUA has unique and distinct eligibility
requirements, applying the improper payment methodology for the regular UI program would
not be appropriate. We also reported that DOL planned to conduct an improper payment risk
assessment after the ﬁrst year of each CARES Act UI program’s operations.297

In July 2021, DOL oﬃcials told us that they were actively planning to conduct an improper
payment risk assessment in accordance with OMB guidance and to then develop improper
payment estimates for CARES Act UI programs.298 Oﬃcials also said that DOL had formed a
working group to develop new sampling and investigative methodologies to estimate improper
payments for the PUA program and that DOL planned to extrapolate regular UI data to the PEUC
program.299 DOL oﬃcials stated that they anticipated the new improper payment estimates would
be included in DOL’s ﬁscal year 2022 agency ﬁnancial report. We have previously reported that the
identiﬁcation of improper payments could suggest that a program is vulnerable to fraud; however,
it is important to note that fraud is a speciﬁc type of improper payment and that improper
payment estimates are not intended to measure fraud in a particular program.

294DOL’s ﬁscal year for reporting improper payment estimates covers July 1 of the previous year through June 30 of the
current year. For example, DOL’s ﬁscal year 2020 improper payment estimate generally covers July 1, 2019, through June
30, 2020. However, the sampling and investigation program was suspended for the quarter April 1, 2020, through June
30, 2020, because of operational ﬂexibilities provided to states in response to the pandemic, according to DOL.
295In addition, DOL’s OIG reported that an independent auditor had concluded that DOL had met all the statutory
criteria for compliance with the Payment Integrity Information Act of 2019. The OIG also reported that DOL had received
direction from OMB to utilize the results from the ﬁrst 3 quarters of the program year for its improper payment
reporting in ﬁscal year 2020 and that DOL’s decision to suspend fourth quarter program year testing was approved
by OMB. Department of Labor, Oﬃce of Inspector General, The U.S. Department of Labor Complied with the Payment
Integrity Information Act for FY 2020, but Reported Unemployment Insurance Information Did Not Represent Total Program
Year Expenses, Report No. 22-21-007-13-001 (Washington, DC: Aug. 6, 2021).
296According to DOL guidance, once a claim was selected for a state’s assessment sample, any dollars provided through
the FPUC program would be calculated as part of the claimant’s Total Whole Amount of Overpayments/Underpayments
as “Federal Supplemental Payment” in addition to dollars provided through the regular UI program.
297Guidance from OMB states that agency management is responsible for managing payment integrity risks to reduce
improper payments and protect taxpayer funds. According to the guidance, all newly established programs should
complete an improper payment risk assessment after the ﬁrst 12 months of the program. See Oﬃce of Management
and Budget, Appendix C to OMB Circular A-123: Requirements for Payment Integrity Improvement, OMB Memorandum
M-18-20 (Washington, DC: June 26, 2018). In March 2021, OMB issued a revised version of OMB Circular A-123, Appendix
C, eﬀective starting in ﬁscal year 2021. See Oﬃce of Management and Budget, Appendix C to OMB Circular A-123,
Requirements for Payment Integrity Improvement, OMB Memorandum M-21-19 (Washington, DC: Mar. 5, 2021).
298According to OMB guidance, agencies are to develop improper payment estimates for those programs determined to
be susceptible to the risk of signiﬁcant improper payments. See Oﬃce of Management and Budget, Appendix C to OMB
Circular A-123, Requirements for Payment Integrity Improvement, OMB Memorandum M-21-19 (Washington, DC: Mar. 5,
2021).
299DOL oﬃcials stated that they do not plan to produce an improper payment estimate for the FPUC program, because
the program augmented beneﬁts under the regular UI program and CARES Act UI programs and did not have separate
eligibility requirements.

Page 181 GAO-22-105051
For more information about fraud risks in the UI programs, DOL’s eﬀorts to address potential
fraud, and the extent to which DOL has comprehensively assessed fraud risks, see the enclosure
on Unemployment Insurance Fraud Risk Management in appendix I.

Methodology
To conduct this work, we analyzed regularly reported DOL data for calendar years 2019, 2020, and
2021, having obtained the most recent data on September 30, 2021. We reviewed relevant federal
laws, DOL guidance, and DOL OIG reports. We interviewed DOL oﬃcials about program data and
agency actions; we also interviewed DOL OIG oﬃcials and National Association of State Workforce
Agencies staﬀ. In addition, we reviewed data ﬁle documentation and written responses from DOL
oﬃcials. Further, we interviewed DOL oﬃcials about the UI database, PUA claims data ﬁles, and
data on outstanding federal loans to pay UI beneﬁts, speciﬁcally related to state-reported data
on claims counts, overpayments, payment timeliness, and loan balance amounts. We examined
the data for outliers, missing values, and errors. We determined that the DOL data we used were
suﬃciently reliable for the purposes of this report.

Agency Comments
We provided DOL and OMB with a draft of this enclosure. DOL provided technical comments on
this enclosure, which we incorporated as appropriate. OMB did not provide any comments.

GAO’s Ongoing Work
We continue to examine the implementation and administration of CARES Act UI programs
and the implications of high claims volumes during the pandemic on the timeliness of beneﬁt
payments and on overall program integrity. We are conducting additional work to examine
selected claimants’ experiences during the pandemic and experiences with accessing CARES
Act UI programs. We are also continuing to analyze selected states’ data on PUA beneﬁt receipt,
by race and ethnicity, as part of our ongoing work on the PUA program. In addition, we are
examining programmatic risks and challenges for the UI program as well as options for program
transformation.

GAO’s Prior Recommendations
The table below presents our recommendations on UI programs from prior bimonthly CARES Act
reports.

Page 182 GAO-22-105051
Prior GAO Recommendations Related to Unemployment Insurance (UI) Programs
Recommendation Status

The Secretary of Labor should ensure the Oﬃce of Open—partially addressed. The Department of
Unemployment Insurance collects data from states on Labor (DOL) agreed with our recommendation and
the amount of overpayments waived in the Pandemic on September 3, 2021, issued PUA program guidance
Unemployment Assistance (PUA) program, similar to the and updated instructions for states to report PUA
regular UI program (March 2021 report). overpayments waived. As of September 27, 2021, this
recommendation remained open, as PUA data were not
yet available for September 2021. We will continue to
monitor state reporting of PUA overpayments waived.

The Secretary of Labor should ensure the Oﬃce of Open—partially addressed. DOL agreed with our
Unemployment Insurance collects data from states on the recommendation and on January 8, 2021, issued
amount of overpayments recovered in the PUA program, PUA program guidance and updated instructions for
similar to the regular UI program (January 2021 report). states to report PUA overpayments recovered. As of
September 27, 2021, only 30 states had begun reporting
some data on the amount of PUA overpayments
recovered. Sustained reporting by more states is
needed to help inform DOL, policymakers, and the
public about the amount of PUA overpayments states
have recovered. We will continue to monitor state
reporting of PUA overpayment recovery data.

The Secretary of Labor should ensure the Oﬃce of Open—partially addressed. DOL partially agreed
Unemployment Insurance pursues options to report the actual with our recommendation. Speciﬁcally, DOL agreed to
number of distinct individuals claiming beneﬁts, such as by pursue options to report the actual number of distinct
collecting these already available data from states, starting individuals claiming UI beneﬁts. However, DOL did not
from January 2020 onward (November 2020 report). agree with the recommended retroactive eﬀective date
of the reporting. In a letter dated March 30, 2021, DOL
stated that it had begun developing a new state report
that would capture data related to distinct individuals
claiming regular UI beneﬁts; DOL estimated that this
data collection might begin in early 2022. DOL also
reiterated its concerns about the feasibility of states’
reporting this information retroactively, including for
the pandemic UI programs, without detracting from
their primary obligation for timely and accurate claims
processing.
We maintain that DOL should pursue options to report
the actual number of distinct individuals claiming
UI beneﬁts, retroactive to January 2020. Even if the
information is unavailable for some time, these data
are vital to understanding how many individuals
received UI beneﬁts as well as the size of the population
supported by the UI system during the pandemic. Given
the substantial investment in UI programs during the
pandemic, an accurate accounting of the size of the
population supported by this funding may be critical
to understanding the eﬃciency and eﬀectiveness of
the nation’s response to unemployment during the
pandemic. An accurate accounting may also be critical
to helping DOL and policy makers identify lessons
learned about the administration and use of regular
and expanded UI beneﬁt programs.

Page 183 GAO-22-105051
Recommendation Status

In August 2021, DOL oﬃcials said their work on the
new state report that would capture data related to
distinct individuals claiming regular UI beneﬁts had
been delayed by other competing priorities related to
implementing the American Rescue Plan Act of 2021.
The oﬃcials said that in spite of this delay, they hoped
to complete the report by early 2022.

The Secretary of Labor should ensure the Oﬃce of Closed—addressed. DOL’s weekly news release of
Unemployment Insurance revises its weekly news releases to December 10, 2020, clariﬁed that the numbers reported
clarify that in the current unemployment environment, the for weeks of UI beneﬁts claimed do not represent the
numbers it reports for weeks of unemployment claimed do not number of unique individuals claiming beneﬁts.
accurately estimate the number of unique individuals claiming
beneﬁts (November 2020 report).

The Secretary of Labor should, in consultation with the Closed—addressed. DOL issued guidance on August
Small Business Administration (SBA) and the Department 12, 2020, clarifying that individuals working full time
of the Treasury, immediately provide information to state and being paid through PPP are not eligible for UI. The
unemployment agencies that speciﬁcally addresses SBA's guidance also clariﬁed that individuals working part
Paycheck Protection Program (PPP) loans, and the risk of time and being paid through PPP would be subject to
improper payments associated with these loans (June 2020 certain state policies, including state policies on partial
report). unemployment, to determine their eligibility for UI
beneﬁts. Further, the guidance clariﬁed that individuals
being paid through PPP but not performing any services
would similarly be subject to certain provisions of
state law. Finally, the guidance noted that an individual
receiving full compensation would be ineligible for UI.

Source: GAO. I GAO-22-105051

Related GAO Products
Management Report: Preliminary Information on Potential Racial and Ethnic Disparities in the Receipt
of Unemployment Insurance Beneﬁts during the COVID-19 Pandemic. GAO-21-599R. Washington, D.C.:
June 17, 2021.

Contact information: Thomas M. Costa, (202) 512-7215, costat@gao.gov

Page 184 GAO-22-105051
Unemployment Insurance Fraud Risk Management
The amount of fraudulent and potentially fraudulent activity in unemployment insurance
programs increased substantially after the three CARES Act temporary expansions, relative to the
amount of such activity in the regular unemployment insurance program before the pandemic.
Improper payments have also been a long-standing concern in the regular unemployment
insurance program, suggesting the program may be vulnerable to fraud. The Department of
Labor continues to identify and implement strategies to address potential unemployment
insurance fraud and has ongoing program integrity activities to identify risks. However, it has not
comprehensively assessed fraud risks in alignment with leading practices identiﬁed in GAO’s Fraud
Risk Framework, which by law must be incorporated in guidelines established by the Oﬃce of
Management and Budget for agencies.

Entity involved: Department of Labor

Recommendations for Executive Action
The Secretary of Labor should designate a dedicated entity and document its responsibilities
for managing the process of assessing fraud risks to the unemployment insurance program,
consistent with leading practices as provided in our Fraud Risk Framework. This entity should
have, among other things, clearly deﬁned and documented responsibilities and authority for
managing fraud risk assessments and for facilitating communication among stakeholders
regarding fraud-related issues.

The Secretary of Labor should identify inherent fraud risks facing the unemployment insurance
program.

The Secretary of Labor should assess the likelihood and impact of inherent fraud risks facing the
unemployment insurance program.

The Secretary of Labor should determine fraud risk tolerance for the unemployment insurance
program.

The Secretary of Labor should examine the suitability of existing fraud controls in the
unemployment insurance program and prioritize residual fraud risks.

The Secretary of Labor should document the fraud risk proﬁle for the unemployment insurance
program.

The Department of Labor neither agreed nor disagreed with our recommendation to designate
a dedicated entity for managing the process for assessing fraud risks within the unemployment
insurance program. The department stated that its Chief Financial Oﬃcer and the Employment
and Training Administration’s Assistant Secretary are responsible for risk assessment and
management in the UI program. However, it is important that, consistent with our Fraud Risk
Framework, the Department of Labor clearly document this designation and these oﬃcials’
antifraud responsibilities.

Page 185 GAO-22-105051
The Department of Labor neither agreed nor disagreed with our other ﬁve recommendations.
The department stated that its current process allows it to identify, evaluate, and manage risks.
However, the department also said it will incorporate the recommended practices and approaches
moving forward.

Background
The unemployment insurance (UI) system is a federal–state partnership that provides temporary
ﬁnancial assistance to eligible workers who become unemployed through no fault of their
own. States design and administer their own UI programs within federal parameters, and the
Department of Labor (DOL) oversees states’ compliance with federal requirements, such as by
ensuring states pay beneﬁts when they are due. Regular UI beneﬁts—those provided under the
state UI programs in place before the CARES Act was enacted—are funded primarily through
state taxes levied on employers and are intended to typically be lower than a claimant’s previous
employment earnings, according to DOL.300

The CARES Act created three federally funded temporary UI programs that expanded beneﬁt
eligibility and enhanced beneﬁts, which were amended by the Consolidated Appropriations Act,
2021 and the American Rescue Plan Act of 2021.301 These programs expired on September 6,
2021, although some states ended their participation before that date.

300To be eligible for regular UI beneﬁts, applicants generally must be able and available to work and actively seeking
work. 42 U.S.C. § 503(a)(12). Administration of the regular UI program is ﬁnanced by a federal tax on employers,
according to DOL.
301The CARES Act also addressed other aspects of the UI system, such as authorizing certain ﬂexibilities for states to
hire additional staﬀ and funding for Short-Time Compensation programs. In addition to the CARES Act, the Families
First Coronavirus Response Act provided up to $1 billion in emergency grant funding to states in ﬁscal year 2020 for UI
administrative purposes. In addition, on August 8, 2020, the President signed a memorandum directing the Department
of Homeland Security’s Federal Emergency Management Agency (FEMA) to provide up to $44 billion in lost wages
assistance. Pursuant to the presidential memorandum, upon receiving a FEMA grant, states and territories could provide
eligible claimants $300 or $400 per week—which included a $300 federal contribution—in addition to their UI beneﬁts.
The White House, Memorandum on Authorizing the Other Needs Assistance Program for Major Disaster Declarations Related
to Coronavirus Disease 2019 (Aug. 8, 2020). FEMA approved 53 states and territories to provide lost wages assistance to
eligible claimants for a maximum of 6 weeks of unemployment experienced from the week ending on August 1, 2020,
through the week ending on September 5, 2020. These 53 states and territories included the District of Columbia. For
the purpose of this enclosure, we categorize the District of Columbia as a state.
302At the time of the program’s expiration, PUA generally authorized up to 79 weeks of beneﬁts. Pub. L. No. 117-2, §
9011(a), (b), 135 Stat. 4, 118; Pub. L. No. 116-260, div. N, tit. II, § 201(a), (b), 134 Stat. 1182, 1950-1951 (2020); Pub. L.
No. 116-136, § 2102, 134 Stat. 281, 313 (2020).

Page 186 GAO-22-105051
2. Federal Pandemic Unemployment Compensation (FPUC) generally authorized an additional
weekly beneﬁt for individuals who were eligible for weekly beneﬁts under the regular UI and
CARES Act UI programs.303

3. Pandemic Emergency Unemployment Compensation (PEUC) generally authorized additional
weeks of UI beneﬁts for those who had exhausted their regular UI beneﬁts.304

In addition, the Consolidated Appropriations Act, 2021 created the Mixed Earner Unemployment
Compensation (MEUC) program, which was extended by the American Rescue Plan Act of 2021.305
According to DOL, the MEUC program was intended to cover regular UI claimants whose beneﬁts
do not account for signiﬁcant self-employment income and who thus may have received a lower
regular UI beneﬁt than the beneﬁt they would have received had they been eligible for PUA.306

See the enclosure on Unemployment Insurance Programs in appendix I for additional background
information.

Federal and state entities continue to investigate and report on high levels of fraud, potential
fraud, and fraud risks in the UI programs.307 At the federal level, DOL’s Oﬃce of Inspector General
(OIG) and the Department of Justice (DOJ) continue to investigate potential UI fraud and highlight
the level of fraud, potential fraud, and fraud risks in the UI programs. As of July 14, 2021, DOL OIG
oﬃcials reported that they had opened more than 21,000 complaints and investigations involving
alleged UI fraud since the pandemic began. The DOL OIG last reported that it had identiﬁed and
recovered more than $160 million in UI fraud, according to the DOL OIG’s website.

303FPUC generally authorized an additional $600 beneﬁt through July 2020 and authorized an additional $300
beneﬁt for weeks beginning after December 26, 2020, through the end of the program. Pub. L. No. 117-2, § 9013,
135 Stat. 4, 119; Pub. L. No. 116-260, div. N, tit. II, § 203, 134 Stat. 1182, 1953; Pub. L. No. 116-136, § 2104, 134 Stat.
at 318.
304At the time of the program’s expiration, PEUC generally authorized an additional 53 weeks of beneﬁts for
claimants who were fully unemployed. Pub. L. No. 117-2, § 9016(a), (b), 135 Stat. 4, 119-120; Pub. L. No. 116-260, div.
N, tit. II, § 206(a), (b), 134 Stat. 1182, 1954; Pub. L. No. 116-136, § 2107, 134 Stat. at 323.
305The MEUC program, which was voluntary for states, authorized an additional $100 weekly beneﬁt for certain UI
claimants who received at least $5,000 of self-employment income in the most recent tax year prior to their application
for UI beneﬁts. Pub. L. No. 117-2, § 9013(a), 135 Stat. 4, 119; Pub. L. No. 116-260, div. N, tit. II, § 261(a)(1), 134 Stat. 1182,
1961. The $100 weekly beneﬁt was in addition to other UI beneﬁts received by claimants; however, individuals receiving
PUA beneﬁts could not receive MEUC payments.
306According to DOL, 51 states and territories elected to participate in the MEUC program, and Idaho and South Dakota
opted not to participate. However, 23 states terminated their participation in June or July 2021, before the program
expiration date, and one state intended to terminate participation but did not because of litigation at the state level,
according to DOL. The remaining 27 states and territories continued participating in the MEUC program until it expired
in September 2021. According to DOL, when the MEUC program expired, the following 18 states and territories had
not begun paying beneﬁts: Colorado, Georgia, Illinois, Iowa, Kansas, Louisiana, Nevada, New Jersey, Ohio, Oklahoma,
Oregon, Rhode Island, Texas, the U.S. Virgin Islands, Vermont, Washington, West Virginia, and Wyoming. [Note: DOL’s
information on states that had yet to make ﬁrst MEUC payments was current as of July 29, 2021]. Before paying any
beneﬁts, Wyoming submitted notice to DOL that it planned to return all MEUC funds and not to operate the program,
according to DOL.
307Fraud involves obtaining something of value through willful misrepresentation. Whether an act is fraudulent is
determined through the judicial or other adjudicative systems. In this report, “fraud risk” includes existing circumstances
that provide an opportunity to commit fraud.

Page 187 GAO-22-105051
DOL OIG eﬀorts have also identiﬁed billions of dollars in potential UI fraud under investigation. In
June 2021, DOL’s OIG reported that it had identiﬁed nearly $8 billion of potentially fraudulent UI
beneﬁts paid from March 2020 through October 2020.308

In addition, since March 2020, DOJ has publicly announced charges in numerous fraud-related
cases related to the UI programs. Speciﬁcally, from March 2020 through July 2021, 71 individuals
pleaded guilty to federal charges of defrauding UI programs and federal charges were pending
against 192 individuals.309 See the enclosure on Federal Fraud-Related Cases in appendix I for
more information about DOJ charges.

Several state auditors have also reported on fraud, potential fraud, and fraud risks in the UI
programs. For example, state auditors in California, Louisiana, and Kansas identiﬁed millions of
dollars in potentially fraudulent payments.310

Overview of Key Issues
Our review of DOL OIG reports, state audits, and DOJ cases identiﬁed several fraud risks in
UI programs. On the basis of our review of DOL OIG reports, state audit reports, and DOJ cases,
we identiﬁed the following fraud risks in UI programs:311

• Applicants’ falsifying information on income or employment eligibility to receive beneﬁts. For
example, an individual who pleaded guilty to falsely reporting income information to receive UI
beneﬁts, including the additional $600 available through FPUC, obtained more than $13,000 in
fraudulent beneﬁts.

• Applicants’ using stolen identities or personally identiﬁable information (PII) to apply for beneﬁts
or receive beneﬁts. For example, an individual who pleaded guilty to ﬁling fraudulent UI claims
used the stolen identities of dozens of individuals to obtain more than $500,000.

• Applicants’ applying for, or receiving, beneﬁts by using fake identity information. For example, an
individual who pleaded guilty to applying for UI beneﬁts by using a fake Social Security number
obtained an unspeciﬁed amount of fraudulent beneﬁts.

308For more information, see Department of Labor, Oﬃce of Inspector General, Alert Memorandum: The Employment
and Training Administration Needs to Issue Guidance to Ensure State Workforce Agencies Provide Requested Unemployment
Insurance Data to the Oﬃce of Inspector General, Report No. 19-21-005-03-315 (Washington, DC: June 16, 2021).
309For the purpose of this enclosure, federal charges are deﬁned as criminal complaints, criminal informations, and
indictments. A charge is merely an allegation, and all defendants are presumed innocent until proven guilty beyond a
reasonable doubt in a court of law. Twenty of the individuals who pleaded guilty had been sentenced as of July 31, 2021;
sentences ranged from 3 months to 10 years in prison.
310Auditor of the State of California, Employment Development Department: Signiﬁcant Weaknesses in EDD’s Approach to
Fraud Prevention Have Led to Billions of Dollars in Improper Beneﬁt Payments, Report 2020-628.2 (Sacramento, CA: Jan. 28,
2021). Louisiana Legislative Auditor, Louisiana Workforce Commission, Improper Payments in the Unemployment Insurance
Program: Ineligible Incarcerated Recipients, (Baton Rouge, LA: Apr. 28, 2021). Kansas Legislative Division of Post Audit,
Evaluating the Kansas Department of Labor’s Response to COVID-19 Unemployment Claims (Part I), Report No. R-21-003
(Topeka, KS: February 2021).
311The fraud risks identiﬁed in this report do not constitute an exhaustive list of all fraud risks aﬀecting the UI programs.

Page 188 GAO-22-105051
• Applicants’ submitting fraudulent claims or erroneously receiving beneﬁts in multiple states. For
example, an individual who pleaded guilty to applying for, and in many cases receiving, UI
beneﬁts from state administrators in at least 40 diﬀerent states and the District of Columbia
obtained approximately $350,000 in fraudulent beneﬁts.

• Prison inmates’ applying for beneﬁts for which they were not eligible. For example, in a
coordinated scheme, three individuals—two of whom have pleaded guilty—collected PII and
ﬁled fraudulent claims on behalf of more than 35 co-conspirators who were ineligible for
beneﬁts, including 15 incarcerated individuals. Using co-conspirators’ PII, the three individuals
ﬁled 37 fraudulent claims that resulted in payment of at least $499,000 in beneﬁts.

• Current or former federal or state or territory employees’ misusing their positions to fraudulently
obtain beneﬁts for themselves or others. For example, a former territory employee pleaded
guilty to federal program theft and accessing a protected computer in furtherance of fraud
in connection with a scheme to defraud UI. He ﬁled for pandemic-related unemployment
beneﬁts on his own behalf, falsely claiming that he was unemployed and eligible. In addition,
he used his position as a customer service representative to submit fraudulent UI claims
on behalf of others, and he directed the beneﬁts from these fraudulent claims into a bank
account he controlled. This individual obtained, or attempted to obtain, a total of $93,000 in
beneﬁts. He was sentenced to 10 months imprisonment followed by 3 years of supervised
release and was ordered to pay more than $14,000 in restitution.

The amount of fraudulent and potentially fraudulent activity in UI programs increased
substantially after the three CARES Act temporary expansions, relative to the amount of such
activity in the regular UI program before the pandemic. The high beneﬁt payment amounts
oﬀered during the pandemic created a signiﬁcant ﬁnancial incentive for fraudsters to target the
UI program, according to oﬃcials from the National Association of State Workforce Agencies
(NASWA).312 In addition, DOL oﬃcials stated that the CARES Act UI programs are a key target for
fraud because fraudsters can receive a large amount of money in one payment.

DOL, DOL OIG, and NASWA have identiﬁed factors that contribute to fraud risk in the UI
programs. Information about fraud schemes in a program is a useful source for understanding
fraud risks. According to DOL oﬃcials, fraud in the regular UI program has historically involved
a misrepresentation of eligibility, such as an employee’s failing to report returning to work,
failing to report earned wages, or failing to fulﬁll work search requirements or an employer’s
failing to report a reason for separation. Oﬃcials stated that although DOL was aware of isolated
occurrences of identity-related fraud before the pandemic, such as the use of false identities, it
has seen an increase in the frequency and volume of identity-related fraud, as well as signiﬁcantly
more sophisticated fraud schemes, since the pandemic began.

According to DOL oﬃcials, the department is aware of fraud risks in CARES Act UI programs,
particularly risks of fraudulent claim schemes and of fraudulent individual claims for beneﬁts. In
its ﬁscal year 2020 agency ﬁnancial report, DOL acknowledged an increase in suspected fraudulent

312NASWA represents all 50 state workforce agencies, the District of Columbia, and U.S. territories.

Page 189 GAO-22-105051
activity—speciﬁcally, organized fraud schemes targeting the CARES Act UI program.313 Moreover,
according to NASWA oﬃcials, the UI system has faced unrelenting attacks by foreign organized
crime groups during the pandemic. According to DOL oﬃcials, the most common fraud schemes
during the pandemic have included the use of stolen PII to ﬁle a claim or multiple claims; the
use of synthetic identities (i.e., real identities mixed with ﬁctitious information); and the use of
bot attacks in attempts to overwhelm state UI systems or launch phishing schemes to obtain
individual PII to perpetrate future fraud.314

DOL oﬃcials also told us that they have observed the use of new fraud schemes targeting CARES
Act UI programs.

• Hijacking of bank accounts. After an individual submits a legitimate application for UI beneﬁts
and provides bank account information for the funds’ direct deposit, a fraudster will hack into
the applicant’s UI system account and reroute the deposit from the applicant’s bank account
to a bank account the fraudster can access.

• Mimicking of state UI websites. When people conduct Internet searches for their state’s UI oﬃce,
they may ﬁnd, and ﬁle claims on, a fraudulent website that looks like the state workforce
agency’s website, thus providing their PII to fraudsters.

Factors contributing to fraud risk include conditions or actions that are most likely to cause or
increase the possibility of fraud. Through our review of DOL OIG reports, state audit reports, and
DOJ cases, we identiﬁed the following fraud risk factors in CARES Act UI programs:315

• Reliance on self-certiﬁcation. The CARES Act allowed PUA applicants to self-certify their
eligibility and did not require them to provide any documentation of self-employment or
prior income. In October 2020, DOL’s OIG reported that states cited the PUA self-certiﬁcation
requirement as a top fraud vulnerability.316 We have previously reported that relying on
program participants to self-report and self-certify information on agency forms, instead of
verifying such information independently, could cause an agency to miss opportunities to
prevent program fraud and abuse. To help address this risk, the Consolidated Appropriations
Act, 2021, enacted in December 2020, included a requirement for individuals to submit
documentation of employment or self-employment when applying for PUA.317

313Department of Labor, Agency Financial Report Fiscal Year 2020 (Washington, DC: Nov. 16, 2020).
314A bot, or bot-net, is a type of cybersecurity threat. Bot-net operators use a network, or bot-net, of compromised,
remotely controlled systems to coordinate attacks and distribute phishing schemes, spam, and malware attacks. The
services of these networks are sometimes available on underground markets (e.g., through the purchase of a denial-of-
service attack or services to relay spam or phishing attacks).
315Factors contributing to fraud risk do not necessarily indicate that fraud exists but are often present when fraud
occurs.
316Department of Labor, Oﬃce of Inspector General, COVID-19: States Cite Vulnerabilities in Detecting Fraud While
Complying with the CARES Act UI Program Self-Certiﬁcation Requirement, Report No. 19-21-001-03-315 (Washington, DC:
Oct. 21, 2020).
317The Consolidated Appropriations Act, 2021 extended the PUA program and included a requirement that
all PUA claimants must submit documentation substantiating employment, self-employment, or the planned

Page 190 GAO-22-105051
• Waiver of waiting period. During the pandemic, states were encouraged to process and pay
claims quickly while experiencing a historic number of claims. In an eﬀort to speed claims
processing, DOL encouraged states to temporarily suspend the existing waiting period for
beneﬁts and the CARES Act provided full federal funding for the ﬁrst week of regular UI
beneﬁts to states that did so. According to DOL oﬃcials, under the regular UI program, DOL
allows states to take up to 21 days to make the ﬁrst payment of beneﬁts, giving them time to
detect potential fraud. Waiving the waiting period meant that some states had less time to
employ tools for fraud prevention and detection, according to NASWA oﬃcials.

• Low staﬃng levels and antiquated IT systems. In the beginning of the pandemic, outdated IT
systems and low staﬃng levels made it diﬃcult for many states to eﬃciently process the
unprecedentedly high volume of claims and conduct internal control activities, according to
DOL OIG oﬃcials. These oﬃcials also told us that state agency staﬃng levels are determined
on the basis of claim volume levels in previous years. At the start of the pandemic, many state
UI programs had been experiencing their lowest claims volume, and thus their lowest staﬃng
and funding levels, since the 1970s. To process the high volume of claims after the pandemic
began, many states reassigned beneﬁt payment control staﬀ to claims processing, with the
result that few staﬀ were working to prevent and detect fraud, according to DOL OIG oﬃcials.
Additionally, DOL OIG oﬃcials stated that some state IT systems were not equipped to handle
the volume of claims and some may not have been easily compatible with the NASWA UI
Integrity Center’s Integrity Data Hub resources.318

• Variation in data analysis across states. States’ use of data mining, cross matching, and identity
veriﬁcation resources varies. According to DOL oﬃcials, the department does not have
authority to require states use the databases available in the UI Integrity Center’s Integrity
Data Hub, such as the Identify Veriﬁcation or Multi-State Cross Match databases. Additionally,
not all states were able to cross-match claims with federal incarceration data and many states
did not have access to state-level incarceration data.

DOL continues to identify and implement strategies to address potential fraud in the UI
programs. DOL has taken steps to prevent and detect fraud in UI programs, including the CARES
Act UI programs. In its ﬁscal year 2020 agency ﬁnancial report, DOL highlighted a number of these
steps, including (1) working with states and DOL’s OIG to employ data mining and data analytics to
detect fraud when a claim is ﬁled, (2) holding routine conference calls with DOL’s OIG and states to
discuss and share information on UI fraud schemes, and (3) establishing training.

In May 2020, DOL began issuing speciﬁc fraud prevention and detection guidance in response to
the challenges faced by state workforce agencies and the increase in suspected fraud, encouraging
states to continue to uphold the integrity of the UI programs and to use the resources provided
through the NASWA UI Integrity Center, including the Integrity Data Hub.319 In addition, the

commencement of employment or self-employment. Pub. L. No. 116-260, div. N, tit. II, § 241(a), 134 Stat. 1182,
1959-60.
318The Integrity Data Hub is a centralized, multistate data system that the UI Integrity Center operates in
partnership with DOL, using DOL funding.
319According to DOL, this guidance included UIPL 23-20, UIPL 28-20, and Training and Employment Notice Nos. 03-20,
04-20, and 05-20. To facilitate the identiﬁcation of potential fraud, states can access the NASWA UI Integrity Center’s

Page 191 GAO-22-105051
guidance highlighted states’ responsibilities to perform required cross-matches to prevent
potential fraud.320 The UI Integrity Center also provides a fraud-alert application to facilitate
information sharing about fraud schemes between states and the DOL OIG. According to DOL
oﬃcials, the department continues to invest in the Integrity Data Hub to enhance cross-matching
functionality and ensure states have access to key fraud detection services. Oﬃcials said that key
investments include the Multi-State Cross-match, the Identity Veriﬁcation solution, and Account
Veriﬁcation Services.

Also, as we have previously reported, DOL made two allotments of $100 million available to
states, in September 2020 and January 2021, respectively, to address potential fraud and identity
theft in the PUA and PEUC programs.321 According to DOL, states have reported using the funds
from the January 2021 allotment to, among other things, hire additional data analytic and fraud
investigator staﬀ, hire third-party vendors to conduct fraud risk and cybersecurity assessments of
states’ UI systems, subscribe to identity veriﬁcation and identity prooﬁng solutions, and coordinate
with ﬁnance departments and law enforcement to aid in the recovery of overpaid beneﬁts.
States also reported using this funding to make updates to their systems, including purchasing
fraud prevention software, fraud detection software, and data mining tools. In addition, states
reported enhancing their UI systems by implementing multifactor authentication, automating
claim validation, and creating portals for individuals to report fraud.322

The American Rescue Plan Act of 2021 provided DOL with $2 billion to combat potential fraud,
among other things.323 In response, in August 2021, DOL oﬃcials issued a four-part plan for using
this $2 billion that includes the following actions:324

• Sending experts directly to states to help them address challenges related to fraud, beneﬁt access,
and beneﬁt timeliness. DOL oﬃcials said they have sent teams of experts in fraud, technology,
and equity to help states identify claims processing deﬁciencies as well as opportunities
to address equity in the UI system.325 According to oﬃcials, these teams will focus on
operations, communications, and process ﬂows. The oﬃcials said the teams will also focus on
technologies that support the timely distribution of beneﬁts, reduction of backlogs, equitable
access to UI beneﬁts, and fraud prevention and detection. Six states—Colorado, Kansas,
Nevada, Virginia, Washington, and Wisconsin—have volunteered to work with the teams, and
oﬃcials said they are recruiting the next group of states. Oﬃcials also plan to develop a federal

Integrity Data Hub, operated in partnership with DOL. The UI Integrity Center operates the Integrity Data Hub, which
allows states to submit claims for cross-matching and supports data analytics on multistate claims.
320The required cross matches include the National Directory of New Hires Cross-match, Quarterly Wage Records Cross-
match, and the Systematic Alien Veriﬁcation for Entitlement.
321We previously reported on how states used the funds from the September allotment.
322DOL monitors states’ use of funds and progress in carrying out these eﬀorts through quarterly reports from states.
323The American Rescue Plan Act of 2021 provided DOL with $2 billion to detect and prevent fraud, promote equitable
access, and ensure the timely payment of UI beneﬁts.
324Department of Labor, “Fact Sheet: Unemployment Insurance Modernization: American Rescue Plan Act Funding
for Timely, Accurate and Equitable Payment in Unemployment Compensation Programs,” Aug. 11, 2021, https://
oui.doleta.gov/unemploy/pdf/FactSheet_UImodernization.pdf.
325We previously reported on potential inequities in the receipt of UI beneﬁts by race and ethnicity in some states.
See GAO, Management Report: Preliminary Information on Potential Racial and Ethnic Disparities in the Receipt of
Unemployment Insurance Beneﬁts during the COVID-19 Pandemic, GAO-21-599R (Washington, DC: June 17, 2021).

Page 192 GAO-22-105051
team that will supervise the state-level teams and maintain a repository of common problems
and solutions. According to DOL oﬃcials, the federal team will also enhance DOL’s technical
assistance capacity.

• Providing states with grants to combat potential fraud and address equity. DOL oﬃcials said they
plan to provide states with additional grants to help them combat potential fraud.326 States
will be able to use up to $140 million in grants provided under one of the allotments for the
regular UI program, unlike previous allotments, which were focused on the PUA and PEUC
programs. In addition, states receiving these grant funds must provide claims data to the DOL
OIG during the duration of the grant for the purposes of investigating fraud and conducting
audits. Oﬃcials also plan to provide states with grants to address equity issues, such as by
improving access to the regular UI program for individuals with disabilities or individuals who
have limited or no internet access.327 According to DOL oﬃcials, states seeking these funds are
required to identify the equity gaps they plan to address and the metrics they plan to use to
measure improvement in equitable access.

• Supporting states in fully modernizing their IT systems. DOL oﬃcials said they plan to support
states eﬀorts to fully modernize their IT systems. The oﬃcials said they plan to select states
that are in various stages of IT modernization and work with them to develop modular
technology solutions that can be integrated with state systems.

• Contracting with vendors that can assist states in preventing identity fraud. According to DOL
oﬃcials, they have awarded purchase agreements to three vendors that states can use to
combat identity-related fraud. Oﬃcials noted that they have concerns about the equity of
states’ processes for ﬂagging claims for potential fraud. The oﬃcials also noted that they
hope to identify any gaps in applicants’ access to beneﬁts that may result from states’ identity
veriﬁcation eﬀorts.

In addition to working with states, DOL oﬃcials said they continue to collaborate with a variety
of other entities to address fraud prevention, detection, and recovery. According to oﬃcials, they
are working with the Social Security Administration to establish a secure data exchange that will
allow states to cross-match UI claims data with incarceration records. Oﬃcials said they also
meet regularly with the DOL OIG to discuss emerging UI fraud issues, streamline communication
with states, and coordinate fraud prevention and recovery eﬀorts. In addition, oﬃcials said they
continue to participate in biweekly banking work group calls to discuss ongoing recovery eﬀorts
and improve communication among banking organizations, federal government agencies, and
law enforcement agencies. Further, DOL has issued guidance requiring states to share claims data
with the DOL OIG relating to the federal pandemic-related UI programs.

326For more information about these fraud prevention grants, see Department of Labor, Additional Funding to
Assist with Strengthening Fraud Detection and Prevention Eﬀorts and the Recovery of Overpayments in the Pandemic
Unemployment Assistance (PUA) and Pandemic Emergency Unemployment Compensation (PEUC) Programs, as well
as Guidance on Processes for Combatting Identity Fraud, UIPL 28-20, Change 2 (Washington, DC: Aug. 11, 2021).
Also see Department of Labor, Grant Opportunity to Support States with Fraud Detection and Prevention, Including
Identity Veriﬁcation and Overpayment Recovery Activities, in All Unemployment Compensation (UC) Programs, UIPL 22-21
(Washington, DC: Aug. 11, 2021).
327For more information about these equity grants, see Department of Labor, Grant Opportunity for Promoting
Equitable Access to Unemployment Compensation (UC) Programs, UIPL 23-21 (Washington, DC: Aug. 17, 2021).

Page 193 GAO-22-105051
Although DOL has taken steps to identify risks through program integrity activities, it has
not comprehensively assessed UI fraud risks. Although DOL has various program integrity
eﬀorts in place to identify risks in the regular UI program and the CARES Act UI programs and is
conducting ongoing eﬀorts that could inform a fraud risk assessment, it has not comprehensively
assessed UI fraud risks in alignment with leading practices or documented a prioritized approach
to managing fraud risks.

DOL conducts various program integrity eﬀorts focused on identifying risks, including fraud risks,
such as the following:

• Enterprise risk management (ERM) eﬀort. According to a DOL oﬃcial, DOL developed a risk
register in 2019 as part of an agency-wide ERM eﬀort.328 This oﬃcial also told us that staﬀ
responsible for managing the UI programs provided input about the risk register and ﬂagged
identity-related fraud as an identiﬁed and targeted risk.

• Entity Level Controls Survey. DOL reported that it conducts the annual Entity Level Controls
Survey of DOL agency management and personnel to review risks related to ﬁnancial and
internal controls. According to DOL, the Entity Level Controls Survey canvasses agencies
to assess fraud at an agency and program level. These surveys assist in assessing and
maintaining a well-informed and fraud-aware culture and, according to DOL, have aﬃrmed the
implementation of antifraud controls.

• Quarterly state risk assessments and reviews. DOL’s regional oﬃces conduct quarterly state risk
assessments and reviews. The risk assessments are designed to identify potential risk factors
that may adversely aﬀect a state’s program administration. During these quarterly reviews,
DOL staﬀ may also conduct program integrity reviews to ensure that states have operational
processes in place to prevent, detect, and recover improper and fraudulent payments and to
keep the UI improper payment rate as low as feasible.

• Beneﬁt Accuracy Measurement Program. Oﬃcials told us that DOL collects information about
UI fraud from states’ independent assessments of the accuracy of UI beneﬁt payments and
their estimates of improper payments, including fraud. See the enclosure on Unemployment
Insurance Programs in appendix I for additional details about DOL’s improper payment
estimates.

• Other antifraud eﬀorts. DOL oﬃcials also cited the ﬁscal year 2020 agency ﬁnancial report for
a signiﬁcant discussion of antifraud eﬀorts. In its most recent agency ﬁnancial report, DOL
reported that in ﬁscal year 2020, it conducted a variety of eﬀorts to ascertain fraud risk. These
eﬀorts included reviews of risks aﬀecting all DOL programs, required by the Payment Integrity
Information Act of 2019.329

328On March 22, 2019, the Secretary of Labor announced the establishment of the DOL Enterprise Risk
Management Council. According to DOL, its ERM initiative provides leaders with an integrated, enterprisewide view
of risk, risk tolerances, and risk mitigation eﬀorts.
329In March 2020, Congress enacted the Payment Integrity Information Act of 2019, which repealed the Improper
Payments Information Act of 2002, the Improper Payments Elimination and Recovery Act of 2010, and the Improper
Payments Elimination and Recovery Improvement Act of 2012, and enacted substantially similar provisions

Page 194 GAO-22-105051
The Oﬃce of Management and Budget (OMB) requires agencies to assess the full spectrum of an
organization’s risks and to identify those that are enterprise-level risks. Agencies are expected to
rate those enterprise risks in terms of impact, build internal controls to monitor and assess the
risk developments at various points in time, and incorporate risk awareness into their culture and
operations.330 While fraud risks can be a part of a broader assessment of enterprise risk such as
ERM, such assessments do not take into account actions that agencies must undertake to manage
fraud risks.

Our Fraud Risk Framework acknowledges that agencies may use initiatives such as ERM to assess
their fraud risks. However, agencies, including DOL, must nonetheless fulﬁll the separate and
independent fraud risk management requirements initially established by the Fraud Reduction
and Data Analytics Act of 2015. The act, enacted in June 2016, required OMB to establish
guidelines for federal agencies to create controls to identify and assess fraud risks and to design
and implement antifraud control activities.331 The act further required OMB to incorporate the
leading practices from the Fraud Risk Framework in the guidelines. In its guidelines, OMB directed
that agencies should adhere to the Fraud Risk Framework’s leading practices as part of their
eﬀorts to eﬀectively design, implement, and operate an internal control system that addresses
fraud risks. Managers are responsible for determining the extent to which the leading practices in
the framework are relevant to their program and for tailoring the practices, as appropriate, to align
with the program’s operations.

Although DOL has various program integrity eﬀorts focused on identifying risks, including fraud
risks, DOL has not clearly assigned deﬁned responsibilities to a dedicated entity for designing
and overseeing fraud risk management activities such as managing the fraud risk assessment
process.332 DOL indicated that it has not previously designated oﬃcials to be dedicated speciﬁcally
to managing fraud risks because it manages these risks within the execution of existing payment
integrity and ERM activities. Speciﬁcally, DOL oﬃcials noted that the agency’s Chief Financial
Oﬃcer and the Employment and Training Administration’s Assistant Secretary are the senior
executive oﬃcials responsible for payment integrity. DOL oﬃcials further clariﬁed that this
responsibility concerns risk management for fraudulent and nonfraudulent improper payments.
Additionally, oﬃcials stated that the Employment and Training Administration has created, and
will continue to build, a designated fraud team that will coordinate antifraud eﬀorts. However,
DOL oﬃcials did not provide documentation on the structure and fraud risk management
responsibilities and authorities planned for this entity.

to replace them. Pub. L. No. 116-117, 134 Stat. 113 (2020) (codiﬁed at 31 U.S.C. §§ 3351-3358). The Payment
Integrity Information Act also enacted some enhancements to improper payments law, including more detailed
requirements for agency risk assessments. 31 U.S.C. § 3352.
330Oﬃce of Management and Budget, OMB Circular No. A-123, Management's Responsibility for Enterprise Risk Management
and Internal Control, M-16-17 (July 2016).
331Although the Fraud Reduction and Data Analytics Act of 2015 was repealed in March 2020, the Payment Integrity
Information Act of 2019 requires these guidelines to remain in eﬀect, subject to modiﬁcation by OMB as necessary and
in consultation with GAO. Pub. L. No. 116-117, § 2(a), 134 Stat. 113, 131 - 132 (2020), codiﬁed at 31 U.S.C. § 3357. OMB
established guidelines in OMB Circular No. A-123.
332Although the UI program is a federal–state partnership, our Fraud Risk Framework states that managers of federal
programs maintain primary responsibility for enhancing program integrity.

Page 195 GAO-22-105051
The Fraud Risk Framework’s leading practices note that enterprise-wide risk management
processes may have functions that overlap with fraud risk management. Thus, enterprise-wide
risk management processes may incorporate the roles and responsibilities of the antifraud entity
speciﬁed in our leading practices in order to implement a strategic approach for assessing and
managing fraud risks.333 The framework notes that such an approach includes designating an
entity with deﬁned responsibilities and the necessary authority across the program to lead fraud
risk management activities.

Without a dedicated entity with deﬁned responsibilities to lead antifraud initiatives, including
the process of assessing fraud risks to UI programs, DOL may not be strategically managing its
UI fraud risks. For example, a dedicated antifraud entity could be held responsible for, among
other activities, managing the fraud risk assessment process and coordinating antifraud initiatives
across an agency’s various programs to assure that agency activities called for by the Fraud Risk
Framework are conducted. These activities include

• identifying inherent fraud risks facing the agency’s programs,

• assessing the likelihood and impact of inherent fraud risks facing these programs,

• determining fraud risk tolerance,

• examining the suitability of existing fraud controls and prioritizing residual fraud risks, and

• documenting the programs’ fraud risk proﬁles.

DOL’s ongoing program integrity eﬀorts could inform a fraud risk assessment; however, the
agency has not comprehensively assessed UI fraud risks in alignment with leading practices. The
second component of the Fraud Risk Framework—“Assess”—calls for federal managers to plan
regular fraud risk assessments and assess risks to determine a fraud risk proﬁle. Speciﬁcally,
leading practices include tailoring the fraud risk assessment to the program and planning to
conduct the assessment at regular intervals and at the time of any changes to the program or
operating environment.

DOL provided, as evidence of its assessment of fraud risks, the template it used to develop the
2019 risk register (i.e., a compilation of risk across the organization). According to DOL oﬃcials,
DOL had developed this template for ERM eﬀorts. In addition, DOL oﬃcials cited the department’s
ﬁscal year 2021 Employment and Training Administration Operating Plan, which incorporates
information from the risk register, according to the oﬃcials. They also pointed to a quarterly
Integrity Strategic Plan progress report and additional ERM documents.

These documents demonstrate that the agency has taken some steps to develop an antifraud
strategy and evaluate risk in business processes at DOL. However, these documents do not
provide evidence that DOL has fully considered the speciﬁc UI fraud risks, analyzed the potential

333The Fraud Risk Framework helps managers meet their responsibilities to assess and manage fraud risks as required
by federal internal control standards.

Page 196 GAO-22-105051
likelihood and impact of fraud schemes, and documented prioritized fraud risks, as called for in
leading practices.

In addition, DOL developed the 2019 risk register before the implementation of the CARES Act
UI programs so the register does not incorporate fraud risks speciﬁc to these programs. A fraud
risk assessment is particularly informative when the volume and nature of fraud risk indicators
signiﬁes an evolving fraud risk landscape, such as that observed in the CARES Act UI programs.
Such an assessment helps program managers determine whether they have the right controls in
place to combat the most likely and impactful risks facing the program.

Without comprehensively assessing UI fraud risks—including risks during the pandemic—in
accordance with leading practices, DOL lacks reasonable assurance that it has identiﬁed the most
signiﬁcant fraud risks for the regular UI program that will exist after the pandemic. For example,
some fraud risks identiﬁed in the CARES Act UI programs may continue to exist in the regular
UI program after the expiration of the temporary UI programs. An analysis of fraud risks across
all UI programs would also help DOL determine whether to establish additional fraud controls
for the regular UI program and whether to make adjustments to existing controls. In addition,
incorporating this analysis in a documented fraud risk proﬁle and its broader antifraud strategy
could position DOL to deal more eﬀectively with any future emergency UI programs.

Continuing to collaborate with state workforce agencies—consistent with leading practices for
stakeholder involvement—could also help DOL communicate information on fraud risks and
emerging fraud schemes and share lessons learned related to fraud control activities. In addition,
DOL could beneﬁt from involving state workforce agencies in assessing fraud risks to UI programs
and sharing information about fraud risks with state auditors. Fraud risk assessments that involve
relevant internal and external stakeholders are more likely to be successful and to reﬂect a
complete understanding of fraud risks and control vulnerabilities.

Eﬀective fraud risk management helps ensure that federal programs’ services fulﬁll their intended
purpose, funds are spent eﬀectively, and assets are safeguarded. If an agency is not managing
fraud risks eﬀectively in a steady state, it will not be positioned to do so during emergencies. The
purpose of proactively managing fraud risks is to facilitate, not hinder, a program’s mission and
strategic goals by ensuring that taxpayer dollars and government services serve their intended
purposes and by maintaining public trust in government. If done well, fraud risk management
would allow DOL to both enhance the eﬃcient and eﬀective delivery of UI beneﬁts to those in
need and help prevent and detect fraudsters seeking beneﬁts to which they are not entitled.

Methodology
To conduct this work, we reviewed information from DOJ to identify federal fraud–related
charges related to unemployment insurance as of July 31, 2021, and analyzed related federal
court documents. We also reviewed relevant DOL OIG and state audit reports. In addition, we
interviewed DOL oﬃcials about fraud risk management eﬀorts; we also interviewed DOL OIG
and NASWA oﬃcials. Further, we reviewed enterprise risk management guidance and written
responses from DOL oﬃcials.

Page 197 GAO-22-105051
Agency Comments
We provided a draft of this enclosure to DOL and OMB. DOL’s Employment and Training
Administration provided written comments, which are reproduced in appendix VII, as well as
technical comments, which we incorporated as appropriate. OMB did not provide comments.

In its comments, DOL neither agreed nor disagreed with our six recommendations. More
speciﬁcally, DOL neither agreed nor disagreed with our recommendation to designate a dedicated
entity with responsibility for managing the process of assessing fraud risks to the UI program,
consistent with leading practices as provided in our Fraud Risk Framework. DOL stated that the
department’s Chief Financial Oﬃcer and the Employment and Training Administration’s Assistant
Secretary are the designated senior executive oﬃcials responsible for risk assessment and
management in the UI program, which includes both fraud and nonfraud improper payments.
DOL stated that these oﬃcials make decisions on fraud risk management.

We acknowledge that the enterprise-wide risk management processes may incorporate the roles
and responsibilities of the dedicated antifraud entity speciﬁed in the Fraud Risk Framework.
However, as the framework discusses, it is a leading practice for this entity to have deﬁned
responsibilities, as well as the necessary authority across the program, to design and oversee
fraud risk management activities. Until DOL has a dedicated antifraud entity, in alignment with
leading practices, the department may not be well positioned to strategically manage UI fraud
risks. We revised the recommendation slightly to reﬂect the importance of documenting the
designation of the dedicated entity and its responsibilities.

In addition, DOL neither agreed nor disagreed with our ﬁve remaining recommendations to

• identify inherent fraud risks facing the unemployment insurance program,

• assess the likelihood and impact of inherent fraud risks in the unemployment insurance
program,

• determine fraud risk tolerance for the unemployment insurance program,

• examine the suitability of existing fraud controls in the unemployment insurance program and
prioritize residual fraud risks, and

• document the fraud risk proﬁle for the unemployment insurance program.

DOL stated that its enterprise-wide risk management process allows it to identify, evaluate,
and manage risks that could signiﬁcantly disrupt the successful achievement of its mission and
objectives. Additionally, DOL stated that it continues to identify fraud risks, assess the likelihood
and impact of these risks, and identify and support controls regarding risks to the UI program as
part of this enterprise-wide risk management process. However, DOL said that it will incorporate
the practices and approaches from the Fraud Risk Framework in its risk assessment activities as
it moves forward and will use these recommendations to build on its current risk assessment
and management process. DOL further stated that the dedicated entity envisioned in the ﬁrst
recommendation will take on these responsibilities.

Page 198 GAO-22-105051
GAO’s Ongoing Work
We continue to examine programmatic risks and challenges for the UI program as well as options
for program transformation.

Related GAO Products
Telecommunications: FCC Should Take Action to Better Manage Persistent Fraud Risks in the Schools and
Libraries Program. GAO-20-606. Washington, DC: Sept. 16, 2020.

A Framework for Managing Fraud Risks in Federal Programs. GAO-15-593SP. Washington, DC: July 28,
2015.

Contact information: Seto J. Bagdoyan, (202) 512-6722, bagdoyans@gao.gov

Page 199 GAO-22-105051
Emergency Rental Assistance
The Department of the Treasury has taken steps to enable Emergency Rental Assistance grantees
to spend their allocations more rapidly, but some grantees will be at risk of losing portions of their
allocation when Treasury recaptures and reallocates excess funds.

Entity involved: Department of the Treasury

Background
Congress appropriated a total of $46.55 billion to the Department of the Treasury (Treasury)
for the Emergency Rental Assistance (ERA) program, which Treasury disburses through two
programs to state, territorial, and local governments (grantees) to provide ﬁnancial assistance
to eligible landlords and renter households.334 Speciﬁcally, in December 2020, the Consolidated
Appropriations Act, 2021 authorized Treasury to disburse about $25 billion to grantees to
remain available until September 30, 2022 (referred to as ERA1 by Treasury).335 In March 2021,
the American Rescue Plan Act of 2021 authorized Treasury to disburse about $21.55 billion in
additional funding to grantees to remain available until September 30, 2025 (referred to as ERA2
by Treasury).336

Treasury allocated ERA funds to grantees based primarily on their population, as required by
the laws.337 Landlords, renter households, and utility providers can then apply to grantees for
assistance.338 As of August 31, 2021, Treasury had disbursed all of the ERA1 appropriation to
grantees (about $25 billion) and about $8.2 billion of the ERA2 appropriation to grantees that
applied for funding, according to data from Treasury.339 Beginning on September 30, 2021, the

334Grantees include states, the District of Columbia and U.S. territories (Puerto Rico, U.S. Virgin Islands, Guam,
Commonwealth of the Northern Mariana Islands, and American Samoa), as well as local governments with more
than 200,000 residents. Grantees for the ﬁrst round of ERA funding also include the Department of Hawaiian Home
Lands and Indian tribes or their tribally designated housing entities. Grantees must use the majority of their allocation
to provide assistance to eligible households for the payment of rent, utilities, and other housing-related expenses.
Grantees may also use a portion of their ERA allocation for administrative costs and housing stability services. In general,
eligible households must (1) have experienced a ﬁnancial hardship due to COVID-19 or qualiﬁed for unemployment
beneﬁts, (2) demonstrate a risk of housing instability or homelessness, and (3) have household income that does not
exceed 80 percent of their area median.
335Pub. L. No. 116-260, div. N, tit. V, § 501, 134 Stat. 1182, 2069-2078 (2020) (codiﬁed at 15 U.S.C. § 9058a). The American
Rescue Plan Act of 2021 extended the period of availability from December 30, 2021, to September 30, 2022. See Pub. L.
No. 117-2, tit. III, § 3201(h), 135 Stat. 4, 58. The Consolidated Appropriations Act, 2021 reserved $15 million of the ERA1
appropriation for administrative expenses of the Secretary of the Treasury.
336American Rescue Plan Act of 2021 § 3201(g), 15 U.S.C. § 9058c(g). The law reserved a total of $33 million of the ERA2
appropriation for administrative expenses of the Secretary of the Treasury and Oﬃce of Inspector General.
337Consolidated Appropriations Act, 2021 § 501(b), 15 U.S.C. § 9058a(b); American Rescue Plan Act of 2021 § 3201(b), 15
U.S.C. § 9058c(b).
338Grantees are required to make reasonable eﬀorts to obtain the cooperation of landlords and utility providers to
accept ERA1 payments before they can be made directly to renter households. For ERA2 funds, grantees can make
payments directly to renter households.
339As required by law, Treasury has made available 40 percent of the ERA2 allocations to grantees. American Rescue
Plan Act of 2021 § 3201(c), 15 U.S.C. 9058c(c). Treasury is disbursing additional ERA2 funds to grantees that have
substantially expended their ERA1 allocation and obligated at least 75 percent of their initial ERA2 allocation. According

Page 200 GAO-22-105051
Consolidated Appropriations Act, 2021 required Treasury to begin recapturing and reallocating
excess unobligated ERA1 funds.340

Overview of Key Issues
ERA grantees have spent about $7.7 billion of their ERA allocations and served about 1.5 million
unique households, based on Treasury data collected from grantees through August 2021.
Speciﬁcally, for ERA1, grantees spent about $7.5 billion of their allocations—about 30 percent
of the total ERA1 appropriation—including about $7.1 billion for assistance to households
through August 2021, as well as about $284 million for administrative expenses, $75 million for
housing stability services, and $76 million for tribal expenses through June 2021.341 For ERA2,
grantees spent about $223 million through August 2021—about 1 percent of the total ERA2
appropriation—including $207 million for assistance to households through August 2021, as
well as about $15 million for administrative expenses and $470,000 for housing stability services
through June 2021 (see table).342

to Treasury staﬀ we interviewed in July 2021, some local governments that were eligible for an ERA2 allocation did not
apply for the additional funds.
340Consolidated Appropriations Act, 2021 § 501(d), 15 U.S.C. 9058a(d). Treasury’s reallocation guidance states that the
agency will consider funds to be obligated that have been spent on ﬁnancial assistance and housing stability services,
needed to pay for assistance promised in a commitment letter, or when assistance has been approved through a
binding agreement but not yet disbursed. Separately, Treasury is required to reallocate ERA2 funds beginning on March
31, 2022. American Rescue Plan Act of 2021 § 3201(e), 15 U.S.C. 9058c(e).
341Assistance to households includes current and future rent and utility payments and arrears. Housing stability
services include those that enable eligible households to maintain or obtain housing, such as eviction prevention and
eviction diversion programs, mediation between landlords and tenants, and housing counseling. Treasury’s data do not
separately identify tribal expenses for assistance to households, administrative expenses, and housing stability services.
The reporting period for certain expenses varies based on whether Treasury collects them monthly or quarterly.
342Grantees generally transition to using ERA2 funds after exhausting their ERA1 allocations, which few have done
based on Treasury’s data. In addition, about 7 percent of ERA1 grantees and 12 percent of ERA2 grantees that were
required to submit compliance reports did not for August 2021 and are excluded from Treasury’s data.

Page 201 GAO-22-105051
Summary of Emergency Rental Assistance (ERA) Laws and Amounts Appropriated, Disbursed, and Spent (dollars
in billions) through Aug. 2021
Amount
appropriated to Amount disbursed Amount spent
Authorizing law Treasury to grantees by grantees

Consolidated Appropriations Act, 2021(ERA1) 25 24.985 7.506

American Rescue Plan Act of 2021 (ERA2) 21.55 8.248 0.223

Total 46.55 33.233 7.729

Source: GAO analysis of the Consolidated Appropriations Act, 2021; the American Rescue Plan Act of 2021; and Department of the Treasury (Treasury) data. | GAO-22-105051

Note: The amounts disbursed are through August 31, 2021. The amount spent includes expenditures for assistance to
households through August 2021, as well as expenses for administration and housing stability services through June 2021.
The Consolidated Appropriations Act, 2021 set aside $15 million for administrative expenses of the Secretary of the Treasury.
The American Rescue Plan Act of 2021 also set aside a total of $33 million for administrative expenses of the Secretary of the
Treasury and Inspector General. As required by law, Treasury has made available 40 percent of the ERA2 allocations to grantees
and is disbursing additional ERA2 funds to grantees that have substantially expended their ERA1 allocation and obligated at
least 75 percent of their initial ERA2 allocation.

Grantees have primarily provided ERA1 funds to the lowest-income renter households. According
to Treasury’s data through June 2021, grantees provided about 63 percent of ERA1 funds to renter
households with incomes at or below 30 percent of their area median (compared to 13 percent
to those with incomes from 51 to 80 percent of their area median).343 In addition, grantees have
provided similar assistance for current and future payments versus arrears. At the end of June
2021, about 622,000 renter households had received assistance for current or future payments
and 667,000 households had received assistance for arrears.

Although grantees were initially slow to spend ERA1 funds as they implemented and scaled up
rental assistance programs, many have accelerated their spending in recent months. In August
2021, grantees provided about $2.3 billion in assistance to 399,000 unique households—a sharp
increase from earlier reporting periods. For example, from January through March 2021, grantees
provided about $282 million in assistance to about 89,000 renter households (see ﬁgure). At the
same time, grantees have also spent more on each household served in recent months. In the ﬁrst
3 months of the program (January–March 2021), grantees provided each household served with
about $3,200 in assistance, on average. In comparison, grantees provided each household served
in August 2021 with about $5,700 in assistance, on average.

343Percentage of households served by income level only reﬂects those households for which grantees reported
income levels. Data on households served and households served by income level are collected from separate reports:
households served is collected from monthly reports from January through August 2021, and households served
by income level is collected from quarterly reports from January through June 2021. Some grantees may also have
served households for which they did not collect income levels. Treasury allows grantees to use fact-speciﬁc proxies to
determine income eligibility, such as the average income in an applicant’s geographic area. For such applicants, grantees
would not have collected speciﬁc income levels.

Page 202 GAO-22-105051
First-Round Emergency Rental Assistance (ERA1) Spending and Unique Renter Households Served, by Reporting
Period (Jan.–Aug. 2021)

Yet, some grantees continue to experience delays in delivering ERA1 funds and may not be
fully meeting the needs of landlords and lower-income renter households. Local grantees have
generally spent their ERA1 allocations more quickly than state and territorial governments.
Through August 2021, local grantees spent about 47 percent of their allocations, on average,
compared to about 23 percent among state grantees.344 Territorial governments spent an average
of 3 percent of their allocations. Further, about 13 of the 350 local grantees and two of the ﬁve
territorial grantees did not report spending any of their ERA1 allocation. In total, spending among
all grantees in each state and territory as a percentage of their total ERA1 allocations ranged from
0 percent to about 78 percent (see ﬁgure).345

344For the purposes of this enclosure, we categorized the Department of Hawaiian Home Lands as a local grantee.
Local grantees exclude grantees that redirected their allocations to county or state governments. Percentages reported
throughout this enclosure are the expenditure ratio—the total spending on assistance to households divided by
90 percent of the total ERA1 allocation (or the proportion of the ERA1 allocation grantees are required to spend on
assistance to households).
345Spending by states and territories includes spending by state and territorial governments, as well as their local
governments that received separate ERA1 allocations, if applicable.

Page 203 GAO-22-105051
Percentage of First-Round Emergency Rental Assistance (ERA1) Allocations Spent by Grantees in U.S. States and
Territories through Aug. 2021

Note: Percentages by U.S. state and territory include spending by state and territorial governments, as well as their local
governments that received separate allocations, if applicable. Percentages are the total spending on assistance to households
divided by 90 percent of the total ERA1 allocations (or the proportion of each ERA1 allocation grantees are required to spend on
assistance to households).

Grantees that continue to experience delays spending their ERA1 allocation may forfeit unused
funds when Treasury recaptures and reallocates excess funds. On September 30, 2021, Treasury
was required to begin recapturing excess ERA1 funds that grantees had not obligated and
reallocate them among those that had obligated at least 65 percent of their allocation.346 On
October 4, 2021, Treasury issued reallocation guidance for ERA1 funds, which clariﬁed that for the
ﬁrst assessment of excess funds,

• Grantees that had not obligated at least 65 percent of their allocation through September 30,
2021 must submit a program improvement plan by November 15, 2021 that describes how
they will accelerate their delivery of assistance;

• Grantees that had not achieved an expenditure ratio of at least 30 percent through September
30, 2021 will be determined to have excess funds equal to the diﬀerence between their
current expenditures on assistance and the amount needed to achieve an expenditure ratio of
30 percent; and

346Although the Consolidated Appropriations Act, 2021 § 501(d), 15 U.S.C. § 9058a(d) requires Treasury to begin
recapturing and reallocating excess ERA1 funds on September 30, 2021, Treasury oﬃcials said the act provides ﬂexibility
regarding when and how the agency must recapture and reallocate the excess funds among grantees.

Page 204 GAO-22-105051
• Grantees can initially limit or avoid recapture if Treasury approves their program improvement
plan, or it they can certify to obligating at least 65 percent or expending 30 percent of their
allocation by November 15, 2021.347

Every 2 months thereafter, Treasury will reassess grantee expenditures and redistribute
recaptured funds based on availability and conﬁrmed need.

Data on expenditures through August 2021 suggest that many grantees may need to develop
program improvement plans and some may be determined to initially have excess funds. Treasury
will make reallocation decisions based on expenditures and obligations from grantees; however,
Treasury has not yet collected data on obligations.348 Nonetheless, based on expenditures
through August 2021, about 74 percent of grantees have spent less than 65 percent of their ERA1
allocation and 42 percent have expenditure ratios below 30 percent. Further, expenditure data
suggest that few state and territorial grantees are on track to initially receive reallocated funds.
Through August 2021, three of the 51 state grantees and no territorial grantees (among the ﬁve)
have spent 65 percent or more of their ERA1 allocation. In comparison, many local grantees—101
of the 350 (about 29 percent)—have spent at least 65 percent of their ERA1 allocation.

Data on applications for assistance and rental payments suggest that signiﬁcant demand for ERA
funds remains unmet. Grantees continue to process a backlog of applications for assistance. For
example, according to its dashboard, the California COVID-19 Rent Relief Program has provided
assistance to about 24 percent of the nearly 204,000 complete applications it received through
September 21, 2021.349 Further, the nearly 421,000 renter households that received assistance
from ERA1 and ERA2 funds in August 2021 would be equal to about one-fourth (23 percent) of the
estimated number of lowest-income renter households that reported being very likely or likely
to be evicted, based on September 2021 survey data from the Census Bureau.350 It is critical that

347Department of the Treasury, Emergency Rental Assistance Under the Consolidated Appropriations Act, 2021
Reallocation Guidance (October 4, 2021). For grantees that submit an approved program improvement plan,
Treasury will assume a one-time addition to their expenditure ratio of 15 percent to help them meet the 30 percent
minimum expenditure ratio.
348Treasury required grantees to submit their ﬁrst detailed quarterly program and ﬁnancial reports in October 2021,
including information on ERA1 obligations and expenditures through September 30, 2021. However, these data,
including data on obligations by grantee, were not available at the time of our review. In comments on this enclosure,
White House oﬃcials said they believed data on obligations and expenditures through September 2021 would
demonstrate improvements in grantee performance, which would limit the number of grantees required to develop
program improvement plans and found to have excess funds and increase the number that are eligible to receive
reallocated funds.
349California Department of Housing and Community Development, California COVID-19 Rent Relief Program
Dashboard, accessed September 25, 2021, https://housing.ca.gov/covid_rr/dashboard.html. The dashboard does not
indicate the extent to which all of the households that completed applications are eligible for assistance through ERA
funds.
350Census Bureau, “Week 37 Household Pulse Survey: September 1 – September 13,” accessed September 25, 2021,
https://www.census.gov/data/tables/2021/demo/hhp/hhp37.html. This estimate had a relative margin of error of about
± 26 percent of the estimate at the 95 percent conﬁdence level. Renter households with the lowest incomes had annual
incomes of less than $25,000.

Page 205 GAO-22-105051
grantees deploy funds in a timely and eﬀective manner to help curtail avoidable evictions and limit
ﬁnancial burdens on landlords, especially given uncertainty about eviction moratoriums.351

Treasury has identiﬁed reasons why spending has lagged among some grantees. According to
Treasury, most grantees could not deploy funds immediately because they needed additional
time to enhance existing rental assistance programs or develop new programs in order to
accommodate the signiﬁcant increase in the scale of operations made possible through ERA1
funding. Speciﬁcally, grantees have had to develop new policies and procedures, hire additional
staﬀ, and develop electronic application and data collection systems. In addition, some grantees
have experienced overwhelming demand for ERA1 funds and lack suﬃcient staﬀ capacity to
address all applications in a timely manner. Further, the distribution of ERA1 funds from grantees
to landlords, renter households, and utility providers may have been delayed in some areas while
state and local legislatures deliberated how to administer the new funds.

Treasury has encouraged grantees to speed up their deployment of ERA1 funds by publishing
promising practices and issuing updated guidance. In June 2021, Treasury published a series of
promising practices based on engagement with grantees that are intended to help grantees speed
up program implementation and improve access to ERA funds, among other goals. For example,
the promising practices encourage grantees to simplify and automate application processes,
partner with community organizations and other grantees, and proactively engage landlords
and utility providers to inform them and their customers about the availability of ERA assistance.
In September 2021, Treasury also published guidelines and examples to help grantees develop
eﬀective ERA program websites and applications.

In addition, Treasury has updated its ERA grantee guidance (known as frequently asked questions)
several times to assist grantees in delivering ERA funds more eﬃciently—most recently in August
2021. Among other updates, Treasury has modiﬁed the guidance to

• encourage greater reliance on self-attestations and fact-speciﬁc proxies (such as average
income in an applicant’s geographic area) in lieu of more stringent documentation
requirements;352

• shorten the required outreach period for landlords and utility providers before grantees can
provide payments directly to renter households;

• allow grantees to make advance payments to landlords, utility providers, and nonproﬁt
organizations while renters complete application and documentation requirements;

351The 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. The agency later issued a new order that
became eﬀective on August 3, 2021, that prohibited evictions of covered renter households in areas with higher rates of
COVID-19 through October 3, 2021. However, on August 26, 2021, the Supreme Court issued an opinion that determined
that the agency did not have authority to issue the eviction order and blocked its enforcement. Separately, as of late
August 2021, 12 states continued to restrict residential evictions, according to the Mortgage Bankers Association. For
more information, see our March 2021 report on eviction moratoriums.
352Treasury has also published examples of self-attestation forms used by ERA grantees.

Page 206 GAO-22-105051
• expand the eligible uses of ERA funds to cover arrears at prior addresses and court costs
associated with an eviction; and

• provide a pathway for households who are no longer renting due to eviction to receive
assistance for relocation and other expenses.

Despite Treasury’s guidance, some grantees may be hesitant to modify their standard practices
for documentation, payments, and other aspects of their ERA programs out of concern that
noncompliance with the federal ERA laws and requirements could result in negative audit ﬁndings
and recoupment of funds. According to data collected and analyzed by the National Low Income
Housing Coalition, documentation and payment policies vary across ERA grantees. For example,
just over half of grantees (57 percent) allow applicants to self-attest that they meet certain
eligibility criteria, about 28 percent provide payments directly to tenants, and fewer than half (48
percent) provide assistance for housing expenses beyond rent and utility payments.353

At the same time, Treasury must balance its goal for grantees to deliver rental assistance quickly
with the competing need to safeguard the use of funds. Treasury has developed and implemented
reporting requirements for grantees and is beginning to collect detailed data on their use of
ERA funds. In August 2021, Treasury oﬃcials told us that an internal monitoring and oversight
group was preparing to review grantee-submitted data to identify potential compliance issues.
In addition, Treasury oﬃcials and staﬀ from Treasury’s Oﬃce of Inspector General told us that
they continued to work together to ﬁnalize plans to monitor and oversee grantees’ use of funds.
However, as of late September 2021, neither had ﬁnalized plans to do so.

Methodology
To conduct this work, we interviewed oﬃcials from Treasury and Treasury’s Oﬃce of Inspector
General on eﬀorts to implement and oversee the ERA program. We also analyzed Treasury’s
compliance report data from January through August 2021 to identify trends and reviewed ERA
program documentation published by Treasury on its website.354 In addition, we interviewed or
reviewed the recent publications of a selection of stakeholder groups that represent grantees,
landlords, and lower income households in order to identify potential successes and challenges
associated with the ERA program.355

353National Low Income Housing Coalition, Treasury Emergency Rental Assistance (ERA) Dashboard, accessed August 31,
2021, https://nlihc.org/era-dashboard.
354We assessed the reliability of these data by interviewing knowledgeable agency oﬃcials and reviewing technical
documentation. We found the data to be suﬃciently reliable for describing spending and other trends.
355These stakeholders included the National Council of State Housing Agencies, National Low Income Housing Coalition,
and National Multifamily Housing Council.

Page 207 GAO-22-105051
Agency Comments
We provided a draft of this enclosure to Treasury and the Oﬃce of Management and Budget
(OMB). Treasury provided technical comments, which we incorporated, as appropriate. OMB did
not provide comments on this enclosure.

GAO’s Ongoing Work
We will continue to monitor Treasury’s administration of the ERA program, including plans to
monitor and oversee grantees’ use of funds and reallocate excess funds.

Related GAO Product
COVID-19 Housing Protections: Moratoriums Have Helped Limit Evictions, but Further Outreach Is
Needed. GAO-21-370. Washington, D.C.: March 15, 2021.

Contact information: John Pendleton, (202) 512-8678, pendletonj@gao.gov

Page 208 GAO-22-105051
Advance Child Tax Credit and Economic Impact Payments
The Department of the Treasury and the Internal Revenue Service quickly issued millions
of advance child tax credit payments while also issuing the third round of economic impact
payments, but they could do more to reach some eligible families.

Entities involved: Internal Revenue Service, within the Department of the Treasury

Recommendations for Executive Action
The Secretary of the Treasury, in coordination with the Commissioner of Internal Revenue, should
estimate the number of individuals, including nonﬁlers, who are eligible for advance child tax
credit payments, measure the 2021 participation rate based on that estimate, and use that
estimate to develop targeted outreach and communications eﬀorts for the 2022 ﬁling season; the
participation rate could include individuals who opt in and out of the advance payments.

The Department of the Treasury neither agreed nor disagreed with this recommendation, stating
that while it supports the goal of the recommendation, it has not estimated the eligible population
for the advance child tax credit. Treasury also stated that it and IRS continue to undertake advance
child tax credit outreach, education, and media campaign eﬀorts.

We maintain that without a comprehensive estimate of eligibility and a participation rate, which
includes more nonﬁlers, Treasury and the Internal Revenue Service are missing an opportunity
to assess the eﬀectiveness of their outreach to nonﬁlers, who are likely experiencing poverty or
hardship and may be in need of the CTC payments.

Background
The American Rescue Plan Act of 2021 (ARPA) authorized the Internal Revenue Service (IRS)
and the Department of the Treasury to issue advance payments of the child tax credit (CTC), as
amended by ARPA.356 Additionally, IRS and Treasury continue to issue direct payments authorized
by ARPA, known as economic impact payments (EIP), to eligible individuals to help address
ﬁnancial stress due to the pandemic.357

ARPA made several temporary changes that expanded eligibility for CTC to additional qualiﬁed
individuals and increased the amount of the credit. These changes included temporarily
eliminating the earned income requirement for eligible individuals to receive the refundable
CTC.358 ARPA also increased the maximum amount of the CTC—raising it from $2,000 per any

356Pub. L. No. 117-2, § 9611, 135 Stat. 4, 144–150 (2021).
357 Pub. L. No. 117-2, § 9601, 135 Stat. 4, 138–144 (2021).
358The elimination of the earned income requirement applies to a taxpayer who has a principal place of abode in
the United States for more than half of the taxable year (2021) or who is a bona ﬁde resident of Puerto Rico for the
taxable year. According to IRS, these taxpayers will claim the refundable child tax credit on Form 1040 and for all other
taxpayers, the earned income requirement continues to apply to the additional child tax credit on Form 1040.

Page 209 GAO-22-105051
qualifying child to $3,600 for a young qualifying child (0 to 5 years old) and $3,000 for an older
qualifying child (6 to 17 years old) (see ﬁgure).359

Income Qualiﬁcations for the Expanded Child Tax Credit, Based on Tax Year 2021

a
During this ﬁrst phaseout, payment amounts are reduced to $2000 ($50 for each $1000 AGI exceeds the applicable
thresholds).
b
During this second phaseout, payment amounts are reduced to $0 ($50 for each $1000 AGI exceeds the applicable thresholds).
c
Other ﬁlers include single and married ﬁling separately.

As required by ARPA, IRS and Treasury are responsible for issuing half of the CTC through periodic
advance payments (advance CTC).360 IRS determines the amount of an individual’s advance
CTC payment by estimating the individuals’ annual advance amount, which is 50 percent of the
estimated amount of the CTC that the individual may claim on a 2021 tax return during the 2022
tax ﬁling season. This annual advance amount for 2021 is generally based on an individual’s
processed 2020 tax return or on the individual’s 2019 tax return if the IRS has not processed the
2020 tax return.

Eligible individuals automatically receive monthly advance CTC payments from July 2021 through
December 2021, unless they unenroll from the payments. Eligible individuals may claim the
remaining portion of their CTC when they ﬁle their 2021 tax return in 2022.

IRS reported that as of September 25, 2021, it had disbursed more than 106 million advance
payments totaling over $45.5 billion—an average payment of $426 (see ﬁgure).

359In 2017, the maximum CTC amount was temporarily raised from $1,000 to $2,000 for tax years 2018 through 2025.
26 U.S.C. § 24(h) (2).
360Pub. L. No. 117-2, § 9611(b), 135 Stat. at 146–150, codiﬁed at 26 U.S.C. 7527A.

Page 210 GAO-22-105051
Advance Child Tax Credit Disbursals, by Month, as of Sept. 25, 2021

a
Number of payments is the volume of advance child tax credit payments IRS issued electronically or by paper check.
b
Number of children is the total volume of qualifying children who received advance payments.
c
Number of unenrollments is the cumulative volume of taxpayers who opted to unenroll from receiving advance payments.

In addition to issuing advance CTC payments, Treasury and IRS issued a third round of economic
impact payments (EIP 3) to most eligible individuals.361 IRS reported that as of September 24,
2021, it had processed more than 174 million EIP 3 payments totaling over $405 billion.362

ARPA directed IRS to make supplemental payments to individuals who received EIP 3 before
their 2020 tax returns were processed but who are eligible for an additional payment based on
their recently processed 2020 tax returns.363 IRS reported that as of September 30, 2021, it had
processed more than 10 million supplemental EIPs totaling over $21 billion.

Overview of Key Issues
IRS is conducting outreach to increase awareness of the advance CTC. IRS is conducting
several outreach eﬀorts to increase the public’s awareness of advance CTC payments.

361The CARES Act (in March 2020) and the Consolidated Appropriations Act, 2021 (in December 2020) respectively
authorized Treasury and IRS to issue EIP 1 and EIP 2 as direct payments to help individuals address ﬁnancial stress due
to the pandemic. Pub. L. No. 116-136, § 2201, 134 Stat. 281, 335–340 (2020); Pub. L. No. 116-260, § 272, 134 Stat. 1182,
1965–1976 (2020).
362IRS will issue EIP 3 through December 31, 2021. Pub. L. No. 117-2, § 9601(a), 135 Stat. at 140–141, codiﬁed at 26 U.S.C.
§ 6428B(g)(5).
363Individuals may be eligible for an additional supplemental payment if their adjusted gross income was lower, or if
they claimed more qualifying dependents, on their 2020 tax return than on their 2019 tax return.

Page 211 GAO-22-105051
• IRS continues to coordinate with community organizations to raise awareness of the advance
CTC payments.

• IRS provides multilingual outreach resources to improve communications to those with limited
English proﬁciency and is urging nonﬁlers to use the CTC Non-ﬁler Sign-Up Tool before it closes
on October 15, 2021.364

• IRS is planning to include advance CTC messaging in its Get Ready campaign for the 2022 tax
ﬁling season. IRS oﬃcials said the annual campaign typically begins in November.

However, IRS and Treasury have not developed a comprehensive estimate of individuals who are
potentially eligible for advance CTC payments and have not set a participation goal. IRS oﬃcials
said the agency developed an updated estimate in April 2021 of 36.5 million ﬁlers. The estimate
is based on processed tax year 2020 returns and includes individuals with dependent children
who ﬁled taxes for the ﬁrst time to receive EIP 1. Yet this estimate is incomplete because it misses
additional eligible nonﬁlers and their dependent children.

IRS oﬃcials said that they cannot estimate the number of eligible nonﬁlers and their dependent
children because these individuals have not ﬁled taxes. However, according to the oﬃcials,
they plan to meet with Census Bureau oﬃcials to assess what data are needed to produce a
CTC participation rate and to discuss next steps. IRS oﬃcials said it currently has a multiyear
interagency agreement with the Census Bureau through March 2024 to obtain data to support
its calculation of an Earned Income Tax Credit participation rate. Federal internal control
standards state that management should obtain relevant data from reliable sources and use that
information to achieve its objectives. This standard is reﬂected in Treasury’s strategic plan, which
sets a goal of improving analytics to accomplish objectives.

With an estimate of all individuals who are eligible for the advance CTC, including nonﬁlers,
Treasury and IRS could measure the tax credit’s participation rate. Participation rates for other
refundable tax credits and federal beneﬁt programs have helped policy makers better understand
whether eligible recipients are aware of the credits’ and programs’ existence and accessibility.365
For example, the Earned Income Tax Credit participation rate—over 75 percent, according to
IRS—is at the high end of the range of participation rates for antipoverty programs. In contrast, we
previously reported that the Temporary Assistance for Needy Families participation rate was about
34 percent.366

An advance CTC eligibility estimate and participation rate, including individuals who have opted in
and out of the advance CTC payments, would provide greater clarity about which populations may

364For individuals who requested an extension, October 15, 2021, is the last day to ﬁle a 2020 tax return.
365Tax credits targeted to individuals and businesses are generally provided to promote certain social and economic
objectives. These tax credits can be nonrefundable, where the amount claimed is limited to the taxpayer’s tax liability, or
refundable, where the amount claimed is payable to the taxpayer as a refund to the extent that the tax credit exceeds
the taxpayer’s tax liability.
366The Earned Income Tax Credit is a refundable tax credit intended to encourage work by oﬀsetting payroll taxes for
low-income taxpayers. Temporary Assistance for Needy Families is a block grant to states that emphasizes work and
time-limited cash assistance.

Page 212 GAO-22-105051
be at risk of not receiving the payment. These populations would beneﬁt from targeted outreach
and communications to learn more about the payments and how to claim the CTC during the 2022
ﬁling season. For example, oﬃcials from one IRS partner organization we spoke with said IRS could
do more to enhance its coordination with federal and state agencies to reach more nonﬁlers.367
Moreover, this information could inform IRS’s administration of other refundable tax credits as
well as any future changes to the CTC that Congress is considering.

IRS conducted in-person events to help eligible taxpayers claim the advance CTC. In June
and July 2021, IRS conducted in-person events in more than 27 cities across the country to help
eligible individuals prepare and ﬁle their 2020 tax returns in order to receive the advance CTC. IRS
conducted these events at Taxpayer Assistance Centers (TAC) and community outreach centers.
Oﬃcials from the IRS Taxpayer Advocate Service and preparers from Low Income Tax Clinics who
attended some of the events said the assistance helped people verify their identity to register
for the advance CTC. They noted that two groups of individuals expressed confusion about their
advance CTC eligibility—separated or divorced parents, who expressed confusion regarding which
parent should receive the advance CTC, and individuals with mixed immigration status.

IRS assists individuals with the advance CTC. IRS and Treasury have provided several online
tools to help individuals manage their CTC payments: the Advance Child Tax Credit Eligibility
Assistant, the CTC Update Portal (CTCUP), and options for nonﬁlers to ﬁle simpliﬁed returns to
register for payments. IRS has also established a dedicated phone line for CTC assistance.

Advance Child Tax Credit Eligibility Assistant. Individuals can use the Advance Child Tax Credit
Eligibility Assistant to check their eligibility for the advance CTC. The tool provides generic
questions and responses based upon the advance CTC qualiﬁcations, such as asking the user for
the number of children claimed on the user’s 2019 or 2020 tax return.

CTCUP. This online portal allows qualiﬁed individuals to check their eligibility for the advance
payments, opt out of receiving the payments, update their bank account information, and
change their mailing address. According to IRS data, as of September 20, 2021, around 1.9 million
individuals had accessed the portal to opt out of payments, around 1.6 million individuals had
made changes to their bank account information, and around 216,000 individuals had made
changes to their mailing address.

IRS oﬃcials said IRS is planning further updates to CTCUP that will allow users to change personal
information, such as marital status, income, and number of children. IRS also plans new portal
features, such as a summary of an individual’s 2021 advance CTC payments, and intends to make
CTCUP available in Spanish.

To access the portal, individuals must log in using an established IRS online account or establish
a new account with ID.me, a company that provides identity prooﬁng and authentication. Users
must upload documentary evidence, such as a driver’s license, to verify their identity. Once ID.me

367IRS partners with nationwide and local organizations by providing outreach material, training, and tax preparation
products for taxpayer assistance and education. IRS partner organizations are meant to serve low- to moderate-income
populations, older Americans, students, military service members, people with disabilities, and other populations. IRS
refers to these organizations as outreach partners.

Page 213 GAO-22-105051
is able to verify the information, the user can access CTCUP to make changes. ID.me provides
support to users who experience challenges verifying their identity.

According to IRS and ID.Me oﬃcials, they are working together to monitor CTCUP fraud and ease
of use. ID.me provides IRS with weekly reports on the percentages of users who successfully
authenticated their identity and users who were unsuccessful after exhausting ID.Me’s customer
service options. According to ID.Me oﬃcials, they review their data with IRS to try to better
understand users’ behavior, including challenges that users encounter in completing the
authentication process. As we have previously reported, designing authentication programs
involves a balancing act—IRS needs to prevent fraudsters from using stolen taxpayer information
to authenticate, but it must balance that against the burden of authentication on legitimate
taxpayers. If IRS makes the authentication process too stringent, legitimate taxpayers may not be
able to successfully authenticate.

Nonﬁler tools. Individuals who do not normally ﬁle a tax return had two online options to ﬁle a
simpliﬁed return in order to receive the advance CTC, EIP 3, and recovery rebate credit.368 The Free
File Alliance, a consortium of tax preparation companies, developed the CTC Non-ﬁler Sign-Up Tool
for IRS in June 2021 in time for the ﬁrst round of payments in July. A Treasury oﬃcial said that the
CTC Non-ﬁler Sign-Up Tool did not fully meet the needs of intended users. Speciﬁcally, the CTC
Non-ﬁler Sign-Up Tool could be diﬃcult to access and navigate, particularly on a mobile device,
which may be the only device available to some individuals to access the internet. Also, it was not
available in Spanish.

On August 13, 2021, Treasury and the White House announced that Code for America, a civic
technology nonproﬁt organization, would create a new tool to assist individuals who do not
regularly ﬁle taxes to claim the advance CTC. On September 1, 2021, Code for America launched
GetCTC, which is mobile friendly and available in English and Spanish. Individuals can also use
GetCTC to receive EIP 3 and claim the recovery rebate credit.

Individuals had the option to use either the CTC Non-ﬁler Sign-Up Tool or GetCTC. According to
IRS oﬃcials, the CTC Non-ﬁler Sign-Up Tool will be taken oﬄine on October 15, 2021. However,
individuals can still use GetCTC until November 15, 2021. Code for America oﬃcials said GetCTC
will be available until then to ensure that qualiﬁed individuals receive the December 2021 advance
CTC payment.

Dedicated CTC phone line. Anticipating high demand for customer support related to the advance
CTC payments, IRS established a dedicated phone line for CTC assistance (1-800-908-4184) and
increased the number of IRS customer service representatives. As of September 30, 2021, IRS
had received 1.8 million phone calls through the hotline since it went live in June. According to
IRS, 80,000 individuals spoke to an IRS customer service representative. IRS oﬃcials said callers
asked questions regarding accessing CTCUP, missing payments, payment amounts, eligibility
requirements, and changing the number of dependent children.

368Eligible individuals can claim the recovery rebate credit on their tax year 2020 income tax return if they did not
receive the ﬁrst or second EIP. According to IRS data, as of September 24, 2021, over 3.9 million 2020 returns had
claimed only a recovery rebate credit. These individuals would not have ﬁled a tax return except to claim a recovery
rebate credit because their income is below the minimum required tax ﬁling threshold or all of their income is derived
from federal beneﬁts.

Page 214 GAO-22-105051
IRS is taking steps to prevent advance CTC identify theft, fraud, and overpayments.

Steps to prevent identity theft and fraud. IRS oﬃcials said that, in addition to authenticating the
identity of individuals who use CTCUP and CTC Non-ﬁler Sign-Up Tool, they revised fraud ﬁlters
that were used on the original EIP Non-ﬁler Tool.369 Additionally, IRS has 15 fraud indicators
targeting high-risk behaviors including questionable bank account behavior. IRS is also sharing all
bank account changes in CTCUP with the Bureau of the Fiscal Service to ensure the bank account
is valid. ID.me also provides IRS reports on fraudulent attempts to access CTCUP.

Steps to prevent overpayments. Individuals who receive advance CTC payments exceeding the
amount of CTC they can properly claim on their 2021 tax return may need to repay some or all
of the excess payment.370 Some lower-income taxpayers will qualify for repayment protection.
Individuals cannot qualify for full repayment protection—and will need to repay any excess
amount—if their modiﬁed adjusted gross income for 2021 is above an established amount.371 IRS
continues to work on guidance and related communications to address taxpayer concerns about
repayment protection. In addition, in preparation for the 2022 ﬁling season, IRS plans to send a
letter to payment recipients in December 2021 after it issues the last advance payment that will
provide the total amount of advance CTC payments received.

IRS oﬃcials said the agency is still determining how it will monitor and report on advance CTC
overpayments. IRS reports an annual improper payment rate, including overpayments, for
the Additional Child Tax Credit.372 The Additional Child Tax Credit has provided tax beneﬁts to
taxpayers who are raising children. The annual improper payment rate for the Additional Child
Tax Credit has been based on data collected through multiyear compliance studies.373 According
to IRS oﬃcials, IRS has not yet determined whether it will include the advance CTC payments with
the Additional Child Tax Credit improper payment estimate or will report a separate estimate for
the advanced payments. IRS oﬃcials said they are discussing options with Treasury and Oﬃce of
Management and Budget oﬃcials.

EIP disbursements for dependents. In June 2021, we tested 79 transactions for the second
round of the economic impact payment (EIP 2) as well as 217 EIP 3 transactions that IRS processed

369The original Non-ﬁler Tool was operational from April 2020 to November 2020 to assist individuals receiving their ﬁrst
EIP. IRS updated the NFT to help eligible individuals receive the advance CTC.
370Pub. L. No. 117-2, § 9611 (b)(2), 135 Stat. 4 at 148–149, codiﬁed at 26 U.S.C. § 24(j)(2)(B).
371Taxpayers will not qualify for any repayment protection if their modiﬁed adjusted gross income is at or above certain
amounts based on the ﬁling status on their 2021 tax return. These amounts are $120,000 for those who are married
and ﬁling a joint return or are ﬁling as a qualifying widow or widower; $100,000 for those who are ﬁling as a head of
household; and $80,000 for single ﬁlers or for those who are married and ﬁling a separate return. Pub. L. No. 117-2, §
9611(b)(2), 135 Stat. at 148–149.
372Improper payments include both over- and underpayments.
373 In ﬁscal year 2020, IRS estimated that 12 percent, or $4.5 billion, of the total Additional Child Tax Credit payments
of $39.1 billion were improper. The most recent data on the Additional Child Tax Credit improper payment rate are
from a National Research Program study of 2017 tax returns ﬁled predominantly in calendar year 2018. The National
Research Program collects data on reporting, payment, and ﬁling compliance that supports strategic decisions about the
placement and types of resources necessary to eﬀectively address taxpayer needs.

Page 215 GAO-22-105051
from December 31, 2020, through May 27, 2021.374 As a result of our testing, we found instances
in which IRS disbursed EIPs to individuals for ineligible dependents, such as dependents with
invalid Social Security numbers, in four of the EIP 2 transactions and nine of the EIP 3 transactions.

IRS limited the processing of EIP 3 dependent credits to a maximum of 15 dependents to help
prevent the disbursement of invalid EIPs.375 IRS set the 15-dependents limitation on the basis of its
experience with disbursing the ﬁrst round of EIP. Speciﬁcally, for processing EIP 3 transactions, IRS
stated that 15 dependents was a reasonable threshold for identifying errors related to dependent
counts without aﬀecting a signiﬁcant number of transactions.

In March 2021, IRS developed program requirements to notify speciﬁc Wage and Investment (W&I)
staﬀ when the 15-dependent limitation was applied, so that the staﬀ could review the accounts
aﬀected and determine whether the taxpayers had additional dependents who were eligible for
the EIP 3 dependent credit. However, in discussions with W&I staﬀ, we learned that they had not
received any notiﬁcations when the 15-dependents limitation was applied.

When testing EIP 3 transactions, we found that IRS had applied the 15-dependents limitation
to 27 transactions but that the aﬀected accounts were not reviewed by W&I staﬀ. As a result,
the taxpayers had not received their additional payments as of June 2021. To address this issue,
IRS stated that it plans to perform a one-time review of all accounts that were subject to the 15-
dependents limitation and process any additional EIP 3 for taxpayers with more than 15 eligible
dependents before the end of calendar year 2021.

Methodology
To review how Treasury and IRS issued and administered advance child tax credits and economic
impact payments, we examined Treasury and IRS data as of September 30, 2021, as well as federal
laws and agency guidance. We interviewed Treasury and IRS oﬃcials to determine whether the
data were suﬃciently reliable to describe the number and amount of payments disbursed, and
we determined that the data were suﬃciently reliable for this purpose. We also interviewed IRS
oﬃcials at four Taxpayer Assistance Centers (TAC)—in Atlanta, GA; Houston, TX; Milwaukee, WI;
and St. Louis, MO—that sponsored in-person events in June and July 2021 to help eligible families
prepare and ﬁle tax returns to receive the advance CTC. We selected the four TACs to represent
both high and low attendance at these events.

374We performed testing of these 79 EIP 2 transactions and 217 EIP 3 transactions in connection with our audit of
IRS’s ﬁscal year 2021 ﬁnancial statements. The EIPs we tested consisted of two monetary unit samples of 61 EIP 2
transactions and 166 EIP 3 transactions. Additionally, we tested all transactions for married taxpayers ﬁling jointly
who claimed at least 15 dependents, which consisted of 18 EIP 2 transactions of $10,200 or more each and 51 EIP 3
transactions of $23,800 each.
375To determine EIP 3 disbursements, IRS used 2020 tax return information, 2019 tax return information (if the 2020
tax return was not available), or information for those who receive certain federal beneﬁts. For married taxpayers
ﬁling jointly who claimed more than 15 dependents on the tax return, IRS limited the EIP 3 disbursement to $23,800
comprising (1) an individual credit of $2,800 for married taxpayers ﬁling jointly, or $1,400 multiplied by two, and (2) a
dependent credit of $21,000 for 15 dependents, or $1,400 multiplied by 15.

Page 216 GAO-22-105051
We asked representatives from 18 selected IRS outreach partners to provide us with their
perspectives on IRS’s outreach coordination with their organizations. We selected eight partners
on the basis of their national outreach to underserved populations such as low-income families,
veterans, and seniors; we also selected 10 partners on the basis of their local outreach to
communities in zip codes with high numbers of potential nonﬁlers (according to IRS data). This
sample is not representative; however, these partners provided illustrative examples of how
organizations worked with IRS to reach traditionally underserved communities and of aspects of
the IRS communications plan that worked well. Their responses also highlighted potential areas
for improvement. In addition, we interviewed representatives from Free File Alliance, Code for
America, and ID.me.

To test IRS’s controls over the processing of EIP 2 and EIP 3 transactions and whether IRS
conformed with the EIP provisions of the Consolidated Appropriations Act, 2021 and ARPA,
we statistically selected two monetary unit samples.376 The ﬁrst sample consisted of 61 EIP 2
transactions, totaling $79,560, that were processed during the period from December 31, 2020,
through January 28, 2021.377 The second sample consisted of 166 EIP 3 transactions, totaling
$536,717, that were processed during the period from March 11, 2021, through May 27, 2021.378
In addition, during our review of the EIP 2 and EIP 3 populations, we identiﬁed transactions for
married taxpayers ﬁling jointly who claimed at least 15 dependents. These consisted of 18 EIP 2
transactions of $10,200 or more, totaling approximately $310,200, and 51 EIP 3 transactions of
$23,800, totaling approximately $1.2 million. We selected all 69 transactions for testing.

Our tests consisted of reviewing IRS tax module records and related individual tax returns to
validate the accuracy of IRS’s records, which IRS used to determine eligibility, calculate the EIPs,
and determine whether any duplicate EIPs were issued. We also reviewed payment conﬁrmations
from the Bureau of the Fiscal Service to determine whether the EIPs were issued for the correct
amounts and to the correct individuals and to conﬁrm that no oﬀsets were applied against the
EIPs.

Agency Comments
We provided a draft of this enclosure to Treasury, IRS, and the Oﬃce of Management and Budget.
Treasury provided written comments, which are reproduced in appendix IX, and technical
comments, which we incorporated as appropriate. IRS also provided technical comments,
which we incorporated as appropriate. The Oﬃce of Management and Budget did not provide
comments.

Treasury neither agreed nor disagreed with our recommendation. Treasury stated that while it
supports the goal of our recommendation, it has not estimated the eligible population for the
advance child tax credit (CTC). Treasury said it has not performed this action because there are
several eligibility requirements for the CTC that are not known to IRS until the taxpayer ﬁles an

376Pub. L. No. 116-260, § 272, 134 Stat. 1182, 1965–1976 (2020) and Pub. L. No. 117-2, § 9601, 135 Stat. 4, 138–144
(2021), respectively.
377We selected the EIP 2 sample from the population of approximately 152 million transactions, totaling $148 billion.
378We selected the EIP 3 sample from the population of approximately 169 million transactions, totaling $397 billion.

Page 217 GAO-22-105051
income tax return to claim the CTC. However, we note that IRS established an estimate of eligible
taxpayers based on tax year 2020 returns. Nevertheless, this estimate is incomplete because
it misses eligible nonﬁlers and their dependent children. An advance CTC eligibility estimate
and participation rate, including individuals who opt in and out of the payments, would provide
Treasury and IRS greater insight regarding populations, particularly nonﬁlers, that may be at risk of
not receive the payments.

Treasury also stated in its comments that, with IRS, it continues to undertake advance CTC
outreach, education, and media campaign eﬀorts. Treasury stated that in June and July 2021, IRS
sent two rounds of letters to inform more than 30 million families who previously ﬁled taxes that
they may be eligible to receive advance CTC payments. At the same time, IRS publicized the CTC
Non-Filer Tool at numerous outreach events. Treasury also published a ﬁle containing zip code
data that show the number of children who may be eligible for the advance CTC but who had not
been claimed on a recent tax return. Treasury stated that this information is designed to help local
governments and nonproﬁts target their outreach work so that this crucial relief gets to all eligible
families. Last, Treasury stated that it and IRS have developed partnerships within and beyond the
tax community to help eligible individuals access their payments.

Without an estimated eligibility rate that includes nonﬁlers and a participation goal, Treasury and
IRS are missing an opportunity to use data to determine whether their outreach strategy, which
will extend throughout the 2022 ﬁling season, is eﬀectively reaching those who are outside the tax
system. These individuals are more likely to be experiencing poverty or hardship and may be more
in need of the CTC payments.

GAO’s Ongoing Work
We will continue to monitor IRS and Treasury's eﬀorts to analyze data that could potentially
improve communication and outreach to nonﬁlers. We will also continue to monitor IRS and
Treasury’s progress to ensure eligible individuals receive the advance payments of the CTC and EIP
3.

GAO’s Prior Recommendations
The table below presents our recommendations on economic impact payments from prior
bimonthly and quarterly CARES Act reports.

Page 218 GAO-22-105051
GAO’s Prior Recommendations Related to Economic Impact Payments
Recommendation Status

The Secretary of the Treasury, in coordination with the Open—not addressed. The Department of the
Commissioner of Internal Revenue, should release interim ﬁndings Treasury neither agreed nor disagreed with this
on the eﬀectiveness of the notices it sent in September 2020 recommendation and stated that it shares the
to nonﬁlers who are potentially eligible for economic impact underlying goal of reaching as many nonﬁlers as
payments; incorporate that analysis into Internal Revenue Service possible to encourage them to claim economic
(IRS) outreach eﬀorts as appropriate; and then, if necessary, impact payments (EIP) online. However, Treasury
release an update based on new analysis after the 2021 ﬁling does not plan to release any interim ﬁndings until
season (July 2021 report). it completes its analysis. According to Treasury
oﬃcials, additional time will be needed to process
and analyze data as the 2021 ﬁling season was
extended to October 15, 2021. Treasury oﬃcials also
said outreach eﬀorts for the advance child tax credit
(advance CTC) included reminders that individuals
who had not ﬁled a 2020 federal income tax return
could be eligible to receive EIP and recovery rebate
credits (RRC).

The Commissioner of Internal Revenue should periodically review Closed—addressed. IRS took steps to implement
control activities for issuing direct payments to individuals to our recommendations, such as updating control
determine that the activities are designed and implemented procedures for issuing direct payments to
appropriately as IRS disburses a third round of economic impact individuals. Additionally, individuals had the
payments (EIP 3) and prepares for advance payments on the child opportunity to update their bank account
tax credit. These control activities should include appropriate information during the 2021 ﬁling season, which ran
testing procedures, quality assurance reviews, and processes that from February 12 through May 17, 2021.
ensure payments distributed by tax partners reach the intended IRS oﬃcials said that the updated procedures
recipients (March 2021 report). resulted in the transmission of a small number of
EIP 3 payments to incorrect bank accounts. The
oﬃcials said they anticipated the same for July 2021
advance CTC payments.
The number of direct payments that were rejected
was over 5.3 million for the second round of EIP
and close to 2.5 million for EIP 3. Additionally, over
500,000 direct payments were rejected for the July
2021 advance CTC payments.

The Secretary of the Treasury, in coordination with the Open—partially addressed. Treasury and IRS
Commissioner of Internal Revenue, should begin tracking and agreed with this recommendation. According to
publicly reporting the number of individuals who were mailed EIP Treasury oﬃcials, Treasury began analyzing data
notiﬁcation letter and subsequently ﬁled for and received an EIP in January 2021 on those individuals who received
and should use that information to inform ongoing outreach and a notice and subsequently ﬁled for, and received,
communications eﬀorts (November 2020 report). a ﬁrst-round EIP (EIP 1). According to Treasury
oﬃcials, additional time will be needed to process
and analyze data, because the 2021 ﬁling season
had been extended to October 15, 2021. Treasury
oﬃcials also said that outreach eﬀorts for the
advance CTC included reminders to individuals who
had not ﬁled a 2020 federal income tax return that
they could be eligible to receive EIPs and RRC.

The Secretary of the Treasury, in coordination with the Open—partially addressed. Treasury and
Commissioner of Internal Revenue, should make estimates of IRS neither agreed nor disagreed with this
eligible recipients who have yet to ﬁle for an EIP, and other relevant recommendation, but they took some actions that

Page 219 GAO-22-105051
Recommendation Status
information, available to outreach partners to raise awareness were consistent with it. For example, in September
about how and when to ﬁle for EIPs (September 2020 report). 2020, the agencies used tax return information to
identify nearly 9 million individuals who had not
received an EIP 1 and then notiﬁed these individuals
that they may be eligible for a payment. The letters
also provided instructions for requesting a payment.
In addition, IRS publicly released detailed zip code
data from the notices to help community outreach
partners with their own outreach eﬀorts.
Treasury oﬃcials said that additional time would
be needed to process and analyze the data to
determine who did or did not claim an EIP or RRC,
because the 2021 ﬁling season had been extended
to October 15, 2021. Oﬃcials also said that a
challenge to conduct this analysis is the information
tax returns used in developing the September
2020 notices do not contain enough information
to determine eligibility, limiting Treasury’s ability to
make a determination.
Treasury oﬃcials said that because of resource
constraints, Treasury and IRS have been focusing on
delivering the advance CTC. In June 2021, Treasury
published a ﬁle containing, by zip code, the number
of children who may be eligible to be claimed for the
advance CTC but who had not been claimed on a
recent tax return.

The Secretary of the Treasury, in coordination with the Open—partially addressed. Treasury and
Commissioner of Internal Revenue, should update and reﬁne the IRS neither agreed nor disagreed with this
estimate of eligible recipients who have yet to ﬁle for an EIP to recommendation, but they took some actions that
help target outreach and communications eﬀorts (September 2020 were consistent with it. For example, in January
report). 2021, Treasury revised its estimate of eligible
recipients who had not yet ﬁled for an EIP 1 to
8 million. According to Treasury oﬃcials, this
estimate is based on the 9 million notices IRS sent
in September 2020. Treasury oﬃcials stated that
some of the 9 million recipients likely have since
claimed the EIP, but Treasury did not provide data
supporting this claim.
Treasury oﬃcials said that additional time would
be needed to process and analyze the data to
determine who did or did not claim an EIP or RRC,
because the 2021 ﬁling season was extended to
October 15, 2021. Oﬃcials also said that a challenge
to conducting this analysis is that the information
tax returns used in developing the September
2020 notices do not contain enough information
to determine eligibility, limiting Treasury’s ability to
make these determinations.
Treasury oﬃcials said that because of resource
constraints, Treasury and IRS have been focusing on
delivering the advance CTC. In June 2021, Treasury
published a ﬁle containing, by zip code, the number
of children who may be eligible to be claimed for the

Page 220 GAO-22-105051
Recommendation Status
advance CTC but who had not been claimed on a
recent tax return.

The Commissioner of Internal Revenue should consider cost- Closed—addressed. Treasury and IRS took steps
eﬀective options for notifying ineligible recipients on how to return to implement our recommendation, such as
payments (June 2020 report). requesting on the IRS website that individuals
voluntarily mail the appropriate EIP amount sent
to decedents back to IRS, for both electronic and
paper check payments. Treasury has also held and
canceled payments made to decedents, along with
payments that have been returned. As of April 30,
2021, around 57 percent (just over $704 million) of
the $1.2 billion in EIP 1 sent to deceased individuals
had been recovered.
As of March 2021, Treasury and IRS had not taken
any further action to recoup unreturned payments
to decedents. IRS oﬃcials determined that further
actions, such as initiating erroneous refund cases
against decedents’ estates that had not returned
payments, could be burdensome to taxpayers,
the federal court system, and IRS. As a result, IRS
oﬃcials concluded that doing so was not prudent at
that time.

Source: GAO. | GAO-22-105051

Related GAO Products
Identity Theft: IRS Needs to Strengthen Taxpayer Authentication Eﬀorts. GAO-18-418. Washington, D.C.:
June 22, 2018.

Refundable Tax Credits: Comprehensive Compliance Strategy and Expanded Use of Data Could
Strengthen IRS's Eﬀorts to Address Noncompliance. GAO-16-475. Washington, D.C.: May 27, 2016.

Standards for Internal Control in the Federal Government. GAO-14-704G. Washington, D.C.:
September 10, 2014.

Contact information: James R. McTigue, Jr., (202) 512-6806, mctiguej@gao.gov

Page 221 GAO-22-105051
Agricultural Debt Relief
The U.S. Department of Agriculture has identiﬁed about 21,564 socially disadvantaged agricultural
producers who qualify for debt relief payments under the American Rescue Plan Act of 2021; as of
September 2021, less than $1.2 million of the $4 billion initially set aside for the program had been
expended pending the outcome of a legal challenge to the program.

Entities involved: Farm Service Agency, within the U.S. Department of Agriculture

Background
The U.S. Department of Agriculture (USDA) provides billions of dollars through loans and other
farm programs to support agricultural producers (that is, farmers and ranchers). For example,
according to USDA’s FY 2021 Budget Summary, in ﬁscal year 2020, the department provided
about $8.4 billion in farm loans. In calendar year 2020, federal direct payments and federal crop
insurance net payments to agricultural producers were projected to be about $52 billion, as we
reported in March 2021.379 Agricultural producers often require loans to buy agricultural real
estate, make capital improvements, and purchase supplies and equipment. Several types of
lenders provide credit to agricultural producers, including commercial banks and USDA’s Farm
Service Agency (FSA). FSA focuses its lending on agricultural producers who are generally unable to
obtain credit elsewhere.

However, as we noted in our July 2019 report on agricultural lending, some demographic groups
are considered underserved by the credit market or have alleged discrimination by USDA and
others in providing agricultural loans. USDA has, through public statements, acknowledged the
department’s history of systemic discrimination in its policies and programs.

The American Rescue Plan Act of 2021 (ARPA) directs the Secretary of Agriculture to provide
payments of up to 120 percent of the outstanding direct and guaranteed farm loan balances
for each socially disadvantaged producer to pay oﬀ the balance of the producer’s loans.380
Direct loans are funded and serviced by FSA as the lender. Guaranteed loans are made by USDA-
approved traditional lenders, such as banks and credit unions, with the backing of FSA.

Under ARPA, socially disadvantaged agricultural producers are those who are members of a
socially disadvantaged group whose members have been subjected to racial or ethnic prejudice.381
According to USDA, under the ARPA debt relief program, payments may represent 120 percent of a
loan balance (as of January 1, 2021) with 20 percent intended to cover tax liabilities and other fees.
FSA is responsible for implementing the program.

379In our March 2021 report (GAO-21-387), the sum of about $52 billion includes about $46 billion in direct payments
and about $6 billion in crop insurance indemnities minus producer-paid premium.
380Pub. L. No. 117-2, § 1005, 135 Stat. 4, 12-13.
381Section 1005 of ARPA references the deﬁnition of socially disadvantaged farmers and ranchers outlined in the
Food, Agriculture, Conservation, and Trade Act of 1990, Pub. L. No. 101-624, § 2501(a), 104 Stat. 3359, 4064 (codiﬁed as
amended at 7 U.S.C. § 2279(a)). This statute deﬁnes them as farmers or ranchers who are members of a group whose
members have been subjected to racial or ethnic prejudice because of their identity as members of a group without
regard to their individual qualities. Under this deﬁnition, according to USDA, members of socially disadvantaged groups
include, but are not limited to, American Indians or Alaskan Natives, Asians, Blacks or African Americans, Hispanics or
Latinos, and Native Hawaiians or other Paciﬁc Islanders.

Page 222 GAO-22-105051
Congress provided such sums as may be necessary for the program, and USDA set aside $4 billion
dollars to implement it. However, as of September 2021, FSA had issued debt relief payments to
only four agricultural producers before legal challenges stopped the agency from making further
payments. In June 2021, a federal judge issued a temporary restraining order prohibiting USDA
from forgiving loans under the debt relief program until such time as the court could rule on
whether a preliminary injunction was warranted. A preliminary injunction would prevent FSA
from issuing debt relief payments pending the outcome of a court case alleging discrimination.
Later in that same month, before the ﬁrst court could consider the matter, a separate court
issued a preliminary injunction preventing USDA from making debt relief payments nationwide.
Subsequently, additional lawsuits challenging the program on the basis of discrimination have
been ﬁled. Generally, the complaints allege that the race of the loan account holder, even a
member of a socially disadvantaged group, should not be the basis in awarding debt forgiveness.
According to USDA’s website, FSA plans to continue to process paperwork for eligible socially
disadvantaged agricultural producers until the preliminary injunction is lifted and the agency is
able to make payments.

Overview of Key Issues
FSA has identiﬁed loan accounts that are eligible or potentially eligible for the program.
Socially disadvantaged agricultural producers do not apply to participate in USDA’s ARPA debt
relief program. Rather, FSA identiﬁes the producers with eligible direct loan accounts and
potentially eligible guaranteed loan accounts.382 In September 2021, FSA oﬃcials stated that the
agency had identiﬁed 21,564 agricultural producers with direct (18,586) or guaranteed (2,978) loan
accounts that are eligible or potentially eligible for the program.383

These agricultural producers have a total unpaid loan balance of about $4.6 billion. When
accounting for tax liabilities and other fees on these loan balances, as of September 2021, FSA
oﬃcials estimated that potential payments would total about $5.6 billion, which exceeds the $4
billion currently set aside to fund the program.384 However, according to USDA, not all of the
agricultural producers with eligible loan accounts may choose to participate in the program. FSA
explained that, in rare instances, eligible agricultural producers elect to not receive a payment for
various personal reasons.

382For this report, we refer to the producers with guaranteed loan accounts as being “potentially eligible” because FSA
plans to issue a separate notice of funding availability specifying the time frames and requirements for guaranteed loan
accounts, according to the agency’s website and a May 2021 notice of funding availability on the program. According
to the May 2021 notice, the separate notice will also cover direct loan accounts that no longer have collateral and have
been previously referred to the Department of the Treasury for debt collection for oﬀset (e.g., are delinquent or in
foreclosure), which were not included among the accounts FSA identiﬁed as eligible.
383The number of socially disadvantaged agricultural producers with eligible loan accounts can change over time. For
example, according to FSA oﬃcials, between December 31, 2020, and July 31, 2021, about 5,500 agricultural producers
updated their identiﬁcation in FSA’s database from non-socially disadvantaged to a member of a socially disadvantaged
group. The total for direct loans, 18,586, includes 457 agricultural producers with Farm Storage Facility Loans, a type of
direct loan eligible for debt relief.
384According to FSA, if the initial amount set aside is insuﬃcient to meet all payments, the ARPA legislation provides
for sums as necessary to execute the program, and an additional amount will be made to make payments to all eligible
accounts.

Page 223 GAO-22-105051
FSA sent notiﬁcation letters to the majority of agricultural producers with eligible direct
loan accounts. Although the preliminary injunction prevents USDA from making payments under
the program, USDA sent “notiﬁcation letters” to the majority of agricultural producers it identiﬁed
as having eligible direct loans; it plans to do so at a later date for guaranteed loans. These
letters speciﬁed the terms of the debt relief, including FSA's calculation of payments for direct
loan accounts. The letters also emphasized that agricultural producers will be sent notiﬁcation
regarding guaranteed loans at a later date and that they are still indebted for any debt not eligible
for a payment. If the agricultural producers agreed to the terms, they were to sign the letter and
send it back to USDA.385

As of September 16, 2021, FSA oﬃcials said that the agency sent notiﬁcation letters to
approximately 14,170 of the agricultural producers with eligible direct loans and received
approximately 9,669 signed letters back. The remaining producers had not responded, as of
September 2021. For those agricultural producers that returned signed letters, FSA is ready to
make payments if the injunction is lifted, according to agency oﬃcials. FSA oﬃcials said that the
agency did not send notiﬁcation letters to all of the agricultural producers with eligible direct loan
accounts at one time because it sends notiﬁcation letters as it determines and validates planned
payments, which oﬃcials said is a time-consuming process.

FSA plans to send notiﬁcation letters related to guaranteed loan accounts after it issues a
subsequent notice of funding availability, according to FSA.

FSA provided some characteristics of the agricultural producers with eligible direct and
potentially eligible guaranteed loan accounts. FSA provided us with some information about
the characteristics of agricultural producers with eligible direct loan accounts or potentially eligible
guaranteed loan accounts, as of September 2021:

• More than 50 percent of producers with eligible accounts are located in Arkansas, Oklahoma,
Texas, or Puerto Rico.

• Most of the eligible direct loan accounts and potentially eligible guaranteed loan accounts are
for debts of less than $600,000.

USDA also provided information on participation in the debt relief program by agricultural
producer group. USDA’s data suggest that the largest group of producers with eligible accounts
were identiﬁed as American Indian or Alaskan Native, followed by producers that were identiﬁed
as Hispanic or Latino, and producers that were identiﬁed as Black or African American. However,
in September 2020, the USDA Oﬃce of the Inspector General found that, in violation of
departmental policy, some of FSA’s data on race and ethnicity were based on USDA employees’
visual assessments of agricultural producers instead of being voluntarily provided by the
agricultural producers themselves. According to FSA, for the ARPA debt relief program, agricultural
producers are required to certify their race and ethnicity to receive payments. Because of this

385Depending on the type of operation and loan terms, more than one producer may be required to sign the
notiﬁcation letters. According to FSA’s Notice of Funds Availability; American Rescue Plan Act of 2021 Section 1005 Loan
Payment (ARPA), agricultural producers also have the options of declining to participate or requesting a meeting with FSA
to discuss the letter’s contents. 86 Fed. Reg. 28,329 (May 26, 2021).

Page 224 GAO-22-105051
certiﬁcation process, we requested information on the number of producers who voluntarily
provided or certiﬁed their race and ethnicity data. These data will allow us to conﬁrm the
distribution of eligible loans by socially disadvantaged agricultural producer group for the loan
programs included in the ARPA debt relief program. However, although FSA provided these data,
the agency did not do so in time to be reviewed and incorporated into the report. We will be
analyzing this information and reporting on it in our future updates.

Methodology
To conduct this work, we reviewed federal laws, agency guidance on USDA’s website, and FSA
oﬃcials’ written responses to our questions about the notiﬁcation letters sent and received
by USDA and participation by socially disadvantaged producers, among other topics. We also
reviewed summary data that USDA provided to us as of July 28 and September 15, 2021, as well
as FSA data for direct loans as of January 1, 2021, and guaranteed loans as of December 31, 2020.
Based on our review of the written responses from USDA oﬃcials, we determined that these
data were reliable for our purposes of supporting the description of the total number, location,
and loan size associated with eligible or potentially eligible accounts as of July 28 and September
15, 2021. We were unable to determine whether the data by socially disadvantaged agricultural
producer group were suﬃciently reliable for race and ethnicity.

Agency Comments
We provided FSA and the Oﬃce of Management and Budget with a draft of this enclosure. FSA
provided technical comments, which we incorporated as appropriate. The Oﬃce of Management
and Budget did not provide comments.

GAO’s Ongoing Work
We will continue to monitor USDA’s spending for and implementation of its ARPA debt relief
program for socially disadvantaged agricultural producers.

Related GAO Products
Agricultural Lending: Information on Credit and Outreach to Socially Disadvantaged Farmers and
Ranchers Is Limited. GAO-19-539. Washington, D.C.: July 11, 2019.

Indian Issues: Agricultural Credit Needs and Barriers to Lending on Tribal Lands. GAO-19-464.
Washington, D.C.: May 9, 2019.

Contact information: Steve D. Morris, (202) 512-3841, morriss@gao.gov

Page 225 GAO-22-105051
Shuttered Venue Operators Grant
The Small Business Administration’s Shuttered Venue Operators Grant program, which had
awarded $9.7 billion out of a total appropriation of $16.25 billion as of September 20, 2021, has
faced information technology challenges and concerns about the agency’s ability to eﬀectively
oversee the program.

Entity involved: Small Business Administration

Background
In 2020 and 2021, businesses in the performing arts and entertainment industries experienced
economic hardship due to the COVID-19 pandemic. To assist these businesses, the Consolidated
Appropriations Act, 2021, enacted in December 2020, authorized and appropriated $15 billion
to the Small Business Administration (SBA) to implement the Shuttered Venue Operators Grant
(SVOG) program. The American Rescue Plan Act of 2021, enacted in March 2021, appropriated an
additional $1.25 billion for the program and modiﬁed certain eligibility requirements.386

• Eligibility. Eligibility is limited to live venue operators, promoters, live performing arts
organization operators, movie theater operators, museums, talent representatives, and
theatrical producers. Entities must have been in operation as of February 29, 2020, and must
have lost at least 25 percent in gross earned revenue during at least one quarter of 2020
(compared to the same quarter in 2019), among other eligibility criteria.387

• Prioritization of grant awards. During the ﬁrst 28 days of the program, SBA was required to
prioritize applicants that lost signiﬁcant gross earned revenue between April and December
2020 (compared to the same period in 2019). Speciﬁcally, during the ﬁrst 14 days of the
program accepting applications, SBA could only award grants to otherwise eligible applicants
that had experienced a 90 percent or greater revenue loss. During the next 14 days, SBA
could only award grants to otherwise eligible applicants that had experienced a 70 percent or
greater revenue loss. After the priority period, SBA was required to award grants to all other
eligible entities. In addition, SBA had to reserve at least $2 billion for eligible entities with up to
50 full-time employees during the ﬁrst 59 days of the program.

• Grant amounts. SBA is authorized to award initial grants equal to 45 percent of an eligible
applicant’s 2019 gross earned revenue. SBA also is authorized to award supplemental grants,
if funding is available, to initial grant recipients that lost at least 70 percent during the most
recent calendar quarter of 2021 (compared to the same quarter in 2019). Grant recipients
can receive a maximum of $10 million in combined funding for their initial and supplemental
awards.

386The American Rescue Plan Act of 2021 modiﬁed grant eligibility requirements to allow entities that received a
Paycheck Protection Program loan on or after December 27, 2020, to apply for SVOG. The loan amount must be
deducted from the SVOG grant amount.
387SBA program guidance deﬁnes gross earned revenue as total earned revenue from sales of goods or services
(and excludes other fund sources, such as donations, governmental assistance, or returns on investments). In
addition, entities must operate primarily in the U.S.; submit a good faith certiﬁcation of need; and demonstrate that
their principal business activity, model, and site meet the characteristics of an eligible type of entity.

Page 226 GAO-22-105051
• Fund usage. Grant recipients may spend initial grants on certain expenses incurred between
March 2020 and December 2021 that enabled ongoing business operations (e.g., payroll costs,
rent/mortgage payments). Grant recipients must return any initial funds not spent within 1
year of disbursement and maintain and submit records within 15 days of expending their
grant to demonstrate compliance with grant requirements.388

Through the program, SBA is responsible for awarding $16.25 billion in grants to replace
revenue that eligible businesses in the performing arts and entertainment industries lost due
to the pandemic. As of September 20, 2021, the agency had issued decisions on 95 percent
of applications. As of this date, applicants had requested 82 percent of the $16.25 billion in
grant funds and SBA had awarded $9.7 billion, of which $9.5 billion had been disbursed. Grant
recipients—located in all 50 states, the District of Columbia, Guam, Puerto Rico, and the U.S. Virgin
Islands—were awarded about $812,000 on average. Most grant recipients (70 percent) received
$500,000 or less (see ﬁgure). In addition, of the funds awarded, $5.6 billion went to entities with up
to 50 full-time employees.

Shuttered Venue Operators Grant (SVOG) Initial Grant Recipients, by Grant Amount, as of Sept. 20, 2021

According to SBA, the program stopped accepting new applications on August 20, 2021. On
August 27, 2021, SBA began inviting grant recipients that meet the 70 percent loss requirement to
request supplemental grants. SBA oﬃcials said that eligible grant recipients would have 2 weeks to
indicate their interest in receiving supplemental grants, after which SBA could not guarantee those

388If an eligible entity receives a supplemental grant, these funds, as well as the initial grant, may be used for costs
incurred between March 2020 and June 2022. Grant recipients must return any supplemental funds (as well as
initial funds) not spent within 18 months of disbursement, dating from the date of the initial award.

Page 227 GAO-22-105051
funds would remain available. On September 24, 2021, SBA announced that it had begun issuing
supplemental award decisions.

Overview of Key Issues
Information technology issues delayed the intake and processing of SVOG applications,
but SBA modiﬁed its review process to speed up award issuance. On March 19, 2021, SBA
announced it would start accepting SVOG applications on April 8, 2021, through an online portal.
However, SBA closed the online application portal within hours of opening due to an error in the
software code that prevented applicants from completing the applications. SBA delayed the start
of the program for 2 weeks while it worked with its contractor to make the portal operable. On
April 23, 2021, SBA announced that it would reopen the portal on April 24, 2021, but shifted the
date to April 26, 2021, due to concerns from stakeholders about reopening on a Saturday.

Upon reopening the portal to applicants on April 26, 2021, SBA had to work with its contractor
to develop and integrate technical features necessary for the agency to review applications
and award grants. As a result, SBA delayed its issuance of awards from the initially announced
timeline of mid-May 2021 to May 26, 2021. According to one performing arts industry group,
its membership had experienced signiﬁcant ﬁnancial hardship due to the pandemic and found
these delays frustrating. Representatives from the industry group told us their members had been
waiting to receive ﬁnancial assistance since the program was authorized in December 2020. They
explained that the delays meant their members’ access to much needed ﬁnancial assistance kept
being prolonged. See the ﬁgure below for a timeline of the SVOG program.

Page 228 GAO-22-105051
Timeline for the Shuttered Venue Operators Grant Program, as of Sept. 20, 2021

Even after additional features were added to the portal, SBA was not meeting internal goals for
processing applications, according to SBA oﬃcials. SBA oﬃcials told us their initial goal was to
issue decisions within the ﬁrst 14 days of grant awards for all applicants that had experienced a 90
percent or greater revenue loss between April and December 2020 (compared to the same period
in 2019). According to SBA data, by June 9, 2021, the program had issued only 90 awards, totaling
$127.9 million. On June 10, 2021, SBA revised its goal for processing applications and set a new
goal of internally ﬁnalizing decisions regarding eligibility for 10,000 applications by July 4, 2021,
according to agency oﬃcials.

To process applications more quickly, SBA oﬃcials told us they revised the review process
after June 10, 2021. According to SBA, the agency modiﬁed its review approach for verifying
applicants’ information to align with standards used in other federal programs, such as the
Restaurant Revitalization Fund. Previously, the agency had assessed and veriﬁed information
that all applicants submitted using external government databases. Beginning in mid-June 2021,
SBA started to rely more on applicants’ attestation and submitted documents to validate certain
eligibility components for selected applicants. In addition, SBA began to disburse funds to grant
recipients in a single lump-sum payment, rather than using its earlier method of disbursing funds

Page 229 GAO-22-105051
in installments based on risk factors. The rate at which SBA awarded grants increased after it
implemented these application revisions, among other changes.

SBA’s Oﬃce of Inspector General (OIG) raised concerns about SBA’s ability to conduct
eﬀective program oversight. On April 7, 2021, the SBA OIG issued a management alert
expressing concerns about SBA’s implementation of the program, including policies and
procedures to address potential misuse of program funds.389 The management alert suggested
four actions for strengthening program oversight (see table). Because the management alert does
not make recommendations, OIG oﬃcials told us they are not following typical procedures to
monitor SBA’s progress in implementing the four suggested actions. However, OIG oﬃcials had
information on SBA’s progress from documents that SBA submitted following the management
alert’s issuance. OIG oﬃcials said SBA had fully addressed two of the suggested actions.

389Small Business Administration, Oﬃce of Inspector General, Management Alert: Serious Concerns About SBA’s Control
Environment and the Tracking of Performance Results in the Shuttered Venue Operators Grant Program, Report Number 21-13
(Apr. 7, 2021). According to the OIG, SBA cites 2 CFR 200 as the applicable regulation for this federal award.

Page 230 GAO-22-105051
SBA’s Implementation of Suggested Actions in the OIG Shuttered Venue Operators Grant Management Alert, as
of Aug. 2021
Oﬃce of Inspector General (OIG) suggested
actions for SBA Status SBA actions

Reassess its audit plan for the program to In progress SBA is revising the program’s audit plan.
strengthen internal controls and reduce risk
of fund misuse. SBA’s audit plan classiﬁed the
majority of awards as low-risk due to the size of
the award (less than $1 million) and planned to
subject them to limited oversight.

Implement federally required performance In progress SBA is developing performance goals and
measures to determine the impact of program measures for the program.
funds.

Ensure suﬃcient resources are available to Completed SBA hired a suﬃcient number of grant
implement and oversee the program. oﬃcers to eﬀectively administer the
program.

Clearly establish federal grant requirements for Completed SBA updated the program description
the program to ensure compliance during the listed on the Catalog of Federal Domestic
implementation and oversight phases. Assistance to provide consistent information
about grant requirements.

Source: Small Business Administration (SBA). | GAO-22-105051

In some instances, SBA’s updates to guidance resulted in confusion for applicants. SBA
has issued several types of program guidance for applicants, such as an application checklist
and frequently asked questions. However, SBA reversed guidance under which many applicants
had operated for multiple months. For example, on July 27, 2021, SBA issued modiﬁed guidance,
stating that applicants could appeal or request a reconsideration of their application decisions.
This reversed earlier guidance, in eﬀect since February 2021, that prohibited appeals or
reconsiderations. SBA has been making funding decisions since late May 2021, which means that
applicants denied funding and operating under the old guidance did not know they would have an
opportunity to appeal the denial. As a result, some applicants operating under the old guidance
may have had to close their businesses or seek other ﬁnancial support in the months before the
guidance changed. According to SBA, the agency is retroactively applying the modiﬁed guidance.
On August 2, 2021, SBA began inviting all declined applicants to appeal their denials. On August 6,
2021, SBA invited grant recipients awarded at least $100 less than they had requested to request
reconsideration of their grant amount.

Representatives from ﬁve industry groups told us SBA’s approach to updating guidance
sometimes made it diﬃcult to understand program requirements, but they added that SBA
has been responsive to their feedback. For example, representatives from two industry groups
noted that SBA did not appear to fully understand their industries’ diverse business models
when creating written descriptions of program eligibility requirements. SBA has also updated
some guidance to improve its clarity in response to feedback from industry groups, according
to representatives from a commercial theater industry group and museum industry groups.
SBA oﬃcials said their iterative approach to disseminating program guidance was necessary for

Page 231 GAO-22-105051
reasons that included wanting to ensure the guidance reﬂected lessons learned as the program
progressed.

Methodology
We reviewed SBA data on applications and grants awarded as of September 20, 2021. We
determined these data were suﬃciently reliable for our purposes by discussing the data collection
and quality-check processes with knowledgeable SBA oﬃcials. We also reviewed relevant laws
and SBA program guidance, policies, and procedures. We interviewed SBA oﬃcials to discuss
their experiences in administering the program and assistance to program participants. Further,
we reviewed the SBA OIG’s management alert for the program and relevant agency documents
that pertain to concerns raised in the alert. In addition, we interviewed OIG oﬃcials to discuss
SBA’s responses to the alert. Finally, we interviewed ﬁve associations that represent SVOG-
eligible businesses. Their views are not generalizable to all SVOG eligible businesses, but oﬀered
important perspectives.

Agency Comments
We provided SBA and the Oﬃce of Management and Budget (OMB) with a draft of this enclosure.
SBA provided technical comments, which we incorporated as appropriate. OMB did not provide
comments on this enclosure.

GAO’s Ongoing Work
Our work assessing SBA’s implementation of the SVOG program is ongoing. We plan to examine
the characteristics and experiences of SVOG participants. We also plan to examine the safeguards
SBA implemented to help ensure that grants were awarded to applicants that met program
requirements and were used to pay for eligible expenses.

Contact information: William B. Shear, (202) 512-8678, shearw@gao.gov

Page 232 GAO-22-105051
Restaurant Revitalization Fund
The Small Business Administration awarded all of the $28.6 billion in Restaurant Revitalization
Fund assistance by June 30, 2021.

Entity involved: Small Business Administration

Background
In 2020 and 2021, restaurants, bars, and other food service businesses experienced substantial
revenue loss due to the COVID-19 pandemic and related business closures, disrupted business
models, and decreased customer demand for indoor dining. On March 11, 2021, the American
Rescue Plan Act of 2021 established the Restaurant Revitalization Fund (RRF) and appropriated
$28.6 billion for the Small Business Administration (SBA) to provide support to eligible entities that
suﬀered revenue losses related to the COVID-19 pandemic.

• Eligibility. Businesses eligible for RRF assistance included restaurants, food stands, caterers,
bars, and similar places of business that serve food or drink.390 Eligibility requirements
included that businesses had to have fewer than 20 locations and had to not have been
permanently closed.391 Assistance may be used for 11 types of business expenses, including
payroll, mortgage, rent, supplies, and food and beverages.392

• Assistance size and requirements. The maximum funding amount was $5 million per
location and $10 million for businesses with multiple locations or aﬃliates. The minimum
funding amount was $1,000. Assistance was awarded based on the recipients’ documented
revenue losses.393 The funds do not need to be repaid as long as they are used for eligible
expenses by March 11, 2023. Recipients that do not fully expend award funds prior to
December 31, 2021, will be required to complete annual reporting submissions until they

390The following businesses were also eligible: bakeries, breweries, microbreweries, brewpubs, distilleries, food
carts, food trucks, inns, lounges, saloons, snack and nonalcoholic beverage bars, taprooms, tasting rooms, taverns,
and wineries.
391Other requirements related to legal structure, ownership, and other federal relief funding. Eligibility was
limited to C corporations, S corporations, partnerships, limited liability companies, sole proprietors, self-employed
individuals, independent contractors, and tribal businesses. Eligibility was limited to businesses that earned at
least 33 percent of 2019 gross receipts from on-site sales of food and beverage to the public. Franchisees were
eligible, but nonproﬁt organizations, businesses operated by state or local governments, and publicly traded
companies were not. Applicants also could have received an SBA Paycheck Protection Program loan or Economic
Injury Disaster Loan, but they could not have received a Shuttered Venues Operators Grant or have had a pending
application for that grant.
392Other eligible expense types are debt service; utility payments; maintenance, including construction of outdoor
seating and maintenance on walls, ﬂoors, deck surfaces, furniture, ﬁxtures, and equipment; covered supplier costs;
and operating expenses. Covered supplier costs are expenditures (1) for goods essential to operations and made
pursuant to a contract, order, or purchase order prior to the RRF funding; or (2) for purchase of perishable goods
pursuant to a contract, order, or purchase order in eﬀect before or at any time during the period beginning on
February 11, 2020, and ending on March 11, 2023.
393Funding amounts were based on revenue losses as documented through gross receipts less any Paycheck
Protection Program loans or other federal COVID-19 relief payments.

Page 233 GAO-22-105051
fully expend the award funding or until March 11, 2023. SBA plans to put a new form in the
application portal for recipients to report on their use of award funds in October 2021.

• Application period. SBA accepted RRF applications from May 3, 2021 to May 24, 2021.394 SBA
began approving applications on May 7, 2021, and disbursing funds on May 10, 2021.

The American Rescue Plan Act required that during the ﬁrst 21 days of the program, eligible
entities owned and controlled by women, veterans, or socially and economically disadvantaged
small business concerns as deﬁned in the Small Business Act receive priority in award
assistance.395 The 21-day period ended on May 24, 2021. After SBA launched the RRF, three federal
lawsuits were ﬁled challenging the 21-day priority period—one in the Eastern District of Tennessee
(Vitolo et. al. v. Guzman) and two in the Northern District of Texas (Greer’s Ranch Café et. al. v.
Guzman and Blessed Cajuns et. al. v. Guzman), which led to three adverse court rulings against
SBA.

With the expiration of the 21-day priority period on May 24, 2021, and in light of certain court
rulings issued on May 27 and May 28, 2021, SBA began processing RRF applications in the order
received, without regard to priority status.396 SBA continued to process applications and issue RRF
assistance until the fund was exhausted on June 30, 2021.

Overview of Key Issues
All RRF funds had been awarded as of June 30, 2021. Speciﬁcally, SBA disbursed funds totaling
$28.5 billion to 100,650 recipients (see ﬁgure). Thirty-six percent of applicants received funding.
According to SBA, an estimated 177,300 applicants were not approved because the fund was
exhausted. Businesses applied for a total of $72.2 billion in funding, which was $43.7 billion more
than was appropriated for the program.

394Applicants could apply online (through an SBA application portal), by telephone, or through SBA Point of
Sale restaurant partners (technology companies that provide software, hardware, and payment services to the
restaurant industry) Square and Toast which provided application submission for their current customers. Other
SBA Point of Sale partners such as Clover, NCR and Oracle provided supporting documentation and knowledge
support to their current customers.
395Pub. L. No. 117-2, § 5003(c)(3), 135 Stat. 4, 88-89 (codiﬁed at 15 U.S.C. § 9009c(c)(3)).
396SBA did not fund 2,964 priority-period applicants that it approved for awards. As a result of the litigation, funding
went to businesses that submitted applications before these 2,964 did. The RRF program’s funding was exhausted
before these 2,964 applications became eligible based on the ﬁrst-come ﬁrst-served basis.

Page 234 GAO-22-105051
Timeline for the Restaurant Revitalization Fund Program, as of July 2021

Funds were distributed across all U.S. states, but were concentrated in urban areas. According to
SBA data,

• 85 percent of recipients were located in urban areas (85,128 recipients, $26 billion);

• 27 percent of recipients operated in low- to moderate-income areas (27,210 recipients, $7.3
billion), which include rural and urban areas;397

• states with the most recipients and funds were California (15,962 recipients, $5.7 billion), New
York (9,753 recipients, $3.7 billion), and Texas (6,372 recipients, $1.7 billion); and

• states with the fewest recipients and funds were South Dakota (193 recipients, $25.6 million),
North Dakota (170 recipients, $26 million), and Wyoming (123 recipients, $18.1 million).

397SBA recipient data for RRF included indicators for rural and urban areas and low- to moderate-income areas,
which SBA identiﬁed using the recipient’s address.

Page 235 GAO-22-105051
Between 12 and 41 percent of potentially eligible recipients per state received funding, as shown
in the ﬁgure below.398 The states with the fewest recipients as a percentage of eligible restaurants
were Idaho (304 recipients, $47.5 million), North Dakota (170 recipients, $26 million), and Wyoming
(123 recipients, $18.1 million), each with 12 percent. The states with the greatest number of
recipients as a percentage of potentially eligible recipients were Hawaii (1,145 recipients, $414.4
million) with 41 percent, the District of Columbia (729 recipients, $400.5 million) with 38 percent,
and Oregon (2,337 recipients, $530.9 million) with 29 percent. Not every potentially eligible
restaurant in each state applied for funding.

398We used the Census Bureau’s 2018 Statistics of U.S. Businesses data to estimate the number of restaurants per state
that were potentially eligible for the RRF program. More speciﬁcally, we designated “total” restaurants in each state as
those with 500 or fewer employees and that had a North American Industry Classiﬁcation System code deﬁnition of
“Food Services and Drinking Places” (NAICS 722) from the most recent County Business Patterns data from the Census
Bureau. While these data do not align with the eligibility criteria for RRF recipients—such as not having more than 20
locations and receiving 33 percent or more of their revenue from food or drink—they provide reasonable estimates for
comparing eligible business populations at the state level.

Page 236 GAO-22-105051
Restaurant Revitalization Fund Recipients, as a Percentage of the Total Number of Restaurants with 500 or
Fewer Employees in Each State, as of June 30, 2021

Note: We used the Census Bureau’s 2018 Statistics of U.S. Businesses data to estimate the number of restaurants per state that
were potentially eligible for the Restaurant Revitalization Fund program. More speciﬁcally, we designated “total” restaurants in
each state as those with 500 or fewer employees and that had a North American Industry Classiﬁcation System code deﬁnition
of “Food Services and Drinking Places” (NAICS 722) from the most recent County Business Patterns data from the Census
Bureau. While these data do not align with the eligibility criteria for Restaurant Revitalization Fund recipients—such as not
having more than 20 locations and receiving 33 percent or more than of their revenue from food or drink—they provide
reasonable estimates for comparing eligible business populations at the state level. This analysis shows recipients as a
percentage of estimated eligible entities in a state, which diﬀers from recipients as a percentage of applicants in a state (which
ranged from an average of 22 percent in Rhode Island and to 48 percent Hawaii in and 36 percent across the U.S.).

Most funding amounts (87 percent) were for less than $500,000 (see ﬁgure). According to SBA
data, approximately 10 percent of recipients were franchisees (businesses that purchase the
right to sell a company’s goods or services under the company’s business model and trademark).
Of those recipients, Subway franchisees were the largest group (2,868 recipients), followed by
Dunkin’ Donuts franchisees (302 recipients) and IHOP franchisees (296 recipients). In addition,
when considering total funding amount per franchise chain, 41 percent of the RRF funds disbursed
to franchisees went to those representing 10 franchise chains.

Page 237 GAO-22-105051
Restaurant Revitalization Fund (RRF) Recipients, by Assistance Amount, as of June 30, 2021

Methodology
To conduct this work, we reviewed SBA documentation on the RRF program and interviewed SBA
oﬃcials. We analyzed SBA’s recipient data and assessed the reliability of these data by interviewing
SBA oﬃcials.399 We determined that the data were suﬃciently reliable for reporting aggregate
recipient data, including the number and dollar amount by geography and other demographic
descriptions. We also reviewed court decisions and other relevant legal documents related to the
program. We interviewed two industry groups that represent eligible businesses; their views are
not generalizable to all eligible businesses.

Agency Comments
We provided SBA and the Oﬃce of Management and Budget with a draft of this enclosure. SBA
and the Oﬃce of Management and Budget did not provide comments on this enclosure.

GAO’s Ongoing Work
Our work assessing SBA’s implementation of the RRF program is ongoing. We continue to
examine the characteristics and experiences of program participants and the safeguards SBA
has implemented to help ensure that funds were disbursed to applicants that met program
requirements and were used to pay for eligible expenses.

399Data on RRF recipients are available at https://data.sba.gov/dataset/rrf-foia.

Page 238 GAO-22-105051
Contact information: William B. Shear, (202) 512-8678, shearw@gao.gov

Page 239 GAO-22-105051
Assistance for Fishery Participants
As of September 15, 2021, of the $298 million in CARES Act funding for ﬁsheries participants,
$214.4 million has been disbursed to eligible ﬁsheries participants; none of the additional $300
million in COVID-19 relief funds made available under the Consolidated Appropriations Act, 2021,
has been disbursed because the agency is still receiving and reviewing spend plans.

Entity involved: National Oceanic and Atmospheric Administration, within the Department of
Commerce

Background
Commercial and recreational marine ﬁsheries are critical to the nation’s economy, contributing
approximately $99.5 billion to the U.S. gross domestic product and supporting approximately 1.7
million jobs in 2016, according to the National Oceanic and Atmospheric Administration (NOAA).400
Widespread restaurant closures in the spring of 2020 led to a decrease in demand for seafood,
adversely aﬀecting the ﬁsheries industry.

The 2020 CARES Act authorizes the Department of Commerce to provide assistance to eligible
tribal, subsistence, commercial, and charter ﬁshery participants aﬀected by COVID-19, which
may include direct relief payments.401 The act appropriated $300 million to the Department of
Commerce to assist ﬁshery participants.402 After $2 million in administrative fees were assessed
by NOAA, $298 million of the $300 million was obligated for eligible ﬁshery participants. These
include tribes, persons, ﬁshing communities, aquaculture businesses not otherwise eligible for
certain assistance, processors, and other ﬁshery-related businesses, who have incurred, as a
direct or indirect result of COVID-19, certain speciﬁed economic revenue losses or other negative
impacts.403 Businesses such as vessel repair businesses, restaurants, and seafood retailers are
not considered ﬁshery-related businesses eligible to receive this CARES Act funding, according to
NOAA’s website.

An additional $300 million for certain COVID-19-related ﬁsheries disaster assistance was
appropriated under the Consolidated Appropriations Act, 2021, signed by the President into law
on December 27, 2020.404 The amount provided is only to be allocated to states bordering the

400U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National Marine Fisheries Service,
Fisheries Economics of the United States, 2016, Technical Memorandum NMFS-F/SPO-187a (Silver Spring, Md.: December
2018). Information on gross domestic product and jobs includes data on commercial seafood harvesters, processors,
dealers, wholesalers, distributors, importers, and retailers, as well as recreational ﬁshing trips and ﬁshing equipment.
Data for 2016 were the most recent available at the time of our review.
401Pub. L. No. 116-136, § 12005(a), 134 Stat. at 518.
402Id. § 12005(d).
403Id. § 12005(b). Speciﬁcally, “ﬁshery participants” are deﬁned as belonging to these categories and as having incurred,
as a direct or indirect result of the COVID-19 pandemic, economic revenue losses greater than 35 percent as compared
with their prior 5-year average revenue or any negative impacts to subsistence, cultural, or ceremonial ﬁsheries.
Additionally, the CARES Act provided that the Department of Commerce may use up to 2 percent of the $300 million for
administration and oversight activities.
404Pub. L. No. 116-260, div. M, tit. I, 134 Stat. 1182, 1909-10. Speciﬁcally, the act appropriated $300 million, to remain
available until September 30, 2021, to prevent, prepare for, and respond to COVID-19, domestically or internationally,
which is to be only for activities authorized under section 12005 of the CARES Act.

Page 240 GAO-22-105051
Atlantic, Paciﬁc, or Arctic Oceans, the Gulf of Mexico, or the Great Lakes, as well as to territories
and certain tribes.405 As of September 7, 2021, all funds appropriated under the Consolidated
Appropriations Act, 2021 have been obligated, according to NOAA oﬃcials.

Overview of Key Issues
CARES Act funding was allocated in May 2020, and most of it was obligated in July 2020. On
May 7, 2020, the Secretary of Commerce announced the allocation of about $298 million of the
$300 million in CARES Act funding for states, tribes, and territories with ﬁshery participants.406

NOAA used these allocations to obligate almost $297 million of the available $298 million in
funding to the three interstate marine ﬁsheries commissions—Atlantic States, Gulf States, and
Paciﬁc States—from June 30 through July 2, 2020, and to the government of Puerto Rico on July 14,
2020. NOAA also obligated $993,000 to the government of the U.S. Virgin Islands on November 13,
2020.407

The commissions play an important role in disbursing funds as part of NOAA’s ﬁshery disaster
assistance program, which provides funding to ﬁshery participants experiencing economic
losses from speciﬁc events, such as hurricanes or oil spills.408 For the CARES Act funding, the
commissions worked with states, tribes, and territories in their regions to develop spend plans
for NOAA’s review and approval. These plans explain how states, tribes, and territories will verify
whether ﬁshery participants meet the requirements of the CARES Act to receive funds.

• As of May 1, 2021, NOAA had received the 30 plans it anticipated from the states and
territories. As of July 1, 2021, all 30 plans had been reviewed and approved.

• Additionally, as of May 1, 2021, 30 tribal spend plans had been submitted to NOAA and all had
been approved.

405The act provides that of the $300 million, $30 million is for COVID-19-related ﬁshing impacts for speciﬁed tribal
ﬁshery participants, and $15 million is for COVID-19-related ﬁshing impacts to nontribal commercial, aquaculture,
processor, and charter ﬁshery participants in states bordering the Great Lakes. Additionally, the act provides that each
eligible state and territory, other than those states only bordering the Great Lakes, is to receive an amount equal to
not less than 1 percent of the total $300 million, and not greater than, from amounts provided under the CARES Act
or this appropriation, that state’s or territory’s total annual average revenue from commercial ﬁshing operations,
aquaculture ﬁrms, the seafood supply chain, and charter ﬁshing businesses.With regard to tribes, the act speciﬁes that
tribes eligible for funding are federally recognized tribes in any of the nation’s coastal states and territories as well as
federally recognized tribes in any of the nation’s Great Lakes states with ﬁsheries on the tribe’s reservation or ceded or
usual and accustomed territory.
406According to NOAA oﬃcials, allocations represent the maximum amount of total funding that ﬁshery participants in
a particular state, tribe, or territory can receive. These allocations are net of administrative fees that NOAA assessed.
Additional administrative fees can be assessed by grantees, such as the interstate marine ﬁsheries commissions,
according to NOAA oﬃcials.
407Puerto Rico and the U.S. Virgin Islands are not part of an interstate marine ﬁsheries commission.
408The commissions, established in the 1940s, partner with NOAA on data collection and the management of ﬁsheries
resources that are shared across states, such as striped bass. Since Puerto Rico and the U.S. Virgin Islands are not part of
the interstate marine ﬁsheries commissions, they handle their own disbursement processes.

Page 241 GAO-22-105051
Once a state, tribe, or territory’s spend plan has been approved by NOAA, that entity solicits and
reviews applications from ﬁshery participants, determines whether participants meet the eligibility
criteria, and determines the direct payment amount based on the methodology outlined in its
spend plan.409 The respective interstate marine ﬁsheries commissions disburse the appropriate
amount of funds directly to the ﬁshery participant consistent with the approved spend plan,
though some states and tribes may disburse funds themselves.410

More than 70 percent of CARES Act funding had been disbursed to ﬁshery participants as of
September 15, 2021. As of September 15, 2021, the three interstate marine ﬁsheries commissions
had disbursed over 70 percent of obligated funds ($214.4 million out of the $298 million) to ﬁshery
participants in 22 states, 17 tribes, and 2 U.S. territories as shown in the table below.411 As we
previously reported in our March 2021 report, NOAA oﬃcials told us that they expected the vast
majority of CARES Act funds to be disbursed to ﬁshery participants by May 2021.

409NOAA oﬃcials said that states, tribes, and territories can also use existing records, such as ﬁshing permits, to identify
eligible recipients.
410As noted previously, Puerto Rico and the U.S. Virgin Islands are not part of an interstate marine ﬁsheries commission,
so they will disburse funds directly to ﬁshery participants.
411Puerto Rico disbursed funds directly to ﬁshery participants, since it is not part of an interstate marine ﬁsheries
commission.

Page 242 GAO-22-105051
Disbursement of CARES Act Funding, as of September 15 , 2021
Funds
disbursed
($ in
Interstate commission State thousands)

Atlantic States Marine Fisheries Commission Massachusetts 27,676

Florida 23,447

Maine 17,129

New Jersey 10,825

New York 6,398

North Carolina 5,255

Virginia 4,483

Rhode Island 3,168

Maryland 3,000

New Hampshire 2,660

Connecticut 1,821

Georgia 1,525

South Carolina 1,483

Delaware 992

a
Puerto Rico 358

Gulf States Marine Fisheries Commission Louisiana 14,397

Alabama 3,254

Mississippi 1,415

Texas 958

Paciﬁc States Marine Fisheries Commission Washington 38,563

California 18,042

Oregon 15,768

Hawaii 4,279

b
West Coast Tribes 6,350

Page 243 GAO-22-105051
Funds
disbursed
($ in
Interstate commission State thousands)

c
Alaska Tribes 932

Guam 176

Total 214,354

Source: GAO analysis of Department of Commerce data. | GAO-22-105051

Note: The CARES Act appropriated $300 million to the Department of Commerce to assist ﬁshery participants. Pub. L. No.
116-136, § 12005(d), 134 Stat. at 518.
a
Puerto Rico is not part of an interstate marine ﬁsheries commission.
b
According to National Oceanic and Atmospheric Administration oﬃcials, 16 of the 30 tribes from this group have received
funding.
c
According to National Oceanic and Atmospheric Administration oﬃcials, one eligible tribe from this group has received
funding.

NOAA oﬃcials acknowledged that the vast majority of funds had not been disbursed by May 2021
for a few reasons. Oﬃcials told us that that the amount of funding provided under the CARES Act
and the number of states involved in the process was larger than any past ﬁshery disaster the
agency has been involved with. In addition, Washington and Alaska, which together account for
about one-third of the allocated funds ($99.3 million), took longer than the other states to submit
spend plans and review applications. According to NOAA oﬃcials, this was due in part to the
number of applications that needed to be reviewed by these states as part of implementing their
spend plans and in part to the review process occurring during the ﬁshing season when ﬁshery
participants were not available to meet with state oﬃcials to discuss identiﬁed issues in their
applications. NOAA oﬃcials said that during conversations with state and commission oﬃcials,
they have shared lessons learned from the disbursement process in an attempt to help other
states disburse the funds in a more expedited manner.

An additional $300 million in COVID-19 relief funds was appropriated in December 2020,
and as of September 7, 2021, all of the funds have been obligated but none have been
disbursed. Of the $300 million in additional COVID-19 relief funds appropriated under the
Consolidated Appropriations Act, 2021, $255 million has been allocated by NOAA to the three
ﬁsheries commissions, Puerto Rico, and the U.S. Virgin Islands. The remaining $45 million was
allocated to speciﬁed tribal ﬁshery participants and to ﬁshery participants bordering the Great
Lakes. As of September 7, 2021, all funding has been allocated and obligated but no funds have
been disbursed, according to agency oﬃcials. Fourteen states have approved spend plans and
have initiated their application processes to disburse funds. According to NOAA oﬃcials, the
agency is still receiving and reviewing spend plans from the remaining marine coastal states and
territories, and the Great Lakes states. As of September 15, 2021, NOAA has received 10 spend
plans to review, and is still waiting to receive 12 spend plans.412

412Of the 10 spend plans, NOAA received seven plans from the marine coastal states and territories, and three plans
from the Great Lakes states.

Page 244 GAO-22-105051
Methodology
To conduct this work, we reviewed NOAA data as of September 15, 2021; examined documents
related to NOAA’s data system; and interviewed oﬃcials familiar with the data system. We
determined that the data were suﬃciently reliable for our purposes by reviewing agency
documents and speaking with agency oﬃcials. We also reviewed relevant laws, including
the CARES Act and the Consolidated Appropriations Act, 2021, and agency documents, and
interviewed NOAA oﬃcials.

Agency Comments
We provided a copy of this enclosure to NOAA and the Oﬃce of Management and Budget
(OMB) for review and comment. NOAA provided us with technical comments, which we have
incorporated as appropriate. OMB did not provide comments on this enclosure.

GAO’s Ongoing Work
We will continue to monitor CARES Act assistance to ﬁshery participants in ongoing and planned
work.

GAO’s Prior Recommendations
The table below presents our recommendation on assistance for ﬁshery participants from prior
bimonthly and quarterly CARES Act reports.

Page 245 GAO-22-105051
Prior GAO Recommendation Related to Assistance for Fisheries Participants
Recommendation Status

The Assistant Administrator for the National Oceanic and Closed—implemented. In February 2021,
Atmospheric Administration (NOAA) Fisheries should develop NOAA developed an electronic tracking tool to
a mechanism to track the progress of states, tribes, and track the disbursement of funds to ﬁsheries
territories in meeting timelines established in spend plans to participants, and as of July 2021, the agency was
disburse funds in an expedited and eﬃcient manner. (January, regularly inputting data into this tool to monitor
2021) the progress of states, tribes, and territories to
disburse funds.

Source: GAO. I GAO-22-105051

Contact information: Cardell Johnson, (202) 512-3146, johnsoncd1@gao.gov

Page 246 GAO-22-105051
Food Safety Inspections
The U.S. Department of Agriculture has taken steps to continue required inspections of meat and
poultry and track the COVID-19 status of inspectors.

Entity involved: Food Safety and Inspection Service, within the U.S. Department of Agriculture

Background
COVID-19 has caused disruptions in the U.S. food supply chain, from the farms that produce raw
agricultural commodities to the food-processing and distribution network that enables consumers
to use these commodities. The 7,799 inspectors and other frontline staﬀ from the U.S. Department
of Agriculture’s (USDA) Food Safety and Inspection Service (FSIS) work in more than 6,500 federally
inspected meat and poultry plants and other establishments.413 These inspectors help ensure
the safety and wholesomeness of meat and poultry that enter interstate commerce. FSIS also
has responsibility for inspecting meat and poultry products at U.S. ports and conducting audits of
foreign food safety inspection systems.

During the pandemic, some FSIS inspectors have been exposed to COVID-19, and the risk of
such exposure has posed a potential challenge for FSIS in continuing food safety inspections.
According to April 2020 interim guidance from the Centers for Disease Control and Prevention
(CDC) and the Occupational Safety and Health Administration, working in close conditions, as is
common in meat and poultry plants, may contribute to exposure to COVID-19. In January 2021,
the Occupational Safety and Health Administration published COVID-19 worker safety guidance,
including recommendations for workplace COVID-19 prevention programs and policies.414 See the
enclosure on Worker Safety and Health in appendix I for more information on the Occupational
Safety and Health Administration’s eﬀorts during the pandemic.

As of July 2021, USDA had obligated and spent $32 million of the $33 million in CARES Act
funds that Congress appropriated to FSIS in March 2020 to prevent, prepare for, and respond
to COVID-19 with regard to food safety inspections.415 FSIS spent almost $24 million for food
safety inspections, including to cover costs to maintain staﬃng of FSIS inspectors and to fund
achievement awards.416 For example, oﬃcials told us that USDA most recently used $13.9 million
of these funds for achievement awards to frontline and other FSIS employees who volunteered to

413The Food and Drug Administration (FDA) covers food safety activities for other types of food production. FDA did not
receive CARES Act funds for food safety-related activities.
414Department of Labor, Occupational Safety and Health Administration, Protecting Workers: Guidance on Mitigating and
Preventing the Spread of COVID-19 in the Workplace (Jan. 29, 2021), accessed February 12, 2021, https://www.osha.gov/
coronavirus/safework. In January 2021, we reported on the Department of Labor Occupational Safety and Health
Administration’s eﬀorts to protect worker safety and health.
415Pub. L. No. 116-136, div. B, tit. I, 134 Stat. 281, 506 (2020).
416As we reported in March 2021, USDA used CARES Act funds to provide FSIS employees with personal protective
equipment. In addition, as we reported in November 2020, USDA used the funds to maintain staﬃng of FSIS inspectors,
including to cover costs for additional hours worked by part-time inspectors, to bring in inspectors from other USDA
oﬃces, and to pay for nonreimbursable overtime. Nonreimbursable overtime occurs when an inspector has already
worked a full shift and needs to work additional hours at another establishment that is not in an overtime status and
subject to paying FSIS an overtime fee, according to USDA oﬃcials. The agency also used almost $400,000 for state
inspections, laboratory testing of pathogens in meat and poultry products, and training.

Page 247 GAO-22-105051
perform in-plant inspection from March 2020 through June 2020.417 Of the remaining $8 million,
FSIS used these funds for supplies and materials, such as personal protective equipment. USDA
plans to use any remaining CARES Act funds for award payouts to in-plant personnel, according to
oﬃcials.

Overview of Key Issues
FSIS continued U.S. inspection activities and conducted remote audits to evaluate foreign
inspection systems during the COVID-19 pandemic. According to USDA oﬃcials, during
the COVID-19 pandemic, FSIS has maintained required inspection services to ensure that
establishments including meat and poultry plants can operate. FSIS also continued to execute
its ﬁscal year 2021 testing plan—sampling of pathogens in meat and poultry products—with no
impact from the COVID-19 pandemic.

In addition to inspections at U.S. plants, FSIS took steps to continue monitoring and verifying
that foreign food safety inspection systems met U.S. standards. In carrying out this responsibility,
FSIS typically uses a three-pronged approach—point-of-entry reinspection of imported meat and
poultry products, reviews of annual self-reporting tools by countries that export to the U.S., and
on-site veriﬁcation audits—and it largely maintained this approach during the pandemic.418

Although FSIS was not able to conduct on-site audits in foreign countries because of COVID-19,
the agency implemented a remote audit process, consisting of virtual meetings and electronic
information exchanges, to evaluate the eﬀectiveness of foreign food safety inspection systems
through document review and interviews.419 According to oﬃcials, FSIS also used available
data streams to monitor for food-related recalls, illness outbreaks, natural disasters, issuance
of inspection system audit reports by third parties, and animal disease outbreaks in countries
exporting to the U.S. The remote audit approach is similar to the one taken by some trading
partners for auditing the U.S. inspection system during this time, according to FSIS oﬃcials.

According to agency documentation, as of December 2020, the remote audit is to assess
inspection records for a designated period within the last 12 months and seek to verify ongoing
eﬀectiveness of foreign inspection programs. Throughout the pandemic, these remote audits,

417According to FSIS oﬃcials, 6,700 frontline employees and other employees qualiﬁed for the award. The value of
the individual awards is based on how many hours qualiﬁed employees worked in the establishments during the time
frame. The minimum award for individual employees who worked during this time frame is $500, and the award value
increases for every in-plant hour worked. The awards ranged from $500 to approximately $4,000.
418According to oﬃcials, FSIS continued to perform ongoing point-of-entry reinspections of all imported meat and
poultry products, as well as reviews of self-reporting tool submissions by countries that export such products to the
U.S. The self-reporting tool is the system FSIS uses to obtain information from foreign countries seeking to demonstrate
that their documented food safety inspection systems achieve a level of public health protection equivalent to the U.S.
inspection system. FSIS has not denied or delayed any reinspection service requests for imported meat, poultry, catﬁsh,
or processed egg products, according to oﬃcials.
419The remote audit approach serves as a temporary measure to fulﬁll FSIS requirements in lieu of conducting on-site
audits of foreign food safety inspection systems. The proposed remote audit schedule includes all ongoing equivalence
audits postponed during ﬁscal year 2020, as well as those countries scheduled for ﬁscal year 2021. The proposed audit
schedule includes the following countries: Australia, Austria, Canada, China, Finland, France, Honduras, Iceland, Italy,
Mexico, Netherlands, New Zealand, Nicaragua, Poland, San Marino, Thailand, and Vietnam.

Page 248 GAO-22-105051
among other things, have not shown evidence of any systemic failures to comply with U.S.
equivalency standards representing an immediate threat to public health, according to FSIS
oﬃcials.

FSIS may leverage performance-monitoring data used by foreign authorities, if available, to assess
the eﬀectiveness of foreign authorities’ inspection systems. This approach is similar to how FSIS
collects and utilizes data to monitor and verify the eﬀectiveness of the U.S. inspection system.
The remote audit process has proven to be a useful tool for reviewing foreign inspection systems
in the absence of being able to travel internationally to perform on-site audits, according to
oﬃcials. FSIS is currently in the process of scheduling on-site audits in locations where it is safe
to do so; however, lessons learned from the remote audit process will be applied to enhance the
eﬃciency and eﬀectiveness of future audits, according to oﬃcials. They noted that such audits will
incorporate a blend of virtual meetings and document review with on-site observations.

USDA has tracked FSIS inspectors’ and other staﬀ’s COVID-19-related absences and
identiﬁed lessons learned. According to USDA documentation, as of August 23, 2021, 1,692
inspectors and staﬀ (or about 22 percent of total inspectors and other frontline staﬀ) had reported
a COVID-19 diagnosis conﬁrmed by a test or medical professional since the ﬁrst reported case in
March 2020. Of these employees, 1,620 had returned to work, seven were self-quarantining, and
14 were deceased.420

Although FSIS does not speciﬁcally require employees to report COVID-19 infections and relies
on staﬀ to self-report, FSIS employees are required to report an absence from work. According to
USDA’s FSIS Human Pandemic Operations Plan, under the agency’s leave directive, FSIS employees
are required to report to their supervisors if they will be absent from duty and, in the event of
a human pandemic, supervisors and managers should initiate a follow-up for any unscheduled
employee absence.421 Upon notiﬁcation of a positive COVID-19 test result from an employee,
FSIS gathers pertinent information such as illness onset, testing date and result, last time in the
building, and contact with other employees at the workplace. The agency uses the information to
determine any necessary notiﬁcations to FSIS and meat and poultry plant personnel, according to
oﬃcials.

In response to the pandemic, USDA and the meat and poultry industry implemented intervention
measures.

420According to USDA oﬃcials, as of June 28, 2021, FSIS continued to collect information related to employees who
tested positive for COVID-19, but this information is reported to USDA through an internal SharePoint system. The total
number of FSIS employees with a conﬁrmed diagnosis of COVID-19 does not equal the total number of employees who
have returned to work, were self-quarantining, or were deceased because the status of some employees is not indicated
in USDA documentation provided to us. According to USDA documentation, the number of employees who had a
COVID-19 diagnosis is cumulative, and this cumulative number includes employees who have returned to work (i.e.,
recovered from the disease) and deceased employees. This number combines inspectors and staﬀ, whose occupational
exposures may vary.
421See U.S. Department of Agriculture, Oﬃce of the Administrator, The Food Safety and Inspection Service Human
Pandemic Operations Plan (March 2020) and U.S. Department of Agriculture, Food Safety and Inspection Service, Leave,
FSIS Directive 4630.2 (Washington, D.C.: Sept. 15, 2010).

Page 249 GAO-22-105051
• Earlier in the pandemic, some meat and poultry plants closed temporarily as an intervention
measure in response to COVID-19 cases among workers, according to CDC.

• In accordance with the department’s COVID-19 workplace safety plan dated July 28, 2021,
USDA began requiring its employees to wear a mask in federal buildings and vehicles owned
or leased by the government, regardless of the employee’s vaccination status, if located in an
area of “substantial” or “high” community COVID-19 transmission.422 In an update to the plan
on August 23, 2021, the department also stated federal employees who work in areas that
have stricter state, local, or tribal mask mandates should follow those masking requirements.

• According to the company’s website, as of August 2021, as an intervention measure, a national
meat and poultry company is requiring all of its employees to be fully vaccinated against
COVID-19 by November 1, 2021.

USDA oﬃcials said that, as of August 2021, there were no establishments that had to close
because of a lack of available FSIS inspectors during the pandemic.

For future pandemics, FSIS intends to update its emergency plan—the FSIS Human Pandemic
Operations Plan—annually and apply lessons learned, according to oﬃcials. The plan will inform
leadership preparation and decision-making for any future pandemics, according to oﬃcials. They
also said that storing up personal protective equipment is one lesson learned from the COVID-19
pandemic experience.

According to these oﬃcials, another lesson learned is to rely on the expertise of federal partners
such as CDC, the Occupational Safety and Health Administration, and state and local health
departments. For example, FSIS recently issued a notice for establishments (meat and poultry
plants) to follow the latest CDC guidance—wearing masks when FSIS personnel are present,
regardless of vaccination status, if located in an area of “substantial” or “high” community
COVID-19 transmission—by August 9, 2021.423 According to FSIS, if an establishment does not
meet these expectations, the agency may take additional measures to reduce COVID-19 exposure
for FSIS plant personnel, such as not providing inspection service to the establishment until it
follows CDC guidance.

Methodology
To conduct this work, we reviewed the most recent USDA data on COVID-19 illnesses and
exposures among FSIS employees available as of August 23, 2021; the CARES Act; agency

422See U.S. Department of Agriculture, USDA COVID-19 Workplace Safety Plan (July 28, 2021, and Aug. 23,
2021), accessed August 6, 2021, and August 27, 2021, respectively, https://www.usda.gov/coronavirus/
workplacesafetyplan. FSIS supplemented the USDA COVID-19 Workplace Safety Plan with an appendix for FSIS-speciﬁc
guidelines. For example, FSIS frontline employees are required to wear face shields in addition to face masks.
423See U.S. Department of Agriculture, The Food Safety and Inspection Service, Measures to Protect FSIS Inspection
Program Personnel (IPP) From COVID-19 Infection, FSIS Notice 30-21 (Washington, D.C.: Aug. 4, 2021). FSIS advised
establishment managers and employees to use the CDC COVID Data Tracker website, which is updated daily, to
determine whether they are located in an area of “substantial” or “high” community COVID-19 transmission.

Page 250 GAO-22-105051
policy and other guidance; USDA expenditure data as of July 23, 2021; and written responses
to questions we emailed FSIS oﬃcials. We assessed the reliability of agency data by reviewing
relevant USDA and FSIS documents, our prior use of the data sources, and written responses from
the agency about the data. We determined the data were suﬃciently reliable for estimating the
number of FSIS staﬀ who had become ill with COVID-19.

Agency Comments
We provided a draft of this enclosure to USDA and the Oﬃce of Management and Budget
for review and comment. USDA and the Oﬃce of Management and Budget did not have any
comments related to this enclosure.

GAO’s Ongoing Work
We plan to assess federal eﬀorts to protect worker safety and health in meat and poultry plants
during the COVID-19 pandemic in ongoing work.

Contact information: Steve D. Morris, (202) 512-3841, morriss@gao.gov

Page 251 GAO-22-105051
Airport Grants
The Federal Aviation Administration is administering grants to help the nation’s airports respond
to and recover from the economic eﬀects of the COVID-19 pandemic.

Entity involved: Federal Aviation Administration, within the Department of Transportation

Background
Historic decreases in passenger demand for air travel due to the COVID-19 pandemic signiﬁcantly
aﬀected U.S. airports’ abilities to generate the revenue needed for operating and infrastructure
costs. According to data ﬁled with the Department of Transportation (DOT), U.S. airlines carried
about 47 percent fewer passengers in March 2021 than in March 2019. One airport association
estimates that U.S. airports will face $40 billion in operating losses and additional costs related
to COVID-19 from March 2020 to March 2022. Airport associations reported that passenger
traﬃc increased over the summer months of 2021, particularly at airports serving domestic
leisure travelers. However, airports serving more international or business travelers have not had
the same levels of recovery. Additionally, it is unclear how the delta variant, or other COVID-19
variants, will aﬀect air travel demand in the fall and winter months of 2021. The CARES Act, the
Consolidated Appropriations Act, 2021, and the American Rescue Plan Act of 2021 provided a
combined total of $20 billion in federal funding for U.S. airports to respond to the COVID-19
pandemic, although funding allocation and certain allowable uses diﬀer under each act.

Obligations and expenditures. Of the $20 billion combined total in federal COVID-19 relief
funding provided by the CARES Act, Consolidated Appropriations Act, 2021, and the American
Rescue Plan Act of 2021 for U.S. airports, about $16 billion has been obligated and $8 billion
expended by FAA as of September 27, 2021, according to FAA oﬃcials.

CARES Act. The CARES Act, signed into law on March 27, 2020, provided $10 billion to support U.S.
airports of all sizes to prevent, prepare for, and respond to the COVID-19 pandemic.424 Airport
owners—also known as airport sponsors—may use CARES Act funds for any purpose for which
airport revenues may be lawfully used, including for airport operating expenses and debt service.
As of September 27, 2021, the Federal Aviation Administration (FAA) had obligated over $9 billion
and expended over $7 billion to reimburse airports for eligible costs and to increase the federal
share for 2020 Airport Improvement Program (AIP) grants, according to FAA oﬃcials (see table). As
of September 27, 2021, FAA had processed CARES Act grant applications for 3,229 U.S. airports.

424Pub. L. No. 116-136, 134 Stat. 281, 596-97. The CARES Act gives the Federal Aviation Administration (FAA) the
authority to retain up to 0.1 percent of the $10 billion (up to $10 million) provided for Grants-in-Aid for Airports to fund
the award and oversight by FAA of grants made under the CARES Act.

Page 252 GAO-22-105051
FAA Obligations and Expenditures for CARES Act Airport Grants, as of September 27, 2021
Obligations Expenditures
Funding group ($ thousands) ($ thousands)

Increase federal share for 2020 Airport Improvement $548,325 $381,345
a
Program (AIP) grants

Commercial service airports
b $7,210,565 $5,901,947

Primary airports
c $1,635,766 $933,169

General aviation airports
d $100,480 $76,162

Reallocated CARES Act funds
e $290,775 $131,683

Total $9,785,911 $7,424,305

Source: GAO analysis of CARES Act, Pub. L. No. 116-136, 134 Stat. 281 (2020) and data from the Federal Aviation Administration (FAA). | GAO-22-105051

a
The CARES Act directed FAA to allocate funding to these groups through various formulas. Approximately 3,300 airports in
the U.S. are part of the national airport system and are eligible to receive federal AIP grants to fund infrastructure projects.
The CARES Act appropriated at least $500 million to increase the federal share to 100 percent for grants awarded for airport
infrastructure projects under ﬁscal year 2020 and supplemental discretionary grants.
b
Commercial service airports are publicly owned airports with at least 2,500 passenger boardings per year and scheduled air
service.
c
Primary airports are large, medium, and small hub and non-hub airports with more than 10,000 passenger boardings per year.
d
General aviation airports are public-use airports with fewer than 2,500 passenger boardings per year and no scheduled air
service.
e
Under the Consolidated Appropriations Act, 2021, unallocated CARES funds as of December 27, 2020 were to be allocated
under the Consolidated Appropriations Act, 2021 using the primary commercial service and certain cargo airports allocation
formula. According to FAA oﬃcials, FAA calculated that $290,774,557 in CARES Act funds are available for reallocation under the
Consolidated Appropriations Act, 2021.

Consolidated Appropriations Act, 2021. The Consolidated Appropriations Act, 2021, enacted
on December 27, 2020, provided $2 billion in additional federal aid to help eligible airports
and certain tenants to prevent, prepare for, and respond to COVID-19.425 Airports must use
Consolidated Appropriations Act, 2021 grant funding for costs related to operations, personnel,
cleaning, sanitization, janitorial services, combating the spread of pathogens at the airport, and
debt service payments.426 Certain amounts are also available to provide relief from rent and
minimum annual guarantees to airport concessions. As of September 27, 2021, FAA had obligated
about $2 billion and expended about $570 million to reimburse airports for eligible costs,

425Pub. L. No. 116-260, div. M, tit. IV, 134 Stat. 1182, 1939-41 (2020). Division M of the Consolidated Appropriations
Act, 2021 is also referred to as the Coronavirus Response and Relief Supplemental Appropriations Act, 2021 (CRRSA).
The Consolidated Appropriations Act, 2021, made up to $5 million of the $2 billion in funding available for the Small
Community Air Service Development Program.
426The 31 airports that received CARES Act funds in excess of four times their annual operating expenses are excluded
from receiving Consolidated Appropriations Act, 2021 funding. See Pub. L. No. 116-260, 134 Stat. at 1939. Speciﬁc
amounts appropriated by the Consolidated Appropriations Act, 2021 are also available to cover lawful expenses to
support FAA contract tower operations and to provide relief from rent and minimum annual guarantees to on-airport car
rental, on-airport parking, and in-terminal concessions.

Page 253 GAO-22-105051
according to FAA oﬃcials (see table). As of September 27, 2021, FAA had processed Consolidated
Appropriations Act, 2021 grant applications for 2,966 U.S. airports.

Page 254 GAO-22-105051
FAA Obligations and Expenditures for the Consolidated Appropriations Act, 2021 Airport Grants, as of September
27, 2021
Obligations Expenditures
Funding group ($ thousands) ($ thousands)

Primary commercial service airports and certain cargo airports
a $1,751,347 $542,290

Non-primary commercial service and general aviation airports
b $38,854 $8,550

Non-primary airports participating in the Federal Aviation $4,783 $660
c
Administration (FAA) Contract Tower program

Tenant relief for primary commercial service airports $199,416 $15,267

Small Community Air Service Development Program
d $4,000 $4,000

Total $1,998,400 $570,767

Source: GAO analysis of the Consolidated Appropriations Act, 2021, Pub. L. No. 116-260, div. M, tit. IV, 134 Stat. 1182, 1939-41 (2020), and data from the Federal Aviation
Administration (FAA). | GAO-22-105051

a
The Consolidated Appropriations Act, 2021 directed FAA to allocate funding to these groups through various formulas. Primary
commercial service airports are publicly owned airports with more than 10,000 passenger boardings per year and scheduled air
service. Cargo airports are airports that, in addition to any other air transportation services that may be available, are served by
aircraft providing air transportation of only cargo with a total annual landed weight of more than 100 million pounds.
b
Non-primary commercial service airports have at least 2,500 and no more than 10,000 passenger boardings each year.
General aviation airports are public-use airports with fewer than 2,500 passenger boardings per year or no scheduled air
service.
c
Non-primary airports are the same as non-primary commercial service airports, and have at least 2,500 and no more than
10,000 passenger boardings each year. The Federal Contract Tower Program, established in 1982, allows the agency to contract
out the operation of certain towers.
d
The Small Community Air Service Development Program is a grant program designed to help small communities address air
service and airfare issues.

American Rescue Plan Act of 2021. The American Rescue Plan Act of 2021, enacted on March 11,
2021, provided an additional $8 billion for airport assistance.427 The allowable uses of funds
are similar to those for the Consolidated Appropriations Act, 2021, and are available for costs
related to operations, personnel, cleaning, sanitization, janitorial services, combating the spread
of pathogens at the airport, and debt service payments. Certain amounts are available to increase
the federal share for AIP grants, and to provide relief from rent and minimum annual guarantees
to airport concessions. As of September 27, 2021, FAA had obligated over $4 billion and expended
about $300 million to reimburse airports for eligible costs, according to FAA oﬃcials (see table). As
of September 27, 2021, FAA had processed American Rescue Plan Act of 2021 grant applications
for 1,147 U.S. airports.

427Pub. L. No. 117-2, § 7102(a), 135 Stat. 4, 96.

Page 255 GAO-22-105051
FAA Obligations and Expenditures for the American Rescue Plan Act of 2021 Airport Grants, as of September 27,
2021
Obligations Expenditures
Funding group ($ thousands) ($ thousands)

Primary commercial service airports and certain cargo $3,814,655 $215,509
a
airports

Increase federal share for 2021 and select 2020 Airport $486,531 $82,933
Improvement Program (AIP) grants

Non-primary commercial service and general aviation $39,206 $2,292
b
airports

Tenant relief for primary commercial service airports
c $0 $0

Total $4,340,391 $300,734

Source: GAO analysis of the American Rescue Plan Act, 2021, Pub. L. No. 117-2, § 7102, 135 Stat. 4, 96-98, and data from the Federal Aviation Administration (FAA). | GAO-22-105051

a
The American Rescue Plan Act of 2021 directed FAA to allocate funding to these groups through various formulas. Primary
commercial service airports are publicly owned airports with more than 10,000 passenger boardings per year and scheduled air
service. Cargo airports are airports that, in addition to any other air transportation services that may be available, are served by
aircraft providing air transportation of only cargo with a total annual landed weight of more than 100 million pounds.
b
Non-primary commercial service airports have at least 2,500 and no more than 10,000 passenger boardings each year.
General aviation airports are public-use airports with fewer than 2,500 passenger boardings per year or no scheduled air
service.
c
Primary commercial airport sponsors may only use these funds to provide relief from rent and minimum annual guarantees
to in-terminal airport tenants, subject to additional conditions. An eligible large airport concession is one that is in-terminal and
has maximum gross receipts, averaged over the previous three ﬁscal years, of more than $56,420,000. An eligible small airport
concession is one that is in-terminal and is a small business with maximum gross receipts, averaged over the previous three
ﬁscal years, of less than $56,420,000, and is a joint venture.

Overview of Key Issues
Airport grant administration, challenges, and oversight. With regard to COVID-19 relief
funding appropriated by the CARES Act, Consolidated Appropriations Act, 2021, and the American
Rescue Plan Act of 2021, FAA is processing airport grant applications, obligating funds, and
reviewing invoices to reimburse airport sponsors. FAA has also provided guidance on airport grant
requirements for these programs, including for workforce retention and tenant relief, which we
discuss further below.

As we previously reported, FAA has identiﬁed challenges to administering CARES Act airport
grants, including the need to process grants for over 3,000 airport sponsors under expedited
time frames, with expanded eligible uses for these funds. To address the increased workload
of processing and monitoring three new airport relief grant programs, FAA has established a
dedicated team, including two full-time employees and three annuitants with prior airport grant
management experience, to review and process airport payment requests. Since we last reported,
FAA identiﬁed additional workload challenges related to administering the airport tenant relief
portions of the Consolidated Appropriations Act, 2021 and the American Rescue Plan Act of 2021.
Airport association representatives also noted that FAA was slow to administer and implement

Page 256 GAO-22-105051
the American Rescue Plan Act of 2021 in the spring and summer of 2021, in part because FAA was
administering other airport grant programs at the same time. According to airport association
representatives, delays and uncertainty around grant rollout for the American Rescue Plan Act
of 2021 funds have created challenges for airports’ ﬁnancial planning eﬀorts. To address some
of these challenges, FAA reported that it has detailed additional personnel to assist in reviewing
concessions rent relief plans, which we will discuss in further detail below under airport tenant
relief.

With regard to monitoring and oversight, FAA oﬃcials also reported that the agency hired a
contractor in the fall of 2020 to review FAA’s reimbursement processes for CARES Act grants
and provide recommendations on auditing policies and procedures. FAA oﬃcials said that
the contractor had completed its review of FAA’s reimbursement processes for CARES Act
funding, and issued several recommendations for improvement.428 Speciﬁcally, the contractor
recommended that FAA (1) provide airport sponsors with job aids (including a guide and checklist)
on how to submit reimbursement requests to ensure completeness and consistency, (2) create
a centralized document to improve reviews of reimbursement requests, and (3) standardize how
the agency communicates its decisions to airports. FAA oﬃcials reported that FAA has taken steps
to streamline its reimbursement process and publish updated guidance on its CARES Act website
for airports, and will continue to improve its processes based on these recommendations. FAA
oﬃcials reported that they have hired the same contractor to conduct reviews of the Consolidated
Appropriations Act, 2021 funds.429 FAA is also currently in the process of searching for and hiring a
contractor to oversee the American Rescue Plan Act of 2021 funds.

As non federal entities, public airport sponsors that receive federal grants are also subject to
the Single Audit Act, and they must undergo a single audit of those awards annually when their
expenditures meet a certain dollar threshold—currently $750,000 or more in a ﬁscal year.430 Single

428FAA oﬃcials also reported that they had hired the contractor to develop an electronic dashboard to monitor and
track funds. However, FAA oﬃcials said that they issued a stop work order for the dashboard when they determined that
the contractor did not have the technical expertise necessary to create it. Similarly, FAA determined that the tracker the
contractor created to monitor CARES Act audits was not eﬃcient, because it required too many steps to use.
429Speciﬁcally, FAA oﬃcials reported that the contractor will assess whether FAA is properly identifying and addressing
any issues in a timely manner, and will conduct sampling and testing of payment requests.
430The Single Audit Act is codiﬁed, as amended, at 31 U.S.C. §§ 7501-06, and implementing Oﬃce of Management and
Budget guidance is reprinted in 2 C.F.R. part 200. The Single Audit Act establishes requirements for non federal entities
(deﬁned as states, localities, and nonproﬁt organizations) that receive federal awards to undergo audits of those awards
annually (unless a speciﬁc exception applies) when their expenditures meet a certain dollar threshold. More speciﬁcally,
nonfederal entities that expend $750,000 or more in federal awards in a ﬁscal year are required to undergo a single
audit—that is, an audit of an entity’s ﬁnancial statements and federal awards, or a program-speciﬁc audit, for the ﬁscal
year. 31 U.S.C. § 7502; 2 C.F.R. § 200.501. If public airport sponsors do not meet this threshold in a ﬁscal year, then
they are subject to applicable local and state audit requirements. Because private owners of public-use airports are
not non federal entities as deﬁned by the Single Audit Act, private owners of public-use airports are not subject to the
Single Audit Act; however, they are still subject to grant assurances that they agree to when they accept federal airport
grants. For example, under one grant assurance, a private owner of a public-use airport would agree to keep all project
accounts and records which fully disclose the amount and disposition by the recipient of the proceeds of this grant, the
total cost of the project in connection with which this grant is given or used, and the amount or nature of that portion of
the cost of the project supplied by other sources, and such other ﬁnancial records pertinent to the project. Additionally,
private owners of public-use airports must make any documents pertinent to a grant available to the Comptroller
General for the purpose of audit and examination. Finally, DOT may require that an appropriate audit be conducted by a
recipient.

Page 257 GAO-22-105051
audits of an entity’s ﬁnancial statements and federal awards can help identify deﬁciencies in an
award recipients’ compliance with applicable laws and regulations, help ensure the appropriate
use of federal funds, and reduce the likelihood of federal improper payments. FAA oﬃcials stated
that, in accordance with Single Audit Act requirements, the agency notiﬁes airport sponsors if the
amount of their expenditures triggers the $750,000 threshold. The airport sponsor must then
conﬁrm annually that the audit was completed and uploaded to the Federal Audit Clearinghouse.
FAA oﬃcials explained that airport sponsors’ independent auditors conduct the Single Audits,
but FAA often provides technical assistance about CARES Act grant expenditures. If the audits
identify a ﬁnding, FAA then follows up with the airport to ensure they have taken corrective action.
Additionally, the DOT Oﬃce of Inspector General may check compliance and review single audit
reports for DOT fund recipients, and is currently reviewing DOT’s processes for verifying that these
audits have been completed, among other things.

Airport grant funding uses. Airport association representatives told us that the federal funding
provided has been critical. FAA has begun to collect and consolidate data from airports on general
spending categories for CARES Act funding through grant close-out reports, which are completed
once all allocated airport funds have been expended. As of September 27, 2021, FAA oﬃcials
said that 765 CARES Act airport grants, totaling $2.47 billion, have been closed out. For these
grants, the majority of airport grant funds have been used for payroll and debt servicing. As
for Consolidated Appropriations Act, 2021 and American Rescue Plan Act of 2021 funding, FAA
oﬃcials reported that they have not yet begun to close out these grants as of September 2021.
However, airport associations said that airport sponsors are generally using these grants to pay
down debt and pay for operating expenses such as payroll.

Workforce retention requirements and monitoring. Certain airport sponsors accepting CARES
Act grant funds were required to continue to directly employ at least 90 percent of the number
of employees onboard as of March 27, 2020, through the end of the calendar year 2020.431 The
Consolidated Appropriations Act, 2021 extended these same workforce retention requirements
through February 15, 2021. Airports that accept American Rescue Plan Act of 2021 grants were
subject to the same workforce retention requirements through September 30, 2021.432 According
to FAA, the 131 largest U.S. airports were subject to this requirement under the Consolidated
Appropriations Act, 2021.

431Certain airport sponsors that are subject to workforce retention requirements include large, medium, and small-
hub airports. As such, airports that were not classiﬁed as large, medium, or small-hub airports were exempt from
this requirement. More speciﬁcally, non-hub and non-primary airports are excluded from the workforce retention
requirement. As a result, non-hub primary commercial service airports (airports with more than 10,000 annual
passenger boardings, but less than 0.05 percent of total annual passenger boardings); non-primary commercial service
airports (airports with at least 2,500 and no more than 10,000 passenger boardings each year); general aviation airports
(public-use airports that do not have scheduled service or have scheduled service with fewer than 2,500 passenger
boardings each year); and reliever airports (airports designated by FAA to relieve congestion at commercial service
airports) are all exempt from the workforce retention requirement. Airports were required to retain 90 percent of full-
time equivalent employees working at the airport as of March 27, 2020, as the baseline comparison. According to
FAA guidance, airport sponsors did not need to count contractors providing services other than airport management,
tenants, or concessionaires. Airport sponsors may make adjustments for retirements or voluntary employee separations
when calculating the workforce retention percentage.
432As with the CARES Act and the Consolidated Appropriations Act, 2021, DOT could waive the workforce retention
requirement under the American Rescue Plan Act of 2021, if DOT determined that the airport was experiencing
economic hardship as a direct result of the requirement, or the requirement reduces aviation safety or security.

Page 258 GAO-22-105051
FAA oﬃcials said at the time that airport sponsors execute a COVID-19 relief grant, sponsors certify
that they will meet the workforce retention requirements and report their employee numbers.
Since we last reported in July 2021, FAA has continued to monitor compliance with workforce
retention requirements for the 131 aﬀected airports. More speciﬁcally, FAA oﬃcials stated that
they receive workforce retention information from airports through a dedicated CARES Act email
and track these employee numbers, comparing them to the March 27, 2020 baseline data, to
ensure airports are meeting the 90-percent threshold. Airport associations told us that airports
have not faced challenges meeting these workforce retention requirements.

Airport tenant relief. As part of the Consolidated Appropriations Act, 2021 requirements,
airport sponsors that accept tenant relief funds will waive rent and minimum annual guarantee
obligations for eligible airport tenants beginning December 27, 2020, until the relief equals the
total tenant relief allocation amount and to the extent permissible under state and local laws.433
Eligible airport tenants include on-airport car rental and parking as well as in-terminal concession
tenants.

To administer the tenant relief portion of the Consolidated Appropriations Act, 2021 funding,
FAA oﬃcials have calculated airport allocations and informed airport sponsors of the amount
of eligible concessions relief, including funds they can use for administration costs.434 FAA is
requesting that airport sponsors provide tenant relief plans with their payment requests when
they are ready to accept their concessions-relief funding. FAA oﬃcials said they are reviewing these
plans to ensure that airport sponsors are providing relief according to the law and FAA guidance
prior to approving payment. According to FAA guidance, FAA requires airport tenants to provide
certiﬁcations of eligibility directly to airport sponsors, who then keep the documentation on ﬁle
for possible audits. As previously mentioned, FAA oﬃcials said that monitoring concessions relief
funding and understanding the contractual structures between an airport and its concessionaires
are new scopes of work for FAA’s Oﬃce of Airports and thus, FAA has detailed personnel from its
Oﬃce of Civil Rights to review concessions relief plans and train other FAA staﬀ on how to conduct
these reviews.

Airport representatives noted that determining tenant eligibility and how to provide
concessionaire relief on a proportional basis under the Consolidated Appropriations Act, 2021
can be complex due to varying tenant agreements. Airport association representatives stated,
however, that FAA has mostly addressed these challenges related to airports’ roles in determining
tenant eligibility and providing relief. For example, FAA has provided guidance on how to
administer concession relief through the FAQs posted to FAA’s website, which the agency updated
with additional clariﬁcations in April 2021. FAA has also held videoconferences to answer airports’
tenant relief questions, and has established a dedicated email for airports to direct questions to
FAA headquarters. Airport association representatives reported that airports have been able to get
their questions answered through FAA’s dedicated email.

433Both the Consolidated Appropriations Act, 2021, and the American Rescue Plan Act of 2021 direct airport sponsors to
provide such relief to the extent permissible under state laws, local laws, and applicable trust indentures.
434Airport tenant relief allocation amounts were calculated based on each airport’s passenger boardings compared to
total passenger boardings of all airports eligible for concessions relief for calendar year 2019.

Page 259 GAO-22-105051
As previously mentioned, the American Rescue Plan Act of 2021 also provides in-terminal
concession tenant relief funding for eligible airport sponsors. FAA oﬃcials reported that they
would not begin processing these tenant relief applications until after October 1, 2021, however,
because of their need to manage ﬁscal year-end operations. Representatives from an airport
association and an airport concessionaires association noted, however, that airports continue
to have questions about the tenant relief portion of the American Rescue Plan Act of 2021.
Speciﬁcally, the concessionaires association noted that some airports are unclear about the
eligibility requirements for concessions that operate through various business models, such as
joint ventures, and whether they would qualify as large or small concessions. To help address
some of these questions, FAA oﬃcials stated that they are considering updating guidance to
airports on this tenant relief funding based on feedback from airports and lessons learned from
administering the Consolidated Appropriations Act, 2021 tenant relief funds.

An airport concessionaires association told us although airport traﬃc has begun to recover, the
dedicated tenant relief continues to be important to help airport businesses respond to the
impacts of COVID-19. They noted that although many concessionaires have started to re-open
and rehire, these businesses are far from returning to full-service and staﬃng. In particular, these
concessionaires face increasing challenges due to the uncertainty around COVID-19’s eﬀect on
passenger traﬃc in the fall and winter of 2021, and diﬃculty recruiting and hiring staﬀ during the
pandemic.

Methodology
To conduct this work, we analyzed FAA data on airport funding as of September 27, 2021. We
determined the data were suﬃciently reliable for the purposes of our reporting by performing
interviews with agency oﬃcials and reviewing relevant documentation. We also reviewed the
CARES Act, the Consolidated Appropriations Act, 2021, and American Rescue Plan Act of 2021 and
related agency guidance, and conducted interviews with agency oﬃcials and representatives from
airport associations, and an airport concessions association, selected to represent a wide variety
of industry and airport types.

Agency Comments
We provided FAA and the Oﬃce of Management and Budget (OMB) with a draft of this enclosure.
FAA provided technical comments that we incorporated as appropriate. OMB did not have any
comments on this enclosure.

GAO’s Ongoing Work
Our work on aviation industry COVID-19 ﬁnancial assistance is ongoing. We will continue to
monitor FAA’s administration of grants under the CARES Act, the Consolidated Appropriations
Act, 2021, and the American Rescue Plan Act of 2021. We will also continue to monitor aviation
operations, impacts, and lessons from the COVID-19 pandemic through other ongoing work.

Page 260 GAO-22-105051
Contact information: Heather Krause, (202) 512-2834 or krauseh@gao.gov

Page 261 GAO-22-105051
Payroll Support Assistance to Aviation Businesses
The Department of the Treasury has nearly completed making payroll assistance payments to the
aviation industry; however, it has not developed policies and procedures to ensure that warrants,
which give the federal government the ability to purchase stock in the companies of certain
program recipients, are acted upon in a manner that will provide appropriate compensation to the
federal government.

Entity Involved: Department of the Treasury

Recommendations for Executive Action
The Secretary of the Treasury should develop policies and procedures to determine when to act on
warrants obtained as part of the Payroll Support Program to provide appropriate compensation to
the federal government.

Treasury agreed with our recommendation and said the agency is in the process of creating a
policy to dispose of the warrants obtained as part of the Payroll Support Program.

Background
The CARES Act established the Payroll Support Program (PSP1) in March 2020, and the
Consolidated Appropriations Act, 2021 established a Payroll Support Program Extension (PSP2)
in December 2020.435 In March 2021, the American Rescue Plan Act of 2021 created a new round
of the program (PSP3).436 PSP1 included up to $32 billion in ﬁnancial assistance for passenger
air carriers, cargo air carriers, and aviation contractors. PSP2 provided up to $16 billion, and
PSP3 provided up to $15 billion in ﬁnancial assistance for passenger air carriers and aviation
contractors.

Payments from PSP1, PSP2, and PSP3 were to be used exclusively for the continuation of wages,
salaries, and beneﬁts. To help ensure recipients’ compliance with the Payroll Support Program
agreement terms, Treasury developed a program agreement that includes various compliance
requirements, and Treasury also developed a compliance monitoring approach with two levels of
testing. Treasury also requires that some recipients provide taxpayer protections through notes
and/or warrants.437 Speciﬁcally, for PSP1, PSP2, and PSP3 recipients, Treasury requires passenger
air carriers that received payments of more than $100 million and contractors that received more

435CARES Act, Pub. L. No. 116-136, § 4112, 134 Stat. 281, 498 (2020) (codiﬁed at 15 U.S.C. § 9072); Pub. L. No. 116-260,
div. N, tit. IV, § 402, 134 Stat. 1182, 2052-61 (2020).
436Pub. L. No. 117-2, § 7301, 135 Stat. 4, 104-107.
437The Secretary has the authority to receive from Payroll Support Program recipients notes and warrants, or other
ﬁnancial instruments that the Secretary determines provide appropriate compensation to the federal government
for providing ﬁnancial assistance. Notes are ﬁnancial instruments whose value is a percentage of the payroll support
provided over a certain threshold. Notes are a form of indebtedness, and notes must be repaid by recipients. Warrants
are an option to buy shares of stock at a predetermined price before a speciﬁed date.

Page 262 GAO-22-105051
than $37.5 million to provide warrants and/or notes as taxpayer protection.438 Further, Treasury
required cargo carriers that received payments through PSP1 to provide notes equal to 56 percent
of the payroll support over $50 million. The notes have a 10-year term, and the warrants must be
exercised by Treasury within 5 years.

Passenger air carriers are experiencing rising demand for domestic leisure travel; however, it
remains uncertain how passenger demand will respond as the pandemic continues. However,
bookings for domestic business travel will likely return at a slower pace, and between 10 to 30
percent of this business travel may not return after the pandemic, according to a ratings agency.
International travel is also expected to return at a slower pace than domestic leisure travel,
according to a consulting ﬁrm’s projections, with international travel not expected to return
to prepandemic levels until 2023 or 2024. Further, the aviation industry is expected to incur
operating losses through the end of 2021, although these losses are likely to be less than the
losses the industry experienced in 2020, according to a ratings agency.

Overview of Key Issues
Treasury made $59 billion in payments out of $63 billion provided for PSP1, PSP2, and PSP3.
As of September 2021, across PSP1, PSP2, and PSP3, Treasury had made payments of $50.1 billion
to the 10 largest passenger air carriers—the remaining passenger air carriers received $3.4 billion.
Aviation contractors received $4.6 billion across PSP1, PSP2, and PSP3. Cargo carriers received
$827.7 million through PSP1.439 Through PSP1, PSP2, and PSP3, Treasury provided payments
to 402 companies—254 passenger air carriers and 148 contractors—as of September 2021.
See ﬁgure below for additional information on the pace and amount of PSP1, PSP2, and PSP3
payments.

438In some cases, for recipients that were related but independently operated, Treasury required a weighted threshold
for providing notes. As a result, some recipients that did not meet the $37.5 million threshold provided notes.
439The CARES Act provided funding for cargo air carriers for PSP1, whereas the Consolidated Appropriations Act, 2021
and the American Rescue Plan Act of 2021 did not provide funding for cargo air carriers for PSP2 and PSP3, respectively.
Of the $4 billion appropriated for cargo air carriers in PSP1, $3.2 billion was not awarded because total demand by cargo
air carriers for PSP1 funds was far below available funds.

Page 263 GAO-22-105051
Payment Amount and Number of Agreements Signed for the Payroll Support Program (PSP1), Payroll Support
Program Extension (PSP2), and Second Payroll Support Program Extension (PSP3), as of September 2021

Treasury has nearly completed making payments for PSP3, and as of September 2021, has
administered 98 percent of PSP3 funds. As of September 2021, Treasury has administered $14.7
billion out of $15 billion available under PSP3. As directed by statute, only recipients that received
PSP2 payments were eligible for PSP3 payments.440 Treasury has made PSP3 payments to 462
recipients—287 passenger air carriers and 175 contractors. According to Treasury oﬃcials, the
remaining PSP3 agreements will be signed and funds released once PSP2 applicants have resolved
outstanding issues. For passenger air carriers, the 10 largest payments averaged $1.3 billion,
and the remaining passenger air carrier payments averaged $3.1 million. For contractors, the 10
largest payments averaged $50.4 million, and the remaining contractor payments averaged $2.3
million.

As we reported in July 2021, according to Treasury oﬃcials, passenger air carriers received PSP3
payments equal to 93 percent of their PSP2 payment and contractors received PSP3 payments
equal to their PSP2 payment.441 Three selected industry associations and two companies we
interviewed said that not having to complete an application for PSP3 streamlined the process
compared to previous iterations of the program.

440Pub. L. No. 117-2, § 7301, 135 Stat. at 104-05.
441Due to the high demand for the program and as authorized in statute, PSP2 payments provided by Treasury were
46 percent of the total approved payment amount for passenger air carriers and 30.4 percent of the total approved
payment amount for contractors.

Page 264 GAO-22-105051
Across PSP1, PSP2 and PSP3, Treasury required warrants from 14 recipients and notes
from 36 recipients; however, Treasury does not have policies and procedures for when to
act upon these warrants. By statute, Treasury may receive warrants, notes, or other ﬁnancial
instruments which the Secretary determines provide appropriate compensation to the federal
government for provision of ﬁnancial assistance.442 As detailed in Treasury guidance and on
Treasury’s website, Treasury requires that recipients of Payroll Support Program payments over
a certain dollar amount provide notes and/or warrants to Treasury.443 Speciﬁcally, passenger
air carriers receiving over $100 million of payroll support are required to provide notes equal to
30 percent of the payroll support provided over $100 million with a 10-year term. Contractors
receiving over $37.5 million of payroll support must provide notes equal to 44 percent of the
payroll support provided over $37.5 million with a 10-year term.444 Payroll Support Program
recipients that received payroll support over the dollar amounts listed above must also provide
warrants, if they are publicly traded companies. The number of warrants a recipient is required
to provide is equal to 10 percent of the principal amount of the note issued by the participant,
divided by an exercise price.445 In total, across PSP1, PSP2, and PSP3, the note principal amount
that must be paid by 36 recipients is $15.1 billion, and 14 recipients provided a total of 58 million
warrants. See table below for additional details.

442Pub. L. No. 116-136, § 4117, 134 Stat. at 500-01; Pub. L. No. 116-260, § 408, 134 Stat. at 2059; Pub. L. No. 117-2, §
7301(b)(4), 135 Stat. at 107.
443The CARES Act, Consolidated Appropriations Act, 2021, and the American Rescue Plan Act of 2021 authorized
Treasury to receive warrants and other ﬁnancial instruments from recipients.
444According to Treasury oﬃcials, Treasury is taking a measured approach to maximizing both the intrinsic and time
value of the notes. The notes do not mature until the 2030-2031 timeframe, and Treasury currently has no plans to sell
these notes and expects that the notes will be repaid in full on or before maturity.
445For passenger air carriers that received ﬁnancial support through PSP1, the exercise price is generally equal to the
value of the shares as of market close on April 9, 2020. For cargo air carriers that received ﬁnancial support through
PSP1, the exercise price is equal to the market value of the shares as of market close on May 1, 2020. For PSP2, the
exercise price is generally equal to the value of the shares as of market close on December 24, 2020. For PSP3, the
exercise price is generally equal to the value of the shares as of market close on March 10, 2021. In some instances, a
recipient provides a note, but does not provide a warrant.

Page 265 GAO-22-105051
Information about Notes and Warrants Provided to the Department of the Treasury for the Payroll Support
Program (PSP1), Payroll Support Program Extension (PSP2), and Second Payroll Support Program Extension
(PSP3), as of September 2021
Number of Total note principal Number of payments Total number of
payments that amount that required warrants
required notes (in millions) warrants (in millions)

PSP1

Passenger air carriers 13 $6,757.4 13 33.4

Cargo air carriers 2 $262.2 1 0.6

Contractors 19 $530.0 0 N/A

PSP2

Passenger air carriers 12 $3,869.3 12 14.1

Contractors 8 $75.9 0 N/A

PSP3

Passenger air carriers 11 $3,568.1 11 9.8

Contractors 8 $75.9 0 N/A

Source: GAO analysis of Department of the Treasury data. | GAO-22-105051

According to Oﬃce of Management and Budget Circular A-129, which provides guidance on
managing federal loan programs and other ﬁnancial assets, agencies are required to establish
appropriate controls over programmatic functions and operations. This guidance also requires
management to employ a reporting framework to inform proactive portfolio management.

Treasury has not exercised any of the warrants for common stock it holds in the 14 businesses
mentioned above. As Treasury continues to hold these warrants for stock purchases, the warrants
may increase in value as the airline industry recovers. For example, based on the current stock
price at market close on October 1, 2021, its warrants from one recipient would be valued at 159
percent above the predetermined price Treasury would have to pay to act on them.

However, Treasury has not documented policies and procedures to guide when to act on the
warrants to fulﬁll the statutory purpose to provide appropriate compensation to the federal
government. As we reported in 2010, acquiring an ownership interest in private companies can
help protect taxpayers by enabling the government to earn returns when it sells its shares and the
institutions repurchase their shares or redeem their warrants. According to Treasury oﬃcials, it is
likely that—if the airline industry continues to recover—the Payroll Support Program, through the
exercising of these warrants, could end up with a proﬁt. According to Treasury oﬃcials, as these
warrants do not mature until the 2024-2026 timeframe, Treasury is taking a measured approach
to disposing of the warrants. By establishing policies and procedures over warrant processes,
Treasury will help ensure that management has the information needed for informed decision

Page 266 GAO-22-105051
making on when to act on the warrants it holds, thereby maximizing the beneﬁt to the federal
government.

Treasury’s compliance monitoring approach is in progress, and some recipients may
undergo compliance monitoring for up to 10 years. Treasury developed a compliance
monitoring approach for PSP1 that involved two levels of compliance testing. All recipients
undergo the ﬁrst level of compliance testing, which is a set of automated testing rules applied
to each submitted compliance report. Speciﬁcally, at the end of each quarter, recipients must
submit a quarterly compliance report to Treasury via an online portal that Treasury has developed.
This report includes information about the amount of payroll support funds expended during
the quarter, ﬁnancial statements, a copy of the recipient’s IRS Form 941, and information about
changes in the number of employees and the amount spent on wages, salary, and beneﬁts, among
other things.

If an issue is identiﬁed during the ﬁrst level of compliance testing, the recipient is elevated to the
second level of compliance testing, which involves a more detailed review by a Treasury analyst.
Treasury oﬃcials said that they are using the same compliance monitoring approach for PSP2 and
PSP3, with some updates. As we previously reported, for PSP2 compliance monitoring, Treasury
developed a portal to assess recipients’ compliance with the employee recall requirements
included in the PSP2 agreement.446

As of July 2021, PSP1 compliance monitoring is underway, and according to Treasury oﬃcials, PSP1
compliance monitoring will end with the second quarter of 2022. For PSP2 recipients, Treasury
began conducting compliance monitoring in the ﬁrst quarter of 2021, and PSP2 compliance
monitoring will end with the ﬁrst quarter of 2023. Treasury oﬃcials said that they expect that PSP3
compliance monitoring will begin with the third quarter of 2021, and PSP3 compliance monitoring
will end with the second quarter of 2023. However, if a recipient received a PSP1, PSP2, or PSP3
payment that required a note or warrant, then the recipient can remain in the program for up
to 10 years, and will continue to undergo compliance testing. According to the Payroll Support
Program agreement, the recipient must submit quarterly compliance reports until no notes or
warrants are outstanding or held by Treasury.

According to Treasury documents, Treasury requires that when recipients submit a quarterly
compliance report, they certify their compliance with the terms and agreements of the Payroll
Support Program agreement and maintenance of eﬀective internal controls, as well as certify the
veracity and accuracy of any data, documents, or information submitted to Treasury. According
to Treasury oﬃcials, while Treasury is not required to validate that the information in the ﬁnancial
statements is accurate, Treasury monitors publicly reported ﬁnancial information for recipients
whose payment requires notes and warrants. Treasury oﬃcials said that monitoring publicly
reported ﬁnancial information is intended to help Treasury conﬁrm that ﬁnancial information
recipients submit as part of Treasury’s compliance monitoring approach is accurate.

446As we reported in July 2021, under the Consolidated Appropriations Act, 2021, PSP2 recipients were required to
recall any employees involuntarily furloughed by the PSP2 recipient between either October 1, 2020 (for recipients that
received PSP1 assistance), or March 27, 2020 (for recipients that did not receive PSP1 assistance), and the date such
passenger air carrier or contractor entered into a PSP2 agreement, among other requirements.

Page 267 GAO-22-105051
The Payroll Support Program agreement requires recipients to establish and maintain eﬀective
internal controls, among other things. Treasury oﬃcials said that recipients can develop a wide
range of internal controls, and that internal control systems diﬀer across recipients. Therefore,
Treasury relies on recipients’ certiﬁcation that their internal control systems are adequate and
appropriate. However, Treasury oﬃcials said that if they ﬁnd evidence that recipients do not
have eﬀective internal controls in place, they could do additional compliance testing and ask
the recipient to provide additional information. For example, Treasury oﬃcials said that if a
recipient reported that they used more funds than they had received from Treasury, indicating an
accounting or reporting weakness, Treasury would conduct additional compliance testing.

Industry associations said that Payroll Support Program payments provided critical
assistance to the industry. Five selected industry associations and three selected companies
we interviewed said that while challenges existed with the program, it provided critical ﬁnancial
assistance for the industry. For example, a member from one industry association said Payroll
Support Program payments helped create lasting jobs in three of its locations, and due to the
payments it received, the company did not have to lay oﬀ employees. Another member from the
same industry association said Payroll Support Program payments helped keep this companies’
employees on payroll and ensured that employees were up-to-date on training requirements. This
helped ensure that the industry could quickly restart, because the company did not have to hire
new employees.

However, in certain cases, increased demand for air travel has exceeded airlines’ ability to return
capacity to service, resulting in rescheduled and canceled ﬂights. Airlines have experienced
workforce shortages because they reduced their employment levels through voluntary measures,
such as early retirement, and involuntary measures, such as furloughs, and retraining and
recertifying some staﬀ can, in some cases, be a lengthy process.

Lessons learned from the implementation of the Payroll Support Program could inform
similar programs in the future. Our reviews of PSP1, PSP2, and PSP3 oﬀer some lessons learned
that Treasury and Congress could consider when establishing and implementing future federal
government relief programs.

Clear communication. Treasury’s communication with applicants about the Payroll Support
Program was not always clear. For example, as we reported in March 2021, recipients raised
concerns about the need for additional guidance. Further, four selected industry associations and
one company we interviewed said that Treasury did not provide status updates or timeframes
about the application process. Five selected industry associations and one company told us
that communication between Treasury and applicants—for example, the ability to get in touch
with Treasury staﬀ and get answers to time-sensitive questions—did not improve throughout
Treasury’s implementation of PSP1, PSP2, and PSP3. Clear communication from Treasury could
help ensure program transparency and keep applicants informed about program requirements
and timeframes.

Identifying which type of assistance would best help achieve a deﬁned goal. Treasury oﬃcials told
us that they implemented the Payroll Support Program by following the requirements laid out
in statute; however, oversight bodies and members of Congress continue to raise concerns that
certain actions taken by recipients may not serve the interests of taxpayers and aviation workers.

Page 268 GAO-22-105051
For example, speciﬁc concerns have been raised about businesses obtaining funds from more
than one federal ﬁnancial assistance program. However, the laws that authorized the three
rounds of the Payroll Support Program did not prohibit businesses from accessing other federal
ﬁnancial assistance. In contrast, new programs authorized in December 2020 and March 2021,
however, including the Coronavirus Economic Relief for Transportation Services program and
Aviation Manufacturing Jobs Protection program, do provide funding limitations in relation to
other COVID-19 relief funding received by eligible recipients. Additionally, as we have stated in
prior work, as Congress contemplates future assistance to aid the aviation industry’s recovery,
Congress may wish to consider how to best deﬁne goals and objectives for future assistance,
which may help determine which tools are needed and most appropriate to support an aviation
industry recovery following the pandemic.

Methodology
To conduct this work, we reviewed the CARES Act, the Consolidated Appropriations Act, 2021,
and the American Rescue Plan Act of 2021. We also reviewed Treasury guidance and documents,
and interviewed Treasury oﬃcials. We reviewed PSP1 data as of September 1, 2021, PSP2 data
as of September 8, 2021, and PSP3 data as of September 22, 2021. We reviewed these data for
outliers and missing information, and we determined that Treasury’s data were suﬃciently reliable
for the purposes of summarizing the number, value, and timing of Payroll Support Program
payments, as well as the number of warrants and the total note principal amount. In addition, we
conducted interviews with a selection of ﬁve companies and six industry associations representing
businesses eligible to apply for Payroll Support Program payments.

Agency Comments
We provided Treasury and the Oﬃce of Management and Budget (OMB) with a draft of this
enclosure. Treasury provided technical comments, which we incorporated as appropriate.
Treasury’s general comments are reproduced in appendix IX. In its management response,
Treasury agreed with our recommendation and said the agency is in the process of creating a
policy that will allow Treasury to evaluate when and how to act to dispose of the warrants obtained
as part of the Payroll Support Program. OMB did not provide comments on this enclosure.

GAO’s Ongoing Work
We will continue to monitor Treasury’s implementation of federal ﬁnancial assistance programs
for the transportation sector authorized by COVID-19 relief laws in December 2020 and March
2021.

GAO’s Prior Recommendations
The table below presents our recommendation on payroll support assistance for aviation
businesses from a prior bimonthly CARES Act report.

Page 269 GAO-22-105051
Prior GAO Recommendation Related to Payroll Support Assistance for Aviation Businesses
Recommendation Status

The Secretary of the Treasury should ﬁnish Closed-implemented. In April 2021, GAO conﬁrmed that
developing and implement a compliance monitoring the Department of the Treasury (Treasury) had developed,
plan that identiﬁes and responds to risks in the documented, and implemented a risk-based approach to monitor
Payroll Support Program (PSP) to ensure program PSP recipients’ compliance with the terms of the assistance.
integrity and address potential fraud, including Treasury’s risk-based approach entails a two level compliance
the use of funds for purposes other than for the review. In the ﬁrst level review, automated testing is conducted on
continuation of employee wages, salaries, and all recipients’ quarterly reports using factors/thresholds that can
beneﬁts (November 2020 report). trigger recipients being moved to the next review. In the second
level review, Treasury analysts conduct a more detailed review
of recipients that failed the ﬁrst level review or were selected
for other reasons. Treasury has also developed penalties and
a process for remediating noncompliance with PSP agreement
terms through Payroll Support Program agreements. As of April
2021, Treasury has identiﬁed noncompliance by recipients and
applied penalties, as appropriate.

Source: GAO. | GAO-22-105051

Related GAO Products
Financial Assistance: Ongoing Challenges and Guiding Principles Related to Government Assistance for
Private Sector Companies. GAO-10-719. Washington, D.C.: August 3, 2010.

Troubled Asset Relief Program: June 2009 Status of Eﬀorts to Address Transparency and Accountability
Issues. GAO-09-658. Washington, D.C.: June 17, 2009.

Contact Information: Heather Krause, (202) 512-2834, krauseh@gao.gov

Page 270 GAO-22-105051
Loans for Aviation and Other Eligible Businesses
As directed by the CARES Act, the Department of the Treasury required certain loan recipients to
provide ﬁnancial assets, such as warrants that give the federal government the ability to purchase
stock, to protect taxpayer interests; however, it has not developed policies and procedures to
ensure these warrants are acted upon in a manner that will beneﬁt taxpayers.

Entity involved: Department of the Treasury

Recommendation for Executive Action
The Secretary of the Treasury should develop policies and procedures to determine when to act on
warrants obtained as part of the loan program for aviation and other eligible businesses to beneﬁt
the taxpayers. Treasury agreed with our recommendation and said the agency is in the process of
creating a policy to dispose of the warrants obtained as part of the loan program.

Background
The CARES Act authorized the Department of the Treasury to provide up to $46 billion in loans
and loan guarantees to certain aviation businesses and other businesses deemed critical to
maintaining national security (national security businesses).447 This loan program was intended to
provide liquidity to these sectors. Treasury executed 35 loan agreements with businesses in these
targeted sectors, totaling about $22 billion. Consistent with the CARES Act, the executed loans
have a duration of 5 years or less. Of these 35 loans, as of October 1, 2021, 10 loans have been
fully repaid and the total value of outstanding loans is about $1.1 billion, as shown in the table
below.448

447CARES Act, Pub. L. No. 116-136, § 4003, 134 Stat. at 470 (codiﬁed at 15 U.S.C. § 9042). The CARES Act did not
provide criteria for which businesses were “critical to maintaining national security.” Treasury established the following
deﬁnition: (1) performing under a “DX”-priority-rated contract or order under the Defense Priorities and Allocations
System regulations (15 C.F.R. pt. 700) or (2) operating under a valid top secret facility security clearance under the
National Industrial Security Program regulations (32 C.F.R. pt. 2004). Treasury guidance further noted that applicants that
did not meet either of these criteria may still be considered for loans, if based on the recommendation and certiﬁcation
by the Secretary of Defense or the Director of National Intelligence, the applicant’s business is critical to maintaining
national security.
448As directed by the CARES Act, Treasury coordinated with the Department of Transportation (DOT) to determine the
eligibility of certain applicants. DOT conﬁrmed applicants held the appropriate air carrier certiﬁcates and the status
of their operations, and reported this information to Treasury, according to DOT. Treasury also coordinated with the
Department of Defense to determine the eligibility of applicants.

Page 271 GAO-22-105051
Executed and Outstanding Loans for the CARES Act Loan Program for Aviation and Other Eligible Businesses as
of October 1, 2021
Total loan
amount
approved by Total outstanding
Number of loans Treasury Number of loans loan amount
a
Loan category executed ($ millions) outstanding ($ millions)

Passenger and cargo air carrier 17 21,116 9 382

Repair station operator 5 19 5 20

Ticket agent 2 21 1 0.6

National security business 11 736 10 744

Total 35 21,891 25 1,147

Source: GAO analysis of Department of the Treasury data. | GAO-22-105051

Note: Section 4003 of the CARES Act authorized maximum assistance available through loans in three categories: passenger air
carrier, repair station operator, and ticket agent ($25 billion); cargo air carrier ($4 billion); and businesses critical to maintaining
national security ($17 billion). CARES Act, Pub. L. No. 116-136, § 4003, 134 Stat. 281, 470 (2020). To match the Department of the
Treasury’s reporting on these loans, and because air carriers that received loans could provide both passenger and cargo air
services, we combined all air carriers into a single category.
a
Includes all loan disbursements and increases in loan principal amount arising from payment-in-kind interest, less any
repayments of principal.

In nearly all cases, borrowers with loans over $300 million had to draw down 10 percent of the
loan amount when the loan was executed, while borrowers with loans less than $300 million had
to draw down the entire loan amount. Eight companies—seven passenger air carriers and one
national security business—had loans over $300 million, and all drew down at least the 10 percent
required by Treasury near the time the loan was executed.449 Under the loan agreements, seven of
these borrowers had until May 28, 2021, to draw down additional loan funds. With the passage of
this deadline, Treasury will not make any additional loan disbursements.

The CARES Act set out conditions for this loan program including that borrowers, among other
things, (1) not reduce employment levels by more than 10 percent from levels as of March 24,
2020, through September 30, 2020, (2) refrain from share buybacks and dividend payments, and
(3) limit compensation to employees with salaries over certain amounts until 1 year after the
date the loan or loan guarantee is no longer outstanding. Additionally, the CARES Act permitted
Treasury to make loans to publicly traded companies only if those businesses provided a

449The eight companies are Alaska Airlines, American Airlines, Frontier Airlines, Hawaiian Airlines, JetBlue Airways,
SkyWest Airlines, United Airlines, and Yellow Corporation—formerly known as YRC Worldwide. When Treasury increased
loan commitments, Treasury did not require borrowers to make additional draws to reach 10 percent of the new
total commitment. Yellow Corporation drew down the full approved loan amount of $700,000,000. The other seven
companies, all passenger air carriers, had drawn down about $1.6 billion of the $20.8 billion in loan funds Treasury
made available to them. As of October 1, 2021, six of these seven passenger air carriers have repaid the loans in full.

Page 272 GAO-22-105051
warrant—an option to buy shares of stock at a predetermined price before a speciﬁed date—or
equity interest—ownership in a company—to Treasury.450

Overview of Key Issues
Treasury has developed and is applying a four-pronged approach to compliance monitoring
and relies on loan recipients to certify the accuracy of submitted information. According
to Treasury oﬃcials, the agency is using a four-pronged approach to monitor loan recipients’
compliance with the terms and conditions of loan agreements, including restrictions on
compensation to employees with salaries over certain amounts and dividend payments.

• On a quarterly basis, recipients answer questions, through an online portal, to determine
compliance with loan agreement terms. These questions focus on employee headcount,
internal controls, and other loan agreement requirements. Treasury conducts two levels of
review on recipients’ answers to these questions. In the ﬁrst level of review, Treasury uses an
automated process to verify that the answers submitted by the recipient indicate compliance
with loan agreement terms and conditions. If the ﬁrst level of review identiﬁes a potential
compliance issue, then a Treasury analyst conducts a second, more detailed review of the
answers submitted by the recipient and the documents submitted. In this second review,
a Treasury analyst will also contact the recipient to discuss the issue, if needed. During this
review, potential compliance issues are cleared or determined to be an actual compliance
issue that requires remediation.

• On a quarterly basis, recipients must submit reports and certiﬁcations on a variety of ﬁnancial
topics. Staﬀ with the Bank of New York Mellon, which serves as Treasury’s ﬁnancial agent, and
Treasury’s recipient monitoring group review these reports and certiﬁcations to ensure all
recipients are complying with the terms and conditions of loan agreements.

• On an ongoing basis, Treasury’s asset management group reviews recipients’ ﬁnancial
statements and relevant media reports, and meets with recipients’ management to discuss the
business environment and any identiﬁed concerns.

• On an ongoing basis, Bank of New York Mellon staﬀ monitor the principal and interest
payments from borrowers. As the ﬁnancial agent for Treasury, the Bank of New York Mellon
collects payments from borrowers and relays to Treasury any issues or problems with
payments.

According to Treasury oﬃcials, recipients certify that the information submitted through the
agency’s online portal is correct. The oﬃcials stated that the reliance on recipient self-certiﬁcation
is an acceptable level of risk since about 90 percent of the outstanding loan funds are to publicly
traded companies. As publicly traded companies, these businesses are subject to ﬁnancial
reporting requirements established by the Securities and Exchange Commission, such as
preparing and making public annual and quarterly reports on ﬁnancial condition that contain
audited ﬁnancial statements.

450For other businesses, Treasury had discretion to obtain a warrant or equity interest or senior debt instrument.

Page 273 GAO-22-105051
Certain terms and conditions of the loan agreements, including restrictions on paying dividends to
shareholders and compensation for employees with salaries over certain thresholds, are in eﬀect
for 1 year after loans have been repaid in full. Therefore, Treasury expects to continue monitoring
all loan recipients’ compliance with loan agreement terms and conditions through at least the
second quarter of 2022. Since the loans can have a duration of up to 5 years, depending on the
repayment status, Treasury may keep this compliance monitoring process in place through 2026.

Treasury’s approach to address noncompliance with loan agreement terms and conditions
varies depending on the situation. Most of the penalties for noncompliance are laid out in the
terms and conditions of each loan agreement; however, according to Treasury oﬃcials, they have
the authority to negotiate amendments to agreements and address compliance violations as
appropriate. In cases of noncompliance with any terms and conditions, Treasury’s practice is to
contact the borrower and instruct it to remedy the violation as soon as possible. For example, if a
recipient does not provide compliance information on time, Treasury would contact the recipient
immediately after the deadline and notify it of the noncompliance and that it will face penalties if
the requested information is not provided within 30 days.

If Treasury determines that a loan recipient has intentionally submitted inaccurate information as
part of its compliance reporting, Treasury may refer the case to the Treasury Oﬃce of Inspector
General for investigation. Under the terms of the loan agreements, in certain circumstances
Treasury has the ability to accelerate the loan—require payment of the loan before the terms of
the loan agreement specify—in whole or in part. According to Treasury oﬃcials, as of September
2021, Treasury found one borrower to be materially noncompliant with loan agreement terms and
conditions; however, the borrower remedied the noncompliance, and Treasury does not plan to
take further action.451

Treasury monitors the performance of outstanding loans using real-time information.
According to Treasury oﬃcials, they use real-time information—such as ﬁlings with the Securities
and Exchange Commission and phone calls with the management teams of loan recipients—to
monitor the operations and overall ﬁnancial health of borrowers. Treasury oﬃcials also use this
information to assess the need for corrective action to avoid defaults or noncompliance. For
example, if a borrower did not meet the required collateral to loan value ratio, Treasury would
work with the borrower to ﬁx the issue. This could include the borrower providing additional
collateral or paying down the loan. According to Treasury oﬃcials, as of September 2021, no
borrowers have missed any loan payments.

451On August 3, 2021, Treasury issued a Notice of Non-Compliance to Caribbean Sun Airlines, Inc., a borrower through
the Treasury loan program, because this borrower did not respond to a request from the Special Inspector General
for Pandemic Recovery to ﬁll out a survey about its experience with the loan program. Pursuant to this Notice of Non-
Compliance, if Caribbean Sun Airlines, Inc. did not respond to the survey by September 2, 2021, an event of default
would occur, which would allow Treasury to pursue remedies including declaring the loan immediately due and payable.
Caribbean Sun Airlines, Inc. did not respond by September 2, 2021, and therefore, an event of default occurred with
the loan. Following the event of default, Treasury again communicated with Caribbean Sun Airlines, Inc. about the need
to complete the survey. Caribbean Sun Airlines, Inc. responded to the survey on September 10, 2021, and therefore,
according to Treasury oﬃcials, is now in compliance with the loan terms and conditions and Treasury does not plan to
take further action on this matter. According to Treasury oﬃcials, Caribbean Sun Airlines, Inc. has not been found to be
out of compliance with other terms and conditions of the loan agreement.

Page 274 GAO-22-105051
Treasury anticipates the loan program may make money for taxpayers, but has not
developed policies and procedures to guide when it will act on the warrants. The CARES
Act required Treasury to receive warrants or equity interest as a condition of making loans to
publicly traded businesses.452 The terms and conditions for these must be designed to provide
for Treasury’s reasonable participation in equity appreciation for the beneﬁt of taxpayers.453 The
law further provided that for the primary beneﬁt of taxpayers, Treasury may sell, exercise, or
surrender ﬁnancial instruments it obtained. Also, according to Oﬃce of Management and Budget
Circular A-129, which provides guidance on managing federal loan programs and other ﬁnancial
assets, agencies are required to establish appropriate controls over programmatic functions and
operations. This guidance also requires management to employ a reporting framework to inform
proactive portfolio management.

As taxpayer protection for the loan program, Treasury received warrants, which allow the
purchasing of common stock, from nine businesses equal to 10 percent of the total loan amount
drawn.454 According to Treasury oﬃcials, these warrants have a 5-year term and expire in 2025.
Before these warrants expire, Treasury may use these warrants to buy stock in these nine
businesses at the agreed upon price. Treasury also received shares equal to 29.6 percent of Yellow
Corporation common stock as taxpayer compensation. Treasury has not acted upon any of the
warrants for stock it holds in these nine businesses. While seven of these nine businesses have
already repaid the loan, Treasury still holds the warrants for stock purchases and—as the airline
industry recovers—these warrants may increase in value.

As stated above, Treasury must use these warrants or any senior debt instruments received as
part of granting these loans for the primary beneﬁt of the taxpayer. However, Treasury has not
developed policies and procedures for determining when to act on the warrants to beneﬁt the
taxpayer. According to Treasury oﬃcials, the agency has not ﬁnalized a policy on when to act upon
or exercise the warrants held through this program because it does not plan to dispose of these
warrants in 2021. The reasons for this include (1) uncertainty due to the ongoing pandemic, (2)
some of the warrants are not proﬁtable at this time, and (3) Treasury has until 2025 to exercise
these warrants, which according to Treasury oﬃcials removes any urgency for having a policy on
disposing of these warrants in 2021.

According to Treasury oﬃcials, it is likely that—if the airline industry continues to recover and
borrowers do not default—the loan program, through the exercising of these warrants, could
end up with a proﬁt. For example, based on the stock price at market close on October 1, 2021,
the warrants from one borrower would be valued at 159 percent above the predetermined price
Treasury would have to pay to act on them. By establishing policies and procedures over warrant

452In the event that the business is not publicly traded, Treasury must receive either a warrant or equity interest in the
business, or a senior debt instrument issued by the business. Pub. L. No. 116-136 § 4003(d)(1)(B), 134 Stat. at 474.
453Speciﬁcally, the CARES Act requires that the terms and conditions for these ﬁnancial instruments be designed to
provide for a reasonable participation by Treasury, for the beneﬁt of taxpayers, in equity appreciation in the case of a
warrant or other equity interest, or a reasonable interest rate premium, in the case of a debt instrument. Pub. L. No.
116-136 § 4003(d)(2), 134 Stat. at 474-75.
454These nine businesses are passenger air carriers, speciﬁcally: Alaska Airlines, American Airlines, Frontier Airlines,
Hawaiian Airlines, JetBlue Airways, Mesa Airlines, Republic Airways, SkyWest Airlines, and United Airlines. As of October
2021, all but one of these businesses are public companies.

Page 275 GAO-22-105051
processes, Treasury will help ensure that management has the information needed for informed
decision-making on when to act on the warrants it holds, in order to beneﬁt the taxpayers.

Methodology
To conduct this work, we reviewed the most recent Treasury data on the status of executed
loans as of October 1, 2021; reviewed the CARES Act; and interviewed Treasury oﬃcials. We are
continuing to use the data on loans that we have found reliable through interviews with agency
oﬃcials on the data’s consistency and completeness during prior reviews. We determined the
data were suﬃciently reliable for the purposes of our enclosure. We also reviewed the loan
transaction summaries prepared by Treasury to conﬁrm the nine businesses that provided
warrants to Treasury through the loan program—Alaska Airlines, American Airlines, Frontier
Airlines, Hawaiian Airlines, JetBlue Airways, Mesa Airlines, Republic Airways, SkyWest Airlines,
and United Airlines—and these businesses’ quarterly ﬁnancial reports (10-Q), if available, to
understand the number of warrants and predetermined price to use them.

Agency Comments
We provided Treasury and the Oﬃce of Management and Budget (OMB) with a draft of this
enclosure. Treasury provided technical comments, which we incorporated as appropriate. OMB
did not provide comments on this enclosure. Treasury’s general comments are reproduced in
appendix IX. In its management response, Treasury agreed with our recommendation and said the
agency is in the process of creating a policy that will allow Treasury to evaluate when and how to
act to dispose of the warrants obtained as part of the loan program.

Related GAO Products
Financial Assistance: Lessons Learned from CARES Act Loan Program for Aviation and Other Eligible
Businesses. GAO-21-198. Washington, D.C.: December 10, 2020.

Contact information: Heather Krause, (202) 512-2834, krauseh@gao.gov

Page 276 GAO-22-105051
Aviation Manufacturing Jobs Protection Program
The Department of Transportation set up a cross-organizational team to distribute payroll support
assistance to aviation manufacturing businesses and announced it will oﬀer $482.3 million in
funding to 313 businesses that applied in the ﬁrst application period.

Entity involved: Department of Transportation

Background
In March 2021, the American Rescue Plan Act of 2021 was enacted, establishing the Aviation
Manufacturing Jobs Protection (AMJP) Program.455 Through this program, the Department of
Transportation (DOT) is to provide up to $3 billion in funding to eligible aviation manufacturing
companies to pay up to half of their compensation costs for certain categories of employees,
for up to 6 months.456 While some aviation manufacturing companies were eligible for other
federal COVID-19 relief programs such as the three rounds of the Payroll Support Program (PSP),
and the Paycheck Protection Program (PPP), others were not eligible due to their size, among
other reasons.457 In response to reduced commercial passenger demand during the COVID-19
pandemic, airlines parked or retired a substantial portion of their aircraft ﬂeet, which in turn
reduced demand for new commercial airplanes, engines, and spare parts, and also aﬀected
companies that provide aircraft maintenance services.

DOT will provide AMJP program funds to eligible applicants to help cover payroll and facilitate the
retention or rehiring of employees.458 For a company to be eligible, among other criteria, it must
have involuntarily furloughed or laid oﬀ at least 10 percent of its workforce in 2020 as compared
to 2019 or experienced a 15 percent decline in 2020 revenues compared to 2019.459 Companies
that received ﬁnancial assistance from the ﬁrst round of PSP (PSP1) or that are still expending PPP
assistance as of the date of application to the AMJP program are ineligible for the program.460

455Pub. L. No. 117-2, §§ 7201-02, 135 Stat. 4, 101-07.
456Under the American Rescue Plan Act of 2021, DOT may use up to 1 percent of the funds appropriated ($30 million)
for implementation costs and administrative expenses. Pub. L. No. 117-2 § 7202(a), 135 Stat. at 103.
457The Payroll Support Programs provided funds to be used by air carriers and aviation contractors for employee
wages, salaries, and beneﬁts. To be an eligible contractor for the Payroll Support Programs, the CARES Act and Treasury
guidance deﬁned a contractor as a person that, under contract with a passenger air carrier conducting operations
under 14 C.F.R. pt. 121, performs catering functions or functions on airport property that are directly related to the air
transportation of persons, property, or mail, such as ground-handling of aircraft, among other things. Eligibility for the
Paycheck Protection Program was limited to small businesses (as deﬁned by provisions authorizing the program).
458AMJP recipients must agree to use funds for statutorily authorized purposes, which are the continuation of employee
wages, salaries, and beneﬁts and to facilitate the retention, rehire, or recall of employees of the employer. Funds cannot
be used for back pay of returning rehired or recalled employees. Pub. L. No. 117-2, § 7202(b), 135 Stat. at 103.
459For a company to be eligible for AMJP funding, it must be a company that actively manufactures an aircraft,
aircraft engine, propeller, or a component, part, or systems of an aircraft or aircraft engine under a Federal Aviation
Administration production approval; hold speciﬁc certiﬁcations; or operate certain certiﬁed processes, among other
requirements. The company must also be established, created, or organized under the laws of the United States, and
have signiﬁcant operations in and the majority of employees engaged in aviation manufacturing or maintenance, repair,
and overhaul activities and services based in the United States.
460The American Rescue Plan Act of 2021 provides that employers that received ﬁnancial assistance under the Payroll
Support Program (PSP1) established by section 4113 of the CARES Act (codiﬁed at 15 U.S.C. § 9073) are ineligible. The

Page 277 GAO-22-105051
When applying to the AMJP program, a company must identify the group of employees that are
eligible for payroll support through the program.461 Employees in the eligible employee group
must meet certain criteria, including: (1) cannot exceed 25 percent of a company’s U.S. workforce
as of April 1, 2020, (2) cannot have a compensation level greater than $200,000 per year, and (3)
must be engaged in aviation manufacturing or maintenance, repair, and overhaul activities or
services. In order to receive AMJP funding, companies also have to make several commitments,
including—but not limited to—a commitment that they will not involuntarily furlough or lay
oﬀ employees for whom they are receiving AMJP funding for the duration of the agreement.462
Once an AMJP agreement is in place with a company, DOT may require continuing disclosure
and reporting in support of requests for disbursement of funds.463 If DOT receives eligible AMJP
applications that exceed the $3 billion appropriated for the AMJP program, it must reduce the
funds provided to companies on a pro-rated basis.

Overview of Key Issues
DOT oﬃcials set up a cross-organizational team to manage the AMJP program and develop
and communicate eligibility and application requirements. The AMJP program is managed
out of DOT’s Oﬃce of the Secretary (OST); however, many oﬃces have been involved with the
set-up of the program. These include the Federal Highway Administration, Federal Motor Carrier
Safety Administration, Pipeline and Hazardous Materials Safety Administration, Federal Rail
Administration, and several oﬃces within OST. The Federal Aviation Administration has provided
technical support as needed.

DOT oﬃcials told us they met with oﬃcials at the Department of the Treasury and the Small
Business Administration (SBA) to discuss how to administer the AMJP program based on
experiences administering other federal COVID-19 relief programs. According to DOT oﬃcials, they
also discussed ways to verify the eligibility of applicants that received funds under Treasury or

Act does not address employers that have received ﬁnancial assistance under two similar programs, the Payroll Support
Program Extension (PSP2) established by the Consolidated Appropriations Act, 2021, and the Payroll Support Program
Extension (PSP3) established by the American Rescue Plan Act of 2021.
461Each company has to identify its eligible employee group and the amount of the total compensation level for the
eligible employee group, supported by sworn ﬁnancial statements or other appropriate data.
462Speciﬁcally, entities with certain certiﬁcates related to passenger airplanes with a seating capacity of 50 or more,
must agree to refrain from conducting involuntary layoﬀs or furloughs or reducing pay rates and beneﬁts for the eligible
employee group until September 30, 2021, or the duration of the agreement and receipt of funds, whichever period
ends later. Entities that do not fall within this group must agree to refrain from these actions for the duration of the
agreement and receipt of funds. If an approved company experiences natural attrition within the eligible employee
group, or terminates any employee in the eligible employee group due to performance or conduct issues in accordance
with employer policy, DOT will not require the company to backﬁll vacancies. However, the company will be required to
disclose any reduction in the total compensation costs for the eligible employee group and DOT may make comparable
reductions in the actual disbursements to the company. For the duration of the agreement and receipt of funds,
companies must provide DOT with immediate notice and justiﬁcation of involuntary furloughs or layoﬀs that exceed 10
percent of the company’s workforce for whom they are not receiving AMJP funding. In addition, for the duration of the
AMJP agreement, a company must also commit to fund all compensation costs for the eligible employee group that are
not paid from AMJP funds; this company-provided funding must be at least 50 percent of compensation costs.
463Process for Eligible Businesses Requesting Support under the Aviation Manufacturing Jobs Protection (AMJP)
Program, 86 Fed. Reg. 31,573 (June 14, 2021).

Page 278 GAO-22-105051
SBA COVID-19 relief programs, such as PSP1 and PPP respectively, during the period that would
disqualify applicants from the AMJP program.

In April 2021, DOT published a request for emergency approval of information collection for the
AMJP program in the Federal Register and released its public AMJP website with information
about the program.464 A second notice, published in the Federal Register in June 2021, outlined
the process for eligible companies to apply for the AMJP program.465 DOT hosted three webinars
about the program, its eligibility requirements, and the application process, and published audio
and video recordings and presentation slides to the AMJP website. It also created and published
a Frequently Asked Questions document that DOT oﬃcials told us was routinely monitored and
considered for potential updates as additional questions were submitted. DOT oﬃcials also told us
they directly addressed questions via telephone and email.

According to DOT oﬃcials, DOT has hired a ﬁnancial support contractor to, among other duties,
manage AMJP application validation, conduct work related to ﬁnancial controls and stratiﬁcation
of risk, and coordinate review of applications for completeness. According to DOT oﬃcials, DOT
oﬃcials will review and validate the process the contractor uses for application review and then
make an independent determination of an applicant’s eligibility and payroll support payment size.
DOT oﬃcials told us that they can request additional information from applicants when they ﬁnd
inconsistencies, deﬁciencies, or defects in the application, and have asked applicants to provide
corrected information when needed.466

While DOT is able to verify some eligibility requirements against government data, for other
requirements, oﬃcials told us they must rely on the self-certiﬁcation of applicants. For example,
DOT oﬃcials stated that disclosure rules on Treasury and Internal Revenue Service’s ability to
share taxpayer information constrains DOT from being able to verify the timing of Employee
Retention Credit use—instead DOT must rely on applicants to self-certify if and when they accrued,
requested, or planned to request the Employee Retention Credit.467 DOT oﬃcials told us that post-
award audits may be considered to verify self-certiﬁed information for some recipients.

464Agency Request for Emergency Approval of an Information Collection of Information Associated With the Aviation
Manufacturing Jobs Protection (AMJP) Program, 86 Fed. Reg. 19,695 (Apr. 14, 2021).
46586 Fed. Reg. 31,573.
466Pursuant to the Federal Register Notice, DOT may seek additional supporting documentation from any applicant at
any time, either during the application review process or subsequently. 86 Fed. Reg. at 31,577.
467See 26 U.S.C. § 6103. According to DOT oﬃcials, they anticipated making the ﬁrst AMJP awards during the third
quarter of calendar year 2021, which ended on September 30, 2021. By statute, a company cannot receive an AMJP
award if the company was allowed an Employee Retention Credit during the quarter proceeding the award—in the
case of AMJP awards made in the third quarter of calendar year 2021, a company cannot be allowed the Employee
Retention Credit in the second quarter of calendar year 2021, which ended on June 30, 2021. The ﬁrst AMJP application
deadline was July 13, 2021, 13 days after the end of the second quarter of calendar year 2021. DOT oﬃcials told us
that most businesses would not have had time to complete the second quarter tax ﬁling prior to the July 13, 2021
AMJP application deadline. As such, DOT relied on companies’ self-certiﬁcation on whether the applicant accrued,
requested, or planned to request an Employee Retention Credit during the second calendar quarter ending June 30,
2021. According to DOT oﬃcials, DOT will conduct post-award reviews that, among other things, will consider corrective
measures if DOT discovers an AMJP recipient was allowed an Employee Retention Credit during the calendar quarter
immediately preceding the AMJP award.

Page 279 GAO-22-105051
In addition, DOT oﬃcials told us that all AMJP recipients are subject to post-award reporting
requirements, performance assessment, and compliance reviews. Standard reporting will occur
at three points—within 10 days of a business entering into an AMJP agreement, when businesses
request an interim disbursement of award funds (if applicable), and a ﬁnal report when the ﬁnal
payment request is made. DOT oﬃcials also stated that selected recipients designated as higher-
risk will be required to submit a payroll report within 10 business days after the businesses’
regular payroll is made to employees.

DOT re-opened the application process due to confusion about one of the AMJP program
limitations. On July 26, 2021, DOT announced that it would re-open the AMJP application for an
additional 4 weeks, starting August 4, 2021, through September 1, 2021, due to confusion about
eligibility requirements for the program. In particular, DOT oﬃcials told us that they had heard
from some companies that thought they had to make a choice between applying for the AMJP
program or claiming the Employee Retention Credit, which is not accurate.468 By statute, DOT
cannot enter into an AMJP agreement with companies that were allowed the Employee Retention
Credit under the CARES Act for the quarter immediately prior to the one in which they enter into
an AMJP agreement.469 DOT initially opened applications to the program for four weeks, from
June 15, 2021, through July 13, 2021. DOT oﬃcials told us they received over 350 applications
during this period, far lower than the approximately 4,900 companies DOT oﬃcials estimated
were eligible to apply. DOT oﬃcials stated that they received 196 applications during the second
application period, which they are in the process of reviewing for eligibility as of late September
2021.

DOT announced it will oﬀer $482.3 million in funding to 313 businesses that applied in
the ﬁrst application period. Oﬃcials told us they entered into the ﬁrst AMJP agreements and
disbursed $196.8 million to 264 businesses as of late September 2021.470 DOT oﬃcials expect to
announce in November 2021 initial oﬀers of funding for businesses that applied in the second
application period. As required by statute, DOT must enter into all AMJP agreements within 6
months of the eﬀective date of the ﬁrst AMJP agreement.

468DOT oﬃcials refer to this as the Employee Retention Tax Credit (ERTC). By statute, DOT cannot enter into an AMJP
agreement with a company that was allowed the Employee Retention Credit (Pub. L. No. 116-136 § 2301, 134 Stat.
281, 347 (2020) (codiﬁed at 26 U.S.C. § 3111 note)) for the immediately preceding calendar quarter from the quarter in
which they enter the AMJP agreement. Pub. L. No. 117-2, § 7202(c), 135 Stat. at 104. As we reported in July 2021, eligible
employers of any size—including tax-exempt entities, certain governmental entities, and self-employed individuals with
employees—can claim the refundable Employee Retention Credit, as established under the CARES Act and as provided
by the Consolidated Appropriations Act, 2021 and American Rescue Plan Act of 2021. The credit amount is based on
qualiﬁed wages paid to employees, including certain health care expenses.
469For example, if a company was allowed the Employee Retention Credit in the quarter ending June 30, 2021, then
they could not enter into an AMJP agreement in the quarter ending September 30, 2021. However, according to DOT’s
application, companies in this situation can tell DOT if they want to enter into an AMJP agreement after October 1, 2021,
and receive AMJP funds.
470For AMJP recipients, DOT anticipates disbursing at least 50 percent of the estimated public contribution of the AMJP
payroll assistance ﬁnancial award shortly after receiving a fully executed AMJP agreement from an eligible company,
subject to certain limitations. The amounts and timing of subsequent funding disbursements will be addressed in the
AMJP agreements with each company and will depend upon companies’ reports and supporting documentation of
allowable costs incurred during the term of the agreement.

Page 280 GAO-22-105051
Selected industry associations were generally complimentary of DOT’s transparency and
communication about the AMJP program and its application requirements, although some
expressed challenges with the clarity of DOT guidance. Representatives from four selected
industry associations we interviewed told us that DOT oﬃcials were responsive and made an
eﬀort to provide answers when the program’s eligibility requirements or application process were
unclear. Representatives told us that DOT’s webinars, Frequently Asked Questions document,
and responsiveness to direct questions from individual companies were all helpful. For example,
representatives from three industry associations told us that their members had submitted direct
questions to DOT and received responses, sometimes within days.

However, representatives also told us that some program requirements were not always
clear, such as how to calculate the eligible employee group for the program. For example, one
industry association told us they received questions from members about whether they could
round up when calculating the number of employees in the eligible group. Industry association
representatives also told us that while the aviation manufacturing industry was starting to see an
increase in demand in summer 2021, there was still a need for the AMJP program. In particular,
representatives said that the assistance would help some companies rehire workers and start
making parts that have longer lead times, allowing the industry to prepare for a rebound in
demand and potentially minimize future disruptions to the supply chain.

Methodology
To conduct this work, we reviewed the American Rescue Plan Act of 2021 and DOT’s Federal
Register notices related to the AMJP program. We also reviewed DOT guidance, documents, and
webinars, and interviewed DOT oﬃcials. In addition, we conducted interviews with a selection of
four industry associations representing companies eligible to apply for the AMJP program.

Agency Comments
We provided DOT and the Oﬃce of Management and Budget (OMB) with a draft of this enclosure.
DOT provided technical comments, which we incorporated as appropriate. OMB did not provide
comments on this enclosure.

GAO’s Ongoing Work
We will continue to monitor DOT’s administration and oversight of the AMJP program, including
the steps it is taking to provide appropriated funds consistent with statutory and program
requirements and to monitor program recipients’ compliance with these requirements.

Contact Information: Heather Krause, (202) 512-2834, krauseh@gao.gov

Page 281 GAO-22-105051
Transit Industry
Transit agencies will continue to use COVID-19 relief funds to cover operating expenses, including
revenue losses, and are taking steps to strengthen rider conﬁdence.

Entity involved: Federal Transit Administration, within the Department of Transportation

Background
Millions of Americans rely on public transportation systems for mobility and access to jobs,
education, and essential services, such as medical care and grocery shopping. Within the
Department of Transportation (DOT), the Federal Transit Administration (FTA) provides grants
to state departments of transportation, local public transit systems, and tribes to support and
expand services. These services may include buses, subways, light rail, commuter rail, trolleys, and
ferries in urban, rural, and tribal areas.

The CARES Act appropriated about $25 billion to FTA to support the transit industry through two
formula programs—the Urbanized Area Formula program ($22.7 billion) and the Formula Grants
for Rural Areas program ($2.2 billion).471 CARES Act grant funds were made available to transit
agencies for COVID-19-related expenses incurred on or after January 20, 2020; FTA allocated
the $25 billion to urbanized areas, states, and tribes on April 2, 2020.472 These funds must now
be directed, to the maximum extent possible, to payroll and operating expenses.473 There is no
limit on the amount of funds that recipients may use for operating expenses, and the funds are
available until expended.474

On December 27, 2020, the Consolidated Appropriations Act, 2021 appropriated an additional $14
billion to FTA for the transit industry to prevent, prepare for, and respond to COVID-19.475 Further,
on March 11, 2021, the American Rescue Plan Act of 2021 (ARPA) appropriated approximately

471Pub. L. No. 116-136, 134 Stat. 281, 599 (2020). Formula programs allocate funding to recipients using a distribution
formula set by statute. Within the funds appropriated for the Rural Area formula program, $30 million is set aside for
tribal transit programs. The CARES Act provided an additional $75 million set-aside for the administration and oversight
of the appropriated funds.
472An urbanized area is an area, deﬁned and designated by the Secretary of Commerce, that encompasses a population
of 50,000 or more. Designated recipients of FTA Urbanized Area formula funds can include entities designated by
local oﬃcials, state governors, and local public transportation operators; alternatively, a state or regional authority
responsible for public transportation may be the designated recipient. 49 U.S.C. § 5302.
473As of December 27, 2020, recipients were required to direct funds, to the maximum extent possible, to payroll and
operations of public transit (including payroll and expenses of private providers of public transportation), unless the
recipient certiﬁes to the Secretary of Transportation that the recipient has not furloughed any employees. If the recipient
has certiﬁed this, the agency may employ relief funds for other uses such as capital or planning expenses. Consolidated
Appropriations Act, 2021, Pub. L. No. 116-260, div. M, tit. IV, 134 Stat. 1182, 1947 (2020).
474These ﬂexibilities are exceptions to the usual process for FTA’s Urbanized Area and Rural Area formula programs.
An additional exception is that there is no requirement for local matching funds for grants provided to large and small
urban areas and rural areas. All other Urbanized Area and Rural Area program requirements apply to CARES Act funds,
with the exception that operating expenses and certain capital expenses do not have to be included in a transportation
improvement program, a long-range transportation or statewide transportation plan, or a statewide transportation
improvement program.
475Pub. L. No. 116-260, div. M, tit. IV, 134 Stat. 1182, 1945. Division M of the Consolidated Appropriations Act, 2021 is
also referred to as the Coronavirus Response and Relief Supplemental Appropriations Act, 2021 (CRRSAA).

Page 282 GAO-22-105051
$30.5 billion to FTA for the same purpose.476 Both of these acts provide funding to urbanized
areas and states diﬀerently than the CARES Act, using a formula based on operating expenses
reported to FTA in 2018.477 FTA is allocating the majority of the funds appropriated by the two acts
through its Urbanized Area and Rural Area formula programs, similar to its allocation of funds
appropriated by the CARES Act.478 Consolidated Appropriations Act, 2021 funds are available for
obligation until expended, but ARPA funds are available for obligation until the end of ﬁscal year
2024.

Overview of Key Issues
Transit agencies have expended the majority of CARES Act funds and are beginning to
obligate and expend funding from the Consolidated Appropriations Act, 2021 and ARPA.
According to FTA data, as of August 31, 2021, transit agencies had obligated about 98 percent
of allocated CARES Act funds and had expended 81 percent of the funds (see table). In addition,
transit agencies had obligated about 49 percent of allocated funds from the Consolidated
Appropriations Act, 2021 and about 22 percent of allocated formula funds from ARPA. FTA oﬃcials
reported that as of August 31, 2021, 102 applications for new grants had been submitted for
Consolidated Appropriations Act, 2021 funds and 149 applications had been submitted for ARPA
funds.479

476Pub. L. No. 117-2, § 3401, 135 Stat. 4, 72.
477The Consolidated Appropriations Act, 2021 limited funding to 75 percent of an urbanized area’s 2018 operating
expenses and 125 percent of a state’s 2018 rural operating expenses when combined with the amount of CARES
Act funding it received. ARPA limited funding to 132 percent of an urbanized area’s 2018 operating expenses when
combined with the amount of other COVID-19 relief funding it has previously received. For an urbanized area that has
already received COVID-19 relief funding exceeding this percentage, ARPA limited funding to 25 percent of the area’s
2018 operating expenses. ARPA limited funding to varying percentages, ranging from 5 to 20 percent, of a state’s 2018
rural operating expenses, also based on the amount of COVID-19 relief funding it has already received. Operating
expenses for 2018 were determined from data reported to FTA’s National Transit Database.
478Of the available funds in both relief acts, $65 million is set aside for tribal transit programs and approximately $100
million is set aside for FTA’s program for the enhanced mobility of seniors and individuals with disabilities. Under ARPA,
an additional $1.675 billion was appropriated for FTA’s Capital Investment Grants program, $100 million for the Intercity
Bus program, $25 million for Competitive Planning grants, and about $2.2 billion for competitive grants for urbanized
and rural area recipients who need additional assistance because of the pandemic. Consolidated Appropriations Act,
2021 and ARPA funds are also provided at 100 percent federal share—that is, no local match is required. In this report,
we refer to these acts as “COVID-19 relief laws” and refer to the funding appropriated by these laws as “COVID-19 relief
funds.”
479FTA oﬃcials review grant applications to ensure that a transit agency’s proposed expenditures are eligible for
reimbursement by FTA. Only transit agencies whose applications are approved may obligate funding.

Page 283 GAO-22-105051
FTA Allocations and Transit Agency Obligations and Expenditures of COVID-19 Relief Funds for Formula
Programs, as of Aug. 31, 2021
e
Allocations Obligations Expenditures
Funding source ($ millions) ($ millions) ($ millions)

a 24,925 24,389 20,142
CARES Act

Consolidated
b 13,990 6,841 2,262
Appropriations, 2021

American Rescue Plan Act of
c 26,547 5,757 342
2021

d 65,462 36,986 22,746
Total

Source: GAO analysis of Federal Transit Administration (FTA) data. | GAO-22-105051

Note: FTA Formula programs include the Urbanized Area Formula program, the Formula Grants for Rural Areas program, the
Tribal Transit program, the Intercity Bus program, and the Enhanced Mobility of Seniors & Individuals with Disabilities program.
a
The CARES Act appropriated $25 billion for transit industry grants.
b
The Consolidated Appropriations Act, 2021 appropriated $14 billion.
c
The American Rescue Plan Act of 2021 appropriated about $30.5 billion, including approximately $26.5 billion to FTA formula
funding programs.
d
Numbers in columns may not sum to totals because of rounding.
e
Allocations do not include funding set aside for FTA administration and oversight.

Transit agency oﬃcials reported using COVID-19 relief funds to cover operating expenses
and mitigate the virus’s spread and also reported steps to strengthen rider conﬁdence.

Use of COVID-19 relief funds for operating expenses. Oﬃcials from each of the 20 transit agencies
where we conducted interviews said they used COVID-19 relief funds to cover operating expenses.
According to the oﬃcials, such expenses included purchasing equipment to mitigate the spread
of COVID-19 (ﬁve agencies) and labor costs (four agencies). Additionally, oﬃcials said they
used COVID-19 relief funds to oﬀset lost revenue, such as revenue from fares (10 agencies).
Oﬃcials from all 20 agencies reported receiving COVID-19 relief funds through the CARES Act.
Oﬃcials from 10 agencies reported also receiving COVID-19 relief funds through the Consolidated
Appropriations Act, 2021, in addition to the CARES Act, and oﬃcials from nine agencies reported
receiving COVID-19 relief funds through ARPA. Oﬃcials from four agencies said they had received
COVID-19 relief funds through all three acts.

Steps to mitigate COVID-19 spread. Oﬃcials from the 20 transit agencies told us they had taken
one or more steps to mitigate the spread of COVID-19 that were generally similar to those we
identiﬁed in November 2020. For example, oﬃcials said they had practiced enhanced cleaning
(18 agencies) and social distancing (17 agencies) and provided personal protective equipment (15
agencies) and barriers for drivers (eight agencies). Additionally, oﬃcials from all of the 20 agencies
said they had followed federal requirements such as mask mandates. Further, oﬃcials said they
generally relied on some combination of tools, guidance, and resources from state and local
health departments (12 agencies), the Centers for Disease Control and Prevention (10 agencies),

Page 284 GAO-22-105051
FTA (seven agencies), and the Transportation Security Administration (ﬁve agencies) to develop
COVID-19 mitigation measures.

Ongoing negative eﬀects of the COVID-19 pandemic. Oﬃcials from the 20 transit agencies reported
ongoing negative eﬀects of the COVID-19 pandemic—including reduced ridership (16 agencies),
lost fare revenue (12 agencies), and service reductions (10 agencies). In addition, oﬃcials
expressed concern about the possibility of continued reduced ridership (12 agencies) and lost fare
revenue (seven agencies). Oﬃcials from ﬁve of the 20 agencies said they planned to delay the use
of COVID-19 relief funds to mitigate future uncertainty such as continued fare revenue shortfalls.
Moreover, oﬃcials from seven agencies said they were concerned there would be an ongoing
shortage of drivers and staﬀ.

Steps to strengthen rider conﬁdence. Transit agency oﬃcials told us they were taking steps to
strengthen rider conﬁdence and increase ridership. Oﬃcials from 13 of the 20 agencies said they
had worked to strengthen rider conﬁdence via public outreach eﬀorts. For example, oﬃcials from
one transit agency said they had undertaken a publicity campaign to increase rider conﬁdence,
highlighting their enhanced sanitation eﬀorts via internet and television advertisements and
providing information resources on the agency website. Additionally, oﬃcials from 14 transit
agencies said they were relying on COVID-19 mitigation measures, such as increased sanitation
and mask requirements, to strengthen rider conﬁdence.

Tribal transit agencies reported similar uses of COVID-19 relief funds as nontribal transit
agencies, but said their service had been severely aﬀected by the pandemic. Four of the ﬁve
tribal transit agencies where we conducted interviews had either suspended or reduced service at
some point during the pandemic. According to a transit association oﬃcial, tribal transit agencies
were more likely than nontribal transit agencies to suspend operations for some period during the
pandemic. Oﬃcials from two of the ﬁve agencies said their agencies had shut down completely for
6 to 15 months, between March 2020 and June 2021. Oﬃcials from one tribal transit agency told
us that as of July 2021, transit service remained suspended because of the diﬃculty of hiring new
staﬀ.

Similar to nontribal transit agencies, oﬃcials from the ﬁve tribal transit agencies we interviewed
reported using COVID-19 relief funds to cover operating expenses, such as salaries (four agencies)
and equipment purchases to mitigate the spread of COVID-19 (two agencies). For example,
oﬃcials from all ﬁve tribal transit agencies reported implementing COVID-19 mitigation measures
on their transit systems that were generally similar to the measures implemented by nontribal
transit agencies. Oﬃcials reported using sanitation measures, physical distancing, and mask
requirements.

Some transit agencies are reserving COVID-19 relief funds because of continued
uncertainty, while some urbanized areas are still determining allocations at a local level.
The CARES Act and the Consolidated Appropriations Act, 2021 provide funding that is available
until expended and transit agencies may reserve funds to cover COVID-19 related expenses for the
next few years. FTA oﬃcials told us that recipients are carefully weighing the short- and long-term
eﬀects of the pandemic when determining how to reserve or spend COVID-19 relief funds. Oﬃcials
from ﬁve transit agencies said that they are planning to reserve the use of COVID-19 relief funds to
mitigate future uncertainty.

Page 285 GAO-22-105051
However, some urbanized areas are still determining how to distribute funding allocated under
the COVID-19 relief laws among local transit agencies. For example, the funding recipient for
the Chicago, Illinois, urbanized area, which encompasses areas in three states, is the Regional
Transportation Authority (RTA). Oﬃcials from the Chicago Transit Authority (CTA), a transit agency
within that urbanized area, told us that dividing allocations requires signiﬁcant discussion between
RTA and transit agencies in the area. CTA oﬃcials told us RTA has had to decide how to split its
funds–which were allocated to it based on the urbanized area’s 2018 operating expenses under
the Consolidated Appropriations Act, 2021–among three large transit agencies.480 Oﬃcials said
RTA is also undergoing the same process to decide how to distribute ARPA funds. According to FTA
oﬃcials, the agency had not identiﬁed any grant administration challenges other than waiting for
recipients to decide how to split allocated funds at the local level.

FTA has integrated COVID-19 relief programs into its existing review processes and
announced that the agency will begin supplemental oversight of these funds. FTA oﬃcials
told us that the agency is including oversight of COVID-19 relief funds in its regular oversight
program for Urbanized Area and Rural Area program funding recipients. Under FTA’s Triennial and
State Management review programs, the agency evaluates how Urbanized Area and Rural Area
funding recipients are meeting statutory and administrative requirements every 3 years. According
to FTA oﬃcials, these reviews will now incorporate supplemental guidance and instructions for the
oversight of COVID-19 relief funds. In March 2021, we reported that FTA had reinitiated Triennial
and State Management oversight activities in October 2020.

In addition, FTA announced on April 20, 2021, that the agency would begin new supplemental
oversight activities for recipients of COVID-19 funds, which will include spot reviews of expenses
charged to FTA grants and the documentation of these expenses. In some cases, FTA will also look
at the grant recipient’s ﬁnancial systems during the spot reviews. FTA oﬃcials told us that they
expect to complete 119 spot reviews this year and said that these reviews will help ensure proper
documentation and the eligibility of expenses charged to COVID-19 relief grants.

Further, according to the oﬃcials, FTA continues to provide webinar series to address areas
of risk in grant management and to educate recipients. FTA oﬃcials said that as of August 31,
2021, they had awarded grants to six recipients that had not previously received any FTA formula
funding. Moreover, the DOT Oﬃce of Inspector General announced a new audit of FTA’s COVID-19
Relief Funding oversight in July 2021. This initial audit will focus on the design of FTA’s controls
addressing relief funding requirements and on oversight risks throughout the grant life cycle.

Methodology
To conduct this work, we analyzed COVID-19 relief laws, applicable regulations, and FTA data on
transit industry grant funding, including tribal transit funding, as of August 31, 2021. We found
these data to be reliable for the purposes of describing federal allocations and transit agency
obligations and expenditures.

480Prior to the pandemic, RTA’s three transit agencies–Metra, Pace, and CTA–transported 2 million riders each day
across three states and six counties. Riders from these three agencies took over 500 million trips in 2019.

Page 286 GAO-22-105051
We reviewed written responses from FTA oﬃcials about how they were implementing provisions of
the CARES Act; the Consolidated Appropriations Act, 2021; and ARPA. In addition, we interviewed
oﬃcials from 20 selected nontribal transit agencies and ﬁve of 10 selected tribal transit agencies.
We interviewed oﬃcials, either by phone or through written questions and responses, regarding
challenges they had experienced related to the pandemic and the provision of COVID-19 relief
funds. We selected the 30 transit agencies on the basis of the amounts of CARES Act funds they
had obligated. We selected agencies that had obligated varying amounts of COVID-19 relief funds,
which we categorized as high, medium, or low amounts of relief funding. Finally, we interviewed
representatives of the Community Transportation Association of America and the American Public
Transportation Association to discuss the impact of COVID-19 relief funds on their members.

GAO’s Ongoing Work
Our work on transit industry COVID-19 relief funds is ongoing. We will continue to work with our
audit partners to monitor these programs.

Contact information: Andrew Von Ah, (213) 830-1011, vonaha@gao.gov

Page 287 GAO-22-105051
Coronavirus Economic Relief for Transportation Services
(CERTS)
The Department of the Treasury worked with federal agencies and industry groups to develop
eligibility guidance and has obligated all of $2 billion available.

Entities involved: Department of Transportation, Department of the Treasury, and United States
Coast Guard

Background
In December 2020, the Consolidated Appropriations Act, 2021 was enacted, establishing the
Coronavirus Economic Relief for Transportation Services (CERTS) Program.481 Through this
program, the Department of the Treasury must provide $2 billion in grant funding to eligible
providers of transportation services that have experienced certain revenue losses because of
COVID-19.482

According to industry associations, revenue losses can be attributed to declines in travel that
aﬀected multiple forms of transportation, including some of these transportation service
providers. For the purposes of the CERTS Program, transportation service providers generally
include private-sector operators of motor coaches (e.g., intercity buses), school buses, and
passenger vessels (e.g., small cruise ships and river boats). They also include pilot organizations
that provide vessels with pilots who navigate trade and cruise ships into and out of the port.483
While some of these companies may have received federal ﬁnancial assistance from other
COVID-19 relief programs, such as the Payroll Protection Program, the CERTS Program provides
funding exclusively to transportation service providers in these industries.

Treasury is awarding CERTS Program grants to eligible applicants to, for example, help cover
payroll, rehire employees who have been laid oﬀ, and cover certain overhead and operational
expenses.484 In addition to meeting other requirements, applicants must certify that they have
experienced revenue losses of 25 percent or more annually as a result of COVID-19. Applicants
must also certify that the amount of assistance sought, when combined with any other pandemic-
related federal ﬁnancial assistance they have received, does not exceed the total revenue they
earned during calendar year 2019.485

481Pub. L. No. 116-260, div. N, tit. IV, subtit. B, § 421, 134 Stat. 1182, 2061-2068 (2020) (codiﬁed at 15 U.S.C. § 9111).
48215 U.S.C. § 9111(b)-(c).
483The CERTS Program guidance’s deﬁnition of a “pilotage transportation service provider” or “pilotage company” is
a private-sector company, organization, group, or association whose principal business is providing pilotage services
regulated by a state in accordance with federal requirements. The deﬁnition does not include companies providing
towing or other tug assist services.
484CERTS Program funding may be used only for statutorily authorized activities such as these activities.
485According to Treasury guidance, pandemic-related federal ﬁnancial assistance includes “assistance under the
Paycheck Protection Program, and the Economic Injury Disaster Loan Program, tax credits under the Employee
Retention Tax Credit due to COVID-19, and, for sole proprietors, Federally funded unemployment beneﬁts under
the CARES Act, as amended, such as the Federal Pandemic Unemployment Compensation (FPUC) and the Pandemic
Unemployment Assistance (PUA) programs.”

Page 288 GAO-22-105051
According to program guidance, all CERTS grant applicants eligible for ﬁnancial assistance
that submit a complete application and are approved by Treasury will receive an award, based
primarily on their annual lost revenue and the aggregate annual lost revenues of all eligible
applicants in that industry.

Overview of Key Issues
As of October 6, 2021, Treasury had obligated all of the $2 billion appropriated for CERTS
Program grants to eligible providers of transportation services. Treasury accepted CERTS
applications through July 19, 2021. Treasury received 2,091 completed applications, including
1,646 from motor coach companies, 224 from school bus companies, 204 from passenger vessel
companies, and 17 from pilot organizations. As of October 6, 2021, Treasury had awarded CERTS
Program grants to 1,447 applicants.

Treasury oﬃcials said they are providing grant funding to approved applicants in one or two
payments. Treasury guidance states that this approach will ensure that grantees receive the
bulk of their funds on a rolling basis instead of waiting until Treasury has ﬁnished reviewing
all applications. For grantees that will receive two payments, the ﬁrst payment averaged
approximately $1.3 million per grantee, representing an estimated 80 percent of the total funding
the grantee will receive, according to Treasury oﬃcials. After reviewing all applications, Treasury
made a second payment to some grantees. .

Treasury oﬃcials said they began making grant payments to eligible companies in mid-August and
ﬁnished reviewing applications and making ﬁnal payments in October 2021.

Treasury developed a process to determine grant eligibility in consultation with the
Department of Transportation (DOT) and U.S. Coast Guard (USCG). Before accepting
applications, Treasury oﬃcials met with DOT and USCG oﬃcials to identify the information
Treasury would need to verify the eligibility of applicants in the motor coach, school bus, and
passenger vessel industries, which DOT and USCG regulate. For example, oﬃcials from DOT’s
Federal Motor Carrier Safety Administration (FMCSA) provided information about the motor
coach industry to Treasury oﬃcials and worked with them to develop a process for verifying
that applicants were authorized operators of motor coaches. To become an authorized motor
coach operator, FMCSA requires companies to obtain a unique USDOT number and comply with
applicable federal and state statutes and regulations.486

After the application period ended, Treasury sent FMCSA applications that did not provide
Treasury with suﬃcient information to verify that the motor coach applicant was an authorized
operator (e.g., the USDOT number was missing or was not an authorized operator’s USDOT
number). FMCSA oﬃcials said they reviewed more than 700 such applications.

486Although DOT, in technical comments, suggested using the term “passenger carriers,” we are using the term “motor
coach operators” for clarity and consistency with CERTS Program guidance. DOT also suggested the term “active and
properly registered,” but we are using the term “authorized” for the same reasons.

Page 289 GAO-22-105051
Treasury also required school bus companies, passenger vessel companies, and pilot
organizations to submit information in their applications that existing federal and state sources
could validate. For example:

• School bus companies. Treasury required applicants to submit, among other information, the
name of a school, school district, or state department of education for which the company
provided school bus services to transport students in 2019. Treasury was able to compare
this information with some information from state departments responsible for student
transportation. Treasury also used applicants’ tax returns to verify that transporting students
was their principal business.

• Passenger vessel companies. Treasury required applicants to submit vessel names and oﬃcial
numbers as documented on a certiﬁcation of inspection issued by USCG. Treasury was able to
compare the applicants’ information with information from USCG.

• Pilot organizations. Treasury required applicants to submit a copy of the operating license or
other document issued by a state pilot commission, board, or oversight body. Treasury was
able to compare these documents with information obtained from those entities.

In addition, Treasury required applicants to submit tax returns to show that they met revenue
loss and other eligibility requirements, such as having a certain number of employees. According
to Treasury oﬃcials, they also coordinated with the Department of Labor (DOL) to draft CERTS
Program guidance related to unemployment insurance, and they compared the applicant list for
the CERTS Program with Small Business Administration information to determine whether an
applicant had received other federal ﬁnancial assistance related to the COVID-19 pandemic.

Treasury plans to monitor awardees’ use of grant funds by requiring them to update Treasury
on their anticipated expenditures of these funds at least once every 90 days from the date of
receipt. Treasury will review these reports to verify that applicants are using grant funds for eligible
activities. In addition, Treasury oﬃcials told us they are conducting a risk assessment on the
design of the CERTS Program and will conduct a similar assessment of the grant award process in
2022.

Most selected industry associations expressed satisfaction with Treasury’s communication
about the CERTS Program but also reported some challenges related to the application time
frames and requirements. Oﬃcials of six of the seven industry associations where we conducted
interviews were generally complimentary of Treasury’s outreach and communication on the CERTS
Program. For example:

• Oﬃcials from one association said that Treasury oﬃcials solicited early and frequent input
from multiple industry groups, which Treasury incorporated into the program guidance and
application requirements.

• Oﬃcials from another industry association said that within weeks after the enactment of the
Consolidated Appropriations Act, 2021, Treasury oﬃcials contacted them to learn about their
industry. Speciﬁcally, oﬃcials of one school bus association we interviewed said that Treasury
met with them regularly from January 2021 through July 2021 to learn about the school bus

Page 290 GAO-22-105051
industry, including characteristics of private contractors and drivers, and the impact of the
pandemic on these groups.

Oﬃcials from one association were more critical of Treasury’s outreach and communication,
saying that they would have liked Treasury to communicate with industry members directly in
addition to communicating through their industry association.

Moreover, oﬃcials from six of the industry associations reported some challenges related to the
application time frames and Treasury’s requirements.487 For example:

• Oﬃcials from three of the industry associations said they would have preferred that Treasury
had begun accepting applications for the CERTS Program sooner.

• Oﬃcials from six of the industry associations we interviewed reported challenges in using
“ID.me,” a company that provides secure identity prooﬁng, authentication, and group aﬃliation
veriﬁcation for government and businesses across industries. Treasury required applicants to
verify their identity through ID.me before they could complete a CERTS Program application.

• Oﬃcials from six of the industry associations said that some of their members were confused
by certain information requirements in the application. Examples included a requirement to
provide a tax form that did not apply to the applicant and a requirement to provide calendar-
year tax information, although the applicant had ﬁled its taxes for a ﬁscal year.488

Nevertheless, industry associations we interviewed said that Treasury oﬃcials had generally
addressed these issues—either by providing instructions or, in some cases, making changes to
the application—and that many of their members had successfully completed the CERTS Program
application.

Methodology
To conduct this work, we reviewed the Consolidated Appropriations Act, 2021; applicable statutes
and regulations; and Treasury guidance and documentation. We conducted interviews with USCG,
DOT, and Treasury oﬃcials to understand Treasury’s approach to implementing, monitoring,
and overseeing the CERTS Program. We also reviewed Treasury data on the number and value of
CERTS Program grants and total obligations as of October 6, 2021, and asked Treasury about these
data’s source and reliability. On the basis of our review and Treasury’s response, we determined
that the data were suﬃciently reliable for the purposes of summarizing the number of grants
awarded and their amounts.

In addition, we interviewed USCG and DOT oﬃcials, asking them to describe how they worked with
Treasury oﬃcials to develop and implement the CERTS Program. Finally, we interviewed oﬃcials

487Oﬃcials from the seventh industry association said that they had not received much feedback on the application
time frame and requirements from members, because most members were not eligible for CERTS funding.
488The tax form mentioned was the Employer's Quarterly Federal Tax Return Form 941.

Page 291 GAO-22-105051
from seven associations representing the motor coach, school bus, passenger vessel, and pilotage
industries to obtain their perspectives on the CERTS Program and application process. We selected
these associations from each of the four industries on the basis of their role in the coalition that
advocated for the program, their inclusion in our prior reports, and Treasury and stakeholder
recommendations.

Agency Comments
We provided Treasury, DOT, and the Oﬃce of Management and Budget with a draft of this
enclosure. DOT provided technical comments, which we incorporated as appropriate. Treasury
and the Oﬃce of Management and Budget did not provide comments.

GAO’s Ongoing Work
We will continue to monitor Treasury’s oversight of the CERTS Program, including steps it takes to
ensure that the obligation and expenditure of appropriated funding meet statutory and program
requirements.

Contact Information: Biza Repko, Director, (202) 512-2834, repkoe@gao.gov

Page 292 GAO-22-105051
Paycheck Protection Program
The Small Business Administration has simpliﬁed its processes as it continues to review and
forgive Paycheck Protection Program loans.

Entities involved: Small Business Administration, Department of the Treasury

Background
Since March 2020, Congress has provided commitment authority of about $814 billion for the
Paycheck Protection Program (PPP) under the Small Business Administration’s (SBA) largest
guaranteed loan program, its 7(a) small business lending program.489 PPP loans, made by lenders
but guaranteed 100 percent by SBA, are low interest (1 percent) and fully forgivable if certain
conditions are met.490

As of July 30, 2021, lenders had made about 11.5 million PPP loans, totaling about $791 billion.491
Of those, lenders made about 5.1 million loans (totaling about $521 billion) during Round 1 (April
through August 2020). SBA relaunched the program (Round 2) on January 11, 2021, following
enactment of the Consolidated Appropriations Act, 2021, which authorized additional PPP loans
and made changes to the program. Among other things, the act expanded the categories of
forgivable nonpayroll costs and allowed PPP borrowers to receive a second PPP loan (second draw
loans) of up to $2 million provided that they meet certain criteria.492 In 2021 (Round 2 of PPP),
lenders made about 6.3 million loans totaling about $271 billion.

As of September 26, 2021, SBA had received about 7.3 million loan forgiveness decisions from
lenders and made payments on about 7 million of those (about $553 billion).493

489See Pub. L. No. 116-136, §§ 1102(b), 134 Stat. 281, 293 (2020); Pub. L. No. 116-139, § 101(a), 134 Stat. 620, 620 (2020);
Pub. L No. 116-260, div. N, tit. III, §323(a) 134 Stat. 1182, 2018-19 (2020); Pub. L. No. 117-2, § 5001(d) 135 Stat. 4, 85
(2021). The program also paid fees to lenders for their participation in the program.
490As originally implemented by SBA, at least 75 percent of the loan forgiveness amount must have been for payroll
costs. In addition, the CARES Act required loans to be used within an 8-week period in order for the loans to be fully
forgiven. However, the Paycheck Protection Program Flexibility Act of 2020 modiﬁed this to at least 60 percent and
allowed borrowers to pay or incur those expenses over a 24-week period. Pub. L. No. 116-142, § 3, 134 Stat. 641, 641-42
(2020). Under the Paycheck Protection Program Flexibility Act of 2020, the loan forgiveness covered period for PPP loans
was to end the earlier of 24 weeks after origination or December 31, 2020. The Consolidated Appropriations Act, 2021
further modiﬁed the covered period for forgiveness to allow the borrower to choose a covered period ending any date
between 8 and 24 weeks after origination. Pub. L. No. 116-260, div. N, tit. III, § 306, 134 Stat. 1182, 1997 (2020).
491New applications were accepted through May 31, 2021, and SBA had until June 30, 2021, to process submitted
applications.
492The Consolidated Appropriations Act, 2021 expanded the categories of forgivable nonpayroll costs to include
certain operations, property damage, supplier, and worker protection expenditures. PPP borrowers were eligible to
receive a second PPP loan of up to $2 million provided that they met certain criteria, such as having not more than 300
employees, having used the full amount of their initial PPP loan only for authorized uses, and having revenue losses of
at least 25 percent in a quarter of 2020 when compared to the same quarter in 2019. Pub. L No. 116-260, div. N, tit. III, §
311, 134 Stat. 1182, 2001 (2020); see also 86 Fed. Reg. 3712 (Jan. 14, 2021).
493Under SBA rules and guidance, the borrower submits the forgiveness application to the lender. The lender then has
60 days from receipt of the application to review and submit its forgiveness decision (approved in full, approved in part,
or denied) to SBA. SBA reviews the lender decision and remits the appropriate forgiveness amount to the lender. In
general, SBA must remit the forgiveness amount to the lender within 90 days of that amount being determined. SBA

Page 293 GAO-22-105051
SBA implemented the program rapidly and millions of small businesses have beneﬁted from PPP.
However, the speed with which SBA implemented the program left it with limited safeguards to
identify and respond to program risks, including susceptibility to improper payments and fraud.
Consequently, we have made four recommendations since June 2020 to ensure program integrity,
achieve program eﬀectiveness, and address potential fraud. As discussed in more detail below,
SBA has partially addressed each of these recommendations.

As of August 31, 2021, SBA had obligated about $810.3 billion across the two rounds of PPP,
including lender fees, and expended about $809.7 billion, according to SBA.

Overview of Key Issues
Status of loan reviews. For Round 1 loans, SBA conducted an initial automated review of all loans
to identify anomalies or attributes that may indicate noncompliance with eligibility requirements,
fraud, or abuse.494 Loans with any identiﬁed issues were subject to manual review. Any issue
identiﬁed would need to be resolved before borrowers could receive a second draw PPP loan or
have their loan forgiven. According to SBA oﬃcials, SBA had completed about 114,000 manual
reviews and 20,500 reviews were pending, including 6,200 loans pending higher authority reviews
by more experienced analysts for partial or full denial of forgiveness, as of July 26, 2021.

In April 2020, SBA and the Department of the Treasury (Treasury) announced that SBA would
review borrower eligibility for all loans of more than $2 million. In October 2020, SBA released
questionnaires to be ﬁlled out by for-proﬁt and nonproﬁt businesses that, together with their
aﬃliates, received PPP loans with an original principal amount of $2 million or greater. According
to SBA, the agency would use the information collected through these questionnaires to inform its
assessment of whether a business’s certiﬁcation that economic uncertainty made the loan request
necessary to support the business’s ongoing operations was made in good faith.495

As we reported in July 2021, SBA was processing loan forgiveness decisions for these large loans
slowly. Almost all loan forgiveness decisions for loans of $2 million or more submitted as of May
17, 2021, were processed in more than 90 days. In contrast, SBA processed about 79 percent of
loans of less than $500,000 in 10 days or less.

In part to review loans more quickly, SBA has discontinued the use of the loan necessity
questionnaires, which were designed to help SBA evaluate the good faith certiﬁcation made
by PPP borrowers on their loan application that economic uncertainty made the loan request

and Treasury oﬃcials told us they interpreted the CARES Act requirement to remit funds within 90 days to be subject
to SBA’s review of loans. Of the approximately 7.3 million loan forgiveness decisions submitted to SBA by lenders as of
September 26, 2021, 4,622,125 were for loans made in 2020 and 2,653,132 were for loans made in 2021.
494For a discussion of potential fraud in PPP, see the enclosure on Federal Fraud-Related Cases in app. I.
495As set forth in the CARES Act, borrowers had to certify in good faith that, among other things, (1) current economic
uncertainty made the loan request necessary to support the applicant’s ongoing operations and (2) the funds would
be used for allowed business-related purposes, such as to retain workers and maintain payroll or to make mortgage
payments, lease payments, or utility payments.

Page 294 GAO-22-105051
necessary to support ongoing operations.496 In a July 29, 2021, response to a frequently asked
question posted on its website, SBA explained that the results of the loan reviews that it had
completed to date showed that audit resources would be more eﬃciently deployed across all
loans if the loan necessity questionnaires were discontinued. The agency also noted that the loan
necessity reviews, including the review of borrowers’ completed questionnaires, were lengthy and
had caused delays beyond the 90-day statutory timeline for forgiveness.

In July 2021, we made two recommendations to SBA to improve its loan review process.

• We found that SBA had not documented policies and procedures for some elevated reviews
conducted when SBA determines that the borrower is ineligible for a PPP loan or for the
loan amount or loan forgiveness amount claimed by the borrower. Consequently, we
recommended that SBA establish time frames for ﬁnalizing and issuing these procedures.
In response, SBA agreed with the recommendation and told us it was working to ﬁnalize its
procedures for these reviews, but had not yet done so as of September 2021.

• We also found that although SBA developed tools such as a web portal and lender hotline,
its system for responding to lender inquiries was ad hoc. Some lenders, lender associations,
and state banking associations also noted that SBA was not responsive to lender inquiries,
including on loan reviews and the status of loan forgiveness determinations. Consequently,
we recommended that SBA develop and implement a process to ensure it responds in a
timely manner to PPP lender inquiries on loan reviews. In response, SBA agreed with the
recommendation and said it would provide procedures to us, but had not yet done so as of
September 2021.

Changes to loan forgiveness process. On July 30, 2021, SBA published an interim ﬁnal rule with
changes meant to simplify the loan forgiveness process for borrowers with the smallest loans.497

• Changes to loan forgiveness application process. Some borrowers now have the option to apply
for forgiveness directly to SBA instead of their lender. This option is available to borrowers
with loans of $150,000 or less whose lenders have opted into using SBA’s direct borrower
forgiveness platform.498 Similar to loan forgiveness applications submitted directly to the
lender, the lender would approve (or deny) the loan forgiveness request. Once the lender
submits its decision to SBA, SBA reviews the decision and sends any loan forgiveness proceeds

496The Associate General Contractors of America, Inc., ﬁled a lawsuit in December 2020 alleging, among other things,
that the necessity questionnaires and the process used to approve them violated the Administrative Procedure Act and
the Paperwork Reduction Act. Among other things, the lawsuit ﬁlings noted that the questionnaires ask borrowers to
describe their business success (or failure) after applying for the loan, not the status of their operations at the time they
applied for the loan.
49786 Fed. Reg. 40921 (July 30, 2021).
498Borrowers will continue to be required to submit their loan forgiveness application to their lender (rather than
through the SBA platform) if (1) the lender does not opt into using the direct borrower forgiveness process; (2) the
borrower’s PPP loan amount is greater than $150,000; (3) the borrower does not agree with the data as provided
by the SBA system of record or cannot validate their identity in the platform (for example, if there is an unreported
change of ownership); or (4) for any other reason the platform rejects the borrower’s submission. As of August 2,
2021, 882 of the approximately 5,500 lenders that issued PPP loans had opted into using the direct forgiveness
platform.

Page 295 GAO-22-105051
to the lender. SBA believes that (1) lenders that opt in will have reduced costs, increased
eﬃciency, and more timely remittance of forgiveness payments from SBA and (2) borrowers
who submit loan forgiveness applications directly through SBA’s direct borrower forgiveness
process will experience less wait time and uncertainty than submitting through their lender.499
The new platform began accepting loan forgiveness applications on August 4, 2021.

• Changes to documentation requirements. Borrowers who received a second draw PPP loan
of $150,000 or less in 2021 and did not submit documentation of revenue reduction at
the time of loan application may use an alternative method of documenting the required
revenue reduction.500 An independent third-party contractor has developed a COVID Revenue
Reduction Score (COVID score) that will be assigned for each second draw PPP loan of
$150,000 or less and that will be visible to the lender and certain borrowers in the platform.501
The lender may review the borrower’s score in the platform. When the COVID score meets
or exceeds the value required for validation of the borrower’s revenue reduction, use of the
COVID score will satisfy the requirement for the borrower to document revenue reduction.
If the COVID score does not validate the required revenue reduction for the borrower, the
borrower must either provide documentation directly to the lender or upload it through the
platform for its lender to review.

In July 2021, we reported that SBA had not issued guidance for key aspects of the forgiveness
process and made two recommendations in this area.

• The CARES Act requires SBA to make an advance purchase for the expected forgiveness
amount of a PPP loan within 15 days of receiving a report on the expected forgiveness amount
from a lender, and outlines a process for lenders to initiate the advance purchase. As of July
2021, SBA had not implemented this provision. Consequently, we recommended that SBA
implement the advance purchase provision in the CARES Act or report to Congress why it has
not complied, including seeking statutory ﬂexibilities or exceptions believed appropriate. In
response, SBA said it would notify Congress of its request to seek statutory ﬂexibility on this
matter or would request that Congress repeal the advance purchase requirement, but had not
yet done so as of September 2021.

• As of early July 2021, SBA had not yet ﬁnalized a process for PPP lenders to claim the loan
guarantee if a borrower ceases operations or defaults on a loan. We recommended that
SBA establish time frames for ﬁnalizing and issuing a PPP-speciﬁc loan guarantee purchase
process, including allowing lenders to claim the SBA guarantee when they have evidence the

499According to SBA, many smaller PPP lenders have expressed concerns that they do not have the technology
or human resources to develop eﬃcient electronic loan forgiveness platforms to process loan forgiveness
applications.
500Among other things, to be eligible for a second draw PPP loan, a borrower had to have experienced a revenue
reduction of not less than 25 percent during one quarter of 2020 compared to the same quarter in 2019.
501The independent third-party contractor will use a Consumer Demand Recovery Index that combines multiple
data sources on the consumption of products or services (such as foot traﬃc, third-party data, and credit card
spending) provided by businesses. Further, using the Business Operations Response Index, the score will measure
the businesses’ return to operational status, which includes employment and unemployment data, business-to-
business payment transactions, mobility, and foot traﬃc at workplace and visitor frequency at physical locations.

Page 296 GAO-22-105051
business ceased operations or declared bankruptcy. In July 2021, SBA issued a procedural
notice on lenders’ servicing responsibilities for PPP loans and SBA’s guarantee purchase
process. According to the notice, SBA will honor its guarantee and purchase 100 percent of
the outstanding balance of the loan in applicable circumstances provided that the lender has
complied with all the PPP requirements, including the lenders’ underwriting requirements and
the document collection and retention requirements. The process outlined in the procedural
notice would also apply to requests for guarantee purchase and charge-oﬀ for loans to
businesses that have permanently closed and do not plan to submit a forgiveness application
or have ﬁled for Chapter 7 bankruptcy protection, among other circumstances.

Methodology
To conduct this work, we reviewed SBA guidance and data on PPP loan forgiveness applications
and obligations and expenditures, and interviewed oﬃcials from SBA. We assessed the reliability
of the SBA data by reviewing documentation and interviewing SBA oﬃcials. We determined that
the data were suﬃciently reliable for reporting the status of loan forgiveness applications and PPP
expenditures.

Agency Comments
We provided SBA, Treasury, and the Oﬃce of Management and Budget (OMB) with a draft of this
enclosure. Treasury provided technical comments, which we incorporated as appropriate. SBA and
OMB did not provide comments on this enclosure.

GAO’s Ongoing Work
Our work on PPP is ongoing. We continue to examine SBA’s loan review and forgiveness processes
and the fraud risks in the program. We also continue to monitor SBA’s progress toward developing
and implementing corrective actions to address the material weaknesses identiﬁed by its ﬁnancial
statement auditor.

GAO’s Prior Recommendations
The table below presents our recommendations on the Paycheck Protection Program from prior
bimonthly and quarterly CARES Act reports.

Page 297 GAO-22-105051
Prior GAO Recommendations Related to the Paycheck Protection Program
Recommendation Status

The Administrator of the Small Business Administration (SBA) Open—partially addressed. SBA agreed with the
should conduct and document a fraud risk assessment for the recommendation, stating that it would work to ensure
Paycheck Protection Program (PPP) (March 2021 report). that a fraud risk assessment for PPP is completed.
According to SBA oﬃcials, as of July 2021, SBA had hired
a contractor to conduct a formal fraud risk assessment
and the assessment was underway.

The Administrator of SBA should develop a strategy that Open—partially addressed. SBA agreed with the
outlines speciﬁc actions to monitor and manage fraud risks in recommendation, stating that it would work to ensure
the Paycheck Protection Program on a continuous basis (March that fraud risks are monitored on a continuous basis.
2021 report). According to SBA oﬃcials, as of July 2021, SBA had
begun conducting a formal fraud risk assessment,
which is an important part of developing a strategy to
manage fraud risks.

The Administrator of SBA should expeditiously estimate Open—partially addressed. SBA neither agreed nor
improper payments and report estimates and error rates disagreed with our recommendation at the time of our
for PPP due to concerns about the possibility that improper report. In response to our recommendation, SBA stated
payments, including those resulting from fraudulent activity, that it was planning to conduct improper payment
could be widespread (November 2020 report). testing for PPP and that it takes improper payments
seriously. SBA oﬃcials stated that SBA had submitted a
sampling plan to the Oﬃce of Management and Budget
in February 2021. In July 2021, they said that SBA would
use this sampling plan to estimate both improper
payments and error rates for PPP in the fourth quarter
of ﬁscal year 2021. They noted in August 2021 that SBA
would oﬃcially report the improper payment rate in its
Fiscal Year 2022 Agency Financial Report, not the same
report for ﬁscal year 2021, because of timing.

The Administrator of SBA should develop and implement plans Open—partially addressed. At the time of our
to identify and respond to risks in PPP to ensure program report, SBA neither agreed nor disagreed with our
integrity, achieve program eﬀectiveness, and address potential recommendation. As we reported in September 2020,
fraud, including in loans of $2 million or less (June 2020 report). SBA had said that it planned to review all PPP loans of
$2 million or more and further stated that it may review
any PPP loan it deems appropriate, including loans
of less than $2 million. In late December 2020, SBA
provided a Loan Review Plan outlining steps it planned
to take to review PPP loans. The document describes
three steps in the process: automated screenings of
all loans, manual reviews of selected loans, and quality
control reviews to ensure the quality, completeness,
and consistency of the review process. In February
and April 2021, SBA provided additional documents
referenced in the plan that give further details on
how SBA and its contractors will conduct the various
reviews. However, SBA is still implementing its oversight
plan and has yet to complete other critical steps to
address potential fraud, including conducting a fraud
risk assessment.

Source: GAO. I GAO-22-105051

Page 298 GAO-22-105051
Related GAO Product
Paycheck Protection Program: SBA Added Program Safeguards, but Additional Actions Are Needed.
GAO-21-577. Washington, D.C.: July 29, 2021.

Contact information: William B. Shear, (202) 512-8678, shearw@gao.gov

Page 299 GAO-22-105051
Coronavirus State and Local Fiscal Recovery Funds
To provide assurance that funds are expended in accordance with program requirements, the
Department of the Treasury should comply with federal internal control standards by establishing
policies and procedures for administering the Coronavirus State and Local Fiscal Recovery Funds
as states and localities plan for using and managing their allocations.

Entities involved: Department of the Treasury and Oﬃce of Management and Budget

Recommendation for Executive Action
The Secretary of the Treasury should design and document timely and suﬃcient policies
and procedures for monitoring recipients of Coronavirus State and Local Fiscal Recovery
Funds to provide assurance that recipients are managing their allocations in compliance with
laws, regulations, agency guidance, and award terms and conditions, including ensuring that
expenditures are made for allowable purposes.

Treasury agreed with the recommendation.

Background
COVID-19 relief laws have appropriated $500 billion to the Department of the Treasury to provide
direct funding to states, the District of Columbia, localities, tribal governments, and U.S. territories
to help them respond to, and recover from, the COVID-19 pandemic.502 This amount includes
$350 billion appropriated to Treasury through the American Rescue Plan Act of 2021 (ARPA) for the
Coronavirus State and Local Fiscal Recovery Funds (CSLFRF) in March 2021.503

The CSLFRF allocates funds to states, the District of Columbia, localities, tribal governments, and
U.S. territories to cover a broad range of costs stemming from the ﬁscal eﬀects of the COVID-19
pandemic.504 Localities consist of metropolitan cities, counties, and smaller local governments,

502COVID-19 relief laws have appropriated other COVID-19 funding, including the $500 billion, to these entities to
address speciﬁc purposes.
503In addition, the CARES Act established the $150 billion Coronavirus Relief Fund (CRF) for Treasury to provide
payments to state, local, territorial, and tribal governments to cover the costs of necessary expenditures incurred
because of the COVID-19 pandemic between March 1, 2020, and December 31, 2021. CARES Act, Pub. L. No. 116-136,
div. A, tit. V, § 5001, 134 Stat. 281, 501-04 (2020), as amended by the Consolidated Appropriations Act, 2021, Pub. L.
No. 116-260, div. N, tit. X, § 1001, 134 Stat. 1182, 2145 (2020) (codiﬁed at 42 U.S.C. § 801). Pursuant to implementing
Treasury guidance on permissible uses of CRF funds, CRF recipients may use their allocations to oﬀset costs incurred
related to either the pandemic’s direct eﬀects (e.g., public health needs) or its indirect eﬀects (e.g., harm to individuals or
businesses as a result of COVID-19-related closures). 86 Fed. Reg. 4,182 (Jan. 15, 2021).
504Pub. L. No. 117-2, tit. IX, subtit. M, § 9901, 135 Stat. 4, 223 (2020) (codiﬁed at 42 U.S.C. §§ 802-803). Section 9901 of
ARPA appropriated $350 billion for two funds—the Coronavirus State Fiscal Recovery Fund and the Coronavirus Local
Fiscal Recovery Fund. For purposes of this report, we discuss these two funds as one—the Coronavirus State and Local
Fiscal Recovery Funds, or CSLFRF. For the purposes of the CSLFRF, ARPA establishes that the District of Columbia is
considered to be a state. 42 U.S.C. §§ 802(g)(5), 803(g)(9).

Page 300 GAO-22-105051
referred to as non-entitlement units of local government.505 According to Treasury data, it had
distributed approximately $240 billion in CSLFRF funds to recipients as of August 31, 2021. The
ﬁgure below shows the amounts of funding the CSLFRF allocates to various recipient types as well
as the amount of funding Treasury had distributed to each recipient type as of August 31, 2021.

Coronavirus State and Local Fiscal Recovery Funds Allocations and Treasury Distributions as of Aug. 31, 2021, by
Recipient Type

a
Non-entitlement units of local government are local governments typically serving populations of less than 50,000, such as
cities, villages, towns, or townships.
b
Metropolitan cities are the central cities or any other cities within a metropolitan area that have a population of 50,000 or
more.
c
ARPA provides that to receive their Coronavirus State and Local Fiscal Recovery Funds allocations, states, the District of
Columbia, and U.S. territories must ﬁrst provide Treasury with a signed certiﬁcation stating that they require their allocations to
carry out allowable activities and will comply with relevant requirements when they use their allocations. According to Treasury,
states that have experienced a net increase of more than 2 percentage points in their unemployment rate from February
2020 to the date of the latest available data will receive their full allocation of funds in a single distribution; other states will
receive funds in two equal tranches, with the second tranche provided within 12 months after certiﬁcation. Governments of
U.S. territories will receive a single distribution. Treasury is required to distribute allocations to metropolitan cities, states (for
distribution to NEUs), and counties in two equal tranches, providing the ﬁrst allocation within 60 days after ARPA’s enactment,
or May 10, 2021, to the extent practicable, and providing the second allocation no earlier than 12 months after the ﬁrst. After
receiving allocations for distribution to NEUs, states have 30 days to make those distributions, unless Treasury grants an
extension. According to Treasury, tribal governments will receive two distributions: the ﬁrst distribution in May 2021 and the
second distribution, based on employment data, in June 2021.

505A metropolitan city is deﬁned as the central city within a metropolitan area (i.e., a standard metropolitan statistical
area as established by the Oﬃce of Management and Budget) or any other city within a metropolitan area that has
a population of 50,000 or more. 42 U.S.C. § 803(g)(4). Non-entitlement units of local government (NEUs), are local
governments typically serving populations of less than 50,000. 42 U.S.C. §§ 803(g)(5), 5302(a)(5). NEUs include cities,
villages, towns, townships, or other types of local governments. NEUs receive their CSLFRF allocation through their state
governments. State governments will receive a speciﬁc allocation of these funds from Treasury for this purpose and are
responsible for distributing these funds to NEUs within their state. 42 U.S.C. § 803(b)(2).

Page 301 GAO-22-105051
ARPA established four key eligible-use categories for which recipients can use the CSLFRF
payments:506

1. Respond to the COVID-19 pandemic or its negative economic impacts, including assistance to
households, small businesses, and nonproﬁts or aid to impacted industries, such as tourism,
travel, and hospitality

2. Respond to workers performing essential work during the COVID-19 pandemic, by providing
premium pay to the recipients’ eligible workers or grants to eligible employers that have
eligible workers who perform essential work

3. Provide government services to the extent of any revenue reduction resulting from the
COVID-19 pandemic relative to revenues collected in the recipient government’s most recent
prepandemic full ﬁscal year

4. Make necessary investments in water, sewer, or broadband infrastructure

Recipients must incur obligations with their CSLFRF payments by December 31, 2024, and must
liquidate those obligations by December 31, 2026.507

On May 10, 2021, Treasury released an interim ﬁnal rule implementing the CSLFRF, which includes
guidance for recipients on the eligible uses of CSLFRF payments.508 The interim ﬁnal rule states
that CSLFRF recipients have ﬂexibility, within the four key eligible-use categories identiﬁed in
ARPA, to determine how best to use the payments to meet the needs of their communities and
populations. The interim ﬁnal rule also urges all recipients to develop plans, in consultation with
their constituents and communities, for spending their allocations across the four categories.
Further, the interim ﬁnal rule establishes a framework for determining the types of programs and
services that are eligible under ARPA and provides examples of uses that recipients may consider.

506ARPA requires recipients receiving CSLFRF payments to provide Treasury with a detailed accounting of the uses
of funds, including, in the case of a state or territory, modiﬁcations to the state’s or territory’s tax revenue sources in
addition to such other information as Treasury requires for the administration of the fund. 42 U.S.C. §§ 802(d)(2), 803(d).
507CSLFRF recipients may not use their allocations for deposit into any pension fund. Furthermore, states, the District
of Columbia, and territories may not use their allocations to either directly or indirectly oﬀset a reduction in net tax
revenue resulting from a change in law, regulation, or administrative interpretation made between March 3, 2021,
and the end of the ﬁscal year in which the recipient expends the last of the funds it receives that reduces any tax or
delays the imposition of any tax or tax increase. However, multiple state Attorneys General have ﬁled suit against the
federal government regarding this provision, alleging that it can be read as prohibiting a state from reducing taxes in
any manner and thus impinges on the sovereignty of the states to set their own tax policy. In July 2021, the U.S. District
Court for the Southern District of Ohio found that the provision exceeds Congress’s power under the Constitution and
enjoined Treasury from enforcing the provision against Ohio. Ohio v. Yellen, No. 1:21-CV-00181 (S.D. Ohio, July 1, 2021).
On August 27, 2021, Treasury appealed the decision to the U.S. Court of Appeals for the Sixth Circuit. Cases in other
states remain ongoing. If a recipient fails to comply with these requirements and restrictions, it must repay an amount
equal to the amount of funds used in violation of these requirements and restrictions. Pub. L. No. 117-2, tit. IX, subtit. M,
§ 9901, 135 Stat. at 223-33 (codiﬁed at 42 U.S.C. §§ 802-803).
508Subsequently, on May 17, 2021, Treasury published the interim ﬁnal rule in the Federal Register. See Coronavirus
State and Local Fiscal Recovery Funds, 86 Fed. Reg. 26,786.

Page 302 GAO-22-105051
Treasury requested public comments on the interim ﬁnal rule by July 16, 2021. Before issuing
the interim ﬁnal rule, Treasury published a document containing answers to frequently asked
questions (FAQs) about the program. Treasury updates this document periodically in response to
questions received from stakeholders.

Treasury’s interim ﬁnal rule also identiﬁes reporting requirements for recipients. For example,
speciﬁed recipients are required to submit reports that provide information on their use of the
funding and projects undertaken with the funding, among other things. On June 24, 2021, Treasury
released the Compliance and Reporting Guidance for the CSLFRF program, which provides
additional detail and clariﬁcation for each recipient’s compliance and reporting responsibilities.
The ﬁgure below shows reporting requirements for the ﬁve types of recipients.

Treasury’s Reporting Requirements for the Coronavirus State and Local Fiscal Recovery Funds, by Recipient
Type, as of October 7, 2021

Overview of Key Issues
States’ and selected localities’ plans for CSLFRF allocations. We surveyed states and localities
(cities and counties) to, among other things, determine the status of their plans for spending
CSLFRF allocations across the ARPA eligible use categories.509 Of the 48 states and 45 localities

509We sent a survey to budget oﬃcials in the 50 states and the District of Columbia. We also sent surveys to city and
county oﬃcials in a randomly selected nongeneralizable sample of 48 localities (24 metropolitan cities and 24 counties)
that had received some of their CSLFRF allocations from Treasury as of June 17, 2021. We administered all three surveys
from July 1 through August 6, 2021. We received responses from 48 states (including the District of Columbia) and 45
localities (22 cities and 23 counties). For the purposes of the CSLFRF, ARPA establishes that the District of Columbia is
considered a state. 42 U.S.C. § 802(g)(5). Therefore, in discussing survey responses, we use “states” to refer to the 50
states and the District of Columbia. Furthermore, unless otherwise noted, we use the single term “localities” to refer to

Page 303 GAO-22-105051
that responded to our survey, 24 states and four localities (two cities and two counties) reported
having a proposed or adopted spending plan as of July 2021 for using their CSLFRF allocations
(see ﬁgure). The remaining 24 states and 41 localities reported that they did not have a proposed
or adopted plan. Oﬃcials from two associations representing state and local governments told
us that many states and localities were still in the process of developing plans for their CSLFRF
allocations because they have until the end of 2024 to obligate them.

Number of States and Localities That Reported Having a Proposed or Adopted Plan for Spending Their CSLFRF
Allocations, as of July 2021

Note: We surveyed state budget oﬃcials in all 50 states and the District of Columbia and surveyed city and county oﬃcials
in a randomly selected nongeneralizable sample of 48 localities (24 metropolitan cities and 24 counties). The results of these
surveys cannot be generalized to the larger population of states and localities receiving CSLFRF allocations. We administered
the three surveys from July 1 through August 6, 2021. We received responses from 48 states (including the District of Columbia)
and 45 localities. The data shown reﬂect states’ and localities’ responses to a survey question asking whether they had a
proposed or adopted plan for how they will use their CSLFRF allocations.

The 24 states and four localities that reported having a proposed or adopted CSLFRF plan provided
documentation for our review. Our review of the documentation provided by the 24 states found
the following:

• Documentation or written survey responses for 19 states showed that their legislatures had
appropriated at least some of the funds from their CSLFRF allocations. However, some of
these states’ documentation and written responses also showed that they will need to take
additional steps to appropriate their full allocations or determine speciﬁc uses for funds
already appropriated. One state appropriated funds from its CSLFRF allocation to various state
programs and initiatives—including transportation, water protection and sustainability, and
reemployment assistance—in appropriations for ﬁscal year 2021–2022. In response to an
open-ended survey question, a budget oﬃcial stated that the state is still determining which
activities the appropriations for these programs and initiatives will support.

• Documentation or written survey responses for ﬁve states identiﬁed proposals or
recommendations for using allocations that their states’ executives had made but their
legislatures had not yet acted on. For example, one state’s governor had developed an
ARPA budget that recommended a range of uses for the state’s CSLFRF allocation, including
assisting those most aﬀected by the pandemic and upgrading the state’s broadband and

metropolitan cities and counties. While the results of these surveys cannot be projected to the full population of states
and localities receiving funding, they provide descriptive information about the spending and management of CSLFRF
allocations and about perceptions of the clarity of Treasury's CSLFRF guidance across a range of geographic areas.

Page 304 GAO-22-105051
water infrastructure. As of July 2021, the state legislature had not yet adopted the governor’s
budget proposal.

The four localities that reported having a proposed or adopted spending plan provided
documentation that varied in the level of detail and in the breadth of projects covered. For
example, one city provided the mayor’s spending proposal, which included appropriations for
various projects, such as increasing aﬀordable housing units, revitalizing tourism, and incentivizing
grocery stores to locate within food desert areas. Further, one county provided meeting minutes
as its plan, which documented county oﬃcials agreeing to use some of its CSLFRF allocation to
improve broadband service to unserved and underserved households and businesses. The county
also reported that it was unsure how it would use all of its allocation at this time.

States and localities that reported having a proposed or adopted CSLFRF spending plan also
reported intending to spend their allocations across multiple eligible use categories. Speciﬁcally,
22 of the 24 states and three of the four localities reported that they planned to spend some
portion of their CSLFRF allocation on public health, on responding to the pandemic’s economic
impact, or on both. States also frequently reported investments in infrastructure, oﬀsets to state
revenue losses, and administrative costs as areas in which they planned to use their CSLFRF
allocations. Two of the four localities that reported having a proposed or adopted plan reported
that they did not have spending plans in place for three-quarters of their allocations.

The 24 states and 41 localities that reported not having a proposed or adopted CSLFRF spending
plan also reported needing to take additional steps before proposing or adopting a plan.
Those steps included collecting public or stakeholder input (e.g., other levels of government
or businesses) or waiting for approval from a legislative body. For example, in response to an
open-ended survey question, one state oﬃcial reported that the governor’s emergency relief and
recovery oﬃce was working with state agencies and other stakeholders (e.g., local governments
and businesses) to ﬁnalize a plan before consulting with the state’s executive leaders and its
legislature to revise and approve it. Similarly, oﬃcials from two cities stated in response to open-
ended survey questions that they needed to develop other plans for their respective cities, such as
infrastructure and expenditure plans, before proposing or adopting a plan for using their CSLFRF
allocations.

States’ and localities’ self-reported capacity for managing allocations. Most states and
localities that responded to our survey reported that they had suﬃcient or more than suﬃcient
capacity to manage their CSLFRF allocations in accordance with federal requirements.510 For
example, 35 of 48 states and 38 of 45 localities reported they had either suﬃcient or more than
suﬃcient capacity to allocate and disburse CSLFRF funds. However, some of the states and
localities that responded to our survey reported that they had insuﬃcient capacity to manage

510For the purposes of this report, “capacity” refers to both the maintenance of appropriate resources and the ability to
eﬀectively manage and use those resources. In a July 2021 report, we identiﬁed several types of capacity: organizational
(i.e., the degree to which an organization is institutionally prepared for grant management and implementation,
including its ability to employ technology for grant oversight and reporting); human capital (i.e., the extent of suﬃcient
staﬀ with the knowledge and technical skills needed to eﬀectively meet its goals and objectives); and ﬁnancial (i.e.,
the extent of an organization’s ability to meet ﬁnancial responsibilities related to federal grants, such as matching
requirements).

Page 305 GAO-22-105051
their CSLFRF allocations in accordance with federal requirements. For example, 17 of 48 states
and eight of 45 localities reported that they had somewhat less than or much less than suﬃcient
capacity to report on the use of their CSLFRF allocations consistent with federal requirements (see
ﬁgure).

States’ and Localities’ Reported Capacity for Managing Their CSLFRF Allocations, as of July 2021

Note: We surveyed state budget oﬃcials in all 50 states and the District of Columbia and surveyed city and county oﬃcials
in a randomly selected nongeneralizable sample of 48 localities (24 metropolitan cities and 24 counties). The results of these
surveys cannot be generalized to the larger population of states and localities receiving CSLFRF allocations. We administered
the three surveys from July 1 through August 6, 2021. We received responses from 48 states (including the District of Columbia)
and 45 localities. The data shown reﬂect states’ and localities’ responses to a survey question asking about the suﬃciency
of their capacity (i.e., number of staﬀ, expertise, ﬁnancial systems, and IT systems) to administer their CSLFRF allocations in
accordance with federal requirements.

Forty-four of 48 states and 30 of 45 localities also reported that they had taken or planned to take
one or more additional steps to help them manage their CSLFRF allocations. These states and
localities most often reported that they had hired or planned to hire new staﬀ (39 states and 13
localities)—including contractors or consultants—or had reassigned existing staﬀ (38 states and
18 localities) to support the management of their CSLFRF allocations. For example, in response
to an open-ended survey question, one state oﬃcial commented that the state was adding ﬁve
staﬀ to a central oﬃce to manage CSLFRF allocations to agencies, respond to statewide inquiries,
and meet the program’s reporting requirements. In addition, this oﬃcial stated that the state’s
agencies would need to add staﬀ resources to manage their own CSLFRF responsibilities.

Additional steps that states and localities reported having taken or planning to take included
implementing new internal controls or amending existing ones (28 states and nine localities);
procuring new ﬁnancial management or information technology systems (16 states and four

Page 306 GAO-22-105051
localities); expanding fraud detection eﬀorts (21 states and 10 localities); and creating a new entity
(e.g., oﬃce or task force) to oversee CSLFRF spending (19 states and four localities). Likewise, one
county oﬃcial stated, in response to an open-ended survey question, that the county had created
a ﬁnancial recovery team and would use contracted auditors to ensure ﬁscal and contractual
CSLFRF compliance.

Six states reported that they may face challenges in expanding staﬀ capacity. For example, one
state, responding to an open-ended survey question, stated that onboarding contract staﬀ is a
long process and is unlikely to help the state meet its staﬃng needs. The state is therefore relying
on existing staﬀ, who have been strained by the demands of responding to the pandemic, to
manage its allocation.

Similarly, some localities expressed concern about their ability to ensure they had suﬃcient staﬀ
capacity for managing their CSLFRF allocations. For example, in response to open-ended survey
questions, oﬃcials from three counties stated that they were concerned about their ability to
maintain routine operations during periods of increased workload when fulﬁlling CSLFRF reporting
requirements. Oﬃcials from three associations representing state and local governments told us
that because of the extensive compliance and reporting requirements, smaller localities that do
not regularly receive federal funding assistance may face capacity challenges when managing their
CSLFRF allocations. Some of the oﬃcials said these jurisdictions generally have fewer staﬀ and
less institutional knowledge and awareness of federal processes than larger localities have, which
could limit the smaller localities’ ability to fully understand and comply with CSLFRF requirements.

States’ and localities’ perspectives on clarity of Treasury’s CSLFRF guidance. Few states and
localities responding to our survey indicated that Treasury’s guidance on the allowable uses of
CSLFRF funds and the program reporting requirements was very clear (see ﬁgure).

States’ and Localities’ Views on Clarity of Treasury’s CSLFRF Guidance on Allowable Uses of Funds and Reporting
Requirements, as of July 2021

Note: We surveyed state budget oﬃcials in all 50 states and the District of Columbia and surveyed city and county oﬃcials
in a randomly selected nongeneralizable sample of 48 localities (24 metropolitan cities and 24 counties). The results of these
surveys cannot be generalized to the larger population of states and localities receiving CSLFRF allocations. We administered
the three surveys from July 1 through August 6, 2021. We received responses from 48 states (including the District of Columbia)
and 45 localities. The data shown reﬂect states’ and localities’ responses to two survey questions asking about the clarity of
the Department of the Treasury’s guidance on, respectively, the allowable uses of Coronavirus State and Local Fiscal Recovery
Funds (CSLFRF) and CSLFRF reporting requirements.

Page 307 GAO-22-105051
In addition, most of the states and localities that responded to our survey reported needing
additional information about the allowable uses of CSLFRF funds. Most of the states and nearly
half of the localities also reported needing additional information about the program’s reporting
requirements.

Allowable uses of CSLFRF funds. The majority of states (29 of 48) and localities (29 of 45) that
responded to our survey reported needing additional information about allowable uses of CSLFRF
funds. For states, examples of such information included guidance on whether incarcerated
persons are considered underserved populations; guidance on the taxability of funds provided to
small businesses; and guidance on water projects, such as those for irrigation, dams, and levees.
Similarly, examples of information that localities reported needing included whether CSLFRF funds
could be used to purchase laptops and software for remote work and emergency communication
radios for emergency and safety employees.

In addition, 12 states and six localities reported that they needed additional information related to
calculating revenue losses. For example, one state reported that it needed additional information
about what could be considered a revenue loss. In a response to an open-ended survey question,
one city oﬃcial stated that the city needed clarity on how local tax increases from previous years
aﬀect the revenue replacement calculation.

Representatives of three associations representing state and local governments told us that, in
response to feedback from their members, the associations had asked Treasury for additional
information on the allowable uses of funds. For example, representatives of two associations
said that their members were uncertain whether they could pool their CSLFRF funds with other
jurisdictions to pursue regional eﬀorts, such as behavioral health and broadband projects. In
addition, in a comment submitted to Treasury about its interim ﬁnal rule for implementing the
CSLFRF, an association representing local governments noted that a lack of clarity on the allowable
uses of the funds might hamper its members’ ability to achieve an equitable economic recovery, a
key goal of the program.

Treasury oﬃcials told us they intend the ﬁnal rule to address recipients’ questions and concerns
and to provide additional information recipients need to eﬀectively manage their allocations. They
said that as of September 2021, Treasury was reviewing comments on the interim ﬁnal rule. Before
ﬁnalizing the rule, Treasury has taken steps to respond to recipients’ questions, including those
related to allowable uses, in updates to its FAQ document. For example, Treasury’s FAQ document
includes a six-page section devoted to questions about revenue loss. Treasury also speciﬁed
in a July 2021 update that jurisdictions could pool their funds to pursue regional projects. One
state reported that it had fewer questions about potential allowable uses with each FAQ update
Treasury published.

CSLFRF reporting requirements. Thirty-two states and 17 localities that responded to our survey
reported needing additional information about CSLFRF reporting requirements. For example,
states reported that they needed to know, among other things, details about the system through
which they would have to report. States cited speciﬁc information that would be helpful for
meeting reporting requirements, such as a CSLFRF reporting system user guide and data
dictionary, information about the data requirements (e.g., system data ﬁelds and business rules),
and reporting templates for the system.

Page 308 GAO-22-105051
Of the 17 localities that reported needing additional information about the reporting
requirements, ﬁve reported that they wanted Treasury to provide relevant training or webinars.
Since mid-June 2021, Treasury has provided seven webinars on CSLFRF reporting requirements,
including overviews of the relevant guidance and speciﬁc technical assistance topics (e.g.,
evidence-based interventions and program evaluations). Of the six localities that requested
information about the reporting system, one county wanted the reporting templates, to help it set
up its own reporting systems, and another county asked whether the portal would allow prime
recipients to upload data ﬁles in addition to entering data manually.

Two localities responding to the survey reported that they had obtained support from associations
representing state and local governments to supplement their understanding of Treasury’s
guidance. One association developed webinars and told us they hosted weekly calls, and another
association created on-line forums where members could share ideas and concerns with other
government oﬃcials.

In August 2021, after we administered our survey, Treasury issued additional guidance on CSLFRF
reporting requirements.511 This guidance includes a user guide for the reporting system, with
detailed instructions for navigating the system and entering required information. Treasury also
issued a template for recipients’ Recovery Plan Performance Reports, which identiﬁes information
that recipients will be required to include in the reports. Treasury also plans to issue a user guide
for submitting Project and Expenditure Reports.

States’ and localities’ concerns about single audits of CSLFRF recipients. CSLFRF recipients
may be subject to single audits as required by the Single Audit Act. The act establishes
requirements for states, the District of Columbia, localities, Indian tribes, U.S. territories, and
nonproﬁt organizations that receive federal awards to undergo single audits (or, in limited
circumstances, program-speciﬁc audits) of those awards annually (unless a speciﬁc exception
applies) when their expenditures meet a certain dollar threshold.512

The Single Audit Act’s Compliance Supplement provides guidance and policy for performing
single audits. Speciﬁcally, the supplement includes information about the programs’ objectives,
procedures, and compliance requirements. Without such information, auditors would need to
research compliance requirements for each program in numerous statues and regulations. Each
year, the Oﬃce of Management and Budget (OMB) works with agencies over multiple months to
compile this information and issues the information in the supplement. According to OMB, CSLFRF
and other programs included as part of ARPA, enacted in March 2021, were not included in the
2021 Compliance Supplement, because agencies did not have suﬃcient time to provide OMB
with the inputs required for those programs to be included in the August 2021 publication of the
Compliance Supplement.

511Department of the Treasury, Treasury’s Portal for Recipient Reporting: State and Local Fiscal Recovery Funds, ver. 1.0
(Aug. 9, 2021), accessed Sept. 30, 2021, https://home.treasury.gov/system/ﬁles/136/SLFRF_Treasury-Portal-Recipient-
Reporting-User-Guide.pdf.
512The Single Audit Act is codiﬁed, as amended, at 31 U.S.C. §§ 7501-06, and implementing Oﬃce of Management
and Budget guidance is reprinted in 2 C.F.R. part 200. Nonfederal entities (states, U.S. territories, Indian tribes, local
governments, or nonproﬁt organizations) that expend $750,000 or more in federal awards in their ﬁscal year are
required to undergo a single audit—that is, an audit of the entity’s ﬁnancial statements and federal awards or a
program-speciﬁc audit, in limited circumstances, for the ﬁscal year. 31 U.S.C. § 7502; 2 C.F.R. § 200.501.

Page 309 GAO-22-105051
In July 2021, we asked states and localities whether their single audits would include CSLFRF and
if they needed additional guidance to complete them. Among survey respondents that expected
their single audits to include CSLFRF, some expressed concern about the timeliness of the 2021
Compliance Supplement. One respondent reported that auditors are not able to adequately plan
their audit procedures for CSLFRF until the Compliance Supplement is available. Respondents
also expressed the need for additional guidance, such as clariﬁcation of requirements to report
CSLFRF data on the Schedule of Expenditures of Federal Awards and to report non-entitlement
unit distributions, revenue loss, and subrecipient monitoring.

These concerns expressed by states and localities are similar to those we previously reported.
Speciﬁcally, in March 2021, we reported that auditors who conduct single audits for entities
whose ﬁscal year ends on June 30 have expressed a need to receive the Compliance Supplement
by no later than April each year to eﬀectively plan their audits and conduct interim testing.513
In our March 2021 report, we recommended that OMB work with federal agencies and the
audit community (e.g., agencies’ inspectors general; the National Association of State Auditors,
Comptrollers, and Treasurers; and the American Institute of Certiﬁed Public Accountants) to
incorporate measures into OMB’s process for preparing single audit guidance to better ensure
that such guidance is issued in a timely manner and is responsive to users’ input and needs.

OMB neither agreed nor disagreed with our recommendation. Although OMB stated that it
shares the draft Compliance Supplement with the grant and audit communities as part of the
Compliance Supplement preparation process, OMB has not taken additional steps to ensure that
the Compliance Supplement and other single audit guidance are issued in a timely manner and
are responsive to users' input and needs. OMB oﬃcials stated that they believe there is currently
an appropriate balance between the timeliness of issuing the annual Compliance Supplement
and the consideration given to inputs and comments from federal agencies and the audit and
grantee community. However, the survey respondents expressed concerns similar to the audit
community’s concerns that we previously reported.

In July 2021, we further recommended that OMB, in consultation with Treasury, issue timely
and suﬃcient single audit guidance for auditing recipients’ uses of CSLFRF payments. OMB
neither agreed nor disagreed with our recommendation. In August 2021, OMB issued the 2021
Compliance Supplement and stated that issuance of an addendum containing CSLFRF guidance
was forthcoming.

Standards for internal control in the federal government require that management identify,
analyze, and respond to change—such as by providing timely guidance— as part of a risk
assessment process. The lack of timely single audit guidance could prevent auditors from
completing and issuing timely audit reports, which could in turn limit federal agencies’ ability
to ensure their awardees’ appropriate use of the CSLFRF and reduce the likelihood of improper
payments.

Treasury’s monitoring of recipients’ internal controls. In April 2021, Treasury established the
Oﬃce of Recovery Programs to oversee programs authorized through the COVID-19 relief laws
(primarily the CARES Act, the Consolidated Appropriations Act, 2021, and ARPA). Given Treasury’s

513Other entities’ ﬁscal years may end on other dates, such as March 31, September 30, or December 31.

Page 310 GAO-22-105051
role in administering the CSLFRF, Treasury’s newly established Oﬃce of Recovery Programs has
an opportunity to create a robust system of CSLFRF recipient monitoring that provides assurance
that funds are expended for allowable purposes and recipients are managing their allocations in
compliance with laws, regulations, agency guidance, and the awards’ terms and conditions.

Treasury’s current eﬀorts to administer the CSLFRF program include developing recipient reporting
guidance and, according to oﬃcials, planning the department’s approach to recipient monitoring.

Developing recipient reporting guidance. Treasury’s Compliance and Reporting Guidance for the
CSLFRF program includes a table listing internal control best practices for recipients. These best
practices relate to written policies and procedures, written standards of conduct, risk-based due
diligence, risk-based compliance monitoring, and record maintenance and retention. In addition,
the guidance includes requirements for recipients to report on their CSLFRF expenditures at
various intervals, beginning with an interim period and followed by quarterly and annual periods.
Further, the guidance notes that recipients and subrecipients will be subject to an audit under the
Single Audit Act when their annual expenditures in federal awards meet a certain dollar threshold.

Planning recipient monitoring. In our interview with CSLFRF program oﬃcials in Treasury’s Oﬃce
of Recovery Programs, the oﬃcials stated that Treasury’s procedures for monitoring of CSLFRF
recipients will include recipient risk assessment, systematic review of relevant single audit reports,
data collection and monitoring, risk-based and event-based compliance reviews, remediation,
escalation, and recoupment. Treasury oﬃcials also indicated that single audit report ﬁndings
related to eligibility will serve as a fundamental compliance control. In addition, according to the
oﬃcials, Treasury plans to conduct reviews of recipients’ interim, quarterly, and annual reporting
data. Further, the oﬃcials said Treasury plans to assess the risk of recipient noncompliance, using
information from sources such as whistleblower reports or media reports and following up as
needed with reviews of selected recipients.

However, as of August 2021, Treasury’s key internal processes and control activities for the
timely monitoring of recipients’ use of their CSLFRF allocations for allowable purposes and for
responding, as appropriate, to CSLFRF internal control and compliance ﬁndings were in the
development phase. According to Treasury oﬃcials, the key internal processes and control
activities had not been ﬁnalized or documented. The oﬃcials noted that program development
had occurred within a short timeframe since the enactment of ARPA in March 2021 and that
ﬁnalizing and documenting internal processes and control activities for the new program required
time and resources. Further, vacancies in top-level leadership positions in the Oﬃce of Recovery
Programs have contributed to uncertainty about how the ﬁnal program policies and procedures
will be implemented.

Federal internal control standards state that management should design control activities
to achieve objectives and respond to risks and should implement control activities through
policies. As part of this process, management designs control activities in response to the
entity’s objectives and risks to achieve an eﬀective internal control system. Management then
documents the internal control responsibilities in policies at an appropriate level of detail to allow
management to eﬀectively monitor the control activity.

Page 311 GAO-22-105051
Until Treasury properly designs and documents policies and procedures to guide CSLFRF program
oﬃcials and other responsible oversight parties in the Oﬃce of Recovery Programs, there is a
risk that key control activities needed to help ensure program management fulﬁlls its recipient
monitoring and oversight responsibilities may not be established or applied eﬀectively and
consistently. This risk may be particularly acute with respect to monitoring state and local
recipients with insuﬃcient capacity to manage their CSLFRF allocations in accordance with federal
requirements, as some survey recipients noted. Given that Treasury had distributed approximately
$240 billion of the CSLFRF as of August 31, 2021, it is important that suﬃcient policies and
procedures be developed and implemented expeditiously to prevent potentially unallowable uses
of CSLFRF by recipients from going undetected and uncorrected.

Methodology
To conduct this work, we surveyed budget oﬃcials in the 50 states and the District of Columbia.514
We also sent surveys to city and county oﬃcials in a randomly selected nongeneralizable sample
of 48 localities (24 metropolitan cities and 24 counties).515 We generated our sample of cities
and counties by using Treasury data on localities that had received at least some of their CSLFRF
allocations as of June 17, 2021. We then grouped the data on the basis of small, medium, and
high allocation amounts. We also used the U.S. Census Bureau’s division of regions to achieve
geographic diversity across four regions. The sampled localities included entities within each of
the 24 strata formed by the combination of allocation amounts, region, and type of locality (city or
county).

The results of the three surveys cannot be generalized to the larger population of states and
localities receiving funding, including to non-entitlement units of local government or to other
recipients of CSLFRF allocations (i.e., territories and tribal governments). However, the survey
results provide useful information about respondents’ planned use and management of state
and local CSLFRF allocations, self-reported capacity to manage allocations, and perceptions of the
clarity of Treasury’s CSLFRF guidance on allowable uses of funds and reporting requirements.

We asked state, city, and county oﬃcials to respond to each question from the perspective of their
state, city, or county as a whole. We did not independently verify whether oﬃcials sought input
from other state, city, or county oﬃces when completing the survey. The survey questions were
designed to obtain state, city, and county oﬃcials’ perspectives on their state’s, city’s, or county’s
planned use and management of its CSLFRF allocations at the time the oﬃcials were completing
the survey. Each of the three surveys also included questions to obtain oﬃcials’ perspectives
regarding the clarity of Treasury’s CSLFRF guidance on allowable uses and reporting requirements
at the time they completed the survey.

We administered the surveys from July 1, 2021, through August 6, 2021. We pretested a draft of
the surveys with oﬃcials in two states, one city, and one county to help ensure that the questions
were understandable and answerable. We received responses from 48 states, including the

514For the purposes of the CSLFRF, ARPA establishes that the District of Columbia is considered a state. 42 U.S.C. § 802(g)
(5). Therefore, in discussing survey responses, we use “states” to refer to the 50 states and the District of Columbia.
515We did not include non-entitlement units of local government in our sample.

Page 312 GAO-22-105051
District of Columbia; 22 cities; and 23 counties. We assessed data reliability by checking for
missing values and survey response errors and followed up with oﬃcials on survey responses
as appropriate. After completing these checks, we determined that the ﬁnal survey data were
suﬃciently reliable for the purpose of obtaining states’ and selected counties’ and cities’
perspectives on their planned use and management of their CSLFRF allocations and the clarity of
Treasury’s CSLFRF guidance on allowable uses and reporting requirements.

In addition, we interviewed a nongeneralizable sample of oﬃcials from ﬁve associations
representing state and local governments to obtain their perspectives on the CSLFRF program.
Furthermore, we reviewed federal laws and Treasury guidance, interviewed Treasury oﬃcials, and
collected Treasury’s written responses to questions that we posed.

Agency Comments
We provided a draft of this report to Treasury and OMB for review and comment.

Treasury provided written comments, which are reproduced in appendix IX. Treasury agreed
with our recommendation in this enclosure and stated that it is in the process of designing,
documenting, and implementing a risk-based compliance program to monitor recipient use
of CSLFRF program funds. Treasury also provided technical comments, which we integrated as
appropriate.

OMB did not provide comments on this enclosure.

GAO’s Ongoing Work
We currently have multiple ongoing or planned reviews of the funding that federal COVID-19 relief
laws appropriated for agencies across the federal government to provide payments to states,
the District of Columbia, localities, territories, and tribal governments for responding to, and
recovering from, the COVID-19 pandemic. Our work on the CSLFRF, in particular, is ongoing. We
will continue to review the extent to which federal agencies provide eﬀective guidance to help
recipients achieve accountability and transparency for their use of payments. We also plan to
examine how CSLFRF recipients spend their payments, address challenges they face in managing
the funds, and evaluate outcomes of their funded projects.

We will continue to monitor OMB and Treasury’s eﬀorts to provide CSLFRF guidance, and we will
work with the audit community to determine whether concerns previously expressed have been
addressed by the 2021 Compliance Supplement. We will also continue to monitor the status of our
open recommendations.

GAO’s Prior Recommendations
The table below presents our recommendations on single audits from prior bimonthly and
quarterly CARES Act reports.

Page 313 GAO-22-105051
Prior GAO Recommendations Related to Single Audits
Recommendation Status

The Director of the Oﬃce of Management and Budget, in Open—not addressed. OMB neither agreed nor
consultation with the Secretary of the Treasury, should issue timely disagreed with our recommendation. OMB oﬃcials
and suﬃcient single audit guidance for auditing recipients' uses stated that OMB is working with Treasury to prepare
of payments from the Coronavirus State and Local Fiscal Recovery audit guidance for the Coronavirus State and Local
Funds. (July 2021 report). Fiscal Recovery Funds and has already shared a
draft document with interested grant and audit
stakeholders for inputs and comments. OMB will
issue the ﬁnal audit guidance for the Coronavirus
State and Local Fiscal Recovery Funds as an
addendum to the 2021 Compliance Supplement,
but did not provide a planned issuance date. We
will continue to monitor the actions OMB takes in
response to our recommendation.

The Director of the Oﬃce of Management and Budget (OMB) Open—partially addressed. OMB neither agreed
should work in consultation with federal agencies and the audit nor disagreed with our recommendation. In
community (e.g., agency Oﬃces of Inspector General; National response to this report, OMB stated that it is
Association of State Auditors, Comptrollers, and Treasurers; actively working, to the extent practicable, to
and American Institute of Certiﬁed Public Accountants), to the update processes to better ensure that the single
extent practicable, to incorporate appropriate measures in OMB’s audit guidance is issued in a timely manner and
process for preparing single audit guidance, including the annual is responsive to users' input and needs. For
Single Audit Compliance Supplement, to better ensure that such example, in the preparation and publication of the
guidance is issued in a timely manner and is responsive to users’ 2021 Compliance Supplement, OMB shared the
input and needs (March 2021 report). draft audit guidance to both the grant and audit
communities for inputs and comments. OMB also
worked with agencies to address the technical
comments and make necessary edits on speciﬁc
programs that are reﬂected in the ﬁnal Compliance
Supplement. For comments on the general sections,
OMB provided the responses through the federal
notice.
However, the audit community continues to express
concerns with the process and provided a detailed
description of such concerns in written feedback in
late August. In addition, as we previously reported,
auditors who conduct single audits for entities with
June 30 year-ends have expressed a need to obtain
the Compliance Supplement by no later than April
of each year in order to eﬀectively plan their audits
and conduct interim testing. The 2021 Compliance
Supplement was not issued until August 2021 and
lacked guidance for several ARPA programs.
OMB oﬃcials stated in August 2021 that they were
working with agencies to prepare audit guidance
for ARPA programs as an addendum to the 2021
Compliance Supplement, but they did not provide
a planned issuance date. We met again with OMB
and the audit community in September 2021 to
further discuss the audit community’s concerns and
the additional single audit guidance needed. We
will continue to monitor the actions OMB takes in
response to our recommendation.

Page 314 GAO-22-105051
Source: GAO. I GAO-22-105051

Related GAO Products
COVID-19: Continued Attention Needed to Enhance Federal Preparedness, Response, Service Delivery,
and Program Integrity. GAO-21-551. Washington, D.C.: July 19, 2021.

State and Local Governments: Fiscal Conditions During the COVID-19 Pandemic in Selected States.
GAO-21-562. Washington, D.C.: July 15, 2021.

COVID-19: Sustained Federal Action Is Crucial as Pandemic Enters Its Second Year. GAO-21-387.
Washington, D.C.: March 31, 2021.

COVID-19: Federal Eﬀorts Could Be Strengthened by Timely and Concerted Actions. GAO-20-701.
Washington, D.C.: September 21, 2020.

COVID-19: Opportunities to Improve Federal Response and Recovery Eﬀorts. GAO-20-625. Washington,
D.C.: June 25, 2020.

Standards for Internal Control in the Federal Government. GAO-14-704G. Washington, D.C.:
September 2014.

Contact Information: Jeﬀ Arkin, (202) 512-6806, arkinj@gao.gov, and Beryl Davis, (202) 512-2623,
davisbh@gao.gov

Page 315 GAO-22-105051
Federal Contracts and Agreements for COVID-19
As of September 30, 2021, federal agencies continued to obligate billions of dollars monthly in
support of COVID-19 response eﬀorts through contracts and other transaction agreements, with
drugs and treatments continuing to be the predominant type of good and service procured.

Entities involved: U.S. Department of Agriculture; Department of Defense; Department of Health
and Human Services; and Department of Homeland Security, among others

Background
Federal agencies have used a variety of contracting mechanisms to provide vital goods and
services in support of federal, state, and local COVID-19 response eﬀorts.516 For example, federal
agencies have reported billions of dollars in obligations on contracts subject to the Federal
Acquisition Regulation—which provides uniform policies and procedures for acquisitions by all
executive agencies.517 Our prior work on disaster contracting has found that contracts play a
key role in federal emergency response eﬀorts, and that contracting during an emergency can
present a unique set of challenges as oﬃcials can face signiﬁcant pressure to provide critical goods
and services as expeditiously and eﬃciently as possible. The January 2021 National Strategy for
the COVID-19 Response and Pandemic Preparedness emphasizes the important role contracts will
continue to play during the response. The strategy states that the federal government will fully
leverage contract authorities to strengthen the vaccine supply chain; staﬀ vaccination sites; and ﬁll
supply shortages for personal protective equipment, drugs, and therapeutics.

In addition, federal agencies like the Department of Defense (DOD) and Department of Health
and Human Services (HHS) have relied on the use of other transaction agreements—which are
not subject to the Federal Acquisition Regulation—for activities such as vaccine development
and manufacturing in response to COVID-19.518 Our prior work has noted that the ﬂexibility to
tailor other transaction agreements can help agencies attract companies that do not typically do
business with the government. However, their use also carries a risk of reduced accountability
and transparency. The CARES Act relaxed certain requirements on the use of other transaction
agreements in response to COVID-19 for HHS and DOD—for example, related to congressional
reporting and who can approve certain transactions.519

As federal contracting activity through contracts and agreements continues to play a critical
role in response to the pandemic, it is important to ensure that the use of these contracts and

516The CARES Act includes a provision for GAO to provide a comprehensive audit and review of federal contracting
pursuant to the authorities provided in the Act. In addition to speciﬁc contracting reviews, we have reported on federal
contracting in response to the pandemic as part of regularly issued government-wide reports on the federal response to
COVID-19.
517For the purposes of this report, “contract obligations” refers to obligations on procurement contracts that are subject
to the Federal Acquisition Regulation and does not include, for example, grants, cooperative agreements, loans, other
transactions for research, real property leases, or requisitions from federal stock.
518Other transaction authorities allow certain agencies to enter into agreements “other than” standard government
contracts or other traditional mechanisms. Agreements under these authorities are generally not subject to federal
laws and regulations applicable to federal contracts or ﬁnancial assistance, allowing agencies to customize their other
transaction agreements to help meet project requirements and mission needs.
519Pub. L. No. 116-136, §§ 3301, 13006, 134 Stat. 281, 383, 522 (2020).

Page 316 GAO-22-105051
agreements in response to COVID-19 are accurately reported and visible to congressional decision
makers, entities with oversight responsibilities, and taxpayers. National Interest Action (NIA)
codes were established in 2005 following Hurricane Katrina to enable the consistent tracking of
emergency or contingency-related contracting actions in the Federal Procurement Data System
(FPDS).520 The COVID-19 NIA code was established on March 13, 2020, to track contract actions
and associated obligations in response to the pandemic in FPDS. The Department of Homeland
Security (DHS) and DOD have subsequently extended the code four times—generally in 6-month
increments, and most recently until March 30, 2022.521

Overview of Key Issues
Agencies obligated $86.9 billion on federal contracts, with DOD and HHS accounting for
three-quarters of obligations as of September 30, 2021. At the beginning of the response, HHS
accounted for the most federal contract obligations. However, as the response has progressed,
DOD’s contract obligations surpassed HHS’s. The increase in DOD’s contract obligations is due, in
part, to DOD’s support of interagency acquisition needs, which has included awarding contracts on
behalf of HHS for vaccine and therapeutic production and medical supplies.

As of September 30, 2021, DOD accounted for about 51 percent and HHS for about 24 percent
of the total obligations made by federal agencies in response to the COVID-19 pandemic. U.S.
Department of Agriculture (USDA) obligations, almost all of which were in support of the Farmers
to Families Food Box Program, accounted for $6.2 billion, or 7 percent of total obligations made in
response to COVID-19 (see ﬁgure).522

520The memorandum of agreement guiding the use of NIA codes does not address tracking of other transaction
agreements. Our prior work has identiﬁed challenges with how the Departments of Defense, Health and Human
Services, and Homeland Security tracked other transaction agreements in response to COVID-19. Our recommendations
related to these ﬁndings are detailed later in this enclosure.
521According to the memorandum of agreement guiding the management of the NIA code, DHS and DOD are
responsible for making determinations about whether to establish or close a code, based on a variety of considerations.
The General Services Administration (GSA)—the agency that operates and maintains FPDS—is responsible for adding
or updating the NIA code in the system based on DHS’s and DOD’s decisions. The extensions of the code are consistent
with our prior recommendations to DHS, DOD, and GSA related to the importance of ensuring federal agencies, the
public, and Congress have visibility into contract actions and associated obligations related to emergency response
eﬀorts.
522In May 2020, USDA implemented the Farmers to Families Food Box Program to assist commodity suppliers impacted
by the pandemic and to provide food assistance to the public. To accomplish these goals, USDA contracted with
hundreds of distributors to purchase billions of dollars in fresh fruits, vegetables, dairy, and meat products, and package
these products into family-sized food boxes for delivery to food banks, community and faith-based organizations, and
other non-proﬁt entities across the country. USDA ended the Farmers to Families Food Box Program on May 31, 2021,
but has reported it would use pandemic assistance funding to purchase fresh produce to be oﬀered in boxes to those in
need through the Emergency Food Assistance Program (TEFAP), administered by Food and Nutrition Service.

Page 317 GAO-22-105051
Contract Obligations in Response to COVID-19 by Federal Agency, as of September 30, 2021

In our July 2021 report, we reported that government-wide contract obligations related to
COVID-19 totaled $61.4 billion through May 31, 2021; by September 30, 2021, those obligations
had increased by about $25.5 billion—to $86.9 billion. DOD accounted for about $19.6 billion, or
about 77 percent of the increase in total contract obligations since May 31, 2021. See ﬁgure for
government-wide obligations and conﬁrmed COVID-19 cases by month.

Page 318 GAO-22-105051
Government-wide COVID-19-Related Contract Obligations and Conﬁrmed COVID-19 Cases by Month, Feb. 2020–
Sept. 2021

Types of goods and services purchased and competition rate changed over the course of
the pandemic. As the response to the pandemic has progressed, the types of goods and services
purchased have shifted from primarily medical equipment and supplies—such as ventilators
and personal protective equipment—to drugs and treatments, such as COVID-19 vaccines and
therapeutics.

• We reported in July 2021 that drugs and treatments were the largest area of government-wide
obligations and that remains the case. This area accounted for 37 percent of total obligations.
These obligations increased about tenfold from $3 billion as of November 2020, prior to the
Food and Drug Administration’s emergency use authorizations for the Pﬁzer, Moderna, and
Janssen vaccines, to about $31.9 billion as of September 30, 2021.523 Obligations for drugs
and treatments increased by $17.6 billion from May 31, 2021, to September 30, 2021, with
over 50 percent, or $9.3 billion, of the increase due to the purchase of additional COVID-19
vaccines from Pﬁzer to support booster shots, pediatric vaccinations, and international vaccine
donations. An additional $4.1 billion of the increase was for the purchase of therapeutics to
treat COVID-19 from Regeneron and Merck, and $3.3 billion of the increase was for additional
doses of COVID-19 vaccines from Moderna.

523Emergency use authorizations allow for the temporary use of unapproved medical products. Janssen
Pharmaceutical Companies are a part of Johnson & Johnson.

Page 319 GAO-22-105051
• Medical equipment and supplies—including personal protective equipment like nitrile
gloves—increased by about $708.1 million since May 31, 2021, and accounted for about $10.1
billion, or 12 percent of government-wide contract obligations as of September 30, 2021.

• Obligations for laboratory equipment and supplies—including COVID-19 test kits—increased
by an additional $965.7 million since May 31, 2021, to $3.6 billion as of September 30, 2021.

See ﬁgure for obligation amounts for the most-procured goods and services over time.

Contract Obligation Amounts for Top Five Goods and Services Procured in Response to COVID-19 by month, Feb.
2020–Sept. 2021

Note: In addition to what is reﬂected in the ﬁgure, agencies canceled, or deobligated, $176.5 million and $335.1 million for
drugs and treatments in July 2020 and April 2021, respectively.

As of September 30, 2021, COVID-19-related contracts for goods continued to be competed less
frequently than contracts for services. About 74 percent of the obligations for goods were on
contracts that were not awarded competitively, compared with about 40 percent of the obligations
for services. For example, about $29.6 billion, or 93 percent, of the $31.9 billion in obligations for
drugs and treatments and about $8.2 billion, or 82 percent, of the $10.1 billion in obligations for
medical and surgical equipment were on contracts awarded noncompetitively.

Since our July 2021 report, the proportion of COVID-19 related contracts identiﬁed as having
been awarded noncompetitively increased from 58 percent as of May 31, 2021, to 62 percent

Page 320 GAO-22-105051
as of September 30, 2021—about $54.3 billion.524 Throughout the course of the pandemic, the
percentage of obligations on these noncompetitive contracts has ﬂuctuated from a low of 25
percent of obligations in February 2020 to a high of 94 percent of obligations in July 2021. The
higher rate of obligations on noncompetitively awarded contracts was driven in part by large
noncompetitive awards for vaccine production.

Agencies must provide for full and open competition when awarding contracts, unless one of
several limited exceptions applies, such as when there is an unusual and compelling urgency for a
needed supply or service. Agencies cited an urgent need for awarding contracts noncompetitively
for about 84 percent, or about $45.8 billion, of the contract obligations associated with
noncompetitive awards.525

Federal agencies’ use of undeﬁnitized contracts increased. Undeﬁnitized contracts are one
technique that agencies have reported using to respond to COVID-19. Undeﬁnitized contracts can
enable the government to quickly fulﬁll requirements that are urgent or need to be met quickly
by allowing contractors to begin work before reaching a ﬁnal agreement with the government on
all contract terms and conditions.526 Since May 31, 2021, undeﬁnitized contract obligations for
COVID-19 have increased from about $5 billion to about $5.5 billion as of September 30, 2021,
totaling about 8 percent of government-wide obligations on contracts awarded in response to
COVID-19. This increase was driven in part by undeﬁnitized contracts made by HHS for vaccine kit
services to support booster shots and pediatric vaccinations.

DOD continued to report the highest amount of undeﬁnitized contract obligations, identifying
about $4.2 billion, or about 10 percent of its overall COVID-19-related contract obligations as being
on undeﬁnitized contracts. DOD’s use of undeﬁnitized contracts included the procurement of
goods and services such as N95 respirator production and constructing alternate care facilities to
treat COVID-19 patients. Our prior work has shown that, while undeﬁnitized contract actions can
allow the government to fulﬁll requirements that are urgent or need to be met quickly, these types
of contracts can pose risks to the government. For example, contractors may lack incentives to
control costs before all contract terms and conditions are deﬁned.527

524Our methodology for identifying noncompetitive contracts is explained in detail at the end of this enclosure.
525For the purposes of this report, obligations on contracts identiﬁed as using the unusual and compelling urgency
exception include those associated with contracts subject to Federal Acquisition Regulation 6.302-2, as well as orders
under multiple award contracts, which are subject to separate requirements under Federal Acquisition Regulation
subpart 16.5. Speciﬁcally, under Federal Acquisition Regulation 16.505(b)(2), orders on multiple award contracts
require contracting oﬃcers to give every awardee a fair opportunity to be considered for a delivery order or task order
exceeding $3,500, with exceptions, including if the agency need for the supplies or services is so urgent that providing
a fair opportunity would result in unacceptable delays. When using the unusual and compelling urgency exception to
full and open competition, agencies still must request oﬀers from as many potential sources as is practicable under the
circumstances.
526Undeﬁnitized contracts include letter contracts, as well as other undeﬁnitized actions. Letter contracts are a
preliminary contract that authorizes the contractor to begin work immediately, and undeﬁnitized contract actions
include any contract action for which the contract terms, speciﬁcations, or price are not agreed upon before
performance has begun under the action. Federal Acquisition Regulation 16.603 and Defense Federal Acquisition
Regulation Supplement 217.74.
527We reviewed the agencies’ use of undeﬁnitized contracts in response to COVID-19 in July 2021, including the use of
CARES Act ﬂexibilities for such contracts.

Page 321 GAO-22-105051
Use of other transaction agreements has stabilized, but transparency of certain awards
could be improved. In addition to contract obligations, DOD, HHS, and DHS have reported using
other transaction agreements in response to COVID-19, with obligations on these agreements
remaining about the same as we reported in July 2021, at $12.5 billion.528 In July 2021, we found
that about $12 billion of the $12.5 billion, or about 96 percent, was used for vaccine development
and manufacturing; medical research and development, such as COVID-19 rapid test kits; and
applied defense research and development, such as COVID-19 therapeutics.

In July 2021, we also found that agencies did not use a systematic approach to consistently
and accurately track other transaction agreements awarded in response to COVID-19 and the
associated dollars obligated. Speciﬁcally, HHS, DOD, and DHS reported obligating $10.9 billion
on COVID-19 other transaction agreements in FPDS. However, our analysis of FPDS data, agency
data, and agreement documents found these agencies obligated at least $12.5 billion—a $1.6
billion diﬀerence.529 In addition to this $1.6 billion, HHS’s National Institutes of Health oﬃcials
told us that as of August 2021, they had obligated about $560 million on COVID-19 agreements,
which were not reported in FPDS.530 Oﬃcials told us that while they do not report their use of
agreements to FPDS (as they are not required to do so by statute or policy), they began reporting
these agreements in a database used by HHS to track grant awards in May 2021.531

Methodology
To identify agencies’ federal contract and other transaction agreement obligations and
competition rate on contracts in response to COVID-19, we reviewed data reported in FPDS
through September 30, 2021.532 We primarily identiﬁed contract obligations related to COVID-19

528Our July 2021 report on COVID-19 other transaction agreements analyzed data as of March 14, 2021.
529HHS was responsible for a majority of the $1.6 billion in inaccurate reporting. We previously found in January 2021
that HHS misreported its COVID-19 agreements as procurement contracts because it reports agreements into the
procurement module of FPDS. In our January 2021 report, we recommended that HHS should accurately report data in
the federal procurement database system and provide information that would allow the public to distinguish between
spending on other transaction agreements and procurement contracts. HHS agreed with our recommendation and is
taking steps to address it.
530We did not include the $560 million in our FPDS data analysis.
531We recommended that HHS, DOD, and DHS—in coordination with other agencies responsible for a government-wide
initiative intended to improve federal contracting systems—consider prioritizing the development and implementation
of a systematic approach to consistently and accurately track other transaction agreements used for national interest
events, such as the COVID-19 pandemic, and the associated dollars obligated. HHS and DOD agreed and partially agreed,
and are exploring options that would address the recommendations. DHS disagreed and stated it is required to report
only certain agreements to FPDS—none of which were COVID-19 related—and that COVID-19 agreements that had been
inaccurately reported were not required to be reported. DHS also noted that it tracks its agreements internally, including
those that are COVID-19 related. However, the primary purpose of our recommendation was to increase accuracy and
transparency for the public to better understand the use of agreements and dollars obligated, regardless of what is
required to be reported.
532FPDS data from SAM.gov accessed through September 30, 2021. For purposes of this report, “competition rate” is
the percentage of total obligations associated with contracts awarded competitively. We calculated competition rates
as the percentages of obligations on competitive contracts and orders over all obligations on contracts and orders.
Competitive contracts included contracts and orders coded in the FPDS as “full and open competition,” “full and open
after exclusion of sources,” and “competed under simpliﬁed acquisition procedures” as well as orders coded as “subject
to fair opportunity,” “fair opportunity given,” and “competitive set aside.” Noncompetitive contracts included contracts

Page 322 GAO-22-105051
using the NIA code. We supplemented the use of the NIA code by searching for “COVID-19” and
“coronavirus” in the contract description ﬁeld to identify a limited number of additional contract
obligations.533 For contract actions over $1 million, we removed obligations that were identiﬁed in
the contract description as not related to COVID-19.

We assessed the reliability of federal procurement data by reviewing existing information
about FPDS and the data it collects—speciﬁcally, the data dictionary and data validation
rules—and by performing electronic testing. For the other transaction agreements that HHS
misreported as contracts, we removed the $1.6 billion associated obligations from our reported
contract obligations and reported them instead as other transaction agreement obligations. We
supplemented our FPDS analysis with analysis of agency-provided data and interviews with agency
oﬃcials. We determined that the data were suﬃciently reliable for the purposes of describing
agencies’ reported contract obligations in response to COVID-19.

Agency Comments
We provided HHS, DOD, DHS, the U.S. Department of Agriculture (USDA), and the Oﬃce of
Management and Budget with a draft of this enclosure. HHS, DOD, USDA, and the Oﬃce of
Management and Budget did not provide comments. DHS provided technical comments, which we
incorporated as appropriate.

GAO’s Ongoing Work
We have work underway related to DOD’s use of advance and progress payments during the
response to COVID-19.

GAO’s Prior Recommendations
The table below presents our recommendations on federal contracts and agreements for
COVID-19 from prior bimonthly and quarterly CARES Act reports.

and orders coded in the FPDS as “not competed,” “not available for competition,” and “not competed under simpliﬁed
acquisition procedures,” as well as orders coded as an exception to “subject to fair opportunity,” including “urgency,”
“only one source,” “minimum guarantee,” “follow-on action following competitive initial action,” “other statutory
authority,” and “sole source.” Even for contracts identiﬁed as noncompetitive, agencies may have solicited more than
one source.
533In November 2019 we identiﬁed some inconsistencies in the information agencies report in the contract description
ﬁeld in the FPDS. Data on DOD contract obligations based on information in the description ﬁeld were available only
through July 1, 2021, due to diﬀerences in the time frames for which DOD data are made publicly available.

Page 323 GAO-22-105051
Prior GAO Recommendations Related to Federal Contracts and Agreements for COVID-19
Recommendation Status

The Secretary of Agriculture should direct the Administrator of Closed-Implemented. The U.S. Department of
the Agricultural Marketing Service to issue guidance—such as Agriculture (USDA) neither agreed nor disagreed with
an acquisition alert or a reminder to contracting oﬃcials—on our recommendation. In February 2021, following our
the use of the COVID-19 National Interest Action code for identiﬁcation of contract data reporting challenges
the Farmers to Families Food Box Program or successor using the COVID-19 National Interest Action code for
food distribution program to ensure it accurately captures the Farmers to Families Food Box Program, Agricultural
COVID-19-related contract obligations in support of the Marketing Service oﬃcials said they conducted training
program (March 2021 report). with staﬀ to review National Interest Action code data
entry protocols. At that time, a senior Agricultural
Marketing Service oﬃcial also sent an email reminder to
procurement division personnel about OMB’s guidance
on the use of the COVID-19 National Interest Action
code. Following this training and email, oﬃcials took
action to retroactively report contract actions for the
program with the National Interest Action code. In
May 2021, the Agricultural Marketing Service updated
its instructions for entering contract actions into
the Federal Procurement Data System to include a
reminder to utilize the proper National Interest Action
code, if applicable.

The Secretary of Agriculture should direct the Administrator Open-Partially Addressed. USDA neither agreed
of the Agricultural Marketing Service to assess the nor disagreed with our recommendation, and as of
contracting personnel needed to fully execute the award September 2021 had not fully assessed the contracting
and administration of existing contracts in support of the personnel needed to execute and administer contracts
Farmers to Families Food Box Program or successor future in support of the Farmers to Families Food Box Program
food distribution program, and take the necessary steps to or successor food distribution program. According
ensure it has adequate contracting staﬀ in place to award and to Agricultural Marketing Service oﬃcials, they have
administer any future contracts for the program (March 2021 discontinued the program, and are using other
report). methods of hunger relief, so they do not anticipate
needing additional permanent staﬀ. Agricultural
Marketing Service oﬃcials are planning to use an
existing contract vehicle to obtain additional staﬀ
support for contract documentation needs for the
awards that have been made under the Farmers to
Families Food Box Program and other food purchasing
eﬀorts. Agricultural Marketing Service oﬃcials have
prepared a statement of work for the contract support
services needed, which includes information on the
number of Farmers to Families Food Box program
contracts needing support services and other tasks.
Agricultural Marketing Service oﬃcials expect interested
vendors to submit staﬃng plans identifying the number
of staﬀ needed to accomplish the work under the
contract, and to award the contract by the end of
calendar year 2021.

The Assistant Secretary for Preparedness and Response Open-Partially Addressed. ASPR agreed with our
(ASPR), in coordination with the appropriate oﬃces within recommendation, and as of April 2021, ASPR oﬃcials
the Department of Health and Human Services (HHS), should stated that they have discussed within ASPR the need
accurately report data in the federal procurement database to consistently identify other transaction agreements
system and provide information that would allow the public in the Federal Procurement Data System (FPDS) and
explored how their contract writing system may

Page 324 GAO-22-105051
Recommendation Status
to distinguish between spending on other transaction interface with the FPDS other transaction agreement
agreements and procurement contracts (January 2021 report). module in the future. In August 2021, ASPR oﬃcials
added that in the meantime, they have issued guidance
to their contracting teams to manually track other
transaction agreements in their contract writing system.
We will continue to monitor ASPR’s eﬀorts to implement
our recommendation.

The Secretary of Homeland Security, in coordination with Closed-Implemented. The Department of Homeland
the Secretary of Defense, should (1) revise the criteria in the Security (DHS) did not agree with our recommendation.
2019 National Interest Action (NIA) code memorandum of However, in March 2021, DHS, in coordination with
agreement to clearly identify steps they will take to obtain the Department of Defense (DOD), issued a revised
input from key federal agencies prior to extending or closing memorandum of agreement. The revised agreement
a National Interest Action code, (2) establish timelines for establishes a process and timelines for communicating
evaluating the need to extend a National Interest Action and evaluating NIA code extensions by requiring the
code, and (3) deﬁne what constitutes a consistent decrease General Services Administration to notify other federal
in contract actions and routine contract activity to ensure the agencies no less than seven days before a NIA code is
criteria for extending or closing the National Interest Action set to expire so that agencies can request an extension
code reﬂect government-wide needs for tracking contract as needed. The revised agreement also more clearly
actions in longer term emergencies, such as a pandemic deﬁnes what constitutes a consistent decrease in
(September 2020 report). contract actions to ensure criteria for extending or
closing a NIA code is consistently applied.

The Secretary of Defense, in coordination with the Secretary Closed-Implemented. DOD did not agree with our
of Homeland Security, should (1) revise the criteria in the 2019 recommendation. However, in March 2021 DOD, in
National Interest Action code memorandum of agreement coordination with DHS, issued a revised memorandum
to clearly identify steps they will take to obtain input from of agreement. The revised agreement establishes
key federal agencies prior to extending or closing a National a process and timelines for communicating and
Interest Action code, (2) establish timelines for evaluating evaluating NIA code extensions by requiring the General
the need to extend a National Interest Action code, and (3) Services Administration to notify other federal agencies
deﬁne what constitutes a consistent decrease in contract no less than seven days before a NIA code is set to
actions and routine contract activity to ensure the criteria for expire so that agencies can request an extension as
extending or closing the National Interest Action code reﬂect needed. The revised agreement also more clearly
government-wide needs for tracking contract actions in longer deﬁnes what constitutes a consistent decrease in
term emergencies, such as a pandemic (September 2020 contract actions to ensure criteria for extending or
report). closing a NIA code is consistently applied.

Source: GAO. I GAO-22-105051

Related GAO Products
COVID-19 Contracting: Actions Needed to Enhance Transparency and Oversight of Selected Awards.
GAO-21-501. Washington, D.C.: July 26, 2021.

COVID-19 Contracting: Observations on Federal Contracting in Response to the Pandemic. GAO-20-632.
Washington, D.C.: July 29, 2020.

Defense Acquisitions: DOD’s Use of Other Transactions for Prototype Projects Has Increased. GAO-20-84.
Washington, D.C.: November 22, 2019.

DATA Act: Quality of Data Submissions Has Improved but Further Action Is Needed to Disclose Known
Data Limitations. GAO-20-75. Washington, D.C.: November 8, 2019.

Page 325 GAO-22-105051
Disaster Contracting: FEMA Continues to Face Challenges with Its Use of Contracts to Support Response
and Recovery. GAO-19-518T. Washington, D.C.: May 9, 2019.

2017 Disaster Contracting: Actions Needed to Improve the Use of Post-Disaster Contracts to Support
Response and Recovery. GAO-19-281. Washington, D.C.: April 24, 2019.

2017 Disaster Contracting: Action Needed to Better Ensure More Eﬀective Use and Management of
Advance Contracts. GAO-19-93. Washington, D.C.: December 6, 2018.

Department of Homeland Security: Further Action Needed to Improve Management of Special
Acquisition Authority. GAO-12-557. Washington, D.C.: May 8, 2012.

Defense Contracting: DOD Has Enhanced Insight into Undeﬁnitized Contract Action Use, but
Management at Local Commands Needs Improvement. GAO-10-299. Washington, D.C.: January 28,
2010.

Department of Homeland Security: Improvements Could Further Enhance Ability to Acquire Innovative
Technologies Using Other Transaction Authority. GAO-08-1088. Washington, D.C.: September 23,
2008.

Contact information: Marie A. Mak, (202) 512-4841, makm@gao.gov

Page 326 GAO-22-105051
Vaccines Provided Abroad
COVID-19 Vaccines Global Access, known as COVAX—a key U.S.-supported initiative to provide
vaccines abroad—has exceeded its fundraising target but, because of supply limitations and other
challenges, is not projected to reach its goal of delivering 2 billion doses in 2021. COVAX now
expects to achieve this goal in early 2022.

Entities involved: Coalition for Epidemic Preparedness Innovations; Gavi, the Vaccine Alliance;
World Health Organization; United Nations Children’s Fund; Department of Health and Human
Services; Department of State; and U.S. Agency for International Development

Background
Providing safe and eﬀective vaccines to protect people from COVID-19 is crucial to mitigating
the public health and economic impacts of the virus and to ending the pandemic.534 In addition
to limiting cases and deaths, slowing transmission of COVID-19 through vaccination is critical to
deterring the emergence of new virus variants. By contrast, uncontrolled transmission anywhere
in the world increases the risk of variants that can evade current vaccines, diagnostics, and
therapeutics, thereby prolonging the pandemic globally and endangering the fragile U.S. recovery.
The World Health Organization’s (WHO) reporting indicates that about 6.1 billion vaccine doses
had been administered worldwide as of September 29, 2021. However, according to WHO
estimates, a total of 11 billion to 12 billion doses will be needed to vaccinate at least 70 percent of
the world’s population in order to end the pandemic.535

In April 2020, WHO and seven other global health organizations created the Access to COVID-19
Tools Accelerator (ACT-Accelerator), an eﬀort to rapidly develop and provide equitable access to
vaccines, diagnostics, and therapeutics.536 The vaccine-related component of the ACT-Accelerator
is known as COVID-19 Vaccines Global Access (COVAX). Several global health organizations—the
Coalition for Epidemic Preparedness Innovations (CEPI);537 Gavi, the Vaccine Alliance (Gavi);538 and
WHO—colead COVAX, with the United Nations Children’s Fund (UNICEF) as a key implementing
partner.539 COVAX is the only global initiative working with governments and manufacturers to
ensure COVID-19 vaccines are available worldwide to both higher-income and lower-income
countries, according to WHO.540 As of September 2021, 193 countries had signed up as members

534As of October 1, 2021, more than 233 million conﬁrmed cases of COVID-19 had been reported worldwide, including
more than 4.7 million deaths, according to the World Health Organization.
535Some COVID-19 vaccines require one dose and others require two doses.
536The seven other organizations are the Coalition for Epidemic Preparedness Innovations; FIND; Gavi, the Vaccine
Alliance; the Global Fund; Unitaid; Wellcome; and the World Bank Group.
537The Coalition for Epidemic Preparedness Innovations is a global partnership between public, private, philanthropic,
and civil society organizations working together to accelerate the development of vaccines against emerging infectious
diseases and to enable equitable access to these vaccines for people during outbreaks.
538Gavi is an international organization, created in 2000, that brings together public and private sectors with the shared
goal of saving lives and protecting people’s health by increasing equitable and sustainable use of vaccines.
539This enclosure refers to CEPI, Gavi, WHO, and UNICEF collectively as COVAX partners.
540According to COVAX partners and documentation, COVAX is not a legal entity but rather a collaborative mechanism
that relies on existing institutions wherever possible.

Page 327 GAO-22-105051
of COVAX.541 This included 101 higher- and middle-income countries paying for vaccines
themselves (i.e., self-ﬁnancing) and 92 eligible low- and middle-income countries relying on donor
funding.542

The U.S. has provided funding in support of COVAX. Through the Consolidated Appropriations Act,
2021, Congress appropriated $4 billion for the Global Health Programs account and directed that
the funds be administered by the U.S. Agency for International Development (USAID) and made
available as a contribution to Gavi.543 In February 2021, the U.S. announced that it would provide
an initial $2 billion contribution to COVAX through Gavi to support access to safe and eﬀective
vaccines for the 92 low- and middle-income countries participating in COVAX. USAID obligated the
initial $2 billion for COVAX to Gavi in March 2021 and the remaining $2 billion in July 2021.

The U.S. has also provided dose donations in support of COVAX. In May 2021, the U.S. announced
plans to donate 80 million vaccine doses abroad from its own surplus supply, providing the
majority of the doses through COVAX. The U.S. later announced that it had donated more than
110 million doses from its own supply. In June 2021, the U.S. announced it would facilitate the
purchase of 500 million vaccine doses to be provided through COVAX—including 200 million to
be delivered by the end of 2021—to the 92 low- and middle-income countries as well as eight
additional self-ﬁnancing countries in the African Union.544 According to the White House, this
represented “the world’s largest-ever purchase and donation of vaccines by a single country.” In
September 2021, the U.S. committed to donating an additional 500 million vaccine doses. These
additional doses will be provided through COVAX to the same countries referenced in the June
announcement, according to administration oﬃcials.545

The U.S. government has also taken steps apart from COVAX to increase the availability of
COVID-19 vaccines abroad. These include the following actions:

• Bilateral vaccine donations. In addition to providing doses through COVAX, the U.S. has donated
millions of vaccines on a bilateral basis, beginning with doses it provided to Canada, Mexico,
South Korea, and Taiwan in June 2021. According to the Department of State, the U.S. will
continue to make donations as it receives more surplus supply.

• Support for local vaccine production. The U.S. has taken a number of steps to bolster
vaccine production overseas. For example, in March 2021, the President announced a
vaccine partnership with India, Japan, and Australia to support increased production
of COVID-19 vaccines for the Indo-Paciﬁc region. Additionally, in June 2021, the U.S.

541In this enclosure, “countries” refers to COVAX participant economies, not all 193 of which are independent states.
542According to Gavi, the 92 low- and middle-income countries relying on donor funding includes all countries with a
gross national income per capita of less than $4,000, based on 2018 and 2019 World Bank data, as well as countries
meeting other World Bank criteria for receiving donor assistance.
543Speciﬁcally, Congress directed that the funds be made available as a contribution to the GAVI Alliance. Pub. L. No.
116-260, Div. K, tit. IX, 134 Stat. 1182, 1822 (2020).
544The announcement stated that the remaining 300 million doses would be delivered by June 2022.
545The U.S. government supports COVAX in a variety of other ways as well. For example, USAID serves as a member on
the Gavi board and is the primary U.S. government liaison to COVAX; the Department of State has been diplomatically
involved in global fundraising eﬀorts for COVAX; and the Department of Health and Human Services sits on the ACT-
Accelerator Facilitation Council, which oversees COVAX.

Page 328 GAO-22-105051
International Development Finance Corporation announced that it would provide €600 million
(approximately $720 million) in joint ﬁnancing with other G7 development ﬁnance institutions
to support vaccine manufacturing in South Africa. According to the announcement, the
vaccines will be distributed primarily to the African Union, South Africa, and COVAX.546

Overview of Key Issues
COVAX has four primary lines of eﬀort, each led by one of the co-leading organizations—CEPI,
Gavi, and WHO—in collaboration and coordination with UNICEF, other organizations,
governments, civil society, and industry, according to COVAX (see table below).

546The administration has also expressed its support for waiving intellectual property rights on COVID-19 vaccines,
which may allow other countries to produce them domestically. See the enclosure on International Trade for more
details.

Page 329 GAO-22-105051
COVAX Key Lines of Eﬀort and Leadership
Key lines of eﬀort and Eﬀorts
lead organization

COVAX Facility and The COVAX Facility is the global procurement mechanism for COVID-19 vaccines through which
COVAX Advance Market COVAX is to ensure fair and equitable access to vaccines for all 193 participating countries. Gavi
Commitment (AMC) manages the facility, including overseeing the operation and governance structures, managing
relationships with participants, coordinating fundraising eﬀorts, and negotiating advance
Lead: Gavi
a
purchase agreements with manufacturers.
Guided by an allocation framework developed by WHO, the COVAX Facility distributes doses
to help protect the most at-risk groups in all participating countries, raising coverage levels as
evenly as possible across countries over time. A key goal of the framework is to ensure an initial
proportional allocation of doses to enable all participating countries to cover 20 percent of their
population. Self-ﬁnancing countries can request vaccine doses suﬃcient to vaccinate 10 to 50
percent of their populations. The amount they pay into the facility serves as an insurance policy
that increases their chances of securing vaccines, even if their own bilateral deals fail.
The COVAX AMC is the ﬁnancing mechanism that supports the participation of 92 low- and
middle-income countries in the COVAX Facility, enabling access to donor-funded doses of
b
safe and eﬀective vaccines. The primary focus of the COVAX AMC is to ensure that these 92
countries have equitable access to the vaccines, and at the same time, as higher-income, self-
ﬁnancing countries. Gavi coordinates the design, operation, and fundraising for the COVAX
AMC.

Development and CEPI invests in research and development (R&D) across a variety of promising vaccine
manufacturing candidates with the aim of supporting the ongoing development of safe and eﬀective COVID-19
vaccines that COVAX can make available to countries participating in the COVAX Facility. As
Lead: CEPI
of August 31, 2021, CEPI had invested $1.7 billion in its portfolio, which comprised at least 12
active candidate vaccines, including four targeting emerging variants. By that date, four of these
candidates had shown clinical eﬃcacy in preventing severe disease, of which two had received
c
WHO’s emergency use listing (i.e., qualiﬁed for emergency use).
In addition, CEPI secured ﬁrst right of refusal to potentially more than 1.7 billion doses for
the COVAX Facility through its R&D contracts, according to CEPI. CEPI also made strategic
investments in vaccine manufacturing, which included reserving dose manufacturing capacity
at a network of facilities, securing glass vials to hold 2 billion doses of vaccine, and securing
d
adjuvant supply. WHO provides R&D technical coordination. In work related to this line of
eﬀort, the COVAX Manufacturing Task Force, which COVAX established in May 2021, addresses
COVID-19 vaccine supply and manufacturing challenges.

Policy and allocations WHO provides guidance to member states on vaccine policy, regulation, safety, R&D,
allocation, and country readiness and delivery. For example, WHO develops evidence-based
Lead: WHO
immunization policy recommendations, issues emergency use listing of vaccines to enable
international procurement and facilitate authorization across member states, and provides
global coordination and member state support for vaccine safety monitoring.
WHO has conducted related work on vaccine access and allocation, including setting up an
allocation team with Gavi and designing an allocation algorithm and general governance
procedures that member states endorsed. The algorithm is continually being adjusted to
respond to, and account for, changing context, unstable supply, and new “features” of the
COVAX Facility (e.g., dose sharing, dose donations, cost-sharing).
The ﬁrst phase of the vaccine allocation started in January 2021, when doses became available
to the COVAX Facility. During this phase, a WHO–Gavi Joint Allocation Taskforce uses the
algorithm to determine the volume of vaccines that should be allocated to each country
participating in COVAX within a given period. An independent advisory group validates the
allocation decisions, which are endorsed by the Director-General of WHO. The procurement

Page 330 GAO-22-105051
Key lines of eﬀort and Eﬀorts
lead organization
agencies—UNICEF and PAHO—then use this information to prepare the shipments to
countries. Allocation decisions can change for various reasons, including lack of preparedness
at the country level or refusal of vaccines because of short shelf life.

Procurement and delivery In addition to administering the COVAX Facility and the COVAX AMC, Gavi makes vaccine
at scale procurement decisions and provides funding for, and oversight of, UNICEF’s and PAHO’s
Lead: Gavi in vaccine purchases on behalf of the facility. UNICEF, in collaboration with PAHO, leads eﬀorts to
collaboration with WHO, purchase and deliver doses of COVID-19 vaccines for COVAX to countries’ ports of entry. As the
UNICEF, and PAHO single largest vaccine buyer in the world, UNICEF leverages its expertise to purchase COVID-19
vaccine doses and manage their freight, logistics, and storage.
In addition, Gavi, UNICEF, and WHO are working with governments to ensure that countries are
ready to receive, distribute, and administer the vaccines with appropriate cold-chain equipment
(e.g., cold rooms, refrigerators, freezers, cold boxes, and vaccine carriers), syringes, vehicles,
health care workers trained to dispense vaccine, and other aspects of the delivery logistical
operations in place.

Source: GAO analysis of COVID-19 Vaccines Global Access (COVAX); Coalition for Epidemic Preparedness Innovations (CEPI); Gavi, the Vaccine Alliance (Gavi); World Health
Organization (WHO); United Nations Children’s Fund (UNICEF); and Pan American Health Organization (PAHO) documents. | GAO-22-105051

Note: “Countries” refers to COVAX participant economies, not all 193 of which are independent states. “COVAX partners” refers
to CEPI, Gavi, WHO, and UNICEF.
a
Because normal liability insurance is not available to manufacturers, each country receiving COVID-19 vaccines through the
COVAX Facility is required to indemnify manufacturers, donors, distributors, and other stakeholders against any losses they
incur from the deployment and use of the vaccines, according to COVAX documents.
b
The COVAX AMC includes a no-fault compensation mechanism, a key beneﬁt for lower-income countries, administered by
WHO. According to COVAX partners, this mechanism provides fair, no-fault, lump-sum compensation to any individual in the 92
low- and middle-income countries covered by the COVAX AMC who suﬀers a serious adverse event from any vaccine procured
or distributed through the COVAX Facility.
c
As of August 31, 2021, the Moderna, University of Oxford–AstraZeneca, Novavax, and Curevac COVID-19 vaccines had shown
clinical eﬃcacy in preventing severe disease, according to CEPI. Of these, the Moderna and University of Oxford–AstraZeneca
vaccines had received WHO’s emergency use listing.
d
An adjuvant is a component of a vaccine that helps the immune system response.

COVAX’s key goal for 2021 is to deliver 2 billion vaccine doses to COVAX participating
countries. To help end the acute phase of the pandemic, COVAX set a goal of delivering 2 billion
vaccine doses to COVAX participating countries—including at least 1.3 billion doses for low-
and middle-income countries—by the end of 2021. According to COVAX, these 2 billion doses
would cover at least 20 percent of populations in all participating COVAX countries and would be
suﬃcient to vaccinate the most at-risk population groups, including health care workers, elderly
adults, and people with underlying health conditions.

As late as June 2021, COVAX partners (i.e., CEPI, Gavi, WHO, and UNICEF) projected that this goal
could be met—and potentially exceeded—if COVAX raised at least $9.3 billion in total funding
commitments by June 2021.547 As of July 2021, COVAX partners said that they were in the process
of updating COVAX’s strategy and goals for 2022.

COVAX exceeded its fundraising target but was no longer projected as of September to
deliver 2 billion doses in 2021, primarily because of supply challenges. By June 2021, COVAX

547Commitments include funds pledged and funds provided.

Page 331 GAO-22-105051
had raised more than $9.6 billion in funding commitments, surpassing the June fundraising target
amount needed to purchase suﬃcient vaccine doses to achieve COVAX’s 2021 goal. According to
Gavi, as of September 2021 at least $6.7 billion of COVAX’s funding commitment pledges had been
fulﬁlled and as of August 2021 its total funding commitments had increased to about $9.8 billion.

However, our analysis of UNICEF data found that COVAX will likely not meet its goal of delivering
2 billion doses by the end of 2021.548 COVAX anticipated in April 2021 that nearly half a billion
doses would be shipped by the middle of 2021. Yet data that UNICEF reported on September
15, 2021, indicated that only about 279 million doses had been delivered through COVAX to
141 participating countries and that relatively few additional doses—about 4.1 million—were in
transit.549 Therefore, as of that date, COVAX was still 1.7 billion doses, or about 86 percent, short of
its goal of delivering 2 billion doses in 2021 (see ﬁgure). As of September 8, 2021, COVAX partners
expected to achieve this goal in the ﬁrst quarter of 2022.

COVAX Progress toward Goal of Delivering 2 Billion Vaccine Doses to Participating Countries in 2021, as of Sept.
15, 2021

COVAX and UNICEF data demonstrate that limited vaccine supply is the primary challenge
preventing COVAX from achieving its key goal. According to the September 8, 2021, COVAX Global
Supply Forecast (the most recent forecast), COVAX had agreements with 11 diﬀerent vaccine

548Data are from UNICEF’s COVID-19 Market Dashboard, which UNICEF regularly produces and processes using data
reported from sources it believes to be reliable.
549Of these approximately 279 million doses, about 213 million doses were delivered to low- and middle-income
countries. The approximately 66 million remaining doses were delivered to self-ﬁnancing countries.

Page 332 GAO-22-105051
developers to purchase up to 4.974 billion doses.550 However, UNICEF-reported data show that
as of September 15, manufacturers had made available fewer than 286 million doses for delivery
through COVAX and that COVAX had delivered 98 percent of these doses. Further, the September
8, 2021, COVAX Global Supply Forecast estimated that under the most likely scenario, only about
1.4 billion doses will be available to COVAX by the end of 2021, including doses it purchased
from manufacturers and donated doses (estimated on the basis of commitments from donor
countries).551 This represents a 26 percent decrease from the COVAX Global Supply Forecast on
July 27, 2021 (the previous forecast), which projected that 1.9 billion doses would be available to
COVAX by the end of 2021.

COVAX partners attributed the initiative’s limited vaccine supply to, among other factors, high-
income countries’ bilateral deals with manufacturers to purchase a signiﬁcant portion of the
available vaccine supply early in the pandemic. Although COVAX received funding pledges early
in the pandemic, it lacked adequate cash resources to purchase suﬃcient doses of the limited
vaccine supply, according to COVAX partners. In addition, COVAX partners said that bilateral
dose donations outside COVAX, manufacturing scale-up challenges, and export restrictions have
contributed to the limited supply of vaccine doses available to COVAX. For example, according
to COVAX and USAID documents as of September 2021, vaccine supply from a major COVAX
manufacturer, the Serum Institute of India, had been constrained since the spring of 2021, when
the Indian government instituted export restrictions in response to a spike of COVID-19 cases in
India.552

COVAX partners said that they had taken steps to try to increase vaccine production. For example,
COVAX developed a manufacturing task force to address delays caused by supply shortages and
manufacturing disruptions. According to COVAX partners, the task force brokers agreements
between stakeholders, supports technology transfers, and examines new manufacturing
opportunities. For example, in July 2021, the task force launched a marketplace to match critical
raw material suppliers with vaccine manufacturers in an eﬀort to accelerate the global production
of vaccine doses for COVAX. COVAX has also solicited donations of excess vaccine doses from
higher-income countries to increase the supply of vaccines available to lower- and middle-income
countries. WHO reported that as of September 17, 2021, COVAX had received pledges for about
722 million dose donations, of which about 114 million had been delivered.553

COVAX partners expect that vaccine supply will increase dramatically later this year. The COVAX
Global Supply Forecast on September 8, 2021, projected that of the 1.4 billion vaccine doses
expected to be available to COVAX by the end of 2021, approximately 1.1 billion doses would

550According to COVAX, these agreements include both committed-dose agreements, in which COVAX is required to
purchase a set number of doses, and optional-dose agreements, which give COVAX the option to purchase a set number
of doses. COVAX’s ability to purchase these vaccine doses is contingent on its ability to raise the funds required to fulﬁll
the purchase agreements.
551COVAX expects that approximately 1.2 billion of the 1.4 billion doses will be available for—but not necessarily
delivered to—low- and middle-income countries, covering about 20 percent of their populations.
552According to COVAX documents, COVAX contracted with the Serum Institute of India to manufacture up to about 1.6
billion vaccine doses (including both committed and optional-dose agreements) for countries participating in COVAX.
553This does not include dose donations announced after September 17, 2021, including the additional 500 million dose
donations the U.S. announced on September 22, 2021. COVAX partners accounted for donated vaccine pledges in their
September 8, 2021, statement that COVAX would not achieve its goal of delivering 2 billion vaccines by the end of 2021.

Page 333 GAO-22-105051
become available from September through December. However, as of September 2021, the extent
to which factors such as Delta and other variants and the increasing demand for booster shots
would aﬀect the limited supply remained unclear. Further, partners predict that once the supply
constraints are alleviated, other factors, such as countries’ abilities to absorb and administer the
vaccines, may signiﬁcantly aﬀect COVAX’s progress in meeting its goal.

COVAX partners have taken some steps to address other key challenges that may aﬀect
progress. COVAX documents and partners identiﬁed other key challenges, in addition to the
limited supply of vaccine doses, that may aﬀect progress in ending the pandemic, including
progress toward COVAX’s goal of delivering 2 billion doses in 2021. Some of these challenges, as
well as steps taken to respond to these challenges, are shown in the table below.

Page 334 GAO-22-105051
Other Key Challenges Aﬀecting COVAX Eﬀorts to Deliver Vaccines and Steps Taken to Address These Challenges,
as Reported by COVAX Partners
Challenge Steps taken

Earmarked vaccine donations. Some high-income COVAX partners are working with donor countries to try
countries, including the U.S., have donated vaccine doses to ensure that COVAX can determine a fair and equitable
to help COVAX progress toward its vaccination coverage distribution of COVID-19 vaccines. Additionally, WHO
goals. However, WHO oﬃcials indicated that some of is adjusting its allocation framework to account for
these donors have earmarked doses for speciﬁc countries. earmarked vaccine doses provided both bilaterally and
This earmarking works against COVAX’s goal of ensuring through COVAX.
equitable distribution of vaccine doses to COVAX countries.

High demand. Some COVAX AMC countries are asking for COVAX partners support higher vaccination coverage and
higher vaccine coverage than originally targeted through have increased COVAX’s fundraising targets accordingly.
COVAX. This high demand is based in part on the countries’ They are also developing a cost-sharing program for doses
desires to receive better protection against variants and that AMC countries request in excess of COVAX’s 2021
to keep pace with the vaccine rollout in high-income goal of making vaccines available for 20 percent of their
countries. Moreover, some high-income countries have populations. Additionally, COVAX partners are working with
begun to oﬀer booster shots to their own populations, the World Bank and other partners to ﬁnance additional
further increasing demand. vaccine doses for AMC countries.

Variation in countries’ health system capacity. COVAX COVAX and others have taken a number of steps to address
partners have reported wide variation in the capacity health system capacity. For example, as of July 2021,
of countries’ health systems to receive and administer COVAX had raised $799 million in pledges to contribute
COVID-19 vaccines on a large scale. For example, a World to the costs of delivering doses. Country governments
Bank report on country readiness between November can use these contributions to fund delivery logistical
2020 and February 2021 found that about 70 percent of operations, including the purchase of hardware such as
all countries surveyed did not have processes in place to cold-chain equipment. UNICEF and the World Bank also
train the large numbers of health care workers needed for separately provide funding for delivery costs. Additionally,
a mass vaccination campaign. UNICEF and the Pan American Health Organization are
oﬀering technical assistance and supporting delivery
logistical operations. As another example, UNICEF
and WHO have been providing technical assistance
for developing and implementing countries’ National
Deployment and Vaccination Plans. This assistance
includes providing support to train health care workers,
strengthen cold-chain and vaccine management capacity,
develop risk communication and community engagement
strategies, and prepare detailed plans to reach local target
populations.

Variants. Virus variants of concern, including the Delta Although COVAX partners’ assessment is that existing
variant, have higher transmissibility and may have higher COVID-19 vaccines have so far proven eﬀective against
morbidity and mortality rates than the original strain of the severe disease from known virus variants, COVAX partners
COVID-19 virus. It is also unclear whether existing vaccines have taken steps to address other potential challenges
may be less eﬀective against future variants, thereby from variants. For example, CEPI is supporting research to
potentially complicating eﬀorts to end the pandemic. develop vaccines that are more eﬀective against variants.
Additionally, COVAX partners are reevaluating existing
manufacturing contracts to allow the COVAX Facility
ﬂexibility to procure vaccines that will be eﬀective against
variants. COVAX partners are also conducting clinical
trials to determine whether mixing diﬀerent vaccine types
increases protection and are continuing to monitor clinical
trials on variants.

Page 335 GAO-22-105051
Challenge Steps taken

Vaccine hesitancy. COVAX partners have identiﬁed vaccine COVAX partners have engaged with policy makers and
hesitancy at both the country and individual levels as a regulators in individual countries to communicate to
signiﬁcant factor aﬀecting COVAX’s progress towards its the public that the beneﬁts of vaccines are greater than
target vaccination coverage. Some countries, including the risks. Additionally, COVAX partners are working to
certain countries within the COVAX AMC, are facing vaccine disseminate accurate information about vaccines and
demand issues related to product preferences. These expectations for vaccine doses and to unify messaging
preferences stem from perceptions regarding vaccine across COVAX. To address vaccine hesitancy among
safety; perceived performance against COVID-19, including individuals, COVAX partners have collaborated with local
variants; and views about the country of origin. Such champions to build public trust and encourage vaccinations
hesitancy increases the possibility that doses will expire by increasing the visibility of successful vaccinations.
before they can be administered. In turn, this increases
demand for other vaccines that are limited in supply.

Source: GAO analysis of Coalition for Epidemic Preparedness Innovations (CEPI); Gavi, the Vaccine Alliance (Gavi); World Health Organization (WHO); United Nations Children’s Fund
(UNICEF); and World Bank documents and interviews. | GAO-22-105051

Notes: “COVAX partners” refers to CEPI, Gavi, WHO, and UNICEF.
“Advance Market Commitment (AMC) countries” refers to 92 low- and middle-income economies that receive donor-
funded doses through COVID-19 Vaccines Global Access (COVAX). “Countries” refers to AMC economies, not all of which are
independent states.
COVAX is part of the Access to COVID-19 Tools (ACT) Accelerator, a larger multilateral eﬀort to combat the COVID-19 pandemic.
The ACT-Accelerator includes the Health Systems Connector, a crosscutting work area aimed at addressing in-country health
systems bottlenecks to facilitate country readiness.

Methodology
To conduct this work, we reviewed documents and data from CEPI, Gavi, WHO, and UNICEF
regarding COVAX’s structure and operations, the goals and status of COVAX eﬀorts, and steps
that COVAX partners have taken to address identiﬁed challenges. We also interviewed CEPI, Gavi,
WHO, and UNICEF oﬃcials knowledgeable about COVAX. We reviewed UNICEF data on the status
of COVAX vaccine supply and deliveries as of September 15, 2021. To assess the reliability of these
data, we performed logic tests and interviewed UNICEF about the reliability and limitations of the
data. We determined that the data were suﬃciently reliable for the purposes of this work.

In addition, we reviewed White House documents and interviewed, or obtained written responses
from, oﬃcials at the Department of Health and Human Services, the Department of State, and
USAID regarding COVAX, the bilateral provision of COVID-19 vaccine doses to other countries, and
other U.S. eﬀorts to increase the availability of the vaccine abroad. We also reviewed relevant laws,
including the Consolidated Appropriations Act, 2021.554

Agency Comments and Third-Party Views
We provided the Department of Health and Human Services, the Department of State, USAID,
the Oﬃce of Management and Budget (OMB), CEPI, Gavi, UNICEF, and WHO with a draft of this
enclosure. All except OMB and CEPI provided technical comments, which we incorporated as
appropriate.

554Pub. L. No. 116-260.

Page 336 GAO-22-105051
GAO’s Ongoing Work
We plan to conduct additional work to examine in-country readiness and delivery of COVID-19
vaccines abroad.

Contact information: Jason Bair, (202) 512-6881, bairj@gao.gov

Page 337 GAO-22-105051
International Trade
U.S. imports of COVID-19-related products, such as face masks, ventilators, gloves, and hand
sanitizers, have ﬂuctuated, while U.S. exports of COVID-19 vaccines to the European Union and the
United Kingdom have increased in recent months.

Background
The COVID-19 pandemic has disrupted businesses around the world as well as international
supply chains. According to the United Nations Conference on Trade and Development, world
merchandise trade grew by 11.6 percent from the ﬁrst quarter through the second quarter of
2021—a signiﬁcant improvement over the 21 percent quarterly decline from the ﬁrst through the
second quarter of 2020.555

U.S. trade of COVID-19-related products. U.S. imports of COVID-19-related products (e.g., face
masks, ventilators, gloves, and hand sanitizers) have ﬂuctuated. U.S. imports of products in
categories related to the COVID-19 response decreased slightly from March through June 2021.
However, imports of such products in June 2021 were 32 percent higher than in February 2020,
the last month before the transmission of COVID-19 started to become widespread in the U.S.,
according to the Centers for Disease Control and Prevention (CDC) (see ﬁgure).556

555For more information about overall trade trends in 2021, see United Nations Conference on Trade and Development,
“Global Merchandise and Services Trade Nowcast June 2021,” accessed Aug. 19, 2021, https://unctad.org/system/ﬁles/
oﬃcial-document/gdsdsimisc2021d4_en.pdf.
556U.S. Census Bureau trade statistics—a widely used source analyzing U.S. international trade—do not contain precise
data on imports of COVID-19-related products. As a result, we estimated the import value of all product types and
categories within those types using Harmonized Tariﬀ Schedule of the United States (HTS) statistical reporting numbers
and associated product groupings listed by the U.S. International Trade Commission (USITC) in COVID-19 Related Goods:
U.S. Imports and Tariﬀs, Investigation No. 332-576, USITC Publication 5073 (Washington, D.C.: June 2020). Revisions to
the HTS on July 1, 2020, and January 1, 2021, provided several new HTS-10 statistical reporting numbers for previously
identiﬁed COVID-19-related product categories. We identiﬁed these product categories and included them in our
analysis.

Page 338 GAO-22-105051
Monthly U.S. Imports of COVID-19-Related Products, by Product Type, Jan. 2018–June 2021

Note: U.S. Census Bureau trade statistics—a widely used source analyzing U.S. international trade—do not contain precise data
on imports of COVID-19-related products. As a result, we estimated the import value of all product types and categories within
those types using Harmonized Tariﬀ Schedule of the United States (HTS) statistical reporting numbers and associated product
groupings listed by the U.S. International Trade Commission (USITC). Examples of products included in the “Other” category
of product type include hospital beds and wheelchairs. See U.S. International Trade Commission, COVID-19 Related Goods: U.S.
Imports and Tariﬀs, Investigation No. 332-576, USITC Publication 5073 (Washington, D.C.: June 2020). Revisions to the HTS on
July 1, 2020, and January 1, 2021, provided several new HTS-10 statistical reporting numbers for previously identiﬁed COVID-19-
related product categories. We identiﬁed these product categories and included them in our analysis. Some HTS categories
represent more than one product, and some categories contain products that are not directly relevant to COVID-19 responses.
Product categories that USITC identiﬁed as COVID-19 related refer only to the subset of goods considered to be COVID-19
related in each HTS-10 statistical reporting number. Therefore, the values shown may overestimate the imports of products
directly relevant to COVID-19 responses. Nevertheless, the values shown are useful indicators for tracking import trends for
such products. For more information about factors inﬂuencing import trends in various types of COVID-19-related products,
see U.S. International Trade Commission, COVID-19 Related Goods: The U.S. Industry, Market, Trade and Supply Chain Challenges,
Investigation No. 332-580 (December 2020).

Total trends in import value are related to changes in both the quantity and price of the imported
goods. For example, the unit value of nasal swabs decreased by 30 percent from May 2020
through May 2021, while the quantity imported increased by 69 percent during the same period.
Overall, the import value of nasal swabs increased from $23 million in May 2020 to $28 million in
May 2021.

Many factors aﬀecting product availability, such as supply chain constraints, export restrictions,
and product demand, may drive trends in imports of COVID-19-related products. For example,
according to the Bureau of Labor Statistics (BLS), drastically reduced shipping capacity and urgent
demand for personal protective equipment pushed the costs of air freight transportation services
for shipments of goods from foreign countries to the U.S. to unprecedented heights in spring
2020. In addition, according to the United Nations Conference on Trade and Development, more
than 80 countries banned exports of medical and personal protective goods in the early phases of
the pandemic, and 60 percent of those restrictions remained in place as of June 11, 2021. Overall,
the need for medical supplies in response to the pandemic explains the increase in imports of
these products since early 2020. Fluctuations in the number of COVID-19 cases may shift the

Page 339 GAO-22-105051
demand for some COVID-19 products, such as pharmaceuticals and diagnostic equipment, over
time.

From March through June 2021, imports of COVID-19-related products from China—which
accounted for 11 percent of such imports overall in May—and from the rest of the world
ﬂuctuated. Speciﬁcally, imports of COVID-19-related products from China decreased by 26 percent
(from $2.75 billion to $2.05 billion) and total imports of such products from other countries
decreased by 12 percent (from $16.2 billion to $14.1 billion) from March through May 2021 before
increasing by 7 percent from May through June 2021.557 Previously, from December 2020 through
March 2021, imports of COVID-19-related products from China had risen by 12 percent (from $2.44
billion to $2.75 billion) and imports of such products from other countries had risen by 7 percent
(from $15.1 billion to $16.2 billion).558

Total U.S. trade of COVID-19 vaccines with European Union (EU) member countries and the United
Kingdom (UK) increased in recent months.559 From January through May 2021, the value of U.S.
COVID-19 vaccine exports more than doubled (from roughly $62.3 million to $125.3 million)
and the value of U.S. COVID-19 vaccine imports rose almost 11-fold (from $1.3 million to $15.3
million) (see ﬁgure). After a decline in the value of exports from January through April 2021,
U.S. export and import value of COVID-19 vaccines to EU countries and UK more than tripled
from April through May 2021. From April through May 2021, U.S. exports of these vaccines to EU
countries and the UK increased from $38.6 million to $125.3 million, and U.S. import values of
these vaccines increased from $3.6 million to $15.3 million. U.S. Census Bureau trade statistics
indicate that EU member countries and the UK exported the majority of vaccines most similar to
the COVID-19 vaccines imported by the U.S. from January through May 2021.560

557Some HTS categories identiﬁed in USITC Publication 5073 represent more than one product, and some categories
contain products that are not directly relevant to COVID-19 responses. Product categories that USITC identiﬁed as
COVID-19 related refer only to the subset of goods considered to be COVID-19 related for each HTS-10 statistical
reporting number. Therefore, the values we present may overestimate the imports of products directly relevant to
COVID-19 responses. Nevertheless, these values are useful indicators for tracking import trends of such products.
558Since 2018, certain imports from China have been subject to tariﬀs imposed by the Oﬃce of the U.S. Trade
Representative (USTR) at the direction of the President under Section 301 of the Trade Act of 1974. In response to the
COVID-19 pandemic, USTR excluded some of these tariﬀs on certain medical-care-related products. In March 2021, USTR
extended 99 tariﬀ exclusions for COVID-19-related products, such as personal protective equipment and other medical
care products, through September 30, 2021.
559We used statistics on international trade in goods published by the EU Commission’s Eurostat and Her Majesty’s
Revenue and Customs to identify U.S. imports and exports of vaccines against SARS-related coronaviruses (which we
refer to as COVID-19 vaccines) from EU member countries and the UK—speciﬁcally, trade recorded under the 3002.20.01
Harmonized System product category in these two databases
560According to the Census Bureau trade statistics, from January through June 2021, 73 percent of imports of vaccines
classiﬁed under the HTS reporting number (3002.20.0080) that includes the FDA-authorized COVID-19 vaccines came
from EU member countries or the United Kingdom. Products entering the U.S. under this HTS statistical reporting
number includes several other vaccines not related to COVID-19, such as shingles, whooping cough, and human
papillomavirus.

Page 340 GAO-22-105051
U.S. Trade of COVID-19 Vaccines with the European Union and United Kingdom, Jan.–May 2021

Note: We used international trade statistics on goods published by the European Union (EU) Commission’s Eurostat and Her
Majesty’s Revenue and Customs trade in goods statistics to identify imports and exports of vaccines against SARS-related
coronaviruses (which we refer to as COVID-19 vaccines) between the U.S. and EU member countries and between the U.S. and
the United Kingdom. We converted euros and pounds sterling into U.S. dollars using monthly exchange rate data from Federal
Reserve Economic Data. Speciﬁcally, we used the listed exchange rate on the earliest available day of data in each month as of
July 21, 2021, from https://fred.stlouisfed.org/series/EXUSEU and https://fred.stlouisfed.org/series/EXUSUK to convert euros
and pounds, respectively, to dollars. As a result, our estimates of COVID-19 vaccine trade ﬂows are dependent on the euro-
dollar and pound-dollar exchange rates in addition to prices and quantities. The values shown may be underestimated, because
the data for some EU member countries were not available. Because of rounding, numbers in columns may not sum to totals
shown.

U.S. exports of all vaccines for human use, including COVID-19 vaccines, averaged $128 million
per month in 2020 but totaled $820 million in May 2021 and $1.5 billion in June 2021. The price
per dose of the COVID-19 vaccines varied. We estimate that the per-dose cost to the federal
government for U.S. and EU-approved vaccines ranged from $4 (AstraZeneca) to $20 (Pﬁzer), on
the basis of reporting from the Congressional Research Service.561

The EU and UK governments have taken steps to regulate COVID-19 vaccine trade.

561We used data from a March 2021 Congressional Research Service report, detailing the number of COVID-19 vaccine
doses for which government agencies contracted with several vaccine manufacturers as well as the value of the
contracts, to estimate the value per dose of U.S.- or EU-approved COVID-19 vaccines. See table 1 in Congressional
Research Service, Operation Warp Speed Contracts for COVID-19 Vaccines and Ancillary Vaccination Materials, CRS Report
No. IN11560 (March 2021), accessed Sept. 7, 2021, https://crsreports.congress.gov/product/pdf/IN/IN11560.

Page 341 GAO-22-105051
• In 2020, the EU and UK governments each allowed exemptions from value-added taxes
on certain imports of COVID-19 vaccines to reduce supply chain constraints on acquiring
COVID-19 vaccines.

• On January 29, 2021, the European Commission of the EU issued a regulation imposing an
export authorization requirement on certain COVID-19 vaccines. On March 11, 2021, the
commission extended this requirement.562

• On March 24, 2021, the European Commission of the EU issued a regulation allowing
authorities to consider whether an export authorization would threaten the security of the
supply of speciﬁed vaccines and active substances used for their manufacture to EU member
countries.

• On June 29, 2021, the European Commission of the EU extended these regulations through
September 2021. According to the commission, these regulations are intended to ensure
timely access to COVID-19 vaccines for all EU citizens and to increase transparency regarding
vaccine exports to countries outside the EU.

EU member countries and the UK increased their global exports of COVID-19 vaccines from
January through April 2021. Exports of COVID-19 vaccines from EU member countries increased
from $180 million in January to $1.44 billion in April, while exports of COVID-19 vaccines from the
UK increased from around $1 million in January to almost $9 million in April. Overall, EU member
countries and the UK exported more than $2.8 billion worth of COVID-19 vaccines from January
through April 2021.

Meanwhile, the U.S. government has taken actions to increase the availability of COVID-19 vaccines
to other countries.

• In July 2021, we reported that the federal government had taken several steps to increase
domestic production of COVID-19 vaccines. An increase in domestic supply, along with a
leveling oﬀ in domestic demand for vaccines, could create more opportunities for COVID-19
vaccine exports.

• The Oﬃce of the U.S. Trade Representative stated in May 2021 that it would begin negotiations
at the World Trade Organization to facilitate waiving intellectual property rights on COVID-19
vaccines, which may allow other countries to produce them domestically.563

• The FDA has allowed AstraZeneca COVID-19 vaccines—which is produced, but not authorized
for use, in the U.S.—to be exported to other countries.

562Generally, under the regulation, the export authorization shall be granted unless the exports pose a threat to
the execution of the EU advance purchase agreements concluded with vaccine manufacturers. These agreements
commit vaccine producers to deliver a set amount of vaccines to EU member countries, according to the European
Commission.
563USTR stated that it plans to negotiate for the intellectual property rights waiver at the World Trade Organization.
A number of countries have yet to support the waiving intellectual property rights for COVID-19 vaccines, and many
multinational pharmaceutical companies oppose the measure.

Page 342 GAO-22-105051
• The U.S. has provided funding, donated surplus doses, and taken other steps in support
of eﬀorts to provide vaccines to other countries, including those that cannot aﬀord to buy
vaccines themselves. (For more information, see the Vaccines Provided Abroad enclosure in
app. I.)

Methodology
To conduct this work, we reviewed the most recent publicly available U.S. trade statistics from
the Census Bureau as well as U.S. International Trade Commission data on product categories
that contain COVID-19-related products.564 We analyzed EU and UK trade data from the European
Commission and Her Majesty’s Revenue and Customs that were available as of July 19, 2021, to
study trends in exports and imports of vaccines against SARS-related coronaviruses (i.e., COVID-19)
from and to EU member countries and the UK.565 We converted euros and pounds sterling into
U.S. dollars, using monthly exchange rate data from Federal Reserve Economic Data.566

We determined that the trade data were suﬃciently reliable for our reporting purposes.
According to an EU Commission report on the quality of EU trade statistics, data on imports from,
and exports to, countries outside the EU from 2016 through 2019 were likely fully accounted
for because they rely on customs declarations.567 Her Majesty’s Revenue and Customs also
conducts several validation and credibility checks of its trade data to ensure their accuracy before
publication.

Agency Comments
We provided a draft of this enclosure to the Oﬃce of Management and Budget, which had no
comments on this enclosure.

564We compared COVID-19-related HTS-10 codes before and after July 1, 2020, and before and after January 1, 2021.
If we found no match, we checked USITC guidance to determine whether the original code had been annotated or
discontinued. If it had been annotated or discontinued, we included imports of those codes after July 1, 2020, or January
1, 2021, in our analysis. For instance, according to guidance provided by the USITC, products under 4818.50.0000 were to
be split into two new HTS-10 product categories, 4818.50.0080 and 4818.50.0020, on July 1, 2020. Therefore, we included
imports for products contained in 4818.50.0080 and 4818.50.0020 after July 1, 2020, in our calculations.
565Companies may produce FDA-approved COVID-19 vaccines in other countries before shipping to importers in the
United States. For example, in September 2020, Pﬁzer announced that it was producing its COVID-19 vaccine at its
production plants in Belgium and that BioNTech’s production would take place in Germany.
566Speciﬁcally, to convert euros and pounds, respectively, to dollars, we used the exchange rate on the ﬁrst day in each
month, shown at https://fred.stlouisfed.org/series/EXUSEU and https://fred.stlouisfed.org/series/EXUSUK as of August
10, 2021.
567Eurostat, Quality Report on European Statistics on International Trade—2016-2019 Data: 2020 Edition, Eurostat Statistical
Reports (European Commission: Dec. 17, 2020), accessed June 9, 2021, https://ec.europa.eu/eurostat/en/web/products-
statistical-reports/-/ks-ft-20-008.

Page 343 GAO-22-105051
GAO’s Ongoing Work
We will continue to monitor U.S. trade of COVID-19-related products and COVID-19 vaccines.

Related GAO Products
COVID-19: Eﬀorts to Increase Vaccine Availability and Perspectives on Initial Implementation.
GAO-21-443. Washington, D.C.: April 14, 2021.

Contact information: Kimberly Gianopoulos, (202) 512-8612, gianopoulosk@gao.gov

Page 344 GAO-22-105051
Federal Fraud-Related Cases
Federal agencies’ enforcement actions on fraud-related charges help protect consumers and
ensure that taxpayer dollars and government services related to COVID-19 serve their intended
purposes.

Entities involved: Government-wide

Background
The public health crisis, economic instability, and increased ﬂow of federal funds associated with
the COVID-19 pandemic present increased pressures and opportunities for fraud.568 By proactively
managing fraud risks, federal oﬃcials can help safeguard taxpayer dollars to ensure they serve
their intended purpose, particularly given that Congress had appropriated about $4.8 trillion as of
August 31, 2021, to fund COVID-19 response and recovery eﬀorts.569

According to GAO’s A Framework for Managing Fraud Risks in Federal Programs, among other things,
eﬀective managers of fraud risks refer instances of potential fraud to Oﬃces of Inspector General
(OIG) or other appropriate parties, such as law enforcement entities or the Department of Justice
(DOJ), for further investigation.

The extent of fraud associated with the COVID-19 relief funds appropriated to date has not
yet been determined. One of the many challenges is that because of fraud’s deceptive nature,
programs can incur ﬁnancial losses related to fraud that are never identiﬁed, and such losses are
diﬃcult to reliably estimate. However, many individuals have already pleaded guilty to federal
charges of defrauding COVID-19 relief programs—including the Small Business Administration’s
(SBA) Paycheck Protection Program (PPP) and Economic Injury Disaster Loan (EIDL) program,
the Department of Labor’s (DOL) unemployment insurance (UI) programs, and economic impact
payments (EIP) issued by the Department of the Treasury and Internal Revenue Service.570 Also,
one individual has been convicted at trial of PPP-related fraud, and four individuals have been
convicted at trial of fraud related to both the PPP and EIDL program.571 In addition, numerous
others faced related federal charges as of July 31, 2021.

Further, federal hotlines have received numerous complaints from the public, many of them
alleging potential fraud involving COVID-19 relief funds. For example, from March 2020 through

568Fraud and “fraud risk” are distinct concepts. Fraud—obtaining something of value through willful
misrepresentation—is challenging to detect because of its deceptive nature. Fraud risk (which is a function of likelihood
and impact) exists when individuals have an opportunity to engage in fraudulent activity, have an incentive or are under
pressure to commit fraud, or are able to rationalize committing fraud. Fraud risk management is a process for ensuring
program integrity by continuously and strategically mitigating the likelihood and impact of fraud. When fraud risks can
be identiﬁed and mitigated, fraud may be less likely to occur. Although the occurrence of fraud indicates there is a fraud
risk, a fraud risk can exist even if actual fraud has not yet been identiﬁed or occurred.
569An appropriation provides legal authority for federal agencies to incur obligations and make payments out of the U.S.
Treasury for speciﬁed purposes.
570In July 2021, we reported that SBA’s initial implementation of PPP contributed to increased risk of improper
payments and extensive fraud. Also in July 2021, we reported on eﬀorts SBA has taken to address risks of fraud in the
EIDL program and provision of funds to ineligible applicants.
571We consider convictions to be cases where an individual was convicted of a fraud-related charge at trial.

Page 345 GAO-22-105051
July 2021, our hotline—known as FraudNet—received over 2,700 complaints related to the CARES
Act, many of which involve SBA’s PPP and EIDL program, DOL’s UI program, and EIPs (see text
box).572

Report Fraud, Waste, and Abuse
GAO’s FraudNet supports accountability across the federal government. Allegations of fraud, waste, or abuse
can be submitted via the FraudNet portal or by calling the hotline at 1-800-424-5454.

Source: GAO | GAO-22-105051

In addition to fraud against federal programs, scammers are also targeting consumers, which
can result in ﬁnancial losses and undermine health and safety. For example, the Federal Trade
Commission (FTC) is tracking complaints related to COVID-19 fraud against consumers. According
to FTC reporting, the agency had received over 347,000 reports about fraud and over 69,000
reports about identity theft as of August 4, 2021.573 Also according to FTC reporting, over a third
of the reports about fraud indicated a ﬁnancial loss and cost Americans over an estimated $518
million.

Overview of Key Issues
Since March 2020, DOJ has publicly announced charges in numerous fraud-related cases involving
COVID-19 relief programs, COVID-19-related consumer fraud schemes, or other types of fraud
related to COVID-19.574 The charges—ﬁled across the U.S. and investigated by a range of law
enforcement agencies—include making false statements and engaging in identity theft, wire and
bank fraud, and money laundering.575 The number of individuals facing fraud-related charges has
continued to grow in the past year and will likely increase, as these cases take time to develop.576

572While not all of the complaints received involve allegations of potential fraud, many of them do.
573According to FTC, the fraud reports reﬂect complaints in the Consumer Sentinel Network that mention COVID,
stimulus, N95, and related terms. The identity theft reports reﬂect complaints that mention COVID, stimulus, or related
terms in the following identity theft subtypes: tax, employment and wage, government beneﬁts, and government
documents.
574A charge is merely an allegation, and all defendants are presumed innocent until proven guilty beyond a reasonable
doubt in a court of law.
575The federal government may enforce laws through civil or criminal action. Such action may be resolved through a
trial, a permanent injunction, a civil settlement, or a guilty plea. Since March 2020, DOJ has resolved three PPP fraud-
related cases involving three individuals and their respective companies through a civil settlement. For example, in
one case, a company and its sole individual owner agreed to pay over $230,000 to settle civil fraud allegations that the
company obtained multiple PPP loans after the owner certiﬁed that the company would not receive multiple loans. In
addition to the federal government, state governments have brought COVID-19-related unemployment insurance fraud
charges.
576The statute of limitations for mail fraud and wire fraud prosecutions is 5 years (18 U.S.C. § 3282), except for mail
and wire fraud schemes that aﬀect a ﬁnancial institution, in which case the statute is 10 years (18 U.S.C. § 3293). Also,
based on our analysis, these cases can take many years to resolve. For example, the Department of Housing and Urban
Development OIG closed cases in 2017–2020 resulting from Hurricane Sandy in 2012.

Page 346 GAO-22-105051
Fraud against federal programs. From March 2020 through July 2021, 198 individuals pleaded
guilty to federal charges of defrauding COVID-19 relief programs, including SBA’s PPP and EIDL
program, DOL’s UI programs, and EIPs.577 For example:

• In one case, an individual was sentenced to more than 11 years in prison after fraudulently
seeking over $24 million in PPP loans. This individual pleaded guilty to wire fraud and money
laundering and admitted that he had fraudulently sought PPP loans on behalf of nearly a
dozen diﬀerent companies. Speciﬁcally, this individual admitted to making several false and
misleading statements about the companies’ respective payroll expenses and submitting
fraudulent documentation in support of his applications when, in fact, none of the businesses
had employees or paid wages consistent with the amounts claimed in the applications. Based
on this individual’s false statements and documentation, lenders paid out over $17 million.
The individual used the proceeds for personal expenses including purchasing multiple homes,
paying oﬀ mortgages, and purchasing multiple luxury cars. In addition to the prison sentence,
this individual was ordered to pay over $17 million in restitution. See the enclosure on the
Paycheck Protection Program in appendix I for more information on the program.

• As part of another case, an individual pleaded guilty to wire fraud associated with a scheme
to defraud the EIDL program. The individual submitted an EIDL application on behalf of a
business he owned, certifying that he would use the loan proceeds solely as working capital to
alleviate economic injury caused by disaster. After obtaining a loan for $150,000, the individual
fraudulently misapplied the loan proceeds by purchasing a boat for his personal use and by
paying for other expenditures that were in violation of EIDL rules and regulations.

• Three individuals pleaded guilty to wire fraud associated with a scheme to defraud UI by using
information belonging to other people, including prison inmates, to ﬁle for pandemic-related
unemployment beneﬁts, falsely stating that these other people had lost their jobs because of
the COVID-19 pandemic. In exchange for cash payments, one of these individuals provided
third parties with the debit cards loaded with UI beneﬁts. Each of these individuals caused the
state unemployment agency to disburse approximately $353,000 to $477,000 in fraudulently
obtained UI beneﬁts. For more information on potential fraud in the UI programs, see the
enclosure on Unemployment Insurance Fraud Risk Management, and for more information on
the UI programs, see the enclosure on Unemployment Insurance Programs in appendix I.

• One individual pleaded guilty to a charge of conspiracy to commit bank fraud in connection
with EIPs. This case involved the individual and his conspirators creating counterfeit checks,
including counterfeit pandemic relief checks; depositing the checks online and at various
bank ATMs; and withdrawing the funds before the victim institutions identiﬁed the checks as
fraudulent.

577One of these individuals pleaded guilty to federal charges of defrauding the Coronavirus Food Assistance Program.
Some of the 198 individuals pleaded guilty to charges related to more than one federal program. Speciﬁcally, 25
individuals pleaded guilty to federal charges related to both PPP and EIDL, two individuals pleaded guilty to federal
charges related to both PPP and UI, three individuals pleaded guilty to federal charges related to both EIDL and UI,
and one individual pleaded guilty to federal charges related to PPP, EIDL, and UI, as of July 31, 2021. Forty-nine of the
198 individuals had been sentenced as of July 31, 2021. Sentences ranged from time served and 2 years of supervised
release to more than 11 years in prison and restitution of over $17 million.

Page 347 GAO-22-105051
As of July 31, 2021, ﬁve individuals had been convicted at trial for COVID-19 relief fraud.578 For
example, a federal jury found four individuals guilty of conspiracy to commit bank fraud and
wire fraud, 11 counts of wire fraud, eight counts of bank fraud, and conspiracy to commit money
laundering for obtaining more than $18 million in PPP and EIDL COVID-19 relief funds. These
individuals used fake, stolen, and synthetic identities to submit fraudulent applications for PPP
and EIDL loans, submitted false and ﬁctitious documents to lenders and SBA, and then used the
fraudulently obtained funds as down payments on luxury homes and to purchase other luxury
items.579

Federal charges were pending against 465 individuals for attempting to defraud COVID-19 relief
programs as of July 31, 2021.580

Consumer fraud. In addition to fraud against federal programs, fraud can result in ﬁnancial
losses to consumers and undermine health and safety. From March 2020 through July 2021,
15 individuals or entities pleaded guilty to federal charges related to consumer fraud.581 For
example, in one case, a company and its owner pleaded guilty to charges related to the unlawful
importation, sale, and mailing of an unregistered pesticide product marketed as a killer of airborne
viruses such as COVID-19. In another case, an individual devised a scheme to defraud an investor
and obtain money under fraudulent pretenses related to the purchase and resale of personal
protective equipment (PPE) to hospitals and retailers in response to the COVID-19 pandemic. This
individual pleaded guilty to wire fraud and money laundering.

There were also federal charges pending against 55 individuals or entities related to consumer
fraud as of July 31, 2021. For example, an individual was indicted for allegedly defrauding a
company that developed and manufactured COVID-19 test kits by falsely representing that his own
company was certiﬁed to perform testing, was an “end user” of the tests, and would not attempt

578One of these individuals had been sentenced as of July 31, 2021. This individual was sentenced to 37 months in
federal prison and 3 years of supervised release and ordered to pay restitution of $2,068,700 for PPP-related fraud. As of
July 31, 2021, there had not been any convictions related to UI, EIP, or other federal COVID-19 relief programs.
579One of these individuals was also found guilty of two counts of aggravated identity theft. Another individual was also
found guilty of one count of aggravated identity theft. A third individual was also found guilty of one count of money
laundering. Two of the individuals were charged with violation of pretrial release in late August 2021 after allegedly
removing their tracking bracelets, and bench warrants have been issued for their arrests.
580The majority of these individuals were charged with attempting to defraud SBA’s PPP and EIDL program, DOL’s
UI programs, or EIPs. Three individuals were charged with attempting to defraud only other federal COVID-19 relief
programs, including the Higher Education Emergency Relief Fund and the Provider Relief Fund. Also, some of these 465
individuals faced charges related to more than one federal program. Speciﬁcally, 55 individuals faced federal charges
related to both PPP and EIDL, eight individuals faced federal charges related to both EIDL and UI, and one individual
faced federal charges related to the Provider Relief Fund and EIDL as of July 31, 2021. Three individuals faced federal
charges related to PPP, EIDL, and UI; one individual faced federal charges related to PPP, the Accelerated and Advance
Payments Program, and the Provider Relief Fund; one individual faced charges related PPP, EIDL, and EIP; and one
individual faced charges related to PPP, EIDL, and Medicare.
581One of the 15 individuals or entities has also pleaded guilty to federal charges of defrauding a COVID-19 relief
program. Also, 11 of the 15 individuals had been sentenced as of July 31, 2021. Sentences for individuals ranged from
1 year of probation to more than 5 years in prison followed by 5 years of supervised release. In addition to these 15
individuals or entities, since March 2020, DOJ resolved a civil complaint of consumer fraud against one entity through a
civil settlement.

Page 348 GAO-22-105051
to resell them. This individual advertised online that he was selling “Food and Drug Administration-
approved COVID-19 test kits” and attempted to sell the tests at a substantial markup.582

As of July 31, 2021, the majority of individuals and entities that had pleaded guilty or faced federal
charges for COVID-19-related consumer fraud schemes were allegedly involved in schemes related
to prevention or treatment or PPE sales (see ﬁgure).

Number of Individuals or Entities Who Have Pleaded Guilty to or Faced Federal Charges for Consumer Fraud, as
of July 31, 2021

a
The “other” category can include individuals or entities who engaged in deceptive business practices, making COVID-19-related
claims that caused consumers to suﬀer or potentially suﬀer ﬁnancial or other losses unrelated to prevention or treatment,
personal protective equipment, or testing. For example, one individual pleaded guilty and was sentenced to 63 months in
prison for, among other things, creating a website to sell other goods that were sought after during the pandemic and not
delivering those goods.

Further, as COVID-19 vaccines have become available, potential consumer fraud related to
vaccines has emerged. While the extent of vaccine-related fraud is unknown, DOJ has publicly
announced charges or other actions in consumer fraud cases involving individuals or entities
that claimed to oﬀer vaccines to prevent COVID-19. For example, DOJ announced charges against
three individuals for allegedly creating a fraudulent replication of a vaccine company’s website and
stating that consumers could buy COVID-19 vaccines. Further, DOJ announced charges against
another individual for selling immunization pellets that this individual allegedly falsely claimed
would provide lifelong immunity to COVID-19. This individual allegedly created and provided
counterfeit COVID-19 vaccination cards to make it appear that customers received an authorized
COVID-19 vaccine.

Other federal cases. The federal government is also pursuing charges including conspiracy,
wire fraud, and theft that are related to COVID-19 but separate from consumer fraud—including
vaccine-related fraud—and fraud against the federal programs discussed earlier. From March 2020

582This individual was also accused of fraudulently obtaining a loan from SBA’s EIDL program.

Page 349 GAO-22-105051
through July 2021, 14 individuals pleaded guilty to these types of federal charges.583 For example,
one individual was sentenced to 46 months in prison and 3 years of supervised release, ordered to
pay over $530,000 in restitution, and ordered to forfeit $527,000 and a vehicle after pleading guilty
to embezzling more than $500,000 from AmeriCorps and to agreeing to accept a bribe for the
administration of grants under the CARES Act. Another individual pleaded guilty to hoarding and
price gouging, admitting that she began accumulating nearly 20,000 N95 respirator masks starting
in February 2020 in anticipation of a shortage that would be caused by the COVID-19 pandemic,
and that she sold the masks for up to $15 each.

There were also other federal charges pending against 29 individuals as of July 31, 2021. For
example, one individual was indicted on charges for his role in a $9.3 million health care kickback
scheme. Among other things, this individual was charged with conspiracy to commit health care
fraud and wire fraud for allegedly conspiring with others to pay and receive kickbacks in exchange
for referring Medicare beneﬁciaries to his clinical laboratory for testing that they did not need.
Speciﬁcally, among other things, this individual submitted at least $260,000 in claims related to
medically unnecessary testing that was improperly bundled with COVID-19 testing.584

Other federal eﬀorts to address and prevent future fraud-related cases. In addition to DOJ
ﬁling fraud-related charges against individuals and entities for actions related to COVID-19, other
federal agencies have undertaken eﬀorts to address fraud-related cases and prevent such cases in
the future.

Taking administrative and other enforcement actions. FTC and the Food and Drug Administration
have issued warning letters to companies for allegedly selling fraudulent COVID-19-related
products, including those making deceptive or scientiﬁcally unsupported claims about their
ability to prevent or treat COVID-19. In addition, FTC has sought civil penalties against individuals
and entities for allegedly engaging in deceptive practices related to the COVID-19 pandemic.585
In one case, FTC brought action against two companies and two oﬃcers of those companies
in connection with the labeling, advertising, marketing, distribution, and sale of products they
claimed would treat, prevent, or cure COVID-19. One of the oﬃcers agreed to settle the charges
and is barred from making similar unsupported health claims in the future and was ordered to
provide refunds to defrauded consumers. In another case, the Environmental Protection Agency

583Eight individuals had been sentenced as of July 31, 2021. Sentences ranged from 2 years of probation and a $75,000
ﬁne to almost 4 years in prison, 3 years of supervised release, and an order to pay over $530,000 in restitution and to
forfeit $527,000 and a vehicle. In addition to these eight individuals, since March 2020, DOJ resolved complaints against
three individuals or entities for activities such as hoarding and price gouging through settlements.
584This individual was also indicted on wire fraud charges for allegedly submitting fraudulent applications seeking a
PPP loan and an EIDL in the name of his clinical laboratory. In addition, two other individuals were charged separately
for their roles in the $9.3 million health care kickback scheme. In May 2021, DOJ announced criminal charges against
14 defendants, including these three individuals, for their alleged participation in various health care fraud schemes
resulting in over $143 million in false billings. In September 2021, DOJ announced charges against ﬁve defendants who
allegedly engaged in the misuse of Provider Relief Fund monies and nine defendants who allegedly engaged in various
health care fraud schemes designed to exploit the COVID-19 pandemic, resulting in the submission of over $29 million in
false billings.
585The COVID-19 Consumer Protection Act, which became law in December 2020 as part of the Consolidated
Appropriations Act, 2021, gives FTC authority to seek civil penalties on the ﬁrst oﬀense for scams and deceptive practices
related to the COVID-19 pandemic. Pub. L. No. 116-260, div. FF, tit. XIV, § 1401, 134 Stat. 1182, 3275-3276 (2020).

Page 350 GAO-22-105051
ordered a website to stop selling pesticides, such as a “disinfection card” that claims to protect the
wearer from coronavirus.

Establishing task forces and working groups. On May 17, 2021, the Attorney General established
the COVID-19 Fraud Enforcement Task Force to marshal the resources of DOJ in partnership with
agencies across the government.586 The task force held its ﬁrst meeting on May 27, 2021, where
task force members discussed priority goals, including increased eﬀorts to combat fraud related to
COVID-19 relief programs. Also on May 17, 2021, the President established the Initiative on Identity
Theft Prevention and Public Beneﬁts to develop recommendations on prevention measures to
respond to the threat of identity theft in public beneﬁts programs. Through the initiative, the
White House’s American Rescue Plan Implementation Coordinator and the Oﬃce of Management
and Budget (OMB) are working with relevant stakeholders to develop recommendations and
proposals.

Further, on July 15, 2021, the Chair of the Pandemic Response Accountability Committee
announced the formation of a new working group focused on preventing and addressing identity
fraud in pandemic response programs. The working group is a joint eﬀort of multiple agency
Inspectors General. In addition, the Federal Bureau of Investigation Springﬁeld Field Oﬃce
organized a PPP and EIDL fraud working group consisting of its federal partners—the Internal
Revenue Service Criminal Investigations and the SBA OIG.

Providing information to the public about emerging fraud schemes. As a result of complaints from the
public alleging potential fraud involving COVID-19 relief funds received through hotlines and other
fraud detection eﬀorts, federal agencies have warned the public about emerging fraud schemes,
which can help prevent future cases of fraud against federal programs and consumers. For
example, in March 2021, DOL launched a website to help the public better understand UI identity
theft. The website also provides resources for those who may have been victims of identity theft,
including a list of contact information for each state to report UI identity theft.587 We previously
reported on examples of agency warnings to the public about emerging fraud schemes in our
quarterly July 2021 report.

Methodology
To conduct this work, we reviewed information from DOJ to identify federal fraud-related charges
related to COVID-19 relief funding as of July 31, 2021. We also analyzed related federal court

586According to a related press release, the task force will augment and incorporate the existing coordination
mechanisms within DOJ. Also, it will work closely with interagency partners to share information and insights gained
from prior enforcement experience to reduce the potential threat to the American people and COVID-19 relief. In
addition, it will help agencies increase their fraud prevention eﬀorts by providing information about fraud trends and
illicit tactics. Further, it will bolster eﬀorts to investigate and prosecute the most culpable domestic and international
criminals, prevent the exploitation of government assistance for personal and ﬁnancial gain, and recover stolen funds.
587Oﬃcials from the National Association of State Workforce and its UI Integrity Center, funded by and operated
in partnership with DOL, said that identity theft remains the biggest challenge for states in addressing potential UI
fraud. According to oﬃcials, states have recently experienced increases in instances of criminals taking over legitimate
claimants’ UI accounts and rerouting beneﬁts to other bank accounts. In January 2021, we reported that states were
working to address these account takeovers through communication campaigns that raise public awareness about
phishing attempts to steal account information, as well as coordinating with law enforcement and banking institutions.

Page 351 GAO-22-105051
documents. In addition, we reviewed FTC reports on complaints related to fraud and identity theft
and press releases from other federal entities, including the Pandemic Response Accountability
Committee, FTC, DOL, and the Environmental Protection Agency, describing COVID-19 fraud-
related eﬀorts.

Agency Comments
We provided OMB with a draft of this enclosure. OMB provided technical comments, which we
incorporated as appropriate.

GAO’s Ongoing Work
We will continue our oversight of government-wide fraud risk management eﬀorts.

Related GAO Products
Economic Injury Disaster Loan Program: Additional Actions Needed to Improve Communication with
Applicants and Address Fraud Risks. GAO-21-589. Washington, D.C.: July 30, 2021.

Paycheck Protection Program: SBA Added Program Safeguards, but Additional Actions Are Needed.
GAO-21-577. Washington, D.C.: July 29, 2021.

A Framework for Managing Fraud Risks in Federal Programs. GAO-15-593SP. Washington, D.C.: July 28,
2015.

Contact information: Rebecca Shea, (202) 512-6722, shear@gao.gov

Page 352 GAO-22-105051
FEMA’s Disaster Relief Fund and Assistance to State, Local,
Tribal, and Territorial Governments
The Federal Emergency Management Agency does not consistently interpret and apply COVID-19
Public Assistance guidance and provide timely approval of COVID-19 applications for Public
Assistance, causing undue burdens and stress for communities.

Entity involved: Federal Emergency Management Agency, within the Department of Homeland
Security

Recommendations for Executive Action
The Federal Emergency Management Agency Administrator should improve the consistency of the
agency’s interpretation and application of the COVID-19 Public Assistance policy within and across
regions by further clarifying and communicating eligibility requirements nationwide.

The Federal Emergency Management Agency Administrator should require the agency’s Public
Assistance program employees in the regions and at its Consolidated Resource Centers to attend
training on changes to COVID-19 Public Assistance policy to help ensure it is interpreted and
applied consistently nationwide.

The Department of Homeland Security concurred with both recommendations and outlined
actions it has taken to improve the consistency of the agency’s interpretation and application of
the COVID-19 Public Assistance Policy and train employees in the regions and at its Consolidated
Resource Centers.

Background
Before the COVID-19 pandemic, the Federal Emergency Management Agency’s (FEMA) Disaster
Relief Fund—the primary source of federal disaster assistance for state, local, tribal, and territorial
governments—had never been used during a nationwide public health emergency.588 As of August
31, 2021, FEMA had obligated more than $80 billion from the Disaster Relief Fund to respond to
COVID-19. As of the same date, the Disaster Relief Fund’s balance was more than $40 billion.

588The Disaster Relief Fund receives an annual appropriation and has routinely received supplemental appropriations.
In March 2020, the CARES Act was enacted, appropriating $45 billion for the Disaster Relief Fund. Pub. L. No. 116-136,
div. B, tit. VI, 134 Stat. 281, 543 (2020). The Consolidated Appropriations Act, 2021 appropriated $17 billion to the
Disaster Relief Fund for major disasters and an additional $2 billion to provide assistance for COVID-19-related funeral
expenses. Pub. L. No. 116-260, div. F, tit. III, 134 Stat. 1182, 1462 (2020); Pub. L. No. 116-260, div. M, tit. II, 134 Stat. at
1910. In March 2021, the American Rescue Plan Act of 2021 appropriated $50 billion to the Disaster Relief Fund. Pub. L.
No. 117-2, § 4005, 135 Stat. 4, 79.

Page 353 GAO-22-105051
FEMA’s Disaster Relief Fund Balance, by Month, Feb. 2020– Aug. 2021

FEMA has used the Disaster Relief Fund to respond to the COVID-19 pandemic by providing three
primary types of disaster assistance.

1. Individual Assistance. FEMA provides Individual Assistance to disaster survivors to cover
necessary expenses and serious needs—such as housing assistance, counseling, or funeral
assistance—which cannot be met through insurance or low-interest loans. For the COVID-19
response, FEMA has provided lost wages assistance, funeral assistance, and crisis counseling.

2. Public Assistance. FEMA provides Public Assistance to state, local, tribal, and territorial
governments, and certain types of private nonproﬁt organizations so that communities can
quickly respond to, and recover from, major disasters or emergencies. After natural disasters,
Public Assistance tends to be used for emergency cleanup and for permanent reconstruction
projects—for example, to rebuild damaged public infrastructure. For all 59 major disaster
declarations for COVID-19, FEMA has authorized Public Assistance for emergency protective
measures only. This may include eligible medical care, purchase and distribution of food,
noncongregate medical sheltering, operation of Emergency Operations Centers, and the
purchase and distribution of personal protective equipment.
On February 2, 2021, the President issued a memorandum that directed FEMA to fully
reimburse state, territorial, and tribal governments for all work eligible for emergency
protective measures assistance through September 30, 2021.589 On August 17, 2021, the
President extended this full reimbursement through December 31, 2021.590 According to
FEMA oﬃcials, FEMA’s Public Assistance Workforce, which manages FEMA’s Public Assistance

589White House, Memorandum on Maximizing Assistance from the Federal Emergency Management Agency (Feb. 2,
2021).
590White House, Memorandum on Maximizing Assistance to Respond to COVID-19 (Aug. 17, 2021).

Page 354 GAO-22-105051
workload, consists of approximately 1,800 employees, of which more than 1,000 have
supported the COVID-19 eﬀort in some capacity from their primary duty station.

3. Mission assignments. FEMA also issues mission assignments—work orders directing
other federal agencies to provide direct assistance to state, local, tribal, and territorial
governments—to support disaster response and recovery. For the COVID-19 response, for
example, FEMA issued a mission assignment to the Department of Defense to fund National
Guard deployments to support state and territorial response eﬀorts.

Overview of Key Issues
FEMA’s use of the Disaster Relief Fund to support COVID-19 activities. As of August 31, 2021,
FEMA had obligated more than $80 billion from the Disaster Relief Fund to respond to COVID-19.

Individual Assistance. On August 8, 2020, a presidential memorandum directed that up to $44
billion be made available from the Disaster Relief Fund to provide lost wages assistance to
supplement unemployment insurance compensation.591 According to FEMA oﬃcials, as of August
31, 2021, FEMA had obligated approximately $39 billion for the Lost Wages Assistance program.
Further, in December 2020, the Consolidated Appropriations Act, 2021, appropriated $2 billion to
the Disaster Relief Fund for eligible funeral expenses for individuals or households with COVID-19-
related funeral expenses.592 On April 12, 2021, FEMA began accepting and processing applications
for COVID-19 funeral assistance via a dedicated call center number.

Public Assistance. FEMA oﬃcials stated that as of August 31, 2021, FEMA had received 25,274
applications for Public Assistance and awarded $30.7 billion. From September 28, 2021 through
December 31, 2021, FEMA anticipates it will receive 5,600 additional public assistance projects for
an estimated additional $8.8 billion. With the emergence of the Delta variant of COVID-19, there
could be an increase in COVID-19-related Public Assistance projects. From September 1, 2020 to
August 31, 2021, FEMA obligated a total of approximately $26.8 billion for Public Assistance for
almost 8,200 projects. The ﬁgure below shows the number of projects and award amount for each
of FEMA’s 10 regions from September 1, 2020, through August 31, 2021.

591White House, Memorandum on Authorizing the Other Needs Assistance Program for Major Disaster Declarations Related to
Coronavirus Disease 2019 (Aug. 8, 2020). Pursuant to the presidential memorandum, upon receiving a FEMA grant, states
and territories may provide eligible claimants $300 or $400 per week—which includes a $300 federal contribution—in
addition to their Unemployment Insurance beneﬁts. The presidential memorandum directed that the program would
end either when $44 billion had been obligated, the balance of the Disaster Relief Fund reached $25 billion, on
December 27, 2020, or upon the enactment of legislation providing supplemental federal unemployment compensation,
whichever comes ﬁrst.
592Pub. L. No. 116-260, div. M, tit. II, 134 Stat. at 1910. The American Rescue Plan Act of 2021 also stated that FEMA is to
provide ﬁnancial assistance for COVID-19-related funeral expenses. Pub. L. No. 117-2, § 4006, 135 Stat. 4, 79.

Page 355 GAO-22-105051
Number of Projects and Amounts Obligated for Public Assistance for COVID-19, by FEMA Region, Sept. 1, 2020–
Aug. 31, 2021

According to FEMA oﬃcials, for COVID-19 declarations, FEMA can only provide Public Assistance
funding for emergency measures to protect public health and safety. A January 21, 2021,
presidential memorandum expanded the deﬁnition of eligible activities under FEMA’s Public
Assistance Program to include measures implemented to facilitate the safe opening and operation
of all eligible facilities, including schools, domestic violence shelters, and transit systems, among
others. Such assistance may include funding for the provision of personal protective equipment
and disinfecting services and supplies.593 According to FEMA oﬃcials, the memorandum will likely
have a signiﬁcant impact on their resources and operations. Speciﬁcally, FEMA oﬃcials stated

593White House, Memorandum to Extend Federal Support to Governors’ Use of the National Guard to Respond to COVID-19
and to Increase other Federal Assistance Provided to States (Jan. 21, 2021). A subsequent presidential memorandum, issued
on February 2, 2021, required FEMA to provide 100 percent cost share reimbursement for work eligible under Public
Assistance for emergency protective measures from January 20, 2020, through September 30, 2021. This requirement
did not include the operational expenses made eligible by the January 21, 2021, memorandum, which receive 100
percent cost share reimbursement from January 21, 2021, through September 30, 2021. White House, Memorandum on
Maximizing Assistance from the Federal Emergency Management Agency (Feb. 2, 2021). On August 17, 2021, the President
extended this full reimbursement for all work eligible under Public Assistance through December 31, 2021. White House,
Memorandum on Maximizing Assistance to Respond to COVID-19 (Aug. 17, 2021).

Page 356 GAO-22-105051
that the eligibility of reopening and operating expenses for schools and other facilities is likely to
signiﬁcantly increase the number of applicants and cost of eligible claims under Public Assistance.

Mission assignments. FEMA has issued mission assignments to multiple federal agencies—the
Department of Agriculture, Department of Labor, Environmental Protection Agency, and
Department of Defense, among others—to assist in the COVID-19 response. For example, FEMA
issued a mission assignment to the Department of Defense to fund National Guard deployments
to assist in recovery eﬀorts. The presidential memorandum issued on January 21, 2021, provided
that FEMA would fully reimburse expenses for National Guard activities to respond to COVID-19,
which may include vaccination distribution.594 According to FEMA, the estimated cost for National
Guard assistance totaled more $5.6 billion as of August 31, 2021.

The ﬁgure below shows FEMA’s obligations for COVID-19, by program and activity.

FEMA’s Disaster Relief Fund Obligations for COVID-19 by Program and Activity through August 2021

Note: The amounts shown include Disaster Relief Fund obligations through August 31, 2021 and projections through
September 30, 2021. Percentages are rounded to the nearest whole number.

FEMA’s Disaster Relief Fund supports a variety of activities.

Mass vaccination sites. The President directed FEMA to establish Pilot Community Vaccination
Centers (CVC) as part of a national eﬀort to speed the pace of COVID-19 vaccination campaigns
and ensure equitable access to vaccinations. Pilot CVCs were established through a partnership
between FEMA; the Centers for Disease Control and Prevention; other federal agencies; and
states. After an initial 8-week operational period, CVC pilot sites are closed or transitioned to the
state entity to fully operate. According to FEMA oﬃcials, the last CVC pilot site closed on June 20,
2021. FEMA told us that as of June 20, 2021, the pilot CVCs had given more than 5.6 million doses

594White House, Memorandum to Extend Federal Support to Governors’ Use of the National Guard to Respond to COVID-19
and to Increase other Federal Assistance Provided to States (Jan. 21, 2021).

Page 357 GAO-22-105051
of vaccines across 39 locations, including sites that had extended their participation in the pilot
program.595 FEMA oﬃcials stated that, as of September 8, 2021, there are no plans to re-establish
pilot CVC sites in response to the COVID-19 Delta variant, as state and local jurisdictions currently
have the capacity to continue providing vaccinations as needed.

Work and associated costs to support the distribution and administration of COVID-19 vaccines
may be eligible for Public Assistance. As such, FEMA is coordinating with other federal agencies
to meet state, local, tribal, and territorial needs. Speciﬁcally, FEMA oﬃcials stated that as of
August 31, 2021, the agency had obligated more than $5.83 billion to states, tribes, and territories
for vaccine distribution through the Public Assistance Program. In addition, FEMA is deploying
additional personnel to vaccination sites where they will assist people seeking access to the
vaccine. According to FEMA oﬃcials, eligible work and costs under Public Assistance for vaccine
distribution may include, but is not limited to,

• leasing facilities or equipment to administer and store the vaccine,

• providing staﬃng and training support,

• providing personal protective equipment and other administrative supplies, and

• using technology to register and track vaccine administration.

FEMA funeral assistance for COVID-19-related deaths. In December 2020, the Consolidated
Appropriations Act, 2021, appropriated $2 billion to the Disaster Relief Fund for eligible funeral
expenses for individuals or households with COVID-19-related funeral expenses.596 FEMA began
accepting applications on April 12, 2021. According to FEMA data, as of August 30, 2021, the call
center had received and was processing 264,544 applications, had approved 165,154 applications,
and had awarded more than $1 billion.

COVID-19 Funeral Assistance provides up to $9,000 per deceased individual to applicants who
incurred COVID-19-related funeral expenses on or after January 20, 2020, and meet eligibility
requirements. The maximum assistance provided to an applicant who incurred expenses for
multiple deaths per state or territory is $35,500. FEMA provides ﬁnancial assistance for eligible
funeral expenses including remains transfer, caskets and urns, burial plots and cremation niches,
and markers and headstones.597 According to FEMA’s interim policy, the death certiﬁcate must
directly or indirectly attribute the death to COVID-19 and the death must have occurred in the U.S.,
including the U.S. territories and the District of Columbia. However, in June 2021, FEMA amended

595The initial pilot period for each location was 8 weeks; however, jurisdictions were able to request an extension of an
additional 4 weeks. During this period, the site would receive federal staﬃng and support but not an additional vaccine
allocation.
596Pub. L. No. 116-260, div. M, tit. II, 134 Stat. at 1910. The American Rescue Plan Act of 2021 also stated that FEMA is to
provide ﬁnancial assistance for COVID-19-related funeral expenses. Pub. L. No. 117-2, § 4006, 135 Stat. 4, 79.
597Eligible expenses also include eligible interment expenses that include transportation to identify the deceased for
up to two people, interment, funeral services, oﬃciant of the services, and costs for up to ﬁve death certiﬁcates. Costs
associated with travel to scatter ashes outside of a memorial service and clothing to attend a funeral service are among
expenses not eligible for reimbursement under the program. Applicants must be U.S. citizens, noncitizen nationals, or
qualiﬁed aliens who paid for funeral expenses that are not covered by other sources, such as burial insurance.

Page 358 GAO-22-105051
its policy to allow applicants more ﬂexibility in attributing a death to COVID-19 without amending
the death certiﬁcate. Speciﬁcally, for deaths that occurred between January 20, 2020, and May
16, 2020, applicants can submit a signed statement or letter from the certifying oﬃcial on the
death certiﬁcate, or from the medical examiner or coroner in the jurisdiction in which the death
occurred, that attributes the death to COVID-19. The signed statement must be accompanied by
a death certiﬁcate. For deaths occurring after May 16, 2020, applicants must include a copy of the
death certiﬁcate that attributes the death to COVID-19.

FEMA does not accept online applications for COVID-19 Funeral Assistance. According to FEMA
oﬃcials, the agency established a dedicated, toll free call center number and hired a contractor to
accept and support the processing of applications for COVID-19 Funeral Assistance. As of July 22,
2021, the call center is staﬀed by approximately 1,900 operators across the U.S. Further, according
to FEMA oﬃcials, as of August 31, 2021, the average time needed to complete an application was
about 20 minutes. FEMA has not established a deadline to apply at this time.

After completing an application with a call center representative, applicants receive an application
number they can use to provide supporting documentation to FEMA online, by fax, or mail.
In addition, FEMA sends applicants a letter requesting additional documents to support their
application for COVID-19 funeral assistance. Applicants approved for COVID-19 Funeral Assistance
will receive a check by mail or direct deposit, depending on the option chosen when applying
for assistance. Applicants who are not approved for COVID-19 Funeral Assistance will receive
a decision letter explaining why they are not approved, their rights to appeal the decision, and
information on how to appeal it. Applicants have 60 days from the date of the decision letter to
appeal FEMA’s decision.

According to FEMA oﬃcials, in June 2021, FEMA identiﬁed a malfunction in its ﬁnancial system
that resulted in duplicate COVID-19 Funeral Assistance payments to 65 applicants in Georgia.
As of September 30, 2021, 49 of the 65 duplicate payments, totaling almost $295,000, had been
successfully recalled by the Department of the Treasury or voluntarily refunded by applicants.
FEMA oﬃcials told us that they were continuing to reach out to applicants to ask them to
voluntarily return the duplicate payments. If applicants do not voluntarily return these payments,
FEMA will consider other means, such as garnishing wages, to retrieve the money. According to
FEMA oﬃcials, as of September 10, 2021, FEMA made technical enhancements to the system to
increase memory and database size to accommodate increased funeral assistance payments.
However, FEMA was still working on other enhancements to the system to repair the underlying
issue that caused duplicate payments to be paid. FEMA oﬃcials stated that they were not aware of
any other duplicate payments going out to applicants.

The scope of FEMA’s COVID-19 Funeral Assistance is unprecedented. In the decade before the
COVID-19 pandemic, FEMA had processed approximately 6,000 applications for funeral assistance
after other natural disasters, according to FEMA oﬃcials. The oﬃcials told us that FEMA has
internal controls to mitigate fraudulent activity and described these internal controls as eﬀective
for preventing and identifying fraud. According to the oﬃcials, to develop the internal controls,
FEMA relied on numerous sources and lessons learned from previous disasters. The oﬃcials said
that as of August 31, 2021, FEMA had ﬂagged 22,275 of the 264,321 applications it had received for
additional review based on its fraud controls. Of the ﬂagged applications, 11,497 were cleared for
continued processing; 1,239 applications remained under review; 9,499 applications were pending

Page 359 GAO-22-105051
additional documentation from the applicant; and 40 applications were pending adjudication for
potential fraud.

Delays in FEMA’s processing of Public Assistance applications. FEMA is experiencing delays
in processing applications for COVID-19 Public Assistance. Applicants and state emergency
management personnel have raised concerns to FEMA about the time it takes to complete and
approve applications for COVID-19 Public Assistance.

Senior state emergency managers we spoke with in one state said that long processing times
have created ﬁnancial challenges for communities that are awaiting reimbursement from
FEMA for expenses they incurred. For example, these oﬃcials told us that they submitted 823
projects to FEMA for Public Assistance. Of these, as of August 26, 2021, 195 applications were
determined to have ineligible expenses, 92 were pending, 86 were withdrawn, 248 were going
through formulation598, and 202 had been approved and obligated funding. According to these
state oﬃcials, on average, it took FEMA 236 days to make ineligibility determinations for the 195
applications with ineligible expenses and about 135 days to approve and make obligations for
the 202 approved applications once they were submitted. The long processing time led some
applicants to seek other funding sources for their projects and also created additional stress and
ﬁnancial burden for applicants already dealing with a disaster; it also created additional strain on
state, territorial, and federal personnel who were assisting applicants.

FEMA oﬃcials stated that FEMA currently has a backlog of 3,002 applications that are pending a
determination memorandum for potentially ineligible expenses.599 As of September 2021, FEMA’s
median time to process an application is 51 days if there is not an issue with expense eligibility.
However, if there is an eligibility issue, the median number of days to process the application is
138 days. 600

FEMA oﬃcials acknowledged concerns that they had heard regarding the application process for
COVID-19 Public Assistance and took actions to address them. In an eﬀort to make the application
process easier and to improve timeliness in making an eligibility determination, FEMA took steps
that included clarifying information collection and documentation requirements and simplifying
and streamlining the application process. In addition, to reduce the backlog of ineligibility
determinations, in February 2021, FEMA established the Determination Memorandum task force
consisting of 27 personnel, including analysts and attorneys, who were experienced determination
memorandum writers. To improve the application process, the task force created a review process
to ensure that the memorandums were written clearly and stated the appropriate determinations
based on the speciﬁcs of the projects.

According to FEMA oﬃcials, as a result of these steps, there has been a reduction in FEMA’s
backlog of ineligibility determinations and number of requests for additional information from

598Project formulation is the process of documenting the eligible facility, eligible work and eligible costs for a project.
599FEMA uses determination memorandums to inform applicants that some of the expenses claimed on their
application are ineligible for reimbursement.
600According to FEMA oﬃcials, the median processing times for COVID-19-related projects is faster than non-COVID-19-
related projects. As of August 30, 2021, the median processing time for COVID-19-related projects from project creation
to initial obligation was 80 days compared to 135 days for non-COVID-related projects.

Page 360 GAO-22-105051
applicants. However, with the emergence of the Delta variant for COVID-19, the number of
COVID-19-related Public Assistance applications could increase, which could aﬀect the backlog of
ineligibility determinations. We will continue to monitor FEMA’s processing of COVID-19-related
public assistance claims.

FEMA inconsistently applied COVID-19 Public Assistance policies. FEMA issued several
COVID-19 related policies for Public Assistance.601 FEMA inconsistently interpreted and applied
its policies for COVID-19 Public Assistance within and across its regions.602 State emergency
managers we spoke with from four of the 10 regions provided multiple examples of FEMA’s
inconsistency in interpreting and applying its policy regarding the eligibility of expenses for
reimbursement from its Public Assistance Program.

Overtime. Oﬃcials we spoke with in two states said that FEMA inconsistently interpreted and
applied its policies for reimbursement of overtime expenses for Emergency Medical Services
workers. For example, according to the oﬃcials, in some cases FEMA said that these expenses
were eligible for reimbursement only if the Emergency Medical Service worker was responding
to a patient who had already tested positive for COVID-19; however, FEMA did not interpret this
policy consistently across communities in the state and across other regions. In addition, based on
ongoing issues that applicants faced, one of these states requested that FEMA clarify its policy on
overtime pay for Emergency Medical Service workers. However, as of August 2021, according to
oﬃcials in one state, FEMA had not provided clarity on its policy, while oﬃcials in the other state
said that the issue had been resolved.603

Personal protective equipment. Oﬃcials in one state said that at one point FEMA had
deemed the provision of personal protective equipment at correctional facilities as ineligible for
reimbursement in their region but that states in other regions had received reimbursement for
the same expense. Furthermore, oﬃcials we spoke with in two states said that FEMA imposed
requirements for reimbursement for personal protective equipment and disinfectant for schools
that they believed were inconsistent with its policy at the time and doing so created challenges for
schools reopening. FEMA oﬃcials we spoke with in August 2021 acknowledged that the agency’s
interpretation and application of its policy related to reimbursement for personal protective
equipment varied across regions.

601FEMA issued a series of policies from March 2020 to April 2021 that describe eligible expenses for the use of Public
Assistance for COVID-19. Some of the key policies issued are: FEMA, Coronavirus (COVID-19) Pandemic: Eligible Emergency
Protective Measures, FEMA Fact Sheet (March 2020); Coronavirus (COVID-19) Pandemic: Work Eligible for Public Assistance
(Interim), FEMA Policy FP 104-009-19 (September 2020); Coronavirus (COVID-19) Pandemic: Medical Care Eligible for Public
Assistance (Interim) (Version 2), FEMA Policy #104-21-0004 (March 2021); and Coronavirus (COVID-19) Pandemic: Safe
Opening and Operation Work Eligible for Public Assistance (Interim) FEMA Policy FP-104-21-0003 (April 2021).
602We have previously reported past challenges with FEMA’s management of the Public Assistance Program. Speciﬁcally,
in 2018 related to the response to Hurricanes Harvey, Irma, and Maria we reported on issues over the lack of
consistency in eligibility determinations and the knowledge and experience of program staﬀ as presenting potential
challenges for recovery.
603FEMA’s guidance includes requirements for determining the eligibility of labor costs, including overtime. See FEMA,
Public Assistance Program and Policy Guide, FP 104-009-2 (June 1, 2020); and Coronavirus (COVID-19) Pandemic: Medical Care
Eligible for Public Assistance (Interim) (Version 2), FEMA Policy #104-21-0004 (March 2021).

Page 361 GAO-22-105051
Based on our conversation with state emergency managers, we found that inconsistent
interpretation and application continues in one state. For example, state oﬃcials told us that
reimbursement for PPE has been inconsistent across counties within the state. That state reported
that although the expenses for PPE were incurred around mid-2020, after which FEMA issued
a clarifying policy, current interpretation of the prior policy as it applies to PPE continues to be
inconsistent.

Noncongregate sheltering. Oﬃcials we spoke with in one state said that FEMA had inconsistently
interpreted and applied its policy across FEMA regions for reimbursement of expenses for
noncongregate sheltering.604 According to FEMA policy, noncongregate sheltering can be approved
for up to 30 days or longer if justiﬁed.605 However, oﬃcials in this state said that FEMA had not
approved reimbursement for noncongregate sheltering costs beyond 14 days but had approved it
for longer periods in states in other regions.

FEMA oﬃcials in headquarters acknowledged these challenges and have taken initial steps to
address them. For example, in December 2020, FEMA established the Request For Information
task force that included subject matter experts from its four Consolidated Resource Centers and
local hires who were experienced determination memorandum writers.606 The intent of this task
force was to improve the consistency of FEMA’s interpretation and application of policy. As part of
the Request For Information task force, FEMA personnel reviewed every instance where FEMA was
planning to request additional supporting documentation from applicants. As a result of the task
force’s eﬀorts, FEMA identiﬁed common scenarios that FEMA could use to train staﬀ and improve
consistency across regions. FEMA oﬃcials said that they communicated the task force’s ﬁndings
in writing and in meetings with FEMA regional personnel and associations of state emergency
managers.

Based on our discussions with FEMA headquarters oﬃcials and state emergency managers, we
identiﬁed four key areas that contributed to the inconsistent interpretation and application of
COVID-19 policy for Public Assistance.

1. Changes in policy. Although FEMA’s Public Assistance Program and Policy Guide deﬁnes policies
and procedures for the Public Assistance Program, it does not speciﬁcally address COVID-19
Public Assistance. FEMA had not previously used the Public Assistance Program to respond
to a nationwide emergency such as the COVID-19 pandemic. As a result, FEMA oﬃcials said
that they quickly developed and issued a series of policy documents clarifying the use of
Public Assistance to respond to COVID-19, which FEMA personnel interpreted and applied

604Noncongregate sheltering helps protect public health and safety—for example, by providing an isolated or individual
shelter for those who test positive for COVID-19 and do not require hospitalization.
605FEMA has traditionally required applicants to request time extensions every 30 days to ensure non-congregate
sheltering operations are still necessary. However, FEMA eliminated the requirement to submit time extensions in 30
day increments for COVID-19 declarations. See FEMA, Update to Non-Congregate Sheltering Delegation of Authority Public
Assistance Program and Policy Guide Waiver, Memorandum for Regional Administrators Regions I – X (Dec. 16, 2020); and
FEMA, Frequently Asked Questions, Coronavirus (COVID-19) Pandemic: Non-Congregate Sheltering (March 2020).
606FEMA’s Consolidated Resource Centers house Public Assistance oﬃcials who conduct peer reviews of completed
projects, depending on the expertise of the specialist developing the project. Other specialists at the centers check
to ensure the accuracy of the scope of work, cost estimates, and supporting documents for the project and provide
feedback on ways to improve work on future projects.

Page 362 GAO-22-105051
diﬀerently.607 FEMA oﬃcials stated that they work hard to retain consistency in policies that
are issued, but it is diﬃcult with every state and region not experiencing the same thing at the
same time.

State emergency managers we interviewed stated that changes in policy were interpreted and
applied inconsistently within and across regions. Similarly, the National Governors Association
letter to FEMA on May 12, 2021, stated that prior to FEMA issuing clarifying policy, states and
territories received inconsistent messaging across FEMA regions about the eligibility of items
for reimbursement, such as masks and disinfectant for schools and increased operating costs
for 24/7 emergency operations center. According to the letter, subsequent changes to FEMA’s
policy guidance resulted in three diﬀerent eligibility requirements based on arbitrary dates,
which led to challenges in verifying duplication of beneﬁts and untangling obligated funds and
caused strain on personnel at the state, territorial, and federal levels. In the letter, the National
Governors Association stated that they continue to be challenged by FEMA’s COVID-19 Public
Assistance guidance and recommended that the FEMA Administrator implement better
communication of policy changes and decisions that are disseminated by the FEMA regions
to ensure consistency of messaging across the country.608 FEMA oﬃcials stated that as of
October 1, 2021, they had not issued a response to this recommendation. In addition, in
September 2021, the National Emergency Management Association made recommendations
to streamline Public Assistance to FEMA and speciﬁcally stated that FEMA’s changing policies
made it challenging for states to determine which policies applied to speciﬁc projects at any
given time, and prolonged the application and review process.609 On September 8, 2021, FEMA
issued a policy that provides retroactive reimbursement at 100 percent federal cost share for
Public Assistance expenses incurred from January 20, 2020 through December 31, 2021.610

2. Delegation of authority to FEMA regions. According to FEMA oﬃcials, FEMA’s 10 regions
have the authority and responsibility to make ﬁnal eligibility determinations, while FEMA’s
Consolidated Resource Centers handle all processing of applications, including cost estimates
and compliance reviews. FEMA oﬃcials said that although they strive to achieve consistency
in interpretation and application of policy, diﬀerences arise because of the delegated
authority. According to FEMA oﬃcials, COVID-19 is unusual because it aﬀects every state
and every region at the same time, unlike the disasters that FEMA policies typically address.

607Some of FEMA’s policy changes were dictated by presidential memorandum. See, e.g., White House,
Memorandum to Extend Federal Support to Governors’ Use of the National Guard to Respond to COVID-19 and to Increase
other Federal Assistance Provided to States (Jan. 21, 2021).
608See National Governors Association, Letter to FEMA on Public Assistance Guidelines (May 12, 2021). The National
Governors Association serves as the voice of leaders of 55 states, territories, and commonwealths. It is the
mechanism used by governors to identify priority issues and deal with matters of public policy and governance at
the state, national, and global levels.
609The National Emergency Management Association is a nonpartisan, nonproﬁt 501(c)(3) association dedicated
to enhancing public safety by improving the nation's ability to prepare for, respond to, and recover from all
emergencies, disasters, and threats to our nation's security. It is the professional association of and for emergency
management directors from all 50 states, eight U.S. territories, and the District of Columbia.
610FEMA issued this policy in response to an August 17, 2021 Presidential Memorandum. See White House,
Memorandum on Maximizing Assistance to Respond to COVID-19 (Aug. 17, 2021). For the FEMA policy, see
Coronavirus (COVID-19) Pandemic: Safe Opening and Operation Work Eligible for Public Assistance (Interim), FEMA
Policy 104-21-0003, Version 2 (Sep, 8, 2021).

Page 363 GAO-22-105051
FEMA oﬃcials stated that because of delegated authority, the 10 regions’ interpretation
and application of COVID-19 policies may not be the same in every circumstance, and they
continue to discuss methods to improve in this area. According to FEMA oﬃcials, some of the
diﬀerent outcomes exist because there are diﬀerent circumstances behind projects in diﬀerent
places.

3. Training of staﬀ. FEMA oﬃcials stated that training on COVID-19 policy is available to but not
required for individuals handling Public Assistance applications. As part of its task force to
reduce inconsistent application of policy, FEMA identiﬁed training gaps among its staﬀ at
Consolidated Resource Centers and among state, local, tribes, and territorial partners and
began to address these gaps through training and mentoring staﬀ. According to FEMA oﬃcials,
FEMA works closely with the regions and state, local, tribal, and territorial partners in an eﬀort
to apply policy consistently across all COVID-19 declarations. Speciﬁcally, FEMA oﬃcials stated
that for each policy developed for COVID-19 Public Assistance, FEMA provided training on an
as needed basis. FEMA also developed frequently asked questions and fact sheets and placed
them on their website for employees to access as needed. Additionally, FEMA hosted a series
of webinars that were available to staﬀ and state, local, tribal, and territorial governments that
addressed eligibility requirements for reimbursable Public Assistance expenses for COVID-19.
However, staﬀ at Consolidated Resource Centers, among others, were not required to take this
training.611

4. Experience and knowledge of staﬀ. Emergency management oﬃcials from two states attributed
inconsistent interpretation and application of COVID-19 Public Assistance policies to
the experience level of the person making the eligibility determination for applications.
In September 2021, the National Emergency Management Association stated that "the
movement of Public Assistance projects through the entire process successfully and eﬃciently
is heavily reliant on the knowledge of the FEMA representatives to ensure proper policy
intent compliance and to avoid future audits. The lack of experience of FEMA representatives
presents constant challenges as they do not have a full understanding of the project scope
nor the policy that would back the project eligibility." In addition, according to National
Emergency Management Association oﬃcials, the rotation of personnel reviewing COVID-19
Public Assistance applications could aﬀect eligibility determinations for that project. For
example, if an application is in process and has to be transferred to a diﬀerent person to
review, that individual may not have a full understanding of the project scope including prior
determinations made by their predecessor. This could result in expenses that were initially
determined to be eligible for Public Assistance to ultimately be deemed ineligible.

According to FEMA’s National Disaster Recovery Framework, the federal government is responsible
for ensuring that information is distributed in an accessible manner and is well understood, so
that all stakeholders are informed and aware of the process.612 FEMA has acknowledged that
in spite of its eﬀorts, inconsistent interpretation and application of Public Assistance policy for
COVID-19 continues to occur within and across regions. Given the current rise in the COVID-19

611FEMA oﬃcials told us that they held several 2-day mandatory training, and the most recent one held in March
2021 included a session on Public Assistance eligibility requirements for COVID-19.
612Department of Homeland Security, National Disaster Recovery Framework, 2nd ed. (Washington, D.C.: June 2016).

Page 364 GAO-22-105051
Delta variant across the nation, FEMA is likely to receive applications for reimbursement for
a larger number of projects than it estimated earlier in 2021. By improving the consistency of
its interpretation and application of COVID-19 Public Assistance policy, FEMA can help ensure
that applicants for Public Assistance receive timely and consistent reimbursement for eligible
expenses.

Methodology
To conduct this work, we reviewed FEMA’s monthly Disaster Relief Fund reports to obtain FEMA
obligations data for Individual Assistance, Public Assistance, and mission assignments for
February 2020 through August 2021 and projected obligations data through September 2021. We
reviewed federal laws and FEMA policies and guidance on how states, local, tribal, and territorial
governments may apply for, and receive, assistance to respond to the COVID-19 pandemic. We
also reviewed data from FEMA’s advisories on mass vaccination sites, Public Assistance, and
funeral assistance related to COVID-19. Although we present FEMA data on Public Assistance and
COVID-19 Funeral Assistance, we did not independently determine the reliability of the data.

In addition, we reviewed presidential memorandums issued on August 8, 2020, January 21,
February 2, and August 17, 2021; FEMA’s Frequently Asked Questions; and previous GAO reports
on FEMA’s response to the COVID-19 pandemic and other natural disasters. Further, we reviewed
letters to FEMA from the National Governors Association and National Emergency Management
Association on FEMA’s Public Assistance Program. We also interviewed state emergency oﬃcials
from Illinois, North Carolina, Pennsylvania, and Washington as well as representatives from the
National Governors Association and the National Emergency Management Association. Finally, we
interviewed FEMA oﬃcials regarding their eﬀorts to implement COVID-19 Funeral Assistance and
the COVID-19 Public Assistance Program.

Agency Comments
We provided a draft of this enclosure to the Department of Homeland Security (DHS), FEMA, and
the Oﬃce of Management and Budget (OMB). In its comments, which are reproduced in appendix
V, DHS concurred with both recommendations and outlined actions it has taken to improve the
consistency of the agency’s interpretation and application of the COVID-19 Public Assistance
Policy and train employees in the regions and at its Consolidated Resource Centers. Speciﬁcally,
FEMA stated that on September 8, 2021, it issued the Safe Opening and Operation Interim Policy
which amends the applicable time period of eligibility retroactively to the beginning of the incident
period for activities related to the safe opening and operation of facilities. This policy also speciﬁes
that work conducted from the beginning of the incident period to December 31, 2021, will be
reimbursed at a federal cost share of 100 percent. According to FEMA’s management response
letter, FEMA believes this policy will improve the consistent interpretation and application of
COVID-19 Public Assistance Policy nationwide. We agree that the newly issued policy and the
period of time it covers should assist FEMA in improving the consistency of its decision-making
on this issue; however, it does not necessarily address the variance in interpretation on other
issues, such as overtime for Emergency Medical Services employees or non-congregate sheltering.
Further, given this is a recently issued policy, we will continue to monitor its implementation

Page 365 GAO-22-105051
to see if it fully addresses our recommendation. Additionally, FEMA stated that it has taken a
number of actions to educate staﬀ on changes to COVID-19 Public Assistance policy to ensure
that interpretation and application are consistent nationwide. For example, FEMA conducted a
Public Assistance Training on September 9, 2021, which included a section on COVID-19 related
policies and guidance, including helping applicants navigate COVID-19 project development. It is
too early to measure the impact of FEMA’s eﬀorts to educate staﬀ on changes to COVID-19 Public
Assistance policy.

GAO’s Ongoing Work
We will continue to monitor issues related to FEMA’s Disaster Relief Fund, workforce, and funeral
assistance for COVID-19 related deaths. Speciﬁcally, we will monitor obligations for Individual
Assistance, Public Assistance, and mission assignments, as well as the balance in the fund. In
addition, we will continue to monitor and assess the controls FEMA is using to prevent fraud,
waste, and abuse in its delivery of COVID-19 funeral assistance. We will continue to report on this
program in our CARES Act reporting through April 2022.

GAO’s Prior Recommendations
The table below presents our recommendations on FEMA’s response to COVID-19 from prior
bimonthly and quarterly CARES Act reports.

Page 366 GAO-22-105051
Prior GAO Recommendations Related to COVID-19
Recommendation Status

The Federal Emergency Management Agency Administrator Open—partially addressed. In March 2021, DHS
should adhere to the agency’s protocols listed in the concurred with our recommendation. DHS stated that
updated 2019 Tribal Consultation Policy by obtaining tribal FEMA’s National Tribal Aﬀairs Adviser, based in the Oﬃce
input via the four phases of the tribal consultation process of External Aﬀairs, will coordinate with other FEMA oﬃces
when developing new policies and procedures related to and directorates, as appropriate, to review the agency’s
COVID-19 assistance. (March 2021 report). adherence to protocols listed in the Tribal Consultation
policy. According to FEMA oﬃcials, in March 2021, FEMA
conducted formal consultation with Tribal Leaders on
COVID-19 Funeral Assistance before ﬁnalizing the interim
policy. In April 2021, FEMA sent letters to tribal leaders
discussing (1) FEMA policy and procedure for ﬁnancial
assistance to individuals and households for COVID-19
related funeral expenses incurred after January 20, 2020;
and (2) framework, policy details and requirements for
determining the eligibility of safe opening and operation
work and costs under the Public Assistance Program.

The Federal Emergency Management Agency Administrator Open—partially addressed. In March 2021, DHS
should provide timely and consistent technical assistance concurred with our recommendation. DHS stated that
to support tribal governments’ eﬀorts to request and FEMA’s Recovery Directorate will publish a memorandum
receive Public Assistance as direct recipients, including that will contain direction to FEMA regions regarding
providing additional personnel, if necessary, to ensure that the assignment of Public Assistance Program delivery
tribal nations are able to eﬀectively respond to COVID-19. managers to promote equitable delivery of Public
(March 2021 report). Assistance to tribal governments. According to FEMA
oﬃcials, on August 6, 2021, FEMA sent a memorandum that
provided updates on how FEMA would deliver assistance.
This guidance provides FEMA’s regional staﬀ the ability to
work with all tribal applicants to understand their capacity
to address issues through their assigned Public Assistance
program delivery manager. As of October 1, 2021, FEMA did
not have data on the outcome of this assistance.

The Administrator of the Federal Emergency Management Open—partially addressed. In September 2020, DHS
Agency—who heads one of the agencies leading the disagreed with this recommendation, noting, among other
COVID-19 response through the Uniﬁed Coordination things, work that FEMA had already done to manage the
Group—consistent with their roles and responsibilities, medical supply chain and increase supply availability.
should work with relevant federal, state, territorial, and Although DHS disagreed with our recommendation, it
tribal stakeholders to devise interim solutions, such began taking some actions in March 2021. As of May
as systems and guidance and dissemination of best 2021, DHS has not demonstrated action to devise interim
practices, to help states enhance their ability to track solutions that would systematically help states, tribes,
the status of supply requests and plan for supply needs and territories eﬀectively track, manage, and plan for
for the remainder of the COVID-19 pandemic response. supplies to carry out the COVID-19 pandemic response in
(GAO-20-701) the absence of state-level end-to-end logistics capabilities
that would track critical supplies required for a response
of this scale. We note that we made this recommendation
to both DHS and HHS with the intent that they would
work together under the Uniﬁed Coordination Group to
address challenges reported by state oﬃcials with both
public health and emergency management responsibilities.
Moreover, we recommended they take actions that were
consistent with the roles and responsibilities that were to
be more clearly deﬁned as HHS took a more central role in
leading supply distribution. The recommendation to deﬁne

Page 367 GAO-22-105051
Recommendation Status
those roles and responsibilities remains open. Moreover,
although both DHS and HHS have reported separate
actions, taken as part of other eﬀorts within each separate
purview, neither has articulated how they worked with the
other nor how they assessed whether the actions changed
the experiences of state oﬃcials who reported issues
during our prior work. Without systematic and deliberate
action to help states ensure they have the support they
need to track, manage, and plan for supplies, states, tribes,
and territories on the front lines of the whole-of-nation
COVID-19 response may continue to face challenges that
hamper their eﬀectiveness.

Source: GAO. | GAO-22-105051

Related GAO Products
FEMA Disaster Workforce: Actions Needed to Address Deployment and Staﬀ Development Challenges.
GAO-20-360. (Washington, D.C.: May 4, 2020).

Disaster Response: Federal Assistance and Selected States and Territory Eﬀorts to Identify Deaths from
2017. GAO-19-486. (Washington, D.C.: Sept. 13, 2019.

Disaster Assistance: Opportunities to Enhance Implementation of the Redesigned Public Assistance Grant
Program. GAO-18-30. (Washington, D.C.: Nov. 8, 2018)

2017 Hurricanes and Wildﬁres: Initial Observations on the Federal Response and Key Recovery
Challenges. GAO-18-472 (Washington, D.C.: Sept. 4, 2017).

Disaster Recovery: FEMA’s Public Assistance Grant Program Experienced Challenges with Gulf Coast
Rebuilding. GAO-09-129. (Washington, D.C.: Dec. 18, 2008).

Contact information: Chris Currie, (404) 679-1875, curriec@gao.gov

Page 368 GAO-22-105051
COVID-19 Impact on IRS Enforcement
In response to the pandemic’s onset in March 2020, the Internal Revenue Service signiﬁcantly
reduced its eﬀorts to enforce taxpayer compliance. In July 2020, IRS started increasing these
enforcement eﬀorts by making changes to address the challenges of operating in a pandemic
environment.

Entity involved: Internal Revenue Service, within the Department of the Treasury.

Background
The Internal Revenue Service’s (IRS) tax enforcement eﬀorts help collect tax revenue from
noncompliant taxpayers and promote voluntary tax compliance. We have included enforcement of
tax laws on our High-Risk List due to the need for IRS to improve tax compliance and address the
gross tax gap, or the diﬀerence between taxes owed by individuals and businesses and the taxes
they paid voluntarily and on time.613

IRS’s eﬀorts to enforce tax compliance include examination programs to help ensure that
taxpayers ﬁle accurate tax returns. IRS’s Wage and Investment (W&I), Small Business and Self-
Employed (SB/SE), and Large Business and International (LB&I) divisions conduct examinations
based on the type of taxpayer. Examinations can be conducted by mail and telephone or in
person, such as at an IRS oﬃce or a taxpayer’s location. IRS also uses automated compliance
checks, such as the Automated Underreporter (AUR) program, to verify taxpayer compliance by
comparing information reported by taxpayers to information reported to IRS by third parties, such
as employers and banks. IRS collection eﬀorts seek to enforce taxpayers’ compliance with tax ﬁling
requirements and payment of taxes.614

Overview of Key Issues
Beginning in March 2020, IRS closed its on-site operations and generally suspended its key tax
enforcement programs to protect staﬀ and taxpayers from transmitting COVID-19 and because
it had limited ability to perform work remotely. In addition, IRS provided tax compliance relief to
taxpayers experiencing COVID-19 hardships. This relief included postponing certain collection
activities and limiting examinations in the ﬁeld.

IRS began to resume some enforcement activities in July 2020 by transitioning staﬀ to telework,
revising the types of work they were to do, and increasing the use of technology. However, IRS
faced signiﬁcant work backlogs along with process challenges and reduced staﬀ availability. IRS
continued to slowly resume enforcement activities in the fall and winter of 2020. As of June 2021,
IRS had made major improvements but had not achieved prepandemic levels in caseload and tax
revenue for all enforcement activities. IRS is considering ways to identify lessons learned and make

613Every 2 years, we report on federal programs and operations that are vulnerable to waste, fraud, abuse, and
mismanagement, or that need broad reform—our High-Risk List. For more information, see our March 2021 report.
IRS enforcement reduces the gross tax gap by collecting unpaid tax revenue, resulting in a “net” tax gap. In 2019, IRS
estimated that the average annual net tax gap was $381 billion for tax years 2011–2013.
614SB/SE Collection administers these programs for taxpayers served by all divisions.

Page 369 GAO-22-105051
permanent changes as appropriate to improve enforcement and prepare for future disruptions to
operations.

Use of Weather and Safety Leave over time. At the beginning of the pandemic, few IRS staﬀ
were working in the oﬃce, while many others began teleworking. Under IRS policy, staﬀ who were
unable to work in either of these settings were placed in Weather and Safety Leave, a type of paid
leave typically used when it is unsafe for staﬀ to come into the oﬃce due to inclement weather or
some other reason (see ﬁgure).

Internal Revenue Service (IRS) Staﬀ Work Status, Mar. 2020–June 2021

Note: The major declines for teleworking and time in the oﬃce generally occurred during holiday periods for which federal staﬀ
generally also use accumulated annual leave

As of June 2021, almost no staﬀ were on full-week Weather and Safety Leave, while more than
60,000 IRS staﬀ performed at least some of their work remotely and approximately 30,000
performed at least some work from the oﬃce.615

The pandemic’s eﬀects on enforcement caseloads and revenue. We found that nearly all IRS
enforcement caseloads dropped substantially from June 2019 to June 2020. Caseloads increased
from June 2020 to June 2021, but not all activities have returned to their prepandemic levels (see
table). Although not all changes can be attributed to the pandemic, it was a key factor during the
period.

615We were not able to analyze the amount of time that IRS staﬀ charged to enforcement activities as comparable work
time measures are not kept across the various divisions and activities.

Page 370 GAO-22-105051
Enforcement Caseload Trends for IRS Examination, Automated Underreporter, and Collection Activities for June
2019, 2020, and 2021
FY 2019–21
Activity/caseload measures June 2019 June 2020 June 2021 % Change

a
SB/SE—Field Examination

Starts 96,870 90,301 75,379 -22.19

Closures 108,379 67,369 85,736 -20.89

a
SB/SE—Correspondence Examination

Starts 164,661 143,068 99,284 -39.70

Closures 156,242 125,404 184,867 18.32

a
SB/SE—AUR

Starts 1,420,125 1,302,067 1,829,166 28.80

Closures 1,361,778 829,538 1,673,051 22.86

a
W&I—Correspondence Examination

Starts 217,242 209,964 262,014 20.61

Closures 236,027 180,895 277,971 17.77

a
LB&I—Examination

Starts 13,727 17,147 17,477 27.32

Closures 16,819 12,853 14,376 -14.53

b
SB/SE—Collection Inventory

Unpaid taxes—automated collection 6,097,332 5,742,953 6,191,968 1.55

Unpaid taxes—ﬁeld collection 699,708 534,807 514,221 -26.51

Unﬁled returns—automated collection 1,317,229 2,222,384 1,313,936 -0.25

Unﬁled returns—ﬁeld collection 49,709 86,113 46,836 -5.78

Legend: AUR = Automated Underreporter; FY = ﬁscal year; LB&I = Large Business and International; SB/SE = Small Business and Self-
Employed; W&I = Wage and Investment
Source: GAO analysis of Internal Revenue Service (IRS) data. | GAO-22-105051

a
Examination and AUR caseloads are tracked by the number of cases started and cases closed.
b
Collection inventory is measured by the number of taxpayer modules being pursued for unpaid taxes or unﬁled returns. A
module is a record for a speciﬁc taxpayer covering one return for one tax period.

Page 371 GAO-22-105051
The table shows that the pandemic’s eﬀect on caseload varied by the type of enforcement
activity. By June 2021, caseloads for AUR and W&I correspondence examinations, which are
highly automated, had recovered from the 2020 decline and exceeded the 2019 levels. In LB&I,
ﬁeld examiners generally were ready to telework, which enabled them to start even more
examinations, but closures declined because they rely on other parts of IRS that were shut down
or at reduced staﬃng.

By June 2021, workloads in collection for both unpaid taxes and unﬁled returns had mostly
returned to or were trending toward the 2019 levels. One exception is ﬁeld collection, where the
June 2021 unpaid tax inventory was below the June 2019 and June 2020 levels. SB/SE collection
oﬃcials said that collecting unpaid taxes in the ﬁeld generally involves face-to-face contact with
taxpayers, unlike the other two activities, which involve collecting unpaid taxes by mail and
telephone.

The pandemic also aﬀected the revenue collected through enforcement actions.616 The table
below shows that enforcement revenue declined across all the major enforcement activities during
the height of the pandemic in June 2020, although the decline was smaller for collection from
unﬁled tax returns. Collection oﬃcials explained that they shifted staﬀ to work more unﬁled tax
return cases and fewer unpaid tax cases in the early part of the pandemic because that work could
be done remotely without face-to-face contact. Collection oﬃcials said they shifted staﬀ back to
more unpaid tax cases as safety procedures were implemented.

616Other factors beyond the pandemic may aﬀect these revenue collection trends. For example, time lags of months if
not years exist between completing examinations and collecting the related revenue.

Page 372 GAO-22-105051
Enforcement Revenue in June 2019, 2020, and 2021 for Selected Major Enforcement Activities
Enforcement activities Amount—June 2019 ($) Amount—June 2020 ($) Amount—June 2021 ($)

All enforcement activities 41,657,106,427 35,760,140,668 47,362,241,265

Selected major activities

Field examination
a 2,836,883,338 2,345,763,320 3,241,727,554

Campus examination
b 1,059,321,674 861,381,325 1,022,024,404

Automated underreporter
c 3,370,695,783 2,873,110,884 3,662,829,596

Collection—taxes owed 29,585,743,434 24,949,583,505 33,525,724,252

Collection—unﬁled returns 2,572,959,696 2,381,635,066 2,980,318,305

Source: GAO analysis of Internal Revenue Service (IRS) data. | GAO-22-105051

a
Involves examinations done by IRS revenue agents in IRS ﬁeld oﬃces.
b
Involves examinations done by IRS examiners through mail from IRS campus locations.
c
Involves matching information returns ﬁled by third parties to report income such as wages to tax returns ﬁled by taxpayers.

Operational challenges and IRS’s actions. According to IRS management, enforcement activities
across IRS generally were not telework ready because the case work largely relied on paper
processes.617 Oﬃce closures limited access to physical case ﬁles; when staﬀ were able to come
to the oﬃce, IRS faced logistical challenges of keeping staﬀ socially distanced. Staﬀ did not have
computers, scanners, and printers to telework from home, and it took IRS time to procure and
distribute this equipment. Enforcement staﬀ were limited in their ability to safely meet taxpayers
face to face. Meanwhile, interruptions to mail, phone, and print operations caused delays and
backlogs in casework. The table below shows these common operational challenges and the
actions IRS took across its divisions.

617As discussed later, LB&I examination and SB/SE ﬁeld collection staﬀ were telework ready, according to IRS oﬃcials.

Page 373 GAO-22-105051
IRS Pandemic-Related Enforcement Challenges and Actions through June 2021

a
Actions are not meant to correspond directly with the challenges on the same lines.

Because staﬀ were not telework ready, IRS enforcement managers stated that many staﬀ
experienced downtime during the summer of 2020 while IRS acquired equipment so they could
work from home. Downtime was also an issue for staﬀ who had been doing much of their work
remotely using IRS equipment before the pandemic, such as highly graded examination and
collection staﬀ who meet with taxpayers in the ﬁeld and do the most complex work. Although
these staﬀ were telework ready, some of them could not do as much work as before the
pandemic. If they did meet with taxpayers, they faced challenges in keeping the interactions safe.

Interruptions in support services, such as mail, phone, and print operations, also hampered
resumption of enforcement activities. Many of these operations were not restarted until mid- or
late summer 2020. As a result, IRS staﬀ and taxpayers could not communicate with each other,
leading to delays and increases in backlogged enforcement casework.

IRS managers and documents indicated that safety concerns among IRS staﬀ and taxpayers were
a major challenge related to the pandemic. When staﬀ did go into the oﬃce to pick up or drop oﬀ
paper case ﬁles, IRS faced logistical challenges in ensuring that staﬀ stayed socially distant. IRS
also had to ensure that taxpayers were kept safe during interactions with staﬀ.

To understand and address enforcement staﬀ concerns, IRS collected feedback in both formal
and informal settings. For example, one division held virtual town hall meetings, while others
sought feedback through an electronic mailbox. IRS managers said that staﬀ feedback helped
them address staﬀ questions and concerns about resuming enforcement activities from safety
and practical standpoints.

Page 374 GAO-22-105051
In addition to these agency-wide changes, individual IRS divisions made changes to address
speciﬁc pandemic-related challenges. We discuss some division-speciﬁc changes below, based on
interviews with enforcement managers and a review of related documents.

LB&I examinations. LB&I managers stated that most of their examiners were already telework
ready, but that LB&I relied heavily on paper processes for case ﬁles located in its oﬃces. LB&I
arranged for staﬀ to go into the oﬃce as needed to retrieve, print, and scan ﬁles and also provided
staﬀ with printers and scanners to work remotely. Additionally, LB&I changed its Information
Document Request process. Normally, when taxpayers do not comply with a request to provide
IRS with documents in a timely manner, IRS is authorized to obtain taxpayer documents through
a summons. LB&I approved changes to make time frames more ﬂexible and reduce the use
of summonses since many corporate taxpayers’ employees were not working in an oﬃce. The
division also established a process by which examiners could virtually close an examination case
without having to go into an oﬃce.

SB/SE ﬁeld and oﬃce examinations. Given challenges with starting new examinations, SB/SE directed
oﬃce examiners to work taxpayer claims and ongoing examinations that could be worked
virtually. SB/SE used more virtual technology to allow ﬁeld examiner staﬀ to safely interact with
taxpayers. Video conferencing proved to be popular with staﬀ, and IRS is obtaining more licenses
to allow its use. If examiners needed to retrieve documents from taxpayers, they arranged to meet
outdoors. Further, SB/SE developed procedures to exchange information virtually through secure
communication channels and set up a virtual Taxpayer Digital Communications pilot project at the
Philadelphia campus.

W&I correspondence examinations. Because W&I examiners had not previously teleworked, IRS
issued a memorandum in May 2020 that allowed work to be done outside of the oﬃce. To make
the work portable, W&I managers scanned examination case ﬁles until safety protocols were
established, and then clerks did the scanning. W&I also arranged for examiners to visit the campus
to pick up paper ﬁles as needed. With the shutdown of mail operations, W&I initiated e-fax to
allow taxpayers to send IRS documents. When phone and mail operations became more available
during fall 2020, W&I changed telephone scripts and call routing to shorten call times and take
more calls, according to IRS oﬃcials. IRS also initiated a call-back function in January 2021 that
allows a taxpayer to receive a call from IRS rather than remain on hold.

AUR. Although AUR document matching is automated, interactions with taxpayers rely on a paper
process. During the early stages of the pandemic, AUR management brought program managers
onto the campuses to assess work to be done based on critical needs and staﬀ availability.
Managers determined what work was portable and which staﬀ should work at the campuses.
Because AUR’s systems did not allow for document scanning, management had to decide how
many clerks to bring in to a campus to prepare paper case ﬁles for AUR staﬀ to pick up and work
remotely.

SB/SE ﬁeld collection. Field collection management said its staﬀ were telework ready but were
generally assigned more unﬁled return work during the early part of the pandemic because
this work is more automated and can be done without ﬁeld visits, unlike unpaid tax cases. Field
collection management provided guidance to carve out exceptions for certain unpaid tax work to
continue, such as egregious employment tax returns, pyramid schemes, high-income taxpayers,

Page 375 GAO-22-105051
and high-risk cases where the government might lose the opportunity to collect money. Since
September 2020, SB/SE has been changing the case mix to get back to a normal distribution of
unﬁled return and unpaid tax cases.

SB/SE campus collection.618 Given that staﬀ generally were not telework ready, management used
the shutdown period to train staﬀ to use new equipment to access phones and online systems.
Automated Collection System (ACS) management said they opened some phone lines by mid-
April of 2020. SB/SE suspended collection actions, but taxpayers could respond to collection
notices and ask for hardship exceptions or to pay in installments. ACS and Compliance Services
Collection Operations (CSCO) managers said that when the print sites reopened at the end of
June 2020, IRS prioritized which notices involving unpaid taxes to send based on factors such as
operational status and staﬀ’s ability to help taxpayers settle their accounts and to respond to
taxpayer inquiries. IRS remained behind on notices until September 2020 after deciding to freeze
sending new notices so that it could catch up on processing older notices. As a result, IRS delayed
issuing some notices until November 2020.

CSCO set up dedicated e-fax lines so that taxpayers could submit documents electronically. SB/
SE also scanned documents received by mail as much as possible to get work to remote staﬀ.
Further, ACS has been piloting an online text chat service with taxpayers, allowing taxpayers to
authenticate their identities, send documents, and settle speciﬁc account issues. Another new
feature allows taxpayers to request a telephone call-back from IRS at a speciﬁc time rather than
remaining on hold. Also, IRS is attempting to reduce the number of people calling by redirecting
taxpayers to online resources. Managers said that expanding self-help options has been a
successful change to operations during the pandemic.

Lessons learned. Our review also looked at how IRS is tracking lessons learned from changes
made to resume enforcement activities. Although IRS has not yet ﬁnalized its plans, enforcement
managers across IRS said they were considering how to track the lessons learned. Similarly, these
IRS managers stated that they are considering ways to assess changes that have been working
well during the pandemic and that should be kept or revised. Recognizing a hesitancy to abandon
processes and procedures that worked prior to the pandemic, IRS oﬃcials said they are asking
staﬀ to consider whether the pandemic-induced changes would help them work enforcement
cases outside of a pandemic environment.

IRS managers stated that the agency recognizes the need to move from paper-based to more
digital processes. Agency oﬃcials said that responding to challenges posed by the pandemic has
encouraged this transition and has demonstrated that it can be done. Virtual tools that allow for
video communication and sharing information electronically have shown signs of being useful.
As a result, the enforcement managers we interviewed said these tools will be made available for
staﬀ if they ﬁnd them useful and if taxpayers are willing to participate. Similarly, IRS is looking for
ways to expand acceptance of digital signatures on electronic documents.

618Campus collection eﬀorts to collect unpaid taxes or unﬁled tax returns include the Automated Collection System,
which sends notices and handles taxpayer calls, and the Compliance Services Collection Operations, which screens and
sorts incoming taxpayer responses and creates inventory batches that are sent to the appropriate unit to be worked.

Page 376 GAO-22-105051
Methodology
To describe the eﬀect of the pandemic on IRS tax enforcement caseloads and revenue, we
reviewed data in IRS systems for reporting on key enforcement programs using data points as
of the end of June in ﬁscal years 2019 (prior to the pandemic), 2020 (during the height of the IRS
shutdown due to the pandemic), and 2021 (the most recent available month). We identiﬁed the
key programs by reviewing data on all of the enforcement activities of the three IRS divisions
responsible for individual and business taxpayers—SB/SE, W&I, and LB&I—and, in consultation
with IRS oﬃcials, we selected those areas generally recognized as having the greatest eﬀects on
the use of IRS resources to enforce tax laws. To describe the eﬀect of the pandemic on staﬃng,
we reviewed data on IRS staﬀ use of Weather and Safety Leave and time working remotely or in
the oﬃce. We assessed the reliability of the data by reviewing existing information about data
IRS collects—speciﬁcally, the data dictionary and data validation methods—and by interviewing
oﬃcials responsible for data reporting. We determined that the data were suﬃciently reliable for
the purpose of describing changes to IRS enforcement activities during the pandemic.

To identify the challenges faced by IRS during the pandemic and changes made to address
those challenges, including any lessons learned, we interviewed division oﬃcials responsible for
managing the programs and reviewed supporting data and documentation. We summarized the
challenges and IRS’s changes and interviewed IRS oﬃcials to corroborate the summary.

Agency Comments
We provided IRS, the Department of the Treasury, and the Oﬃce of Management and Budget with
a draft of this enclosure. They did not have any comments on this enclosure.

GAO’s Ongoing Work
We will continue to monitor IRS’s administration of tax enforcement programs.

Related GAO Product
High Risk Series: Dedicated Leadership Needed to Address Limited Progress in Most High-Risk Areas.
GAO-21-119SP. Washington, D.C.: March 2, 2021.

Contact information: James R. McTigue Jr., (202) 512-6806, mctiguej@gao.gov

Page 377 GAO-22-105051
Postal Service
The U.S. Postal Service’s mail volume increased, its on-time performance declined, and its ﬁnances
declined in the third quarter of ﬁscal year 2021 as compared to the same period in 2020, which
was just after the onset of the COVID-19 pandemic.

Entity involved: U.S. Postal Service

Background
The U.S. Postal Service (USPS) plays a critical role in the nation’s communication and commerce,
a role highlighted in 2020 as USPS delivered billions of pieces of mail throughout the COVID-19
pandemic including ballots, Census forms, and recovery rebate checks, in addition to an
unprecedented surge of packages. As an independent establishment of the executive branch,
USPS is expected to provide aﬀordable, quality, and universal postal service. USPS is also expected
to be ﬁnancially self-suﬃcient by covering its expenses through revenues generated from the
sale of its products and services. However, USPS has not been able to cover its expenses since
ﬁscal year 2007 due to (1) long-term declines in its most proﬁtable mail products and (2) rising
expenses, such as for compensation and beneﬁts. As a result, USPS’s ﬁnancial viability has been
on our High-Risk List since 2009.

USPS’s mail volume, on-time delivery performance, revenue, and expenses changed signiﬁcantly
since the onset of the COVID-19 pandemic. In addition, we reported in April 2021 that USPS took
actions to reduce its costs that may have also reduced its on-time performance, such as reducing
late and extra mail transportation trips. We reported that some of these actions were suspended
by USPS prior to the general election on November 3, 2020. We also reported in April 2021 that
when comparing 2020 to 2019, USPS’s

• overall mail volume declined even with increases in the volume of packages;

• on-time performance fell with a steep decline nationwide in December 2020; and

• net loss grew even though revenue increased by $4.3 billion.619

To help USPS respond to the COVID-19 emergency, the CARES Act, as amended in late 2020,
provided USPS up to $10 billion in additional funding for COVID-19-related operating expenses.620

619We compared calendar year 2020 to calendar year 2019 in our April 2021 report.
620The Consolidated Appropriations Act, 2021 amended CARES Act authority to borrow this amount by providing that
USPS shall not be required to repay such amounts borrowed. Pub. L. No. 116-260, div. N, tit. VIII, § 801, 134 Stat. 1182,
2119 (2020) (amending Pub. L. No. 116-136, div. A, tit. VI, § 6001, 134 Stat. 281, 504-05 (2020)). As of July 29, 2021, USPS
had received all of the $10 billion available. The funds available under the CARES Act, as amended, are in addition
to USPS’s borrowing authority under 39 U.S.C. § 2005, which authorizes USPS to borrow up to $3 billion in any one
ﬁscal year and not more than $15 billion in total. As of June 30, 2021, USPS has borrowed $11 billion of the $15 billion
available.

Page 378 GAO-22-105051
Overview of Key Issues
Mail volume. In the third quarter of ﬁscal year 2021, overall mail volume increased by about
4.0 billion pieces (about 15.1 percent) when compared to the same period in ﬁscal year 2020
(see ﬁgure). This change is almost entirely attributable to Marketing Mail volumes rebounding
from signiﬁcant declines that occurred in the beginning of the COVID-19 pandemic. Marketing
Mail volume increased 4.3 billion pieces (39 percent) year over year. Mail volume trended toward
prepandemic levels, which meant increases for market-dominant products such as First Class
Mail (e.g., letters, cards, billing statements) and Marketing Mail (e.g., advertisements, ﬂyers,
newsletters), and decreases for competitive products (primarily packages), which spiked during
the early parts of the pandemic. For example, in the third quarter of ﬁscal year 2021, competitive
package volume was about 14 percent lower than it was in the same period in ﬁscal year 2020.

Change in Mail Volume for Selected Mail Types between Third Quarters in Fiscal Year 2020 and 2021 (in
thousands of piece of mail)

On-time performance. Nationally, while on-time performance for market-dominant products
was slightly lower during the third quarter of ﬁscal year 2021 as compared to the same period in
ﬁscal year 2020, performance continued to rebound from USPS’s December 2020 lows. On-time
performance for First Class Mail averaged 87.6 percent nationally from April through June 2021, as
compared to 90.8 percent for the same period in ﬁscal year 2020. Moreover, since the ﬁrst quarter
of ﬁscal year 2021 ended with a December on-time performance for First Class Mail of 69 percent,
the third quarter showed continued improvement as on-time performance for First Class Mail
increased from 87.27 percent in April 2021, to 87.59 percent in May 2021, to 87.95 percent in June
2021, but remained well below 2020 levels.

Revenue and expenses. USPS reported a net loss of $3 billion for the third quarter of ﬁscal year
2021, as compared to a net loss of $2.2 billion for the same period in ﬁscal year 2020.621 USPS
earned about $845 million more in operational revenue in the third quarter of ﬁscal year 2021

621Excluding non-cash workers’ compensation adjustments for each period that vary signiﬁcantly based on interest
rate and other actuarial revaluations, the loss for the 2021 third quarter would have been approximately $2.3 billion,
compared to a loss of approximately $2.4 billion for the same quarter last year.

Page 379 GAO-22-105051
when compared to the third quarter of ﬁscal year 2020. Revenue from USPS’s market-dominant
products increased by about $1.3 billion, mainly due to increased volume of marketing mail,
while revenue for USPS’s competitive products declined by about $466 million as package volume
decreased.

USPS’s expenses increased by approximately $1.6 billion for the third quarter in ﬁscal year 2021
when compared to the same period in ﬁscal year 2020. USPS attributed the increase to several
factors, such as

• a $900 million increase in worker’s compensation expenses due to changes in interest rates;

• a retiree health beneﬁts increase of $387 million, also due to changes in interest rates; and

• an increase in transportation expenses of $182 million, reﬂecting higher package volumes on
trucks, changes in transportation modes, and higher fuel prices.

COVID-19 funding. As of July 2021, USPS had received all $10 billion of its CARES Act funding.
USPS requested this funding for operating expenses incurred due to the COVID-19 emergency
from March 2020 to May 2021.622 These expenses included categories such as overtime, hiring and
training costs of new employees, absenteeism, and costs of sanitizing work areas (see table).

622USPS can request this funding for COVID-19 related expenses under its memorandum of understanding with the
Department of the Treasury. As required by statute, USPS must determine that, due to the COVID-19 emergency, USPS
will not be able to fund operating expenses without borrowing money. CARES Act, Pub. L. No. 116-136, div. A, tit. VI, §
6001, 134 Stat. 281, 504-05 (2020).

Page 380 GAO-22-105051
Amounts Requested by USPS under the CARES Act, as Amended, for COVID-19-Related Operating Expenses
Incurred Mar. 2020–May 2021
Dollars
requested
Expense category (in millions)

Supplies & services $303

COVID leave 492

Transportation 73

Hiring/training costs for new employees 159

Increase in carriers out after 6 p.m. 66

Overall overtime cost increase 2,000

Ineﬃciency factor—general ineﬃciency of 2 percent of salaries and beneﬁts for time spent 1,495
managing personal protective equipment, sanitizing work areas, social distancing in postal facilities,
among other things.

Ineﬃciency Factor—New Hires/Absenteeism 401

Additional expenses incurred in providing USPS’s statutorily mandated infrastructure and operations 5,011
a
during the COVID-19 emergency

Total (rounded): $10,000

Source: GAO analysis of U.S. Postal Service (USPS) data. | GAO-22-105051

Note: These funds are made available by the CARES Act, Pub. L. No. 116-136, div. A, tit. VI, § 6001, 134 Stat. 281, 504-05 (2020),
as amended by the Consolidated Appropriations Act, 2021, Pub. L. No. 116-260, div. N, tit. VIII, § 801, 134 Stat. 1182, 2119
(2020). USPS requests these funds pursuant to a memorandum of understanding with the Department of the Treasury.
a
According to USPS’s memorandum of understanding with the Department of the Treasury, these additional expenses are due
to changes in product volumes attributable to the COVID-19 emergency versus pre-COVID forecasts in USPS’s ﬁnancial plans,
taking into account (1) the costs captured above for Supplies and Services, (2) the revenue and expenses of all products and
services, and (3) the actions USPS was unable to undertake to account for the reduction in volumes because of its statutory
mandates, such as the requirement to provide 6-days-a-week delivery. See, e.g., Pub. L. No. 116-260, div. E, tit. V, 134 Stat. 1182,
1423 (2020).

As stated above, USPS has not been able to cover its expenses with its revenues since ﬁscal year
2007 due to long-term declines in its most proﬁtable mail products and rising expenses, such
as for compensation and beneﬁts. USPS has maintained its operations since then through a
combination of actions, such as taking on debt and not making required payments for pensions
and retiree health care beneﬁts. USPS stated that it took those actions to preserve cash to
continue to provide universal postal service.

USPS’s cash balance has decreased from $25.5 billion at the end of March 2021 to $22.8 billion at
the end of June 2021. USPS oﬃcials stated that the main reason for this decrease was a $3 billion
repayment of short-term debt to the federal government and the expenditure of USPS’s CARES
Act funds. As of the end of June 2021, USPS had expended $9.6 billion of its CARES Act funds to
cover its operating expenses such as payroll and transportation costs. As of the end of July 2021,

Page 381 GAO-22-105051
USPS had received all $10 billion of its CARES Act funding. As we reported in April 2021 and July
2021, USPS continued to preserve cash in 2021 by forgoing making required retiree health care
and pension payments. However, USPS reported that its current cash balance is insuﬃcient to
support USPS’s $82 billion in annual operating expenses, its capital investments, and to prepare
for unexpected contingencies.

USPS’s cash balance could be reduced in the coming years. As stated above, USPS has received
all of its CARES Act funds. USPS’s volume of packages is now decreasing, which has reduced its
revenue growth. USPS also deferred payments of about $1.8 billion for the employer’s share of the
Social Security payroll tax on wages from March 27, 2020, through December 31, 2020, as allowed
under the CARES Act. Payment of half of this deferred amount is due on December 31, 2021, and
payment of the other half is due on December 31, 2022. 623

Methodology
To conduct this work, we analyzed USPS data on volume, on-time performance, and revenue and
expenses, as of June 2021, which were the latest data available. We used USPS’s 10-Q ﬁnancial
statement for the third quarter of ﬁscal year 2021. To determine the reliability of the data we used,
we interviewed relevant USPS oﬃcials about volume, on-time performance, and ﬁnancial data that
they either provided us directly or that we obtained from publicly available reports. They described
where the data came from, how they were collected, and controls in place to provide assurance
the data were complete and accurate. Based on this interview and relevant USPS documents we
reviewed, we determined all data used were suﬃciently reliable for the purposes of reporting on
USPS mail volumes, on-time performance levels, revenues, and expenses.

We also reviewed applicable federal laws and interviewed USPS oﬃcials.

Agency Comments
We provided USPS and the Oﬃce of Management and Budget (OMB) with a draft of this enclosure
for review and comment. USPS provided technical comments that we incorporated as appropriate.
USPS also provided general comments, which are reproduced in appendix X. In its comments,
USPS provided additional context about its ﬁnancial condition, recent service performance, and
anticipated plans to address these issues. OMB did not provide any comments.

Related GAO Products
U.S Postal Service: Better Use of Climate Data Could Enhance the Resilience of Postal Facilities to
Climate Change Eﬀects. GAO-21-104152. Washington, D.C.: September 30, 2021.

U.S. Postal Service: Further Analysis Could Help Identify Opportunities to Reduce Injuries among
Non-Career Employees. GAO-21-556. Washington, D.C.: August 17, 2021.

623CARES Act, Pub. L. No. 116-136, div. A, tit. II, § 2302, 134 Stat. 281, 351–52 (2020).

Page 382 GAO-22-105051
U.S. Postal Service: Volume, Performance, and Financial Changes since the Onset of the COVID-19
Pandemic. GAO-21-261. Washington, D.C.: April 29, 2021.

High Risk Series: Dedicated Leadership Needed to Address Limited Progress in Most High-Risk Areas.
GAO-21-119SP. Washington, D.C.: March 2, 2021.

High Risk Series: Restructuring the U.S. Postal Service to Achieve Sustainable Financial Viability.
GAO-09-937SP. Washington, D.C.: July 28, 2009.

Contact information: Jill Naamane, Acting Director, (202) 512-2834, naamanej@gao.gov

Page 383 GAO-22-105051
Appendix II: Highlights Pages from Recently Issued GAO
COVID-19 Products

Contracting Flexibilities
We issued COVID-19 Contracting: Actions Needed to Enhance Transparency and Oversight of Selected
Awards, GAO-21-501, on July 26, 2021.

Page 384 GAO-22-105051
Page 385 GAO-22-105051
VA CLC Infection Prevention
We issued COVID-19: VA Should Assess Its Oversight of Infection Prevention and Control in Community
Living Centers, GAO-21-559, on July 28, 2021.

Page 386 GAO-22-105051
Contractor Paid Leave Reimbursements
We issued COVID-19 Contracting: Contractor Paid Leave Reimbursements Could Provide Lessons
Learned for Future Emergency Responses, GAO-21-475, on July 28, 2021.

Page 387 GAO-22-105051
Child Welfare
We issued Child Welfare: Pandemic Posed Challenges, but also Created Opportunities for Agencies to
Enhance Future Operations, GAO-21-483, on July 29, 2021.

Page 388 GAO-22-105051
Implementation of Paycheck Protection Program
We issued Paycheck Protection Program: SBA Added Program Safeguards, but Additional Actions Are
Needed, GAO-21-577, on July 29, 2021.

Page 389 GAO-22-105051
BOP Response to COVID-19
We issued Bureau of Prisons: BOP Could Further Enhance its COVID-19 Response by Capturing and
Incorporating Lessons Learned, GAO-21-502, on July 29, 2021.

Page 390 GAO-22-105051
Assessing Vendors and Contracting Lessons Learned
We issued COVID-19 Contracting: Opportunities to Improve Practices to Assess Prospective Vendors and
Capture Lessons Learned, GAO-21-528, on July 29, 2021.

Page 391 GAO-22-105051
Oversight of COVID-19 Assistance
We issued COVID-19 Pandemic: Actions Needed to Improve Federal Oversight of Assistance to
Individuals, Communities, and the Transportation Industry, GAO-21-105202, on July 29, 2021.

Page 392 GAO-22-105051
Economic Injury Disaster Loans
We issued Economic Injury Disaster Loan Program: Additional Actions Needed to Improve
Communication with Applicants and Address Fraud Risks, GAO-21-589, on July 30, 2021.

Page 393 GAO-22-105051
Page 394 GAO-22-105051
Biodefense Preparedness
We issued Biodefense: After-Action Findings and COVID-19 Response Revealed Opportunities to
Strengthen Preparedness, GAO-21-513, on August 4, 2021.

Page 395 GAO-22-105051
Hospital Capacity Data Collection
We issued COVID-19: HHS’s Collection of Hospital Capacity Data, GAO-21-600, on August 5, 2021.

Page 396 GAO-22-105051
Federal Debt Management
We issued Federal Debt Management: Treasury Quickly Financed Historic Government Response to the
Pandemic and Is Assessing Risks to Market Functioning, GAO-21-606, on August 17, 2021.

Page 397 GAO-22-105051
Immigration Courts
We issued COVID-19: Improvements Needed in Guidance and Stakeholder Engagement for Immigration
Courts, GAO-21-104404, on August 31, 2021.

Page 398 GAO-22-105051
Medicaid HCBS Waivers
We issued Medicaid Home and Community-Based Services: Evaluating COVID-19 Response Could Help
CMS Prepare for Future Emergencies, GAO-21-104401, on September 8, 2021.

Page 399 GAO-22-105051
Food Box Program
We issued USDA Food Box Program: Key Information and Opportunities to Better Assess Performance,
GAO-21-353, on September 8, 2021.

Page 400 GAO-22-105051
Exposure Notiﬁcation
We issued Exposure Notiﬁcation: Beneﬁts and Challenges of Smartphone Applications to Augment
Contact Tracing, GAO-21-104622, on September 9, 2021.

Page 401 GAO-22-105051
Page 402 GAO-22-105051
Characteristics of Paycheck Protection Program Borrowers
We issued Paycheck Protection Program: Program Changes Increased Lending to the Smallest
Businesses and in Underserved Locations, GAO-21-601, on September 21, 2021.

Page 403 GAO-22-105051
DOD Pandemic Response
We issued COVID-19: Defense-Wide Working Capital Fund Cash Management and Defense Logistics
Agency Pandemic Response, GAO-21-104590, on September 22, 2021.

Page 404 GAO-22-105051
Dedicated User Fees
We issued COVID-19: Reviewing Existing Policies Could Help Selected Agencies Better Prepare for
Dedicated User Fee Revenue Fluctuations, GAO-21-104325, on September 29, 2021.

Page 405 GAO-22-105051
Telework Cybersecurity
We issued COVID-19: Selected Agencies Overcame Technology Challenges to Support Telework but Need
to Fully Assess Security Controls, GAO-21-583, on September 30, 2021.

Page 406 GAO-22-105051
HUD Oversight
We issued COVID-19: Additional Risk Assessment Actions Could Improve HUD Oversight of CARES Act
Funds, GAO-21-104542, on September 30, 2021.

Page 407 GAO-22-105051
IHS COVID-19 Contracting
We issued COVID-19 Contracting: Indian Health Service Used Flexibilities to Meet Increased Medical
Supply Needs, GAO-22-104745, on October 14, 2021.

Page 408 GAO-22-105051
Federal Reserve Lending Programs
We issued Federal Reserve Lending Programs: Credit Markets Served by the Programs Have Stabilized,
but Vulnerabilities Remain, GAO-22-104640, on October 19, 2021.

Page 409 GAO-22-105051
IRS IT Investments
We issued Information Technology: Cost and Schedule Performance of Selected IRS Investments,
GAO-22-104387, on October 19, 2021.

Page 410 GAO-22-105051
Appendix III: List of Ongoing GAO Work Related to COVID-19
as of September 30, 2021
Accountability for Bureau of Indian Education Spending of COVID-19 Funds

Agencies' Human Capital Flexibilities in Response to Coronavirus Pandemic

Agencies' Readiness and Use of Telework for COVID-19 Response

Air Travel Disease Research and Development

Aviation Contact Tracing

Aviation Operations in a Pandemic Environment

CARES Act Assistance to Farmers

CARES Act International Humanitarian Assistance

CARES Act Title IV Federal Reserve Facilities II

U.S. Customs and Border Protection Trade Facilitation

Contact Tracing App Technology Assessment

Behavioral Health Impacts

Defense Production Act Capstone

COVID-19 Diagnostic Testing

COVID-19 Disparities

COVID-19 in Nursing Homes: Data and Challenges

COVID-19 in Nursing Homes: Federal Policies

Meat and Poultry Worker Safety

COVID-19 Services for Older Adults

COVID-19 Testing Lessons Learned

Unemployment Assistance for Contingent Workers

Vaccine Allocation Methodology

Vaccine Distribution and Communication

Page 411 GAO-22-105051
Employer Tax Provisions

Critical Manufacturing Supply Chain

Department of State Overseas Operations

COVID-19: State Repatriation

Department of Defense Advance and Progress Payments during COVID-19 National Emergency

Election Administration during the COVID-19 Pandemic

Election Assistance Commission Guidance and Grants Oversight during the COVID-19 Pandemic

Federal Agencies' Reentry

Financial Regulatory Oversight during COVID-19

Freedom of Information Act Processing during COVID-19

Fraud Risks in Small Business Administration (SBA) Pandemic Relief Programs

Health Insurance Loss

Department of Health and Human Services (HHS) Medicare Telehealth Waivers for COVID-19

HHS Public Relations Campaign

HHS's Public Health Situational Awareness Capability

Housing Finance System in the Pandemic

IHS Response to COVID-19

Department of the Interior and Department of the Treasury's COVID-19 Response to Tribes

K-12 Disconnected Students during the COVID-19 Pandemic

Medicaid Telehealth during COVID-19

Operation Warp Speed

Oversight of Unemployment Insurance

Pandemic Learning Loss

Political Interference at Selected HHS Agencies

Post-COVID-19 Federal Space Planning

Page 412 GAO-22-105051
Regulatory Flexibilities for COVID-19 Response

Regulatory Flexibilities Timeline

SBA Assistance to Venues and Restaurants

Scientiﬁc Integrity at Selected HHS Agencies

Social Security Administration Service Delivery during COVID-19

State Small Business Credit Initiative Funds for COVID-19 Response

Strategic National Stockpile Contents and Management Review

Tax Policy Eﬀects on Businesses by Sex, Race, and Ethnicity

Tax Policy Eﬀects on Households by Sex, Race, Ethnicity

Tribal Epidemiological Data Access

Transportation Security Administration COVID-19-Related Directives for Transportation Systems

Unemployment Insurance Risks and Transformation

U.S. Department of Agriculture Human Pandemic Preparedness Plan for Food Safety Inspections

Vaccine Development

Page 413 GAO-22-105051
Appendix IV: Comments from the Department of Health and
Human Services

Page 414 GAO-22-105051
Page 415 GAO-22-105051
Page 416 GAO-22-105051
Appendix V: Comments from the Department of Homeland
Security

Page 417 GAO-22-105051
Page 418 GAO-22-105051
Page 419 GAO-22-105051
Page 420 GAO-22-105051
Appendix VI: Comments from the Internal Revenue Service

Page 421 GAO-22-105051
Page 422 GAO-22-105051
Appendix VII: Comments from the Department of Labor,
Employment and Training Administration

Page 423 GAO-22-105051
Page 424 GAO-22-105051
Page 425 GAO-22-105051
Appendix VIII: Comments from the Department of Labor,
Occupational Safety and Health Administration

Page 426 GAO-22-105051
Appendix IX: Comments from the Department of the Treasury

Page 427 GAO-22-105051
Page 428 GAO-22-105051
Page 429 GAO-22-105051
Appendix X: Comments from the United States Postal Service

Page 430 GAO-22-105051
Page 431 GAO-22-105051
Appendix XI: Comments from the Department of Veterans
Aairs

Page 432 GAO-22-105051
Contacts
Report Director(s)
Jessica Farb
Managing Director, Health Care, (202) 512-7114, farbj@gao.gov

Congressional Relations
A. Nicole Clowers, Managing Director, clowersa@gao.gov, (202) 512-4400

Public Aairs
Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800

Strategic Planning and External Liaison
Stephen J. Sanford, Managing Director,spel@gao.gov, (202) 512-4707

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 cost of production and distribution. Pricing and ordering
information is posted on our website.

To Report Fraud, Waste, and Abuse in Federal Programs
Website: http://www.gao.gov/fraudnet/fraudnet.htm
Automated answering system: (800) 424-5454 or (202) 512-7470

Connect with GAO
Connect with GAO on Facebook, Flickr, LinkedIn, Twitter, and YouTube. Subscribe to our RSS Feeds or E-mail
updates. Listen to our Podcasts. Visit GAO on our website and read the Watchblog.

(105051) Page 433 GAO-22-105051

File and source

File: covid-19-additional-actions-needed-to-improve-accountability-and-progr-f867cf0a.pdf
Size: 17,484,635 bytes
SHA-256: f867cf0af0604c839ce485cfe782e542f69fa352ed71dda3717738ebffec0240
Our copy: covid-19-additional-actions-needed-to-improve-accountability-and-progr-f867cf0a.pdf
Original: www.gao.gov