PRAC Fraud Prevention Alert — $79 Billion in Preventable Pandemic Fraud (June 2025)

Record facts

Summary

A fraud prevention alert issued in June 2025 by the Pandemic Response Accountability Committee (PRAC) on SBA's COVID-19 EIDL program and PPP and DOL's Pandemic Unemployment Insurance programs. The alert describes how the PRAC randomly sampled 662,000 identity records from 67.5 million funded applications and had the Social Security Administration verify the SSNs, names and dates of birth. It reports that 23,854 records had questionable verification status and estimates that $79 billion was disbursed to applicants using over 1.4 million potentially stolen or invalid SSNs. The alert breaks the estimate into $55.8 billion, $13.8 billion and $9.8 billion for COVID-19 EIDL, PPP and Pandemic UI, respectively, and includes a table of verification codes by program. It closes by describing how pre-award vetting with PRAC data analytics tools could flag such applications.

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

Full text

FRAUD PREVENTION ALERT:  
Pre-Award Vetting Using Data 
Analytics Could Have Prevented Over 
$79 Billion in Potentially Fraudulent 
Pandemic Relief Payments
The PRAC Analyzed Applications Across Major Pandemic Relief Programs 
to Identify Potential Fraud Using Stolen or Invalid Social Security Numbers 
As part of our independent oversight of the $5 trillion in COVID-19 relief programs, the Pandemic Response 
Accountability Committee (PRAC) conducts in-depth data analysis across multiple federal programs to identify 
potential fraud, waste, abuse, and mismanagement and to provide agencies with lessons learned to prevent future 
program fraud. 
This PRAC fraud prevention alert focused on some of the largest pandemic relief programs: the Small Business 
Administration’s (SBA) COVID-19 Economic Injury Disaster Loan (COVID-19 EIDL) program and Paycheck Protection 
Program (PPP), and the Department of Labor’s (DOL) pandemic-related Unemployment Insurance (UI) programs. 
In 2023, the SBA Office of Inspector General (OIG) and the DOL OIG estimated that the total amount of fraud and 
improper payments for these programs is nearly $400 billion.1
In this alert, the PRAC estimates the amount of potential fraud across these programs stemming from the use 
of stolen or invalid Social Security numbers (SSNs), and illustrates how pre-award vetting using the PRAC’s data 
analytics tools could have mitigated this risk.
Estimate of over 
1.4 million questionable 
Social Security numbers
SBA COVID-19 
EIDL Program 
SBA Paycheck 
Protection Program
DOL Pandemic 
Unemployment 
Insurance Programs
Totaling 
$79 billion in 
potentially fraudulent 
payments
$79 
billion
Random sample 
of 662,000 
identity records
23,854 records 
with questionable 
verification status
Statistical
Modeling
The PRAC has previously issued numerous reports detailing the widespread identity fraud that occurred in 
connection with pandemic relief programs and the need for the federal government to address the issue.2 
Leveraging our cross-agency partnerships and applying innovative data analytics tools and resources, the PRAC 
estimates that these pandemic programs disbursed approximately $79 billion in potential fraudulent payments due 
1	
SBA OIG “COVID-19 Pandemic EIDL and PPP Loan Fraud Landscape” report, and DOL “OIG Oversight of the Unemployment Insurance Program” report.
2	
Examples of PRAC reports include Key Insights: Identity Fraud Reduction and Redress in Pandemic Response Programs and Best Practices and Lessons 
Learned from the Administration of Pandemic-Related Unemployment Benefits Programs.
June 2025

2
Pandemic Response Accountability Committee | June 2025
FRAUD PREVENTION ALERT: Pre-Award Vetting Using Data Analytics Could Have 
Prevented Over $79B in Potentially Fraudulent Pandemic Relief Payments
to the use by applicants of over 1.4 million potentially stolen or invalid SSNs. The findings in this fraud prevention 
alert overlap with some of the findings of prior DOL OIG and other OIG reports.3
PRAC’s Efforts to Estimate the Amount of Potentially Fraudulent Payments in the 
COVID-19 EIDL, PPP, and UI Programs
To conduct our analysis, the PRAC’s Pandemic Analytics Center of Excellence (PACE) randomly sampled 662,000 
identity records from a population of 67.5 million funded applications across the COVID-19 EIDL, PPP, and Pandemic 
UI programs.4 Using our authorities in the Coronavirus Aid, Relief, and Economic Security Act of 2020 (CARES Act), 
and pursuant to a Memorandum of Understanding (MOU) with the Social Security Administration (SSA), we shared 
the sampled SSNs and the associated names and dates of birth (DOB) from the 662,000 records with the SSA for 
verification. 
Using those results, our data scientists then employed a statistical sampling method to estimate that 1.4 to 1.5 
million potentially stolen or invalid SSNs were used to obtain benefits from the programs.5 These SSNs were either 
never issued by SSA or had mismatched identifying information (name and/or DOB) when compared to SSA records, 
indicating that they were potentially stolen or being used without authorization.6 The PRAC estimates that $79.41 
billion to $79.54 billion was disbursed to applicants using these potentially stolen or invalid SSNs. 
As detailed later in this fraud prevention alert, this type of fraud is readily preventable using the authorities and 
analytics platform that Congress has provided to the PRAC. Moreover, having an identity verification capability using 
a recognized source such as SSA enables agencies to effectively focus their efforts on a significantly smaller subset 
of the applications for further examination (approximately 4 percent of the records in our analysis). Additionally, 
identifying applications that are misusing SSN, name, and DOB information enables the PRAC to use its analytics 
capabilities to determine if fraudsters have improperly obtained benefits from other federal programs using that 
identifying information or other linked identifying information (such as IP address, bank account, and address).
Without the PRAC’s unique data analytics capabilities and the tools authorized by Congress, we would not have 
been able to identify this significant amount of potential fraud across these programs. However, in order for these 
unique capabilities to be effective, the PRAC needs to be able to obtain accurate and complete data from agencies. 
For example, because PPP applicants were not required to provide DOB information, the PRAC was able to obtain 
such information for only approximately 13 percent of PPP loan applicants.7 Requiring applicants to provide data like 
DOBs is essential for detecting identity fraud, conducting oversight analysis, and strengthening program integrity. 
3	
For instance, the potentially invalid SSNs we identified for this fraud prevention alert matched DOL OIG’s previous analysis. Specifically, 95 percent of 
questionable SSNs in our random sample for Pandemic UI were previously identified by DOL OIG. Additionally, 23 percent were already included in DOL OIG 
suspected fraud totals in their prior published reports.
4	
An identity record contains key identifying details such as an individual’s SSN, first name, last name, and date of birth submitted in their applications for 
the COVID-19 EIDL, PPP, or Pandemic UI programs. While individuals may have applied to various programs or submitted multiple applications within the 
same program, we consolidated their information into a single record to be verified by the SSA and avoid duplications. Additionally, a single application can 
include multiple individuals. For these reasons, we use the term “records” rather than “applications” throughout this fraud prevention alert for clarity.
5	
The number of individuals that used these potentially stolen or invalid SSNs to apply for funds across these programs is likely to be less than 1.4 to 1.5 
million given that some fraudsters are presumably responsible for the use of more than one of the potentially invalid SSNs.
6	
We would expect some false positives in these verification results. Performing an automated match is subject to error, including unrecognized nicknames 
or misspellings, transposed numbers, and data entry and clerical mistakes. For example, we observed examples in which records were flagged as an 
incorrect name when an individual’s first name and last name were transposed. We cannot estimate the exact proportion of false positives without an 
in-depth understanding of the verification algorithm used by SSA or/and without addressing the data cleanliness issues in the pandemic program data. 
To maintain the security of its verification process, SSA does not release the agency’s matching logic concerning how it verifies SSNs.
7	
Date of birth was not required on the PPP application form; however, PPP lenders may have included the applicant DOB in the loan submission to the SBA.

3
Pandemic Response Accountability Committee | June 2025
FRAUD PREVENTION ALERT: Pre-Award Vetting Using Data Analytics Could Have 
Prevented Over $79B in Potentially Fraudulent Pandemic Relief Payments
Further, as evidenced in this fraud prevention alert, applying our innovative tools and the subject matter expertise 
of our data scientists can enhance information-sharing and identity verification before disbursing funds, which can 
strengthen program integrity, ensuring funds are paid to legitimate applicants, as Congress intends, including during 
future federal emergency programs, such as natural disasters or financial crises, or regular annual appropriations. 
Details of our Analysis
How We Sampled and Verified Records from Pandemic Programs
To conduct this analysis, we randomly sampled 662,000 records from a population of 67.5 million funded 
applications across COVID-19 EIDL, PPP, and Pandemic UI programs. The sample sizes for COVID-19 EIDL, PPP, and 
Pandemic UI source tables were set to 256,000, 256,000, and 150,000, respectively. We only considered individual 
applicants who received funding. Each individual and transaction combination had an equal probability of being 
sampled from its respective programs. 
Using the legal authorities included in the CARES Act of 2020, and pursuant to an MOU between the PRAC and 
SSA, the PRAC requested that SSA verify the validity of the SSNs supplied in the 662,000 records. The PRAC did 
not request or obtain SSN data from SSA in connection with this analytics review. Rather, as described below, 
we requested that SSA simply verify information that applicants had provided in those COVID-19 EIDL and PPP 
applications and Pandemic UI claims, which SSA performed and then provided us with the results. Specifically, 
the PRAC provided SSA with the SSN, name, and (where available) DOB information from the randomly selected 
662,000 records. It took SSA less than one week to provide these responses to the PRAC for the 662,000 records.
Specifically, for the 662,000 records in our random sample set, we asked SSA to verify the following:
1.	Is the SSN valid?
2.	If the SSN is valid, does the name associated with the SSN on the application match SSA records?8
3.	If the SSN is valid, does the date of birth associated with the SSN on the application match SSA records? 
4.	Is the SSN used on the application associated with a deceased individual?
SSA processed the data and provided the PRAC with an identity verification code for each record (including SSN, 
name, DOB) they were asked to verify. The codes provided by SSA indicated that the identifying information (SSN, 
name, DOB) provided by applicants in almost 24,000 of the 662,000 did not fully match the information in SSAs 
records, indicating potential fraud (See Table 1 below for a breakdown by verification code).
8	
As noted above, SSA does not share its matching methodology given the sensitive nature of the data and to maintain the security of its SSN verification 
process. This includes its matching methodology concerning maiden names, hyphenation issues, or other matching scenarios to determine the verification 
outcome.

4
Pandemic Response Accountability Committee | June 2025
FRAUD PREVENTION ALERT: Pre-Award Vetting Using Data Analytics Could Have 
Prevented Over $79B in Potentially Fraudulent Pandemic Relief Payments
Table 1. Disbursed COVID-19 EIDL and PPP Loans, and Pandemic UI Claims Associated with SSNs with 
Verification Codes Indicating Potential Fraud 
SSA Identity Verification Codes  
PPP 
COVID-19 EIDL 
Pandemic UI 
Total 
SSN Not Issued 
1,509 
482 
139 
2,130 
SSN with Incorrect Name 
11,476 
5,603 
1,852 
18,931 
SSN with Incorrect Date of Birth 
541 
1,803 
439 
2,783 
SSN Not Verified – Other Reason9 
1 
5 
4 
10 
Subtotal – Questionable SSNs
13,527
7,893
2,434
23,854
SSN Matched SSA’s Records 
242,473 
248,107 
147,566 
638,146 
Total 
256,000 
256,000 
150,000 
662,000 
Source: PRAC analysis of SSA’s verification of the randomly sampled SSNs and associated names and DOBs resulting from SSA’s Enumeration 
Verification System (EVS).
In addition, SSA informed the PRAC that 11,514 of the 638,146 fully verified sample records (the SSN, name, and 
DOB provided by the applicant fully matched the information in SSA’s records) were associated with individuals who 
were deceased as of the date that SSA responded to the PRAC’s request.10 We have not included those SSNs and 
the associated loans in this fraud prevention alert because we first need to determine whether those individuals 
were alive at the time the COVID-19 EIDL or PPP applications or Pandemic UI claims were submitted. Because SSA 
told us that it is not permitted by law to provide date of death information to the PRAC, determining date of death 
requires the PRAC to obtain information from the Treasury Department’s Do Not Pay service for these deceased 
individuals. However, the need to confirm date of death does not impact this report’s finding that 23,854 records 
were potentially fraudulent because the identifying information provided by those applicants did not fully match 
SSA’s records.
How We Estimated the Amount of Potential Fraud Within and Across Pandemic Programs  
As a result of the SSA’s verification process, the PRAC was able to determine that 23,854 of the randomly selected 
records were potentially fraudulent. We then applied statistical modeling that involved taking repeated random 
samples, with replacement, followed by a subsequent analysis to infer the entire population for two parameters—
total payment amount at risk and potentially stolen or invalid SSN count. 
Overall, our statistical modeling estimated that the two agencies disbursed $79 billion to recipients whose 
applications used potentially stolen or invalid SSNs.11 Specifically, as indicated in Graph 1 below, the PRAC estimated 
this resulted in disbursements of $55.8 billion, $13.8 billion, and $9.8 billion in COVID-19 EIDL, PPP, and Pandemic 
9 
SSA provided no additional information on the verification code “SSN Not Verified – Other Reason.”
10 
SSA can only provide death indicator information for fully verified SSNs.
11 
As noted above, this estimated amount of potential fraud across COVID-19 EIDL, PPP, and Pandemic UI overlaps in part with fraud identified by previous 
reports issued by SBA OIG, DOL OIG, and the U.S. Government Accountability Office (GAO), which highlighted identity theft as one of the leading risk factors 
in pandemic fraud. The $79 billion is a result of our statistical estimate and therefore we do not have the actual records to cross reference other OIGs’ 
findings to determine the exact overlap.

5
Pandemic Response Accountability Committee | June 2025
FRAUD PREVENTION ALERT: Pre-Award Vetting Using Data Analytics Could Have 
Prevented Over $79B in Potentially Fraudulent Pandemic Relief Payments
UI programs, respectively. While the estimated count of potentially stolen or invalid SSNs used in applications for the 
Pandemic UI programs is higher than those for the other two programs, the total dollar amount of potential fraud is 
lower than the other two programs. One possible explanation could be the higher-volume and lower-amount nature 
of individual Pandemic UI claims when compared with COVID-19 EIDL and PPP applications.  
Graph 1. Estimated COVID-19 EIDL, PPP, and Pandemic UI Payments Associated with SSNs with 
Verification Codes Indicating Potential Fraud
Amount (Billions $)
$55.84B
$13.80B
$9.83B
PPP
EIDL
UI
SSN Count*
479,518
230,449
769,438
PPP
EIDL
UI
* Note: The amounts and counts shown here represent the population totals estimated from verified samples. The cross-program SSN total is 
not a direct sum of program SSN counts, as there are SSNs that were used for applications in multiple programs.  
Statistical Considerations and Assumptions 
Estimating the amount of potential fraud attributed to particular applicants is complex due to scenarios where 
multiple applicants could appear on the same pandemic loan application, or a single applicant could apply to 
multiple pandemic programs. Due to this complexity, we estimated the total amounts of potential fraud taking into 
account the combination of program participation, co-applicants, and funding amounts simultaneously. 
Since the underlying true population distributions for each of these are complex, we employed a “Bootstrap 
Sampling” statistical approach.12 We transformed transactions to individual applicant records by attributing amounts 
to applicants proportionally distributed based on the number of co-applicants, then summing funding amounts 
associated with each applicant. 
For example, if a COVID-19 EIDL application with a payment amount of $10,000 had five individual owners, the 
transformation would create five records listing each individual owner with $2,000 as the funding amount.13 We then 
resampled from SSA-verified sample records, with replacement, to reproduce the record-count of the full population 
and estimate the total payment amount for each verification category. The resulting categorized counts and amounts 
were used to calculate our population estimates within each program and their corresponding confidence intervals.  
By aggregating results across all bootstrap samples, we obtained more robust and reliable estimates, highlighting 
potential fraud without having to make any rigid assumptions about the data.
12	
See An Introduction to the Bootstrap, by B. Efron & R.J. Tibshirani.
13	
In COVID-19 EIDL applications, an individual owner is defined as (a) Proprietor, or (b) Limited partner or LLC member who owns 20 percent or more interest 
and each general partner or managing member, or (c) Stockholder or entity owning 20 percent or more voting stock.

6
Pandemic Response Accountability Committee | June 2025
FRAUD PREVENTION ALERT: Pre-Award Vetting Using Data Analytics Could Have 
Prevented Over $79B in Potentially Fraudulent Pandemic Relief Payments
The Amount of Potentially Fraudulent Payments Identified in this Fraud Prevention Alert 
Likely Would Have Been Higher than $79 Billion Had SBA Required PPP Applicants 
Provide DOB Information 
As noted previously, PPP applicants were not required to provide DOB information and 87 percent of our sampled 
PPP records did not include DOB information (i.e., 221,951 out of 256,000 PPP records). Of those 221,951 records 
with missing DOB information, approximately 95 percent were associated with a blank verification code, indicating 
that the SSN and name on the application matched SSA’s SSN and name records. However, for the sampled PPP 
records that included DOB information, the percentage of records associated with a blank verification code 
(meaning all three data indicators on the records—SSN, name, DOB—matched SSA’s data) dropped to 93 percent.
Had the SBA required PPP applicants provide DOB information in the PPP application, as COVID-19 EIDL and 
Pandemic UI programs did, we would expect to have identified even more potentially stolen or invalid SSNs, as well 
as higher potentially fraudulent payment amounts in our PPP estimates, because there would have been one more 
data point (i.e., DOB) to check against SSA’s records.  
Potential Synthetic IDs Used to Apply for Pandemic Programs 
A portion of the potentially stolen or invalid SSNs we uncovered could be linked with Synthetic Identity Fraud, which 
involves creating an artificial identity by combining real (often stolen) information, such as an SSN, with real or 
invented details, such as a made-up name and date of birth. This is different from traditional identity theft, in which 
fraudsters impersonate a victim using real personal information to apply for pandemic loans or to submit 
unemployment insurance claims. 
It is challenging to quantify how much of the $79 billion we estimated in potential fraud is associated with synthetic 
IDs, in part because SSA only returns verification codes for each processed record without the underlying name and 
DOB information in their system of records. Estimations of this type of criminal activity would require knowledge of 
both the incorrect record and the true data associated with an SSN. 
The PRAC’s Innovative Graph Analytics Approach to Find Common Connections Between 
Fraudsters
The PRAC will continue to conduct further analysis using SSA-verified results for lead identification efforts, including 
by using graph technology to find connections among entities that used shared attributes on applications, such as 
common SSNs, Tax IDs, phone numbers, and email addresses. 
Graph technology offers new, innovative methods of uncovering organized crime, fraud rings, and other complex 
schemes with a high level of accuracy through advanced contextual link analysis. Our graph analytics capabilities 
examine relations among entities and determine strength and direction of relationships between entities and their 
associated attributes. This enables the PRAC to generate actionable leads to share with our federal OIG and law 
enforcement partners. 
Ensuring Program Integrity and Preventing Payments to Fraudsters
In addition to our ongoing law enforcement activities, we urge benefit programs to continue to work with SSA to 
explore information-sharing agreements that will facilitate pre-award identity verifications across grant, loan, and 
benefit programs vulnerable to identity-based fraud. Program offices and oversight organizations need accurate and 

7
Pandemic Response Accountability Committee | June 2025
FRAUD PREVENTION ALERT: Pre-Award Vetting Using Data Analytics Could Have 
Prevented Over $79B in Potentially Fraudulent Pandemic Relief Payments
complete data to ensure that funds are being used as Congress intended. To assist identity verification, we urge 
program offices to require the collection of DOB information in benefit applications.
When program guardrails were removed during the pandemic, a substantial amount of funds were rapidly disbursed 
without proper identity verification. Implementing pre-award verification helps streamline the vetting process before 
disbursal, preventing fraudulent payments from going out and ensuring that funds are disbursed with additional 
program integrity controls.
Advantages of Implementing SSN Verification Agreement Early to Strengthen Program 
Integrity – Protecting Taxpayers and Honoring Congressional Intent
SSA is a recognized source to provide identity verification for a variety of federal programs. However, as noted in our 
January 2023 Fraud Alert, the process to implement new SSN verification agreements among agencies and address 
legal questions regarding the permissibility of information-sharing can be lengthy.14
Prior to the next natural disaster, health crisis, financial failure, or other emergency, it is vital for program 
administrators to establish SSN verification agreements in a timely manner so that an information exchange can 
be set up as a part of program integrity controls to prevent fraud, waste, abuse, and mismanagement and protect 
taxpayer funds from improper payments, as Congress intends in creating such emergency relief programs. 
Pandemic Program Background
Pandemic Unemployment Insurance Programs 
Reacting to the job losses caused by the pandemic, Congress created five new federally funded, temporary 
unemployment insurance programs, designed to broaden eligibility, increase benefit amounts, and extend the 
duration of benefits: 
•	 Pandemic Unemployment Assistance (PUA): provided benefits to individuals who were not eligible for regular 
unemployment compensation, such as self-employed workers, independent contractors, and gig workers. PUA 
offered up to a maximum of 79 weeks of benefits. 
•	 Federal Pandemic Unemployment Compensation (FPUC): provided an additional $600 per week to individuals 
receiving benefits from eligible unemployment compensation programs until July 31, 2020.15 Later, it was 
resumed and reduced to $300 per week.
•	 Pandemic Emergency Unemployment Compensation (PEUC): extended unemployment benefits for those who 
had exhausted their regular unemployment benefits. With all the legislative extensions, claimants could receive 
up to 53 weeks of PEUC payments. 
•	 Mixed Earner Unemployment Compensation (MEUC): offered an extra $100 per week on top of the weekly 
benefit amount and the $300 FPUC benefit to those who earned at least $5,000 in self-employment income in 
the most recent taxable year before applying for regular unemployment benefits.
14	
See PRAC Fraud Alert: PRAC Identifies $5.4 Billion in Potentially Fraudulent Pandemic Loans Obtained Using Over 69,000 Questionable Social Security 
Numbers, January 2023. According to SSA, it conducts over two billion SSN verifications annually. These exchanges include verifications specifically 
authorized under statutorily mandated programs (e.g., E-Verify) and consent-based processes (e.g., eCBSV). They also include disclosures made in 
accordance with the Privacy Act, Section 1106 of the Social Security Act, and SSA’s privacy regulations at 20 C.F.R. § 401.150, by which SSA will provide 
information to another federal agency when the information is necessary to determine eligibility in a health or income maintenance program which is 
compatible with SSA’s Social Security programs.
15	
UI programs eligible for FPUC were regular UI, Unemployment Compensation for Federal Employees, Unemployment Compensation for Ex-Servicemembers, 
Pandemic Emergency Unemployment Compensation, Pandemic Unemployment Assistance, among others.

8
Pandemic Response Accountability Committee | June 2025
FRAUD PREVENTION ALERT: Pre-Award Vetting Using Data Analytics Could Have 
Prevented Over $79B in Potentially Fraudulent Pandemic Relief Payments
• Lost Wages Assistance (LWA): was funded by Federal Emergency Management Agency (FEMA) to provide $300
per week to supplement unemployment benefits to eligible claimants.
During the UI pandemic period, March 27, 2020, through September 6, 2021, as defined by DOL, more than $888 
billion in total federal and state UI benefits were paid to applicants.16
The Pandemic UI programs were susceptible to fraud, waste, abuse, and mismanagement due to the unprecedented 
volume of claims and the use of antiquated state Information Technology (IT) systems, along with process- and 
technology-related challenges that state workforce agencies faced to quickly implement the new UI programs.17
The expanded benefits required states to implement major changes to their UI IT processing systems. Faced 
with challenges from IT modernization efforts, states experienced delays in implementing new UI programs and 
associated antifraud controls, which hampered their ability to quickly and effectively prevent fraud. 
DOL OIG issued an audit report in September 2022 highlighting improper payments including fraud in sampled and 
tested claims.18 The report found that many states struggled with verifying the identity of claimants, leading to 
improper and fraudulent payments.  
SBA’s COVID-19 Economic Injury Disaster Loan and Paycheck Protection Programs 
SBA’s COVID-19 EIDL and EIDL Advance programs provided funding to help small businesses recover from the 
economic impacts of the COVID-19 pandemic. SBA has provided over $378 billion in COVID-19 EIDL loans and 
COVID-19 EIDL advances (grants). The program closed to new applications on January 1, 2022. PPP provided SBA-
backed loans to businesses to keep their workforce employed during the COVID-19 crisis. The program ended on 
May 31, 2021, after more than 11 million PPP loans were approved totaling about $800 billion. 
COVID-19 EIDL and PPP were more susceptible to fraud due to the elevated urgency for agencies to provide timely 
relief to applicants in response to the COVID-19 pandemic. SBA’s initial approach to implement these programs 
quickly made billions of dollars available to millions of borrowers affected by the pandemic, but used few program 
controls to verify applicants’ eligibility prior to disbursing funds. 
In June 2023, the SBA OIG released a fraud landscape report offering a detailed estimate of potential fraud within 
the PPP and COVID-19 EIDL programs.19 This analysis highlighted how the weakening or removal of critical controls 
by the SBA enabled fraudsters to exploit these programs more easily.
For information about SBA’s COVID-19 EIDL and PPP, see the “Background” section in our January 2023 Fraud 
Alert.20
16	
See DOL OIG report, OIG Oversight of the Unemployment Insurance Program, December 15, 2023.
17	
See GAO report, Unemployment Insurance - Estimated Amount of Fraud during Pandemic Likely Between $100 Billion and $135 Billion, September 2023.
18	
See DOL OIG report, COVID-19: ETA and States did not Protect Pandemic-Related UI Funds from Improper Payments Including Fraud or from Payment 
Delays, September 30, 2022.
19	
See SBA OIG report, COVID-19 Pandemic EIDL and PPP Loan Fraud Landscape, June 27, 2023.
20	
See PRAC Fraud Alert: PRAC Identifies $5.4 Billion in Potentially Fraudulent Pandemic Loans Obtained Using Over 69,000 Questionable Social Security 
Numbers, January 2023.

9
Pandemic Response Accountability Committee | June 2025
FRAUD PREVENTION ALERT: Pre-Award Vetting Using Data Analytics Could Have 
Prevented Over $79B in Potentially Fraudulent Pandemic Relief Payments
About the PRAC 
The PRAC, created by the CARES Act in March 2020, supports and coordinates independent oversight of the $5 
trillion in coronavirus pandemic relief and response programs. Through in-depth data analysis across multiple federal 
programs, we identify potential fraud, waste, abuse, and mismanagement, collaborating across the government 
oversight community to identify cross-cutting issues and risks to hold wrongdoers accountable, preserve program 
integrity, and help prevent misuse of future funds. 
The PRAC has the processes, infrastructure, and relationships that enable us to conduct pre-award vetting to identify 
potential fraud before funds are disbursed.
In addition to SSN verification, the focus of this fraud prevention alert, the PRAC could also leverage existing 
investigative data on over 100,000 known fraud schemes from across a variety of agencies, including names, bank 
accounts, dates of birth, internet protocol and email addresses, and other identifiers, to flag new attempts by bad 
actors to steal American taxpayer dollars. The PRAC could also use its unique graph analytics tools to identify hidden 
networks and patterns among claims data. This pre-award vetting data could be retained and later be used for 
future pre-award vetting to serve as an early warning system for emerging fraud threats.
PRAC’s Commitment to Quality and Integrity 
We conducted this review using agile oversight principles that require we adhere to the professional standards 
of independence, due professional care, and quality assurance and follow procedures to ensure accuracy of the 
information presented. 
The PRAC relied on SSA’s verification results. To maintain the security of SSA’s SSN verification process, SSA does 
not release the agency’s matching logic concerning how it verifies SSNs. We believe the evidence obtained provides 
a reasonable basis for the results of our analysis. 
For more on protecting program integrity, read our Chapter 1 of the Blueprint for Enhanced Program Integrity.21
For more on preventing payments to fraudsters, read our Chapter 3 of the Blueprint for Enhanced Program 
Integrity.22 
21	
See PRAC Blueprint for Enhanced Program Integrity Chapter 1: Best Practices for Strengthening Federal Programs.
22	
See PRAC Blueprint for Enhanced Program Integrity Chapter 3: Fraud Prevention and Detection.

For more information:
Lisa Reijula
Associate Director of Outreach and Engagement, PRAC
Lisa.Reijula@cigie.gov
Visit us at:
PandemicOversight.gov
Follow us at:
Report Fraud, Waste, Abuse, or Misconduct:
To report allegations of fraud, waste, abuse, or misconduct regarding 
pandemic relief funds or programs please go to the PRAC website at
PandemicOversight.gov.
Pandemic Response 
Accountability Committee logo. A Committee of the Council of the Inspectors General 
on Integrity and Efficiency

File and source

File

2023-79b-preventable-fraud-alert.pdf

Size

417,752 bytes

SHA-256

507104ef9e12f1ab104cbd74bd33e08a987e3914202fc58e5f376c21d7b8960c

Our copy

2023-79b-preventable-fraud-alert.pdf

Original

www.pandemicoversight.gov