Pandemic Darlings The pandemic economy, in original documents
Home Court filings GAO-21-589 — Economic Injury Disaster Loan Program: Additional Actions Needed to Improv…

Court filing

GAO-21-589 — Economic Injury Disaster Loan Program: Additional Actions Needed to Improve Communication with Applicants and Address Fraud Risks (July 2021)

Record facts

CourtU.S. Government Accountability Office
Filed2021-07-01

Summary

A report to congressional addressees by the United States Government Accountability Office, GAO-21-589, issued July 2021, on communication with applicants and fraud risks in the Economic Injury Disaster Loan program. It reports that between March 2020 and February 2021 SBA provided about 3.8 million EIDL loans and 5.8 million grants totaling $224 billion. It finds applicants cited a lack of information and uncertainty about application status, and that SBA data showed 5.3 million duplicate applications. On fraud risk it states that SBA had approved at least 3,000 loans totaling about $156 million to businesses its policies made ineligible as of September 30, 2020, and that more than 20,000 suspicious activity reports were filed between May and October 2020. GAO recommends a comprehensive communications strategy for disaster response programs and states SBA agreed.

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

ECONOMIC INJURY 
DISASTER LOAN 
PROGRAM 
Additional Actions 
Needed to Improve 
Communication with 
Applicants and 
Address Fraud Risks 
 
 
 
Report to Congressional Addressees 
July 2021 
 
GAO-21-589 
 
 
United States Government Accountability Office 

 
 
 United States Government Accountability Office 
  
Highlights of GAO-21-589, a report to 
congressional addressees 
 
July 2021 
ECONOMIC INJURY DISASTER LOAN 
PROGRAM 
Additional Actions Needed to Improve 
Communication with Applicants and Address Fraud 
Risks  
What GAO Found 
Economic Injury Disaster Loan (EIDL) applicants and recipients varied in terms of 
business size, years in operation, and industry, based on GAO’s analysis of 
Small Business Administration (SBA) data from March 2020 through February 
2021: 
• 
Business size. A majority of EIDL applicants (about 81 percent) and 
EIDL recipients (about 86 percent) were smaller businesses (10 or fewer 
employees). 
• 
Years in operation. A majority of EIDL applicants (about 63 percent) 
had been in operation for less than 5 years. However, businesses in 
operation for more than 5 years received the majority of total EIDL loan 
dollars and had higher approval rates compared to newer businesses.  
• 
Industry. Businesses in the personal services and transportation 
industries made up the largest share of applicants, while those in the 
legal services and lodging industries were approved for loans at the 
highest rates (see figure).  
 
Top Loan Applicants and Approval Rates by Business Industry 
 
 
 
In addition, small businesses in counties with higher median household income, 
better internet access, and more diverse populations generally received more 
loans per 1,000 businesses and larger loans.  
 
EIDL applicants have faced a number of challenges, according to applicants and 
other business stakeholders GAO interviewed between August 2020 and 
February 2021. For example, applicants from five discussion groups and several 
stakeholders cited lack of information and uncertainty about application status as 
major concerns. In addition, until February 2021, SBA did not provide important 
information to potential applicants, such as limits on loan amounts and definitions 
of certain program terms. Lack of important program information and application 
status put pressure on SBA’s resources and negatively affected applicants’ 
experience. For example, SBA’s customer service line experienced call surges 
that resulted in long wait times, and SBA’s data showed that 5.3 million 
applications were duplicates. SBA’s planning documents describe in general 
View GAO-21-589. For more information, 
contact William B. Shear at (202) 512-8678 or 
ShearW@gao.gov. 
Why GAO Did This Study 
Between March 2020 and February 
2021, SBA provided about 3.8 million 
low-interest EIDL loans and 5.8 million 
grants (called advances) totaling $224 
billion to help small businesses 
adversely affected by COVID-19. 
Borrowers can use these low-interest 
loans and advances to pay for 
operating and other expenses.   
The CARES Act includes a provision 
for GAO to monitor funds provided for 
the COVID-19 pandemic. This report 
examines, among other objectives, the 
characteristics of program applicants 
and recipients; the challenges EIDL 
applicants experienced and the extent 
to which SBA has addressed them; 
and the steps SBA has taken to 
address risks of fraud and provision of 
funds to ineligible applicants. 
GAO reviewed documents from SBA, 
an EIDL contractor, and two of its 
subcontractors. In addition, GAO 
analyzed loan application data, 
conducted five discussion groups with 
applicants, and interviewed staff from 
SBA, six Small Business Development 
Centers, and six business 
associations. GAO also analyzed 
socioeconomic, demographic, and 
geographic data on EIDL program 
participants.  
What GAO Recommends 
GAO recommends that SBA develop a 
comprehensive communications 
strategy that includes guidelines for the 
type and timing of information to be 
provided to potential and actual 
applicants of its disaster response 
programs. SBA agreed with the 
recommendation. 

 
 
terms the public outreach to be conducted following disasters, but they do not 
detail the type or timing of the information to be provided. Developing and 
implementing a comprehensive communication strategy that includes these 
details could improve the quality, clarity, and timeliness of information SBA 
provides to its applicants and resource partners following catastrophic disasters. 
 
GAO’s ongoing review of the EIDL program related to COVID-19 has found that 
the program is susceptible to providing funding to ineligible and fraudulent 
applicants. For example, as GAO reported in January 2021, SBA had approved 
at least 3,000 loans totaling about $156 million to businesses that SBA policies 
state were ineligible for the EIDL program, such as real estate developers and 
multilevel marketers, as of September 30, 2020. In addition, GAO found that 
between May and October 2020, over 900 U.S. financial institutions filed more 
than 20,000 suspicious activity reports related to the EIDL program with the 
Financial Crimes Enforcement Network. Further, GAO’s analysis of 51 
Department of Justice cases involving fraud charges for EIDL loans as of March 
2021 found that these cases involved identity theft, false attestation, fictitious or 
inflated employee counts, and misuse of proceeds. 
 
Over the course of its COVID-19 response, SBA has made some changes to 
address these risks. For example, beginning in June 2020, SBA took actions to 
improve loan officers’ ability to withhold funding for applicants suspected of fraud. 
However, SBA has not yet implemented recommendations GAO has previously 
made to address EIDL program risks.  
• 
In January 2021, GAO recommended that SBA conduct data analytics 
across the EIDL portfolio to detect potentially ineligible and fraudulent 
applications (GAO-21-265). SBA did not agree or disagree with this 
recommendation. However, in May 2021, SBA officials stated the agency 
was in the process of developing analysis to apply certain fraud 
indicators to all application data.   
• 
In March 2021, GAO recommended that SBA (1) implement a 
comprehensive oversight plan to identify and respond to risks in the EIDL 
program, (2) conduct and document a fraud risk assessment, and (3)  
develop a strategy to address the program’s assessed fraud risks on a 
continuous basis (GAO-21-387). SBA agreed with all three 
recommendations. In May 2021, SBA officials stated that the agency had 
started to assess fraud risk for the program.  
Fully implementing these recommendations would help SBA to safeguard billions 
of dollars of taxpayer funds and improve the operation of the EIDL program.  
  
 
 

 
 
 
 
 
 
 
 
 
 
 
Page i 
GAO-21-589  Economic Injury Disaster Loan Program 
Letter 
 
1 
Background 
4 
SBA Took Steps to Address Processing Challenges but Lacks a 
Comprehensive Communications Strategy 
13 
Program Participants Generally Found the EIDL Program Helpful 
26 
SBA Has Begun to Address Eligibility and Fraud Concerns but 
Has Not Fully Implemented GAO Recommendations 
28 
Smaller and Older Businesses Received Larger Loans and Were 
Approved at Higher Rates, and Wealthier Communities 
Received More Funding 
41 
Conclusions 
60 
Recommendation for Executive Action 
60 
Agency Comments and Our Evaluation 
61 
Appendix I 
Objectives, Scope, and Methodology 
64 
 
Appendix II 
Socioeconomic and Demographic Characteristics of EIDL Applicants  
and Recipients 
72 
 
Appendix III 
Analysis of EIDL Lending Outcomes for Selected Metropolitan  
Statistical Areas by ZIP Code 
82 
 
Appendix IV 
Comments from the Small Business Administration 
87 
 
Appendix V 
GAO Contact and Staff Acknowledgments 
91 
 
Tables 
Table 1: Key Legislation Affecting the Economic Injury Disaster 
Loan (EIDL) Program in Response to COVID-19 
5 
Table 2: Economic Injury Disaster Loan Program Eligibility Criteria 
and Validation 
35 
Table 3: Economic Injury Disaster Loan Approval Rates by 
Business Size, March 2020–February 2021 
42 
Contents 

 
 
 
 
 
 
 
 
 
 
 
Page ii 
GAO-21-589  Economic Injury Disaster Loan Program 
Table 4: Average Economic Injury Disaster Loan (EIDL) Amount 
and Percentage of Total Loan Dollars, by Business Size, 
March 2020–February 2021 
42 
Table 5: Economic Injury Disaster Loan (EIDL) Approval Rate, 
Average Loan Amount, and Percentage of Total Loan 
Dollars, by Age of Business, March 2020–February 2021 
43 
Table 6: Top 10 Industries by Number of Economic Injury Disaster 
Loan (EIDL) Applicants, March 2020–February 2021 
44 
Table 7: Top 10 Industries by Economic Injury Disaster Loan 
Approval Rate, March 2020–February 2021 
44 
Table 8: Top 10 Industries by Economic Injury Disaster Loan 
Average Loan Amount, March 2020–February 2021 
45 
Table 9: Economic Injury Disaster Loan (EIDL) Applications, Loan 
Size, and Share of Total Loan Dollars, by County 
Unemployment Rate 
72 
Table 10: Economic Injury Disaster Loan (EIDL) Applications, 
Loan Size, and Share of Total Loan Dollars, by County 
Share of Households Receiving Public Assistance Income 
74 
Table 11: Economic Injury Disaster Loan (EIDL) Applications, 
Loan Size, and Share of Total Loan Dollars, by County 
Poverty Rate 
76 
Table 12: Economic Injury Disaster Loan (EIDL) Applications, 
Loan Size, and Share of Total Loan Dollars, by County 
Foreign-Born Population 
78 
Table 13: Economic Injury Disaster Loan (EIDL) Applications, 
Loan Size, and Share of Total Loan Dollars, by County 
Households with Limited English Proficiency 
80 
 
Figures 
Figure 1: Timeline of the Economic Injury Disaster Loan Program 
in Response to COVID-19 
9 
Figure 2: SBA Processing of Economic Injury Disaster Loans and 
Advances 
11 
Figure 3: Weekly SBA Processing and Disbursement Center 
Staffing Levels, March 20, 2020–February 26, 2021 
15 
Figure 4: Economic Injury Disaster Loans (EIDL) Cumulative 
Application Volume and Percentage of Applications 
Processed, March 15, 2020–February 27, 2021 
16 
Figure 5: Number of Economic Injury Disaster Loan Applications 
and Average Processing Times, by Month Application 
Accepted, March 2020–February 2021 
17 

 
 
 
 
 
 
 
 
 
 
 
Page iii 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 6: SBA Customer Service Line Average Wait Times and 
Call Volumes, Week of March 15, 2020, through Week of 
February 14, 2021 
19 
Figure 7: SBA Customer Service Center Staffing Levels, March 
20, 2020–February 26, 2021 
20 
Figure 8: Department of Justice Fraud Charges Related to the 
Economic Injury Disaster Loan Program, as of March 
2021 
30 
Figure 9: Economic Injury Disaster Loan (EIDL) Lending 
Outcomes by Median Household Income 
47 
Figure 10: Economic Injury Disaster Loan (EIDL) Lending 
Outcomes by Household Internet Access 
49 
Figure 11: Economic Injury Disaster Loan Program Lending 
Outcomes by County Minority Population 
51 
Figure 12: Economic Injury Disaster Loan Program Lending 
Outcomes for Urban and Rural Counties 
53 
Figure 13: Economic Injury Disaster Loan Program Lending 
Outcomes for Tribal Lands 
55 
Figure 14: Approved Economic Injury Disaster Loans (EIDL) Per 
1,000 Small Businesses by County, March 2020–
February 2021 
57 
Figure 15: Economic Injury Disaster Loan (EIDL) Amount per 
Small Business Employee by County, March 2020–
February 2021 
58 
Figure 16: Approval Rate for Economic Injury Disaster Loans 
(EIDL) by County, March 2020–February 2021 
59 
Figure 17: Economic Injury Disaster Loan Program Lending 
Outcomes by County Unemployment Rate 
73 
Figure 18: Economic Injury Disaster Loan Program Lending 
Outcomes by County Share of Households Receiving 
Public Assistance Income 
75 
Figure 19: Economic Injury Disaster Loan Program Lending 
Outcomes by County Poverty Rate 
77 
Figure 20: Economic Injury Disaster Loan Program Lending 
Outcomes by Foreign-Born Population Percentage 
79 
Figure 21: Economic Injury Disaster Loan Program Lending 
Outcomes by County Share of Households with Limited 
English Proficiency 
81 
Figure 22: Approval Rates for Economic Injury Disaster Loans in 
the New York-Newark-Jersey City Metropolitan Statistical 
Area for Selected Indicators, March 2020–February 2021 
83 

 
 
 
 
 
 
 
 
 
 
 
Page iv 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 23: Approval Rates for Economic Injury Disaster Loans in 
the Chicago-Naperville-Elgin Metropolitan Statistical Area 
for Selected Indicators, March 2020–February 2021 
84 
Figure 24: Approval Rates for Economic Injury Disaster Loans in 
the San Jose-Sunnyvale-Santa Clara Metropolitan 
Statistical Area for Selected Indicators, March 2020–
February 2021 
85 
Figure 25: Approval Rates for Economic Injury Disaster Loans in 
the New Orleans-Metairie Metropolitan Statistical Area for 
Selected Indicators, March 2020–February 2021 
86 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 
 
 
Page v 
GAO-21-589  Economic Injury Disaster Loan Program 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Abbreviations  
 
COVID-19 
 
Coronavirus Disease 2019 
DOJ 
 
 
Department of Justice 
EIDL  
 
Economic Injury Disaster Loan 
FAQ   
 
frequently asked questions   
Federal Reserve 
Board of Governors of the Federal Reserve System 
FinCEN 
 
Financial Crimes Enforcement Network 
MSA 
 
 
metropolitan statistical areas 
OIG  
 
 
Office of Inspector General   
SBA   
 
Small Business Administration   
SBDC  
 
Small Business Development Center 
USDA  
 
U.S. Department of Agriculture 
ZCTA  
 
ZIP code tabulation area 
 
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 1 
GAO-21-589  Economic Injury Disaster Loan Program 
441 G St. N.W. 
Washington, DC 20548 
July 30, 2021 
Congressional Addressees 
Coronavirus Disease 2019 (COVID-19) has adversely affected most of 
the more than 30 million small businesses in the United States through 
decreased revenue or business closures. To assist small businesses and 
nonprofits affected by the pandemic, Congress appropriated funding and 
eased borrowing requirements for the Economic Injury Disaster Loan 
(EIDL) program. The EIDL program, which is administered and 
implemented by the Small Business Administration (SBA), provides low-
interest loans to help borrowers meet obligations or pay ordinary and 
necessary operating expenses. In addition, Congress appropriated 
funding to create grants—known as advances—for EIDL applicants; 
these advances do not have to be repaid. The program provided about 
$230 billion in loans and advances to small businesses and nonprofits 
between March 2020 and May 2021.  
We were asked to review how SBA administered the EIDL program in 
response to the COVID-19 pandemic. In addition, the CARES Act 
includes a provision for us to monitor and oversee the use of funds made 
available to prepare for, respond to, and recover from the COVID-19 
pandemic.1 For this report, we examined (1) challenges SBA and EIDL 
applicants experienced as part of SBA’s implementation of the EIDL 
program in response to COVID-19 and the extent to which SBA has 
addressed these challenges; (2) the EIDL program’s effects on its 
participants, including the effects of loan limits SBA put in place; (3) steps 
SBA has taken to address risks of fraud and provision of funds to 
ineligible applicants; and (4) the characteristics of program applicants and 
recipients. The EIDL loan program existed prior to the pandemic, but 
Congress has made legislative changes to modify the program in 
response to COVID-19, including the addition of EIDL advances. This 
report focuses on SBA’s implementation of EIDL since the start of the 
pandemic.2 
                                                                                                                       
1Pub. L. No. 116-136, § 19010, 134 Stat. 281, 579-80 (2020).  
2We regularly issue government-wide reports on the federal response to COVID-19. For 
the latest report, see 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). Our next government-wide report will be issued in 
October 2021 and will be available on GAO’s website at https://www.gao.gov/coronavirus. 
Letter 
 

 
 
 
 
 
 
Page 2 
GAO-21-589  Economic Injury Disaster Loan Program 
To address the first objective, we reviewed relevant legislation and SBA’s 
guidance, policies, procedures, contracts, and other material for the EIDL 
program. In addition, we reviewed SBA data on EIDL program processing 
times, staffing, and Customer Service Center call wait times. To assess 
the reliability of these data, we interviewed SBA officials. We determined 
these data were reliable for the purpose of reporting on SBA’s loan 
processing volumes and times, staffing levels, and Customer Service 
Center call wait times and call volumes. We also interviewed officials from 
SBA’s Office of Disaster Assistance. 
In addition, we interviewed representatives from a nongeneralizable 
sample of six state or regional Small Business Development Centers 
(SBDC), the national association representing SBDCs, two associations 
representing small businesses nationwide, and four industry-specific 
business associations.3 Collectively, we refer to the SBDCs and 
associations as stakeholders. We selected the SBDCs based on a few 
factors, including that the Census Bureau’s Small Business Pulse Survey 
showed that the state or region in which they are located experienced a 
large negative impact from the pandemic and had a high share of small 
businesses receiving EIDL funding.4 We chose industry-specific 
associations that reflect industries that were substantially affected by the 
pandemic and had large numbers of EIDL borrowers, based on our 
review of the Census Bureau’s Small Business Pulse Survey and SBA 
loan data. 
We also held five discussion groups with EIDL program applicants to 
learn about their experience with the program—two groups of applicants 
that were declined for loans and three groups that were approved. Each 
group consisted of four to six participants, and we spoke with 22 EIDL 
applicants in total. Discussion groups are intended to generate in-depth 
information about the reasons for participants’ views on specific topics. 
While we report our findings by the number of discussion groups in which 
a topic was discussed, this does not necessarily mean that there was a 
consensus or agreement among all discussion group participants on a 
                                                                                                                       
3SBA partially funds SBDCs, which offer counseling, training, and technical assistance to 
current and prospective small businesses. 
4The Small Business Pulse Survey measures the effect of changing business conditions 
during the pandemic on small businesses, including whether small businesses received 
EIDL funding.  

 
 
 
 
 
 
Page 3 
GAO-21-589  Economic Injury Disaster Loan Program 
given topic. The opinions expressed by the participants represent their 
points of view and may not represent the views of all EIDL applicants. 
To address the second objective, we obtained information from SBA 
about the number of borrowers affected by the loan limits SBA put in 
place. We interviewed SBA representatives and the stakeholders noted 
above about the impact of these limits and the impact of the program. We 
also asked participants in the discussion groups described above about 
these topics. 
To address the third objective, we reviewed prior GAO reports and 
reports from SBA’s Office of Inspector General and SBA’s independent 
financial statement auditor. We interviewed SBA officials about their 
controls for ensuring eligibility and addressing fraud risk. We also 
obtained written responses and other documentation from SBA’s EIDL 
contractor and subcontractors about the services they provided and their 
interactions with SBA. To characterize fraud cases and schemes used to 
obtain EIDL funds, we analyzed 51 fraud-related cases based on 
information provided in Department of Justice (DOJ) and federal court 
documents from May 2020 to March 2021. 
To address our fourth objective, we analyzed SBA data on loan 
applications submitted between March 14, 2020, and February 28, 2021. 
We combined these data with county-level data from the Census Bureau 
and Department of Agriculture to examine the socioeconomic, 
demographic, and geographic characteristics of the counties in which 
approved and denied applicants were located. For four metropolitan 
statistical areas, we also combined SBA’s EIDL applicant data with 
Census Bureau data at the ZIP code tabulation area level to examine the 
socioeconomic and demographic characteristics of ZIP code tabulation 
areas in which applicants and borrowers were located.5 We reviewed 
documentation related to the data used and determined that the data 
were reliable for describing the characteristics of the communities in 
which EIDL applicants were located. For more information about our 
objectives, scope, and methodology, see appendix I. 
                                                                                                                       
5The Census Bureau defines metropolitan statistical areas as having at least one 
urbanized area of 50,000 or more population, plus adjacent territory that has a high 
degree of social and economic integration with the core as measured by commuting ties. 
ZIP code tabulation areas are generalized areal representations of U.S. Postal Service 
ZIP code service areas, which are not areal features but a collection of mail delivery 
routes.  

 
 
 
 
 
 
Page 4 
GAO-21-589  Economic Injury Disaster Loan Program 
We conducted this performance audit from July 2020 to July 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 that 
the evidence obtained provides a reasonable basis for our findings and 
conclusions based on our audit objectives. 
 
 
The Small Business Act authorizes SBA to make EIDL loans to eligible 
small businesses and nonprofit organizations located in a disaster area. 
In response to COVID-19, Congress provided additional funding for the 
existing EIDL program to support small businesses through several 
legislative actions and temporarily changed certain EIDL program 
application requirements that were in place under the Small Business Act 
(see table 1). On March 6, 2020, Congress deemed COVID-19 a disaster 
under the Coronavirus Preparedness and Response Supplemental 
Appropriations Act of 2020, making financial obligations that could not be 
met as a result of COVID-19 an eligible expense for the EIDL program. At 
that time, SBA had about $1.1 billion in loan credit subsidy available to 
support about $7–$8 billion in disaster loans.6 On April 16, 2020, SBA 
announced that this funding had been exhausted. On April 24, 2020, 
Congress provided $50 billion in loan credit subsidy, which would support 
about $470 billion in EIDL loans. 
 
 
                                                                                                                       
6Loan credit subsidy covers the government’s cost of extending or guaranteeing credit 
and takes into consideration the estimated cash flows to and from the government. Loan 
credit subsidy represents the estimated long-term cost of providing loans and takes into 
account expected future loan performance, including loan repayments, prepayments, 
defaults, recoveries, and the timing of these events. The loan credit subsidy cost was 
13.62 percent and 8.92 percent for fiscal years 2020 and 2021, respectively. For fiscal 
year 2020, this means that the estimated cost was $13.62 per $100 of disaster loans 
provided. 
Background 
Legislative Changes to 
EIDL in Response to 
COVID-19 

 
 
 
 
 
 
Page 5 
GAO-21-589  Economic Injury Disaster Loan Program 
Table 1: Key Legislation Affecting the Economic Injury Disaster Loan (EIDL) Program in Response to COVID-19 
Legislation and date enacted  
Appropriation 
amount 
Key provisions and changes 
Coronavirus Preparedness and 
Response Supplemental 
Appropriations Act of 2020 
March 6, 2020 
None 
• 
Deemed COVID-19 a disaster under the Small Business Act, making 
economic injury caused by COVID-19 eligible for EIDL loans  
CARES Act 
March 27, 2020 
 
$10 billion for EIDL 
advances 
• 
Authorized the Small Business Administration (SBA) to provide 
eligible applicants with EIDL advances of up to $10,000 that do not 
need to be repaid 
• 
Removed the requirement that applicants must not be able to obtain 
credit elsewhere 
• 
Expanded eligibility for the EIDL program 
• 
Restricted SBA from obtaining federal tax transcripts as part of the 
EIDL application process 
Paycheck Protection Program and 
Health Care Enhancement Act 
April 24, 2020  
$10 billion for EIDL 
advances and $50 
billion for loan 
subsidies  
• 
Expanded EIDL eligibility to agricultural enterprises previously 
ineligible  
Consolidated Appropriations Act, 
2021 
December 27, 2020  
$20 billion for 
targeted EIDL 
advances 
• 
Removed restriction that SBA cannot obtain federal tax transcripts as 
part of the EIDL application process 
• 
Extended deadline to apply for EIDL loans and advances by 1 year, 
from December 31, 2020, to December 31, 2021 
• 
Provided targeted advances up to the full amount of $10,000 to 
certain eligible businesses and nonprofits that have not more than 
300 employees, are located in low-income areas, and have 
experienced greater than 30 percent loss in income 
American Rescue Plan Act of 
2021 
March 11, 2021 
$10 billion for 
targeted EIDL 
advances and $5 
billion for 
supplemental 
targeted EIDL 
advances 
• 
Provided additional funding for targeted advances of up to $10,000 for 
the same entities eligible under the Consolidated Appropriations Act, 
2021 
• 
Provided an additional $5,000 in supplemental targeted advances for 
eligible entities in low-income communities that suffered economic 
loss of greater than 50 percent and employed not more than 10 
employees 
Source: GAO analysis of legislation. | GAO-21-589 
 
In addition to funding for loans, Congress provided funding for EIDL 
advances—grants that do not need to be repaid—which were previously 
not an element of the EIDL program. In March and April of 2020, 
Congress authorized SBA to provide $20 billion in advances of up to 
$10,000 to EIDL applicants, even if they did not qualify for a loan. In 
implementing this provision, SBA limited the advances to $1,000 per 
employee up to a total of $10,000. Funding for the advances was 
depleted in July 2020. In December 2020 and March 2021, Congress 
provided a total of $30 billion for targeted EIDL advances and $5 billion 

 
 
 
 
 
 
Page 6 
GAO-21-589  Economic Injury Disaster Loan Program 
for supplemental targeted EIDL advances available for businesses that 
meet specific criteria, including being located in low-income communities 
and meeting certain thresholds for economic loss and number of 
employees.7 
Small businesses eligible to borrow from the EIDL program in response to 
COVID-19 include small agricultural cooperatives, Employee Stock 
Ownership Plans, tribal concerns, sole proprietorships, independent 
contractors, and agricultural enterprises.8 Eligible small businesses 
consist of those that have no more than 500 employees or are defined as 
small according to SBA size standards.9 Most private nonprofit 
organizations are also eligible. Loan, advance, and targeted or 
supplemental targeted advance recipients can use the funds to cover 
expenses such as payroll, rent, utilities, and fixed debt payments, among 
other normal operating expenses. 
                                                                                                                       
7A low-income community is defined as a census tract where the poverty rate is at least 
20 percent, or in the case of a tract not located within a metropolitan area, where the 
median family income does not exceed 80 percent of the statewide median family income. 
For census tracts in metropolitan areas, the median family income for the tract must not 
exceed 80 percent or greater of the statewide median family income or the metropolitan 
area median family income. For areas not within census tracts, equivalent county divisions 
will be used for the purpose of determining poverty rates and median family income. 
Additionally, a population census tract with population of less than 2,000 shall be treated 
as a low-income community if the tract is within an area of high poverty and 
unemployment designated under 26 U.S.C. § 1391, and is contiguous to one or more low-
income communities. For a census tract within high migration rural counties to be defined 
as a low-income community, the median family income must not exceed 85 percent of the 
statewide median family income. Economic loss is defined as the amount by which the 
gross receipts of the covered entity declined during an 8-week period between March 2, 
2020, and December 31, 2021, relative to a comparable 8-week period immediately 
preceding March 2, 2020, or during 2019. For seasonal businesses, SBA shall determine 
the economic loss as appropriate. 
8Most agricultural enterprises were not eligible for targeted EIDL advances or 
supplemental targeted EIDL advances.  
9Applicants to the targeted EIDL advances must have not more than 300 employees, and 
applicants for the supplemental targeted EIDL advances must have not more than 10 
employees. SBA’s Table of Size Standards matches industries to the North American 
Industry Classification System to specify the largest a business can be and still qualify as 
a small business. Based on SBA’s size standards, qualifying as a small business can be 
based on the number of employees or gross receipts. However, the CARES Act made all 
small businesses with 500 or fewer employees eligible for EIDL. 
EIDL Program Terms in 
Response to COVID-19 

 
 
 
 
 
 
Page 7 
GAO-21-589  Economic Injury Disaster Loan Program 
For EIDL loans, recipients have a repayment period of up to 30 years, 
and SBA deferred repayment for all EIDL loans until 2022.10 For EIDL 
loans in response to COVID-19, SBA set the interest rate at 3.75 percent 
for businesses and 2.75 percent for nonprofits. Interest accrues on the 
outstanding balance of the loan during deferment. 
The loan amount SBA offers to the EIDL applicant depends on the 
amount of the applicant’s economic injury, as well as on loan limits that 
SBA put in place.11 Economic injury is the change in the applicant’s 
financial condition attributable to the effect of a disaster, resulting in the 
inability of the applicant to meet its obligations or to pay ordinary and 
necessary operating expenses. SBA calculated the economic injury using 
gross revenue and cost of goods sold between January 31, 2019, and 
January 31, 2020, two pieces of information requested on the EIDL 
application. SBA initially calculated economic injury related to COVID-19 
by presuming 6 months of lost gross profit. For example, a business with 
$120,000 in annual revenue and $60,000 in cost of goods sold has 
$60,000 in annual gross profit, or $5,000 per month.12 The economic 
injury for this business would be $5,000 multiplied by six, or $30,000. 
From March 16, 2020 through May 3, 2020, SBA limited the maximum 
loan amount to $500,000, even if the calculated economic injury 
exceeded that amount. From May 4, 2020, through April 5, 2021, SBA 
limited the maximum loan amount to $150,000. SBA previously told us 
that this decision was made in order to provide more loans to small 
businesses.13 Beginning on April 6, 2021, SBA increased the loan limit 
back to $500,000, revised its economic injury calculation to presume 24 
                                                                                                                       
10For loans made in calendar year 2020, the first payment due date was extended from  
12 months to 24 months from the date the loan was executed. For loans made in 2021, 
the first payment due date was extended from 12 months to 18 months from the date the 
loan was executed.  
11The statutory maximum for an EIDL loan is $2 million, but SBA can set a lower 
maximum amount.  
12In February 2021, SBA defined the cost of goods sold as the direct costs of producing 
the goods sold by a company, including the cost of materials and labor directly used to 
create the goods. It excludes indirect expenses, such as sales force and distribution costs. 
SBA used different information to determine economic injury for agricultural enterprises 
and nonprofits. SBA asked these applicants to provide operating expenses for the 12 
months prior to the disaster. To calculate 6 months of economic injury, SBA divided the 
12-month operating expenses by two.  
13See GAO, COVID-19: Opportunities to Improve Federal Response and Recovery 
Efforts, GAO-20-625 (Washington, D.C.: June 25, 2020). 

 
 
 
 
 
 
Page 8 
GAO-21-589  Economic Injury Disaster Loan Program 
months of lost gross profit, and began to accept requests for loan 
increases. Using the same example of a business with a gross profit of 
$5,000 per month, the calculated economic injury under the revised 
economic injury calculation would be $120,000 ($5,000 multiplied by 24) 
rather than $30,000. After SBA makes a loan offer, the applicant can 
select an amount that is lower than the offered amount.14 
SBA’s Office of Disaster Assistance is responsible for administering the 
EIDL program in response to COVID-19, primarily through the following 
offices: 
• 
The Office of Disaster Assistance Headquarters creates policies, 
procedures, and guidelines for all office operations and coordinates 
with Congress. 
• 
The Customer Service Center is a single nationwide point of contact 
for SBA disaster loans. It provides a call center, email response, and 
disaster application mailings. 
• 
Field Operations Centers coordinate disaster field operations and 
publicize the Office of Disaster Assistance’s disaster lending 
programs before and after disasters. 
• 
The Processing and Disbursement Center screens all applications, 
reviews and processes those that are complete for approval or 
declination, closes all approved loans, and disburses loan proceeds. 
 
To implement the EIDL program in response to COVID-19, SBA took 
several key steps, including declaring businesses in specific areas 
eligible, implementing a new system for processing applications, and 
creating a new application website (see fig. 1). At times, SBA temporarily 
stopped receiving and processing applications because of program 
changes, legislative changes, and exhaustion of funding. 
                                                                                                                       
14SBA stated that if a borrower accepts a loan for less than the full amount offered, the 
borrower would have up to 2 years after the date of the loan was executed to request 
additional funds.  
SBA’s Office of Disaster 
Assistance 
SBA’s Implementation of 
the EIDL Program in 
Response to COVID-19 

 
 
 
 
 
 
Page 9 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 1: Timeline of the Economic Injury Disaster Loan Program in Response to COVID-19 
 
 
SBA declared disaster areas eligible for the EIDL program. After 
Congress deemed COVID-19 a disaster, SBA began to declare states 
and territories eligible for EIDL loans beginning on March 16, 2020.15 The 
declarations allowed SBA to begin using about $1.1 billion of its existing 
disaster loan credit subsidy to make EIDL loans. By March 21, 2020, SBA 
had declared all states, U.S. territories, and the District of Columbia to be 
disaster areas eligible for EIDL funding. 
SBA implemented a new system for processing COVID-19 EIDL 
applications. SBA had an existing disaster loan processing system in 
                                                                                                                       
15SBA changed its requirement that a state or territory provide documentation certifying 
that at least five small businesses have suffered substantial economic injury as a result of 
the disaster, with at least one business located in each declared county or parish. Under a 
new criterion, states and territories are required to certify that at least five small 
businesses within the state or territory have suffered economic injury, regardless of where 
businesses are located. 

 
 
 
 
 
 
Page 10 
GAO-21-589  Economic Injury Disaster Loan Program 
place at the start of the pandemic. However, that system did not have the 
capacity to handle the number of EIDL applications SBA received in 
response to COVID-19. SBA turned to an existing contractor, RER 
Solutions, for a system to provide automated initial recommendations to 
approve or decline EIDL applications and flag applications with issues for 
further review by SBA.16 RER Solutions had an existing subcontractor—
Rocket Loans—that supported SBA’s existing systems used to provide 
disaster loans to individuals for home and personal property repairs. To 
process EIDL loans, Rocket Loans added a second-tier contractor—
Rapid Finance—to use Rapid Finance’s underwriting system to perform 
data validation and conduct automated checks for potential fraud and 
eligibility as part of the application review process.17 We describe these 
checks further below. 
SBA created new EIDL application forms and a new loan application 
website. In late March 2020, SBA released a streamlined loan 
application. The streamlined application did not require applicants to 
provide documents previously required for EIDL loan applications, 
including profit and loss statements and a copy of the most recent federal 
income tax return. Working with Rapid Finance, SBA also implemented a 
new application website that included an option for loan applicants to 
apply for advances. 
EIDL temporarily stopped accepting applications as a result of 
exhaustion of funding and legislative changes. On April 15, 2020, 
SBA stopped accepting new applications and announced the following 
day that the $10 billion in funding for advances and the $1.1 billion in 
disaster loan credit subsidy had been exhausted. After the passage of the 
Paycheck Protection Program and Health Care Enhancement Act on April 
24, 2020, SBA resumed accepting applications on May 4, 2020, but only 
from agricultural enterprises. SBA resumed accepting applications from 
all eligible applicants on June 15, 2020. On July 11, 2020, SBA 
announced that funding for EIDL advances had been exhausted and 
subsequently removed the option to request EIDL advances from the 
application website. 
                                                                                                                       
16SBA modified the original contract from $100 million to $600 million on April 17, 2020, 
and again to $750 million on August 12, 2020. SBA also extended the term of the contract 
as part of the modification.  
17RER Solutions told us that the company receives 51 percent of the contract’s funding 
while the subcontractors combined receive 49 percent.  

 
 
 
 
 
 
Page 11 
GAO-21-589  Economic Injury Disaster Loan Program 
SBA processed EIDL loans and advances differently (see fig. 2). 
Applicants that applied to the EIDL program before July 2020 (prior to the 
depletion of funding for advances) could apply for advances and loans on 
the same application. SBA approved and declined advances solely based 
on automated validation of certain applicant information. To approve and 
decline loans, SBA combined automated validation and further manual 
review by SBA staff. 
Figure 2: SBA Processing of Economic Injury Disaster Loans and Advances 
 
Note: This figure does not depict processing of targeted or supplemental advances. 
 
Rapid Finance’s underwriting system accesses various third-party 
databases and public domain information to identify certain indicators of 
potential fraud and to validate bank account information. Examples of 
potential fraud indicators the system identifies include suspicious online 
behavior, high-risk internet protocol addresses, bank account ownership 
that does not match the business, and owner information that cannot be 
validated. Additionally, the system compares application responses to 
SBA’s eligibility criteria, such as type of business activity and the date on 
EIDL Loan and Advance 
Processing Steps 

 
 
 
 
 
 
Page 12 
GAO-21-589  Economic Injury Disaster Loan Program 
which the business was established.18 SBA automatically approved or 
declined applicants for advances that passed fraud and eligibility 
validations without further action.19 To determine eligibility for loans, the 
system also pulls credit reports to validate whether the applicant meets 
the minimum credit score, has an open bankruptcy, or has delinquent 
child support. The validation system does not automatically approve loan 
applications but does automatically decline loan applications that do not 
meet the minimum credit score. The system records fraud alerts and 
other alerts identified through the validation process for further review by 
SBA staff. 
SBA loan officers at the Processing and Disbursement Center are to 
review fraud and other alerts recorded by the automated validation 
system and approve or decline the application based on whether they can 
mitigate the alerts through research or obtain information from applicants. 
Supervisory officials—team leaders—review the loan officers’ decisions 
and make the final decision. In cases where the system does not identify 
any alerts and therefore does not require the loan officers to mitigate 
them, the team leaders are to review the application and approve the 
loan. 
Applicants that pass the automated validation process for loans receive 
invitations to create an account for an online loan portal, where the 
applicant can select the desired loan amount up to the amount of the 
calculated economic injury or the maximum loan limit. Applicants that 
were automatically declined do not have access to this online loan portal. 
Declined loan applicants may request that SBA reconsider their 
applications. Additionally, approved applicants may also file a 
reconsideration request for SBA to increase the loan amount up to the 
loan limit set by SBA.20 
                                                                                                                       
18Certain business activities are ineligible for EIDL loans and advances. These include, for 
example, applicants engaged in illegal activities, political or lobbying activities, and 
production of products or services determined to be obscene. Additionally, businesses 
must have been established on or before January 31, 2020.  
19This process differs for targeted and supplemental advances, which we discuss later in 
the report. Additionally, we provide more information on fraud and eligibility checks later in 
the report. 
20Approved applicants that selected a loan amount less than the amount SBA offered may 
appeal for increases. Additionally, approved applicants that believe their economic injury 
should be higher may also appeal for increases.  

 
 
 
 
 
 
Page 13 
GAO-21-589  Economic Injury Disaster Loan Program 
 
 
 
 
 
 
A key challenge to implementing the EIDL program in response to 
COVID-19 was the historically high volume of applications SBA received, 
according to agency officials. Between March 15, 2020, and February 27, 
2021, SBA accepted about 17.9 million EIDL applications and processed 
17.7 million of them. Of the processed applications, SBA approved about 
3.8 million loan applications totaling about $203.5 billion, declined about 6 
million applications, withdrew about 2.6 million applications, and found 5.3 
million applications to be duplicates.21 Additionally, between March 29, 
2020, and July 15, 2020, SBA approved about 5.8 million EIDL advances 
for about $20 billion and declined about 2.5 million requests for advances. 
For comparison, from SBA’s inception in 1953 until its COVID-19 
response in March 2020, SBA approved a total of about 2.2 million 
disaster loans for $67 billion, according to one SBA official. 
Our review of SBA’s weekly summary processing reports showed that 
SBA missed its goal of approving or declining 85 percent of the COVID-
19 loan applications within 60 days.22 As of February 27, 2021, SBA 
processed about 63 percent of the approved applications within 60 days 
and about 90 percent of the declined applications within 60 days. The 
cumulative average processing times were longer for approved loans (47 
days) than for declined loans (22 days). The shorter processing time for 
declined loans may be the result of automatic declines, such as when an 
applicant does not meet the minimum credit score. 
                                                                                                                       
21SBA officials told us the agency withdraws applications because of applicant 
nonresponse (after 120 days without a response) or applicant requests to withdraw.  
22SBA has four response levels that correlate staffing levels with the anticipated number 
of applications and processing goals. The pandemic triggered a level IV response, which 
indicates that SBA anticipates more than 500,000 applications and has a processing goal 
of more than 4 weeks.  
SBA Took Steps to 
Address Processing 
Challenges but Lacks 
a Comprehensive 
Communications 
Strategy 
High Volume of 
Applications Led to Longer 
Processing Times Early 
On, but SBA Took Steps to 
Address Them 

 
 
 
 
 
 
Page 14 
GAO-21-589  Economic Injury Disaster Loan Program 
The average processing time for advances was shorter than that for 
loans, and SBA did not establish a processing goal for the advances. 
SBA’s summary processing reports showed that the cumulative average 
processing time for all advance applications was about 18 days, with an 
additional average disbursement time of 7 days for all approved 
advances. SBA processed advances using only automated validation, 
resulting in shorter processing times. 
To process the large volume of applications, SBA worked with Rapid 
Finance to modify its existing validation system and increased processing 
staff: 
• 
Modifying an existing automated validation system to help 
process EIDL applications. SBA asked Rapid Finance to modify its 
existing software and hardware to incorporate SBA’s requirements 
related to fraud checks and eligibility criteria, rather than building a 
new validation system specifically for SBA. An SBA official stated that 
the agency began processing EIDL applications using this new 
system on April 7, 2020. Prior to implementing the new system, SBA 
officials told us loan officers manually reviewed all applications. SBA 
officials stated that the daily number of applications processed was 
616 prior to the use of the validation system and 45,050 after the use 
of the system. 
• 
Increasing staff and establishing processing time goals. SBA 
data for March 20, 2020, to February 12, 2021, show that staffing for 
the Processing and Disbursement Center increased five-fold from late 
March 2020 to its peak in December 2020 (see fig. 3). SBA officials 
told us the center initially drew detailees from other SBA offices and 
other federal agencies to help provide support. Additionally, SBA used 
contractors to quickly ramp up staffing starting from the end of April 
2020. SBA officials said the use of detailees and contractors allowed 
SBA time to recruit and hire new employees. In April 2020, SBA 
projected it would need about 6,900 staff at the center to handle 5 
million EIDL applications. The center reached this level of staffing in 
September 2020. SBA officials stated that fully staffing the center 
helped SBA address the increased volume of work. Loan officers 
were expected to approve or decline at least four loan applications per 
hour, and team leaders were expected to review and confirm loan 
officers’ decisions on 10 to 12 loan applications per hour. 

 
 
 
 
 
 
Page 15 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 3: Weekly SBA Processing and Disbursement Center Staffing Levels, March 20, 2020–February 26, 2021 
 
 
SBA’s data show that backlogs in processing loan applications and 
application processing times declined between March 2020 and February 
2021. 
• 
Application backlog. SBA’s summary processing report data for loan 
applications show that initial surges of applications resulted in a 
processing backlog (see fig. 4). By mid-April 2020, when SBA stopped 
accepting new applications due to exhaustion of funds, SBA had 
received about 5.5 million applications and processed about 12 
percent of them.23 The number of new applications then declined 
because SBA limited the program to agricultural enterprises, but it 
increased again when SBA reopened the program to all eligible 
businesses in mid-June 2020. By July 11, 2020, when SBA 
announced that the funding for advances had been exhausted, SBA 
had received about 10.9 million applications and processed 68 
percent of them. SBA’s data show that the number of new loan 
applications decreased starting in mid-July. Subsequently, SBA 
reduced its backlog of loan applications. 
                                                                                                                       
23This total includes all applications SBA received, which may have been approved, 
declined, withdrawn, or found to be duplicates. 

 
 
 
 
 
 
Page 16 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 4: Economic Injury Disaster Loans (EIDL) Cumulative Application Volume and Percentage of Applications Processed, 
March 15, 2020–February 27, 2021 
 
Note: The figure reflects the status of the applications at the time SBA generated the data. 
Processing status may change over time. For example, SBA could reactivate a withdrawn application 
and approve or decline the application later. 
 
• 
Application processing times. Our analysis of SBA’s applicant data 
from March 2020 through February 2021 showed that applicants who 
applied early experienced the longest processing times, up to 76 days 
on average for approved applicants and 58 days on average for 
declined applicants.24 In particular, applicants who applied in March 
and April 2020 experienced longer processing times compared to later 
applicants (see fig. 5). 
                                                                                                                       
24We analyzed the processing times for applications that SBA funded after approval; see 
app. I for more information.  

 
 
 
 
 
 
Page 17 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 5: Number of Economic Injury Disaster Loan Applications and Average Processing Times, by Month Application 
Accepted, March 2020–February 2021 
 
Note: Data for number of days to process applications for approved loans include only applications 
that SBA funded after approval. Number of applications includes applications that were funded or 
were waiting to be funded after approval, or were declined, under processing, or withdrawn. 
 
Some applicants experienced long wait times when calling SBA’s 
customer service line, according to SBA’s data for its Customer Service 
Center from March 21, 2020, through February 20, 2021. For example, 
the daily maximum time that callers were placed on hold between March 
2020 and July 2020 reached a high of 4 hours and 10 minutes.25 
Participants from three of five discussion groups we held, as well as a 
                                                                                                                       
25We obtained this information from the SBA Office of Inspector General’s report 
published in October 2020. See Small Business Administration, Office of Inspector 
General, Inspection of Small Business Administration’s Initial Disaster Assistance 
Response to the Coronavirus Pandemic (Washington, D.C.: Oct. 28, 2020). SBA officials 
told us that the agency does not currently keep track of maximum call times.  
Some Applicants 
Experienced Long Wait 
Times on SBA’s Customer 
Service Line as Call 
Volumes Exceeded SBA’s 
Expectations 

 
 
 
 
 
 
Page 18 
GAO-21-589  Economic Injury Disaster Loan Program 
majority of the stakeholders, told us that applicants experienced long wait 
times to speak to an SBA customer service representative.26 
SBA received about 10.8 million calls between March 2020 and February 
2021. As shown in figure 6, the weekly average wait time ranged from 11 
seconds to 39.5 minutes, with an overall average wait time of 7 minutes. 
Additionally, SBA’s data show that abandoned calls made up about 15 
percent of all calls. About 69 percent of the abandoned calls occurred 
between the weeks of June 14, 2020, and July 12, 2020, when SBA 
received the highest volume of calls. This timing coincided with SBA 
reopening the EIDL program to all eligible applicants on June 15, 2020, 
and SBA announcing that funding for EIDL advances was depleted on 
July 11, 2020. SBA officials told us that a majority of customer service 
calls SBA received were applicants checking on their application status.27 
However, SBA officials also said that as application processing time 
decreased, the number of such calls declined. SBA officials told us that 
due to the high volume of calls related to the pandemic, the Customer 
Service Center did not project call volumes for fiscal year 2020. 
 
 
                                                                                                                       
26We conducted five discussion groups with a total of 22 EIDL program applicants. Two of 
these groups consisted of applicants declined for loans, and three groups consisted of 
applicants that were approved. Each group consisted of four to six participants. The 
opinions expressed by the participants represent their points of view and may not 
represent the views of all EIDL applicants. We refer to the six SBDCs and seven 
associations we spoke with collectively as stakeholders. We use the terms “few,” 
“several,” “a majority,” and “almost all” to characterize stakeholder responses. We define 
“few” as 1–3 responses, “several” as 4–6 responses, “a majority” as 7–10 responses, and 
“almost all” as 11–13 responses.  
27SBA’s customer service representatives may be able to provide information about 
applications depending on the stage of the application. 

 
 
 
 
 
 
Page 19 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 6: SBA Customer Service Line Average Wait Times and Call Volumes, Week of March 15, 2020, through Week of 
February 14, 2021 
 
Note: The customer service calls were primarily, but not exclusively, related to COVID-19. 
 
SBA’s data show that the volume of calls exceeded the capacity of the 
Customer Service Center in late March 2020 and in mid-June 2020 (when 
the program reopened to all eligible applicants) (see fig. 7). In April 2020, 
SBA projected that its Customer Service Center would need about 1,650 
staff. SBA reached this level of staffing in late June 2020. However, SBA 
officials told us that SBA adjusted staffing at the center based on actual 
incoming calls. SBA first used contractors to provide support and then 
increased its own staff, primarily through new hiring. SBA maintained a 
higher level of customer service staffing after June 2020, and the volume 
of calls declined beginning in late July 2020. As a result, average wait 
times and the percentage of abandoned calls also subsequently declined. 
 
 

 
 
 
 
 
 
Page 20 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 7: SBA Customer Service Center Staffing Levels, March 20, 2020–February 26, 2021 
 
 
As SBA began to implement the EIDL program in response to COVID-19, 
its initial communications with potential applicants and other stakeholders 
excluded key information about program terms, creating confusion among 
applicants. SBA also did not communicate information about application 
status to applicants in a timely manner. SBA’s current disaster-related 
plans do not provide a comprehensive strategy for addressing these 
communication issues. 
 
SBA communicated EIDL program information, such as eligibility criteria 
and how to apply, to potential applicants using various channels. 
However, SBA’s outreach and published information during the initial 
months of its COVID-19 response did not provide certain key information 
about the program effectively or in a timely manner. 
SBA distributes information about the EIDL program through media, its 
website, and other channels. For SBA’s COVID-19 response, the 
agency’s Field Operations Centers disbursed information about the 
program by hosting webinars, coordinating with media contacts, and 
interacting with trade associations, chambers of commerce, and resource 
SBA’s Communication 
about Program Terms and 
Applicants’ Status Created 
Confusion and 
Uncertainty, and SBA 
Lacks a Comprehensive 
Communications Strategy 
SBA’s Initial Communication 
Excluded Key Information 

 
 
 
 
 
 
Page 21 
GAO-21-589  Economic Injury Disaster Loan Program 
partners (such as SBDCs). SBA officials also created a frequently asked 
questions (FAQ) document for SBA’s website to provide details about the 
EIDL program in response to COVID-19, including eligibility criteria, steps 
in the application process, and reasons for declining an application. 
However, SBA officials said SBA did not publish the FAQ until September 
2020 because of delays in reviewing the document internally. Prior to the 
publication of the September 2020 FAQ, SBA’s website information was 
limited to the purpose of the program, eligible entities, and loan terms. 
SBA communicated changes to advance and loan limits through some 
channels but not others, sometimes several months after these changes 
occurred: 
• 
Advance limits. SBA revised the limits on advances to $1,000 per 
employee and a maximum of $10,000 per applicant on April 7, 2020, 
and informed EIDL applicants that had applied about the revised limit 
by email on April 13, 2020. SBA also provided the revised advance 
limits during webinars. However, SBA did not communicate this 
information on its website or within the EIDL application. SBA included 
the limits on advances in a press releases issued on June 15, 2020, 
but as of early July 2020—almost 3 months after SBA revised the 
advance limits—SBA’s website still did not include this information.28 
• 
Loan limits. SBA lowered the maximum EIDL loan amount to 
$150,000 beginning on May 4, 2020. However, SBA did not provide 
this information on its website, in its press releases, or within the EIDL 
application. The first time SBA broadly announced the $150,000 loan 
maximum was in February 2021 when it updated its FAQ document. 
Prior versions of the FAQ stated that the maximum loan amount was 
6 months of working capital and did not specify a maximum amount. 
 
Almost all stakeholders we interviewed said that SBA did not make 
information about limits it imposed on advances and loans available to 
applicants in a timely manner. Several stakeholders said this caused 
confusion and frustration for small businesses. For example, a 
representative from one business association said that some small 
business owners did not understand why they received less than they 
                                                                                                                       
28In the June 15, 2020, press release, SBA announced that it would resume accepting 
applications from all eligible applicants and included information about the limit on 
advances. SBA announced on July 11, 2020, that funding for the advances had been 
exhausted. 

 
 
 
 
 
 
Page 22 
GAO-21-589  Economic Injury Disaster Loan Program 
expected, and a representative from an SBDC said some applicants 
viewed the program as false advertising. 
SBA also did not provide other key information to help applicants estimate 
their potential loan amount. Specifically, the EIDL application form 
requests that applicants provide the dollar amounts for 12 months of 
gross revenues and cost of goods sold. These amounts are part of the 
calculation SBA uses to determine an applicant’s economic injury and 
ultimately the loan amount SBA could offer the applicant. However, as of 
April 2021, SBA had not clarified in its application, on its website, or in its 
most recent FAQ how it uses gross revenues and cost of goods sold to 
calculate the loan amount that it could offer. 
Discussion group participants and stakeholders we spoke with provided 
mixed views about the clarity of information presented on the EIDL 
application. Participants from each of the five discussion groups said that 
the application was clear and straightforward. However, a few participants 
from three of the five discussion groups and a majority of the 
stakeholders said some applicants did not understand certain financial 
terms on the application, such as the cost of goods sold. These 
stakeholders told us this resulted in some applicants providing incorrect 
information on their applications and deterred others from applying. 
Providing incorrect information may result in declined applications or 
inaccurate estimates of applicants’ economic injury. 
In February 2021, 11 months after the CARES Act revised the EIDL 
program in response to COVID-19, SBA defined “cost of goods sold” in its 
updated FAQ and provided a link to an Internal Revenue Service website 
for further information.29 However, the link takes the applicant to a 
webpage with over 30 links on a variety of topics, none of which are 
labeled “cost of goods sold.”30 
After submitting their applications, EIDL applicants did not receive clear 
information from SBA about their application status—whether by email, 
from SBA’s online loan portal, or from its Customer Service Center. 
Applicants generally apply online on SBA’s website. As discussed earlier, 
if the application successfully passes the validation process, SBA sends 
applicants an email inviting them to access the loan portal. In the loan 
                                                                                                                       
29https://www.irs.gov/businesses/small-businesses-self-employed.  
30One of these links leads the applicant to another webpage that has a link labeled “cost 
of goods sold.”  
Stakeholders Cited a Lack of 
Information about Application 
Status and Processing Times 

 
 
 
 
 
 
Page 23 
GAO-21-589  Economic Injury Disaster Loan Program 
portal, applicants can view their application status for the first time (for 
example, their screen would display a status such as “application is being 
processed”). Alternatively, applicants who fail the validation process 
would receive an email informing them that SBA has declined their 
application. This email either declining the application or inviting the 
applicant to access the loan portal is generally the first communication 
applicants receive from SBA about the status of their application. 
Applicants may call SBA’s Customer Service Center to check the status 
of their application, but the information the center’s representatives can 
provide may be limited and may depend on the stage of application 
review. Prior to an application undergoing the automated validation 
process, customer service representatives would not be able to access 
system information about the application. After an application has 
completed the validation process, customer service representatives can 
inform an applicant if SBA has approved or declined the application, if the 
applicant has to take additional steps to move the application forward, or 
if the application is under review. For applications under review, SBA’s 
training guide directs customer service representatives not to provide 
more detailed information on application status. SBA officials told us this 
was because the agency could not share certain information with the 
applicant. Participants from three discussion groups said that customer 
service representatives could not provide more than general information 
about the status of their applications. 
Participants from all five discussion groups and several stakeholders we 
spoke with cited lack of information and uncertainty about application 
status as major concerns for applicants. Participants from all five 
discussion groups told us they wanted to know if they had submitted their 
application successfully and about the progress of their application 
throughout the entire application process. A few stakeholders said status 
information would have helped applicants make business decisions that 
were contingent on the EIDL application outcome, such as whether to 
apply for other funding. One discussion group participant explained that 
knowing in advance when she could expect to receive funds would have 
helped her decide if she should identify other funding sources to pay for 
her inventory. Participants from one discussion group also told us that 
they reapplied to the program because SBA did not confirm the receipt of 
their applications or did not provide clear directions when it made 
application processing changes. SBA data show that 5.3 million of the 

 
 
 
 
 
 
Page 24 
GAO-21-589  Economic Injury Disaster Loan Program 
17.9 million EIDL applications SBA accepted from March 15, 2020, to 
February 27, 2021, were duplicates.31 
SBA officials acknowledged that not knowing their application status 
alarmed applicants, particularly early in the pandemic when applicants 
had to wait several weeks. Additionally, they said that because SBA 
quickly built the EIDL application website specific to the COVID-19 
response, the website did not incorporate the functionality to provide 
applicants with confirmation emails and status updates. 
SBA tracks high-level information that could provide approximate 
processing times for various stages of the application process, but it has 
not made this information publicly available to applicants, such as through 
its website. SBA also collects detailed data on loan applications, including 
the number of applications submitted, approved, and declined, and 
develops weekly statistics such as the average number of days to 
approve and decline applications. Similarly, SBA develops processing 
time statistics for other EIDL-related funding and actions, including for 
reconsiderations of declined applications and requests for loan increases. 
The nature and scale of the COVID-19 pandemic have presented 
unprecedented challenges for SBA’s operations, including its 
communications. Unlike prior disasters, which were regional and episodic, 
the pandemic is nationwide and SBA has operated in disaster recovery 
mode for over a year—longer than it has for past disasters. As SBA 
continues to implement the EIDL program in response to COVID-19, clear 
and timely communication with applicants remains important. 
SBA has taken some steps to improve its communication efforts. For 
example, in contrast to SBA’s implementation of the EIDL loans in 
response to COVID-19, its implementation of targeted advances included 
a FAQ document detailing eligibility requirements, targeted advance 
program criteria, and the eligible amount. Additionally, SBA publicized its 
decision to increase the maximum loan limit to $500,000 through a press 
release on March 24, 2021, and announced the effective date. However, 
SBA has not yet provided information on processing times related to EIDL 
loans. For example, while SBA has begun processing appeals for loan 
                                                                                                                       
31Some duplicates may have occurred because of early changes in the program. SBA 
updated its website to include an option to request advances on March 29, 2020. SBA 
informed applicants that applied for a COVID-19 EIDL loan prior to March 30, 2020, that 
they had to reapply in order to obtain the advances. SBA’s data showed that as of March 
28, 2020, SBA had received about 93,500 applications.  
SBA Has Communication 
Plans but Lacks a 
Comprehensive 
Communications Strategy 

 
 
 
 
 
 
Page 25 
GAO-21-589  Economic Injury Disaster Loan Program 
increases to the limit of $500,000, it has not yet provided information on 
processing times for such appeals. 
Federal internal control standards state that management should 
externally communicate the necessary quality information to achieve the 
entity’s objectives. Management is to select appropriate methods to 
communicate externally and, in doing so, to consider the intended 
audience, the availability of information for the audience when needed, 
and the costs of communicating the information, including the resources 
used to communicate.32 Additionally, the Paperwork Reduction Act has a 
broad requirement that an agency disseminate information in a manner 
that is efficient, effective, and economical.33 
SBA has two plans to guide its predisaster preparation and post-disaster 
response efforts, including communications: 
• 
Disaster Preparedness and Recovery Plan. SBA’s 2019 Disaster 
Preparedness and Recovery Plan identifies roles and responsibilities 
for functional areas, such as public communications, as part of the 
disaster assistance response. The plan outlines how SBA is to 
publicly communicate with its target audiences—such as disaster 
survivors, government officials, media outlets, and resource 
partners—about the agency’s recovery operations and services. The 
plan also broadly outlines the type of information to provide disaster 
survivors, such as how to contact SBA and its offices to obtain 
assistance in applying for disaster loans. 
• 
Outreach and Marketing Plan. SBA’s Office of Disaster Assistance 
maintains an Outreach and Marketing Plan for State, Local, and Tribal 
Leadership and Emergency Personnel, which provides broad 
guidance for raising stakeholder awareness about SBA’s disaster 
assistance programs. These stakeholders consist of federal, state, 
local and tribal entities; resource partners; and media contacts. 
 
                                                                                                                       
32GAO, Standards for Internal Control in the Federal Government, GAO-14-704G 
(Washington, D.C.: Sept. 10, 2014).  
33The Paperwork Reduction Act of 1995 requires agencies to minimize the paperwork 
burden on individuals and small businesses and maximize the utility of information 
collected by agencies. Pub. L. No. 104-13, 109 Stat. 163 (codified, as amended, at 44 
U.S.C. §§ 3501-3521). 

 
 
 
 
 
 
Page 26 
GAO-21-589  Economic Injury Disaster Loan Program 
However, these plans do not identify the type of information that 
applicants would need in order to understand and participate in disaster 
loan programs, guidelines on what information to provide to potential 
applicants and when to provide it, and considerations for how applicants 
could access the information in a manner that is efficient, effective, and 
economical. The lack of existing guidelines may have amplified the 
communications challenges applicants faced when combined with the 
magnitude and longevity of the pandemic response. 
The scope of the COVID-19 pandemic may have exceeded the bounds of 
what SBA has previously experienced, but by developing a more 
comprehensive strategy for communicating with disaster relief program 
participants, SBA would be better prepared in the future to respond to 
longer-term, large-scale disasters. Such a strategy could help ensure 
SBA more consistently communicates with program participants in a 
clear, timely manner, facilitating program implementation. In turn, this 
could also reduce demand on SBA’s resources—for example, by 
providing timely information on application status, SBA could reduce the 
number of applicants who contact the Customer Service Center or 
reapply. 
 
 
 
As previously discussed, SBA decreased the loan limit from $500,000 to 
$150,000 in May 2020 in order to assist more small businesses, which 
increased the number of EIDL borrowers with unmet financial needs. For 
the initial $500,000 loan limit SBA had in place from March 16, 2020, 
through May 3, 2020, SBA data show that the agency approved about 
7,000 EIDL loans in the amount of $500,000 for applicants with an 
economic injury greater than $500,000, which indicates that these 
borrowers may have had some degree of unmet need. As of April 1, 
2021, SBA had approved 567,553 EIDL loans in the amount of $150,000 
for applicants with an economic injury greater than $150,000, which 
represents about 15 percent of all applicants approved between March 
15, 2020, and April 3, 2021. As mentioned previously, SBA began to 
accept requests for loan increases on April 6, 2021. 
Participants from our three discussion groups consisting of approved loan 
applicants and stakeholders we interviewed expressed mixed views on 
Program Participants 
Generally Found the 
EIDL Program Helpful 
Loan Limits Affected More 
Borrowers after SBA 
Decreased the Limit in 
May 2020 

 
 
 
 
 
 
Page 27 
GAO-21-589  Economic Injury Disaster Loan Program 
the effect of the loan limit. For example, discussion group participants 
said that they accepted a smaller loan amount than SBA offered to cover 
expenses for a short duration, which indicates that the limit did not affect 
the program’s ability to meet their short-term needs. However, several 
stakeholders said that loan amount was insufficient for some businesses, 
particularly larger businesses and especially as the pandemic continued 
beyond 6 months. 
Discussion groups and stakeholders identified various reasons for 
applicants’ interest in EIDL loans and advances: 
• 
Program provides needed funds for operating expenses. 
Businesses were interested in using the EIDL program for continued 
operating expenses, such as rent and utilities, and new operating 
expenses, such as personal protective gear and technology to 
facilitate online operations, according to a majority of the stakeholders 
and participants from four discussion groups. 
• 
Program terms are attractive. Loan terms—including the low 
interest rate, long repayment period, and flexibility in how businesses 
could use EIDL loans—made the program attractive, according to 
participants from three discussion groups and a majority of the 
stakeholders. 
• 
Credit with similar terms not available elsewhere. The lack of 
available lending or difficulty in obtaining credit elsewhere, such as 
from a bank, led some applicants to apply to EIDL, according to 
participants from two discussion groups and several stakeholders. For 
example, a few stakeholders noted businesses faced restrictive 
underwriting criteria because of the uncertainties of the pandemic. 
 
Participants in three discussion groups consisting of approved EIDL 
applicants and a majority of the stakeholders told us that the EIDL 
program has been beneficial for small businesses (other stakeholders did 
not comment on the impact of the program). Discussion group 
participants provided examples of how they used loans, such as paying 
business partners for goods and maintaining relationships with business 
partners, pivoting their business models (such as toward online sales), 
and helping the business to cover expenditures. Representatives from all 
six SBDCs said EIDL loans helped to keep businesses open, in part by 
providing them with funding to pay for operating and other expenses. 
Participants from our two discussion groups of EIDL applicants whose 
loans were declined provided examples of how the declination negatively 
Stakeholders Said EIDL 
Loan Terms Are Attractive 
and the Loans Have Been 
Helpful to Businesses 

 
 
 
 
 
 
Page 28 
GAO-21-589  Economic Injury Disaster Loan Program 
affected their business and personal finances. For example, they accrued 
debt with interest rates higher than EIDL or used retirement or other 
savings to keep their businesses open, laid off personnel, or cut back on 
certain spending. However, they also said that funding obtained 
elsewhere—such as forgivable loans from the Paycheck Protection 
Program—mitigated the negative impact.34 
Finally, several stakeholders noted that some businesses may not be 
interested in new loans because of their unwillingness to incur debt and 
the longer-term uncertainty associated with the pandemic. One declined 
loan applicant told us that she was only interested in the advances and 
did not want to take on a large loan. 
 
 
 
 
 
We and others have identified potentially fraudulent and ineligible 
applicants within the EIDL program. Additionally, fraud schemes identified 
by law enforcement officials and our analysis of EIDL fraud-related court 
cases charged by the Department of Justice (DOJ) point to significant risk 
of fraud in the program. 
As we reported in January 2021, SBA had provided about 5,000 
advances totaling about $26 million to potentially ineligible businesses in 
three types of industries—adult entertainment, casino gambling, and 
marijuana retail—as of July 14, 2020.35 In addition, SBA had approved at 
least 3,000 loans totaling about $156 million to businesses that SBA 
policies state were ineligible for the EIDL program, such as real estate 
developers and multilevel marketers, as of September 30, 2020. In 
                                                                                                                       
34Loans from the Paycheck Protection Program, another SBA program, may be fully 
forgiven if they meet certain conditions, including that at least 60 percent of the loan 
forgiveness amount be for payroll costs.  
35See GAO, 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). 
SBA Has Begun to 
Address Eligibility and 
Fraud Concerns but 
Has Not Fully 
Implemented GAO 
Recommendations 
EIDL Program Is at High 
Risk of Funding 
Fraudulent and Ineligible 
Applicants 

 
 
 
 
 
 
Page 29 
GAO-21-589  Economic Injury Disaster Loan Program 
October 2020, SBA’s Office of Inspector General (OIG) reported that it 
found strong indicators of fraud, such as loans going to potentially 
fraudulent accounts; loans to applicants with the same internet protocol 
address, email address, bank account, or mailing address; and loans to 
potentially ineligible businesses. In December 2020, SBA’s independent 
financial statement auditor noted discrepancies including more than one 
loan or advance approved and disbursed to the same borrower. The 
auditor also identified over 6,000 disbursed EIDL loans (over $212 million 
total) issued to potentially ineligible borrowers. 
Our review of aggregate data from the Financial Crimes Enforcement 
Network (FinCEN) also indicated potential fraud in the EIDL program. Our 
January 2021 report found that between May and October 2020, over 900 
U.S. financial institutions filed more than 20,000 suspicious activity 
reports related to the EIDL program with FinCEN.36 These reports 
identified multiple types of potentially suspicious activity related to EIDL, 
such as indicators of identity theft, the rapid movement of funds, and 
forgeries. 
In addition, law enforcement officials have reported systemic patterns of 
fraud across investigations involving EIDL loans. We reported in March 
2021 that officials at SBA OIG and the Federal Bureau of Investigation 
told us they had identified systemic patterns of potential fraud including 
identity theft, false attestations on loan documents, fictitious and inflated 
employee counts, and misuse of proceeds.37 Additionally, a senior official 
with SBA OIG told us that, as of January 2021, SBA OIG had opened 
over 260 investigations related to CARES Act loans, including EIDL 
                                                                                                                       
36See GAO-21-265. Institutions are required to electronically file a suspicious activity 
report when a transaction involves or aggregates at least $5,000 in funds or other assets, 
and the institution knows, suspects, or has reason to suspect that the transaction meets 
certain criteria qualifying as suspicious. Banks are also required to file a report for known 
or suspected criminal violations involving insider abuse of any amount, as well as 
violations aggregating $5,000 or more when a suspect can be identified, and $25,000 or 
more without a potential suspect. See, e.g., 31 C.F.R. § 1020.320. See also 12 C.F.R. §§ 
21.11(c)(1)-(3), 163.180(d)(3)(i)-(iii) (OCC); 12 C.F.R. § 208.62(c)(1)-(3) (Federal 
Reserve); 12 C.F.R. § 353.3(a)(1)-(3) (FDIC). Although the filing of a suspicious activity 
report does not necessarily mean that fraud has occurred, law enforcement agencies 
query these reports and use them to support investigations, such as those related to EIDL 
fraud. 
37See GAO, COVID-19: Sustained Federal Action Is Crucial as Pandemic Enters Its 
Second Year, GAO-21-387 (Washington, D.C.: Mar. 31, 2021).  

 
 
 
 
 
 
Page 30 
GAO-21-589  Economic Injury Disaster Loan Program 
loans—at least three times the number of investigations the office would 
typically open in a year. 
In our analysis of 51 DOJ cases involving fraud charges for EIDL loans as 
of March 2021, we identified characteristics similar to those identified by 
SBA OIG and Federal Bureau of Investigation. Specifically, across the 51 
cases in which DOJ filed fraud-related charges, we found that 19 involved 
identity theft, 39 involved false attestation, 39 involved fictitious or inflated 
employee counts, and 29 involved misuse of proceeds.38 These cases 
included charges of bank fraud, wire fraud, false statements, money 
laundering, and identity theft (see fig. 8). 
Figure 8: Department of Justice Fraud Charges Related to the Economic Injury 
Disaster Loan Program, as of March 2021 
 
Note: A case could include multiple charges, so these categories are not mutually exclusive. 
 
The 90 defendants in these 51 cases sought at least $15.4 million and 
received at least $12.2 million in EIDL funds. In 29 of the 51 cases, 
defendants allegedly spent EIDL funds on ineligible purchases, such as 
real estate, automobiles, cryptocurrency, and luxury goods. Defendants in 
six cases allegedly diverted at least $860,000 in EIDL funds overseas, 
                                                                                                                       
38As of March 2021, 13 of the 51 cases were closed. Of the 13 closed cases, 12 reached 
conclusion through a guilty plea and one was dismissed. Thirty-eight cases remained 
ongoing. We identified similar characteristics among closed and ongoing cases. 

 
 
 
 
 
 
Page 31 
GAO-21-589  Economic Injury Disaster Loan Program 
including transfers to the Dominican Republic, Ghana, Singapore, South 
Korea, Taiwan, Uganda, and the United Kingdom. 
EIDL funds also allegedly went to fake and nonoperating businesses. 
Specifically, across the 51 cases, 18 involved funds going to fictitious 
entities, and 14 cases involved funds going to shell companies, which do 
not have employees or business operations (see text box for illustrative 
examples).39 
Illustrative Examples of Fraud against the Economy Injury Disaster Loan Program (EIDL) 
Ineligible business received disaster funds. A defendant who pleaded guilty to fraud against the EIDL program had falsely 
claimed eligibility for an EIDL loan, despite (1) his business not operating prior to the pandemic, (2) his criminal history, and 
(3) listing an ineligible address for his business. Although the defendant claimed to have operated a retail apparel business 
since 2015, he did not register it with his state until 2020. The defendant also falsely certified that he had no prior convictions 
despite being on parole. Lastly, the defendant listed his business address as a UPS store located in a strip mall, despite the 
EIDL application specifying that the primary business address could not be a post office box. The Small Business 
Administration disbursed $196,900 in EIDL funds to the defendant. 
Defendant used shell companies and stolen personal information to obtain pandemic assistance. Another defendant 
applied for an EIDL loan for a shell company that had no operations or employees. The defendant used his mother’s name 
to apply for an Employer Identification Number for the shell company. On the application, he falsely represented that the 
business had been operating since 2017 and had multiple employees and gross receipts of over $1,500,000. In total, the 
defendant applied for $750,000 and the Small Business Administration issued $299,800 in EIDL funds. 
Source: GAO analysis of public federal court documents. | GAO-21-589 
 
Our analysis indicated that many defendants were alleged to have 
committed other crimes in addition to EIDL fraud. For example, 30 of the 
51 cases also involved alleged fraud against the Paycheck Protection 
Program, and eight cases involved alleged fraud against other pandemic-
related programs, such as unemployment insurance fraud (see text box 
for illustrative example). Some cases also involved allegations of other 
crimes, such as romance scams, business email compromise schemes, 
and health care fraud.40 
 
                                                                                                                       
39Applicants in these examples applied for EIDL loans between March 2020 and July 
2020. 
40Romance scams occur when a criminal adopts a fake online identity to gain a victim’s 
affection and trust. The scammer then uses the illusion of a romantic or close relationship 
to manipulate or steal from the victim. 

 
 
 
 
 
 
Page 32 
GAO-21-589  Economic Injury Disaster Loan Program 
Fraudster Targeted Multiple Federal Programs 
A defendant pleaded guilty to fraud against the Economic Injury Disaster Loan (EIDL) program and other federal pandemic 
relief programs. On his EIDL application, the defendant fraudulently inflated the number of employees and salary information 
for his business, which provided goods and services to federal agencies. The defendant also targeted the Paycheck 
Protection Program, another Small Business Administration pandemic relief program. In total, the defendant received over 
$1 million in pandemic relief funds and spent those funds on ineligible expenses. 
The defendant also pleaded guilty to making false statements for falsely claiming his business had N95 masks available to 
sell to other federal agencies seeking to purchase personal protective equipment. In addition, the defendant pleaded guilty to 
theft of government funds for attempting to obtain veterans’ benefits by falsely claiming to have served in the U.S. Marine 
Corps. 
Source: GAO analysis of public federal court documents. | GAO-21-589 
 
Fraud against the EIDL program can have broad and serious effects on 
the government, the economy, and individuals. In addition to the financial 
losses and lost program delivery potential from unknown fraud, the 
government bears the burden of investigating and prosecuting the alleged 
fraud, which redirects resources law enforcement agencies could use to 
combat other crimes. Additionally, fraud resulting from internal control 
failures can have a reputational effect on government services, negatively 
affecting the public’s trust in government. Fraud can also affect the 
broader economy. Fraud against public programs can distort markets by 
giving fraudsters unfair advantages, diverting fraudulently obtained 
money into the economy, and undermining efforts to maintain the integrity 
of the financial system. Finally, experiencing fraud, such as identity theft, 
can be a traumatic experience for victims, causing financial losses as well 
as serious psychological and emotional harm. 
Over the course of its COVID-19 response, SBA has made some 
changes to enhance its application review process and to identify 
potential fraud. These changes applied to both the automated validation 
system and the manual review process. SBA officials told us that some of 
the changes resulted from interactions with SBA OIG and that others 
resulted from loan officers who raised awareness about issues. For 
example: 
• 
Validating application inputs. SBA made several changes to 
validate applicant information. For example, in May 2020, SBA 
changed the application such that the applicant must check each 
eligibility criterion before being able to proceed, added validation of 
bank account routing numbers, and added a function to identify 
mismatches between ZIP codes and states. In July 2020, SBA began 
to validate the types of tax identification numbers associated with the 
SBA’s Application Review 
Process for EIDL-Related 
Programs Changed over 
Time to Address Fraud 
and Other Risks 

 
 
 
 
 
 
Page 33 
GAO-21-589  Economic Injury Disaster Loan Program 
types of entity. In August 2020, SBA began to revalidate bank account 
information whenever the loan applicant changes this information. 
• 
Changing the application review process. SBA OIG found that until 
August 2020, applications that did not contain certain fraud alerts 
flagged by the automated validation system were being approved by 
team leaders in batches and with little to no additional review by the 
team leaders.41 However, these applications contained other issues 
that SBA did not review, such as the inability to confirm business 
registration. After August 2020, SBA stopped approving loans in 
batches and began requiring SBA staff to review all applications prior 
to approval and to mitigate all system alerts. SBA data showed that 
between April and August 2020, SBA approved about 3.2 million 
applications. 
• 
Addressing suspected fraud. Beginning in June 2020, SBA took 
actions to improve loan officers’ ability to withhold funding for 
applicants suspected of fraud. SBA set up teams to address 
suspected EIDL fraud, identity theft complaints, and financial 
institution inquiries, among other things, and added staffing to those 
teams. In particular, SBA created an email account for loan officers to 
report potential fraud internally to a team that reviews these cases 
and refers them to SBA OIG. SBA also created a reporting 
mechanism for victims of identity theft to request release from EIDL 
loan debt, as well as internal processes to review and validate those 
requests and refer those cases to SBA OIG.42 
 
SBA has also made changes to the loan application review process in 
response to new legislation. SBA officials told us the CARES Act’s 
restriction on using applicants’ tax information presented a challenge for 
                                                                                                                       
41Small Business Administration, Office of Inspector General, Inspection of Small 
Business Administration’s Initial Disaster Assistance Response to the Coronavirus 
Pandemic. 
42In May 2021, SBA OIG made five recommendations to SBA related to identity theft and 
the EIDL program. The recommendations related to (1) developing a process to maintain 
and track all identity theft complaints; (2) developing a process to provide status updates 
to each complainant alleging identity theft; (3) completing and formalizing a process to 
restore identity theft victims to their conditions prior to the fraud; (4) developing a process 
to remove any fraudulent loans and fees; and (5) reviewing over 150,000 loan billing 
statements sent to EIDL borrowers that had been returned to SBA to resolve any that 
involved identity theft and refer fraudulent loans to SBA OIG. As of the report’s publication 
in May 2021, SBA had addressed the recommendation related to review of loan billing 
statements. Small Business Administration, Office of Inspector General, SBA’s Handling 
of Identity Theft in the COVID-19 Economic Injury Disaster Loan Program (Washington, 
D.C.: May 6, 2021). 

 
 
 
 
 
 
Page 34 
GAO-21-589  Economic Injury Disaster Loan Program 
validating applications. The agency had relied on self-certification of 
applicant information and the controls put in place as part of the 
automated validations and manual review. The Consolidated 
Appropriations Act, 2021 removed this restriction. SBA officials told us 
that beginning in April 2021, the agency had started incorporating tax 
information as part of its validation process for loan applications to 
confirm that businesses existed on or before January 31, 2020, a 
requirement for program eligibility, and to verify business revenue. 
For targeted and supplemental advances, SBA built on the automated 
validations and manual review processes that existed for loan 
applications but incorporated additional requirements specific to targeted 
and supplemental advances. For example, to confirm eligibility, SBA 
began calculating applicants’ income losses to ensure that applicants met 
legislative criteria for income loss, and it created a map to confirm that 
applicants reside in low-income areas. SBA obtains specific tax return 
information to confirm that applicants existed on or before January 31, 
2020, and that applicants meet the income loss criteria imposed by 
legislation. Additionally, in March 2021, SBA began to use the 
Department of the Treasury’s Do Not Pay service to verify recipients’ 
eligibility for payment in order to reduce improper payments.43 
SBA continues to rely on applicant self-certification for certain eligibility 
criteria (see table 2), as allowed by the CARES Act. The EIDL application 
informs applicants they are self-certifying under penalty of perjury. 
 
 
                                                                                                                       
43The Department of the Treasury’s Do Not Pay service is an analytics tool that helps 
federal agencies detect and prevent improper payments made to vendors, grantees, loan 
recipients, and beneficiaries. Agencies can check multiple data sources to make payment 
eligibility decisions.  

 
 
 
 
 
 
Page 35 
GAO-21-589  Economic Injury Disaster Loan Program 
Table 2: Economic Injury Disaster Loan Program Eligibility Criteria and Validation 
Eligibility criteria 
Loans 
Advances 
Targeted and 
supplemental 
advances 
Applicant must meet employee size limits.  
Self-certification 
Self-certification 
Self-certification 
Applicant cannot be a state, local, or municipal government 
entity, or a member of Congress. 
Self-certification 
Self-certification 
Self-certification 
Applicant cannot be engaged in illegal activity or production of 
products or services that have been determined to be obscene 
by a court. Applicant also cannot be primarily engaged in political 
or lobbying activities; present live performance of a prurient 
sexual nature; or derive more than one-third of gross annual 
revenue from legal gambling activities. Additionally, businesses 
whose purpose for being is gambling are not eligible.a 
Self-certification 
Self-certification 
Self-certification 
Business must have been established on or before January 31, 
2020. 
Self-certification until 
April 2021; 
thereafter, validation 
using tax information 
Self-certification 
Validation using tax 
information 
Applicant cannot be engaged in lending or investment, 
speculative activities, multilevel sales distribution, real estate 
development or investment, or life insurance. Applicant also 
cannot earn more than one-third of its gross revenue packaging 
Small Business Administration (SBA) loans, have a business 
considered as a hobby, or be a pawn shop that derived more 
than 50 percent of prior year’s income from interest. 
Self-certification 
Criteria not 
applicable 
Self-certification 
Business owners with 50 percent or more ownership interest 
must not be more than 60 days delinquent on child support. 
Self-certification, 
credit report 
validation for child 
support delinquency 
Self-certification 
Self-certification, 
credit report 
validation for child 
support delinquency 
Applicant must meet certain character requirements, including 
not having a felony within the last 5 years involving fraud, 
bribery, embezzlement, or a false statement in a loan application 
or an application for federal financial assistance. Additionally, 
applicant must not be currently incarcerated. 
Self-certification 
Criteria not 
applicable 
Self-certification 
Applicant must be a U.S. citizen, noncitizen national, or qualified 
alien. 
Self-certification 
Criteria not 
applicable 
Self-certification 
Applicant must be at least 18.  
Self-certification 
Criteria not 
applicable 
Self-certification 
Applicant must not be debarred from contracting with the federal 
government or receiving federal grants or loans. 
Self-certification 
Criteria not 
applicable 
Self-certification 
Applicant must meet certain SBA criteria related to franchises. 
Self-certification 
Criteria not 
applicable 
Self-certification  
Applicant must meet a minimum credit score. 
Credit report 
validation 
Criteria not 
applicable 
Criteria not 
applicable 
Applicant must not have open bankruptcy. 
Credit report 
validation 
Criteria not 
applicable 
Criteria not 
applicable 

 
 
 
 
 
 
Page 36 
GAO-21-589  Economic Injury Disaster Loan Program 
Eligibility criteria 
Loans 
Advances 
Targeted and 
supplemental 
advances 
Applicant’s economic injury must not be zero or negative. 
Self-certification until 
April 2021; 
thereafter, validation 
using tax information  
Criteria not 
applicable 
Criteria not 
applicable 
Source: GAO analysis of Small Business Administration documents. | GAO-21-589 
aThe restriction against businesses whose purpose for being is gambling applies to loans and 
targeted and supplemental targeted advances, but not advances. 
 
GAO, SBA OIG, and SBA’s financial statement auditor have made 
recommendations to address weaknesses in SBA’s controls that may 
have led SBA to approve and disburse funds to potentially ineligible or 
illegitimate entities. SBA has not yet addressed many of these 
recommendations. 
 
As discussed earlier, in January 2021, we reported that our analysis of 
SBA data showed that the agency approved EIDL loans and advances for 
potentially ineligible businesses. We therefore recommended that SBA 
develop and implement data analytics across EIDL loans and advances 
made in response to COVID-19 as a means to detect potentially ineligible 
and fraudulent applications.44 Data analytics can be used to detect 
potential fraud and can include a variety of techniques, such as data 
mining, data matching, and predictive analytics, to identity particular types 
of behavior. SBA neither agreed nor disagreed with this recommendation. 
In May 2021, SBA officials stated the agency was in the process of 
developing analysis to apply certain fraud indicators to all application 
data. These data include applicants’ physical address, internet protocol 
address, email address, phone numbers, bank accounts, and tax 
identification numbers. 
Additionally, in March 2021, we reported that SBA did not have a 
comprehensive plan to proactively assess controls and mitigate risks in 
the EIDL program in a timely manner and that SBA had not conducted a 
formal fraud risk assessment for the program.45 We recommended that 
SBA 
                                                                                                                       
44GAO-21-265. 
45GAO-21-387.  
A Number of 
Recommendations by 
GAO and Others to 
Address EIDL Fraud Risks 
Remain Unaddressed 
GAO Findings and 
Recommendations 

 
 
 
 
 
 
Page 37 
GAO-21-589  Economic Injury Disaster Loan Program 
• 
implement a comprehensive oversight plan to identify and respond to 
risks in the EIDL program to help ensure program integrity, achieve 
program effectiveness, and address potential fraud; 
• 
conduct and document a fraud risk assessment for the EIDL program; 
and 
• 
develop a strategy that outlines specific actions to address assessed 
fraud risks in the EIDL program on a continuous basis. 
 
SBA agreed with the recommendations, stating that it would work to 
implement a comprehensive oversight plan for EIDL and ensure that a 
fraud risk assessment for EIDL is completed and that fraud risks are 
monitored on a continuous basis. In May 2021, SBA officials told us that 
the agency had begun the process of assessing fraud risk for its CARES 
Act programs, including EIDL. Officials also told us that SBA had not yet 
designated an anti-fraud entity to lead fraud risk management activities 
across the agency, but that it plans to do so by the end of fiscal year 
2021. 
In October 2020, SBA OIG reported weaknesses in SBA’s controls that 
could have led SBA to disburse payments to ineligible entities.46 These 
findings include that SBA did not adequately address alerts raised by the 
contractor’s automated validation system and that SBA provided duplicate 
loans to applicants. SBA OIG made 10 recommendations to SBA. Among 
other things, SBA OIG directed SBA to review (1) loans with bank 
account number changes; (2) duplicate loans made to the same internet 
protocol addresses, email addresses, business addresses, and bank 
accounts; and (3) advances to determine the recipients’ eligibility and 
legitimacy. SBA OIG also directed SBA to cancel all ineligible loans and 
recover funds from ineligible applicants. Additionally, SBA OIG directed 
SBA to strengthen its verification controls. In its report, SBA OIG stated 
that SBA partially agreed with nine recommendations and disagreed with 
one. The report stated that SBA took corrective actions to implement 
                                                                                                                       
46Small Business Administration, Office of Inspector General, Inspection of Small 
Business Administration’s Initial Disaster Assistance Response to the Coronavirus 
Pandemic. 
SBA OIG Findings and 
Recommendations 

 
 
 
 
 
 
Page 38 
GAO-21-589  Economic Injury Disaster Loan Program 
seven of the recommendations and that additional actions would be 
needed to address the remaining three recommendations.47 
In December 2020, SBA’s independent financial statement auditor 
identified, among other deficiencies, two material weaknesses in internal 
controls related to EIDL concerning (1) approval of EIDL loans and 
advances and (2) oversight of the contractor that maintains the EIDL 
application system.48 As stated above, the auditor noted duplicate 
payments and payments to potentially ineligible borrowers. The auditor 
also stated that SBA did not adequately design and implement controls 
for the evaluation and monitoring of the contractor that developed and 
maintains the EIDL applications system. Specifically, SBA did not assess 
the contractor’s controls relevant to the processing of EIDL loans and 
advances, including the operating effectiveness of its validation controls, 
such as duplicate application checks, bank account verification, and 
business owner identity checks. The auditor noted that the deficiencies 
were primarily caused by SBA prioritizing the implementation of CARES 
Act provisions and related legislation as quickly and efficiently as possible 
over designing and implementing effective internal control processes. In 
the case of EIDL, this included quickly distributing loans and advances 
without adequately designing and implementing controls to help ensure 
borrower eligibility and accurate recording of transactions. 
The auditor made seven recommendations to SBA to address these two 
material weaknesses. Among other things, the auditor recommended that 
SBA (1) review the EIDL portfolio and determine which transactions were 
made to ineligible recipients; (2) implement controls to detect or prevent 
                                                                                                                       
47SBA disagreed with the OIG’s recommendation to check employee identification number 
dates for EIDL to flag erroneously approved loans and advances as improper payments. 
SBA did not fully resolve two recommendations. To close those recommendations, SBA 
OIG stated that SBA should (1) cancel ineligible loans that were not disbursed, recover 
funds from all loans already disbursed, and flag those loans for the improper payments 
estimates process, and (2) provide evidence that the agency reviewed approved loans to 
determine if there are undisbursed funds remaining that should be suspended until the 
business establishment date is verified and the applicant is deemed eligible.  
48Small Business Administration, Office of Performance Management and the Chief 
Financial Officer, Agency Financial Report, Fiscal Year 2020 (Washington, D.C.: Dec. 18, 
2020). A material weakness is a deficiency, or combination of deficiencies, in internal 
control over financial reporting, such that there is a reasonable possibility that a material 
misstatement of the entity’s financial statements will not be prevented, or detected and 
corrected, on a timely basis. A deficiency in internal control exists when the design or 
operation of a control does not allow management or employees, in the normal course of 
performing their assigned functions, to prevent, or detect and correct, misstatements on a 
timely basis. 
Financial Statement Auditor 
Findings and 
Recommendations 

 
 
 
 
 
 
Page 39 
GAO-21-589  Economic Injury Disaster Loan Program 
loans from being approved for ineligible applicants; (3) train loan officers 
and supervisory loan officers to execute their responsibilities using 
established guidance and standard operating procedures; and (4) 
enhance its review and evaluation of contractor controls. 
The auditor also reported a material weakness in SBA’s entity-level 
controls, citing that SBA management did not properly design and 
implement overall effective management controls to account for new and 
expanded programs, such as EIDL, under the CARES Act and related 
legislation. The auditor issued five recommendations to improve entity-
level controls, including that SBA hold accountable individuals 
responsible for overseeing management’s design, implementation, and 
operation of SBA’s internal control system; developing and implementing 
monitoring controls; and documenting internal controls related to 
implementing the CARES Act and related legislation. 
SBA disagreed with the material weaknesses related to approval of EIDL 
loans and advances and SBA’s entity-level controls and partially agreed 
with the material weakness related to EIDL contractor oversight. SBA 
neither agreed nor disagreed with the recommendations. In May 2021, 
SBA officials stated that the agency had reviewed the findings identified in 
the financial statement auditor’s report and had been working with RER 
Solutions to implement the recommendation related to contractor 
oversight. In particular, officials stated that the contractor had engaged a 
firm to develop a report for the controls relevant to the processing of EIDL 
loans and advances.49 SBA officials stated that this report will allow SBA 
to evaluate the adequacy of the contractor’s controls related to 
processing EIDL loans and advances. 
SBA relies on RER Solutions—the prime contractor—to oversee and 
evaluate subcontractor performance. SBA officials said they focus on 
whether RER Solutions meets performance requirements and do not 
have insight into how the company evaluates subcontractors’ 
performance. A representative from RER Solutions told us that given the 
urgency of the pandemic, the company’s evaluations of subcontractors 
have been on an informal basis as issues have arisen. 
While SBA does not directly oversee the subcontractors, SBA officials 
said that the agency worked closely with the subcontractors to implement 
EIDL program requirements. For example, SBA provided program 
                                                                                                                       
49SBA’s contract with RER Solutions did not require such a report.  
SBA’s Oversight of the 
Contractor Used to 
Process Loans and 
Advances 

 
 
 
 
 
 
Page 40 
GAO-21-589  Economic Injury Disaster Loan Program 
requirements, such as eligibility criteria, to Rapid Finance. Additionally, 
SBA officials told us the agency selected certain fraud checks from a 
suite of Rapid Finance’s options. Similarly, representatives from RER 
Solutions and the subcontractors told us SBA requested certain changes 
or additions to Rapid Finance’s validation system and provided guidance 
through frequent emails and conference calls. Additionally, a 
representative from RER Solutions told us SBA officials interacted directly 
with Rapid Finance—such as through daily calls—to ensure its platform 
met SBA’s needs. As part of ongoing program implementation, SBA 
officials said RER Solutions and the subcontractors continue to hold daily 
calls with SBA. 
SBA’s contract requires that RER Solutions provide certain documents, 
including documents that specify expected risk and risk mitigation and 
project plans. However, SBA officials told us that due to the urgency of 
the pandemic, the agency has not requested certain documents specified 
in the contract. Rather, SBA receives other documentation from RER 
Solutions that is not specified in the contract. Specifically, RER Solutions 
provides a work plan that records the progress of work tasks and goals 
and technical issues. Additionally, RER Solutions provides SBA with 
daily, weekly, and monthly reports that monitor application intake and 
processing.50 A representative from RER Solutions told us that the 
company and the subcontractors have provided these reports and other 
additional information as requested by SBA, beyond the requirements 
specified in the contract. SBA officials said that such reports help SBA to 
determine whether the contract meets performance requirements stated 
in the contract and whether the contractor is maintaining efficiency and 
effectiveness.51 
                                                                                                                       
50RER Solutions and Rapid Finance both stated that Rapid Finance produces these 
reports.  
51SBA’s contract with RER Solutions states that the contractor is to provide 
recommendations to approve or decline applicants within 10 minutes of receiving data 
from the application. However, SBA staff make all final decisions related to applications.  

 
 
 
 
 
 
Page 41 
GAO-21-589  Economic Injury Disaster Loan Program 
 
 
 
 
 
 
 
 
Our analysis of the 12.5 million nonduplicate EIDL loan applications 
received between March 2020 and February 2021 showed that EIDL 
applicants and loan recipients varied in terms of business size (measured 
by the number of employees), years in operation, and industry.52 
Generally, the smallest and oldest businesses were most successful in 
obtaining loans. 
Businesses with fewer than 10 employees, which make up about 94 
percent of the businesses in the United States, also made up the majority 
of EIDL applicants. Most loan applicants (81.1 percent), as well as the 
majority of approved applicants (86.3 percent), had fewer than 10 
employees. SBA approved loans for businesses with 10 or fewer 
employees at a higher rate than for larger businesses (see table 3). 
Businesses with over 500 employees had a lower approval rate. 
 
 
                                                                                                                       
52The data consisted of about 3.7 million approved applications, 6 million declined 
applications, 2.6 million withdrawn applications, and 123,000 other applications, such as 
applications SBA is currently processing. Where we describe all applicants, we analyzed 
the data for all applicants. We analyzed approval rates using only the data for approved 
and declined applications. We analyzed the characteristics of approved applicants using 
only the data for approved applications. For the purposes of this analysis, approved loans 
consist of applications that SBA has funded or for which SBA is obligating or has obligated 
funds.  
Smaller and Older 
Businesses Received 
Larger Loans and 
Were Approved at 
Higher Rates, and 
Wealthier 
Communities 
Received More 
Funding 
Characteristics of EIDL 
Applicants and Loan 
Recipients Varied 
Business Size 

 
 
 
 
 
 
Page 42 
GAO-21-589  Economic Injury Disaster Loan Program 
Table 3: Economic Injury Disaster Loan Approval Rates by Business Size, March 2020–February 2021 
Business size 
Approval rate (percentage) 
0–10 employees 
38.5  
11–49 employees 
35.3  
50–500 employees 
34.6  
501 or more employees 
11.2  
Overall 
38.2 
Source: GAO analysis of Small Business Administration data. | GAO-21-589 
 
Larger businesses generally received larger loans than smaller 
businesses, but businesses with 10 or fewer employees received a 
majority of EIDL loan dollars (see table 4). The average approved loan 
amount was greatest for businesses with between 50 and 500 employees 
($144,320), although it was less for the largest businesses. Several 
factors may have affected the loan amounts business received, including 
SBA’s calculation of economic injury for businesses using gross revenue 
and cost of goods sold, applicants self-selecting the loan amount, and the 
$150,000 loan limit SBA had in place from May 2020 through April 
2021.53 
Table 4: Average Economic Injury Disaster Loan (EIDL) Amount and Percentage of Total Loan Dollars, by Business Size, 
March 2020–February 2021 
Business size 
Average loan amount ($) 
Percentage of total loan dollars 
0–10 employees 
45,793 
76.6 
11–49 employees 
118,467 
20.5 
50–500 employees 
144,320 
2.9 
501 or more employees 
84,999 
0.0 
Overall 
53,593 
100 
Source: GAO analysis of Small Business Administration (SBA) data. | GAO-21-589 
Note: Overall, SBA provided about $200.4 billion in EIDL loans between March 2020 and February 
2021. SBA provided loans to 994 businesses with 501 or more employees. 
 
The majority of EIDL applicants had been in operation for less than 5 
years (62.5 percent), but older businesses had a higher approval rate and 
                                                                                                                       
53SBA increased the loan limit from $150,000 to $500,000 on April 6, 2021. The data 
needed to analyze the effect of this increase were not available within our reporting time 
frames.  
Age of Business 

 
 
 
 
 
 
Page 43 
GAO-21-589  Economic Injury Disaster Loan Program 
received larger loans (see table 5). For example, the average loan size 
for businesses in operation for less than 5 years was about $40,300, 
compared to $65,353 for businesses in operation for between 5 and 10 
years. In addition, businesses in operation for more than 5 years received 
the majority of total loan dollars. 
Table 5: Economic Injury Disaster Loan (EIDL) Approval Rate, Average Loan Amount, and Percentage of Total Loan Dollars, 
by Age of Business, March 2020–February 2021 
Years in operation 
Approval rate 
(percentage) 
Average loan amount ($) 
Percentage of total loan 
dollars 
2 years or less 
23.8 
32,290 
13.1 
Greater than 2 years to 5 years 
34.8  
47,263 
22.1 
Greater than 5 years to 10 years 
48.7  
60,243 
22.6 
Greater than 10 years 
58.0  
68,454 
42.3 
Overall 
38.2  
53,621 
100 
Source: GAO analysis of Small Business Administration (SBA) data. | GAO-21-589 
Note: Overall, SBA provided about $200.4 billion in EIDL loans between March 2020 and February 
2021. 
 
Loan application volume, approval rates, and average loan size varied 
widely by industry.54 SBA received the largest number of applications 
from businesses in the personal services (such as hair and nail salons 
and dry cleaning), transportation, construction and contractors, retail, and 
business services industries (see table 6).55 An early study of the 
industry-by-industry effects of COVID-19 showed that several of these 
industries—including personal services, restaurants and bars, and travel 
and transportation—were the most exposed to economic damage.56 
                                                                                                                       
54SBA’s EIDL application allowed applicants to select from a list of industry types. For 
some industries, such as retail, SBA provided a list of subcategories under a primary 
category. For other industries, such as agriculture, SBA did not provide a list of 
subcategories under the primary category. We analyzed the primary categories shown on 
SBA’s application. 
55We excluded from our analysis industries with fewer than 1,000 applications or 
applications that did not list a business activity. These applications represented less than 
1 percent of all applications.  
56Joseph S. Vavra, Shutdown Sectors Represent Large Share of All U.S. Employment 
(Chicago, IL: Becker Friedman Institute for Economics at the University of Chicago, Mar. 
27, 2020), accessed April 12, 2021,  
https://bfi.uchicago.edu/insight/finding/shutdown-sectors-represent-large-share-of-all-us-e
mployment/.  
Industry 

 
 
 
 
 
 
Page 44 
GAO-21-589  Economic Injury Disaster Loan Program 
Table 6: Top 10 Industries by Number of Economic Injury Disaster Loan (EIDL) Applicants, March 2020–February 2021 
Industry 
Number of EIDL loan applicants 
Percentage of total applicants 
Personal services 
1,373,434 
14.0 
Transportation 
952,644 
9.8 
Construction and contractors 
887,001 
9.0 
Retail 
856,545 
8.8 
Business services 
694,447 
7.1 
Miscellaneous services 
679,061 
7.0 
Real estate 
590,179 
6.0 
Health services 
551,818 
5.6 
Agriculture 
538,295 
5.5 
Entertainment services 
378,821 
3.9 
Other industries 
2,271,946 
23.3 
Total 
9,774,191 
100 
Source: GAO analysis of Small Business Administration data. | GAO-21-589 
 
Applicants in the legal services field had the highest approval rate of any 
industry (63.2 percent), followed by hotel and lodging, real estate, mining, 
and manufacturing (see table 7).57 In contrast, applicants in agriculture 
and personal services had the lowest approval rates (20.7 and 27.1 
percent, respectively). 
Table 7: Top 10 Industries by Economic Injury Disaster Loan Approval Rate, March 2020–February 2021 
Industry 
Approval rate (percentage) 
Legal services 
63.2 
Hotels and lodging 
57.2 
Real estate 
55.9 
Mining and natural resources extraction 
51.9 
Manufacturing 
50.7 
Health services 
49.9 
Religious services 
49.9 
Insurance 
49.5 
                                                                                                                       
57To improve data reliability for this analysis, we excluded about 9,900 applications, 
including applications for businesses that did not clearly fall under the industry categories 
SBA established. As a result, the overall approval rate for industries differs from the 
approval rate for the entire population of applicants.  

 
 
 
 
 
 
Page 45 
GAO-21-589  Economic Injury Disaster Loan Program 
Industry 
Approval rate (percentage) 
Automotive sales and gas stations 
48.8 
Finance 
48.0 
All industries 
38.1 
Source: GAO analysis of Small Business Administration (SBA) data. | GAO-21-589 
Note: To improve data reliability for this analysis, we excluded about 9,900 applications, including 
applications for businesses that did not clearly fall under the industry categories SBA established. As 
a result, the overall approval rate for industries differs from the approval rate for the entire population 
of applicants. 
In terms of average loan size, eating and drinking places (such as bars 
and restaurants) received the largest loans ($92,618) (see table 8). 
Businesses in personal services received the smallest average loans 
($29,323), followed by transportation ($39,172) and real estate ($40,572). 
Table 8: Top 10 Industries by Economic Injury Disaster Loan Average Loan Amount, March 2020–February 2021 
Industry 
Average loan amount ($) 
Eating and drinking places 
92,618 
Manufacturing 
85,280 
Automotive sales and gas stations 
82,177 
Wholesale trade 
76,980 
Hotels and lodging 
75,770 
Mining and natural resources extraction 
75,550 
Health services 
68,027 
Food and beverage stores 
66,314 
Storage 
65,981 
Legal services 
63,883 
All industries 
53,634 
Source: GAO analysis of Small Business Administration data. | GAO-21-589 
 
Industries with the largest shares of EIDL applicants tended to have large 
numbers of businesses with either no employees or one employee. For 
example, more than half of applicants in the personal services, 
transportation, and business services industries each had one employee 
or none. However, in each of these industries, more than half of total loan 
dollars went to larger businesses, as larger businesses received larger 
loans on average. 

 
 
 
 
 
 
Page 46 
GAO-21-589  Economic Injury Disaster Loan Program 
Our analysis of EIDL applicant and Census Bureau data found that small 
businesses located in counties with higher median household income and 
better internet access generally had better EIDL lending outcomes as 
measured by the number of loans per 1,000 businesses, amount of loans 
per employee, and approval rates.58 In addition, more diverse 
communities, as measured by the share of the county’s minority 
population, had more loans per 1,000 small businesses.59 Businesses in 
more diverse communities also had larger loan amounts on a per- 
employee basis, but lower approval rates. See appendixes II and III for 
analysis of other socioeconomic and demographic variables at the county 
level and by tabulated ZIP code areas. 
EIDL applicants in counties with median household incomes above the 
national estimate (about $63,000) had a higher approval rate than those 
located in counties with lower median household incomes (see fig. 9).60 
Applicants in these counties received larger loans on a per-employee 
basis. Additionally, counties with higher median household income had 
more loans per 1,000 small businesses. 
                                                                                                                       
58We combined SBA’s EIDL applicant data with socioeconomic, demographic, and 
geographic data at the county level to examine characteristics of the counties in which 
applicants and borrowers were located. We used data from the Census Bureau’s 2015–
2019 American Community Survey to define and compute socioeconomic and 
demographic characteristics. We used data from the U.S. Department of Agriculture’s 
(USDA) Economic Research Service to classify counties as metro or nonmetro and data 
from the Census Bureau to identify tribal lands. We focused on median household income 
and internet access as socioeconomic indicators because they may be closely related to 
EIDL eligibility and access—for example, SBA requires that business submit applications 
online only. 
59For the purposes of this analysis, we define “minorities” as the non-White share of a 
county’s population.  
60According to our analysis of data from the 2015–2019 American Community Survey, the 
national 5-year estimate for median household income in the United States based on 2019 
inflation-adjusted dollars was $62,843. 
Small Businesses in More 
Diverse and Wealthier 
Communities Received 
More Loans 
Median Household Income 

 
 
 
 
 
 
Page 47 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 9: Economic Injury Disaster Loan (EIDL) Lending Outcomes by Median 
Household Income 
 
Note: According to our analysis of data from the 2015–2019 American Community Survey, the 
national 5-year estimate for median household income in the United States based on 2019 inflation-
adjusted dollars was $62,843. 
 
Variation in median household income may have affected EIDL 
application outcomes and the distribution of loans across counties. 
Research by the Board of Governors of the Federal Reserve System 
(Federal Reserve) has found some correlation between income and credit 
scores.61 SBA used credit score as a key underwriting criterion to approve 
or decline a loan application. Thus, differences in median household 
income, and potentially credit scores, might help to explain lower approval 
rates in counties with lower median income as compared to counties with 
higher median income. Additionally, recent work by the National Bureau 
of Economic Research found that pandemic-related business closures 
resulted in larger revenue losses for small businesses located in wealthier 
ZIP codes because of larger overall declines in consumer spending by 
                                                                                                                       
61Rachel Beer, Felicia Ionescu, Geng Li, “Are Income and Credit Scores Highly 
Correlated?” FEDS Notes, accessed April 28, 2021, 
https://www.federalreserve.gov/econres/notes/feds-notes/are-income-and-credit-scores-hi
ghly-correlated-20180813.htm.  

 
 
 
 
 
 
Page 48 
GAO-21-589  Economic Injury Disaster Loan Program 
high-income individuals.62 The larger decline in spending and relatively 
higher rate of business closures in more affluent areas may have resulted 
in a greater need in those areas. This could, in part, account for 
businesses in counties with higher median income receiving more loans 
per 1,000 small businesses. 
EIDL applicants in counties with better internet access—that is, where the 
share of households with no internet access is below the national 
estimate (13.9 percent)—had better EIDL lending outcomes as compared 
to applicants in counties where that share is above the national estimate 
(see fig. 10).63 Applicants in counties with better internet access had 
higher approval rates and received larger loans on a per-employee basis. 
Additionally, these counties had more loans per 1,000 small businesses. 
                                                                                                                       
62Raj Chetty, John N. Friedman, Nathaniel Hendren, Michael Stepner, Opportunity 
Insights Team, How Did COVID-19 and Stabilization Policies Affect Spending and 
Employment? A New Real Time Economic Tracker Based on Private Sector Data 
(Cambridge, MA: National Bureau of Economic Research, 2020). The authors found that 
small business revenues fell by 50 percent for businesses located in the richest 5 percent 
of ZIP codes, compared to revenue losses of 35 percent for small businesses located in 
the poorest 5 percent of ZIP codes. Small business revenue data came from Womply, 
which records revenues from credit card transactions for small businesses (as defined by 
SBA), and income data for ZIP codes came from the Census Bureau’s 2014–2018 
American Community Survey. The analysis compared revenues during a baseline period 
prior to the COVID-19 shock (Jan. 4, 2020–Jan. 31, 2020) to the period immediately after 
the COVID-19 shock but prior to the stimulus program (Mar. 25, 2020–Apr. 14, 2020). 
63According to our analysis of data from the 2015–2019 American Community Survey, the 
national 5-year estimate of the share of households without internet access is 13.9 
percent.  
Internet Access 

 
 
 
 
 
 
Page 49 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 10: Economic Injury Disaster Loan (EIDL) Lending Outcomes by Household 
Internet Access 
 
Note: The national 5-year estimate for share of households with no internet access is 13.9 percent, 
according to our analysis of data from the 2015–2019 American Community Survey. 
 
Differences in counties’ internet connectivity may be linked to other 
socioeconomic and geographic characteristics affecting EIDL application 
outcomes. According to the 2018 Census Bureau American Community 
Survey, a larger percentage of households in urban areas reported 
owning any type of computer and having broadband subscriptions 
compared with households in rural areas across almost every region of 
the country.64 Differences in internet access across counties may have 
influenced who did and did not apply to the EIDL program because the 
EIDL application could only be accessed online and the distribution of the 
information about the program was generally limited to online forums, 
such as webinars, as a result of the pandemic. For example, 
representatives of one SBDC noted that many small businesses in their 
region were in rural areas and lacked adequate technology infrastructure 
to navigate the internet. Accordingly, our analysis found that the absolute 
number of total approved and declined applications was higher in 
counties with better internet connectivity. The number of loans per 1,000 
                                                                                                                       
64Census Bureau, Computer and Internet Use in the United States: 2018 (Washington, 
D.C.: 2021). “Any type of computer” includes at least one of the following: desktop or 
laptop, smartphone, or tablet or other portable wireless computer. A “broadband” internet 
subscription refers to having at least one type of internet subscription other than a dial-up 
subscription alone.  

 
 
 
 
 
 
Page 50 
GAO-21-589  Economic Injury Disaster Loan Program 
small businesses was also lower in counties with lower rates of internet 
access. 
As discussed earlier, counties with higher median incomes had higher 
approval rates and wealthier areas had larger pandemic-related losses. 
Wealthier areas also have better internet connectivity. Specifically, the 
Census Bureau found that 99 percent of households earning $150,000 or 
more had some sort of computing device, compared to only 77 percent of 
households making less than $25,000 per year.65 This may help to 
explain similarities in EIDL lending outcomes between communities with 
high median household incomes and high rates of internet connectivity. 
EIDL lending outcomes were similar for counties where the minority 
population made up the majority and counties where the percentage of 
minority population was greater than the national percentage of minority 
population (27.5 percent) (see fig. 11). Applicants in counties with larger 
minority populations had lower approval rates but received larger loan 
amounts per employee. Additionally, these counties received more loans 
per 1,000 small businesses than counties with smaller minority 
populations. 
                                                                                                                       
65Census Bureau, Computer and Internet Use.   
Community Diversity 

 
 
 
 
 
 
Page 51 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 11: Economic Injury Disaster Loan Program Lending Outcomes by County 
Minority Population 
 
Note: According to our analysis of data from 2015–2019 American Community Survey, 27.5 percent 
of the U.S. population identifies as non-White (or minority). 
 
Underlying correlations between demographic measures and 
socioeconomic indicators at the county level may contribute to differences 
in EIDL application outcomes across counties. According to the Census 
Bureau’s 2015–2019 American Community Survey, a county’s median 
household income was negatively correlated with its share of minority 
residents. The American Community Survey also found in 2018 that that 
median income among Black and Hispanic Americans was lower than for 

 
 
 
 
 
 
Page 52 
GAO-21-589  Economic Injury Disaster Loan Program 
White and Asian Americans.66 As stated earlier, SBA used credit score as 
a key underwriting criterion to approve or decline loan applications, and 
some correlation exists between income and credit score. According to 
the Federal Reserve’s 2018 Small Business Credit Survey, Black and 
Hispanic small business owners reported lower credit scores on average 
compared to White and Asian small business owners.67 Lower median 
income and lower credit scores may help explain why approval rates for 
businesses were lower in counties with higher minority shares, especially 
counties with larger Black and Hispanic populations. 
Differences in access to financing that exist among different 
demographics may also contribute to variations in EIDL lending 
outcomes. In the 2020 Small Business Credit Survey, the Federal 
Reserve found that, compared to White-owned small businesses, a larger 
percentage of Black, Hispanic, and Asian-owned businesses self-reported 
being in poor financial condition, and a smaller percentage reported 
receiving the full amount of funding sought from banks and other sources 
of credit.68 Prior to the pandemic, Black and Hispanic small business 
owners were also more likely to apply for SBA financing.69 Taken 
together, these factors may have increased interest in the EIDL program 
in these counties and driven higher levels of program activity, as our 
analysis found that counties with larger minority populations had more 
total approved and declined EIDL applications than those with smaller 
minority populations. 
                                                                                                                       
66Census Bureau, Household Income: 2018 (Washington, D.C.: 2019).  
67Federal Reserve Bank of Atlanta, Small Business Credit Survey: Report on Minority-
Owned Firms (Atlanta, GA: 2019). 
68Board of Governors of the Federal Reserve System, Small Business Credit Survey: 
2021 Report on Employer Firms (2021). The 2020 Small Business Credit Survey collected 
responses from over 9,600 small businesses throughout the United States. While the 
survey respondents are not drawn from a random sample, survey results are weighted by 
firm age, industry, number of employees, and geography to ensure that results are 
nationally representative. The Federal Reserve conducted the survey in September and 
October 2020, approximately 6 months after the onset of the pandemic.   
69Federal Reserve Bank of Atlanta, Small Business Credit Survey: Report on Minority-
Owned Firms. The 2018 Small Business Credit Survey collected responses from over 
6,600 small businesses throughout the United States. While the survey respondents are 
not drawn from a random sample, survey results are weighted by firm age, industry, 
number of employees, and geography to ensure that results are nationally representative. 

 
 
 
 
 
 
Page 53 
GAO-21-589  Economic Injury Disaster Loan Program 
Our analysis of EIDL applicant, Census Bureau, and other data found that 
EIDL program lending outcomes varied by geographic location. We 
examined lending outcomes for urban and rural counties and tribal lands, 
as well as nationwide trends across geographic regions. 
EIDL lending outcomes were mixed for urban and rural counties (see fig. 
12).70 Applicants in urban counties had lower approval rates than 
applicants in rural counties, but received larger loans on a per-employee 
basis. Additionally, urban counties received more loans per 1,000 small 
businesses as compared to rural counties. 
Figure 12: Economic Injury Disaster Loan Program Lending Outcomes for Urban 
and Rural Counties  
 
 
Underlying correlations between a county’s urban or rural classification 
and its demographic and socioeconomic characteristics may contribute to 
differences in EIDL lending outcomes across counties. For example, 
minority-owned small businesses tend to be more highly concentrated in 
                                                                                                                       
70USDA’s Economic Research Service designated each U.S. county as metro or 
nonmetro. Metro areas include all counties with urban areas containing 50,000 people or 
more. Metro areas also include outlying counties that are economically tied to the central 
counties, as measured by the share of workers commuting on a daily basis to the central 
counties. Nonmetro counties are outside the boundaries of metro areas and have no cities 
with 50,000 residents or more. For the purposes of this report, rural counties are 
“nonmetro” counties and urban counties are “metro” counties designated by USDA in 
2017. 
EIDL Lending Outcomes 
Varied by Geography 
Urban and Rural Areas 

 
 
 
 
 
 
Page 54 
GAO-21-589  Economic Injury Disaster Loan Program 
urban areas relative to White-owned businesses.71 This may explain 
similar EIDL lending outcomes for urban counties and counties with larger 
shares of minority populations. Differences in rural or urban 
characteristics may also contribute to differences in approval rates. 
According to the 2016 Small Business Credit Survey, small businesses in 
rural areas self-reported higher credit scores on average compared to 
businesses in urban areas.72 As discussed previously, credit scores might 
help to explain differences in approval rates across counties, as reflected 
in higher approval rates for businesses located in rural areas. 
Small businesses located on tribal lands had lower approval rates and 
smaller average EIDL loans as compared to all U.S. counties (see fig. 
13).73 SBA approved 25,606 EIDL loans to small businesses on tribal 
lands, which represented less than 1 percent of all EIDL lending. 
                                                                                                                       
71Federal Reserve Bank of Atlanta, Small Business Credit Survey: Report on Minority-
Owned Firms.  
72Federal Reserve Bank of Richmond, Small Business Credit Survey: Report on Rural 
Employer Firms (Richmond, VA: 2017). The 2016 Small Business Credit Survey collected 
responses from over 10,300 small businesses throughout the United States. While the 
survey respondents are not drawn from a random sample, survey results are weighted by 
firm age, industry, number of employees, and geography to ensure that results are 
nationally representative. 
73We analyzed information for loans to businesses located on tribal lands, regardless of 
whether the businesses themselves were Indian-owned. Tribal lands we analyzed 
included federally recognized American Indian reservations and off-reservation trust land 
areas, state-recognized American Indian reservations, and Hawaiian home lands for 
which the Census Bureau publishes data. We were unable to scale EIDL lending 
outcomes for tribal areas by the loan amount per employee or number of loans per 1,000 
small businesses because data on the number of businesses in tribal areas were 
unavailable.  
Tribal Areas 

 
 
 
 
 
 
Page 55 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 13: Economic Injury Disaster Loan Program Lending Outcomes for Tribal 
Lands 
 
 
Underlying correlations between demographic measures and 
socioeconomic indicators contribute to differences in EIDL lending 
outcomes for businesses located on tribal lands. For example, Census 
Bureau data have shown median income among American Indians and 
Alaska Natives to be lower than for White and Asian Americans, which 
may have contributed to lower approval rates for businesses on tribal 
lands. Additionally, several factors can make successful business 
development on tribal lands more difficult, including limited access to 
capital and a lack of technology infrastructure (e.g., telecommunications 
lines and internet access). 
We found that counties within four geographic regions of the United 
States tended to have similar EIDL lending outcomes for certain 
measures.74 
Of the 1,229 counties with more than the average number of EIDL loans 
per 1,000 small businesses, 726 were located in the Census Bureau’s 
                                                                                                                       
74For this analysis, we used the Census Bureau’s regional definitions. The Census Bureau 
divides the United States into four geographic regions: Midwest, Northeast, South, and 
West. Counties differ in size by land area across both states and regions and, thus, 
reported results should not necessarily be interpreted as being proportionately, or 
disproportionately, allocated across regions 
Nationwide Trends in Lending 
Outcomes 

 
 
 
 
 
 
Page 56 
GAO-21-589  Economic Injury Disaster Loan Program 
South region (see fig. 14).75 The Northeast region had the highest 
percentage of counties with more than the average number of loans per 
1,000 small businesses (58.1 percent), while the Midwest region had the 
smallest percentage of such counties (12.6 percent).76 
                                                                                                                       
75The Census Bureau’s South region includes the following states: Delaware, Florida, 
Georgia, Maryland, North Carolina, South Carolina, Virginia, West Virginia, Alabama, 
Kentucky, Mississippi, Tennessee, Arkansas, Louisiana, Oklahoma, and Texas. The 
region also includes Washington, D.C.  
76The Census Bureau’s Northeast region includes the following states: Connecticut, 
Maine, Vermont, Massachusetts, New Hampshire, Rhode Island, New York, New Jersey, 
and Pennsylvania. The Midwest Region includes the following states: Indiana, Illinois, 
Michigan, Ohio, Wisconsin, Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, 
and South Dakota.  

 
 
 
 
 
 
Page 57 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 14: Approved Economic Injury Disaster Loans (EIDL) Per 1,000 Small Businesses by County, March 2020–February 
2021 
 
 
Similarly, of the 1,174 counties with above-average loan amounts per 
employee, 718 were located in the Census Bureau’s South region (see 
fig. 15). The West region had the highest percentage of counties with 
above-average loan amounts per employee (56.3 percent), while the 

 
 
 
 
 
 
Page 58 
GAO-21-589  Economic Injury Disaster Loan Program 
Midwest region had the lowest percentage of such counties (11.2 
percent). 
Figure 15: Economic Injury Disaster Loan (EIDL) Amount per Small Business Employee by County, March 2020–February 
2021 
 
 
Of the 1,569 counties with above-average approval rates, 656 were 
located in the Census Bureau’s Midwest region (see fig. 16). The West 

 
 
 
 
 
 
Page 59 
GAO-21-589  Economic Injury Disaster Loan Program 
region had the highest percentage of counties with above-average 
approval rates (82.3 percent), while the South region had the lowest 
number of such counties (26 percent). 
Figure 16: Approval Rate for Economic Injury Disaster Loans (EIDL) by County, March 2020–February 2021 
 
 
We also found that the West region had the highest share of counties that 
had better EIDL lending outcomes in all three measures. Of the 383 

 
 
 
 
 
 
Page 60 
GAO-21-589  Economic Injury Disaster Loan Program 
counties with all three EIDL lending outcome measures (loans per 1,000 
small businesses, loan amounts per employee, and approval rates) above 
the national average, the West region had the highest number of these 
counties (170 counties), while the Midwest region had the lowest number 
(42 counties).77 
In response to the COVID-19 pandemic, SBA moved quickly under 
challenging circumstances to provide EIDL funding to applicants, 
benefiting many small businesses. However, SBA did not communicate 
with potential and actual applicants key information—such as processing 
times and loan limits—and their loan status in an effective, consistent, or 
timely manner. SBA has two plans to guide the agency’s communication 
efforts about loan programs, but these plans do not cover what 
information program participants, including applicants, should receive and 
when. Developing a comprehensive strategy for communicating with 
potential and actual applicants in the event of a disaster would benefit 
both SBA and its program participants. For example, by providing clear 
information on the time frames and status of applications, SBA could 
reduce applicant confusion and uncertainty. In turn, this could lessen the 
burden on SBA’s resources created by applicants submitting duplicate 
applications and contacting SBA’s Customer Service Center. 
In addition, we and others have found the EIDL program to be susceptible 
to the risk of fraud and provision of funding to ineligible applicants, and 
have made a number of recommendations to address these risks, such 
as to conduct a fraud risk assessment for the EIDL program. SBA has not 
fully implemented our recommendations, but officials stated the agency 
has begun the process of conducting a fraud risk assessment of the EIDL 
program. Fully implementing these recommendations would help SBA to 
safeguard billions of dollars of taxpayer funds and improve the operation 
of the EIDL program. 
The Associate Administrator of SBA’s Office of Disaster Assistance 
should develop a comprehensive strategy for communicating with 
potential and actual program applicants in the event of a disaster. Such a 
strategy should provide guidelines for the types of information and timing 
of information to be provided to program participants throughout a 
disaster. The types of information to be addressed in the strategy could 
                                                                                                                       
77The Census Bureau’s West region includes the following states: Arizona, Colorado, 
Idaho, New Mexico, Montana, Utah, Nevada, Wyoming, Alaska, California, Hawaii, 
Oregon, and Washington.  
Conclusions 
Recommendation for 
Executive Action 

 
 
 
 
 
 
Page 61 
GAO-21-589  Economic Injury Disaster Loan Program 
include processing steps and time frames applicants might experience 
through different stages of the loan process. (Recommendation 1) 
We provided a draft of this report to SBA for review and comment. In 
written comments, reproduced in appendix IV, SBA stated that it agreed 
with the report’s recommendation and plans to develop a comprehensive 
strategy for communicating with potential and actual disaster loan 
applicants, which will include information such as processing steps and 
corresponding time frames applicants might experience through different 
stages of the loan process. SBA also stated that it is moving operation of 
the EIDL program related to COVID-19 from the Office of Disaster 
Assistance to the Office of Capital Access in order to better serve small 
businesses.  
SBA also provided technical comments, which we incorporated as 
appropriate. 
We are sending copies of this report to appropriate congressional 
committees, the SBA Administrator, and other interested parties. 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 
William B. Shear at (202) 512-8678 or ShearW@gao.gov. Contact points 
for our Offices of Congressional Relations and Public Affairs may be 
found on the last page of this report. Major contributors to this report are 
listed in appendix V. 
 
William B. Shear 
Director, Financial Markets and Community Investment 
 
 
Agency Comments 
and Our Evaluation 

 
 
 
 
 
 
Page 62 
GAO-21-589  Economic Injury Disaster Loan Program 
List of 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 
Chair 
The Honorable Rob Portman 
Ranking Member 
Committee on Homeland Security and Governmental Affairs 
United States Senate 
The Honorable Benjamin Cardin 
Chairman 
The Honorable Rand Paul 
Ranking Member 
Committee on Small Business and Entrepreneurship 
United States Senate 
The Honorable Rosa L. DeLauro 
Chairwoman 
The Honorable Kay Granger 
Ranking Member 
Committee on Appropriations 
House of Representatives 

 
 
 
 
 
 
Page 63 
GAO-21-589  Economic Injury Disaster Loan Program 
The Honorable Frank Pallone, Jr. 
Chair 
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 
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 Nydia Velázquez 
Chairwoman 
The Honorable Blaine Luetkemeyer 
Ranking Member 
Committee on Small Business 
House of Representatives 
The Honorable Richard Neal 
Chair 
The Honorable Kevin Brady 
Republican Leader 
Committee on Ways and Means 
House of Representatives 
The Honorable Elizabeth Warren 
United States Senate 
The Honorable Ayanna Pressley 
House of Representatives 

 
Appendix I: Objectives, Scope, and 
Methodology 
 
 
 
 
Page 64 
GAO-21-589  Economic Injury Disaster Loan Program 
For this report, we examined (1) challenges the Small Business 
Administration (SBA) and Economic Injury Disaster Loan (EIDL) program 
applicants experienced as part of SBA’s implementation of the EIDL 
program in response to Coronavirus Disease 2019 (COVID-19) and the 
extent to which SBA has addressed these challenges; (2) the EIDL 
program’s effects on its participants, including the effects of loan limits 
SBA put in place; (3) steps SBA has taken to address risks of fraud and 
provision of funds to ineligible applicants; and (4) the characteristics of 
program applicants and recipients. The EIDL loan program existed prior 
to the pandemic, but Congress has made legislative changes to modify 
the program in response to COVID-19, including the addition of EIDL 
advances. This report focuses on SBA’s implementation of EIDL since the 
start of the pandemic. 
To address the first objective, we reviewed relevant legislation and SBA’s 
guidance, policies and procedures, training documents, contracts with 
vendors, and other material for the EIDL program in response to COVID-
19. We also interviewed officials from SBA’s Office of Disaster Assistance 
about the implementation of the program. In addition, we interviewed 
representatives from a nongeneralizable sample of six state or regional 
Small Business Development Centers (SBDC) located in California, 
Illinois, Georgia, New York, Oklahoma, and Puerto Rico about their 
clients’ experiences with the EIDL program. We selected these SBDCs 
because each of them met at least two of our several criteria: (1) the state 
or region in which it is located experienced a large negative impact from 
the pandemic based on the Census Bureau’s Small Business Pulse 
Survey; (2) it had a high share of small businesses receiving EIDL 
funding based on the same survey; (3) a large difference existed between 
the percentage of applicants that requested versus received EIDL funding 
based on the survey; and (4) it had a high share of minority-owned 
businesses.1 We also selected SBDCs that are located in different 
geographic regions of the country. We also interviewed America’s 
SBDCs, an industry association representing SBDCs. 
We also interviewed representatives of two business associations 
representing small businesses nationally (the National Federation of 
Independent Businesses and National Small Business Association) and 
four business associations representing industries in specific sectors 
                                                                                                                       
1The Small Business Pulse Survey measures the effect of changing business conditions 
during the pandemic on small businesses, including whether small businesses requested 
or received EIDL funding. The survey asked small businesses the magnitude of the 
effect—positive or negative—of the pandemic.  
Appendix I: Objectives, Scope, and 
Methodology 

 
Appendix I: Objectives, Scope, and 
Methodology 
 
 
 
 
Page 65 
GAO-21-589  Economic Injury Disaster Loan Program 
(Americans for Arts, American Hotel and Lodging Association, National 
Association of Realtors, and the National Retail Federation). We identified 
these associations through internet searches. We selected the industry-
specific associations because the industries they represent experienced a 
larger-than-average negative effect from the pandemic, according to the 
Census Bureau Small Business Survey, and had large numbers of EIDL 
borrowers (based on our review of SBA loan data). 
Collectively, we refer to representatives from SBDCs and associations we 
interviewed as stakeholders. We used “few,” “some,” “a majority,” and 
“almost all” to characterize stakeholder responses. We define “few” as 1–
3 responses, “several” as 4–6 responses, “a majority” as 7–10 responses, 
and “almost all” as 11–13 responses. We conducted these interviews 
between August 2020 and February 2021. 
We also conducted five discussion groups with EIDL program applicants 
and asked them about their experience with the program. Two groups 
consisted of applicants whose loans were declined, and three groups 
consisted of applicants whose loans were approved. Each group 
consisted of four to six participants, and we spoke with 22 EIDL 
applicants in total. We recruited these participants through the National 
Small Business Association, which emailed its members on our behalf. 
Those association members interested in participating responded using a 
web form. We conducted the discussion groups in January 2021. 
Discussions were guided by a GAO moderator who used a standardized 
list of questions to encourage participants to share their thoughts about 
and experiences with the EIDL program. 
We spoke to these participants using a video conferencing tool. Each 
group discussion was digitally recorded and transcribed by an outside 
vendor, and we used the transcripts to summarize participants’ 
responses. An initial coder provided an explanation of the types of 
statements that should be assigned to a particular code and assigned a 
code that best summarized the statements from discussion group 
participants. A separate individual reviewed and verified the accuracy of 
the initial coding. The initial coder and reviewer discussed orally and in 
writing any disagreements about code assignments and documented 
consensus on the final results. Discussion groups are intended to 
generate in-depth information about the reasons for participants’ views on 
specific topics. While we report our findings by the number of discussion 
groups in which a topic was discussed, this does not necessarily mean 
that there was a consensus or agreement among all discussion group 
participants on a given topic. The opinions expressed by the participants 

 
Appendix I: Objectives, Scope, and 
Methodology 
 
 
 
 
Page 66 
GAO-21-589  Economic Injury Disaster Loan Program 
represent their points of view and may not represent the views of all EIDL 
applicants. 
To identify trends in processing EIDL loans and advances, such as 
processing times and volume, we reviewed SBA’s weekly summary 
processing reports for loans and advances. For loans, we reviewed 
reports dating from March 15, 2020, through February 27, 2021. For 
advances, we reviewed reports dating March 29, 2020, through July 15, 
2020. We interviewed SBA officials in order to assess the reliability of the 
data contained within those reports and determined that they were 
reliable for the purpose of reporting on SBA’s loan processing volumes 
and times. 
To further analyze trends in processing times, we obtained and analyzed 
SBA data for individual EIDL loan applications. These data were derived 
from about 12.5 million applications submitted to SBA between March 14, 
2020, and February 28, 2021. To assess the reliability of these data, we 
conducted electronic testing for missing data, outliers, and obvious errors, 
and we interviewed SBA officials. We determined that the data were 
sufficiently reliable for the purpose of describing application volumes and 
processing times. Within the application data, each application was 
assigned a processing status depending on the processing stage the 
application was in at the time data were produced. Based on information 
from SBA, we categorized applications with statuses listed as “funded,” 
“funding,” “obligated,” and “obligating” as applications that had been 
“approved.”2 
To analyze average processing times by the month that applicants 
submitted their applications, we analyzed the applications that were 
“funded” or declined. We used “funded” rather than the entirety of the 
“approved” category to provide a consistent measurement for the 
processing time. Using these data, we approximated the processing time 
from the time applicants submitted their applications to the time SBA 
funded or declined their applications. 
To identify trends in customer service calls received, we analyzed SBA’s 
weekly data for its Customer Service Center from the week of March 15, 
                                                                                                                       
2Applications in these stages have been approved by SBA. Specifically, “obligating” 
indicates that SBA has made a final approval for the application; “obligated” means that 
the borrower may sign the closing documents; “funding” means that borrowers have 
signed closing documents; and “funded” means the funds have been disbursed to the 
borrower.  

 
Appendix I: Objectives, Scope, and 
Methodology 
 
 
 
 
Page 67 
GAO-21-589  Economic Injury Disaster Loan Program 
2020, through the week of February 14, 2021. The calls from this period 
were primarily related to COVID-19. SBA distinguishes COVID-19 calls 
from non-COVID-19 calls using certain coding, but COVID-19-related 
calls are not always coded as such, according to SBA. To identify trends 
in staffing, we analyzed SBA’s weekly staffing data from March 20, 2020, 
through February 12, 2021. To assess the reliability of SBA’s call center 
and staffing data, we interviewed SBA officials. We determined that these 
data were sufficiently reliable for the purpose of identifying trends in 
customer service calls and staffing levels. 
To address the second objective, we reviewed information from SBA 
about when it implemented the loan limits, and we interviewed officials 
from SBA’s Office of Disaster Assistance about the loan limits put in 
place. Additionally, we obtained information from SBA about the number 
of borrowers affected by SBA’s loan limits. We asked discussion group 
participants and representatives of SBDCs and business associations 
identified above about the effects of the program and loan limits. 
To address the third objective, we reviewed prior GAO reports and 
reports from SBA’s Office of Inspector General and independent financial 
statement auditor.3 We interviewed SBA officials from the Office of 
Disaster Assistance and SBA’s contractor about controls in place to 
ensure eligibility and to address fraud risk. We also obtained written 
responses to questions we submitted to SBA’s contractor and two 
subcontractors regarding services they provided in implementing EIDL 
and their interactions with SBA, among other things. Additionally, we 
reviewed documentation on SBA’s policies, procedures, and training 
related to reviewing EIDL applications. We also reviewed documents 
related to EIDL program eligibility criteria and modifications made to the 
contractors’ validation system. 
To characterize fraud cases and schemes used to obtain EIDL funds, we 
analyzed 51 fraud-related cases based on information provided in 
                                                                                                                       
3See GAO, COVID-19: Sustained Federal Action Is Crucial as Pandemic Enters Its 
Second Year, GAO-21-387 (Washington, D.C.: Mar. 31, 2021) and 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); Small 
Business Administration, Office of Inspector General, SBA’s Handling of Identity Theft in 
the COVID-19 Economic Injury Disaster Loan Program (Washington, D.C.: May 6, 2021) 
and Inspection of Small Business Administration’s Initial Disaster Assistance Response to 
the Coronavirus Pandemic (Washington, D.C.: Oct. 28, 2020); and Small Business 
Administration, Office of Performance Management and the Chief Financial Officer, 
Agency Financial Report, Fiscal Year 2020 (Washington, D.C.: Dec. 18, 2020). 

 
Appendix I: Objectives, Scope, and 
Methodology 
 
 
 
 
Page 68 
GAO-21-589  Economic Injury Disaster Loan Program 
Department of Justice (DOJ) and federal court documents from May 2020 
to March 2021. We identified the 51 cases included in our analysis by 
subscribing to alerts from a legal news subscription site, using the key 
phrase “Economic Injury Disaster Loan.” For identified cases, we used an 
electronic public access service for U.S. federal court documents to 
access and download documents used in the court process, such as 
indictments, criminal information, and plea agreements. We conducted a 
thematic analysis of case information, using a federal fraud ontology 
currently under development at GAO as a schema. This thematic analysis 
was structured and organized using WebProtégé, an ontology modeling 
tool developed by the Stanford Center for Biomedical Informatics 
Research at the Stanford University School of Medicine. We then 
analyzed the aggregate data to describe the characteristics of EIDL-
related DOJ cases. To identify illustrative cases of potential fraud and 
abuse involving EIDL funds, we selected from a list of cases that had 
reached conclusion through guilty plea. We selected cases to achieve 
variation across the following criteria: whether the cases involved (1) 
eligibility misrepresentation, (2) fraud schemes identified by law 
enforcement, and (3) fraud related to COVID-19 or other crimes. 
To address our fourth objective, we (1) analyzed SBA data on EIDL loan 
applicants to determine applicants’ demographic and other 
characteristics; (2) combined these applicant data with Census Bureau 
and U.S. Department of Agriculture (USDA) data at the county level to 
examine the socioeconomic, demographic, and geographic 
characteristics of the counties in which applicants and borrowers were 
located; and (3) combined SBA’s EIDL applicant data with Census 
Bureau data at the tabulated ZIP code level to examine the 
socioeconomic and demographic characteristics of tabulated ZIP code 
areas in which applicants and borrowers were located for four 
metropolitan statistical areas (MSA).4 
We analyzed the characteristics of EIDL applicants using SBA’s data on 
EIDL applications from March 14, 2020, through February 28, 2021, 
                                                                                                                       
4The Census Bureau defines metropolitan statistical areas as having at least one 
urbanized area of 50,000 or more population, plus adjacent territory that has a high 
degree of social and economic integration with the core as measured by commuting ties. 
ZIP code tabulation areas are generalized areal representations of U.S. Postal Service 
ZIP code service areas, which are not areal features but a collection of mail delivery 
routes.  

 
Appendix I: Objectives, Scope, and 
Methodology 
 
 
 
 
Page 69 
GAO-21-589  Economic Injury Disaster Loan Program 
discussed above. The data covered about 12.5 million applications.5 We 
analyzed the number of applicants, the loan approval rate, and the loan 
amount by the size of a business (as measured by the number of 
employees reported by applicants), the age of the business, and the 
business industry reported by the applicant.6 Although we determined that 
these data were reliable for our use in describing trends in applicant and 
loan recipient characteristics, SBA collected this information from 
applicants who self-certified the accuracy of the information provided. 
Therefore, the reliability of the data is dependent on the accuracy of the 
information provided by the applicants. 
We combined SBA’s EIDL applicant data with socioeconomic, 
demographic, and geographic data at the county level to examine 
characteristics of the counties in which applicants and borrowers were 
located.7 We used data from the Census Bureau’s 2015–2019 American 
Community Survey to define and compute socioeconomic and 
demographic characteristics. Specifically, for socioeconomic 
characteristics, we examined median household income, unemployment 
rate, poverty rate, percentage of households receiving public assistance 
income, and percentage of households without internet access. For 
demographic characteristics, we examined the percentage of non-White 
residents, the percentage of foreign-born residents, and the percentage of 
households with limited English proficiency. For geographic 
characteristics, we used 2017 data from USDA’s Economic Research 
                                                                                                                       
5The data consisted of about 3.7 million approved applications, 6 million declined 
applications, 2.6 million withdrawn applications, and 123,000 other applications, such as 
applications SBA is currently processing. 
6The SBA data included data on business activity, number of employees, the date the 
business was established, whether the loan was approved, and the loan amount. We 
created additional variables to facilitate our analysis of the data. Specifically, we created 
the variable for the age of business by calculating the length of time between the date that 
applicants reported they established their business and January 31, 2020. We chose to 
use January 31, 2020, because applicants were generally eligible for EIDL in response to 
COVID-19 if they established businesses on or before this date. For industry, SBA’s data 
generally consisted of business activities that applicants could select from a list on the 
application. However, some business activities did not align with those that appear on the 
application. Where possible, we grouped business activities that did not align with those 
that appear on the application. 
7For the purposes of this analysis, we defined “community” as a U.S. county and a small 
business as a business with fewer than 500 employees.  

 
Appendix I: Objectives, Scope, and 
Methodology 
 
 
 
 
Page 70 
GAO-21-589  Economic Injury Disaster Loan Program 
Service to classify counties as rural or urban.8 Additionally, we used data 
from the Census Bureau to identify tribal lands.9 
In our analysis, we examined the relationship between EIDL program 
outcomes and county-level socioeconomic, demographic, and geographic 
characteristics. Specifically, we analyzed the total number of EIDL 
applications, the total number of approved EIDL applications, and the 
total number of declined EIDL applications. We also analyzed the loan 
approval rate and the total and average loan amounts for approved 
applicants.10 Additionally, we used data from two Census Bureau 
reports—the Statistics of U.S. Businesses and Nonemployer Statistics—
to determine the number of small business establishments by county and 
the number of small business employees by county. We used these data 
to compute the number of approved EIDL loans per small business and 
the loan amount approved on a per-employee basis.11 We reviewed 
documentation related to the data from the Census Bureau and USDA 
and determined that the data were sufficiently reliable for our use in 
describing the characteristics of the communities in which EIDL 
applicants and loan recipients were located. 
We combined SBA’s EIDL applicant data with Census Bureau data at the 
ZIP code tabulation area (ZCTA) level to examine the socioeconomic and 
demographic characteristics discussed above for ZCTAs within which 
applicants and borrowers were located for four MSAs. For each MSA, we 
                                                                                                                       
8USDA’s Economic Research Service designates each U.S. county as metro or nonmetro. 
Metro areas include all counties containing one or more urbanized areas: high-density 
urban areas containing 50,000 people or more. Metro areas also include outlying counties 
that are economically tied to the central counties, as measured by the share of workers 
commuting on a daily basis to the central counties. Nonmetro counties are outside the 
boundaries of metro areas and have no cities with 50,000 residents or more. For the 
purposes of this report, rural counties are “nonmetro” counties and urban counties are 
“metro” counties designated by USDA in 2017.  
9We used data from the Census Bureau to define tribal areas, which consist of American 
Indian, Alaska Native, and Native Hawaiian areas. These areas contain both legal and 
statistical American Indian, Alaska Native, and Native Hawaiian entities for which the 
Census Bureau publishes data. The legal entities consist of federally recognized American 
Indian reservations and off-reservation trust land areas, state-recognized American Indian 
reservations, and Hawaiian home lands. American Indian tribal subdivisions and Alaska 
Native Regional Corporations are additional types of legal entities. 
10We calculated the loan approval rate using the data for approved and declined 
applicants.  
11We calculated loans per small business by adjusting for the number of approved, 
nonagricultural EIDL loans in a particular county by the total number of small businesses 
in a county to facilitate comparison across geographic areas.  

 
Appendix I: Objectives, Scope, and 
Methodology 
 
 
 
 
Page 71 
GAO-21-589  Economic Injury Disaster Loan Program 
compared the socioeconomic and demographic characteristics of the 
ZCTAs to the national estimates. To select MSAs for analysis, we 
established population densities for each MSA and grouped the MSAs 
into four size categories: mega, large, medium, and small. From these, 
we judgmentally selected four MSAs for further analysis at the ZCTA level 
using information about (1) population; (2) percentage of businesses that 
requested assistance based on Census Bureau data (a proxy of level of 
need for the EIDL program); (3) the difference between the percentage of 
businesses that requested assistance versus those that received 
assistance based on Census Bureau data; and (4) distribution across the 
Census Bureau’s four geographic regions. We conducted analysis for 
New York-Newark-Jersey City (mega), Chicago-Elgin-Naperville (large), 
San Jose-Sunnyvale-Santa Clara (medium), and New Orleans-Metairie 
(small). 
Aggregate business statistics from the Census Bureau are unavailable at 
the ZCTA level. Therefore, we are unable to scale EIDL lending outcomes 
by the number of EIDL loans per 1,000 small businesses and the loan 
amount approved on a per-employee basis. We therefore limited our 
analysis to loan approval rate for ZCTAs. Additionally, we excluded the 
percentage of foreign-born residents from our analysis because no 
estimate for the ZCTAs is above the national average. 
To provide context for our observations from the data analysis, we 
conducted a narrow, limited search for reports and studies related to 
certain socioeconomic and demographic variables (median household 
income, household internet access, demographic trends, and urban or 
rural location). We focused on reports and studies that contained 
information about interactions among those variables, the EIDL 
application process, and COVID-19. We identified these reports and 
studies using a Google search. We restricted our search to scholarly and 
government material published within the last 5 years. 
We conducted this performance audit from July 2020 to July 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 that 
the evidence obtained provides a reasonable basis for our findings and 
conclusions based on our audit objectives. 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 72 
GAO-21-589  Economic Injury Disaster Loan Program 
This appendix contains information on Economic Injury Disaster Loan 
(EIDL) program lending outcomes by unemployment rate, share of 
households receiving public assistance income, poverty rate, foreign-born 
population, and share of households with limited English proficiency. 
Businesses in counties with unemployment rates below or equal to the 
national unemployment rate received fewer loans on average and a 
smaller overall share of all EIDL dollars (see table 9).1 However, the 
average loan size was larger as compared with loans received by 
businesses in counties with unemployment rates above the national rate. 
 
Table 9: Economic Injury Disaster Loan (EIDL) Applications, Loan Size, and Share of Total Loan Dollars, by County 
Unemployment Rate 
Variable 
Counties above 
national 
unemployment rate 
Counties below or equal to 
national unemployment rate 
All U.S. counties 
Average number of EIDL applications 
3,954 
2,150 
2,927  
Average number of declined EIDL applications 
2,524 
1,222 
1,782  
Average number of approved EIDL applications 
1,431 
928 
1,144  
Average loan size (in dollars) 
50,710 
56,721 
53,487  
Percentage of all approved EIDL loan dollars  
51.0  
49.0 
100  
Source: GAO analysis of Small Business Administration (SBA) and Census Bureau data. | GAO-21-589 
Note: The total number of counties or similar entities that can be matched to EIDL loan data is 3,139. 
According to the Census Bureau, 1,351 counties had rates of unemployment above the national rate, 
and 1,788 counties had rates below the national rate. 
 
Figure 17 illustrates EIDL lending outcomes associated with county 
unemployment rate, by quartile. Counties falling into the lower 
unemployment quartiles had higher loan approval rates and larger 
average loan sizes than other counties, although they also had fewer 
loans per 1,000 small businesses and smaller loans on a per-employee 
basis. 
 
 
                                                                                                                       
1The national 5-year estimate of unemployment rate was 5.3 percent, according to our 
analysis of data from the 2015–2019 American Community Survey.  
Appendix II: Socioeconomic and 
Demographic Characteristics of EIDL 
Applicants and Recipients 
EIDL Applicant and 
Recipient Outcomes 
by County 
Unemployment Rate 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 73 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 17: Economic Injury Disaster Loan Program Lending Outcomes by County Unemployment Rate 
 
Note: We ordered the unemployment rates for all counties from the lowest to the highest value. We 
then divided the ordered distribution of counties into four equally sized groups (quartiles). The 
counties with the lowest rates fall within the “<25th” quartile while those with the highest rates fall 
within the “>75th” quartile. We calculated the average for each lending outcome by quartile. 
 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 74 
GAO-21-589  Economic Injury Disaster Loan Program 
Businesses in counties with shares of residents receiving public 
assistance income at a level below or equal to the national estimated 
share received fewer EIDL loans on average, but they received a larger 
overall share of loan dollars (see table 10).2 The average dollar amount of 
these loans was slightly smaller than that of loans received by businesses 
in counties with larger shares of the population receiving public 
assistance income. 
 
Table 10: Economic Injury Disaster Loan (EIDL) Applications, Loan Size, and Share of Total Loan Dollars, by County Share of 
Households Receiving Public Assistance Income 
Variable 
Counties above 
national rate 
Counties below or equal to 
national rate 
All U.S. counties 
Average number of EIDL applications 
3,533 
2,620 
2,927 
Average number of declined EIDL applications 
2,144 
1,600 
1,782 
Average number of approved EIDL applications 
1,389 
1,020 
1,144  
Average loan size (in dollars) 
53,785 
53,281 
53,487  
Percentage of all approved EIDL loan dollars  
41.0  
59.0  
100  
Source: GAO analysis of Small Business Administration (SBA) and Census Bureau data. | GAO-21-589 
Note: The total number of counties or similar entities that can be matched to EIDL loan data is 3,139. 
According to the Census Bureau, 1,055 counties had shares of residents receiving public assistance 
income above the national rate, and 2,084 counties had rates below the national rate. 
 
Figure 18 illustrates EIDL lending outcomes associated with county share 
of households receiving public assistance income, by quartile. All 
measures were generally comparable across quartiles. 
                                                                                                                       
2The national 5-year estimate of the share of households receiving public assistance 
income was 2.4 percent, according to our analysis of data from the 2015–2019 American 
Community Survey.  
EIDL Applicant and 
Recipient Outcomes 
by County Share of 
Households 
Receiving Public 
Assistance Income 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 75 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 18: Economic Injury Disaster Loan Program Lending Outcomes by County Share of Households Receiving Public 
Assistance Income 
 
Note: We ordered the shares of households receiving public assistance income for all counties from 
the lowest to the highest value. We then divided the ordered distribution of counties into four equally 
sized groups (quartiles). The counties with the lowest share fall within the “<25th” quartile while those 
with the highest share fall within the “>75th” quartile. We calculated the average for each lending 
outcome by quartile. 
 
 
 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 76 
GAO-21-589  Economic Injury Disaster Loan Program 
Businesses in counties with poverty rates below or equal to the national 
rate received more and larger loans on average, as well as a larger share 
of all EIDL dollars (see table 11).3 
 
 
Table 11: Economic Injury Disaster Loan (EIDL) Applications, Loan Size, and Share of Total Loan Dollars, by County Poverty 
Rate 
Variable 
Counties above 
national rate 
Counties below or equal to 
national rate 
All U.S. counties 
Average number of EIDL applications 
2,799 
3,095 
2,927  
Average number of declined EIDL applications 
1,806 
1,752 
1,782  
Average number of approved EIDL applications 
994 
1,343 
1,144  
Average loan size (in dollars) 
51,305 
55,623 
53,487  
Percentage of all approved EIDL loan dollars  
47.5 
52.5 
100  
Source: GAO analysis of Small Business Administration (SBA) and Census Bureau data. | GAO-21-589 
Note: The total number of counties or similar entities that can be matched to EIDL loan data is 3,139. 
According to the Census Bureau, 1,788 counties had poverty rates above the national rate, and 1,351 
counties had rates below the national rate. 
 
Figure 19 illustrates EIDL lending outcomes associated with county 
poverty rate, by quartile. Businesses in the lowest quartile of counties in 
terms of poverty rate generally had higher loan approval rates, loan 
amounts on a per-employee basis, and average loan sizes. In contrast, 
counties in the highest quartile in terms of poverty rate had somewhat 
more loans per 1,000 businesses. 
                                                                                                                       
3The national 5-year estimate of poverty rate was 12.9 percent, according to our analysis 
of data from the 2015–2019 American Community Survey.   
EIDL Applicant and 
Recipient Outcomes 
by County Poverty 
Rate 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 77 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 19: Economic Injury Disaster Loan Program Lending Outcomes by County Poverty Rate 
 
Note: We ordered the poverty rates for all counties from the lowest to the highest value. We then 
divided the ordered distribution of counties into four equally sized groups (quartiles). The counties 
with the lowest poverty rates fall within the “<25th” quartile while those with the highest rates fall 
within the “>75th” quartile. We calculated the average for each lending outcome by quartile. 
 
 
 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 78 
GAO-21-589  Economic Injury Disaster Loan Program 
Businesses in counties with foreign-born populations above the national 
estimated share of the population received more and larger loans on 
average, as well as a larger share of all EIDL dollars (see table 12).4 
 
 
 
Table 12: Economic Injury Disaster Loan (EIDL) Applications, Loan Size, and Share of Total Loan Dollars, by County Foreign-
Born Population 
Variable 
Counties above 
national rate 
Counties below or equal to 
national rate 
All U.S. counties 
Average number of EIDL applications 
23,088 
1,482 
2,927  
Average number of declined EIDL applications 
13,624 
933 
1,782  
Average number of approved EIDL applications 
9,464 
548 
1,144  
Average loan size (in dollars) 
55,112 
51,473 
53,487  
Percentage of all approved EIDL loan dollars  
57.0 
43.0 
100  
Source: GAO analysis of Small Business Administration (SBA) and Census Bureau data. | GAO-21-589 
Note: The total number of counties or similar entities that can be matched to EIDL loan data is 3,139. 
According to the Census Bureau, 210 counties had percentages of foreign-born residents above the 
national rate, and 2,929 counties had rates below the national rate. 
 
Figure 20 illustrates EIDL lending outcomes associated with foreign-born 
population, by quartile. Businesses in the highest quartile of counties in 
terms of poverty rate had higher loan approval rates, loan amounts on a 
per-employee basis, and average loan sizes. Counties with more foreign-
born residents also received more loans per 1,000 small businesses. 
                                                                                                                       
4The national 5-year estimate of the share of the population born outside of the United 
States was 13.6 percent, according to our analysis of data from the 2015–2019 American 
Community Survey.   
EIDL Applicant and 
Recipient Outcomes 
by County Foreign-
Born Population 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 79 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 20: Economic Injury Disaster Loan Program Lending Outcomes by Foreign-Born Population Percentage 
 
Note: We ordered the percentage of foreign-born population for all counties from the lowest to the 
highest value. We then divided the ordered distribution of counties into four equally sized groups 
(quartiles). The counties with the lowest percentages fall within the “<25th” quartile while those with 
the highest percentages fall within the “>75th” quartile. We calculated the average for each lending 
outcome by quartile. 
 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 80 
GAO-21-589  Economic Injury Disaster Loan Program 
Businesses in counties with shares of households with limited English 
proficiency above the national estimate of 4.4 percent received more and 
larger loans on average, as well as a larger share of all EIDL dollars (see 
table 13). 
 
 
 
Table 13: Economic Injury Disaster Loan (EIDL) Applications, Loan Size, and Share of Total Loan Dollars, by County 
Households with Limited English Proficiency 
Variable 
Counties above 
national rate 
Counties below or equal to 
national rate 
All U.S. counties 
Average number of EIDL applications 
14,465 
1,617 
2,927  
Average number of declined EIDL applications 
8,610 
1,007 
1,782  
Average number of approved EIDL applications 
5,856 
609 
1,144  
Average loan size (in dollars) 
54,686 
52,178 
53,487  
Percentage of all approved EIDL loan dollars  
53.3 
46.7 
100  
Source: GAO analysis of Small Business Administration (SBA) and Census Bureau data. | GAO-21-589 
Note: The total number of counties or similar entities that can be matched to EIDL loan data is 3,139. 
According to the Census Bureau, 320 counties had percentages of households with limited English 
proficiency above the national rate, and 2,819 counties had rates below the national rate. 
 
Figure 21 illustrates EIDL lending outcomes associated with county share 
of households with limited English proficiency, by quartile. Businesses in 
the highest quartile of counties in terms of limited English proficiency had 
larger loan amounts on a per-employee basis and average loan sizes. 
These counties also received more loans per 1,000 small businesses. 
However, loan approval rates across quartiles were similar. 
EIDL Applicant and 
Recipient Outcomes 
by Share of 
Households with 
Limited English 
Proficiency 

 
Appendix II: Socioeconomic and Demographic 
Characteristics of EIDL Applicants and 
Recipients 
 
 
 
 
Page 81 
GAO-21-589  Economic Injury Disaster Loan Program 
Figure 21: Economic Injury Disaster Loan Program Lending Outcomes by County Share of Households with Limited English 
Proficiency 
 
Note: We ordered the shares of households with limited English proficiency for all counties from the 
lowest to the highest value. We then divided the ordered distribution of counties into four equally 
sized groups (quartiles). The counties with the lowest shares fall within the “<25th” quartile while 
those with the highest shares fall within the “>75th” quartile. We calculated the average for each 
lending outcome by quartile. 

 
Appendix III: Analysis of EIDL Lending 
Outcomes for Selected Metropolitan Statistical 
Areas by ZIP Code 
 
 
 
 
Page 82 
GAO-21-589  Economic Injury Disaster Loan Program 
We analyzed Economic Impact Disaster Loan (EIDL) lending outcomes 
for ZIP code tabulation areas (ZCTA) within four selected metropolitan 
statistical areas (MSA): New York-Newark-Jersey City, Chicago-
Naperville-Elgin, New Orleans-Metairie, and San Jose-Sunnyvale-Santa 
Clara.1 Because of a lack of aggregate business statistics at the ZCTA 
level, we are unable to analyze the number of EIDL loans per 1,000 small 
businesses or the loan amount approved on a per-employee basis. We 
limited our analysis of ZCTAs to calculating and comparing differences in 
loan approval rates across various socioeconomic and demographic 
factors. 
For each MSA, we examined EIDL loan approval rates for selected 
socioeconomic and demographic characteristics by ZCTA.2 We identified 
these MSAs based on factors such as varied population densities, 
geographic regions, and the number of businesses in each MSA 
requesting assistance (see app. I for more information). Based on our 
analysis of loan approval rates at the ZCTA level, our findings from the 
MSA analysis are generally consistent with those from our county-level 
analysis. Thus, our results do not appear to be sensitive to the 
geographic unit used.3 
The New York-Newark-Jersey City MSA spans 25 counties.4 Between 
March 2020 and February 2021, the Small Business Administration (SBA) 
approved 361,881 EIDL loans with an average loan amount of $59,171 
across 918 ZCTAs within the MSA. Our analysis of the demographic 
indicators showed that the approval rates were lower for ZCTAs with 
percentages of non-White population and number of households with 
                                                                                                                       
1The Census Bureau defines metropolitan statistical areas as having at least one 
urbanized area of 50,000 or more population, plus adjacent territory that has a high 
degree of social and economic integration with the core as measured by commuting ties.  
2ZIP code tabulation areas are generalized areal representations of U.S. Postal Service 
ZIP code service areas, which are not areal features but a collection of mail delivery 
routes. 
3We excluded the indicator for “percentage of foreign-born residents” from the analysis for 
MSAs because there was no ZCTA above the 13.6 percent national estimate within the 
MSAs we analyzed.  
4The counties in the New York-Newark-Jersey City MSA are Duchess County, NY; 
Putnam County, NY; Nassau County, NY; Suffolk County, NY; Essex County, NJ; 
Hunterdon County, NJ; Morris County, NJ; Somerset County, NJ; Sussex County, NJ; 
Union County, NJ; Pike County, PA; Bergen County, NJ; Hudson County, NJ; Middlesex 
County, NJ; Monmouth County, NJ; Ocean County, NJ; Passaic County, NJ; Bronx 
County, NY; Kings County, NY; New York County, NY; Orange County, NY; Queens 
County, NY; Richmond County, NY; Rockland County, NY; and Westchester County, NY.  
Appendix III: Analysis of EIDL Lending 
Outcomes for Selected Metropolitan 
Statistical Areas by ZIP Code 
New York-Newark-
Jersey City MSA 

 
Appendix III: Analysis of EIDL Lending 
Outcomes for Selected Metropolitan Statistical 
Areas by ZIP Code 
 
 
 
 
Page 83 
GAO-21-589  Economic Injury Disaster Loan Program 
limited English proficiency above the national estimate (see fig. 22). Our 
analysis of socioeconomic indicators showed that approval rates were 
higher for ZCTAs with median household income above the national 
estimate, and lower for ZCTAs with unemployment rate, poverty rate, and 
percentage of households with no internet access above the national 
estimate. 
Figure 22: Approval Rates for Economic Injury Disaster Loans in the New York-Newark-Jersey City Metropolitan Statistical 
Area for Selected Indicators, March 2020–February 2021 
 
Note: Traditionally underserved ZIP code tabulation areas consist of areas where the following 
indicators are above the national estimate: percentage of non-White population, percentage of 
population with limited English proficiency, unemployment rate, poverty rate, and percentage of 
population with no internet access. Additionally, traditionally underserved ZIP code tabulation areas 
consist of areas where the median household income is below the national estimate. We use 
“traditionally underserved” and “traditionally advantaged” to facilitate understanding of the data. 
 
The Chicago-Naperville-Elgin MSA spans 14 counties.5 Between March 
2020 and February 2021, SBA approved 126,929 loans with an average 
loan amount of $50,236 across 413 ZCTAs within the MSA. Our analysis 
of the demographic indicators showed that the approval rates were lower 
for ZCTAs with percentages of non-White population above the national 
estimate (see fig. 23). Additionally, approval rates were similar for ZCTAs 
with numbers of households with limited English proficiency above and 
                                                                                                                       
5The counties in the Chicago-Naperville-Elgin MSA are Cook County, IL; DuPage County, 
IL; Grundy County, IL; Kendall County, IL; McHenry County, IL; Will County, IL; DeKalb 
County, IL; Kane County, IL; Jasper County, IN; Lake County, IN; Newton County, IN; 
Porter County, IN; Lake County, IL; and Kenosha County, WI.  
Chicago-Naperville-
Elgin MSA 

 
Appendix III: Analysis of EIDL Lending 
Outcomes for Selected Metropolitan Statistical 
Areas by ZIP Code 
 
 
 
 
Page 84 
GAO-21-589  Economic Injury Disaster Loan Program 
below the national estimate. Our analysis of socioeconomic indicators 
showed that approval rates were higher for ZCTAs with a median 
household income above the national estimate, and lower for ZCTAs with 
unemployment rates, poverty rates, and percentages of households with 
no internet access above the national estimate. 
Figure 23: Approval Rates for Economic Injury Disaster Loans in the Chicago-Naperville-Elgin Metropolitan Statistical Area 
for Selected Indicators, March 2020–February 2021 
 
Note: Traditionally underserved ZIP code tabulation areas consist of areas where the following 
indicators are above the national estimate: percentage of non-White population, percentage of 
population with limited English proficiency, unemployment rate, poverty rate, and percentage of 
population with no internet access. Additionally, traditionally underserved ZIP code tabulation areas 
consist of areas where the median household income is below the national estimate. We use 
“traditionally underserved” and “traditionally advantaged” to facilitate understanding of the data. 
 
The San Jose-Sunnyvale-Santa Clara MSA spans two counties.6 
Between March 2020 and February 2021, SBA approved 22,047 loans 
with an average amount of $67,264 across 69 ZCTAs within the MSA. 
Our analysis of the demographic indicators showed that the approval 
rates were lower for ZCTAs with percentages of non-White population 
above the national estimate (see fig. 24). Approval rates were similar for 
ZCTAs with numbers of limited English proficiency households above and 
below the national estimate. In addition, our analysis of socioeconomic 
indicators showed that approval rates were higher for ZCTAs with median 
                                                                                                                       
6The counties in the San Jose-Sunnyvale-Santa Clara MSA are San Benito County and 
Santa Clara County.  
San Jose-Sunnyvale-
Santa Clara MSA 

 
Appendix III: Analysis of EIDL Lending 
Outcomes for Selected Metropolitan Statistical 
Areas by ZIP Code 
 
 
 
 
Page 85 
GAO-21-589  Economic Injury Disaster Loan Program 
household income and levels of internet access above the national 
estimate, and lower for ZCTAs with unemployment rates and poverty 
rates above the national estimate. 
Figure 24: Approval Rates for Economic Injury Disaster Loans in the San Jose-Sunnyvale-Santa Clara Metropolitan Statistical 
Area for Selected Indicators, March 2020–February 2021  
 
Note: Traditionally underserved ZIP code tabulation areas consist of areas where the following 
indicators are above the national estimate: percentage of non-White population, percentage of 
population with limited English proficiency, unemployment rate, poverty rate, and percentage of 
population with no internet access. Additionally, traditionally underserved ZIP code tabulation areas 
consist of areas where the median household income is below the national estimate. We use 
“traditionally underserved” and “traditionally advantaged” to facilitate understanding of the data. 
 
The New Orleans-Metairie MSA spans eight parishes.7 Between March 
2020 and February 2021, SBA approved 30,491 loans with an average 
amount of $45,917 across 80 ZCTAs within the MSA. Our analysis of the 
demographic indicators showed that the approval rates were lower for 
ZCTAs with percentages of non-White population above the national 
estimate (see fig. 25). Additionally, approval rates were similar for ZCTAs 
with numbers of households with limited-English proficiency above and 
below the national estimate. Our analysis of socioeconomic indicators 
showed that approval rates were higher for ZCTAs with median 
household income above the national estimate, and lower for ZCTAs with 
                                                                                                                       
7The parishes in the New Orleans-Metairie MSA are Jefferson Parish, LA; Orleans Parish, 
LA; Plaquemines Parish, LA; St. Bernard Parish, LA; St. Charles Parish, LA; St. James 
Parish, LA; St. John the Baptist Parish, LA; and St. Tammany Parish, LA.  
New Orleans-Metairie 
MSA 

 
Appendix III: Analysis of EIDL Lending 
Outcomes for Selected Metropolitan Statistical 
Areas by ZIP Code 
 
 
 
 
Page 86 
GAO-21-589  Economic Injury Disaster Loan Program 
unemployment rates, poverty rates, and percentages of households with 
no internet access above the national average. 
Figure 25: Approval Rates for Economic Injury Disaster Loans in the New Orleans-Metairie Metropolitan Statistical Area for 
Selected Indicators, March 2020–February 2021 
 
Note: Traditionally underserved ZIP code tabulation areas consist of areas where the following 
indicators are above the national estimate: percentage of non-White population, percentage of 
population with limited English proficiency, unemployment rate, poverty rate, and percentage of 
population with no internet access. Additionally, traditionally underserved ZIP code tabulation areas 
consist of areas where the median household income is below the national estimate. We use 
“traditionally underserved” and “traditionally advantaged” to facilitate understanding of the data. 
 

 
Appendix IV: Comments from the Small 
Business Administration 
 
 
 
 
Page 87 
GAO-21-589  Economic Injury Disaster Loan Program 
 
 
Appendix IV: Comments from the Small 
Business Administration 

 
Appendix IV: Comments from the Small 
Business Administration 
 
 
 
 
Page 88 
GAO-21-589  Economic Injury Disaster Loan Program 
 
 

 
Appendix IV: Comments from the Small 
Business Administration 
 
 
 
 
Page 89 
GAO-21-589  Economic Injury Disaster Loan Program 
 
 

 
Appendix IV: Comments from the Small 
Business Administration 
 
 
 
 
Page 90 
GAO-21-589  Economic Injury Disaster Loan Program 
 
 

 
Appendix V: GAO Contact and Staff 
Acknowledgments 
 
 
 
 
Page 91 
GAO-21-589  Economic Injury Disaster Loan Program 
William B. Shear, (202) 512-8678 or ShearW@gao.gov. 
 
In addition to the contact name above, Marshall Hamlett (Assistant 
Director), Weifei Zheng (Analyst in Charge), Vida Awumey, Irina 
Carnevale, Chelsea Carter, Marissa Esthimer, Jacob Fender, Daniel 
Flavin, Jill Lacey, Andrew Lobel, Ying Long, Marc Molino, Angela Pun, 
Jessica Sandler, and Jennifer Schwartz made significant contributions to 
this report. 
 
Appendix V: GAO Contact and Staff 
Acknowledgments 
GAO Contact 
Staff 
Acknowledgments 
(104417) 

 
 
 
 
 
 
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. 
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. 
The price of each GAO publication reflects GAO’s actual cost of production and 
distribution and depends on the number of pages in the publication and whether 
the publication is printed in color or black and white. Pricing and ordering 
information is posted on GAO’s website, https://www.gao.gov/ordering.htm.  
Place orders by calling (202) 512-6000, toll free (866) 801-7077, or  
TDD (202) 512-2537. 
Orders may be paid for using American Express, Discover Card, MasterCard, 
Visa, check, or money order. Call for additional information. 
Connect with GAO on Facebook, Flickr, Twitter, and YouTube. 
Subscribe to our RSS Feeds or Email Updates. Listen to our Podcasts. 
Visit GAO on the web at https://www.gao.gov. 
Contact FraudNet: 
Website: https://www.gao.gov/about/what-gao-does/fraudnet 
Automated answering system: (800) 424-5454 or (202) 512-7700 
Orice Williams Brown, Managing Director, WilliamsO@gao.gov, (202) 512-4400, 
U.S. Government Accountability Office, 441 G Street NW, Room 7125, 
Washington, DC 20548 
Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800 
U.S. Government Accountability Office, 441 G Street NW, Room 7149  
Washington, DC 20548 
Stephen J. Sanford, Managing Director, spel@gao.gov, (202) 512-4707 
U.S. Government Accountability Office, 441 G Street NW, Room 7814, 
Washington, DC 20548 
GAO’s Mission 
Obtaining Copies of 
GAO Reports and 
Testimony 
Order by Phone 
Connect with GAO 
To Report Fraud, 
Waste, and Abuse in 
Federal Programs 
Congressional 
Relations 
Public Affairs 
Strategic Planning and 
External Liaison 
Please Print on Recycled Paper.

File and source

File
21-589-eidl-communication-fraud.pdf
Size
4,758,286 bytes
SHA-256
a1b5d3a52da4a1ef2d7be88a96d93348a096c90bcb5a91e6b2784a65db255445
Our copy
21-589-eidl-communication-fraud.pdf
Original
www.gao.gov
Back to top