GAO-21-601 — Paycheck Protection Program: Program Changes Increased Lending to the Smallest Businesses and in Underserved Locations…

Record facts

Summary

GAO-21-601, a report to congressional addressees issued by the U.S. Government Accountability Office in September 2021 on the Paycheck Protection Program. The report describes how small business and lender participation in PPP changed across three phases, for which Congress provided commitment authority of about $814 billion. It states that 42 percent of Phase 1 loans, approved April 3-16, 2020, went to larger businesses of 10 to 499 employees, which accounted for 4 percent of all U.S. small businesses, and that rural businesses received 19 percent of Phase 1 loans while representing 13 percent of small businesses. It reports that SBA admitted about 600 new lenders for Phase 2 and targeted funding to minority-owned businesses through Community Development Financial Institutions. GAO analyzed SBA loan-level data and surveyed 1,383 lenders, receiving 781 responses.

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

PAYCHECK 
PROTECTION 
PROGRAM 
Program Changes 
Increased Lending to 
the Smallest 
Businesses and in 
Underserved 
Locations 
 
 
 
Report to Congressional Addressees 
September 2021 
 
GAO-21-601 
 
 

 
 
 United States Government Accountability Office 
  
Highlights of GAO-21-601, a report to 
congressional addressees 
 
September 2021 
PAYCHECK PROTECTION PROGRAM 
Program Changes Increased Lending to the Smallest 
Businesses and in Underserved Locations 
What GAO Found 
The Paycheck Protection Program (PPP) supports small businesses through 
forgivable loans for payroll and other eligible costs. Early lending favored larger 
and rural businesses, according to GAO’s analysis of Small Business 
Administration (SBA) data. Specifically, 42 percent of Phase 1 loans (approved 
from April 3–16, 2020) went to larger businesses (10 to 499 employees), 
although these businesses accounted for only 4 percent of all U.S. small 
businesses. Similarly, businesses in rural areas received 19 percent of Phase 1 
loans but represented 13 percent of all small businesses. Banks made a vast 
majority of Phase 1 loans.   
In response to concerns that some underserved businesses—in particular, 
businesses owned by self-employed individuals, minorities, women, and 
veterans—faced challenges obtaining loans, Congress and SBA made a series 
of changes that increased lending to these businesses. For example,  
• 
SBA admitted about 600 new lenders to start lending in Phase 2 (which ran 
from April 27–August 8, 2020), including nonbanks (generally, lending 
institutions that do not accept deposits).  
• 
SBA developed guidance after Phase 1 helping self-employed individuals 
participate in the program. 
• 
SBA targeted funding to minority-owned businesses in part through 
Community Development Financial Institutions in Phases 2–3. (Phase 3 ran 
from January 12–June 30, 2021.) 
By the time PPP closed in June 2021, lending in traditionally underserved 
counties was proportional to their representation in the overall small business 
community (see figure). While lending to businesses with fewer than 10 
employees remained disproportionately low, it increased significantly over the 
course of the program. 
Paycheck Protection Program Loans, by Type of Business or County 
 
View GAO-21-601. For more information, 
contact John Pendleton at (202) 512-8678 or 
pendletonj@gao.gov. 
Why GAO Did This Study 
The COVID-19 pandemic resulted in 
significant turmoil in the U.S. economy, 
leading to temporary and permanent 
business closures and high 
unemployment. In response, in March 
2020, Congress established PPP 
under the CARES Act and ultimately 
provided commitment authority of 
approximately $814 billion for the 
program over three phases. When 
initial program funding ran out in 14 
days, concerns quickly surfaced that 
certain businesses were unable to 
access the program, prompting a 
series of changes by Congress and 
SBA.  
The CARES Act includes a provision 
for GAO to monitor the federal 
government’s efforts to respond to the 
COVID-19 pandemic. GAO has issued 
a series of reports on this program, 
and has made a number of 
recommendations to improve program 
performance and integrity. This report 
describes trends in small business and 
lender participation in PPP. 
GAO analyzed loan-level PPP data 
from SBA and county-level data from 
four U.S. Census Bureau products and 
surveyed a generalizable sample of 
PPP lenders, stratified by lender type 
and size. GAO also reviewed 
legislation, interim final rules, agency 
guidance, and relevant literature, as 
well as interviewed SBA officials.  

 
 
 
 
 
 
 
 
 
 
 
Page i 
GAO-21-601  Paycheck Protection Program 
Letter 
 
1 
Background 
4 
Program Changes Increased Lending to the Smallest Businesses 
and in Underserved Locations 
11 
Agency Comments 
34 
Appendix I 
Objectives, Scope, and Methodology 
40 
 
Appendix II 
GAO Contacts and Staff Acknowledgments 
47 
 
Tables 
Table 1: Paycheck Protection Program (PPP) Phases 
5 
Table 2: Key Paycheck Protection Program Changes, by Program 
Phase 
10 
Table 3: Selected Paycheck Protection Program Changes 
21 
 
Figures 
Figure 1: Paycheck Protection Program Lending Volume by 
Approval Date, April 2020–June 2021 
5 
Figure 2: Overview of Paycheck Protection Program Lenders 
7 
Figure 3: Overview of the Paycheck Protection Program (PPP) 
Application Process 
10 
Figure 4: Percentage of First Draw Paycheck Protection Program 
Loans to Larger Businesses Relative to Percentage of All 
Small Businesses, by Program Phase 
12 
Figure 5: Paycheck Protection Program Loans per Small Business 
in Phase 1, by County 
14 
Figure 6: Percentage of Paycheck Protection Program Recipients 
Relative to the Percentage of All Small Businesses in the 
Hardest-Hit Sectors 
17 
Figure 7: Number and Total Dollar Amount of Paycheck Protection 
Program Loans Made in Phase 1, by Lender Type 
18 
Figure 8: Percentage of First Draw Paycheck Protection Program 
Loans to Self-Employed Individuals, by Program Phase 
22 
Figure 9: Percentage of First Draw Paycheck Protection Program 
Loans to Businesses in Counties with Large Shares of 
Women-Owned Businesses, by Program Phase 
24 
Contents 

 
 
 
 
 
 
 
 
 
 
 
Page ii 
GAO-21-601  Paycheck Protection Program 
Figure 10: Percentage of First Draw Paycheck Protection Program 
Loans to Businesses in High-Minority Counties, by 
Program Phase 
26 
Figure 11: Percentage of First Draw Paycheck Protection Program 
Loans to Businesses in Metro Counties Relative to the 
Percentage of All Small Businesses, by Program Phase 
and Share of Minority Residents 
28 
Figure 12: Percentage of Paycheck Protection Program Loans, by 
Lender Type and Program Phase 
29 
Figure 13: Percentage of Paycheck Protection Program Loans 
Relative to the Share of Small Businesses, by Type of 
Business or County 
33 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Abbreviations 
 
CDFI  
 
Community Development Financial Institution 
Census 
 
U.S. Census Bureau 
COVID-19   
 
Coronavirus Disease 2019 
MDI 
 
 
Minority Depository Institution 
PPP 
 
 
Paycheck Protection Program 
SBA 
 
 
Small Business Administration 
 
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-21-601  Paycheck Protection Program 
441 G St. N.W. 
Washington, DC 20548 
September 21, 2021 
Congressional Addressees 
Congress passed the CARES Act in March 2020 in response to the 
Coronavirus Disease 2019 (COVID-19) pandemic.1 The CARES Act 
provided more than $2 trillion in emergency assistance and health care 
response for individuals, families, and businesses affected by COVID-19. 
Stay-at-home orders, social distancing requirements, and reduced 
consumer demand early in the pandemic caused both temporary and 
permanent business closures, particularly among small businesses. 
Congress included the Paycheck Protection Program (PPP) in the 
CARES Act, providing commitment authority of $349 billion to support 
small businesses affected by the pandemic and to keep small business 
workers employed.2 
Administered by the Small Business Administration (SBA) through a 
network of participating lenders, PPP provided forgivable loans to small 
businesses for payroll and certain other eligible costs. To provide relief 
quickly, lenders began approving loans 7 days after the CARES Act was 
enacted. When the program’s initial round of funding ran out in 14 days, 
policy makers and small business organizations raised concerns that the 
smallest businesses and those owned by minorities, women, and 
veterans had difficulty accessing the program. In response, Congress 
made program changes, renewed the program twice, and provided 
approximately an additional $465 billion to meet continued demand. 
                                                                                                                       
1Pub. L. No. 116-136, 134 Stat. 281 (2020). 
2PPP was one of the programs Congress created, expanded, or funded under the CARES 
Act to support workers and small businesses in the economic downturn. Congress also 
eased borrowing requirements for the Economic Injury Disaster Loan program and 
created and provided appropriations for Economic Injury Disaster Loan Advances. For 
more information, see GAO, Economic Injury Disaster Loan Program: Additional Actions 
Needed to Improve Communication with Applicants and Address Fraud Risk, GAO-21-589 
(Washington, D.C.: July 30, 2021). In addition, the CARES Act created three federally 
funded temporary unemployment insurance programs that expanded benefit eligibility and 
enhanced benefits. This report does not assess the efficacy of PPP in addressing the 
needs of small businesses during the pandemic, nor does it compare outcomes of 
recipient businesses to nonrecipient businesses or businesses that utilized other small 
business lending programs during the pandemic. 
Letter 
 

 
 
 
 
 
Page 2 
GAO-21-601  Paycheck Protection Program 
The CARES Act includes a provision for us to monitor and oversee the 
federal government’s efforts to prepare for, respond to, and recover from 
COVID-19.3 We have issued a series of reports on this program and 
made a number of recommendations to improve program performance 
and integrity. For example, in July 2021, we issued a report on the 
safeguards SBA put in place during the PPP loan approval and 
forgiveness processes and SBA’s oversight of PPP loans and lenders.4 
We also included emergency loans for small businesses—which include 
PPP—as a new area on our High-Risk List in March 2021 because of the 
potential for fraud, significant program integrity risks, and need for much 
improved program management and oversight.5 This report describes 
how small business and lender participation in PPP evolved over time. 
To describe how small business and lender participation in PPP evolved 
over time, we analyzed loan-level PPP data from SBA for loans approved 
from April 3, 2020, to June 30, 2021, and county-level data from four U.S. 
Census Bureau (Census) products: the 2019 release of the American 
Community Survey 5-year estimates; the 2018 Annual Business Survey; 
the 2017 Nonemployer Statistics dataset; and the 2017 Statistics of U.S. 
Businesses dataset. Specifically, we used PPP data to describe the size, 
type, and sector of recipient businesses. Because SBA does not have 
complete data on the race, gender, and veteran status of business 
owners, we used Census data and estimates to describe the 
demographic and socioeconomic characteristics of counties in which 
recipient businesses were located.6 To assess the reliability of the data 
sources we used in our analyses, we primarily reviewed related 
documentation. In some cases, we also interviewed agency officials and 
                                                                                                                       
3We 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. 
4See GAO, Paycheck Protection Program: SBA Added Program Safeguards, but 
Additional Actions Are Needed, GAO-21-577 (Washington, D.C.: July 29, 2021). 
5See GAO, High Risk Series: Dedicated Leadership Needed to Address Limited Progress 
in Most High-Risk Areas, GAO-21-119SP (Washington, D.C.: Mar. 2, 2021). 
6As we reported in July 2021, SBA did not require demographic information—for example, 
the race, gender, and veteran status of the business owner—in the PPP borrower 
application. However, SBA revised its PPP borrower application form in January 2021 to 
allow for the optional disclosure of demographic information and included such a section 
in all subsequent revisions and an optional demographic information form as part of the 
PPP loan forgiveness application. See GAO-21-577. 

 
 
 
 
 
Page 3 
GAO-21-601  Paycheck Protection Program 
tested data for missing values and obvious errors. Because PPP 
applicants self-certified the accuracy of the information they provided to 
lenders and SBA, the reliability of the PPP data depends on the accuracy 
of the information they provided. We determined the data sources we 
used for our analyses were sufficiently reliable for the purposes of 
describing small business and lender participation in PPP. 
We did not assess PPP loans for fraud, and SBA officials told us they did 
not flag fraudulent loans or remove them from the PPP dataset they 
provided to us.7 While the full extent of fraud in PPP is not yet known, as 
of March 2021, the Department of Justice had announced charges in 134 
cases involving 431 loans to borrowers who were charged with fraud 
(such as bank fraud, wire fraud, or identity theft). Similarly, the SBA Office 
of Inspector General found that, of the 5.2 million total loans made from 
April 3 to August 9, 2020, lenders made more than one loan each to 
4,260 borrowers and 57,473 loans to individuals on the Department of the 
Treasury’s “Do Not Pay” list.8 
We also surveyed a generalizable sample of 1,383 PPP lenders, stratified 
by lender type and size, to obtain their perspectives on the program.9 We 
administered the survey from February 2021 to April 2021 and received 
781 responses. We obtained a weighted response rate of 57.3 percent.10 
Finally, we reviewed legislation, interim final rules, agency guidance, and 
relevant literature, and we interviewed SBA officials. For more information 
on our scope and methodology, see appendix I. 
                                                                                                                       
7SBA officials told us they track Department of Justice cases related to PPP and the 
corresponding loans. 
8Small Business Administration, Office of Inspector General, Flash Report: Duplicate 
Loans Made Under the Paycheck Protection Program, Report No. 21-09 (Washington, 
D.C.: Mar. 15, 2021); and Management Alert: Paycheck Protection Program Loan 
Recipients on the Department of Treasury’s Do Not Pay List, Report No. 21-06 
(Washington, D.C.: Jan. 11, 2021). 
9We also used this survey to inform our July 2021 report on SBA’s implementation of PPP 
(GAO-21-577). See app. I of that report for more details on the survey’s methodology and 
app. II for the survey’s full results. 
10We used a weighted response rate because our survey sample incorporated strata with 
different probabilities of selection. A weighted response rate may more accurately reflect 
the level of participation. For example, large units that contribute relatively more to the 
estimate of a total would have a larger “weight” on the response rate. 

 
 
 
 
 
Page 4 
GAO-21-601  Paycheck Protection Program 
We conducted this performance audit from July 2020 to September 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 CARES Act, signed into law by the President on March 27, 2020, 
established PPP under section 7(a) of the Small Business Act to provide 
forgivable loans to small businesses and nonprofit organizations, referred 
to collectively as “small businesses.”11 Under the law, qualifying 
businesses (in general, these include businesses with 500 or fewer 
employees or that meet SBA’s industry-based size standard) could obtain 
loans equal to 2.5 months of average total monthly payments for payroll 
costs up to $10 million, and were required to self-certify their need for the 
loan.12 
When loan approvals ended in June 2021, Congress had provided 
commitment authority of about $814 billion to PPP in three phases (see 
table 1).13 
 
 
                                                                                                                       
11SBA administers the 7(a) guaranteed loan program, which provides small businesses 
access to capital that they would not be able to access in the competitive market. In fiscal 
year 2019, SBA approved $23 billion in loans through the 7(a) program. 
12As set forth in the CARES Act, borrowers had to certify in good faith that (1) current 
economic uncertainty made the loan request necessary to support the applicant’s ongoing 
operations and (2) funds would be used to retain workers and maintain payroll or make 
payments for other covered expenses. Pub. L. No. 116-136, § 1102(a)(1)(B), 134 Stat. 
281, 291 (2020) (codified at 15 U.S.C. § 636(a)(36)(G)(i)(I),(II)). To streamline the 
process, SBA required minimal review from lenders—such as confirming receipt of 
borrower certifications and supporting payroll documentation.  
13SBA’s loan-level PPP data included the date SBA approved the loan but not the date 
the application was received. As a result, Phase 1 data include loans approved from April 
3–16, 2020, Phase 2 data include loans approved from April 27–August 8, 2020, and 
Phase 3 data include loans approved from January 12–June 30, 2021. The dataset 
includes 10 loans with approval dates on April 20-22, 2020, which we categorized as 
Phase 1, and one loan with an approval date of August 9, 2020, which we categorized as 
Phase 2. 
Background 
PPP Commitment 
Authority and Lending 
Volume 

 
 
 
 
 
Page 5 
GAO-21-601  Paycheck Protection Program 
Table 1: Paycheck Protection Program (PPP) Phases 
Phase 
Legislation  
Date enacted 
Commitment authority 
provided 
Application period 
1 
CARES Act 
March 27, 2020 
$349 billion 
April 3, 2020–April 16, 
2020 
2 
Paycheck Protection Program and 
Health Care Enhancement Act 
April 24, 2020 
$310 billion April 27, 2020–August 8, 
2020 
3 
Consolidated Appropriations Act, 
2021  
December 27, 2020 
$147.45 billion 
January 11, 2021–May 
31, 2021a 
American Rescue Plan Act of 2021 
March 11, 2021 
$7.25 billion 
Source: GAO analysis of relevant laws.  |  GAO-21-601 
aThe PPP Extension Act of 2021 extended the application period for Phase 3 from March 31, 2021, to 
May 31, 2021, and allowed the Small Business Administration until June 30, 2021, to process those 
applications. On May 4, 2021, the PPP general fund was exhausted and closed to new applications, 
except those processed by a community financial institution lender. 
 
Early demand for PPP loans was high. Daily lending volume—that is, the 
number of loans approved per day—was highest from April 3–May 1, 
2020, when 31 percent of all PPP loans were approved (see fig. 1). 
Figure 1: Paycheck Protection Program Lending Volume by Approval Date, April 2020–June 2021 
 

 
 
 
 
 
Page 6 
GAO-21-601  Paycheck Protection Program 
The CARES Act authorized SBA to use lenders already approved to 
participate in the 7(a) program and permitted SBA and Treasury to 
approve new lenders to issue PPP loans, provided they met certain 
requirements.14 Participating lenders included depository institutions (for 
example, banks and credit unions) and non-depository institutions (for 
example, nonbanks) (see fig. 2).15 For the purposes of this report, we 
refer to depository institutions as “banks” and characterize them by size: 
• 
large banks: total assets of $10 billion or more; 
• 
medium banks: total assets from $1 billion to $10 billion; and 
• 
small banks: total assets of less than $1 billion. 
 
Further, some lenders are designated Community Development Financial 
Institutions (CDFI) and Minority Depository Institutions (MDI). CFDIs and 
MDIs target loans to businesses and individuals in traditionally 
underserved areas with the goal of expanding economic opportunity. 
CDFIs include both banks and nonbanks, while MDIs include only 
banks.16 
                                                                                                                       
14Pub. L. No. 116-136, § 1102(a), 134 Stat. 281, 290 (2020) (codified at 15 U.S.C. § 
636(a)(36)(F)(iii)). In an interim final rule published April 2, 2020, SBA announced that any 
federally insured depository institution, credit union, or farm credit institution in good 
standing with its regulator would automatically qualify to participate in PPP upon 
submission of SBA’s PPP Lender Agreement. 85 Fed. Reg. 20,811, 20,815 (Apr. 15, 
2020). SBA and the Department of the Treasury were jointly responsible for approving 
lenders new to SBA to issue PPP loans. According to SBA officials, SBA approved new 
federally regulated lenders, and only new non-federally regulated and insured lenders 
required joint SBA and Treasury approval. 
15In general, nonbanks provide lending services but do not accept deposits. Nonbanks 
include SBA Small Business Lending Companies, SBA Microlenders, SBA Certified 
Development Companies, state-regulated financial companies, Business and Industrial 
Development Corporations, and Farm Credit System lenders. 
16The Department of the Treasury administers the CDFI Fund, which certifies lenders as 
CDFIs that share a common goal of expanding economic opportunity in low-income 
communities by providing access to financial products and services for local residents and 
businesses. MDIs are defined as depositories that (1) if a privately owned institution, 51 
percent is owned by socially and economically disadvantaged individuals; (2) if publicly 
owned, 51 percent of the stock is owned by socially and economically disadvantaged 
individuals; and (3) in the case of a mutual institution where the majority of the board of 
directors, account holders, and the community that the institution serves is predominantly 
minority. The term “minority” means any Black American, Native American, Hispanic 
American, or Asian American. 12 U.S.C. § 1463 note; 15 U.S.C. § 636(a)(36)(A)(xi)(II). 
We assigned to the CDFI/MDI group all lenders SBA flagged as CDFIs, MDIs, or both, 
and all remaining lenders to one of the other groups. 
PPP Lenders 

 
 
 
 
 
Page 7 
GAO-21-601  Paycheck Protection Program 
Figure 2: Overview of Paycheck Protection Program Lenders 
 
Note: All lenders that the Small Business Administration (SBA) flagged as Minority Depository 
Institutions, Community Development Financial Institutions, or both, were assigned to the Community 
Development Financial Institutions/Minority Depository Institutions group, and all remaining lenders 
were assigned to one of the other groups. 
 
As PPP was initially implemented, small businesses or nonprofit 
organizations were generally eligible for PPP loans if they had 500 or 
PPP Borrowers 

 
 
 
 
 
Page 8 
GAO-21-601  Paycheck Protection Program 
fewer employees.17 These businesses comprise approximately 99 
percent of businesses of all sizes.18 
Businesses that may not have paid employees (sole proprietors, 
independent contractors, self-employed individuals, qualified joint 
ventures, and single-member limited liability companies) were eligible for 
a PPP loan. For the purposes of this report, we refer to these business 
types collectively as “self-employed individuals.” Based on the statutory 
definition of payroll costs for self-employed individuals, the calculation of 
loan amounts for these businesses differed from the calculations used by 
other businesses, and the program requirements related to these 
calculations changed over the course of the program.19 
We used information from the 2020 Bureau of Labor Statistics’s Business 
Response Survey to identify six “hardest-hit sectors,” or sectors that were 
most likely to experience adverse effects to business operations as a 
result of the pandemic.20 These six sectors are (1) accommodation and 
food services; (2) arts, entertainment, and recreation; (3) educational 
services; (4) health care; (5) manufacturing; and (6) retail trade.21 
Finally, the CARES Act included a Sense of the Senate statement noting 
the Senate’s preference that SBA should issue guidance to lenders to 
ensure that processing and disbursement of loans prioritizes businesses 
in underserved and rural markets, including veterans and members of the 
                                                                                                                       
17Under the CARES Act, eligible borrowers included: (1) businesses that are small under 
SBA’s size standards (either the industry size standard or the alternative size standard); 
(2) any business, 501(c)(3) nonprofit organization, 501(c)(19) veterans organization, or 
Tribal business concern with 500 or fewer employees or that meets the applicable SBA 
industry size standard if more than 500; (3) any business in the accommodations and food 
services sector that has more than one physical location and employs not more than 500 
employees per physical location; and (4) sole proprietors, independent contractors, and 
eligible self-employed individuals. 
18U.S. Census Bureau, Nonemployer Statistics, 2017; and Statistics of U.S. Businesses, 
2017. 
1915 U.S.C. § 636(a)(36)(A)(viii)(I)(bb). 
20Adverse effects to business operations included a shortage of supplies or inputs, 
decreased demand for products or services, difficulty moving or shipping goods, and 
government-mandated closure of a business location. 
21Sector names refer to their classification codes within the North American Industry 
Classification System, which is the standard used by federal statistical agencies in 
classifying business establishments according to industry. 

 
 
 
 
 
Page 9 
GAO-21-601  Paycheck Protection Program 
military community, small business concerns owned and controlled by 
socially and economically disadvantaged individuals, women, and 
businesses in operation for less than 2 years.22 For the purposes of this 
report, we refer to “traditionally underserved businesses and counties” 
more broadly as very small businesses; those owned by minorities, 
women, or veterans; or those located in rural counties, counties with high 
minority populations, or Tribal areas.23 
To apply for a PPP loan, potential borrowers first submitted applications 
directly to a participating lender (see fig. 3). The lender then reviewed the 
application documentation and submitted its loan approval decision to 
SBA through SBA’s loan processing portal.24 Once received, SBA issued 
a loan number to the lender.25 The lender then disbursed the loan amount 
to the borrower. 
                                                                                                                       
22Pub. L. No. 116-136, § 1102(a), 134 Stat. 281, 293 (2020) (codified at 15 U.S.C. § 
636(a)(36)(P)(iv)). 
23In our analyses, we used businesses with fewer than 10 employees and those owned by 
self-employed individuals to represent very small businesses. To represent minority-, 
women-, and veteran-owned businesses, we used counties with high shares of minority 
residents and high shares of women- and veteran-owned businesses. We include Tribal 
areas when collectively referring to traditionally underserved counties. 
24SBA guidance stated that PPP was “first come, first served.” 85 Fed. Reg. 20,811, 
20,813 (Apr. 15, 2020). 
25SBA issued a loan number when it agreed to guarantee the loan. Initially, SBA reviewed 
loan and borrower information to look for duplicate applications before issuing a loan 
number to the lender. For Phase 3, SBA began conducting upfront compliance checks 
before issuing a loan number to the lender. For more information, see GAO-21-577. 
Application and Lending 
Processes 

 
 
 
 
 
Page 10 
GAO-21-601  Paycheck Protection Program 
Figure 3: Overview of the Paycheck Protection Program (PPP) Application Process 
 
 
As the program continued, Congress and SBA made a series of changes 
to PPP to increase lender participation and availability, target funding, 
and modify loan calculations (see table 2). In addition, for Phase 3, 
Congress allowed certain businesses that had already received a PPP 
loan to receive a second loan, referred to throughout this report as 
“second draw” loans. 
Table 2: Key Paycheck Protection Program Changes, by Program Phase 
Phase 
Legislation or program 
guidance and date 
Description of change 
1 
Technical event (April 8, 2020) The Small Business Administration (SBA) released the nonbank lender application form 
allowing new nonbank lenders to participate in the Paycheck Protection Program 
(PPP). 
2 
Interim final rule (April 20, 
2020) 
SBA provided instructions for self-employed individuals filing a Form 1040 Schedule C 
to apply by calculating the loan amount based on the business’s net profit from 2019. 
Guidance (April 23, 2020) 
SBA issued clarifying guidance stating that public companies with access to capital 
markets would be unlikely to have made a certification of need for a PPP loan in good 
faith, and allowed companies 2 weeks to return loan funds if they applied for the loan 
prior to the issuance of the guidance. 
Interim final rule (April 30, 
2020) 
SBA adjusted portfolio requirements for Community Development Financial Institutions 
(CDFI), majority minority-, women-, or veteran/military-owned financial institutions, and 
certain other nonbank lenders, which allowed smaller lenders to participate in PPP. 
Technical event (May 28, 
2020) 
SBA established a $10 billion set-aside for businesses that applied through CDFIs to 
target lending to minority-owned businesses and underserved communities.a  
3 
Consolidated Appropriations 
Act, 2021 (December 27, 
2020) 
Congress allowed some businesses to receive a second loan and targeted eligibility for 
second loans to businesses that met certain criteria, including having 300 or fewer 
employees and documenting revenue losses in 2020. 
Key Program Changes 

 
 
 
 
 
Page 11 
GAO-21-601  Paycheck Protection Program 
Phase 
Legislation or program 
guidance and date 
Description of change 
Guidance (January 11, 2021) 
SBA dedicated the first few days of Phase 3 to processing loans made through 
community financial institutions, which include CDFIs, Minority Depository Institutions, 
Certified Development Companies, and Microloan Intermediaries.  
Guidance (February 22, 2021) 
SBA established a 14-day loan application period exclusively for businesses or 
nonprofits with fewer than 20 employees. 
Interim final rule (March 4, 
2021) 
SBA revised the maximum loan amount calculation for self-employed individuals filing a 
Form 1040 Schedule C based on either gross income or net profit, which allowed for 
larger loans. 
Source: GAO analysis of SBA guidance and relevant laws.  |  GAO-21-601. 
Note: Phase 1 includes the application period from April 3–16, 2020, Phase 2 from April 27–August 8, 
2020, and Phase 3 from January 11–May 31, 2021. 
aCongress established additional set-asides throughout the course of the program. These included a 
total $60 billion set-aside for businesses that applied through community financial institutions, certain 
small insured depository institutions, and certain small credit unions established under the Paycheck 
Protection Program and Health Care Enhancement Act as well as $15 billion for lending by 
community financial institutions and $35 billion to new first draw PPP borrowers established under the 
Consolidated Appropriations Act, 2021. 
 
PPP lending in Phase 1 of the program favored larger and rural 
businesses, and banks made the vast majority of Phase 1 loans. Partly as 
a result of program changes made by Congress and SBA, the share of 
loans to most traditionally underserved businesses and counties we 
analyzed substantially increased in Phases 2 and 3, and generally was 
proportional to their representation in the overall small business 
community when the program closed in June 2021. 
 
 
 
 
Larger businesses. Larger businesses—small businesses with 10 to 499 
employees—received a higher share of loans in Phase 1 relative to their 
share of all small businesses. Specifically, 42 percent of loans in Phase 1 
went to these larger businesses, despite accounting for only 4 percent of 
all small businesses in the U.S. (see fig. 4). The share of first draw loans 
to these businesses decreased over time, suggesting that much of their 
demand for loans was met early in the program. Further, as we reported 
in September 2020, during Phases 1–2, a larger share of loans over $2 
Program Changes 
Increased Lending to 
the Smallest 
Businesses and in 
Underserved 
Locations 
Phase 1 Loans Flowed to 
Larger Businesses and 
Rural Areas Primarily 
through Banks 
Larger, Rural, and Certain 
Hardest-Hit Businesses 
Received Early Loans 

 
 
 
 
 
Page 12 
GAO-21-601  Paycheck Protection Program 
million were canceled compared to smaller loans following reports that 
publicly traded companies had received loans and SBA’s subsequent 
guidance clarifying the economic necessity requirements. As a result, the 
demand for larger loans may have diminished over time because of 
increased scrutiny from the public, the Department of the Treasury, and 
SBA.26 Finally, because the loan amount for each recipient business was 
based on average payroll, larger businesses generally received larger 
loans, likely precipitating the exhaustion of Phase 1 funding after just 2 
weeks.27 
Figure 4: Percentage of First Draw Paycheck Protection Program Loans to Larger 
Businesses Relative to Percentage of All Small Businesses, by Program Phase 
 
Note: Larger businesses are those with 10 to 499 employees. This analysis excludes loans that were 
approved and subsequently canceled, and second draw loans. Generally, Phase 1 data include loans 
approved from April 3–16, 2020, Phase 2 data include loans approved from April 27–August 8, 2020, 
and Phase 3 data include loans approved from January 12–June 30, 2021. 
                                                                                                                       
26GAO-20-701. 
27The average (median) loan size for recipient businesses with fewer than 10 employees 
was $21,697 ($18,115), compared to $296,292 ($146,000) for recipient businesses with 
10 to 499 employees. 

 
 
 
 
 
Page 13 
GAO-21-601  Paycheck Protection Program 
Research on small business lending and PPP suggests businesses with 
pre-existing relationships with banks were able to access PPP earlier 
than businesses without such relationships. For example, one study on 
PPP found that among businesses that applied for PPP loans from banks, 
approval rates were generally higher for businesses with a pre-existing 
relationship with a bank.28 Also, larger businesses were more likely to 
have these relationships than the smallest businesses.29 
Other research found that the smallest businesses were less aware of 
PPP and less likely to apply early in the program.30 If they did apply, the 
smallest businesses applied later, faced longer processing times, and 
were less likely to have their applications approved, which may also partly 
explain why early lending skewed toward larger businesses during Phase 
1. 
Businesses in rural areas. Rural counties generally received a higher 
number of loans per small business in Phase 1 than metro counties (see 
fig. 5).31 Additionally, while only 13 percent of small businesses are 
                                                                                                                       
28Alexander W. Bartik et al., “The Targeting and Impact of Paycheck Protection Program 
Loans to Small Businesses,” (working paper 27623, National Bureau of Economic 
Research, July 2020). Additionally, as we found in GAO-21-577, because lenders were 
required to apply relevant Bank Secrecy Act program requirements, they had to collect 
additional information from PPP loan applicants with which they did not have a pre-
existing relationship. Because pre-existing clients required the lender to conduct less due 
diligence, their applications could be processed more quickly. The Bank Secrecy Act 
requires banks and other financial institutions to take precautions against money 
laundering and other illicit financial activities by conducting due diligence activities and 
informing the Department of the Treasury of suspicious activity by their customers. 
29According to the Federal Reserve Banks’ 2020 Small Business Credit Survey, larger 
businesses (those with annual revenue greater than $1 million) were more likely to have 
existing banking relationships. See Federal Reserve Banks of Atlanta, Boston, Chicago, 
Cleveland, Dallas, Kansas City, Minneapolis, New York, Philadelphia, Richmond, St. 
Louis, and San Francisco, “Small Business Credit Survey: 2020 Report on Employer 
Firms (New York, NY: 2020). The survey was conducted in the third and fourth quarters of 
2019 and yielded 5,514 responses from small employer firms with one to 499 full or part-
time employees, in the 50 states and the District of Columbia. 
30Christopher Neilson, John Eric Humphries, and Gabriel Ulyssea. “Information Frictions 
and Access to the Paycheck Protection Program,” (working paper 27624, National Bureau 
of Economic Research, July 2020). 
31We used data from the Department of Agriculture’s Economic Research Service to 
classify counties as “metro” if they have one or more high-density urban areas with 50,000 
or more residents or are outlying counties that are economically tied to these central 
counties, as measured by the share of workers commuting on a daily basis to the central 
counties. Rural counties are outside the boundaries of metro areas and have no cities with 
50,000 or more residents. 

 
 
 
 
 
Page 14 
GAO-21-601  Paycheck Protection Program 
located in rural counties, 19 percent of all Phase 1 loans went to 
businesses in these counties. 
Figure 5: Paycheck Protection Program Loans per Small Business in Phase 1, by County 
 
Note: This analysis excludes loans that were approved and subsequently canceled. Generally, Phase 
1 data include loans approved from April 3–16, 2020. 
 

 
 
 
 
 
Page 15 
GAO-21-601  Paycheck Protection Program 
Businesses in rural counties also received loans earlier on average than 
businesses in metro areas. Specifically, in the period from April 3–May 1, 
2020, when loan volume was at its highest, loans were approved an 
average of 3 days earlier for businesses in rural areas than loans for 
businesses in metro areas. 
Additionally, we found that businesses in rural counties collectively 
received a higher share of loans in Phase 1 than the share of small 
businesses in rural counties, even in high-minority rural counties and rural 
counties with poorer socioeconomic outcomes.32 More specifically, 
businesses in rural counties received a high share of Phase 1 loans 
potentially because many of their loans were approved by small banks, 
which approved loans earlier than large banks. For example, of all loans 
approved by small banks from April 3–May 1, 2020 (when daily approvals 
were highest), 59 percent were approved before initial funding ran out on 
April 16, 2020. Conversely, of all loans approved by large banks from 
April 3–May 1, 2020, only 34 percent were approved before initial funding 
ran out. In total, 44 percent of loans approved by small banks went to 
businesses in rural areas compared to 6 percent of loans approved by 
large banks. 
Businesses in hardest-hit sectors. Collectively, businesses in the 
hardest-hit sectors—accommodation and food services; arts, 
entertainment, and recreation; educational services; health care; 
manufacturing; and retail trade—received a higher share of loans early in 
the program, as well as a high share of second draw loans in Phase 3, 
than the share of small businesses in those sectors.33 Specifically, 
businesses in these sectors received 40 percent of loans in Phase 1, and 
account for 27 percent of all small businesses nationwide. Moreover, like 
                                                                                                                       
32For this analysis, we define minority counties as those in which the percentage of 
minority residents was higher than the national share of 27.5 percent. To determine a 
county’s socioeconomic status relative to the national estimate, we used five indicators: 
median household income, unemployment rate, use of public assistance income, poverty 
rate, and no internet access. Among all five indicators, small businesses in rural counties 
with a county-level estimate worse than the national estimate still received a relatively high 
share of loans compared to small businesses in those counties. Based on 2015–2019 5-
year estimates from Census’s American Community Survey data, median household 
income was $62,843, the national unemployment rate was 5.3 percent, the national share 
of households receiving public assistance income was 2.4 percent, the national poverty 
rate was 12.9 percent, and the national share of households without internet access was 
13.9 percent. 
33Based on our analysis of data from the U.S. Bureau of Labor Statistics’s 2020 Business 
Response Survey, businesses in these sectors were most likely to experience adverse 
effects to their business operations as a result of the COVID-19 pandemic. 

 
 
 
 
 
Page 16 
GAO-21-601  Paycheck Protection Program 
loans to larger businesses, the share of loans to businesses in hardest-hit 
sectors declined over time, suggesting that demand for loans among 
these businesses was met early in the program. 
However, businesses in some of the hardest-hit sectors received a higher 
share of loans than others. In particular, in Phase 1, businesses in the 
retail trade, health care, accommodation and food services, and 
manufacturing sectors received a higher share of loans relative to the 
number of small businesses in those sectors (see fig. 6). Conversely, 
businesses in the arts, entertainment, and recreation and educational 
services sectors received a lower share of Phase 1 loans than the share 
of small businesses in those sectors. 
Businesses in the retail trade sector that received loans in Phase 1 were 
more concentrated in rural areas, and businesses in the educational 
services sector were less concentrated in rural areas than recipient 
businesses overall. Because businesses in rural counties received a 
higher share of Phase 1 loans than businesses in metro counties, location 
may partly explain the higher share of early loans to businesses in the 
retail trade sector and the lower share of early loans to businesses in the 
educational services sector. Additionally, businesses in the 
accommodation and food services, educational services, and 
manufacturing sectors that received loans in Phase 1 were larger, on 
average, than recipient businesses overall, which may further explain the 
high share of loans to these sectors. Finally, researchers found that early 
application rates were particularly high in the accommodation and retail 
sectors, which likely resulted in higher rates of loans to businesses in 
these sectors.34 
Businesses in the hardest-hit sectors also received a high share of 
second draw loans (35 percent) in Phase 3 than the share of small 
businesses in those sectors (27 percent), demonstrating their ongoing 
demand for financial assistance as the pandemic continued. However, as 
with early loans, businesses in certain hardest-hit sectors received a 
higher share of second draw loans than others. Businesses in the health 
care, accommodation and food services, and manufacturing sectors 
received a relatively high share of second draw loans, while those in retail 
                                                                                                                       
34Bartik et al., “The Targeting and Impact of Paycheck Protection Program Loans.” 

 
 
 
 
 
Page 17 
GAO-21-601  Paycheck Protection Program 
trade; arts, entertainment, and recreation; and educational services 
sectors received a relatively low share.35 
Figure 6: Percentage of Paycheck Protection Program Recipients Relative to the 
Percentage of All Small Businesses in the Hardest-Hit Sectors 
 
Note: This analysis excludes loans that were approved and subsequently canceled. Generally, Phase 
1 data include loans approved from April 3–16, 2020, and second draw applies to Phase 3 loans 
(those approved from January 12–June 30, 2021). 
                                                                                                                       
35Congress created two sector-specific SBA programs to target aid to venue operators 
and food services businesses. The Shuttered Venue Operators Grant program, 
established in 2020 by the Consolidated Appropriations Act, 2021, and amended by the 
American Rescue Plan Act, includes over $16 billion in grants to shuttered venues. The 
Restaurant Revitalization Fund, established in 2021 by the American Rescue Plan Act, 
provides funding to restaurants and other food service establishments to compensate for 
pandemic-related revenue losses. Businesses eligible for these programs generally fall 
into the accommodation and food services and arts, entertainment, and recreation 
sectors. Businesses that chose to receive both a PPP loan in Phase 3 and a Shuttered 
Venue Operators Grant had the amount of their PPP loan deducted from their Shuttered 
Venue Operators Grant award. 

 
 
 
 
 
Page 18 
GAO-21-601  Paycheck Protection Program 
Banks collectively made more than 93 percent of all loans in Phase 1. 
Large banks made 41 percent of loans (663,577 loans), medium banks 
made 30 percent of loans (481,063 loans), and small banks made 23 
percent of loans (366,030 loans) (see fig. 7). CDFIs and MDIs collectively 
made 4 percent of loans (66,419 loans) and nonbank lenders made 3 
percent of loans (42,112 loans).36 When measuring lending activity by the 
total dollar amount of loans made in Phase 1, the distribution among 
large, medium, and small banks was similar. 
Figure 7: Number and Total Dollar Amount of Paycheck Protection Program Loans 
Made in Phase 1, by Lender Type 
 
Note: For this analysis, all depository lenders are categorized as “banks,” including banks, credit 
unions, and savings and loan associations. Large banks are as those with at least $10 billion in 
                                                                                                                       
36As discussed later in this report, limited lending among CDFIs, MDIs, and nonbanks in 
Phase 1 corresponded with low participation among certain traditionally underserved 
businesses and counties during the same period. 
Banks Made the Vast Majority 
of Phase 1 Loans 

 
 
 
 
 
Page 19 
GAO-21-601  Paycheck Protection Program 
assets, medium banks are those with $1 to $10 billion, and small banks are those with less than $1 
billion based on publicly available call reports. All nondepository lenders are categorized as 
“nonbanks,” including SBA Small Business Lending Companies, SBA Certified Development 
Companies, SBA Microlenders, Business and Industrial Development Corporations, Farm Credit 
System lenders, and state-regulated financial companies. Generally, Phase 1 data include loans 
approved from April 3–16, 2020. 
 
Although large banks made more loans than all other lender types in 
Phase 1, researchers found they underperformed compared to small 
banks—with underperformance defined as having made a smaller share 
of PPP loans relative to their share of non-PPP small business loans.37 
Further, bank performance affected whether PPP borrowers were able to 
access loans before funding ran out at the end of Phase 1. Specifically, 
only 25 percent of all PPP borrowers located in areas with 
underperforming banks (regardless of size) obtained PPP approval prior 
to the end of Phase 1. By contrast, approximately 42 percent of all PPP 
borrowers in areas with banks that over-performed (or made more PPP 
loans than their share of other small business loans) had access to funds 
in Phase 1. 
Small banks may have been more experienced with small business 
lending, and therefore better equipped to process PPP applications than 
other lenders, according to researchers and our analysis. For example, 
research from the Federal Deposit Insurance Corporation found that 
community banks (which are generally smaller banks) held a high share 
of small business loans before the COVID-19 pandemic.38 It also found 
that community banks’ participation in PPP was proportionately larger 
than their size in the banking industry, meaning that community banks 
made a larger share of PPP loans than other types of loans. Further, 
based on results from our survey of PPP lenders, small banks approved 
the highest percentage of PPP applications on average (approximately 96 
                                                                                                                       
37João Granja et al., “Did the Paycheck Protection Program Hit the Target?.” (working 
paper 27095, National Bureau of Economic Research, November 2020). 
38Margaret Hanrahan and Angela Hilton, “The Importance of Community Banks in 
Paycheck Protection Program Lending.” FDIC Quarterly, vol. 14, no. 4 (2020). For the 
purposes of this research, community banks include banks with assets less than $1 billion, 
banks that primarily engage in basic banking activities rather than specialty activities, and 
banks that operate within a limited geographic scope. 

 
 
 
 
 
Page 20 
GAO-21-601  Paycheck Protection Program 
percent), compared to an estimated 93 percent for large and medium 
banks, and an estimated 86 percent for CDFIs and MDIs.39 
 
 
 
 
 
Following concerns raised by small business associations that some 
traditionally underserved businesses faced challenges accessing loans in 
Phase 1, Congress and SBA modified PPP to increase participation 
among certain businesses, such as self-employed individuals, women-
owned businesses, and businesses in minority counties.40 These changes 
included increasing the number of lenders in the program, targeting 
funding and processing timeframes to certain businesses, and expanding 
program eligibility (see table 3). 
 
 
                                                                                                                       
39We surveyed a generalizable sample of 1,383 PPP lenders, stratified by lender type and 
size, to obtain their perspectives on the program. The survey closed on April 15, 2021, 
and we received 781 responses. We obtained a weighted response rate of 57.3 percent. 
The 95 percent confidence intervals for these estimates are (95, 97) for small banks, (91, 
94) for medium and large banks, and (84, 89) for CDFIs and MDIs. 
40Changes were also intended to target PPP lending to veteran-owned businesses. The 
share of loans to businesses in counties with large shares of veteran-owned businesses 
was generally consistent over time. 
Program Changes, 
Including Lender Pool 
Expansion, Increased 
Lending to Traditionally 
Underserved Businesses 
and Counties in Phases 2 
and 3 
Lending Increased to 
Traditionally Underserved 
Businesses and Counties 

 
 
 
 
 
Page 21 
GAO-21-601  Paycheck Protection Program 
Table 3: Selected Paycheck Protection Program Changes  
Type of change 
Phase 
Description of change  
Lender participation 
1 
The Small Business Administration (SBA) released the nonbank lender application 
form allowing new nonbank lenders to participate in the Paycheck Protection Program. 
2 
SBA adjusted portfolio requirements for Community Development Financial Institutions 
(CDFI), majority minority-, women-, or veteran/military-owned financial institutions, and 
certain other nonbank lenders, which allowed smaller lenders to participate in PPP. 
Targeted funding and 
processing timelines 
2 
SBA established a $10 billion set-aside for businesses that applied through CDFIs to 
target lending to minority-owned businesses and traditionally underserved 
communities.a 
3 
SBA dedicated the first few days of Phase 3 to processing loans made through 
community financial institutions, including CDFIs, Minority Depository Institutions, 
Certified Development Companies, and Microloan Intermediaries. SBA also 
established a 14-day loan application period exclusively for businesses or nonprofits 
with fewer than 20 employees. 
Expanded business eligibility 
2 
SBA issued a rule instructing self-employed individuals to apply by calculating the loan 
amount based on the business’s net profit from 2019.  
3 
SBA revised the maximum loan amount calculation for self-employed individuals filing 
a Form 1040 Schedule C based on either gross income or net profit, which allowed for 
larger loans and enabled increased loan access for businesses with very little or 
negative net profit. 
Source: GAO analysis of SBA guidance.  |  GAO-21-601 
Note: Phase 1 includes the application period from April 3–16, 2020, Phase 2 from April 27–August 8, 
2020, and Phase 3 from January 11–May 31, 2021. 
aCongress established additional set-asides throughout the course of the program. These included a 
$60 billion set-aside for businesses that applied through community financial institutions, certain small 
depository institutions, and certain small credit unions established under the Paycheck Protection 
Program and Health Care Enhancement Act as well as $15 billion for lending by community financial 
institutions and $35 billion to new first draw PPP borrowers established under the Consolidated 
Appropriations Act, 2021. 
 
Self-employed individuals. While the share of loans to self-employed 
individuals was low in Phase 1, program changes helped increase their 
share substantially in later phases of the program. Specifically, loans to 
self-employed individuals increased from 9 percent of all loans in Phase 1 
to 86 percent in Phase 3 (see fig. 8).41 This increase corresponded with 
SBA issuing guidance for self-employed individuals to apply for loans and 
increasing the number and types of lenders in the program at the end of 
                                                                                                                       
41Census’s Statistics on U.S. Businesses includes data on the number of small 
businesses based on characteristics including business size and industry sector, but it 
does not include data on business ownership type that are comparable to SBA’s data on 
PPP recipients. As a result, we were not able to provide information on the share of these 
businesses relative to all small businesses in the U.S. The share of loans to businesses 
with fewer than 10 employees also increased across all three phases. 

 
 
 
 
 
Page 22 
GAO-21-601  Paycheck Protection Program 
Phase 1, and allowing for larger loan amounts for self-employed 
individuals in Phase 3. 
Figure 8: Percentage of First Draw Paycheck Protection Program Loans to Self-
Employed Individuals, by Program Phase 
 
Note: Self-employed individuals include independent contractors, sole proprietorships qualified joint 
ventures, and single-member limited liability companies. This analysis excludes loans that were 
approved and subsequently canceled. Generally, Phase 1 data include loans approved from April 3–
16, 2020, Phase 2 data include loans approved from April 27–August 8, 2020, and Phase 3 data 
include first draw loans approved from January 12–June 30, 2021. The U.S. Census Bureau’s 
Statistics on U.S. Businesses does not include data on the share of these businesses relative to all 
small businesses in the U.S. 
 
Women-owned businesses. The share of loans to businesses in 
counties with a large share of women-owned businesses also increased, 
potentially because of Congress’s and SBA’s modifications related to 
increasing lender participation and expanding business eligibility. 
Because SBA does not have complete data on the gender of business 
owners who obtained PPP loans, we used estimates from Census’s 
Annual Business Survey to assess the extent to which businesses in 
counties with large shares of women-owned businesses were able to 

 
 
 
 
 
Page 23 
GAO-21-601  Paycheck Protection Program 
access PPP.42 For the purposes of this analysis, we examined lending in 
counties with shares of women-owned businesses that were higher than 
the national estimate.43 We refer to these counties as having a “large 
share” of women-owned businesses. 
The share of loans to businesses in counties with large shares of women-
owned businesses doubled between Phase 1 and Phase 2, from 9 
percent to 18 percent (see fig. 9).44 This increase corresponded with the 
increase in the number and types of lenders after Phase 1 and targeted 
funding for traditionally underserved businesses and communities in 
Phases 2 and 3. In addition, approximately 90 percent of women 
business owners are self-employed individuals, according to Census 
data.45 As a result, program changes aimed at increasing participation 
among self-employed individuals likely also contributed to the increase in 
lending to counties with large shares of women-owned businesses. 
                                                                                                                       
42As we reported in July 2021, SBA did not require demographic information—for 
example, the race, gender, and veteran status of the business owner—in the PPP 
borrower application. However, SBA revised its PPP borrower application form in January 
2021 to allow for the optional disclosure of demographic information and included such a 
section in all subsequent revisions and an optional demographic information form as part 
of the PPP loan forgiveness application. See GAO-21-577. The Annual Business Survey 
provides information on selected economic and demographic characteristics for employer 
businesses (those with paid employees) and business owners by sex, ethnicity, race, and 
veteran status. We used county-level data from 2017, the most recently available data at 
the time of our review. While these data include employer businesses of all sizes, small 
businesses comprise more than 99 percent of all businesses in the U.S., and therefore are 
representative of small businesses.  
43For our analysis, we excluded counties whose estimated shares had a relative standard 
error greater than 20 percent. We determined counties to be similar to the national 
estimate if the 95 percent confidence interval surrounding their estimated share fell within 
the 95 percent confidence interval surrounding the national share estimate. Based on data 
from the 2017 Annual Business Survey, approximately 20 percent of all businesses were 
women-owned. 
44Sixteen percent of all small businesses are located in counties with above-national 
shares of women-owned businesses. 
45U.S. Census Bureau, Number of Women-Owned Employer Firms Increased 0.6% From 
2017 to 2018 (Washington D.C.: Mar. 29, 2021).  

 
 
 
 
 
Page 24 
GAO-21-601  Paycheck Protection Program 
Figure 9: Percentage of First Draw Paycheck Protection Program Loans to 
Businesses in Counties with Large Shares of Women-Owned Businesses, by 
Program Phase 
 
Note: Counties with large shares of women-owned businesses include counties where the share of 
women-owned businesses is higher than the national share at the 95 percent confidence level, based 
on the U.S. Census Bureau’s Annual Business Survey data. This analysis excludes loans that were 
approved and subsequently canceled and loans that could not be matched to data from the U.S. 
Census Bureau. Generally, Phase 1 data include loans approved from April 3–16, 2020, Phase 2 
data include loans approved from April 27–August 8, 2020, and Phase 3 data include first draw loans 
approved from January 12–June 30, 2021. Sixteen percent of all small businesses are located in 
counties with above-national shares of women-owned businesses. 
 
Businesses in high-minority counties. Small businesses in counties 
with a high share of minority residents (“high-minority counties”) received 
a relatively low share of loans in Phase 1, but they experienced greater 
access over time, potentially resulting from a number of program 
changes.46 Because SBA does not have complete data on the race of 
business owners who obtained PPP loans, we used Census’s American 
Community Survey to measure the percentage of loans in each phase 
                                                                                                                       
46Forty-seven percent of all small businesses are located in high-minority counties. 

 
 
 
 
 
Page 25 
GAO-21-601  Paycheck Protection Program 
that went to businesses in high-minority counties, or counties with a share 
of minority residents above the national share.47 
Based on our analysis, 36 percent of Phase 1 loans went to small 
businesses located in high-minority counties, but this increased to 50 
percent in Phase 2 (see fig. 10).48 Similarly, researchers found that 
businesses in high-minority counties received a lower share of Phase 1 
loans than low-minority counties, but a higher share of Phase 2 loans.49 
This trend corresponds with a general increase in the number and types 
of lenders in the program after Phase 1 and program changes to target 
funds to lenders that lend in traditionally underserved counties in Phases 
2 and 3. For example, in Phase 2, SBA established a $10 billion set-aside 
for businesses that applied through CDFIs, seeking to target lending to 
minority-owned businesses and traditionally underserved counties. 
                                                                                                                       
47As we reported in July 2021, SBA did not require demographic information—for 
example, the race, gender, and veteran status of the business owner—in the PPP 
borrower application. However, SBA revised its PPP borrower application form in January 
2021 to allow for the optional disclosure of demographic information and included such a 
section in all subsequent revisions and an optional demographic information form as part 
of the PPP loan forgiveness application. See GAO-21-577. The American Community 
Survey is an ongoing survey of about 3.5 million U.S. households that uses a series of 
monthly samples to produce annually updated estimates for census tracts across the U.S. 
The survey collects data on the economic, social, housing, and demographic 
characteristics of communities at various geographic levels, including metropolitan areas, 
states, and counties. We used county-level 2015–2019 5-year estimates, the most 
recently available data at the time of our review.  
48For this analysis, we define high-minority counties as counties where the percentage of 
minority residents was higher than the national share of 27.5 percent. 
49This research used zip code-level data from Census’s County Business Patterns on 
employer businesses to analyze PPP loans. R. Fairlie and F.M. Fossen, “Did the 
Paycheck Protection Program and Economic Injury Disaster Loan Program get disbursed 
to minority communities in the early stages of COVID-19?.” Small Business Economics 
(2021). 

 
 
 
 
 
Page 26 
GAO-21-601  Paycheck Protection Program 
Figure 10: Percentage of First Draw Paycheck Protection Program Loans to 
Businesses in High-Minority Counties, by Program Phase 
 
Note: High-minority counties were identified based on data from the U.S. Census Bureau’s American 
Community Survey and include counties with a share of minority residents greater than the national 
share of 27.5 percent. This analysis excludes loans that were approved and subsequently canceled 
and loans that could not be matched to the data from the U.S. Census Bureau. Generally, Phase 1 
data include loans approved from April 3–16, 2020, Phase 2 data include loans approved from April 
27–August 8, 2020, and Phase 3 data include first draw loans approved from January 12–June 30, 
2021. Forty-seven percent of all small businesses are located in high-minority counties. 
 
Loans to minority-owned businesses may also have been low in Phase 1 
because these businesses lacked pre-existing relationships with banks, 
particularly small banks. According to the Federal Reserve Banks’ 2019 
Small Business Credit Survey, prior to the pandemic, Black and Hispanic 
business owners were more likely to seek financing from large banks and 
online lenders than from small banks.50 The 2020 update to the survey 
found this trend continued during Phases 1 and 2: regardless of where 
businesses applied for PPP loans, those that had a relationship with a 
                                                                                                                       
50Federal Reserve Bank of Atlanta, Small Business Credit Survey: 2019 Report on 
Minority-Owned Firms (New York, NY: 2019). The 2018 survey was conducted in the third 
and fourth quarters of 2018 and generated 6,614 responses from employer firms with 
information on the race or ethnicity of a firm’s owner(s). 

 
 
 
 
 
Page 27 
GAO-21-601  Paycheck Protection Program 
bank were more likely to apply for PPP loans than those that did not.51 
The survey also demonstrated that minority-owned businesses were less 
likely to have a relationship with a bank. Our survey of PPP lenders found 
CDFIs and MDIs were more likely than other types of lenders to report 
that they accepted applications from borrowers with whom they had no 
prior relationship.52 
Additionally, when comparing high-minority counties by metro and rural 
status, we found that businesses in high-minority metro counties were 
underrepresented in Phase 1 relative to their share of total small 
businesses, even in high-minority metro counties with better 
socioeconomic outcomes. By contrast, businesses in low-minority metro 
counties received a share of loans higher than their share of small 
businesses in Phase 1 (see fig. 11). However, participation among 
businesses in high-minority metro counties increased in Phases 2–3. This 
suggests that program changes successfully targeted lending to 
businesses in high-minority counties. 
 
 
                                                                                                                       
51Federal Reserve Banks of Atlanta, Boston, Chicago, Cleveland, Dallas, Kansas City, 
Minneapolis, New York, Philadelphia, Richmond, St. Louis, and San Francisco, Small 
Business Credit Survey: 2021 Report on Firms Owned by People of Color (New York, NY: 
2021).  
52We surveyed a generalizable sample of 1,383 PPP lenders, stratified by lender type and 
size, to obtain their perspectives on the program. The survey closed on April 15, 2021, 
and we received 781 responses. We obtained a weighted response rate of 57.3 percent. 
Ninety percent of CDFI/MDIs reported accepting PPP applications from businesses with 
whom they had no prior relationship, compared to 73 percent of large banks. The 95 
percent confidence interval for these estimates are (83, 95) for CDFI/MDIs and (67, 79) for 
large banks. 

 
 
 
 
 
Page 28 
GAO-21-601  Paycheck Protection Program 
Figure 11: Percentage of First Draw Paycheck Protection Program Loans to 
Businesses in Metro Counties Relative to the Percentage of All Small Businesses, 
by Program Phase and Share of Minority Residents 
 
Note: We used data from the Department of Agriculture’s Economic Research Service to classify 
counties as “metro” if they have one or more high-density urban areas with 50,000 or more residents 
or are outlying counties that are economically tied to these central counties, as measured by the 
share of workers commuting on a daily basis to the central counties. Rural counties are outside the 
boundaries of metro areas and have no cities with 50,000 or more residents. High-minority counties 
were identified based on data from the U.S. Census Bureau’s American Community Survey and 
include counties with a share of minority residents greater than the national share of 27.5 percent. 
Forty-seven percent of all small businesses are located in high-minority counties. This analysis 
excludes loans that were approved and subsequently canceled and loans that could not be matched 
to the data from the U.S. Census Bureau. Generally, Phase 1 data include loans approved from April 
3–16, 2020, Phase 2 data include loans approved from April 27–August 8, 2020, and Phase 3 data 
include first draw loans approved from January 12–June 30, 2021. 
 
To increase lending to traditionally underserved businesses and counties, 
Congress and SBA modified PPP to include more lenders, particularly 
Nonbanks, CDFIs, and MDIs 
Substantially Increased 
Lending in Later Phases 

 
 
 
 
 
Page 29 
GAO-21-601  Paycheck Protection Program 
those that lend to traditionally underserved businesses.53 Nonbanks, 
CDFIs, and MDIs each made a greater percentage of PPP loans in 
Phases 2 and 3 (see fig. 12). Specifically, the percentage of loans made 
by nonbanks increased significantly, from 3 percent of loans in Phase 1 to 
23 percent of loans in Phase 3. In addition, the percentage of loans made 
by CDFIs and MDIs rose from 4 percent of loans in Phase 1 to 24 percent 
of loans in Phase 3. Further, beginning in Phase 2, SBA added 623 new 
lenders to make PPP loans, increasing the total number of lenders from 
4,837 to 5,460. CDFIs, MDIs, and nonbanks made 97 percent of the 
nearly 1.5 million loans issued by new lenders in the last two phases of 
the program. 
Figure 12: Percentage of Paycheck Protection Program Loans, by Lender Type and 
Program Phase 
 
Note: For this analysis, all depository lenders are categorized as “banks,” including banks, credit 
unions, and savings and loan associations. Large banks are as those with at least $10 billion in 
assets, medium banks are those with $1 to $10 billion, and small banks are those with less than $1 
billion based on publicly available call reports. All nondepository lenders are categorized as 
“nonbanks,” including SBA Small Business Lending Companies, SBA Certified Development 
Companies, SBA Microlenders, Business and Industrial Development Corporations, Farm Credit 
                                                                                                                       
53SBA and the Department of the Treasury were jointly responsible for approving lenders 
new to SBA to issue PPP loans. According to SBA officials, SBA approved new federally 
regulated lenders, and only new non-federally regulated and insured lenders required joint 
SBA and Treasury approval. 

 
 
 
 
 
Page 30 
GAO-21-601  Paycheck Protection Program 
System lenders, and state-regulated financial companies. This analysis excludes loans that were 
approved and subsequently canceled. Generally, Phase 1 data include loans approved from April 3–
16, 2020, Phase 2 data include loans approved from April 27–August 8, 2020, and Phase 3 data 
include first draw loans approved from January 12–June 30, 2021. 
 
Nonbanks, CDFIs, and MDIs made a higher proportion of loans to 
traditionally underserved businesses and counties than other types of 
lenders, particularly small banks. Specifically, self-employed individuals 
received 85 percent of nonbanks’ loans, but only 47 percent of loans 
overall, which suggests that nonbank lenders’ increased participation over 
time could have helped increase lending to these borrowers. In addition, 
21 percent of loans made by CDFIs or MDIs and 23 percent of loans 
made by nonbanks went to businesses in counties with large shares of 
women-owned businesses, compared to 4 percent of loans made by 
small banks.54 Similarly, 64 percent of loans made by nonbanks and 69 
percent of loans made by CDFIs and MDIs went to businesses in high-
minority counties, compared to 22 percent of loans made by small 
banks.55 These findings suggest the changes SBA and Congress made to 
increase lending to traditionally underserved businesses through 
nonbanks, CDFIs, and MDIs helped these businesses better access PPP. 
By the end of Phase 3, PPP lending to businesses in traditionally 
underserved counties was proportionate to their representation in the 
overall small business community, in part because of the changes 
Congress and SBA made to the program.56 For example, while loans to 
businesses in rural counties were high from the start of the program, 
loans to businesses in high-minority counties and counties with large 
shares of women-owned businesses reached proportionate levels by the 
end of Phase 3. However, while lending to businesses with fewer than 10 
                                                                                                                       
54By comparison, 19 percent of loans made by large banks and 10 percent of loans made 
by medium banks went to businesses in counties with large shares of women-owned 
businesses. 
55Fifty-four percent of loans made by large banks and 36 percent of loans made by 
medium banks went to businesses in counties with high minority populations. 
56For the purposes of this report, we refer to “traditionally underserved businesses and 
counties” more broadly as very small businesses; those owned by minorities, women, or 
veterans; or those located in rural counties, counties with high minority populations, or 
Tribal areas. In our analyses, we used businesses with fewer than 10 employees and 
those owned by self-employed individuals to represent very small businesses. To 
represent minority-, women-, and veteran-owned businesses, we used counties with high 
shares of minority residents and high shares of women- and veteran-owned businesses. 
We include Tribal areas when collectively referring to traditionally underserved counties. 
By Program Close, 
Lending in Traditionally 
Underserved Counties 
Was Comparable to Their 
Share of Small 
Businesses 

 
 
 
 
 
Page 31 
GAO-21-601  Paycheck Protection Program 
employees increased significantly over time, it remained 
disproportionately low by the end of Phase 3. 
Specifically, the share of loans to businesses in rural counties exceeded 
their share of all small businesses (see fig. 13). Overall, 15 percent of 
loans went to small businesses in rural counties, while 13 percent of all 
small businesses are located in these counties. 
Despite initial access challenges, lending to high-minority counties 
exceeded and lending to counties with large shares of women-owned 
businesses reached their share of small businesses by the program’s 
end. Businesses in high-minority counties received 50 percent of all loans 
and account for 47 percent of all small businesses nationwide. Overall, 
businesses in counties with large shares of women-owned businesses 
accounted for 16 percent of loans and represent 16 percent of small 
businesses. In addition, the share of loans to businesses in counties with 
large shares of veteran-owned businesses was generally commensurate 
with the share of small businesses in those counties. Overall, counties 
with large shares of veteran-owned businesses accounted for 4 percent 
of loans and 5 percent of small businesses. 
Businesses across the hardest-hit sectors received 30 percent of all 
program loans but account for 27 percent of small businesses. 
However, the share of loans to businesses with fewer than 10 employees 
remained below their share of small businesses overall, despite program 
changes to increase their participation over time. These businesses 
received 84 percent of all loans, but represent 96 percent of all small 
businesses. In particular, self-employed individuals received 47 percent of 
loans.57 By comparison, businesses with 10 to 499 employees received 
16 percent of all loans, despite accounting for just 4 percent of all small 
businesses. 
The proportionately low program participation for businesses with fewer 
than 10 employees and self-employed individuals could be, in part, 
because they were able to obtain other forms of assistance. For example, 
in July 2021, we reported that businesses with fewer than 10 employees 
accounted for 86 percent of approved applications in the Economic Injury 
                                                                                                                       
57Data were not available to calculate the percentage of all businesses that are self-
employed individuals.  

 
 
 
 
 
Page 32 
GAO-21-601  Paycheck Protection Program 
Disaster Loan program—an SBA-administered program that provided 
$230 billion in loans and advances to small businesses and nonprofits 
between March 2020 and May 2021.58 Further, many self-employed 
individuals may have been eligible for unemployment insurance.59 
Finally, businesses on Tribal lands received 1 percent of all PPP loans.60 
                                                                                                                       
58See GAO-21-589. 
59SBA has noted that a self-employed individual’s participation in PPP may affect eligibility 
for unemployment compensation. 
60Although SBA’s loan-level data include the type of businesses that received loans, these 
data did not consistently record whether businesses were Tribal businesses. Because 
these data were not available, we estimated the percentage of PPP loans that went to 
businesses located on Tribal lands. Tribal lands in our analysis included federally 
recognized American Indian reservations and off-reservation trust land areas, state-
recognized American Indian reservations, and Hawaiian homelands for which Census 
publishes data. Businesses located in these Tribal areas may or may not be Tribal 
businesses. In addition, data were not available to calculate the percentage of all 
businesses located on Tribal lands.  

 
 
 
 
 
Page 33 
GAO-21-601  Paycheck Protection Program 
Figure 13: Percentage of Paycheck Protection Program Loans Relative to the Share of Small Businesses, by Type of Business 
or County 
 
aSelf-employed individuals include independent contractors, sole proprietorships, qualified joint 
ventures, and single-member limited liability companies. Data were not available to calculate the 
percentage of all small businesses that were self-employed individuals. 
bHardest-hit sectors include businesses holding the sector classification code within the North 
American Industry Classification System identified by the Bureau of Labor Statistics’s 2020 Business 
Response Survey as those that were most likely to experience adverse effects to their business 
operations as a result of the COVID-19 pandemic. Collectively, businesses in these sectors received 
a proportionate share of PPP loans relative to their share of all small businesses, but businesses in 
certain sectors including educational services and arts, entertainment, and recreation received a 
lower share of PPP loans than their share of all small businesses. The Consolidated Appropriations 
Act, 2021, enacted December 27, 2020, established the Shuttered Venue Operators Grant program 
to target over $16 billion in grants to venue operators. Eligible businesses generally fall into the arts, 
entertainment, and recreation sector. 
cCounties with large shares of women- and veteran-owned businesses include counties in which the 
share of business ownership is higher than the national share at the 95 percent confidence level, 
based on Census’s Annual Business Survey. 
dHigh-minority counties include counties with a share of minority residents greater than the national 
share based on Census’s American Community Survey. 
eData were not available to calculate the percentage of all businesses that are Tribal businesses. 
fWe define all small businesses as employer businesses with fewer than 500 employees and all non-
employer businesses based on 2017 data from Census’s Statistics of U.S. Businesses and Non-
Employer Statistics. 

 
 
 
 
 
Page 34 
GAO-21-601  Paycheck Protection Program 
gThis analysis excludes loans that were approved and subsequently canceled and includes both first 
and second draw loans. For counties and Tribal areas, loans that could not be matched to Census 
data were excluded. 
 
Our analysis suggests that program changes made by Congress and 
SBA helped to increase access for the intended businesses and 
counties.61 We will continue to monitor and report on the use of PPP and 
related small business loan and grant funds through our government-wide 
reports. 
We provided a draft of this report to the Small Business Administration 
and the Department of the Treasury for review and comment. Both 
provided technical comments, which we incorporated as appropriate. 
We are sending copies of this report to the appropriate congressional 
committees, the Administrator of the Small Business Administration, and 
the Secretary of the Treasury. In addition, the report is available at no 
charge on the GAO website at https://www.gao.gov. 
If you or your staff have any questions about this report, please contact 
me at (202) 512-8678 or pendletonj@gao.gov. Contact points for our 
Offices of Congressional Relations and Public Affairs may be found on 
the last page of this report. GAO staff who made key contributions to this 
report are listed in appendix II. 
 
John H. Pendleton 
Director, Financial Markets and Community Investment 
 
 
                                                                                                                       
61Our analysis did not account for the total number of businesses that may have 
permanently closed as a result of the pandemic, or those that may have utilized other 
assistance instead of PPP. 
Agency Comments 
 

 
 
 
 
 
Page 35 
GAO-21-601  Paycheck Protection 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 
 
 

 
 
 
 
 
Page 36 
GAO-21-601  Paycheck Protection Program 
The Honorable Kyrsten Sinema 
Chair 
The Honorable James Lankford 
Ranking Member 
Subcommittee on Government Operations and Border Management 
Committee on Homeland Security and Governmental Affairs 
United States Senate 
The Honorable Rosa L. DeLauro 
Chairwoman 
The Honorable Kay Granger 
Ranking Member 
Committee on Appropriations 
House of Representatives 
The Honorable Frank Pallone, Jr. 
Chairman 
The Honorable Cathy McMorris Rodgers 
Republican Leader 
Committee on Energy and Commerce 
House of Representatives 
The Honorable Bennie G. Thompson 
Chairman 
The Honorable John Katko 
Ranking Member 
Committee on Homeland Security 
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 M. Velázquez 
Chairwoman 
The Honorable Blaine Luetkemeyer 
Ranking Member 
Committee on Small Business 
House of Representatives 

 
 
 
 
 
Page 37 
GAO-21-601  Paycheck Protection Program 
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 Earl Blumenauer 
House of Representatives 
The Honorable Steve Cohen 
House of Representatives 
The Honorable Peter A. DeFazio 
House of Representatives 
The Honorable Diana DeGette 
House of Representatives 
The Honorable Mike Doyle 
House of Representatives 
The Honorable John Garamendi 
House of Representatives 
The Honorable Raúl M. Grijalva 
House of Representatives 
The Honorable Jahana Hayes 
House of Representatives 
The Honorable Henry C. Johnson, Jr. 
House of Representatives 
The Honorable Barbara Lee 
House of Representatives 
The Honorable Mike Levin 
House of Representatives 

 
 
 
 
 
Page 38 
GAO-21-601  Paycheck Protection Program 
The Honorable James P. McGovern 
House of Representatives 
The Honorable Grace F. Napolitano 
House of Representatives 
The Honorable Eleanor Holmes Norton 
House of Representatives 
The Honorable Chellie Pingree 
House of Representatives 
The Honorable Ayanna Pressley 
House of Representatives 
The Honorable Mary Gay Scanlon 
House of Representatives 
The Honorable Jan Schakowsky 
House of Representatives 
The Honorable Adam Smith 
House of Representatives 
The Honorable Darren Soto 
House of Representatives 
The Honorable Jackie Speier 
House of Representatives 
The Honorable Rashida Tlaib 
House of Representatives 
The Honorable Paul Tonko 
House of Representatives 
The Honorable Norma J. Torres 
House of Representatives 
The Honorable Bonnie Watson Coleman 
House of Representatives 

 
 
 
 
 
Page 39 
GAO-21-601  Paycheck Protection Program 
The Honorable Frederica S. Wilson 
House of Representatives 

Appendix I: Objectives, Scope, and 
Methodology  
 
 
 
 
Page 40 
GAO-21-601  Paycheck Protection Program 
This report examines how small business and lender participation in the 
Paycheck Protection Program (PPP) evolved over time. 
To address this objective, we used loan-level PPP data provided by the 
Small Business Administration (SBA) on the nearly 12.5 million loans 
approved from April 3, 2020, through June 30, 2021. The dataset 
included 70 variables describing loans, small business borrowers, and 
lenders. In addition, SBA provided a list of lenders approved to participate 
in PPP. This dataset included 10 additional variables, including 
information on the lender type and size. 
For all analyses, we excluded loans that were approved and 
subsequently canceled (approximately 6 percent of all loans). To 
understand the extent to which different types of businesses and 
business owners were able to access the program, our primary unit of 
analysis was the number of loans (i.e., whether a business received a 
loan) rather than the loan amount, which is determined by a formula 
based on a business’s payroll and other expenses and can be sensitive to 
geographical variations in the cost of doing business. 
SBA’s loan-level PPP data included the date SBA approved the loan but 
not the date the application was received. As a result, we categorized 
loans into three phases based on the date SBA approved the loan. Phase 
1 is based on the program’s initial funding from the CARES Act and 
includes loans approved from April 3–16, 2020.1 Phase 2 is based on the 
program’s second round of funding through the Paycheck Protection 
Program and Health Care Enhancement Act and includes loans approved 
from April 27–August 8, 2020.2 Phase 3 is based on the program’s third 
round of funding through the Consolidated Appropriations Act, 2021, and 
American Rescue Plan Act of 2021, and includes loans approved from 
January 12–June 30, 2021.3 
                                                                                                                       
1CARES Act, Pub. L. No. 116-136, §§ 1102(b), 1107(a)(1), 134 Stat. 281, 301 (2020). The 
dataset includes 10 loans with approval dates on April 20-22, 2020, which we categorized 
as Phase 1. 
2Paycheck Protection Program and Healthcare Enhancement Act, Pub. L. No. 116-139,§ 
101(a) 134 Stat. 620, 620 (2020). The dataset includes 1 loan with an approval date of 
August 9, 2020, which we categorized as Phase 2. 
3Consolidated Appropriations Act, 2021, Pub. L. No. 116-260, div. N, tit. III, §323, 134 
Stat. 1182, 2018-22 (2020); American Rescue Plan Act of 2021, Pub. L. No. 117-2, § 
5001(d), 135 Stat. 4, 85. 
Appendix I: Objectives, Scope, and 
Methodology 
Small Business 
Administration Data 

Appendix I: Objectives, Scope, and 
Methodology  
 
 
 
 
Page 41 
GAO-21-601  Paycheck Protection Program 
We categorized lenders by type and size. In its list of approved PPP 
lenders, SBA indicated the lender category (depository or nondepository) 
and subcategory (lender type). SBA indicated whether a lender was a 
Community Development Financial Institution (CDFI) by matching its data 
to a list of CDFIs provided by the Department of the Treasury. SBA also 
indicated whether a lender was a Minority Depository Institution (MDI) by 
matching its data to MDI lists from the Federal Deposit Insurance 
Corporation, the Office of the Comptroller of the Currency, and the 
National Credit Union Administration. CFDIs and MDIs share a goal of 
expanding economic opportunity by providing loans to businesses and 
individuals in traditionally underserved areas. CDFIs include both 
depository institutions and nonbanks, while MDIs include only depository 
institutions. We categorized any lender that SBA indicated was a CDFI, 
an MDI, or both, as “CDFI/MDI.” 
For the remaining lenders, we categorized all depository lenders—banks, 
credit unions, and savings and loan associations—as “banks.” SBA 
provided the size of depository lenders using publicly available call 
reports, which we used to define large banks as those with at least $10 
billion in assets, medium banks as those with at least $1 billion but less 
than $10 billion in assets, and small banks as those with less than $1 
billion in assets. We categorized the remaining nondepository lenders as 
“nonbanks,” and these included SBA Small Business Lending 
Companies, SBA Certified Development Companies, SBA Microlenders, 
Business and Industrial Development Corporations, Farm Credit System 
lenders, and state-regulated financial companies. 
To understand the characteristics of businesses that received PPP loans, 
we generally analyzed each characteristic by program phase and lender 
type category. To provide a benchmark, we determined the total number 
of small businesses for each characteristic using U.S. Census Bureau 
(Census) data. Specifically, we defined small businesses as employer 
businesses with fewer than 500 employees and all non-employer 
businesses based on 2017 data from Census’s Statistics of U.S. 
Businesses and Non-Employer Statistics. Based on the data available, 
this provides the best estimate of the number of small businesses that 
were eligible for PPP loans, but it does not precisely capture the full 
universe of eligible businesses because businesses with 500 employees 
are excluded and we do not account for variations in SBA’s size 
standards and eligibility criteria across industries and business types. 
For business size, we defined the smallest businesses as those with 
fewer than 10 employees and analyzed those businesses by program 
Analysis of Business and 
Lender Characteristics 

Appendix I: Objectives, Scope, and 
Methodology  
 
 
 
 
Page 42 
GAO-21-601  Paycheck Protection Program 
phase. We also analyzed loans to self-employed individuals—which also 
include sole proprietors, independent contractors, qualified joint ventures, 
and single member limited liability companies—by program phase and 
lender category type. 
For business sector, we analyzed data from the U.S. Bureau of Labor 
Statistics’s 2020 Business Response Survey and found that businesses 
in the following six sectors were most likely to experience adverse effects 
to their business operations as a result of the COVID-19 pandemic: 
accommodation and food services; arts, entertainment, and recreation; 
educational services; health care; manufacturing; and retail trade. We 
considered these industries “hardest-hit sectors” and analyzed them by 
program phase and location. 
While the dataset provided by SBA included variables indicating the race, 
gender, and veteran status of the business owner, the value was missing 
or unreported for most loans (76 percent missing for race, 61 percent for 
gender, and 67 percent for veteran status). We determined these 
variables were not complete enough to analyze for the purposes of our 
analysis. Instead, we used estimates from Census’s 2018 Annual 
Business Survey to assess the extent to which counties with a high share 
of these businesses were able to access PPP. The survey provides 
information on selected economic and demographic characteristics for 
businesses and business owners by sex, ethnicity, race, and veteran 
status.4 We used county-level data covering 2017, the most recently 
available data at the time of our review. While these data include 
employer businesses of all sizes, small businesses comprise more than 
99 percent of all businesses in the U.S. 
We geocoded the loan-level PPP dataset and merged it with Census’s 
Annual Business Survey data to indicate the share of small businesses 
owned by women and veterans for the county in which the PPP recipient 
                                                                                                                       
4Census’s Annual Business Survey is an electronic survey that sampled approximately 
850,000 employer businesses in 2017. The sample is stratified by state, frame, and 
industry and is systematically sampled within each stratum. Business ownership is defined 
as having more than 50 percent of the stock or equity in the business and is categorized 
by (1) sex (male, female, or equally male/female); (2) ethnicity (Hispanic, equally 
Hispanic/non-Hispanic, or non-Hispanic); (3) race (White, Black or African American, 
American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, 
minority, equally minority/nonminority, or nonminority); and (4) veteran status (veteran, 
equally veteran/nonveteran, or nonveteran). Publicly held and other firms are not 
classifiable by sex, ethnicity, race, and veteran status. Firms equally male-/female-owned, 
equally minority-/nonminority-owned, and equally veteran-/nonveteran-owned are counted 
and tabulated as separate categories. 

Appendix I: Objectives, Scope, and 
Methodology  
 
 
 
 
Page 43 
GAO-21-601  Paycheck Protection Program 
business was located. Specifically, we compared each county’s share of 
women- and veteran-owned businesses to the national estimate and 
categorized each county as above, below, or similar to the national 
estimate. We then compared the share of PPP loans to businesses in 
counties with above-national shares of women and veteran business 
ownership to those same counties’ share of small businesses. For our 
analysis, we excluded counties whose share estimates had a relative 
standard error greater than 20 percent. We determined counties to be 
similar to the national average if the 95 percent confidence interval 
surrounding their estimated share fell within the 95 percent confidence 
interval surrounding the national estimate. For all surveys, we excluded 
Puerto Rico and U.S. territories from our analysis. 
To identify the characteristics of the counties served by PPP, we merged 
the geocoded loan-level PPP data with data from Census’s American 
Community Survey—an ongoing survey of around 3.5 million households 
across the U.S. We used county-level 2015–2019 5-year estimates, the 
most recently available data at the time of our review.5 We analyzed five 
socioeconomic indicators at the county level: median household income, 
poverty rate, unemployment rate, percentage of households receiving 
public assistance income, and percentage of households with no internet 
access. 
We compared each measure for the county in which a PPP recipient 
small business was located to the national measure, and categorized 
each small business as being located in county whose estimate is above 
or below the national estimate. To facilitate comparison across 
geographic areas, we calculated the number of loans per small business 
in a county using data from the 2017 Statistics of U.S. Businesses and 
2017 Nonemployer Statistics. We also used American Community Survey 
data to analyze the share of PPP loans that went to businesses in 
counties with minority population shares above or below the national 
estimate and the share that went to rural counties compared to metro 
counties. We referred to counties classified as “non-metro” by the U.S. 
Department of Agriculture’s Economic Research Service as “rural” and all 
                                                                                                                       
5The American Community Survey uses a series of monthly samples to produce annually 
updated estimates for different geographic units, including counties, across the U.S. The 
survey collects data on the economic, social, housing, and demographic characteristics of 
communities at various geographic levels, including metropolitan areas, states, and 
counties.  
Analysis of County 
Characteristics 

Appendix I: Objectives, Scope, and 
Methodology  
 
 
 
 
Page 44 
GAO-21-601  Paycheck Protection Program 
other counties as “metro.”6 We also analyzed the percentage of PPP 
loans that went to businesses located on Tribal lands, which may or may 
not be Tribal businesses.7 
To learn more about lenders’ experience with PPP applications and the 
recipients they served, we administered a web-based survey to a 
representative sample of PPP lenders.8 In the survey, we asked lenders 
about the number of applications received and approved, reasons for not 
approving an application, and the extent to which PPP borrowers had an 
existing relationship with the lender. We administered the survey from 
February 2021 to April 2021, and we collected information for the 14-
month period from March 2020 to April 2021. 
To identify the universe of PPP lenders, we used data provided by SBA 
on August 3, 2020, which contained 5,466 lenders. We stratified our 
sample (seven strata) and used a stratified random sample of each: (1) 
small banks and savings and loans, (2) medium banks and savings and 
loans, (3) large banks and savings and loans, (4) extra-large banks, (5) 
credit unions, (6) nonbank lending institutions, and (7) minority-owned 
depository institutions and community development financial institutions.9 
Our initial sample size was designed to achieve a stratum-level margin of 
error of no greater than plus or minus 8 percentage points for an attribute 
                                                                                                                       
6The U.S. Department of Agriculture’s Economic Research Service defines counties as 
either metro or non-metro areas based on population and commuting patterns. Metro 
counties are defined as having at least one urban area with a population of 50,000 or 
more, plus adjacent territory that has a high degree of social and economic integration 
with the urban core, as measured by commuting ties. Non-metro areas do not have an 
urban area with at least 50,000 people and do not have measured commuting ties to a 
metro county. 
7Tribal lands in our analysis included federally recognized American Indian reservations 
and off-reservation trust land areas, state-recognized American Indian reservations, and 
Hawaiian homelands for which Census publishes data.  
8We also used this survey to inform our July 2021 report on SBA’s implementation of 
PPP. See GAO, Paycheck Protection Program: SBA Added Program Safeguards, but 
Additional Actions Are Needed, GAO-21-577 (Washington, D.C.: July 29, 2021). 
Specifically, see app. I of that report for more details on the survey’s methodology and 
app. II for the survey’s full results. 
9We defined small banks and savings and loans as those with less than $250 million in 
total assets, medium banks and savings and loans as those with $250 million or more and 
less than $1 billion in total assets, large banks and savings and loans as those with $1 
billion or more and less than $250 billion in total assets, and extra-large banks as those 
with $250 billion or more in total assets.  
Lender Survey 

Appendix I: Objectives, Scope, and 
Methodology  
 
 
 
 
Page 45 
GAO-21-601  Paycheck Protection Program 
level at the 95 percent level of confidence. Our resulting sample size was 
1,383 and we received 781 survey responses. We obtained a weighted 
response rate of 57.3 percent.10 Because we followed a probability 
procedure based on random selections, our sample is only one of a large 
number of samples that we might have drawn. Since each sample could 
have provided different estimates, we express our confidence in the 
precision of our particular sample’s results as a 95 percent confidence 
interval (for example, plus or minus 8 percentage points). This is the 
interval that would contain the actual population value for 95 percent of 
the samples we could have drawn. Confidence intervals are provided with 
each sample estimate in the report. All survey results presented in the 
body of this report are generalizable to the estimated population of 5,451 
in-scope depository institutions, except where otherwise noted. 
To assess the reliability of the SBA data, we identified potential variables 
for use in our analyses and output statistics on these variables (e.g., 
frequencies of values, number of blanks or zero values, minimum, 
maximum, and mean) to identify any potential reliability concerns, such as 
outliers or missing values. We met with relevant SBA officials to discuss 
each of the variables to understand how SBA collected, used, and 
maintained the data; the reliability and completeness of key variables; 
reasons for any potential discrepancies we identified; and whether our 
understanding of the data and approach to analyzing them were accurate 
and reasonable. After these meetings, we requested updated versions of 
the data and updated our analyses accordingly. In the instances in which 
we identified potential outliers or errors, we discussed them with SBA, 
and when necessary, requested a correction. We also reviewed related 
interviews and data reliability checks conducted for previous reports in 
which we analyzed PPP data.11 We determined that all data elements we 
assessed were sufficiently appropriate and reliable for this report’s 
objectives. 
To assess the reliability of all Census data, we reviewed technical 
information for each data source. We determined the surveys were 
sufficiently reliable for the purposes of reporting business characteristics 
                                                                                                                       
10We used a weighted response rate because our survey sample incorporated strata with 
different probabilities of selection. A weighted response rate may more accurately reflect 
the level of participation. For example, large units that contribute relatively more to the 
estimate of a total would have a larger “weight” on the response rate. 
11For example, GAO, COVID-19: Federal Efforts Could Be Strengthened by Timely and 
Concerted Actions, GAO-20-701 (Washington, D.C.: Sept. 21, 2020). 
Data Reliability 

Appendix I: Objectives, Scope, and 
Methodology  
 
 
 
 
Page 46 
GAO-21-601  Paycheck Protection Program 
on the county level. Findings from each survey are subject to sampling 
errors. 
To further inform our work, we reviewed legislation, interim final rules, 
agency guidance, and academic studies, and interviewed SBA officials. 
We also reviewed and summarized relevant literature. 
We conducted this performance audit from July 2020 to September 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 I: Objectives, Scope, and 
Methodology  
 
 
 
 
Page 47 
GAO-21-601  Paycheck Protection Program 
John H. Pendleton, (202) 512-8678 or pendletonj@gao.gov 
 
In addition to the contact named above, Cory Marzullo (Assistant 
Director), Christopher Forys (Analyst in Charge), Vida Awumey, Irina 
Carnevale, Chelsea Carter, Rachel DeMarcus, Gita DeVaney, Jacob 
Fender, Toni Gillich, Ying Long, John Mingus, Julia Robertson, Rebecca 
Shea, Jena Sinkfield, Shenandoah Sowash, Tyler Spunaugle, Farrah 
Stone, and Seyda Wentworth made key contributions to this report. 
Appendix II: GAO Contacts and Staff 
Acknowledgments 
GAO Contact 
Staff 
Acknowledgments 
(104427) 

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

21-601-ppp-smallest-businesses.pdf

Size

1,810,103 bytes

SHA-256

017456fa09e84766978cae4bca5c4866e2f58e6c5ebdd39c1d6a2a02a28b56ff

Our copy

21-601-ppp-smallest-businesses.pdf

Original

www.gao.gov