Granja Makridis Yannelis Zwick Nber W27095 Did Ppp Hit the Target

Document type: Notice
Full text

NBER WORKING PAPER SERIES
DID THE PAYCHECK PROTECTION PROGRAM HIT THE TARGET?
João Granja
Christos Makridis
Constantine Yannelis
Eric Zwick
Working Paper 27095
http://www.nber.org/papers/w27095
NATIONAL BUREAU OF ECONOMIC RESEARCH
1050 Massachusetts Avenue
Cambridge, MA 02138
May 2020, Revised July 2020
We thank seminar participants at the University of Chicago Booth School of Business, the Stigler
Center Economic Effects of COVID-19 Workshop, the Federal Reserve Bank of New York, the
Federal Reserve Bank of Philadelphia and the Bank of Portugal as well as Scott Baker, Sylvain
Catherine, Jediphi Cabal, Raj Chetty, Mike Faulkender, Sam Hanson, Steve Kaplan, Mike
Minnis, Tiago Pinheiro, Larry Schmidt, Adi Sunderam, and Luigi Zingales for comments.
Laurence O’Brien and Igor Kuznetsov provided excellent research assistance. João Granja
gratefully acknowledges support from the Jane and Basil Vasiliou Faculty Scholarship and from
the Booth School of Business at the University of Chicago. Yannelis and Zwick gratefully
acknowledges financial support from the Booth School of Business at the University of Chicago.
We are grateful to the Small Business Administration, Womply, and Homebase for providing
data. This draft is preliminary and comments are welcome. The views expressed herein are those
of the authors and do not necessarily reflect the views of the National Bureau of Economic
Research.
At least one co-author has disclosed a financial relationship of potential relevance for this
research. Further information is available online at http://www.nber.org/papers/w27095.ack
NBER working papers are circulated for discussion and comment purposes. They have not been
peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies
official NBER publications.
© 2020 by João Granja, Christos Makridis, Constantine Yannelis, and Eric Zwick. All rights
reserved. Short sections of text, not to exceed two paragraphs, may be quoted without explicit
permission provided that full credit, including © notice, is given to the source.

Did the Paycheck Protection Program Hit the Target?
João Granja, Christos Makridis, Constantine Yannelis, and Eric Zwick
NBER Working Paper No. 27095
May 2020, Revised July 2020
JEL No. E6,E62,G2,G21,G28,G38,H25,H32,H81,I38
ABSTRACT
This paper takes an early look at the Paycheck Protection Program (PPP), a large and novel small
business support program that was part of the initial policy response to the COVID-19 pandemic.
We use new data on the distribution of the first round of PPP loans and high-frequency micro-
level employment data to consider two dimensions of program targeting. First, we do not find
evidence that funds flowed to areas more adversely affected by the economic effects of the
pandemic, as measured by declines in hours worked or business shutdowns. If anything, funds
flowed to areas less hard hit. Second, we find significant heterogeneity across banks in terms of
disbursing PPP funds, which does not only reflect differences in underlying loan demand. The
top-4 banks alone account for 36% of total pre-policy small business loans, but disbursed less
than 3% of all PPP loans in the first round. Areas that were significantly more exposed to low-
PPP banks received much lower loan allocations. We do not find evidence that the PPP had a
substantial effect on local economic outcomes—including declines in hours worked, business
shutdowns, initial unemployment insurance claims, and small business revenues—during the first
round of the program. Firms appear to use first round funds to build up savings and meet loan and
other commitments, which points to possible medium-run impacts. As data become available, we
will continue to study employment and establishment responses to the program and the impact of
PPP support on the economic recovery. Measuring these responses is critical for evaluating the
social insurance value of the PPP and similar policies.
João Granja
Booth School of Business
University of Chicago
5807 S. Woodlawn Avenue
Chicago, IL 60637-1610
joaogranja@chicagobooth.edu
Christos Makridis
MIT Sloan School of Management
christos.a.makridis@gmail.com
Constantine Yannelis
Booth School of Business
University of Chicago
5807 S. Woodlawn Avenue
Chicago, IL 60637
and NBER
constantine.yannelis@chicagobooth.edu
Eric Zwick
Booth School of Business
University of Chicago
5807 South Woodlawn Avenue
Chicago, IL 60637
and NBER
ezwick@chicagobooth.edu

1
Introduction
The COVID-19 pandemic triggered an unprecedented economic freeze and a massive immedi-
ate policy response. Among the ﬁrms most affected by the freeze were millions of small busi-
nesses without access to public ﬁnancial markets or other ways to manage short-term costs.
Without an existing system of social insurance to support these ﬁrms, policymakers rushed to
develop new programs to help contain the damage, culminating in the CARES Act.
This paper takes an early look at a large and novel small business support program that
was part of the initial crisis response package, the Paycheck Protection Program (PPP). The
PPP offered guaranteed loans to small businesses through the Small Business Administration
(SBA) in order to stimulate lending to liquidity constrained ﬁrms. The loans are forgivable if
ﬁrms do not lay off workers or change their compensation. Our ﬁrst goal is to describe the
targeting of the ﬁrst round of PPP funding. We then build on our targeting results to evaluate
the short-term economic impacts of the ﬁrst round (i.e., from April 2nd to May 2nd). As
data become available, we will continue to expand this work by studying the second round of
funding, the medium-term economic impacts, and ultimately the social insurance value of the
program.
We bring new data from two sources to study the PPP. First, we obtained data from the SBA
on the number and amount of PPP loans disbursed by each participating lender during the ﬁrst
round of the program. The data offer a clear look at which lenders were most active in disburs-
ing loans and at the geographic distribution of PPP lending across the U.S. economy. Second,
we obtained high-frequency employment data from Homebase, a software company that pro-
vides free scheduling, payroll reporting and other services to small businesses, primarily in the
retail and hospitality sectors. The granularity of the data, coupled with the focus on sectors
most adversely affected by the pandemic, allows us to trace out the response of employment,
wages, hours worked, and business closures in almost real-time and evaluate the effects of PPP
support.
In the ﬁrst part of the paper, we consider two dimensions of program targeting. First,
did the funds ﬂow to where the economic shock was greatest? A central policy goal is to
prevent unnecessary mass layoffs and ﬁrm bankruptcies by injecting liquidity into ﬁrms. These
potential beneﬁts are likely greatest in areas with more pre-policy economic dislocation and
disease spread. We ﬁnd no evidence that funds ﬂowed to areas that were more adversely
2

affected by the economic effects of the pandemic, as measured by declines in hours worked
or business shutdowns. If anything, we ﬁnd some suggestive evidence that funds ﬂowed to
areas less hard hit. The fraction of establishments receiving PPP loans is greater in areas with
better employment outcomes, fewer COVID-19 related infections and deaths, and less social
distancing.
Second, given that the PPP used the banking system as a conduit to access ﬁrms, what role
did the banks play in mediating policy targeting? Anecdotal evidence suggests some banks
were eager to participate in the program, while others were unable or unwilling to process large
numbers of loans in the short program window.1 Given the magnitude and pace of the evolving
pandemic and the resulting disruptions in the marketplace, it is important for policymakers to
understand whether banks of different sizes and lending strategies had equal access to the
lending program. In particular, we ask whether there are systematic differences in program
targeting at the aggregate level driven by bank behavior and then quantify the resulting bank
allocation effect on the labor market.
Lender heterogeneity in PPP participation appears to be one reason why we ﬁnd a weak
correlation between economic declines and PPP lending. We ﬁnd signiﬁcant heterogeneity
across banks in terms of disbursing PPP funds, which reﬂects more than mere differences in
underlying loan demand. For example, because of an asset cap restriction in place since 2018,
Wells Fargo disbursed a signiﬁcantly smaller portion of PPP loans relative to their market share
of small business loans. We construct a measure of geographic exposure to bank performance
in the PPP using the distribution of deposits across geographic regions. The measure exploits
the fact that most small business lending is local (Brevoort, Holmes and Wolken, 2010; Granja,
Leuz and Rajan, 2018), comparing lenders that did more PPP lending, relative to other small
business lending, versus their counterparts who did less. We ﬁnd that areas that were signiﬁ-
cantly more exposed to banks whose PPP lending shares exceeded their small business lending
market shares received disproportionately larger allocations of PPP loans.
Overall, our ﬁndings shed light on the nature of and mechanism for how the ﬁrst round
of PPP loans were distributed. PPP loans were disproportionately allocated to areas least af-
fected by the crisis: ﬁfteen percent of establishments in the regions most affected by declines in
1For example an article from Forbes notes that in the days preceding the launch of the program, Fifth-Third
Bank did not initially participate in the program, while Old National only processed loans for customers that
had an existing account. Bank of America was the ﬁrst bank to process PPP loans, and they only took loans
from customers with “pre-existing business lending and business deposit relationship with Bank of America, as of
February 15, 2020.”
3

hours worked and business shutdowns received PPP funding; in contrast, thirty percent of all
establishments received PPP funding in the least affected regions. A major factor behind this
pattern was the signiﬁcant heterogeneity in the intensity of PPP participation across lenders.
Our ﬁndings indicate that underperforming banks—whose participation in the PPP underper-
formed their share of the small business lending market—account for two-thirds of the small
business lending market, but only twenty percent of total PPP disbursements in the ﬁrst round.
The top-4 banks in the U.S. economy (JPMorgan Chase, Bank of America, Wells Fargo, and
Citibank) alone accounted for 36% of the total number of small business loans but disbursed
less than 3% of all PPP loans. These banks were disproportionately located in areas that re-
ceived less PPP funding. We also ﬁnd that, while the regional funding gap narrowed during
the second round, a large gap persisted across states for at least another month after the initial
funds were exhausted.
In the second part of the paper, we study the effect of the PPP on economic outcomes during
the ﬁrst round of the program. Our results on bank participation motivate a research design
to evaluate the PPP using bank-driven differences in regional exposure to the program. This
variation across regions allows us to isolate the effect of the PPP from differences in loan de-
mand or confounding correlations between PPP funding and local economic outcomes. Our
research design relies on the assumption that pre-policy bank deposit shares in particular re-
gions are not correlated with the various outcomes we study. This assumption holds once we
condition on relevant observables, such as the relationship between PPP funding and the initial
severity of the crisis. We use this research design to study business shutdowns, reductions in
hours worked, initial unemployment insurance (UI) claims, and small business revenues at the
congressional district and county levels.
We present two ﬁndings. First, we do not ﬁnd evidence that the PPP had a substantial
effect on local economic outcomes during the ﬁrst round of the program. We establish this
ﬁnding using weekly ﬁrm-level employment and shutdown data from Homebase, which covers
small retail establishments especially hard hit by the crisis. We conﬁrm the ﬁrm-level evidence
using initial unemployment insurance claims at the county level. Given that one motivation
for the program was to provide an escape valve for congested state UI systems, the absence
of a signiﬁcant effect on UI claims during the initial weeks of the program is striking. Our
conﬁdence intervals on employment outcomes are wide enough to permit modest effects of
the program, but precise enough to reject large effects. The fact that the program disbursed
4

signiﬁcant funds, yet had little effect on employment, leads to the natural question of what
ﬁrms did with additional cash on hand.
Our second ﬁnding draws on the Census Small Business Pulse Survey to show that PPP
funds during the ﬁrst round allowed ﬁrms to build up liquidity and to meet loan and other
non-payroll spending commitments. For these ﬁrms, the PPP may have strengthened balance
sheets at a time when shelter-in-place orders prevented workers from doing work and when
unemployment insurance was more generous than wages for a large share of workers. This
ﬁnding is important because it implies that, while employment effects are small in the short run,
they may well be positive in the medium run because ﬁrms are less likely to close permanently.
At the same time, because program eligibility was deﬁned broadly, many less affected ﬁrms
received PPP funding and may have continued as they would have in the absence of the funds,
either by spending less out of retained earnings or by borrowing less from other sources. Our
results likely reﬂect a mix of these factors.
This paper is part of a broader research program to evaluate the impacts of COVID-19
economic policy responses that target private ﬁrms and households. As data become available,
we will build on these ﬁndings to evaluate the employment responses, to look at ﬁrm closures,
and to study the speed and nature of the economic recovery. Measuring these responses is
critical for evaluating the social insurance value of the PPP and similar policies.
Related Literature.
This paper joins a literature focusing on how government interventions
following crises impact recovery and the broader economy. Agarwal, Amromin, Ben-David,
Chomsisengphet, Piskorski and Seru (2017) and Ganong and Noel (2018) study the impact of
mortgage modiﬁcations following the Great Recession. House and Shapiro (2008) and Zwick
and Mahon (2017) study the effect of ﬁscal stimulus in the form of temporary tax incentives
for business investment, and Zwick (Forthcoming) documents the role of delegated agents in
mediating take-up of tax-based liquidity support for small ﬁrms. Mian and Suﬁ(2012), Parker,
Souleles, Johnson and McClelland (2013), Kaplan and Violante (2014) and Baker, Farrokhnia,
Meyer, Pagel and Yannelis (2020b) study how stimulus payments following recessions affect
household consumption. This paper evaluates a very large stimulus program aimed at provid-
ing liquidity and support to small ﬁrms.
Speciﬁcally, this paper also joins a rapidly growing literature studying the impact of the
2020 COVID-19 pandemic on the economy.
Jones, Philippon and Venkateswaran (2020),
5

Barro, Ursua and Weng (2020), Eichenbaum, Rebelo and Trabandt (2020), and Berger, Herken-
hoff and Mongey (2020) study the macroeconomics of infectious disease outbreaks and Gorm-
sen and Koijen (2020) use asset prices to back out growth expectations.2 Baker, Bloom, Davis
and Terry (2020) study changes in risk expectations induced by the COVID-19 pandemic.
Barrios and Hochberg (2020), Allcott, Boxell, Conway, Gentzkow, Thaler and Yang (2020),
and Makridis and Rothwell (2020) study how partisan afﬁliation impacts the response to the
pandemic, and Coibion, Gorodnichenko and Weber (2020), Cajner, Crane, Decker, Grigsby,
Hamins-Puertolas, Hurst, Kurz and Yildirmaz (2020), and Bartik, Betrand, Lin, Rothstein and
Unrath (2020) use a combination of administrative and survey data to study short term la-
bor market impacts. Baker, Farrokhnia, Meyer, Pagel and Yannelis (2020a), Chetty, Friedman,
Hendren and Stepner (2020), and Cox, Ganong, Noel, Vavra, Wong, Farrell and Greig (2020)
study consumption during the COVID-19 pandemic using high-frequency household transac-
tion data, documenting a deeper decline in and slower recovery of consumption among higher
income areas and earners. Taking a more aggregate approach, Mulligan (2020) and Makridis
and Hartley (2020) estimate baseline annual GDP effects of COVID-19 of $7 and $2.14 trillion,
respectively. Guerrieri, Lorenzoni, Straub and Werning (2020) show how supply-side shocks
can generate substantial shocks to demand and aggregate output. Similarly, Papanikolaou and
Schmidt (2020) explore how industry differences in remote work can explain revenue fore-
casts and other supply-side disruptions, building on approaches to measuring the task content
of occupations as in Dingel and Neiman (2020) and Gallipoli and Makridis (2018). Humphries,
Neilson and Ulyssea (2020) take a different approach using a social media survey on nearly
3,000 small and medium size businesses to document large differences in expectations and
awareness of the PPP. We join this emerging literature by providing early microeconomic ev-
idence on how ﬁrms and employees were affected as a function of credit supply in the ﬁrst
stages of the pandemic.
Finally, the paper joins work studying loan guarantees, an important and widely used form
of government intervention in credit markets. Classic work such as Smith (1983), Gale (1990)
and Gale (1991) focus on modeling government credit interventions such as loan guarantees.
2A related emerging empirical literature also explores the role of non-pecuniary factors, like social capital, as
mediating factors during the pandemic. For example, Makridis and Wu (2020) show how counties with greater
social capital have had fewer infections and a slower spread of infection, conditional on local demographic factors.
Similarly, Barrios, Benmelech, Hochberg and Zingales (2020) and Ding, Levine, Lin and Xie (2020) both show
how areas with greater social capital were more likely to engage in disease-mitigating activities because their
residents have greater trust in institutions and collective efﬁcacy.
6

Early empirical work focused on loan guarantee programs in France (Lelarge, Sraer and Thes-
mar, 2010). Recent theoretical work has focused on government guarantees to banks (Atke-
son, d’Avernas, Eisfeldt and Weill, 2018; Kelly, Lustig and Van Nieuwerburgh, 2016), economic
stimulus (Lucas, 2016) and a burgeoning empirical literature examines the effects of loan guar-
antees on credit supply, employment and small business outcomes (Bachas, Kim and Yannelis,
2020; Barrot, Martin, Sauvagnat and Vallee, 2019; Mullins and Toro, 2017; Gonzalez-Uribe
and Wang, 2019). Barrios, Minnis, Minnis and Sijthoff (2020) provide an empirical frame-
work and Elenev, Landvoigt and Van Nieuwerburgh (2020) focus on theoretical considerations
for assessing the optimal targeting of PPP loans during the pandemic and its role in expanding
credit supply to small versus larger ﬁrms that differ in their liquidity constraints. Cororaton and
Rosen (2020) examine the ﬁrm characteristics of public ﬁrms that have received PPP loans, un-
derscoring the importance of targeting loans towards the ﬁrms that need liquidity most. These
patterns contrast in some ways with the rollout of a similar lending program in Italy where
smaller ﬁrms and those in more adversely affected areas were more likely to receive support
(De Marco, 2020). Along these lines, we provide a comprehensive assessment of an impor-
tant and large loan guarantee program, evaluating its impacts during a period of economic
contraction and uncertainty.
The remainder of this draft is organized as follows. Section 2 describes the PPP. Section
3 discusses the main data sources used. Section 4 describes how the distribution of relative
performance in the PPP is correlated with bank and other characteristics. Section 5 documents
how differences across banks in PPP activity imply geographic differences in PPP exposure.
Section 6 explores the implications for PPP targeting to different geographic areas. Section
7 analyzes the early effects of the PPP on employment, and section 8 explores mechanisms.
Section 9 concludes.
2
The Paycheck Protection Program (PPP)
The Paycheck Protection Program (PPP) began on April 3rd, 2020 as part of the 2020 CARES
Act as a temporary source of liquidity for small businesses, authorizing $349 billion in forgiv-
able loans to help small businesses pay their employees and additional ﬁxed expenses during
the COVID-19 pandemic. Firms apply for support through banks and the Small Business Ad-
ministration (SBA) is responsible for overseeing the program and processing loan guarantees
7

and forgiveness. An advantage of using the banking system (including FinTech) as a conduit
for providing liquidity to ﬁrms is that, because nearly all small businesses have pre-existing
relationships with banks, this connection could be used to ensure timely transmission of funds.
The lending program is generally targeted toward small businesses of 500 or fewer employ-
ees.3 Although the initial round of funding was exhausted on April 16th, funds were drawn
from the Economic Injury Disaster Loan Program (EIDL) in the interim to continue funding
small businesses until the second round of $310 billion in PPP funding was passed by Congress
as part of the fourth COVID-19 aid bill.4 Small businesses were eligible as of April 3rd and
independent contractors and self-employed workers were eligible as of April 10th.
The terms of the loan are the same for all businesses. The maximum amount of a PPP
loan is the lesser of 2.5 times the average monthly payroll costs or $10 million. The average
monthly payroll is based on prior year’s payroll after subtracting the portion of compensation
to individual employees that exceeds $100,000.5
The interest rate on all loans is 1% and
their maturity is two years. The loans will be forgiven if two conditions are met. First, the
loan proceeds must be used to cover payroll costs, mortgage interest, rent, and utility costs
over the eight-week period following the provision of the loan, but not more than 25 percent
of the loan forgiveness amount may be attributable to non-payroll costs. Second, employee
counts and compensation levels must be maintained. If companies cut pay or employment
levels, loans may not be forgiven.6 However, if companies lay off workers or cut compensation
between February 15th and April 26th, but subsequently restore their employment levels and
employee compensation, their standing can be restored.
An important feature of the program is that the SBA waived its standard “credit elsewhere”
test used to grant regular SBA 7(a) loans. This test determines whether the borrower has the
ability to obtain the requested loan funds from alternative sources and amounts to a signiﬁcant
barrier in the access to regular SBA loans. Instead, in the PPP, applicants were only required to
3A notable exception was made for ﬁrms operating in NAICS Code 72 (accommodations and food services),
which are eligible to apply insofar as they employ under 500 employees per physical location. Firms whose
maximum tangible net worth is not more than $15 million and average net income after Federal income taxes
(excluding any carry-over losses) of the business for the two full ﬁscal years before the date of the application is
not more than $5 million can also apply. See the SBA for further information about the program.
4Recipients of an EIDL loan can receive a $10,000 loan advance that does not need to be paid back. The EIDL
loan itself is capped at a maximum of $2 million, is not forgivable, and the funds can be used ﬂexibly for operating
expenses.
5Payroll costs include wages and salaries but also payments for vacation, family and medical leave, healthcare
coverage, retirement beneﬁts, and state and local taxes.
6Loan payments on the remainder of the loan can be deferred for six months and interest accrues at 1%.
8

provide documentation of their payroll and other expenses, together with a simple two-page
application process where they certify that the documents are true and that current economic
uncertainty makes this loan request necessary to support ongoing operations. In sum, the PPP
program was designed to be a “ﬁrst-come-ﬁrst-served” program with eligibility guidelines that
allowed it to reach a broad spectrum of small businesses.
During the ﬁrst weeks of April, demand for PPP loans outstripped supply, which was limited
by statute. Between April 3 and 16 all of the initial $349 billion was disbursed, and the program
stopped issuing loans for a period of time. The House and Senate passed a bill to add an
additional $320 billion in funding on April 21 and 23 respectively, which was signed into law
on the 24th. The PPP began accepting applications on April 27 for the second round of funding.
Under the second round, funds were disbursed much less quickly, with unallocated PPP funds
being available in late June.
3
Data
We obtained conﬁdential data on the number of approved PPP loans and approved PPP amounts
from the Small Business Administration during the ﬁrst round through a Freedom of Informa-
tion Act request. The data set contains information on the amounts and number of loans
approved by each lender, amounts and number of loans received by small businesses in each
state, and total amounts and number of PPP loans received by small businesses in each con-
gressional district as of April 15, 2020. The PPP loan amounts in our records account for 336
billion of the 349 billion allocated to the program under the ﬁrst round of the CARES Act.
We hand-match this data set with the Reports of Condition and Income (Call Reports) ﬁled
by all active commercial banks as of the fourth quarter of 2019. We are able to match 4,228
out of 4,980 distinct participants in the PPP program to the Call Reports data set. We did not
match 1,031 commercial and savings banks that ﬁled a Call Report in the fourth quarter of
2019. We assume that these banks did not participate in the PPP program and made no PPP
loans. We further classiﬁed 631 PPP program participants as credit unions and the remaining
121 participants as non-bank PPP lenders. This group includes small community development
funds, as well as some large non-bank Fintech lenders. The commercial banks in the PPP
sample that we matched to the Call Report account for 93.7% of all PPP loans and 96.8% of
the total amount of loans disbursed under the PPP. By contrast, the group of PPP lenders that
9

we classiﬁed as credit unions accounted for 3.3% of all loans and 1.4% of the total PPP lending
amounts. Non-bank PPP lenders accounted for 3% of all loans and 1.8% of total PPP lending
amounts.
We obtain ﬁnancial characteristics of all banks from the Call Report ﬁled by commercial
and savings banks in the fourth quarter of 2019. The Call Report provides detailed data on the
size, capital structure, and asset composition of each commercial and savings bank operating
in the United States. Importantly, we obtain information on the number and amount of small
business loans outstanding of each commercial and savings bank from the Loans to Small
Business and Small Farms Schedule of the Call Reports. Using this information, we benchmark
the participation of all commercial and savings banks in the PPP program relative to their share
of the small business lending market.
To compute measures of exposure of each state, congressional district, and county to PPP
lenders, we match the matched-PPP-call reports data set with Summary of Deposits data con-
taining the location of all branches and respective deposit amounts of all depository institutions
operating in the United States as of June 30th, 2019. A signiﬁcant number of depository insti-
tutions merged in the second half of 2019, which means that some branches are assigned to
commercial and savings banks that no longer exist as stand-alone institutions. Notably, Sun-
Trust Banks, Inc. merged with Branch Banking and Trust Company (BB&T) to create the sixth
largest ﬁnancial institutions in the United States. We use the bank mergers ﬁle from the Na-
tional Information Center to adjust the branch network of merged institutions and account for
these mergers. We take advantage of the idea that most small business lending is mostly local
(e.g. Granja, Leuz and Rajan (2018)) and use the distribution of deposits across geographic
regions to create a measure of exposure of these regions to lenders that did more versus less
PPP lending than the expected small business lending share benchmark.
To evaluate whether PPP amounts were allocated to areas that were hardest-hit by the
COVID-19 crisis, we use data from multiple available sources on the employment, social dis-
tancing, and health impact of the crisis. We obtained detailed data on hours worked among
employees of ﬁrms that use Homebase to manage their scheduling and time clock. Homebase
processes exact hours worked by the employees of a large number of businesses in the United
States. We use information obtained from Homebase to track employment indicators at a daily
frequency across different states and congressional districts. The Homebase data set dispro-
portionately covers small ﬁrms in food service, retail, and other sectors (Bartik, Bertrand, Lin,
10

Rothstein, and Unrath, 2020).7 We complement the Homebase data set with ofﬁcial weekly
state unemployment insurance ﬁlings from the Department of Labor. We conduct our main
analyses at the congressional district level because that is the ﬁnest level of geographic disag-
gregation for which we have aggregate PPP lending. We use data from the County Business
Patterns dataset to approximate the number of establishments in the congressional district and
compute measures of the average amount of PPP lending per establishment and the fraction
of establishments receiving PPP loans in the region.
Because Homebase primarily covers small retail, beverage, and food service establishments,
it is not representative of aggregate employment. Nevertheless, the Homebase data are quite
useful for evaluating the employment impacts of the PPP speciﬁcally, as many hard-hit ﬁrms
are in the industries Homebase does cover and much of the early employment losses came from
these ﬁrms.8
To broaden our picture of the employment effects, we supplement the Homebase data with
two additional data sources. First, we obtain county-by-week initial unemployment insurance
claims from state web sites or by contacting state employment ofﬁces for data. We utilize
initial unemployment insurance claims as a measure of ﬂows into unemployment. We addi-
tionally supplement the Homebase data with data from Womply, a company that aggregates
data from credit card processors. The Womply data includes card spending at small businesses
at the county level, deﬁned by the location where a transaction occurred. Small businesses are
deﬁned as businesses with revenues below SBA thresholds.
To understand the mechanisms underlying our results, we also draw on newly-available
data from the Census Bureau’s Small Business Pulse Survey (SBPS), launched within seven
weeks of the national emergency declaration in March (Bufﬁngton, Dennis, Dinlersoz, Foster
and Klimek, 2020). To obtain real-time information tailored towards small businesses, the
SBPS is run weekly and entirely online that takes less than 10 minutes to complete and relies
largely on checkboxes, rather than numerical responses. Rather than contacting businesses
with an ofﬁcial letter, they were reached via email based on the Census Bureau’s Business
Register that has been populated using responses to the Economic Census across the 50 states
7More information about Homebase can be found at www.joinhomebase.com.
8We note that we were made aware of an error in the current raw formulation of the HomeBase data. This
resulted in doubling the number of hours three sectors: retail, professional services and transportation. The error
only affects the level, not the relative difference, across industries or states and hence should not change our two
primary outcomes, business shutdowns and the ratio of hours worked. However, these results should be treated
with some caution and the issue is scheduled to be corrected in the next data release.
11

(and D.C. and Puerto Rico). Furthermore, the SBPS focuses on businesses with receipts that
are greater than or equal to $1,000, but retain 500 employees or fewer. Each week, the sample
weights are adjusted to maintain representativeness.
While we focus our analysis on aggregations of small businesses across states and metropoli-
tan areas over time, we observe a wide array of information. For example, we observe changes
in revenues (increase, decrease, and no change) and the actual level of revenues. We also
observe how ﬁrms are changing their labor force through temporary business closures, em-
ployment, or hours worked. Finally, and most importantly for our purposes, we draw upon
responses to questions about liquidity (e.g., cash on hand), loans, defaults, and applications
for ﬁnancial assistance. We observe separate responses to questions about use of the PPP, EIDL,
and SBA loan forgiveness programs. We note that Humphries, Neilson and Ulyssea (2020) and
Bartik, Bertrand, Cullen, Glaeser, Luca and Stanton (2020) have pursued similar approaches
through other channels aimed at gathering information about small businesses.
Finally, we obtain counts of COVID-19 cases by county and state from the Center for Dis-
ease Control and use data on the effectiveness of social distancing from Unacast. Unacast
provides a social distancing scoreboard that describes daily changes in average mobility. Un-
acast measures the change in average distance travelled using individual’s GPS signals. The
data is availably on a daily basis, at the county level. We obtain information on the effective
dates of statewide shelter-in-place orders from the New York Times.9
4
PPPE and Bank Characteristics
4.1
Paycheck Protection Program Exposure
Table 1 shows summary statistics for the 20 largest ﬁnancial institutions in the United States,
as measured by total assets. The left-most column gives the institution’s name, while the ﬁrst
column of the table gives total assets as of the fourth quarter of 2019. The second and third
columns, respectively, show the share of total PPP volume and the share of the small business
loan (SBL) market of each institution. The ﬁfth column presents relative bank performance
9The New York Times presents some aggregates on shelter-in-place orders.
12

which is measured as
PPPEb = ShareVol.PPP −ShareSBLMarket
(ShareVol.PPP + ShareSBLMarket) × 0.5
(1)
where ShareVol.PPP the share of PPP volume held by bank b, and ShareSBLMarket is their
total small business loan volume market share. The next three columns present similar infor-
mation to columns (2) through (4), using the market share of total number of loans rather
than their volume, where
PPPEb = ShareN br.PPP −ShareN br.SBLMarket
(ShareN br.PPP + ShareN br.SBLMarket) × 0.5.
(2)
Here ShareN br.PPP the share of the number of PPP loans held by bank b, and the term
ShareN br.SBLMarket is their total small business loan market share, based on the number
of loans outstanding in each bank’s balance sheet as of the fourth quarter of 2019.
Figure 1 shows the distribution of relative performance in the PPP comparing PPP market
share against the overall stock of small business loans. The top panel shows the distribution
of relative bank performance in the PPP, based on the total volume of PPP loans granted. The
bottom panel shows the distribution of relative bank performance in the PPP, based on the
number of PPP loans granted. Both ﬁgures show a wide dispersion of relative performance,
with the distribution of performance based on the total number of loans showing greater mass
at the tails.
Figure 2 plots the bank measure of relative performance in the PPP by percentile of bank
size. The top panel shows relative performance based on the total volume of PPP loans granted,
the middle panel shows relative performance based on the number of PPP loans issued, while
the bottom panel shows average PPP loan size. All three panels indicate a similar pattern—
larger banks issued more PPP loans than expected given their share of the small business mar-
ket, whether by volume, number of loans, or average loan size. This pattern could be consistent
with larger banks being better suited to take advantage of the PPP program as it was rolled
out. This pattern reverses at the very top of the bank size distribution. The very largest banks,
those in the top percentile of bank size, signiﬁcantly underperformed in PPP lending relative
to their pre-policy share of small business lending. This underperformance is clear both in
terms of lending volume (Panel A) and number of loans (Panel B). Panel C suggests that the
underperformance of the top percentile of banks occurs despite their making the largest PPP
13

loans in the sample in terms of average loan amount.
Figure 3 provides evidence concerning the signiﬁcant dislocations between the share of PPP
lending of underperforming banks and the share of PPP that we would expect had these banks
issued PPP loans in proportion to their share of the small business lending market. The blue
hollow triangles and red hollow circles represent, respectively, the cumulative share of the PPP
and small business lending of banks whose PPPE is below a certain threshold. The ﬁgure shows
that commercial and savings banks, representing 20% of the small business lending market,
simply did not participate in the PPP lending program, (PPPE = −0.5). The plot further shows
that the group of banks whose PPP share is below their share of the small business lending
market, (PPPE < 0), made less than 20% of the PPP loans, but account for approximately
two-thirds of the entire small business lending market. Overall, the evidence is consistent with
substantial heterogeneity across lenders in their responses to the program’s rollout.
The fact that lenders were signiﬁcantly heterogeneous in accepting and processing PPP
loans would not necessarily result in aggregate differences in PPP lending across geographic
areas if small businesses could easily substitute and place their PPP applications to lenders
that were willing to accept and quickly expedite them. If many lenders, however, prioritize
their existing business relationships in the processing of PPP applications, ﬁrms’ pre-existing
relationships might determine to a large extent whether they are able to tap into PPP funds. In
this case, the exposure of geographic areas to banks that over/underperformed in the deploy-
ment of the PPP might signiﬁcantly determine the aggregate PPP amounts received by small
businesses located in these areas. Next, we examine if geographic areas that were exposed to
banks with weak PPP performance received less PPP lending overall.
5
Geographic Exposure to Bank PPP Performance
We next explore how the geography of the PPPE is related to PPP lending outcomes. Figure 4
presents a map of county level exposure to PPPE based on the share of deposits of each bank
in the county. Exposure varies across the United States with some Western areas containing
a large Wells Fargo presence that exhibit lower levels of PPPE, suggesting greater exposure
to lenders that underperformed in the PPP program relative to their small business lending
benchmark. By contrast, the counties with lower median household income (ρ = −0.13), a
lower share of college educated individuals (ρ = −0.16), and a smaller COVID-19 shock were
14

more likely to be exposed to lenders that overperformed in the PPP roll-out.
Figure 5 explores the relationship between PPPE exposure and PPP lending. The top panel
of Figure 5 plots aggregate PPP volume per small business by exposure to PPPE for each state,
while the bottom panel shows the fraction of all small businesses receiving PPP loans in the
state. Both panels present a similar pattern—there is a strong positive relationship between
PPP lending and PPPE exposure at the state level. States with higher exposure to banks that
performed well in terms of PPPE also saw greater PPP lending. Figure 7 presents a similar
pattern at the congressional district level, and a similar correlation emerges.
Table 2 makes this graphical evidence explicit. The top panel shows the relationship be-
tween PPPE and aggregate lending, at the congressional district level. Column (1) shows
the correlation between aggregate PPP lending and PPPE at the congressional district level.
The relationship is highly statistically signiﬁcant, with an F-statistic of approximately 45. A
one-standard deviation increase in the congressional district exposure to PPPE based on total
amounts of outstanding PPP and small business loans and weighted by the share of deposits
of each bank leads to a 16.1 percent increase in PPP lending. Column (2) adds in aggregate
employment and payroll controls, and column (3) adds in industry shares. The results remain
highly statistically signiﬁcant at the 1 percent level. Column (4) shows that the correlation
holds even within states when we add state ﬁxed effects, although the coefﬁcient remains
signiﬁcant at only the 10 percent level.
The bottom panel repeats the analysis, replacing aggregate lending per business with the
fraction of establishments receiving PPP loans in each congressional district. This panel also
indicates a very strong relationship between our PPPE measure and lending. In column (1),
the ﬁrst stage F-statistic is now above 200, and even with state ﬁxed effects in column (4) the
relationship is highly statistically signiﬁcant at the 1 percent level. A one-standard deviation
increase in our measure of congressional district exposure to bank PPP performance is associ-
ated with an increase of 4.3 percentage points in the fraction of establishments receiving loans
in a congressional district. These results suggest that businesses were much more likely to re-
ceive a PPP loan simply because they were located closer to banks that processed a large share
of PPP loans relative to their benchmark share of small business loans.
A potential concern with the above results is that the causality runs reverse. That is, banks
do relatively better where demand for PPP loans is abundant. To address this concern, we
15

compare survey measures on ﬁrm applications and PPP receipt.10 The Census Small Business
Pulse survey measures small businesses conditions in relation to the COVID-19 crisis at the
state by sector level, including PPP application and receipt. We compare the difference between
application and receipt to our measure of bank exposure.
The top panel of Figure 6 shows volume-based PPPE exposure and the difference between
the percentage of businesses reporting having applied to PPP and percentage of business that
received PPP in each state. The bottom panel shows PPPE based on the number of loans and
the difference between the percentage of businesses reporting having applied to PPP and the
percentage of business that received PPP in each state. Both panels show a very similar pattern,
the difference between PPP application and receipt is much lower in states with higher exposure
to banks that allocated more PPP funds. In other words, conditional on applying, businesses
were more likely to receive PPP funds in states where there were more banks allocating funds.
Appendix A presents a case study, which demonstrates the importance of ﬁnancial interme-
diation in the allocation of PPP funds. Wells Fargo was severely constrained from expanding
its balance sheet as a result of an asset cap imposed by the Fed in the aftermath of the fake
accounts scandal. This asset cap was only lifted on April 10 when the Fed excluded PPP loans
from the formula it uses to restrict Wells Fargo’s growth. The asset cap limited Wells Fargo’s
ability to lend under the PPP in the early days for the ﬁrst phase of the program. Wells Fargo
performed poorly in terms of allocating PPP funds during the ﬁrst wave and ﬁrms located in
areas with higher Wells Fargo market share see lower PPP allocations, both in terms of overall
aggregate loan volume per business and in the fraction of businesses receiving PPP loans.
The collection of results in this section suggests that exposure to bank-speciﬁc heterogeneity
in their willingness and ability to extend PPP loans was a signiﬁcant determinant of the allo-
cation of PPP loans in the economy. Next, we examine how the PPP allocation and exposure
to over/underperforming banks correlated with the local magnitude of the epidemic.
6
Are PPP Allocations Targeted to the Hardest Hit Regions?
Were PPP funds disbursed to geographic areas that were most affected by the epidemic? Figure
8 shows the relationship between PPP allocations, exposure to Bank PPP performance, and the
fraction of businesses in each congressional district that shut down during the week of March
10Census Small Business Pulse survey data can be found here.
16

29th to April 4th, just before PPP funds were disbursed.11 We estimate business shutdowns
in the congressional district using the high-frequency data set obtained from Homebase. The
ﬁgure indicates little if any correlation between PPP allocation or relative bank performance
and hours worked or business shutdowns. In Figure 9, we follow Bartik, Bertrand, Cullen,
Glaeser, Luca and Stanton (2020) and repeat the analysis using the ratio of hours worked on
March 31st, 2020 relative to a baseline of the average hours worked in the same weekday of
the last two weeks of January. Again, we ﬁnd that PPP allocations across congressional districts
are very weakly correlated to the impact of the epidemic crisis on labor markets and aggregate
ﬁrm outcomes. We have also investigated the robustness of these results to alternative cutoffs
when measuring the decline in hours worked and employment (e.g., as of April).12
To better illustrate the relation between ﬁrm and employment outcomes at the congres-
sional district level and PPP allocations, we stratify congressional districts into 20 bins based
on the impact of the COVID-19 epidemic on the fraction of businesses that shut down and on
the average decline in hours worked in the congressional district. Figure 10 plots the average
fraction of business receiving PPP loans in each business shutdown bin (top panel) and hours
worked bin (bottom panel). The plots suggest that approximately 15% percent of businesses
located in the most affected congressional districts were able to obtain PPP funding until April
15th, 2020. By contrast, more than 30% of all businesses operating in the least affected con-
gressional districts were able to tap into PPP funding.
The results suggest that PPP funds were not targeted towards geographic areas that were
most affected by the pandemic, at least in terms of small business employment drops. This fact
could be a result of the pre-existing bank relationships across counties, rather than a problem
with implementation: banks were caught off guard by the pandemic and the corresponding
actions taken to social distance. A related factor likely inﬂuencing these geographic patterns is
differential loan demand in harder hit areas. Because PPP support is more generous for ﬁrms
that maintain their payroll, the program likely appealed more to ﬁrms with smaller reductions
in their business. To the extent these geographic patterns reﬂect such differences in loan de-
11Following Bartik, Betrand, Lin, Rothstein and Unrath (2020), we deﬁne a business shutdown as businesses
that report zero hours worked during a week.
12In Tables A.2 and A.3, we employ cross-sectional speciﬁcations at the congressional district to further support
the idea that the PPP funding did not ﬂow to areas with largest pre-PPP declines in employment and ratios of
shutdown businesses. The tables report the results of ordinary least squares (OLS) regressions examining the
relation between the allocation of PPP funds and the share of businesses that shut down operations in the last
week of March, and the decline in hours worked between January and the last week of March.
17

mand, the evidence suggests the PPP functioned less as social insurance to support the hardest
hit areas and more as liquidity support for less affected ﬁrms. Nevertheless, our bank-level
results point to an important loan supply channel distorting the distribution of PPP loans.
The appendix presents suggestive evidence that, if anything, funds were disproportionately
allocated to geographic areas that were less hard hit by the virus. Figures A.2 and A.3 repeat
the analyses of Figures 8 and 9 at the state-level. Figures A.4 and A.5 show that there is a
slight negative correlation between loans and PPPE with COVID-19 conﬁrmed cases and deaths.
This fact is consistent with Figure A.6, which indicates that states with earlier shelter-in-place
orders—which were presumably harder hit by the epidemic—saw lower fund allocations. Fig-
ure A.7 shows that there is little correlation between the magnitude of social distancing at the
state level and PPP allocation and bank exposure. Finally, Figure A.8 conﬁrms our ﬁndings
using the Homebase data with another public data source—we ﬁnd no consistent relationship
between PPP allocation and bank exposure with state UI claims. The totality of the evidence
suggests that there was little targeting of funds to geographic areas that were harder bit by
the epidemic and, if anything, areas hit harder by the virus and subsequent economic impacts
received smaller portions of PPP funds.
7
The Early Effects of PPP on Employment and Local Eco-
nomic Activity
7.1
Research Design
Our results on PPP performance differences across banks motivate a research design for evalu-
ating the PPP. The basic idea is to use differences in local area PPP exposure (PPPE) to partition
geographies and compare the evolution of local outcomes for high versus low PPPE regions.
By exploiting differential exposure to banks that performed poorly in distributing PPP funding
during the ﬁrst round of the program, we can isolate the effect of the PPP from other differ-
ences across regions that may drive differences in PPP loan demand. As described above, we
map bank level aggregates for PPP lending from the SBA data onto local geographies using
measures of local bank branch presence. The research design is akin to a Bartik instrument
and therefore relies on the assumption that pre-policy bank deposit shares in particular regions
are not correlated with the various outcomes we study, conditional on observables.
18

We focus our analysis in the time period between the third week of January and the last
week of April to study the short-term effects of the PPP in the immediate aftermath of the
pandemic when the injection of liquidity was thought to matter the most. We choose this time
window for three reasons. First, starting our sample period in January allows to establish a
baseline period prior to the full economic impact of the virus hitting US ﬁrms. Second, the PPP
began disbursing funds on April 3 and all of the initial $349 billion was allocated by April 16.
During this period, banks played a key role in allocating limited funds, creating the variation
we exploit to identify the effects of the program. Third, beginning on April 27 a second round
of PPP funding began. Partly because PPP funding limits were no longer binding, the second
wave closed some of the gap between initially high and low PPPE exposure regions. Thus, were
we to use PPPE to study the effect of the program later in May and June, we would need to
interpret the research design as assigning some ﬁrms funding with a lag, instead of as assigning
some ﬁrms no funding at all.13
Given the rapid nature and size of the economic shock, we highlight an important consid-
eration when analyzing data from this time period. Our targeting analysis shows that regions
receiving more PPP funding were less hard hit by the initial shock, in part due to the banking
channel we emphasize. Thus, it is important to properly condition on this non-random assign-
ment of PPP funding. For example, if one does not break out the data ﬁnely enough or con-
dition properly for targeting differences—for example by treating all of March as a pre-period
benchmark or by looking at the correlation between PPP receipt and employment outcomes—
then one might detect a spurious effect of the program. We show this issue is very clear in
week-by-week outcomes around the policy window.
To account for these targeting differences, we estimate the effects of the program using
the period prior to large scale lockdowns as a base period and separately report “effects” for
the “Pre-Lockdown Covid” period, “Post-Lockdown Pre-PPP” period and “Post-PPP” period. We
then compare the post-PPP period effects to the ‘Post-Lockdown Pre-PPP” period to estimate
the causal effect of the program. We also include time-varying controls and state-by-time-
by-industry ﬁxed effects to estimate treatment effects under weaker versions of the Bartik as-
sumption. Once we adjust for targeting differences, including these more restrictive controls
has little effect on our estimates.
In our main analysis, we present reduced form regressions of local economic outcomes on
13We plan to extend our analysis to study the second round as new data become available.
19

PPPE. We choose this approach because we only have PPP funding for certain geographic levels,
namely, state and congressional district.14 In particular, we estimate variants of the following
speciﬁcation
Eisnt = αi + δsnt + β11[Pre-Lockdown] × PPPEc
+ β21[Post-Lockdown & Pre-PPP] × PPPEc
+ β31[Post-PPP] × PPPEc + γXc + ϵisnt,
where Eisnt is an outcome, business shutdowns or the decline in hours worked in our main
speciﬁcation for ﬁrm i operating in state s and industry n in week t. The term αi captures ﬁrm
ﬁxed effects, while δsnt are state-by-industry-by-week ﬁxed effects, PPPEc is PPP exposure at
the congressional district level measured using the number of ﬁrms exposed, and ϵisnt is an
error term. 1[Pre-Lockdown] is an indicator variable that takes the value of one between the
weeks of March 1st–March 7th and March 15th–March21st, 1[Post-Lockdown, Pre-PPP] is an
indicator variable that takes the value of one during the weeks of March 22nd–March 28th
and March 29th–April 4th, and 1[Post-PPP] is an indicator variable that takes the value of one
following the week of April 5th–April 12th.
The coefﬁcients β1, β2, and β3 capture the differential effect of PPP exposure on the out-
come of interest during each time period relative to the baseline period. The identifying as-
sumption is that the ﬁrms differentially exposed would have trended similarly in the absence
of the PPP, conditional on ﬁxed effects and controls. Due to the inclusion of state-by-industry-
by-week ﬁxed effects, we are comparing trajectories for ﬁrms within state-by-industry groups
and allowing general time trends within these groups. The coefﬁcient β1 captures differences
between ﬁrms after the initial spread of COVID-19 but prior to subsequent lockdowns. The co-
efﬁcient β2 captures targeting effects—differences in outcomes related to PPP exposure prior
to the disbursement of funds—while the coefﬁcient β3 captures the differential effect of PPP
exposure on the outcome of interest after the PPP launched. The difference between the coef-
ﬁcients β3 −β2 identiﬁes the effect of PPP exposure conditional on targeting.
14Appendix Table A.4 reports instrumental variable regressions to study Homebase outcomes at the congres-
sional district level, and we brieﬂy interpret these results below. We also use instrumental variable regressions to
study Census survey outcomes at the state level.
20

7.2
Small Business Employment
A signiﬁcant portion of the policy and media attention regarding the PPP focused on the pro-
gram’s potential employment effects. We examine business shutdowns (i.e., hours worked
reduced to zero during the entire week), declines in hours worked, and unemployment insur-
ance claims and ﬁnd no evidence of substantial effects of the PPP on these outcomes.
Recall that our targeting analysis revealed that if anything a larger fraction of businesses
located in areas that were less hard hit received PPP funding during the ﬁrst round. Specif-
ically, the plots of Figure 10 indicate that in congressional districts that saw least decline in
hours worked and the lowest rates of business shutdowns saw a greater fraction of their small
businesses receiving small business loans.
Figure 11 presents simple difference-in-difference graphs for each of our employment and
local activity outcomes. We divide all ﬁrms in the sample based on whether they are located
in congressional districts with above- or below-median high and low PPPEs. We use vertical
markers to demarcate the post-lockdown, pre-PPP period; the post-PPP launch; when the ﬁrst
round of PPP funds are exhausted; and when the second round of PPP funding begins. For both
measures of small business employment, we see a dramatic decline in employment starting in
the week prior to the lockdowns. Consistent with our targeting results, this decline is modestly
larger for regions with low PPPE. Importantly, during the ﬁrst round of PPP, the gap between
high and low PPPE areas does not widen further, indicating little incremental impact of PPP
during this time.
Figure 12 plots weekly coefﬁcients and standard error bands for regressions of each out-
come on congressional district PPPE. Speciﬁcally, we estimate regressions of the form
Eisnt = αi + δsnt + βtPPPEc × 1[Week=t] + ϵisnt
where Eit is business shutdowns or the decline in hours worked for ﬁrm i operating in state
s and industry n in week t, αi are ﬁrm ﬁxed effects, αsnt are state-by-industry-by-week ﬁxed
effects, PPPEc is PPP exposure at the congressional district level measured using the number of
ﬁrms exposed, 1[Week=t] is an indicator variable that takes the value of one during week t and
ϵisnt is an error term. The coefﬁcients βt capture the effect of PPP exposure on the outcome of
interest under the identifying assumption the ﬁrms differentially exposed would have trended
similarly in the absence of the PPP. The baseline week is the ﬁrst week of the sample period
21

(January 19th–January 25th). The state-by-industry-by-week ﬁxed effects imply that we are
comparing trajectories for ﬁrms within state-by-industry groups and allowing general time
trends within these groups.
The results align with the raw differences across high and low PPPE regions in Figure 11.
Common trends hold until the week of the declaration of the national emergency and lock-
downs take effect. At this point, higher PPPE is associated with better employment outcomes
in terms of fewer business shutdowns and smaller reductions in hours worked. When the PPP
is launched and funds start to reach ﬁrms, we see no incremental effects of the program on
these outcomes relative to pre-PPP levels.15
Table 3 presents our difference-in-difference estimates, in which we pool the weekly effects
into four time intervals reﬂecting the pre-Covid-19 crisis period; the pre-lockdown - post-Covid-
19 period; the post-lockdown, pre-PPP period; and the post-PPP period. We measure the effect
of the PPP by comparing the coefﬁcients for the latter two intervals. In the top panel, the
outcome of interest is business shutdowns, while in the bottom panel it is the decline in hours
worked. The ﬁrst two columns and second two columns exclude and include ﬁrm ﬁxed ef-
fects, respectively. The ﬁrst and third columns use the fraction of ﬁrms receiving PPP as the
program exposure variable. The second and fourth columns use congressional district PPPE as
the program exposure variable, and thus can be thought of as reduced form regressions.16
The table conﬁrms the ﬁnding of no statistically or economically signiﬁcant relationship
between PPP loans or PPP bank exposure and these employment outcomes. Moreover, our
least squares estimates are not simply statistically insigniﬁcant with large conﬁdence intervals;
rather, they are precise zeros. To address potential concerns about the strength of the research
design, Appendix Table A.4 instruments the fraction of ﬁrms receiving PPP loans with the PPPE
measure. Again, we ﬁnd nearly identical results. While one concern is that our ﬁrst-stage effect
is weak, we show that our F-statistic is above the rule-of-thumb and highly signiﬁcant: a 10
percentage point increase in PPPE is associated with a 3.4 to 3.8 percentage point increase in
the share of ﬁrms receiving PPP.
As another way of interpreting our magnitudes, consider the following comparison. The
15Appendix Figure A.9 shows that our instrument improves the reliability of inferences by mitigating pre-trends,
likely driven by differences in loan demand. However, the results when we use PPP funding instead of PPPE as
the right-hand-side variable are consistent with our preferred speciﬁcation: we see little evidence of substantial
effects of PPP on these outcomes relative to pre-PPP outcomes.
16Appendix Table A.4 estimates 2SLS regressions in ﬁrst differences comparing the post-PPP period to the post-
lockdown, pre-PPP period. The effects are qualitatively identical but with modestly larger standard errors.
22

difference between PPPE for top versus bottom quartile congressional districts is 0.42. This
difference implies an increase in the share of establishments receiving PPP funding of 14.3
percentage points.17 Using the reduced form estimates in Table 3, column (4), this change
in funding implies a change in the probability of ﬁrm shutdown of -0.08 percentage points
(= (−0.042 −0.040) × 0.42). The lower bound of the 95% conﬁdence interval is well below
a one percentage point effect. An analogous calculation for the decline in hours worked gives
similarly small effect sizes. Thus, relative to the aggregate patterns in Figure 11—a 40 per-
centage point increase in the probability of ﬁrm shutdown and 60 percentage point reduction
in the ratio of hours worked relative to January—we can reject modest effect sizes during this
time period.
Table 4 presents similar results using another separate measure of employment outcomes:
UI claims. The dependent variable is the ratio between the weekly initial unemployment in-
surance ﬁlings hand-collected from several state labor departments and county employment
measured using the Bureau of Labor and Statistics January County Employment data. The re-
sults paint a very similar picture to those in Table 3. Again, there is no statistically or econom-
ically signiﬁcant relationship between PPP loans or PPP bank exposure and UI claims. Again,
the ﬁrst and third columns use the fraction of ﬁrms receiving PPP as the program exposure
variable while the second and fourth columns use the respective county PPPE as the program
exposure variable. While the results are slightly less precise than those using outcomes from
the Homebase data, the coefﬁcient on the interaction post the rollout of the PPP program indi-
cates no discernable changes in the number of UI ﬁlings, relative to the post-lockdown period.
7.3
Local Economic Activity
While much of the focus of the PPP was in countering a surge in unemployment and busi-
ness closures, we now explore the potential effects of PPP on the evolution of small business
revenues as a proxy for local consumption activity. Table 5 presents analogous results. For ex-
ample, in columns (1) and (3), we ﬁnd no statistically signiﬁcant association between county
PPPE and consumption expenditures. Here, we control for state × industry × week ﬁxed ef-
fects, exploiting differences in the exposure of ﬁrms to PPP across counties within the same
state and industry. This rules out potential concerns about the passage of stay-at-home and
17This calculation comes from 0.42×0.34 from Appendix Table A.4, Panel A, establishment-level results, column
(1).
23

nonessential business closure laws at the state-level, which could otherwise confound our esti-
mates. Columns (2) and (4) subsequently add county × industry ﬁxed effects, thus absorbing
invariant characteristics of small businesses at the county×industry level. Again, we ﬁnd no ev-
idence of a positive association with consumption expenditures. Moreover, column (4) controls
for weekly infections and deaths per capita, as well as an index of social distancing (measured
by distance traveled among residents) to control for other time-varying county-speciﬁc shocks
present during these weeks.
8
Interpretation and Mechanisms
Additional funds from the PPP do not have signiﬁcant effects on employment or local economic
activity during the ﬁrst round of the program. If ﬁrms did not maintain or increase employment,
what did they do with these funds?
There are several non-mutually exclusive channels through which businesses may have ab-
sorbed the funds without immediate employment effects. First, program eligibility was deﬁned
broadly, so many less affected ﬁrms likely received funds and continued as they would have
in the absence of the funds. In these cases, the program’s beneﬁts accrue to the ﬁrm’s owners.
Related to this idea, it is possible that PPP funds crowded out private market borrowing, as
ﬁrms or banks may substitute publicly guaranteed PPP loans for other lending that would have
happened otherwise (Holmstrom and Tirole, 1998).
Second, despite restrictions on the share of funds ultimately forgiven that needed to be
spent on payroll, ﬁrms retained signiﬁcant ﬂexibility in how they could use the funds over
time. Firms could elect to use funds for required payments initially, to leave the funds in the
bank, or to use only the share that they are able to use when they can. For these ﬁrms, the
PPP may have strengthened balance sheets at a time when shelter-in-place orders prevented
workers from doing work and when unemployment insurance was more generous than wages
for a large share of workers. An implication of this channel is that, while employment effects
are zero in the short run, they may well be positive in the medium run because ﬁrms are less
likely to close permanently.
There was also likely considerable uncertainty about how to use the funds, either because
of evolving regulatory guidance or because small businesses struggled to understand the rules
in the ﬁrst place. Moreover, ﬁrms may respond to periods of high uncertainty, such as the
24

ongoing pandemic (Baker, Bloom, Davis and Terry, 2020), by holding cash as a precautionary
buffer.
Third, some ﬁrms may have increased employment or called back workers. Small effects
for these ﬁrms are indeed consistent with our null overall result. These effects may be par-
ticularly muted given historic expansion of UI beneﬁts, causing former employees to prefer
unemployment over a return to work (Chetty, 2008). Nevertheless, it is unlikely that the PPP
was responsible for large changes in employment at the end of April and into the ﬁrst week in
May. However, we caution that more data are likely needed to better support this claim.
Overall, we ﬁnd evidence that PPP funds during the time period we study are used as
liquidity support and to meet loan and other spending commitments.
8.1
Payments
While we do not ﬁnd short term effects on employment and ﬁrm revenue, the PPP may have
had signiﬁcant effects promoting ﬁnancial stability and the long-term recovery by extending
a lifeline to small businesses and preventing a rapid wave of small business bankruptcies and
liquidations. In Table 6 and Figure 13, we use information from the Census Small Business
Pulse Survey to relate the state PPPE with the percentage of businesses reporting having re-
ceived PPP in each state-industry group. The results suggest that between April 26th and June
6th, the period covered by the weekly surveys, state PPPE is strongly and positively related to
the percentage of businesses reporting receiving PPPE within a state and industry. Over the
entire sample period, a one-unit increase in state PPPE is associated with a 35% increase in the
share of businesses who report receiving PPP. Consistent with the idea that the second round of
funding lessened some of the geographic imbalance in the allocation of PPP disbursements, we
also see that the effect of state PPPE on the percentage of businesses reporting receiving PPP is
more pronounced in the early waves of the survey. Between April 26 to May 2, a comparable
unit increase in PPPE is associated with an 85% increase in PPP received. By May 31 to June
6, the coefﬁcient declines to 24%, although it remains statistically signiﬁcant.
We repeat this exercise using the percentage of businesses reporting requesting, but not
receiving PPP in each state and industry. As discussed in Section 5, this exercise also assuages
concerns that the relation between state PPPE and the percentage of businesses receiving PPP
could be driven by demand effects whereby ﬁrms and industries that did not request PPP funds
25

were disproportionately located in areas whose local banks have low PPPE. We report the
results of this analysis in Table A.5 and Appendix Figure A.10. The results indicate that states
with low PPPE had a greater proportion of businesses applying for but not receiving funds
under the program thus suggesting that the lower fraction of ﬁrms receiving PPP in those
areas is driven not by lower demand for funds from the program, but rather by their lower
access to banks that processed applications during the ﬁrst round.18
Having conﬁrmed that PPPE is signiﬁcantly related with access to PPP funds, we further
explore the information in the Small Business Pulse Survey to examine whether receipt of PPP
allowed ﬁrms to avoid becoming delinquent on scheduled payments (either loan or non-loan).
In Table 8, we show regressions examining the relationship between the allocation of PPP
funds and the percentage of ﬁrms reporting missing loan payments at the state-industry level.
In Table 7, we show similar regressions using the percentage of ﬁrms missing other scheduled
payments such as rent, utilities, supplier payments, and payroll.19 In light of our targeting
results, these regressions add controls for pre-PPP measures of crisis severity, including the
pre-PPP decline in hours worked from Homebase, the pre-PPP counts of COVID cases and
deaths per capita, and the pre-PPP social distancing index.
The results in Table 8, column (1), indicate that a percentage point increase in the share
of ﬁrms reporting receiving PPP is not signiﬁcantly associated with a decline in the percentage
of ﬁrms missing loan payments. This result, however, could indicate that areas and industries
with a lower percentage of businesses receiving PPP had a larger fraction of businesses that
did not apply because they were not eligible to apply or decided they did not need the funds.
To address this issue, we use state PPPE to capture geographic differences in access to the
supply of PPP funds resulting from differences across regions in their exposure to bank PPP
performance. These differences are plausibly unrelated to demand factors and therefore less
likely to be confounded by them. In column (2), we run a similar speciﬁcation, but instead
focus on the relation between percentage of ﬁrms reporting missing payments with state PPPE.
In this case, an increase in state PPPE is associated with a signiﬁcantly lower percentage ﬁrms
18In the Appendix, we further repeat these empirical exercises at the level of the Metropolitan Statistical Area
(MSA) rather than at the State-Industry level and we ﬁnd similar results in that MSAs with lower exposure to PPP
see a lower percentage of small businesses receiving PPP.
19Unfortunately, the Pulse survey does not separate this category of other payments into payroll versus non-
payroll components. However, it does focus on “required” payments, which ﬁrms may interpret as referring to
payments for past labor rather than discretionary payments based on retaining workers going forward. Results
for this measure should be interpreted with some uncertainty about respondents’ interpretation of the question.
26

reporting missing loans payments. In columns (3) and (4), we present results of an IV strategy
whereby we instrument for the percentage of ﬁrms receiving PPP using state PPPE. Using this
strategy, we ﬁnd that a ten percentage point increase in ﬁrms receiving PPP is associated with
1.7 percentage point decline in missing loan payments.
In Table 7, we ﬁnd that a ten percentage point increase in the share of ﬁrms receiving PPP
is associated with an even larger effect on missed non-loan payments. This result might not be
surprising when we consider that a large fraction of small businesses do not necessarily have
loans. We ﬁnd that, across all speciﬁcations, an increase in the percentage of ﬁrms receiving
PPP or in the access to PPP funds is associated with a lower percentage of ﬁrms reporting
missing these types of payments. Speciﬁcally, the results of the IV strategy in columns (3) and
(4) suggest that a ten percentage point increase in ﬁrms receiving PPP is associated with a
5.5 percentage point decline in the number of ﬁrms reporting missing any type of scheduled
payments. The relationship is highly statistically signiﬁcant at the 1 percent level.
Finally, we show in Appendix Table A.7 that the share of ﬁrms reporting missing payments
is positively related to the fraction of small businesses reporting having applied, but not having
received PPP funds. The percentage of ﬁrms that applied, but not did not receive, PPP funding is
plausibly less related to differences in demand for PPP funds across geographies and industries
and likely better captures timely access of small businesses to the supply of PPP funds. These
results further suggest that PPP might have been crucial in allowing small businesses to make
scheduled payments and survive the economic crisis without permanently closing.20
8.2
Liquidity Support and Precautionary Savings
We additionally ﬁnd that the PPP funds increased ﬁrms’ cash on hand. We use the Census Small
Business Pulse Survey to investigate whether access to PPP funds is associated with a greater
percentage of ﬁrms reporting having greater liquidity support. Table A.16 shows regressions
results examining the relation between access to PPP funds during the ﬁrst round and the
percentage of businesses reporting having at least two months of cash-on-hand. This exercise
also offers a useful sanity check of the informativeness of the survey data. As above, these
regressions include controls for pre-PPP measures of crisis severity.
Similar to results presented above, the coefﬁcients reported in column (1) of Table 9 do
20We also show in the appendix that these results are economically and statistically similar when we use MSA-
level observations.
27

not indicate an economically or statistically signiﬁcant relation between cash-on-hand and the
percentage of ﬁrms in that state and industry that reported receiving PPP. However, when
we turn our attention to the relation between state PPPE, which better isolates access to the
supply of PPP funds, access to PPP is economically and signiﬁcantly related to the share of
ﬁrms reporting signiﬁcant liquidity. In the reduced form speciﬁcation presented in column (2),
we ﬁnd that a unit increase in state PPPE is associated with a ten percentage point increase in
the share of ﬁrms reporting signiﬁcant liquidity. In the IV regression, which uses state PPPE
to instrument for the share of ﬁrms receiving PPP, a ten percentage point increase in the share
of ﬁrms receiving PPP is associated with a 2.9 percentage point increase in the share of ﬁrms
reporting at least two months of cash to cover business operations. Moreover, in the appendix
we report similar results when we use the share of small businesses that applied, but did not
receive, PPP as our main right-hand-side variable of interest.
Overall, these results are consistent with the idea that the PPP provided ﬁrms with an
important liquidity cushion that they used to navigate the turmoil of the initial months of
the pandemic. Another possible interpretation, given our evidence that PPP did not induce
increases in employment or declines in initial unemployment ﬁlings, is that these businesses
are maintaining the loaned PPP funds in bank accounts as precautionary savings until they are
ready to resume activities, perhaps as demand for their goods and services return to normal or
as relaxed shelter-in-place orders permit them to reopen for business. Generally, the results are
not consistent with the idea that the PPP served as a large-scale alternative to unemployment
insurance for delivering funds directly to affected workers.
9
Concluding Remarks and Next Steps
This paper takes an early look at a large and novel small business support program that was
part of the initial crisis response package, the Paycheck Protection Program (PPP). We consider
two dimensions of program targeting. First, did the funds ﬂow to where the economic shock
was greatest? Second, given the PPP used the banking system as a conduit to access ﬁrms,
we ask what role did the banks play in mediating policy targeting? We also explore effects on
employment and other ﬁrm outcomes.
We ﬁnd little evidence that funds were targeted towards geographic regions more severely
affected by the pandemic. If anything, preliminary evidence indicates that the opposite is
28

true and funds were targeted towards areas less severely affected by the virus. We do ﬁnd
that bank heterogeneity played an important role in mediating funds. We construct a new
measure of geographic exposure of regions to banks that over or underperformed in terms of
PPP allocation relative to their share of small business lending. States with higher exposure
to banks that performed well in terms of bank PPP exposure also saw higher levels of PPP
lending.21
Using a number of data sources, we do not ﬁnd evidence that the PPP had a substantial
effect on employment during the ﬁrst round of the program. Our estimates are precise enough
to rule out modest employment effects. While there appears to be little effect of the program on
employment, the program may have played an important role in promoting ﬁnancial stability.
Firms with greater exposure to the PPP hold more cash on hand, and are more likely to make
loan and other scheduled payments. Measuring these responses is critical for evaluating the
social insurance value of the PPP and similar policies and designing them effectively. As data
become available, we will continue to examine whether the program merely delayed ﬁrms’
inability to meet commitments or whether there are other medium-term and long-term effects.
21The analysis here focuses on ex ante targeting of the PPP, that is, the distribution of funding provided at
the start of the program. Ultimate targeting will depend on the extent of loan forgiveness and defaults, as well
as subsequent changes to the PPP, including conditions for recoupment based on ex post economic hardship
and changes to program eligibility criteria going forward. See Hanson, Stein, Sunderam and Zwick (2020) for a
discussion of these dynamic policy considerations in the design of business liquidity support during the pandemic.
29

References
Agarwal, Sumit, Gene Amromin, Itzhak Ben-David, Souphala Chomsisengphet, Tomasz
Piskorski, and Amit Seru. 2017. “Policy Intervention in Debt Renegotiation: Evidence from
the Home Affordable Modiﬁcation Program.” Journal of Political Economy, 125(3): 654–712.
Allcott, Hunt, Levi Boxell, Jacob Conway, Matthew Gentzkow, Michael Thaler, and David Y
Yang. 2020. “Polarization and Public Health: Partisan Differences in Social Distancing during
the Coronavirus Pandemic.” NBER Working Paper, , (w26946).
Atkeson, Andrew, Adrien d’Avernas, Andrea Eisfeldt, and Pierre-Olivier Weill. 2018. “Gov-
ernment Guarantees and the Valuation of American Banks.” NBER Macroeconomics Annual.
Bachas, Natalie, Olivia Kim, and Constantine Yannelis. 2020. “Loan Guarantees and Credit
Supply.” Journal of Financial Economics.
Baker, Scott R, Nicholas Bloom, Steven J Davis, and Stephen J Terry. 2020. “Covid-Induced
Economic Uncertainty.” National Bureau of Economic Research.
Baker, Scott R, Robert A Farrokhnia, Steffen Meyer, Michaela Pagel, and Constantine Yan-
nelis. 2020a. “How Does Household Spending Respond to an Epidemic? Consumption Dur-
ing the 2020 COVID-19 Pandemic.” National Bureau of Economic Research.
Baker, Scott R, Robert A Farrokhnia, Steffen Meyer, Michaela Pagel, and Constantine Yan-
nelis. 2020b. “Income, Liquidity, and the Consumption Response to the 2020 Economic Stim-
ulus Payments.” National Bureau of Economic Research.
Barrios, John, and Yael Hochberg. 2020. “Risk Perception Through the Lens Of Politics in the
Time of the COVID-19 Pandemic.” Working Paper.
Barrios, John M., Efraim Benmelech, Yael V. Hochberg, and Luigi Zingales. 2020. “Civic
Capital and Social Distancing During the COVID-19 Pandemic.” NBER working paper.
Barrios, John, Michael Minnis, William Minnis, and Joost Sijthoff. 2020. “Assessing the
Payroll Protection Program: A Framework and Preliminary Results.” Working paper.
Barro, Robert J, José F Ursua, and Joanna Weng. 2020. “The Coronavirus and the Great
Inﬂuenza Epidemic.”
30

Barrot, Jean-Noel, Thorsten Martin, Julien Sauvagnat, and Boris Vallee. 2019. “Employ-
ment Effects of Alleviating Financing Frictions: Worker-level Evidence from a Loan Guaran-
tee Program.” Working Paper.
Bartik, Alexander W., Marianne Bertrand, Zoe B. Cullen, Edward L. Glaeser, Michael Luca,
and Christopher T. Stanton. 2020. “How Are Small Businesses Adjusting to COVID-19?
Early Evidence from a Survey.” Proceedings of the National Academy of Sciences.
Bartik, Alexander W., Marianne Betrand, Feng Lin, Jesse Rothstein, and Matt Unrath.
2020. “Measuring the labor market at the onset of the COVID-19 crisis.” Brookings Papers on
Economic Activity.
Berger, David W, Kyle F Herkenhoff, and Simon Mongey. 2020. “An SEIR Infectious Disease
Model with Testing and Conditional Quarantine.” National Bureau of Economic Research
Working Paper 26901.
Brevoort, Kenneth P., John A. Holmes, and John D. Wolken. 2010. “Distance Still Matters:
The Information Revolution in Small Business Lending and the Persistent Role of Location,
1993-2003.” Board of Governors of the Federal Reserve System.
Bufﬁngton, Catherine, Carrie Dennis, Emin Dinlersoz, Lucia Foster, and Shawn Klimek.
2020. “Measuring the Effect of COVID-19 on U.S. Small Businesses: The Small Business
Pulse Survey.” Census for Economic Studies, working paper.
Cajner, Tomaz, Leland Crane, Ryan A. Decker, John Grigsby, Adrian Hamins-Puertolas,
Erik Hurst, Christopher Kurz, and Ahu Yildirmaz. 2020. “The U.S. Labor Market During
the Beginning of the Pandemic Recession.” NBER working paper.
Chetty, Raj. 2008. “Moral Hazard Versus Liquidity and Optimal Unemployment Insurance.”
Journal of Political Economy, 116(2): 173–234.
Chetty, Raj, John N. Friedman, Nathaniel Hendren, and Michael Stepner. 2020. “How did
COVID-19 and Stabilization Policies Affect Spending and Employment? A New Real-time
Economic Tracker Based on Private Sector Data.” NBER working paper.
Coibion, Olivier, Yuriy Gorodnichenko, and Michael Weber. 2020. “Labor Markets During
the COVID-19 Crisis: A Preliminary View.” Fama-Miller Working Paper.
Cororaton, Anna, and Samuel Rosen. 2020. “Public Firm Borrowers of the US Paycheck Pro-
tection Program.” SSRN working paper.
31

Cox, Natalie, Peter Ganong, Pascal Noel, Joseph Vavra, Arlene Wong, Diana Farrell, and
Fiona Greig. 2020. “Initial Impacts of the Pandemic on Consumer Behavior: Evidence from
Linked Income, Spending, and Savings Data.” Brookings Papers on Economic Activity.
De Marco, Filippo. 2020. “Public Guarantees for Small Businesses in Italy during COVID-19.”
SSRN working paper.
Dingel, Jonathan I., and Brent Neiman. 2020. “How Many Jobs Can Be Done at Home.”
Journal of Public Economics.
Ding, Wenzhi, Ross Levine, Chen Lin, and Wensi Xie. 2020. “Social Distancing and Social
Capital: Why U.S. Counties Respond Differently to COVID-19.” NBER working paper.
Eichenbaum, Martin S, Sergio Rebelo, and Mathias Trabandt. 2020. “The Macroeconomics
of Epidemics.” National Bureau of Economic Research.
Elenev, Vadim, Tim Landvoigt, and Stijn Van Nieuwerburgh. 2020. “Can the Covid Bailouts
Save the Economy?” National Bureau of Economic Research.
Gale, William. 1990. “Federal Lending and the Market for Credit.” Journal of Public Economics,
, (42): 177–93.
Gale, William. 1991. “Economic Effects of Federal Credit Programs.” American Economic Re-
view, 81(1): 133–52.
Gallipoli, Giovanni, and Christos Makridis. 2018. “Structural Transformation and the Rise
of Information Technology.” Journal of Monetary Economics, 97: 91–110.
Ganong, Peter, and Pascal Noel. 2018. “Liquidity vs. Wealth in Household Debt Obligations:
Evidence from Housing Policy in the Great Recession.” National Bureau of Economic Re-
search.
Gonzalez-Uribe, Juanita, and Su Wang. 2019. “Dissecting the Effect of Financial Constraints
on Small Firms.” Working Paper.
Gormsen, Niels Joachim, and Ralph SJ Koijen. 2020. “Coronavirus: Impact on Stock Prices
and Growth Expectations.” University of Chicago, Becker Friedman Institute for Economics
Working Paper, , (2020-22).
Granja, João, Christian Leuz, and Raghuram Rajan. 2018. “Going the Extra Mile: Distant
Lending and Credit Cycles.” National Bureau of Economic Research.
32

Guerrieri, Veronica, Guido Lorenzoni, Ludwig Straub, and Ivan Werning. 2020. “Macroe-
conomic Implications of COVID-19: Can Negative Supply Shocks Cause Demand Shortages.”
Working paper.
Hanson, Samuel, Jeremy Stein, Adi Sunderam, and Eric Zwick. 2020. “Business Conti-
nuity Insurance and Business Continuity Loans: Keeping America’s Lights on During the
Pandemic.” Policy Brief.
Holmstrom, Bengt, and Jean Tirole. 1998. “Private and Public Supply of Liquidity.” Journal
of Political Economy, 106(1): 1–40.
House, Christopher, and Matthew Shapiro. 2008. “Temporary Investment Tax Incentives:
Theory with Evidence from Bonus Depreciation.” American Economic Review, 98(3): 737–
768.
Humphries, John Eric, Christopher Neilson, and Gabriel Ulyssea. 2020. “Information Fric-
tions and Access to the Paycheck Protection Program.”
Jones, Callum, Thomas Philippon, and Venky Venkateswaran. 2020. “Optimal Mitigation
Policies in a Pandemic.” Working Paper.
Kaplan, Greg, and Gianluca Violante. 2014. “A Model of the Consumption Response to Fiscal
Stimulus Payments.” Econometrica, 82(4): 1199–1239.
Kelly, Bryan, Hanno Lustig, and Stijn Van Nieuwerburgh. 2016. “Too-Systemic-to-Fail: What
Option Markets Imply about Sector-Wide Government Guarantees.” American Economic Re-
view, 106(6): 1278–1319.
Lelarge, Clair, David Sraer, and David Thesmar. 2010. “Entrepreneurship and Credit Con-
straints: Evidence From a French Loan Guarantee Program.” International Differences in En-
trepreneurship, 243–273.
Lucas, Deborah. 2016. “Credit Policy as Fiscal Policy.” Brookings Papers on Economic Activity,
, (Spring): 1–57.
Makridis, Christos A., and Cary Wu. 2020. “Ties that Bind (and Social Distance): How Social
Capital Helps Communities Weather the COVID-19 Pandemic.” SSRN working paper.
Makridis, Christos A., and Jonathan Hartley. 2020. “The Cost of COVID-19: A Rough Esti-
mate of the 2020 GDP Impact.” Mercatus Center, Policy Brief Special Edition.
33

Makridis, Christos A., and Jonathan Rothwell. 2020. “The Real Cost of Political Polarization:
Evidence from the COVID-19 Pandemic.” SSRN working paper.
Mian, Atif, and Amir Suﬁ. 2012. “The Effects of Fiscal Stimulus: Evidence from the 2009 Cash
for Clunkers Program.” Quarterly Journal of Economics, 127: 1107–1142.
Mulligan, Casey B. 2020. “Economic Activity and the Value of Medical Innovation During a
Pandemic.” NBER Working paper.
Mullins, William, and Patricio Toro. 2017. “Credit Guarantees and New Bank Relationships.”
Policy Research Working Paper No, 8241.
Papanikolaou, Dimitris, and Lawrence D. W. Schmidt. 2020. “Working Remotely and the
Supply-side Impact of COVID-19.” Working paper.
Parker, Jonathan A, Nicholas S Souleles, David S Johnson, and Robert McClelland. 2013.
“Consumer Spending and the Economic Stimulus Payments of 2008.” American Economic
Review, 103(6): 2530–53.
Smith, Bruce. 1983. “Limited Information, Credit Rationing, and Optimal Government Lend-
ing Policy.” American Economic Review, 73(3): 305–18.
Zwick, Eric. Forthcoming. “The Costs of Corporate Tax Complexity.” American Economic Jour-
nal: Economic Policy.
Zwick, Eric, and James Mahon. 2017. “Tax Policy and Heterogeneous Investment Behavior.”
American Economic Review, 107(1): 217–248.
34

Figure 1: Histogram of Bank Paycheck Protection Program Exposure (PPPE)
Panel A of Figure 1 plots the distribution of the measure of relative bank performance in the PPP based on the volume of PPP loans granted and
stock of small business loans at the bank as of fourth quarter of 2019. We compute this measure as: PPPEb =
ShareVol.PPP−ShareSBLMarket
(ShareVol.PPP+ShareSBLMarket)×0.5 .
Panel B of Figure 1 plots the distribution of the measure of relative bank performance in the PPP based on the number of PPP loans
granted and number of small business loans held by the bank as of fourth quarter of 2019.
We compute this measure as: PPPEb =
ShareN br.PPP−ShareN br.SBLMarket
(ShareN br.PPP+ShareN br.SBLMarket)×0.5 . Data is obtained from the SBA and call reports.
Panel A: Histogram of Volume-based PPPE
0
5
10
15
20
25
30
35
40
45
50
Percent
-.5
-.4
-.3
-.2
-.1
0
.1
.2
.3
.4
.5
PPPE (Volume of Loans)
Panel B: Histogram of Number Loans-based PPPE
0
5
10
15
20
25
30
35
40
45
50
Percent
-.5
-.4
-.3
-.2
-.1
0
.1
.2
.3
.4
.5
PPPE (Number of Loans)
35

Figure 2: PPPE and Commercial Bank Size
Figure 2 plots average PPPE based on volume of PPP loans (Panel A) number of PPP loans (Panel B) and average amount of PPP loan (Panel
C) in each percentile size bin. The size bins stratify all commercial banks operating as of the fourth quarter of 2019 based on their total assets.
Data is obtained from the SBA and call reports.
Panel A: Volume-based PPPE and Size
-.2 -.175-.15-.125 -.1 -.075-.05-.025 0 .025 .05 .075 .1
Bank PPPE (Volume of Lending)
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95 100
Size Bin (Percentile)
Panel B: Number of Loans-based PPPE and Size
-.1
-.05
0
.05
.1
.15
.2
.25
.3
.35
Bank PPPE (Number of Loans)
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95 100
Size Bin (Percentile)
Panel C: Average Amount of Loan and Size
50
100
150
200
250
300
350
400
450
500
Av. Loan Amount (in $000)
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95 100
Size Bin (Percentile)
36

Figure 3: PPPE and PPP Allocation
Figure 3 plots the cumulative share of PPP and SBL lending by all banks whose PPPE is below x, where x ∈(−0.5,0.5). Data is obtained from the SBA and call reports.
0
.2
.4
.6
.8
1
Cumulative Shares of PPP Lending and SBL Lending
-.5
0
.5
Bank PPPE
Cumulative Share of PPP Lending
Cumulative Share of SBL Lending
37

Figure 4: Map of County Exposure to PPPE
Figure 4 plots the average exposure of each county to the volume-based PPPE. County exposure to PPPE is computed as the average of the PPPE of each bank with a branch presence in the county.
The PPPE of each bank is weighted by the share of deposits of the bank in the county as of June 30th, 2019. Data is from the SBA, Call Reports, and FDIC’s Summary of Deposits.
38

Figure 5: State Exposure to PPPE and PPP per Establishment
Figure 5 are scatterplots of the total PPP allocation per small business establishment in the state and the state exposure to the volume-based
PPPE (Panel A) and fraction of small business establishments receiving a PPP loan and the state exposure to the PPPE based on the number
of loans (Panel B). Data comes from SBA, Call Reports, Summary of Deposits, and County Business Patterns.
Panel A: State Exposure to Volume-Based PPPE and PPP Allocation per Small Business
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
30000
40000
50000
60000
70000
PPP allocation per Small Business Establishment (State)
-.3
-.2
-.1
0
.1
.2
State PPPE
Panel B: State Exposure to Number-Based PPPE and Fraction of Small Businesses receiving PPP
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MNMO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
.1
.2
.3
.4
.5
Fraction of Small Businesses receiving PPP (State)
-.2
0
.2
.4
State PPPE
39

Figure 6: State Exposure to PPPE and Unmet PPP Demand
Figure 5 are scatterplots of the state exposure to the volume-based PPPE and the difference between percentage of businesses reporting having
applied to PPP and percentage of business that received PPP in each state (Panel A) and the state exposure to the PPPE based on the number
of loans and the difference between percentage of businesses reporting having applied to PPP and percentage of business that received PPP
in each state (Panel B). Data comes from the Census Bureau, SBA, Call Reports, Summary of Deposits, and County Business Patterns.
Panel A: State Exposure to Volume-Based PPPE and Difference between % Businesses Applying and Receiving PPP
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
10
20
30
40
50
Diff. between % of businesses applying and receiving PPP
-.2
-.1
0
.1
.2
State PPPE (Volume)
Panel B: State Exposure to Numer-Based PPPE and Difference between % Businesses Applying and Receiving PPP
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
10
20
30
40
50
Diff. between % of businesses applying and receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
40

Figure 7: Congressional District Exposure to PPPE and PPP per Establishment
Figure 7 are scatterplots of the total PPP allocation per establishment in the congressional district and the congressional district exposure to
the volume-based PPPE (Panel A) and fraction of establishments receiving a PPP loan and the congressional district exposure to the PPPE
based on the number of loans (Panel B). Data comes from the SBA, Call Reports, Summary of Deposits, and County Business Patterns.
Panel A: Congressional District Exposure to Volume-Based PPPE and PPP Allocation per Small Business
0
50000
100000
150000
PPP allocation per Establishment (Congressional District)
-.4
-.2
0
.2
.4
Congressional District PPPE
Panel B: Congressional District Exposure to Number-Based PPPE and Fraction of Small Businesses receiving PPP
0
.1
.2
.3
.4
.5
Fraction of Establishments receiving PPP (Congressional District)
-.4
-.2
0
.2
.4
Congressional District PPPE
41

Figure 8: Business Shutdowns and PPP Allocation by Congressional District
Figure 8 presents four scatterplots of the share of businesses in each state that shutdown that in the week of March 29th–April 4th and four alternative measures of allocation of PPP funds across
states. The ﬁgure on the top left plots the amount of PPP loans received by small businesses in each state divided by the total number of small businesses in the state. The ﬁgure on the top right
corner plots the fraction of small businesses in each state that received a PPP loan. The ﬁgure on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan
and the state exposure to the PPPE measured in terms of the total volume of loans. The ﬁgure on the bottom right corner plots the fraction of small businesses in each state that received a PPP
loan and the state exposure to the PPPE measured in terms of the total number of loans.
0
.2
.4
.6
.8
Share of Business Shutdowns (Cong. Dist.)
0
50000
100000
150000
PPP amount per establishment
(Date: 31/03/2020)
0
.2
.4
.6
.8
Share of Business Shutdowns (Cong. Dist.)
0
.1
.2
.3
.4
.5
Fraction of Businesses receiving PPP
(Date: 31/03/2020)
0
.2
.4
.6
.8
Share of Business Shutdowns (Cong. Dist.)
-.4
-.2
0
.2
.4
Congressional District PPPE (Volume)
(Date: 31/03/2020)
0
.2
.4
.6
.8
Share of Business Shutdowns (Cong. Dist.)
-.4
-.2
0
.2
.4
Congressional District PPPE (Nbr. Loans)
(Date: 31/03/2020)
42

Figure 9: Decline in Hours Worked and PPP Allocation by Congressional District
Figure 9 presents four scatterplots of the decline in hours worked in each congressional district relative to a January baseline and four alternative measures of allocation of PPP funds across states.
The ﬁgure on the top left plots the amount of PPP loans received by small businesses in each state divided by the total number of small businesses in the state. The ﬁgure on the top right corner
plots the fraction of small businesses in each state that received a PPP loan. The ﬁgure on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan and the
state exposure to the PPPE measured in terms of the total volume of loans. The ﬁgure on the bottom right corner plots the fraction of small businesses in each state that received a PPP loan and
the state exposure to the PPPE measured in terms of the total number of loans.
0
.25
.5
.75
1
Ratio of Hours Worked relative to January Baseline
0
50000
100000
150000
PPP amount per establishment
(Date: 31/03/2020)
0
.25
.5
.75
1
Ratio of Hours Worked relative to January Baseline
0
.1
.2
.3
.4
.5
Fraction of Businesses receiving PPP
(Date: 31/03/2020)
0
.25
.5
.75
1
Ratio of Hours Worked relative to January Baseline
-.4
-.2
0
.2
.4
Congressional District PPPE (Volume)
(Date: 31/03/2020)
0
.25
.5
.75
1
Ratio of Hours Worked relative to January Baseline
-.4
-.2
0
.2
.4
Congressional District PPPE (Nbr. Loans)
(Date: 31/03/2020)
43

Figure 10: PPP Allocation by Employment Shock Bin
Figure 10 stratiﬁes congressional districts on 20 bins based on the share of Homebase businesses that shutdown in the week of March29th–
April 4th (Panel A) and on their decline in hours worked relative to a January baseline. The y-axis represents the fraction of businesses
receiving PPP funds in each bin computed as total number of PPP loans in that bin divided by total number of establishments of congressional
districts in that bin. Data is from SBA, Homebase, and County Business Patterns.
.15
.18
.21
.24
.27
.3
Fraction of Businesses receiving PPP
Most Affected Districts
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Least Affected Districts
Business Shutdown Shock
.15
.18
.21
.24
.27
.3
Fraction of Businesses receiving PPP
Most Affected Districts
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Least Affected Districts
Decline in Hours Worked Shock
44

Figure 11: PPPE and Post-PPP Outcomes
Figure 11 shows the ratio of hours worked over time, Data is from SBA, the percent of businesses shut down, initial unemployment ﬁlings and small business revenue by above and below PPPE
exposure. Homebase, and County Business Patterns.
Initial Lockdown Orders
PPP Launch
PPP Runs out of Funds
PPP 2nd Round begins
0
.2
.4
.6
.8
1
% Decline in Hours Worked relative to January
Week 1: Jan19 - Jan25
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
Low PPPE Cong. Dist.
High PPPE Cong. Dist.
Ratio Hours Worked over Time and Congressional District PPPE (Nbr-Based)
Initial Lockdown Orders
PPP Launch
PPP Runs out of Funds
PPP 2nd Round begins
0
.1
.2
.3
.4
% Businesses Shutdown
Week 1: Jan19 - Jan25
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
Low PPPE Cong. Dist.
High PPPE Cong. Dist.
% Businesses Shutdown and Congressional District PPPE (Nbr-Based)
Initial Lockdown Orders
PPP Launch
PPP Runs out of Funds
PPP 2nd Round begins
0
1%
2%
3%
4%
Initial Unemployment Insurance Claims (as % of Jan 2020 BLS County Emp.)
Week 1: Jan19-Jan25
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
Low PPPE County
High PPPE County
Initial Unemployment Filings and County PPPE (Nbr.-Based)
Initial Lockdown Orders
PPP Launch
PPP Runs out of Funds
PPP 2nd Round begins
-.6
-.4
-.2
0
.2
Decline in Y/Y Change in Total Consumer Spending relative to January
Week 1: Jan19-Jan25
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
Week 16: May3-May10
Low PPPE County
High PPPE County
Small Business Revenue and County PPPE (Nbr.-Based)
45

Figure 12: Weekly Elasticities of Employment and Firm Outcomes to PPPE
Figure 12 plots coefﬁcients and respective standard errors of regression analyses investigating the week-by-week elasticities between employment and ﬁrm outcomes and exposure to PPPE. The
top-left ﬁgure plots the coefﬁcients and respective standard errors of an empirical speciﬁcation that augments the speciﬁcation of Panel A of Table 5 column (4) to include a full set of interactions
between the Congressional District PPPE and week dummy variables. The top-left ﬁgure plots the coefﬁcients and respective standard errors of an empirical speciﬁcation that augments the
speciﬁcation of Panel B of Table 5 column (4) to include a full set of interactions between the Congressional District PPPE and week dummy variables. The bottom-left ﬁgure plots the coefﬁcients
and respective standard errors of an empirical speciﬁcation that augments the speciﬁcation of Table 6 column (4) to include a full set of interactions between the County PPPE and week dummy
variables. The bottom-right ﬁgure plots the coefﬁcients and respective standard errors of an empirical speciﬁcation that augments the speciﬁcation of Table 7 column (4) to include a full set of
interactions between the County PPPE and week dummy variables.
PPP Launch
Vertical bands represent +/- 1.96 * St. Error of each point estimate
Initial Lockdown Orders
PPP Runs out of Funds
PPP 2nd Round begins
-.06
-.04
-.02
0
Beta
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
State-Industry-Week and Firm Fixed Effects
Weekly Elasticities of Shutdown and Congressional District PPPE
PPP Launch
Vertical bands represent +/- 1.96 * St. Error of each point estimate
Initial Lockdown Orders
PPP Runs out of Funds
PPP 2nd Round begins
-.02
0
.02
.04
.06
Beta
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
State-Industry-Week and Firm Fixed Effects
Weekly Elasticities of Weekly Hours Worked and Congressional District PPPE
PPP Launch
Vertical bands represent +/- 1.96 * St. Error of each point estimate
Initial Lockdown Orders
PPP Runs out of Funds
PPP 2nd Round begins
-.4
-.2
0
.2
Beta
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
Year
State-by-Week and County Fixed Effects
Weekly Elasticities of Initial County UI Claims and County PPPE
PPP Launch
Vertical bands represent +/- 1.96 * St. Error of each point estimate
Initial Lockdown Orders
PPP Runs out of Funds
PPP 2nd Round begins
-.02
0
.02
.04
.06
.08
Beta
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
Year
State-Industry-Week and County-Industry Fixed Effects
Weekly Elasticities of Small Business Revenue and County PPPE
46

Figure 13: % Receiving PPP and State Exposure to PPPE (Industry×State Level)
Figure 13 are scatterplots of the state exposure to the number-based PPPE and the percentage of businesses reporting receiving PPP at the
state-by-industry level. The plots represent the evolution of the two variables for each survey week. Data comes from the Census Bureau,
SBA, Call Reports, Summary of Deposits, and County Business Patterns.
AR
CA
CO
FL
IA
IL
IN
KY
MA
MD
MI
MN
MO
NC
NH
NJ
NY
OH
OK
OR
PA
TX
VA
WA
WI
AL
AL
AL
AL
AL
AR
AR
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
IA
IA
IA
IA
IA
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KY
KY
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
ME
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MNMO
MO
MO
MO
MO
MO
MO
MS
MS
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NH
NJ
NJ
NJ
NJ
NJ
NJ
NM
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
70
80
90
100
% Businesses reporting receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 15: April26 - May2nd
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
70
80
90
100
% Businesses reporting receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 16: May3rd - May9
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NE
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NV
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PR
PR
PR
SC
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
70
80
90
100
% Businesses reporting receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 17: May10 - May16
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NE
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PR
PR
SC
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
70
80
90
100
% Businesses reporting receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 18: May17 - May23
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MNMO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NE
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PR
PR
PR
SC
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
70
80
90
100
% Businesses reporting receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 19: May24 - May30
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NE
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PR
PR
PR
SC
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
70
80
90
100
% Businesses reporting receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 20: May31 - Jun6
47

Table 1: Top 20 Banks by Total Assets and PPPE
Table 1 reports individual bank statistics and the PPPE index for the 20 largest ﬁnancial institutions in the United States. Total Assets is computed using information from fourth quarter 2019 call
reports. Share of total vol. PPP program is the total amount disbursed by each ﬁnancial institution relative to the total amount disbursed under the ﬁrst wave of the PPP. Share of SBL market is the
share of the total outstanding amount of small business loans held by each ﬁnancial institution relative to the total outstanding amount of small business loans as of 2019:Q4. PPPE (volume) is
the volume-based bank PPP index. Total assets are in $ millions. Share of loans in PPP program is the total number of loans processed by each ﬁnancial institution relative to the total number of
loans processed in the ﬁrst wave of the PPP. Share of loans in SBL market is the share of the total number of outstanding small business loans held by each ﬁnancial institution relative to the total
outstanding number of small business loans as of 2019:Q4. PPPE (Nbr. Loans) is the number-based bank PPP index.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
Financial Institution Name
Total
Assets
Share
of total
vol.
PPP
Share
of SBL
Market
PPPE
(vol.)
Share
of loans
in PPP
Share
of loans
in SBL
Market
PPPE
(Nbr.
Loans)
JPMORGAN CHASE BANK, NATIONAL ASSOCIATION
2,337,707
3.892%
6.547%
-0.127
1.432%
10.47%
-0.380
BANK OF AMERICA, NATIONAL ASSOCIATION
1,866,841
1.199%
9.510%
-0.388
.5565%
11.86%
-0.455
WELLS FARGO BANK, NATIONAL ASSOCIATION
1,736,928
.0362%
6.502%
-0.494
.0664%
4.308%
-0.485
CITIBANK, N.A.
1,453,998
.3348%
2.121%
-0.364
.4431%
9.729%
-0.456
U.S. BANK NATIONAL ASSOCIATION
486,004
.6953%
3.327%
-0.327
1.120%
5.643%
-0.334
TRUIST BANK
461,256
3.160%
2.011%
0.111
2.078%
1.732%
0.045
CAPITAL ONE, NATIONAL ASSOCIATION
453,626
.0212%
2.822%
-0.493
.0134%
10.38%
-0.499
PNC BANK, NATIONAL ASSOCIATION
397,703
2.765%
1.124%
0.211
1.235%
1.373%
-0.027
BANK OF NEW YORK MELLON, THE
342,225
0%
.0024%
-0.500
0%
.0001%
-0.500
TD BANK, N.A.
338,272
1.837%
.6876%
0.228
1.698%
.5698%
0.249
STATE STREET BANK AND TRUST COMPANY
242,148
0%
0%
0.000
0%
4.493%
-0.500
CHARLES SCHWAB BANK
236,995
0%
.0745%
-0.500
0%
.0039%
-0.500
MORGAN STANLEY BANK, N.A.
229,681
0%
.1445%
-0.500
0%
.0089%
-0.500
GOLDMAN SACHS BANK USA
228,836
0%
.0032%
-0.500
0%
.0001%
-0.500
HSBC BANK USA, NATIONAL ASSOCIATION
172,888
.1411%
.0845%
0.125
.0697%
.0140%
0.332
FIFTH THIRD BANK, NATIONAL ASSOCIATION
167,845
.9991%
.4589%
0.185
.5948%
.1920%
0.256
ALLY BANK
167,492
.2639%
2.118%
-0.389
.0631%
1.382%
-0.456
CITIZENS BANK, NATIONAL ASSOCIATION
165,742
1.072%
.8077%
0.070
1.148%
.5274%
0.185
KEYBANK NATIONAL ASSOCIATION
143,390
2.370%
.7295%
0.265
2.236%
.2743%
0.391
BMO HARRIS BANK NATIONAL ASSOCIATION
137,588
1.385%
1.959%
-0.086
.7133%
.5413%
0.069
ALL OTHER BANKS
6,889,908
79.82%
58.96%
-0.048
86.52%
40.97%
0.218
48

Table 2: PPPE and PPP Allocation
Table 2 reports the results of ordinary least squares (OLS) regressions examining the impact of the congressional district exposure to PPPE
on the cross-sectional allocation of PPP funds to congressional districts. The dependent variable of the speciﬁcations, Ln(Total PPP Allocation
per establishment), is the natural logarithm total amount of PPP funds disbursed to small businesses in each congressional district divided by
the number of establishments in the congressional district. The dependent variable of the speciﬁcations in Panel B, Fraction of Establishments
receiving PPP is the total number of PPP loans made to small businesses in each congressional district divided by the number of establishments
in the congressional district. Cong. Dist. Exposure to PPPE (Vol) is the congressional district average of the PPPE based on total amounts of
outstanding PPP and small business loans, weighted by the share of deposits of each bank in each congressional district. Cong. Dist. Exposure
to PPPE (Nbr.) is the congressional district average of the PPPE based on the number of outstanding loans, weighted by the share of deposits
of each bank in each congressional district. Ln(Payroll) is the natural logarithm of the sum of payroll of all establishments in the congressional
district. Ln(Employment) is the natural logarithm of total employment in the congressional district. Industry Shares are additional controls
for the share of establishments in each two-digit NAICS code industry. The speciﬁcation of column (4) includes state ﬁxed effects. Standard
errors are presented in parentheses, and are clustered at the level of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and
10% levels, respectively.
Panel A: Total PPP Allocation per Establishment at the Congressional District
(1)
(2)
(3)
(4)
Ln(Total PPP Allocation per Establishment)
Cong. Dist. PPPE (Vol)
0.160∗∗∗
0.174∗∗∗
0.126∗∗∗
0.148∗∗
(0.023)
(0.023)
(0.038)
(0.067)
Ln(Total Payroll)
0.290
0.279
0.191
(0.223)
(0.233)
(0.242)
Ln(Employment)
-0.089
-0.113
0.112
(0.397)
(0.351)
(0.376)
Observations
436
436
436
436
Adjusted R2
0.071
0.109
0.135
0.102
Industry Shares
No
No
Yes
Yes
State Fixed Effects
No
No
No
Yes
Panel B: Fraction of Establishment receiving PPP at the Congressional District
(1)
(2)
(3)
(4)
Fraction of Establishments receiving PPP
Cong. Dist. PPPE (Nbr.)
0.068∗∗∗
0.067∗∗∗
0.056∗∗∗
0.047∗∗∗
(0.005)
(0.005)
(0.007)
(0.009)
Ln(Total Payroll)
-0.017
-0.045
-0.029
(0.038)
(0.036)
(0.037)
Ln(Employment)
0.028
0.051
0.050
(0.062)
(0.059)
(0.066)
Observations
436
436
436
436
Adjusted R2
0.414
0.412
0.449
0.482
Industry Shares
No
No
Yes
Yes
State Fixed Effects
No
No
No
Yes
49

Table 3: Evolution of Homebase Employment Outcomes and Exposure to PPP
Table 3 reports the results of ordinary least squares (OLS) regressions examining the relation between the geographic allocation of PPP funds
during the ﬁrst round and weekly ﬁrm employment outcomes measured using the Homebase data set between the second week of March until
the last week of April. The dependent variable in Panel A, Bus. Shutdowns, is an indicator variable that takes the value of one if the business
reported zero hours worked over the course of the week. The dependent variable in Panel B, Decline Hours Worked, is the decline in hours
worked in each establishment relative to the average hours worked in the same establishment during the last two weeks of January. Pre-Covid
is an indicator variable that takes the value of prior to the ﬁrst week of March. This period represents the base period in the speciﬁcations
that include ﬁrm ﬁxed effects and as a result this variable is omitted in such speciﬁcations. Pre-Lockdown is an indicator variable that takes
the value of one prior to the week of March 22nd – March 28th (exclusive) Post-Lockdown & Pre-PPP is an indicator variable that takes the
value of one during the weeks of March 22nd – March 28th and March 29th – April 4th. Post-PPP is an indicator variable that takes the value
of one following the week of April 5th – April 12th (inclusive), Fraction receiving PPP is the percentage of businesses that received PPP loans
during the ﬁrst round in the congressional district where the ﬁrm is located. Congressional District PPPE (Nbr.) is the congressional district
average of the PPPE based on the number of outstanding loans, weighted by the share of branches of each bank in each congressional district.
Standard errors are presented in parentheses, and are double-clustered at the level of the state and week. ***, **, and *, represent statistical
signiﬁcance at 1%, 5%, and 10% levels, respectively.
Panel A: Business Shutdowns
(1)
(2)
(3)
(4)
Bus. Shutdown
Pre-Covid × Fraction receiving PPP
-0.005
(0.004)
Pre-Lockdown × Fraction receiving PPP
-0.032∗∗∗
-0.027∗∗∗
(0.010)
(0.002)
Post-Lockdown & Pre-PPP × Fraction receiving PPP
-0.104∗∗
-0.098∗∗
(0.042)
(0.043)
Post-PPP × Fraction receiving PPP
-0.103∗
-0.097∗
(0.049)
(0.052)
Pre-Lockdown × Congressional District PPPE (Nbr.)
-0.008∗∗
-0.007∗∗∗
(0.003)
(0.002)
Post-Lockdown & Pre-PPP × Congressional District PPPE (Nbr.)
-0.041∗∗∗
-0.040∗∗∗
(0.007)
(0.007)
Post-PPP × Congressional District PPPE (Nbr.)
-0.043∗∗∗
-0.042∗∗∗
(0.009)
(0.010)
Observations
726868
726868
726868
726868
Adjusted R2
0.261
0.262
0.522
0.522
P-Value Diff. in Coefﬁcients
.8864
.5061
.9106
.6163
Firm Fixed Effects
No
No
Yes
Yes
State×Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
Panel B: Decline in Hours Worked
(1)
(2)
(3)
(4)
Decline Hours Worked
Pre-Covid × Fraction receiving PPP
-0.007
(0.009)
Pre-Lockdown × Fraction receiving PPP
0.062∗∗
0.068∗∗∗
(0.025)
(0.024)
Post-Lockdown & Pre-PPP × Fraction receiving PPP
0.109∗∗
0.113∗∗
(0.044)
(0.047)
Post-PPP × Fraction receiving PPP
0.112∗∗
0.116∗∗
(0.046)
(0.049)
Pre-Covid × Congressional District PPPE (Nbr.)
0.000
(0.002)
Pre-Lockdown × Congressional District PPPE (Nbr.)
0.026∗∗∗
0.026∗∗∗
(0.004)
(0.003)
Post-Lockdown & Pre-PPP × Congressional District PPPE (Nbr.)
0.040∗∗∗
0.040∗∗∗
(0.009)
(0.010)
Post-PPP × Congressional District PPPE (Nbr.)
0.044∗∗∗
0.043∗∗∗
(0.009)
(0.009)
Observations
726868
726868
726868
726868
Adjusted R2
0.419
0.420
0.659
0.660
P-Value Diff. in Coefﬁcients
.813
.1661
.7928
.1718
Firm Fixed Effects
No
No
Yes
Yes
State×Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
50

Table 4: Evolution of County Unemployment Outcomes and Exposure to PPP
Table 4 reports the results of ordinary least squares (OLS) regressions examining the relation between the geographic allocation of PPP funds
during the ﬁrst round and weekly initial unemployment insurance ﬁlings at the county level between the second week of March until the last
week of April. The dependent variable, County Initial UI Filings Ratios, is the ratio between the weekly initial unemployment insurance ﬁlings
hand-collected from several state labor departments and county employment measured using the Bureau of Labor and Statistics January
County Employment data. Pre-Covid is an indicator variable that takes the value of prior to the ﬁrst week of March. This period represents the
base period in the speciﬁcations that include county ﬁxed effects and as a result this variable is omitted in such speciﬁcations. Pre-Lockdown
is an indicator variable that takes the value of one prior to the week of March 22nd – March 28th (exclusive) Post-Lockdown & Pre-PPP is
an indicator variable that takes the value of one during the weeks of March 22nd – March 28th and March 29th – April 4th. Post-PPP is
an indicator variable that takes the value of one following the week of April 5th – April 12th (inclusive), County PPPE (Nbr.) is the county
average of the PPPE based on the number of outstanding loans, weighted by the share of branches of each bank in each county. Weekly County
Covid-19 Cases (per capita) are the weekly per capita number of reported Covid-19 cases. Weekly County Covid-19 Deaths (per capita) are the
weekly per capita number of reported Covid-19 deaths. Social Distancing Index is the change in average distance travelled in the county using
individuals’ GPS signals. Standard errors are presented in parentheses, and are double-clustered at the level of the state and week. ***, **,
and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
County Initial UI Filings Ratio
Pre-Covid × County PPPE (Nbr.)
0.009∗
0.009∗
(0.005)
(0.005)
Pre-Lockdown × County PPPE (Nbr.)
-0.021
-0.019
-0.032∗∗∗
-0.029∗∗∗
(0.014)
(0.011)
(0.004)
(0.006)
Post-Lockdown & Pre-PPP × County PPPE (Nbr.)
-0.089∗
-0.077
-0.100
-0.096
(0.049)
(0.052)
(0.063)
(0.064)
Post-PPP × County PPPE (Nbr.)
-0.076∗∗
-0.050
-0.085∗
-0.070
(0.034)
(0.037)
(0.048)
(0.048)
Weekly County Covid-19 Cases (per capita)
3.389
3.324∗∗
(2.050)
(1.274)
Weekly County Covid-19 Deaths (per capita)
8.186
-32.817∗∗
(20.726)
(11.587)
Social Distancing Index
-0.787∗∗∗
-0.654∗∗∗
(0.199)
(0.145)
Observations
23971
23889
23971
23887
Adjusted R2
0.758
0.762
0.830
0.832
P-Value Diff. in Coefﬁcients
.3909
.2305
.4621
.2591
County Fixed Effects
No
No
Yes
Yes
State×Week Fixed Effects
Yes
Yes
Yes
Yes
51

Table 5: Evolution of Small Business Revenues and Exposure to PPP
Table 5 reports the results of ordinary least squares (OLS) regressions examining the relation between the geographic allocation of PPP funds
during the ﬁrst round and weekly small business revenues at the county×industry level between the second week of March until the last
week of April collected from Womply, a company specializing in processing revenue for small businesses. The dependent variable, Y/Y Change
in Total Consumer Spending, is the average year-on-year change in small business revenues in a given county and industry. Pre-Covid is an
indicator variable that takes the value of prior to the ﬁrst week of March. This period represents the base period in the speciﬁcations that
include county×industry ﬁxed effects and as a result this variable is omitted in such speciﬁcations. Pre-Lockdown is an indicator variable that
takes the value of one prior to the week of March 22nd – March 28th (exclusive) Post-Lockdown & Pre-PPP is an indicator variable that takes
the value of one during the weeks of March 22nd – March 28th and March 29th – April 4th. Post-PPP is an indicator variable that takes the
value of one following the week of April 5th – April 12th (inclusive), County PPPE (Nbr.) is the county average of the PPPE based on the
number of outstanding loans, weighted by the share of branches of each bank in each county. Weekly County Covid-19 Cases (per capita) are
the weekly per capita number of reported Covid-19 cases. Weekly County Covid-19 Deaths (per capita) are the weekly per capita number of
reported Covid-19 deaths. Social Distancing Index is the change in average distance travelled in the county using individuals’ GPS signals.
Standard errors are presented in parentheses, and are double-clustered at the level of the state and week. ***, **, and *, represent statistical
signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
Y/Y Change in Total Consumer Spending
Pre-Covid × County PPPE (Nbr.)
-0.068∗∗∗
-0.075∗∗∗
(0.009)
(0.009)
Pre-Lockdown × County PPPE (Nbr.)
-0.032∗∗
-0.045∗∗∗
0.033∗∗∗
0.028∗∗∗
(0.012)
(0.011)
(0.010)
(0.009)
Post-Lockdown & Pre-PPP × County PPPE (Nbr.)
-0.019∗∗∗
-0.035∗∗∗
0.042∗∗∗
0.033∗∗∗
(0.006)
(0.007)
(0.006)
(0.007)
Post-PPP × County PPPE (Nbr.)
-0.014∗∗
-0.033∗∗∗
0.046∗∗∗
0.029∗∗
(0.005)
(0.008)
(0.008)
(0.010)
Weekly County Covid-19 Cases (per capita)
-0.017
0.163
(0.306)
(0.349)
Weekly County Covid-19 Deaths (per capita)
4.020
-7.032
(3.781)
(8.209)
Social Distancing Index
0.255∗∗∗
0.288∗∗∗
(0.071)
(0.063)
Observations
210076
193328
209950
193216
Adjusted R2
0.362
0.361
0.665
0.668
P-Value Diff. in Coefﬁcients
.1093
.6282
.3683
.3787
County×Industry Fixed Effects
No
No
Yes
Yes
State×Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
52

Table 6: % Receiving PPP and Exposure to State PPPE
Table 6 reports the results of ordinary least squares (OLS) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst round and the percentage of ﬁrms
reporting having received PPP funds at the state-by-industry level in each week of the survey collected from the Census Small Business Pulse Survey. The dependent variable is the percentage
of businesses that received PPP funds in a state-by-industry group during each week of the survey. State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans.
Standard errors are presented in parentheses, and are double-clustered at the level of the state and week. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
% PPP Received
State PPPE (Nbr.)
35.865∗∗∗
89.549∗∗∗
51.603∗∗∗
23.899∗∗∗
28.381∗∗∗
17.061∗∗∗
24.315∗∗∗
(3.201)
(8.169)
(4.298)
(3.594)
(5.024)
(3.610)
(3.879)
Observations
2230
277
386
390
390
393
394
Adjusted R2
0.653
0.629
0.529
0.397
0.416
0.387
0.516
Industry Fixed Effects
No
Yes
Yes
Yes
Yes
Yes
Yes
Industry×Week Fixed Effects
Yes
No
No
No
No
No
No
Week:
Full Sample
Apr26–May2
May3–May9
May10–May16
May17–May23
May24–May30
May31–Jun6
53

Table 7: % Receiving PPP and Missed Loans Payments
Table 7 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst round
and the percentage of ﬁrms reporting missing payments at the state-by-industry level collected from the Census Small Business Pulse Survey. The dependent variable is the percentage of ﬁrms
reporting a missed scheduled loan payment. % PPP Received is the percentage of businesses reporting having received PPP funds in a state-by-industry group. State PPPE (Nbr.) is the state average
of the PPPE based on the number of outstanding loans. Pre-PPP Decline Hours Worked is the average decline in hours worked in each state between January and the last week of March. Pre-PPP
State Covid-19 Cases (per capita) are per capita number of reported Covid-19 cases in the state. Pre-PPP State Covid-19 Deaths (per capita) are the weekly per capita number of reported Covid-19
deaths in the state. Pre-PPP State Social Distancing Index is the change in average distance travelled in the state until the end of March using individuals’ GPS signals. All speciﬁcations include
industry×week ﬁxed-effects. Standard errors are presented in parentheses, and are clustered at the level of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels,
respectively.
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Missing Loan Payments
% Missing Loan Payments
% PPP Received
% Missing Loan Payments
%PPP Received
-0.005
-0.149∗∗
(0.014)
(0.045)
State PPPE (Nbr.)
-4.401∗∗∗
34.402∗∗∗
(1.380)
(3.055)
Pre-PPP Decline Hours Worked
-3.536∗∗
-3.242∗∗∗
8.034∗∗
-1.740
(1.330)
(0.977)
(3.041)
(0.904)
Pre-PPP State Covid-19 Cases (per capita)
0.853∗
0.801∗
1.639
1.001∗
(0.466)
(0.409)
(0.995)
(0.473)
Pre-PPP State Covid-19 Deaths (per capita)
-26.308∗
-26.047∗
-7.200
-23.355
(13.770)
(13.396)
(39.721)
(15.393)
Pre-PPP Social Distancing Index
-5.795∗∗
-6.787∗∗∗
14.786
-2.501
(2.803)
(2.363)
(8.984)
(2.421)
Observations
2218
2487
2218
2218
Adjusted R2
0.573
0.554
0.664
-0.064
FStat
17.386
Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
54

Table 8: % Receiving PPP and Missed Other Scheduled Payments
Table 8 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst round
and the percentage of ﬁrms reporting missing payments at the state-by-industry level collected from the Census Small Business Pulse Survey. The dependent variable is the percentage of ﬁrms
reporting a missed other scheduled payment such as rent, utilities, and payroll. % PPP Received is the percentage of businesses reporting having received PPP funds in a state-by-industry group.
State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans. Pre-PPP Decline Hours Worked is the average decline in hours worked in each state between January
and the last week of March. Pre-PPP State Covid-19 Cases (per capita) are per capita number of reported Covid-19 cases in the state. Pre-PPP State Covid-19 Deaths (per capita) are the weekly
per capita number of reported Covid-19 deaths in the state. Pre-PPP State Social Distancing Index is the change in average distance travelled in the state until the end of March using individuals’
GPS signals. All speciﬁcations include industry×week ﬁxed-effects. Standard errors are presented in parentheses, and are clustered at the level of the state. ***, **, and *, represent statistical
signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Missing Schd. Payments
% Missing Schd. Payments
% PPP Received
% Missing Schd. Payments
%PPP Received
-0.091∗∗∗
-0.435∗∗∗
(0.022)
(0.067)
State PPPE (Nbr.)
-14.807∗∗∗
34.662∗∗∗
(2.651)
(3.053)
Pre-PPP Decline Hours Worked
-12.380∗∗∗
-12.176∗∗∗
7.824∗∗
-8.192∗
(3.587)
(4.094)
(3.097)
(4.094)
Pre-PPP State Covid-19 Cases (per capita)
4.390∗∗∗
4.189∗∗∗
1.674∗
4.754∗∗∗
(0.736)
(0.577)
(0.988)
(0.462)
Pre-PPP State Covid-19 Deaths (per capita)
-107.446∗∗∗
-101.315∗∗∗
-8.416
-100.872∗∗∗
(25.205)
(20.723)
(39.435)
(17.170)
Pre-PPP Social Distancing Index
-15.797∗∗
-13.084∗∗
14.833
-7.857
(6.450)
(6.317)
(9.059)
(5.592)
Observations
2206
2448
2206
2206
Adjusted R2
0.708
0.698
0.663
-0.028
FStat
128.871
Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
55

Table 9: % Receiving PPP and Cash-on-Hand
Table 9 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst round
and the percentage of businesses at the state-by-industry level reporting at least two months of cash on hand to sustain operations as collected from the Census Small Business Pulse Survey. The
dependent variable is the fraction of businesses with cash on hand to sustain operations for two months or more. % PPP Received is the percentage of businesses reporting having received PPP
funds in a state-by-industry group. State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans. Pre-PPP Decline Hours Worked is the average decline in hours
worked in each state between January and the last week of March. Pre-PPP State Covid-19 Cases (per capita) are per capita number of reported Covid-19 cases in the state. Pre-PPP State Covid-19
Deaths (per capita) are the weekly per capita number of reported Covid-19 deaths in the state. Pre-PPP State Social Distancing Index is the change in average distance travelled in the state until the
end of March using individuals’ GPS signals. All speciﬁcations include industry×week ﬁxed-effects. Standard errors are presented in parentheses, and are clustered at the level of the state. ***,
**, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Cash 3 months
% Cash 3 months
% PPP Received
% Cash 3 months
%PPP Received
-0.000
0.270∗
(0.035)
(0.125)
State PPPE (Nbr.)
8.920∗∗
32.502∗∗∗
(3.768)
(3.116)
Pre-PPP Decline Hours Worked
11.966∗
15.861∗∗∗
4.354
11.007∗
(6.170)
(5.378)
(5.855)
(4.844)
Pre-PPP State Covid-19 Cases (per capita)
-1.979∗∗
-2.003∗∗
2.095∗∗
-2.476∗∗
(0.733)
(0.818)
(0.937)
(0.726)
Pre-PPP State Covid-19 Deaths (per capita)
75.244∗∗
76.722∗∗
-26.761
76.491∗
(28.941)
(31.278)
(44.659)
(35.062)
Pre-PPP Social Distancing Index
-6.850
-12.334
13.236
-12.138
(9.277)
(11.326)
(11.661)
(9.223)
Observations
903
918
903
903
Adjusted R2
0.629
0.640
0.797
-0.089
FStat
19.527
Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
56

A
Case Study: Wells Fargo
This section explores a case study, illlustrating the importance of ﬁnancial intermediation in
allocating PPP funds. Figure A.1 and Table A.1 present a case study of a particular bank with
a very low share of PPP loans relative to overall market share—Wells Fargo. Wells Fargo was
severely constrained from expanding its balance sheet as a result of an asset cap imposed by
the Fed in the aftermath of the fake accounts scandal. This asset cap was only lifted on April
10, when the Fed excluded PPP loans from the formula it uses to restrict Wells Fargo’s growth.
The asset cap limited Wells Fargo’s ability to lend under the PPP in the early days for the ﬁrst
phase of the program. Table 1 shows that Wells Fargo, the third largest bank in the nation
by total assets, held a 6.5% share of the total outstanding small business loans but processed
only 0.04% of the total volume of loans in the PPP program until April 15. Figure A.1 shows
PPP allocations by the market share of Wells Fargo. The top panel shows volume, while the
bottom panel shows the number of loans. Both ﬁgures show a similar pattern—ﬁrms located in
areas with higher Wells Fargo market share see lower PPP allocations, both in terms of overall
aggregate loan volume per business and in the fraction of businesses receiving PPP loans.
Table A.1 presents similar information, regressing the log total volume and number of loans
on the share of Wells Fargo branches in congressional districts. The relationship between PPP
allocations and Wells Fargo branches echoes the results seen in Table 2. Areas with higher Wells
Fargo exposure (and lower PPPE exposure) see lower PPP allocations. The effect is statistically
signiﬁcant at the 5 percent level or higher in columns (1) through (3), but loses signiﬁcance
when state ﬁxed effects are included. More broadly, Figure A.1 and Table A.1 illustrate an
example of the variation underlying overall PPPE exposure.
57

Figure A.1: Wells Fargo Exposure and PPP per Establishment
Figure A.1 are scatterplots of the total PPP allocation per establishment in the congressional district and the share of branches of Wells Fargo
in the Congressional District (Panel A) and fraction of establishments receiving a PPP loan and the share of Branches of Wells Fargo in the
Congressional District (Panel B). Data comes from the SBA, Summary of Deposits, and County Business Patterns.
0
50000
100000
150000
PPP allocation per Establishment (Congressional District)
0
.1
.2
.3
.4
Share of Wells Fargo Branches
0
.1
.2
.3
.4
.5
Fraction of Establishments receiving PPP (Congressional District)
0
.1
.2
.3
.4
Share of Wells Fargo Branches
58

Figure A.2: Decline in Hours Worked and PPP Allocation by State
Figure A.3 presents four scatterplots of the decline in hours worked relative to a January baseline and four alternative measures of allocation of PPP funds across states. The ﬁgure on the top left
plots the amount of PPP loans received by small businesses in each state divided by the total number of small businesses in the state. The ﬁgure on the top right corner plots the fraction of small
businesses in each state that received a PPP loan. The ﬁgure on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to the
PPPE measured in terms of the total volume of loans. The ﬁgure on the bottom right corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to
the PPPE measured in terms of the total number of loans.
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
ILIN
KS
KYLA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
0
.25
.5
.75
1
Ratio of Hours Worked relative to January Baseline
30000
40000
50000
60000
70000
PPP Amount per Small Business in State
(Date: 31/03/2020)
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL IN
KS
KY
LA
MA
MD
ME
MI
MN
MO MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
0
.25
.5
.75
1
Ratio of Hours Worked relative to January Baseline
.1
.2
.3
.4
.5
Fraction of Businesses receiving PPP
(Date: 31/03/2020)
AK
ALAR
AZ
CA
CO
CT
DC
DE
FLGA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
0
.25
.5
.75
1
Ratio of Hours Worked relative to January Baseline
-.3
-.2
-.1
0
.1
.2
State PPPE (Volume)
(Date: 31/03/2020)
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FL GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
0
.25
.5
.75
1
Ratio of Hours Worked relative to January Baseline
-.4
-.2
0
.2
.4
State PPPE (Nbr. Loans)
(Date: 31/03/2020)
59

Figure A.3: Business Shutdowns and PPP Allocation by State
Figure A.3 presents four scatterplots of the share of businesses in each state that shutdown that in the week of March29th–April 4th and four alternative measures of allocation of PPP funds across
states. The ﬁgure on the top left plots the amount of PPP loans received by small businesses in each state divided by the total number of small businesses in the state. The ﬁgure on the top right
corner plots the fraction of small businesses in each state that received a PPP loan. The ﬁgure on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan
and the state exposure to the PPPE measured in terms of the total volume of loans. The ﬁgure on the bottom right corner plots the fraction of small businesses in each state that received a PPP
loan and the state exposure to the PPPE measured in terms of the total number of loans.
AK
AL
AR
AZ
CA
COCT
DC
DE
FL
GA
HI
IA
ID
ILIN
KS
KYLA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
0
.2
.4
.6
.8
Share of Business Shutdowns (State)
30000
40000
50000
60000
70000
PPP Amount per Small Business in State
(Date: 31/03/2020)
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FL
GA
HI
IA
ID
IL IN
KS
KY
LA
MA
MD
ME
MI
MN
MO MSMT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
0
.2
.4
.6
.8
Share of Business Shutdowns (State)
.1
.2
.3
.4
.5
Fraction of Businesses receiving PPP in State
(Date: 31/03/2020)
AK
AL
AR
AZ
CA
COCT
DC
DE
FL GA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
0
.2
.4
.6
.8
Share of Business Shutdowns (State)
-.4
-.2
0
.2
.4
State PPPE (Nbr. Loans)
(Date: 31/03/2020)
AK
AL
AR
AZ
CA
CO
CT
DC
DE
FLGA
HI
IA
ID
IL
IN
KS
KY
LA
MA
MD
ME
MI
MN
MO
MS
MT
NC
ND
NE
NH
NJ
NM
NV
NY
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VA
VT
WA
WI
WV
WY
0
.2
.4
.6
.8
Share of Business Shutdowns (State)
-.3
-.2
-.1
0
.1
.2
State PPPE (Volume)
(Date: 31/03/2020)
60

Figure A.4: COVID-19 Cases and PPP Allocation by State
Figure A.4 presents four scatterplots of the number of conﬁrmed COVID-19 cases per thousand as of April, 3rd 2020 and four alternative measures of allocation of PPP funds across states. The
ﬁgure on the top left plots the amount of PPP loans received by small businesses in each state divided by the total number of small businesses in the state. The ﬁgure on the top right corner plots
the fraction of small businesses in each state that received a PPP loan. The ﬁgure on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan and the state
exposure to the PPPE measured in terms of the total volume of loans. The ﬁgure on the bottom right corner plots the fraction of small businesses in each state that received a PPP loan and the
state exposure to the PPPE measured in terms of the total number of loans. Data comes from the Center for Disease Control, SBA, Call Reports, and FDIC Summary of Deposits.
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
0
1
2
3
Covid-19 Cases per Thousand (04/03/2020)
30000
40000
50000
60000
70000
PPP volume per Small Business Establishment
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
0
1
2
3
Covid-19 Cases per Thousand (04/03/2020)
.1
.2
.3
.4
.5
Fraction of Small Businesses Receiving PPP loan
ALAK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
IDIL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
0
1
2
3
Covid-19 Cases per Thousand (04/03/2020)
-.2
-.1
0
.1
.2
State PPPE (Volume)
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL
IN
IAKS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
0
1
2
3
Covid-19 Cases per Thousand (04/03/2020)
-.2
0
.2
.4
State PPPE (Nbr. Loans)
61

Figure A.5: COVID-19 Deaths and PPP Allocation by State
Figure A.5 presents four scatterplots of the number of COVID-19 deaths per thousand as of April, 3rd 2020 and four alternative measures of allocation of PPP funds across states. The ﬁgure on the
top left plots the amount of PPP loans received by small businesses in each state divided by the total number of small businesses in the state. The ﬁgure on the top right corner plots the fraction of
small businesses in each state that received a PPP loan. The ﬁgure on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to
the PPPE measured in terms of the total volume of loans. The ﬁgure on the bottom right corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure
to the PPPE measured in terms of the total number of loans. Data comes from the Center for Disease Control, SBA, Call Reports, and FDIC Summary of Deposits.
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
ILIN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
0
.02
.04
.06
.08
Covid-19 Deaths per Thousand (04/03/2020)
30000
40000
50000
60000
70000
PPP volume per Small Business Establishment
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA RI
SC
SD
TN
TX UT
VT
VA
WA
WV
WI
WY
0
.02
.04
.06
.08
Covid-19 Deaths per Thousand (04/03/2020)
.1
.2
.3
.4
.5
Fraction of Small Businesses Receiving PPP loan
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
0
.02
.04
.06
.08
Covid-19 Deaths per Thousand (04/03/2020)
-.2
-.1
0
.1
.2
State PPPE (Volume)
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL
IN
IAKS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
0
.02
.04
.06
.08
Covid-19 Deaths per Thousand (04/03/2020)
-.2
0
.2
.4
State PPPE (Nbr. Loans)
62

Figure A.6: Shelter-in-Place Orders and PPP Allocation by State
Figure A.6 presents four scatterplots of the timing of statewide shelter-in-place orders and four alternative measures of allocation of PPP funds across states. The ﬁgure on the top left plots the
amount of PPP loans received by small businesses in each state divided by the total number of small businesses in the state. The ﬁgure on the top right corner plots the fraction of small businesses
in each state that received a PPP loan. The ﬁgure on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to the PPPE measured
in terms of the total volume of loans. The ﬁgure on the bottom right corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to the PPPE measured
in terms of the total number of loans. Data comes from the New York Times, SBA, Call Reports, and FDIC Summary of Deposits.
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WAWV
WI
WY
30000
40000
50000
60000
70000
PPP allocation per Small Business Establishment
18mar2020
19mar2020
20mar2020
21mar2020
22mar2020
23mar2020
24mar2020
25mar2020
26mar2020
27mar2020
28mar2020
29mar2020
30mar2020
31mar2020
01apr2020
02apr2020
03apr2020
04apr2020
05apr2020
06apr2020
07apr2020
No State Order
Date Shelter Order
AL
AK
AZ
AR
CA
CO
CTDE
DC FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
.1
.2
.3
.4
.5
Fraction of Small Businesses Receiving PPP Loans
18mar2020
19mar2020
20mar2020
21mar2020
22mar2020
23mar2020
24mar2020
25mar2020
26mar2020
27mar2020
28mar2020
29mar2020
30mar2020
31mar2020
01apr2020
02apr2020
03apr2020
04apr2020
05apr2020
06apr2020
07apr2020
No State Order
Date Shelter Order
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WVWI
WY
-.2
-.1
0
.1
.2
State PPPE (Volume)
18mar2020
19mar2020
20mar2020
21mar2020
22mar2020
23mar2020
24mar2020
25mar2020
26mar2020
27mar2020
28mar2020
29mar2020
30mar2020
31mar2020
01apr2020
02apr2020
03apr2020
04apr2020
05apr2020
06apr2020
07apr2020
No State Order
Date Shelter Order
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WVWI
WY
-.2
0
.2
.4
State PPPE (Nbr. Loans)
18mar2020
19mar2020
20mar2020
21mar2020
22mar2020
23mar2020
24mar2020
25mar2020
26mar2020
27mar2020
28mar2020
29mar2020
30mar2020
31mar2020
01apr2020
02apr2020
03apr2020
04apr2020
05apr2020
06apr2020
07apr2020
No State Order
Date Shelter Order
63

Figure A.7: Social Distancing Index and PPP Allocation by State
Figure A.7 presents four scatterplots of and four alternative measures of allocation of PPP funds across states. The ﬁgure on the top left plots the amount of PPP loans received by small businesses
in each state divided by the total number of small businesses in the state. The ﬁgure on the top right corner plots the fraction of small businesses in each state that received a PPP loan. The ﬁgure
on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to the PPPE measured in terms of the total volume of loans. The ﬁgure
on the bottom right corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to the PPPE measured in terms of the total number of loans. Data
comes from the SBA, Call Reports, and FDIC Summary of Deposits.
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
-.6
-.5
-.4
-.3
-.2
Social Distance Index (03/28/2020)
30000
40000
50000
60000
70000
PPP volume per Small Business Establishment
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
-.6
-.5
-.4
-.3
-.2
Social Distance Index (03/28/2020)
.1
.2
.3
.4
.5
Fraction of Small Businesses Receiving PPP loan
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
-.6
-.5
-.4
-.3
-.2
Social Distance Index (03/28/2020)
-.2
-.1
0
.1
.2
State PPPE (Volume)
AL
AK
AZ
AR
CA
CO
CT
DE
DC
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
-.6
-.5
-.4
-.3
-.2
Social Distance Index (03/28/2020)
-.2
0
.2
.4
State PPPE (Nbr. Loans)
64

Figure A.8: State Unemployment Insurance and PPP Allocation by State
Figure A.8 presents four scatterplots of the ratio of state unemployment insurance claims to employment covered by unemployment insurance and four alternative measures of allocation of PPP
funds across states. State unemployment insurance claims are the sum of ﬁled claim in the weeks ended March 21st, March 28th, and April 4th, 2020. The ﬁgure on the top left plots the amount
of PPP loans received by small businesses in each state divided by the total number of small businesses in the state. The ﬁgure on the top right corner plots the fraction of small businesses in each
state that received a PPP loan. The ﬁgure on the bottom left corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to the PPPE measured in
terms of the total volume of loans. The ﬁgure on the bottom right corner plots the fraction of small businesses in each state that received a PPP loan and the state exposure to the PPPE measured
in terms of the total number of loans. Data comes from the Department of Labor, SBA, Call Reports, and FDIC Summary of Deposits.
AL
AK
AZ
AR
CA
COCT
DE
FL
GA
HI
ID
IL
IN
IA
KS
KYLA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
.05
.1
.15
.2
State Unemployment Claims by State (% Cov. Employment)
30000
40000
50000
60000
70000
PPP volume per Small Business Establishment
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NYNC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX UT
VT
VA
WA
WV
WI
WY
.05
.1
.15
.2
State Unemployment Claims by State (% Cov. Employment)
.1
.2
.3
.4
.5
Fraction of Small Businesses Receiving PPP loan
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL
IN
IA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NY
NC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
.05
.1
.15
.2
State Unemployment Claims by State (% Cov. Employment)
-.2
-.1
0
.1
.2
State PPPE (Volume)
AL
AK
AZ
AR
CA
CO
CT
DE
FL
GA
HI
ID
IL
INIA
KS
KY
LA
ME
MD
MA
MI
MN
MS
MO
MT
NE
NV
NH
NJ
NM
NYNC
ND
OH
OK
OR
PA
RI
SC
SD
TN
TX
UT
VT
VA
WA
WV
WI
WY
.05
.1
.15
.2
State Unemployment Claims by State (% Cov. Employment)
-.2
0
.2
.4
State PPPE (Nbr. Loans)
65

Figure A.9: Weekly Elasticities of Employment Outcomes to % Businesses Receiving PPP
Figure A.9 plots coefﬁcients and respective standard errors of regression analyses investigating the week-by-week elasticities between employment and ﬁrm outcomes and exposure to PPPE. The
left ﬁgure plots the coefﬁcients and respective standard errors of an empirical speciﬁcation that augments the speciﬁcation of Panel A of Table 5 column (3) to include a full set of interactions
between the fraction of businesses receiving PPP in the Congressional District during the ﬁrst round and week dummy variables. The ﬁgure on the right plots the coefﬁcients and respective standard
errors of an empirical speciﬁcation that augments the speciﬁcation of Panel B of Table 5 column (3) to include a full set of interactions between the Congressional District PPPE and week dummy
variables.
PPP Launch
Vertical bands represent +/- 1.96 * St. Error of each point estimate
Initial Lockdown Orders
PPP Runs out of Funds
PPP 2nd Round begins
-.2
-.15
-.1
-.05
0
.05
Beta
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
State-Industry-Week and Firm Fixed Effects
Weekly Elasticities of Shutdown and Fraction Receiving PPP
PPP Launch
Vertical bands represent +/- 1.96 * St. Error of each point estimate
Initial Lockdown Orders
PPP Runs out of Funds
PPP 2nd Round begins
-.1
-.05
0
.05
.1
.15
.2
Beta
Week 2: Jan26-Feb1
Week 3: Feb02-Feb8
Week 4: Feb9-Feb15
Week 5: Feb16-Feb22
Week 6: Feb23-Feb29
Week 7: Mar1-Mar7
Week 8:  Mar8-Mar14
Week 9:  Mar15-Mar21
Week 10:  Mar22-Mar28
Week 11:  Mar29-Apr4
Week 12: Apr5-Apr11
Week 13: Apr12-Apr18
Week 14: Apr19-Apr25
Week 15: Apr26-May2
State-Industry-Week and Firm Fixed Effects
Weekly Elasticities of Shutdown and Fraction Receiving PPP
66

Figure A.10: Unmet PPP Demand and Exposure to State PPPE (Industry×State)
Figure A.10 are scatterplots of the state exposure to the number-based PPPE and the percentage of businesses in each industry within a state
that applied but did not receive PPP. The plots represent the evolution of the two variables for each survey week. Data comes from the Census
Bureau, SBA, Call Reports, Summary of Deposits, and County Business Patterns.
AR
CA
CO
FL
IA
IL
IN
KY
MA
MD
MI
MN
MO
NC
NH
NJ
NY
OH
OK
OR
PA
TX
VA
WA
WI
AL
AL
AL
AL
AL
AR
AR
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
IA
IA
IA
IA
IA
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KY
KY
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
ME
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MS
MS
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NH
NJ
NJ
NJ
NJ
NJ
NJ
NM
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 15: April26 - May2nd
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 16: May3rd - May9
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NE
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NV
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PR
PR
PR
SC
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 17: May10 - May16
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NENH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NM
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PR
PR
SC
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 18: May17 - May23
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NE
NH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NM
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PR
PR
PR
SC
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 19: May24 - May30
AL
AL
AL
AL
AL
AL
AL
AR
AR
AR
AR
AR
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
AZ
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CA
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CO
CT
CT
CT
CT
CT
CT
CT
CT
DC
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
FL
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
GA
HI
IA
IA
IA
IA
IA
IA
IA
ID
ID
ID
ID
ID
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IL
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
IN
KS
KS
KS
KS
KS
KS
KS
KY
KY
KY
KY
KY
LA
LA
LA
LA
LA
LA
LA
LA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MA
MD
MD
MD
MD
MD
MD
MD
MD
MD
MD
ME
ME
ME
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MI
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MN
MO
MO
MO
MO
MO
MO
MO
MO
MO
MS
MS
MS
MS
MT
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NE
NE
NE
NE
NENH
NH
NH
NH
NH
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NJ
NM
NM
NV
NV
NV
NV
NV
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
NY
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OK
OK
OK
OK
OK
OK
OK
OK
OK
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
OR
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PA
PR
PR
PR
SC
SC
SC
SC
SC
SC
SC
SC
TN
TN
TN
TN
TN
TN
TN
TN
TN
TN
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
TX
UT
UT
UT
UT
UT
UT
UT
UT
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VA
VT
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WA
WI
WI
WI
WI
WI
WI
WI
WI
WI
WI
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
0
.2
.4
State PPPE (Nbr. Loans)
Week 20: May31 - Jun6
67

Figure A.11: % Receiving PPP and Exposure to MSA PPPE
Figure A.11 are scatterplots of the MSA exposure to the number-based PPPE and the percentage of businesses in each MSA that applied but
did not receive PPP in each week. The plots represent the evolution of the two variables for each survey week. Data comes from the Census
Bureau, SBA, Call Reports, Summary of Deposits, and County Business Patterns.
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
70
80
90
100
% of Businesses reporting receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 15: April26 - May2nd
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CODetroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
70
80
90
100
% of Businesses reporting receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 16: May3rd - May9
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
70
80
90
100
% of Businesses reporting receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 17: May10 - May16
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
70
80
90
100
% of Businesses reporting receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 18: May17 - May23
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
70
80
90
100
% of Businesses reporting receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 19: May24 - May30
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
70
80
90
100
% of Businesses reporting receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 20: May31 - Jun6
68

Figure A.12: Unmet PPP Demand and Exposure to MSA PPPE
Figure A.12 are scatterplots of the MSA exposure to the number-based PPPE and the percentage of businesses in each MSA that applied but
did not receive PPP in each week. The plots represent the evolution of the two variables for each survey week. Data comes from the Census
Bureau, SBA, Call Reports, Summary of Deposits, and County Business Patterns.
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 15: April26 - May2nd
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 16: May3rd - May9
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FLPhiladelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 17: May10 - May16
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CODetroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FLPhiladelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 18: May17 - May23
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 19: May24 - May30
Atlanta-Sandy Springs-Alpharetta, GA
Austin-Round Rock-Georgetown, TX
Baltimore-Columbia-Towson, MD
Birmingham-Hoover, AL
Boston-Cambridge-Newton, MA-NH
Buffalo-Cheektowaga, NY
Charlotte-Concord-Gastonia, NC-SC
Chicago-Naperville-Elgin, IL-IN-WI
Cincinnati, OH-KY-IN
Cleveland-Elyria, OH
Columbus, OH
Dallas-Fort Worth-Arlington, TX
Denver-Aurora-Lakewood, CO
Detroit-Warren-Dearborn, MI
Hartford-East Hartford-Middletown, CT
Houston-The Woodlands-Sugar Land, TX
Indianapolis-Carmel-Anderson, IN
Jacksonville, FL
Kansas City, MO-KS
Las Vegas-Henderson-Paradise, NV
Los Angeles-Long Beach-Anaheim, CA
Louisville/Jefferson County, KY-IN
Memphis, TN-MS-AR
Miami-Fort Lauderdale-Pompano Beach, FL
Milwaukee-Waukesha, WI
Minneapolis-St. Paul-Bloomington, MN-WI
Nashville-Davidson--Murfreesboro--Franklin, TN
New Orleans-Metairie, LA
New York-Newark-Jersey City, NY-NJ-PA
Oklahom
Orlando-Kissimmee-Sanford, FL
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD
Phoenix-Mesa-Chandler, AZ
Pittsburgh, PA
Portland-Vancouver-Hillsboro, OR-WA
Providence-Warwick, RI-MA
Raleigh-Cary, NC
Richmond, VA
Riverside-San Bernardino-Ontario, CA
Sacramento-Roseville-Folsom, CA
St. Louis, MO-IL
Salt Lake City, UT
San Antonio-New Braunfels, TX
San Diego-Chula Vista-Carlsbad, CA
San Francisco-Oakland-Berkeley, CA
San Jose-Sunnyvale-Santa Clara, CA
Seattle-Tacoma-Bellevue, WA
Tampa-St. Petersburg-Clearwater, FL
Virginia Beach-Norfolk-Newport News, VA-NC
Washington-Arlington-Alexandria, DC-VA-MD-WV
0
10
20
30
40
50
60
Diff. between % of businesses applying and receiving PPP
-.2
-.1
0
.1
.2
.3
MSA PPPE (Nbr. Loans)
Week 20: May31 - Jun6
69

Table A.1: Wells Fargo and PPP Allocation
Table A.1 reports the results of ordinary least squares (OLS) regressions examining the impact of the exposure of the Congressional District
to Wells Fargo on the allocation of PPP funds. The dependent variable of the speciﬁcations, Ln(Total PPP Allocation per establishment),
is the natural logarithm total amount of PPP funds disbursed to small businesses in each congressional district divided by the number of
establishments in the congressional district. The dependent variable of the speciﬁcations in Panel B, Ln(Total PPP Allocation per establishment)
is the total number of PPP loans made to small businesses in each congressional district divided by the number of establishments in the
congressional district. Share of Wells in the Congressional District is share of branches owned by Wells Fargo in the Congressional District.
Ln(Payroll) is the natural logarithm of the sum of payroll of all establishments in the congressional district Ln(Employment) is the natural
logarithm of total employment in the congressional district Industry Shares are additional controls for the share of establishments in each
two-digit NAICS code industry. The speciﬁcation of column (4) includes state ﬁxed effects Standard errors are presented in parentheses, and
are clustered at the level of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
Panel A: Total PPP Allocation per Establishment at the Congressional District
(1)
(2)
(3)
(4)
Ln(Total PPP Allocation per Establishment)
Share of Wells Branches in Cong Dist.
-1.871∗∗∗
-1.977∗∗∗
-1.277∗∗
-2.317
(0.526)
(0.594)
(0.629)
(1.721)
Ln(Total Payroll)
0.078
0.116
0.127
(0.299)
(0.253)
(0.251)
Ln(Employment)
0.259
0.105
0.183
(0.521)
(0.354)
(0.367)
Observations
436
436
436
436
Adjusted R2
0.047
0.088
0.125
0.097
Industry Shares
No
No
Yes
Yes
State Fixed Effects
No
No
No
Yes
Panel B: Fraction of Establishment receiving PPP at the Congressional District
(1)
(2)
(3)
(4)
Fraction of Establishments receiving PPP
Share of Wells Branches in Cong Dist.
-0.588∗∗∗
-0.492∗∗∗
-0.272∗∗
-0.287
(0.148)
(0.137)
(0.108)
(0.276)
Ln(Total Payroll)
-0.162∗∗∗
-0.119∗∗∗
-0.071∗
(0.061)
(0.043)
(0.041)
Ln(Employment)
0.233∗∗
0.172∗∗∗
0.099
(0.098)
(0.062)
(0.070)
Observations
436
436
436
436
Adjusted R2
0.151
0.210
0.376
0.455
Industry Shares
No
No
Yes
Yes
State Fixed Effects
No
No
No
Yes
70

Table A.2: Business Shutdowns and PPP Allocation
Table A.2 reports the results of ordinary least squares (OLS) regressions examining the relation between the allocation of PPP funds and the share of businesses that shutdown operations in the last
week of March. The dependent variable, Share of Firms Shutdown (March 31st, 2020), is the share of businesses in the congressional district that did not operate in the week of March29th–April 4th.
Ln(Total PPP Allocation per establishment) is the natural logarithm total amount of PPP funds disbursed to small businesses in each congressional district divided by the number of establishments in
the congressional district. Fraction of Establishments receiving PPP is the total number of PPP loans made to small businesses in each congressional district divided by the number of establishments
in the congressional district. Congressional District PPPE (Vol) is the congressional district average of the PPPE based on total amounts of outstanding PPP and small business loans, weighted by the
share of deposits of each bank in each congressional district. Congressional District PPPE (Nbr.) is the congressional district average of the PPPE based on the number of outstanding loans, weighted
by the share of deposits of each bank in each congressional district. Ln(Payroll) is the natural logarithm of the sum of payroll of all establishments in the congressional district Ln(Employment)
is the natural logarithm of total employment in the congressional district Industry Shares are additional controls for the share of establishments in each two-digit NAICS code industry. Standard
errors are presented in parentheses, and are clustered at the level of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
Share of Firms Shutdown (March 31st, 2020)
Fraction receiving PPP
-0.078
-0.025
(0.049)
(0.045)
Ln(PPP loan per establishment)
0.008
0.008
(0.007)
(0.006)
Congressional District PPPE (Vol)
0.020∗
-0.008
(0.010)
(0.008)
Congressional District PPPE (Nbr.)
0.019
-0.015
(0.012)
(0.011)
Ln(Total Payroll)
0.062
0.011
0.071∗
0.014
0.096∗∗
0.006
0.095∗∗
-0.004
(0.039)
(0.022)
(0.041)
(0.021)
(0.044)
(0.025)
(0.044)
(0.030)
Ln(Employment)
0.023
0.010
-0.000
-0.001
-0.038
0.016
-0.038
0.028
(0.050)
(0.043)
(0.053)
(0.040)
(0.064)
(0.046)
(0.066)
(0.051)
Observations
436
436
436
436
436
436
436
436
Adjusted R2
0.317
0.583
0.315
0.584
0.335
0.584
0.325
0.586
Industry Shares
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
State Fixed Effects
No
Yes
No
Yes
No
Yes
No
Yes
71

Table A.3: Decline in Hours Worked and PPP Allocation
Table A.3 reports the results of ordinary least squares (OLS) regressions examining the relation between the allocation of PPP funds and the share of businesses that shutdown operations in the last
week of March. The dependent variable, Decline in Hours Worked (March 31st, 2020), is the decline in hours worked at establishments located in the congressional district related to the average
hours worked in the same weekdays of the last two weeks of January at the same congressional district. Ln(Total PPP Allocation per establishment) is the natural logarithm total amount of PPP
funds disbursed to small businesses in each congressional district divided by the number of establishments in the congressional district. Fraction of Establishments receiving PPP is the total number
of PPP loans made to small businesses in each congressional district divided by the number of establishments in the congressional district. Congressional District PPPE (Vol) is the congressional
district average of the PPPE based on total amounts of outstanding PPP and small business loans, weighted by the share of deposits of each bank in each congressional district. Congressional District
PPPE (Nbr.) is the congressional district average of the PPPE based on the number of outstanding loans, weighted by the share of branches of each bank in each congressional district. Ln(Payroll)
is the natural logarithm of the sum of payroll of all establishments in the congressional district Ln(Employment) is the natural logarithm of total employment in the congressional district Industry
Shares are additional controls for the share of establishments in each two-digit NAICS code industry. Standard errors are presented in parentheses, and are clustered at the level of the state. ***,
**, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
Decline in Hours Worked (March 31st, 2020)
Fraction receiving PPP
0.121∗
0.051
(0.067)
(0.068)
Ln(PPP loan per establishment)
-0.005
-0.005
(0.007)
(0.007)
Congressional District PPPE (Vol)
-0.022∗∗
-0.001
(0.010)
(0.011)
Congressional District PPPE (Nbr.)
-0.024∗
-0.002
(0.014)
(0.017)
Ln(Total Payroll)
-0.073∗∗
0.014
-0.087∗∗
0.009
-0.114∗∗∗
0.008
-0.117∗∗∗
0.007
(0.032)
(0.043)
(0.034)
(0.041)
(0.038)
(0.047)
(0.038)
(0.052)
Ln(Employment)
-0.011
-0.065
0.017
-0.054
0.063
-0.058
0.069
-0.056
(0.043)
(0.069)
(0.046)
(0.068)
(0.059)
(0.076)
(0.063)
(0.083)
Observations
436
436
436
436
436
436
436
436
Adjusted R2
0.274
0.462
0.267
0.461
0.286
0.461
0.280
0.461
Industry Shares
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
State Fixed Effects
No
Yes
No
Yes
No
Yes
No
Yes
72

Table A.4: Employment Outcomes and PPPE: 2SLS Estimator
Table A.4 reports the results of 2SLS regression in ﬁrst differences comparing the Post-PPP period to the post-lockdown period prior to the passage of the PPP. The dependent variable in Panel A, ∆
Bus. Shutdown is either deﬁned as the average percentage change in business shutdowns in the Congressional District between the pre- and post-PPP period (ﬁrst three columns) or the percentage
change in the fraction of weeks in which a business decided to shutdown between the pre- and post-PPP period (last three columns). The dependent variable in Panel B, ∆Hours Worked is either
the average change in ﬁrm hours worked in the Congressional District between the pre- and post-PPP period (ﬁrst three columns) or the percentage change in weekly hours worked at the ﬁrm-level
between the pre- and post-PPP period (last three columns). Fraction receiving PPP is the percentage of businesses that received PPP loans during the ﬁrst round in the congressional district where
the ﬁrm is located. Congressional District PPPE (Nbr.) is the congressional district average of the PPPE based on the number of outstanding loans, weighted by the share of branches of each bank
in each congressional district. Standard errors are presented in parentheses, and are clustered at the level of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels,
respectively.
Panel A: Business Shutdowns
(1)
(2)
(3)
(1)
(2)
(3)
Congressional District Level
Establishment Level
First-Stage
Reduced Form
Second-Stage
First-Stage
Reduced Form
Second-Stage
Dep. Variable:
Fraction PPP
∆Bus. Shutdown
∆Bus. Shutdown
Fraction PPP
∆Bus. Shutdown
∆Bus. Shutdown
Congressional District PPPE (Nbr.)
0.378∗∗∗
-0.017
0.335∗∗∗
-0.009
(0.037)
(0.031)
(0.041)
(0.032)
Fraction receiving PPP
-0.044
-0.027
(0.083)
(0.095)
Observations
428
428
428
48174
48174
48174
Adjusted R2
0.466
0.324
-0.012
0.535
0.012
-0.000
F-Stat
103.207
67.367
State Fixed Effects
Yes
Yes
Yes
No
No
No
State×Industry Fixed Effects
No
No
No
Yes
Yes
Yes
Panel B: Decline in Hours Worked
(1)
(2)
(3)
(1)
(2)
(3)
Congressional District Level
Establishment Level
First-Stage
Reduced Form
Second-Stage
First-Stage
Reduced Form
Second-Stage
Dep. Variable:
Fraction PPP
∆Hours Worked
∆Hours Worked
Fraction PPP
∆Hours Worked
∆Hours Worked
Congressional District PPPE (Nbr.)
0.378∗∗∗
0.014
0.335∗∗∗
0.024
(0.037)
(0.017)
(0.041)
(0.019)
Fraction receiving PPP
0.038
0.070
(0.046)
(0.059)
Observations
428
428
428
48174
48174
48174
Adjusted R2
0.466
0.397
-0.011
0.535
0.024
-0.001
F-Stat
103.207
67.367
State Fixed Effects
Yes
Yes
Yes
No
No
No
State×Industry Fixed Effects
No
No
No
Yes
Yes
Yes
73

Table A.5: Unmet Demand for Loans and Exposure to PPP
Table A.5 reports the results of ordinary least squares (OLS) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst round and the percentage of
businesses reporting applying but not receiving PPP funds at the state-by-industry level. The dependent variable is the difference between the percentage of businesses that applied for PPP funds
and the percentage of businesses that received PPP funds at the state-by-industry level. State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans. All speciﬁcations
include industry×week ﬁxed-effects. Standard errors are presented in parentheses, and are clustered at the level of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10%
levels, respectively.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
% PPP Requested - % PPP Received
State PPPE (Nbr.)
-24.564∗∗∗
-78.463∗∗∗
-42.087∗∗∗
-16.478∗∗∗
-15.676∗∗∗
-8.184∗∗∗
-6.107∗∗∗
(2.525)
(7.743)
(3.943)
(2.851)
(1.975)
(1.789)
(1.333)
Observations
2229
277
386
390
390
393
393
Adjusted R2
0.757
0.562
0.421
0.247
0.170
0.074
0.096
Industry Fixed Effects
No
Yes
Yes
Yes
Yes
Yes
Yes
Industry×Week Fixed Effects
Yes
No
No
No
No
No
No
Week:
Full Sample
Apr26–May2
May3–May9
May10–May16
May17–May23
May24–May30
May31–Jun6
74

Table A.6: Unmet Demand for PPP and Missed Loan Payments
Table A.6 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst
round and the percentage of businesses reporting missing scheduled payments at the state-by-industry level. The dependent variable is the percentage of ﬁrms reporting a missed scheduled loan
payment. % PPP Requested - % PPP Received is the difference between the percentage of businesses that applied for PPP funds and the percentage of businesses that received PPP funds at a
state-by-industry level. State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans. Pre-PPP Decline Hours Worked is the average decline in hours worked in each
state between January and the last week of March. Pre-PPP State Covid-19 Cases (per capita) are per capita number of reported Covid-19 cases in the state. Pre-PPP State Covid-19 Deaths (per
capita) are the weekly per capita number of reported Covid-19 deaths in the state. Pre-PPP State Social Distancing Index is the change in average distance travelled in the state until the end of
March using individuals’ GPS signals. All speciﬁcations include industry×week ﬁxed-effects. Standard errors are presented in parentheses, and are clustered at the level of the state. ***, **, and
*, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Missing Loan Payments
% Missing Loan Payments
% PPP Request - % PPP Received
% Missing Loan Payments
% PPP Requested - % PPP Received
0.056∗∗∗
0.219∗∗∗
(0.020)
(0.043)
State PPPE (Nbr.)
-4.401∗∗∗
-23.458∗∗∗
(1.380)
(2.264)
Pre-PPP Decline Hours Worked
-3.084∗∗
-3.242∗∗∗
-6.071∗∗
-1.610∗
(1.204)
(0.977)
(2.764)
(0.922)
Pre-PPP State Covid-19 Cases (per capita)
0.838∗
0.801∗
-0.248
0.810∗∗
(0.437)
(0.409)
(0.552)
(0.374)
Pre-PPP State Covid-19 Deaths (per capita)
-26.712∗∗
-26.047∗
24.368
-27.610∗∗
(13.179)
(13.396)
(17.188)
(11.926)
Pre-PPP Social Distancing Index
-5.507∗∗
-6.787∗∗∗
-1.591
-4.365∗
(2.580)
(2.363)
(5.317)
(2.328)
Observations
2217
2487
2217
2217
Adjusted R2
0.576
0.554
0.762
-0.060
FStat
107.388
Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
75

Table A.7: Unmet Demand for PPP and Missed Non-Loan Payments
Table A.7 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst
round and the percentage of businesses reporting missing scheduled payments at the state-by-industry level. The dependent variable is the percentage of ﬁrms reporting a missed other scheduled
payment such as rent, utilities, and payroll. % PPP Requested - % PPP Received is the difference between the percentage of businesses that applied for PPP funds and the percentage of businesses
that received PPP funds at a state-by-industry level. State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans. Pre-PPP Decline Hours Worked is the average
decline in hours worked in each state between January and the last week of March. Pre-PPP State Covid-19 Cases (per capita) are per capita number of reported Covid-19 cases in the state. Pre-PPP
State Covid-19 Deaths (per capita) are the weekly per capita number of reported Covid-19 deaths in the state. Pre-PPP State Social Distancing Index is the change in average distance travelled in
the state until the end of March using individuals’ GPS signals. All speciﬁcations include industry×week ﬁxed-effects. Standard errors are presented in parentheses, and are clustered at the level
of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Missing Schd. Payments
% Missing Schd. Payments
% PPP Request - % PPP Received
% Missing Schd. Payments
% PPP Requested - % PPP Received
0.229∗∗∗
0.639∗∗∗
(0.029)
(0.068)
State PPPE (Nbr.)
-14.807∗∗∗
-23.577∗∗∗
(2.651)
(2.211)
Pre-PPP Decline Hours Worked
-11.404∗∗∗
-12.176∗∗∗
-6.119∗∗
-7.682∗∗
(3.328)
(4.094)
(2.770)
(3.127)
Pre-PPP State Covid-19 Cases (per capita)
4.261∗∗∗
4.189∗∗∗
-0.270
4.199∗∗∗
(0.666)
(0.577)
(0.556)
(0.361)
Pre-PPP State Covid-19 Deaths (per capita)
-110.471∗∗∗
-101.315∗∗∗
24.296
-112.755∗∗∗
(25.156)
(20.723)
(17.605)
(18.614)
Pre-PPP Social Distancing Index
-16.215∗∗∗
-13.084∗∗
-1.645
-13.250∗∗∗
(5.683)
(6.317)
(5.344)
(4.319)
Observations
2205
2448
2205
2205
Adjusted R2
0.718
0.698
0.762
0.086
FStat
113.656
Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
76

Table A.8: Unmet Demand for Loans and Cash-on-Hand
Table A.8 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst
round and the percentage of businesses reporting at two months or more of cash-on-hand at the state-by-industry level. The dependent variable is the fraction of businesses with cash on hand to
sustain operations for two months or more. % PPP Requested - % PPP Received is the difference between the percentage of businesses that applied for PPP funds and the percentage of businesses
that received PPP funds at a state-by-industry level. State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans. Pre-PPP Decline Hours Worked is the average
decline in hours worked in each state between January and the last week of March. Pre-PPP State Covid-19 Cases (per capita) are per capita number of reported Covid-19 cases in the state. Pre-PPP
State Covid-19 Deaths (per capita) are the weekly per capita number of reported Covid-19 deaths in the state. Pre-PPP State Social Distancing Index is the change in average distance travelled in
the state until the end of March using individuals’ GPS signals. All speciﬁcations include industry×week ﬁxed-effects. Standard errors are presented in parentheses, and are clustered at the level
of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Cash 3 months
% Cash 3 months
% PPP Request - % PPP Received
% Cash 3 months
% PPP Requested - % PPP Received
-0.172∗∗∗
-0.411∗∗
(0.047)
(0.152)
State PPPE (Nbr.)
8.920∗∗
-21.380∗∗∗
(3.768)
(2.019)
Pre-PPP Decline Hours Worked
10.973∗
15.861∗∗∗
-6.302
9.596
(5.868)
(5.378)
(4.006)
(5.830)
Pre-PPP State Covid-19 Cases (per capita)
-1.912∗∗
-2.003∗∗
0.219
-1.820∗∗
(0.743)
(0.818)
(0.381)
(0.756)
Pre-PPP State Covid-19 Deaths (per capita)
75.800∗∗
76.722∗∗
17.796
76.570∗∗
(29.392)
(31.278)
(17.170)
(31.054)
Pre-PPP Social Distancing Index
-6.618
-12.334
5.529
-6.293
(9.077)
(11.326)
(5.524)
(9.132)
Observations
903
918
903
903
Adjusted R2
0.637
0.640
0.846
-0.077
FStat
112.191
Industry×Week Fixed Effects
Yes
Yes
Yes
Yes
77

Table A.9: % Receiving PPP and % Missing Payments: Week-by-Week
Table A.9 reports the results of instrumental variables (IV) regressions split week by week examining the relation between the geographic allocation of PPP funds during the ﬁrst round and the
percentage of businesses reporting missing scheduled payments at the state-by-industry level. In the top panel, the dependent variable is the percentage of ﬁrms reporting a missed scheduled loan
payment. In the bottom panel, the dependent variable is the percentage of ﬁrms reporting a missed other scheduled payment such as rent, utilities, and payroll. % PPP Received is the percentage
of businesses reporting having received PPP funds in a state-by-industry group. State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans. All speciﬁcations
include industry×week ﬁxed-effects. Standard errors are presented in parentheses, and are clustered at the level of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10%
levels, respectively.
Panel A: Missing Loan Payments
(1)
(2)
(3)
(4)
(5)
(6)
(7)
% Missing Loan Payments
%PPP Received
-0.173∗∗∗
-0.068∗
-0.148∗∗∗
-0.278∗∗∗
-0.210∗∗∗
-0.349∗∗∗
-0.208∗∗
(0.041)
(0.034)
(0.048)
(0.097)
(0.075)
(0.122)
(0.091)
Observations
2229
277
386
390
389
393
394
Adjusted R2
-0.154
-0.021
-0.076
-0.370
-0.336
-0.665
-0.185
FStat
126.358
120.162
144.125
44.222
30.384
22.335
39.285
Industry Fixed Effects
No
Yes
Yes
Yes
Yes
Yes
Yes
Industry×Week Fixed Effects
Yes
No
No
No
No
No
No
Week:
Full Sample
Apr26–May2
May3–May9
May10–May16
May17–May23
May24–May30
May31–Jun6
Panel B: Missing Other Scheduled Payments (Rent, Utilities, Payroll)
(1)
(2)
(3)
(4)
(5)
(6)
(7)
% Missing Schd. Payments
%PPP Received
-0.555∗∗∗
-0.282∗∗∗
-0.468∗∗∗
-0.880∗∗∗
-0.651∗∗∗
-0.998∗∗∗
-0.630∗∗∗
(0.110)
(0.060)
(0.092)
(0.187)
(0.226)
(0.291)
(0.218)
Observations
2216
277
385
387
388
389
390
Adjusted R2
-0.346
0.153
-0.067
-1.089
-0.687
-1.932
-0.403
FStat
131.427
120.162
144.126
44.423
29.598
25.182
39.759
Industry Fixed Effects
No
Yes
Yes
Yes
Yes
Yes
Yes
Industry×Week Fixed Effects
Yes
No
No
No
No
No
No
Week:
Full Sample
Apr26–May2
May3–May9
May10–May16
May17–May23
May24–May30
May31–Jun6
78

Table A.10: % Receiving PPP and Cash-on-Hand: Week-by-Week
Table A.10 reports the results of instrumental variables (IV) regressions split week-by-week examining the relation between the geographic allocation of PPP funds during the ﬁrst round and the
percentage of businesses reporting at least two months or more of cash-on-hand at the state-by-industry level. The dependent variable is the difference between the percentage of businesses that
applied for PPP fundings, minus the percentage of businesses that received PPP funds. % PPP Received is the percentage of businesses reporting having received PPP funds in a state-by-industry
group. State PPPE (Nbr.) is the state average of the PPPE based on the number of outstanding loans. All speciﬁcations include industry×week ﬁxed-effects. Standard errors are presented in
parentheses, and are clustered at the level of the state. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
% Cash 3 months
%PPP Received
0.283∗∗
0.091
0.153
0.327∗
0.285
0.619∗∗
0.555
(0.121)
(0.079)
(0.105)
(0.188)
(0.310)
(0.279)
(0.348)
Observations
903
95
141
136
177
180
174
Adjusted R2
-0.247
-0.079
-0.166
-0.291
-0.280
-0.678
-0.480
FStat
144.087
50.909
84.021
49.197
16.342
32.496
Industry Fixed Effects
No
Yes
Yes
Yes
Yes
Yes
Yes
Industry×Week Fixed Effects
Yes
No
No
No
No
No
No
Week:
Full Sample
Apr26–May2
May3–May9
May10–May16
May17–May23
May24–May30
May31–Jun6
79

Table A.11: % Receiving PPP and Exposure to PPP: MSA Level
Table A.11 reports the results of ordinary least squares (OLS) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst round and the percentage of
businesses reporting applying but not receiving PPP funds at the MSA level. The dependent variable is the difference between the percentage of businesses that applied for PPP funds and the
percentage of businesses that received PPP funds at the MSA level. MSA PPPE (Nbr.) is the MSA average of the PPPE based on the number of outstanding loans. Standard errors are presented in
parentheses, and are clustered at the level of the MSA. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
% PPP Received
MSA PPPE (Nbr.)
32.714∗∗∗
68.793∗∗∗
47.107∗∗∗
19.004∗∗∗
24.207∗∗∗
16.938∗∗∗
20.232∗∗∗
(6.579)
(6.202)
(4.106)
(5.071)
(4.977)
(3.356)
(4.363)
Observations
300
50
50
50
50
50
50
Adjusted R2
0.083
0.656
0.657
0.171
0.284
0.242
0.241
Week Fixed Effects
Yes
No
No
No
No
No
No
Week:
Full Sample
Apr26–May2
May3–May9
May10–May16
May17–May23
May24–May30
May31–Jun6
80

Table A.12: % Receiving PPP and Missed Payments: MSA-Level
Table A.12 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst
round and the percentage of ﬁrms reporting missing payments at the MSA level collected from the Census Small Business Pulse Survey. In the top panel, the dependent variable is the percentage
of ﬁrms at the MSA reporting a missed scheduled loan payment. In the bottom panel, the dependent variable is the percentage of ﬁrms reporting a missed other scheduled payment such as rent,
utilities, and payroll. % PPP Received is the percentage of businesses reporting having received PPP funds in a state-by-industry group. MSA PPPE (Nbr.) is the MSA average of the PPPE based on
the number of outstanding loans. All speciﬁcations include week ﬁxed-effects. Robust standard errors are presented in parentheses. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and
10% levels, respectively.
Panel A: Missing Loan Payments
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Missing Loan Payments
% Missing Loan Payments
% PPP Received
% Missing Loan Payments
% PPP Received
-0.154∗∗∗
-0.181∗∗∗
(0.013)
(0.042)
MSA PPPE (Nbr.)
-5.902∗∗∗
32.652∗∗∗
(1.583)
(6.579)
Observations
298
298
298
298
Adjusted R2
0.357
0.039
0.083
0.347
FStat
24.635
Week Fixed Effects
Yes
Yes
Yes
Yes
Panel B: Missing Other Scheduled Payments (Rent, Utilities, Payroll)
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Missing Schd. Payments
% Missing Schd. Payments
% PPP Received
% Missing Schd. Payments
% PPP Received
-0.235∗∗∗
-0.451∗∗∗
(0.019)
(0.085)
MSA PPPE (Nbr.)
-14.753∗∗∗
32.714∗∗∗
(2.438)
(6.579)
Observations
300
300
300
300
Adjusted R2
0.320
0.099
0.083
0.046
FStat
24.728
Week Fixed Effects
Yes
Yes
Yes
Yes
81

Table A.13: % Receiving PPP and Cash-on-Hand: MSA-Level
Table A.13 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst
round and the percentage of businesses at the MSA level reporting at least two months of cash on hand to sustain operations as collected from the Census Small Business Pulse Survey. The
dependent variable is the fraction of businesses in the MSA reporting cash on hand to sustain operations for two months or more. % PPP Received is the percentage of businesses reporting having
received PPP funds at the MSA level. MSA PPPE (Nbr.) is the MSA average of the PPPE based on the number of outstanding loans. All speciﬁcations include week ﬁxed-effects. Robust standard
errors are presented in parentheses. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Cash 3 months
% Cash 3 months
% PPP Received
% Cash 3 months
% PPP Received
0.172∗∗∗
-0.172∗
(0.018)
(0.097)
MSA PPPE (Nbr.)
-5.628∗∗
32.714∗∗∗
(2.468)
(6.579)
Observations
300
300
300
300
Adjusted R2
0.194
0.014
0.083
-0.595
FStat
24.728
Week Fixed Effects
Yes
Yes
Yes
Yes
82

Table A.14: Unmet Demand and Exposure to PPP: MSA Level
Table A.14 reports the results of ordinary least squares (OLS) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst round and the percentage of
businesses reporting applying but not receiving PPP funds at the MSA level. The dependent variable is the difference between the percentage of businesses that applied for PPP funds and the
percentage of businesses that received PPP funds at the MSA level. MSA PPPE (Nbr.) is the MSA average of the PPPE based on the number of outstanding loans. Robust standard errors are
presented in parentheses. ***, **, and *, represent statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
% PPP Requested - % PPP Received
MSA PPPE (Nbr.)
-22.987∗∗∗
-60.564∗∗∗
-36.165∗∗∗
-13.531∗∗∗
-11.946∗∗∗
-9.384∗∗∗
-6.331∗∗∗
(6.526)
(7.053)
(3.150)
(2.527)
(2.570)
(1.500)
(1.845)
Observations
300
50
50
50
50
50
50
Adjusted R2
0.043
0.585
0.643
0.360
0.270
0.298
0.121
Week Fixed Effects
Yes
No
No
No
No
No
No
Week:
Full Sample
Apr26–May2
May3–May9
May10–May16
May17–May23
May24–May30
May31–Jun6
83

Table A.15: Unmet Demand and Missed Payments:MSA Level
Table A.15 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst
round and the percentage of businesses reporting missing scheduled payments at the MSA level. In the top panel, the dependent variable is the percentage of ﬁrms reporting a missed scheduled
loan payment. In the bottom panel, the dependent variable is the percentage of ﬁrms reporting a missed other scheduled payment such as rent, utilities, and payroll. % PPP Requested - % PPP
Received is the difference between the percentage of businesses that applied for PPP funds and the percentage of businesses that received PPP funds at the MSA level. MSA PPPE (Nbr.) is the
MSA average of the PPPE based on the number of outstanding loans. All speciﬁcations include week ﬁxed-effects. Robust standard errors are presented in parentheses. ***, **, and *, represent
statistical signiﬁcance at 1%, 5%, and 10% levels, respectively.
Panel A: Missing Loan Payments
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Missing Loan Payments
% Missing Loan Payments
% PPP Request - % PPP Received
% Missing Loan Payments
% PPP Requested - % PPP Received
0.166∗∗∗
0.258∗∗∗
(0.013)
(0.067)
MSA PPPE (Nbr.)
-5.902∗∗∗
-22.911∗∗∗
(1.583)
(6.526)
Observations
298
298
298
298
Adjusted R2
0.380
0.039
0.043
0.265
FStat
12.325
Week Fixed Effects
Yes
Yes
Yes
Yes
Panel B: Missing Other Scheduled Payments (Rent, Utilities, Payroll)
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Missing Schd. Payments
% Missing Schd. Payments
% PPP Request - % PPP Received
% Missing Schd. Payments
% PPP Requested - % PPP Received
0.258∗∗∗
0.642∗∗∗
(0.020)
(0.155)
MSA PPPE (Nbr.)
-14.753∗∗∗
-22.987∗∗∗
(2.438)
(6.526)
Observations
300
300
300
300
Adjusted R2
0.355
0.099
0.043
-0.434
FStat
12.406
Week Fixed Effects
Yes
Yes
Yes
Yes
84

Table A.16: Unmet Demand and Cash-on-Hand: MSA Level
Table A.16 reports the results of ordinary least squares (OLS) and instrumental variables (IV) regressions examining the relation between the geographic allocation of PPP funds during the ﬁrst
round and the percentage of businesses reporting at least two months or more of cash-on-hand at the MSA level. The dependent variable is the fraction of businesses with cash on hand to sustain
operations for two months or more. % PPP Requested - % PPP Received is the difference between the percentage of businesses that applied for PPP funds and the percentage of businesses that
received PPP funds at the MSA level. All speciﬁcations include week ﬁxed-effects. Robust standard errors are presented in parentheses. ***, **, and *, represent statistical signiﬁcance at 1%, 5%,
and 10% levels, respectively.
(1)
(2)
(3)
(4)
OLS
Reduced Form
IV
First-Stage
Second-Stage
Dep. Variable:
% Cash 3 months
% Cash 3 months
% PPP Request - % PPP Received
% Cash 3 months
% PPP Requested - % PPP Received
-0.203∗∗∗
0.245
(0.019)
(0.156)
MSA PPPE (Nbr.)
-5.628∗∗
-22.987∗∗∗
(2.468)
(6.526)
Observations
300
300
300
300
Adjusted R2
0.247
0.014
0.043
-0.972
FStat
12.406
Week Fixed Effects
Yes
Yes
Yes
Yes
85
File and source

File: Granja-Makridis-Yannelis-Zwick_NBER-w27095_did-ppp-hit-the-target.pdf
Size: 911,020 bytes
SHA-256: 17a153121eaa58337ee318dd42940b1e6ce7c0d0ae819e134dff54e113228164
Our copy: Granja-Makridis-Yannelis-Zwick_NBER-w27095_did-ppp-hit-the-target.pdf
Original: www.nber.org
Granja Makridis Yannelis Zwick Nber W27095 Did Ppp Hit the Target

Full text

File and source

Related records
