Did the Paycheck Protection Program Support Small Business Activity? Research Brief
1
Research Brief
Did the Paycheck Protection Program Support
Small Business Activity?
Chris Wheat and Chi Mac
December 2021
The COVID-19 pandemic has been a widespread and prolonged
disruption to life in the U.S. since a national emergency was
declared on March 13, 2020. Small businesses saw substantial
revenue declines in the initial months (Farrell, Wheat, and Mac
2020a, 2020b). Expenses also declined commensurately,
reflecting lower revenues as well as efforts to preserve liquidity
(Farrell et al. 2020). Many small businesses have adapted to current
economic conditions, and while some have fully recovered, others
have not. Figure 1 shows that among small businesses that were
operating before the pandemic, median expenses in August 2021
were 10 percent below their January 2020 levels. However, median
balances remain elevated relative to their January 2020 levels.
Several relief and stimulus programs have contributed to deposit
account balances during the pandemic, including the Paycheck
Protection Program (PPP), Economic Injury Disaster Loan (EIDL)
advances, Economic Impact Payments (EIP), and the expanded
advance Child Tax Credit (CTC).
1
The signature program for small businesses was the PPP, which
distributed nearly $800 billion in 2020 and 2021 through 11.8
million loans to small businesses nationwide.
2
Box 1 provides
additional summary details about the program. This research will
help policymakers evaluate the large fiscal expenditure required
to support the program and provide insight to the design
considerations of future relief programs. In particular, our research
offers insights about two features of the PPP: the large number of
relatively small loans and the duration of what was initially
short-term relief. Our findings shed light on the effects of smaller
loans, as nearly 69 percent of PPP loans in 2020 and 87
percent of those in 2021 were $50,000 or less (Small Business
Administration 2020a, 2021). Other studies have focused on loans
greater than $150,000. In addition to the PPP, state and local
programs often offered smaller amounts or limited eligibility to
smaller small businesses, and our research could have implications
for their programs as well. For example, the New York State
Pandemic Small Business Recovery Grant Program offered grants
ranging from $5,000 to $50,000 to small businesses with 100 or
fewer employees.
3
Ohio’s Small Business Relief Grant provided
$10,000 to small businesses with no more than 25 employees.
4
Our results inform not only the magnitude but also the
duration of any PPP effects. The program was initially designed
with an 8-week covered period under the expectation that firms
would spend their proceeds during this time. However, as the
pandemic evolved, this covered period was revised to 24 weeks,
although the maximum loan amounts remained unchanged. Our
analysis can provide insight into the duration of effects as well as
whether the revised covered period provided the flexibility small
businesses needed.
In this brief, we analyzed the magnitude and duration of any
effects the PPP had on business operating activity, as measured by
expenses and in comparison to a control group. While expenses
materially increased among small businesses that received PPP in
the week loan proceeds were received (Wheat and Mac 2021), there
is less evidence about whether expenses increased more among
PPP recipients than non-recipients.
Key Findings
• Finding 1: Upon PPP receipt, small business expenses increased by over 40 percent relative to a comparison group, with
significant but declining effects over four months.
• Finding 2: The impact of PPP loans on expenses was largest in April and May 2020, when small business expenses were
particularly depressed.
• Finding 3: The smallest firms experienced larger spending effects upon loan receipt, perhaps because they were more liquidity
constrained than larger firms.
• Finding 4: Restaurants may have used PPP loan proceeds to frontload expenses.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
2
Figure 1: Small business revenues and expenses have not rebounded to January 2020 levels, although balances are higher
Source: JPMorgan Chase Institute
Nov
2019
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
Jan
2021
Mar
2021
May
2021
Jul
2021
−50%
−25%
0%
25%
50%
Median changes in firm balances, revenues, and expenses relative to January 2020
Percent change
Note: Sample includes small businesses operating January 2019 to February 2020.
Balances
Revenues
Expenses
View text version
Based on de-identified administrative data from bank
accounts, our research provides new evidence of the effect
of PPP loans—particularly smaller loans—on small business
activity. Other research on PPP has focused on employment
outcomes, often among firms receiving loans of more than
$150,000 compared to firms ineligible for PPP. For example,
there were small positive employment effects and reduced
business closure among employer firms that applied for PPP
loans greater than $150,000, relative to a control group of
firms with 500-1,000 employees (Hubbard and Strain 2020).
Firms receiving PPP loans increased employment by 7.5
percent relative to firms ineligible for PPP (Autor et al.
2020). Employment rates at firms with fewer than 500
employees, which would have been eligible for PPP,
increased by 2 percentage points compared to firms with
more than 500 employees (Chetty et al. 2020).
Our research differs in several aspects: First, our sample
consists largely of smaller loans to nonemployer and small
employer firms. Second, our control group also consists of
small businesses, so the comparison is between firms with
and without observed PPP funds in the given period, not
between the small businesses that were eligible and the
large businesses that were ineligible for the program. Third,
we focus on business expenses as the outcome variable, as
a proxy for business activity, which is particularly relevant
because we do not observe electronic payroll payments for
most of our sample.
5
Fourth, our period of analysis extends
through the end of 2020. As the pandemic extends into its
second year, policymakers may be interested in the duration
of any PPP effects.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
3
Box 1: Background on the PPP
The PPP was a complex program with guidelines that changed over time. The following provides a summary
of key dates and program features. It is not an exhaustive reference of relevant legislation or rules.
Key dates
• March 27, 2020: PPP established by CARES
Act with initial funding of $349 billion
a
• April 27, 2020: PPP loan applications resumed
b
after an additional $310 billion authorized
c
• June 5, 2020: Paycheck Protection Program
Flexibility Act
d
revised full loan forgiveness
terms to require at least 60 percent of
proceeds to be spent on payroll during
24 weeks after loan disbursement
• August 8, 2020: Application period closed
e
• December 27, 2020: Consolidated
Appropriations Act authorized $284 billion
f,g
• January 11, 2021: First draw loans reopened,
and second draw loans reopened two days later
h
• March 4, 2021: Nonemployers may use
gross income to calculate loan amount
i
• May 31, 2021: Applications closed; banks have
additional 30 days to process applications
j
Summary guidelines
• Eligibility: Entities with 500 or fewer
employees, including self-employed
individuals, sole proprietorships, and
independent contractors;
k
for second draw
loans in 2021, no more than 300 employees
and can demonstrate at least a 25 percent
reduction in gross receipts between
comparable quarters in 2019 and 2020
l
• Loan forgiveness: Initially, at least 75 percent
of proceeds must be spent on payroll costs
during the 8-week covered period following
loan disbursement; revised June 5, 2020 to
60 percent and 24 weeks, respectively
m
• Maximum loan amount: 2.5x average
monthly 2019 payroll costs or 2.5x net profit
n
for nonemployers; after March 4, 2021,
nonemployers may use gross income
o
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
4
We analyzed a de-identified sample of small businesses with
Chase Business Banking accounts that received exactly one
PPP loan inflow between 2020 and 2021. Within this sample,
we identified a treated group consisting of firms that received
PPP loans between April and August of 2020, which we
divided into cohorts based on the month in which they
received PPP loan proceeds.
6
May was the largest cohort,
comprising 77 percent of the 2020 loans in this analysis. We
compared each monthly loan cohort to a control group of
small businesses that did not receive PPP loans in 2020 but
did receive them in 2021, suggesting that this group was likely
eligible in 2020 as well and therefore comparable to those
with loans in 2020.
We used firm expenses as a measure of activity because it
may be less dependent upon customer demand or pandemic-
related operating restrictions. Earlier analysis showed that
expenses were more responsive to PPP loan inflows than
revenues (Wheat and Mac 2021). Figure 2 shows the median
change in expenses for the May cohort and the control
group, relative to their respective January 2020 values,
during the months between November 2019 and December
2020. Some overall trends are evident, such as decreasing
expenses in April followed by increasing expenses. While the
May cohort experienced a sharp increase in expenses when
PPP funds were received, the entirety of that increase cannot
be attributed to PPP, as the control group’s expenses also
increased in May but to a lesser extent. In subsequent months,
the difference between the groups narrowed. Our model was
estimated over this period
7
to include several months prior
to the pandemic and several after receipt of PPP loans. This
brief focuses on the estimated effects across these 14 months.
Methodology
Figure 2: Firms receiving PPP increased expenses more sharply than the control group
Nov
2019
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
−30%
−20%
−10%
0%
10%
Median change in small business expenses relative to January 2020
Percent change
Source: JPMorgan Chase Institute
Note: Sample includes small businesses that received PPP in May 2020 or January to June 2021 (control).
May cohort
Control
Firms receiving PPP in May 2020 sharply
increased spending in month of receipt
View text version
To estimate the effects of PPP for each loan cohort relative
to the control group that is also responding to the pandemic,
we used a difference-in-difference event study. In this
framework, we look not at the difference between the
average expenses of the treated and control groups, but
rather the difference in how each group’s expenses changed
after the event in question. That is, the first difference was
the change each group experienced after PPP, some of
which may have been due to the pandemic’s evolution.
The difference between the treated and control groups’
differences therefore provided an estimate of how receiving
PPP loans changed the trend in business activity.
All of our estimates of PPP’s impact on expenses used the
difference-in-difference framework, which also implies that
any estimates prior to the treatment—receipt of PPP loans—
should be near zero despite potential level differences
between the two groups. The Methodological Appendix
discusses the sample and model in greater detail.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
5
Finding One: Upon PPP receipt, small business expenses
increased by over 40 percent relative to a comparison
group, with significant but declining effects over four
months. Across the loan cohorts, small businesses with
PPP loans increased their spending significantly for several
months after loan proceeds were received, relative to the
control group, although the effect diminished over time.
We summarized the overall effects of PPP over time by
calculating the weighted average effect across cohorts
for each month relative to the month of loan receipt. We
weighted each cohort’s estimate by its sample size, using
a method proposed by Callaway and Sant’Anna (2020).
As the May cohort was our largest, those estimates drive
the overall shape of the effects.
Figure 3 shows that small businesses with PPP loans
increased their expenses by more than those in the control
group for several months after receiving the loans. During
the initial month when small businesses received loan
proceeds, expenses changed by over 40 percent more than
changes in the control group’s expenses. Firms quickly
deployed loan inflows to pay expenses, suggesting that
cash liquidity was an imminent concern that was alleviated
by PPP. This effect declined over time: by the fifth month,
the effect was near zero. While the decline over time may
not be surprising, our estimates show that the effect on
firm expenses persists for approximately four months after
loan receipt. Notably, this duration was less than the
24-week covered period.
The PPP was designed to offer short-term cash liquidity to
small businesses, although the guidelines for using the
funds were later revised from 8 to 24 weeks.
8
Our results
suggest that there was a relatively large initial effect on
expenses, followed by an effort to stretch PPP funds over
several months, but any effects had diminished before
24 weeks had elapsed. Firms may have timed their
expenditures to occur within the guidelines, but the
declining effects suggest that the 24-week guideline
provided adequate flexibility.
Figure 3: Upon receipt of PPP, small business expenses increased by 42 percent relative to the control group, with declining
effects over time
Source: JPMorgan Chase Institute
42%
25%
11%
6%
3%
0 1 2 3 4
Effect of PPP on small business expenses among 2020 recipients
Number of months since PPP receipt
Note: Estimates are for 2020 PPP recipients compared to a control group.
View text version
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
6
Finding Two: The impact of PPP loans on expenses was
largest in April and May 2020, when small business
expenses were particularly depressed. The PPP was
established relatively quickly and emphasized the rapid
disbursement of funds under the assumption that small
businesses needed immediate relief. Many small businesses
applied soon after applications opened, but others did not.
We analyzed differences between the cohorts to understand
how contemporaneous circumstances can affect the
magnitude or duration of the impact on expenses. The
differences between cohorts also provides insight into
potential anticipatory effects, which can have implications
on how future relief programs are structured.
During April and May of 2020, the effect of receiving loans
on expenses—relative to the control group—was larger than
the effect for firms receiving loans in subsequent months.
The April and May cohorts comprised 84 percent of 2020
loans in our sample, and the initial effects observed in
these cohorts are relatively large. The upper panels of
Figure 4 show that changes to these cohorts’ expenses
in May 2020 were about 50 percent higher than the
changes in the expenses of the control group. In
hindsight, April and May 2020 represented two months
with particularly low expenses in the downturn related
to the pandemic, as shown in Figure 1 and Farrell et al.
(2020). This suggests the cash infusion was particularly
effective in supporting expenses during a period in which
firms would have otherwise restrained their spending.
However, there may be some selection effects, as we do not
know why some firms applied later. Early in the program,
not all details, especially about loan forgiveness, were
clear.
9
Firms applying for loans earlier may have been more
severely affected by the pandemic, while those waiting
until later could have had less immediate need or other
funding options. Alternatively, firms applying earlier may
have been more confident that they would be able to repay
a loan in the event they did not qualify for forgiveness.
We do not observe these or other potential contributing
factors. The Appendix provides details about the cohorts.
In contrast, the lower panels of Figure 4 show that during
the month of loan receipt, changes to the June and July
cohorts’ expenses were 15 and 22 percent higher,
respectively, than the changes to the control group’s
expenses. In the summer of 2020, small businesses saw
some recovery as capacity constraints eased or as they
found new ways to serve their customers. The effect of
receiving PPP funds could have been lower in these months
because they were measured relative to small businesses
whose circumstances were also improving despite not
receiving loans.
Figure 4: PPP loans had larger effects on small business expenses in April and May 2020
Source: JPMorgan Chase Institute
Effect of PPP on small business expenses by 2020 cohort
Note: Sample for each cohort includes firms that received PPP in the given month and the control group. Error bars indicate 95% confidence interval.
Month of PPP receipt
Nov
2019
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
−10%
0%
10%
20%
30%
40%
50%
60%
Nov
2019
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
−10%
0%
10%
20%
30%
40%
50%
60%
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
−10%
0%
10%
20%
30%
40%
50%
60%
Nov
2019
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
−10%
0%
10%
20%
30%
40%
50%
60%
Effect of PPPEffect of PPP
April
May
June
July
Four months after PPP receipt
Nov
2019
View text version
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
7
For each cohort, we saw no statistically significant
differences between the treated and control groups in
the pre-treatment period. In each panel of Figure 4, the
estimated differences in the months before loan receipt is
near zero, with one exception, which we discuss below. This
is consistent with observing parallel trends between the
treated and control groups prior to receipt of the loan.
That is, their expenses were exhibiting similar trends before
PPP loan proceeds were received.
The May cohort exhibited increasing expenses relative to
the control group in April, the month before loan receipt.
This raises questions about the existence of potential
anticipatory effects—effects that can have important
implications for policymakers designing future programs.
There are two potential channels for anticipatory effects.
First, upon the establishment of the PPP, firms that
expected to be eligible for loans may have started to
spend even before applying. If that were the case, then a
program announcement would begin to impact behavior,
implying that timely announcements are critical to the
policy response. We did not find evidence consistent with
this type of anticipatory effect.
A second, narrower, channel is possible in the period
between loan approval and the receipt of loan proceeds.
Firms may have been comfortable spending knowing
their loans were approved but before the funds were
posted to their accounts. Our analysis suggests the
anticipatory effect seen in the May cohort was consistent
with the latter, due to the timing of the reopening of the
program, the loan application and approval process, and
the receipt of funds.
The PPP reopened on Monday, April 27, 2020, and
applications were processed and approved in the subsequent
days. The loan proceeds from this set of applications were
often posted to accounts in early May, but firms would
have learned of their loan approvals prior to receiving the
funds and some may have begun spending based on
that knowledge. To investigate this, we estimated our
model specification using four-week periods instead of
calendar months. For example, one four-week period
was Sunday, April 26 through Saturday, May 23, 2020
so the entire period encompassed both the reopening
of PPP and the funding of those applicants. There was
no statistically significant anticipatory effect in the
prior period, supporting the interpretation that small
businesses started spending following approval, not
simply because they were eligible for the program.
We also did not find anticipatory effects in subsequent
cohorts who were eligible for PPP, but applied and received
loans later. Similar to the May cohort, firms in other cohorts
could have begun spending after receiving loan approvals
and before receiving loan funds. However, in other months,
loans were not clustered at the beginning of the month,
so this pattern was less apparent.
The PPP required full disbursement of the loan within 10
calendar days of loan approval,
10
limiting this type of
anticipatory effect. This channel may be relevant for
programs with longer funding timelines. For example,
firms that have been approved for loans but are liquidity
constrained may nevertheless need to wait until they receive
the funds before they can start spending. Prior to the
pandemic, the typical small business had about two weeks
of cash liquidity, and long waits for funds may strain the
finances of many small businesses (JPMorgan Chase
Institute 2020).
The PPP provided cash liquidity to small businesses during
the pandemic and emphasized rapid disbursement of funds.
Our results suggest the cash infusion was particularly
effective in supporting expenses during April-May 2020,
when firms experienced severe disruptions to other inflows.
Our results also imply that some firms changed their
spending behavior before they received the funds, an effect
we will discuss further in the next finding.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
8
Finding Three: The smallest firms experienced larger
spending effects upon loan receipt, perhaps because
they were more liquidity constrained than larger firms.
Only small businesses were eligible for the PPP, but the
small business sector spans a wide range of firm sizes.
Among both PPP and non-PPP recipients, the vast majority
of small businesses—81 percent—were nonemployers and
another 17 percent were employers with fewer than 20
employees (Small Business Administration 2020b). The
effects of receiving loans may vary based on firm size. For
example, Hubbard and Strain (2020) found some evidence
that PPP was more effective among smaller firms in their
sample.
11
We investigated the differential effects of PPP
on business expenses by firm size and found that smaller
firms experienced larger effects in the month loan proceeds
were received. However, smaller firms exhibited lower
anticipatory effects in April, consistent with being more
liquidity constrained than their larger counterparts.
We used our largest cohort to analyze differences by firm
size. We separated the firms in our May sample, including
both the May loan recipients and the control group, into
quartiles based on their annual expenses in 2019
12
and
estimated the same model. The first quartile included firms
with annual expenses up to $191,000, while the fourth
quartile consisted of firms with annual expenses over
$991,000.
Figure 5 shows that smaller firms experienced larger
effects on expenses in the month loan proceeds were
received. For firms in the first size quartile, the increase
in May expenses was 61 percent higher than the change
experienced by the control group. In comparison, the
effect was 38 percent for the firms in the fourth quartile.
In addition, the positive effect on expenses lasted fewer
months for smaller firms: three months for the first quartile
compared to four months for the fourth quartile firms.
Figure 5: The smallest firms had larger effects on expenses in the month of PPP loan receipt and smaller anticipatory effects than
the largest firms
Source: JPMorgan Chase Institute
Effect of PPP on small business expenses by firm size within May cohort
Note: Firms in quartile 1 had 2019 expenses below $191,000. Firms in quartile 4 had expenses greater than $991,000. Treated sample includes firms that received PPP
in May 2020, subsampled by 2019 expenses. Error bars indicate 95% confidence interval.
Effect of PPP
Quartile 1 Quartile 4
Nov
2019
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
−10%
0%
10%
20%
30%
40%
50%
60%
70%
80%
Nov
2019
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
−10%
0%
10%
20%
30%
40%
50%
60%
70%
80%
Larger anticipatory
effect in April
Month of PPP receipt
Four months after PPP receipt
Larger effect among
small firms in month
of PPP receipt
View text version
The anticipatory effect in April was smaller for firms in
the first quartile, about 8 percent, compared to the 19
percent estimated for firms in the fourth quartile in the
same month. One explanation for this difference could be
that the smallest firms were less able to start spending
prior to receiving the loan proceeds, perhaps because
they did not have the cash liquidity from other sources
to draw upon. Larger small businesses could have been
more willing or able to start spending prior to actually
receiving loan proceeds. This suggests that cash liquidity
was more critical for the smallest small businesses.
Larger effects on expenses after the receipt of loan
proceeds and the smaller anticipatory effect both suggest
that cash liquidity may be more critical to the spending
decisions of smaller small businesses. The PPP required
loan disbursements within 10 calendar days of approval,
limiting this type of anticipatory effect. However, it could
be a consideration in the design of future programs.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
9
Finding Four: Restaurants may have used PPP loan
proceeds to frontload expenses. Restaurants have been
particularly hard hit during the pandemic. While many
restaurant owners adapted their business models to
changing economic conditions related to the COVID-19
pandemic, these changes may not have been sufficient to
compensate for evolving consumer demand. Their recovery
has lagged behind small businesses in other industries.
13
The challenges faced by small restaurants provided an
opportunity to better understand how short-term relief may
have been used by firms facing longer term disruption more
broadly. We used the May cohort, which offers the largest
sample size, and restricted it to restaurants in this model.
Similar to what we saw across our sample, small restaurants
also increased their expenses relative to the control group
during the initial months after receiving their loans, as
shown in Figure 6. However, the significant positive effect
did not last long—just two months after receiving the
loan proceeds—although there was an anticipatory effect
in April. However, beginning in September 2020, the change
in expenses for restaurants with PPP loans was significantly
lower than the control group.
Figure 6: Restaurants initially increased expenses upon PPP loan receipt, but changes to spending later in 2020 were lower than
the control group
Source: JPMorgan Chase Institute
Effect of PPP on expenses of restaurants in the May cohort
Note: Sample includes restaurants that received PPP in May 2020 compared to restaurants in the control group. Error bars indicate 95% confidence interval.
Effect of PPP
Nov
2019
Jan
2020
Mar
2020
May
2020
Jul
2020
Sep
2020
Nov
2020
−50%
−25%
0%
25%
50%
75%
100%
View text version
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
10
Our difference-in-difference framework does not
estimate differences in expense levels between groups
but rather differences in how each group changes.
Over the entire period, restaurants with PPP had higher
average expenses than the control group. However, our
results suggest they frontloaded those expenses in the
initial months after receiving their loan proceeds.
Figure 7 shows the ratio of average monthly expenses to
the average over the April to December 2020 period for
each group. If monthly expenses were perfectly smooth
over time, then this ratio would equal 1 in each month. That
is, each month’s average expense is the same as in other
months. For restaurants with PPP loans in May, monthly
expenses were higher early in the period and shortly after
receiving loan proceeds. Later in 2020, expenses were lower
than their period average. Among restaurants in the control
group, we saw a different pattern. Average expenses in May
were lower than their period average and more gradually
increased. In autumn of 2020, average expenses for the
control group were increasing when those of the treated
group were decreasing, resulting in the negative difference-
in-difference estimates. This suggests that restaurants
receiving PPP funds in May were able to frontload more
expenses in the initial weeks, perhaps investing in outdoor
seating or an online presence. In contrast, firms without PPP
loans could have delayed or spread out similar expenses.
View text version
Figure 7: Restaurants with PPP spent more soon after receiving funds, while the control group ramped up spending later in 2020
Source: JPMorgan Chase Institute
Ratio of monthly restaurant expenses to April through December average
Note: Sample includes restaurants that received PPP in May 2020 and restaurants in the control group.
Ratio
Apr
2020
May
2020
Jun
2020
Jul
2020
Aug
2020
Sep
2020
Oct
2020
Nov
2020
Dec
2020
0.6
0.8
1.0
1.2
1.4
Control group:
Higher than average
expenses later in period
PPP recipients:
Higher than average
expenses earlier in period
Small businesses in other industries may have also used
their PPP loan proceeds to frontload expenses. However,
that pattern may be obscured in industries that have
experienced more robust recoveries, such as health care
and professional services. It is perhaps not a coincidence
that we observe this pattern in restaurants and, to a lesser
extent, personal services, two of the hardest hit industries
during the pandemic.
Our model estimates illustrate that the experiences of small
businesses during the pandemic—including experiences
with relief programs such as PPP—varied by industry.
Their experiences suggest the need for industry-specific
considerations alongside broader policy interventions.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
11
The PPP was an unprecedented program designed to
support small businesses through the pandemic. This brief
provides insight into the magnitude and duration of the
effect of PPP loans on small business expenses. With this
context in mind, we offer the following implications for
leaders and decision makers seeking to support small
businesses during periods of severe economic disruptions:
Small businesses need flexibility in how and when to
use relief funds. Policymakers want to ensure that relief
funds are used promptly and effectively. However, the wide
range of small businesses implies that they may face a wide
range of challenges resulting from the same disruption.
Policymakers did recognize this during the pandemic and
revised some PPP rules, such as increasing the time for
using the loan proceeds from 8 to 24 weeks and decreasing
the share of funds to be used on payroll expenses from 75
to 60 percent. As the pandemic evolved, some sectors may
have needed different strategies for how and when to use
their loan proceeds. More generally, relief programs should
consider the need for flexibility along with the need for
structure and guidelines.
Timeliness is an important factor in effective relief. Our
research demonstrates that PPP’s largest impact on business
expenses was in April and May 2020, when business
spending was particularly depressed. The CARES Act was
passed relatively quickly, establishing several different relief
programs, including the PPP. The timeliness of this action
may have been critical given the sudden disruption caused
by the pandemic. Prior to the pandemic, typical small
businesses had about two weeks of cash liquidity, with
Black- and Latinx-owned businesses having lower cash
buffers than White-owned ones (JPMorgan Chase Institute
2020; Farrell, Wheat, and Mac 2020c). Nevertheless, in
designing relief programs, policymakers must balance
the need to disburse relief funds quickly with competing
goals, such as considering those who have been most
severely affected, collecting data and documentation from
applicants, or mitigating the risk of fraud.
Targeted policies could be appropriate, depending on
the circumstances. A widespread emergency such as the
pandemic required a relief program that was just as broad.
However, the related economic disruption was not
experienced evenly across the small business sector, as
restaurants and personal services have been more severely
affected than other industries. During the pandemic, the
Restaurant Revitalization Fund
14
and Shuttered Venue
Operator Grants
15
offered additional assistance, but they
were not available until 2021. Depending on the
circumstances of the economic disruption, policymakers
should consider a possible role for more targeted
interventions. Examples include industry-specific programs,
perhaps as complements to broader programs, or using
maximum loan amounts to focus relief efforts.
Conclusions and Implications
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
12
To analyze the effects of PPP on small business expenses,
we created a sample of de-identified small businesses that
were similar but for their observed receipt of PPP loans.
Firms included in our sample, either in a treated or control
group, have Chase Business Banking accounts and exactly
one observed PPP loan. The PPP lender could be Chase,
as it was in 78 percent of our sample, or another lender.
16
To reduce confounding factors, we excluded any firms
that received either more than one PPP loan or an EIDL
advance. It is possible that firms in our sample had access
to other sources of funding during the period of analysis.
Our estimates used expense data derived from the de-
identified deposit account transactions from November
2019 through December 2020. To ensure that we were
not measuring expenses during firms’ startup phases,
we required that the accounts were open for at least a
year prior to November 2019 with activity during each
month of the pre-pandemic period. Accounts also must
have stayed open through the end of December 2020,
but accounts did not need to have activity in every
month of 2020.
17
Satisfying the criteria for inclusion in
our sample necessarily means that we neither included
every PPP loan originated by Chase nor did we include
only those originated by Chase.
18
A careful reader of our
reports will notice that the typical firm in this sample
was somewhat larger than those in our other reports.
Using this sample, we separated the 2020 PPP loan
recipients into cohorts based on the month the loan
proceeds were posted into the deposit accounts. Each
cohort represents a treated group. The control group
was the same for each treated group: firms with their
first and only PPP loan in 2021. Therefore, they were
not-yet-treated during the estimation period which ends
in 2020. However, the fact they received a PPP loan in
2021 implies they likely would have qualified in 2020.
Firms that did not apply by August 2020 would not have
known of future opportunities to apply until legislation
reauthorizing the PPP was signed in late December 2020.
There may have been differences between firms receiving
PPP loans earlier, compared to those receiving loans later.
However, for our research design, differences between
treated and control groups are acceptable as long as they
followed parallel trends prior to the treatment. Firms
applying earlier could have had greater need; firms applying
later could have been waiting for clearer guidelines. Table
A1 shows the median loan sizes across cohorts. The typical
loan amount in our May cohort was $27,000, compared to
$11,000 in the July cohort. The median loan amount among
the 2021 loans in our sample was $17,000. The publicly
available data from the SBA show similar trends. Loans
approved in May 2020 had a median size of $19,000, while
the median among those approved in July was $13,000.
Limited to the loans under $150,000, which is more
comparable to our sample, a similar pattern of larger loans
earlier emerges. However, some nonemployer firms would
have qualified for larger loans in 2021 than in 2020. A rule
change allowed them to calculate their maximum loan
amount based on gross income instead of net income.
19
We did not include firms that never received a PPP loan
in the control group because while some of these firms
may have served as appropriate controls, others may have
been ineligible for the PPP. Survey data indicated that
nonemployer firms were less likely to apply for PPP than
employers, with the most common reason cited as the
expectation that they would not qualify either for the loan
or for loan forgiveness (Federal Reserve 2021a, 2021b).
Other reasons included confusion about the program, not
needing funding, or seeking other funding. The Federal
Reserve also noted that 30 percent of nonemployers
collected unemployment insurance benefits (2021b),
20
which could have been an alternate avenue for relief.
Methodological Appendix
Sample
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
13
Table A1: Median loan amounts and share of loans, by sample and cohort
Table A2: Industry composition, by sample and cohort
Table A2 shows that the industry distribution in our sample
was similar to the distribution across those industries for
all loans reported by the SBA. However, our sample was
more heavily weighted in industries common in greater
metropolitan areas. For example, our sample included a
larger share of firms in health care and professional services.
Our sample did not include industries for which we did not
have a representative sample, including transportation and
warehousing, as well as agriculture, forestry, fishing, and
hunting.
JPMCI sample SBA data
Firms Median loan amount Share of loans
2020 PPP loans Median loan amount Count Share $150k and under All loans $150k and under All loans
April $59,000 3,508 7% $27,000 $38,000 52% 56%
May $27,000 36,303 77% $18,000
$19,000 32% 29%
June $14,000 4,106 9% $13,000 $13,000 9% 8%
July $11,000 2,641 6% $13,000 $13,000
5% 4%
August $11,000 728 2% $13,000 $14,000 2% 2%
Total 47,286 100% 100% 100%
2021 PPP loans $17,000 6,657 $19,000 $20,000
JPMCI sample SBA data
2020 loans
April May June July August 2021 loans Overall
$150k and
under
All loans
Other professional services 20% 17% 13% 14% 14% 17% 17% 14% 14%
Health care services 16% 16% 15% 16% 14% 12% 15% 11% 11%
Retail 11% 13% 12% 12% 13% 12% 13% 13% 13%
Construction 12% 13% 14% 14% 15% 16% 13% 14% 14%
Restaurants 9% 10% 10% 8% 9% 10% 10% 9% 10%
Repair & maintenance 7% 8% 8% 9% 8% 9% 8% 9% 9%
Personal services 5% 7% 13% 13% 12% 11% 8% 16% 15%
Wholesalers 7% 7% 5% 5% 5% 5% 6% 5% 5%
Real estate 5% 5% 5% 5% 5% 3% 5% 7% 6%
High-tech services 5% 4% 3% 2% 3% 3% 3% 2% 2%
Metal & machinery 2% 1% 1% 1% <1% 1% 1% 1% 2%
High-tech manufacturing <1% <1% <1% <1% <1% <1% <1% 0% 0%
100%
100% 100% 100% 100% 100% 100% 100% 100%
Selected industries as share of total 62% 62%
JPMCI sample includes firms with one PPP loan from Chase or another lender and meet our criteria for deposit account activity. Loan cohorts are based on the date loan
proceeds were posted to deposit accounts. Firms with loans approved in the last week of April 2020 received their funds in early May.
SBA tabulation is based on available data as of June 30, 2021. Loan cohorts are based on loan approval date.
JPMCI sample includes firms with one PPP loan from Chase or another lender and meet our criteria for deposit account activity.
SBA tabulation is based on available data as of June 30, 2021.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
14
Model
We used a difference-in-difference event study to estimate the effect of receiving PPP loans on expenses over time.
We were particularly interested in how long any effects might persist, as the pandemic has lasted longer than expected.
An advantage of the difference-in-difference framework is that level differences between the treated and control groups
are acceptable as long as the two groups are following parallel trends prior to the event of interest.
Using our May cohort and control group as an example, the average log expenses of May cohort was larger than that of
the control group, but the difference between them was stable. After May 2020, small business expenses could have been
recovering for both the treated group of firms with PPP loans as well as for the control group, but it is the difference in
how each group’s expenses changes—the difference in the differences—after May that is of interest.
The equation above shows the regression model estimated. We chose expenses as a proxy for business activity for two
reasons. First, unlike revenues, it is possible for expenses to be independent of consumer demand or pandemic-related
restrictions. Consider a business that was not permitted to operate or perhaps chose not to operate during the pandemic
and therefore generated no revenue. Another possibility is if the business did continue to operate, but customer demand
was limited due to the pandemic. However, the firm could have—using PPP loan proceeds or other funds—continued to pay
its employees, utilities, and other expenses. To the extent that PPP facilitated this continuity despite a disruption in
revenues, we would want to consider its expense activity. Moreover, expenses are often highly correlated with revenues
(Farrell et al. 2020), making it an appropriate proxy for business activity. Expenses, which include payroll expenses, are
defined as outflows that are not related to financing. Financing could include transfers to personal accounts or loan
repayment.
21
We observed electronic payroll payments in 37 percent of firms in our sample.
The model for each loan cohort in 2020 was estimated separately with the same control group of firms with 2021 PPP
loans.
22
For example, the May cohort included firms with May 2020 PPP loans in the treated group and firms with 2021
PPP loans in the control group.
23
Groups were weighted so that the treated and control groups each represented half of
the sample. Our model was estimated at the firm-month level using the above equation, where f indexed the firm and m
indexed the month. The variable PPP
f
is a dummy variable equal to one if the firm was in the PPP month cohort, and Month
m
is a dummy variable for the month.
24
Also included are controls for industry, state, and month.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
15
Data Explanation
Figure 1: Median changes in firm balances, revenues, and expenses relative to January 2020
Line chart showing median changes in small business balances, revenues, and expenses relative to January 2020 in each calendar month from
November 2019 through August 2021. The balance line shows that median balances increased in April 2020 and stayed at elevated levels through
August 2021, relative to January 2020. The revenues and expenses lines show that median revenues and expenses fell to -30 percent of January
2020 levels in April 2020 and remained below January 2020 levels in most months through August 2021.
View chart version
Figure 2: Median change in small business expenses relative to January 2020
Line chart showing the median change in expense relative to January 2020 for the May 2020 PPP recipient cohort and control group from
November 2019 through December 2020. The chart shows that expenses for each group were similar until April 2020, when the control group had a
larger drop in expenses, relative to January 2020, than the May 2020 PPP recipient cohort. The gap in expenses between the two groups is greatest
in May 2020, when the PPP recipients increased spending more sharply than the control group.
View chart version
Figure 3: Effect of PPP loans on small business expenses among 2020 recipients
Bar chart showing the estimated effect of PPP on small business expenses compared to the control group. Chart shows the estimates as the number
of months since PPP receipt on the x-axis, and ranges from 0 to 4 months. In the month of PPP receipt, expenses increased by 42 percent relative to
the control group, followed by 25 percent, 11 percent, 6 percent, and 3 percent in the subsequent four months.
View chart version
Figure 4: PPP had larger effects on small business expenses in April and May 2020
Four scatter plot charts with error bars at the 95 percent confidence interval showing the effect of PPP on small business expenses for each monthly
cohort: April 2020, May 2020, June 2020, and July 2020. Estimated effect shown for every calendar month between November 2019 and December
2020. Chart shows almost no difference between each cohort and the control group prior to PPP receipt for the April, June, and July cohorts. The
May cohort shows expenses increased by 13 percent more than the control group in April 2020 compared to the control group. Expenses for the
April cohort increased by 37 percent relative to the control group in April 2020, expenses for the May cohort increased by 47 percent more than the
control group in May 2020, expenses for the June cohort increased by 14 percent more than the control group in June 2020, and expenses for the
July cohort increased by 22 percent more than the control group in July 2020. Charts show that in all cohorts the effect of PPP on small business
expenses diminishes to zero within four months of PPP receipt and then no difference with the control group through the end of the year.
View chart version
Figure 5: Effect of PPP on small business expenses by firm size within May cohort
Scatter plot charts with error bars at 95 percent confidence interval showing the effect of PPP on small businesses in the May cohort sized by 2019
expenses. Estimated effect shown for every calendar month between November 2019 and December 2020 for the first and fourth quartiles of firm
expenses. Among small businesses in both quartiles there is almost no difference in expenses until April 2020. Small businesses in quartile 1 have
changes to expenses that are 8 percent higher than the control group in April 2020, and small businesses in quartile 4 have changes that are 19 per-
cent higher than the control group in April 2020. Expenses for small businesses that received PPP in the first quartile of 2019 expenses increased by
61 percent more in May 2020 than the control group, expenses for small business that received PPP in the fourth quartile of 2019 expenses changed
by 38 percent more in May 2020 than the control group. The differences in effects between small businesses in both quartiles 1 and 4 and the con-
trol group diminish over the next four months, and there are no differences in effects at the end of 2020.
View chart version
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
16
Figure 6: Effect of PPP on expenses of restaurants in the May cohort
Scatter plot chart with error bars at 95 percent confidence interval showing the effect of PPP on restaurants in the May cohort. Estimated effect
shown for every calendar month between November 2019 and December 2020. Almost no difference in expenses between May PPP recipient
restaurants and the control group prior to PPP receipt. Expenses for restaurants that received PPP in May 2020 increased by 61 percent more than
expenses for the control group in May 2020. Changes to expenses for restaurants that received PPP in May 2020 were lower than those for the con-
trol group from September 2020 through December 2020.
View chart version
Figure 7: Ratio of monthly restaurant expenses to April through December average
Line chart showing monthly expenses from April 2020 through December 2020 as a ratio of the average monthly expenses over the entire period,
separated by restaurants that received PPP in May 2020 compared to the control group. Restaurants that received PPP in May 2020 had high-
er-than-average expenses from May 2020 through September 2020. The control group had higher-than-average expenses from July 2020 through
the end of the year.
View chart version
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
17
References
Autor, David, David Cho, Leland D. Crane, Mita Goldar, Byron
Lutz, Joshua Montes, William B. Peterman, David Ratner,
Daniel Villar, and Ahu Yildermaz. 2020. “An Evaluation of the
Paycheck Protection Program Using Administrative Payroll
Microdata.” Working paper. https://economics.mit.edu/
files/20094.
Callaway, Brantly and Pedro H.C. Sant’Anna. 2020. “Difference-
in-Differences with Multiple Time Periods.” Journal of
Econometrics. https://doi.org/10.1016/j.jeconom.2020.12.001.
Chetty, Raj, John N. Friedman, Nathaniel Hendren, Michael
Stepner, and the Opportunity Insights Team. 2020. “The
Economic Impacts of COVID-19: Evidence from a New Public
Database Built Using Private Sector Data.” National Bureau of
Economic Research. Working Paper 27431. https://www.nber.
org/papers/w27431.
Farrell, Diana, Chris Wheat, and Chi Mac. 2020a. “Small
Business Financial Outcomes during the Onset of COVID-19.”
JPMorgan Chase Institute.
Farrell, Diana, Chris Wheat, and Chi Mac. 2020b. “Small
Business Financial Outcomes during the COVID-19 Pandemic.”
JPMorgan Chase Institute.
Farrell, Diana, Chris Wheat, and Chi Mac. 2020c. “Small
Business Owner Race, Liquidity, and Survival.” JPMorgan
Chase Institute.
Farrell, Diana, Chris Wheat, Chi Mac, and Bryan Kim. 2020.
“Small Business Expenses during COVID-19.” JPMorgan Chase
Institute.
Federal Reserve Banks. 2021a. “Small Business Credit
Survey: 2021 Report on Employer Firms.” https://www.
fedsmallbusiness.org/medialibrary/FedSmallBusiness/
files/2021/2021-sbcs-employer-firms-report.
Federal Reserve Banks. 2021b. “Small Business Credit Survey:
2021 Report on Nonemployer Firms.” https://www.fed-
smallbusiness.org/medialibrary/FedSmallBusiness/
files/2021/2021-sbcs-nonemployer-firms-report.
Goodman-Bacon, Andrew. 2021. “Difference-in-differences with
Variation in Treatment Timing.” Journal of Econometrics.
https://doi.org/10.1016/j.jeconom.2021.03.014.
Hubbard, R. Glenn and Michael R. Strain. 2020. “Has the
Paycheck Protection Program Succeeded?” National Bureau
of Economic Research. Working Paper 28032. http://www.
nber.org/papers/w28032.
JPMorgan Chase Institute. 2020. “Small Business Cash
Liquidity in 25 Metro Areas.” https://www.jpmor-
ganchase.com/institute/research/small-business/
small-business-cash-liquidity-in-25-metro-areas.
Opportunity Insights Economic Tracker, Small Business
Revenue. 2021. https://tracktherecovery.org/?s91z8.
U.S. Small Business Administration. 2020a. “Paycheck
Protection Program (PPP) Report: Approvals through
08/08/2020.” 2020. https://www.sba.gov/sites/default/
files/2021-09/PPP_Report%20-%202020-08-10-508.pdf.
U.S. Small Business Administration. 2020b. “2020 Small
Business Profile.” https://cdn.advocacy.sba.gov/wp-content/
uploads/2020/06/04144224/2020-Small-Business-Econom-
ic-Profile-US.pdf.
U.S. Small Business Administration. 2021. “Paycheck Protection
Program (PPP) Report: Approvals through 05/31/2021.”
https://www.sba.gov/sites/default/files/2021-06/PPP_Report_
Public_210531-508.pdf.
Wheat, Chris and Chi Mac. 2021. “The Paycheck Protection
Program: Small Business Balances, Revenues, and Expenses
in the Weeks after Loan Disbursement.” JPMorgan Chase
Institute.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
18
Box 1: Endnotes
a CARES Act. https://www.con-
gress.gov/116/plaws/publ136/
PLAW-116publ136.pdf.
b Small Business Administration
Press Release Number 20-34,
April 24, 2020. https://www.
sba.gov/article/2020/apr/24/
joint-statement-sba-administra-
tor-jovita-carranza-treasury-secre-
tary-steven-t-mnuchin-resumption.
c Paycheck Protection Program and
Health Care Enhancement Act.
https://www.congress.gov/116/plaws/
publ139/PLAW-116publ139.pdf.
d Paycheck Protection Program
Flexibility Act of 2020. https://
www.congress.gov/116/plaws/
publ142/PLAW-116publ142.pdf.
e PPP Extension Act of 2020. https://
www.govinfo.gov/content/pkg/PLAW-
116publ147/pdf/PLAW-116publ147.pdf.
f https://home.treasury.gov/
policy-issues/coronavirus/
about-the-cares-act.
g https://www.congress.gov/116/bills/
hr133/BILLS-116hr133enr.pdf.
h Department of the Treasury
Press Release, January 8, 2021.
https://home.treasury.gov/
news/press-releases/sm1230.
i “Business Loan Program
Temporary Changes; Paycheck
Protection Program-Revisions
to Loan Amount Calculation
and Eligibility.” Small Business
Administration. March 8, 2021.
https://www.federalregister.gov/
documents/2021/03/08/2021-04795/
business-loan-program-tem-
porary-changes-paycheck-pro-
tection-program-revi-
sions-to-loan-amount.
j PPP Extension Act of 2021. https://
www.whitehouse.gov/brief-
ing-room/legislation/2021/03/30/
press-release-bill-signing-h-r-1799/.
k PPP Information Sheet. https://
home.treasury.gov/system/
files/136/PPP--Fact-Sheet.pdf.
l “Paycheck Protection Program
Second Draw Loans: Top-Line
Overview.” Department of the
Treasury. https://home.treasury.
gov/system/files/136/Top-line-
Overview-of-Second-Draw-PPP.pdf.
m Department of the Treasury
Press Release, June 8, 2020.
https://home.treasury.gov/
news/press-releases/sm1026.
n “Paycheck Protection Program:
How to Calculate Maximum Loan
Amounts—By Business Type.” Small
Business Administration. June 26,
2020. https://www.sba.gov/sites/
default/files/2020-06/How-to-
Calculate-Loan-Amounts-508_1.pdf.
o Effective March 4, 2021. “Business
Loan Program Temporary Changes;
Paycheck Protection Program-
Revisions to Loan Amount Calculation
and Eligibility.” Small Business
Administration. March 8, 2021.
https://www.federalregister.gov/
documents/2021/03/08/2021-04795/
business-loan-program-temprary-
changes-paycheck-protection-pro-
gram-revisions-to-loan-amount.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
19
1 EIP and CTC were made to eligi-
ble individuals as opposed to firms.
However, the majority of small busi-
nesses at large and in our sample
are nonemployer firms. Pass-through
entities, such as sole proprietor-
ships, report net income on their
owners’ tax returns. If those tax
returns used the business’s direct
deposit information, the payments
could have been deposited into
the business checking accounts.
2 Based on approvals through
May 31, 2021 (Small Business
Administration 2021).
3 New York Small Business Pandemic
Recovery Initiative Fact Sheet (New
York State Senate, updated August
1, 2021). https://www.nysenate.gov/
newsroom/articles/2021/brad-hoyl-
man/new-york-small-business-pan-
demic-recovery-initiative-fact-sheet.
4 Small Business Relief Grant Fact
Sheet (Ohio Development Services
Agency). https://businesshelp.
ohio.gov/pdf/10292020-Small-
business-relief-grant.pdf.
5 We observed electronic payroll pay-
ments for 37 percent of firms in our
sample. However, not all firms with
employees use electronic payroll
processors, and there may have
been employer firms for whom we
did not observe payroll payments.
For example, they may have pro-
cessed payroll independently using
paper checks, as opposed to using
an electronic payroll processor.
6 We used the dates on which trans-
actions were posted to accounts.
Parties may be aware of pending
transactions prior to the post date.
7 The omitted period is February 2020.
8 Revised June 5, 2020 by the Paycheck
Protection Program Flexibility Act of
2020. https://www.congress.gov/116/
plaws/publ142/PLAW-116publ142.pdf
9 The interim final rule with details
on loan forgiveness was pub-
lished June 1, 2020. https://home.
treasury.gov/system/files/136/
PPP-IFR-Loan-Forgiveness.pdf.
10 Federal Register, Vol. 85, No. 86.
May 4, 2020. https://home.trea-
sury.gov/system/files/136/Interim-
Final-Rule-on-Disbursements.pdf.
11 Their sample consisted of firms
applying for PPP loans greater than
$150,000 (Hubbard and Strain 2020).
12 Firm size can be measured in
different ways. Commonly used
measures include the number of
full-time employees or annual
revenues. Since our model spec-
ification is based on expenses,
and revenues and expenses
are highly correlated, we used
expenses to categorize firm size.
13 Median small restaurant expenses
were 18 percent lower in September
2020 compared to the prior year;
among all small businesses, expenses
were 7 percent lower (Farrell et al.
2020). The Opportunity Insights
Economic Tracker reported that
small business revenue in the lei-
sure and hospitality industry was
59 percent lower at the end of
June 2021 than in January 2020.
In comparison, revenues among
all small businesses was 43 per-
cent lower than January values.
14 The $28.6 billion Restaurant
Revitalization Fund was estab-
lished by the American Rescue Plan.
Applications opened May 3, 2021
(Small Business Administration
Release Number 21-33). https://
www.sba.gov/article/2021/apr/27/
sba-administrator-guzman-announc-
es-application-opening-286-bil-
lion-restaurant-revitalization-fund.
15 The $16.2 billion Shuttered
Venue Operators Grants pro-
gram opened April 8, 2021 (Small
Business Administration Release
Number 21-28). https://www.sba.
gov/article/2021/apr/07/sba-open-
shuttered-venue-operators-grants-
applications-april-8-12-pm-edt.
16 We identified PPP loan inflows
from other lenders based on
transaction characteristics.
17 The resulting sample reflects sur-
viving firms in that their deposit
accounts remained open through
the end of 2020. However, firms
without transaction activity in their
open accounts were included in the
sample. This requirement was con-
sistently applied across treated and
control groups. The control group,
consisting of firms with 2021 PPP
loans, necessarily kept their accounts
open through 2020, as they depos-
ited their loan proceeds in 2021.
18 For example, Chase PPP loans
deposited into accounts other
than Business Banking accounts
would not be included. Also
excluded are firms with accounts
that do not meet our activity cri-
teria and firms that received PPP
loans in both 2020 and 2021.
19 Small Business Administration.
March 8, 2021. “Business Loan
Program Temporary Changes;
Paycheck Protection Program-
Revisions to Loan Amount
Calculation and Eligibility.” https://
www.federalregister.gov/docu-
ments/2021/03/08/2021-04795/
business-loan-program-temporary-
changes-paycheck-protection-pro-
gram-revsions-to-loan-amount.
20 Self-employed workers are usu-
ally not eligible for unemploy-
ment benefits, but the CARES
Act gave states the option to
extend benefits to them.
21 Deposit account transactions are
classified based on various trans-
action characteristics. Transactions
Endnotes
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
20
are aggregated to the month
based on the date the transac-
tion posted to the account.
22 Including all cohorts in one model
with staggered treatment may lead
to unexpected weighting across
cohorts (Goodman-Bacon 2021).
23 Within a month cohort, the loans
may have been received at differ-
ent times of the month, but this is
not modeled at our level of aggre-
gation. For our May cohort, which
is the largest in our sample, most
loan proceeds were deposited in
early May 2020 after PPP applica-
tions resumed on April 27, 2020.
24 February 2020 is the omitted month.
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
21
Acknowledgements
We thank Nicholas Tremper and Noah Forougi for their hard work and vital contributions to this research. Additionally,
we thank Emily Rapp and Robert Caldwell for their support. We are indebted to our internal partners and colleagues,
who support delivery of our agenda in a myriad of ways, and acknowledge their contributions to each and all releases.
We would like to acknowledge Jamie Dimon, CEO of JPMorgan Chase & Co., for his vision and leadership in establishing
the Institute and enabling the ongoing research agenda. We remain deeply grateful to Demetrios Marantis, Head of
Corporate Responsibility, Heather Higginbottom, Head of Research & Policy, and others across the firm for the
resources and support to pioneer a new approach to contribute to global economic analysis and insight.
Suggested Citation
Wheat, Chris and Chi Mac. 2021. “Did the Paycheck Protection Program Support Small Business Activity?”
JPMorgan Chase Institute. https://www.jpmorganchase.com/institute/research/small-business/did-ppp-support-small-
business-activity.
For more information about the JPMorgan Chase Institute or this report, please see our website
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
22
The JPMorgan Chase Institute has adopted rigorous security protocols and checks and balances to ensure all customer data
are kept confidential and secure. Our strict protocols are informed by statistical standards employed by government agencies
and our work with technology, data privacy and security experts who are helping us maintain industry-leading standards.
There are several key steps the Institute takes to ensure customer data are safe, secure and anonymous:
• The Institute’s policies and procedures require that data it receives and processes for research purposes do not identify
specific individuals.
• The Institute has put in place privacy protocols for its researchers, including requiring them to undergo rigorous back-
ground checks, and enter into strict confidentiality agreements. Researchers are contractually obligated to use the data
solely for approved research, and are contractually obligated not to re-identify any individual represented in the data.
• The Institute does not allow the publication of any information about an individual consumer or business. Any data point
included in any publication based on the Institute’s data may only reflect aggregate information or information that is
otherwise not reasonably attributable to a unique, identifiable consumer or business.
• The data are stored on a secure server and only can be accessed under strict security procedures. The data can-
not be exported outside of JPMorgan Chase’s systems. The data are stored on systems that prevent them from being
exported to other drives or sent to outside email addresses. These systems comply with all JPMorgan Chase Information
Technology Risk Management requirements for the monitoring and security of data.
The Institute prides itself on providing valuable insights to policymakers, businesses, and nonprofit leaders. But these
insights cannot come at the expense of consumer privacy. We take all reasonable precautions to ensure the confidence and
security of our account holders’ private information.
JPMC Institute Public Data Privacy Notice
Did the Paycheck Protection Program Support Small Business Activity? Research Brief
23
This material is a product of JPMorganChaseInstitute and is provided to you solely for general information purposes. Unless
otherwise specifically stated, any views or opinions expressed herein are solely those of the authors listed and may differ from
the views and opinions expressed by J.P. Morgan Securities LLC (JPMS) Research Department or other departments or divisions
of JPMorganChase&Co. or its affiliates. This material is not a product of the Research Department of JPMS. Information has
been obtained from sources believed to be reliable, but JPMorganChase&Co. or its affiliates and/or subsidiaries (collectively
J.P. Morgan) do not warrant its completeness or accuracy. Opinions and estimates constitute our judgment as of the date of
this material and are subject to change without notice. The data relied on for this report are based on past transactions and
may not be indicative of future results. The opinion herein should not be construed as an individual recommendation for any
particular client and is not intended as recommendations of particular securities, financial instruments, or strategies for a
particular client. This material does not constitute a solicitation or offer in any jurisdiction where such a solicitation is unlawful.
Disclaimer
©2021 JPMorgan Chase & Co. All rights reserved. This publication or any portion hereof may not be reprinted, sold, or
redistributed without the written consent of J.P. Morgan. www.jpmorganchaseinstitute.com
