Murat Tasci et al., PPP Loans & State-level Employment Growth (2021)
Summary
Economic Commentary Number 2021-20 of the Federal Reserve Bank of Cleveland, dated November 22, 2021, by Murat Tasci, Bezankeng Njinju and Hana Braitsch. It examines the first round of Paycheck Protection Program loans, from March 2020 to early August 2020, and their effect on state-level employment growth. The authors find that PPP loans helped mitigate the negative impact of the pandemic recession on state employment, and that states receiving most of their funding early had smaller employment declines than states receiving comparable funds later. The paper notes that about 89 percent of all loans and 91 percent of total PPP loan value had been forgiven in full or in part.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
Cite this document
“Murat Tasci et al., PPP Loans & State-level Employment Growth (2021).”
Original at www.clevelandfed.org. Archived at Pandemic Darlings, https://pandemicdarlings.com/source-documents/murat-tasci-et-al-ppp-loans-state-level-employment-growth-2021-0830989cc5/
, SHA-256 624e86f735448a0012632e206259c78f660642460ed538ed706c4cd1ac567e1a.Full text
Number 2021-20
November 22, 2021
PPP Loans & State-level Employment Growth
Murat Tasci, Bezankeng Njinju, Hana Braitsch
In this Economic Commentary, we focus on the first round of Paycheck Protection Program (PPP) loans granted
beginning in March 2020 until early August 2020, when turbulence in the labor market was pronounced, in order to
demonstrate the PPP’s effects on local labor markets. We find that PPP loans helped mitigate the negative impact of
the pandemic recession on state-level employment growth. States that received most of their funding early in the loan
period had smaller employment declines than did states that received comparable funds later in the period.
The SARS-CoV-2 (COVID-19) pandemic and the mitigation of employment and PPP loans across states. We rely on the
efforts instituted by state and local governments slowed aggregate impact of the loans at the state level to address this
economic activity drastically in spring 2020. Unlike in a question because though we now have information on the
typical recession, social distancing measures and widespread loans made to individual businesses, corresponding data on
shutdowns forced many businesses to cease activity without employment levels for individual businesses are not available.
officially going out of business. This unique situation motivated
Previous research suggests that employment subsidies in the
policymakers to institute a large-scale employment subsidy
wake of a serious recession could have more long-lasting
program, the Paycheck Protection Program (PPP), in the
positive effects on employment than would alternative policies
effort to save small businesses from going under. The United
that provide hiring subsidies and payroll tax deductions, albeit
States Congress quickly passed the Coronavirus Aid, Relief,
at a higher fiscal cost (Kitao, Sahin, and Song, 2010). PPP loans
and Economic Security (CARES) Act and instituted the PPP
act as inexpensive employment subsidies, from the government
geared toward small businesses to weather the adverse effects
and facilitated by lenders, to individual businesses since these
of the pandemic. The PPP was the largest spending item in the
loans are, in principle, forgivable. To date, about 89 percent of
CARES Act and was in effect between March 27 and August 8
all loans and 91 percent of the total PPP loan value from 2020
in 2020. The program was later reinstated on January 11, 2021,
has been forgiven in full or in part.4
with new funding through the Consolidated Appropriations
Act, 2021, which was, after an extension, active until the end of We find that PPP loans helped mitigate the negative impact
June 2021.1 of the pandemic-related recession on state-level employment
growth. This effect was more pronounced during the initial
In this Economic Commentary, we focus on the first round of PPP
stage of the pandemic in the United States, when uncertainty
loans granted beginning in March 2020 until early August
surrounding the pandemic and its effects prevailed and
2020, when turbulence in the labor market was pronounced,
unemployment rates peaked across the country through April
in order to demonstrate its effects on local labor markets.2
2020. The positive effect becomes less significant during
The PPP was designed to “provide small businesses with
the early part of the recovery starting in May and ending
the resources they need to maintain their payroll, hire back
with the expiration of the PPP in August 2020. Overall, our
employees who may have been laid off, and cover applicable
analysis suggests that one additional week of payroll support
overhead.”3 Given these stated intentions, the main question
in the form of PPP loans during this initial distribution period
we address is whether PPP loans supported employment
would have mitigated the state-level employment decline by
during this period. We conduct an analysis using comparisons
Murat Tasci is a senior research economist at the Federal Reserve Bank of Cleveland. Bezankeng Njinju and Hana Braitsch are research analysts at the Federal Reserve
Bank of Cleveland. The authors thank Mark Schweitzer and Garrett Borawski for providing them with the geocoded version of the loan level PPP data from the Small
Business Administration. The views authors express in Economic Commentary are theirs and not necessarily those of the Federal Reserve Bank of Cleveland or the
Board of Governors of the Federal Reserve System or its staff.
Economic Commentary is published by the Research Department of the Federal Reserve Bank of Cleveland. Economic Commentary is also available on the Cleveland
Fed’s website at www.clevelandfed.org/research. To receive an email when a new Economic Commentary is posted, subscribe at www.clevelandfed.org/subscribe-EC.
ISSN 2163-3738 DOI: 10.26509/frbc-ec-202120
between around 1.5 percentage points to 2.3 percentage points. larger total loan amounts distributed. By the same token, the
Moreover, conditional on receiving the same amount of PPP small-business sector in two similarly sized states, say, Maryland
loans over the entire period, states whose small businesses and Missouri, might incur different payroll costs because of
received the bulk of their overall PPP loans early on had much geographical differences in wages and benefits.9 To alleviate this
smaller net employment declines over the entire period, from anomaly, we normalize the total PPP loan amount in a state
February to August, than did states that received the bulk of with the total weekly payroll of the small-business sector in that
their PPP loans later in the period. Overall, our results indicate state. For instance, if all the small businesses in a state apply
that the PPP had significant benefits. But it remains for future for PPP loans and obtain the maximum eligible amount, the
research to compare these benefits to the costs of the program state-level measure of the PPP loan will be slightly more than
and the efficacy of possible alternatives. 10, that is, somewhat greater than 10 weeks of average payroll
costs in a reference year. In our case, the reference year will be
2018, which is the year for which we have the most recent data
Data on PPP loans and labor market outcomes from the Census Bureau’s Statistics of US Businesses (SUSB)
We aggregate individual loans from the PPP data at the state to identify a measure of total payrolls for firms with fewer than
level and focus on two distinct episodes over the course of the 500 employees.10
first round of the PPP applications based on the evolution of Based on this normalization, our data indicate that states
the pandemic and labor markets.5 The first episode covers the and the District of Columbia (DC) obtained PPP loans
period just before the pandemic in the United States through totaling somewhere between 6 weeks (District of Columbia)
its earliest months, from February 2020 through April 2020, to 11.3 weeks (Hawaii) of their 2018 average payroll costs.11
when the unemployment rate peaked nationally. This episode The second lowest amount in the sample is 8.5 weeks for
essentially covers the initial shock to the US economy and labor Massachusetts. It appears that most of the states received
markets in general.6 The second episode covers the period loans equivalent to 85 percent to 115 percent of 10 weeks of
that immediately follows and ends by August 2020, when the their 2018 payroll averages. This result echoes Schweitzer and
first-round PPP expired.7 We interpret this episode as the initial Borawski’s (2021a) analysis at the industry level.
labor market recovery. As a result, we have two measures of
PPP utilization for each state, one during the COVID-19 shock Figure 1 shows for each state and DC how the composition
and one during the recovery period. of the total PPP loans was distributed over time. States that
received large shares of PPP loans as a fraction of 2018
It is, however, hard to interpret the amount of total PPP payrolls through April 2020 received smaller loans during the
loans at the state level on its own. Larger states or states second phase after April. For most locations, when adding up
with small businesses that on average have higher revenue the weeks of loans in the two phases, PPP borrowing was about
streams could have applied for and received larger amounts 10 weeks of 2018 payrolls. That is, most firms maxed out their
of PPP loans irrespective of how severely they were affected PPP borrowing. Two more patterns also emerge from this
by the pandemic. The level of the PPP loans is particularly picture. First, most of the locations that received a larger share
uninformative because of the legislative design of the PPP. of their total PPP loans during the first wave were relatively
Since businesses could apply for loans that amount to between small states.12 Interestingly, three large states by population
10 and 11 weeks’ worth8 of their average payroll expenses, received the larger share of their loans or received more than 5
states with larger populations will appear in our data with weeks of 2018 payroll averages during the recovery period.13
Figure 1. Total PPP Loans in Terms of 2018 Weekly Payrolls
10
NE
SD
ND KS
HI
ME
Loans Approved through April 2020
OK IA
MS
8 WI MT
MO
MN
VT KY OH ID
AR WY IN
AL
UT
NH PA LA
WV TN
RI CO
IL
AK MI GA
6 DE
TX SC
MA FL
MD NC
VA
NM CT
OR AZ
WA NJ
NY NV
4 CA
DC
Source: Small Business
Administration, Census Bureau,
2 authors’ calculations
2 3 4 5 6
Loans Approved after April 2020 2
Figure 2. Employment Growth − Initial Shock vs the Recovery
-10 NE
SD
UT
Employment Growth % − from February to April 2020
WY AR
ID
OK
IA AZ KS
TX
ND
WA TN
VA MO
AL
NC WI
DC NM IL
-15 MN GA
CO
MS MT
FL
WV SC
AK LA OR DE
ME
MD
KY
CA IN
OH
NH
CT
NJ
-20 MA
PA
NY VT
RI
-25 NV
HI MI
Source: Bureau of Labor Statistics
0 5 10 15 20 25
Employment Growth % − from April to August 2020
In order to evaluate the effects of the PPP loans on labor Table: State-Level Employment Growth and PPP Loans
market outcomes, we focus on payroll employment at the
Dependent Variable: Employment growth
state level in the private sector.14 As we did with the total PPP
loans, we compute employment growth rates in two distinct (1) (2) (3) (4)
February April to February February
periods: first from February to April 2020 and second from to April August to August to August
April to August 2020. The first measure gauges the severity 2020 2020 2020 2020
of the labor market shock, and the second quantifies the Total PPP 0.928**
strength of the early recovery. Based on previous research, loans (0.387)
one would expect to see that states with larger employment through April
declines at the onset of the pandemic recession would Total PPP 0.414
bounce back stronger during the recovery episode (Tasci loans after (0.570)
and Zevanove, 2019). As Figure 2 shows, this was indeed the April
case. States that faced larger drops in employment early on New case -0.246**
rate through (0.109)
later experienced stronger growth in employment in the early
April
phase of the recovery. Nevertheless, none of the states or DC
New case -0.220*
recovered fully by August 2020 from the steep employment rate after (0.110)
losses experienced in the initial shock. Data presented in April
Figures 1 and 2 cover all 50 states and DC. However, for the Total PPP 2.280*** 1.471**
formal analysis in the rest of this Commentary, we omit from loans (0.509) (0.549)
our sample data from Hawaii and DC, the two outliers.15 Overall -0.049 0.084
average new (0.109) (0.111)
case rate
Did PPP loans lessen employment loss across states? Fraction -0.861** 0.685 -0.590* -0.647**
of HCI (0.405) (0.437) (0.296) (0.275)
Our analysis suggests that the answer is yes, PPP loans did workforce
lessen employment loss across states. For a given loan amount, Employment -0.011 -0.174 -0.034 -0.061
the effects of the PPP were strongest at the beginning of the share (0.087) (0.110) (0.072) (0.068)
recession; loans received earlier in the pandemic seem to of small
business
have had a disproportionately larger positive effect on net
employment loss at the state level. The prevalence of both Fraction of 0.109***
early PPP (0.038)
high-contact-intensity (HCI) jobs and COVID-19 infections loans
within a state negatively affected employment growth,
Observations 49 49 49 49
especially early in the pandemic. R-squared
0.379 0.144 0.379 0.479
Note: Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
3
PPP loans and employment decline through April 2020 covariates. Column 1 in the table shows that both the total
The CARES Act became law on March 27, 2020, and the PPP loan amount and the average number of new cases
US Department of the Treasury and the US Small Business were significant variables correlated with payroll growth
Administration launched the PPP immediately. Private from February through April 2020. Our coefficient estimates
financial institutions accepted applications and administered suggest that one additional week of PPP loans mitigated the
the approval process. Despite the brief window between decline in employment by almost 1 percentage point (0.93),
the launch of the program in late March and the significant a number which is statistically significant and economically
contraction in total employment by April 2020, many states meaningful.17
were able to obtain large sums during the first wave of We also control for some other observable features of the
lending. Figure 3 shows that there was quite a bit of variation states to understand the evolution of employment within
on this dimension, represented on the x-axis by equivalent each state during the pandemic periods outlined. We know,
number of weeks of average 2018 payroll. There is a positive for instance, the pandemic and related mitigation efforts
correlation between total PPP loans and employment growth disproportionately affected occupations and jobs that require
in this first period. In other words, states that received larger high social contact with other individuals, such as grocery
PPP loans experienced smaller employment declines by April store clerks, hairstylists, servers, and so on. Hence, states with
2020, as represented on the y-axis. a larger share of employment in HCI jobs were expected to
In Figure 3, each state is represented with a circle; the size be especially vulnerable to employment loss.18 As column 1 of
of the circle reflects the average new case rate per day from the table shows, we find a significant and negative relationship
February through April 2020.16 We use data on new case between employment growth during the early days of the
rates to control for the severity of the pandemic at the state pandemic and the share of employment in HCI occupations.
level. It appears that most of the states with large outbreaks The average state had about 21.7 percent of its workforce
did in fact experience more employment contraction. This employed in HCI occupations. Based on our results, this
is expected as consumers and workers started to avoid their implies that a state with a slightly larger employment share
usual economic activities, either voluntarily or because of for HCI occupations, at 23 percent, would have suffered
local government mandates and stay-at-home orders. It is 1 percent more employment loss. Finally, we find no
also interesting to note that states that obtained larger PPP significant effect of a larger small-business sector in a state
loans were not the areas hardest hit by COVID-19 during this on the employment loss experienced at the onset of the
first loan dispersal period. We formally test these correlations pandemic recession.
within a regression framework and control for additional
Figure 3. PPP Loans through April 2020 vs Initial Employment Shock
-10 UT
SD
NE *Weighted by COVID Case Rate
WY
AR ID
OK **Case Rate: confirmed
AZ
TX
IA KS COVID−19 cases per 100,000
WA ND
TN people, seven−day moving
Employment Growth %
VA MO
AL
average − monthly rate is a
NC WI
NM IL
-15 GA MNMT MS
FL
CO
WV
simple average of weekly rate
OR SC AK LA
DE
ME
MD KY
CA IN
OH
NH
CT
NJ
-20 MA
PA
NY VT Source: Bureau of Labor Statistics,
RI Small Business Administration,
-25 Census Bureau, Opportunity
NV
Insights Economic Tracker, authors’
MI
calculations
4 5 6 7 8 9
Weeks of Average 2018 Payrolls
4
PPP loans and the early labor market recovery until August 2020 larger employment rebound between April and August. This
After the initial shock of the recession, labor markets started was a period when outbreaks affected multiple states over the
to recover, as demonstrated in the incoming data for May course of the summer, and unlike in the early phase of the
2020, and they continued to do so at a varying pace for pandemic during which caseloads were driven by a handful
each state throughout August 2020. For instance, Wyoming of states, cases on average stayed relatively high across the
and Alaska barely registered growth during this period, 0.9 country. We can observe this by comparing Figures 3 and 4
percent and 2.8 percent, respectively, whereas Michigan and the relative size of the circles for each state. As infections
witnessed a payroll growth above 20 percent from April to became more widespread geographically, the explanatory
August. Of course, since the employment base in April was power of outbreaks declined during this period (May through
much lower than the level in February, even an otherwise August), though they are still statistically somewhat significant.
incredible growth rate of 20 percent within a month was Other covariates such as the fraction of HCI jobs or the
not enough to erase the pandemic-induced employment loss size of the small-business sector did not seem to matter
experienced earlier. much for explaining differences across states in terms of
In Figure 4, we visually explore whether employment employment recovery during this period of the recovery.
growth over this early recovery phase was correlated with In all, our regression framework seems to provide a poor
PPP lending. Even though there seems to be significant fit for this period. One potential explanation for this is the
variation in terms of PPP loan receipts and employment underlying heterogeneity in terms of mitigation measures and
growth during this episode, the relationship of PPP loans to government policy at the state level. Unlike in the first phase
employment growth appears weaker in this period than in up to April 2020, when the entire country was experiencing
the early months of the pandemic. Column 2 in the table a lockdown of varying degrees, some states eased their
presents our results from the regression for this episode, coronavirus restrictions starting in early summer. In this
results which confirm the visual observation and suggest that analysis, we did not control for this as we could not find a
the correlation is not statistically significant. Thus, higher clear summary measure describing the whole episode (May
amounts of PPP loans were not associated with significantly through August 2020) uniformly for each state.
Figure 4. PPP Loans May through August 2020 vs Early Employment Recovery
25 *Weighted by COVID Case Rate
MI **Case Rate: confirmed
COVID−19 cases per 100,000
20 people, seven−day moving
Employment Growth %
RI
average − monthly rate is a
simple average of weekly rate
15 VT
IN NV
PA
MT OH
NH KY
SC CT NY
MS
DE ID
10 WI AL
MA
WV CO NJ FL
MO NC GA OR
ME KS TN MD UT
MN VA
SD AZ
IA AR IL LA CA
OKNE
5
ND TX
WA Source: Bureau of Labor Statistics,
NM
Small Business Administration,
WY
Census Bureau, Opportunity
AK Insights Economic Tracker, authors’
0 calculations
2 3 4 5 6
Weeks of Average 2018 Payrolls
5
PPP loans and net employment loss from February to August 2020 larger than our cutoff, to apply for PPP loans if they satisfied
Our study of the recessionary shock and the early recovery certain criteria.19 In Figure 5, we see some clustering around
period in isolation suggests that, to assess the overall impact 10 weeks, but there is still a lot of variation across states with
of PPP loans from 2020, we need to pay close attention to the respect to payroll equivalent.
early disbursement of loans at the onset of the recession. As When we analyze the cumulative employment change from
small businesses were contemplating what to do in the face February to August, the significant correlation between
of a somewhat novel threat to their survival, having financial PPP loan amounts and employment growth returns. As
support readily available could have mitigated the negative column 3 in the table shows, one additional week of PPP
impact of the shock. It is also important to keep in mind loans obtained throughout the whole period is associated
that for each individual eligible business, a PPP loan had a with a statistically significant 2.3 percent increase in the net
maximum feasible limit of 2.5 months of average payroll employment change over this period. The overall new case
costs. Thus, in our state-level analysis, there will be a very rate average over the sample period does not seem to be
close relationship between the PPP loans from the two periods correlated with employment growth. To understand the role
(loans received through April versus those obtained from May of PPP loans that states received before May 2020, we also
through August). For instance, Utah received PPP loans that look at an empirical specification in which we control for
were equivalent to 7 weeks of payrolls by the end of April how much of the overall PPP loans a state received within
2020, a disbursement amount which meant that about 3 to 4 that timeframe. As column 4 of the table indicates, even for
weeks of additional support could have been available for the states that received identical loan amounts, a state receiving a
reminder of the program. Indeed, as Figure 4 shows, Utah larger portion of the total amount earlier experienced smaller
received an additional 3.7 weeks’ worth of PPP loans from employment declines across the two periods together.
May to August. Since total loan amounts across the two periods The average state in our sample received 68 percent of its
should be close to 10 weeks of the aggregate payroll, PPP loan total PPP loans during April 2020, with significant variation
amounts at the state level for the first period should predict over this number across states. Consider two states, one of
the later amount obtained because of the limitations on the which at the end of April 2020 had received total PPP loans
number of weeks of payroll costs available (see Figure 1). placing the state at the tenth percentile of states ranked
Nevertheless, not all states obtained PPP loans that equal in terms of available funding percentages received and
exactly 10 to 11 weeks of payrolls in the small-business another one at the ninetieth percentile. The former received
sector, and there are several reasons for this. First, some 52 percent of its total loans in the early phase. The latter,
small businesses by our definition (fewer than 500 employees) however, obtained about 81 percent of its PPP loans at the
might have declined to apply for PPP loans. Second, those beginning of the PPP. Even if both states had received PPP
that applied could have requested loan amounts that were loans equivalent to 10 weeks of 2018 payroll over the course
lower than the maximum for which they were eligible. Finally, of the whole lending period, the latter would have suffered an
our definition for the eligible group of establishments (based employment loss that was 3.2 percentage points lower than
on size) misses other potential recipients. The CARES Act the former, a number which is a meaningful and statistically
allowed some businesses larger than 500 employees, that is, significant difference.
Figure 5. Total PPP Loans vs Net Employment Growth from February to August 2020
0 *Weighted by COVID Case Rate
ID
**Case Rate: confirmed
UT
COVID−19 cases per 100,000
people, seven−day moving
Employment Growth %
SD
MT
NE average − monthly rate is a
-5 AR
KS simple average of weekly rate
IN MS
TN AZ AL OK MO
NC SC WI IA
KY
VA OH
DE CO GA
TX
NH WV FL
WA ND
WY MN
IL OR
CT
PA ME
-10 MD
MI
LA
NM RI
NJ
CA VT
MA Source: Bureau of Labor Statistics,
Small Business Administration,
NY Census Bureau, Opportunity
-15
NV
Insights Economic Tracker, authors’
AK
calculations
8 9 10 11
Weeks of Average 2018 Payrolls
6
Conclusion 11. See section 2.c, for a list of reasons why state level total loans
The COVID-19 pandemic presented extraordinary could differ from 2.5 months of payrolls.
challenges for the US labor market. These extraordinary 12. Nebraska, North Dakota, Kansas, Vermont, South Dakota, and
challenges were met with a set of extraordinary responses by so on.
policymakers. In this Commentary, we analyzed the effect 13. California, New York, and Florida.
of timely distribution of Paycheck Protection Program loans 14. Even though PPP loans are offered mostly to small businesses,
and how such timing ultimately affected a state’s individual data on payroll employment for this firm class size are not
employment recovery. Our results suggest that loans provided available at a high frequency. Hence, we focus on total private
under the PPP umbrella in fact mitigated the negative effects employment. Data on private employment for states are from the
Current Employment Statistics for State and Metro Areas: SAE
of the COVID-19-induced recession to some extent. State- Home Page: US Bureau of Labor Statistics (bls.gov).
level data suggest that PPP loans equivalent to an additional
15. DC stands out as an outlier in terms of PPP usage (Figure 1).
week of payroll kept the state-level employment declines This status is likely a result of the composition of establishments
lower by somewhere between 1.5 percentage points to 2.3 in DC: the public sector is relatively large, and nonprofit
percentage points. Moreover, conditional on receiving the establishments constitute a larger fraction of the establishments
same amount, states that received the bulk of their overall in the class size on which we focus. For this reason, DC has been
omitted. Hawaii appears to be an outlier in terms of employment
PPP loans early on had much smaller net employment
growth performance. After contracting almost 26.5 percent in
declines over the entire period, February through August two months, employment grew only a very small amount, leaving
2020. More research is needed to understand the true it 24.7 percent below February’s level. This is likely related to the
economic mechanisms behind these results and how state- industry composition of employment in the state, which relies
level pandemic restrictions played a role. to a large degree on the tourism industry. Beyond Hawaii, net
employment change by the end of August ranged between -15
percent and -2 percent. For the empirical analysis we present
here, we ignore both Hawaii and DC. This omission does not
effect our formal results qualitatively.
Footnotes 16. Data on new case rates at the state level come from the
1. The Paycheck Protection Program Extension Act of 2021 Opportunity Insights Economic Tracker (Economic Tracker
was signed into law by President Biden on March 26, 2021, (tracktherecovery.org)).
extending the application period from March 31 to June 30, 17. During the Great Recession, average employment decline in a
2021. state was 7.8 percent. Thus, mitigating the employment decline
2. This Commentary uses the total amount lent per state rather than by a few percentage points could be an economically meaningful
the number of loans made in each state because we think the intervention.
actual dollar amounts are, on the whole, more informative than 18. We use the HCI definition of Leibovici, Santacreu, and
the number of loans made to small businesses per state. Famiglietti (2020) and use their estimates for state-level share of
3. See Assistance for Small Businesses | U.S. Department of the these occupations.
Treasury. 19. See Schweitzer and Borawski (2020a) for a detailed discussion on
4. See Forgiveness Platform Lender Submission Metrics, September this topic.
27 (sba.gov).
5. The loan level data we use come from SBA’s PPP data, geocoded
and appended with ACS data as described in Schweitzer and
Borawski (2021a and 2021b). References
6. With the exception of seven states (CT, DE, MA, MN, NJ, NM, Leibovici, Fernando, Ana Maria Santacreu, and Matthew Famiglietti.
NY) all of the US states experienced their peak unemployment 2020. “Social Distancing and Contact-Intensive Occupations.” On
rates in April 2020 and labor market started recovering the Economy (blog). March 24, 2020. https://www.stlouisfed.org/
afterwards. on-the-economy/2020/march/social-distancing-contact-intensive-
7. PPP applications resumed in January after the US Congress occupations.
passed the fiscal package, Consolidated Approprations Act of Schweitzer, Mark E., and Garrett Borawski. 2021a. “Which Industries
2021, in late December 2020 and added another $284 billion of Received PPP Loans?” Economic Commentary (Federal Reserve Bank of
funding. Cleveland), no. 2021–08 (April). https://doi.org/10.26509/frbc-
8. To be precise, 2.5 times average monthly payroll, the ec-202108.
measurement given, could correspond to a loan amount as large Schweitzer, Mark E., and Garrett Borawski. 2021b. “How Well Did
as 10.9 weeks of payrolls. PPP Loans Reach Low- and Moderate-Income Communities?”
9. In this example, both states employ about 1.2 million workers Economic Commentary (Federal Reserve Bank of Cleveland), no. 2021–13
in the small-business sector by our definition in this Commentary. (May): 1–5. https://doi.org/10.26509/frbc-ec-202113.
However, small businesses in Maryland incurred about $59 Tasci, Murat, and Nicholas Zevanove. 2019. “Do Longer Expansions
billion in payroll costs in 2018 compared to $46 billion in Lead to More Severe Recessions?” Economic Commentary (Federal
Missouri. Hence, larger total PPP loans for Maryland does not Reserve Bank of Cleveland), no. 2019–02 (January): 1–6. https://doi.
necessarily reflect a larger shock to small businesses or more org/10.26509/frbc-ec-201902.
generous support for the state compared to Missouri.
10. We follow a similar logic presented in Schweitzer and Borawski
(2021a) for picking our 500-employee threshold. Data can be
accessed at SUSB Datasets (census.gov).
7
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International
License. This paper and its data are subject to revision; please visit clevelandfed.org for
updates.
File and source
- File
- tasci-2021-ppp-loans-state-level-employment-growth.pdf
- Size
- 339,237 bytes
- SHA-256
- 624e86f735448a0012632e206259c78f660642460ed538ed706c4cd1ac567e1a
- Original
- www.clevelandfed.org