Steven Hamilton
POLICY PROPOSAL 2020-14 | SEPTEMBER 2020
From Survival to Revival:
How to Help Small Businesses
through the COVID-19 Crisis
The Hamilton Project seeks to advance America’s promise
of opportunity, prosperity, and growth.
We believe that today’s increasingly competitive global economy
demands public policy ideas commensurate with the challenges
of the 21st Century. The Project’s economic strategy reflects a
judgment that long-term prosperity is best achieved by fostering
economic growth and broad participation in that growth, by
enhancing individual economic security, and by embracing a role
for effective government in making needed public investments.
Our strategy calls for combining public investment, a secure social
safety net, and fiscal discipline. In that framework, the Project
puts forward innovative proposals from leading economic thinkers
— based on credible evidence and experience, not ideology or
doctrine — to introduce new and effective policy options into the
national debate.
The Project is named after Alexander Hamilton, the nation’s
first Treasury Secretary, who laid the foundation for the modern
American economy. Hamilton stood for sound fiscal policy,
believed that broad-based opportunity for advancement would
drive American economic growth, and recognized that “prudent
aids and encouragements on the part of government” are
necessary to enhance and guide market forces. The guiding
principles of the Project remain consistent with these views.
MISSION STATEMENT
The Hamilton Project • Brookings 1
SEPTEMBER 2020
From Survival to Revival:
How to Help Small Businesses
through the COVID-19 Crisis
Steven Hamilton
The George Washington University
is policy proposal is a proposal from the author(s). As emphasized in e Hamilton Project’s original
strategy paper, the Project was designed in part to provide a forum for leading thinkers across the nation to
put forward innovative and potentially important economic policy ideas that share the Project’s broad goals
of promoting economic growth, broad-based participation in growth, and economic security. e author(s)
are invited to express their own ideas in policy papers, whether or not the Project’s sta or advisory council
agrees with the specic proposals. is policy paper is oered in that spirit.
2 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
Abstract
e COVID-19 pandemic poses an existential threat to small businesses, with more than 400,000 lost since the crisis began.
Many small businesses are nancially fragile and not equipped to weather a prolonged period of substantially reduced revenues.
Further widespread business failures would destroy jobs and rm-specic capital, and hamstring the recovery. e main existing
source of support, the Paycheck Protection Program, has had mixed success, and is not well suited to what now looks to be a
prolonged contraction. In its place, we should signicantly expand the Employee Retention Credit to help cover small businesses’
payroll costs, and introduce a new Small Business Survival Credit to help cover small businesses’ xed costs. Looking to the
future, we should signicantly invest in the capabilities of the IRS so it may better support small businesses in future crises.
The Hamilton Project • Brookings 3
Table of Contents
ABSTRACT 2
INTRODUCTION 4
BACKGROUND 6
THE CHALLENGE 8
THE PROPOSAL 16
QUESTIONS AND CONCERNS 22
CONCLUSION 23
AUTHOR AND ACKNOWLEDGMENTS 24
ENDNOTES 25
REFERENCES 26
4 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
Introduction
W
ith so much tragedy to absorb over the past six
months, it has been easy to overlook an emerging
crisis. While our attention has rightly been
occupied by the mounting death toll and record unemployment
rate, millions of America’s small business owners have been
facing the greatest threat to their survival in living memory.
But they are not in a silo. ese businesses employ almost half
of all Americans. eir fate will aect how well our economy
bounces back when the pandemic subsides.
e impact so far has been catastrophic. During the rst two
months of the crisis, from mid-March to mid-May 2020, more
than 40 percent of all small businesses were closed. ese
closures caused revenues to plummet, down 40 percent on
average. In leisure and hospitality revenues were down more
than 70percent. Businesses drew on what little cash they had
to stay aoat. But by June, just three months into the crisis,
more than 400,000 small businesses had already permanently
closed—more than typically close in an entire year.
is recession is highly unusual. e scale and speed of the
contraction are, of course, unprecedented. But more critical
is the unevenness of its eects. Some businesses were forced
to close, or lost customers who chose to stay home. Some were
spared from the immediate eects, while others experienced
a surge in demand.
In an ordinary recession, the textbook approach is to pump
consumers’ wallets full of cash and rely on the economy’s
plumbing to get it to the businesses and workers in need.
But the pandemic has shut down much of that plumbing. No
amount of household stimulus is going to open a bar that local
authorities shut down, get its bartenders back to work, or pay
its rent. And as those businesses and workers lose income,
the contraction spills over onto the businesses and workers
spared from the rst-round eects.
It is possible to arrest this downward spiral, but doing so
requires scal support of the small businesses aected. e
case for acting is clear. is was an uninsurable risk for
small businesses with eects that are highly uneven. Many
small businesses lack the access to credit that would help
them bridge the crisis. When large businesses fail, they tend
to proceed through an orderly reorganization with much of
their capital preserved. When small businesses fail, they tend
to dissolve.
A large volume of simultaneous business failures constitutes
a systemic risk. e load would swamp bankruptcy processes,
and add to an already over-burdened unemployment
insurance (UI) system. And the businesses themselves
represent tremendous value that would be lost. Much of their
capital is intangible, and thus nontransferable. e rm-
specic human capital, the matches between businesses and
their workers, suppliers, and customers—all would be lost.
Ending a business is far easier than starting one—a loss of
businesses on a large scale would have a scarring eect that
would slow the recovery.
And this is among otherwise-viable businesses that, with
the support of insurance and adequate capital, would have
been spared from such a fate. We should of course be wary
of propping up otherwise-unviable businesses. e revenue
required to support small businesses is not free—it comes at
some cost to our future prosperity. And the longer the period
of depressed activity goes on, the less generous the level of
support should be. All of these considerations suggest some
restraint.
e support provided to date has had mixed success. In late
March 2020 Congress and the White House authorized the
$350 billion Paycheck Protection Program (PPP). By any
normal standard, that is a lot of money. But it was insucient
to cover the demand for the program, and an additional
$310billion had to be authorized just a few weeks later. And
in total, that paid for a program that gave too much to those
that did not need it, and not enough to those that did.
As problematic as the PPP and its rollout were, it succeeded
in undergirding many small businesses. It replenished cash
reserves. It brought condence. Early estimates indicate it
saved at least 2.3 million jobs for several months, with the
nal number likely to be far higher. But it was a program
designed in a more optimistic time—back when we thought
we could freeze every small business in America to buy time
to suppress the virus, and then thaw them all out again as we
resumed normal life.
The Hamilton Project • Brookings 5
But that is not what happened. Instead, it looks as though we
will be living with the virus for some time. Small business
revenues have plateaued substantially below precrisis levels.
Many businesses are operating at much-reduced capacity,
while facing higher costs in adapting to life with the virus.
e single-most-eective measure to help small businesses
remains suppressing the virus. ere is broad agreement
among public health experts on how to do that. Every other
advanced country suppressed the virus at some point—we
can too, if we choose.
Another critical measure to support small businesses is
further broad stimulus. As the crisis has spread from the
directly aected businesses to the broader economy, broad
stimulus measures gain potency. Generous supplementary UI
support, signicant funding for state and local governments,
and household cash stimulus all would boost demand,
providing indirect support to many of the small businesses
still operating.
As for direct support, the optimal policy today is very
dierent from what it was in mid-March. It is feasible to fully
cover businesses’ revenue shortfalls for the duration of a short
lockdown. But doing so for a year or more until we roll out a
vaccine would be imprudent. e longer the support must last,
the less generous it must be. Unfortunately, this will mean not
every otherwise-viable business will be saved. But with more
tightly targeted support, the businesses that remain will have
a ghting chance.
Accordingly, the PPP should not be extended. In its place, the
Employee Retention Credit (ERC), a refundable credit against
the employer’s payroll tax obligations, should be signicantly
expanded. e credit would apply for three quarters starting
October 1, 2020. Any small business that has experienced a
30percent year-on-year decline in revenues in a given quarter
would be eligible. e credit would cover 80 percent of all
wages up to $15,000 per employee per quarter. In addition,
eligible businesses would receive a Small Business Survival
Credit (SBSC), which would provide $5,000 per employee up
to a maximum of $50,000 per business per quarter to cover
non-payroll expenses such as rent, interest, utilities, and
COVID-19 mitigation costs.
By comparison, the PPP covered 100 percent of all wages
up to the equivalent of $25,000 per quarter, with additional
support for non-payroll costs. But that support lasted only
two months and was granted to almost three-quarters of
all businesses. e support proposed here would be less
generous, as demanded by a longer period of subdued activity,
and it would apply to only the worst-aected businesses. But
it would last almost four times longer, it would be easier to
administer, and it would draw in and drop o businesses as
local conditions change.
Many of the problems with the PPP arose from funneling
the money through private banks. is is another respect
in which the United States is an international outlier. In
other countries, wage supports for small businesses were
administered by their tax authorities. In the United States,
we relied on private banks because we wanted to provide
immediate liquidity during a crisis. ere is no good reason
why we should not be able to rely on the IRS to ll that role in
future crises: it is the largest nancial institution in the world
and carries out trillions of dollars in transactions each year
with hundreds of millions of counterparties.
e IRS has been starved of funding for decades, diminishing
even its core functions. It lacks the systems necessary to
implement a program like the PPP on short notice. Tax
authorities in many other countries have such systems. is
is yet another example of America’s moribund state capacity
being laid bare by the crisis. To better prepare for future
crises—indeed, to help the IRS perform its core functions even
in normal times—we should provide signicant, sustained
additional funding to the IRS. In particular, this should
include funding for a real-time electronic payroll reporting
system covering every American business and worker.
e rst round of stimulus was an act of uncharacteristic
bipartisanship, and one that was remarkably eective in
helping to safeguard the livelihoods of millions of Americans.
Now that those initial measures have run their course, it is
past time for us to act again. Both the House and the Senate
have passed bills that contain commendable elements. ere
is much common ground on small business support in
particular. is proposal improves on those plans, exhibiting
good qualities of both—generous support for payroll and
adaptation costs, and strong hiring incentives—while adding
the support for non-payroll costs that many businesses have
called for.
6 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
Background: Before the Crisis, Small Businesses Were
Financially Fragile
T
here are around 6million small business employers in
America, together responsible for more than 60million
jobs (US Census Bureau 2020a).
1
ese small businesses
make up more than 99percent of all businesses, but account for
47percent of employment due to their smaller size. is policy
proposal does not consider the tens of millions of nonemployer
small businesses such as the self-employed. ey are best served
by expanded UI, which this proposal recommends extending.
Contrary to some commentary, the sectors most exposed
to COVID-19 are not served disproportionately by small
businesses.
2
If anything, the opposite is true. While the
overwhelming majority of businesses in the aected sectors
are indeed small businesses, that is also true for the economy
generally. Among big businesses, 63 percent serve the
directly aected sectors, while among small businesses, only
46percent do.
e special focus on small businesses is not because they
are more exposed to COVID-19, but rather because they are
much more nancially fragile and therefore vulnerable to a
large and sustained hit to their protability. Small businesses
lack the access to capital markets of large businesses, and
are much more likely than larger businesses to be dissolved
instead of reorganized under bankruptcy.
Even in normal times, many small businesses face perilous
conditions. While on net there is typically small business
creation, this masks substantial turnover. Around a third of
small businesses in US cities are unprotable at any given
time (Farrell, Wheat, and Grandet 2019), and around a third
do not survive beyond the rst four years (Farrell, Wheat, and
Mac 2018).
Many small businesses have only limited access to credit and
very little cash on hand to nance unexpected losses. Around
half of small businesses in US cities have two weeks or less of
cash on hand (Farrell, Wheat, and Grandet 2019). is diers
considerably by the race of the business owner, with White-
owned businesses having 19 days of cash on hand on average,
compared to just 12 days for Black-owned businesses (Farrell,
Wheat and Mac 2020).
FIGURE 1.
Total Number of Small Businesses, 1988–2017
Source: U.S. Census Bureau 2020e.
Note: Shading corresponds to periods of disrupted business growth coinciding with recessions.
0
7
1988 1992 1996 2000 2004 2008 2012 2016
Count (in millions)
6
5
4
3
2
1
The Hamilton Project • Brookings 7
Fiy-six percent of all small businesses have relied on funds
from their personal savings, friends, or family to support
operations in the past ve years, and 47 percent say they
would rely on personal funds if they needed to ll a two-
month gap in revenues (Federal Reserve System 2020).
Eighty-eight percent of small business owners rely on their
personal credit score to secure nancing, and only 44percent
of small businesses have obtained funds from a bank in the
past ve years.
In an economic crisis this fragility has economic
consequences. During the Great Recession there was a net
loss of 6 percent (around 375,000) of all small businesses
(see gure 1) (US Census Bureau 2020e). At the same time,
nancial constraints are responsible for having reduced
employment growth among small businesses by 4 to 8
percentage points relative to large businesses (Siemer 2019).
e sharpness and scale of the COVID-19 crisis makes it
dicult to draw lessons from past experience about how small
businesses might weather this crisis. However, the resilience
of small businesses during natural disasters provides at
least some context for their ability to withstand a sharp and
systemic but temporary shock to revenues.
In 2017 Hurricanes Harvey and Irma hit Houston and Miami.
e storms caused reductions in cash inows of 63 percent
and 82percent for at least half of all small businesses, but most
recovered within one to two weeks (Farrell and Wheat 2018).
ese cash shortfalls were substantially, but not completely,
oset by reductions in cash outows of 54 percent and
62percent, with cash balances on net falling by 7.5 percent
and 7.4percent at the height of the disaster.
ough severe, the temporary nature of these disasters meant
cash ows for most small businesses returned to normal
within one to two weeks. And because the reduction in cash
outows lasted around a week longer than the reduction in
inows, most businesses exited the crisis with more cash on
hand than they entered it with.
One lesson from those hurricanes is that many businesses
can sustain a very large reduction in revenues provided it is
short-lived. e exibility that rms have in cutting variable
costs can shield them substantially from revenue shortfalls,
and modest net losses can be weathered by drawing on
cash reserves if the disruption lasts only a matter of weeks.
e initial impact of the COVID-19 crisis on small business
revenues is similar in magnitude to that of a natural disaster.
But rather than bouncing back in one to two weeks, revenues
were well down for at least three months and even now
remain substantially below their pre-crisis levels on average
(see gure 2).
Source: Womply 2020.
Note: Percent changes in revenue are indexed to January 10 revenue.
-75
-60
-45
-30
-15
0
15
30
Jan Feb Mar Apr May Jun Jul
Percent change in revenue
Retail and
transportation
Total
Education and
health
Leisure and
hospitality
PPP round 1
PPP exhausted
PPP round 2
CARES Act
Lockdowns
FIGURE 2.
Change in Small Business Revenue for Selected Industries, January–July
8 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
The Challenge: COVID-19 Is an Existential Threat to
Many Small Businesses
O
ver the six months since the crisis began, a remarkable
volume of real-time data on the state of small
businesses in America has emerged. e US Census
Bureau has produced a new weekly survey gauging the
experiences of more than 20,000 small businesses, as well as a
number of other high-frequency data sets. Several academics
have rolled out surveys gauging small business experiences,
and a range of nancial technology rms with access to small
business data have made these available to researchers. Many
of these data overlap; some cover certain time periods and
not others. Once collated, a coherent story emerges about the
experience of small businesses during the COVID-19 crisis.
On March 16, 2020, the day before the rst lockdowns began,
11percent of small businesses had already closed (Waldman
2020). e following day, closures rose to 20 percent. By
late March, more than 40 percent of small businesses were
closed (Bartik, Bertrand, Cullen, Glaeser, Luca, and Stanton
2020). A month later, small business closures remained
above 40 percent (US Census Bureau 2020d). Including
nonemployer businesses, the number of active business
owners had fallen 22 percent by April, the largest drop on
record (Fairlie 2020). e fall was most extreme among Black
business owners, down 41percent, because the industries in
which those business owners were more likely to operate were
those hit the hardest by the pandemic.
Although the closures were more prevalent in the areas hit
worse by the virus, no region was spared (see gure 3). In late
March 54percent of small businesses were closed in the Mid-
Atlantic region (the most aected area), but 39percent were
closed even in the Mountain region (the least-aected area)
(Bartik, Bertrand, Cullen, Glaeser, Luca, and Stanton 2020).
is was still the case for both regions a month later (US
Census Bureau 2020d).
ese widespread closures led to widespread revenue losses.
At the end of March small business revenues were already
down more than 40 percent (see gure 2) (Womply 2020).
e drop was most extreme in the leisure and hospitality
FIGURE 3A.
Share of Small Businesses Experiencing
a Temporary Closure Last Week by State,
May 2
Source: U.S. Census Bureau 2020d.
FIGURE 3B.
Share of Small Businesses Experiencing
a Temporary Closure Last Week by State,
June 27
Source: U.S. Census Bureau 2020d.
Share of small businesses experiencing temporary closures (percent)
10 to 25 25 to 35 35 to 50 50 to 75
Share of small businesses experiencing temporary closures (percent)
Less than 10 10 to 25 25 to 35 35 to 50
The Hamilton Project • Brookings 9
industries, where revenues were down almost 70percent. At
the end of April, around six weeks aer the lockdowns began,
74percent of small businesses reported revenues were down
(US Census Bureau 2020d), and by more than 30percent on
average (Wompley 2020). In hospitality and leisure, revenues
were still down by almost 60percent.
ese revenue losses substantially depleted cash reserves
through April. In late March, around a fortnight aer the rst
lockdowns began, 25 percent of small businesses reported
having insucient cash on hand to cover more than a month
of expenses, while 53percent had only between one and two
months’ worth of cash on hand (Bartik, Bertrand, Cullen,
Glaeser, Luca, and Stanton 2020). A month later, 41percent
of rms had less than a month’s of cash on hand, and only
29percent had between one and two months’ of cash on hand
(US Census Bureau 2020d).
e closures also resulted in widespread layos. Between
March 28 and April 20, 65percent of small businesses laid o
at least one worker (Humphries et al. 2020b). In mid-April,
on net, small business employment was down by around
60percent (Homebase 2020). At the end of April, 28percent
of small businesses were continuing to lay o workers (US
Census Bureau 2020d), with small business employment still
down by more than 50percent on net (Homebase 2020).
Based on all of these data, it is clear the COVID-19 crisis poses
the greatest existential threat to American small businesses in
memory. As of August, more than 18percent of all U.S. small
businesses—and more than 27percent of those in leisure and
hospitality—remained closed. e critical question for the
medium-term prospects for the US economy and the path of
further scal responses is just how many will remain closed
forever.
GOVERNMENT SHOULD SUPPORT SMALL
BUSINESSES WHERE MARKETS CANNOT
In an ordinary recession, broad cash stimulus can be
dispatched to arrest the vicious circle of falling demand
causing layos causing falling demand, and so on. If the
economy’s plumbing is functional, that cash can ow to
where it is needed through the ordinary course of trade and
commerce. e eectiveness of broad cash stimulus was
a valuable lesson coming out of the Great Recession. In a
standard recession, the justication for direct government
support for businesses is lessened by the fact that cash
stimulus serves to support businesses and their employees
indirectly.
e COVID-19 crisis has produced a very dierent kind
of recession. In this environment, the standard suites of
economic analysis and policy tools have been found wanting
(Hamilton and Veuger 2020a). Critically, the pandemic has
clogged the economy’s plumbing. e rst-round eects of
social distancing (both government-enforced and voluntary)
on economic activity in certain sectors cannot be oset by
scal stimulus—that activity is impossible. Some sectors are
not aected directly, while others in fact have experienced a
surge in demand, causing shortages and price rises, and an
increase in employment.
But for those rms initially spared from the demand
contraction, the reprieve is short-lived. As the directly
aected workers and business owners lose income, they
reduce spending on goods and services across the whole
economy, including in sectors not directly aected. Workers
are then shed in those sectors, too, and on the vicious circle
goes (Guerrieri et al. 2020). Just as the virus is passed from
person to person, so too the economic contagion spreads
from sector to sector. In the end, none is spared.
Because the contractions in demand are unevenly spread
across the economy, conventional broad cash stimulus will
nd its way back to some businesses and workers but not
to others. e clogged plumbing limits the ability of cash
stimulus to arrest the downward spiral. Unlike in a normal
recession, the only way to help directly aected businesses
and their employees is with direct support.
Another unusual feature of this crisis is the very sharp but
temporary nature of the economic contraction. e rst phase
of full-scale lockdowns lasted less than three months. ere
have been renewed restrictions in some cities, but these too
will li eventually. Given the temporary nature of the crisis,
there is no clear reason why aer the crisis much of the supply
side of the economy could not in principle return to its pre-
crisis state. Many businesses that were viable before should be
viable aer.
If private insurance to cover such a signicant reduction in
revenue were available, business owners would have been
able to purchase such insurance, allowing them to bridge
the crisis. Insurance coverage would have subsidized those
businesses unlucky enough to have been adversely aected
by a once-in-a-century pandemic, the eects of which were
dicult to anticipate. In that case, signicant government
support would be less defensible. However, no such insurance
was available.
3
is lack of coverage prevents pandemic risk
from being disbursed throughout the economy. Many of these
same issues arise in insuring against natural disasters, but
the pandemic is like a natural disaster occurring across the
country for months on end.
Government-funded loans would help some rms, in
particular those suering as a result of limited access to
credit, but for many they would not be enough. Millions of
small businesses have taken a signicant hit to their net
worth, which will render many unviable. ese businesses
will rightly deem the resultant debt too great a burden to
carry forward.
10 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
If these were large businesses, the equity holders would be
dissolved, with the remaining assets reorganized under new
ownership. Many smaller rms, on the other hand, would
simply disappear. ey might otherwise have grown into
larger businesses, spurred innovation, and contributed to
job and productivity growth (Decker et al. 2014). e lack of
private insurance to cover these losses calls for the provision
of social insurance that at least partly disperses them across
the economy and over time.
Without a subsidy, the destruction of capital resulting from
an economic contraction of this size and duration would
be unprecedented. Much of this capital is rm-specic and
thus nontransferable. e matches between rms and their
customers, suppliers, and employees would be dissolved.
Many unique products would vanish. Much of the learning-
by-doing that is specic to each business would disappear. If
we believe the value of all of this capital at stake exceeds the
economic cost of the taxes and subsidies necessary to save it,
then we should provide the necessary scal support.
If provided in a way that encourages businesses to retain their
workers (Bishop and Bartik 2009), business support can have
strong macroeconomic benets too. Discouraging directly
aected businesses from shedding workers, and encouraging
them to maintain wages, would help arrest the downward
spiral that would precipitate a very deep and long recession.
ose workers would also take pressure o the UI system. In
preserving the productive capacity of the economy, it would
ensure a speedier transition to a steeper long-run growth
trajectory. e businesses lost during the Great Recession, for
example, le a persistent dent in employment (Sedláček 2020).
ere will no doubt be some permanent changes in demand
that will necessitate permanent changes in supply. Some
businesses that were unviable before the crisis will be pushed
over the edge. Such Schumpeterian creative destruction is
one of the few silver linings of a recession. e nature of the
crisis today means scal policy will not be able to save every
small business from failure. In March it was reasonable to
believe that most rms could be saved by very generous but
sharply temporary support, formulated on an expectation of
an eective public health response. But with the virus still
spreading six months later, we must be realistic. e length
of the crisis suggests some restraint in support for the aected
businesses.
THE PAYCHECK PROTECTION PROGRAM
In mid-March, in light of the impending calamity, Congress
and the White House formulated the $2.2trillion Coronavirus
Aid, Relief, and Economic Security (CARES) Act, which
provided three planks of scal support: broad cash stimulus
to households, expanded UI, and direct support to businesses.
e CARES Act was signed into law on March 27, roughly
two weeks aer the rst lockdowns began.
e most prominent form of business support was the PPP,
a subsidized small business loan program. e PPP was a
bipartisan initiative to give small businesses relief through the
crisis in exchange for retaining their workers and maintaining
payroll. e program launched on April 3 with $349 billion in
initial funding.
e PPP was implemented by the Small Business
Administration (SBA). It applied to small businesses with 500
or fewer employees, sole proprietors, independent contractors,
self-employed persons, nonprots, veterans’ organizations,
and tribal businesses that had been in operation on February
15.
4
Businesses for which an owner was on probation or
parole, had been convicted of a felony within the past ve
years, or was an undocumented alien were ineligible.
5
To
be eligible, an applicant had to certify “that the uncertainty
of current economic conditions makes necessary the loan
request to support the ongoing operations of the eligible
recipient” (CARES Act 2020).
Under the program, the SBA guaranteed loans made by
banks and other nancial institutions to eligible recipients.
e loan amount was limited to two and a half months of
the recipient’s average prior-year payroll costs (excluding any
annual per employee compensation in excess of $100,000),
capped at $10 million.
6
e loans had a term of ve years
and an interest rate of 1percent.
7
Applicants did not have to
provide collateral, and the loans were non-recourse.
e key feature of the PPP was that, under certain conditions,
the loans would be forgiven entirely.
8
e recipient had to
spend at least 60percent of the loans on payroll costs,
9
and the
remainder on only interest, rent, and utilities, all over a 24-
week period.
10
e proportion of the loans forgiven was equal
to the number of full-time equivalent employees on payroll
during the 24 weeks aer the loan proceeds were disbursed
as a proportion of those on payroll in 2019. If a business
maintained full-time equivalent hours, 100 percent of the
loan amount was forgivable. Allowances in forgiveness were
given if a business faced diculty rehiring or hiring. Any
salary reductions in excess of 25percent were deducted from
the forgiven amount.
e initial funding allocation of $349 billion was widely
understood to be inadequate to meet the needs of the program.
Two and a half months of payroll for all of America’s small
businesses totals more than $500billion (US Census Burear
2020a), and the program also applied to a range of larger
businesses in certain industries.
On April 16, less than two weeks aer the program
commenced, the initial funding allocation was exhausted. In
response, on April 24 an additional $310billion in funding
was added, which became available to applicants from April
27. e program was amended again on June 5 in response to
criticisms of the loans’ lack of exibility.
The Hamilton Project • Brookings 11
AN ASSESSMENT OF THE DESIGN OF THE PAYCHECK
PROTECTION PROGRAM
e COVID-19 economic crisis has a single source: a
temporary fall in business revenues caused by a contraction
in demand, due initially to government-imposed lockdowns
and voluntary social distancing. e impact varies widely
across aected businesses. And the aected businesses have
a variety of cost structures. e problem is that revenues are
too low, not that payroll costs are too high. Because of this,
a subsidy equal to payroll inevitably oversubsidizes some
businesses and under-subsidizes others. An alternative option
would have been to cover revenue losses directly (Hamilton
and Veuger 2020b).
Nevertheless, many countries, including Australia, the
Netherlands, New Zealand, the United Kingdom, and the
United States, tied business support to payroll. at is not
ideal, but it still has the potential to help many businesses,
and to do so in a way that encourages them to maintain links
to their workers. It preserves the stock of businesses and
worker–rm matches, which in turn preserves the productive
capacity of the economy during the recovery phase. It also
provides indirect support to millions of workers, which would
spare them from an already overburdened UI system.
A drawback of tying the subsidy to payroll is that the support
will inevitably be insucient to keep some businesses aoat.
Many businesses were operating at reduced capacity so they
did not need to maintain payroll, but their ability to use the
loans to defray other costs was limited. Some businesses
could scale back to reduce their variable costs, but still faced
large, unavoidable xed costs like rent. For businesses with
minimal sta, high xed costs, and low margins, the program
would have been of limited use.
Another drawback is an almost complete lack of targeting.
While businesses had to declare in good faith that they
required the support to maintain operations, this was a vague
declaration and dicult to enforce. As a result, the PPP is
likely to have made some recipients more protable during
the crisis than before it. To the extent those businesses might
otherwise have laid o workers, the program will still have
served a purpose. However, the lack of targeting would have
been straightforward to address with an eligibility threshold
tied to public health orders or revenue losses.
e United States is the only country in the world to
implement a payroll subsidy via banks and nancial
institutions. In other countries support has been provided
via the tax system. In Australia, for example, businesses
experiencing revenue declines of more than 30 percent
receive a per employee subsidy of around $500 per week for
six months (Hamilton 2020). e program relied on banks
because the IRS is not capable of paying out large amounts of
money on very short notice to millions of businesses on the
basis of certain criteria. Operating via the banks avoided the
liquidity shortfall that would have resulted from any delay.
THE TROUBLED ROLLOUT OF THE PAYCHECK
PROTECTION PROGRAM
e initial rollout of the PPP was plagued with problems
(Morrell et al. 2020). e SBA, which in 2019 facilitated just
$28billion in loans, was asked to expand more than tenfold
in a matter of weeks. JPMorgan Chase, the largest US bank,
initially indicated it would delay its launch because it had
not received the necessary guidance from the Treasury
Department. Bank of America, the second-largest US bank,
initially said it would provide PPP loans only to its existing
customers. Other big lenders such as Wells Fargo, Citigroup,
and PNC delayed their launches.
Despite these initial hiccups, over the 13 days following the
April 3 launch the SBA processed 1.7million loans via 4,975
lenders (SBA 2020a). e average loan amount was $206,000,
and 74percent of the loans were for $150,000 or less. However,
businesses in the leisure and hospitality industries (i.e.,
accommodation and food services), seemingly the industries
most adversely aected by the crisis, received 9percent of all
loans, while 13 percent went to businesses in construction;
13percent to professional, scientic, and technical services;
and 12percent to manufacturing.
Moreover, the entire $349 billion initial funding allocation
was exhausted in less than two weeks. Predictably, it proved
wildly inadequate, leaving more than 2 million small
businesses hanging (US Census Bureau 2020d). And the
access to rst-round funding was strongly related to size (see
gures 4a and 4b). As the rst round was exhausted, almost
three-quarters of the businesses with more than 100 workers
that would eventually receive funding had received it. By
contrast, fewer than a quarter of those with four or fewer
workers had received funding, and fewer than half of those
with between ve and twenty workers had received funding.
Overall, when the rst-round funding ran out, 75 percent
of small businesses had requested PPP funding, and only
38percent had received it.
At least in the rst round, funds did not ow on the basis
of need. Among those worst aected, the proportion of
applicants denied or still waiting for approval was more
than double that among those unaected (Bartik, Bertrand,
Cullen, Glaeser, Luca, Stanton, and Sunderam 2020).
Businesses with more cash on hand were more likely to be
approved. And areas that experienced greater declines in
hours worked and more business closures in fact received
fewer PPP loans (Granja et al. 2020).
12 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
With the rst round of funding so limited, frictions were
critical to the rationing process. As intermediaries, the banks
played the role of gatekeeper. e intensity of PPP lending
varied widely among banks. If a business was lucky enough
to be located near a bank processing a high volume of PPP
loans relative to other kinds of loans, it was much more likely
to obtain a loan (Granja et al. 2020). Having a preexisting
loan with a bank raised the probability of being approved
by 4.4 percent (Bartik, Cullen, Glaeser, Luca, Stanton, and
Sunderam 2020).
Firm size was also an important factor. On March 28, the
day aer the CARES Act was passed, businesses with nine
or fewer employees were much less likely to know about the
PPP than those with 10–50 employees (Humphries, Neilson,
and Ulyssea 2020a). By April 5, two days aer applications
opened, awareness among businesses with ve to nine and
a half employees had rapidly increased. Among businesses
with four or fewer employees, awareness had increased only
modestly, remaining below 80percent through April 16 when
rst-round funding was exhausted. Smaller businesses were
then much less likely to apply for the PPP, they applied later,
they waited longer to be approved, and they were less likely to
be approved.
Some small businesses simply were not interested in the
PPP. Twenty-eight percent indicated they would not accept
a PPP loan if it were oered to them, despite the generous
terms (Bartik, Bertrand, Cullen, Glaeser, Luca, and Stanton
2020). irty-ve percent of those who would refuse a loan
said they did not need the cash, 30percent said they did not
think they would qualify, 19percent said they did not trust
the government to forgive the debt, and 11percent thought it
would be too much hassle.
FIGURE 4A.
Total Number of Approved PPP Loans and Total
Number of Small Businesses, by Firm Size
Source: U.S. Department of the Treasury 2020; U.S. Census Bureau 2020a.
Note: Firm size for the PPP loans is based on how many employees an applicant indicated it
would retain under the program.
FIGURE 4B.
Distribution of Approved PPP Loans, by
Employment and Approval Date
Source: U.S. Department of the Treasury 2020.
Note: Firm size for the PPP loans is based on how many employees an
applicant indicated it would retain under the program.
Having exhausted the rst round of funding on April 16, the
second round became available on April 27. In its rst week,
loans were disbursed to more than a million small businesses
(US Census Bureau 2020d). Funding continued to roll out
rapidly over the following two weeks. e second round
went to much smaller businesses, with an average loan size
of $112,000, around half that in round 1 (see gure 4) (SBA
2020b). To date, 72 percent of small businesses—or around
4.5million businesses—have received a total of $512billion
in funding under the PPP. Less than 3 percent of small
businesses that applied were not approved. In the end, around
$130billion in funds remained unallocated.
ere was some controversy about large public companies
receiving funding. Under public pressure, Shake Shack
returned the $10million it had received under the program.
Following the public discontent, the Treasury Department
released guidance advising that public companies receiving
funding under the PPP were likely to have violated their good
faith declaration of need and would be penalized if found to
have improperly accessed the program. In reality, only 424
public rms accessed the PPP across both rounds, receiving
a total of $1.4billion in funding through July 15 (Cororaton
and Rosen 2020). Despite the public outrage, this constituted
just 0.2percent of funds disbursed.
THE EFFICACY OF THE PAYCHECK PROTECTION
PROGRAM
It is still too early to comprehensively assess how many
businesses and jobs were saved by the PPP. But the evidence
to date is positive. On being told about the PPP ahead of
its rollout, small businesses responded that they would lay
o only 6 percent of their employees by December rather
1−4 5−9 10−19 20−49 50−99 100−249 250−500
0
500
1,000
1,500
2,000
2,500
3,000
Employees per Firm
Number of firms (in thousands)
PPP round 2
PPP round 1
Total rms
0
25
50
75
100
Cumulative share of loans (percent)
PPP round 1
PPP round 1
exhausted
PPP round 2
Apr May Jun
Jul
1-4 employees
5-19 employees
20-99 employees
100-500 employees
The Hamilton Project • Brookings 13
than 40 percent without the PPP (Bartik, Bertrand, Cullen,
Glaeser, Luca, and Stanton 2020). Learning of the PPP also
led them to increase their expected probability of being
open in December from 62percent to 85percent. In a survey
conducted aer the rst round of funding but before the
second, receiving funding increased a business’ self-reported
probability of survival by 14–30 percentage points (Bartik,
Cullen, Glaeser, Luca, Stanton, and Sunderam 2020).
Only a few studies to date have considered the eect of the
PPP on actual outcomes. e most compelling study, relying
on high-quality, representative ADP payroll data covering
26 million workers, nds that the PPP saved 2.3 million
jobs through the rst week of June (Autor et al. 2020).
11
is
implies a scal cost of $224,000 per job directly supported for
those months. Because the true number of jobs saved is likely
to be much higher, the true scal cost is likely much lower.
12
ere is clear evidence that the rollout of the second funding
round substantially improved the cash holdings of small
businesses. rough the rst three weeks of May, as second-
round funding was being disbursed, an additional 31percent
of small businesses received funding (see gure 5) (US Census
Bureau 2020d). Over those same three weeks, an additional
16 percent of small businesses had more than a month’s
worth of cash on hand. Four percent fewer rms had no cash
on hand, while 11 percent fewer had less than two weeks of
cash. Meanwhile, 6percent fewer businesses reported missing
a loan payment and 7 percent fewer reported missing other
payments.
During the rollout of the second round, the self-reported
outlook of small businesses was deteriorating substantially.
As the rollout completed, the outlook stabilized. Limiting the
rst round of funding to $349billion withheld funding from
around 2million of the smallest businesses for weeks. By the
time the funding arrived, the initial lockdowns had been
going for six weeks. Small businesses’ limited cash holdings
and access to credit is well documented. Many will not have
been able to bridge that gap.
Moreover, while the PPP replenished small business cash
holdings drawn down during the lockdowns, the support
was temporary. e program was reformed to allow rms
more exibility in using the loans, and that is welcome. In
particular, the SBA will be more lenient in forgiving loans
where small businesses have faced diculty in rehiring
workers. And the previous Treasury guideline requiring rms
to spend at least 75percent of the money on payroll has been
loosened to 60percent. Businesses are now allowed to spread
their loans over a much longer period.
But, ultimately, the subsidy each business received was
limited. As the PPP was being designed, many policymakers
did not expect the lockdowns to last as long as they did and
public health capacity was expected to be built so the virus
could be suppressed as the lockdowns lied. is was the
experience in many countries hard-hit by the virus, including
Italy, Spain, and the United Kingdom. In the United States,
the lockdowns went on for longer than many expected and
the environment businesses have returned to is far worse than
many expected. As the lockdowns have lied, there have been
renewed outbreaks across large swathes of the country. While
the PPP will have helped many businesses get by at the height
of the initial lockdowns, the support it provided will not have
been nearly enough nor for long enough.
FIGURE 5A .
Distribution of Small Businesses’
Description of Their Current Cash
Holdings, May–September
Source: U.S. Census Bureau 2020d.
Note: These are responses from small business owners to the question, “How would you
describe the current availability of cash on hand, including any financial assistance or loans?”
FIGURE 5B.
Distribution of Small Businesses’ Expectations
Regarding When Their Business Will Return
to Normal, May–September
Source: U.S. Census Bureau 2020d.
Note: These are responses from small business owners to the question, “In your opin-
ion, how much time do you think will pass before this business returns to its normal?”
0
25
50
75
100
Percent of respondents
Under 2 weeks
3-4 weeks
Over 1 month
None
0
25
50
75
100
Percent of respondents
Never
More than 6 months
1 month or less
2-6 months
14 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
OTHER SMALL BUSINESS MEASURES
While the PPP was by far the highest prole support for
small businesses, a range of other programs were available. A
little-known alternative to the PPP was the ERC, which was
unavailable to small businesses that opted for the PPP. e
ERC is a refundable tax credit equal to 50percent of qualied
wages up to $10,000 per employee paid between March 12 and
the end of the calendar year. To be eligible, a business must
either have been ordered to shut down or have experienced at
least a 50percent decline in revenues.
To receive funds immediately, rms could draw on their
federal tax withholdings. Where these were insucient to
fund the eligible credit, employers could apply to the IRS for
an advance. is program applied to businesses of all sizes,
but rms with more than 100 employees received the ERC
only against the wages of workers not currently working.
ose with 100 or fewer employees were allowed to receive the
ERC also for those still working.
e ERC provided a maximum of $5,000 per employee
versus a maximum of $20,833 per employee under the PPP.
is modesty will have limited its impact on employment
and rm viability. But some small businesses may have
preferred the simplicity and speed of delivery of the ERC
relative to the PPP, particularly those without an established
banking relationship, and the ERC is not subject to some of
the eligibility criteria that might have excluded some small
businesses from the PPP. ere is not yet any available
evidence on the ecacy of the ERC. But to date fewer than
0.4percent of small businesses have received assistance under
the program (US Census Bureau 2020d).
e Economic Injury Disaster Loan (EIDL) program,
administered by the SBA, is typically used to provide liquidity
to small businesses aected by natural disasters. Following the
president’s COVID-19 emergency declaration on March 13,
EIDLs were extended to small businesses adversely aected
by the pandemic. e loans may be used by small businesses
to pay xed debts, payroll, accounts payable, and other bills
they cannot pay because of the disaster. e interest rate is
3.75 percent with available terms up to 30 years. To date,
22 percent of small businesses have received an EIDL (US
Census Bureau 2020d), suggesting this has been an important
source of liquidity for small businesses during the crisis.
e CARES Act also set aside $10 billion to fund an
immediate $10,000 advance to small businesses applying for
an EIDL, which they would not have to repay. e advance
would be received within three days of applying for an EIDL,
and the EIDL would not have to be approved in order for the
advance to be paid. e amount of the advance would then
be deducted from any loan amount approved under the
program.
Businesses also received a payroll tax deferral. e employer’s
share of Social Security tax contributions (6.2 percent of
wages up to $137,700 per year) on wages paid during March
27–December 31 could be deferred, with half to be paid by
December 31, 2020, and the other half by December 31, 2021.
is is eectively over $140 billion of interest-free loans of
$22,000 per business on average, or $2,200 per employee.
e Federal Reserve has taken a range of actions to support
small business liquidity during the crisis. It introduced the
PPP Liquidity Facility extending credit to eligible nancial
institutions originating PPP loans, taking the loans as
collateral at face value. As of August 5, 2020, the Federal
Reserve held around $70billion of these loans on its balance
sheet (Federal Reserve Board 2020).
e Federal Reserve also introduced the Main Street Lending
Program, which provided ve-year loans to small and mid-
sized businesses with up to 15,000 employees.
13
Interest is
deferred for a year and repayment of the principal is deferred
for two years. e interest rate is around 3.2 percent.
14
Loans may be between $250,000 and $300 million.
15
Banks
retain 5percent of the value of the loans, selling the rest to
the Federal Reserve, which has agreed to purchase up to
$600billion of the loans. Under the CARES Act, the Treasury
Department provided $75billion in equity to cover potential
losses. As of August 5, 2020, the Federal Reserve held only
around $38billion of these loans on its balance sheet (Federal
Reserve Board 2020), and only 0.2percent of small businesses
report having received a loan under the program (US Census
Bureau 2020d).
DESPITE SUPPORT, SMALL BUSINESSES HAVE BEEN
DECIMATED, AND THE OUTLOOK IS BLEAK
By the end of March, a fortnight or so aer the rst lockdowns
began and right aer the CARES Act was passed, 1.8percent
of small businesses had already permanently closed due to
COVID-19 (Bartik, Bertrand, Cullen, Glaeser, Luca, and
Stanton 2020). By June 15 that had risen to 6.8percent, and by
July 10 to 7.1percent, or more than 420,000 small businesses.
16
If these businesses are representative of national employment,
this means we have lost at least 4million jobs that will only
return with the creation of new businesses. e situation is
particularly bleak in certain industries. As of July 10, more
than 57,000 restaurants (more than 13percent of restaurants
nationally), employing roughly 1.4 million workers, had
already permanently closed. Another 42,000 restaurants
remained at least temporarily closed.
In normal times, there is typically signicant turnover among
small businesses. From 2012 to 2014, aer rm destruction
during the Great Recession had stabilized, around 380,000
small businesses closed each year (US Census Bureau 2020c).
is is consistent with the long-run average going back
The Hamilton Project • Brookings 15
decades (see gure 6). Over that same period, more than
400,000 new small businesses were created each year, with the
stock of small businesses growing by about 25,000 per year
on net.
In just the three months from March to June, 2020, more
small businesses were lost than is typical during an entire
year. Even if for the remainder of the year losses simply keep
pace with those in previous years, we will see a doubling of
the ordinary annual rate of small business losses to more
than 700,000 (or 12 percent). at likely optimistic scenario
would see around 50 percent more business losses than at
the peak of the Great Recession, and the largest loss of small
businesses since records began in 1977. During the lockdown
period, there was a signicant pause in the formation of new
businesses, but this has since reversed. rough the year to
date, the formation of new likely employer businesses of all
sizes is consistent with the trend in the years since the Great
Recession (US Census Bureau 2020b).
In net terms, therefore, we look set to lose at least as many
small businesses in this year alone as over the four-year
period from peak to trough during the Great Recession (see
gure 1). at net loss was partly driven by exits, but more
substantially by a large drop in entries (see gure 6), while
the current crisis looks set to do the opposite. Given that the
current crisis is more extreme than the Great Recession, we
should be prepared for net business losses to mount in the
months ahead. is is all the more likely if the deteriorating
outlook among small businesses continues, which could
dampen the formation of new businesses.
At the end of March, a fortnight aer the lockdowns began and
right aer the CARES Act was passed, many small businesses
reported what has turned out to be optimism about the path
of the crisis. Twenty percent of small businesses expected
the crisis to be over by the end of May, 30 percent between
May and July, and 50percent beyond July (Bartik, Bertrand,
Cullen, Glaeser, Luca, and Stanton 2020). But as of mid-June,
almost 40 percent of small businesses were reporting that
the crisis was still having a large negative eect and almost
45 percent reported a moderate negative eect (US Census
Bureau 2020d).
And the outlook deteriorated considerably in April, with
25percent of small businesses reporting they did not expect
to recover within a year, and 5percent reporting a 90percent
chance they would permanently close or go bankrupt within
six months (Humphries et al. 2020b). Between March 28 and
April 20, the proportion of small businesses expecting to ever
recover fell by 10 percentage points.
e outlook deteriorated further in May as businesses moved
into their third month of lockdown (see gure 5b). Over the
rst three weeks of May, the proportion of small businesses
expecting to recover within two to six months fell by 15
percentage points, from 52 to 37percent (US Census Bureau
2020d). Meanwhile, the proportion expecting their recovery
to take more than six months rose by 11 percentage points
from 31 to 42 percent, and the proportion expecting never
to recover rose by 4 percentage points from 6 to 10 percent.
While from late May to mid-June the outlook stabilized,
from mid-June through mid-August it resumed its decline,
with more than half of all small businesses expecting not to
recover within six months.
FIGURE 6.
Small Business Births, Deaths, and Net Rate of Creation, 1978–2014
Source: U.S. Census Bureau, 2020c.
-5
-4
-3
-2
-1
0
1
2
3
4
5
-600
-400
-200
0
200
400
600
1978 1982 1986 1990 1994 1998 2002 2006 2010 2014
Births and deaths of rms (in thousands)
Net rate of creation (percent)
Firm birth
Firm death
Net rate of creation
16 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
The Proposal
T
he more time that passes, the less we should consider
the policies we would institute in an “ideal” pandemic.
American public health outcomes have been a disaster,
and we should not pin our hopes on that being comprehensively
addressed any time soon. Hopes for a V-shaped recovery
should long have been abandoned. Ambitions to freeze every
small business in the country while suppressing the virus, and
then thawing them all out as the economy reopens, are well
past their expiration date. We must respond to the crisis as it
stands today.
Many hundreds of thousands of small businesses—and the
millions of jobs they are responsible for—are gone. ere is
nothing we can do about that now. But there are millions of
other small businesses teetering on the brink, and there are
lots of things we can do to help them. In doing so, we must
recognize that every dollar of scal support comes at a cost.
And we should be sure our policy responses do not unduly
constrain the recovery. In time, economic resources must be
allowed to ow to where they will be of most use.
In conguring small business support, we should aim for
responses at two margins. First, we should aim to minimize
the failure of otherwise-viable small businesses, which will
preserve valuable rm-specic capital, reduce employment
losses today and in the medium term, and mitigate a systemic
shock caused by a large volume of simultaneous rm exits.
And second, we should aim to maximize employment by
small businesses, which account for roughly half of all jobs.
ese goals apply to those businesses directly aected by the
pandemic, but also to those hit by the second-round eects of
the economic crisis.
ECONOMY-WIDE MEASURES TO SUPPORT SMALL
BUSINESSES
e single-most-eective measure to support small
businesses would be to suppress the virus. Evidence from the
United States suggests that much of the economic contraction
has come from voluntary social distancing measures rather
than from the lockdowns themselves (Goolsbee and Syverson
2020). As shown in gure 7, countries that took more-
FI GURE 7.
Change in GDP and COVID-19 Deaths for Selected OECD Countries
Source: Hassel 2020; author’s calculations.
Note: Data include the top 22 OECD countries by GDP per capita (excluding New Zealand and Luxembourg due to data limitations). Observations are
colored by region. Change in GDP reflects the change in GDP between the second quarter of 2019 and the second quarter of 2020.
Ireland
Switzerland
Norway
United States
Iceland
Netherlands
Austria
Denmark
Australia
Germany
Sweden
Belgium
Finland
Canada
United Kingdom
France
Japan
Italy
South Korea
Israel
Czech Republic
Spain
-30
-25
-20
-15
-10
-5
0
-50 50 150 250 350 450 550 650 750 850
Percent change in GDP
Conrmed deaths per million
The Hamilton Project • Brookings 17
aggressive steps to suppress the virus—meanwhile nancially
supporting people and businesses—have had less-severe
economic contractions (Hamilton 2020). Because the United
States did not take adequate steps to suppress the virus, we
have the worst of both worlds: many deaths and a severe
recession.
ere is broad agreement among public health experts on
strategies to suppress the virus. ere is a general consensus
on the value of improved testing—in scale but also in
speed. Barriers to improving testing and therapies should
be dismantled. Mask use should be mandated where social
distancing is impossible. State governments should roll out
contact tracing apps that leverage the Apple–Google API,
which has been designed to maximize utility while protecting
privacy. Some have proposed a renewed temporary national
lockdown to bring the contagion down to a manageable level
(Osterholm and Kashkari 2020). ese are just a few among
many measures recommended by experts.
Another critical measure to support small businesses is to
support the broader economy via aggressive scal stimulus.
e federal government should continue to provide signicant
additional support to UI recipients, in the order of $400 per
week. Incentivizing states to develop their systems to better
match workers’ replacement rates is a good idea, but should
not prevent the support from being provided. is additional
amount should be phased out slowly over time at a constant
rate so that support is gradually withdrawn as the economy
recovers. e government should also provide additional cash
support to households, and a signicant funding boost to
state and local governments.
e economic contraction has spread beyond the small
businesses directly aected to those in the broader economy.
Broad stimulus can support demand for these businesses’
products and services. With many businesses operating
at necessarily reduced capacity, the demand for labor is
unavoidably weak. ere is no evidence that businesses
are struggling to nd workers, much less that a shortage of
available labor is driven by the generosity of UI. is will
become an emerging concern if the current, very high levels
of UI are maintained well into the recovery phase, but it is
too soon to worry about that. In this climate, we should be
far more concerned about the welfare of the unemployed and
the macroeconomic eects of a loss of income than about any
disincentive to work.
EXPANDED REFUNDABLE SMALL-BUSINESS TAX
CREDITS
e current Employee Retention Credit (ERC) should be
expanded as follows:
• e ERC, which provides a refundable credit against the
employer’s payroll tax obligations, should apply to all small
business employers (those with 500 or fewer employees)
with revenue during the relevant quarter down at least
30percent relative to the most recent corresponding pre-
crisis quarter.
• e condition tying the ERC to public health orders should
be removed.
• Businesses that received forgivable loans under the PPP
should be eligible.
• It should cover all workers, regardless of whether or not
they are “providing services,” which is a restriction under
the current ERC for businesses with 100 or more employees.
• It should cover 80percent of eligible wages up to $15,000 per
quarter (for a maximum subsidy of $12,000 per employee
per quarter).
• e ERC should operate for the three quarters beginning
October 1, 2020. e end date of the existing credit should
be brought forward to September 30.
• e IRS should continue to allow businesses to request an
advance, and it must build the capacity necessary to fulll
what may be a large volume of such requests.
• An amount equal to the businesses’ regular Social Security
tax payments that are credited should be paid into the
Social Security Trust Fund out of general revenues.
In addition, a new Small Business Survival Credit (SBSC)
should be introduced to provide additional support
for non-payroll costs. Under the same revised eligibility
conditions as above, this would provide an additional $5,000
per employee per quarter, up to a maximum of $50,000 per
business per quarter, to cover rent, utilities, interest, and
COVID-19 mitigation costs.
e existing ERC is eectively a wage subsidy, covering
50 percent of wages up to $10,000 between March 12 and
December 31, 2020. But this amounts to a maximum subsidy
of just $5,000 per worker over a nine-month period and
could not be used in conjunction with the PPP. Moreover, for
businesses with more than 100 workers, it could be used only
to cover the wage costs of idled workers.
e changes I propose would transform the ERC into the
business support policy we need today. In its rst iteration, it
was ill-suited to serve as immediate but temporary life support
for locked-down businesses. But with these proposed changes,
it oers generous, targeted support for the small businesses
suering through the crisis. is new design is similar to
that used in other countries, including Australia, where it
has been found to be eective in supporting employment and
minimizing business exits (Australian Treasury 2020).
18 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
e PPP has monopolized discussion of small business
support during the crisis. ough imperfectly designed and
implemented, the PPP was in principle the right kind of tool
for providing substantial up-front support to small businesses
during a national lockdown to suppress the virus. But outside
that environment, even if lockdowns still occur in certain
locations and to varying degrees, the PPP is inappropriate. As
such, it should not be extended.
By contrast, the existing ERC received almost no attention.
But with improvements it is far better placed to support small
businesses in this new environment. Rather than providing
money up front via banks and for a prespecied time, with
rigid employee retention and other forgiveness conditions,
the existing ERC piggybacks on the payroll tax system to
provide ongoing quarterly support for payroll, and much
more exibility. is expanded ERC would do the same.
Targeting is critical. Any future scal response will inevitably
be capped. We cannot aord to provide aid to the almost
three-quarters of all businesses that received funding under
the PPP. Many of those businesses will in fact have proted
from the pandemic with this support. Meanwhile, those most
in need of support did not receive enough. But the 50percent
revenue-loss threshold under the existing ERC is too high
for an ongoing program intended to support a broader set of
rms.
Tying support to revenue losses each quarter ties that support
to local conditions. e PPP was predicated on the basis of a
temporary, national lockdown, but the virus does not spread
uniformly across the country. What appears at a national
level to have been a second wave was in fact mostly a series
of initial signicant waves in a number of dierent locations
(see gure 8). In the period ahead there will inevitably be
renewed outbreaks in certain locations. Some businesses may
not qualify in earlier quarters, but will enter the program in
later quarters; as local conditions improve, businesses will
drop out.
e credit would provide support that is much more generous
than the existing ERC, covering 80percent of wages for most
workers. is provides businesses with a strong incentive
to retain their existing workers and to hire new ones. To
the extent that an increased demand for labor leads to a
tightening of the labor market, workers will benet both in
greater employment and higher wages. By covering a large
portion of payroll, the program frees up cash to defray other
costs, which will help stem business exits.
For businesses with a high level of non-payroll costs relative
to payroll costs, the SBSC provides additional support to
cover those costs. is was a major criticism of the PPP by
small businesses. e contraction in demand has led many
businesses to scale back operations, necessitating layos. But
other expenses such as rent, utilities, interest, and COVID-19
mitigation costs are unavoidable.
Provided an employer has sucient non-payroll costs
to exhaust the SBSC, for the rst 10 employees the ERC
and SBSC combine to provide a $5,000 per employee base
subsidy, rising at a rate of 80 cents per dollar of wages up to a
maximum of $17,000 (see gure 9a). Providing the SBSC on a
per employee basis generates a powerful retention and hiring
incentive for the 80percent of small businesses with nine or
fewer employees. A full-time worker on the federal minimum
FIGURE 8.
COVID-19 Cases by U.S. Region, March–July
Source: Centers for Disease Control and Prevention 2020; author’s calculations.
0
5
10
15
20
25
luJnuJyaMrpAraM
Rolling 7-day average (in thousands)
South
West
Midwest
Northeast
The Hamilton Project • Brookings 19
wage earns $3,872 per quarter—the SBSC would more than
oset the cost of such a worker.
By being capped, the design also allocates support
disproportionately to businesses with fewer employees or
employees on low wages. A business with ve full-time
employees earning the federal minimum wage and with
sucient non-payroll costs would receive $25,000 per quarter
to cover non-payroll costs and around $15,000 per quarter
to cover payroll costs. is amounts to around $8,000 of
quarterly support per employee, or around $2.70 in support
for every dollar of payroll. If each of those workers earned
the national median wage, which is more than double the
minimum wage, the business would receive around $12,000
per employee, or only around $1.70 in support for every dollar
of payroll.
For most businesses, the 30percent revenue-drop threshold
would require a decline in revenue per employee per quarter of
at least $15,000, and for many eligible businesses the drop will
be much larger. Some will be made up for by cost reductions
in payroll and other variable costs, and by negotiating lower
rent. But the ERC and SBSC would together help substantially
narrow—though not completely close—that revenue gap.
Quarterly payroll per employee is around $10,000 on average,
so the ERC would provide a bit less than $8,000 per employee
per quarter on average (it covers 80percent of wages only up to
$15,000 per quarter). e SBSC would provide businesses with
10 or fewer employees up to an additional $5,000 per employee
per quarter, taking average total support for these rms to
$13,000 per employee per quarter, or around 80percent of the
minimum revenue loss. For the 20percent of businesses with
11 or more employees, the SBSC would provide less support
per employee (a business with 50 employees would receive
only an additional $1,000 per employee, for example) so larger
rms would not have as much of their revenue losses covered
(see gure 9b).
FIGURE 9A .
Maximum Quarterly Tax Credit Per
Employee, by Salary
Since the measure depends on the distribution of revenue
losses and expenses across rms, it is dicult to predict take-
up. But if 10percent of all small businesses suered revenue
losses exceeding 30 percent, the ERC would cost around
$47 billion per quarter.
17
If all eligible small businesses had
sucient expenses to exhaust the SBSC, it would cost around
$13billion per quarter.
18
Accordingly, for every 10percent of
small businesses that qualify for all three quarters, the ERC
and SBSC combined would cost up to $180billion. e $130
billion in unused PPP funds could be used to help fund this
spending.
Even if 30percent of small businesses were to qualify in all
three quarters, the two credits would cost around the same
as the PPP. is is because in this scenario the credits would
go to fewer than half the number of rms that received PPP
funding. While the PPP replaced 100 percent of salaries up
to the equivalent of $25,000 per quarter (versus this ERC
proposal, which replaces 80 percent up to $15,000) and
provided more-generous support for xed costs on a monthly
basis, it covered less than a quarter of the timespan of this
proposal. By prolonging the spread of the virus, we have
prolonged the contraction in demand, which necessitates
a longer duration of support. is reduces the generosity of
support that can be provided.
e ERC and SBSC are refundable tax credits. Businesses
could draw on their own employer-side payroll tax payments
to partially nance them in advance, or apply to the IRS for
an advance as under the existing ERC. e IRS must be given
the capacity to fulll what may be a large volume of such
requests. Without an advance, businesses would receive the
payments trailing each quarter. e CARES Act also allowed
businesses to defer their 2020 Social Security tax payments
from March 12, with the rst half to be paid on December 31.
is ERC proposal would then implicitly be partly funded by
these deferred tax payments.
Source: Author’s calculation.
0
5
10
15
20
Maximum quarterly tax credit
per employee (thousands of dollars)
$0 $5,000 $10,000 $15,000 $20,000
Quarterly salary
1-10 employees
11-500 employees
FIGURE 9B.
Average Quarterly Tax Credit Per
Employee, By Firm Size
Source: Author’s calculations.
0
2
4
6
8
10
12
14
Average quarterly tax credit
per employee (thousands of dollars)
0 50 100 150 200 250 300 350 400 450 500
Number of employees
20 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
BUILDING CAPACITY IN THE IRS OVER THE LONG
TERM
ere should be a signicant, sustained increase in
investment in the IRS, in particular in a real-time electronic
payroll reporting system.
Having reected on what went wrong in the crisis, it is
important to consider how we might better prepare our
infrastructure for the future. Many of the problems with the
PPP stemmed from the delivery mechanism. e Treasury
Department was too slow in providing guidance to the banks,
and then updated their guidance repeatedly. Having the
banks act as intermediaries introduced frictions. Many of the
biggest banks extended loans only to their existing customers.
e forgiveness process is only beginning, but is sure to be
fraught. Under the circumstances, the SBA and the banks
performed about as well as can be expected. But we should
not have had to rely on them.
e IRS is the largest nancial institution in the world,
collecting more than $3trillion in annual revenue. rough
the withholding system, the IRS lends to and borrows from
hundreds of millions of businesses and people every year. e
IRS holds the nancial records of all of these taxpayers. It
knows their bank details. With the necessary infrastructure,
the IRS is uniquely placed to implement a large-scale wage
subsidy program, and provide immediate liquidity to every
small business in the country.
e United States was unique in taking the private bank route
to deliver a wage subsidy. Other countries, such as Australia,
the Netherlands, New Zealand, and the United Kingdom all
delivered their wage subsidies via the tax authority. In fact,
the United States did as well, albeit in the limited form of the
original ERC. e reason the ERC could not be the primary
delivery mechanism—why we instead had to rely on the
banks—is liquidity, or a lack thereof.
Under the ERC, the IRS directed businesses in the rst
instance to draw on their tax withholdings (for both the
employers’ and employees’ shares of the payroll tax and the
employees’ income tax) to fund the subsidy. Payroll taxes
constitute around 15.3percent of payroll, and income taxes
a little more. But these add up to far less than the liquidity
required to fund a more ambitious program like the PPP. If a
business participating in the ERC wanted an advance, it had to
ll out and submit a form to the IRS. Given the cumbersome
nature of this process, it seems unlikely the IRS could have
managed a large volume of such requests in the mere days it
took to get the PPP up and running.
Ultimately, the IRS should have been capable of implementing
the PPP, remitting the necessary funds to small businesses in
advance. In an ideal world, the IRS would simply have shied
business tax withholding into reverse—instead of receiving
money from businesses in advance (implicit borrowing), they
would have paid it out to them in advance (implicit lending).
With knowledge of prior-year payroll, it would have been
straightforward to remit two and a half months’ worth to
every eligible small business in the country.
If in possession of real-time payroll information, the IRS
could then easily have assessed loan forgiveness (based on
worker retention and pay) over the relevant eight-week period.
Any non-forgiveness (because workers were not retained
or were underpaid) could be reconciled at tax time, which
happens quarterly for most businesses. e amount not used
for payroll could be taxed back through the ordinary business
tax ling process.
But the value of having the IRS run something like the PPP
is not only a matter of ecient administration, important
as that is. A key advantage the IRS has over the SBA and
private banks is that it is far better placed to overcome the
information asymmetries that make it dicult for the banks
and SBA to assess loan forgiveness. Much of the necessary
information is already known by the IRS as a matter of
course. Obtaining additional information could only enable
it to better perform its collection and enforcement functions.
And, importantly, the IRS is better placed than the SBA to
enforce the loan forgiveness terms.
Why did every other country in the world that delivered
a wage subsidy rely on its tax authority while the United
States relied on private banks? is choice reveals a lack of
capacity in the IRS to deliver such a program. is is not to
denigrate the IRS, which among other functions performed
admirably in distributing stimulus payments in record time.
But for decades, it has been hamstrung by a lack of funding.
Since 2010 the IRS budget has declined by 20 percent in
real terms (Weinberger 2020). is has diminished even its
core functions, with the audit rates on both personal and
corporate returns having nearly halved over that decade.
Critically, information technology spending at the IRS has
lagged behind private-sector nancial rms and even the
Federal Reserve (see gure 10).
In Australia, as in many other countries, payroll information
for every employee is transmitted to the tax authority in
real time (Australian Taxation Oce 2020). As soon as an
employee is paid, all of the relevant payroll information is
accessible by the taxpayer on the tax authority’s website. is
real-time information is critical to delivering a program like
the PPP quickly. It also supports compliance and enforcement,
and provides a ow of real-time data to the national statistical
authority, which are then made available to the public. All
of these should be well within the capabilities of the tax
authority of the world’s richest nation.
The Hamilton Project • Brookings 21
FIGURE 10.
Change in Annual Technology Spending for Selected Institutions, 2001–19
Source: Internal Revenue Service (IRS) 2001–19; Federal Reserve System 2001–18; Securities and Exchange Commission (SEC) 2001–19a; SEC 2001–19b; SEC
2001–19c; author’s calculations.
Note: Relative spending growth adjusted for inflation.
-20
30
80
130
180
230
2001 2004 2007 2010 2013 2016 2019
IRS
Citi
Bank of
America
Federal
Reserve Board
JP Morgan
0
Percent change (relative to 2001 spending)
22 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
Questions and Concerns
1. What are the drawbacks of tying eligibility to revenue
losses?
e revenue-loss threshold will inevitably encourage some
businesses at the margin to reduce their revenues in order to
qualify for the credit. A broad literature studying bunching
around tax thresholds suggests this is likely to be conned
to only a limited set of rms located close to the threshold,
and that this manipulation will mainly occur via reporting
or time-shiing of revenues rather than changes in real
output. e other drawback is that small business revenues
are volatile even in the absence of the pandemic, which means
support will inevitably be provided to businesses for which
revenues would have declined anyway. When targeting scal
support, there is always a tradeo between eciency and
equity. e policy proposed here is a far better tradeo in
this regard than the PPP, and is more easily implemented and
more transparent than more elaborate eligibility schemes.
2. Is there a risk that the program will cost more than
indicated?
Because eligibility is tied to revenue losses within a given
quarter, any cost estimate will be subject to signicant
uncertainty. In the three months from mid-March to mid-
June, small business revenues were down by around 20 to
25 percent on average, while around 40 percent of small
businesses on average indicated revenues were at or up
(US Census Bureau, 2020d). is puts a very conservative
ceiling of perhaps 50 percent on the proportion of businesses
qualifying during a worst-case-scenario quarter. If 50 percent
of businesses were to qualify in all three quarters, the program
would cost around $900 billion, but this is highly unlikely. If
50 percent qualied in one quarter, 40 percent in another,
and 30 percent in another, the program would cost around
$720 billion. With 30 percent qualifying in each quarter, the
program would cost $540 billion, almost exactly the cost of
the PPP.
3. How will new businesses that have not been around for at
least a year but have been adversely aected qualify?
Roughly 7 percent of businesses are less than 12 months
old (US Census Bureau, 2020c), so this is likely to be fairly
limited. But for this small fraction, an alternative eligibility
measure could be dened. For example, in the Australian
wage subsidy, such businesses were allowed to provide
evidence to the tax authority that revenues were down relative
to the period immediately preceding the crisis, and something
similar could be applied in the United States.
4. e PPP was applied at the establishment level for some
industries—should the ERC and SBSC be applied in that way
as well?
e PPP was primarily an employee-retention program,
designed for a short lockdown period. e program was
made available to larger businesses in certain industries (e.g.,
hotels and restaurants) as long as their employee count at the
establishment level was below the 500-employee threshold.
is was designed to encourage the retention of workers
in these industries. e ERC and SBSC proposed here have
dierent goals to the PPP, and as such the same conditions
should not be applied. Rather than employee retention, the
primary goal of the ERC and SBSC is to maximize small
business survival through the prolonged period of reduced
revenue ahead. Businesses in the industries given special
treatment under the PPP do not face the risks to which many
small businesses are exposed, discussed in great detail in this
paper. It is important also to emphasize the funding limits
that are likely to constrain any program—these limited funds
must be allocated to those most in need and those most likely
to be saved by them.
The Hamilton Project • Brookings 23
Conclusion
G
iven the necessary resolve, there is still time to avert
further catastrophe. We can suppress the virus. We
can roll out additional stimulus. We can build a bridge
to help small businesses traverse the abyss. But we must act
quickly—the window is closing rapidly. e second wave of
cases has peaked; this time we must not allow them to plateau
at an unacceptably high level. e massive boost to incomes
that came with the rst round of stimulus payments—in
expanded UI, cash transfers, and small business support—has
enabled consumption to coast in recent months, but it will
soon ebb. e PPP and emergency lending have been a crutch
for small businesses, but before long they will resume drawing
down their cash reserves.
e rst round of stimulus was born from uncharacteristic
bipartisanship—Congress and the White House bridged the
partisan divide to deliver for America in its hour of need. It
was a historic achievement, safeguarding the livelihoods of
millions. Now we must do it again. Both parties—in both
chambers of Congress—have put forward plans containing
commendable elements. On small business support, at least,
the gap between the two is not that great.
My key proposal—to provide radically expanded refundable
tax credits for small businesses—improves on the plans
put forward to date. It oers both the generosity of payroll
support advocated by Democrats and the support for
COVID-19 mitigation costs and hiring incentives advocated
by Republicans. It will support millions of small business
owners and their tens of millions of employees. It will
encourage small businesses to retain their current employees
and hire new ones. It will help protect the economy so that we
can bounce back once we have beaten the virus. It is modest
in scal terms relative to many of the other proposals oated.
e case is clear. We owe it to the millions of struggling small
businesses to act.
24 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
Author
Steven Hamilton
Assistant Professor of Economics, e George Washington
University
Steven Hamilton is assistant professor of economics at e
George Washington University. Steven’s primary area of
research is public nance, where he studies the eects of taxes
on behavior with a view to designing better tax policy. Steven
has provided extensive commentary on the economy and
small business support during the COVID-19 crisis to outlets
such as the New York Times, Slate, e Hill, e Atlantic, the
Los Angeles Times, Time Magazine, Newsweek, e Bulwark,
e Dispatch, and NPR. Steven is a former economist at the
Australian Treasury, where he worked on the federal budget,
corporate and international taxation, and government reviews
of climate change policy and ood insurance. Steven holds a
doctorate and a master’s in economics from the University
of Michigan, and a bachelor of economics with First Class
Honours and bachelor of business management from the
University of Queensland.
Acknowledgments
e author wishes to thank Mitchell Barnes for his outstanding research assistance, e Hamilton Project team for their support
with all aspects of the paper, Alison Hope for her meticulous copy-editing (though any errors that remain are the author’s own),
and the “author’s conference” attendees for their valuable feedback and insights.
The Hamilton Project • Brookings 25
Endnotes
1. e standard denition of a small business is one with 500 or fewer
employees. ese businesses are sometimes known as small and medium-
sized enterprises, or SMEs.
2. ose sectors are retail trade; education; health care and social assistance;
arts, entertainment, and recreation; accommodation and food services; and
other services (except public administration).
3. See French (2020) for legal arguments for and against business interruption
insurance coverage of pandemic losses, and Organisation of Economic Co-
operation and Development (OECD) for policy proposals on extending
pandemic insurance coverage (OECD 2020).
4. e program did not apply to passive businesses such as hedge funds or
private equity rms. For businesses operating in certain industries, the
500-employee threshold was applied on a per establishment basis. is
extended eligibility to many large hotel and restaurant chains that would
otherwise have been ineligible.
5. Veuger and Grawert (2020) argue against these restrictions.
6. e payroll calculation included health insurance and retirement benets,
and withheld federal, state, and local income taxes but not employer-side
payroll taxes.
7. e loan term was originally two years.
8. F8. or an examination of the forgiveness process, see Congressional Research
Service (2020).
9. e program originally required 75percent to be spent on payroll costs.
10. e program originally covered expenses for eight weeks.
11. Chetty et al. (2020) nd that the PPP saved 1.64 million jobs in April and
May with a scal cost per job saved of $319,000, but they rely on data from
Earnin that is highly unrepresentative of the population of rms. Bartlett
and Morse (2020) consider the eect of the PPP on small businesses in
Oakland, California, and nd that application success increased the
probability of survival by 20.5 percent, but only among the smallest
businesses.
12. e study did not account for the fact that employers that did not participate
in the PPP were instead eligible for an ERC subsidizing 50percent of wages
up to $10,000 per employee, which will have encouraged some rms not
eligible for the PPP to lay o fewer workers. is would bias downward the
apparent eect of the PPP. Also, the estimate does not include the jobs saved
in the medium term by preventing permanent business closures, nor the
jobs saved throughout the broader economy due to the stimulatory eect of
the 2.5percent of GDP in PPP support. ese all result in scal savings due
to reduced UI payments.
13. e loan term was originally four years.
14. e interest rate is LIBOR (currently around 0.2percent) plus 3percent.
15. e loan range was originally $500,000–$200million.
16. Womply (2020) reports that, as of June 15, 16.7percent of all businesses were
closed, either temporarily or permanently, while Yelp (2020) reports that
41percent of businesses that were closed had indicated it was permanent.
As of July 10, these gures were 12.9percent and 55percent, respectively.
17. e revised ERC covers 80 percent of wages up to $15,000 per quarter, which
is around the 70th percentile of the U.S. wage distribution. Assuming this
applies to small businesses, 1.8 million of the 18 million employees earning
above this amount would attract a subsidy of $12,000 per quarter, totaling
$21.6 billion. en 4.2 million of the 42 million employees earning less than
this amount would attract a subsidy equal to 80 percent of their salary. If
we assume that the wage distribution rises linearly to that point, the scal
cost would be $25.4 billion. Given employment has fallen signicantly, and
not all of the laid o workers will be rehired, this is likely an overestimate
of the scal cost.
18. ere are 1.3million businesses with 10–500 workers, 10percent of which
would receive $50,000 per quarter, totaling $6.3 billion. Pre-pandemic
there were 13 million workers at businesses with nine or fewer workers,
10percent of whom would attract a credit of $5,000 per quarter, totaling
$6.3billion.
26 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
Australian Taxation Oce. 2020. “Single Touch Payroll.”
Australian Taxation Oce, Canberra, Australia.
Australian Treasury. 2020. e JobKeeper Payment: ree-Month
Review. Canberra, Australia: Australian Treasury.
Autor, David, David Cho, Leland D. Crane, Mita Goldar, Byron
Lutz, Joshua Montes, William B. Peterman, David Ratner,
Daniel Villar, and Ahu Yildirmaz. 2020. “An Evaluation of
the Paycheck Protection Program Using Administrative
Payroll Microdata.” Massachusetts Institute of Technology,
Cambridge, MA.
Bartik, Alexander W., Marianne Bertrand, Zoë B. Cullen, Edward
L. Glaeser, Michael Luca, and Christopher Stanton. 2020.
“How Are Small Businesses Adjusting to COVID-19?
Early Evidence from a Survey.” Proceedings of the National
Academy of Sciences 117 (30): 17656–66.
Bartik, Alexander W., Marianne Bertrand, Zoë B. Cullen, Edward
L. Glaeser, Michael Luca, Christopher Stanton, and Adi
Sunderam. 2020. “e Targeting and Impact of Paycheck
Protection Program Loans to Small Business.” Working
Paper 21-021, Harvard Business School, Cambridge, MA.
Bartlett, Robert P., III, and Adair Morse. 2020. “Small Business
Survival Capabilities and Policy Eectiveness: Evidence
from Oakland.” Working Paper 27629, National Bureau of
Economic Research, Cambridge, MA.
Bishop, John H., and Timothy J. Bartik. 2009. “e Job Creation
Tax Credit.” Brieng Paper 248, Economic Policy Institute,
Washington, DC.
CARES Act. 2020. “H.R.748 – CARES Act.” 116th Congress
(2019- 2020).
Centers for Disease Control and Prevention (CDC). 2020. “United
States COVID-19 Cases and Deaths by State over Time.”
Dataset, Centers for Disease Control and Prevention,
Atlanta, GA.
Chetty, Raj, John Friedman, Nathaniel Hendren, Michael Stepner,
and e Opportunity Insights Team. 2020. “How Did
COVID-19 and Stabilization Policies Aect Spending and
Employment? A New Real-Time Economic Tracker Based
on Private Sector Data.” Opportunity Insights, Cambridge,
MA.
Congressional Research Service, 2020. “SBA Paycheck Protection
Program (PPP) Loan Forgiveness: In Brief.” Report R46397,
Congressional Research Service, Washington, DC.
Cororaton, Anna, and Samuel Rosen. 2020. “Public Firm Borrowers
of the US Paycheck Protection Program.” Available at
SSRN.
Decker, Ryan, John Haltiwanger, Ron Jarmin, and Javier Miranda.
2014. “e Role of Entrepreneurship in US Job Creation and
Economic Dynamism.” Journal of Economic Perspectives 28
(3): 3–24.
Fairlie, Robert. 2020. “e Impact of COVID-19 on Small Business
Owners: Evidence of Early-Stage Losses from the April
2020 Current Population Survey.” Working Paper 20-
022, Institute for Economic Policy Research, Stanford
University, Stanford, CA.
Farrell, Diana, and Chris Wheat. 2018. “Bend, Don’t Break: Small
Business Financial Resilience Aer Hurricanes Harvey
and Irma.” JPMorgan Chase Institute. Available at SSRN:
https://papers.ssrn.com/sol3/papers.cfm?abstract_
id=3138573
Farrell, Wheat, and Grandet. 2019. Facing Uncertainty: Small
Business Cash Flow Patterns in 25 U.S. Cities. Washington,
DC: JPMorgan Chase Institute.
Farrell, Diana, Chris Wheat, and Chi Mac. 2018. Growth, Vitality,
and Cash Flows: High-Frequency Evidence from 1 Million
Small Businesses. Washington, DC: JPMorgan Chase
Institute.
———. 2020. Small Business Owner Race, Liquidity, and Survival.
Washington, DC: JPMorgan Chase Institute.
Federal Reserve Board. 2020. “Factors Aecting Reserve Balances,
ursday August 6, 2020.” Federal Reserve Board,
Washington, DC.
Federal Reserve System. 2001–18. “Annual Report.” Publications,
Board of Governors of the Federal Reserve System,
Washington, DC.
———. 2020. Small Business Credit Survey: 2020 Report on
Employer Firms. New York, NY: Federal Reserve Bank of
New York.
French, Christopher C. 2020. “Covid-19 Business Interruption
Insurance Losses: e Case For and Against Coverage.”
Connecticut Insurance Law Journal 27, Research Paper 14-
2020, Penn State Law, Penn State University, State College,
PA.
Goolsbee, Austan, and Chad Syverson. 2020. “Fear, Lockdown,
and Diversion: Comparing Drivers of Pandemic Economic
Decline 2020.” Becker Friedman Institute for Economics at
the University of Chicago, Chicago, IL.
Granja, Joao, Christos Makridis, Constantine Yannelis, and Eric
Zwick. 2020. “Did the Paycheck Protection Program Hit
the Target?” Becker Friedman Institute for Economics at
the University of Chicago, Chicago, IL.
References
The Hamilton Project • Brookings 27
Guerrieri, Veronica, Guido Lorenzoni, Ludwig Straub, and
Ivan Werning. 2020. “Macroeconomic Implications of
COVID-19: Can Negative Supply Shocks Cause Demand
Shortages?” Harvard University, Cambridge, MA.
Hamilton, Steven. 2020 “A Tale of Two Wage Subsidies: e
American and Australian Fiscal Responses to COVID-19.”
National Tax Journal 73(3): 829–846.
Hamilton, Steven, and Stan Veuger. 2020a. “A Recession Is a Public
Health Necessity. Here’s How to Make It Short and Sharp.”
e Bulwark, March 14, 2020.
———. 2020b. “How to Help American Businesses Endure and Jobs
Survive.” Working paper.
Hassel, Joe. 2020. “Which Countries Have Protected Both Health
and the Economy in the Pandemic?” Our World in Data,
University of Oxford, Oxford, UK.
Homebase. 2020. “Percent Change in Employment.” Opportunity
Insights, Cambridge, MA. https://tracktherecovery.org
Humphries, John E., Christopher A. Neilson, and Gabriel Ulyssea.
2020a. “Information Frictions and Access to the Paycheck
Protection Program.”
———. 2020b. “e Evolving Impacts of COVID-19 on Small
Businesses Since the CARES Act.” Working Paper 2230,
Cowles Foundation for Research in Economics, Yale
University, New Haven, CT.
Internal Revenue Service (IRS). 2001–19. “Costs Incurred by Budget
Activity.” IRS Databook, Internal Revenue Service, U.S.
Department of the Treasury, Washington, DC.
Morrell, Alex, Rebecca Ungarino, Dominick Reuter, and Jennifer
Ortakales. 2020. “How Big Banks Decided the Futures of
America’s Small Businesses: e Inside Story of How $349
Billion in Government Cash Was Doled Out in Just 12
Days, Leaving ousands of Entrepreneurs Without Relief.”
Business Insider, April 25, 2020.
Organisation of Economic Co-operation and Development
(OECD). 2020. “Responding to the COVID-19 and
Pandemic Protection Gap in Insurance.” Organisation of
Economic Co-operation and Development, Paris, France.
Osterholm, Michael T., and Neel Kashkari. 2020. “Here’s How to
Crush the Virus Until Vaccines Arrive.” e New York
Times, August 7, 2020.
Securities and Exchange Commission (SEC). 2001–19a. “EDGAR
Filings - Citigroup.” EDGAR System, Securities and
Exchange Commission, Washington, DC.
———. 2001–19b. “EDGAR Filings - JP Morgan Chase.”
EDGAR System, Securities and Exchange Commission,
Washington, DC.
———. 2001–19c. “EDGAR Filings – Bank of America.”
EDGAR System, Securities and Exchange Commission,
Washington, DC.
Sedláček, Petr. 2020. “Lost Generations of Firms and Aggregate
Labor Market Dynamics.” Journal of Monetary Economics
111:16–31.
Siemer, Michael. 2019. “Employment Eects of Financial
Constraints during the Great Recession.” Review of
Economics and Statistics 101 (1): 16–29.
US Department of the Treasury. 2020. “SBA Paycheck Protection
Program Loan Level Data.” US Department of the
Treasury, Washington, DC.
US Census Bureau. 2020a. “2017 SUSB Annual Data Tables by
Establishment Industry.” US Census Bureau, Washington,
DC.
———. 2020b. “Business Formation Statistics: Weekly Data.” US
Census Bureau, Washington, DC.
———. 2020c. “Legacy BDS Firm Characteristics Data Tables
1977–2014.” US Census Bureau, Washington, DC.
———. 2020d. “Small Business Pulse Survey: Tracking Changes
During the COVID-19 Pandemic.” US Census Bureau,
Washington, DC.
———. 2020e. “SUSB Historical Data.” US Census Bureau,
Washington, DC.
US Small Business Administration (SBA). 2020a. Paycheck
Protection Program (PPP) Report. Approvals through 12 PM
EST 4/16/2020. Washington, DC: US Department of the
Treasury.
———. 2020b. Paycheck Protection Program (PPP) Report.
Approvals through 06/12/2020. Washington, DC: US
Department of the Treasury.
Veuger, Stan, and Ames C. Grawert. 2020. “Don’t Deny Pandemic
Relief Loans to Second-Chance Entrepreneurs.” e
Bulwark, April 21, 2020.
Waldman, John. 2020. “e Real-Time Impact Coronavirus
(COVID-19) Is Having on Small Businesses and Workers.”
Homebase, San Francisco, CA.
Weinberger, Robert A. 2020. “Budget Blues for Tax
Administration.” TaxVox. Tax Policy Center, Washington,
DC.
Womply. 2020. “Percent Change in Number of Small Businesses
Open.” Opportunity Insights. https://tracktherecovery.org
Yelp. 2020. “Q2 2020 Yelp Economic Average report” Yelp, San
Francisco, CA.
28 From Survival to Revival: How to Help Small Businesses through the COVID-19 Crisis
ADVISORY COUNCIL
STEPHANIE AARONSON
Vice President and Director, Economic Studies
Senior Fellow, Economic Studies, The Brookings
Institution
GEORGE A. AKERLOF
University Professor
Georgetown University
ROGER C. ALTMAN
Founder & Senior Chairman
Evercore
KAREN L. ANDERSON
Senior Director of Policy & Communications
Becker Friedman Institute for
Research in Economics
The University of Chicago
ALAN S. BLINDER
Gordon S. Rentschler Memorial Professor of
Economics & Public Affairs,
Princeton University;
Visiting Senior Fellow,
The Brookings Institution.
STEVEN A. DENNING
Chairman, General Atlantic
JOHN M. DEUTCH
Institute Professor
Massachusetts Institute of Technology
CHRISTOPHER EDLEY, JR.
Co-Founder and President Emeritus
The Opportunity Institute
BLAIR W. EFFRON
Partner
Centerview Partners LLC
DOUGLAS W. ELMENDORF
Dean & Don K. Price Professor
of Public Policy
Harvard Kennedy School
JUDY FEDER
Professor & Former Dean
McCourt School of Public Policy
Georgetown University
JASON FURMAN
Professor of the Practice of
Economic Policy
Harvard University
Senior Fellow
Peterson Institute for International Economics;
Senior Counselor
The Hamilton Project
MARK T. GALLOGLY
Cofounder & Managing Principal
Centerbridge Partners, L.P.
TED GAYER
Executive Vice President
Senior Fellow, Economic Studies
The Brookings Institution
TIMOTHY F. GEITHNER
President, Warburg Pincus
Senior Counselor, The Hamilton Project
JOHN GRAY
President & Chief Operating Officer
Blackstone
ROBERT GREENSTEIN
Founder & President
Center on Budget and Policy Priorities
MICHAEL GREENSTONE
Milton Friedman Professor in Economics & the
College
Director of the Becker Friedman Institute for
Research in Economics
Director of the Energy Policy Institute
University of Chicago
GLENN H. HUTCHINS
Co-founder, North Island;
Co-founder, Silver Lake
JAMES A. JOHNSON
Chairman; Johnson Capital Partners
LAWRENCE F. KATZ
Elisabeth Allison Professor of Economics
Harvard University
MELISSA S. KEARNEY
Neil Moskowitz Professor of Economics
University of Maryland;
Nonresident Senior Fellow
The Brookings Institution
LILI LYNTON
Founding Partner
Boulud Restaurant Group
HOWARD S. MARKS
Co-Chairman
Oaktree Capital Management, L.P.
ERIC MINDICH
Founder
Everblue Management
SUZANNE NORA JOHNSON
Former Vice Chairman
Goldman Sachs Group, Inc.
Co-Chair
The Brookings Institution
PETER ORSZAG
CEO, Financial Advisory
Lazard Freres & Co LLC
RICHARD PERRY
Managing Partner & Chief Executive Officer
Perry Capital
PENNY PRITZKER
Chairman & Founder, PSP Partners
38th Secretary of Commerce
MEEGHAN PRUNTY
Managing Director, Blue Meridian Partners
Edna McConnell Clark Foundation
ROBERT D. REISCHAUER
Distinguished Institute Fellow&
President Emeritus
Urban Institute
NANCY L. ROSE
Charles P. Kindleberger Professor of Applied
Economics, MIT Department of Economics
DAVID M. RUBENSTEIN
Co-Founder & Co-Executive Chairman
The Carlyle Group
ROBERT E. RUBIN
Former U.S. Treasury Secretary;
Co-Chair Emeritus
Council on Foreign Relations
LESLIE B. SAMUELS
Senior Counsel
Cleary Gottlieb Steen & Hamilton LLP
SHERYL SANDBERG
Chief Operating Officer, Facebook
DIANE WHITMORE SCHANZENBACH
Margaret Walker Alexander Professor
Director
The Institute for Policy Research
Northwestern University;
Nonresident Senior Fellow
The Brookings Institution
STEPHEN SCHERR
Chief Executive Officer
Goldman Sachs Bank USA
RALPH L. SCHLOSSTEIN
President & Chief Executive Officer, Evercore
ERIC SCHMIDT
Technical Advisor, Alphabet Inc.
ERIC SCHWARTZ
Chairman & CEO, 76 West Holdings
JAY SHAMBAUGH
Professor of Economics and International Affairs,
Elliott School of International Affairs at The
George Washington University;
Nonresident Senior Fellow
The Brookings Institution
THOMAS F. STEYER
Business Leader & Philanthropist
MICHAEL R. STRAIN
Director of Economy Policy Studies and
Arthur F. Burns Scholar in Political Economy
American Enterprise Institute
LAWRENCE H. SUMMERS
Charles W. Eliot University Professor
Harvard University
LAURA D’ANDREA TYSON
Distinguished Professor fo the Graduate School
University of California, Berkeley
WENDY EDELBERG
Director
The Hamilton Project • Brookings 29
ADVISORY COUNCIL
Highlights
The COVID-19 pandemic poses an existential threat to small businesses, with more than
400,000 lost since the crisis began. Many small businesses are financially fragile and not
equipped to weather a prolonged period of substantially reduced revenues. In this proposal,
Steven Hamilton of The George Washington University calls for a significant expansion of
refundable tax credits to help support small businesses through this crisis.
The Proposal
Engage in important economy-wide measures to support small businesses. Noting
that the single-most-effective measure to support small businesses would be to suppress
the virus, the author supports calls for effective social distancing measures, improved testing
capacity, and increased mask use. Another critical measure to support small businesses is
to support the broader economy via aggressive fiscal stimulus. Broad stimulus can support
demand for these businesses’ products and services.
Significantly expand the Employee Retention Credit (ERC) to help cover small
businesses’ payroll costs. The ERC should apply to all small business employers (those with
500 or fewer employees) with revenue during the relevant quarter down at least 30 percent
relative to the most recent corresponding pre-crisis quarter. The credit should cover 80 percent
of eligible wages up to $15,000 per quarter (for a maximum subsidy of $12,000 per employee
per quarter).
Introduce a new Small Business Survival Credit (SBSC) to help cover small
businesses’ fixed costs. To help address small businesses’ need to cover non-payroll costs,
the SBSC would provide an additional $5,000 per employee per quarter, up to a maximum
of $50,000 per business per quarter, to cover rent, utilities, interest, and COVID-19 mitigation
costs.
Invest in the capabilities of the IRS so it may better support small businesses in
future crises. Following decades of underfunding, the IRS was unable to administer large-
scale small business support, like the PPP. The proposal calls for increasing IRS funding—
especially on technology necessary to accurately and efficiently administer some of these
support programs—so that the IRS is prepared to act quickly in a future downturn.
Benefits
Through this proposal, the author provides policymakers with a bridge to help small businesses
get through these difficult times. The proposal offers both generous payroll support as well as
support for the non-payroll costs that are a burden for many small businesses. It will encourage
small businesses to retain their current employees and hire new ones. Lastly, it will help protect
the economy so that we can bounce back once we have beaten the virus.
WWW.HAMILTONPROJECT.ORG
1775 Massachusetts Ave., NW
Washington, DC 20036
(202) 797-6484
Printed on recycled paper.
