Full text of Economic Report of the President : 2021

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

Economic Report
of the President
Together with
The Annual Report
of the
Council of Economic Advisers
January 2021

Contents
Economic Report of the President. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
The Annual Report of the Council of Economic Advisers. . . . . . . . . . . . . . . . . . . . . 9
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Part I: Confronting the Largest Postwar Economic Shock:
The Federal Response to Mitigate the COVID-19 Pandemic
Chapter 1:

Creating the Fastest Economic Recovery. . . . . . . . . . . . . . . . . . . . . 35

Chapter 2:

Prioritizing America’s Households. . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Chapter 3:

Assisting Entrepreneurs and Workers through Aid
to Businesses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Chapter 4:

Advancing the Quality and Efficiency of America’s
Healthcare System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Part II: The Renaissance of American Greatness

Chapter 5:

Assessing the Early Impact of Opportunity Zones.. . . . . . . . . . . . 145

Chapter 6:

Empowering Economic Freedom by Reducing
Regulatory Burdens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173

Chapter 7:

Expanding Educational Opportunity through Choice
and Competition.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199

Chapter 8:

Exploring New Frontiers in Space Policy and Property Rights .. 225

Chapter 9:

Pursuing Free, Fair, and Balanced Trade.. . . . . . . . . . . . . . . . . . . . 251
Part III: An Effort to Rebuild Our Country

Chapter 10: The Year in Review and the Years Ahead.. . . . . . . . . . . . . . . . . . . . 287
Chapter 11: Policies to Secure Enduring Prosperity. . . . . . . . . . . . . . . . . . . . . . 319
References

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395

Appendix A Report to the President on the Activities of the Council of

Economic Advisers During 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
Appendix B Statistical Tables Relating to Income, Employment, and

Production.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 447

____________
*For a detailed table of contents of the Council’s Report, see page 23.

iii

Economic Report of the President

x
Economic Report of the President
To the Congress of the United States:
Four years ago, on the steps of the United States Capitol, I pledged to the citizens of this country to return our Nation to greatness, and to thereby enable,
secure, and enhance the prosperity of all Americans. While this would require
us to confront challenges and hardships, I knew that the American spirit,
set free from overbearing taxation and regulation, would drive the country
forward to unimaginable economic success. I was right. Over the course of
this Administration, the American people demonstrated an indomitable will to
prevail and drove our economy to record heights.
Since my first day in office, I have been steadfast in my commitment
to put America First. The economic pillars of this movement—sweeping tax
reform, extensive deregulatory actions, and fair and reciprocal trade agreements—promoted a robust middle class and led to the longest expansion and
strongest recovery in history.
Before the coronavirus came to our country from China, our economy
created more than 7 million jobs, nearly 12,000 factories returned to our
shores, and wealth for Americans hit all-time highs. Inequality decreased as
wage growth for blue-collar workers outpaced that of their managers, and
earnings for those in the bottom 10 percent grew faster than earnings for the
top 10 percent, reversing the trends of past Administrations. Since the start of
my Administration, median household income grew by more than $6,000, lifting up people no matter their race, ethnicity, educational background, or age
group. In 2019 alone, median household income rose $4,400—more in one year
than in the entire 16 years through 2016. Even as more than 2 million people
returned from the sidelines to enter the labor force, the unemployment rate
plummeted to 3.5 percent in February 2020—the lowest level in more than half
a century.
This past year, our Nation has faced trials the likes of which many have
never experienced. Through these challenges, however, we have witnessed
once again the resilience of the American people. In just 2 months this year,
more than 23 million people saw their livelihoods threatened through no fault
of their own, as the unemployment rate peaked at 14.7 percent. Since this
spring, we have seen more than 12 million jobs return and gross domestic
product increase by a record-shattering 33 percent in the third quarter alone,
thanks to the largest and fastest economic policy response in history.
When the virus hit, my Administration launched the largest industrial
mobilization since World War II. We have created the world’s most innovative testing system, pioneered groundbreaking therapies and treatments,

Economic Report of the President | 3

and, most importantly, we have developed and manufactured gold standard
vaccines in record time. The pandemic may have begun in China, but we are
ending it in America.
Thanks to the pro-growth policies of my Administration, our Nation’s
economy has exceeded expectations at every turn, and despite the economic
shock due to the China Virus, our great American comeback is well underway.

The Renaissance of American Greatness: Rebuilding Our
Country
For decades, political leaders and privileged elites worked to silence American
workers and families even as opportunity—and with it hope—slipped from the
shores of our great Nation. Almost 5 million manufacturing jobs and 60,000
factories fled our country following the establishment of Permanent Normal
Trade Relations with China in 2000. Massive tax burdens and overregulation
encouraged businesses to invest elsewhere. For decades, multinational corporations flooded our Nation with imported goods, stripping millions of American
families of their livelihoods and their dignity. Decades of these damaging
policies led to the prevalence of “Made in China,” as China’s leaders (and those
from other countries) took advantage of establishment politicians who did not
have the best interests of American workers at heart.
Before I took office, politicians and their adherence to a globalist doctrine
converted our borders and national sovereignty into mere negotiable concepts
to be traded away or simply ignored when in conflict with establishment interests. Anti-American ideology flooded into schools, universities, and media,
while American wealth, intellectual property, and innovation rushed out of our
country. Cities corroded by years of neglect and mismanagement became commonplace, each complete with an allotment of lawless streets that were devoid
of prospect, educational choice, and liberty. International financial crises
became matters not of “if” but of “when,” yet the attention of those charged
with governing turned elsewhere and the American people were forgotten.
Since taking office, rather than apologize for America, I have stood
up for America. From day one of my Presidency, I have put America First,
and I have fought for the American worker harder than anyone ever has. My
Administration has adhered to the two simple rules of “Buy American” and
“Hire American,” we have built the most secure border in history, and I took the
toughest-ever action to stand up to China.
I have worked every day to restore promise to our Nation through an economic agenda that lifts up all Americans. In just 3 years, my Administration’s
policies brought more than 6.6 million people out of poverty; created prosperity through record low unemployment rates for Black Americans, Hispanic
Americans, Asian Americans, and those without a college degree; reduced
homelessness among the general population and our veterans; and saved

4 |

Economic Report of the President

thousands of lives by stemming the tide of opioid-related deaths. We committed to breaking a cycle that for too long held children’s education hostage
on the basis of affluence and class background—denying children knowledge
and unrealized potential. And my Administration returned economic freedom
to the American people as we have cut nearly eight regulations for every new,
significant rule—weakening the power of the regulatory state and stifling
stealth taxation. The power of fracking has forged the path for American energy
independence and delivered personal prosperity alongside national security,
contributing to a 10 percent decline in the global price of oil. We also created
the U.S. Space Force—the sixth branch of the military—and have given new
meaning to “Peace through Strength” by expanding our capabilities and restoring American leadership in space.
We unleashed record prosperity at home, while also negotiating fair
and reciprocal trade agreements. The passage of the Tax Cuts and Jobs Act
increased wages for blue-collar workers, and the implementation of the United
States–Mexico–Canada Agreement elevated American competitiveness with
respect to our regional trading partners. We took the toughest, boldest, and
strongest actions against China in American history, and the United States is
now collecting billions of dollars in tariff revenue on imports of Chinese goods.
I took unprecedented action to reduce drug prices and ensure that
Americans never pay more for life-saving medicines than consumers in other
countries. And my Administration took action to end surprise medical billing. We also eliminated the harmful individual mandate from the so-called
Affordable Care Act, as between 1.2 and 4.6 million Americans gained employment sponsored health coverage from 2018 to 2019. These actions, along with
countless other taken by my Administration, have not only boosted economic
growth and wage gains for all Americans, but they have also protected the
American people from foreign competitors trying to take advantage of them.

A Great American Comeback Underway
The virus from China required us to close up the greatest economy in the history of the world. Understanding the risks our Nation faced, I took bold action
to ban travel from China and then later Europe, saving countless American
lives.
In a matter of days and weeks, the global economy ceased to exist as
we knew it. Nations around the world locked down as uncertainty generated
tremendous fear. In order to prepare our frontline responders in hospitals
and health facilities across the United States, we prioritized the safety of the
American people over the strength of our economy. In March of this year, we
implemented an initial plan to slow the spread of the virus, in coordination with
governors across the Nation. During that time, my Administration facilitated
the delivery of thousands of ventilators and millions of gloves, masks, and

Economic Report of the President | 5

protective gear to States and territories, working to get Americans life-saving
medical equipment.
We promised that no patient suffering from the virus would have to
pay for their treatment out of pocket, and we provided billions of dollars to
hospitals and healthcare providers so that uninsured patients would have
access to critical care. By the end of March, the Food and Drug Administration
issued Emergency Use Authorizations to fast-track more than 20 diagnostic
tests and life-saving treatments. Thanks to our efforts, the case fatality rate
today is more than 85 percent lower than its April peak. Meanwhile, Operation
Warp Speed has harnessed the innovation of the private sector to develop and
manufacture millions of doses of life-saving vaccines, decreasing the average
development time from 3 years to less than 9 months. This record-shattering
work has cleared the path for an end to this cruel pandemic, saving millions
of lives around the world and trillions of dollars in health and other economic
costs.
Through no fault of their own, Americans from all walks of life have been
forced to confront the Invisible Enemy. To help the Nation through this difficult time, I championed and signed four pieces of legislation. These laws kept
Americans connected to their jobs and reduced the economic harm to families
and workers. In March, the Coronavirus Aid, Relief, and Economic Security
(CARES) Act—the largest piece of economic relief legislation in history—authorized direct payments to citizens, expanded unemployment insurance, and
deferred loans for those who needed it most. When Congress later abdicated its
duty to expand and extend this relief, for short-sided political gain, I signed four
executive actions to continue providing for the families, students, and workers
of our country. Additionally, my Administration worked hand-in-hand with the
private sector, invoking the Defense Production Act and related authorities
more than 100 times to surge production and distribution of ventilators, protective equipment, and other materials, including therapeutics and vaccines.
The Paycheck Protection Program, a core piece of the CARES Act, saved
or supported more than 51 million American jobs by providing more than 5.2
million critical loans to small businesses. These loans were crucial lifelines to
business owners and their employees that averted widespread bankruptcies.
This effort alone meant that more than 80 percent of March and April layoffs
were temporary, as over 80 percent of these businesses that received them are
still open today. In addition, Economic Impact Payments sent to more than 159
million Americans surged liquidity to every corner of our Nation, and provided
nearly three months of income to households in the bottom 10 percent of the
income distribution. And our Farmers to Families Food Box Program delivered
more than 90 million boxes to families, children, and businesses, protecting
millions of Americans from food insecurity.

6 |

Economic Report of the President

For the last 4 years, I have fought for you, the American people, in all that I
do. The Economic Report of the President that follows describes the policies that
have made our country so successful and lays out steps we can take to continue the great American comeback in response to the China Virus. This year’s
Report is a testament to the resolve of the American people, who never falter in
the face of adversity, and whose courage and relentless drive forge our destiny.

The White House
January 2021

Economic Report of the President | 7

The Annual Report
of the

Council of Economic Advisers

Letter of Transmittal
Council of Economic Advisers
Washington, January 15, 2021
Mr. President:
The Council of Economic Advisers herewith submits its 2021 Annual
Report in accordance with the Employment Act of 1946, as amended by the Full
Employment and Balanced Growth Act of 1978.
Sincerely yours,

Rachael Seidenschnur Slobodien
Chief of Staff

Economic Report of the President | 11

Introduction
In 2020, the U.S. economy experienced its worst macroeconomic shock since
the Great Depression. As a direct result of the arrival of COVID-19—and consequent measures to contain and mitigate viral transmission, real output was
on pace to contract by as much as 12.3 percent in 2020, which would have
constituted the worst economic contraction since 1932. Professional forecasters projected that the unemployment rate would reach as high as 25.0 percent
in May 2020, its worst level since the Great Depression and more than twice its
peak in the aftermath of the 2008–9 global financial crisis. The Congressional
Budget Office (CBO) forecasted a contraction of almost 6 percent during the
four quarters of 2020, and that the unemployment rate would remain over 11
percent through the end of the year.
In the face of this exogenous economic shock of historically unprecedented scale and speed that abruptly terminated the U.S. economy’s record
expansion, the Trump Administration responded with equally unprecedented
scale and speed. As a result of this response, real gross domestic product (GDP)
in the third quarter of 2020 was down 3.5 percent from its prepandemic level—
less than half the drop in the early projections—and high-frequency forecasts

$"0- Ҋрѵ/0' 1 -.0.4спсп-*% /$*)
/0'
Dollars (billions, chained $2012)
ршѶфпп

4спсп
спспѷу

ршѶппп
рчѶфпп
рчѶппп
рцѶфпп
рцѶппп
рхѶфпп
спршѷр
спршѷт
спспѷр
спспѷт
*0- .ѷ 0- 0*!*)*($)'4.$.Ѹ*)"- ..$*)'0" /!!$ җҘѸ'0 #$+Ѹ
'0'/$*).ѵ
*/ ѷспспѷу..0( ./# *1ѵспѶспсп'0 #$+*). ).0.!*- ./!*-- '
"-*2/#ѵ

Economic Report of the President | 13

$"0- Ҋсѵ/0'4-*''(+'*4( )/1 -.0. 4спсп
Projection
/0'

Total nonfarm employees (millions)
рфф

4спсп

спспѷу

рфп
руф
руп
ртф
ртп
рсф
спршѷр
спршѷт
спспѷр
спспѷт
*0- .ѷ0- 0 *! *-//$./$.Ѹ*)"- ..$*)'0" /!!$ җҘѸ
'0'/$*).ѵ
*/ ѷ# спспѷу' 1 '$. /# 1 -" *!/# /* -)*1 ( -' 1 '.ѵ

for the fourth quarter imply a calendar-year decline of 2.3 percent—less than
one-third the projected decline (figure I-1). In seven months, the U.S. labor
market recovered 12.3 million jobs, or 56 percent of job losses in March and
April (figure I-2). The unemployment rate declined from a peak of 14.7 percent
in April to 6.7 percent in November, almost 5 percentage points below the yearend unemployment rate projected by the CBO in May. After peaking at 22.8
percent in April, U-6, the broadest measure of labor market underutilization,
had declined to 12.0 percent, a level lower than that of July 2014, more than
five years into the previous recovery. Aided by unparalleled fiscal support for
households, by July 2020 retail and new and existing home sales had regained
their prepandemic levels.
The COVID-19 pandemic brought to an end the longest economic
expansion in recorded U.S. history—which, for the first time since the 2008–9
financial crisis, was exceeding expectations and delivering real economic
gains across the income and wealth distributions (figure I-3). In the three years
before the pandemic, the U.S. economy added 7 million jobs—5 million more
than projected by the nonpartisan CBO in August 2016. In the first 2 months
of 2020 alone, the U.S. economy added more jobs (465,000) than the CBO projected would be created in the entire 12 months of 2020 (figure I-4).
Through 2019, real median household income rose $6,000—more than
five times total gains under the preceding eight years—while wage, income,
and wealth inequality declined, and the wage gap between African Americans
14 |

Economic Report of the President

$"0- Ҋтѵ/0'1 -.0.0"0./спрх-*% /$*)
/0'

0"0./спрх

Dollars (billions, chained $2012)
ршѶфпп

спршѷу

ршѶппп

рчѶфпп

рчѶппп

рцѶфпп
спрхѷр спрхѷт спрцѷр спрцѷт спрчѷр спрчѷт спршѷр спршѷт
*0- .ѷ0- 0 *!*)*($)'4.$.Ѹ*)"- ..$*)'0" /!!$ җҘѸ
'0'/$*).ѵ

$"0- Ҋуѵ/0'4-*''(+'*4( )/1 -.0.0"0./спрх
Projection
/0'

Total nonfarm payrolls (millions)
рфт

0"0./спрх

спспѷр

рфр

руш
руц

руф
рут
спрхѷр
спрцѷр
спрчѷр
спршѷр
*0- .ѷ0- 0 *! *-//$./$.Ѷ*)"- ..$*)' 0" /!!$ җҘѶ
'0'/$*).ѵ
*/ ѷ # спспѷр' 1 '..0( ./# )0-4– -0-41 -" + ѵ

спспѷр

Economic Report of the President

| 15

$"0- Ҋфѵ )$/$') (+'*4( )/ ).0-) '$(.Ѷспсп
Initial UI claims, regular programs (millions)

/s$")

*1ѵсч

ц
х
ф
у
т
с
р

- / ..$*) &
)ѵр

-ѵр

4р

*0- .ѷ +-/( )/*! *-Ѹ'0'/$*).ѵ

0'ѵр

+ѵр

*1ѵр

and White Americans narrowed. After landmark tax reform in 2017, real wealth
for the bottom 50 percent of households had risen three times faster than
that of the top 1 percent, while real wages for the bottom 10 percent grew
almost twice as fast as for the top 10 percent—marking stark reversals from
the preceding expansion, when wage, income, and wealth inequality all rose.
Although wealth rose across the income distribution, the bottom 50 percent’s
share of real net worth increased—while that of the top 1 percent decreased,
labor’s share of income rose, and capital’s share decreased. In February 2020,
just before the pandemic hit in force, the unemployment rate declined to 3.5
percent—its lowest level in more than 50 years, and a full 1.5 percentage points
below the CBO’s final 2016 forecast.
COVID-19 constituted an exogenous shock that abruptly terminated this
record expansion, though it was met with a similarly swift policy response.
Within a week of the first reported COVID fatality, Congress passed, and
President Trump signed into law, the Coronavirus Preparedness and Response
Supplemental Appropriations Act. Within four weeks, the President signed into
law two more pieces of economic legislation, including the Coronavirus Aid,
Relief, and Economic Security (CARES) Act, which provided $2.2 trillion in direct
financial support to American firms, households, medical facilities, and State
and local governments. These historic policy responses to the adverse shock
of COVID-19, as well as the historic strength of the pre-COVID U.S. economy,

16 |

Economic Report of the President

$"0- Ҋхѵ ) 3*! *0. #*' )*( 4 - )/$' Ѷспсп
$/#- '$ !
Index of household income (100=February 2020)
сфп

$/#*0/- '$ !

спп

рфп

рпп

+4
0)
0'4
0"0./

фп

рп

сф
фп
Household income percentile

цф

шп

*0- .ѷ ).0.0- 0Ѹ0-- )/*+0'/$*)0-1 4Ѹ*)/#'4)))0' *$')*)*($
0++' ( )/Ѹ'0'/$*).ѵ
*/ ѷ җспспҘ!*-( /#**'*"4ѵ

mitigated what was on pace to be a macroeconomic contraction on par with
the Great Depression. In particular, measures designed to preserve employeremployee relationships, most significantly the Paycheck Protection Program
(PPP) and employee retention tax credit, played a key role in allowing firms
to retain workers on leave. By limiting eligibility to small and medium-sized
enterprises, the PPP targeted aid to those employers that were most at risk of
needing to terminate employees (figure I-5).
Meanwhile, income replacement and cost mitigation helped to cushion
the shock to household incomes and thereby facilitated stabilization and
recovery of consumer spending, which alone constitutes 70 percent of the U.S.
economy. Federal assistance programs, including expanded and enhanced
Unemployment Insurance and Economic Impact Payments to households
earning below set income thresholds, largely offset declines in household compensation due to economic shutdowns. Income replacement rates were highest
at the lower end of the income distribution, indicating that relief was targeted
toward households that were more vulnerable to an adverse income shock (figure I-6). Upon expiration of these provisions and in the absence of Congressional
action, the Trump Administration extended further relief through four executive
actions, providing supplemental payments through lost wages assistance to
unemployed Americans, temporary payroll tax relief, and extended relief and
protection for student borrowers and renters at risk of eviction.

Economic Report of the President | 17

$"0- Ҋцѵ *- -& / *1 -4*(+-$.*).4 ..$*).
Level of employment (index, prerecession peak = 100)
рпс
ршчп

рпп

ршшп–шр

сппр

сппч–ш

шч
шх

шу
шс
спсп

шп

чч
чх

чу
п

рп

рф сп сф тп тф уп уф фп фф
Months from prerecession employment peak

хп

хф

цп

цф

*0- .ѷ 0- 0*! *-//$./$.Ѹ/$*)'0- 0*!*)*($ . -#Ѹ'0'/$*).ѵ

Though the pace of the recovery vastly exceeded expectations and constituted the most rapid economic recovery on record (figure I-7), in the face
of the continuing global COVID-19 pandemic, U.S. employment and production remained below prepandemic levels at the end of 2020. Recognizing the
remaining challenges to full economic recovery, most importantly the ongoing
COVID-19 pandemic, the Administration supports and consistently supported
additional fiscal relief, including an additional round of the PPP to support
small business payroll retention, an expanded employee retention tax credit, a
continuation of enhanced Unemployment Insurance benefits, a second round
of Economic Impact Payments, targeted aid to schools and State and local
governments, additional nutritional support, and temporary relief to specific
adversely affected industries. At the time of this writing, the U.S. Congress had
not agreed to these measures.
In chapter 1 of this Report, we discuss the historic economic gains in the
United States on the eve of the pandemic, before quantifying the magnitude of
the economic shock that hit the U.S. economy in 2020 and situating this shock
in historical context. In chapters 2, 3, and 4, we then document and estimate
the economic effect of the Administration’s response to the shock, focusing
first on households and labor markets, then on business and financial markets,
and finally on healthcare. We then turn, in the next five chapters, to the role of
pre-COVID-19 Administration policies—specifically in the areas of Opportunity

18 |

Economic Report of the President

Zones, deregulation, school choice, space innovation and exploration, and
international trade—in establishing the foundations for longer-run potential
economic growth. Finally, we review the U.S. economy in 2020 and discuss the
economic outlook, including potential risks, before concluding with a discussion of potential future policies to promote further economic recovery and
subsequent growth.
In chapter 1, we document that the beginning of 2020 ushered in a strong
U.S. economy that was delivering job, income, and wealth gains to Americans
of all backgrounds, with historically low unemployment and poverty along
with record gains in median income for workers across the socioeconomic
spectrum. Moreover, the robust state of the economy led forecasters to expect
healthy growth through 2020 and beyond. However, the arrival of COVID-19,
with origins in the People’s Republic of China, brought with it an unprecedented economic and public health crisis. This chapter describes the health
of the pre-COVID Trump economy and the nature of the economic shock from
COVID-19, and gives an overview of the swift and bold fiscal response undertaken by the Trump Administration to provide relief and lay the foundation for
the most rapid economic rebound to date in modern U.S. history.
We analyze these issues in greater depth in chapter 2. We find that the
Trump Administration’s pro-growth policies contributed to substantial gains
for U.S. households between 2016 and 2019. Median net worth increased by
18 percent, median income increased by 9.7 percent, and poverty reached
a record low. As the COVID-19 pandemic brought the historic expansion to a
potentially catastrophic halt, the Trump Administration helped protect the
livelihoods of Americans through legislation and executive actions. Even as the
unemployment rate climbed from a 50-year low of 3.5 percent in February 2020
to a record high of 14.7 percent in April, household incomes increased across
the distribution, especially for lower-income households, thanks to Economic
Impact Payments and expanded and enhanced Unemployment Insurance.
Protections against evictions and student loan defaults helped keep people
in their homes and out of default. The ultimate success of these efforts will
depend on how quickly the economy recovers. Between April and November,
the unemployment rate fell by 8.0 percentage points, the fastest six-month
decline on record, paving the way to attaining the same strong economy that
prevailed during the first three years of the Trump Administration.
Chapter 3 analyzes the effects of the economic policy response on businesses and employer-employee ties. The CARES Act, which was signed into
law by President Trump only two weeks after he issued a National Emergency
Declaration, provided record economic relief to families and businesses to
mitigate the shock from the COVID-19 crisis. In total, the fiscal response to
COVID-19 stands out as the most rapid and robust crisis-related economic
policy mobilization in the post–World War II era. Two central objectives have
constituted the Trump Administration’s approach to combating the economic
Economic Report of the President | 19

fallout from COVID-19: the alleviation of financial distress to reduce hardship,
and the preservation of underlying economic health to facilitate a faster recovery. Ensuring the vitality and resilience of small businesses plays an essential
role in achieving these objectives. This chapter describes the fiscal relief provisions aimed at helping small businesses and their workers—principally the
PPP—and their success in fueling what has been thus far the fastest employment and GDP rebound in U.S. history.
In chapter 4, we examine how the COVID-19 pandemic constituted a
rapidly evolving health and economic crisis in the healthcare sector and for
working families across the Nation. The Trump Administration’s response
to address this multifaceted crisis involved a complementary two-pronged
policy approach. First, by enacting several pieces of bipartisan legislation, the
Administration secured significant funding to alleviate the financial burden
experienced by hospitals, offered tax credits to private employers with fewer
than 500 employees to enable them to provide emergency paid family and
sick leave for their workers, and fully covered the cost of COVID-19 testing and
treatment for many low-income and uninsured individuals. Second, through
a series of deregulatory actions, the Administration expanded the use of
telemedicine for both COVID-19 screenings and many other health concerns,
supported the relaxation of occupational licensing requirements for nurse
practitioners, issued Emergency Use Authorizations for COVID-19 diagnostic
tests, and accelerated the development, authorization, and deployment of
therapeutics and vaccines for COVID-19. This included the Administration’s
Operation Warp Speed, which the CEA estimates could result in as much as $2.4
trillion in economic benefit through the accelerated availability of an effective
vaccine. This chapter explores the various effects of these healthcare policy
innovations and achievements, some of which are likely to pay dividends after
the COVID-19 pandemic is resolved.
Beginning with chapter 5, we discuss the role of pre-COVID Administration
policies in establishing the foundations for longer-run potential economic
growth. The Tax Cuts and Jobs Act of 2017 not only broadly lowered taxes for
businesses and individuals but also made targeted cuts to spur investment in
economically distressed communities designated as Opportunity Zones (OZs).
This chapter compares the advantages of OZs with those of other Federal antipoverty programs and documents the characteristics of the nearly 8,800 lowincome communities designated as OZs. The CEA finds that $75 billion has been
invested in funds for OZs, and that this investment is already benefiting OZ residents and potentially having only a small effect on the current Federal budget.
In chapter 6, we revisit the issue of economic regulation. During the Trump
Administration, Federal agencies have demonstrated a sustained commitment
to regulatory reform. As a result, the Administration’s regulatory efforts have
helped reduce red tape for small businesses and middle-income households.
One of the most important deregulatory actions the Administration finalized
20 |

Economic Report of the President

in 2020 is the Safer Affordable Fuel Efficient (SAFE) Vehicles Rule, which we
estimate will lead to an increase in real incomes, and raise GDP by $53 billion
annually, or about 0.3 percent. The CEA also finds that the benefits of deregulations, such as the SAFE Vehicles Rule, tend to favor the lower income quintiles,
suggesting that lower-income households may have benefited the most, relative to household income, from the Administration’s deregulatory actions.
In chapter 7, we examine the topic of school choice. During the past 30
years, school choice programs have undergone dramatic expansion in the
United States. These programs—organized at the Federal, State, and local
levels—share a common goal of expanding access to education options that
exist alongside and ultimately improve public school options for primary and
secondary education. The programs have altered primary and secondary education in fundamental ways by increasing competition in the school system
and by enhancing educational opportunities for all students, especially those
from disadvantaged groups. We document the development and expansion of
school choice programs and discuss the role of Federal policy, including recent
actions by the Trump Administration to further this expansion. We explain
how educational competition empowers families and incentivizes schools to
deliver more value, and we document the growing empirical evidence that
carefully crafted school choice programs do improve educational outcomes
for all students.
Chapter 8 analyzes important developments in a frontier of economic
potential, namely, innovation and opportunity in the space economy. We
review advancements in spaceflight and space policy made during the past
year, including the first commercial human spaceflight in history and implications for the private sector’s role in the space economy. We also discuss
the role of the Administration’s policies—specifically the Executive Order
on “Encouraging International Support for the Recovery and Use of Space
Resources” and the Artemis Accords—in strengthening investor confidence
in the space economy and thereby enabling expansion of the private space
sector. After an extensive review of the economic theory of property rights
and the empirical property rights literature, we find substantial evidence that
improving investor expectations in a novel economic sector such as space
increases investment in that sector. In addition, we estimate that private
space investment could as much as double in the next eight years, due to the
Administration’s executive actions and other enhancements of property rights
in space.
In chapter 9, we examine how the Administration has promoted U.S.
interests in international trade by forging new bilateral trade agreements
with China, Japan, and South Korea, and reshaped regional trade by modernizing the trade agreement with our most important trading partners, Canada
and Mexico. The United States–Mexico–Canada Agreement achieves new
safeguards for U.S. interests across a range of areas including digital services,
Economic Report of the President | 21

intellectual property, and labor protections. These agreements go well beyond
formal tariff barriers that have been the focus of past trade agreements by
addressing structural and technical barriers to free and fair trade. We also
review how the COVID-19 pandemic reduced international trade overall and
has brought into focus underappreciated risks of global supply chains.
In chapter 10, we build on chapters 1, 2, and 3 by summarizing the main
macroeconomic developments of 2020, and discuss the economic outlook for
the years ahead, with particular attention to upside and downside risks. We
find that though the U.S. economy in 2020 was hit with the biggest adverse
macroeconomic shock since the Great Depression—with effects on output,
labor, capital, housing, and energy markets all of historic magnitudes—the
subsequent recovery to date has also been of historic speed, breadth, and
magnitude. We highlight that though official and private forecasters currently
project continued strong recovery in 2021—aided by an unprecedented economic policy response in 2020, a strong pre-COVID economy, and the availability of vaccines through Operation Warp Speed—substantial risks remain,
including both pandemic and policy risks.
In the near term, the single greatest downside economic risk is rising
COVID-19 cases before the widespread availability of vaccines, and the policy
and behavioral responses to viral resurgences. Already, in December 2020,
several State and local governments have reimposed shelter-in-place orders
in response to rising cases in November and December. For this reason, the
Administration continues to articulate support for additional fiscal measures
to provide a bridge to the widespread availability of vaccine candidates developed under Operation Warp Speed. Over the longer term, failure to maintain
or implement the types of pro-growth policies discussed in this Report and
in the 2018, 2019, and 2020 editions of the Economic Report of the President
would constitute additional potential downside risks. But the continuation and
expansion of the Administration’s pro-growth policies in support of full labor
market recovery offer the upside potential for a rapid return to the levels of
employment, production, and real income growth that prevailed on the eve of
the pandemic.
We conclude this Report, in chapter 11, by reviewing a collection of policy
areas highlighted by the COVID-19 pandemic, and we analyze reforms that
might meet the ongoing economic challenges faced by the United States. In
particular, we review potential policies to strengthen connections to the labor
force, support a balance between work and family, advance international coordination to address 21st-century challenges, create a more effective healthcare
system, build a dynamic economy through infrastructure improvement, and
generate a more skilled and resilient workforce. We find that solving these challenges can ensure that the United States not only recovers to its prepandemic
levels of prosperity but also builds a fairer, more dynamic, and more resilient
economy that benefits all Americans.
22 |

Economic Report of the President

Contents
Part I: Confronting the Largest Postwar Economic Shock:
The Federal Response to Mitigate the COVID-19 Pandemic. . . . . . . . 33
Chapter 1: Creating the Fastest Economic Recovery. . . . . . . . . . . . . . . . . . . . . . 35
The Historic Strength of the U.S. Economy before COVID-19. . . . . . . . . . . . . . .
The Early Economic Effects of COVID-19.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The U.S. Economy’s Resilience in Weathering the COVID-19 Shock. . . . . . . . . .
Comparing the COVID-19 Recession and the Great Recession. . . . . . . . . . . . . .
The State of the Economy before the Crises. . . . . . . . . . . . . . . . . . . . . . . . . .
The Origins and Progression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fiscal and Monetary Responses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Federal Support for Low-Income Households. . . . . . . . . . . . . . . . . . . . . . . .

39
45
51
52
53
55
56
60

Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Chapter 2: Prioritizing America’s Households. . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
The Strength of the Pre-COVID Economy and the COVID-19 Shock. . . . . . . . . .
Policy Responses Providing Household Relief. . . . . . . . . . . . . . . . . . . . . . . . . . .
March 2020 Legislative Acts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
President Trump’s Executive Actions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

72
74
74
76

The Impact of Policies in Providing Relief to Households. . . . . . . . . . . . . . . . . . 77
Spurring a Return to the Pre-COVID Economy. . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

Chapter 3: Assisting Entrepreneurs and Workers through Aid to Businesses. 89
Summary of Policies to Assist American Businesses and Their Workers. . . . . . 91
The Paycheck Protection Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Employee Retention Tax Credits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Economic Injury Disaster Loans and Advances. . . . . . . . . . . . . . . . . . . . . . . 92
The Federal Reserve’s Lending Facilities. . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Other Programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Measuring Small Business’s Utilization of Selected CARES Act Business
Provisions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
Comparing Expected and Actual Small Business Bankruptcies during the
COVID-19 Pandemic. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
The CARES Act’s Role in Facilitating Small Business Survival. . . . . . . . . . . . . . 101
The Coronavirus Food Assistance Program’s Impact on Farm Incomes. . . . . . 105
The Pandemic’s Impact on the Financial Sector and Lending Facilities’ Role . 106
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

Chapter 4: Advancing the Quality and Efficiency of America’s
Healthcare System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Expediting Research and Development for Novel Therapies and Tests for
COVID-19.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Economic Report of the President | 23

Emergency Use Authorizations.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
Operation Warp Speed. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Supporting the Healthcare System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Deregulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Financial Support for Healthcare Providers. . . . . . . . . . . . . . . . . . . . . . . . . 124
Subsidizing Beneficial Behaviors and the Cost of COVID-19 Care. . . . . . . . . . . 127
Emergency Paid Sick and Medical Leave. . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Subsidizing the Cost of COVID-19 Care. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
COVID-19 and Future Healthcare Reform. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
FDA Reform.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Telemedicine Deregulation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Scope-of-Practice Deregulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Additional Changes to Promote Choice and Competition. . . . . . . . . . . . . .

129
130
134
137
140

Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

Part II: The Renaissance of American Greatness. . . . . . . . . . . . . . . . . . 143
Chapter 5: Assessing the Early Impact of Opportunity Zones.. . . . . . . . . . . . . 145
Comparing Opportunity Zones with Other Antipoverty or Place-Based
Programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Antipoverty Transfer Policies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Federal Place-Based Policies: The New Markets Tax Credit Program. . . . .
Other Federal Place-Based Development Programs.. . . . . . . . . . . . . . . . .

149
149
150
152

Characteristics of Opportunity Zones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
The Opportunity Zone Selection Process .. . . . . . . . . . . . . . . . . . . . . . . . . . 153
The Economic State of Opportunity Zones. . . . . . . . . . . . . . . . . . . . . . . . . . 154
Opportunity Zones’ Effect on Total Investment. . . . . . . . . . . . . . . . . . . . . . . . . 157
Capital Raised by Qualified Opportunity Funds. . . . . . . . . . . . . . . . . . . . . . 157
Estimated Investment Growth Caused by the Opportunity Zone Incentive . 160
The Industry Focus of Qualified Opportunity Funds.. . . . . . . . . . . . . . . . . . 162
Opportunity Zones’ Effects on Business Investment and Housing Values. . . . 163
Equity Investments in Opportunity Zone Businesses. . . . . . . . . . . . . . . . . . 163
Opportunity Zone Designation and Housing Values . . . . . . . . . . . . . . . . . . 164
Opportunity Zones’ Effects on Poverty and the Budget. . . . . . . . . . . . . . . . . . 168
Projected Effects of Opportunity Zones on Poverty. . . . . . . . . . . . . . . . . . . 168
Budgetary Effects of Opportunity Zones. . . . . . . . . . . . . . . . . . . . . . . . . . . 168
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

Chapter 6: Empowering Economic Freedom by Reducing
Regulatory Burdens.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Regulation in Review. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Safer Affordable Fuel Efficient (SAFE) Vehicles Rule. . . . . . . . . . . . . . . . . .
GHG Credit Transaction Data. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Estimating the Curve for the Marginal Cost of Compliance . . . . . . . . . . . .

176
178
180
182

The Potentially Regressive Nature of Regulation. . . . . . . . . . . . . . . . . . . . . . . . 186
Progressive and Regressive Tax Structures . . . . . . . . . . . . . . . . . . . . . . . . . 186
The Harm Regressive Regulation Systems Pose. . . . . . . . . . . . . . . . . . . . . 188
Lower-Income Households Often Gain the Most from Regulatory Reform.. . . 193

24 |

Economic Report of the President

The Regressivity of Federal Regulation Offsets the Progressivity of Federal
Taxes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198

Chapter 7: Expanding Educational Opportunity through Choice
and Competition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Overview of School Choice Programs and Federal Policy. . . . . . . . . . . . . . . . .
Private Choice Programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Public Choice Programs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Examples of School Choice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Role of Federal Policy in School Choice. . . . . . . . . . . . . . . . . . . . . . . . .

203
204
206
207
210

School Choice and Competition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
Competition between School Districts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
Designing School Choice Systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Evidence on the Impact of School Competition.. . . . . . . . . . . . . . . . . . . . . . . . 217
Direct Value-Added Effects.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Indirect Procompetitive Effects.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222

Chapter 8: Exploring New Frontiers in Space Policy and Property Rights .. . 225
Current Issues in Space Policy and the Space Economy. . . . . . . . . . . . . . . . . .
Space Policy Developments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Economics of Property Rights. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Historical Examples of Property Rights Evolution. . . . . . . . . . . . . . . . . . . .
Investment Responses to Property Right Enhancement. . . . . . . . . . . . . . .

228
233
235
237
238

The Effects of Policies on Investment in Space Industries.. . . . . . . . . . . . . . . .
Looking Ahead. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Flag of Choice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Incentivizing the Private Sector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

242
245
245
246

Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249

Chapter 9: Pursuing Free, Fair, and Balanced Trade. . . . . . . . . . . . . . . . . . . . . . 251
The Phase One Agreement with China. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Background.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Major Provisions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Intellectual Property. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Technology Transfer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Agriculture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Financial Services. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Currency. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chinese Purchase Commitments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Dispute Resolution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
What Is Not in Phase One?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

254
254
256
256
257
257
259
259
260
261
262

The United States–Mexico–Canada Agreement. . . . . . . . . . . . . . . . . . . . . . . . .
Rules of Origin for Automobile Production.. . . . . . . . . . . . . . . . . . . . . . . . .
Digital Trade. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Intellectual Property Protection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Labor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

263
263
264
264
265

Economic Report of the President | 25

Reform of the Investor-State Dispute Mechanism . . . . . . . . . . . . . . . . . . . .
Agricultural Provisions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Trade Facilitation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Overall Economic Effects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

265
266
267
267

Other Trade Agreements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
U.S.-Japan Trade Agreements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The U.S.-South Korea Free Trade Agreement. . . . . . . . . . . . . . . . . . . . . . . .
Limited Trade Agreements with Brazil and Ecuador. . . . . . . . . . . . . . . . . .
U.S.-U.K. Negotiations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

268
269
269
270
270

The Rise of Global Supply Chains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
China and the Emergence of Global Supply Chains.. . . . . . . . . . . . . . . . . .
U.S. Firms Begin to Hedge the Risks of Global Supply Chains. . . . . . . . . . .
The Trade Slowdown in Response to COVID-19. . . . . . . . . . . . . . . . . . . . . .
The Decline in Imports of Intermediate Goods.. . . . . . . . . . . . . . . . . . . . . .
Evidence That Firms Are Reducing Their Exposure to China. . . . . . . . . . . .
Drivers of Shifting Supply Chains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

271
271
272
272
276
277
281

Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

Part III: An Effort to Rebuild Our Country. . . . . . . . . . . . . . . . . . . . . . . . . 285
Chapter 10: The Year in Review and the Years Ahead. . . . . . . . . . . . . . . . . . . . . 287
The Year in Review. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Components of Economic Output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Labor Market. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Inflation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Housing Market. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Financial Markets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Interest Rates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Oil Markets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Global Macroeconomic Situation. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

288
288
293
296
296
300
302
303
303

The Future Economic Outlook.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Forecasts from the Blue Chip Consensus, the Congressional Budget
Office, and the Federal Reserve Open Market Committee. . . . . . . . . . . . . .
Economic Objectives and Policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Near-Term Upside and Downside Risks. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Long-Term Upside and Downside Risks. . . . . . . . . . . . . . . . . . . . . . . . . . . .

305
307
307
311
314

Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

Chapter 11: Policies to Secure Enduring Prosperity. . . . . . . . . . . . . . . . . . . . . . 319
Strengthening Connections to the Labor Force. . . . . . . . . . . . . . . . . . . . . . . . .
Dual-Earner Couples.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Marriage Penalty and the Second-Earner Penalty. . . . . . . . . . . . . . . .
Tax Reforms to Mitigate the Second-Earner Penalty and Boost the
Labor Supply.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
High Marginal Rates for Low Earners. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

321
322
324
327
328

Supporting a Balance between Work and Family. . . . . . . . . . . . . . . . . . . . . . . 334
Unequal Access to Paid Family and Medical Leave. . . . . . . . . . . . . . . . . . . 336
Effects of Paid Leave on Employment and Earnings. . . . . . . . . . . . . . . . . . 338

26 |

Economic Report of the President

Implementation of Paid Leave. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
The Lack of Childcare. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
Increasing Access to Childcare. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
Enhancing International Coordination to Meet 21st-Century Challenges. . . .
The Economics of Networks, Coordination, and Standard Setting. . . . . . .
The Current Paradigm for Coordination and Standard Setting.. . . . . . . . .
Opportunities for Advancing Coordination. . . . . . . . . . . . . . . . . . . . . . . . .
Prospects for U.S.-U.K. Coordination. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

345
346
348
351
352

Creating a More Effective Healthcare System.. . . . . . . . . . . . . . . . . . . . . . . . . .
Rationalizing the Provision of Healthcare Professionals. . . . . . . . . . . . . . .
Balance Billing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Medicare Inpatient Rates. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

353
354
356
358

Building a Dynamic Economy through Infrastructure Improvement. . . . . . . .
The Federal Government’s Role in Infrastructure Investment. . . . . . . . . . .
Infrastructure and Productivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Costs of Building Infrastructure.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Critical Importance of User Fees. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Investing in Port Infrastructure.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

359
359
362
365
366
373

Generating a More Skilled and Resilient Workforce. . . . . . . . . . . . . . . . . . . . . .
Points-Based Immigration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Estimated Economic Benefits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Estimated Effects on the Wages of Domestic Workers. . . . . . . . . . . . . . . . .
Estimated Effects on Government Revenue and Expenditures. . . . . . . . . .
Improving Postsecondary Education and Skill Development. . . . . . . . . . .
Increasing Incentives for Schools to Improve the Economic Gains of
Students. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Improving Support for Educational Programs That Promote Skill
Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

378
378
378
382
384
385
386
388

Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392

References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395

Appendixes
A.
B.

Report to the President on the Activities of the Council of
Economic Advisers During 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 435
Statistical Tables Relating to Income, Employment, and Production . . . . . 447

I-1
I-2
I-3
I-4
I-5
I-6
I-7
1-1

Actual GDP versus May 2020 CBO Projection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Actual Payroll Employment versus May 2020 CBO Projection. . . . . . . . . . . . . . 14
Actual GDP versus August 2016 CBO Projection. . . . . . . . . . . . . . . . . . . . . . . . . . 15
Actual Payroll Employment versus August 2016 CBO Projection . . . . . . . . . . . 15
Initial Unemployment Insurance (UI) Claims, 2020 . . . . . . . . . . . . . . . . . . . . . . . 16
Index of Household Income by Percentile, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . 17
Labor Market Recovery Comparison by Recessions. . . . . . . . . . . . . . . . . . . . . . . 18
Evolution of 2020’s Gross Domestic Product Forecast, 2020. . . . . . . . . . . . . . . 38

Figures

Economic Report of the President | 27

1-2
1-3
1-4
1-5
1-6
1-7
1-8
1-9
1-10
1-11
1-12
1-13
1-14
1-15
1-16
2-1
2-2
2-3
2-4
2-5
2-6
2-7
2-8
2-9
3-1
3-2
3-3
3-4
3-5
3-6
3-7
3-8
3-9
3-10
3-11
28 |

Real Gross Domestic Product Fell and Rose More Sharply Now
Than during the Great Recession. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
The Unemployment Rate versus Preelection Forecasts, 2011–19. . . . . . . . . . 40
Real Median Household Income by Householder Race, 1967–2019. . . . . . . . . 42
Share of Total Net Worth by Percentile, 2007–19 . . . . . . . . . . . . . . . . . . . . . . . . . 43
Poverty Rates by Race and Hispanic Origin, 1959–2019 . . . . . . . . . . . . . . . . . . . 44
U.S. Real Gross Domestic Product, 2014–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Percentage of the U.S. Population under Statewide Restrictions, 2020 . . . . . 46
Retail Spending during the Early Stages of the Pandemic, Seven-Day
Average, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Traffic Congestion during the Early Stages of the Pandemic, Median
across All States, Seven-Day Average, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
Seated Diners in U.S Restaurants, per OpenTable, during the
Early Stages of the Pandemic, Seven-Day Average, 2020. . . . . . . . . . . . . . . . . . 49
Weekly U.S. Hotel Occupancy Rate during the Early Stages of the
Pandemic, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
Percent Change in Small Businesses That Are Open and Hourly
Employees Who Are Working in the Early Stages of the Pandemic, 2020. . . . 50
Unemployment and COVID-19: The Trump Economy versus the 2016
Counterfactual, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
Ratio of Household Sector and Nonfinancial Business Debt to Gross
Domestic Product, 2006:Q4–2020:Q2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Timelines for the Fiscal and Monetary Responses to COVID-19 and the
Financial Crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
Monthly Disposable Personal Income, 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
Index of Household Income by Percentile, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . 79
Index of Household Income for Example Households, 2020 . . . . . . . . . . . . . . . 80
Percent Below Federal Poverty Line, 2019–2020. . . . . . . . . . . . . . . . . . . . . . . . . . 81
Household Receipt and use of Economic Impact Payment, 2020. . . . . . . . . . . 82
Evictions in Selected Cities by Month, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Eviction Filings in Select Cities, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Labor Market Recovery Comparisons by Recessions, 1945–2020. . . . . . . . . . . 85
Private Payroll Job Losses by Sector Since February 2020. . . . . . . . . . . . . . . . . 86
Share of Total PPP Loan Counts and PPP Loan Amounts by Round. . . . . . . . 95
Percentage of Small Businesses Receiving a PPP Loan, through
August 8, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Total Chapter 11 Bankruptcy Filings by U.S. Small Businesses, 2020 . . . . . . . 98
Percent Change in Small Business Chapter 11 Bankruptcies,
FY 2020, from FYs 2017–19 Average . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Small Business Chapter 11 Bankruptcies, FYs 2008–20. . . . . . . . . . . . . . . . . . 100
Actual versus Predicted Small Business Chapter 11 Bankruptcy
Filings, 2007–20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
Change in the Number of Small Business Hourly Employees
Working, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Change in the Number of Small Businesses Open, Seven-Day
Average, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
Change in Small Business Revenue, Seven-Day Average, 2020. . . . . . . . . . . 104
U.S. Net Farm Income, 2016–20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Dow Jones Industrial Average Change during Various Pandemics. . . . . . . . 107
Economic Report of the President

3-12
3-13
4-1
4-2
4-3
4-4
4-5
4-6
4-7
4-8
4-9
5-1
5-2
5-3
5-4
5-5
5-6
5-7
5-8
5-9
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-i
6-8
6-9
6-10
6-11
7-1
7-2
7-3

Market Volatility Index (VIX), 2007–20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
BBB Corporate Bond and 10-Year Treasury Note, Spread, 2006–20. . . . . . . 110
Percentage of U.S. Population under Statewide Restrictions, 2020. . . . . . . 117
Daily Health and Economic Costs of COVID-19 to the United States
If No Vaccine Is Found. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Value of Speeding Up a COVID-19 Vaccine Starting September 1, 2021 . . . 122
National Medicare Utilization, Jan. 10–Oct. 30, 2020. . . . . . . . . . . . . . . . . . . . 126
Monthly Employment by Sector, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Average Drug Life Cycle during Patent Period. . . . . . . . . . . . . . . . . . . . . . . . . . 131
Average Annual Social Surplus by Approval Time Decrease . . . . . . . . . . . . . 133
Cumulative Aggregate Net Present Value of Increased Social Welfare
by Approval Time Decrease. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
State Scope-of-Practice Deregulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
The Geography of Opportunity Zones. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Demographics of Opportunity Zones (OZs), 2012–16 . . . . . . . . . . . . . . . . . . . 155
Average Median Household Income by Census Tract Designation,
2000–16. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Average Median Household Income by Tract Designation and State,
2012–16. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Population by Poverty Rates and Census Tract Designation. . . . . . . . . . . . . 157
Growth in Qualified Opportunity Funds, Novogradac and SEC Data. . . . . . 159
Opportunity Zone (OZ) Investment Supply-and-Demand Model . . . . . . . . . 162
Percentage of Qualified Opportunity Funds, by Industry. . . . . . . . . . . . . . . . 163
Private Equity Investment by Tract Group, 2016–19. . . . . . . . . . . . . . . . . . . . . 165
Significant Final Rules Excluding Deregulatory Actions, by Presidential
Years 2001–19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177
Vehicle Market Equilibrium with GHG Standards. . . . . . . . . . . . . . . . . . . . . . . 179
GHG Credit Market Equilibrium for Various Standards . . . . . . . . . . . . . . . . . . 183
Consumer Savings from the SAFE Vehicles Rule as a Percentage of
Income, by Income Quintile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Illustrative Impact of a 15 Percent Grocery Tax, by Income Quintile . . . . . . 187
The Impact of Regulation in a Competitive Market . . . . . . . . . . . . . . . . . . . . . 189
Percentage of Income Spent on Outlay Category, by Income Quintile . . . . 189
Economically Significant Final Rules with Regulatory Flexibility
Analysis for Small Businesses, by Presidential Years 2001–19. . . . . . . . . . . . 190
Individual Mandate Penalties as a Percentage of Income, by
Income Quintile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Consumer Savings on Internet Access from the Restoring Internet
Freedom Rule, by Income Quintile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
Consumer Savings from Selected Deregulatory Actions and the SAFE
Vehicles Rule, by Income Quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
Consumer Savings from Deregulatory Actions Compared with the
Federal Tax Rate, by Income Quintile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Number of ESAs, Vouchers, and Tax Credit Scholarships, 1996–2020. . . . . 201
Enrollment in Public Choice Programs over Time . . . . . . . . . . . . . . . . . . . . . . 202
Percentage Distribution of Students Age 5–17 Years Attending
Kindergarten through 12th Grade, by School Type, 1999 and 2016. . . . . . . 203

Economic Report of the President | 29

7-4
7-5
8-1
8-2
8-3
8-4
9-1
9-2
9-3
9-4
9-5
9-6
9-7
10-1
10-2
10-3
10-4
10-5
10-6
10-7
10-8
10-9
10-10
10-11
10-12
10-13
10-14
11-1
11-2
11-3
11-4
11-5
11-i
11-ii
11-iii
11-6

30 |

Federal Funding per Student for Public Primary and Secondary
Schools by State, 2016–17 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
Number of Charter Schools Opened, 2006–7 to 2016–17. . . . . . . . . . . . . . . . 212
NASA Outlays and U.S. Private Investment, 2010–19. . . . . . . . . . . . . . . . . . . . 229
Nongovernmental Equity Investment in Commercial Space
Companies, 2010–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Marginal Cost and Benefit of Property Rights Specification. . . . . . . . . . . . . . 237
Nongovernmental Equity Investment in Commercial Space
Companies in the United States, 2010–28 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
U.S. Imports, 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
U.S. Exports, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
U.S. Imports of Goods: Percent Change from Selected
Countries, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
U.S. Imports of Goods from the World by Type of Good, 2020 . . . . . . . . . . . 278
U.S. Manufacturing Import Ratio, 2008–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
U.S. Near-to-Far Ratio, 2009–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
Average Monthly Nominal Wages in U.S. Dollars, 2000–2020. . . . . . . . . . . . . 281
Components of Real GDP, 1990–2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Consumption and Wealth Relative Share of Disposable Personal
Income (DPI), 1952–2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
Components of Investment, 1990–2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Government Purchases as a Share of GDP, 1948–2020. . . . . . . . . . . . . . . . . . 293
Unemployment Rate, 1990–2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Labor Force Participation Rate, 1990–2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Nominal Compensation and Earnings for Private Industry Workers,
2006–20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
Consumer Price Inflation (PCE price index), 2012–20 . . . . . . . . . . . . . . . . . . . 297
U.S. Housing Price Index, 1990–2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
U.S. New Housing Formation and Single-Family Home Sales,
1990–2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
U.S. Homeownership Rate, 1990–2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Percentage of Rent Payments Made, 2019–2020. . . . . . . . . . . . . . . . . . . . . . . . 300
Market Volatility Index (VIX), 2007–20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
Ten-Year Minus Three-Month Treasury Constant Maturity Rate,
1990–2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
Employment-to-Population Ratio, 1960–2020. . . . . . . . . . . . . . . . . . . . . . . . . . 322
Distribution of Joint Filers, 1962–2019 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
Average Tax Rate on the Pretax Wages of Single Persons and
Second-Earners (without Children), 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
Average Tax Rate on the Pretax Wages of Single Persons and
Second-Earners (with Children), 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
The Marginal Effective Rate on Low Earners . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
Pre-TCJA Distribution of MIDs by Adjusted Gross Income. . . . . . . . . . . . . . . . 330
Pre-TCJA Distribution of SALT by Adjusted Gross Income. . . . . . . . . . . . . . . . 332
Homeownership Rates in High-Tax Areas Compared with
ow-Tax Areas, 2018–20 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
Change in Employed Workers with Actual Hours Lower Than
Usual Hours, by Family Type, 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335

Economic Report of the President

11-7
11-8
11-9
11-10
11-11
11-12
11-13
11-14
11-15
11-16
11-17
11-18
11-iv
11-v
11-vi

Access to Paid Leave by Average Wage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Cost of Paid Family and Sick Leave as a Share of State Wages, 2019. . . . . . 341
The Free Movement of Health Care Labor, 2020. . . . . . . . . . . . . . . . . . . . . . . . 355
Percent Change in Medicare FFS DRG Payments Compared with
Medicare Advantage and Private Insurers, Top 25 Diagnosis-Related
Groups, 2015. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359
U.S. Productivity, 1949–2019. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
Number of Seaports by Country among the Top 50 Busiest Seaports
in the World, 2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
Cargo Volumes of the Port of Seattle Compared with the Port of
Prince Rupert, 2015–18. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
Share of Permanent Legal Immigration Based on Employment, 2017 . . . . 379
Estimated Educational Attainment of Recent Immigrants Age 25
Years and Older, 2016–18. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
Employment by Educational Attainment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
Average Federal, State, Local, Institutional, or Other Sources of
Grant Aid Awarded to Undergraduate Students, 2010–18 . . . . . . . . . . . . . . . 386
Growth in Four-Year Public and Private Universities’ Expenses by
Selected Category, 1997–2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 388
Composition of Student Body Population by Income Quintile. . . . . . . . . . . 389
Comparative Rates of Return for a Four-Year Degree. . . . . . . . . . . . . . . . . . . . 390
Comparison of Real Cumulative Median Net Earnings . . . . . . . . . . . . . . . . . . 391

Tables
1-1
1-2
1-3
3-1
3-2
3-3
4-1
5-1
6-1
7-1
8-1
8-2
9-1
9-2
9-3
9-4
9-5
9-6

Growth in Earnings, 2009–20. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
The Federal Response during the Great Recession and COVID-19 . . . . . . . . . . 59
Funding of Major Workforce Program Initiatives under the America
Reinvestment and Recovery Act, 2009. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Liquidity Programs for Small Businesses during COVID-19 . . . . . . . . . . . . . . . . 94
PPP Loan Size by Amount, First and Second Round Combined
(data as of Aug. 8, 2020) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
PPP Loans by Industry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Estimated Social Surplus by FDA Approval Time Reduction, 2025–40
(billions of real 2019 dollars). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
The Effect of Opportunity Zone Designation on Home Value
Appreciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Selected Deregulatory Actions’ Annual Impact on Real Income. . . . . . . . . . 195
Demographics of CSP-Funded Schools and District Public Schools,
2016–17. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
Composition of Global Space Economy, 2019. . . . . . . . . . . . . . . . . . . . . . . . . . 230
Summary of Effects of Property Rights Improvement. . . . . . . . . . . . . . . . . . . 243
U.S. Tariff Actions against China, 2018–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
U.S. Bilateral Trade Deficit with China in Goods and Services. . . . . . . . . . . . 255
Agriculture Phase One Agreement Provisions. . . . . . . . . . . . . . . . . . . . . . . . . . 258
Additional U.S. Exports to China on Top of the 2017 Baseline. . . . . . . . . . . . 260
The Impact of Canada and Mexico Increasing De Minimis Thresholds . . . . 266
U.S. Employment Sector Effects of the United States-Mexico-Canada
Agreement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Economic Report of the President | 31

9-7
9-8
9-9
10-1
10-2
10-3
10-4
11-1
11-2
11-3
11-4
11-5
11-6
11-7

Effects of the United States–Mexico–Canada Agreement on U.S. Trade
(percent changes relative to the baseline in 2017). . . . . . . . . . . . . . . . . . . . . . 267
Imports and Exports of Services by Sector, through September 2020. . . . . 274
U.S. Imports: Change from 2019 Levels for Selected Countries and
Regions, by Type of Good, through October 2020 . . . . . . . . . . . . . . . . . . . . . . 277
Cumulative Real Gross Domestic Product Losses through 2020:Q3. . . . . . . 305
Economic Forecasts, 2019–31. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
Policy-Inclusive Economic Projections, 2019-31. . . . . . . . . . . . . . . . . . . . . . . . 308
Supply-Side Components of Actual and Potential Real Output Growth,
1953–2031 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
Congressional FMLA Tax Proposals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
Annual Cost of Paid Parental and Medical Leave by Take-Up Scenario
(billions of dollars). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
Durations of GATT/WTO Rounds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Congestion Pricing Initiatives Worldwide. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
Lawful Permanent Resident (LPR) Status Obtained by Broad Class of
Admission, Fiscal Year 2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
National Income Increase from a Merit-Based Immigration System . . . . . . 382
75-Year Net Present Value of Fiscal Benefits from Proposed Changes to
the Permanent Legal Immigration System (billions of dollars). . . . . . . . . . . 384

Boxes
6-1
7-1
7-2
8-1
8-2
9-1
11-1
11-2
11-3
11-4
11-5
11-6

32 |

Effects of Regulation on Small Businesses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 190
The Supreme Court’s Espinoza v. Montana Department of Revenue
Decision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
The School Choice Now Act. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
Public-Private Partnerships for Human Spaceflight. . . . . . . . . . . . . . . . . . . . . 232
National Security and Space. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Bahrain’s and the United Arab Emirates’ Agreements with Israel. . . . . . . . . 268
Limiting Tax Expenditures to Facilitate Pro-Growth Reform: the
SALT+MID Deduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330
Continuing the Historic Health Information Technology Modernization
Started during COVID-19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
5G Infrastructure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
Reforming the NEPA Process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 367
Digital Infrastructure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
Historically Black Colleges and Universities. . . . . . . . . . . . . . . . . . . . . . . . . . . . 389

Economic Report of the President

Part I

Confronting the Largest
Postwar Economic
Shock: The Federal
Response to Mitigate the
COVID-19 Pandemic

x
Chapter 1

Creating the Fastest
Economic Recovery
The beginning of 2020 ushered in a strong U.S. economy that was delivering
job, income, and wealth gains to Americans of all backgrounds. By February
2020, the unemployment rate had fallen to 3.5 percent—the lowest in 50 years—
and unemployment rates for minority groups and historically disadvantaged
Americans were at or near their lowest points in recorded history. Wages were
rising faster for workers than for managers, income and wealth inequality were
on the decline, and median incomes for minority households were experiencing especially rapid gains. The fruits of this strong labor market expansion from
2017 to 2019 also included lifting 6.6 million people out of poverty, which is
the largest three-year drop to start any presidency since the War on Poverty
began in 1964. These accomplishments highlight the success of the Trump
Administration’s pro-growth, pro-worker policies.
The robust state of the U.S. economy in the three years through 2019 led almost
all forecasters to expect continued healthy growth through 2020 and beyond.
However, in late 2019 and the early months of 2020, the novel coronavirus
that causes COVID-19, with origins in the People’s Republic of China, began
spreading around the globe and eventually within the United States, causing a
pandemic and bringing with it an unprecedented economic and public health
crisis. Both the demand and supply sides of the economy suffered sudden
and massive shocks due to the pandemic. During the springtime lockdowns
aimed at “flattening the curve,” the labor market lost 22.2 million jobs, and
the unemployment rate jumped 11.2 percentage points, to 14.7 percent—the
largest monthly changes in the series’ histories.

The healthy foundation of the Trump Administration’s prepandemic economy,
coupled with strong and decisive action during the crisis, helped the Nation
weather the catastrophic COVID-19 shock and rebound faster than either
official or private forecasters had projected. After a sharp contraction in the
second quarter of 2020, the U.S. economy posted a 33.1 percent annualized
gain in gross domestic product (GDP) in the third quarter—the largest jump on
record, and nearly double the previous record from 70 years ago. As a result,
the U.S. economy has recovered two-thirds of the GDP damage from COVID-19
in just one quarter.
This chapter first documents the strength and resilience of the U.S. economy
leading up to the COVID-19 pandemic, both in absolute and relative senses. The
chapter demonstrates that the U.S. economy under the Trump Administration
suffered from fewer macroeconomic vulnerabilities than the pre–Great
Recession economy and that the economic experience during the pandemic
would have been even worse if it had not been for the economic improvement
from 2017 to the beginning of 2020.
In addition, this chapter details how, relative to the Great Recession, the
Federal Government acted with greater speed and provided more robust relief
in response to the COVID-19 crisis. In particular, the $2.2 trillion Coronavirus
Aid, Relief, and Economic Security (CARES) Act—passed by Congress within two
weeks of the President’s National Emergency Declaration—delivered the most
extensive fiscal relief in U.S. history. Moreover, it was targeted primarily to vulnerable families, workers, and small businesses, in stark contrast to the larger
focus on banks and big businesses in the fiscal response to the Great Recession.
Two overarching objectives have characterized the Federal Government’s
approach to combating the economic consequences of COVID-19: the alleviation of financial distress to reduce hardship, and the preservation of underlying
economic health to facilitate a faster recovery. For example, enhanced unemployment insurance benefits and eviction moratoriums supported household
balance sheets, and the Paycheck Protection Program strengthened the

36 | Chapter 1

connective tissue of the labor market by helping maintain matches between
employers and furloughed employees, setting the stage for the fastest employment rebound in U.S. history.
Chapters 2, 3, and 4 of this Report analyze the specific responses that this
Administration has implemented to address the dual public health and economic crises resulting from the COVID-19 pandemic.

he U.S. economy entered 2020 with historically low unemployment
and poverty, declining inequality, and some of the strongest household
income and wealth gains on record. In short, the American economy
was delivering greater opportunity to people across the socioeconomic
spectrum. At the time, leading forecasters were predicting this prosperity to
continue in 2020 and beyond with healthy GDP growth. However, COVID-19
interrupted this boom after it spread beyond the borders of China and instigated the most severe global public health and economic crisis in almost a
century. This chapter describes the healthy state of the U.S. economy before
COVID-19 reached American shores, the evolution of what has become the
largest shock to the U.S. economy since the Great Depression, and the historic
range of policies that were quickly passed into law to support the economy and
lay the foundation for a robust recovery.
Before delving into each of these issues individually, it is worth taking stock of the broader economic account of 2020 and just how far the U.S.
economy has recovered since the peak crisis period of the spring shutdowns.
As shown in figure 1-1, leading forecasters had been forecasting healthy 2 percent GDP growth for 2020 at the beginning of the year. Then, as the pandemic
worsened, they sharply revised their forecasts down, predicting the worst
contraction in annual GDP in the post–World War II period. However, in the
face of a much stronger recovery to date than almost anyone had predicted,
forecasters have responded by substantially revising their predictions for the
year upward, especially in light of the 33.1 percent annualized GDP rebound in
the third quarter that eclipsed the prior record from 70 years ago.
Figure 1-2 puts into stark relief the differences in economic behavior
during the COVID-19 pandemic versus during the Great Recession. Each curve
plots real GDP indexed to its level five quarters before the trough of each
downturn. As shown by the time-0 point on the horizontal axis, the onset of
COVID-19 led to a drop in indexed GDP more than twice as large as that of the
Great Recession. However, the figure also reveals the much more dramatic
rebound in economic fortunes during the pandemic thus far, driven by the
Federal government’s swift and bold economic interventions to deliver relief

Creating the Fastest Economic Recovery | 37

$"0- р-рѵ 1*'0/$*)*!спсп’.-*..*( ./$-*0/*- ./Ѷ
спсп
Percent (Q4-over-Q4)
у
с
п
Ҋс

'0 #$+

Ҋу
Ҋх

Ҋч
)ѵ

-ѵ

+-ѵ

0)ѵ

0'ѵ

0"ѵ

+ѵ

/ѵ

*1ѵ

*0- .ѷѸѸ'0 #$+*)*($ )$/*-.ѵ
*/ ѷۙ -'+ )-& /*(($// Ѹۙ-")$5/$*)!*-*)*($
**+ -/$*)) 1 '*+( )/.ҡ# Ѿ.!*- ./$./# 4 -Ҋ*1 -Ҋ4 -
+ - )/#)" ѵ

$"0- р-сѵ '-*..*( ./$-*0/ '')*. *-
#-+'4*2T#)0-$)"/# - / ..$*)
Index (100 = five quarters before recession trough)
рпф

- /
..$*)

Ҋрш
..$*)

рпп

шф

шп
–ф

–у

–т

–с

–р

Quarters (recession trough = 0)
*0- .ѷ0- 0*! *)*($)'4.$.Ѹ'0'/$*).ѵ

38 |

Chapter 1

particularly to households and small businesses. Provided that the economy
continues to receive appropriate and responsive fiscal support, the recovery
is poised to remain on a healthy trajectory. In contrast, in the aftermath of the
Great Recession, the economy suffered from a weaker and more protracted
recovery—especially when viewed through the lens of the labor market, as this
chapter discusses later.

The Historic Strength of the U.S.
Economy before COVID-19
Before the COVID-19 pandemic, the U.S. economy under President Trump
was surpassing milestone after milestone, delivering broad-based economic
gains to Americans of all backgrounds. After years of historically slow recovery
following the Great Recession, the unemployment rate fell below 4 percent for
the first time since December 2000, reaching 3.5 percent at the end of 2019.
The more comprehensive “U-6” unemployment rate—which includes people
not looking for work but wanting a job and people working part-time who
would prefer to have a full-time job—reached an all-time low of 6.7 percent in
December 2019.
Moreover, the advances in labor market opportunity extended to all
corners of American society. The unemployment rate for African Americans fell
to 5.4 percent in late 2019, down from 7.5 percent when President Trump took
office and the lowest level on record. For reference, the lowest rate achieved
under any previous administration was 7.0 percent in April 2000. Hispanic
Americans also enjoyed the lowest unemployment rate on record, with the
rate dropping to 3.9 percent in late 2019. Those with a less formal education
were also beneficiaries of a labor market of unparalleled strength, with the
unemployment rate for people with less than a high school diploma reaching
4.8 percent in late 2019, and Americans with only a high school degree facing
a 3.6 percent rate.
These strong pre-COVID labor market conditions were no mere coincidence; nor were they a passive continuation of economic momentum carried
over from the preceding years of the expansion. Although the unemployment
rate had managed to fall below 5 percent after six years of the slowest labor
market recovery in recorded history, the Congressional Budget Office and the
Federal Open Market Committee issued forecasts before the 2016 election
showing that the unemployment rate would flatten and stay well above 4 percent, as shown in figure 1-3. However, the combination of the landmark Tax
Cuts and Jobs Act in 2017 and the implementation of President Trump’s progrowth deregulatory agenda laid the groundwork for the economy to surpass

Creating the Fastest Economic Recovery | 39

Figure 1-3. The Unemployment Rate versus Preelection
Forecasts, 2011–19
Unemployment rate (percent)
11

2019:Q3

9
Unemployment
rate
7
FOMC median
forecast

CBO
forecast

3
2010:Q1

2012:Q1

2014:Q1

2016:Q1

2018:Q1

Sources: Congressional Budget Office; Bureau of Labor Statistics; Federal Reserve.
Note: CBO = Congressional Budget Office; FOMC = Federal Open Market Committee.
The CBO forecast is from August 2016; the FOMC forecast is from September 2016.

these expectations by boosting economic competitiveness and dynamism (CEA
2019, 2020a).1
Besides increasing the abundance of job opportunities, a low unemployment rate also confers greater bargaining power on workers when they are
negotiating pay with employers. Both when looking to recruit new workers and
retain existing talent, employers must offer a compelling pay package when
unemployment is low or else risk losing valuable workers to their competitors.
In fact, 2019 data from the Job Openings and Labor Turnover Survey (JOLTS)
shows the highest quit rate since 2001—a sign of a challenging environment
for employers to retain workers who were availing themselves of the tight
competition for their services. Table 1-1 compares the magnitude of earnings
growth for different types of workers under the pre-COVID Trump economy
with the expansion period from the previous administration. Table 1-1 shows
that earnings growth was higher across the board in the period since 2017 to
before COVID-19, and on top of that, workers’ earnings were outpacing those of
managers, and the bottom 10 percent of wage earners were experiencing more
rapid earnings growth than the top 10 percent.
1 Chapter 1 in both the 2019 and 2020 editions of the Economic Report of the President provides
a comprehensive analysis of the pro-growth benefits of the Tax Cuts and Jobs Act. Chapter 3 of
the 2020 Report discusses the benefits of the Trump Administration’s focus on deregulation for
household income.

40 | Chapter 1

Table 1-1. Growth in Earnings, 2009–20
-*0+
*-& -.
)" -.
*b# '*-’.
# '*-’.*-(**//*(рп‫ڿ‬2" -) -.
*+рп‫ڿ‬2" -) -.

- Ҋ *)*(4 - 1$*0.3+).$*) -$*
(Jul. 2009–Dec. 2016)
(Jan. 2017–Feb. 2020)
тѵт
сѵц

сѵт
сѵф

җспрцѷр–спршѷуҘ
тѵп
сѵш
уѵш
тѵт

җсппшѷт–спрхѷуҘ
рѵт
рѵф
рѵш
сѵу

*0- .ѷ0- 0*! *-//$./$.Ѹ'0'/$*).ѵ
Note: Data represent a compound annual growth rate for 2009:Q3–2016:Q4 or July 2009–December
2016 and 2017:Q1–2019:Q4 or January 2017–January 2020. For workers and managers, earnings are
!$) .1 -" 2 &'4 -)$)".ѵ*-''*/# -/ "*-$ .Ѷ -)$)".- !$) .( $)0.0'
2 &'4 -)$)".ѵ

The CEA finds that higher earnings growth among low-wage workers is
a result of rising labor demand in the Trump economy. Although some assert
the importance of State-level minimum wage increases based on cross-state
comparisons of wage growth since 2016 (Van Dam and Siegel 2020; Nunn and
Shambaugh 2020; Tung 2020; Tedeschi 2020), there are serious limitations
and flaws in these analyses that undermine their conclusions. In particular,
the limitation of these studies is that they do not show that the timing of wage
increases aligns with the timing of minimum wage hikes in States that have
instituted such hikes. Thus, the studies do not distinguish wage growth that
occurred before a minimum wage hike from wage growth that occurred after a
hike. Because of their failure to consider this timing issue, these studies do not
provide strong evidence that minimum wage hikes are responsible for wage
growth. Additionally, wage growth could have been higher in the States that
increased their minimum wages even without the increases.
In contrast, the CEA’s analysis uses detailed microdata from the Current
Population Survey to identify workers with direct exposure to minimum wage
hikes based on their position in the wage distribution. The CEA then calculates
the effect of the minimum wage by estimating what wage growth for the
directly-affected group would have been had no minimum wage hike occurred.
Based on these calculations and a sensitivity analysis, the CEA attributes as
an upper bound only 0.2 percentage points of wage growth among workers in
the bottom third of the wage distribution to minimum wage hikes. To put this
number in perspective, such workers experienced total annual wage growth of
3.8 percent between 2017 and 2019.
In support of the view that strong labor market conditions—not minimum wages—drove the observed wage gains, research by the Federal Reserve
Bank of Atlanta compares wage growth in States that increased their minimum

Creating the Fastest Economic Recovery | 41

$"0- рҊуѵ ' $) *0. #*' )*( 4 *0. #*' - Ѷ
ршхц–спрш
Annual household income (thousands of 2019 CPI-U-RS adjusted dollars)
рпп
.$)

чп

хп

спрш

''- .

#$/ Ѷ)*/
$.+)$
$.+)$җ)4- Ҙ

уп
'&
сп

п
ршхц

ршцс

ршцц

ршчс

ршчц

ршшс

ршшц

сппс

сппц

спрс

спрц

*0- ѷ ).0.0- 0Ѷ0-- )/*+0'/$*)0-1 4Ѷршхч/*спсп))0'*$')
*)*($0++' ( )/.ѵ
*/ ѷ ҊҊۙ*).0( --$ ) 3 . -# -$ ..

wages with those that did not. Robertson (2019) examines the ratio of the
12.5th percentile wage (i.e., the median wage of the lowest quartile) relative
to the median wage for all workers. Between 2014 and 2019, this ratio was
increasing, indicating faster wage growth at the bottom of the distribution.
Notably, the ratio was increasing at about the same rate among States that
increased their minimum wages and among States that did not. Robertson
(2019) concludes, “The increased tightness of labor markets, or some other
factor than hikes in State minimum wages, is playing a role in pushing up the
pay for those in lower-wage jobs.”
Looking back further than just the previous administration, the $4,400
jump in real median income in 2019 marked the largest one-year increase on
record, capping a nearly 10 percent increase since 2016 after adjusting for the
U.S. Census’s redesign in 2017. Moreover, figure 1-4 reveals that the boost
to household incomes occurred for all races, with minorities experiencing
outsized gains. Specifically, in 2019 real median income for Black households
rose by 7.9 percent, Hispanic Americans saw a 7.1 percent boost, and Asian
Americans enjoyed an even larger 10.6 percent increase, while White households experienced a smaller but still substantial 5.7 percent jump. Each of
these figures represents record increases and record absolute levels.
The broad-based income and employment gains before COVID-19 also
fueled rising household net worth, lower income and wealth concentration,

42 |

Chapter 1

Figure 1-5. Share of Total Net Worth by Percentile, 2007–19
Share of total net worth (percent)

Share of total net worth (percent)

2019:Q3

Share held by top 1% of
households (left axis)

Declined 0.5 percentage
point since TCJA

Increased 0.4 percentage
point since TCJA

23
Share held by bottom 50% of
households (right axis)

20
2007

2009

2011

2013

2015

2017

2019

Sources: Federal Reserve Board; CEA calculations.
Note: TCJA = Tax Cuts and Jobs Act.

and a record fall in the official poverty rate. Through the fourth quarter of
2019, the net worth of the bottom 50 percent increased by 38.9 percent during
President Trump’s first term, while it increased by 20.1 percent for the top 1
percent. Since the Tax Cuts and Jobs Act passed, the wealth share of the top
1 percent fell by 0.5 percentage point, while that of the bottom 50 percent
rose by 0.4 percentage point, as shown in figure 1-5. This broad increase in net
worth partly reflects the stark turnaround in the homeownership rate, which
reached 65.1 percent in 2019 after recovering from a 2016 trough of 62.9 percent. Income concentration also fell, with the Gini coefficient—a widely used
measure of concentration that ranges between 0 and 1—declining from 0.489 in
2017 to 0.484 in 2019. Data from the 2019 Survey of Consumer Finances reveal
broad wealth increases driven by the lower earners, with median net worth in
the lower two income quintiles up by over 30 percent since 2016. Hispanics
and African Americans enjoyed respective gains of 64 percent and 32 percent.
At the bottom of the income distribution, the robust labor market expansion between 2016 and 2019 lifted 6.6 million people out of poverty, which
is the largest three-year reduction to start any presidency since the War on
Poverty began in 1964. As a proportion of the population, the poverty rate
fell to an all-time low of 10.5 percent in 2019—with especially large poverty
declines for African Americans, Hispanics, and Asians—as figure 1-6 makes

Creating the Fastest Economic Recovery | 43

$"0- р-хѵ*1 -/4/ .4 ) $.+)$-$"$)Ѷ
ршфш–спрш
Percent
фп
спрш
уп

тп

сп

$.+)$җ)4- Ҙ

.$)

''- .
рп
*)Ҋ $.+)$#$/
п
ршфш ршху ршхш ршцу ршцш ршчу ршчш ршшу ршшш сппу сппш спру спрш
*0- ѷ ).0.0- 0ѵ

Figure 1-7. U.S. Real Gross Domestic Product, 2014–19
2012 dollars (trillions)
20

2019:Q4
Actual

19
CBO
forecast
(Aug. 2016)

FOMC forecast
(Sep. 2016)

16
2014

2015

2016

2017

2018

2019

Sources: Bureau of Economic Analysis; Congressional Budget Office (CBO); Federal
Open Market Committee (FOMC).

44 |

Chapter 1

evident. Moreover, 2.8 million children were lifted out of poverty between 2016
and 2019, driving the child poverty rate down to a 50-year low of 14.4 percent.
In the years immediately preceding the pandemic, the United States
experienced robust GDP growth that exceeded what the Congressional Budget
Office and the Federal Open Market Committee had previously forecast for
those years, as seen in figure 1-7. Real GDP grew 2.5 and 2.3 percent in 2018
and 2019, respectively, faster than any other Group of Seven country. Entering
2020, many forecasters slated U.S. output to grow at a healthy pace of about
2 percent in 2020, though it is entirely plausible that the U.S. economy could
have continued exceeding projections if the global economy had not been hit
with the COVID-19 pandemic—the largest exogenous shock since the Great
Depression.

The Early Economic Effects of COVID-19
On January 7, 2020, Chinese researchers announced the discovery of the
severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)—which causes
the disease COVID-19—in the travel hub city of Wuhan, China.2 On January 21,
the first case of a person contracting the new coronavirus after traveling from
Wuhan was reported in the United States.3 By late February, the Centers for
Disease Control and Prevention had confirmed the first possible instance of
community transmission in the United States, and the Standard & Poor’s 500
began a sharp sell-off that continued through March 23, losing 33.9 percent of
its value compared with its peak just before the outbreak.4
The Trump Administration responded by promptly putting in place nonpharmaceutical intervention policies to contain the virus.5 Travel restrictions
on China were imposed on January 31, and the restrictions were subsequently
expanded to 26 countries in Europe and several other countries by mid-March
(White House 2020a, 2020b). On March 13, President Trump declared COVID-19
a national emergency (White House 2020c). The adoption of a host of socialdistancing measures—which included school closures, bans on group gatherings, and closures of restaurants—became prevalent across States shortly
thereafter. By March 23, Statewide school closures and restrictions on bars
and restaurants had affected over 90 percent of the U.S. population (figure
2 Chinese researchers isolated and confirmed a novel coronavirus after identifying a cluster of
acute respiratory illnesses in Wuhan on December 31, 2019 (Patel, Jernigan, and 2019-nCov CDC
Response Team 2020).
3 The CDC announced the first case in the United States when a traveler sought treatment after
returning from Wuhan to Washington State a few days earlier (CDC 2020a).
4 The first case of COVID-19 with no prior travel to infected regions was confirmed by the CDC
(2020b).
5 The CDC defines nonpharmaceutical interventions as actions, apart from vaccination and taking
medicine, that people and communities can take to slow the spread of illnesses like the COVID-19
pandemic (CDC 2020c).

Creating the Fastest Economic Recovery | 45

$"0- р-чѵ - )/" *!/# ѵѵ*+0'/$*)0) -// 2$
./-$/$*).Ѷспсп
ѵѵ
. .
Population (percent)
+..спѶппп
рпп
$($/-.)
шп
- ./0-)/.
чп

ѵѵ
/
. .
.$")
+..сппѶппп

+-ѵш

цп
# '/ -
*- -.

хп
фп
уп
тп

#**'
'*.0- .

сп
рп
п
-ѵрт

-ѵсп

-ѵсц

+-ѵт

*0- .ѷ New York TimesѸ// +*'$4))*0) ( )/.Ѹ'0'/$*).ѵ

1-8). By March 30, 30 States had issued stay-at-home orders, with an additional
13 States having issued these orders for State sections. By early April, over 90
percent of the U.S. population lived in a State that had issued a stay-at-home
order.6
Studies of the economic effects of past pandemics indicate that there are
three main channels through which pandemics affect economic activity:7 (1)
increased mortality, (2) illness and absenteeism, and (3) avoidance behavior
to reduce infection. These shocks reduce the size of the labor force, aggregate
productivity, and aggregate demand. Consistent with these observations, the
economy has experienced sudden, large, and simultaneous shocks to supply
and demand as a result of the COVID-19 outbreak in the United States.
On the supply side, many businesses were shuttered by social-distancing measures that States and local authorities put in place or businesses

6 After the Administration’s efforts to inform the American public, States began introducing
restrictive mandates and regulations dictating protective behavior. The CEA finds that 67 to 100
percent of the observed total increases in a variety of protective behaviors appears to have been
driven by the American people’s voluntary decisions and the Administration’s efforts to encourage
these voluntary decisions, and only 33 percent to be accounted for by restrictive State mandates.
7 See Jonas (2013); Kilbourne (2006); Burns, van der Mensbrugghe, and Timmer (2006); Verikios et
al. (2011); McKibbin and Sidorenko (2006); CEA (2019); and McKibbin (2009).

46 | Chapter 1

$"0- р-шѵ /$'+ )$)"0-$)"/# -'4/" .*!/#
) ($Ѷ 1 )Ҋ41 -" Ѷспсп
Year-over-year change (percent)
рп

# '/ -Ҋ$)Ҋ+'
)$)$)" CARES Act
signed
- ./-$/$*). "$)
+-ѵрт

–рп

–сп

–тп

–уп
ѵр

ѵрф

ѵсш

-ѵру

-ѵсч

+-ѵрр

*0- ѷ-*+-$ /-4 .+ )$)"/ѵ
*/ ѷ# /2..$") $)/*'2*)-#сцѶспспѵ

voluntarily adopted to stop the spread of the virus and “flatten the curve.”8
Those that remained open faced supply disruptions that prevented them
from operating normally. On the demand side, many consumers faced stayat-home orders or voluntarily limited their economic activity to reduce the
risk of contracting the disease.9 Consumers also changed the composition of
their demand; for example, they replaced restaurant meals with home-cooked
meals and increased their demand for cleaning supplies.
High-frequency indicators that serve as proxies for demand across various economic activities show that the downturn began in early March, in some
cases before Statewide social-distancing measures were implemented, and
reached its trough at the end of April. Daily retail spending data started plunging in mid-March and bottomed out at a 30 percent year-over-year decline at
the end of March (figure 1-9). By the time shelter-in-place orders and dining
8 E.g., on March 11 (before President Trump’s announcement of COVID-19 as a national emergency),
the NBA had already suspended basketball games indefinitely. The following day, Major League
Baseball delayed the start of its season, the National Hockey League suspended games, and March
Madness was canceled.
9 Baqaee and Farhi (2020) model the distinct shocks to supply and demand and study how
the combination of supply and demand shocks explains the data. They argue that without the
negative shock to aggregate demand, the United States could have experienced stagflation, or a
combination of rising unemployment and rising prices. Instead, the negative shock to aggregate
demand has limited inflation.

Creating the Fastest Economic Recovery | 47

$"0- р-рпѵ-!!$*)" ./$*)0-$)"/# -'4/" .*!/#
) ($Ѷ $)-*..''// .Ѷ 1 )Ҋ41 -" Ѷспсп
Year-over-year change (percent)
уп

# '/ -Ҋ$)Ҋ+' ) $)$)"
- ./-$/$*). "$)

+-ѵрт

сп
п
–сп
–уп
–хп
–чп
-ѵх

-ѵрт

-ѵсп

-ѵсц

+-ѵт

+-ѵрп

*0- .ѷ*(*(Ѹ3)/ Ѹ'0'/$*).ѵ

restrictions began, daily traffic congestion (figure 1-10) and seated diners (figure 1-11) across all States had already dropped over 20 percent year-over-year.
Similarly, weekly hotel occupancy had dropped 56 percent year-over-year in
the week these shelter-in-place measures began (figure 1-12).
Supply indicators—the number of small businesses that were open,
the number of hourly employees who were working, and number of hours
worked—also saw the steepest year-over-year contraction in March and April.
Figure 1-13 illustrates how these indicators compared with a January preCOVID-19 baseline, as reported by Homebase.10 After shelter-in-place orders
became widespread in mid-March, the proportion of employees working fell
from about 15 percent below normal conditions to about 55 to 60 percent.
As the indicators discussed above show, the restrictions on mobility and
the shift toward social distancing played a major role in limiting economic
activity. Academic research conducted since the COVID-19 pandemic began
attempts to quantify the extent to which government restrictions versus voluntary mitigation behaviors can account for the decline in mobility during the
10 Homebase is a company that provides software to help small business owners manage employee
timesheets. Since the start of the pandemic, Homebase has maintained a database of U.S. small
business employment using data from more than 60,000 businesses that use its software. The
data cover more than 1 million employees that were active in the United States in January 2020.
Most Homebase customers are businesses that are individually owned or operator-managed
restaurants, food and beverages businesses, retail outlets, and service establishments.

48 |

Chapter 1

Figure 1–11. Seated Diners in U.S. Restaurants, per
OpenTable, during the Early Stages of the Pandemic,
Seven-Day Average, 2020
Year-over-year change (percent) Shelter-in-place
and dining
restrictions begin
20

Apr. 13

0
–20
–40
–60
–80
–100
Feb. 24

Mar. 9

Mar. 23

Apr. 6

Sources: OpenTable; New York Times; CEA calculations.

spring. For example, Goolsbee and Syverson (2020) examine cellular phone
records data on customer visits to individual businesses across contiguous
boundaries with different policies. They conclude that consumer traffic started
to decline before State and local restrictions were put in place, that the degree
of private mitigation behavior was tied to the local severity of the virus (i.e.,
number of deaths in the county), and that, overall, legal restrictions explained
only a small fraction of the total decline in activity. However, they do find that
the shutdown orders caused a reallocation of consumer activity from “nonessential” to “essential” businesses and from restaurants and bars to groceries.
Another study by Cronin and Evans (2020) contains similar findings, concluding
that private, self-regulating behavior explained more than three-quarters of
the decline in foot traffic but that regulations had large effects on foot traffic to
restaurants, hotels, and nonessential retail.
The pandemic also caused significant disruptions to the labor market and
to macroeconomic activity. Due to their short reporting lag, initial claims for
Unemployment Insurance (UI) provide timely information on how the COVID19 pandemic and containment measures have affected the labor market. In
March, job losses occurred at a level not seen since the Great Depression, with
initial UI claims spiking from 282,000 the week ending March 14 to 6.9 million
two weeks later.
Creating the Fastest Economic Recovery | 49

Figure 1-12. Weekly U.S. Hotel Occupancy Rate during the Early
Stages of the Pandemic, 2020
Shelter-in-place
and dining
restrictions begin

Year-over-year change (percent)
10

Apr. 11

0
–10
–20
–30
–40
–50
–60
–70
Jan. 4

Jan. 25

Feb. 15

Mar. 7

Mar. 28

Sources: STR; CEA calculations.

$"0- р–ртѵ - )/#)" $)(''0.$) .. . #/- + )
) *0-'4(+'*4 .#*- *-&$)"$)/# -'4/" .*!/#
) ($Ѷспсп
Percent change (relative to January base)

# '/ -Ҋ$)Ҋ+' )$)$)" /
- ./-$/$*). "$)
.$")
рп

+-ѵрт

п
–рп
–сп
–тп

*0-.2*-& 4
#*0-'4 (+'*4 .

0( -*!
0.$) .. .*+ )

–уп
–фп
–хп
–цп
-ѵс

-ѵрх

-ѵтп

+-ѵрт

*0- ѷ *( . ѵ
*/ ѷ ''-/ .*(+- "$1 )/ 1 -.0./# ( $)!*-/#/4*!/# 2 &0-$)"/#
+ -$* )0-4у–трѶспспѵ# )0( -*!#*0-'4 (+'*4 .2*-&$)"/- .#*0-.2*-&
4#*0-'4 (+'*4 .ѵ

50 |

Chapter 1

Data on total economic output also reflect the enormous negative shock
that the COVID-19 pandemic and containment measures had on the economy.
First-quarter real GDP declined at an annualized rate of 5.0 percent—itself significant—but this drop would later be dwarfed by the annualized 31.4 percent
collapse in second-quarter GDP. In early June, the Organization for Economic
Cooperation and Development (OECD) estimated that the COVID-19 pandemic
and containment measures would decrease U.S. Q4-over-Q4 GDP by 7.4 percent in 2020 in the absence of a second wave in the fall (single-hit scenario), or
12.3 percent if such a wave were to occur (double-hit scenario). This forecast
was more pessismistic than those provided by the Congressional Budget Office
and the Blue Chip survey of the private sector in July, which were still large, at
5.9 and 5.6 percent decreases, respectively.

The U.S. Economy’s Resilience in
Weathering the COVID-19 Shock
Beyond the immediate prosperity that Americans were enjoying before COVID19, the vibrant state of the U.S. economy rendered it more resilient and better
prepared to weather the COVID-19 shock than if it had occurred in earlier years.
To quantify this resilience, the CEA simulates the likely path of the unemployment rate if the COVID-19 shock had occurred under the weaker economic
conditions of 2016 instead of the stronger actual 2020 pre-COVID conditions.
To construct this simulation, the CEA uses Current Population Survey data to
measure the monthly probability that workers transit between employment,
unemployment, and not being in the labor force. The CEA’s analysis assumes
that any year-over-year deterioration in transition probabilities from 2019 to
2020 is attributable to COVID-19, which makes it possible to isolate the magnitude of the COVID-19 shock to labor flows. Then, the CEA applies this measured
COVID-19 shock to monthly 2016 labor market transition probabilities to arrive
at likely counterfactual labor market flows and ultimately unemployment
dynamics if COVID-19 had occurred under 2016 economic conditions.
The solid blue line in figure 1-14 shows the actual observed path of unemployment, and the solid green line shows the simulated path of the unemployment response to COVID-19 under full 2016 conditions—specifically, starting
from the 4.9 percent February 2016 unemployment rate (compared with 3.5
percent in February 2020) and with the worse baseline (without COVID-19)
labor dynamism from 2016. As the figure shows, if COVID-19 had arrived with
the U.S. economy in its 2016 state, the unemployment rate would likely have
peaked at a higher rate and been nearly 2 percentage points above the actual
level in October. If, instead, the U.S. economy had entered the COVID-19 crisis
with the 2016 level of unemployment but the healthier Trump labor market
flows—as shown in the red dashed curve in the figure—the dynamics of unemployment would not have looked substantially different than what has actually

Creating the Fastest Economic Recovery | 51

$"0- рҊруѵ) (+'*4( )/) Ҋршѷ# -0(+*)*(4
1 -.0./# спрх*0)/ -!/0', 2020
/0'
спрх!'*2.2$/#спсп
Unemployment rate (percent)
рц

-0(+!'*2.2$/#спрх
0''спрх*0)/ -!/0'
Sep. 2020

ру

рр

с
ѵ

-ѵ

+-ѵ

0)ѵ

0'ѵ

0"ѵ

+ѵ

*0- .ѷ ).0.0- 0Ѹ'0'/$*).ѵ
*/ ѷ ۙ )$/$'0) (+'*4( )/-/ ѵ

occurred. In other words, the difference in initial unemployment rates is not
the crux of the superior resilience of the Trump economy. To the contrary, the
gold dashed curve shows that, holding fixed the initial February unemployment rate at 3.6 percent, the unemployment rate would have followed a much
worse trajectory if the economy had suffered from the worse underlying dynamism of the 2016 economy.

Comparing the COVID-19 Recession
and the Great Recession
The pre-COVID U.S. economy possessed fewer macroeconomic vulnerabilities
than it had in the lead-up up to the Great Recession, when overextended
household borrowers and a highly leveraged financial sector precipitated the
Great Recession. Unlike the previous recession, the COVID-19 crisis was not the
consequence of underlying economic imbalances, and the greater resilience
of the pre-COVID U.S. economy coupled with the superior fiscal response
augurs well for the continuing prospects of a much more robust recovery. This
section sheds light on the comparative health of the U.S. economy before the
current crisis relative to the years before the 2007–9 financial crisis and Great
Recession.

52 |

Chapter 1

The State of the Economy before the Crises
This subsection looks at various sectors of the U.S. economy before the crises.
We consider households, nonfinancial businesses, and banks.
Households. The financial situation of the household sector was stronger in early 2020 than at the start of the Great Recession. From 2000 to 2008,
household liabilities as a share of personal disposable income rose from 96
percent to 136 percent before falling back to below 100 percent before COVID19, according to the Federal Reserve’s Flow of Funds data. However, examining only aggregates can obscure the true level of risk, which is captured more
accurately by the tails of the distribution. Even along this dimension, however,
the U.S. economy was in a stronger position before the COVID-19 crisis than it
was back in 2006 before the start of the financial crisis. The share of mortgages
with debt-to-income ratios above 50 percent fell from 11.0 percent in 2006 to
only 6.9 percent in 2018. Though the loan-to-value ratio for new mortgages
was similar to what it was in 2006, credit had shifted toward borrowers with
high credit scores. Whereas 14.1 percent of borrowers taking out a mortgage
had below a 620 credit score in 2006, that share was only 3.3 percent in 2018.
Borrowers were also taking out safer loans by 2018. The share of mortgages
with less than full amortization fell from 29.2 percent in 2006 to 0.6 percent
in 2018, and mortgages for which borrowers were only required to provide
minimal documentation at origination saw their share drop from 34.5 percent
in 2006 to 1.8 percent in 2018 (Davis et al. 2019). Looking beyond mortgages,
the share of credit card volume going to subprime borrowers was under 2.5
percent in 2019, compared with 3.4 percent before the financial crisis, according to the Consumer Financial Protection Bureau. The bureau also shows that,
for automobile loans, the share going to subprime borrowers was under 15
percent in 2019 before COVID-19, versus nearly 20 percent in 2006.
Before COVID-19, researchers ran stress tests on households to examine
how negative shocks to the economy would translate into defaults on household debt. One study simulates a fall in house prices similar to what occurred
in the Great Recession and generates a much smaller peak in foreclosures; the
average shocked stressed default rate—which represents, for a particular loan,
its expected default rate if it were hit shortly after origination with a replay of
the financial crisis—was 9.7 percent in 2018 compared with 34.8 percent in
2006 (Davis et al. 2019). Another study simulates a large house price decline
and unemployment spike meant to mimic the financial crisis. When faced with
the same shocks from 2007 to 2009, the simulated 2020 economic response
generates fewer defaults because of healthier household balance sheets
(Bhutta et al. 2019). Although the COVID-19 economic shock differs from that
of the last crisis, the combined effect of stronger household balance sheets
and a bolder fiscal response has greatly reduced the amount of actual financial
distress that one would expect from such a large disruption.

Creating the Fastest Economic Recovery | 53

$"0- р–рфѵ/$**! *0. #*' /*-)*)!$))$'
0.$) .. //*-*..*( ./$-*0/Ѷсппхѷу–спспѷс
Debt-to-GDP (ratio)

спспѷс

рѵп
пѵш

*0. #*'. /*- /

пѵч
*)!$))$'0.$) .. /

пѵц
пѵх
пѵф
сппхѷу

сппшѷс

спррѷу

спруѷс

спрхѷу

спршѷс

*0- .ѷ0- 0*!*)*($)'4.$.Ѹ -' . -1 *-Ѹ'0'/$*).ѵ
*/ ѷ *0. #*'. /*- /$)'0 . /*2) 4)*)+-*!$/*-")$5/$*).ѵ
/$. !$) .*0./)$)"'*).) /. 0-$/$ .ѵ#$)" )*/ .
- ..$*)ѵ

Nonfinancial businesses. Although households were in good shape before
the COVID-19 pandemic, the nonfinancial business sector had become more
leveraged. By early 2020, the aggregate debt-to-GDP ratio for nonfinancial
businesses had reached levels not seen since the financial crisis (figure 1-15).11
One reason nonfinancial business debt has risen, however, is that interest rates are at historic lows. This reduces the burden of servicing debt. A basic
measure of the debt burden is the ratio of company earnings to their interest
payments, or the interest coverage ratio. In the years leading up to the pandemic, the interest coverage ratio for the median firm remained high (Federal
Reserve 2020). The sales-weighted shares of nonfinancial public corporations
that use more than 30 percent, 40 percent, or 50 percent of their earnings to
make interest payments were all declining; and in early 2020, at the onset of
COVID-19, these shares were all lower than at the start of the Great Recession
(Crouzet and Gourio 2020).
Despite historically low costs of borrowing, the Federal Reserve and the
International Monetary Fund have expressed concern about the quality of
corporate debt. In early 2020, about 50 percent of investment-grade debt was
rated BBB, an amount that was near a historical high. BBB is the lowest rating category for investment-grade debt, and thus carries more risk of default
than higher-grade debt. Another concern is that in recent years, loans to large
11 These ratios spiked in the second quarter of 2020 as GDP contracted sharply.

54 |

Chapter 1

corporations have increasingly focused on highly leveraged firms. In February
2020 at the onset of the pandemic recession, the rate was higher than at the
start of the Great Recession (Federal Reserve 2020; IMF 2019, 2020). Overall,
the second quarter of 2020 had the highest quarterly volume of defaults in
leveraged loans since the first quarter of 2009 (LCD News 2020).
Banks. The banking sector was well capitalized at the start of the COVID19 pandemic. According to data compiled by the Federal Deposit Insurance
Corporation, as of the fourth quarter of 2019, the commercial banking and savings and loan sector stood at a record, or near-record, in various measures of
industry solvency and liquidity. This status was largely attributable to the continuous growth in the economy since the end of the Great Recession and the
passage and continuing implementation of the Dodd-Frank Act of 2010, which
dramatically raised regulatory oversight and capital standards for the industry.
The number of banks on the Federal Deposit Insurance Corporation’s
“Problem Bank List” leading up to COVID-19 was exceptionally low. The number of problem banks fell from 76 in 2007:Q4 to 51 by 2019:Q4, the lowest number of problem banks since 2006:Q4. Total assets of problem banks increased
from $22 billion in 2007 to $46 billion in 2019. The commercial banking sector
also entered the crisis with stable indicators of asset quality.

The Origins and Progression
This subsection reviews the different origins of the COVID-19-induced recession
and the Great Recession, and the important differences in how these shocks
played out over time. The financial crisis and resulting Great Recession of
2007–9 started with an overheated housing market. In 2006, housing market
weakness began to emerge, first in the form of longer selling delays—indicating
a deterioration in housing liquidity—followed by deceleration and reversal in
house price growth. The weakness in housing then spilled over into the rest
of the economy because of the damage it wreaked on household and bank
balance sheets alike.
By March 2007, there were reports that the housing slump had hit some
hedge funds hard. In their book First Responders, Bernanke, Geithner, and
Paulson (2020, 12) state that “if we had to pick the date that the crisis began, it
would be August 9, 2007, when the French bank BNP Paribas froze withdrawals
from three funds that held securities backed by U.S. subprime mortgages.” By
the late summer of 2007, the investment bank Bear Stearns was liquidating two
hedge funds that were heavily invested in subprime mortgages. Over the next
year, the contagion spread to every corner of financial markets and turned into
a full-blown crisis. Facing deteriorating balance sheets and frozen markets,
lenders cut the supply of credit to the economy, which caused households
and businesses to curtail spending. As the economy hemorrhaged jobs, higher
unemployment accelerated the collapse in the housing market, which further
fueled the cascading spiral of economic misery.
Creating the Fastest Economic Recovery | 55

The unemployment rate increased from 4.7 percent in November 2007
to a peak of 10.0 percent in October 2009. Moreover, unemployment remained
above 9 percent for two years after the technical end of the recession (i.e.,
when GDP stopped contracting), and the average duration of unemployment
for jobless workers stayed near historic highs. Households saw their housing
wealth evaporate as prices fell by nearly 30 percent on average—with larger
declines on the coasts and in several Sun Belt States—at the same time that
their retirement portfolios suffered a 50 percent drop in the Dow Jones from
peak to trough on March 9, 2009. In addition, 3.8 million homes were foreclosed
between 2007 and 2010 (Dharmasankar and Mazumder 2016). Even with all the
major interventions that were considered unprecedented at the time, it took
years for the U.S. economy to fully recover as scars from the crisis persisted.
Both the origins of the COVID-19 recession and the progression of the
recovery have been quite different from those of the Great Recession. First, as
discussed above, the pre-COVID U.S. economy was in a much healthier state,
lacking the household balance sheet vulnerabilities that exacerbated the wave
of defaults and financial distress during the 2007–9 financial crisis. House
prices have also remained remarkably stable—likely buoyed by the surge
in personal income fueled by the CARES Act—and these prices are boosting
family finances and have helped prevent a repeated wave of foreclosures like
the one that ripped through the economy during the Great Recession. Most
important, the speed of the recovery to date has been dramatically faster,
with the unemployment rate spending only 4 months above 9 percent during
the COVID-19 pandemic, compared with the over two years it hovered above 9
percent during the sclerotic recovery from the last recession. In the 7 months
of data since the trough of employment during COVID-19, the U.S. economy
has already recovered 56 percent of the lost jobs. By comparison, it took 30
months to gain back more than half the jobs lost in the aftermath of the Great
Recession. Moreover, the broader “U-6” unemployment rate spent five years
above 13 percent during the slow recovery from the Great Recession, whereas
during COVID-19, the rate fell below that level after just 5 months.

Fiscal and Monetary Responses
Despite the health and resilience of the U.S. economy at the beginning of
2020, the initial negative shock was unprecedented. Moreover, even though
the immediate economic losses were concentrated in the second quarter of
2020, when shutdowns were widely in place throughout the United States, the
Federal Government took action to combat the short-term liquidity crisis and
minimize the extent to which it could turn into a widespread solvency crisis
for families and businesses with long-lasting negative effects on bankruptcies,
unemployment, and production. This subsection compares the speed and
scale of the Federal response to COVID-19 with the actions taken to combat

56 | Chapter 1

the Great Recession. Later chapters analyze the economic effect of the specific
COVID-19 economic interventions.
The Federal Government’s policies to address the financial crisis of
2007–9 evolved over a number of years, and they ranged from the fiscal stimulus of increased government expenditures for infrastructure, health, education,
energy independence, tax rebates targeting low- and middle-income families
and tax incentives for business investment; to assistance on refinancing or
modifying mortgages to monetary open market operations and liquidityenhancing programs to bailouts and subsidies of various entities; and, finally,
to substantial regulatory changes. On the monetary policy side, the Federal
Reserve employed open market operations and later a program of largescale asset purchases (commonly referred to as quantitative easing) after the
Federal Funds rate hit the zero lower bound. The Federal Reserve also took a
variety of approaches to help provide liquidity to various markets and market
participants, primarily through the creation of several funding, credit, liquidity,
and loan facilities.
Besides these and other Federal Reserve interventions, Congress passed
significant stimulus bills over the course of the crisis. In February 2008, in an
effort to ameliorate the growing crisis, the Economic Stimulus Act of 2008 was
passed, offering tax recovery rebates to individuals and their dependents, and
targeting low- and middle-income taxpayers. The act also created incentives
for business investment by permitting the accelerated depreciation or immediate expensing for certain assets. In October 2008, the Emergency Economic
Stabilization Act of 2008 was passed, allocating $700 billion to address the
financial crisis by purchasing or insuring troubled assets and attempting to
avert the failure of financial institutions identified as systemically important.
This established the Troubled Asset Relief Program, known as TARP. In 2009,
the American Recovery and Reinvestment Act was passed, which included
tax cuts and government expenditures totaling over $800 billion, for national
infrastructure, energy independence, education, health care, and tax relief.
The Federal Government also stepped in to bail out the automobile industry.
In 2010, the Dodd-Frank Wall Street Reform and Consumer Protection Act was
enacted, entailing substantial changes to the regulatory architecture of U.S.
financial markets.
In addition, the Federal Government took several actions to directly aid
the housing market. It instituted the First-Time Homebuyer Tax Credit between
2008 and 2010, with the goal of stimulating home buying and house prices.
The government also created the Home Affordable Modification Program
(HAMP) and Home Affordable Refinance Program (HARP) to prevent distressed
or underwater borrowers from going into foreclosure. The main distinction
between the two was that HAMP modified a borrower’s existing mortgage contract—often by extending the term or lowering the rate to reduce payments—
whereas HARP loosened underwriting requirements to allow underwater
Creating the Fastest Economic Recovery | 57

<BILL: PHOTO-REDUCE TO FIT?>
$"0- рҊрхѵTimelines for the $.')*) /-4 .+*). ./*
Ҋрш)/# $))$'-$.$.
$))$'c-$.$.t$( '$) , 2007–9

Ҋршt$( '$) , 2020
#$)- +*-/.$/.!$-./ Ҋрш /#, Jan. рр
'- ."'*'# '/# ( -" )4,Jan. тп
- .$ )/-0(+- ./-$/./-1 '!-*(#$),Jan. тр
Diamond Princess "$).,0-)/$) , Feb. ф

2007
*0.$)"0' 0-./.,2$/#/# '-" ./.$)"' Ҋ4 -
- . $)#*( .' .$)/2* ., Feb. 7

)$/ // .- +*-/.!$-./ Ҋрш /#,Feb. сш
*-*)1$-0.- +- ) ..) .+*). 0++' ( )/'
++-*+-$/$*)./.$") $)/*'2,Mar. х

2 )/0-4$))$'Ѷ/# '-" ./ѵѵ.0+-$( ' ) -Ѷ!$' s
!*-)&-0+/4, Apr. 7

- .$ )/-0(+ '- .n/$*)'
e( -" )4,Mar. рт
($'$ .$-./*-*)1$-0. .+*). /.$")
$)/*'2,Mar. рч
$-./“.# '/ -Ҋ$)Ҋ+' ”*- -4
// ,Mar. рш

The - )#)&-$.!-ee5 s2$/#-2'.!-*(
/#- !0)./#/#o'. 0-$/$ .& 4ѵѵ.0+-$(
(*-/"" ., Aug. 7

-$(-4 ' -- $/$'$/4*+ -/$*)',
Mar. сп

*) 4-& /0/0'0) $,0$$/4$'$/4
*+ -/$*)',Mar. ст
/$") $)/*'2, Mar. сц
*(( -$'+ -0)$)"$'$/4*+ -/$*)',
Apr. ру

-(0/$*)$'$/4җҘ$)$/$/ , Dec. 7

сппч

4# &-*/ /$*)-*"-( $,0$$/4
$'$/4*+ -/$*)',Apr. рх

*)*($/$(0'0./ )/ , Feb. 8

4# &-*/ /$*)-*"-() '/#-
)#) ( )//.$") $)/*'2, Apr. су
-$(-4 ' -- $/$'$/4җҘand -(
0-$/$ . )$)"$'$/4җ Ҙ- / , Mar. 8
-/ -).collapses)$.*0"#/4
*-")#. , Mar. 8

))$ and- $ +' $)/*
c*). -1/*-.#$+., Sep. 8
Leh()-*/# -. '- .bankruptcy, Sep. 8
( -" )4*)*($/$'$5/$*)/signed into lawѶ
./'$.#$)", Oct. 8
-(.. /Ҋ& 0-$/$ . *)$'$/4җ Ҙ
$)/-*0 , Nov. 8

сппш

‫ڦ‬рѵсф/-$''$*)" )4*-/"" Ҋ& 0-$/$ .0-#.
-*"-( "$)., Jan. 9
( -$) *1 -4) $)1 ./( )//җҘsigned
into law, Feb. 9

*0- .ѷNew York TimesѸ -))& Ѷ $/#) -Ѷ)0'.*)җспспҘѵ
Note: WHO = World Health Organization.

Ѷ Ѷ) "4 0-$/$ .0'$Ҋ-$1/
)1 ./( )/-*"-(җ Ҙ ./'$.# , Mar. 9

borrowers with negative home equity to take advantage of lower interest rates
through refinancing.
Relative to the Great Recession, the Federal Government has responded
with even greater speed and coordination to COVID-19, and with an even
more expansive range of policies (figure 1-16). The Federal Reserve rapidly cut
the Federal Funds rate target range to 0 percent at the effective lower bound
(0.00–0.25 percent), and it began to reactivate liquidity facilities that it had set
up during the 2007–9 financial crisis. In a matter of just a couple of months, the
Federal Reserve’s balance sheet jumped by over $3 trillion compared with the
five years it took to expand by that amount during the Great Recession. The

58 |

Chapter 1

Creating the Fastest Economic Recovery | 59

‫ڦ‬сѶупп/*
%*$)/!$' -.
)‫ڦ‬фпп+ -
+ ) )/
җۡ‫ڦ‬тпп
$''$*)Ҙ

/$(0'0.
# &.
‫ڦ‬рѶспп/*
%*$)/!$' -.
)‫ڦ‬тпп+ -
+ ) )/
җ‫ڦ‬схп
$''$*)Ҙ

*0. #*'
/ '$ !
*(
!!*-'
*$!$/$*)
-*"-(;
*(
!!*-'
!$))
-*"-(

*- '*.0-
‫ڦ‬хпп2 &'4
.0++' ( )/Ѹрт- (*-/*-$0(;
2 & 3/ ).$*)Ѹ 1$/$*)
(*-/*-$0(
3+) /*
$)'0 . '!Ҋ
(+'*4 )
*/# -.2#*
* )o/,0'$!4!*-
- "0'-
җ‫ڦ‬схч$''$*)Ҙ

) (+'*4( )/
).0-)
‫ڦ‬сф2 &'4
.0++' ( )/Ѹ0+
/*шш2 &
3/ ).$*)

-' . -1
f$'$/$ .
җ‫ڦ‬уфу$''$*)Ҙ;
$-'$) '*)
"0-)/ )
+4-*''"-)/;
/$*)'. 0-$/4
-$/$'0.$) ..
'*)"0-)/
җ‫ڦ‬тс$''$*)Ҙ

$Ҋ)Ҋ -"
0.$) .. '$ !

җ‫ڦ‬цпп$''$*)
''*/ Ѷ‫ڦ‬усп
$''$*)$.+ -. Ҙ;
-' . -1
f$'$/$ .;
)&Ѷ$).0-)
)- $/(-& /
$).0-) ;
0/*$)0./-4
.0++*-/

4# &
-*/ /$*)
-*"-(
җ‫ڦ‬хуш$''$*)Ҙ;
(+'*4
/ )/$*)3
- $/;
'*)
3+).$*)ҝ
(*$!$/$*)

(''
0.$) ..
'$ !
'*)
%0./( )/.

+0'$
# '/#)
- . -#
++-*+-$/$*);
-*1$ -
'$ !0)
җ‫ڦ‬рцф$''$*)Ҙ

0/$*)
/$'$5/$*)
0)
җ‫ڦ‬тр$''$*)Ҙ;
/0 )/
/ ! --'

// and *'
*1 -)( )/$
0).!*-+-*"-(.
җ.*( #$"#'$"#/
/#-*0"#*0//#is
/' Ҙ
($)$./ - 4
S// .)'*'
"*1 -)( )/.Ѷ+$
*0/4/# F -'
"*1 -)( )/
җ‫ڦ‬счс$''$*)Ҙ

*-*)1$-0. '$ !
(+*--4
(+'*4 -Ҋ.$
0)
!0''+4-*''/3
җ‫ڦ‬рфп$''$*)Ҙ
! --';
(+*--4
3+).$*)*! /
+ -/$)" *..
--4&.;
! -red
+4( )/.*!
*$' 0-$/4
/3 ./#-*0"#
( -спсп

30/.Ѷ- $/.
'/#0/$*)
) ! --'.
'/#-
Ҋрс.#**' (+*--4
.0.$$5/$*) +4-*'')
(+'*4 Ҋ.$ ,с-
җ‫ڦ‬рсч$''$*)Ҙ !$'$/4
+ - )/" -+*$)/
.+ )$)"
+4-*''/30/
җ‫ڦ‬рпш$''$*)Ҙ )*( /3
- /

$2.7 trillion
($192 billion
FFCRA, $2.2
trillion CARES,
$321 billion
PPHCEA)

Total
Legislative
Fiscal Relief
$1.4 trillion
($152 billion
ESA 2008,
$787 billion
ARRA, $475
billion TARP)

Sources: Great Recession: Economic Stimulus Act of 2008 (ESA 2008); American Recovery and Reinvestment Act, 2009; Troubled Asset Relief Program (TARP, 2008); COVID-19: Families First
Coronavirus Response Act; CARES Act; Paycheck Protection and Health Care Enhancement Act.

COVID-19
Pandemic
(Nominal)

Relief
Category
Great
Recession
(Nominal)

' рҊсѵThe -' .+*). 0-$)"/# - / ..$*)) Ҋрш

<BILL: MAKE THIS A TURNED TABLE, FULL PAGE, AND PHOTO-REDUCE IF NECESSARY>

Federal Reserve has also created Main Street Lending Facilities to direct relief
to a larger swath of small and mid-sized firms.
The fiscal response to COVID-19 has also been swifter and larger (figure
1-16). During the Great Recession, a fiscal stimulus was rolled out in phases
over the course of a year: the Economic Stimulus Act (ESA) in February 2008,
the Emergency Economic Stabilization Act in October 2008, and the American
Reinvestment and Recovery Act (ARRA) in February 2009. By contrast, during COVID-19 the Federal government passed the Families First COVID-19
Response Act and the CARES Act both within March 2020 (along with the smaller
Coronavirus Preparedness and Response Supplemental Appropriations Act).
Moreover, the CARES Act delivered $2.2 trillion in fiscal relief, compared with
a bit over $800 billion by the ARRA (or about $970 billion after adjusting for
inflation). In terms of composition, both fiscal packages delivered direct aid
to households in the form of rebates and unemployment insurance. The ARRA
also contained a payroll tax cut and direct aid to States to address revenue
shortfalls. Unlike in the Great Recession, however, the CARES Act during COVID19 established the Paycheck Protection Program (PPP), which has disbursed
$525 billion in loans to small businesses to help them maintain payrolls and
avoid insolvency.
Table 1-2 provides a summary comparison of the fiscal response to
COVID-19 to that of the Great Recession. As is evident, not only has the magnitude of legislative fiscal relief during COVID-19 been nearly twice as large overall, but the increased aid has also gone primarily to households and small businesses, with more generous unemployment insurance and Economic Impact
Payments to the former and the novel PPP to the latter. The next subsection
provides a more detailed account of how the policy response to COVID-19 has
been unprecedented in the support provided to low-income workers.

Federal Support for Low-Income Households
A primary focus of the CARES Act and other relief bills has been the provision
of cash and economic support to economically vulnerable households. This
subsection compares these unprecedented measures with those adopted during the Great Recession.
Economic Impact Payments and other tax provisions. In both the COVID-19
recession and the Great Recession, the Federal Government used tax provisions to provide economic support to households. The Economic Stimulus Act
of 2008 (ESA), passed during the Bush Administration, included an individual
income tax “recovery rebate.” The rebates were sent to taxpayers in the form
of stimulus checks. The typical tax filer received a credit of up to $600 for single
filers or up to $1,200 for joint filers. Eligible individuals received an additional
$300 per dependent child. Individuals without a net tax liability were still eligible for the rebate, but only if they had earnings of at least $3,000 annually.
The rebate was phased out at a rate of 5 percent for incomes over $75,000, and
60 |

Chapter 1

$150,000 for those filing jointly (the same as the CARES Act). ARRA, passed in
2009 under the Obama Administration, authorized a Making Work Pay personal
tax credit for 2009 and 2010, which provided a refundable tax credit of up to
$400 for single working individuals and up to $800 per couple. The credit was
phased out for incomes over $75,000 (or $150,000 for joint filers) at a rate of 2
cents per $1 of higher income. ARRA also included one-time stimulus payments
of $250 for seniors, persons with disabilities, and veterans.
During the COVID-19 recession, the Federal Government has also used tax
provisions to provide economic relief to households. The support was larger in
monetary value than in the ESA or ARRA, and it was not limited to households
with Federal income tax liability, so it thereby extended relief to the lowestincome households. In the CARES Act, the U.S. government provided swift
Economic Impact Payments to individuals generally based on 2018 and 2019
tax return information. Those not receiving the advance payments in 2020 can
file for them as a tax credit on 2020 taxes. Although the phase-out rate and
income thresholds are the same as under ESA and ARRA, the CARES Act payments were significantly larger, offering up to $1,200 to individuals and $2,400
to joint filers (El-Sibaie et al. 2020). The CARES Act payments were also larger
for eligible individuals with children. ESA offered an extra $300 tax credit per
dependent child, while ARRA expanded eligibility for the child tax credit. The
CARES Act, by comparison, provided a $500 tax rebate per dependent child
using the same eligibility criteria for dependent children as the child tax credit.
Unlike the ESA tax credit, the CARES tax rebate does not require a minimum tax
liability to receive the full rebate (Marr et al. 2020), meaning that those at the
very lowest end of the income distribution received income support.
Some types of tax relief enacted under ARRA were not paralleled in the
CARES Act. ARRA enhanced the Earned Income Tax Credit by expanding its
coverage and raising the credit claimed by workers with three or more children.
Although these changes were initially enacted on a temporary basis, Congress
later made them permanent. ARRA also subsidized the purchase of cars and
first-time homeowners through an automobile sales tax credit ($1.7 billion
total) and a homeownership tax credit ($6.6 billion).
Workforce programs. In its response to both recessions, the Federal
Government provided support for the Nation’s workforce. Overall, the CARES
Act provided significantly more support. The support was also targeted to
reflect the different nature of the crisis. In the Great Recession, out of the $787
billion ARRA stimulus package, about $12 billion helped finance various public
workforce programs to accommodate expanded participation (table 1-3). State
unemployment insurance agencies received $500 million in administrative
support funding and $7 billion in modernization funds to address increased
demand (BLS 2014). By comparison, the Families First COVID-19 Response
Act authorized $1 billion in additional funding to support UI administration

Creating the Fastest Economic Recovery | 61

' рҊтѵ0)$)"*!%*-*-&!*- -*"-(
)$/$/$1 .0) -/# ( -$ $)1 ./( )/)
*1 -4/Ѷсппш
0)$)"A(*0)/
җ$''$*)s of*''-.Ҙ
($)$./-/$*)
пѵф
* -)$5/$*)
цѵп
Wagner–Peyser Act grants to States
пѵрф
Wagner–Peyser Act reemployment services
пѵсф
0'/
пѵф
$.'*/ *-& рѵсф
$.'*/ *-& -/$*)' . -1
пѵс
$"#-*2/#)( -"$)" )0./-4"-)/.
пѵцф
*0/#
рѵс
**-+.
пѵсф
*0/#0$'
пѵпф
)$*-*((0)$/4 -1$ (+'*4( )/-*"-(
пѵрс
/F0)$)"C/ "*-4

*0- .ѷ0- 0*! *-//$./$.Ѹ-' 4) *- ()җсппшҘѵ
*/ ѷ ۙ*-&!*- )1 ./( )//ѵ

to assist States with processing increased caseloads and expanded programs
(Emsellem and Evermore 2020; Goger, Loh, and George 2020).
Congress also funded additional enhancements and extensions to the
Unemployment Insurance program. In response to the rise in the number of
workers unemployed for more than 26 weeks, Congress enacted a temporary
extension UI. The Emergency Unemployment Compensation Act of 2008 and
its extensions included additional tiers of benefit weeks to supplement regular
State UI and expanded Extended Benefits programs. In combination, between
November 2009 and September 2012, these programs extended the maximum
number of weeks UI recipients could receive benefits for up to 99 weeks.
In 2009, ARRA added to these benefits, providing both for expanded UI
duration and an additional benefit of $25 per eligible worker in weekly UI benefits through temporary Emergency Unemployment Compensation. This benefit enhancement cost the Federal Government $20.1 billion during the period
2009–11. The permanent Extended Benefits program became completely
federally funded through January 1, 2010, and State eligibility rules were
relaxed to make more unemployed workers eligible. These Extended Benefits
cost the Federal Government $24 billion during 2009–11. ARRA also temporarily
suspended the taxation of the first $2,400 of UI benefits.
In response to the COVID-19 recession, Congress both temporarily
extended the duration of UI benefits and increased their level considerably relative to the Great Recession. Under the CARES Act, UI benefits were extended
for up to an additional 13 weeks and States were allowed to eliminate the

62 |

Chapter 1

mandatory 1-week waiting period before benefits can be released to recipients.
The CARES Act also offered a considerable increase in additional UI income—24
times greater than the additional benefit of $25 that was offered during the
Great Recession. Workers claiming UI received a $600 weekly supplement
through July 15, 2020. Furthermore, unlike the Recovery Act, the CARES Act
added a new program to expand eligibility for UI benefits to include the selfemployed, gig workers, workers with limited work history, and other types of
workers who would not otherwise qualify for regular UI benefits. After the $600
weekly supplement expired in July and in the absence of Congressional action,
the Trump Administration extended relief to unemployed workers by issuing a
Presidential Memorandum creating the Lost Wages Assistance Program, which
authorized the use of Disaster Relief Funds to make supplemental payments
of up to $400 ($300 Federal contribution, $100 optional State contribution) per
week for lost wages. Forty-nine states along with Washington, DC and some
US territories ultimately signed up for the program, which provided six weeks’
worth of benefits to every State and territory that applied by September 10.
During the Great Recession, under ARRA, individuals eligible for UI were
referred to the Employment Service for job referral and reemployment services. ARRA allocated an additional $250 million in Reemployment Services
Grants to local employment offices to better serve UI recipients. The Bureau
of Labor Statistics notes that, despite increased funding, the local offices still
faced major constraints, which resulted in increased enrollment in low-cost
services (e.g., orientations, assessments), but smaller increases in expensive
and labor-intensive services (e.g. counseling, education, training). Other
employment services, such as the Workforce Innovation Dislocated Worker
program and the Workforce Innovation Adult program, also received increased
funding (table 1-3).
The CARES Act does not have a parallel to ARRA’s increase in funding for
Reemployment Services Grants and Workforce Innovation and Opportunity Act
formula programs. As outlined in a previous CEA report (2019), many government training programs lack rigorous evidence-based results that demonstrate
their effectiveness in training or retraining workers and helping them find
employment. The CARES Act does, however, provide $345 million in Dislocated
Worker Grants to prevent, prepare for, and respond to COVID-19. In addition,
the act offers incentives to States to adopt or make better-use of short-time
compensation programs, which would allow employers to avoid laying off their
employees by reducing their hours. Under these programs, workers would still
be eligible for UI benefits to make up for their reduced working hours.
The CARES Act goes far beyond ARRA to support the workforce through
its funding of the PPP. The program was designed to support small business employers and their employees during the pandemic. The CARES Act
authorized $349 billion in PPP loans to support payroll and other expenses
for America’s small businesses, self-employed individuals, Tribal business
Creating the Fastest Economic Recovery | 63

concerns, and nonprofit/veterans’ organizations. As part of the PPP and Health
Care Enhancement Act, an additional $310 billion was authorized, bringing
the total amount authorized for the PPP to $659 billion. While the funds will
be used to guarantee and forgive loans, a condition for making the loans fully
forgivable is that no less than 60 percent (originally 75 percent) of the funds be
spent on payroll expenses within a 24-week (originally 8-week) period.
Healthcare. The Federal response to support healthcare during the
COVID-19 recession has been much different from its response in the Great
Recession because of the need to directly address the effects of the COVID-19
health crisis. There was no parallel to this in the Great Recession, which was
driven by a financial crisis rather than a health-related crisis.
During the Great Recession, the Federal response for healthcare focused
on temporarily increasing healthcare benefits for people who lost their jobs.
Before the Great Recession, the Consolidated Omnibus Budget Reconciliation
Act (COBRA) required many employers to provide continued healthcare coverage to workers (and their dependents) who lost their jobs, but it did not require
employers to continue subsidizing the premium payments. ARRA provided a 65
percent subsidy for employers to help cover the premium payments of most
COBRA-eligible workers who lost their jobs between September 2008 and May
2010. This subsidy covered workers and their dependents for up to 9 months
(later extended to 15 months). The CARES Act did not change the terms of
COBRA, but the Department of Labor temporarily extended deadlines for workers who lost their jobs to sign up for coverage and pay premiums.
To respond directly to the COVID-19 health-crisis, the CARES Act established the Provider Relief Fund to support healthcare providers in the midst of
the pandemic. The CARES Act, through the Department of Health and Human
Services, allotted $100 billion to hospitals and other healthcare providers. The
Paycheck Protection Program and Health Care Enhancement Act provided an
additional $75 billion for the Provider Relief Fund to healthcare providers to
reimburse heightened costs and lost revenues that are attributable to COVID19. The Department of Health and Human Services is currently allocating this
$175 billion in aid. The aid includes specific programs to provide safety net
relief to hospitals that serve the most vulnerable segment of the population as
well as rural hospitals and those in small metropolitan areas.
Although this aid is substantial, the portion going to hospitals is unlikely
to fully offset the losses that hospitals have experienced during the pandemic.
The American Hospital Association estimates that the pandemic imposed over
$200 billion in losses on the American healthcare system in the four-month
period between March 1 and June 30. Over 80 percent of this estimated cost is
due to revenue losses from canceled surgeries and other services. This includes
both elective and nonelective procedures, outpatient treatments, and emergency department services. The remaining 20 or so percent of estimated losses
are based on the direct costs of COVID-19 to hospitals: losses from COVID-19
64 | Chapter 1

hospitalizations, additional purchases of personal protective equipment, and
additional support that hospitals provide to their front-line workers.
The CARES Act also provided $25 billion to help increase COVID-19 testing. This includes up to $1 billion to reimburse the cost of testing uninsured
individuals, in addition to the $1 billion previously appropriated for this
purpose by the Families First Coronavirus Relief Act (FFCRA). The FFCRA also,
as amended by the CARES Act, requires Medicare Part B, State Medicaid and
Children’s Health Insurance Programs, and group health plans and health
insurance issuers to cover COVID-19 diagnostic testing without cost sharing for
patients. Uninsured individuals may also obtain COVID-19 diagnostic testing
free of charge under State Medicaid programs, if a State offers this option. The
Centers for Medicare & Medicaid Services has made an accessible and easy-touse toolkit for States to amend their Medicaid programs in order to offer this
service.
Education. During the Great Recession, the Federal Government directed
a considerable portion of stimulus spending to education, allocating $100
billion in additional spending under ARRA. A central goal of the funding was
to avert layoffs in school districts and universities. About half the funding was
allocated to State governors for use in primary, secondary, and higher education through the State Fiscal Stabilization Fund. An additional $10 billion was
targeted to low-income students and about $12 billion was designated to
support students with disabilities. About $17 billion was used to increase the
funding available for Pell Grants for higher education that support students
from low-income households. ARRA also established the American Opportunity
Tax Credit, which modified an existing education credit (the HOPE credit) by
relaxing income-based eligibility limits to cover more students, qualifying
more expenses for the credit, and allowing the credit to be claimed not only for
study at two-year institutions but also for study at four-year higher education
institutions.
Under the CARES Act, the Federal Government provided $31 billion in
emergency relief to educational institutions. This includes about $13 billion
for K-12 schools allocated mainly in proportion to a State’s enrollments of
low-income students. Another $14 billion is allocated to higher education,
with most of the allocation based on an institution’s share of Pell Grant
recipients, but with about $1 billion allocated to Historically Black Colleges
and Universities and other institutions serving students of color, which are
discussed further in chapter 11 of this Report. Another $3 billion in relief is for
governors to distribute to schools or higher educational institutions that have
been particularly affected by COVID-19.
A major difference between the Great Recession and the current crisis
is the large number of school closures across the country in response to the
pandemic. Between the first and third weeks of March, close to 100 percent of
kindergarten, primary, and secondary schools were shut down. These closures
Creating the Fastest Economic Recovery | 65

have had a substantial negative effect both on the U.S. economy and on children themselves. Prorated estimates based on analyses by Angrist and Krueger
(1992) and Bhuller, Mogstad, and Salvanes (2017) suggest that children are
likely to experience a persistent 2.3–3.7 percent decline in future earnings as
a result of lower human capital accumulation from the shortened school year.
Meanwhile, parents who have had to miss work entirely because of childcare
duties induced by school closures may also experience a reduction in lifetime
earnings. The CEA estimates that 18 percent of the workforce may experience
a persistent 1 percent drop in lifetime earnings because of lost job experience
due to school closures. The effects are likely to be particularly severe for earlycareer single mothers, who will experience not just lower earnings but also
less secure job prospects. Accordingly, the safe reopening of schools will help
to boost the economy and support economically vulnerable students and their
families.
Supplemental Nutrition Assistance Program. The Federal Response in
both recessions included support for the Supplemental Nutrition Assistance
Program (SNAP), the Federal program that provides nutritional assistance to
help America’s neediest families purchase food. During the Great Recession,
ARRA allocated $40 billion in additional SNAP benefits for all participants and
raised the minimum benefits. As a result of these changes, in 2009, the average
monthly SNAP benefit increased by $21. In addition to increasing the monthly
benefit, ARRA suspended work requirements for nondisabled, childless adults
between April 2009 and September 2010.
The Families First COVID-19 Response Act provided authority for work
requirement waivers and SNAP benefit increases up to the maximum allotment
for households not already receiving the maximum. The CARES Act provided
over $15 billion in additional contingency funding for the increased costs
associated with the FFCRA provisions, as well as anticipated increased participation in SNAP. As provided by the FFCRA and the CARES Act, the Department
of Agriculture also provided waivers of certain requirements so that nutrition
programs could reach families and children during the social-distancing
restrictions. The FFCRA also suspended work requirements for nondisabled,
childless adults through the month after the end of the COVID-19 public health
emergency.
Housing assistance programs. During the Great Recession, the Federal
response under ARRA provided $13.6 billion for programs administered by
the Department of Housing and Urban Development (HUD), including $1.5
billion for the Homelessness Prevention and Rapid Re-Housing Program. As
discussed in chapter 2 of this Report, the CARES Act provided housing relief to
homeowners and renters in the form of forbearance for federally backed mortgages and a 120-day eviction moratorium that was subsequently extended
by the Trump Administration via Executive Order 13945, Fighting the Spread
of COVID-19 by Providing Assistance to Renters and Homeowners. The CARES
66 | Chapter 1

Act also allocated $12.4 billion for programs administered by HUD for fiscal
year 2020. The funding includes $4 billion for the homeless who are among the
most vulnerable and hardest hit by the pandemic. These funds will support the
Emergency Solutions Grants program, which assists homeless populations or
populations at risk of becoming homeless. About $3 billion of these funds are
being used to operate emergency shelters (covering food, rent, security, etc.),
make even more emergency shelters available, provide essential services to
homeless populations (including childcare, employment assistance, and mental health services), and prevent individuals from becoming homeless through
rapid rehousing.

Conclusion
The COVID-19 pandemic has had a profound effect on what had been a robust
U.S. economy at the start of 2020. The Blue Chip panel of professional forecasters immediately began to sharply revise down its 2020 GDP projections
in March as the pandemic was taking hold, as did the Federal Reserve and
the OECD when updating their forecasts. Instead of predicting GDP growth
of about 2 percent for 2020, all three issued dire warnings of a GDP contraction of about 6 percent to as much as 12 percent—which would have marked
the steepest contraction since the 1930s. However, the swift and dramatic
fiscal interventions implemented in late March and early April by the Federal
Government paid dividends throughout the summer, and the U.S. economy
consistently outperformed expectations.
As a result, as of the fall of 2020, all three leading forecasters were taking
a much more sanguine view of GDP growth for the year, predicting that GDP will
end up falling by less than 4 percent. Whether this robust recovery maintains a
healthy pace depends partly on the progression of virus mitigation efforts and
the continuation of appropriate and responsive levels of fiscal support. The
chapters that follow provide an in-depth discussion of the major components
of the fiscal response and their ensuing effects on different aspects of the U.S.
economy.

Creating the Fastest Economic Recovery | 67

x
Chapter 2

Prioritizing America’s Households
The economic and health crises stemming from the COVID-19 pandemic
required a coordinated response from all levels of government to protect the
livelihoods of Americans. The Trump Administration took decisive action and
worked with Congress to pass and sign three major bills in March 2020—the
largest of which was the Coronavirus Aid, Relief, and Economic Security
(CARES) Act—to address the economic fallout from the pandemic. As CARES Act
provisions began to expire or dissipate in August, and in the absence of further
Congressional action, President Trump followed up with a series of executive
actions that extended further relief to American households.
A key goal of these policies was to provide financial support to American
households weathering the sharp pandemic-fueled economic contraction.
These policies were highly successful against this unprecedented event.
Even as the unemployment rate climbed from a 50-year low of 3.5 percent
in February 2020 to 14.7 percent two months later in April 2020, household
incomes increased, thanks to Economic Impact Payments and to expanded
and enhanced Unemployment Insurance. Lower-income households generally experienced the largest percentage income increases, and their monthly
income in every month through at least August 2020 exceeded pre-COVID levels.
In addition to providing direct financial relief, the CARES Act and follow-up
executive branch actions from the Trump Administration protected Americans
against the risk of eviction and student loan defaults. Evictions fell below preCOVID levels in cities across the United States, averting bouts of homelessness
or shared housing that could pose additional health risks in the midst of the
pandemic.

The long-run success of actions taken to support households will depend on
the pace and depth of the economic recovery. Between April and November,
the unemployment rate fell by 8.0 percentage points, from 14.7 percent to 6.7
percent, the largest seven-month decline on record. Almost 60 percent of all
jobs lost between February and April had been recovered by November, as
employment increased by 12 million over this period.
Continued economic recovery, supported by President Trump’s executive
actions designed to extend assistance beyond the expiration of CARES Act
provisions, can pave the way back to the same strong economy that prevailed
during the first three years of the Trump Administration, which was spurred
in part by the Tax Cuts and Jobs Act and other pro-growth policies. Between
2016 and 2019, median net worth increased by 18 percent, with an increase of
32 percent for Black-headed households and 64 percent for Hispanic-headed
households. Median income increased by 9.7 percent between 2016 and 2019,
and the one-year 6.8 percent increase in 2019 was the largest one-year increase
ever recorded. Poverty hit a record low in 2019 for all racial and ethnic groups,
and fell by the largest amount (1.3 percentage points) in over 50 years. With
continued pro-growth policies, including deregulation and the continued
benefits of the Tax Cuts and Jobs Act, the pre-COVID economy can be attained
again, and households will continue to benefit from the gains experienced during the first three years of the Trump Administration.

he partial shutdown of the U.S. economy in response to COVID-19 was
unprecedented. Over 90 percent of Americans were affected by statewide school closures and restrictions on bars and restaurants by late
March and were subject to State-level stay-at-home orders by early April. As a
result of these events, between February 2020 and April 2020 the unemployment rate increased from the lowest level in over 50 years (3.5 percent) to the
highest level since the Great Depression (14.7 percent). Aggregate, pretransfer
disposable income in the United States fell by 9 percent between February and
April, the largest two-month reduction ever recorded. Because job losses were
concentrated among lower-wage workers, the reduction in pretransfer income
hit lower-income households the hardest, threatening their ability to pay for
rent, food, and other basic necessities.
70 | Chapter 2

Due to the rapid and unprecedented actions taken by President Trump
and Congress, these harmful effects on American households were strongly
mitigated. On March 20, 2020, President Trump delayed Tax Day, providing
liquidity to Americans with tax liabilities. And following two other important pieces of legislation, on March 27, 2020, President Trump signed the
Coronavirus Aid, Relief, and Economic Security (CARES) Act into law, providing
$2.2 trillion in relief for households and businesses. A family with two children
and an income below $150,000 received an Economic Impact Payment of
$3,400, almost twice as much as the maximum $1,800 stimulus checks provided
during the Great Recession. And unlike the Great Recession stimulus payments,
full Economic Impact Payments were available to the lowest-income households with no tax liability. The CARES Act also provided unprecedented relief to
those workers who lost their jobs. A supplemental $600 weekly Unemployment
Insurance (UI) payment ensured that most workers who lost their jobs did not
experience a reduction in income, and eligibility was expanded to workers not
typically eligible for UI benefits. The CARES Act further placed a moratorium on
foreclosures and evictions in homes with federally backed mortgages. Earlier
legislation extended paid leave benefits for families that could not work due to
illness or to care for children affected by school closures.
President Trump provided additional relief to households when CARES
Act provisions expired, and Congress was unable to reach a consensus on
extensions. He issued several important executive actions on August 8, 2020,
providing $300 a week in supplemental Federal assistance to unemployed
workers; deferring the employee portion of payroll taxes through the end of
2020; issuing an order to assist renters unable to pay their rent, and ultimately
imposing a moratorium on evictions from all rental housing through the end
of 2020; and extending the deferral of student loan payments with no interest
through the end of 2020. These executive branch actions ensured that many
households would continue to receive relief in the absence of further legislative
packages.
Due to the CARES Act and subsequent executive action by the Trump
Administration, poverty and income inequality fell, and most workers who lost
their jobs experienced no income loss while receiving supplemental unemployment benefits. In the months immediately after passage of the CARES Act,
households across the income distribution saw an increase in income relative
to pre-COVID levels. The gains were largest for the lowest-income households.
For example, for households at the 25th percentile, monthly incomes spiked
by 127 percent in April, largely due to Economic Impact Payments; and these
households were still above pre-COVID levels from May through August, largely
due to expanded UI benefits. In fact, Economic Impact Payments alone were
large enough to keep a family of four out of poverty for 1.5 months even if
they lost all other income. Expanded UI benefits ensured that the vast majority of unemployed workers received at least as much from UI as they did from
Prioritizing America’s Households

| 71

working. Though these UI payments would typically create strong work disincentives, the partial economic shutdown between April and July mitigated
such concerns (Altonji et al. 2020; Bartik et al. 2020; Marinescu, Skandalis, and
Zhao 2020). The somewhat reduced emergency lost wages assistance issued
under President Trump’s executive action in August alleviated some of the
work disincentives of the $600 payments as the recovery from March and April
proceeded, while continuing to provide additional support to unemployed
workers.
These actions helped pave the way for a strong economic recovery.
Between April and November, the unemployment rate fell by 8.0 percentage
points, from 14.7 percent to 6.7 percent, the largest seven-month decline
on record. Continued economic recovery, combined with President Trump’s
executive actions extending assistance to many households beyond the expiration of key CARES Act provisions, can pave the way to attaining the historically
strong pre-COVID labor market and overall economy.
A strong economy is the most effective tool for lifting up households in
the long term. From when President Trump was elected in 2016 until 2019,
median net worth increased by 18 percent, with the biggest gains for minority
groups. Median net worth increased by 32 percent for Black-headed households and by 64 percent for Hispanic-headed households (Federal Reserve
Board of Governors (2020a, 2020c). Median income increased by 9.7 percent
between 2016 and 2019, and increased by 6.8 percent in 2019 alone, the largest
one-year increase ever recorded. In 2019, poverty fell by the largest amount
(1.3 percentage points) in over 50 years and hit a record low. Black poverty
fell below 20 percent for the first time ever. Continued pro-growth policies,
including deregulation and the continued benefits of the Tax Cuts and Jobs
Act (TCJA), can help ensure that the pre-COVID economy can be attained again,
allowing households to continue seeing the gains experienced during the first
three years of the Trump Administration.1

The Strength of the Pre-COVID
Economy and the COVID-19 Shock
The Trump Administration’s policies have focused on spurring economic
growth and job creation. Deregulation has reduced the costs for businesses to
invest and hire workers. Tax reform has encouraged new capital investment
and has reduced taxes on households that impose high effective tax rates on
work, particularly at the lower end of the income distribution. Other policies—
such as expanded childcare assistance for low-income workers, Opportunity
Zones, and record investments to lessen the opioid epidemic—have helped
1 The CEA previously released research on some of the topics discussed in this chapter. The text of
this chapter builds on the CEA report “Evaluating the Effects of the Economic Response to COVID19” (CEA 2020).

72 |

Chapter 2

spur job growth for those remaining on the sidelines of the labor market. Until
COVID-19 struck, the result of these policies was higher economic growth and a
strong labor market, especially for the most disadvantaged Americans.
Between January 2017, when President Trump took office, and February
2020, the U.S. unemployment rate fell from 4.7 to 3.5 percent, the lowest level
in 50 years. Traditionally, disadvantaged Americans experienced the largest
labor market gains. Between January 2017 and February 2020, the unemployment rate for Black Americans fell by 1.7 percentage points and for Hispanic
Americans by 1.4 percentage points, which was even larger than the overall
decline of 1.2 percentage points.
The rise in labor demand not only brought more workers into the
workforce but also increased wages. Real average hourly earnings rose 3.2
percent between January 2017 and February 2020. Wage growth was fastest
for the lowest-wage workers, who through the first three years of the Trump
Administration saw a nominal wage increase of 11.7 percent—4.2 percentage
points higher than the growth of median wages for all workers during the same
period.
Increased employment and wages translated into large income gains for
households. Between 2016 and 2019, U.S. median pretax household income
increased by 9.7 percent. Due to the TCJA, after-tax income grew even faster.
For example, a family of four with an income of $82,500 now pays $2,300 less in
taxes than before the TCJA, according to the Tax Policy Center’s Tax Cuts and
Jobs Act Calculator.
Households across the income distribution experienced income gains
during the first three years of the Trump Administration. Pro-growth policies
reduced poverty by 2.2 percentage points (6.6 million people) between 2016
and 2019. And the poverty rate reached an all-time record low of 10.5 percent
in 2019. All racial and ethnic groups reached record low poverty rates, with the
Black poverty rate falling below 20 percent for the first time ever.
Family wealth also increased during the Trump Administration. Between
2016 and 2019, overall median net worth increased by 18 percent, with
increases of 32 percent for Black-headed families and 64 percent for Hispanicheaded families. Net worth increased the most for lower-middle-wealth
families (those between the 25th and 50th percentiles of net worth), who saw
a 22 percent increase in mean net worth. Homeownership increased by 1.2
percentage points between 2016 and 2019, the first three-year increase documented by the Survey of Consumer Finances since 2004 (Federal Reserve Board
of Governors 2020b, 2020d).
The strong economic growth that lifted up all households between
2016 and 2019 was disrupted by the COVID-19 pandemic. Between February
2020 and April 2020, the unemployment rate spiked, from 3.5 percent to 14.7
percent—and this 11.2-percentage-point increase alone was larger than the
peak unemployment rate reached during the Great Recession. Job vacancies
Prioritizing America’s Households

| 73

fell more than 40 percent by late April (Forsythe et al. 2020). The Congressional
Budget Office forecasted in May 2020 that the unemployment rate would be
15.8 percent in 2020:Q3 and 11.5 percent in 2020:Q4, which would mean an
unemployment rate that continued to exceed the Great Recession’s peak for
the remainder of 2020. In reality, the unemployment rate fell lower than these
predictions; in November, it was only 6.7 percent, well below what the office
(CBO 2020) had predicted.

Policy Responses Providing Household Relief
In response to the sudden and severe shock caused by COVID-19, Congress
quickly passed and President Trump signed into law three pieces of legislation
in March 2020: the Coronavirus Preparedness and Response Supplemental
Appropriations Act, on March 6, 2020; the Families First Coronavirus Response
Act, on March 18, 2020; and the CARES Act, on March 27, 2020 (which was supplemented in April by the Paycheck Protection and Health Care Enhancement
Act). After certain provisions of these acts expired, and in the absence of forthcoming legislation, President Trump issued a series of important executive
actions on August 8, 2020: an Executive Order on Fighting the Spread of COVID19 by Providing Assistance to Renters and Homeowners (White House 2020a);
a Memorandum on Deferring Payroll Tax Obligations in Light of the Ongoing
COVID-19 Disaster (White House 2020d); a Memorandum on Authorizing the
Other Needs Assistance Program for Major Disaster Declarations Related
to Coronavirus Disease 2019 (White House 2020b); and a Memorandum on
Continued Student Loan Payment Relief During the COVID-19 Pandemic (White
House 2020c). This section summarizes the provisions of these laws and executive actions that have provided direct relief to American households.

March 2020 Legislative Acts
The Coronavirus Preparedness and Response Supplemental Appropriations
Act, which was signed into law by President Trump on March 6, 2020, provided
$8.3 billion to fund the initial health response to COVID-19. The funding focused
on vaccines, therapeutics, testing, and general responses to the health emergency, in addition to funding international relief efforts.
The Families First Coronavirus Response Act (FFCRA), which was signed
into law by President Trump on March 18, 2020, provided assistance for households and State governments at an estimated cost of $192 billion. The FFCRA
required certain employers with fewer than 500 employees to provide their
employees with paid sick and family leave for COVID-related work absences,
which would be fully reimbursed by the Federal government through refundable tax credits. To help fund the provision of leave benefits up front, firms
were allowed to access tax withholdings that would otherwise be required
to be deposited with the Internal Revenue Service (IRS), or to receive the tax

74 |

Chapter 2

credit as an advanced payment from the IRS for the amount not covered by
previously withheld taxes. Workers were entitled to 2 weeks of paid sick leave
covering up to 100 percent of wages, and to an additional 10 weeks of paid
family and medical leave covering up to 67 percent of wages, with certain
caps on wages. The FFCRA also increased Federal funding for Unemployment
Insurance Extended Benefits and Medicaid, suspended work requirements
in the Supplemental Nutrition Assistance Program, and provided no-fee
COVID-19 testing and emergency care for all Americans covered by Medicare,
Medicaid, and the Children’s Health Insurance Program.
The largest legislative package was the CARES Act, which was signed into
law by President Trump on March 27, 2020. The CARES Act provided $2.2 trillion
in relief to households and businesses affected by COVID-19. For context, the
major legislative response to the Great Recession, the American Recovery and
Reinvestment Act of 2009, provided $836 billion over 10 years (in 2009 dollars).
Economic Impact Payments, at a cost of $292.4 billion (JCT 2020), were a
key provision of the CARES Act intended to provide immediate relief to households. Each eligible adult could receive up to $1,200 and $500 for each qualifying child, and these payments were phased out at higher incomes. A family
making less than $150,000 a year with two parents and two children would
receive $3,400, even if they had no tax liability. By contrast, stimulus payments
during the Great Recession offered a family of four a maximum of $1,800 and
offered no payment at all to those with no tax liability and less than $3,000 in
qualifying income. Economic Impact Payments were distributed quickly, with
the IRS reporting that it had sent out nearly $267 billion in payments to 159
million Americans by June 3.
The CARES Act provided an additional $347 billion in targeted relief to
Americans who lost their jobs. Federal Pandemic Unemployment Compensation
(FPUC) offered every beneficiary of unemployment insurance an additional
$600 a week in unemployment benefits from March 29, 2020 through July 31,
2020. For example, a worker typically earning $400 a week may receive $200
in regular UI benefits upon becoming unemployed. Under FPUC, the worker
would receive an additional $600, for a total of $800 per week. Pandemic
Emergency Unemployment Compensation provided an additional 13 weeks
of UI benefits for workers who exhaust their regular State benefits, for a total
of 39 weeks of coverage in most States (in addition to potential coverage by
Extended Benefits). Pandemic Unemployment Assistance granted UI benefits
to workers not eligible for regular State unemployment insurance benefits,
such as self-employed workers, gig workers, business owners, independent
contractors not participating in the UI elective coverage program, and workers with insufficient work history to normally receive unemployment benefits.
This assistance for unemployed workers was complemented by the Paycheck
Protection Program (PPP), which helped ensure that businesses could keep
their workers on payroll and avoid the need to draw unemployment assistance.
Prioritizing America’s Households | 75

The CARES Act also provided loan repayment assistance. Homeowners
with federally backed mortgages who experienced financial hardship due to
COVID-19 were allowed to suspend payments for up to 180 days, with the possibility of an extension of up to 180 more days. During this period, no interest
or fees would accrue. The CARES Act also prohibited foreclosures on homes
with federally backed single-family mortgages for at least 60 days starting on
March 18, 2020, and prohibited evictions of tenants in certain federally supported rental properties for 120 days starting March 27, 2020. To allow families
to borrow money if needed, holders of individual retirement accounts (IRAs)
that were adversely affected by COVID-19 could take a distribution from their
IRA and treat this distribution as a tax-free rollover, provided they recontribute
the amount within three years. The CARES Act also ensured that consumers’
credit did not suffer due to the virus; if consumers had an agreement with their
lender to delay payments or make a partial repayment, they would not receive
a negative credit report.
The CARES Act also included provisions to protect student loan borrowers. Employers were provided with the ability to make up to $5,250 in student
loan payments through December 31, 2020 for each employee without incurring taxes. In addition, through September 30, 2020, student loan payments
and interest accruals for Department of Education-held Federal student loans
were suspended, and involuntary collections related to student loans through
wage garnishments, tax refund reductions, and negative credit reporting were
also suspended for loans held by the Department of Education.
Finally, the CARES Act also allowed for $150 billion in State and local government aid. Because many State and local governments, particularly those
without savings or whose revenues rely heavily on sales taxes, have struggled
to retain employees during the pandemic, this measure is estimated to have
saved over 400,000 public sector jobs (Green and Loualiche 2020).

President Trump’s Executive Actions
Key CARES Act provisions expired in July, and, although the economy was in
the midst of a strong recovery, a substantial share of Americans had yet to
return to work. The last week of July was the final week for which the CARES Act
provided enhanced UI benefits to unemployed workers. The moratorium on
evictions and foreclosures in homes with federally backed mortgages expired
on July 24. In the absence of Congressional action, the Trump Administration
took a series of executive actions on August 8, 2020.
President Trump’s Memorandum on Authorizing the Other Needs
Assistance Program for Major Disaster Declarations Related to Coronavirus
Disease 2019 directed up to $44 billion to be provided in Federal lost wages
assistance. In order to be eligible for Federal lost wages assistance, claimants
were required to self-certify that they were unemployed or partially unemployed due to COVID-19 and that they had already received at least $100 a
76 | Chapter 2

week in benefits. As a result, in addition to their regular UI benefits, claimants
were eligible for up to another $400 a week, $300 of which was provided by the
Federal Government. These benefits were set to terminate when Federal funds
were exhausted, but no later than December 6, 2020.
Executive Order 13945 aimed to minimize evictions and foreclosures
and thereby prevent homelessness or shared housing situations during the
pandemic. The Centers for Disease Control and Prevention (CDC) reported
that some racial and ethnic groups were disproportionately more likely to
be evicted, and that homeless shelters and shared housing are particularly
susceptible to COVID outbreaks. The order authorized the CDC Director to
temporarily halt evictions for failure to pay rent, the Secretary of the Treasury
and the Secretary of Housing and Urban Development (HUD) to identify available Federal funds for temporary financial assistance to homeowners suffering
financial hardship resulting from COVID, and the Secretary of HUD to aid homeowners and renters in avoiding foreclosure, for example, by providing housing
authorities or landlords with financial assistance.
In addition, the Director of the Federal Housing Finance Agency (FHFA)
was directed to review resources that might be used to prevent evictions and
foreclosures due to COVID-19. In response to Executive Order 13945, FHFA
extended the moratorium on foreclosures in homes with federally backed
mortgages through the end of 2020, and CDC declared that eligible renters in
any type of property facing potential homelessness or shared housing situations could not be evicted.
President Trump’s Memorandum on Deferring Payroll Tax Obligations
in Light of the Ongoing COVID-19 Disaster authorized the Secretary of the
Treasury to defer certain payroll tax obligations for Americans in need, relaxing temporary liquidity constraints for workers. At the employer’s discretion,
this deferral was available to employees with pretax biweekly wages below
$4,000. In addition, President Trump instructed the Secretary of the Treasury to
explore additional avenues to eliminate the obligation to eventually pay these
deferred taxes.
Finally, President Trump’s Memorandum on Continued Student Loan
Payment Relief During the COVID-19 Pandemic extended the deferment of
payments and waived interest on student loans held by the Department of
Education through December 31, 2020.

The Impact of Policies in Providing
Relief to Households
Due to their magnitude and coverage, the legislative acts and executive actions
taken to counter the negative consequences of COVID-19 had large effects on
U.S. households. As shown in figure 2-1, in 2020 real disposable income excluding government transfers experienced the largest two-month decline on record
Prioritizing America’s Households | 77

$"0- сҊрѵ*)/#'4 $.+*.' -.*)' )*( Ѷспсп
Dollars (billions)

рѶхпп

) (+'*4( )/ ).0-) /. сп уѵр+ - )/

$)- .
!-*(
-0-4

*)*($
(+/4( )/.

рѶупп

*1 -)( )//-).! -.җ' .. and Ҙ

рѶспп

рѶппп

$.+*.' + -.*)'$)*( җ' ../-).! -.Ҙ

чпп

хпп
)Ҋсп

-Ҋсп

4Ҋсп

0'Ҋсп

+Ҋсп

*0- .ѷ0- 0*!*)*($)'4.$.Ѹ '0'/$*).ѵ
Note: UI = Unemployment Insurance benefits; EIPs = Economic Impact Payments.

between February and April (8.7 percent) and remained suppressed through
August. However, real disposable income including government transfers
experienced the largest two-month increase on record between February and
April (13.1 percent) and remained elevated through the time of publication.
The historic rise in posttransfer disposable income was a result of
CARES Act provisions that provided relief to households. In combination,
total Economic Impact Payments and Unemployment Insurance benefits
paid between April and August were over twice as large as the loss in pretransfer disposable income incurred over the same period. Economic Impact
Payments alone replaced 79 percent of the total reduction in pretransfer
disposable income, and UI benefits on their own replaced 126 percent of the
total reduction in pretransfer disposable income. This is largely a result of the
$600 Federal UI weekly supplement. Ganong, Noel, and Vavra (2020) estimate
that 76 percent of workers who were eligible for regular UI benefits in April
through July received more in unemployment assistance than they would have
received from their typical earnings. Though such assistance would normally
create severe employment disincentives, these concerns were mitigated by the
health benefits of staying home during the pandemic.
As a result of the Federal Government’s unprecedented response to the
pandemic-induced economic crisis, lower-income households experienced the
largest income gains during the COVID crisis. Figure 2-2 simulates the trajectory
of household income at different points of the income distribution—with and

78 | Chapter 2

$"0- сҊсѵ ) 3*! *0. #*' )*( 4 - )/$' Ѷспсп
$/#- '$ !

$/#*0/- '$ !

Index of household income (100 = February 2020)
сфп
спп
рфп
рпп

+4
0)
0'4
0"0./

фп

рп

сф
фп
цф
Household income percentile

шп

*0- .ѷ ).0.0- 0Ѷ0-- )/*+0'/$*)0-1 4Ѷ*)/#'4)))0' *$')
*)*($0++' ( )/Ѹ'0'/$*).ѵ
Note: For the methodology used here, see CEA (2020).

without Economic Impact Payments and expanded UI. Without these provisions, a household at the 10th percentile of the income distribution would have
experienced a 10 percent reduction in income in April 2020 compared with its
February 2020 level, and its income would have remained 7 percent lower by
August. However, because of expanded UI and the Economic Impact Payments,
its monthly income was 127 percent higher in April, 42 percent higher in May, 15
percent higher in June and July, and 7 percent higher in August compared with
February 2020. The spike in income in April is largely a result of the Economic
Impact Payments, while the continued elevated income in May through August
is largely a result of expanded UI.
Because figure 2-2 includes all households, it does not show how important the CARES Act and later executive actions were for preserving the income
of households experiencing job losses. Figure 2-3 provides a more specific
example of a household with two adults and two children, with one worker
who loses their job starting in April 2020 and where all income is assumed to
come from earnings. The worker in the “low-wage” household is assumed to
earn $500 a week, and the worker in the “high-wage” household is assumed to
earn $1,500 a week.
Without expanded UI and the Economic Impact Payment, the illustrative
low-wage household would have experienced a 50 percent reduction in income

Prioritizing America’s Households

| 79

$"0- 2-тѵ ) 3*! *0. #*' )*( !*- 3(+' *0. #*'.Ѷ
спсп
With relief
W$/#*0/- '$ !

Household income (100 = February 2020)

упп
тпп
спп

*2Ҋ2" #*0. #*'

0"0./

0'4

0"0./

0'4

рпп

$"#Ҋ2" #*0. #*'

*0- .ѷ ).0.0- 0Ѷ0-- )/*+0'/$*)0-1 4*)/#'4)))0'*$')*)*($
0++' ( )/Ѹ'0'/$*).ѵ
Note: A low-wage household earns $500 a week, and a high-wage one earns $1,500. For the
methodology used here, see CEA (2020).

$"0- ѵ ) 3*! *0. #*' )*( !*- 3(+'

in April through August, while the illustrative high-wage household would have
experienced a 68 percent reduction in these two months. As a result of Federal
action, the low-wage household instead experiences a 240 percent increase in
income in April, a 70 percent increase in May through July, and a 10 percent
increase in August relative to its income in February. The high-wage household
instead experiences a 28 percent increase in April, a 28 percent decrease in May
through July, and a 48 percent decrease in August. Thus, not only did Federal
action provide greater income protection for both households relative to the
counterfactual scenario with no CARES Act, but also provided greater income
protection for lower-wage households than higher-wage households.
The especially large increases in incomes for lower-income households
can also be seen in reductions in poverty in the months immediately following
the CARES Act. Han, Meyer, and Sullivan (2020) use near real-time data from
the monthly Current Population Survey to estimate poverty rates each month
based on the previous 12 months of income. In updated analysis, they find that
the poverty rate in every month between March and September was near or
below the pre-COVID poverty rate of 11.0 percent in February 2020 (see figure
2-4). Han, Meyer, and Sullivan estimate that CARES Act provisions reduced
the poverty rate by 4.0 percentage points in the 12 months ending June 2020.
Parolin, Curran, and Wimer (2020) project that CARES Act provisions could
significantly reduce the poverty rate for calendar year 2020.

80 |

Chapter 2

*0. #*'

$"0- сҊуѵ - )/ '*2 -'*1 -/4 $) Ѷспрш–сп
Population below 100 percent of the Federal poverty line (percent)
ру

+. сп

рт

рс

рр

рп

ч
). рш

+-. рш

0'. рш

/. рш

). сп

+-. сп

0'. сп

Source: Meyer and Sullivan (2020).

As of June 16, data from the Census Bureau Household Pulse Survey
indicate that the vast majority (85.3 percent) of households had received an
Economic Impact Payment (figure 2-5). The quick timing of Economic Impact
Payments helped ensure that households could meet their basic needs.
Indeed, as of June 16, only 12.8 percent of households reported mostly using
their Economic Impact Payment to add to their savings accounts, and 13.5 percent mostly used it to pay off debt such as car loans, student loans, and credit
cards. By contrast, about 59.1 percent of households used their Economic
Impact Payment to pay for expenses such as food, clothing, and shelter. Baker
and others (2020) show that over 20 percent of Economic Impact Payments
were spent within 10 days of receipt, and spending increased the most for
food, rent, bills, and nondurables. Chetty and others (2020) find that Economic
Impact Payments had a large effect on spending by low-income households,
allowing them to return their spending levels to pre-COVID levels by late April.
In addition to helping ensure that households did not experience income
losses, the CARES Act attempted to help households maintain housing stability
by halting all foreclosures and evictions for properties with federally backed
mortgages. Estimates by the Federal Reserve Bank of Atlanta suggest that this
covered between 28 and 46 percent of all rental units. This partial eviction
moratorium, in combination with local eviction moratoriums in many cities,

Prioritizing America’s Households

| 81

$"0- сҊфѵ *0. #*'s’ $+/)U. *!*)*($
(+/4( )/Ѷспсп
$)*/)* .)*/ 3+ //*- $1 ./$(0'0.+4( )/
*./'4/*/*.1$)".
*./'4/*+4*!! /җ-'*).Ѷ./0 )/'*).Ѷ- $/-.Ҙ
*./'4/*+4!*- 3+ ). .җ!**Ѷ'*/#$)"Ѷ.# '/ -Ѷ /ѵҘ
Households (percent)
рпп
шп
чп
цп
хп
фп
уп
тп
сп
рп
п

0) рх

0) ст

0) тп

10
12

0'4ц

0'4ру

0'4ср

*0- .ѷ ).0.0- 0Ѷ *0. #*'0'. 0-1 4Ѹ'0'/$*).ѵ

helped reduce evictions to below pre-COVID levels. According to data from the
Eviction Lab, evictions in all the cities that it tracks were on average 66 percent
lower in April through August 2020 than in February (figure 2-6).
The Trump Administration’s temporary nationwide moratorium on evictions for eligible renters beginning on September 4 appears to have helped
reduce evictions as well. Figure 2-7 shows evictions in the nine cities tracked
by the Eviction Lab that did not have a local eviction moratorium at the time
of the CDC order. Relative to the total number of evictions in these nine cities
during the week beginning August 30 (before the CDC order), evictions were 41
percent, 11 percent, and 30 percent lower during the next three weeks.
One risk with an eviction moratorium is that nonpayment of rent could
have increased, leaving landlords unable to pay mortgages and other costs.
However, unprecedented income support via Economic Impact Payments
and expanded UI benefits may have mitigated this problem. In fact, data from
the National Multifamily Housing Council (2020) show that the rate of missed
rental payments in multifamily housing properties had increased by only 1 to 2
percentage points in May through September of 2020, compared with the same
month one year earlier.

82 | Chapter 2

$"0- сҊхѵ1$/$*).$) ' / $/$ .4*)/#Ѷспсп
Eviction filings (thousands)
уп
тф
тп
сф
сп
рф
рп
ф
п
)ѵ

-ѵ

+-ѵ

0)ѵ

0'ѵ

0"ѵ

+ѵ

*0- .ѷ +0-)Ѷ *0$.Ѷ) .(*)җспспҘѸ '0'/$*).ѵ
Note: Eviction counts are for Austin, Boston, Bridgeport, Charleston, Cincinnati, Cleveland,
Columbus, Fort Worth, Gainesville, Hartford, Houston, Jacksonville, Kansas City, Memphis,
Milwaukee, Phoenix, Pittsburgh, Richmond, Saint Louis, Tampa, and Wilmington.

рфп

$"0- сҊцѵ1$/$*) $'$)". $) ' /ed$/$ .Ѷспсп
$)$))/$

рпп

' 1 ')
фп

*0./*)

)..$/4

*'0(0.

*-/*-/#

$'20&

*/'

Index of eviction filings (base period week beginning Aug. 30 = 100)
0'ф
0'рш
0"с
*-
-җ +ѵуҘ
тфп

0"рх

*0- .ѷ +0-)җспспҘѸ '0'/$*).ѵ
+ѵ спѶспсп- ! - ) 2 &ѵ #
*/ ѷ/ $.2 &'4Ѹ/ .*-- .+*)/*!$-./4*!/#
*!0"тп– +хѶ$. ,0'/*рпп!*-''. -$ .ѵ) +уѶ*- -/ (+*--$'4
-*../# *0)/-42 )/$)/* !! /ѵ

тпп
сфп
спп
рфп
рпп
фп
п
0'ѵф

0'ѵрш

0"ѵс

0"ѵрх

0"ѵтп

+ѵрт

*0- .ѷ +0-)Ѷ *0$.Ѷ) .(*)җспспҘѸ '0'/$*).ѵ
*/ ѷ/ are2 &'4Ѹ/ .*-- .+*)/*the !$-./4*!/# - ! - ) 2 &ѵ# . + -$*Ѷ/#
2 &*!0"0./тп– +/ ( -хѶ$. ,0'/*рпп!*-''. -$ .ѵ) +/ ( -уѶ*- -
/ (+*--$'4#'/$)" 1$/$*).-*../# *0)/-42 )/$)/* !! /ѵ

Prioritizing America’s Households | 83

Spurring a Return to the Pre-COVID Economy
Provisions in the CARES Act and other legislation, along with President Trump’s
executive actions, provided immediate relief to U.S. households. However, the
most important way to ensure long-term gains in living standards is a rapid
economic recovery.
A central CARES Act component intended to promote a rapid recovery
was the Paycheck Protection Program, which helped keep workers employed
by authorizing $349 billion to support payroll and other expenses for small
businesses, self-employed individuals, Tribal business concerns, and nonprofit
or veterans’ organizations during the COVID-19 crisis. As part of the PPP and
Health Care Enhancement Act, an additional $310 billion was appropriated to
the program. Although the funds were issued as loans, they could be fully forgiven if no less than 60 percent (originally, 75 percent) of the funds were used
for payroll. Other expenses eligible for loan forgiveness included mortgage
interest, rent, and utilities. To further encourage employers to maintain ties
with workers, employers whose operations were disrupted by COVID-19 but did
not receive a PPP loan were offered an employee retention tax credit worth up
to $5,000 per retained employee.
As discussed further in chapter 3 of this Report, these policies helped
hasten the economic recovery. For example, Autor and others (2020) use
administrative payroll data to compare employment changes at firms that
were somewhat below and somewhat above the 500-employee cutoff for PPP
loan eligibility. They find that the PPP saved between 1.4 million and 3.2 million jobs through the first week of June, based on an assumption that firms
somewhat below the eligibility cutoff would have seen employment changes
similar to those experienced by firms somewhat above the eligibility cutoff.
If, however, smaller firms would have experienced larger employment losses
than larger firms in the absence of the PPP, then the true impact of the PPP
would be significantly larger than that estimated by the authors. Other studies
find a range of early effects of PPP on employment (Bartik et al. 2020; Chetty et
al. 2020; Granja et al. 2020). It is important to note that because PPP stemmed
business closures, the total employment effect is likely to be considerably
larger over time as those salvaged businesses rehire furloughed workers. In
total, Standard & Poor’s U.S. Chief Economist Beth Ann Bovino estimates that
PPP could have saved upward of 13.6 million jobs, and JPMorgan Chase’s
Jamie Dimon estimates that PPP saved 35 million jobs (Fox et al. 2020; Ruhle,
Miranda, and Capetta 2020).
After the unemployment rate rose from 3.5 percent in February 2020 to
14.7 percent in April 2020, forecasters expected that it would continue increasing and remain above 10 percent for the remainder of 2020. However, contrary
to expectations, the unemployment rate fell to 6.7 percent just seven months
later. By comparison, after the unemployment reached its peak in October
84 |

Chapter 2

$"0- сҊчѵ *--& / *1 -4*(+-$.*).4 ..$*).Ѷ
ршуф–спсп
Indexed employment level (pre-recession peak = 100)
рпс

ршуф/*сппч
1 -"

рпп
шч
шх
шу

сппч–пш

спсп

шс
шп
чч
чх
чу
п

рс

рч

су

тп

тх

ус

уч

фу

хп

хх

цс

Months from pre-recession employment peak
*0- .ѷ0- 0*! *-//$./$.Ѷ0-- )/(+'*4( )///$./$.Ѹ/$*)'0- 0*!*)*($
. -#Ѹ'0'/$*).ѵ

2009 during the Great Recession, in over a decade it had still not fallen by as
much as it did between April and November 2020. Figure 2-8 shows that the
recovery from the COVID shock has been much faster than that from the Great
Recession and all other postwar recessions in regaining lost employment.
Workers in every major private sector industry have experienced employment gains. Between April and November, the leisure and hospitality industry
regained 59 percent (4.9 million) of jobs lost; trade, transportation and utilities
regained 71 percent (2.4 million); and education and health services regained
55 percent (1.5 million) (figure 2-9). Black workers have regained 53 percent,
Hispanic workers have regained 66 percent, Asian American workers have
regained 66 percent, and White workers have regained 66 percent of jobs lost.
The policy response to the most sudden and severe economic downturn
since the Great Depression was unprecedented. It provided extensive relief to
households that would otherwise have suffered substantial losses of income,
and it set the economic recovery on a strong footing. Of course, continued
progress is needed to return to the strong pre-COVID economy. In addition,
there is near-term risk that policy and behavioral responses to a resurgence
of COVID-19 could disrupt the considerable labor market recovery observed to
date. For this reason, in late 2020 the Administration continued to articulate
support for additional fiscal measures to provide a bridge to the widespread
availability of vaccine candidates developed under Operation Warp Speed
(Goodspeed and Navarro 2020).
Prioritizing America’s Households

| 85

$"0- сҊшѵ-$1/ 4-*'' * *.. .4 /*-s$) -0-4спсп
*..!-*( -0-4/*+-$'

$)!-*(+-$'/**1 ( -

Employment change (millions)
ш

$.0- )#*.+$/'$/4

ч
ц

х
ф
у
т

- Ѷ
/-).+*-//$*)Ѷ
)0/$'$/$ .
0/$*))
# '/#. -1$ .

-*! ..$*)')
0.$) ... -1$ .

**.Ҋ+-*0$)"
''*/# -. -1$ .

с
р
п

*0- .ѷ0- 0*! *-//$./$.Ѷ0-- )/(+'*4( )///$./$.Ѹ'0'/$*).ѵ
*/ ѷ“''*/# -. -1$ .”$./# .0(*!/# !$))$'/$1$/$ .Ѷ$)!*-(/$*)Ѷ)“*/# - . -1$ .”
/ "*-$ .ѵ

Conclusion
The COVID-19 pandemic and subsequent lockdowns caused a historic shock
to the U.S. economy. The unemployment rate increased from a 50-year low
of 3.5 percent in February 2020 to 14.7 percent just two months later. Experts
forecasted that the unemployment rate would remain above 10 percent for the
remainder of 2020. Without policy action, the loss in employment and earnings
threatened the ability of millions of American households to pay for food, housing, and other basic necessities.
Fortunately, the policy response from the Federal Government was
immediate, unprecedented in scale, and targeted to the most vulnerable
households and firms. The Trump Administration worked with Congress to
pass and sign three major bills in March 2020, providing over $2 trillion in
relief to fund direct payments to U.S. households, assistance to employers to
keep workers on payroll, expanded and enhanced unemployment assistance,
measures to prevent housing foreclosures and evictions, and a number of other
measures intended to provide relief to Americans and bolster the economy’s
recovery. In August 2020, President Trump followed up with a series of executive actions that extended further relief to American households, including
providing lost wages assistance and a moratorium on residential evictions.
These actions led to historic increases in household incomes, especially
at the lower end of the income distribution, along with decreases in poverty

86 | Chapter 2

and evictions. In the long run, however, the success of actions taken to support
households will depend on how quickly the economy recovers. Between April
and November, the unemployment rate fell by 8.0 percentage points, from 14.7
to 6.7 percent, the largest seven-month decline on record. Although more progress is needed, the economy has recovered much more quickly than forecasters initially expected. Fortunately, the strong pre-COVID economy—in which
unemployment fell to a 50-year low, labor force participation grew stronger,
the poverty rate reached a record low, and median income experienced the
largest one-year increase in 50 years—together with unprecedented Federal
action during the COVID-19 crisis, has paved the way for a strong economic
recovery.

Prioritizing America’s Households | 87

x
Chapter 3

Assisting Entrepreneurs and
Workers through Aid to Businesses
The Coronavirus Aid, Relief, and Economic Security (CARES) Act, which was
signed into law by President Trump in March 2020, has authorized unprecedented levels of financial support, providing much-needed relief to families
and workers during the COVID-19 pandemic. The Federal Government has also
implemented, through the CARES Act and additional means, several critical
measures to support small businesses as they have weathered this unique and
exogenous economic shock. With businesses facing a significant loss in revenue and demand as a result of stay-at-home orders and social distancing by
households, the Trump Administration has taken decisive actions to keep small
businesses afloat, thus far averting a swath of bankruptcies and a collapse of
the financial system similar to the Great Recession.
In one of the CARES Act’s most significant provisions to help American small
businesses, the Paycheck Protection Program (PPP) authorized hundreds of
billions of dollars in forgivable loans to boost the retention of employees and
preserve the relationships between employers and their workers. This program
was successful in targeting the small businesses that needed loans the most;
5.21 million small businesses and nonprofit organizations received a loan, and
the vast majority of loans went to small businesses and organizations with very
few employees. Estimates find that the PPP has saved or supported tens of
millions of jobs.
In addition to the PPP, the CARES Act funded several measures to provide
liquidity to businesses in the midst of this economic shock. As a result, the
economy in 2020 did not see a wave of bankruptcies like during the Great

Recession. Indeed, though bankruptcies did spike during the pandemic, they
increased at a much lower rate than during the Great Recession, despite the fact
that the pandemic caused an economic shock significantly larger than that of
2008. The robust labor market before the pandemic helped enable businesses
and households to better weather this crisis, further contributing to the lower
rates of small business bankruptcies during the pandemic.
The Department of the Treasury and the Federal Reserve also played critical
roles in easing financial strain and ensuring access to liquidity for businesses.
To accomplish this, the CARES Act authorized emergency actions to stabilize
the financial system and to provide direct support for credit. In doing so, the
act has helped avert a potential collapse of the financial system similar to what
occurred during the Great Recession, thereby greatly improving the outlook for
a swift economic recovery.
In analyzing numerous financial and economic indicators related to the
pandemic, this chapter demonstrates that the CARES Act and the Trump
Administration’s actions have preserved American businesses and saved millions of jobs. These actions have been historic in their speed and magnitude,
and the data show that small businesses have benefited the most from these
unprecedented actions. As a result, the economy has recovered more rapidly
than many people anticipated and is poised to continue its return to the level
of prosperity it experienced before the COVID-19 pandemic provided that fiscal
policy continues to provide appropriate support.

his chapter focuses on the effects that the extraordinary actions taken
by the Federal Government in response to the COVID-19 pandemic
have had on businesses, with a particular focus on small businesses.
Just as forecasters have significantly revised their expectations for 2020 gross
domestic product (GDP) upward, so too did businesses experience an uptick in
their outlook. Between April and October, the small business optimism index
compiled by the National Federation of Independent Business rose from 90.9 to
104.0—one of the steepest annual increases in the index’s history. The federation attributes some of the improvement to the loan forgiveness feature of the

90 |

Chapter 3

Paycheck Protection Program (PPP), which is discussed here, along with other
business-oriented provisions of the CARES Act.

Summary of Policies to Assist American
Businesses and Their Workers
Much of the analysis in this chapter focuses on the PPP and the Federal
Reserve’s lending facilities—given that these are the largest business provisions and have received the most attention by academic researchers—but here
we also include a summary of several of the main provisions in the CARES Act
directed at small, mid-sized, and large businesses. In total, $1.6 trillion was
allocated (excluding Federal Reserve lending facilities), and about $930 billion
has been disbursed as of October 2020.

The Paycheck Protection Program
The CARES Act authorized $349 billion in forgivable PPP loans to support
payroll and other expenses for America’s small businesses, self-employed
individuals, Tribal businesses, and nonprofit/veterans’ organizations. As part
of the Paycheck Protection Program and Health Care Enhancement Act, which
was signed into law by President Trump in April 2020, an additional $310 billion
was authorized, bringing the total amount for the PPP to $659 billion. Though
the funds are used to guarantee and forgive loans, a condition for full loan
forgiveness is that recipient businesses must spend no less than 60 percent of
the loaned funds on payroll expenses within a 24-week span called the covered
period.1 Other expenses eligible for loan forgiveness include interest on mortgages, rent, and utilities.

Employee Retention Tax Credits
The CARES Act provided refundable tax credits against payroll taxes for
employers that either were required to shut down because of COVID-19-related
government mandates or suffered at least a 50 percent decline in year-overyear revenue during a quarter. The credit amount is 50 percent of qualified
wages, up to a maximum of $5,000. Like the PPP, these credits act as a wage
subsidy to boost retention, and the Joint Committee on Taxation estimated a
$55 billion cost of the provision in March, although it is unclear how high the
uptake will end up being, in light of eligibility restrictions. In addition, qualified
wages of firms with more than 100 workers include only wages paid to inactive
employees (e.g., those furloughed).

1 Before the Paycheck Protection Flexibility Act of 2020, which was signed into law by President
Trump in June 2020, the forgiveness criteria required a 75 percent payroll requirement and an
eight-week covered period.

Assisting Entrepreneurs and Workers through Aid to Businesses | 91

Economic Injury Disaster Loans and Advances
The Economic Injury Disaster Loans and Advances (EIDL) program of the U.S.
Small Business Administration (SBA), which predates the COVID-19 crisis,
provides relief to small businesses and nonprofit organizations experiencing
a temporary loss of revenue. Relative to PPP loans, EIDL loans tend to be
longer in duration, have a higher interest rate, and are not forgivable outside
the advance itself—that is, the loan must be repaid in full. The Coronavirus
Preparedness and Response Supplemental Appropriations Act, which preceded the CARES Act, made COVID-19 losses an eligible disaster under the
SBA disaster program, allowing affected businesses to apply for the program’s
loans. The CARES Act, along with the Paycheck Protection Program and
Healthcare Enhancement Act, then expanded eligibility and eased application
requirements. The acts also added $20 billion in funds to allow more entities
to receive the $10,000 EIDL advances. As of November 23, $194 billion in EIDL
loans had been approved for just over 3.6 million loans.

The Federal Reserve’s Lending Facilities
The U.S. Department of the Treasury made available $454 billion via the CARES
Act to backstop some of the emergency lending facilities set up by the Federal
Reserve under Section 13(3) of the Federal Reserve Act. The purpose of this
Treasury backing was to ensure that the Federal Reserve would not be put in
a position to need to absorb losses. For the facilities created by this collaboration, Federal Reserve Banks lend to private firms, to nonprofit organizations, or
to State and local governments.
The facilities can broadly be divided into two groups: those that are
aimed at supplying credit to the macroeconomy (which rely on CARES
Act capital funding)—the Primary and Secondary Market Corporate Credit
Facilities, the Term Asset-Backed Securities Loan Facility, the Municipal
Liquidity Facility, and the Main Street Lending Program—and those that are
aimed at funding markets (which are backed by funding from the Treasury’s
Exchange Stabilization Fund (ESF) or are secured by collateral)—the Money
Market Mutual Fund Liquidity Facility and the Commercial Paper Funding
Facility. The Federal Reserve also created the Primary Dealer Credit Facility and
the Paycheck Protection Program Liquidity Facility, neither of which received
economic support or investments from the Treasury. This chapter goes into
greater depth on these facilities in a later section.

Other Programs
Small Business Administration debt relief. The CARES Act appropriated $17 billion to go toward debt relief for new and existing SBA borrowers. Specifically,
the SBA is required under this provision to pay the principal, interest, and fees
owed on specified loans for six months. This debt relief is distinct from the
conditional forgiveness offered for newly created PPP loans.
92 | Chapter 3

Deferral of employer payroll taxes. The CARES Act allowed employers to
defer payment of their portion of payroll taxes incurred from March 27, 2020,
through December 31, 2020. Businesses will repay their deferred liabilities in
two installments, in December 2021 and 2022. This deferral is a de facto loan
to businesses, giving them short-term liquidity without directly altering their
long-term financial situation, except to the extent that the injection of liquidity
is necessary for some of them to survive the crisis.
Modifications for net operating losses. The CARES Act permitted taxpayers
to offset 100 percent of taxable income in taxable years beginning after 2017
and before 2021 with net operating losses (NOLs). Before the CARES Act, taxpayers’ NOLs from taxable years beginning after 2017 were limited to offsetting
80 percent of taxable income in such years. The CARES Act also temporarily
allowed taxpayers to carry back recently computed NOLs to offset income (and
potentially claim refunds of all or part of their tax liabilities) for the previous
five taxable years, if the losses were incurred in taxable years beginning after
2017 and before 2021.2 The Joint Committee on Taxation estimated that this
provision would reduce revenues by $154 billion in 2020 and by $161 billion
during the 2020–30 window.
Direct sector-specific aid. The CARES Act authorized $46 billion in loans,
loan guarantees, and other investments for certain affected industries.
Specifically, the CARES Act authorized $25 billion for passenger air carriers, $4
billion for cargo air carriers, and $17 billion for businesses deemed critical to
maintaining national security. In exchange for receiving these funds, passenger
air carriers agreed to certain conditions, including a requirement not to reduce
employment by more than 10 percent through September 2020.

Measuring Small Business’s Utilization of Selected CARES Act
Business Provisions
Table 3-1 illustrates the percentage of small businesses that have accessed
different programs since March 13, 2020, as reported in the Census Small
Business Pulse Survey. Small businesses may also have accessed Economic
Impact Payments and Unemployment Insurance or Pandemic Unemployment
Assistance.
As of August 8, at the time of the PPP’s closing, the SBA had approved
over 5.2 million small business loans worth more than $525 billion, supporting
an estimated 51 million jobs and representing 80 percent of the small business
payroll in all 50 States. The average loan size was about $101,000. The loans
were overwhelmingly distributed to small businesses with few employees.
More than 87 percent of the total approved loans, totaling over one-quarter of
the total approved loan amount, were for $150,000 or less (table 3-2). Over 94
2 In addition to the NOL rule changes, the CARES Act included a separate rule change that
deferred limits on owners’ losses from pass-through businesses, which could have an impact on
individuals’ computation of their NOLs.

Assisting Entrepreneurs and Workers through Aid to Businesses | 93

' тҊрѵ $,0$$/4-*"-(.!*-(''0.$) .. .
0-$)" Ҋрш
- )/age*!(''
0.$) .. .
$1$)"..$./)

-*"-(
4# &-*/ /$*)-*"-(җҘ

цтѵс

*)*($ )%0-4$../ - *).җ Ҙ

стѵч

(''0.$) ..($)$./-/$*)җҘ /- '$ !

рпѵс

/# - -'+-*"-(.

сѵу

#$.0.$) ..#.)*/- $1 !$))$'
..$./) !-*()4 -'+-*"-(

ссѵп

*0- .ѷ ).0.0- 0Ѹ'0'/$*).ѵ
*/ ѷ# .0(*!/# . *)*/ ,0'рппbecause0.$) .. .*0'. ' /
(0'/$+' ).2 -.ѵ0-1 4- .0'/.- !*-/# + -$**1 ( -ст–сшѶспспѵ#
ѵѵ ).0.defines a small business as “a single location business with
employment between р)ушш)- $pts of at least $1,000.” The sample
consists of about 885,000 0.$) .. .ѵ

' тҊсѵ *)$5 4(*0)/Ѷ$-./) *)
*0)*($) җ/.*!0"ѵчѶспспҘ
*)$5
җ/#*0.).*!
*''-.Ҙ

++-*1
*).
җ*0)/Ҙ

фп)0) -

++-*1 *)
*/'
- )/" - )/"
(*0)/
*!*0)/
*!(*0)/
җ$''$*)*''-.Ҙ

тѶфцуѶррп

хсѵц

хчѵх

рсѵп

фп–рпп

хчтѶцчф

учѵц

ртѵр

шѵт

рпп–рфп

сшуѶффц

тхѵр

фѵц

хѵш

рфп–тфп

тццѶцшц

чуѵч

цѵс

рхѵр

тфп–рѶппп

ршшѶхцш

рртѵх

тѵч

срѵх

рѶппп–сѶппп

фтѶсрч

цтѵш

рѵп

руѵр

сѶппп–фѶппп

суѶсуч

цсѵс

пѵф

ртѵц

*- /#)фѶппп

уѶцту

ттѵр

пѵр

хѵт

фѶсрсѶрсч

фсфѵп

*/'

*0- ѷѵѵ(''0.$) ..($)$./-/$*)ѵ
*/ ѷۙ4# &-*/ /$*)-*"-(ѵ

percent of the total approved loans, totaling more than 44 percent of the total
approved loan amount, were for $350,000 or less. Because the maximum loan
for which a business can apply is a function of its total payroll costs, the vast
majority of PPP loans were approved for small businesses and organizations
with very few employees.
The first round of the PPP, which ended April 16 when funds ran out,
approved fewer loans but consisted of a larger share of the total loan amounts
(figure 3-1). Round 2 had a change in the composition of which firms received
the loans, with a shift toward smaller businesses. Conditional on participation,

94 |

Chapter 3

$"0- тҊрѵ#- *!*/' *)*0)/.) *)(*0)/.4
*0)
Percent

*0)р

*0)с

чп
хчѵш

цп

хпѵш

хп
фп
уп

тшѵр
трѵр

тп
сп
рп
п
*)*0)/

*)(*0)/

*0- .ѷU.S. Small Business Administration ()Ѹ'0'/$*).ѵ
*/ ѷ# '*)*0)/.)(*0)/.!*--*0)1- -$1 0.$)"the ’.0"0./ч
0+/ , whichshares data specific to round 2.

the average loan size fell from $197,462 in round 1 to $57,095 in round 2. Overall,
73.2 percent of small businesses received PPP loans, as indicated in table 3-1.
Moreover, at the time the program closed, over $130 billion in authorized
PPP funds remained unspent, which provides suggestive evidence that most
small businesses that were eligible and applied for a PPP loan received one.
The presence of leftover funds does not indicate that PPP demand is satiated,
however. Rather, the leftover funds are more likely in part a consequence of the
restriction against businesses receiving more than one PPP loan—for example,
against businesses receiving a second loan to carry themselves through the
fall after receiving an initial loan to get through the spring and early summer.
Further legislation would need to provide such authorization.
Recent research by Autor and others (2020) provides evidence that,
through just the first week of June 2020, the PPP saved between 1.4 and 3.2
million jobs. The rehiring of furloughed workers in the months since, in order
to qualify for PPP loan forgiveness, is likely to result in a much higher total
number of jobs saved attributable to the PPP. In total, Standard & Poor’s (S&P)
U.S. Chief Economist Beth Ann Bovino estimates that PPP could have saved
upward of 13.6 million jobs, while JPMorgan Chase CEO Jamie Dimon suggests
an even larger figure of 35 million jobs (Fox et al. 2020; Ruhle, Miranda, and
Cappetta 2020). As of August 8, healthcare and social assistance; professional,
scientific, and technical services; construction; and manufacturing accounted

Assisting Entrepreneurs and Workers through Aid to Businesses | 95

for 48 percent of the total amount of approved dollars in both rounds of PPP
(table 3-3).
Figure 3-2 shows PPP coverage for each State as a percentage of total
payroll expenses incurred over a 2.5-month period by businesses with fewer
than 500 employees (the duration used to determine PPP loan size and the
firm size cutoff for eligibility).3 Those States with most if not all their small

3 An early analysis of the PPP program found that some funds initially flowed to geographic
regions that were less adversely impacted by the pandemic (Granja et al. 2020).

96 | Chapter 3
фуѵр
усѵф
упѵх
трѵц
сцѵц

стчѶушу
тчтѶфхр
уцсѶурч
фчтѶтчф
рцуѶцпц

*((*/$*))!**. -1$ .

/$'/-

/# -. -1$ .җ 3 +/+0'$($)$./-/$*)Ҙ

#*' .' /-

схѵх
рцѵф
рфѵц
рсѵс
рсѵр
шѵц
шѵт
чѵс
чѵр
уѵф
рѵч
рѵх
рѵф
фсфѵп

сфчѶшпц

рчрѶушт
ччѶпсс
сршѶфпс
цтѶчсу
ртпѶцхп
рушѶфтф
ссѶфпт
руѶсшр
шѶуцс
чѶттр
фѶсрсѶрсч

' .// )- )/')' .$)"

$)) )$).0-)

0/$*)'. -1$ .

)'..$!$ ./'$.#( )/.

-/.Ѷ )/ -/$)( )/Ѷ)- - /$*)

"-$0'/0- Ѷ!*- ./-4Ѷ!$.#$)",)#0)/$)"

$)$)"

0'$($)$./-/$*)

)" ( )/*!*(+)$ .) )/ -+-$. .

/$'$/$ .
*/'

*0- .ѷѵѵ(''0.$) ..($)$./-/$*)Ѹ'0'/$*).ѵ
*/ ѷNAICS = North American Industry Classification System; ۙ4# &-*/ /$*)-*"-(ѵ

)!*-(/$*)

-).+*-//$*))2- #*0.$)"

)- ( $/$*). -1$ .
ссшѶфхф
схсѶшср

($)$./-/$1 ).0++*-/)2./ ()" ( )/

*)./-0/$*)

)0!/0-$)"

хцѵч
ххѵч
хфѵр

фтсѶццф
хчрѶррр
ушхѶффр

++-*1 *''-.
җ$''$*).Ҙ

++-*1 *).

-*! ..$*)'Ѷ.$ )/$!$Ѷ)/ #)$'. -1$ .

'/#- ).*$'..$./)

0. /*- .-$+/$*)

' тҊтѵ *).4 )0./-4

<BILL: PLEASE TURN THIS TABLE, AND PHOTO-REDUCE TO FIT>

пѵс

пѵу
пѵт

сѵш

сѵф

уѵс
рѵу

рѵц

тѵф

уѵу
фѵп

фѵп

тѵу

шѵр
ррѵс

цѵу

уѵх

ртѵр
шѵф

рпѵс

- )/"
*! *).

пѵт

пѵш
пѵт

рѵх

рѵч
рѵч

сѵт

тѵт
тѵп

фѵр

фѵт

цѵц
хѵп

чѵр

рпѵт

рсѵц
рсѵу

рсѵш

- )/"
*!(*0)/

$"0- тҊсѵ - )/age*!(''0.$) .. .
$1$)" *)Ѷ/#-*0"#0"0./ч, 2020

ццѵх‫ڿ‬/*чфѵр‫ڿ‬
чфѵр‫ڿ‬/*чшѵц‫ڿ‬
чшѵц‫ڿ‬/*шсѵш‫ڿ‬
шсѵш‫ڿ‬/*рпсѵш‫ڿ‬
*0- .ѷU.S. Small Business AdministrationѸU.S. ).0.Ѹ'0'/$*).ѵ
*/ ѷ# (*0)/*!+4-*''*( .!-*( ).0.’s спрц.0-1 4%0./ /*спрш*''-.and /# )
- 0 /*сѵф(*)/#.’2*-/#ѵ(''0.$) .. .- !$) . (+'*4( )/*!рto ушш peopleѵ*'
+-*+-$ / -.#$+.'.*- $1 '*).Ѷ.*.*( .// .- $1 (*- /#)рпп percent*!/#$.+4-*''ѵ

businesses payroll covered by PPP are predominantly in the Southeast and
Midwest. As data on retail consumer spending reveal, PPP loans went especially to States that saw more drastic declines in consumer spending. The
Census Small Business Pulse Survey reports that 73.2 percent of small businesses nationwide have received a PPP loan since the spring.

Comparing Expected and Actual Small Business
Bankruptcies during the COVID-19 Pandemic
There are signs that the disruptive effects of the COVID-19 pandemic have led
to greater business churn, with the Census Bureau reporting that business
applications in 2020 are twice as high as they have been at any point in the
past decade. However, measuring business exits is far more challenging, both
because of an absence of live-tracking data, and because of the unique nature
of the economic shock from COVID-19.4 In particular, it is difficult to distinguish between businesses that have exited permanently from those which
have closed temporarily. Bankruptcies can serve as a proxy for the subset of
business exits caused by the inability of owners to meet debt obligations. This
4 Crane et al. (2020) discuss some of these measurement challenges as well as a range of
alternative measures and indicators of business exit.

Assisting Entrepreneurs and Workers through Aid to Businesses | 97

$"0- тҊтѵ*/'#+/ -рр)&-0+/4$'$)". by U.S. Small BusinessesѶ
2020
Percent change in 2020 from average of 2017 to 2019
10пѵп

чсѵт

чпѵп

хшѵт

хпѵп

учѵр

упѵп
спѵп

хт.0

ушѵр
тцѵт
руѵф

хѵп

рчѵс

пѵп
–спѵп
–стѵу

–упѵп

*0- .ѷU.S. +-/( )/*! 0./$ Ѹ'0'/$*).ѵ
*/ ѷ0 /*2 &'4- +*-/$)", /# 0'4+ -$*$. 0) сч–0"0./с)/# 0"0./+ -$*$.
August 3–30.

section compares observed bankruptcies with what an empirical analysis of
historical economic relationships predicts are the volume of bankruptcies
we would expect to observe in the United States, given the magnitude of the
COVID-19 economic shock.
Data from the Department of Justice make it possible to monitor weekly
changes in small business bankruptcies (specifically, those filed under Chapter
11) and also to compare the monthly totals for 2020 with the same month in
prior years to see if small business bankruptcies have spiked as a result of the
crisis.5 Figure 3-3 shows a spike in February and March, before COVID-19 had
begun to wreak havoc, likely due to a new Subchapter 5 provision that came
into effect in February.6 The data then show a sharp decline in April, when most
of the country was under lockdowns, which may have prevented business owners from physically filing for bankruptcy due to social-distancing measures, or
courts being unable to accept filings for the same reason (Tett 2020). Businesses
may also have been waiting to see how economic uncertainty would unfold
before filing for bankruptcy (Keshner 2020). In addition, the strong economy
before the pandemic likely put businesses in a better position to survive for
5 When referred to as “small business,” the data reflect businesses that classify themselves as small
when they are filing for a Chapter 11 bankruptcy.
6 Subchapter 5 of Chapter 11 makes it easier for smaller businesses to reorganize under Chapter
11 bankruptcy. Under the CARES Act, the threshold debt level for businesses that could apply for
Subchapter 5 bankruptcy was raised further, allowing more small businesses to be eligible for this
chapter.

98 |

Chapter 3

$"0- тҊуѵ - )/#)" $)(''0.$) ..#+/ -рр
)&-0+/$ .Ѷ спспѶ!-*(.спрц–рш1 -"

–учѵч‫ڿ‬/*п‫ڿ‬
п‫ڿ‬/*сп‫ڿ‬
сп‫ڿ‬/*рпшѵр‫ڿ‬
рпшѵр‫ڿ‬/*с,тпп‫ڿ‬
*0- .ѷѵѵ +-/( )/*! 0./$ Ѹ'0'/$*).ѵ
*/ ѷ'()*-/#-*'$)#1 )*/1$'' ./# 4- )*/0) -/# %0-$.$/$*)*!
/# )$/ // .-0./ -*"-(ѵ# ' " ).#*2. #,0-/$' ѵۙ!$.'4 -ѵ

some time, even in the face of COVID-related disruptions. Regardless of the
reasons for the April decline, bankruptcy filings started to rise in the late spring
and early summer, before showing a larger rise in July, August, September, and
October of, respectively, 69.3, 49.1, 82.3, and 63 percent. However, this level
was still 34 percent below that during the peak of the 2009 financial crisis.
Figure 3-4 shows which States saw an uptick in small business Chapter
11 bankruptcies in fiscal year (FY) 2020 through September 30, compared with
their 2017–19 averages. There are 36 States with higher FY 2020 small business
Chapter 11 bankruptcies than their averages between 2017 and 2019. These
States account for 72.9 percent of total small business Chapter 11 bankruptcies
in FY 2020.
Figure 3-5 uses the Department of Justice’s data on small business
bankruptcy filings to compare bankruptcy dynamics with those during the
Great Recession and its aftermath. There was a large increase in small business
Chapter 11 bankruptcies during 2009 and 2010, and small business Chapter 11
bankruptcies fell as the recovery continued. The data for 2020 show a much
smaller initial uptick in bankruptcies, which is striking, given that secondquarter gross domestic product fell by an annualized 31.4 percent during the
spring, compared with only 8.4 percent during the worst quarter of the Great
Recession (2008:Q4).

Assisting Entrepreneurs and Workers through Aid to Businesses | 99

$"0- тҊфѵ(''0.$) ..#+/ -рр)&-0+/$ .Ѷ.сппч–сп
Bankruptcies per 100,000 small businesses
фп

спсп

уп

тп

сп

рп
сппч

спрп

спрс

спру

спрх

спрч

спсп

*0- .ѷ +-/( )/*! 0./$ Ѹ'0'/$*).ѵ
Note: FY = fiscal year.

Assessing how these observed bankruptcy filings compare with what one
would expect given the disruption from COVID-19 requires a deeper analysis.
The foremost explanation of why the increase in bankruptcies was not sharper
is that the swift passage and implementation of record fiscal relief through the
CARES Act helped businesses absorb the shock to cash flows that in the past
would have forced them to declare bankruptcy; as we discuss below, in the
absence of such a vigorous policy response, multitudes of companies would
have been forced to permanently close.
One way to forecast small business Chapter 11 bankruptcies is through
a vector autoregression estimate of unemployment insurance claims with
three-month lags from January 2006 to December 2019. An advantage of this
approach is that it can determine the lag between the negative economic
shock and its effect on bankruptcies. In figure 3-6, the gap between actual and
predicted bankruptcies represents “averted bankruptcies.” Small business
bankruptcies from April to September as a whole were predicted to increase
by 181.8 percent, while actual filings rose by a much smaller 28.4 percent. This
analysis suggests that the historic economic policy response in the spring and
summer mitigated the macroeconomic shock and concomitant financial distress. However, this success does not preclude the possibility of a large future

100 |

Chapter 3

$"0- тҊхѵ/0'1 -.0.- $/ (''0.$) ..#+/ - рр
)&-0+/4$'$)".Ѷсппц–сп
Year-over-year change (percent)

тфп

/. спсп

тпп
сфп

- $/

спп
рфп
рпп

/0'

фп
п
–фп
). сппц

). спрп

). спрт

). спрх

). спрш

*0- .ѷU.S. Department of Justice; U.S. +-/( )/*! *-Ѹ'0'/$*).ѵ
*/ ѷ- $/ !$'$)".- . *)/# 1 /*-0/*- "- ..$*)- .0'/.*!/#- Ҋ(*)/#
'"0.$)"$)$/$' U) (+'*4( )/I).0-) '$(.ѵ

bankruptcy spike, especially if the U.S. Congress fails to pass additional relief
and recovery legislation in the coming months.

The CARES Act’s Role in Facilitating
Small Business Survival
The gap between predicted and actual bankruptcies through September (see
figure 3-6) could arise from a number of factors. First, earlier in the pandemic,
social-distancing mechanisms may have affected bankruptcy filing rates, both
for the court systems and debtors. If business owners are unable to connect
with lawyers or face difficulties submitting electronic filings, this could lead
to filing delays that could show up as higher filings later in the data. At the
same time, the courts’ ability to take on cases was likely affected by State
restrictions. A second important factor is the PPP’s role in enabling businesses
to stay afloat. By giving businesses loans that can be forgiven, the PPP allows
them to meet expenses while facing a shock to cash flows. With this liquidity,
many businesses that would otherwise have filed for bankruptcy are able to
sustain themselves. The eligibility criterion disallowing firms in the process of
filing for bankruptcy from accessing PPP loans also acts as a strong bankruptcy

Assisting Entrepreneurs and Workers through Aid to Businesses | 101

disincentive with considerable quantitative significance in light of the fact that
nearly three-quarters of small businesses report having accessed PPP loans.
Finally, other elements of the CARES Act might have helped businesses
avoid bankruptcy. For instance, Pandemic Unemployment Assistance extends
unemployment insurance to the self-employed, sole proprietors, and others
who may not qualify for traditional unemployment benefits, providing liquidity
to help small businesses meet their monthly expenses. Their employees would
be able to claim expanded unemployment insurance as well if they are placed
on temporary furlough. The loan forbearance provision additionally enables
businesses to defer certain expenses, such as rental and mortgage expenses. In
other words, the PPP and other elements of the CARES Act have likely played a
significant role in helping businesses avoid bankruptcy.
Recent academic estimates also highlight the promise and success of
the PPP in keeping small businesses afloat. For example, Elenev, Landvoigt,
and Van Nieuwerburgh (2020) simulate the effect of a PPP-type program in a
general equilibrium macroeconomic model and show that it, along with a Main
Street Lending–type program, successfully stems corporate bankruptcies.
Using an instrumental variables approach, Bartik and others (2020) find that
PPP loans led to a 14- to 30-percentage-point increase in a business’ expected
survival. Clearly, to the extent that the PPP helped mitigate business closures,
it saved many of these businesses’ jobs.
Although important, bankruptcies are only one measure of small business health during the crisis. The remainder of this section discusses data from
Homebase that gives insight into a range of relevant small business metrics.
Homebase is a company that provides software to help small business owners
manage employee timesheets. Since the pandemic started, Homebase has
maintained a database of U.S. small business employment using data from
more than 60,000 businesses that use its software. The data cover more than 1
million employees who were active in the United States in January 2020. Most
Homebase customers are restaurant, food and beverage, retail, and service
businesses that are individually owned or managed by their operators.
The Homebase data show the dramatic effect of the COVID-19 pandemic
on small businesses. Figure 3-7 illustrates the daily change in the number of
hourly employees working at small businesses using Homebase compared
with a January baseline. After shelter-in-place orders became widespread in
mid-March, the number of employees working fell to a level about 55 to 60
percent lower than normal conditions. However, the passage of the CARES Act
marked a significant inflection point that reversed this decline. The data reveal
that small businesses have regained significant ground since then, though the
number of hourly employees working at small businesses has plateaued, at
about 20 percent below normal conditions, suggesting that a full small business recovery is still very much a work in progress.

102 |

Chapter 3

$"0- тҊцѵ#)" $)/# 0( -*!(''0.$) .. *0-'4
(+'*4

.*-&$)"Ѷспсп

Percent change (relative to January base)
рп

Oct. тп

CARES Act signed

–рп
–сп
–тп
–уп
–фп

*'$4.

–хп

–цп
Mar. с

Apr. рт

May сф

Jul. х

Aug. рц

Sep. сч

*0- ѷ *( . ѵ
*/ ѷ''/# -/ .*(+- /#/41er.us/# ( $)!*-/# 4*!/# 2 &!*-/# + -$*
January 4–31, 2020.

Company-specific data on PPP loan receipts and employment provide an
insight into the role the program played in keeping workers attached to their
employers, before which the extent was unclear.7 Recent research by Autor
and others (2020) using administrative payroll data finds that the PPP saved
between 1.4 and 3.2 million jobs through just the first week of June. However,
because the PPP has also stemmed business closures, the total employment
effect is likely to be considerably larger over time as these businesses rehire
furloughed workers. As stated above, the U.S. Chief Economist for S&P Global
Ratings Services, Beth Ann Bovino, estimates that the PPP could have saved
upward of 13.6 million jobs.
Opportunity Insights, a nonprofit research organization based at Harvard
University, has also developed a data set to track the impact of COVID-19 on
small businesses since January 2020. It pulls data from different sources of
“credit card processors, payroll firms, and financial services firms” to construct a time series to track the effect of COVID-19 (Chetty et al. 2020). Figure
3-8 shows that by mid-April, the number of open small businesses had fallen
over 40 percent compared with January. This trend began to increase in midApril as initial PPP loans were disbursed and as States began to gradually lift
restrictions on mobility and economic activity. As with the hourly employees
7 Analysis by Chetty et al. (2020) shows a limited impact of PPP on employment levels at small
businesses. However, their analysis is also constrained by the lack of firm-level data on PPP loan
receipts and employment.

Assisting Entrepreneurs and Workers through Aid to Businesses | 103

$"0- тҊчѵ#)" $)/# 0( -*!(''0.$) .. .+ )Ѷ
1 )Ҋ41 -" Ѷспсп
Percent change (relative to January base)

рп

/ѵтп

/s$")
п
–рп
–сп
–тп
–уп
–фп
ѵр

-ѵсс

4рр

0)ѵтп

0"ѵрш

/ѵч

*0- ѷ++*-/0)$/4 ).$"#/.ѵ
*/ ѷ''/# -/ .*(+- /#/41 -.0./# ( $)!*-/# 4*!/# 2 &!*-/#
+ -$* )0-4у–трѶспспѵ

$"0- тҊшѵ#)" $)(''0.$) .. 1 )0 Ѷ 1 )Ҋ4
1 -" Ѷспсп
Percent change (relative to January base)
сп

/ѵтп

/s$")

рп
п
–рп
–сп

–тп
–уп
–фп
–хп

ѵр

-ѵсс

4рр

0)ѵтп

0"ѵрш

/ѵч

*0- ѷ ++*-/0)$/4 ).$"#/.ѵ
*/ ѷ''/# -/ .*(+- /#/41 -.0./# ( $)!*-/# 4*!/# 2 &!*-
/# + -$* )0-4у–трѶспспѵ

104 |

Chapter 3

data discussed above, the number of open small businesses remains about 25
percent down from its pre-COVID levels. However, it is not currently possible to
tell how many of these businesses have closed permanently versus how many
remain temporarily inactive as a result either of residual or reimposed restrictions from State and local governments, or from continued lower-than-usual
levels of demand.
Opportunity Insights also tracks how much total small business revenue
has fallen (figure 3-9). At its trough in late March, small business revenue had
fallen by nearly 50 percent. Between the end of April and early June, revenues
recovered substantially, to about 25 percent below pre-COVID levels, but there
has been no further progress since then for these data.

The Coronavirus Food Assistance
Program’s Impact on Farm Incomes
The CARES Act contained provisions that authorized support to farmers
who were harmed by the consequences of the COVID-19 pandemic. These
provisions took the form of the U.S. Department of Agriculture’s Coronavirus
Food Assistance Program (CFAP). The COVID-19 pandemic and the associated
economic response disrupted food and agricultural markets, resulting in a

$"0- тҊрпѵѵѵ /-( )*( Ѷспрх–сп
Billions of dollars

2020 forecast, pre-COVID
2020 forecast, with COVID impact
2020 forecast, with COVID impact and CFAP
Actual net farm income

рпп

шп

чп

цп

хп
спрх

спрц

спрч

спрш

спсп

*0- .ѷѵѵ +-/( )/*!"-$0'/0- Ѹ**)"-$0'/0-'*'$4 . -# )./$/0/ Ѹ
'0'/$*).ѵ
Note: “2020F” denotes a forecast; CFAP = Coronavirus Food Assistance Program.

Assisting Entrepreneurs and Workers through Aid to Businesses | 105

dramatic drop in farm income for a wide array of agricultural products. CFAP
made available $16 billion in financial assistance for producers of affected
commodities, including $9.5 billion to compensate for losses due to commodity price reductions between mid-January and mid-April 2020 and another
$6.5 billion for ongoing market disruptions. In early February 2020, before the
extent of the pandemic’s impact on agricultural markets was fully apparent,
U.S. net farm income for 2020 was forecast to be $99 billion, which would have
been a 4 percent increase over 2019 and the highest net farm income since
2013. By June, as the magnitude of the pandemic became evident, analysts had
revised the forecast for 2020 net farm income down by more than $24 billion
(25 percent), when CFAP payments are excluded. Including the $16 billion in
emergency farm payments raises forecasts for net farm income to $99 billion
(figure 3-10).

The Pandemic’s Impact on the Financial
Sector and Lending Facilities’ Role
As the threat of the COVID-19 pandemic increased, the financial system came
under stress in February and March 2020. Stock prices plummeted and market volatility rose. A recent analysis by Baker and others (2020) shows that
COVID-19 has had an unprecedented effect on the stock market, especially in
comparison with other infectious disease outbreaks, including that of Spanish
Influenza in 1918–20 (figure 3-11). Businesses, which already held a historically
high level of debt at the beginning of 2020, were suddenly at higher risk of
default, leading to a decline in the availability of credit, and households’ ability
to repay their debts in the face of job and income loss became uncertain. With
potential defaults looming, the risk of the real economic shock impairing assets
of lenders and thereby infecting credit markets more broadly was substantial.
Under these conditions, monetary or fiscal problems abroad—especially in
Europe, China, and emerging market economies—could have spilled over to
the United States, compounding the stress on the financial system.
The CARES Act, together with emergency powers under section 13(3)
of the Federal Reserve Act, authorized the Federal Reserve and the Treasury
Department to take actions to stabilize the system and prevent a financial crisis
like that of the Great Recession. In accord with the Federal Reserve’s Financial
Stability Report, the Federal Reserve undertook aggressive monetary policy
interventions and also took actions to stabilize short-term funding markets
and provide direct support for credit. In addition, Chairman Jerome Powell
testified that the Federal Reserve “took measures to allow and encourage
banks to use their substantial capital and liquidity levels built up over the past
decade to support the economy during this difficult time.” Here we summarize
some of the findings from the Financial Stability Report (Federal Reserve 2020).

106 |

Chapter 3

$"0- тҊррѵ*2 *) . )0./-$'1 -" #)" during
-$*0.) ($.
Index (100 = month of first confirmed case)
рфп

+)$.#InfluenzaѶршрч

$-./*)!$-( ѵѵ
. $)ршфц

руп
ртп
рсп

Ѷсппт

ррп
рпп
шп

.$)InfluenzaѶршфц–фч

ҊршѶспсп

чп
цп
п

рп

рр

рс

рт

ру

Months since first confirmed case
*0- .ѷ*2 *) .Ѹ'0'/$*).ѵ
*/ ѷ# t ۙ п(*)/#.we- those with the !$-./*)!$-( "'*'. .,)were+-$'ршрчѶ
-0-4 ршфцѶ -0-4 спптѶ ) )0-4 спспѵ '' the pandemics 3 +/ /# .$)
Infl0enza.2 /# ir!$-./ѵѵ. 2$/#$)/# .( (*)/#.as their first global case.

Monetary policy interventions. The Federal Reserve lowered its policy
rate close to zero to make borrowing less expensive. The Federal Open Market
Committee began buying longer-term Treasury securities as well as agency
mortgage-backed securities and commercial mortgage-backed securities after
investors moved toward cash and short-term government securities because
of volatility and uncertainty, which had the effect of smoothing and improving
market conditions.
Stabilizing short-term funding markets and providing direct credit support.
In a dash for liquidity, investors stopped purchasing commercial paper and
pulled out of money market mutual funds that hold such paper along with
other short-term debt instruments, leading to a cash shortage for businesses
that rely on commercial paper to help fund their operations. In response,
and pursuant to Section 13(3) of the Federal Reserve Act and with Treasury’s
approval, the Federal Reserve established a number of emergency lending
facilities to support the flow of credit to businesses, nonprofit organizations,
States, and municipalities.
Several facilities commenced before the CARES Act. The Primary Dealer
Credit Facility allows primary dealers to support smooth market functioning
and facilitate the availability of credit to businesses and households. The
Commercial Paper Funding Facility (CPFF) provides liquidity to short-term
funding markets. The Money Market Mutual Fund Liquidity Facility (MMLF)
Assisting Entrepreneurs and Workers through Aid to Businesses | 107

makes loans available to eligible financial institutions secured by high-quality
assets purchased by the financial institution from money market mutual funds.
A number of additional facilities commenced after enactment of the
CARES Act. The Primary Market Corporate Credit Facility (PMCCF) and the
Secondary Market Corporate Credit Facility (SMCCF, together with the PMCCF,
the CCF) provide liquidity for investment grade corporate bonds (or the bonds
of certain companies that were investment grade as of March 22, 2020) as well as
for exchange traded funds (ETFs) whose objective is to provide broad exposure
to the market for U.S. corporate bonds. The Term Asset-Backed Securities Loan
Facility (TALF) supports the provision of credit to consumers and businesses
by enabling the issuance of asset-backed securities backed by private student
loans, automobile loans and leases, consumer and corporate credit card
receivables, certain loans guaranteed by the Small Business Administration,
and other assets. The Municipal Lending Facility (MLF) supports lending to
State, city and county governments, certain multistate entities, and other issuers of municipal securities. The Paycheck Protection Program Liquidity Facility
offers a source of liquidity to financial institution lenders that lend to small
businesses through the Small Business Administration’s Paycheck Protection
Program. The Main Street Lending Program (MSLP) supports lending to
small and medium-sized businesses and nonprofit organizations that were in
sound financial condition before the onset of the COVID-19 pandemic. MSLP
operates five subfacilities: the Main Street New Loan Facility, the Main Street
Priority Loan Facility, the Main Street Expanded Loan Facility, the Nonprofit
Organization New Loan Facility, and the Nonprofit Organization Expanded
Loan Facility. In support of these facilities, the Treasury Department has made
equity investments in the CPFF, CCF, TALF, MLF, and MSLP, and has provided a
backstop commitment to the MMLF.
These facilities resulted in a drop in the issuance of overnight commercial paper and redemptions from money market funds, easing market strains.
Commercial and industrial loans from the Nation’s commercial banks grew by
$726 billion during the nine weeks from March 4 through May 6, far in excess of
the growth during any similar interval since records were first collected in 1973.
Li, Strahan, and Zhang (2020) argue that banks were able to accommodate this
demand because of Federal Reserve bank liquidity programs, strong preshock
bank capital, and coincident inflows from depositors.
In addition, a variety of indicators of financial distress that spiked early
in the COVID-19 pandemic period have receded as of October 2020. Although
many other shocks have hit the economy, including news about the pandemic
itself, one can argue that public policy has mitigated the contagion of the pandemic into financial markets.
The VIX (the Chicago Board Options Exchange’s Volatility Index), an index
of expected stock market volatility derived from options prices, spiked from 27
in late February to a peak of 83 on March 16 (figure 3-12). It has generally fallen
108 |

Chapter 3

$"0- тҊрсѵ-& / *'/$'$/4 ) 3җ ҘѶсппц–сп
Index

шп

Oct. тп, спсп

чп
цп
хп
фп
уп
тп
сп
рп
п

2007

2009

2011

2013

2015

2017

2019

*0- ѷWall Street Journalѵ
Note: VIX = Chicago Board Options Exchange’s Volatility Index. Shading denotes a recession.

since then, but remains somewhat elevated (as of October 30, the VIX was 38).
Similarly, corporate bond spreads, such as the spread between BBB bonds
relative to Treasury notes, show a similar pattern, peaking on about March 23
and then receding (figure 3-13).
The trends in these indicators, and others, suggest that lending facilities
likely played an important role in easing financial market strain and ensuring
access to liquidity for businesses in 2020. Recent academic work also confirms
these findings. For example, Cox, Greenwald, and Ludvigson (2020) analyze
stock market behavior in the early weeks of COVID-19 and attribute the large
market swings to fluctuations in the pricing of risk, driven either by shifts in
risk aversion or sentiment. They also find evidence that the Federal Reserve’s
“unconventional” monetary policy announcements outlining steps to support the economy played a role in the market turnabout, leading to gains of 8
percent in the S&P 500 and 12 percent in the Russell 2000 index. Importantly,
they conclude that much of the benefit came from the signaling value of the
Federal Reserve’s statements, even as only a fraction of the credit that it has
stood ready to provide has been extended. Haddad, Moreira, and Muir (2020)
trace the recovery in the corporate bond market to the unprecedented actions
the Fed took to purchase bonds, finding that the announcement on March 23
to buy investment-grade debt boosted prices and lowered bond spreads. The
announcement on April 9 had a large effect both on investment-grade and

Assisting Entrepreneurs and Workers through Aid to Businesses | 109

$"0- тҊртѵ *-+*-/ *) ) рпҊ -- .0-4*/ +- Ѷ
сппх–сп
Percentage points
ц

Oct. тп, спсп

х
ф

у
т
с

р
п
сппц
сппш
*0- ѷ'**( -"ѵ

спрр

спрт

спрф

спрц

спрш

high-yield bonds, even for the riskier end. Gilchrist and others (2020) come to
similar conclusions, finding that the announcements regarding the Secondary
Market Corporate Credit Facility led to a reduction in credit spreads on bonds
eligible for purchase by 70 basis points.
Finally, the large-scale income and small business revenue replacement,
debt relief, and cost mitigation provisions in the CARES Act had a salutary
impact on financial market stability. Such extensive relief helped ameliorate
financial distress that could have created widespread default activity that
would have threatened the proper functioning of secondary debt markets.
Such liquidity provision helped buffer households and small businesses
against the immediate shock to cash flows from COVID-19 and also helped
preserve the health of their underlying balance sheets, thereby improving their
long-term ability to service their financial obligations.

Conclusion
The health of U.S. businesses has seen a dramatic turnaround from the
depths of the COVID-19 crisis during the spring 2020 lockdowns. The number
of employees working at small businesses, the number of open businesses,
and small business revenue have all recovered more than half the losses
experienced in April. The Federal Government’s decisive and bold action to

110 |

Chapter 3

inject more than $2 trillion into targeted income replacement and other key
programs helped bolster the ability of businesses to remain financially viable,
to retain or rehire their employees, and ultimately to restart activities. At this
stage, there is still considerable ground to regain to return to the historic level
of prosperity that existed just before the pandemic, but all indicators thus far
point to a much more rapid recovery with fewer scars than in the aftermath
of the Great Recession, despite an adverse shock several orders of magnitude
greater than that experienced in 2008–9.
Additional fiscal support to mitigate business closures, stem layoffs, and
accelerate the pace of hiring would further reduce the risk of economic scars.
For example, allowing hard-hit and at-risk small businesses to receive secondround PPP loans would help them to survive and emerge healthy from the pandemic. In fact, an October 29 survey by the National Federation of Independent
Business found that 75 percent of small businesses would apply or consider
applying for a second-round PPP loan if allowed to do so. Also, expanding
the eligibility of the Employee Retention Tax Credit (ERTC), but targeting it to
reward only net retention and hiring, could significantly accelerate hiring. With
regard to eligibility, if the main objective of the ERTC is to maximize hiring, then
restricting access to businesses suffering significant revenue losses leaves out
many employers that are willing and able to hire workers out of unemployment but may need the extra push from the ERTC incentive in light of ongoing
economic uncertainty.
This problem is especially salient given that businesses in the direst
straits may or may not be those with the highest propensity to hire, depending
on the outlook for them individually and for their industry. As for the design of
the incentive itself, subsidizing the retention of a business’s entire workforce
means paying the business to retain inframarginal workers that it likely would
have kept anyway, unless it was on the verge of shutting down. Redesigning the
ERTC to only subsidize net expansion above a company’s head count or wage
bill in some statutorily specified benchmark would specifically reward firms for
returning to or exceeding the size of their pre-COVID workforce, thus accelerating the reabsorption of unemployed workers into gainful employment. This
design especially benefits hard-hit firms that have the deepest employment
hole from which to recover while also mobilizing other businesses to step up
their recruiting. Recent research by Hamilton (2020) also finds promising labor
market benefits from ERTC expansion. Both these policy actions would support
the continuation of the fastest recovery on record as the economy regains its
full potential.

Assisting Entrepreneurs and Workers through Aid to Businesses | 111

x
Chapter 4

Advancing the Quality and
Efficiency of America’s
Healthcare System
In the face of the global COVID-19 pandemic, the Trump Administration has
taken decisive action to address the strain the health and economic crisis
placed on the healthcare sector and on working families. This response has
been twofold: financial support for hospitals and workers, and deregulation
within the healthcare sector to accelerate the availability of testing and the
development of vaccines and advanced therapeutics.
In March 2020, President Trump signed the bipartisan CARES Act, which
appropriated $100 billion for healthcare providers, and which has alleviated
the financial burden hospitals are experiencing during the COVID-19 pandemic.
This was supplemented by an additional $75 billion for the Provider Relief Fund
as part of the Paycheck Protection Program and Health Care Enhancement Act,
and also funding for testing provided by the Families First Coronavirus Relief
Act, resulting in $175 billion in direct aid to the healthcare sector. As a result,
the CEA finds that the healthcare system has been one of the most resilient
industries during the COVID-19 pandemic. The Administration also established emergency paid family and sick leave through tax credits available to
private employers with fewer than 500 employees for leave payments through
December 31, 2020. This has served to protect public health by encouraging
workers to stay home rather than working while ill, and has allowed employees
to care for sick family members without trading off work hours. In addition, the
Administration provided funds to offer COVID-19 testing and treatment at no
cost to uninsured patients, removing cost barriers for low-income and high-risk

113

individuals—and, in turn, helped the United States identify positive COVID-19
cases and mitigate the effects of the COVID-19 pandemic.
When the United States needed to ramp up its testing capabilities for the virus
at the onset of the COVID-19 pandemic, the Trump Administration, through
the Food and Drug Administration (FDA), took action to issue Emergency Use
Authorizations for COVID-19 diagnostic tests. As a result, the FDA permitted
the use of over 20 diagnostic COVID-19 tests by the end of March 2020, helping public health officials track the spread of the coronavirus throughout the
United States.
Similarly, the Centers for Medicare & Medicaid Services relaxed many of the
regulations surrounding the use of telemedicine to allow patients seeking
COVID-19 screening or advice on non-life-threatening conditions to do so from
the safety of their homes. This reduced nonessential in-person healthcare visits, decreasing the strain on overburdened healthcare facilities and diminishing
the potential transmission of COVID-19 throughout hospitals and healthcare
facilities.
In one of the largest efforts during the pandemic, the Trump Administration
mobilized the public and private sectors through Operation Warp Speed (OWS)
in order to accelerate the development, production, and distribution of a safe
and effective COVID-19 vaccine. OWS accomplishes this by identifying promising vaccines earlier in development, standardizing testing protocols, preparing
manufacturing capacity, and funding infrastructure for vaccine distribution.
Not only will the accelerated vaccine timeline provide an enormous benefit
to public health, but the CEA estimates that OWS could provide an economic
benefit of $155 billion if it pushes the arrival of the vaccine one month earlier,
or $2.4 trillion if scientists were to deliver the vaccine by January 1, 2021. As
of mid-November 2020, four vaccine candidates had entered Phase III clinical
trials. The highly promising results of interim analyses of these candidates raise
the possibility that researchers may develop a vaccine before the end of 2020
for widespread use among a set of targeted populations.

114 | Chapter 4

The deregulatory actions of the Trump Administration can continue to improve
healthcare outcomes for the American people far beyond the scope of the
COVID-19 pandemic. For example, the CEA estimates that more widespread
adoption of telemedicine would allow rural Americans to save $130 per visit in
travel-related opportunity costs while increasing their access to high-quality
healthcare nationwide. In addition, the CEA estimates that a permanent reduction in FDA approval times by one, two, or three years for new drugs would
provide trillions of dollars in social surplus. Moreover, the CEA calculates
that expanding occupational licensing deregulation for nurse practitioners
nationwide could result in $62 billion in cost savings annually. Also, this chapter
explores the effects of several healthcare policy achievements beyond the
response to the COVID-19 pandemic that will promote additional choice and
competition in the market. Permanently deregulating aspects of the healthcare
sector will provide better healthcare options and higher monetary savings for
Americans as the Nation emerges from the COVID-19 pandemic.

he United States endured a major adverse health and economic shock
in 2020 due to the arrival of the SARS-CoV-2 virus in the United States.
The impact of this pandemic is likely to persist past 2020 as widespread
mitigation takes hold. COVID-19—the disease stemming from the novel coronavirus—led to a global pandemic that, as of November 2020, has resulted in
over 50 million confirmed cases worldwide and a global death toll of at least
1.25 million people. In the United States, there have been over 10 million
confirmed cases and over 230,000 deaths. This disease has taken a toll on the
American people that has been manifested not just as a tremendous mortality
and morbidity burden, but also as a significant economic burden that affects
the Nation at every level. In the first and second quarters of 2020, the U.S.
economy contracted by 10.2 percent, and total employment declined by 14.5
percent between February and April 2020 after a record 20.8 million decrease
in employment in April. At its peak, the unemployment rate was 14.7 percent
in April. Initial claims for regular State unemployment insurance peaked in
the week ending March 28, at 6.9 million, whereas insured unemployment in
regular State programs peaked in the week ending May 9, at 24.9 million. This
unprecedented level of economic disruption resulted in the highest levels of
unemployment since the Great Depression, and had a direct impact on the
economic well-being of millions of Americans.

Advancing the Quality and Efficiency of America’s Healthcare System | 115

COVID-19’s dual effects on public health and the economy necessitated a
response on two fronts. The first one, as discussed in the previous chapters of
this Report, has consisted of efforts to address the economic effects of the crisis.
The second front, which this chapter discusses, is the Trump Administration’s
efforts to address the underlying health crisis itself.
The resolution of any healthcare crisis relies largely on the efforts of three
groups of people. First, it relies on the efforts of scientists to develop new treatments and tests for the disease. Second, it relies on the efforts of healthcare
providers and healthcare systems to treat affected patients. And third, it relies
on the efforts of the public to take appropriate actions during the crisis. These
efforts require coordinated governance at the local, State, and Federal levels.
At the Federal level, the Trump Administration moved to eliminate
regulatory barriers that could hinder the development of new treatments or
the ability of healthcare providers to care for their patients. The CEA finds
that these deregulatory efforts have had tremendous economic value. For
example, the Centers for Medicare & Medicaid Services (CMS) relaxed many of
the regulations surrounding the use of telemedicine and the share of telemedicine Medicare primary care visits increased dramatically, from 0.1 percent in
February to 43.5 percent in April.
In addition, understanding that healthcare during a pandemic requires
an economically strong healthcare system, the Administration moved to
ensure the financial security of the healthcare system. Under the CARES Act
and the Paycheck Protection Program and Health Care Enhancement Act (PPP/
HCE Act), Congress made up to $175 billion available for healthcare providers
to support their financial health and livelihood. As a result of this and other
Administration actions, the CEA finds that the healthcare system has been one
of the most resilient industries during the first three quarters of 2020 based on
employment, and indeed appears to be one of the industries that recovered
most quickly from the initial shock caused by COVID-19. A key threat to the
healthcare system early during the pandemic was sudden surges in demand for
healthcare services that overwhelmed locally available resources. To combat
this risk and slow the spread of the virus more broadly, local and State governments began implementing lockdown orders and other restrictions to combat
the spread at the cost of economic activity. As the pandemic spread throughout the country, lockdown measures expanded commensurately, with over 99
percent of the population residing in States that had closed schools and limited
bar and restaurant activity by March 24, and with over 90 percent residing in
States that had issued shelter-in-place orders by April 4 (figure 4-1).
Finally, the Trump Administration’s efforts focused on protecting
Americans from the costs of care related to COVID-19 and on providing incentives for Americans to engage in appropriate behaviors during the crisis. For
example, the Administration established emergency paid family and sick leave
for COVID-19 patients to encourage these patients to stay at home instead of
116 |

Chapter 4

$"0- 4-рѵ - )/age*!ѵѵ*+0'/$*)u) -// 2$
./-$/$*).Ѷспсп
шш‫ ڿ‬#1 .#**''*.0- .
)'$($/ -.)
Population (percent)
- ./0-)/.җтҝсуҘ
рпп
$($/-.)- ./0-)/.
шп

шп‫ڿ‬#1 .# '/ -
+-. ш
*- -.җуҝуҘ

чп
цп

# '/ -*- -.

хп
фп
уп

#**''*.0- .

тп
сп
рп
п
Mar. рт

Mar. сп

Mar. сц

Apr. тп

*0- .ѷ New York TimesѸ// +*'$4))*0) ( )/.Ѹ'0'/$*).ѵ

working while ill. This also allowed family members to take leave so they could
look after those affected by COVID-19. Similarly, though much has been written
on the Administration’s effort to increase testing capacity, from an economic
perspective, other important—and overlooked—parts of its approach were its
efforts to decrease the barriers for Americans to receive testing. In the absence
of treatment, testing may be of limited value to the individual, because a
positive test will have little impact on disease management. However, testing
does provide social value from a public health perspective, because it enables
public health approaches that can limit the spread of the disease such as
quarantining and contact tracing for infected individuals. Because individuals
do not face the full social incentives for testing, making COVID-19 testing free
at point-of-care by requiring that it be covered by insurers and reimbursing
providers for the cost of testing for the uninsured are an important way to align
the individual and social incentives for testing. The Kaiser Family Foundation
found that, in July 2020, data from 78 hospitals revealed that COVID-19 diagnostic test prices ranged from $20 to $850 per diagnostic test, with a median
cost of $127. The Administration’s subsidies probably increased the likelihood
of COVID-19 testing, especially for lower-income Americans.
The President’s response to the unique dual health and economic crises
caused by COVID-19 include an agenda for healthcare reform and deregulation.
Although regulation is intended to benefit the public, whether it actually does
so in practice is an empirical question, one that has been partly answered by

Advancing the Quality and Efficiency of America’s Healthcare System | 117

the Administration’s efforts to suspend and relax many regulations to address
COVID-19. The benefits of deregulation to bolster the pandemic response are
clear. For example, effective treatments and vaccines for COVID-19 have been
and will be introduced at an extremely fast pace, and healthcare providers face
fewer restrictions in providing care. If the absence of many regulations has
improved social welfare, a natural question is why these regulations need to
be reimposed when the pandemic subsides. Indeed, the CEA finds substantial
benefits from extending many of the existing deregulatory efforts. For example,
the CEA finds that expanding occupational licensing deregulation nationwide
could result in $62 billion in cost savings annually.
This chapter begins with an overview of the Administration’s efforts to
promote research and development for COVID treatments and vaccines, followed by a discussion of the Administration’s efforts to support the healthcare
system. Next, we discuss the Administration’s effort to protect the broader
American public by subsidizing appropriate behaviors and the cost of COVID
care. Finally, we conclude with an analysis of how healthcare can be improved
by extending COVID-19 related reforms.

Expediting Research and Development for
Novel Therapies and Tests for COVID-19
One important aspect of research and development for COVID-19 treatments
and vaccines is the issuance of Emergency Use Authorizations to facilitate
availability of pharmaceutical products in the event of an emergency. In
addition, to accelerate the availability of effective COVID-19 therapeutics and
vaccines, the Administration launched Operation Warp Speed, a public-private
partnership to support the development, production, and distribution of treatments, diagnostics, and vaccines.

Emergency Use Authorizations
Ultimately, the solution to any healthcare crisis is to find a treatment for the
underlying disease, and the Trump Administration moved aggressively to field
treatments as quickly and in as widespread a manner as possible. A key roadblock in the development of treatments is the heavily regulated drug and vaccine development processes. On average, it takes 10 years to bring a new drug
or vaccine to market, with just the preclinical phase of vaccine development
taking six months to three years (André 2002; CEA 2019; DiMasi, Grabowski, and
Hansen 2016; Grady et al. 2020; Mullard 2020; Plotkin et al. 2017; Pronker et al.
2013). These timelines are not tenable in the face of a global pandemic.
Early returns from these efforts appear promising. For example,
Remdesivir, an antiviral, received an Emergency Use Authorization (EUA) from
the Food and Drug Administration (FDA) on May 1—within 3.5 months of the
first reported case of COVID-19 in the United States. By October 22, Remdesivir
118 |

Chapter 4

had been approved by the FDA for treatment of COVID-19. Similarly, the Trump
Administration quickly solved early COVID-19 testing capacity problems.
Pre-pandemic FDA rules required that the FDA provide premarket clearance,
approval, or EUA review for COVID-19 diagnostic tests before their use in
clinical labs, which led to significant delays in adequate testing capacity at
the onset of the COVID-19 pandemic. Indeed, in February, only CDC’s COVID19 diagnostic test had been authorized by the FDA for emergency use in labs
across the nation. While it can take years for the FDA to ultimately approve new
diagnostic tests, by the end of March 2020, the FDA had issued EUAs permitting
the emergency use of over 20 diagnostic tests for COVID-19 (FDA 2020; Ivanov
2013). This rapid access to numerous COVID-19 tests was made possible by
FDA granting unprecedented flexibility to manufacturers and labs, including
allowing labs to begin developing and using their own tests before FDA review
of their validation data. And finally, as of September 2020, four vaccine candidates had entered Phase III clinical trials, raising the possibility that a vaccine
may be developed before the end of 2020 (Milken Institute 2020).
Emergency Use Authorization is an authority granted to the FDA by
the Federal Food Drug and Cosmetic Act, and it allows the FDA to permit the
production and distribution of an unapproved product or temporarily allow
an unapproved use of an approved product during a state of emergency. This
does not constitute approval of the new product or use and can be revoked by
the FDA once the emergency has ended or evidence arises that suggests that
the EUA is not in accordance with public health. EUAs have been employed in
previous pandemics, including for the development of influenza testing and
treatment as well as the test for the Novel Coronavirus 2012, more commonly
known as Middle East Respiratory Syndrome (MERS).

Operation Warp Speed
The Trump Administration also worked to expedite the development and largescale production of new vaccine treatments. Operation Warp Speed (OWS) is
a public-private partnership that encompasses most of these Administration
efforts to expedite the availability of vaccines. OWS accelerated vaccine
deployment by identifying promising vaccines earlier in development, standardizing safety and efficacy protocols, preparing manufacturing capacity, and
funding infrastructure for vaccine distribution.
Under a traditional timeline, a COVID-19 vaccine would likely not be
ready until September 2021. But under OWS, initial doses of the vaccine could
become available as early as the end of December 2020 or beginning of January
2021. If OWS accelerates initial vaccine deployment by these 8 months, the CEA
estimates that OWS would save $2.4 trillion in economic and health costs. Even
if OWS only accelerates a vaccine by one month, OWS still provides an expected
benefit of $155 billion.

Advancing the Quality and Efficiency of America’s Healthcare System | 119

Traditionally, vaccine candidates are developed individually by different
firms and are not compared with each other until after they are approved and
commercialized. However, under OWS, animal studies of candidate vaccines
were compared with each other (before additional testing in humans) to
ensure that resources were directed toward the most promising candidates.
As of August 31, the Federal government financially supported and approved
additional testing for seven vaccine candidates. Notably, OWS does not change
the number or types of trials required for vaccines, nor their safety and efficacy
tests, but it does change when they can occur.
Moreover, manufacturing and distribution infrastructure are typically
not established until a vaccine has demonstrated safety and efficacy in clinical trials, leading to additional delays in vaccine deployment. But under OWS,
the Federal government invested in manufacturing capacity for the promising
vaccine candidates while they were still being tested, rather than waiting until
they were approved. Manufacturing capacity that is developed will be used for
whatever vaccine is eventually successful, if possible given the nature of the
successful product, regardless of which firms have developed the capacity.
OWS also preemptively expands the supplies of materials that are necessary
to scale up production of any vaccine, such as glass vials. On October 16, the
President announced that the department of Health and Human Services
(HHS) and the Department of Defense will form a partnership with CVS and
Walgreens to deliver the vaccine once it is available to vulnerable Americans in
long-term-care facilities, free of charge.
The CEA estimates that OWS has the potential to bring tremendous
economic benefits, given COVID-19’s unprecedented costs. Figure 4-2 provides
an estimate of the daily cost to the United States of not having a vaccine, separated into the costs due to COVID-19 deaths (health costs) and the costs due to
lower economic activity (economic costs). As is common for many infectious
diseases, the economic costs of preventing a disease are often of comparable
magnitudes to the direct mortality and health costs induced by the disease.
Daily costs were highest in early April due to the peak of COVID-19 deaths at
that time. However, one prominent model, that of the Institute for Health
Metrics and Evaluation (IHME), projects a second wave in 2021, which suggests
the possibility of additional high future costs. Though IHME is just one among
several COVID-19 forecasting models currently used by public health authorities, it is the only one that has released 2021 projections.
Figure 4-2 demonstrates why even small delays in vaccine deployment
can be costly. Consider a vaccine that has initial doses deployed on January
1, 2021, which is shown by the gray vertical line. In this case, the value of the
vaccine is equal to the sum of the daily health costs for all days January 2, 2021,
or later, plus the sum of the daily gross domestic product costs through April
1, 2021, or later—assuming that it will take 90 days for the economy to return
to normal. However, the vaccine cannot reverse damage that has already
120 |

Chapter 4

$"0- 4-сѵ$'4 '/#)*)*($*./.*! Ҋрш/*/#
)$/ // .I!*$) I.*0)
Dollars (billions)
су

*'
v$) a1$'' d/

.трѶспср

сп
рх
рс

'/#
c*./

у
п
). спсп

*)*($
c*./
4 спсп

+. спсп

). спср

4 спср

+. спср

*0- .ѷ )./$/0/ !*- '/# /-$.)1'0/$*)Ѹ )/ -s !*-$. . *)/-*')
- 1 )/$*)Ѹ*)"- ..$*)' 0" /!!$ Ѹ'0'/$*).ѵ

occurred, so the costs to the left of the gray line cannot be recovered, even with
the introduction of a vaccine in January 2021.
Figure 4-3 demonstrates the value of faster vaccine development. We
assume that without OWS, a vaccine would be available in September 2021,
based on internal HHS projections. However, this should be viewed as a lowerbound estimate of the benefits of OWS, given that vaccines traditionally take 10
years to develop. The vertical axis gives the dollar value of an earlier vaccine,
depending on the date at which it becomes initially available (horizontal axis).
If OWS could accelerate vaccine deployment by 8 months (from September 1,
2021, to January 1, 2021), then the CEA estimates that the benefits would be
$2.4 trillion above traditional deployment (the intersection of the red line and
the left vertical gray line in figure 4-3).
The full value of the vaccine on January 1, 2021, would be $3.8 trillion.
Some estimates suggest that traditional vaccine development processes
would not result in a COVID-19 vaccine until September 2021, at which point it
would provide benefits of $1.4 trillion. The benefit of the eight-month acceleration from OWS ($2.4 trillion) is the difference between the $3.8 trillion value in
January and the $1.4 trillion value in September.
The CEA’s methodology to create figures 4-2 and 4-3 has two aspects.
First, for the value of lives lost (the health cost), the CEA used a widely cited
model developed by the IHME. The model’s most recent update reports the
actual number of COVID-19 deaths in the United States for each day between

Advancing the Quality and Efficiency of America’s Healthcare System | 121

$"0- 4-тѵ'0 *!+
+/ ( -рѶспср

$)"+ Ҋрш$) Starting
Vaccine
availability
2$/#*0/ OWS

*' for
v$) a1$'' d/

Dollars (trillions)
ф
у

$) 1'0
"$) !-*(
.+ $)"0+
+'*4( )/

р
п
0'. спсп

/. спсп
). спср
+-. спср
Date when vaccine becomes available

0'. спср

*0- .ѷ )./$/0/ !*- '/# /-$.)1'0/$*)Ѹ )/ -s !*-$. . *)/-*')
- 1 )/$*)Ѹ*)"- ..$*)' 0" /!!$ Ѹ'0'/$*).ѵ
Note: OWS = Operation Warp Speed.

February 4 and October 19, 2020, and then projects the daily number of deaths
for each day through February 1, 2021. The CEA lacks information on what will
happen after February 1, and thus assumes, for this exercise (absent a vaccine),
a 1 percent daily decline in deaths after February 1, 2021, recognizing that costs
would be greater or less if the future path of pandemic mortality were more or
less severe. The CEA then converted the number of deaths for each day to an
economic cost by using the age-adjusted value of a statistical life, which is the
standard way of evaluating economic costs of mortality (CEA 2019). The CEA
assumes that as soon as the vaccine becomes available, it will immediately
eliminate the health costs of COVID-19. However, because the vaccine will take
time to deploy, only critical populations will get access to it first, and many will
not take the vaccine at all, the CEA notes that this is a very optimistic scenario.
Second, to estimate the value of forgone gross domestic product (the
economic cost), the CEA used the Congressional Budget Office’s forecasts (CBO
2020) through 2022 to calculate the output losses between the current and preCOVID baseline (January 2020) projections. These projections only take into
account current law, meaning that the projections do not take any additional
fiscal relief packages into account. Once a vaccine is available, for the sake of
simplicity, the CEA optimistically assumes that the economy will return to preCOVID conditions after 90 days, although it is likely that COVID-19 may have
inflicted some permanent scarring on the economy.
Although the CEA makes these optimistic assumptions for simplicity,
they do not significantly bias the estimate of the value of OWS. This is because
they apply equally to both the case that a vaccine is developed by January
122 |

Chapter 4

2021 and the counterfactual comparison without OWS that it is not developed
until September. The CEA’s analysis likely underestimates the true value of a
COVID-19 vaccine because it does not include harder-to-measure factors such
as loss of human capital and non-COVID negative health effects or the value of
a vaccine to countries other than the United States.

Supporting the Healthcare System
Along with the Administration’s efforts directly related to the COVID-19
pandemic, it is undertaking deregulatory initiatives to support the healthcare
system more broadly. In addition, providing financial support to healthcare
providers is critical to avoid exacerbating health risks for Americans.

Deregulation
Beyond working toward a vaccine, the Trump Administration has expanded
short-term supply of healthcare services to meet the needs of the pandemic
by enacting a variety of deregulatory actions across Federal agencies. Some
of the larger changes, such as granting nurse practitioners more autonomy
by loosening scope-of-practice regulations and removing restrictions on the
provision of telemedicine, are dealt with more thoroughly later in this chapter
because they represent significant opportunities for long-term improvements
in the regulatory space. In addition to these major actions, regulators at
various agencies within HHS took a number of less quantifiable but significant
actions that increased the capacity of healthcare providers to meet the needs
of their communities.
One of the primary public health concerns at the onset of the pandemic
was the dearth of testing capabilities. To quickly expand diagnostic capacity, the FDA utilized EUA procedures and allowed for the production of tests
earlier in their life cycle. To supplement these actions on the production side,
the Trump Administration increased consumers’ ability to access COVID-19
diagnostic testing by relaxing scope-of-practice regulations with regard to
which healthcare providers were able to administer testing and by reducing
or eliminating the out-of-pocket cost of testing through the CARES Act. The
National Institutes of Health expanded on diagnostic efforts by investing in
improvements in rapid testing technology.
As some localities began to be hit hard by COVID-19 outbreaks, one of
the key public health risks was the limited supply of healthcare providers. To
address this concern, CMS relaxed a plethora of occupational licensing restrictions to increase the number of providers. The supply of doctors and nurses
was increased by allowing those with licenses that had expired or were still
under review to practice. CMS also used deregulatory action to increase the
supply of other healthcare workers by waiving certain licensing requirements
for positions like nurse aides and paid feeding assistants. Such actions were

Advancing the Quality and Efficiency of America’s Healthcare System | 123

particularly beneficial for hard-hit long-term-care facilities, whose patients are
disproportionately at risk from COVID-19. CMS also encouraged out-of-State
practitioners to assist in harder-hit areas by removing Federal restrictions on
their ability to provide care to Medicare beneficiaries outside their State of
licensure.
The Administration also helped to mitigate dangerous shortages of personal protective equipment (PPE). During the early months of the pandemic,
a key risk to healthcare workers was the limited supply of PPE and stringent
Federal regulations on how it must be used. To provide a temporary increase in
the supply of PPE and protect healthcare providers working in settings that put
them at high risk of contracting COVID-19, the FDA’s EUA and the Families First
Coronavirus Relief Act (FFCRA) allowed for highly protective facemasks initially
designed for use in industrial settings to also be used in medical settings.
Furthermore, CMS removed regulations that limited the ability of healthcare
providers to store and reuse masks, which gave hospitals increased autonomy
in determining what PPE policies they wanted to implement and substantially
decreased demand for new masks in facilities that chose to capitalize on the
deregulation.
In addition to using deregulation to increase the number of healthcare
providers and the supply of PPE, the Trump Administration loosened regulations of hospital classifications and facilities. To reduce the spread of COVID-19
within hospitals, HHS allowed hospitals to screen potential patients offsite to
prevent the spread of COVID-19. As hotspots arose in large cities, CMS allowed
for the expansion of patient care areas to respond to sudden increases in
demand for medical services. CMS also waived eligibility requirements for
several classifications of rural hospitals to allow them to expand their capacity
and serve their communities during the pandemic. Many of CMS’s deregulatory actions for facilities benefited long-term-care facilities, including waiving
resident group requirements for in-person meetings, statutory limitations
on transfers and discharges, and requirements to honor resident roommate
requests. All these actions were undertaken to decrease the risk of COVID-19
spreading among both the patient and provider populations.
Finally, CMS temporarily waived a number of paperwork and bureaucratic requirements during the pandemic to allow healthcare providers to
make informed decisions about how to prioritize their time and best meet their
patients’ needs. These included regulations of the time frame for reporting
requirements, the necessity of verbal orders, discharge planning, emergency
preparedness plans, patient privacy, utilization reviews, and food plans.

Financial Support for Healthcare Providers
The COVID-19 pandemic represented a threat to the financial solvency of healthcare providers across the country, restricting their ability to ensure high-quality

124 |

Chapter 4

care for patients in their communities. In response, the Administration worked
with Congress to pass the CARES Act, which established the Provider Relief
Fund to help healthcare providers in the midst of the pandemic. The CARES Act,
through HHS, made up to $100 billion available to eligible hospitals and other
healthcare providers, which constituted about 4.5 percent of spending from
the bill. The PPP/HCE Act provided an additional $75 billion for the Provider
Relief Fund to reimburse healthcare providers for expenses related to healthcare and lost revenues that are attributable to COVID-19. In addition, the PPP/
HCE Act provided $25 billion to help increase COVID-19 testing. This includes up
to $1 billion to reimburse the cost of testing uninsured individuals, in addition
to the $1 billion previously appropriated for this purpose by the FFCRA.
The FFCRA also, as amended by the CARES Act, requires Medicare Part
B, State Medicaid, Children’s Health Insurance Programs (CHIP), and group
health plans and health insurance issuers to cover COVID-19 diagnostic testing without cost sharing for patients. Uninsured individuals may also obtain
COVID-19 diagnostic testing free of charge under the State Medicaid programs,
if the State offers this option. CMS has developed an accessible, easy-to-use
toolkit for States to amend their Medicaid programs so they can offer this service. The CARES Act also appropriated $150 billion for the Coronavirus Relief
Fund, which is administered by the Department of the Treasury, to reimburse
expenses incurred by State, local, and Tribal governments as part of their
response to the COVID-19 pandemic.
With funding allocated by the CARES Act and the PPP/HCE Act, HHS can
allocate up to $175 billion of aid to eligible hospitals and other healthcare
providers to offset these costs. Over $100 billion had been paid to hospitals and
other providers by early October. This includes relief to hospitals that serve the
most vulnerable segment of the population as well as rural hospitals and those
in small metropolitan areas.
Canceling elective surgeries played a major role in declining revenue
for many providers. Following the advice of both State-level policymakers
and the surgeon general, in mid-March, elective surgeries were canceled or
postponed as part of the effort to curb the spread of COVID-19 and prevent the
potential straining of healthcare infrastructure and resources during the pandemic. Figure 4-4 shows the decline and subsequent recovery of five types of
visits of Medicare patients relative to the comparable week in 2019, with total
knee arthroplasties reaching as low as 3.2 percent of their baseline volume in
mid-April. As restrictions were lifted throughout the summer, elective surgery
volumes rebounded, with most at or near their baseline figures by early July.
This likely represents a temporary surge in volume for those who rescheduled
surgeries immediately after the end of restrictions but an overall lower demand
for elective surgeries in the Medicare population.
However, due in part to the financial support that was provided to providers, healthcare has proven to be one of the most resilient labor markets

Advancing the Quality and Efficiency of America’s Healthcare System | 125

$"0- у-4ѵ/$*)' $- /$'$5/$*)ѶJan. рп–Oct. тп, спсп
Knee replacement
Non-COVID ER visit
Cholecystectomy
Visits (percentage of 2019 baseline)

Hip replacement
Colonoscopy

140

Oct. 30

120
100
80
60
40
20
0
Jan.

Feb. Mar.

Apr.

May

Jun.

Jul.

Aug.

Sep.

Oct.

.Sources: Department of Health and Human Services; CEA calculations.
Note: ER = emergency room.

during the pandemic. Figure 4-5 shows employment by sector for each month
of 2020 as a percentage of the 2019 baseline using data from the Bureau of
Labor Statistics (BLS). Healthcare employment fell to 92.2 percent of its 2019
level in April, the second-smallest decline of any sector. In contrast, average
employment in all sectors in April was 86.6 percent and employment in leisure
and hospitality was particularly volatile, falling to 51.8 percent. Healthcare has
so far remained the second-most-resilient sector, after financial services, for
the duration of the recovery and has steadily regained employment, rising to
97.2 percent of its 2019 level in October.
One major concern from the rapid job losses in March and April due to
COVID-19 was the loss of health insurance for those obtaining benefits through
employment. As of May 2, the Kaiser Family Foundation estimated that 47.5
million people who were covered by employer-sponsored insurance (ESI) were
part of a family in which someone had lost a job (CBO 2020; Garfield et al. 2020).
Of this group, about 26.8 million could potentially lose their health insurance,
with the remaining 20.8 million retaining ESI though another worker in their
family or another source of coverage. Given this consideration, all but 5.7 million would then be eligible for publicly subsidized coverage via Medicaid or
marketplace subsidies, significantly reducing the share of job losses that result
in a lack of health insurance.

126 |

Chapter 4

$"0- 4-фѵ*)/#'4 (+'*4( )/4 /*-Ѷспсп
Healthcare
Manufacturing
Business services

Mining/logging
Trade
Education

Construction
Finance
Leisure/hospitality

Employment (percentage of 2019 baseline)
110

Oct. 20

100
90
80

70
60
50
Jan.

Feb.

Mar.

Apr.

May

Jun.

Jul.

Aug.

Sep.

Oct.

. ources: Bureau of Labor Statistics; CEA calculations.
S

However, these projections have not been borne out in the data thus far.
Data from Americans in the Household Pulse Survey from the Census Bureau
showed minimal changes in ESI coverage between the end of April and the
end of September, as Americans reported being both insured and uninsured
at slightly lower rates, with a substantial increase in those who did not report
or reported “don’t know.” In fact, between the end of April and the end of
August, Pulse results showed that uninsurance rates had actually declined
by 0.6 percentage point. The disparity between the observed changes in ESI
coverage and initial projections may in part be due to the PPP/HCE Act allowing
forgivable loans to employers to cover payroll costs, including employer contributions to health insurance coverage. Ultimately, although microsimulation
modeling can be used to approximate the decline in health insurance coverage
due to COVID-19, survey data to quantify the effect remains inconclusive at this
time.

Subsidizing Beneficial Behaviors
and the Cost of COVID-19 Care
Testing is essential to identifying positive COVID-19 cases, quarantining and
treating sick patients, and implementing contact tracing protocols. Test costs
may be a barrier to some members of the public, which could thwart efforts
to contain a pandemic. Passage of the FFCRA on March 18, 2020, reduced this

Advancing the Quality and Efficiency of America’s Healthcare System | 127

potential cost barrier for American families. Nearly all public and private insurance plans are required by this legislation to cover FDA-approved COVID-19
tests and any costs associated with diagnostic testing with no cost sharing, as
long as the test is deemed medically appropriate by an attending health care
provider and the federally declared public health emergency is in effect. The
CARES Act, which was enacted on March 27, 2020, further mandated that private plans reimburse out-of-network COVID-19 tests up to a publicly reported
cash price. The FFCRA Relief Fund includes up to $2 billion ($1 billion appropriated through the FFCRA, and up to $1 billion appropriated through the PPP/
HCE Act) to reimburse healthcare providers who conduct COVID-19 testing for
uninsured individuals, which could raise the likelihood that these individuals
seek testing when they feel ill and therefore contribute to the nation’s public
health objective of mitigating the COVID-19 pandemic. As of September 22,
2020, the CDC has awarded over $12 billion to States, Tribes, localities, and territories. This total includes $10.25 billion for critical support to enhance COVID19 testing and related activities at the State and local levels. All these Federal
protections have reduced the cost barriers of COVID-19 testing—which, in turn,
has helped the United States identify positive COVID-19 cases and deliver care
to individuals who have contracted COVID-19.

Emergency Paid Sick and Medical Leave
To slow the spread and contain the COVID-19 pandemic, the Administration
has encouraged members of the public to stay home when they are sick or
caring for a family member who is sick. At the same time, the Administration
has firmly acted to prevent American workers from trading off work hours for
their own or a family member’s health and the broader public’s health protection. As provided by the FFCRA, on April 1, 2020, the U.S. Department of Labor
announced that private employers with fewer than 500 employees are eligible
for tax credits for costs associated with providing paid leave for COVID-19 until
December 31, 2020. These dollar-for-dollar reimbursements through tax credits
enable employers to keep their workers on the payroll when their employees
become sick or are caring for someone with COVID-19 and are unable to work,
which promotes public health and maintains the flow of financial support to
both employers and employees. For employers that could not cover the cost of
paid leave with funds they would otherwise pay to the Internal Revenue Service
in payroll taxes, the FFCRA enabled employers to seek an expedited advance
from the Internal Revenue Service through streamlined reimbursement claims.

Subsidizing the Cost of COVID-19 Care
In addition to financing the detection of COVID-19 in order to implement
containment and mitigation procedures, the Administration has also provided
Federal support to reduce the cost of COVID-19 treatment. The Administration
has responded in several ways to ensure that individuals seek the care that
they need.
128 |

Chapter 4

Many private Medicare health plans, known as Medicare Advantage plans,
have expanded coverage to meet the unique needs of Medicare beneficiaries
during a pandemic, including telehealth and medical transportation benefits.
These types of support are especially important for lower-income individuals
in the elderly population who would otherwise face cost or mobility constraints
that would make obtaining medical care for COVID-19 difficult.
In addition, through the use of “1135 waivers,” the Administration has
created greater flexibility for Medicaid, Medicare, and CHIP requirements that
can sometimes pose challenges for healthcare providers to provide medical
care and for States to manage their Medicaid and CHIP programs during a
national emergency such as the COVID-19 pandemic. The reduced administrative burden facilitated by these waivers has helped providers deliver medical
care in these high-risk medical populations. When granted, the ultimate goal
of these is to improve the ability of States and the healthcare sector to meet
the needs of Medicare, Medicaid, and CHIP beneficiaries and expand access to
medical services for these beneficiaries during the COVID-19 pandemic.
Finally, the Administration has taken actions to address the significant
out-of-pocket medical cost burden faced by uninsured individuals when they
become ill. Life during a pandemic is especially daunting for the uninsured
because they do not have an insurance buffer in the event that they are
exposed to COVID-19 and end up suffering from it. As noted above, a total of
up to $2 billion in Federal funds appropriated by the FFCRA and the PPP/HCE
Act reduce testing cost barriers among the uninsured population. However,
the Administration has also acted to address treatment cost barriers for these
Americans. HHS is providing claims reimbursement to healthcare providers
that treat uninsured patients with COVID-19. As of November 9, $1.76 billion
had been distributed to providers to reimburse the cost of testing and treating uninsured COVID-19 patients. Of this amount, representing almost 25,000
claims, $677 million was for testing and $1.1 billion was for treatment. The
CARES Act established and appropriated a total of $100 billion to the Provider
Relief Fund, and the PPP/HCE Act appropriated an additional $75 billion in
relief funds. A portion of the Provider Relief Fund was used to reimburse
providers that are treating uninsured individuals with COVID-19. In April 2020,
the Administration began requiring providers to certify that, as a condition for
supplemental COVID-19 funding, they would not seek to collect out-of-pocket
expenses from a patient in an amount greater than what the patient would
have otherwise been required to pay for in-network care.

COVID-19 and Future Healthcare Reform
Several other key initiatives are related to COVID-19 and the future of healthcare reform. These include reform of the FDA drug approval process, the

Advancing the Quality and Efficiency of America’s Healthcare System | 129

expansion of telemedicine, and the deregulation of scope-of-practice requirements for nurse practitioners.

FDA Reform
The pandemic has also shown the value of speed in the development of new
medical breakthroughs and the key role that deregulation can play in such
efforts. At the onset of COVID-19, one of the reasons that testing was limited
was extensive Federal regulations, including the long FDA approval process. To
combat this, the Trump Administration took action through the FDA to issue
EUAs for COVID-19 diagnostic tests. Such decisive actions played a key role in
quickly ramping up testing capacity after initial delays, and they demonstrate
the value of expedited the approval of medical breakthroughs. Currently, the
United States has some of the most stringent regulations of new drugs in the
world, with some approvals taking roughly 12 years from FDA application to
market entry. As with COVID-19 testing and treatment, other new drugs have
the potential to save lives and substantially improve well-being, which creates
high opportunity costs for a long approval process. The CEA estimates that
the net present value of the social surplus gained by decreasing FDA drug
approval times by one, two, or three years would be $1.9 trillion, $3.9 trillion,
and $5.9 trillion, respectively. Experience with the Prescription Drug User Fee
Act (PDUFA) in the 1990s suggests that changes in policy can reduce approval
times on this scale.
To estimate the value of shorter approval times, the CEA first estimates
the annual social surplus generated by a drug for each year it is under patent
protection. Because the FDA’s approval time does not directly affect the patent expiration date of the average drug, the utility gained after postpatent
expiration is assumed to be unchanged. Furthermore, the CEA’s estimates of
the value produced by such a policy change likely understates the true value
because the number of new drugs introduced is treated as exogenous. In reality, shorter approval times increase the profitability of new entrants and would
lead to further advances in medical technology, providing additional value for
both consumers and pharmaceutical companies. (All dollar amounts are 2019
dollars.)
Figure 4-6 shows an average drug’s life cycle, broken down into costs,
producer surplus, and consumer surplus. The model updates the average
drug revenue profile described by Philipson and others (2008)—using data
from the FDA, BLS, and the Saint Louis Federal Reserve on the change in the
number and prices of new drug approvals. Using this updated drug revenue
profile, the CEA applies further calculations (described below) to estimate the
producer and consumer surplus generated by the average drug. Of course, in
reality most drugs will have very different revenue profiles, but the constructed
average drug in the model uses data on average total revenue over the course

130 |

Chapter 4

$"0- 4-хѵ1 -" -0" $! C4' d0-$)"the / )/ -$*
*./.

-*0 -.0-+'0.

*).0( -s0-+'0.

Dollars (billions)
тѵп
сѵф
сѵп
рѵф

рѵп
пѵф

п
п

ц
Year

рп

рр

рс

рт

*0- .ѷ#$'$+.*) /'ѵҗсппчҘѸ**)-0"($)$./-/$*)Ѹ 0- 0*! *-//$./$.Ѹ
aint *0$. -' . -1 BankѸ'0'/$*).ѵ

of the patent period and average share of revenue in each year to construct a
representative example.
Although overall revenue profiles can be easily estimated using publicly
available data on consumer expenditures, it is more difficult to calculate precise measures of producer and consumer surpluses, in large part due to the
wide variation of producers and products in the pharmaceutical industry. The
CEA estimates that the producer surplus in each year of the patent period is 80
percent of revenues, based on the finding that marginal costs are roughly 20
percent of revenue (Berndt, Cockburn, and Griliches 1996; Caves, Whinston,
and Hurwitz 1991; Grabowski and Vernon 1992; Philipson et al. 2012). Of
course, pharmaceutical companies also face high fixed costs early on in the life
cycle of a drug in the form of research-and-development costs for both successful and unsuccessful products, approval application fees, and marketing
expenditures (Kennedy 2018). A reduction in approval time may result in lower
costs associated with the approval process if the preapproval time frame has
nonnegligible marginal costs over time. However, to ensure that the result
represents a true lower bound, the CEA does not include any reduction of fixed
costs in the total benefit estimate.
To arrive at an estimate of total social surplus, the CEA conservatively
assumes that consumer surplus is equal to producer surplus. It is well documented that consumers enjoy greater benefits from the development of new
drugs than the profits made by their producers (CBO 2006; Lichtenberg 2014;
Philipson and Jena 2006; Philipson et al. 2012; Roebuck et al. 2011). In fact, the

Advancing the Quality and Efficiency of America’s Healthcare System | 131

literature suggests that consumers capture the vast majority of the social surplus generated by new drugs, meaning that the CEA may substantially underestimate the total value to consumers of reducing drug approval times. Under
these assumptions, the CEA finds that once an average drug has reached maturity in the market, it will generate about $2.1 billion in social surplus annually.
Figure 4-7 demonstrates how decreasing drug approval time by one, two,
or three years would affect this annual social surplus. The figure also accounts
for the time value of money by using an annual discount rate of 3 percent. That
is, $1 in year one is worth 97 cents in year zero. Using a discount factor accounts
for the fact that both the consumers and producers of a product would rather
have it sooner rather than later. By allowing earlier entry into the market, drugs
reach maturity in the market and provide maximum social surplus earlier than
in the status quo. The maximum social surplus is reached earlier and attains
a higher value due to the discounting of future periods, which represents the
increased value for both consumers and producers.
Some critics of FDA reform suggest that decreased approval times would
result in more unsafe products being brought to market and therefore an
increase in approval withdrawals. However, approval times decreased by over
one year under PDUFA, and Phillipson and others (2008) found no evidence of
an increase in withdrawals after the reduction in approval times, but did not
account for potential adverse effects on safety that do not result in withdrawal.
Qureshi and others (2011) found that safety-related withdrawals accounted
for less than a quarter of all withdrawals between 1980 and 2009. The CEA’s
analysis using an expanded data set of safety-related withdrawals also did not
find an increase in withdrawals after the decreased approval times of PDUFA.
Given the absence of data on the distribution of withdrawals by drug revenue,
the CEA applies the overall drug withdrawal rate of 15.9 percent as a reduction
to the potential increase in social surplus. This likely overstates the extent to
which withdrawals would decrease potential benefit due to the skewed distribution of revenue by different drugs. Although the FDA’s approval is withdrawn
for a small share of drugs for safety reasons, almost 80 percent are voluntarily
withdrawn by their producers for commercial reasons. In reality, the more successful drugs that generate larger surpluses for both producers and consumers
are less likely to be withdrawn, resulting in a conservative estimate of the
overall benefit.
Using the estimate of the net present value of a drug’s life cycle shown
in figure 4-7, the CEA calculates the marginal cumulative net present value of
social surplus generated by reducing FDA approval times, as shown in figure
4-8. The model uses the five-year average from 2015 to 2019 of 44 new drugs
per year by the FDA. As noted above, by increasing the returns on investment
in research, reducing FDA approval times would likely increase the number of
new applicants, and hence approvals. Therefore, the static model that holds
new drugs constant at 44 a year results in a conservative estimate of the value
132 | Chapter 4

$"0- 4-цѵ1 -" ))0'*$'0-+'0.4++-*1'$( - .
. c.

рy -

сy -.

тy -.

Dollars (billions)
сѵф
сѵп
рѵф
рѵп
пѵф
п
п

ч
Year

рп рр рс рт ру рф рх

*0- .ѷ#$'$+.*) /'ѵҗсппчҘѸ** )-0"($)$./-/$*)Ѹ0- 0*! *-//$./$.Ѹ
ain/ *0$. -' . -1 BankѸ'0'/$*).ѵ

$"0- 4-чѵ0(0'/$1 ""- "/ /- . )/'0 *! )- .
*$' '!- 4++-*1'$( - .
р4 -

Dollars (billions)

сy -.

Dollars (billions)
т

тy -.

сѵф

сфп
с
спп

рѵф
р

рфп

пѵф

рпп

п
п

фп

*0- .ѷ #$''$+.*)
*0$. -' . -

п
п

Year

рп

рр

рс

рт

ру

рф

рх

*0- .ѷ#$''$+.*) /'ѵҗсппчҘѸ** )-0"($)$./-/$*)Ѹ0- 0*! *-//$./$.Ѹ
Saint *0$. -' . -1 Ѹ'0'/$*).ѵ

Advancing the Quality and Efficiency of America’s Healthcare System | 133

' уҊрѵ./$(/ *$'0-+'0.4++-*1'$(
Reduction, 2025–40 (billions of real 2019 dollars)
/p- . )/v'0

р р,шпфѵч

++-*1'$( 0/$*)
с -.
т,чцпѵф

т -.
ф,чшхѵп

спсф

пѵ0

пѵп

уѵш

спсх

пѵп

уѵш

ссѵр

спсц

уѵш

ссѵр

учѵт

спсч

рцѵс

утѵу

цхѵш

спсш

схѵс

фшѵц

шцѵт

сптп

ттѵф

црѵс

рруѵф

сптр

тцѵх

чрѵп

рттѵт

сптс

утѵу

шфѵц

рфцѵп

сптт

фсѵт

рртѵц

рчрѵх

спту

хрѵт

рсшѵс

спцѵч

сптф

хцѵш

рухѵу

ссуѵш

сптх

цчѵф

рфцѵп

стуѵч

сптц

цчѵф

рфхѵс

стчѵп

сптч

ццѵц

рфшѵф

сурѵт

сптш

чрѵч

рхтѵх

суфѵу

спуп

чрѵч

рхтѵх

суфѵу

*0- .ѷ#$'$+.*)Ѷ /'ѵҗсппчҘѸ**)-0"($)$./-/$*)Ѹ0- 0*! *-//$./$.Ѹ$)/
*0$. -' . -1 )&Ѹ'0'/$*).ѵ

of deregulation, especially considering the fact that new approvals have been
trending upward since 2005. The results, given in figure 4-8, represent the
increase in social surplus for one year of drug approvals depending on whether
the approval time for the drugs is reduced by one, two, or three years.
To calculate aggregate gain in social surplus, it is necessary to sum
the gains in social surplus associated with quicker drug approvals over time.
Because policies to reduce approval time may be difficult to implement immediately, the CEA assumes that the reductions in approval time would begin
applying to drugs that would otherwise be approved in 2028. Under these
assumptions, table 4-1 displays the nondiscounted gain in social surplus from
a one-, two-, or three-year reduction in approval times for each year from 2025
to 2040, as well as the net present value in 2020 of such a policy change. The
CEA estimates that the net present value of the increase in social surplus from
a permanent reduction in approval times by one, two, or three years for new
drugs would be $1.9 trillion, $3.9 trillion, or $5.9 trillion, respectively.

Telemedicine Deregulation
One of the most substantial deregulatory opportunities for long-term healthcare
improvement that has been highlighted during the pandemic is telemedicine.
134 | Chapter 4

Early during the pandemic, HHS took four key deregulatory actions to increase
the availability of telemedicine opportunities. First, the Office for Civil Rights
(OCR) announced that it would relax enforcement of HIPAA regulations to allow
health professionals to communicate with patients and provide telehealth
services via remote communication technologies that may not fully comply
with HIPAA privacy rules. Though the laws remain unchanged, OCR used its
enforcement discretion to allow any covered health professionals to use a
wide array of commercially available communication technology (e.g., Zoom
or Skype) as part of a good faith effort to provide telehealth services during
the pandemic, regardless of whether the services are directly related to the
diagnosis or treatment of COVID-19.
Second, President Trump’s emergency declaration allows HHS to relax
Federal licensing restrictions so many health professionals can provide care
virtually to patients in other States. This has created a large pool of potential
health professionals available to any given patient who is seeking telehealth
services, increasing access to medical services in the States with the greatest need. Finally, CMS took two significant deregulatory actions to promote
telehealth by temporarily expanding the scope of Medicare telehealth to allow
Medicare beneficiaries across the country—not just in rural areas—to receive
telehealth services from any location, including their homes, as well as adding
over 135 allowable services, more than doubling the number of services that
beneficiaries could receive via telehealth (Verma 2020). The CMS temporarily waived statutory and regulatory provisions that restrict reimbursement
for telemedicine services to those furnished in certain healthcare facilities,
allowing healthcare professionals to be paid for providing telehealth services
regardless of location. CMS also allowed for a broader range of services to be
provided via video or audio call, including emergency department visits, therapy services, and initial nursing facility and discharge visits. These measures
are designed to promote the use of telemedicine and ensure that patients have
access to healthcare while remaining safely at home.
During the beginning stages of the pandemic, quick deregulatory action
mitigated disruptions in care for patients in hotspot areas and those in the
greatest need. Mann and others (2020) found that telemedicine visits increased
almost sevenfold during the period of maximal COVID-19 active cases in New
York City. Many of these online visits were directly related to COVID-19, which
advanced three key public health goals. First, telemedicine allows for comparatively inexpensive and efficient screening for patients before they arrive in the
emergency room. This lowers costs and prevents unnecessary healthcare visits, which decrease the strain on already-overburdened healthcare providers
and the potential transmission of COVID-19 to other patients and healthcare
workers. Second, expanding access to telemedicine provides useful data to
public health officials who are trying to track the spread of the disease and predict future hotspots, an approach that has been shown in the past to provide a

Advancing the Quality and Efficiency of America’s Healthcare System | 135

useful picture of the spread of influenza (Chauhan et al. 2020). Third, provision
of telehealth services that is not directly related to COVID-19 is particularly
necessary for patients who are actively quarantining and require healthcare,
because in-person visits with such patients increase the risk of exposure for
healthcare workers and their patients.
Telemedicine visits have also been useful in maintaining access to
essential care services when physical access to medical services has been
limited. For seniors who are at a heightened risk of serious illness from COVID19, telemedicine has offered an appealing substitute due to the deregulatory
actions of CMS. Telehealth visits constituted 43.5 percent of Medicare primary
care visits in April, compared with just 0.1 percent of such visits before the
pandemic in February. Urban areas that have had higher levels of COVID-19
hospitalizations have utilized telehealth services at a higher rate, suggesting
that this uptake has been at least partly driven by concerns over COVID-19.
With uncertainty and unemployment rising during the pandemic, telehealth
services have also provided a safe and efficient method to meet rising demand
for mental health services among patients of all ages. During the February-toApril period, increases in Medicare telehealth utilization for primary care visits
were dramatic in every State; for example, visits went from 0.20 percent to 43.9
percent in Texas and from 0.03 percent to 69.7 percent in Massachusetts.
According to survey data from McKinsey & Company, 11 percent of U.S.
consumers used telehealth services in 2019 (Bestsennyy et al. 2020). As of
April 2020, 46 percent of U.S. consumers reported that they had already used
telehealth to replace canceled in-person healthcare visits in 2020. Though
telehealth has helped expand access to care at a time when COVID-19 has
restricted patients’ ability to see their doctors, there has been strong interest in
making telehealth services a permanent option; 76 percent of U.S. consumers
report being interested in using telehealth in the future. The enthusiasm for
telehealth on the demand side is matched by favorable reviews of telehealth
on the supply side; 57 percent of providers view telehealth more favorably than
they did before COVID-19, and 64 percent are more comfortable using it. The
positive reaction to exercising telehealth options is likely to increase over time
as awareness and experience with virtual healthcare services grow and existing
challenges (e.g., lower mobile and computer capabilities in lower-income communities and security concerns) are resolved.
The immediate and pressing nature of the COVID-19 pandemic has
demanded that the healthcare system embrace telemedicine on a greatly
accelerated timeline. Though the availability of telehealth services has been
increasing consistently over time, the additional infrastructure built and
deregulatory actions taken provide an opportunity to more strongly embrace
telehealth as a key part of the future of healthcare. In 2019, the American
Hospital Association identified Medicare reimbursement differentials and
regulatory barriers as two key barriers to wider adoption of telemedicine in
136 |

Chapter 4

the United States. Many of these regulatory burdens have been temporarily
removed, and healthcare systems have already implemented telemedicine
programs in response to the pandemic, so they can use them beyond COVID-19
without incurring additional setup costs if HHS’s deregulatory actions become
permanent. Although the benefits to individuals in quarantine and those at a
high risk of contracting COVID-19 will decrease once the threat of the pandemic
has passed, other benefits will remain. Studies of telemedicine programs
have found that they increase patient satisfaction, decrease the loss of work
time (which decreases the opportunity costs for patients to seek care they
need), and decrease the unnecessary use of the emergency department due to
prescreening arrivals, which lowers costs and improves the quality of care for
patients who need it most.
In addition, though the greatest beneficiaries of increased availability
of telemedicine during the pandemic have been patients in urban areas, the
long-term benefits of normalizing telemedicine will be highest among rural
Americans who do not reside near major medical centers. The Department
of Veterans Affairs found that 45 percent of its telemedicine utilization came
from rural veterans. Telemedicine would allow greater access to specialists
with knowledge in a particular area of medicine, even when doctors are not at
the same hospital or region of the country. Furthermore, rural populations are
particularly subject to high opportunity costs for medical care, including lost
wages, transportation costs, and childcare expenses. On the basis of a study
of this phenomenon by Bynum and others (2003), the CEA estimates that rural
Americans would on average save $130 per visit in opportunity costs such
as fuel, wages, and other family expenses if their visits could be replaced by
telemedicine. Rural patients who would otherwise make the national average
2.8 physician’s office visits a year would therefore save up to $362 annually.
Though rural patients may empirically make fewer physician visits per year
(Spoont et al. 2011), the increased access provided by telemedicine may
reduce the geographic disparity between rural and urban Americans.
Given both consumers’ and providers’ interest in continued access to
telemedicine, it is a potentially significant source of future economic value.
McKinsey & Company estimates that before the COVID-19, the total annual revenue of U.S. telehealth players was about $3 billion, with the largest vendors
being focused on virtual urgent care (Bestsennyy et al. 2020). They estimate
that going beyond this segment of virtual healthcare may allow up to $250 billion, or $1 in $5 current healthcare dollars, to be virtualized.

Scope-of-Practice Deregulation
During the COVID pandemic, relaxing stringent scope-of-practice (SOP)
requirements allowed hospitals and other health providers to increase the
amount of care that they could provide for their communities. Before the
outbreak of COVID-19, 22 States and 2 territories allowed full practice for
Advancing the Quality and Efficiency of America’s Healthcare System | 137

nurse practitioners (NPs), meaning that NPs in those States and territories are
authorized by their boards of nursing to evaluate and diagnose patients, order
and interpret diagnostic tests, and manage treatments (including prescribing
medication) without a physician. Increased demand from virus patients combined with decreased supply due to practitioners being out sick threatened to
overwhelm hospital systems across the country. In contrast, States with more
restrictive SOP guidelines place restrictions on NPs in one or more of these
areas, generally in the form of prohibitions or physician supervision requirements. In response, State governments and Federal agencies relaxed SOP
guidelines that prevented nurse practitioners from performing certain routine
tasks without the supervision of a licensed physician. By April 24, 2020, another
22 States had temporarily relaxed their SOP requirements. In addition, CMS
temporarily relaxed its SOP guidelines in March 2020. Medicare and Medicaid
reimbursement payments are critical for the survival of many hospitals, and
State regulations are always binding. Because of this, hospitals tend to operate under the more rigid regulations when their State and CMS regulations are
in conflict. This has enabled providers in areas that have been hit hardest by
COVID-19 to respond with increased labor flexibility in meeting the needs of
their communities.
Existing SOP restrictions on NPs display a strong geographic correlation
(figure 4-9). This is likely due to the greater benefits associated with broadening
SOP in rural areas relative to urban communities, given that full practice was
primarily allowed in New England, the northern Great Plains, the Mountain
West, and the Pacific Northwest. Rural areas rely more heavily on NPs and grant
them greater autonomy than urban areas because they tend to have fewer
physicians to oversee the NPs (Rosenblatt and Hart 2000). This shortage of
physicians can prevent the opening of community health centers (CHCs). The
opening of new CHCs in rural areas was associated with relaxed SOP requirements. Furthermore, CHCs in States with relaxed SOP guidelines have more
NPs relative to physicians than CHCs in States with rigid SOP guidelines (Shi
and Samuels 1997). More CHCs mean better access to care in rural areas. And
because relaxing SOP allows more CHCs to open and more CHCs mean better
access to care, deregulating SOP would improve the ability for rural populations to access healthcare.
In addition to expanding access, relaxing SOP regulations drives down
healthcare costs. Such restrictions increase the cost of healthcare, because
NPs are unable to perform certain tasks without the supervision of a physician
and physicians’ time is expensive. Rigid regulations requiring physicians to
perform some tasks increased the cost of well-child medical exams by 3-16 percent (Kleiner et al. 2016) Another analysis found that costs were lower in States
with reduced and full SOP than in States with restrictive SOP (Spetz et al. 2013).
To estimate the economic benefit of relaxing SOP guidelines for NPs
nationwide, the CEA uses interstate cost comparisons from Poghosyan and

138 |

Chapter 4

$"0- 4-шѵ// *+ -*!--/$ - "0'/$*)

*0- ѷ( -$)..*$/$*)*!0-. -/$/$*) -.ѵ

3+$- o- (+*--4w$1 *a/$*)
3$./$)"f0'' p-/$

others (2019), who estimate the difference in outpatient and prescription
drug costs for Medicaid patients between States that allow for full, reduced,
and restricted practice for NPs. Using these figures, along with data from
BLS and the Kaiser Family Foundation, the CEA estimates that allowing full
practice nationwide would reduce outpatient costs by $33.96 billion a year
and prescription drug costs by $27.73 billion a year across patients enrolled
in employer health plans, nongroup plans, or Medicaid. This would lead to a
reduction in national prescription drug spending of 5.3 percent and, combined,
represent a reduction in national healthcare expenditures of 1.7 percent. Due
to the limited supply of NPs, this number represents the potential long-run
benefit once the labor market for NPs has expanded to match the increased
demand. However, the supply of NPs has been flexible, more than doubling the
past 15 years as States have removed SOP restrictions.
The CEA’s estimate likely understates the total benefit in two ways. First,
Medicaid spending per capita is lower than the privately insured population,
so the savings for the general population in dollar terms may be larger than for
Medicaid enrollees. Second, the CEA’s analysis only accounts for individuals
who are members of employer health plans, nongroup plans, or Medicaid. It
is likely that relaxing SOP for NPs would also reduce costs for other groups,
including those insured by military plans or Medicare, as well as the uninsured
population.

Advancing the Quality and Efficiency of America’s Healthcare System | 139

The impact of relaxing SOP on health outcomes could go one of three
ways. If relaxing SOP restrictions causes NPs to provide lower-quality care in
the absence of physician supervision, then relaxing SOP would have a negative
effect on health outcomes. If, instead, NPs performed just as well as doctors,
then there would be no effect on health outcomes. In addition, if NPs could
now perform more critical health actions, which previously could not have
been performed due to a shortage of physicians to provide supervision, then
one would expect health outcomes to improve when SOP restrictions are
relaxed.
Empirical evidence suggests that allowing nurse practitioners full practice nationwide would not compromise the quality of patient care. State-level
SOP restrictions had no effect on infant mortality or malpractice insurance
premiums (Kleiner et al. 2016). Taking a broader approach, another study
found that
the considerable variation in the results for the measures included in each of
the domains of primary care quality indicators we assessed—chronic disease
management, cancer screening, ambulatory care–sensitive hospital admissions, and adverse outcomes—did not reveal a consistent pattern or relationship with state-level SOP. (Perloff et al. 2017)

In rural areas, the results of one analysis suggested a positive relationship between health outcomes and relaxed SOP guidelines (Ortiz et al. 2018).
A wealth of literature analyzing the difference in patient outcomes between
NPs and physicians has consistently found that, for most patients, NPs provide
equivalent or better care at a lower cost (Lenz et al. 2004; Martin-Misener et
al. 2015; Mundinger et al. 2000; Oliver et al. 2014; Stanik-Hutt et al. 2013). The
States and Federal agencies that have temporarily relaxed their SOP guidelines
during the COVID-19 pandemic could seize this opportunity to improve the
access and affordability of healthcare for their citizens.

Additional Changes to Promote Choice and Competition
Beyond the response to the COVID-19 health crisis, the Trump Administration
has championed several healthcare reforms to promote additional choice
and competition in the market. These policies will provide tangible reform to
Americans and play a critical part in the swift comeback for the U.S. economy.
First, CMS introduced site-neutral payment in 2019 for clinic services
delivered by hospitals. Site-neutral payments were part of the 2019 Hospital
Outpatient Prospective Payment System final rule and address unnecessary
increases in utilization of clinic visits in off-campus, hospital-based departments. Medicare and beneficiaries often pay more for the same type of clinic
visit in the hospital outpatient setting than in the physician office setting. The
rule was challenged by a coalition of hospitals led by the American Hospital

140 |

Chapter 4

Association in Federal court. In September 2019, the U.S. District Court for the
District of Columbia ruled that CMS had overstepped its statutory authority in
making the changes. However, a July 2020 decision issued by the U.S. Court of
Appeals for the District of Columbia Circuit overturned the lower court’s ruling,
clearing the path for implementation. Site-neutral payments are estimated to
generate healthcare savings that have a direct and positive impact on beneficiaries, the Medicare program, employers, and American taxpayers. An evaluation by CMS that has been extrapolated by the CEA shows that site-neutral
payments for evaluation and management services are projected to save the
Medicare program an estimated $330 million and lower patient copayments by
$88 million in 2021.
Second, prescription drugs saw their largest annual price decrease
in nearly half a century in 2019. For three consecutive years, the FDA has
approved a record number of generic drugs. The CEA estimates that these
approvals saved patients $26 billion in 2017 and 2018. The 2020 Creating and
Restoring Equal Access to Equivalent Samples Act will also create opportunities for greater savings from generic drugs by increasing access to samples for
testing. The CEA estimates that the projected savings to American taxpayers
will be $3.5 billion from 2020 to 2030.
Also, in July 2019, the Trump Administration issued an Executive Order
aimed at improving the care of patients with chronic kidney disease. In 2020,
the Department of Health and Human Services published multiple rules that
attempted to streamline the renal care system by removing regulatory barriers,
increasing oversight of Organ Procurement Organizations, and encouraging
living kidney donors. HHS estimates that its changes to the system of these
organizations alone could generate up to 4,500 additional kidney transplants
by 2026. The CEA estimates that these initiatives could have substantial health
and economic benefits. Because each kidney transplant reduces lifetime
medical spending by an estimated $136,000 and creates health benefits, such
as increased longevity, that are worth an estimated $1.8 million, the net present value of these kidney transplants would be roughly $8.8 billion a year.
Moreover, efforts to promote peritoneal dialysis could result in savings of $130
million to $450 million annually. When combined with the value of health gains
and savings from kidney transplants, the CEA finds that the Administration’s
initiatives could provide societal benefits with a net present value of nearly
$9.3 billion.

Conclusion
Although COVID-19 has imposed significant health and economic costs
throughout 2020, the Trump Administration has been able to take decisive
actions to mitigate its effects. Expediting the development of testing and
treatment capabilities has played a key role in curbing the human cost of the

Advancing the Quality and Efficiency of America’s Healthcare System | 141

virus, while the removal of burdensome regulation and provision of financial
support have helped the healthcare sector adjust to the adverse shock. The
Nation’s experience with COVID-19 provides opportunities for extending the
suspension of harmful regulations, which will further encourage economic
recovery and provide long-term health and financial benefits. In particular, the
CEA finds that reforming the FDA drug approval process to reduce approval
times, encouraging the widespread continuation of telemedicine, and removing harmful scope-of-practice regulations would generate significant savings
and improve the health of Americans in the future.

142 |

Chapter 4

Part II

The Renaissance of
American Greatness

143

x
Chapter 5

Assessing the Early Impact
of Opportunity Zones
The Tax Cuts and Jobs Act of 2017 not only cut taxes for businesses and
individuals broadly but also made targeted cuts to spur investment in economically distressed communities designated as Opportunity Zones (OZs). In this
chapter, the Council of Economic Advisers (CEA) compares the advantages
of OZs with those of other Federal antipoverty programs and documents the
characteristics of the nearly 8,800 low-income communities designated as OZs.
It also quantifies the effect of OZs investment and finds that a large increase is
already benefiting OZ residents while potentially having only a small effect on
the Federal budget.
OZs chart a new course in Federal policy aimed at uplifting distressed communities. Antipoverty transfer programs subsidize the consumption of goods such
as housing and healthcare but can lead to reduced economic activity by raising
taxes and discouraging eligible, working-age participants from seeking jobs.
Also, under other existing place-based development programs, the Federal
government selects who receives grants or tax credits and narrowly prescribes
their use. By comparison, OZs cut taxes to increase economic activity by spurring private sector investment, job creation, and self-sufficiency. They also give
greater scope for market forces to guide entrepreneurs and investors because
they have no cap on participation and require no government approval.
The CEA finds that OZs, which are census tracts nominated by State governors
and certified by the U.S. Department of the Treasury to be eligible for the investment tax cuts, are among the poorest communities in the United States. These
communities have an average poverty rate more than double that of all other

145

communities and are home to a higher share of African Americans, Hispanics,
and high school dropouts. Even among all the communities eligible to be an
OZ under Federal law, every State selected communities that, on average,
had a median household income less than that of communities that were not
selected.
The CEA also finds that the OZ tax cuts have spurred a large investment
response. This chapter estimates that Qualified Opportunity Funds raised $75
billion in private capital by the end of 2019, most of which would not have
entered OZs without the incentive. This new capital represents 21 percent of
total annual investment in OZs and helps explain why the CEA also finds that
private equity investment in OZ businesses grew 29 percent relative to the
comparison group of businesses in eligible communities that were not selected
as OZs.
The growth in investment has already made OZs more attractive to their
residents, as reflected in what buyers are willing to pay for homes located in
OZs. The CEA estimates that Opportunity Zone designation alone has caused
a 1.1 percent increase in housing values. Greater amenities and economic
opportunity behind the housing value increase will be broadly enjoyed, and for
the nearly half of OZ residents who own their homes, the increase provides an
estimated $11 billion in new wealth.
With regard to effects on the Federal budget, the CEA finds that each $1 raised by
Qualified Opportunity Funds through 2019 has a direct forgone Federal revenue
effect of 15 cents. By comparison, each $1 in investment spurred by the New
Markets Tax Credit, an existing Federal program with similar goals, results in 18
cents of forgone revenue. Including indirect effects, the CEA estimates that the
OZ incentive could be revenue neutral, with economic growth in low-income
communities reducing transfer payments and offsetting forgone revenues from
taxes on capital gains. Thus, the CEA projects that the capital already raised by
Qualified Opportunity Funds could lift 1 million people out of poverty and into
self-sufficiency, decreasing poverty in OZs by 11 percent.

146 |

Chapter 5

The COVID-19 pandemic slowed investment everywhere in the second quarter
of 2020, including in Opportunity Zones, but the initial evidence suggests that
the OZ model has power to mobilize investors; engage State, local, and tribal
stakeholders; and improve the outlook for low-income communities—all with
limited prescription from the Federal Government. This chapter’s findings
highlight the potential for the Opportunity Zone model to help spur the postCOVID-19 recovery in thousands of distressed communities across the United
States.

ne of the main provisions of the Tax Cuts and Jobs Act, which was
signed in December 2017, reduced U.S. corporate income tax rates to
bring them in line with international levels. Lowering the corporate
tax rate decreases the cost of capital, thereby stimulating investment and
growth in gross domestic product and wages (CEA 2017). The Opportunity
Zones (OZs) provision of the act mirrored this effort to lower capital taxes
but with a focus on distressed communities. By reducing taxes on the capital
gains invested in such communities, the provision lowers the cost of capital for
businesses, which is expected to lead to new investment, jobs, and economic
opportunity that has been lacking for decades. This CEA chapter compares
the advantages of OZs relative to other Federal antipoverty programs, and it
documents the characteristics of the nearly 8,800 low-income communities
designated as OZs. The CEA also quantifies the effect of OZs on investment,
finding a large increase that is already benefiting residents while potentially
having only a small effect on the Federal budget.
To stimulate investment in OZs, the provision provides three potential
tax benefits to investors that invest capital gains in Qualified Opportunity
Funds, which are vehicles for investing in qualified OZ properties. The first
benefit of investing in these funds is that the investor can defer paying taxes
on capital gains rolled into OZs until potentially as late as 2026. Second, when
these taxes are paid, the investor may omit 10 percent (15 percent) of the original gain if the investment is held there for at least five (seven) years.1 Finally,
and most important, any capital gains that accrue to investments in a Qualified
Opportunity Fund are tax free if the investment is held for at least 10 years.
Funds can make equity investments in partnerships or corporations that
operate in OZs as determined by various tests, such as where they generate
1Because an investor must pay capital gains taxes on the original gain by 2026, the original option
to pay taxes on only 85 percent of the original has expired and would not apply to investments
made in 2020. This is because the investments could not be held for the original seven years before
having to pay the tax.

Assessing the Early Impact of Opportunity Zones

| 147

income or where their assets lie. A Qualified Opportunity Fund can also directly
purchase tangible property for use in the fund’s trade or business, but the
property must have its original use begin with the fund or the fund must substantially improve the property. For example, a Qualified Opportunity Fund
could purchase and install new solar panels in an OZ, or it could buy an apartment building and substantially improve it.
Although the Federal tax incentive described here is at the core of OZs,
all levels of government have worked to complement this incentive. At the
Federal level, on December 12, 2018, President Trump signed Executive Order
13853, which established the White House Opportunity and Revitalization
Council.2 The order gave the council the mission of leading efforts across
executive departments and agencies “to engage with State, local, and tribal
governments to find ways to better use public funds to revitalize urban and
economically distressed communities.” In its one-year report to the President,
the council made 223 recommendations to this end and, as of August 2020 has
taken more than 270 related actions.
Complementary efforts have also occurred at the State and local levels.
For example, the Alabama Incentives Modernization Act provides additional
State tax breaks for Qualified Opportunity Funds, and the State of New Jersey
has created an OZ website and data tool with resources for local governments,
investors, and businesses. The city of Erie, Pennsylvania, along with local
businesses and nonprofit leaders, has created the Flagship Opportunity Zone
Development Company to encourage investment in the city’s OZs. And the city
of Cleveland has taken a similar approach by creating the Opportunity CLE
initiative to promote local OZ investments.
The CEA finds that OZs, which are census tracts selected by governors
to be eligible for the investment tax cuts, are among the poorest communities
in the United States. These communities have an average poverty rate that is
more than double that of other communities and are home to a higher share
of African Americans, Hispanics, and high school dropouts. Even among all
the communities that were eligible to be an OZ under Federal law, every State
selected communities that, on average, had a lower median household income
than did eligible communities that were not selected.
The CEA also finds that the OZ tax cuts have spurred a large investment
response. The chapter estimates that Qualified Opportunity Funds raised $75
billion in private capital by the end of 2019, most of which would not have
entered OZs without this incentive. This new capital represents 21 percent of
total annual investment in OZs and helps explain why the CEA also finds that
private equity investment in OZ businesses grew 29 percent relative to eligible
communities that were not selected as OZs and thus act as a control group.

2 The council’s various efforts are highlighted on the interagency website OpportunityZones.gov.

148 |

Chapter 5

This growth in investment has already made OZs more attractive to their
residents as reflected in the prices buyers are willing to pay for homes located
in OZs. The CEA estimates that OZ designation alone has caused a 1.1 percent
increase in housing values. The greater amenities and economic opportunity
behind this housing value increase will be broadly enjoyed, and for the nearly
half of OZ residents who own their homes, the increase provides an estimated
$11 billion in new wealth.
With regard to effects on the Federal budget, the CEA finds that each $1
raised by Qualified Opportunity Funds through 2019 has had a direct forgone
Federal revenue effect of 15 cents. By comparison, each $1 in investment
spurred by the New Markets Tax Credit, an existing Federal program with similar goals, results in 18 cents in forgone revenue. Including indirect effects, the
CEA estimates that the Opportunity Zone incentive could be revenue neutral,
with economic growth in low-income communities reducing transfer payments
and offsetting forgone revenues from taxes on capital gains. Also, the CEA
projects that the capital already raised by Qualified Opportunity Funds could
lift 1 million people out of poverty into self-sufficiency, decreasing poverty in
OZs by 11 percent. These findings are complemented by recent research by
Arefeva and others (2020) showing that in metropolitan areas, the OZ designation boosted employment growth relative to comparable tracts by between 3.0
and 4.5 percentage points, creating new jobs across a wide range of industries
and education levels.

Comparing Opportunity Zones with Other
Antipoverty or Place-Based Programs
Unlike antipoverty transfer programs—which raise taxes and reduce the incentive for program recipients to participate in productive economic activity—OZs
lower taxes to stimulate economic activity in distressed areas. Relative to other
place-based policies, the OZ incentives are more open-ended and less topdown in their design, which makes OZs more effective at attracting investment
to communities most in need.

Antipoverty Transfer Policies
Antipoverty transfer programs provide cash grants or subsidies for the consumption of goods. Notable examples are housing vouchers, food stamps, cash
assistance for needy families, and Medicaid. Although these programs support
many Americans in need, they can also weaken the incentive for working-age
adults to find employment. Because of eligibility requirements linked to
income, taking a job or working more hours can cause a participant to become
ineligible if his or her income exceeds a program’s threshold. Considerable
evidence confirms that such programs typically discourage employment

Assessing the Early Impact of Opportunity Zones | 149

(e.g., Hoynes and Schanzenbach 2012; Jacob and Ludwig 2012; Bloom and
Michalopoulos 2001).
Antipoverty transfer programs also raise taxes to fund these transfers.
Even if the transfers and associated eligibility requirements did not discourage
work, they would still come at a cost. Each $1 raised through taxes costs society
more than $1 because of the positive marginal cost of public funds. This cost
captures the effect of a tax in driving a wedge between the market value of
what an extra hour of labor produces and the worker’s value of that hour (i.e.,
his or her opportunity cost). Given this tax wedge, each $1 in funds raised by
taxes costs society an estimated 50 cents in forgone value (Dahlby 2008; CEA
2019).
The rules governing OZs do not create a disincentive to work because
eligibility is based on community-wide measures of poverty and income rather
than those of any particular individual. Nor does the OZ incentive have the
same marginal cost of public funds associated with transfers funded by tax
revenues. The incentive cuts taxes on capital supplied to low-income communities, which reduces the tax wedge associated with the supply and demand for
capital. The forgone Federal revenue might be made up through higher taxes
elsewhere, or it could be offset by declines in government transfers because of
rising incomes in poor neighborhoods, which is considered in a later section.
OZs, nonetheless, are not a substitute for cash grants or subsidies. Not
everyone can work, and most people living in poverty do not live in OZs. To
the extent that transfer programs have appropriate work requirements for
those who are able to work, OZs complement such programs by fostering job
creation.
OZs also complement the Earned Income Tax Credit (EITC), which is an
antipoverty tax incentive. The EITC targets low-income workers, especially
those with children, and is phased out as a family’s income rises. Because the
EITC is only provided to low-income families with earnings, it encourages people to enter the workforce. Empirical research confirms that the EITC increases
workforce participation for single mothers, who benefit the most from the
credit (Nichols and Rothstein 2015). In this sense, the EITC increases the supply
of labor, while OZs stimulate demand for it.

Federal Place-Based Policies: The New Markets Tax Credit
Program
The Federal program most comparable to Opportunity Zones is the New
Markets Tax Credit (NMTC), though OZs offer improvements over the NMTC
program. Both use tax incentives to encourage private investment in lowincome communities, but the total tax benefit available through the NMTC

150 |

Chapter 5

program is capped, limiting how much investment it can spur.3 In most years
since 2007, Congress has authorized the NMTC program to award tax credits
to support about $3.5 billion in place-based investments. On average, these
credits account for about half of total project costs, so the program supports
roughly $7 billion in investment annually. As of 2016, nearly 3,400 census tracts
have received NMTC program credits since the program’s inception in the early
2000s (Tax Policy Center 2020).
In addition to being smaller in scale than the OZ initiative, the NMTC
program has a top-down approach to distributing tax benefits. The U.S.
Department of the Treasury administers the NMTC program through its
Community Development Financial Institutions Fund (CDFI), which ultimately
selects what applicants can receive tax credits. Community development entities must first apply to the CDFI to be qualified for the program. Those that are
qualified then identify investment opportunities and submit applications to
compete for a limited pool of credits. In 2018, development entities requested
$14.8 billion in NMTC funds, but only $3.5 billion were available, and only about
a third of all applicants received funding (Lowry and Marples 2019).
Even for approved applicants, the NMTC program places greater restrictions on investors. Funds must remain invested and compliant with program
requirements for seven years or else forgo all their tax benefits (with interest
and penalties). With OZs, funds can liquidate one investment and roll the
proceeds into a new one without penalty, though standard taxes apply to any
capital gains. OZs are also flexible in other ways; investors can contribute funds
up to any size, and they can pool their funds with any number of other investors
(Vardell 2019; Bernstein and Hassett 2015).
Many of the participants in the NMTC program are large financial intermediaries equipped to navigate the CDFI’s application process and manage
compliance risk (Vardell 2019; Hula and Jordan 2018). To manage the risk,
most NMTC transactions use a complex leverage model that combines debt
and equity. According to Hula and Jordan (2018, 23), the model requires “a
team of accountants and attorneys” with relevant expertise to structure the
investment. By contrast, any investor with eligible capital gains can invest in
a Qualified Opportunity Fund. These funds, in turn, need only self-certify their
investments on their tax returns and follow the broad guidelines provided by
the Department of the Treasury’s regulations.4
Although the NMTC program is more prescriptive than OZs, it is more
flexible than the economic development grants given by the CDFI Fund. Harger,
Ross, and Stephens (2019) find that the tax credits—but not the grants—
increased the number of new businesses in low-income communities. They
attribute the difference in part to the greater flexibility of the tax credit relative
3 NMTCs are a limited allotment of tax credits that reduce investors’ Federal tax obligations. Tax
credits differ from tax deductions, which decrease the amount of income subject to being taxed.
4 The final regulations are available at www.irs.gov/pub/irs-drop/td-9889.pdf.

Assessing the Early Impact of Opportunity Zones | 151

to the grants. At the same time, the authors found that even the NMTC program
may not have had much effect on local employment.

Other Federal Place-Based Development Programs
Along with Opportunity Zones, in recent decades three other Federal programs
have also relied on tax policy to spur economic development in specific places:
empowerment zones (EZs), enterprise communities (ECs), and renewal communities (RCs). EZs and ECs date to 1993, while RCs were authorized in 2000.
These programs extended a mix of tax benefits and grants to businesses in
designated census tracts. These programs had a smaller geographic reach,
with many States having little or no participation in them. A key tax benefit
among these programs was an employment tax credit of up to $3,000 on the
wages paid to people who lived and worked in the designated tract. Other tax
benefits included increased limits for expensed deductions, tax-exempt bond
financing, and exemptions from certain capital gains taxes (CRS 2011). The EC
and RC programs have both ended, and only the tax benefits associated with
the EZ program continue. Early research on the effects of the programs showed
little evidence of success, but more recent studies have documented beneficial
effects on unemployment, wages, and poverty (CRS 2011; Ham et al. 2011;
Busso, Gregory, and Kline 2013).
The Federal Government also supports place-based economic development through grant programs, with the largest being the Community
Development Block Grant program. The U.S. Department of Housing and
Urban Development (HUD) administers the program and provides about $3 billion a year in block grants. The program’s structure makes rigorous evaluation
difficult, and few systematic evaluations have been done, especially in recent
years (Theodos, Stacy, and Ho 2017). HUD allocates funds using a formula
based on population, poverty, housing conditions, and other factors. State and
local government grantees have considerable discretion, within broad guidelines, on how to use the funding, such as that at least 70 percent of the funds
must be used to benefit low- and moderate-income persons. The flexibility of
the program is similar to OZs, but its design is very different in that it relies
solely on public funding and does not seek to incentivize private investment.
The Economic Development Administration (EDA) of the U.S. Department
of Commerce also administers grants for economic development. EDA’s 2019
appropriation was roughly $300 million, but the Coronavirus Aid, Relief, and
Economic Security Act (CARES Act) appropriated an additional $1.5 billion to
administer grants to States and communities adversely affected by the COVID19 pandemic. As with the HUD grants, few rigorous evaluations have been done
of EDA’s grants (Markusen and Glasmeier 2008).

152 |

Chapter 5

Characteristics of Opportunity Zones
The census tracts designated as OZs have some of the most entrenched poverty in the United States. These communities had an average median income
just over half of the U.S. average in 2000 and they fell further behind over the
subsequent 16 years.

The Opportunity Zone Selection Process
As prescribed by law, governors nominated which census tracts should be
designated as Opportunity Zones by the U.S. Department of the Treasury. To
be eligible for designation, a census tract must:
• Have a poverty rate of at least 20 percent; or
• Have a median income below 80 percent of that in the State or metropolitan area, or for rural census tracts, 80 percent of that in the entire State; or
• Be contiguous with a census tract meeting one of the above conditions
and have a median income less than 125 percent of the qualifying contiguous
census tract.
Governors could designate up to 25 percent of their qualifying census
tracts, or up to 25 tracts for those States with fewer than 100 eligible tracts.
Eligible, contiguous tracts were restricted to make up no more than 5 percent
of designated OZs in any State.
Aside from these restrictions, States could determine how, and which,
census tracts would be designated as OZs, thereby drawing on State and local

Figure 5-1. The Geography of Opportunity Zones

Sources: U.S. Department of the Treasury; U.S. Census Bureau.

Assessing the Early Impact of Opportunity Zones

| 153

expertise. With this Federal design, States took diverse approaches in nominating their OZs. Arizona, for example, tasked the Arizona Commerce Authority
with meeting with city, county, and tribal governments to select tracts. Kansas
took a different approach, with its Department of Commerce requesting
“Letters of Interest” from communities seeking OZ designation, allowing communities to explain their need and their ability to attract investment.
All governors submitted tracts for consideration to the U.S. Department
of the Treasury by the end of April 2018. The Treasury ultimately designated
a total of 8,766 tracts as OZs, with nearly all designations occurring between
April and June 2018. Almost all OZs (8,537 tracts) met one of the criteria for
low-income communities; the remaining 229, or 2.6 percent of all designated
census tracts, were eligible for selection based on contiguity with a low-income
tract. Figure 5-1 highlights the OZ tracts (in green) and the eligible tracts that
were not selected (in gray).

The Economic State of Opportunity Zones
This subsection reports on the CEA’s overall findings that census tracts
selected as Opportunity Zones are among the poorest communities in the
United States. The CEA finds that they have an average poverty rate more than
double that of all other census tracts and are home to a higher share of African
Americans, Hispanics, and high school dropouts (figure 5-2).
The economic woes of OZs are not new. In 2000, census tracts that later
became OZs had an average median household income that was 57 percent of
the average in other tracts, $39,305 compared with $68,726 as given in the 2000
Decennial Census. In real terms, median household income in the average OZ
fell by 11 percent from 2000 to 2012–16, compared with a 6 percent drop in the
average non-OZ census tract (figure 5-3).
The poverty and income criteria for eligibility explain some of the lower
income in selected census tracts; but even among eligible tracts, States consistently nominated low-income tracts. In each of the 50 States and in the District
of Columbia, median household income in OZs was lower than in eligible-butnot-selected tracts and considerably lower than in ineligible tracts (figure 5-4).
Figures 5-2 through 5-4 indicate that, as a whole, OZs encompass economically distressed areas. Although average values can mask diversity within
the OZ group, only 3.2 percent of OZs experienced rapid socioeconomic change
according to a metric developed by the Urban Institute (2018). This metric considers changes in income, demographics, educational attainment, and housing
affordability.
The patterns shown in figure 5-5 suggest that States selected tracts that
were both economically distressed and demonstrated a potential to attract
fruitful investments. They selected tracts with varying levels of poverty, not
focusing solely on those with the least poverty (among eligible tracts) nor on
those with the highest poverty rates. The strategy has an economic rationale:
154 |

Chapter 5

Figure 5-2. Demographics of Opportunity Zones (OZs), 2012–16
Percent
35
30

Share of OZ population

28.9%
25.4%

24.3%

25
20
15

Share of non-OZ population

13.8%

16.4%
11.4%

14.1%
8.1%

10
5
0
People living in
poverty

Hispanics

African
Americans

Adults without a
high school
diploma

Sources: 2016 American Community Survey (ACS) five-year estimates; CEA calculations.
Note: This analysis excludes census tracts in Puerto Rico, American Samoa, the U.S. Virgin
Islands, Guam, and the Northern Mariana Islands. In addition, the 2016 ACS is based on a
five-year estimate from 2012 to 2016.

Figure 5-3. Average Median Household Income by Census Tract
Designation, 2000–2016
Household income index (2000 = 100)
102
100
98
96

Non-OZ tracts
(6% decrease)

94
92
90
88
86
84
82
2000

OZ tracts
(11% decrease)

2016
2012

Sources: 2000 Decennial Census; 2016 American Community Survey five-year estimates; U.S.
Department of the Treasury; CEA calculations.
Note: This analysis excludes census tracts in Puerto Rico, American Samoa, the U.S. Virgin
Islands, Guam, and the Northern Mariana Islands. The 2016 ACS is based on a five-year
estimate from 2012 to 2016.

Assessing the Early Impact of Opportunity Zones

| 155

Figure 5-4. Average Median Household Income by Tract
Designation and State, 2012–16
OZ tracts

Eligible tracts, but not selected

Ineligible tracts

United States
Georgia
Ohio
Nevada
Alabama
Kentucky
Illinois
Tennessee
Louisiana
Pennsylvania
Florida
Missouri
Arkansas
Mississippi
South Carolina
Michigan
Indiana
North Carolina
Wisconsin
Oklahoma
Arizona
Nebraska
Montana
West Virginia
Rhode Island
California
Texas
Kansas
Iowa
New Mexico
Washington, D.C.
Connecticut
Idaho
Maine
New York
South Dakota
Minnesota
Oregon
North Dakota
Massachusetts
Washington
New Jersey
Delaware
Vermont
Colorado
Utah
Virginia
Wyoming
New Hampshire
Maryland
Hawaii
Alaska

25,000

45,000

65,000

85,000

105,000

Median household income
Sources: 2012–16 American Community Survey (ACS), five-year estimates; U.S. Department
of the Treasury ; CEA calculations.
Note: This analysis excludes census tracts in Puerto Rico, American Samoa, the U.S. Virgin
Islands, Guam, and the Northern Mariana Islands. The 2016 ACS is based on a five-year
estimate from 2012 to 2016. Eligible but not selected tracts include those eligible based on
low-income status or on contiguity with low-income tracts.

156 | Chapter 5

Figure 5-5. Population by Poverty Rates and Census Tract
Designation
Opportunity Zones

Eligible tracts, but not selected

Ineligible tracts

Population (in millions)
18
16
14
12
10
8
6
4
2
0
0

20
25
30
35
40
Poverty rate (percent)

Sources: 2016 American Community Survey, five-year estimates; U.S. Department of the
Treasury; CEA calculations.

States would benefit little from OZs if they selected tracts where a designation
was unlikely to spur investment.

Opportunity Zones’ Effect on Total Investment
The CEA estimates that by the end of 2019, Qualified Opportunity Funds had
raised $75 billion in private capital. Although some of this capital may have
occurred without the incentive, the CEA estimates that $52 billion—or 70 percent—of the $75 billion is new investment.

Capital Raised by Qualified Opportunity Funds
The $75 billion estimate for private capital raised is based on two different
samples that track these funds over time. To extrapolate from sample values to
population values, we rely on the total number of these funds in existence, as
estimated by the Department of the Treasury based on tax filings (1,500 funds
in 2018).5 Both samples and estimation approaches give a roughly similar estimate for the capital raised by these funds, with the average being $75 billion.

5 The count of Qualified Opportunity Funds in the population (1,500) is based on a Treasury
Department’s estimate based on preliminary counts of filings of Form 8996. The Treasury may
adjust this count as more information becomes available.

Assessing the Early Impact of Opportunity Zones | 157

The first sample covers Qualified Opportunity Funds voluntarily reporting data to Novogradac, a national professional services organization that
has tracked funds since May 2019. As of January 17, 2020, the sample had
513 of these funds, a small subset of all funds, which had collectively raised
$7.6 billion in capital.6 Qualified Opportunity Funds voluntarily reporting data
might not be representative of the general population of funds. However, comparisons with a non-voluntary sample, as discussed below, suggests that it is
reasonably representative.
The second sample is based on data from the Securities and Exchange
Commission (SEC). The SEC considers investment interests in Qualified
Opportunity Funds as securities, which means that funds must register with
the SEC unless they file for an exemption. Qualified Opportunity Funds seeking
an exemption can file Form D within 15 days of the first sale of securities in an
offering. In filing Form D, these funds provide information such as the amount
sold in the offering, but they are not asked to identify themselves as funds. To
create a sample of these funds from the Form D data, we select all funds with
“Opportunity Zone” or similar words (e.g., “OZ Fund” or “QOZF”) in their name.
This yields 197 Qualified Opportunity Funds that had filed Form D by the close
of 2019, 153 of which had raised capital, totaling about $2.9 billion. If Qualified
Opportunity Fund names are uncorrelated with other fund characteristics,
our sample should be reasonably representative of the broader population of
funds seeking an exemption from SEC registration.7
The Novogradac and SEC samples show similar growth in the number
of Qualified Opportunity Funds and capital raised. From May 2019, when
Novogradac began tracking these funds, until Novogradac’s January 17, 2020,
report, the number of funds increased by 277 percent. The SEC data show a
271 percent increase in the number of these funds from 2018 to 2019, based
on information on when each fund was incorporated. Additionally, the capital
reported by Novogradac Qualified Opportunity Funds increased by 858 percent
over the reporting period, while the capital raised by the SEC sample of funds
increased by 1,523 percent from 2018 to 2019. See figure 5-6.
The two samples of Qualified Opportunity Funds inform two different
approaches for estimating the total capital raised by funds. The first approach,
based on the self-reported Novogradac data, is to multiply the Novogradac
total equity amount ($7.6 billion) by an expansion factor, defined as the number of Qualified Opportunity Funds in the population divided by the number of
funds in the Novogradac database. This factor reflects how much of the fund
6 Although our analysis is for the close of 2019, more recent data from Novogradac show a 31
percent increase in capital raised from January to April 2020 and a roughly 20 percent increase
for April to August 2020, which represents a 79 percent growth over the first eight months of 2020.
7 Funds seeking to make public offerings of securities are generally not exempt from SEC
registration and would not file a Form D. We expect such funds to be larger, on average, than those
focused on private offerings.

158 | Chapter 5

subset of all funds, which had collectively raised $7.6 billion in capital. Qualified
Opportunity Funds voluntarily reporting data might not be representative of the
general population of funds. However, comparisons with a non-voluntary sample, as
discussed below, suggests that it is reasonably representative.
The5-6.
second
sample
is based Opportunity
on data fromFunds,
the Securities
and Exchange
Figure
Growth
in Qualified
Novogradac
Commission (SEC). The SEC considers investment interests in Qualified Opportunity
and SEC Data
Funds as securities, which means that funds must register with the SEC unless they file
(percent) Qualified Opportunity Funds seeking an exemption can file Form D
forGrowth
an exemption.
1,600
within
15 days of the first sale of1,523
securities in an offering. In filing Form D, these funds
provide information such as the amount sold in the offering, but they are not asked to
1,400 themselves as funds. To create a sample of these funds from the Form D data,
identify
we1,200
select all funds with “Opportunity Zone” or similar words (e.g., “OZ Fund” or
“QOZF”) in their name. This yields 197 Qualified Opportunity Funds that had filed
Form
D by the close of 2019, 153 of which had raised capital, totaling about $2.9 billion.
1,000
858
If Qualified Opportunity Fund names are uncorrelated with other fund characteristics,
800
our sample
should be reasonably representative of the broader population of funds
seeking an exemption from SEC registration.7
600
The Novogradac and SEC samples show similar growth in the number of
Qualified
when Novogradac
400 Opportunity Funds and capital raised. From
277May 2019,271
began tracking these funds, until Novogradac’s January 17, 2020, report, the number
200 increased by 277 percent. The SEC data show a 271 percent increase in the
of funds
number of these funds from 2018 to 2019, based on information on when each fund
0
was incorporated. Additionally, the capital reported by Novogradac Qualified
Capital raised
Number of funds
Opportunity Funds increased by 858 percent over the reporting period, while the
Novogradac
data
SEC
data
(2018–19)
capital raised by the SEC sample of funds increased by
1,523
percent from 2018 to
(May
2019–January
2020)
2019. See figure 5-6.
TheNovogradac;
two samples
of Qualified
Opportunity
Funds
inform
two different
Sources:
Securities
and Exchange
Commission
(SEC);
CEA calculations.
approaches for estimating the total capital raised by funds. The first approach, based
on the self-reported Novogradac data, is to multiply the Novogradac total equity
amount ($7.6
by an
factor,database.
defined asThe
the number
Qualified
population
is billion)
captured
by expansion
Novogradac’s
estimateof of
capital
Opportunity Funds in the population divided by the number of funds in the
raised is then:
Novogradac database. This factor reflects how much of the fund population is
captured by Novogradac’s database. The estimate of capital raised is then:
𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑜𝑜𝑜𝑜 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓
𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 (𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁. ) = 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅!"#". 𝑥𝑥 4
:
𝑁𝑁𝑁𝑁𝑁𝑁𝑁𝑁. 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑜𝑜𝑜𝑜 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓
1,500 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓
= 7.6 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 𝑥𝑥 4
:
136 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓
= $84 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏

Theour
number
Opportunity
(1,500)
in the
Although
analysis isofforQualified
the close of 2019,
more recentFunds
data from
Novogradac
showpopulation
a 31 percent
increasefrom
in capital
from January
AprilTreasury
2020 and a and
roughly
20 percent increase
for end
April to
comes
theraised
Department
oftothe
corresponds
to the
of
August 2020, which represents a 79 percent growth over the first eight months of 2020.
2018,
and
the
number
of
funds
in
the
Novogradac
database
(136)
is
from
May
7
Funds seeking to make public offerings of securities are generally not exempt from SEC registration
and would
file a Form
D. We expect
such
funds to be larger,
onThis
average,
than those approach
focused on
2019,
the not
earliest
reporting
of the
Novogradac
data.
estimation
private offerings.
assumes that Qualified Opportunity Funds reporting to Novogradac are
similar in size to funds not reporting to Novogradac. It also assumes that our
expansion factor accurately reflects Novogradac’s coverage of the Qualified
Opportunity Fund population in January 2020.
The second estimation approach, which draws on the SEC sample, multiplies an estimate of the number of Qualified Opportunity Funds in existence
6

Assessing the Early Impact of Opportunity Zones | 159

Opportunity Funds reporting to Novogradac are similar in size to funds not reporting
to Novogradac. It also assumes that our expansion factor accurately reflects
Novogradac’s coverage of the Qualified Opportunity Fund population in January 2020.
The second estimation approach, which draws on the SEC sample, multiplies
anatestimate
of the
number
Fundsamount
in existence
at the close
of per
the close
of 2019
by of
anQualified
estimateOpportunity
of the average
of capital
raised
2019
by among
an estimate
of having
the average
amount
of capital
raised per fund,
fund,
those
raised
capital.
More specifically,
it is:among those
having raised capital. More specifically, it is:
𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 (𝑆𝑆𝑆𝑆𝑆𝑆)
= "𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑜𝑜𝑜𝑜 𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹"!"#$ 𝑥𝑥 "𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺ℎ 𝑖𝑖𝑖𝑖 𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐"!"#$%#&
𝑥𝑥 "𝑆𝑆ℎ𝑎𝑎𝑎𝑎𝑎𝑎 𝑜𝑜𝑜𝑜 𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 𝑤𝑤𝑤𝑤𝑤𝑤ℎ 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝑙𝑙"!"#& 𝑥𝑥 "𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝑝𝑝𝑝𝑝𝑝𝑝 𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹"!"#&
= 1,500 𝑥𝑥 3.71 𝑥𝑥 0.60 𝑥𝑥 0.019
= $63 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏

The population count of Qualified Opportunity Funds is again from the
Department of the Treasury, the growth in the fund count is based on the 2018 to 2019
The
countincorporated
of Qualified(asOpportunity
Funds
is again
growth in
thepopulation
number of funds
reported in the
SEC data);
the from
share the
of theis Treasury,
the2020
growth
in the from
fundthe
count
is baseddatabase;
on the 2018
ofDepartment
funds with capital
as of January
and comes
Novogradac
and
per fundin
comes
from the SEC
data (0.019
billion per(as
fund).
For the share
to capital
2019 growth
the number
of funds
incorporated
reported
in theofSEC
Qualified
Opportunity
Funds
with
capital
(0.60),
we
use
the
Novogradac
data
data); the share of funds with capital is as of January 2020 and comesinstead
from the
of the SEC data, which primarily cover funds that have already raised capital since that
Novogradac database; and capital per fund comes from the SEC data (0.019 bilis what triggers their filing of the SEC form that generates the data. As such, funds that
lion per fund). For the share of Qualified Opportunity Funds with capital (0.60),
have raised at least some capital are likely to be overrepresented in the SEC data. In
we use the
Novogradac
data
instead
the SECare
data,
which
cover
summary,
the key
assumptions
of the
secondofapproach
that the
SECprimarily
data provide
haveofalready
raised
capital
since
that is what
triggersFunds
theirinfiling
a funds
reliablethat
estimate
the growth
in the
number
of Qualified
Opportunity
the of
population
and, among
those withthe
capital,
their
capitalthat
raised.
In line
withatthe
the SEC form
that generates
data.
As average
such, funds
have
raised
least
Novogradac
data,
the
approach
also
assumes
that
60
percent
of
all
funds
raised
some
some capital are likely to be overrepresented in the SEC data. In summary, the
capital
by the close ofof2019.
key assumptions
the second approach are that the SEC data provide a reliThe standard error of the average amount of capital raised per Qualified
able estimate of the growth in the number of Qualified Opportunity Funds in
Opportunity Fund permits providing a confidence interval around the SEC-based
the population
among
those
with capital,
theirconfidence
average capital
8
The resulting
90 percent
intervalraised.
is $33 In
estimate
of the totaland,
capital
raised.
line with
Novogradac
data,
the at
approach
also
assumes
thatincludes
60 percent
billion
at thethe
lower
end and $93
billion
the higher
end.
It therefore
the of
all funds raised some
capital
byaverage
the close
of 2019.
Novogradac-based
estimate
and the
of the
two estimates, which is about $75
billion and
our preferred
estimate.
This is 21 amount
percent ofofbaseline
Theis standard
error
of the average
capitalannual
raisedinvestment
per Qualified
inOpportunity
OZs, which is reported
in the next
subsection.
Fund permits
providing
a confidence interval around the SEC-

based estimate of the total capital raised.8 The resulting 90 percent confidence
interval is $33 billion at the lower end and $93 billion at the higher end. It
therefore includes the Novogradac-based estimate and the average of the two
8 estimates, which is about $75 billion and is our preferred estimate. This is 21
The resulting confidence interval reflects uncertainty over the population value of capital per fund. It
percent
of baseline
in inOZs,
which isofreported
thebynext
does
not capture
uncertaintyannual
over otherinvestment
parameters used
the calculation
total capitalin
raised
funds
in the population.
subsection.

Estimated Investment Growth Caused by the Opportunity Zone
Incentive
Not all the capital raised by Qualified Opportunity Funds is necessarily new to
Opportunity Zones—some of it may have occurred without the incentive, and
it is now occurring through a fund. In this subsection, the CEA draws from the
8 The resulting confidence interval reflects uncertainty over the population value of capital per
fund. It does not capture uncertainty over other parameters used in the calculation of total capital
raised by funds in the population.

160 |

Chapter 5

academic literature to estimate how much new investment is likely given the
lower tax rates caused by the OZ incentive. We estimate that the incentives
have brought $52 billion in new investment in OZs through 2019, representing
70 percent of the $75 billion raised by Qualified Opportunity Funds.
To estimate new investment, we calculate the reduction in the cost of
capital caused by the cuts to capital gains tax rates. We then link the cost of
capital to investment elasticities from the academic literature. This modeling
of the OZ incentive illustrates how the incentive is similar to the corporate
tax rate cuts resulting from the Tax Cuts and Jobs Act. These cuts were also
projected to increase investment through a decline in the user cost of capital
(CEA 2017).
The investment estimates come from first calculating the pretax rate
of return needed to attract investors to supply funds in OZs. To achieve the
same post-tax return inside OZs as outside them, investors would be willing
to accept a lower pretax return because of lower effective tax rates in OZs. The
second step of the estimation then calculates the increased investment from
OZ businesses that occurs as they have access to new funding at a lower capital
cost. Figure 5-7 illustrates the concepts behind the calculation, showing how
the reduction in taxes makes investors willing to accept a lower pretax rate of
return and still invest in OZs.
The numerical estimates rely on three parameters: baseline investment
in OZ census tracts that predates the incentives, the post-tax rate of return
that is required to attract funds, and the effective tax rate that prevails in OZs
with the incentive. For the first parameter, we estimate baseline investment of
$243 billion by apportioning national investment to counties based on gross
domestic product, and then from counties to census tracts based on income
and population. Second, using data that show a pretax 9.8 percent rate of
return earned by investors outside OZs—which then face a capital gains tax
rate of 21.3 percent—the required post-tax rate of return is 7.7 percent. We find
that, to receive the same post-tax 7.7 percent rate of return in OZs—which feature only a 6.9 percent effective tax rate, as described below in the “Budgetary
Effects of Opportunity Zones” subsection—investors only require a pretax rate
of return equal to 8.3 percent (= 7.7/(1 – 0.069)) in 2019. Finally, we assume a
–9.55 semielasticity of investment to the cost of capital, from Ohrn (2019). Over
a one-and-a-half-year period, the increase to investment is then calculated as:
1.5 years x ($243 billion) x (8.3% – 9.8%) x (–9.55) = $52.2 billion.

The one-and-a-half-year period is used to reflect the time between the designation of Opportunity Zones (mid-2018) and the end of 2019.

Assessing the Early Impact of Opportunity Zones | 161

Figure 5-7. Opportunity Zone (OZ) Investment
Supply-and-Demand Model
Rate of return

Investment demand

Pretax required rate of return
(without OZs)

8
7

Pretax required rate
of return (with OZs)

Normal
tax 5
wedge

Posttax required
rate of return

OZ tax
wedge

3
2
1
0
Δ Investment

Investment

The Industry Focus of Qualified Opportunity Funds
Recent data from the Securities and Exchange Commission allow us to describe
the sectoral focus of a sample of Qualified Opportunity Funds, the same one
described above. The SEC form completed by Qualified Opportunity Funds
requires them to select one industry group. The selections, shown in figure 5-8,
reveal the diverse focus of funds. Slightly less than half of them focus on real
estate, with the majority targeting commercial real estate.9 Another 45 percent
describe their industry as a “Pooled Investment Fund,” which suggests that
they have investments across various industries. Finally, about 10 percent are
in the “other” category, which includes funds that reported a focus on health
care, technology, construction, and investing, and as well as those selecting
the “other” option on the form.
The industry focus indicated by the SEC data are consistent with the
types of projects seeking to attract Qualified Opportunity Fund investment, as
evidenced by data from the Opportunity Exchange, which is a private organization that helps entities showcase OZ businesses and properties to stakeholders
locally and nationally. As of February 2020, The Opportunity Exchange hosted
$45 billion in proposed projects across 24 States. About 30 percent of the projects on the Opportunity Exchange are businesses seeking equity investments,
9 Form D does not provide definitions for the industry categories that filers can select.

162 |

Chapter 5

Figure 5-8. Percentage of Qualified Opportunity Funds, by
Industry

Other
Commercial
9%
real estate
18%
Residential real estate
Pooled
6%
investment
fund
Other real
45%
estate
22%

Sources: Securities and Exchange Commission; CEA calculations.
Note: "Other real estate" includes real estate inestment trusts and finance. "Other" includes
healthcare, technology, construction, and investing.

26 percent are real estate projects with a development plan, and the rest are
properties for sale without a development plan.

Opportunity Zones’ Effects on Business
Investment and Housing Values
The CEA finds that receiving an OZ designation led to a 29 percent relative
increase in equity investment. Such communities have also benefited from
larger house price appreciation, which creates $11 billion in additional housing
wealth for homeowners and improved local amenities for renters.

Equity Investments in Opportunity Zone Businesses
Qualified Opportunity Funds can invest in Opportunity Zones by directly
purchasing property or by making equity investments in operating businesses.
In this subsection, we present data regarding private equity investment in
businesses located in OZs compared with those located elsewhere. Investment
data from the Securities and Exchange Commission show that OZ designation
led to a 29 percent increase in equity investments in businesses whose principal place of business is in an OZ, compared with businesses in eligible-but-notselected census tracts.

Assessing the Early Impact of Opportunity Zones | 163

Many businesses pursuing equity investments must file the same SEC
Form D that Qualified Opportunity Funds file. We use address information
from this form, which gives the location of the principal place of business, to
determine whether the business is located in an OZ census tract, an eligiblebut-not-selected tract, or an ineligible tract. To capture nonfinancial operating
businesses, we exclude entities that identified themselves as banks or investment funds.10 To better measure systematic investment trends, as opposed
to variation in the behavior of a few large firms, we focus on filings that raised
less than $50 million in any quarter, which captures more than 96 percent of
filings.11 We then compile the total investment raised by businesses in each
census tract type by quarter.
In figure 5-9, we present the four-quarter moving average of the total
equity investment in each group of tracts, with values indexed to their value
in the first quarter of 2018. The three groups had similar investment trends
until the first half of 2018, when investment in OZ businesses spiked.12 All
States nominated census tracts in March and April 2018, and the Department
of the Treasury finalized its formal designation of OZs by the second quarter
of 2018. Over the seven quarters 2018:Q2–2019:Q4, equity investment in OZs
was 41 percent higher than it was in the prior seven quarters. By comparison,
investment was only 13 percent higher in eligible-but-not-selected tracts. This
suggests that OZ designation led to a 29 percent increase in equity investment
relative to comparable tracts (41.4–12.6 percent).13

Opportunity Zone Designation and Housing Values
Evidence from real estate markets suggests that the Opportunity Zones incentive is making many OZs more attractive for both residents and investors. This
increase in housing value has led to an estimated $11 billion in additional
wealth for the nearly half (47 percent) of OZ residents who own their housing.
Real Capital Analytics tracks commercial real estate properties and portfolios valued at $2.5 million or more. Its data show that year-over-year growth
10 Specifically, we exclude all firms that identified their industry or their fund as “pooled
investment fund,” “commercial banking,” “investment banking,” “other banking and financial
services,” or “investing.”
11 Bauguess, Gullapalli, and Ivanov (2018) report that more than 96 percent of filings have an
offering size of $50 million or less. An even larger percentage would actually raise less than $50
million.
12 Not every businesses in an OZ is necessarily a Qualified Opportunity Zone Business as defined
by statute and regulation.
13 The location of a business in a particular OZ does not mean that the business’s activities
must be concentrated in that particular OZ. A business can achieve the status of a Qualified
Opportunity Zone Business if 50 percent of its gross income is derived from its business activities
in any OZ. Thus, a business could have multiple income-earning centers spread across various
OZs. Alternatively, the business can qualify if at least 50 of the services purchased and used by
the business (measured by hours or dollars) occur in OZs or if at least 50 percent of its tangible
property and management functions are in OZs.

164 |

Chapter 5

Figure 5-9. Private Equity Investment by Tract Group, 2016–19
Opportunity Zones
Eligible tracts, but not selected
Ineligible tracts
Index (1 = 2018:Q1)
1.5

TCJA signed into law
2019:Q4

1.4
1.3
1.2
1.1
1.0
0.9
0.8

Opportunity Zones
nominated and
designated

0.7
2016:Q1 2016:Q3 2017:Q1 2017:Q3 2018:Q1 2018:Q3 2019:Q1 2019:Q3
Sources: Securities and Exchange Commission; U.S. Department of the Treasury; CEA
calculations.
Note: TCJA = Tax Cuts and Jobs Act.

in development site acquisitions surged in OZs by more than 50 percent late
in 2018 after the Department of the Treasury had designated the OZs, greatly
exceeding growth in the rest of the United States. Similarly, Sage, Langen, and
Van de Minne (2019) use the same data and find that OZ designation led to a
14 percent increase in the price of redevelopment properties and a 20 percent
increase in the price of vacant development sites as of early 2019.
Sage, Langen, and Van de Minne (2019) find a price increase only for
particular property types and conclude that the OZ incentive is having limited
economic spillovers in communities. Their data, however, only include commercial properties valued at $2.5 million or more. An analysis by Zillow, which
was based on transactions of varying property types and values, suggests that
the OZ incentive is having broader effects. After designation, the year-over-year
change in the average sales price for properties in OZs rose to more than 25
percent while falling to below 10 percent in eligible-but-not-selected census
tracts.
The Zillow analysis is limited in that it is based on changes in sales prices
over time, without controlling for any changes in the composition of properties
being sold. It is not based on price per square foot or, more ideally, on price
changes for homes that are similar in many other dimensions. Chen, Glaeser,
and Wessel (2019) provide a more rigorous assessment of effects on housing
Assessing the Early Impact of Opportunity Zones | 165

prices, though only through 2018. For a measure of housing prices, they use
the Federal Housing Finance Agency (FHFA) repeat sales index for single-family
homes. Their analysis centers on comparing OZs with eligible but not selected
low-income tracts (thus excluding tracts whose eligibility was based solely on
contiguity with low-income tracts). Across the two groups, they compare the
growth in housing values in 2018 relative to that of prior years (2014–17). Their
estimated effects are much smaller than those suggested by the Zillow analysis: their base model gives an estimate of 0.25 percent higher appreciation,
with the estimates across models ranging from 0.09 to 0.74.
We replicate and extend the analysis done by Chen, Glaeser, and Wessel.
First, we replicate the results from their base model and find a similar result
(table 5-1, first and second columns). Then we reestimate the model with
updated FHFA data released in May 2020. The update improves data from prior
years and adds 2019 data.14
With the updated and expanded data, we estimate that OZ designation
led to a higher annual appreciation of 0.53 percent. Over two years, this implies
a roughly 1.1 percent (= 1.0053^2 – 1) increase in values. This is a notable finding because it is based on OZ designation, not on whether a tract has actually
received investment. Moreover, much of the investment raised by Qualified
Opportunity Funds was probably not invested by the end of 2019. By comparison, Freedman (2012) looked at census tracts that had actually received
investment through the New Markets Tax Credit and failed to find a statistically
significant effect of investment on housing values over about five years, with
the point estimate implying an annual effect of at most 0.5 percent.
The extra 1.1 percent appreciation implies $11 billion in additional
wealth for the nearly half (47 percent) of OZ residents who own their housing. Homeowners can access newly found equity without selling their homes
through cash-out refinancing, which has been common in the last two years.
This does not mean that rising values only benefit homeowners. The causes
of higher values—more local amenities and anticipated economic opportunities—will benefit many renters as well. The renovation of a blighted building,
for example, benefits all who live nearby. Brummet and Reed (2019) draw a
similar conclusion from a thorough analysis of Census microdata, finding that
less exposure to poverty and rising values tend to benefit original residents and
led to better outcomes for their children. Using a different data source from
Medicaid records, Dragan, Ellen, and Glied (2019) draw a similar conclusion
14 The data are available at www.fhfa.gov/DataTools/Downloads/Pages/House-Price-IndexDatasets.aspx; see “Census Tracts (Development Index; Not Seasonally Adjusted).” We also
normalize the housing price index to make 2013 the base year (= 100). The renormalization ensures
that that changes in the index are approximate percentage changes, with a 1-point change in
the index corresponding to a 1 percent increase in values. If index values are about 300, which
is typical in the original index, a 1-point increase represents a 0.3 percent increase in values. The
renormalized values are also much less skewed than the original index values.

166 |

Chapter 5

Table 5-1. The Effect of Opportunity Zone
Designation on Home Value Appreciation

Characteristic

Chen et al.
(2019)

CEA Estimates
Chen et al.
Updated
Data
Data

Opportunity Zone
effect on housing
values (percent)

0.25

0.53

Standard error

0.22

0.19

Number of
Opportunity Zones

2,674

2,700

10,198

10,288

Number of eligible
zones that were not
selected

Sources: Chen, Glaeser, and Wessel (2019); Census Bureau, American Community
Survey, 2012–16; Federal Housing Finance Agency; U.S. Department of the Treasury;
CEA calculations.
Note: The estimated effect is based on comparing Opportunity Zones with low-income
tracts that were eligible but not selected .

about the effects of rising housing values and neighborhood improvement on
residents and their children.
Within Opportunity Zones, the distribution of the benefits from improved
amenities is unclear. In some instances, the benefits may go primarily to lowincome households. For example, Gamper-Rabindran and Timmins (2013)
find that cheaper homes benefit the most from the cleanup of hazardous
waste sites because such homes tend to be closer to such sites. In the same
vein, the renovation of an abandoned warehouse would mostly benefit the
residents in the immediate vicinity, who may also be among the poorest in the
neighborhood.
Residents who rent their housing will generally benefit from improved
amenities as long as the full value of the amenities enjoyed by residents is not
passed on in the form of higher rents. Improved neighborhood conditions do
not always result in rent increases for all renters (Brummet and Reed 2019),
and sometimes improved amenities increase housing values more than they
increase rents (e.g., Granger 2012).

Assessing the Early Impact of Opportunity Zones | 167

Opportunity Zones’ Effects on
Poverty and the Budget
The CEA’s estimate of new investment suggests that Opportunity Zones
may lift about 1 million people out of poverty, an 11 percent decrease in the
baseline population in poverty in OZs. This decline in poverty, and with it a
reduction in transfer payments, may be sufficient to make the OZ incentive
nearly revenue neutral.

Projected Effects of Opportunity Zones on Poverty
Census-tract-level data on poverty for 2019 will not be available for several
years. The CEA therefore projects the effects on poverty using a prior study
linking investment to poverty. Freedman (2012) uses tract-level data to estimate the effects of investment subsidized by the New Markets Tax Credit on
tract-level outcomes. His empirical approach exploits the program’s eligibility
cutoffs to address the potential that subsidized investment went to tracts that
would have performed better even without the subsidy. His most conservative
estimate indicates that each $1 million in subsidized investment (in 2018 dollars) lifts 20 people out of poverty in the tract receiving it. Applying this finding
to our estimate of new investment in Opportunity Zones ($52,000 million) suggests that 1 million people will be lifted out of poverty (= 52,000 x 20).
This effect is arguably applicable to OZ investment. The NMTC program
has similar eligibility requirements for census tracts and rules to ensure that
the subsidized investment happens in qualified tracts. The main difference is
that community development entities must apply to and be selected by the
Treasury Department, which only selects a portion of applicants. The Treasury
scores applications using several criteria, including the expected effect of the
project on jobs and economic growth in the community. It is possible that
applicant reporting and Treasury selections result in the investments having
larger effects on poverty. Conversely, the long-term net effects of a particular
project on low-income populations is arguably hard to discern with consistency. In any case, our poverty projections are arguably conservative; we use
the smallest estimated effect from Freedman (2012), which is about half the
main estimate reported, and apply it to new investment as opposed to all subsidized investment, which is the basis of Freedman’s estimate.

Budgetary Effects of Opportunity Zones
The CEA estimates that the Federal Government forgoes $0.15 for every $1 in
capital gains invested in a Qualified Opportunity Fund before 2020, or about
$11.2 billion for the $75 billion raised through the end of 2019. The forgone
revenues stem from the deferment on the capital gains tax on the original gain,
the reduction in taxes on the original gains when paid, and the lack of taxes
on the gains earned while invested in the Qualified Opportunity Fund. In our
168 |

Chapter 5

calculation, we assume that taxpayers maximize their tax savings by waiting
until 2026 to pay taxes on the original gains, the latest date allowed by law,
and that they keep their money in the Qualified Opportunity Fund for at least
10 years.
Our calculations assume that capital gains would normally be taxed
at a 21.3 percent rate, as opposed to an effective rate of 6.9 percent in 2019.
This lower effective rate arises from the tax deferral and step-up in basis on
funds that are invested in OZs to begin with, as well as the exclusion of capital
gains taxes on the returns that accrue to those investments after they are held
for at least 10 years. For funds invested in 2019, the present values of taxes
paid on investments in an OZ are less than one-third what they would be if
invested outside an OZ. These calculations are then repeated for each year
to incorporate the dynamic nature of the OZ tax incentives, as discussed in a
Congressional Research Service report (Lowry and Marples 2019).
When estimating overall revenue effects, any static calculation that uses
only the difference in rates while assuming a fixed tax base gives an inflated
measure of tax revenue losses. Therefore, in our approach, we incorporate the
response of investment—and hence the tax base—to the incentive. Specifically,
we estimate how much of the observed $75 billion would have occurred anyway—whether in an OZ or elsewhere in the country—versus how much is new
investment. Investment that would have occurred anyway and been taxed at a
21.3 percent rate but that is now taxed at a lower rate because of the incentive
unambiguously lowers revenues. However, new investment creates offsetting
revenue gains, even when taxed at the lower OZ rate.
We employ a similar elasticity-based approach as in the investment section of this chapter. The approach suggests that of the $75 billion in Qualified
Opportunity Fund capital, $22.8 billion would have occurred anyway in OZs,
even without the incentive. Of the $52.2 billion balance, another $24.9 billion is new to OZs but was shifted from elsewhere in the country, based on
calculations using the elasticity-of-investment movement done by Koby and
Wolf (2019). Thus, the incentive results in revenue losses from this $47.7 billion
($22.8 billion + $24.9 billion) but creates revenue gains from the entirely new
$27.3 billion ($75 billion – $47.7 billion) in investment. On net, we estimate
the present value of tax revenue losses on capital invested through 2019 to
be $11.2 billion, which is 15 percent of the $75 billion in Qualified Opportunity
Fund capital.
By comparison, the CEA estimates that for each $1 in investment associated with the New Markets Tax Credit, the Federal government forgoes $0.18,
more than the amount for OZs. Based on estimates from the Joint Committee
on Taxation, the lost tax revenue for each $1 in tax credit authority is $0.26.15
15 In December 2019, the Joint Committee on Taxation estimated the dynamic
revenue effects from a $5 billion allocation for the NMTC (see the relevant line at
www.jct.gov/publications.html?func=startdown&id=5237).

Assessing the Early Impact of Opportunity Zones | 169

However, credit authority typically represents only 69 percent of total private
investment associated with projects (Abravanel et al. 2013).16 This implies
about $0.18 in forgone revenue for each $1 in associated investment (= 0.26 x
0.69).
The previous calculations only consider the effect of the Opportunity
Zone incentive on capital gains tax revenues. However, the incentive will have
an offsetting effect on the Federal budget by stimulating the economies of lowincome areas that receive a large share of transfer payments from the Federal
Government. Using county-level data on transfer payments and poverty rates,
the CEA estimates that an additional person living in poverty in a county is
associated with about $8,240 additional Federal transfer payments to the
county, including transfers related to income maintenance, unemployment
insurance, and medical assistance (mainly Medicaid).17 At this rate, economic
growth that lifts 1 million people out of poverty for a little more than one year
would save the Federal Government enough to offset the revenues forgone
from the capital gains tax cuts (savings of $11.2 billion = a 1 million person
reduction in poverty x 1.36 years x $8,240 per person).18

Conclusion
Much remains to study regarding the effects of Opportunity Zones on real
estate markets, entrepreneurship, poverty, and income. In coming years,
researchers will have ample data to assess the effects of OZs on diverse
community outcomes. As of the 2019 tax year, the Internal Revenue Service’s
revised Form 8996 will collect detailed information on Qualified Opportunity
Fund activity. This information will enable researchers to learn how much
Qualified Opportunity Fund investment is occurring in particular census tracts
and economic sectors. These data will permit the same rigorous empirical
studies that have been done for the New Markets Tax Credit (Freedman 2012;
Harger and Ross 2016), and they will add to the rigorous work already being
done by Arefeva and others (2020) using other data sets to evaluate the impact
of OZs.
16 This is based on footnote 7 in a paper by Abravanel et al. (2013), which reports that qualified
equity investments represent 53 percent of total project costs, while public funds represent 23
percent of project cost. This implies that qualified equity investment represents 69 percent of
private project cost (= 0.53 / (1–0.23).
17 This estimate is based on Bureau of Economic Analysis county-level data on Federal Government
transfers and county-level population and poverty data from the Census Bureau. The average
transfer per person in poverty, defined as total transfers in the county divided by the county
population in poverty, over a seven-year period was about $11,500. However, regressing countylevel transfers per capita on the poverty rate suggests that, at the margin, an extra person living in
poverty is associated with $8,240 in greater transfers to residents of the county.
18 Of course, this calculation should be viewed as illustrative because we lack an estimate of the
causal impact of poverty reduction (via investment incentives) on total Federal spending.

170 |

Chapter 5

The available evidence shows that Qualified Opportunity Funds are
well positioned to invest in communities in 2020: they have raised considerable capital, and the final regulations from the Department of the Treasury,
which were published in December 2019, have given further clarity on how
the incentive and associated investments will function. However, numerous
State-mandated restrictions and preventive behavior to slow the spread of the
COVID-19 pandemic have prevented business as usual and have slowed investment everywhere, including in OZs.
A sizable amount of capital will enter Qualified Opportunity Funds in
2020. As noted above, the capital raised by these funds in the Novogradac
sample grew by about 79 percent in the first eight months of 2020. Late in the
first quarter, the pandemic prompted a massive selloff that likely generated
capital gains for many investors exiting what had been a long bull market. And
the rapid rebound in stock values has created the potential for more gains.
Pre-COVID-19 evidence suggests that the OZ model can help spur economic recovery in thousands of distressed communities across the United
States. It has the power to mobilize investors, engage State and local stakeholders, and improve the outlook for low-income communities—all with
limited prescription from the Federal Government. In other words, the OZ
provision of the Tax Cuts and Jobs Act of 2017 is working as intended.
In nominating communities as Opportunity Zones, States selected places
in need that had the potential to attract investment. The provision’s incentives have helped mobilize the investment of $75 billion in private capital in
Qualified Opportunity Funds, and some of this capital has already spurred
growth in direct equity investments in businesses and real estate. Finally, OZ
designation and the associated investment (both anticipated and realized)
have made people more optimistic about these communities as places to live
and to work in, with designation causing a 1.1 percent increase in housing
values as of the close of 2019.
Such initial benefits underscore the potential of the Opportunity Zone
model, which rests on private initiative; on engaged State, local, and tribal
governments; and on limited Federal prescription—all to further prosperity
and self-sufficiency in those areas that most lack it. This dynamic process will
be important for helping the relatively poorer part of the population that has
been most affected by the economic slowdown from the COVID-19 pandemic.

Assessing the Early Impact of Opportunity Zones | 171

x
Chapter 6

Empowering Economic Freedom
by Reducing Regulatory Burdens
Throughout the Trump Administration, Federal agencies have demonstrated a
sustained commitment to regulatory reform. As a result, the Administration’s
regulatory efforts have reduced red tape for small businesses and the middle
class. Although the Administration set the goal of eliminating two existing
regulations for every one new regulation, it has far exceeded it. Between fiscal
years 2017 and 2019, the executive branch agencies have issued roughly seven
deregulations for every one significant regulatory action. The Administration’s
actions have served to lower costs for businesses and households while
increasing competition and productivity in the American economy, leading to
real gains, particularly at the middle and lower ends of the income distribution.
One of the most important deregulatory actions that the Trump Administration
finalized in 2020 is the Safer Affordable Fuel Efficient (SAFE) Vehicles Rule. This
joint rule from the Environmental Protection Agency and the U.S. Department
of Transportation establishes tough, but reasonable, light vehicle carbon
dioxide (CO₂) and fuel economy requirements for the 2021–26 model years.
This regulatory approach continues to improve fuel economy year over year,
while balancing efficiency, economic, and safety goals in a manner that gives
the automobile industry greater flexibility to produce products that meet
consumer demand and also creates meaningful savings for both manufacturers and customers. The Council of Economic Advisers (CEA) estimates that the
SAFE Vehicles Rule will lead to $26 billion a year in savings for producers and
consumers, and will deliver roughly 300,000 more new vehicles annually than
the previous standards at a similar total cost. Taking market distortions into

173

account, the CEA finds that the broader benefit of the SAFE Vehicles Rule is
$39 billion a year, leading to an increase in real incomes and gross domestic
product of $53 billion a year, or about 0.3 percent.
The CEA finds that the benefits of deregulation tend to skew toward the
lower-income quintiles, suggesting that lower-income households may have
benefited most, relative to household income, from the Administration’s
deregulatory actions. This finding is driven by the fact that deregulation often
reduces the prices of economic necessities—such as groceries, electricity,
prescription drugs, health insurance, and telecommunications—thereby making deregulatory actions progressive because lower-income quintiles spend a
disproportionately larger fraction of their income, relative to higher-income
quintiles, on necessities. Specifically, the gains from the deregulatory actions
discussed in this chapter amount to 3.7 percent of the average income of the
poorest fifth of households, compared with only 0.8 percent of the richest fifth,
suggesting that they benefited the poorest households four times as much as
the richest ones.
When the CEA examined the effect of a subset of the Trump Administration’s
deregulatory agenda for the 2020 Economic Report of the President, it estimated
that, after 5 to 10 years, these deregulations would lead to an increase in real
incomes of $3,100 per household a year. These previous findings, combined
with our distributional analysis, suggest that the prioritization of sensible regulatory reform has particularly benefited the lowest-income households and
allowed the U.S. economy to reach record-setting levels before the COVID-19
pandemic. A persistent focus on regulatory reform will play a critical role in the
U.S. economy’s return to the levels of economic prosperity it achieved before
the COVID-19 pandemic.

174 |

Chapter 6

n this chapter, we briefly review the Administration’s regulatory reform
progress and find that the Administration has slowed the pace of significant
regulations issued compared with previous Administrations.1 While executive agencies added an average of 275 significant regulations a year between
presidential years (PYs) 2001 and 2016, President Trump added an average of
only 74 per year, excluding deregulatory actions.2 We also find that in fiscal
year (FY) 2020, the Trump Administration is likely to achieve additional cost
savings for a fourth consecutive year. We also discuss Executive Order 13891
(EO 13891), which directs executive branch agencies to publish their guidance
documents on easily searchable public websites, marking an important step
toward increasing the transparency and accessibility of the documentation
that regulates all sectors of the U.S. economy.
In the next section, the CEA estimates the benefits associated with the
SAFE Vehicles Rule, one of the Trump Administration’s most significant deregulatory actions. This rule right-sizes CO₂ emissions standards for automobile
manufacturers and establishes a slower rate of stringency increase through
2026. The CEA finds that compared with the 2012 rule, the SAFE Vehicles Rule
will lead to $26 billion in savings a year for car manufacturers and consumers,
and will deliver roughly 300,000 more new vehicles annually than the previous
rule at a similar total cost. In addition, accounting for the effects of the rule on
factor markets, the CEA estimates that the SAFE Vehicles Rule will increase the
real incomes of Americans by $53 billion a year, or $416 per household a year,
over the 2021–29 period.
Finally, the CEA examines how the gains from regulatory reform are
distributed across income quintiles. Federal agencies must analyze whether
a proposed deregulatory action reduces regulatory costs and whether the
cost savings are larger than the benefits forgone from removing the regulation. Earlier, the CEA (2019) analyzed deregulatory actions that yield cost
savings that are larger than the benefits forgone. In this chapter, the CEA finds
that the cost savings from those regulations were distributed progressively.
Specifically, we find that though regulatory reform benefits all households,
those in the lowest income quintile likely benefit the most as a proportion of
their income. The cost savings from the deregulatory actions we study amount
to 3.7 percent of the average income of the lowest income quintile of households compared with only 0.8 percent for the highest income quintile of households. Our findings reaffirm that the Administration’s regulatory reform efforts
are helping consumers in low-income households, in part, because low-income
1 The Office of Information and Regulatory Affairs deems a regulation significant when it may have
an impact on the economy of at least $100 million, adversely affect the economy in a material way,
raise novel legal or policy issues, or otherwise meet the criteria set forth in Section 3(f) of Executive
Order 12866 from 1993. Among regulations deemed significant, those that are expected to have an
impact on the economy of at least $100 million or adversely affect the economy in a material way
are deemed economically significant.
2 Presidential years begin on February 1 and end January 31 of the following year.

Empowering Economic Freedom by Reducing Regulatory Burdens | 175

households spend a relatively large share of their budgets on necessities like
groceries and medical care that are produced by heavily regulated sectors of
the economy.

Regulation in Review
The Trump Administration’s regulatory reform agenda has reduced unnecessary regulatory burdens while continuing to protect workers, public health,
safety, and the environment. This section discusses three major executive
orders that implement this agenda. As directed by Executive Order (EO) 13771
and EO 13777, executive branch agencies have sharply cut the rate at which
they introduce new regulations and have adhered to regulatory budgets.
Under EO 13891, executive branch agencies have improved public access to
their regulatory guidance documents.
EO 13771, which was signed on January 30, 2017, requires executive
branch agencies to remove two regulations for each new regulatory action.3
EO 13777, which was signed on February 24, 2017, further requires agencies to
evaluate their regulations on a periodic basis and to make recommendations
to repeal, replace, or modify them to alleviate unnecessary regulatory burdens.
The Administration surpassed its obligations under these EOs in FY 2019, with
executive agencies issuing 150 deregulatory actions while issuing only 35
new significant regulatory actions. Between FYs 2017 and 2019, the Trump
Administration achieved roughly a 7:1 ratio of deregulatory to significant
regulatory actions. Focusing on significant regulations, the Administration has
achieved a ratio of 2.5 significant deregulatory actions to 1 significant regulatory action between FYs 2017 and 2019.
Figure 6-1 shows the total numbers of significant rules, which include
economically significant rules and other significant rules that meet part of
the definition for economic significance or are important for other reasons
described in EO 12866 (see note 1). During the Trump Administration, the average number of economically significant regulations, excluding deregulatory
actions, was only 26 per PY. The Trump Administration’s average number of
economically significant regulations remains below the average of 52 economically significant regulatory actions per year between PYs 2001 and 2016.
Including both economically significant and other significant rules, executive
branch agencies added an average of 275 significant regulatory actions per year
between PYs 2001 and 2016. Between PYs 2017 and 2019, the average number
of significant regulations each year was only 74—excluding deregulatory
3 The Office of Management and Budget defines an EO 13771 regulatory action as (1) a significant
regulatory action as defined in Section 3(f) of EO 12866 that has been finalized and that imposes
total costs greater than zero; or (2) a significant guidance document (e.g., significant interpretive
guidance) reviewed by the Office of Information and Regulatory Affairs under the procedures of EO
12866 that have been finalized and that impose total costs greater than zero.

176 |

Chapter 6

$"0- хҊрѵ$")$!$)/$)'0' .3'0$)"
- "0'/*-4/$*).Ѷ4- .$ )/$' -sсппр–рш
*)*($''4.$")$!$)/

/# -.$")$!$)/

Final rules
упп
тфп
тпп
сфп
спп
рфп
рпп
фп
спрш

спрч

спрц

спрх

спрф

спру

спрт

спрс

спрр

спрп

сппш

сппч

сппц

сппх

сппф

сппу

сппт

сппс

сппр

*0- .ѷ *-" .#$)"/*))$1 -.$/4 "0'/*-4/0$ . )/ -Ѹ!!$ *! )!*-(/$*)
) "0'/*-4!!$-.Ѹ'0'/$*).ѵ
*/ ѷ- .$ )/$'4 -. "$)$) -0-4) )$) )0-4*!/# !*''*2$)"4 -ѵ0'
*0)/.!*-спрцѶспрчѶ)спрш 3'0 -0' .that were *).$ - .$")$!$)/
- "0'/*-4/$*).; beforeспрцѶthe CEA ./$(/ sthat there was about *) such
/$*)a4 -ѵ

actions. This illustrates that the Trump Administration has slowed the pace of
significant regulations more than any administration since 2001.
In addition to the two-for-one requirement, EO 13771 required executive
branch agencies to adhere to annual regulatory budgets with cost savings
targets set by the Office of Management and Budget. In FY 2019, the Trump
Administration reached its cost savings targets for the third year in a row,
with executive branch agencies eliminating $13.5 billion of regulatory costs.
Between FYs 2017 and 2019, these agencies eliminated nearly $51 billion in
regulatory costs. In FY 2020, the Administration is likely to achieve additional
regulatory cost savings for a fourth year. This four-year stretch of regulatory
reform significantly reduced the regulatory burdens that these agencies
impose.
In 2019, President Trump issued EO 13891 to address the accumulation
of regulatory guidance documents that Federal agencies use to clarify their
regulations. EO 13891 requires executive branch agencies to make guidance
documents more accessible to the public by building a “single, searchable,
indexed website that contains, or links to, all of the agencies’ respective guidance documents.” To comply, agencies were given until June 27, 2020, after
which they needed to submit any existing guidance documents that they had
failed to publicly post as if they were new guidance. Crews (2020) estimates

Empowering Economic Freedom by Reducing Regulatory Burdens | 177

that agencies posted more than 54,010 documents as of July 7, 2020. Though
many Federal agencies have asked for waivers on compliance deadlines, EO
13891 is a significant step toward bringing transparency and oversight to
Federal guidance documents.
The Administration’s regulatory agenda has differed from that of previous administrations due to its emphasis on limiting the burden of Federal
government regulation. After four years of regulatory reform, there has been
an observable change in the cost of and rate of regulation. The establishment
of a regulatory budget and the commitment to removing two regulations for
every one new significant regulatory action have led to significant cost savings
for American firms and consumers. Supported by the improved public access
to agency guidance, these changes are enhancing the Nation’s economic efficiency and competitiveness. The CEA discussed the impact of many deregulatory actions on real income growth in an earlier report (CEA 2019). The next
section examines one of the largest deregulatory actions finalized in 2020: the
SAFE Vehicles Rule.

The Safer Affordable Fuel Efficient
(SAFE) Vehicles Rule
The largest deregulatory action finalized under the Trump Administration has
been the SAFE Vehicles Rule. This rule, which amended CO₂ emission standards
for light vehicles and appropriately increased stringency, now gives automakers greater freedom to build and sell vehicles as demanded by consumers. It
accomplishes this goal by reducing the CO₂ emission requirements for light
vehicles produced by a manufacturer. Given the inherent relationship between
CO₂ emissions and fuel economy, this has also had the effect of reducing the
required minimum fuel economy standards (in miles per gallon, mpg). Though
the SAFE Vehicles Rule promotes fuel efficiency, the fuel economy standards
grow in stringency through 2026 at a lower rate than was prescribed by prior
policy to appropriately balance policy considerations. This section estimates
the potential cost savings associated with the SAFE Vehicles Rule as well as its
distributional effects.
The corporate average fuel economy (CAFE) and greenhouse gas (GHG)
regulations are written jointly by the U.S. Department of Transportation (DOT)
and the Environmental Protection Agency (EPA) to ensure harmonization
between the two standards, given the direct relationship between fuel used
and GHG emissions.4 More stringent GHG standards increase quality-adjusted
automobile prices. In a supply-and-demand diagram, such as figure 6-2, the
gold line represents the marginal cost of producing another vehicle and the red
line represents consumers’ willingness to pay for vehicles. The GHG standard
4 Given the harmonization of the standards, we refer to these standards as GHG standards for
brevity.

178 | Chapter 6

$"0- хҊсѵ #$' -& /,0$'$-$0(2$/# /)-.
2.5

Price of new vehicles (quality
2 adjusted)
1.5
1
-$ 2$/#
./-$/er
./)-.
-$ 2$/#
' .. ./-$/
./)-.

0.5
0

∆𝒑𝒑

2000

2001

C
-"$)'
*./
җ 3'0$)"
*(+'$) Ҙ

()

Qhigh

Qlow

Quantity of new vehicles (quality adjusted)
Note: GHG = greenhouse gas.

drives a wedge between the marginal cost of producing a vehicle (excluding
regulatory compliance costs) and the marginal willingness of consumers to
purchase one, raising the price of the vehicle above the marginal cost of production. The 2012 rule would have increased the wedge by about $2,200 per
vehicle by model year 2026 relative to the SAFE Vehicles rule, as represented
by ∆p in figure 6-2.
The EPA and DOT rules generally allow firms to comply by purchasing
credits from other firms that have overcomplied, thus leading to the lowest
overall cost of compliance for the industry. The approach to this analysis
assumes that the price at which automakers buy and sell compliance credits
reveals the private cost of meeting the standards, because it should incorporate both the cost of building marginally more efficient vehicles and the willingness of consumers to buy them. To estimate prices of compliance credits, the
CEA draws from public records on nearly $700 million in credit transactions
that occurred over seven years (2012–18), which provide a simple and transparent basis for our cost estimates.
Inferring costs and benefits based on actual firm behavior—in this case
the price at which automakers buy and sell GHG compliance credits—eliminates a great deal of guesswork. Credit prices incorporate a wealth of information that is otherwise hard to observe, such as the extra cost of building a more

Empowering Economic Freedom by Reducing Regulatory Burdens | 179

efficient vehicle and the willingness of consumers to pay for such vehicles. This
approach, also known as the revealed preference approach, differs from much
of the existing literature on the costs of CAFE and GHG standards, which examines volumes of automotive engineering data and assesses consumer’s driving
habits, fuel-purchasing routines (including attempts to value consumers’ time
spent pumping fuel), and decisions about when to scrap a vehicle.5
In the revealed preference approach, we replace engineering assumptions with economic assumptions such as cost-minimization and pass-through
of costs, in which case credit prices convey the information needed to estimate
the private costs and benefits of complying with the standards.6 To the extent
that manufacturers minimize the cost of producing a given model and can
freely trade credits, the observed credit price is equal to the marginal cost
of reducing the manufacturer’s fleet-wide emissions.7 To the extent that the
cost of GHG credits is reflected in the prices consumers pay for vehicles (i.e.,
pass-through), the cost also reflects consumers’ willingness to have vehicles
with more weight or other attributes that produce additional emissions as
measured by the GHG program. This includes many dimensions of consumer
preferences, including the value that consumers place on fuel savings over the
life of a vehicle.
The costs and valuations permit quantifying the private net costs of
changing the standards because the market complies with a stricter standard
through some combination of changing vehicle attributes and adjusting prices
to shift sales to lower-emission vehicles. These private net costs are pivotal
for understanding the effects of the SAFE Vehicles Rule. Prior analyses of the
standards show that private costs and benefits dwarf environmental costs and
benefits (Bento et al. 2018).
The value of compliance credits equals the private net costs of changing
the standards, which arise through some combination of changing vehicle
attributes and skewing sales to lower-emission vehicles. These private net
costs are pivotal for understanding the effects of the SAFE Vehicles Rule: prior
analyses of the standards show that private costs and benefits dwarf environmental costs and benefits (Bento et al. 2018).

GHG Credit Transaction Data
The price at which automakers buy or sell GHG compliance credits is not
publicly available. However, because credit revenue is significant for Tesla,
it reports the revenues in its financial reports to the Securities and Exchange
5 See, e.g., regulatory impact analyses (EPA/DOT 2012, 2020).
6 EPA/DOT (2016, 2020) assume a one-for-one pass-through of compliance costs to consumer
prices, as we do.
7 Note that trading was quite limited in the initial years of the program, that these data are not
widely available for every trade, and that some companies announced intentions to not trade, even
when it represented a lower-cost way to comply.

180 |

Chapter 6

Commission. The reports indicate that Tesla earned $695 million in revenues
(in 2018 dollars) from the sale of GHG credits over the years 2012–18.8 For the
same period, EPA data show that Tesla was the second largest seller of GHG
credits, after Honda, since GHG credit trading began in 2012. Tesla’s sales have
accounted for nearly a quarter of all sales in the U.S. credit market (EPA 2019).
These revenue and sales numbers suggest that roughly $3 billion in credit
transactions have occurred across the industry since the GHG credit trading
program began.
Using Tesla’s credit sales and revenues, we calculate the average credit
price over the 2012–16 period.9 We associate this price with the standards of
the 2012–21 period because GHG credits earned during model years 2010–16
are used through model year 2021. Because credits are banked and traded
across automakers and fleets, all model years 2012 through 2021 are effectively
a single fleet for GHG compliance purposes.10 Focusing on the 2012–16 price
also has the advantage of the period being before President Trump’s election,
which would have changed expectations about the value of the credits later in
the 2012–21 period.
When calculating the credit price, we adjust Tesla’s 2012–16 credit revenues to incorporate their timing, using a 7 percent interest rate to standardize
all revenues as if they were earned in 2016, which is when the industry’s fleet
shifted from performing above the standard and accumulating credits to performing below the standard and drawing down credits. Dividing total revenues
by the quantity of credits sold over the period gives an average price of $86 per
ton of CO2 emissions, or $116 per mpg per vehicle (in 2018 dollars).11 12
The $116 credit price is a lower-bound estimate of the actual average
price at which Tesla sold its credits. Automakers are not required to report
the timing of transactions, which complicates efforts to identify credit sales
in individual years. However, automakers cannot sell credits that they do not
have. Over the 2012–16 period, at most Tesla could have sold all the credits
8 For several years, Tesla’s annual filing with the Securities and Exchange Commission did not
report revenues separately for zero emissions vehicle credits and GHG credits, but this breakout
is available from the company’s quarterly filings with the commission and was reported by Forbes
(2017). This allows us to ensure that we are not including zero emissions vehicle revenues in our
GHG revenues.
9 We note that Leard and McConnell (2017) were the first to match Tesla credit revenue with trade
volumes to infer credit prices.
10 Because the GHG standard increased in each of the years 2012–21, we expect manufacturers to
accumulate GHG credits in the early years and spend them in the later years. EPA records show
this to be the case, with most manufacturers having a credit shortfall in model year 2017; see EPA
(2019, figure 5.17).
11 In 2014 Kia and Hyundai forfeited credits in a settlement with the EPA, which were valued at $51
per ton (in 2018 dollars and with interest until 2016). Because the price is not based on a market
transaction, we do not include it in our estimation of the 2012–16 price.
12 When calculating the credit price, we take into account the small number of GHG credits that
Tesla sold in the Canadian GHG market and whose revenues would presumably be included in the
credit revenues reported to the Securities and Exchange Commission.

Empowering Economic Freedom by Reducing Regulatory Burdens | 181

that it earned through model year 2015, which is the quantity that we used to
estimate the 2012–16 price. If Tesla sold any less, the estimated price would
be higher because the same revenue would be divided by a smaller number of
credits.

Estimating the Curve for the Marginal Cost of Compliance
Our credit price data and a prior study provide two relevant points that allows
us to project what the market equilibrium price of compliance credits would
be for any given standard.13 The Tesla credit data described above provide one
observation on compliance costs: credits cost $116 per mpg per vehicle when
the standard was about 35 mpg, the average over the 2012–21 period.14 The
second data point is for model year 2006, for which Anderson and Sallee (2011)
estimate the average marginal cost of tightening CAFE standards by 1 mpg to
be $18 per vehicle. The CAFE standard during that year was 24.8 mpg.15
With two observations on compliance costs at different standards, we
can project the relationship between the standard measured in mpg and the
marginal effect of the standard on the marginal (production and opportunity)
cost of manufacturing a vehicle (figure 6-3).16 The horizontal axis measures
the standard, while the vertical axis measures additions to the marginal cost
of each vehicle. The area under the curve measures the additional cost of the
standard per vehicle. The SAFE Vehicles Rule will raise standards for 2021–26 at
a rate of 1.5 percent a year. Using fleet data from the 2012 rule rather than the
SAFE Vehicles final rule, the standards reach 45.6 in 2026, while the 2012 rule
prescribed a standard of 54.5 for model year 2025, which we assume will also
apply to model year 2026.
If going from 24.8 to 35.8 mpg increased the marginal cost of tightening
the standard from $18 to $116, then the marginal cost of further increasing the
standard must be greater than $116. From the linear credit-supply assumption,
the CEA projects that the credit price would be $203 per mpg for model year
2026 under the standards established in the SAFE Vehicles Rule (a standard of
45.6 mpg), as compared with about $283 per mpg for model year 2026 under
13 The CEA’s theoretical analysis of models with constant elasticity of substitution between
types of vehicles has shown a linear credit-supply schedule (with respect to mpg) to be a good
approximation of the actual schedule, except when the standard is especially tight, in which case
linear supply underestimates compliance costs. This suggests that our estimate of the marginal
cost of complying with the 2012 rule is likely conservative.
14 Some manufacturers let credits expire in 2014, which may suggest that the standard may not
have been binding at that time. However, 2009 credits could not be traded among automakers. In
addition, the credits that expired were 2009 credits that could only be banked for five years, unlike
credits earned in model years 2010–16, which could be banked and used through model year 2021.
15 Although this estimate of the marginal cost of compliance is for CAFE standards, it remains our
best estimate of the cost of compliance of a GHG standard of 24.8 mpg, given that there was not a
GHG standard at the time.
16 Figure 6-3 is labeled with fuel economy standards rather than emissions standards because mpg
are more familiar to readers than tons of GHG.

182 |

Chapter 6

$"0- хҊтѵ - $/-& /,0$'$-$0(!*--$*0./)-.
$/mpg/vehicle

sспрс–ср, спсхѶ
спсхѶ
1 -"
-0' +- 1$*0.-0'

‫ڦ‬счт+-*% /
0) -
+- 1$*0.-0'

-"$)'*./
*!/$"#/ )$)"
/# ./)-

‫ڦ‬спт+-*% /
0) -
-0'
‫ڦ‬ррх
( .0-
!-*(
/- .

Area = ∆𝑝𝑝 ۙ
(+/*)
0)$/*./

∆

‫ڦ‬рч

( .0- 4
) -.*)and
'' җспррҘ

E
D
тх

уф

фф

Mpg-equivalent of GHG standard
Sources: National Highway Traffic Safety Administration; Environmental
Protection Agency; Leard and McConnell (2017); CEA calculations.
Note: GHG = greenhouse gas; MY = model year.

the tighter standard originally put in place by the 2012 rule (a standard of about
54.5 mpg). For each year of the 2021–29 period, we use the average of the two
marginal costs, which can then be multiplied by the mpg difference in the standards to give the savings per vehicle from the SAFE Rule. The resulting value is
equivalent to the green area in figure 6-3.
The CEA estimates that areas A, B, and C of figure 6-2 represent $26 billion
a year in costs to new automobile consumers and producers. Relative to the
SAFE Vehicles Rule, the 2012 rule results in roughly 300,000 fewer new vehicles
delivered to consumers every year at a similar total cost, including fuel costs
and the opportunity costs of vehicle features.
The rectangular area A of figure 6-2 accounts for the largest portion and
is the product of the number of vehicles sold and the effect of changing the
standards on costs per vehicle. The marginal cost of compliance curve shown
in figure 6-3 allows us to calculate the cost of the 2012 rule per vehicle (for
model year 2025) compared with the cost of the SAFE Vehicles Rule. Doing
so indicates that phasing in the higher standard would eventually increase

Empowering Economic Freedom by Reducing Regulatory Burdens | 183

average quality-adjusted prices by about $2,200.17 For the years 2021–29, the
average annualized quality-adjusted price increase would be about $1,600.
This amount corresponds to in figure 6-2.18
Applying the $1,600 average annual savings to the more than 16 million
new vehicles sold annually in the United States gives an annualized average
increase in consumer benefits of $25 billion each year for model years 2021–29,
equivalent to area A in figure 6-2.19
Areas B and C of figure 6-2 are also part of the cost of increasing the
standards. Estimating them requires an estimate of the impact of increasing
the standards on vehicle sales. To identify the new quantity of vehicles sold
annually, the CEA uses a price elasticity of demand for new vehicles of –0.4
(Berry, Levinsohn, and Pakes 2004), model-year-specific increments to vehicle
costs (derived as above) relative to the average 2018 vehicle sales price, and
model-year-specific projections of vehicle sales.20 The sales impact is roughly
300,000 vehicles a year, which makes area B about $0.3 billion a year. Area C
requires an estimate of the effect of the SAFE Vehicle Rule standards, relative to
no standards, costs per vehicle. This baseline private cost per vehicle is shown
in figure 6-3 as areas D, E, and F. Applying it to the change in vehicle sales gives
an estimate of figure 6-2’s area C of roughly $0.4 billion a year.
Because the emissions and fuel-efficiency requirements are imposed on
the supply chain rather than on the final consumer, it follows from the passthrough assumption that costs of the regulation are reflected in consumer
prices. The $26 billion in annual private costs in the market for vehicles is therefore measured as a productivity loss, in the sense that the economy produces
less private value when assessed at market prices, using the same factors of
production—capital and labor.
The productivity loss is experienced by market participants that supply less capital in the long run and less labor in the short run.21 This means
even less real income and, to the extent that factor markets are distorted by
taxes, additional private costs. Using a marginal cost of public funds of 0.5, the
decline in labor and capital supplied adds $13 billion in private costs (0.5 x $26
17 To the extent that compliance with tighter standards is achieved entirely by adding or changing
model designs in ways that reduce emissions and increase fuel economy without other perceptible
effects on consumers’ valuation of the vehicles, the average price increase is the same as the
average quality-adjusted price increase.
18 If we assume a flat $116 per mpg per vehicle in compliance costs, the SAFE rule saves consumers
$1,032 per car, which is similar to the EPA/DOT (2020) regulatory impact analysis estimate.
19 We use a 7 percent real discount rate for the purposes of annualizing 0-year cost profiles. All
amounts are in 2018 dollars.
20 The average vehicle price is from the Kelley Blue Book. Model year 2020–29 sales forecasts are
from EPA/DOT (2020, table VI-189).
21 We adopt the “balanced growth” assumption that productivity has income and substitution
effects on labor supply that offset in the long run. As people earn more, they demand more leisure
(the income effect); but rising wages has the opposite effect, of increasing the value of work relative
to leisure, which encourages more work and less leisure (the substitution effect).

184 |

Chapter 6

$"0- хҊуѵ*).0( -1$)".!-*(/# #$' .0' .
- )/age*! )*( Ѷb4 )*( 0$)/$'
Reduction in new car expenditures as a percentage of income
пѵфп

пѵуф
пѵуп
пѵтф
пѵтп
пѵсф

пѵсп
пѵрф
пѵрп
пѵпф
пѵпп

$-./

#$-

*0-/#

$!/#

Quintile
*0- .ѷ0- 0*! *-//$./$.Ѹ '0'/$*).ѵ

billion). If the full market value of the factors supplied is considered, assuming
a marginal tax rate of 0.48 (CEA 2019), the total gross domestic product loss in
factor markets is about $27 billion ($13 billion / 0.48).
In total, the higher standards reduce real income and gross domestic
product by $53 billion a year ($26 billion in the regulated market and $27 billion
in factor markets), which is about 0.3 percent.22 This makes the SAFE Vehicles
Rule one of the single most effective deregulatory actions that the Trump
Administration has finalized thus far (CEA 2019). The estimated $26 billion in
consumer savings from the SAFE Vehicles Rule can be distributed among different household income groups. We allocate the savings across income quintiles
based on each quintile’s share of aggregate spending on new vehicles, as
reported in the Consumer Expenditure Survey. Figure 6-4 depicts the savings as
a percentage of the posttax income of each group. The savings from the SAFE
Vehicles Rule disproportionately benefit lower-income consumers, with the
savings in the lowest income quintile exceeding those of the highest quintile

22 As with many of the other regulations that the CEA has analyzed previously (CEA 2019), the SAFE
Vehicles Rule has an effect on real income whose dollar amount significantly exceeds the dollars
of net (private and social) benefits. This is primarily because net benefits account for opportunity
costs—for example, the value of leisure if not working—while real income does not.

Empowering Economic Freedom by Reducing Regulatory Burdens | 185

by 66 percent. This is because a larger share of the posttax income of lowerincome consumers goes toward the purchase of new vehicles.

The Potentially Regressive Nature of Regulation
Our analysis of the SAFE Vehicles Rule illustrates that the burden of regulatory costs can fall disproportionately on low-income households. And though
a standard question in public finance is who bears the burden of the taxes
needed to fund government expenditures, much less is known about who
bears the burden of the costs of regulations. We find that deregulation can help
consumers in low-income households by easing restrictions that disproportionately increase the prices of the goods and services they purchase. Because
high-income households spend proportionately less on economic necessities
than low-income households, the deregulation of such goods and services has
progressive benefits.23
In 2019, the CEA studied 20 deregulatory actions of the Trump
Administration and estimated that, after 5 to 10 years, they will together raise
real incomes by 1.3 percent. In this section, we revisit 10 of these regulations
to assess their distributional effect. We find that many of them will lower the
prices of necessities—such as groceries, electricity, prescription drugs, health
insurance, telecommunications, and other consumer goods and services—
and will likely benefit lower-income households more than higher-income
households. Specifically, we find that the cost savings from this subset of
deregulatory actions—together with the SAFE Vehicles Rule—amount to 3.7
percent of the average income of the lowest income quintile of households
compared with only 0.8 percent for the highest-income quintile of households
(figure 6-10). This suggests that these deregulations benefited, relative to their
income, the lowest-income quintile households four times as much as those in
the highest-income quintile.

Progressive and Regressive Tax Structures
To evaluate how a tax burden is shared, public finance economists examine
whether the burden increases with an individuals’ capacity to pay (Duclos
2008). When the burden of a tax relative to income is higher for high-income
individuals, the tax is described as progressive. In the United States, for example, Congress designed the Federal income tax to impose progressively higher
marginal rates on earners with higher incomes. In tax year 2017, the lowest
half of filers accounted for 11 percent of the adjusted gross income share while
the highest quintile accounted for 63 percent. However, due to the progressive
structure of the Federal income tax, the lowest half of filers represented less
than 3 percent of total Federal income taxes, while filers with an adjusted gross
23 The concept of economic necessities defined this way is broader than the way the word
“necessity” is commonly used outside economics.

186 |

Chapter 6

$"0- хҊфѵ ''0./-/$1 (+/*!a рф - )/-* -43Ѷ
4 )*( 0$)/$'
Percentage of income
у

п
$-./

#$-

*0-/#

$!/#

Quintile
*0- .ѷ0- 0*! *-//$./$.Ѹ'0'/$*).ѵ

income in the highest quintile accounted for over 82 percent. Conversely, when
the burden of a tax relative to income is lower for high-income individuals, the
tax is considered regressive.
Sales taxes and other consumption-based taxes, such as the value-added
tax, tend to be regressive. According to the technical economic definition, a
good or service is a necessity when the income-elasticity of demand is less than
1—for example, when a 10 percent increase in income leads to an increase in
consumption of less than 10 percent. Because low-income households spend
a higher proportion of their incomes on necessities like groceries and medical
care, sales taxes on these goods are regressive. Figure 6-5 illustrates the regressivity of a 15 percent sales tax on groceries. Households in the lowest fifth of
the income distribution would pay 3.5 percent of their income in grocery sales
taxes, while households in the top fifth would pay 0.6 percent. The grocery
tax would have an impact on consumers in the lowest income quintile, which,
relative to their income, is over five times larger than the impact on the highestincome quintile. To reduce the regressivity of sales taxes, most States exempt
groceries and some other necessities from the sales tax (Figueroa and Waxman
2017). Other States offer credits or rebates to low-income households to help
offset some of the regressivity of their sales taxes.

Empowering Economic Freedom by Reducing Regulatory Burdens | 187

The Harm Regressive Regulation Systems Pose
Many regulations may be regressive because they increase the costs of producing goods and services that are necessities (e.g., groceries and medical care).
When complying with regulations increases the costs of production, firms
increase the prices charged to consumers. Because low-income households
spend proportionately more of their income on necessities, these regulationinduced price hikes on necessities are similar to regressive sales taxes.
However, the magnitude of the effect of a regulation on consumer prices
depends on how firms respond to production cost increases, which in turn
depends on market conditions. After a regulation, the market reaches a new
equilibrium, where consumers pay a higher price for the good (figure 6-6).
In the case shown in figure 6-6, firms are able to pass their regulatory
costs fully through to consumers through higher consumer prices. In other
cases with different market structures (not shown), a full pass-through of
regulatory costs does not always occur. For example, in response to a $1
increase in the cost of production, a firm might only raise prices by 50 cents
due to competitive constraints. Figure 6-6 can also be reinterpreted to show
another possible effect of regulation, where the regulation acts as a barrier to
entry that limits new competition, resulting in a higher equilibrium price with
above-normal profits or “economic rents” for established firms. In general,
regulations will have effects on consumers and firms, conventionally measured
by changes in consumer and producer surpluses. Tracing through the producer
surplus effects to the distribution of the incomes of factors of production can
be complex. (See box 6-1.)
Low-income households spend more of their income on goods and
services in general because they have lower savings rates, making regulations
that increase the price of these goods and services more regressive (Dynan,
Skinner, and Zeldes 2004). Households in the lower-income quintiles spend
larger fractions of their incomes for almost all the categories of goods and services tracked by the Consumer Expenditure Surveys (CEX). Thus, deregulation
is often progressive because it removes regressive regulatory cost burdens that
inflate the prices of necessities. Figure 6-7 shows spending patterns for some
important categories of goods and services. Even when regulations do not
intentionally target necessities, they can have the unintended consequence of
imposing a regressive cost burden. Chambers, Collins, and Krause (2019) find
that regulatory compliance costs increase the prices of necessities including
energy, food, healthcare and health insurance, housing, and transportation.
Unlike sales taxes, however, policymakers typically do not exempt the production of necessities from regulations.
Other regulations can be regressive because they intentionally target
consumer choices that vary with income. Product standards are a common
example because they mandate that products must have certain features or

188 | Chapter 6

$"0- хҊхѵThe (+/ *! "0'/$*)$)*(+ /$/$1 -& /
Price

-"$)'
*./0-1
2$/#
- "0'/$*)
-"$)'
*./0-1
2$/#*0/
- "0'/$*)

()

Quantity

$"0- хҊ7ѵ - )/age*! )*( + )/*)0/'4/ "*-4Ѷ
4 )*( 0$)/$'
**/#*(

/$'$/$ .

'/#$).0-)

#$' +0-#. .

Percentage of income
30
25

20
15
10
ф

п
$-./

#$-

*0-/#

$!/#

Quintile
*0- .ѷ0- 0*! *-//$./$.Ѹ '0'/$*).ѵ
*/ ѷ/$'$/$ . $)'0 .0/$'$/$ .Ѷ!0 ',)+0'$. -1$ .ѵ

Empowering Economic Freedom by Reducing Regulatory Burdens | 189

Box 6-1. Effects of Regulation on Small Businesses
Regulations can have regressive effects on small business because of economies of scale. For example, if a regulation requires that firms establish retirement accounts, larger firms’ average costs will be lower because they can
spread the fixed costs over a larger pool of employees. Given that the cost of
retirement accounts are already lower for larger firms than for small firms,
large firms, all else being equal, will be more likely to already have retirement
accounts established before the regulation, thereby causing the regulation to
have more of an impact on small than large firms.
Policymakers often attempt to offset regulatory burdens for small businesses by exempting businesses with a certain level of revenue or number of
employees. However, the threshold exemptions distort the market and can
cause businesses to cluster near the threshold limit. In France, where many
regulations apply after a firm reaches 50 employees, Garicano, LeLarge, and
Reene (2016) find that firms cluster below the employee threshold to enjoy
regulatory exemptions. Though clustering reduces firm’s regulatory burden, it
also reduces total welfare and the productivity of the economy. In the United
States, the Affordable Care Act used a similar approach by reducing the
requirements imposed on businesses with fewer than 50 full-time employees.
Congress has passed several pieces of legislation that attempt to reduce
the regulatory burden placed on small entities. The Regulatory Flexibility
$"0- хҊiѵ*)*($''4$")$!$)/$)'0' .2$/# "0'/*-4
' 3$$'$/4)'4.$.!*-(''0.$) .. .Ѷ4- .$ )/$' -s
сппр–рш
)'4.$.)*/- ,0$-

)'4.$.- ,0$-

Economically significant final rules

рпп
шп
чп
цп
хп
фп

уп
тп
сп
рп
спрш

спрч

спрц

спрх

спрф

спру

спрт

спрс

спрр

спрп

сппш

сппч

сппц

сппх

сппф

сппу

сппт

сппс

сппр

*0- .ѷ *-" .#$)"/*))$1 -.$/4 "0'/*-4/0$ . )/ -Ѹ!!$ *! )!*-(/$*))
"0'/*-4!!$-.Ѹ'0'/$*).ѵ
*/ ѷ- .$ )/$'4 -. "$)$) -0-4) )$) )0-4*!/# !*''*2$)"4 -ѵ

190 |

Chapter 6

Act of 1980 requires that agencies perform regulatory flexibility analyses for
regulations that may have an effect on small entities, with specific attention
to competitiveness and fairness; see figure 6-i. In 1996, Congress passed the
Small Business Regulatory Enforcement Fairness Act, creating panels that
enable small entities and regulatory agencies to interact with regulators during the regulatory process. In 2019, the Trump Administration issued EO 13891
and EO 13892, which required Federal agencies to make their guidance easily
accessible and make sure that all enforcement actions are transparent and
fair. These EOs are especially important for small businesses that may otherwise lack the capability to understand regulations relevant to their business.

attributes, whose desirability can depend on one’s income. To the extent that
the mandated features are normal goods (i.e., demand for the feature increases
as income increases), high-income consumers would purchase more of the
mandated feature even without the product standard. If their demand for the
regulated feature is strong enough, the products purchased by high-income
consumers will likely already meet the product standard. Although product
standards are less binding or even nonbinding on high-income consumers,
these standards impose costs on low-income households, which are required
to pay higher prices for features they do not highly value.
Energy efficiency standards are another example of regulations that can
be regressive due to the consumer choices they target. For instance, consumers who use their air conditioners on most days of the summer might find
that energy savings pay back the higher price of a more efficient appliance
within a few years. Low-income consumers who can only afford to use their
air conditioners infrequently face a longer payback period and might be better
off purchasing a lower-price and less-efficient appliance. Therefore, energyefficient appliances and vehicles are more valuable to consumers who use their
appliances and vehicles regularly. The CAFE standards, discussed above, have
a similar effect. Levinson (2019) finds that high-income households purchase
more fuel-efficient cars. Levison estimates that the CAFE standards disproportionately burden low-income households, which are less likely to prioritize
fuel efficiency, absent CAFE. In other words, the CAFE standards may have less
impact on high-income households because they already prefer to purchase
more fuel-efficient cars.
Some health insurance regulations include product standards that can
also be regressive. Health insurance regulations related to the Affordable Care
Act (ACA) are notable examples. The ACA’s individual mandate requires nonexempt consumers to have one of several enumerated forms of health insurance coverage. Through tax year 2018, the Internal Revenue Service enforced
the individual mandate with a monetary penalty; the Tax Cuts and Jobs Act

Empowering Economic Freedom by Reducing Regulatory Burdens | 191

$"0- хҊ8ѵ )$1$0')/ )'/$ .. - )/age*!
)*( Ѷ4 )*( 0$)/$'
Percentage of income
пѵру
пѵрс
пѵрп
пѵпч
пѵпх
пѵпу
пѵпс
пѵпп
$-./

#$-

*0-/#

$!/#

Quintile
Sources: Bureau of Labor Statistics; Internal Revenue Service; CEA calculations.

of 2017 set the mandate penalty to zero, becoming effective in the 2019 tax
year. Because most high-income households already had coverage through
ACA-compliant insurance plans, the mandate penalty fell disproportionately
on lower- and middle-income households (figure 6-8). Households in the lowest income quintile bore a proportionately smaller burden than households in
the second-lowest quintile because households in the lowest income quintile
were more likely to be covered by Medicaid or receive subsidies to purchase
ACA-compliant insurance. After the second lowest income quintile, the burden
of the individual mandate penalty was steeply regressive. Other regulations—
including the 2016 short-term, limited duration insurance rule—banned a number of insurance options that were popular among low-income households
that made choices based on what was best for them.
Academic research provides several explanations for why the regulatory
process leads to product standards and other forms of regulations that intentionally target certain consumer choices. Instead of always serving the general
public interest, the regulatory apparatus may be prone to capture by special
interests (Stigler 1971). Regulatory capture could cause policymakers to enact
legislation and regulators to issue regulations that privilege certain groups, to
the detriment of other groups, such as the public or their competitors. Mulligan
and Philipson (2000) argue that wealthier portions of society may advocate for

192 |

Chapter 6

regulations that impose their preferences on the general population and offset
some of those costs through a progressive tax system. Similarly, Thomas (2012)
suggests that regulators may focus on prioritizing regulations that reduce risks
for wealthier households at the cost of low-income households.

Lower-Income Households Often Gain
the Most from Regulatory Reform
In an earlier report, the CEA estimated that the effect on real incomes associated with 20 deregulatory actions under the Trump Administration will total
$235 billion a year (CEA 2019, 2020), which we also discussed in chapter 3 of the
2020 Economic Report of the President. We estimated that these 20 deregulatory actions will raise real incomes by reducing the prices of consumer goods,
and by increasing competition, productivity, and wages. An important part
of our earlier analysis was to account for the excess burden that regulatory
actions impose on factor markets for labor and capital. In this section, we focus
on the distributional implications of the reductions in the prices of consumer
goods. The narrower scope of the analysis means that some of the deregulatory actions considered in the earlier CEA report are not part of this study (table
6-1). Though the CEA has not studied all the deregulatory actions taken since
2017, our analysis in this section builds upon our previous work, which used
a sampling procedure to identify the largest deregulatory actions in terms of
economic impact (CEA 2019, 2020).
We combine our estimates of the cost savings from deregulatory actions
with data from the CEX. We attribute the reduction in industry costs to an
expenditure category listed in the CEX shares of annual expenditures by
income quintile. For example, we estimated that the Federal Communication
Commission’s (FCC) repeal of the Protecting and Promoting the Open Internet
and issuance of Restoring Internet Freedom Order would provide $16.1 billion
in cost savings to Internet users.24 The expenditure category of the CEX for consumers most affected by Internet prices is the computer information services
(Internet access) category. We used the expenditure shares by income quintile
to calculate the reduction in Internet access expenditures as a fraction of total
income (after tax) for each quintile. The results, shown in figure 6-9, show that
relative to their incomes, the FCC’s deregulation of internet access has an
effect on consumers in the lowest income quintile that is five times larger than
the effect on the highest income quintile.
24 The CEA’s distributional analysis focuses on regulatory cost savings that we predict are
passed through to consumers who pay lower prices for the goods and services produced by the
deregulated industries. Our earlier study finds substantial additional cost-savings in the markets
for factors of production, i.e., in the labor and capital markets, as reported by the CEA (2019, table
6-1). Tracing through the factor market effects to their effects on the distribution of household
incomes is a complex and challenging task that is beyond the scope of this Report. This narrower
focus is only a portion of the total cost savings than we estimated in the earlier CEA report.

Empowering Economic Freedom by Reducing Regulatory Burdens | 193

$"0- хҊ9ѵ*).0( -1$)".*) )/ -) / ..!-*(/#
./*-$)" )/ -) /- *(0' Ѷ4 )*( 0$)/$'
Percentage of income
пѵч

пѵц
пѵх
пѵф
пѵу
пѵт

пѵс
пѵр
п
$-./

#$-

*0-/#

$!/#

Quintile
*0- .ѷ 0- 0*! *-//$./$.Ѹ0- 0*!*)*($)'4.$.Ѹ'0'/$*).ѵ
*/ ѷ'0 .- +- . )/the ’. ./$(/ .*!*).0( -.1$)"...#- *!/# $-
$)*( ,2#$#++'$ the *).0( -3+ )$/0- 0-1 4’.,0$)/$' ) 3+ )$/0-
//*)/$*)'$)*( /ѵ

We performed similar calculations for the set of deregulatory actions
enacted since 2017 that reduced consumer prices (table 6-1). For some deregulatory actions, the distribution of the cost savings across income quintiles
exactly follows the distribution of consumer expenditures in the relevant CEX
category. Examples include deregulatory actions that we estimate will reduce
the prices of electricity, prescription drugs, and Internet access. For other
deregulatory actions, the distribution of cost savings across income quintiles
reflects the fact that the original regulation targeted consumer choices that
were more common among low-income households. Examples include health
insurance deregulations and the deregulation of the short-term loan industry.
We used additional information about consumer behavior in those markets
to refine our estimates of the distribution of the cost savings across income
quintiles.
When we total the results for the complete set of regulations we analyze,
we find that the deregulatory actions are strongly progressive and reduce
the disproportionate burden regulations impose on low-income households.
We find that the gains from the deregulatory actions we study amount to 3.7
percent of the average income of the poorest fifth of households, compared
with only 0.8 percent for the richest fifth (figure 6-10). The deregulatory actions
have an effect on consumers in the lowest income quintile, which relative to
194 |

Chapter 6

Table 6-1. Selected Deregulatory Actions’ Annual Impact on
' )*(
(+/*) ' )*(
җ$)$''$*) dollar.a4 -Ҙ
-*/ /$)"/# -$14*!0./*( -.*!-*))
‫ڦ‬сс
/# - ' *((0)$/$*). -1$ .җ+/Ҋ )Ҙ

( ҝ .-$+/$*)

/- (-*/ /$*)0'
Definition of “Employer” Under Section 3(5) of ERISA..*$/$*) '/#').җ 0' Ҙ
#*-/Ҋ -(Ѷ $($/ Ҋ0-/$*) ).0-) ҡҗ Ҙ
44Ѷ #$' $/' Ѷ) -/$) $"#Ҋ*./ )./''( )/
*).
*+ *! /$*).спсҗҘ)җҘ*!& -.)
/*&4-./
./ - 1 )/$*)Ѷ-*0/$*)0% //**4'/$ .Ѷ)
.*0- *). -1/$*)Ѹ .$..$*)*- 1$.$*)ҡ
+ '*!-*/ /$)")-*(*/$)"/# + ) )/ -) /
) ..0) *! ./*-$)" )/ -) /- *(
) * -)$5/$*)!!*-/.
# 30/.) *./Ҋ 0 /# )$1$0'
)/ )'/4/* -*
! -!!*-' 0 '!!$$ )/җҘ #$' .0'
*/'

‫ڦ‬с
‫ڦ‬рц
‫ڦ‬рт
‫ڦ‬ц
‫ڦ‬п
‫ڦ‬п
‫ڦ‬фу
‫ڦ‬тс
‫ڦ‬сч
‫ڦ‬фт
‫ڦ‬ссч

*0- ѷ'0'/$*).ѵ
Note: An asterisk (*) signifies the use of a shortened name for the regulation. All annual
effects on real income are rounded to the nearest billion dollars. The impact on real
incomes is estimated based on the full impact of the regulation, which may be realized
in the future.

their income is over four times larger than the effect on the highest one. Above,
we noted that we find that, with a hypothetical 15 percent tax on groceries,
households in the lowest fifth of the income distribution would pay 3.5 percent
of their income in grocery sales taxes, over five times larger than the effect on
the highest-income quintile. The deregulatory actions we study removed cost
burdens that were similar to a regressive tax on groceries.
Our analysis focuses on the distribution of the gains from regulatory
reform that reduced the burdens costly regulations impose on consumers.
Regulatory and deregulatory actions have both benefits and costs. The net
effect of the actions on consumer welfare depends on the difference between
benefits and costs, or the net benefits. The distribution of the net benefits of
an action depends on the relative sizes of the benefits and costs and on the
relative progressivity of how the benefits and costs are distributed (Bento,
Freedman, and Lang 2015). Under Executive Order 12866 and Executive Order
13771, Federal agencies must analyze whether a proposed deregulatory action
reduces regulatory costs and whether the cost savings are larger than the
benefits forgone from removing the regulation. The CEA (2019, 2020) analyzed

Empowering Economic Freedom by Reducing Regulatory Burdens | 195

$"0- хҊр0ѵ*).0( -1$)".!-*( ' / - "0'/*-4/$*).
)/# #$' .0' Ѷ4 )*( 0$)/$'
Percentage of income
уѵп
тѵц
тѵф
тѵп

сѵф

сѵф
сѵп

рѵц

рѵф

рѵс
пѵч

рѵп

пѵф
пѵп
$-./

#$-

*0-/#

$!/#

Quintile
*0- .ѷ 0- 0*! *-//$./$.Ѹ0- 0*!*)*($)'4.$.Ѹ'0'/$*).ѵ

deregulatory actions that yield cost savings that are larger than the benefits
forgone, and we find that the cost savings are distributed progressively. Unless
the forgone benefits of these rules were distributed more progressively than
the costs, the distribution of the net benefits from these deregulations were
progressive.

The Regressivity of Federal Regulation
Offsets the Progressivity of Federal Taxes
Despite the deregulatory actions taken by the Trump Administration, a large
amassed body of Federal regulations remains. The total cost of Federal regulations is difficult to estimate with precision. As of September 1, 2020, Federal
agencies estimate that their regulations require that the U.S. public complete
roughly 11.6 billion hours of paperwork at a cost of $150 billion each year.
Between 2006 and 2018, the Federal Government issued an average of 3,600
regulations each year, not including guidance and other documents that
some observers describe as “regulatory dark matter,” which is another form
of regulation that does not always include public participation (Crews 2017).
Recent estimates of the total annual costs of Federal regulations range from

196 | Chapter 6

$"0- хҊр1ѵ*).0( -1$)".!-*( - "0'/*-4 /$*).
*(+- with the -'3/ Ѷ4 )*( 0$)/$'
Consumer savings from deregulatory actions
(percentage of income)
тп

Federal tax rate (rate)

сф

сп

рф

рп

5
0

п
$-./

#$-

*0-/#

$!/#

Quintile
*0- .ѷ*)"- ..$*)'0" /!!$ Ѹ '0'/$*).ѵ
*/ ѷThe ’. ./$(/ . *! #$)*( ,0$)/$' ’. 1 -" -' /3-/ $)'0 .
$)$1$0'$)*( Ѷ+4-*''Ѷ*-+*-/ $)*( Ѷ) 3$. /3 ..*!спрхѵ

almost half a trillion dollars into the trillions of dollars (CEA 2019).25 Crain
and Crain (2014) use a proxy measure of regulation to conclude that Federal
regulations imposed a burden of roughly $2 trillion in 2012. Coffey, McLaughlin,
and Peretto (2020) estimate the effect of regulations on 22 industries, between
1977 and 2012, and find that if regulations were held at 1980 levels, then the
economy would have been $4 trillion larger in 2012.
The magnitude of regulatory burdens, in combination with their potential regressivity, implies that regulatory costs could largely offset the progressivity of Federal taxes. Regulatory cost estimates that range into the trillions of
dollars are substantial compared with the $3.6 trillion in Federal tax revenues
in fiscal year 2020. In fact, the distribution of the gains from the subset of
deregulations we examined is almost the mirror image of the distribution of
the burden of Federal taxation (figure 6-11). Thus, we find that continued regulatory reform has the potential to shift more of the total burden the Federal
government imposes on businesses and consumers away from low-income
households with less capacity to pay.
25 No Federal agency attempts to estimate the cumulative cost of all Federal regulation; however,
the Regulatory Right-to-Know act tasked the Office of Management and Budget to estimate the
total cost and benefits of a subset of Federal rules that have been designated major rules. Federal
regulatory agencies only monetize the costs and benefits of less than 1 percent of all the rules they
issue.

Empowering Economic Freedom by Reducing Regulatory Burdens | 197

Conclusion
This chapter has highlighted the Trump Administration’s commitment to
reducing the regulatory burden on households and businesses. The CEA finds
that the benefits associated with one of fiscal year 2020’s biggest deregulatory
actions (the SAFE Vehicles Rule) will reduce prices for consumers by almost
$2,200 per vehicle by 2026. Moreover, we find that the Administration’s regulatory reform efforts may have benefited those in the lowest income quintile the
most as a proportion of their income. Specifically, we conclude that the costs
savings from the SAFE Vehicles Rule and other deregulatory actions we have
studied amount to 3.7 percent of the average income of the lowest income
quintile of households compared with 0.8 percent for the highest income
quintile of households. Our findings provide evidence of the benefits of regulatory reform and reaffirm that deregulation can help consumers in low-income
households—who spend a relatively large share of their budgets on necessities
that are often in heavily regulated sectors of the economy—the most.
The regulatory reforms we have reviewed in this chapter were enacted
before the COVID-19 pandemic and its effects on health and the U.S. economy.
Although the longer-term consequences are hard to predict, evidence on the
scope and nature of COVID-19’s near-term effects are beginning to emerge.
Unfortunately, the COVID-19 pandemic has hit low-income households particularly hard, in their health outcomes and in economic consequences
including lost jobs and wages. The cost savings and distributional effects from
the deregulations we discuss may have somewhat cushioned the blow to
low-income households. Moreover, regulatory reform may help position the
United States for a robust economic recovery and be a powerful tool to help
lift up middle- and low-income Americans as the economy recovers from the
COVID-19 pandemic.

198 |

Chapter 6

x
Chapter 7

Expanding Educational Opportunity
through Choice and Competition
During the last 30 years, school choice programs have undergone dramatic
expansion in the United States. These programs—organized at the Federal,
State, or local level—share a common goal of expanding access to education
options that exist alongside and ultimately improve public school options for
primary and secondary education. Under a district public school (DPS) system,
students are assigned to schools based on where they live, and the only form
of school choice requires physically moving to an area with better schools for
those families that can afford to do so. School choice programs have altered
this landscape in fundamental ways by increasing competition in the school
system and enhancing educational opportunities for all students, especially
those from disadvantaged groups.
One rapidly growing school choice option is charter schools. Charter schools
are public schools that educate millions of students using public funding, but
with operational autonomy from the local public school system. Additionally,
scholarship programs—funded both publicly and privately—assist hundreds of
thousands of students with tuition at private schools and can provide access to
courses, work-based learning opportunities, concurrent and dual enrollment
for college credit, home education, special education services and therapies,
tutoring, and more. These and other choice programs are providing opportunities for families that lack them, thereby ensuring that all schools have an
incentive to deliver a high-quality education.
This chapter documents the development and expansion of school choice programs since 1990, when the first major school choice program was introduced

199

in this country. We provide an overview of school choice programs, describing the main types of programs with examples from around the country. We
also discuss the role of Federal policy, including recent actions of the Trump
Administration to further expand school choice.
We next explain the key benefit of expanding school choice policies: more educational competition that empowers families and pressures schools to deliver
more value. School choice programs can extend competition to all areas,
including those where families with lower incomes have little ability to move to
more affluent areas in search of better schools. The programs enable families
to hold accountable what in some cases is a failing local education monopoly.
This can benefit the children who use school choice programs as well as the
children who remain in a DPS, because all schools must compete for student
enrollment by providing a higher-quality educational experience. We discuss
the growing empirical evidence that carefully crafted school choice programs
do improve educational outcomes for all students. In other words, competition
can be the tide that lifts all boats.

chool choice refers to policies, legislation, and organizations that foster
alternatives to residentially assigned district public school (DPS) education. This chapter provides an overview of school choice programs in
the United States and the role of Federal policy in helping to foster them. It also
discusses the economic theory of competition that motivates these programs.
And finally, it reviews the empirical research on the programs’ impact.
The first major school choice programs were introduced in the early
1990s. The programs originated from concerns that students from low-income
families had no alternatives to residentially assigned DPSs, especially in places
where the local DPS had a poor performance record. In 1990, the State of
Wisconsin enacted the Milwaukee Parental Choice Program, the first major
voucher program in the Nation. This program, which continues to operate
today, offers publicly funded vouchers to eligible students in Milwaukee who
choose to attend private schools. In 1991, the State of Minnesota enacted the
first charter school law, with the first public charter school opening in Saint
Paul in 1992. Over time, demand for alternatives to residentially assigned DPSs
has increased, and school choice programs have been introduced in many
school districts throughout the country.

200 |

Chapter 7

Chart Title
$"0- цҊрѵ0( -*!.Ѷ*0# -.Ѷ)3Ҋ- $/#*'-.#$+.Ѷ
600000
ршшх–спсп
0/$*).1$)".*0)/.
*0# -.
3Ҋ- $/.#*'-.#$+.
500000

Number (thousands)
400000
хпп
300000

фпп

200000
100000
упп

тпп

спп
Education savings accounts

рпп
п
ршшх

ршшч

сппп

сппс

сппу

Vouchers

сппх

Tax-credit scholarships

сппч

спрп

спрс

спру

спрх

спрч

спсп

*0- ѷ#*$ ѵ

Today, private choice programs provide financial support that enables
hundreds of thousands of students to attend private schools. As we discuss
in more detail below, these programs include vouchers, tax-credit scholarships, and State-funded education savings accounts (ESAs) (figure 7-1), which
together now number about 539,000 (EdChoice 2020a).
Public choice programs further allow millions of students to attend
schools other than the DPS they would attend based on geographic residency.
In addition to establishing public charter schools, States and local governments have introduced other public choice programs, such as magnet schools,
which are public schools with specialized programs of study. Enrollment in
public charter schools and magnet schools has grown over time (figure 7-2). In
2017–18, charter schools and magnet schools enrolled 3.1 million and 2.7 million students, respectively (NCES 2019a). Many public school systems have also
introduced open enrollment programs that permit students to attend public
schools other than a residentially assigned DPS.
Although Federal funding plays a relatively minor role in K-12 school
funding, Federal policy does support State and local governments seeking
to expand school choice. Below, we discuss the main Federal programs that
support school choice, including the Magnet Schools Assistance Program and
the Charter Schools Program. We also report on recent policies implemented
under the Trump Administration to enhance this support, including the expansion of 529 ESAs to primary and secondary education under the 2017 Tax Cuts
and Jobs Act (TCJA).
Although school choice programs have grown dramatically, the majority
of students in K-12 school continue to attend a residentially assigned DPS.
As shown in figure 7-3, the proportion of students attending a residentially
assigned DPS fell by about 5 percentage points between 1999 and 2016. Over

Expanding Educational Opportunity through Choice and Competition

| 201

$"0- цҊсѵ)-*''( )/$)0'$#*$ -*"-(.o1 -$(
#-/ -.#**'.

") /s#**'.

Students enrolled (millions)

тѵф
тѵп
сѵф
сѵп

рѵф
рѵп
пѵф
пѵп
сппп–20пр

сппф–х

спрп–рр

спрф–рх

*0- ѷ /$*)' )/ -!*-0/$*)//$./$.ѵ

the same period, the proportion of students attending a public school of their
choosing rose by about 4 percentage points.
After providing an overview, we next discuss the economic theory of competition that motivates school choice. School choice programs are founded on
the idea that when schools are exposed to increased competition, educational
outcomes will improve. The intellectual foundation is often credited to the
Nobel laureate Milton Friedman (1955), who argued that when public schools
face competition, they have stronger incentives to provide a high-quality,
cost-effective education. As described by Hoxby (2003), the beneficial effects
of competition come into force through the power of choice. Because charter
schools are not guaranteed any enrollment, they are spurred to provide a better education than competing DPSs in order to attract students. Private schools
face similar incentives but, because they charge tuition, they are not financially
accessible to some families. Private choice programs seek to expand access
to private schools and make them more competitive with DPSs by subsidizing
the cost. Far from acting as a one-way street, however, competition may also
induce a response by DPSs. When faced with the threat of losing students to a
competitive charter or private school, a DPS may work to improve its performance in order to retain students and the funding that comes with them.
As we explain in this chapter, the design of school choice programs can
help ensure that competition leads to benefits for all students. In theory, choice
programs could disproportionately entice away more advanced or more motivated students from the DPS system, leaving behind struggling students who
lose the benefits of interacting with their high-performing peers. Similar to the
default method of school choice—the ability to move to a more affluent area
with better schools—this might create segmentation by family background
that harms some students who remain in the DPS system. However, as we
discuss, these theoretical concerns are not borne out by empirical research, in

202 |

Chapter 7

$"0- цҊтѵ - )/" $./-$0/$*)*!/0 )/." ф–рцYears
// )$)" $) -"-/ )/#-*0"#рс/#- Ѷ4#**'4+ ,
ршшш ) спрх
Percent
чп

цп

ршшш
цуѵр

спрх

хчѵч

хп
фп
уп
тп

руѵт

сп

рчѵц

рп

рпѵп

шѵс
рѵц

тѵт

п
0'$.#**'Ѷ..$")

0'$.#**'Ѷ#*. )

-$1/ .#**'

*( .#**'

*0- ѷ/$*)' )/ -!*-0/$*)//$./$.ѵ

part because most school choice programs include design features that avoid
such outcomes by targeting eligibility or providing more generous resources to
relatively disadvantaged students.
In the last section of the chapter, we review the empirical evidence on
the impact of school choice programs. For students participating in these
programs, achievement results as measured by test scores are mixed, although
several studies find large positive results for minority and low-income students. We explain that some positive outcomes of school choice emerge later in
a child’s development through higher educational attainment, and studies of
these longer-term outcomes are generally more positive. Thus, policymakers
should consider a broad range of outcomes when evaluating efforts to promote higher education quality through school choice. We also discuss studies
of school choice relating to racial and ethnic integration, longer-term nonacademic outcomes, and fiscal effects. Finally, in terms of the impact on students
who remain in their residentially assigned DPS, we discuss emerging empirical
research suggesting that all students can benefit from the expansion of school
choice programs in their local district, regardless of whether they participate
themselves in the programs or decide to remain in their DPS. Thus, choice can
be a tide that lifts all boats, not because of the inherent superiority of any one
school type over another, but rather because competition pressures all schools
to improve quality and deliver value.

Overview of School Choice
Programs and Federal Policy
In this section, we describe the main types of private and public choice
programs. Each type has unique advantages, which are highlighted in their

Expanding Educational Opportunity through Choice and Competition

| 203

respective subsections. We then provide examples of school choice in five
regions of the country. Finally, we discuss Federal actions, including actions
taken by the Trump Administration, to support school choice.

Private Choice Programs
There are three basic types of private choice programs: vouchers, tax-credit
scholarships, and education savings accounts. Together, these programs enroll
hundreds of thousands of students (figure 7-1) (EdChoice 2020a). Eight States
also offer an individual tax deduction or tax credit for certain educational
expenses, which are typically expended on one’s own child.
Voucher programs provide a subsidy to parents to enroll their child in a
private school. The voucher subsidy is typically set at a specific share of the
public funding intended for that child’s education. Though vouchers may not
necessarily cover the full cost of tuition, they make private education more
affordable for parents. During the 2018–19 school year, 28 voucher programs
operated in 17 States, the District of Columbia, and Puerto Rico. Together,
these programs served more than 188,000 students (EdChoice 2019). Many programs are targeted at low-income students or students with disabilities: seven
States and the District of Columbia provide vouchers to low-income students,
and eleven States provide vouchers for students with disabilities.
Tax-credit scholarships allow students to receive private funding from
nonprofit organizations to attend private schools or may pay for other educational expenses, including tutoring, online learning, concurrent and dual
enrollment for college credit, and homeschooling expenses. Individuals and
businesses may donate to these nonprofit organizations and receive a State
income tax credit in return. States limit the total amount of tax credits that will
be offered for the year and/or the amount that each business or individual can
claim. Tax-credit scholarships often target low-income students, students with
disabilities, or students assigned to a low-performing DPS. During the 2018–19
school year, 23 programs operated in 18 States and served nearly 275,000
students (EdChoice 2019; Kaplan and Owings 2018). One of these programs,
the Montana tax-credit scholarship program, became inoperable in late 2018
when the Montana Supreme Court ruled that the program violated the State’s
constitution. However, that decision was reversed by the Supreme Court of the
United States in June 2020 (see box 7-1).
Education savings accounts (ESAs) are multiuse scholarships that allow
participating parents to pay for current educational expenses, such as private
school tuition, homeschool expenses, contracted services provided by a public
school or school district, courses, concurrent and dual enrollment for college
credit, special education services and therapies, and tutoring, as well as for
future educational expenses such as college. During the 2018–19 school year,
five programs operated in five States supporting more than 18,700 students.
Most States restrict eligibility to students with disabilities. Arizona’s program
204 |

Chapter 7

Box 7-1. The Supreme Court’s Espinoza v.
Montana Department of Revenue Decision
In June 2020, the Supreme Court of the United States (SCOTUS) reversed
the decision of the Montana Supreme Court in Espinoza v. Montana
Department of Revenue (Supreme Court of the United States 2019). The central judgment, taken from Chief Justice Roberts’s majority opinion, is that “a
State need not subsidize private education. But once a State decides to do so,
it cannot disqualify some private schools solely because they are religious.”
The decision automatically reinstated a Montana tax credit scholarship program that the State court had ended because it violated a “no-aid”
provision, also referred to as a Blaine amendment, in the State’s constitution.
SCOTUS found this particular application of the no-aid provision discriminatory against religious schools and their prospective pupils’ families, and thus
the application violated the U.S. Constitution’s free exercise clause.
The original Blaine amendment was a proposed amendment to the
U.S. Constitution that narrowly failed Senate approval in 1875. The Blaine
amendment, founded in anti-Catholic immigrant sentiment with the purpose of keeping schools Protestant, would have expanded the reach of the
Establishment Clause in the First Amendment to explicitly prohibit public
spending on religious schools or organizations. Subsequently, a majority of
States added Blaine language to their State constitutions; Montana became a
State in 1889 and included a Blaine clause in its original constitution.
Montana’s scholarship program, passed in 2015, provided for tax
credits worth a maximum of $150 for gifts made to organizations that provide
private school scholarships. All Montana students are eligible for the program.
However, due to the litigation and the $150 maximum gift, only 25 students in
Montana received funds from the program in the 2016–17 school year.

also includes other groups of students, including low-income students, students assigned to a low-performing DPS, students who are adopted or in foster
care, students who live on a tribal land, and children of active-duty military or
who were killed in the line of duty. The funding amount varies by State, with
some States setting a flat amount per pupil, such as $6,500 in Mississippi, and
others giving a portion of the State’s education funds per pupil to the parents.
In some States, higher scholarship amounts are provided to low-income
students or students with disabilities. For example, in Arizona, students from
households with income up to 250 percent of the Federal poverty line receive
100 percent of State per-pupil funding, whereas other eligible students receive
90 percent (EdChoice 2020a).

Expanding Educational Opportunity through Choice and Competition

| 205

Public Choice Programs
Public choice programs include charter schools, magnet schools, and open
enrollment policies. Charter schools are tuition-free public schools that operate
independently from district public schools and thus have more autonomy over
their educational programs, hiring, operations, and budget in exchange for
greater accountability. Although the programs differ by State, many charter
schools receive per-pupil funding that follows students from their residentially
assigned DPS to the charter school. In addition to providing funding for charter
schools, State governments grant entities the role of charter authorizers.
Charter authorizers come in a variety of types, including independent chartering boards, nonprofit organizations, institutions of higher education, and
State and local education agencies. The charter authorizers can then approve
charter operators to run schools.
Most charter school operators are independent entities (e.g., a group of
teachers or parents, or a local nonprofit or community organization), but about
35 percent of charter schools are operated by nonprofit or for-profit management organizations as part of larger networks of schools. Examples of such
networks include the Knowledge Is Power Program (KIPP) and Charter Schools
USA. Charter school operators are accountable to the organization that grants
their charter, and the schools are subject to periodic reviews (Kaplan and
Owings 2018; David 2018).
Because public charter schools operate outside the restraints that bind
a DPS system, they can be more innovative than DPSs. Grube and Anderson
(2018) discuss innovations such as Montessori schools and dual-language
immersion schools. Charter schools based on the “no-excuse” approach have
proved popular in urban settings. These schools include features such as
uniforms, strong discipline, extended classroom hours, and intensive tutoring
(Angrist et al. 2016). The KIPP Foundation is known for these schools.
In terms of enrollment, charter schools typically have a mandate to
accept all eligible applicants and to use a lottery to select students if they are
oversubscribed. The charter school segment has grown rapidly. Between the
2000–1 and 2017–18 school years, the number of charter schools increased by
about 260 percent, with student enrollment increasing about 600 percent. As
of the fall of 2017, more than 6 percent of students enrolled in public elementary and secondary schools attended these institutions. Enrollment in charter
schools is generally higher in urban areas where minority and low-income
student populations are higher (NCES 2019a).
Magnet schools are public schools that offer specialized programs meant
to bring together students with common interests or skillsets. These schools
specialize in specific areas, such as mathematics, science, and the performing
arts. Some magnet schools also include niche subjects, such as the culinary
arts or aerospace engineering. Originally, magnet schools were created to

206 |

Chapter 7

foster desegregation by intentionally enrolling students from diverse populations (OII 2004). Many magnet schools continue to serve this mission.
Enrollment in magnet schools is handled through various application
frameworks. Some magnet schools have attendance zones, where part of the
student population is enrolled based on geographic location while the remaining slots are filled by applicants from throughout the rest of the district. Other
magnet schools do not have attendance zones and instead grant all seats
through an application process. Admission may also be handled through a
lottery system; some magnets use random lottery systems, while others use
weighted lottery systems that prioritize students with certain qualifications
(OII 2004; Ayscue et al. 2015). As of the 2017–18 academic year, there were 3,421
magnet schools in operation, enrolling 2.7 million students (NCES 2019a).
Open enrollment school districts facilitate interdistrict or intradistrict public school choice, which allows students to select the school they wish to attend
instead of taking a DPS assignment. Intradistrict policies allow choice among
schools within a student’s designated district, while interdistrict policies give
students the option to attend schools within a State or larger defined region
(EdChoice 2020b). Open enrollment programs help households by giving
students access to higher-quality public schools while also providing competition between public schools. Still, not all States and school districts cover the
costs of traveling to nonneighborhood schools, and this may pose a barrier to
some families, limiting their ability to exercise choice. As of 2018, most States
had enacted policies related to open enrollment. In 34 States, school districts
choose whether to participate, while 28 States mandate open enrollment in
some cases (ECS 2018).
Other types of competition with DPSs include homeschooling and virtual
school. The Department of Education defines homeschooling to include students who attend less than 25 hours of public/private school weekly. As of
2016, 1.7 million students were homeschooled, representing 3.3 percent of all
students, up from 1.7 percent in 1999 (figure 7-3; NCES 2019b). Virtual school
may include a hybrid of in-person and online instruction or be a fully online curriculum run by either private or public schools. In 2017, about 280,000 students
were enrolled in virtual school (NCES 2019a). However, during the coronavirus
pandemic, many more students are experiencing some form of virtual learning
(EdSurge 2020).

Examples of School Choice
Although school choice has grown in communities throughout the United
States, it is instructive to compare how school choice developed in specific
places. Here, we consider several examples where school choice has come to
play a particularly prominent role in the education environment: Milwaukee,
Florida, New Orleans, Massachusetts, and the District of Columbia.

Expanding Educational Opportunity through Choice and Competition

| 207

Milwaukee. The Milwaukee Parental Choice Program was established
in 1990 as a voucher program targeted at low-income students. Initially, the
program was restricted to families with incomes below 175 percent of the
Federal poverty line. Enrollment was also limited to 1 percent of students in
the Milwaukee public school district (MPS), with a randomized selection process for most students. The program did not initially include religious private
schools, which made up about 80 percent of private school student enrollment
in the area (Witte 1998). After 1998, when the Wisconsin Supreme Court ruled
that the inclusion of religious schools in the voucher program did not violate
the Wisconsin Constitution, the voucher program grew more quickly. Today,
the program has expanded to allow families with incomes up to 300 percent of
the poverty level, has no enrollment cap, and uses a lottery system for selection
when particular schools are oversubscribed. As of 2019, there were 120 schools
participating and more than 28,900 students enrolled (EdChoice 2020c).
Milwaukee introduced charter schools in 1996. Today, Milwaukee has 44
charter schools with an enrollment of more than 18,000 students (Wisconsin
Department of Public Instruction 2020). Some of the schools are authorized
by the MPS, while others are independent of it and are authorized by the
University of Wisconsin–Milwaukee and the City of Milwaukee. The independent charter schools have more autonomy than those run by the MPS and
make up almost half of all charter schools in Milwaukee. Milwaukee’s choice
schools have spurred new educational approaches. In 1999, the MPS began to
introduce changes, such as opening new Montessori schools in response to the
competitive pressure (Grube and Anderson 2018).
Florida. In 2001, the Florida legislature established the Florida Tax Credit
Scholarship program. The program offers State tax credits to corporations that
donate to nonprofit scholarship-funding organizations. The scholarships can
be used for tuition and fees at private schools or for transportation to a public
school outside a student’s residential school district. When the program was
initially passed, only students with household incomes below 185 percent of
the Federal poverty line were eligible, and the program expenditures were
capped at $50 million annually. The program has since been expanded so that
students with household incomes between 200 and 260 percent of the Federal
poverty line are eligible for partial scholarships, while students from lowerincome families are eligible for full scholarships. In the 2018–19 school year,
nearly $645 million in scholarships was awarded to 104,091 students attending
1,825 participating private schools (Florida Department of Education 2019c).
Florida has also established other State-funded school choice programs.
The McKay Scholarship Program was established in 2000 as the Nation’s first
school voucher program for students with special needs. The program provides
scholarships for students to attend a private school or to transfer to a different
public school. In 2018–19, the program awarded about $220 million in scholarships to 30,695 students. The Gardiner Scholarship Program, established in
208 | Chapter 7

2014, is an ESA program that provides funds for special needs students to purchase products and services to support their learning. In 2018–19, the program
awarded about $125 million in scholarships to about 12,188 students. The
Family Empowerment Scholarship program, which was established this past
year, provides scholarships for up to 18,000 students from disadvantaged families to attend private schools, with priority for students from households with
incomes less than 185 percent of the Federal poverty line. Florida also operates
the Statewide virtual school known as the Florida Virtual School. This is the
largest virtual school in the country, with 215,505 students enrolled during the
2018–19 school year. Finally, Florida has a large public charter school sector
that enrolled 313,000 students in the 2018–19 school year (Florida Department
of Education 2019a, 2019b).
New Orleans. Before Hurricane Katrina, the New Orleans public schools
had one of the worst performance records in the country. In the 2004–5 school
year, only 35 percent of students in the New Orleans schools achieved proficient scores on State exams, and high school graduation rates were about 54
percent (Teach New Orleans 2020). In 2003, the New Orleans Recovery School
District (RSD) was created as a way to reform the public schools. In 2005, in
response to the devastation left by Katrina, the RSD assumed control of 114
low-performing schools. With the help of $20.9 million in funding from the
Department of Education, New Orleans began to open new charter schools.
Over time, the RSD eliminated some of the low-performing schools and converted others to charters. By 2014, all the RSD schools were charter schools,
and nearly all educators had been replaced. Furthermore, district public school
attendance zones were eliminated, making New Orleans the only all-choice
school system in the country. The reforms led to dramatic gains among New
Orleans schoolchildren. By the 2013–14 school year, student proficiency on
State exams had increased to 62 percent. High school graduation rates, college entry rates, and college graduation rates all rose substantially (Harris and
Larsen 2018).
In 2008, Louisiana also launched a voucher program for students in New
Orleans, known today as the Louisiana Scholarship Program. The program was
targeted to children in failing schools with family incomes at or below 250 percent of the poverty line. Over the first four years, it grew slowly, reaching about
1,900 in annual vouchers in the 2011–12 school year. In 2012, the program was
expanded to the rest of the State, and by 2014, more than 6,500 vouchers were
awarded.
Massachusetts. In 1993, the Massachusetts legislature passed the
Education Reform Act, increasing the State’s role in education. The act allowed
for the creation of charter schools, reserving the right to authorize them for
the Massachusetts Department of Elementary and Secondary Education. In the
2019–20 school year, 81 charter schools operated in Massachusetts, educating
just under 48,000 students. The schools have proved popular, and spots are
Expanding Educational Opportunity through Choice and Competition

| 209

typically allocated by lottery. In the 2019–20 school year, 73 charter schools
had waiting lists, and there were nearly 28,000 students on one or more of
the lists (Massachusetts Department of Elementary and Secondary Education
2016, 2019). In 2010, Massachusetts passed legislation allowing charter schools
with a successful track record to expand. In Boston, the number of charter
schools doubled as a result. Despite the large expansion, a recent study finds
that the new schools generated achievement gains on par with the original
charter schools (Cohodes, Setren, and Walters 2019).
For grades 1 through 8, Boston’s public school system also facilitates
school choice through an assignment system, known as the Home-Based
School Assignment Policy (Boston School Finder 2020). The policy seeks to
find a balance between allowing students to attend a neighborhood school
and giving more students a chance to attend high-quality schools. Families
choose from a list of schools and are allowed to express preferences. A lottery
mechanism is used to assign students to schools, taking these preferences into
account. Over time, the mechanism has been adjusted in response to concerns
about student travel times and persistent racial inequities (Abdulkadiroğlu et
al. 2006).
District of Columbia. The voucher program in the District of Columbia is
the only private school choice program run by the Federal Government. Signed
into law by President George W. Bush in 2004, the D.C. School Choice Incentive
Act created the D.C. Opportunity Scholarship Program, which is intended to
improve education in the District of Columbia, especially for disadvantaged
students. Although Congress has continuously funded the program, the
Obama Administration sought to phase it out and prevented new students
from enrolling in the 2009–10 and 2010–11 school years. However, the Trump
Administration has strongly supported the program and helped increase participation by more than 40 percent between the 2016–17 and 2017–18 school
years to more than 1,600 students (CRS 2019). In the 2020–21 school year,
vouchers are worth up to $9,161 for K-8 students and up to $13,742 for high
school students. In addition to the voucher program, Washington has a large
public charter school sector that dates back to 1996. In the 2019–20 school
year, these schools enrolled more than 43,500 students in grades pre–K-12 and
adult learning programs (DCPCSB 2020).

The Role of Federal Policy in School Choice
In this subsection, we discuss the role of Federal policy in school choice. We
first provide an overview of the organization of K-12 education, its funding,
and the Federal contribution. We then describe the main Federal policies that
are related to school choice. Finally, we highlight recent actions of the Trump
Administration to further support and expand school choice.
State and local governments have the primary responsibility for K-12
education in the United States. Along with public and private organizations,
210 |

Chapter 7

$"0- цҊуѵ -'0)$)"+ -/0 )/!*-0'$-$(-4
) *)-4#**'.4// Ѷспрх–рц

‫ڦ‬цшш*-' ..
‫ڦ‬чпп/*‫ڦ‬шшш
‫ڦ‬рѶппп/*‫ڦ‬рѶршш

‫ڦ‬рѶспп/*‫ڦ‬рѶтшш
‫ڦ‬рѶупп/*‫ڦ‬рѶфшш
‫ڦ‬рѶхпп*-(*-

*0- ѷ/$*)' )/ -!*-0/$*)//$./$.ѵ

they establish new schools, determine graduation standards, and develop curricula (U.S. Department of Education 2017). State and local governments also
provide the primary financing for K-12 education. Of the $736 billion allocated
for public elementary and secondary education for the 2016–17 school year,
the majority was allocated from State and local governments—47 percent and
45 percent, respectively. Only $60 billion (8 percent) was from Federal sources
(NCES 2020a).
More than half of Federal funding from the Department of Education supports students with low family incomes or disabilities. A large share of Federal
funds, about 26 percent, is spent on Title I grants, which supplement State and
local funding in school districts with a high share of low-income students. An
additional 20 percent of Federal funding focuses on children with disabilities
(NCES 2020b). Because States have different demographics, Federal funding
per student varies by State. Figure 7-4 shows the Federal funding per student
for public primary and secondary schools by State (NCES 2019c).
The Federal Government also provides resources to support school
choice programs. The Department of Education oversees the key Federal
programs that promote choice. These include the Magnet Schools Assistance
Program and the Charter Schools Program, as well as the District of Columbia
Opportunity Scholarship Program discussed above (U.S. Department of
Education 2019a, 2019b, 2020a).
The Magnet Schools Assistance Program provides funding for magnet
schools that are part of an approved plan to desegregate schools. These
magnet schools are designed to bring students from different backgrounds
Expanding Educational Opportunity through Choice and Competition

| 211

$"0- цҊфѵ0( -*!#-/ -#**'.+ ) Ѷ сппх–ц/*спрх–рц
Number of charter schools opened

*/!0) 4/#

Not funded by the CSP

Funded
by the CSP
0) 4/#

500

*0- ѷU.S. +-/( )/*!0/$*)ѵ
*/ ѷۙ#-/ -#**'.-*"-(ѵ

фпп

1000

рѶппп

1500

2000

2500

3000

рѶфпп
сѶппп
сѶфпп
Number of schools

3500

тѶппп

4000

тѶфпп

уѶппп

together to reduce minority group isolation in schools with many minority students. As discussed, magnet schools are typically focused on specific academic
areas—such as science, technology, engineering, and mathematics, known
as STEM; the arts; language immersion—or implement alternative teaching
philosophies, such as international baccalaureate programs or Montessori
methods (U.S. Department of Education 2020a).
The Charter Schools Program (CSP) was established in 1994 to encourage the formation of new charter schools. Its initial budget of $4.5 million
had grown to to $440 million as of 2019. In 2019, 85 percent of CSP funding
went toward the creation of new charter schools, although the program also
supports the expansion of existing charter schools. Between the 2006–7 and
2016–17 school years, the majority of new charter schools were started using
funding from the CSP (figure 7-5). Charter schools funded by the CSP serve a
higher percentage of Hispanic and Black students than residentially assigned
district public schools (table 7-1) (U.S. Department of Education 2015, 2019b).
Some Federal programs interact with other Federal initiatives to enhance
the total impact of Federal Government action. For example, earlier in 2020,
the Department of Education provided an additional $65 million through the
CSP, targeting charter schools in Opportunity Zones. These distressed areas
have high poverty rates and low incomes. Currently, less than 30 percent
of Opportunity Zones have at least one charter school. Additional funding
for charter schools will complement tax incentives in Opportunity Zones to
spur much-needed economic opportunity in these areas (U.S. Department of
Education 2020b). A recent CEA report provides an initial assessment of the
Opportunity Zone program (CEA 2020).
The Trump Administration has sought to further aid the expansion of
State-based school choice. President Trump supports the proposed Education
Freedom Scholarships and Opportunity Act, which would give Federal tax-credits

212 |

Chapter 7

Table 7-1. Demographics of CSP-Funded Schools and District Public Schools, 2016–17
Ҋ0) S#**'.җnۙтѶрсшҘ
/0 )/.
- )/
уфуѶхпф
ттѵш
$.+)$
'&
упшѶпрп
тпѵф
#$/
тцрѶухс
сцѵц
.$)
фпѶхтц
тѵч
2**-(*- - .
утѶурп
тѵс
( -$) )$) ҝ '.&/$1
цѶцфш
пѵх
2$$)/$1 ҝ $!$ .') уѶфшц
пѵт
*/')0( -*!./0 )/.
рѶтурѶучп
рпп
(*"-+#$.

.$ )/$''4A..$") җnۙччѶтспҘ
/0 )/.
- )/
рсѶсчсѶпчф
схѵп
хѶшпрѶпут
руѵх
стѶсхчѶуут
ушѵс
сѶуупѶшчх
фѵс
рѶцршѶццу
тѵх
учсѶпчч
рѵп
рцсѶрст
пѵу
уцѶсххѶфус
рпп

*0- ѷU.S. +-/( )/*!0/$*)ѵ
*/ ѷۙ#-/ -#**'.-*"-(ѵ

to individuals and businesses that contribute to organizations that grant scholarships to students. States would designate qualified expenses, which can be
for primary, secondary, career, and technical education. Expenses that might
be recognized include advanced, remedial, and elective course fees; tuition at
private schools; apprenticeships and industry certifications; concurrent and
dual enrollment for college credit; private and home education; special education services and therapies; transportation to education providers outside of
a family’s zoned school; tutoring, especially for students in low-performing
schools; and summer and after-school education programs (U.S. Department
of Education 2019c). Individual taxpayers would be allowed to redeem a credit
of up to 10 percent of their adjusted gross income, and corporations would be
allowed to redeem a credit of up to 5 percent of their taxable income. States
would have the responsibility to recognize scholarship-granting organizations
and to decide the nature of the scholarships and eligibility criteria. For States
with existing tax-credit scholarship programs, the program would incentivize
additional private donations. State participation would be voluntary (Office
of Senator Ted Cruz 2019). For a discussion of related legislation, see box 7-2.
The Trump Administration has also supported school choice as part of
the 2017 Tax Cuts and Jobs Act (TCJA). The TCJA expanded the purview of
the Qualified Tuition Program, known as a 529 program. Dating back to the
1990s, these programs have long allowed an individual to contribute money
to an investment account from which a student’s expenses for higher education (college or vocational school) can be paid without the earnings on the
contributions being taxed. The TCJA allows 529 plans to be used to pay for
primary and secondary education as well. Funds can be withdrawn tax-free
to pay for up to $10,000 in tuition per beneficiary per year at any public, private, or parochial school (CRS 2018). The 529 program offers an alternative to
Coverdell Education Savings Accounts. Coverdell ESAs similarly allow individuals to contribute money to an account that grows tax-free for use on qualified
educational expenses for a student. However, there is a limit on how much can

Expanding Educational Opportunity through Choice and Competition

| 213

Box 7-2. The School Choice Now Act
As part of ongoing COVID-relief legislative efforts, Senator Tim Scott and
Senator Lamar Alexander introduced the School Choice Now Act on July
22, 2020. If passed, the bill would authorize the Department of Education
to allocate 10 percent of its emergency CARES funding to States in the form
of emergency education grants. The bill would also encourage donations to
certain scholarship-granting organizations by establishing Federal tax credits.
Bill proponents point out that more than 100, mostly Catholic, private
schools—which enrolled more than 16,000 total students—have permanently
closed because of the coronavirus. Adding all, or even a significant portion, of
these 16,000 students to the public school population could strain alreadydelicate local government budgets at a critical time (McShane 2020).

be contributed per year (currently $2,000) and a phase-out by annual income
of the contributor (CRS 2018).

School Choice and Competition
As we have seen, school choice programs have expanded significantly in the
United States over the last three decades. In this section, we explain how
school choice promotes competition among schools for students, which can
ultimately lead to improved educational experiences for all children. We begin
by explaining how mobility-based school choice has long existed in the DPS
system for affluent families that can afford to move to higher-quality school
districts. We next discuss how school choice programs introduce a different
form of competition that can raise the educational quality received by all students, including those in less affluent neighborhoods, whether they participate
in a choice program or remain in a DPS.

Competition between School Districts
Even before the school choice movement, parents exercised substantial discretion over where their children attend school, as families with the means to do
so can move to higher-quality school districts. Tiebout (1956) describes how
local governments compete with one another by adjusting levels of public
good provision to meet the demand of potential residents. The model implies
that families with the financial means to do so will live in certain kinds of
communities based on their preferences for various public goods, including
education. Localities will generally face pressure to increase quality, subject to
a given level of cost, in order to increase their number of taxpaying residents.
In the United States, school quality is an important factor for many parents
in determining where to live. According to a 2018 survey, more than half (52
percent) of recent homebuyers with children considered the quality of schools

214 |

Chapter 7

when making their decision on the neighborhood in which to live (National
Association of Realtors 2018). Studies of parental choices find that preferences
for school quality are multidimensional, with parents placing different weights
on factors such as academic performance, teaching quality, class size, safety,
and discipline (Chakrabarti and Roy 2010).
Although this form of competition allows more affluent families to seek
out a higher-quality education, the benefits may be financially out of reach for
many less-affluent families. In fact, a Senate committee report finds that the
median home price in neighborhoods with the highest-quality schools is about
$486,000, which is four times larger than the median price of $122,000 in neighborhoods with the lowest-quality schools (JEC 2019). In addition, this form of
competition can potentially have negative consequences for lower-income
and minority students when it leads to increased segregation along racial and
income lines (Urquiola 2005; Rothstein 2006).
Two important lessons emerge from the research on traditional mobilitybased competition. First, because of the financial barriers that lower-income
students’ families face to moving, competition between districts for residents
is likely insufficient on its own to substantially improve school quality for such
students. Thus, additional school choice within a district may be needed to
foster the procompetitive effects that hold schools accountable for delivering better outcomes. Second, to avoid negative consequences for students
who remain in a residentially assigned DPS, school choice programs should
be designed so that schools compete on their value added and not on their
ability to siphon off students with more advantaged backgrounds from DPSs.
Nonetheless, even when some selective sorting occurs, school choice programs may still outperform the traditional mobility-based system that relies on
sorting across neighborhoods based in part on family affluence.

Designing School Choice Systems
We next explain why carefully designed school choice programs can improve
the quality of education for all students within a district. In early research
on school choice, Hoxby (2003) developed models to explain why exposing
DPSs to competition can lead to better schools. The models characterize
decisionmaking for different types of schools, but all of them rest on the basic
assumption that for a given out-of-pocket cost, parents choose the school they
value the most. This power of choice is the mechanism by which the beneficial
effects of competition come in to play.
To illustrate, consider the case of a charter school that can receive a government payment for each student it attracts away from a DPS, although it is
prevented from charging additional fees. Because the school cannot compete
on price, it must compete on quality. By increasing its quality, the school can
attract more students from DPSs and thus earn more revenue. Because quality
is expensive to produce, the school faces a trade-off between attracting more

Expanding Educational Opportunity through Choice and Competition

| 215

students and keeping costs down. Where the school falls on this trade-off will
depend on the structure of both demand (how much parents and students
value quality) and supply (how expensive quality is to supply). At a minimum,
however, a charter school must provide parents and students with at least as
much value as the DPS.
This direct effect of school choice on program participants is distinguished from the indirect competitive effect that arises when a DPS responds
to the competitive pressure from a charter or other choice school. As discussed
by Hoxby (2003), if a DPS loses funding when students enroll in a different
school, it should have an incentive to retain students. The strength of the
effect will depend both on how much funding is lost and how many students
are threatening to leave. If the competitive pressure is not robust, a DPS may
not react to it. For example, choice programs that cap enrollment at low levels
may not place much pressure on DPSs that are to a large extent guaranteed
enrollment. This is quite different from the direct effect of school choice: any
charter school, no matter how small, must compete with DPSs in order to
enroll students.
When DPSs respond to competitive pressure from choice programs by
increasing their value added, their students will benefit. However, it is not axiomatic that school choice will inevitably improve educational quality through
competition in the same way that is often true in other markets. This is because
of the important role that peer effects play in education provision. In most markets, the quality of a service that one person receives is not directly affected by
the characteristics of the other people that consume that service. However, in
education, students are both consumers and producers of education quality—
not merely passive consumers—because of how peers have a direct impact on
the education quality received by their classmates. As a result, school choice
reforms that induce shifts in student sorting across schools could have large
effects both on the students who leave and on those who remain in residentially assigned DPSs because of changes in peer composition. If a school choice
program attracts highly motivated or affluent students, it might have negative
effects on students who remain in DPSs surrounded by less-motivated, lessaffluent peers.
However, as actually implemented, school choice programs often incorporate features designed to prevent such an outcome from occurring, and
the empirical evidence finds little support for these theoretical concerns.
For example, voucher programs often restrict participation to students from
low-income families or provide resources that are more generous to relatively
disadvantaged students, which directly limits the potential for income-related
sorting. Designs that require oversubscribed schools to use lotteries to allocate
slots also limit the potential for selective sorting. The result is that schools will
compete based on value added rather than on their ability to select students.
In the United States, many school choice programs incorporate these design
216 |

Chapter 7

features (Epple, Romano, and Urquiola 2017). Expanding school choice to more
students while providing more generous assistance to disadvantaged families
could further build upon these benefits (Epple and Romano 2008).

Evidence on the Impact of School Competition
Finally, we turn to the important question of whether the theoretical procompetitive benefits of school choice are borne out empirically. This evidence can
help determine whether school choice program designs in different contexts
have in fact promoted positive student outcomes, and it can help inform
better school choice policies in the future. We first discuss the literature that
examines the direct effects of school choice programs on the students who
participate in them. We then discuss the literature that examines the indirect
competitive effects of school choice programs on DPS students.

Direct Value-Added Effects
That charter schools must compete with DPSs is so basic that there is a sense
in which we should not be surprised by evidence that they do. Simply put,
if a charter school or voucher program does not offer parents and students
additional value relative to the DPS, then it is unlikely to thrive (Hoxby 2003).
Consistent with this, a recent survey finds that a larger share of children in private schools (77 percent) or in chosen public schools (60 percent) have parents
who report being “very satisfied” with their schools compared with children in
assigned public schools (54 percent) (NCES 2019b). That said, there is value
for policymakers in understanding which types of school choice programs
have worked best and for whom, as they test new approaches and encourage
expansion of the most promising programs. Accordingly, a large body of literature has arisen to study the direct effects of school choice programs on their
students. In our discussion, we divide the literature into studies of academic
achievement as measured by test scores, studies of academic attainment such
as graduation rates or college enrollment, studies of racial and ethnic integration, and studies of longer-term, nonacademic outcomes. We also discuss
studies of the fiscal impact of choice programs.
Much of the literature on academic achievement focuses on student performance on national or Statewide tests. This may be surprising in light of the
small role such tests play in the educational experiences of students. However,
these test scores are more amenable to empirical study than measures that
vary widely across schools, such as grades or pass rates for courses. Tests that
are taken by most students are also more amenable for study than scores on
national tests, such as the Scholastic Aptitude Test, that are taken by only a
subset of high-performing students (Hoxby 2003). National accountability systems also emphasize performance on national and Statewide tests. However,

Expanding Educational Opportunity through Choice and Competition

| 217

test scores have been criticized as being a poor measure of the educational
experience (Hitt, McShane, and Wolf 2018).
For private school choice programs, although results vary by individual
study, Epple, Romano, and Urquiola (2017) conclude that most studies of
voucher programs in the United States have not revealed large or statistically
significant test score improvements for students in general. However, multiple
studies uncover substantial test score improvements among Black students.
For example, Mayer and others (2002) find that the School Choice Scholarship
Foundation voucher program in New York City did not yield higher test scores
on average for all students; but for Black students, test scores increased by
about 6 percentage points. One notable exception to the positive findings for
Black students is the Louisiana Scholarship Program (LSP). Abdulkadiroğlu,
Pathak, and Walters (2018) find that the LSP led to a large decrease in math
test scores among participants. They link this to the selection of low-quality,
low-tuition schools into the program, pointing to the importance of program
design for the success of private school choice.
For charter school programs, Epple, Romano, and Zimmer (2016) conclude that researchers have not reached a consensus on their effectiveness
for academic achievement. Broad studies, including those by the Stanford
Center for Research on Education Outcomes (CREDO 2009, 2013), do not reveal
large or statistically significant test score improvements for charter school
students on average (these studies compare test scores of students in charter
schools with those of students that have similar observable attributes—“virtual
twins”—in the fallback DPS). However, numerous studies of programs in urban
areas find large, statistically significant gains. In particular, most studies of
oversubscribed charter schools find positive effects on test scores. These studies are noted for the strength of their research design (they compare students
who win the lottery with students that have similar observable attributes
who lose the lottery; Epple, Romano, and Zimmer 2016). A recent study of a
Massachusetts law allowing charter schools with a successful track record
to expand finds that the new schools generate gains on par with the original
schools (Cohodes, Setren, and Walters 2019). A study of Texas charter schools
suggests that the effectiveness of charter schools is increasing over time, as
successful charter schools expand and poorly performing schools exit (Baude
et al. 2020). In both Massachusetts and Texas, many of the successful charter
schools that have expanded use the “no excuses” approach, which features
strong discipline, extended classroom instruction, and intensive tutoring. Of
note, in a Boston study, Walters (2018) finds evidence that charter expansion
programs are particularly effective when they target students who are unlikely
to apply, including low achievers, as these students have the most to gain.
In comparison with test score studies, studies of educational attainment
on the whole find more encouraging results. This is true for both voucher
programs and charter programs (Epple, Romano, and Urquiola 2017; Epple,
218 |

Chapter 7

Romano, and Zimmer 2016). For example, in a study of the Washington
Opportunity Scholarship Program, Wolf and others (2010, 2013) estimate that
vouchers raise high school graduation rates by 21 percentage points. As with
test scores, vouchers appear to have an even more beneficial impact on the
graduation rates of Black students. Shifting attention to college, Chingos and
Peterson (2015) identify a 6-percentage-point boost to enrollment rates among
Black students offered vouchers in New York City, although they find less evidence of an effect among a broader group of students. Sass and others (2016)
find that students in Florida’s charter schools stay in college longer than students in DPSs, reinforcing related findings by Booker Sass, and Zimmer (2011).
Dobbie and Fryer (2015) find that students admitted to a high-performing
charter school in the Harlem neighborhood of New York City are more likely
to graduate from high school on time and enroll in college immediately after
graduation, although they ultimately attain about the same amount of college
education as DPS students.
The divergence between the results on test scores and the results on
educational attainment leads some researchers to question whether test
scores are a useful yardstick for evaluating school performance. Hitt, McShane,
and Wolf (2018) review studies of a wide variety of school choice programs that
measure test scores and educational attainment as part of the same study.
They find little within-study correlation between results on test scores and
educational attainment. Epple, Romano, and Zimmer (2016) also comment on
this divergence, pointing to Wolf and others (2010) as an example of a study
that finds no significant effects on test scores but strong positive effects on
high school graduation rates.
Some studies address the impact of school choice on racial and ethnic
integration. By separating the decision about where to attend school from the
decision about where to live, school choice has the potential to reduce the role
of income and race disparities in providing educational opportunity. Many
school choice programs began in areas with high concentrations of minority and low-income students specifically to serve the needs of underserved
communities with often poorly performing DPSs. As a result, charter schools
educate a disproportionate number of such students relative to the national
average. In the 2017–18 school year, Black and Hispanic students accounted
for 26 percent and 33 percent of charter school enrollment, respectively, while
accounting for only 15 and 27 percent, respectively, of enrollment across all
public schools (NCES 2019d).
Regarding the impact of school choice on racial and ethnic stratification, Epple, Romano, and Zimmer (2016) discuss a large body of research
and conclude that charter schools and public schools exhibit similar degrees
of racial and ethnic segregation, with charter schools more likely to have a
disproportionately nonwhite student population and DPSs more likely to have
a disproportionately White student population. Moreover, Zimmer and others
Expanding Educational Opportunity through Choice and Competition

| 219

(2009) conclude that charter schools have only modest effects on the racial
mix of schools. Butler and others (2013) analyze the decision to attend charter
schools and find a role for socioeconomic characteristics but not race as a
driving factor. In terms of voucher programs, a recent study by Egalite, Mills,
and Wolf (2017) assessing the Louisiana Scholarship Program finds that most
students using the vouchers reduce racial stratification in the public schools
that they leave and have only small effects on racial stratification in the schools
to which they transfer. In addition, in school districts under Federal desegregation orders, voucher transfers cause a large drop in DPSs’ racial stratification
levels but have no impact on private schools.
Another body of literature focuses on the long-run benefit of school
choice programs on outcomes such as civic engagement and criminal behavior. These studies are relatively rare because they require data from a period
of many years. Two studies of the Milwaukee Parental Choice voucher program
are worthy of note. DeAngelis and Wolf (2019) use data from the Milwaukee program to compare young adult voting behavior between program participants
and similar students in DPS. They find no evidence of statistically different voting patterns, helping to allay potential concerns that private choice programs
might provide less instruction in citizenship skills. DeAngelis and Wolf (2020)
use the data from the Milwaukee program to analyze the prevalence of criminal
activity. They find some evidence that voucher program participants are less
likely to be involved in criminal activity relative to DPS students, including a
large and statistically significant reduction in property damage convictions.
Dobbie and Fryer (2015) find that for students admitted by lottery to a highperforming charter school in Harlem, New York City, female students are less
likely to be pregnant as teenagers and male students are less likely to be incarcerated in comparison with similar students who are not admitted.
The discussion so far has focused on student outcomes in school choice
programs. However, another relevant question is how much money is spent to
achieve these outcomes. Several studies find that charter schools and voucher
programs educate program participants at a lower cost per-pupil than DPSs.
DeAngelis and others (2018) find that across 14 metropolitan areas, public
charter schools received on average $5,828 less revenue per pupil than DPSs in
the 2016 fiscal year. In a study of 16 voucher programs, Leuken (2018) finds that
the voucher programs generated average savings of almost $3,100 per voucher
recipient for State and local budgets in fiscal year 2015.

Indirect Procompetitive Effects
Finally, we turn to studies of the indirect procompetitive effects of school
choice programs on DPSs. Such studies face several challenges. First, the penetration of choice programs in many areas of the United States is simply too
small for robust procompetitive effects to have a reasonable chance of emerging. A DPS is little affected if it is only at risk of losing a handful of students
220 |

Chapter 7

to a choice program, in which case it does not face much market pressure to
improve. Where school choice programs have reached sufficient penetration
to enable study, researchers must try to distinguish competitive efforts by a
DPS to improve its quality from effects related to changes in the DPS student
composition and effects related to changes in DPS funding. However, the
recent growth of school choice programs is enabling a growing number of welldesigned empirical evaluations.
Figlio and Hart (2014) and Figlio, Hart, and Karbownik (2020) study the
impact on DPS students of the Tax Credit Scholarship program in Florida. The
latter study focuses on the scaling up of the program in recent years. They
exploit the fact that some students are more exposed to this expansion due to
the differing availability of nearby private schools before the policy is implemented. Public school students who are more heavily exposed to competition
experience improved test scores as well as fewer suspensions and absences.
Positive effects of increased competition from private schools are largest for
students from low-income families whose parents have lower educational
attainment. In addition, procompetitive effects on public schools increase
over time as the program scales up. Similarly, Chakrabarti (2008) finds that the
expansion of the private voucher program to religious schools in Milwaukee
leads to larger increases in public school test scores. In a review of the literature, Epple, Romano, and Urquiola (2017) conclude that studies generally find
that private school vouchers improve the performance of students in DPSs.
They also find little evidence that school choice gives rise to adverse sorting.
There are examples where the students who leave DPSs for a voucher program
are of higher, lower, or equal ability relative to the peers they leave behind.
Moreover, because voucher programs are often targeted to lower-income families, voucher students tend to come from families with lower or equal income
to their peers in the DPSs that they leave behind. The empirical evidence on the
positive returns from scaling up voucher programs and the absence of adverse
sorting effects suggests that many more students could benefit from an expansion of voucher programs.
Positive competitive effects also arise for charter schools. Gilraine,
Petronijevic, and Singleton (2019) find that when North Carolina lifted caps
on new charter schools, students who lived closer to new charter schools
experienced larger gains in test scores. Ridley and Terrier (2018) find that the
Massachusetts reform that raised the cap on charter schools led to increased
spending per pupil in DPSs and a shift in DPS spending from support services
to instruction, and they similarly find small positive effects of charter schools
on the test scores of DPS students. Dispelling concerns related to sorting and
peer effects, Epple, Romano, and Zimmer (2016) synthesize the findings from
several studies showing that students who transfer to charter schools have a
similar or slightly lower ability relative to the DPS from which they are drawn.
In a survey of further research on this topic vis-à-vis charter schools, Anderson

Expanding Educational Opportunity through Choice and Competition

| 221

(2017) finds that charter schools often serve lower- or similarly performing
students than DPSs.
Thus, the evidence from voucher and charter school studies alike suggest that there is little evidence to warrant fears about DPS students being left
behind. Instead, such students tend to benefit from the improvement in their
own schools that comes about from choice-induced educational competition.
As a final note, we return to the question of the fiscal effect of school
choice programs on DPSs. As discussed, several studies find that public charter
schools and voucher programs educate students with less per-pupil public
funding than DPSs. This implies that when a student switches from the DPS
to a choice program, the school district realizes savings that could be used to
improve DPS education, although there may be an adjustment period before
a district can realize those savings (Epple, Romano, and Zimmer 2016). To
date, there has been relatively little research on this topic; but see Bifulco and
Reback (2014) and Ladd and Singleton (2020) for case studies of New York and
North Carolina. Buerger and Bifulco (2019) find that New York State school
districts with larger charter school enrollments experience decreases in the
cost of providing DPS education, both in the short run and the long run, though
districts with only a small charter school presence can experience short-run
increases in costs that are subsequently offset by efficiency gains. Some States
provide temporary funding increases to DPSs to help them adjust to charter
school expansion, including Massachusetts, as documented by Ridley and
Terrier (2018).

Conclusion
School choice programs have grown dramatically over the past 30 years
as evidence has accumulated about the benefits they provide. Parents are
increasingly choosing alternatives to their assigned DPS as they seek out a
higher-quality educational experience for their children. Federal policy has
long supported school choice, both in the Trump Administration and in earlier
administrations on both sides of the political aisle.
School choice can level the playing field and provide enhanced educational opportunity to all families, particularly when implemented to maximize
competition and facilitate participation by disadvantaged students. The
alternative to this modern form of school choice for everyone is the traditional
system of school choice for the affluent and mobile, whereby those with financial means relocate to districts with better schools. In the traditional system,
lower-income and minority students are disproportionately left behind in
lower-performing schools, while other families may move away from neighborhoods that they enjoy solely to gain access to better schools. School choice
programs that provide students with choices of public, charter, magnet,
private, or home school can improve quality for all students, including those

222 |

Chapter 7

who remain in DPSs that are forced to adapt because of competitive pressure.
Emerging empirical evidence has identified these positive effects at work in the
United States.
As school choice continues to expand, lessons from existing programs
can inform ways to maximize the benefits for children from all backgrounds.
Research suggests that low-income and minority students tend to enjoy the
greatest benefits, and the evidence on procompetitive effects suggests that
substantial gains are possible from scaling up school choice. As a result,
continuing to grow school choice programs is a promising way to reduce
opportunity gaps and create a level playing field for all children. Research also
suggests that a broader set of metrics should be used to assess school choice
programs beyond standardized test scores in light of evidence that such choice
programs can improve outcomes later in life. Parents themselves are also a
source of wisdom in that they can incorporate other aspects of quality into
their decisionmaking than the criteria that are officially measured. Ultimately,
a focus on expanding opportunity for all students combined with a commitment to innovation that is grounded in evidence can help improve educational
quality for all children.

Expanding Educational Opportunity through Choice and Competition

| 223

x
Chapter 8

Exploring New Frontiers in Space
Policy and Property Rights
The United States has been on the cutting edge of space exploration since the
dawn of the space age and has become the world leader in commercial activity
in space. In the 20th century, the United States became the first and only nation
to send individuals to the Moon. After the end of the Apollo Program, the United
States pioneered the Space Shuttle, the world’s first reusable spacecraft. Now
American engineers have become the first to demonstrate and operationalize
the capabilities of commercial spacecraft for orbital cargo delivery, first-stage
reusability, and human spaceflight.
In the 21st century, the United States has ushered in a new era of space exploration based on public-private partnerships and the success of private sector
investment in space technologies. The Trump Administration recognizes the
opportunities and benefits afforded by this new era and has advanced policies that encourage private sector innovation, collaboration with commercial
companies, and a regulatory environment more conducive to investment in
space. In doing so, this Administration is not only accelerating the development
of the today’s space industry; it is also laying the foundation for a viable space
economy that can continue to develop and expand in the coming decades.
This past year has seen historic advances in spaceflight and space policy,
even in the midst of the global COVID-19 pandemic. After the reestablishment
of USSPACECOM as a combatant command for the space domain on August
19, 2019, President Trump established the U.S. Space Force (USSF), the sixth
branch of the U.S. military, on December 20, 2019. The mission of USSF is to
organize, train, and equip space forces to “protect U.S. and allied interests

225

in space and to provide space capabilities to the joint force” (USSF n.d.). In
addition, on May 30, 2020, and November 15, 2020, in major milestones for
the partnership between the National Aeronautics and Space Administration
(NASA) and the private sector, SpaceX launched a total of six astronauts from
Cape Canaveral to the International Space Station (ISS). These missions, which
represent the first commercial human spaceflights in history, are an important
step for the private sector’s role in the space economy.
In support of these achievements, the Trump Administration has advanced policies that strengthen investor confidence in the space economy to enable the
private space sector to flourish. These new policies are creating an environment
that spurs investment in innovation and encourages the responsible and sustainable use of space resources. In this spirit, the Administration has released
the Artemis Accords, a practical set of principles that will create a safe, peaceful,
prosperous, and open future in space. The initial tranche of signatories to these
accords was announced on October 13, 2020, and included several other major
spacefaring nations and international partners, with more to follow.
With regard to the economic theory of property rights and the large and diverse
empirical literature on property rights, the Council of Economic Advisers
finds substantial evidence that improving investors’ expectations in a novel
economic sector—like space—increases investment in that sector, leading to
more innovation and greater benefits. The CEA estimates that private space
investment could potentially double in the next eight years, due to President
Trump’s executive actions and other enhancements of property rights in space.

uch of the economic growth over the last five hundred years has
occurred because economic actors have forgone present consumption to invest in the future. In support of this, a core tenet of the common law tradition is to ensure that future gains from investment accrue to the
entities or individuals that make the investment and take on the subsequent
risk. A fundamental role of government in this process is to set rules that create
expectations about what the future holds for investors. Property rights form a
legal and economic basis to support investment and provide a structure for the

226 |

Chapter 8

allocation and management of resources. Although new norms and systems
will evolve with the growth of space exploration, the institution of property
rights will be critical to encourage investment for the long-term development
of the space economy.
This chapter highlights the Trump Administration’s actions to enhance
space property rights and maintain the United States’ position on the frontier of innovation and economic development in space. A cornerstone of the
Administration’s policy is to encourage private investment in partnership with
the Federal Government. The venture capital firm Space Capital estimated that
companies invested $18 billion in space activities in 2019. The CEA projects that
private investment in the space sector will reach $46 billion a year by 2028 as
a result of the policies undertaken to clarify and improve the enforcement of
property rights in space.
Property rights can be thought of as a “bundle of sticks,” with each stick
providing an aspect of the underlying rights that the owner can expect to
receive (Barzel 1997). Sticks, in this case, could refer to the ability to transfer
ownership of an asset, the right to earn income from the asset, or the right to
restrict others from performing certain acts near the asset. As more sticks are
added to the bundle, property rights are further specified, so that the owner
can form more precise expectations of the value of a given investment. As
activity has developed in space resources, new questions have arisen about
property rights in space. The current system of international agreements does
not require major changes, but it does need “carefully drafted additions and
amendments” for clarification (Hertzfeld and von der Dunk 2005, 82). Recent
actions by the Trump Administration seek to provide this clarification.
This chapter illustrates how recent U.S. space policy focuses on ensuring
certainty and predictability for private investments in opportunities beyond
Earth. The first section discusses current issues in space policy, and the second
one addresses recent policy efforts and explains how they provide enhanced
security and enforcement of property rights. The subsequent sections explain
the economics of property rights theory and review the economic literature on
how improving property rights affects investment. The chapter then projects
future investment into space activities accounting for the effect of Federal
Government policies on investment behavior. The chapter concludes by discussing the benefits of selecting the United States as the flag of choice—that
is, the country whose frameworks a business finds most desirable—for space
activity and how regulatory reform makes the market more competitive and
innovative.

Exploring New Frontiers in Space Policy and Property Rights

| 227

Current Issues in Space Policy
and the Space Economy
Today, most economic activity in space consists of satellites transmitting telecommunications and remote-sensing data to devices on Earth and the rockets
launching these satellites into orbit. This orbital network of satellites has
facilitated a variety of civil and economic activity on Earth, including weather
forecasting, climate modeling, city planning, emergency response, precision
agriculture, satellite television, satellite radio, global broadband Internet, and
even app-based ridesharing services.
At this point in time, predicting some future industries in space is possible, but history suggests that anticipating all the emerging industries within
the space economy is impossible. However, we can use recent developments
in current technologies, such as in the satellite and rocket launch industries,
to hypothesize what the future of the space economy could look like. For
example, the process of mineral extraction on the Moon and other celestial
bodies may become profitable as the costs of extraction fall and innovations in
space manufacturing, habitation, and propulsion create a demand for resource
availability in space. Space-based solar power is also a possibility, because
orbiting solar panels can harness the sun’s rays before they dissipate in Earth’s
atmosphere and can generate more electricity than terrestrially based solar
panels. Finally, private companies are hoping to create a market for space tourism through partnerships with the Federal Government and innovations that
lower costs, providing an experience quite literally like none other on Earth.
The beginnings of the space economy date to the mid–20th century,
when the Soviet Union sent Sputnik 1 into orbit in 1957, spurring a flurry of
investment into the space race from national governments. The National
Aeronautics and Space Administration (NASA) began operations in 1958,
sent the first American into space by 1961, and landed the first human on the
Moon in 1969. These accomplishments occurred in parallel with a number of
new United Nations treaties as countries around the globe contemplated the
prospect of widespread activity in space. Because there were few profitable
opportunities for the private sector at the time, the U.S. Government laid the
groundwork for a space economy, and the industry developed based on projects funded by taxpayers (Weinzierl 2018).
Most space activity in the 1970s and 1980s involved the launch and
operation of satellites for commercial telecommunications, reconnaissance,
and surveillance purposes. In 1974, the first satellite of the forthcoming Global
Positioning System (GPS) was launched into orbit (Pace et al. 1995). The
Department of Defense initially utilized the GPS constellation purely for military purposes. However, in 1983, the United States announced that it would
make GPS’s standard positioning service available to the general public at no
cost. This event initiated private, civilian uses of GPS that have since led to the
228 |

Chapter 8

Figure 8-1. NASA Outlays and U.S. Private Investment, 2010–19
Dollars (billions)
25

NASA outlays

U.S. private investment

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

Sources: Office of Management and Budget; Space Capital; CEA calculations.
Note: NASA = National Aeronautics and Space Administration.

creation of countless new firms, technologies, and applications. O’Connor and
others (2019) estimate that GPS has generated about $1.4 trillion in economic
benefits since being made available for civilian and commercial use.
Although governments initially funded all space activities, the private
satellite industry grew throughout the second half of the 20th century as
companies realized the market opportunity for satellite services, such as telecommunications, broadcasting, and data transmission. More recently, private
industry has begun to offer other products and services that had been primarily
owned and operated by the Federal Government, such as space launches, crew
transportation, and remote sensing. For example, since the Space Shuttle was
retired in 2011, private companies such as SpaceX and United Launch Alliance
have provided launch services for civil, commercial, and national security
space systems. Figure 8-1 shows that nongovernmental equity investment in
U.S. space companies is rising in relation to the level of NASA outlays.
This shift from government launch vehicles toward commercial space
launch services accelerated in 2005, when NASA began the $500 million
Commercial Orbital Transportation Services (COTS) program. The COTS program operated on fixed-price payments rather than cost-plus procurement,
which is intended to incentivize innovation and shift NASA’s role from being

Exploring New Frontiers in Space Policy and Property Rights

| 229

Table 8-1. Composition of Global Space Economy, 2019
Percentage of
Spending
Space
(billions of dollars)
Economy

Industry

Good or Service

Satellite

Total

270.7

74.0

Satellite ground equipment

130.3

35.6

Television

92.0

25.1

Fixed satellite services

17.7

4.8

Satellite manufacturing

12.5

3.4

Satellite radio

6.2

1.7

Launch services

4.9

1.3

Broadband
Commercial remote
sensing
MSS

2.8

0.8

2.3
2.0

0.6

Total
U.S. Government’s space
budget
European space budget

95.3

26.0

57.9

16.0

12.0

3.3

China’s space budget

11.0

3.0

Russia’s space budget
Rest of world’s space
budget
Japan’s space budget

4.1

1.1

4.0

1.1

3.1

0.8

Commercial space flight

1.7

0.5

India’s space budget

1.5

0.4

Nonsatellite

0.5

Sources: Bryce Space and Technology; CEA calculations.

an owner and operator to being a customer for resupply services to the
International Space Station. This method of procurement and the use of other
contracting mechanisms have been gaining traction elsewhere in the space
industry, in an effort to decrease costs and benefit from private innovation (box
8-1). Market competition has since provided stronger incentives for innovation
than the existing government monopoly on launches. NASA estimates that the
use of commercial services for ISS resupply services alone has saved taxpayers
between $20 billion and $30 billion since 2011.
Annual estimates of the size of the space economy, incorporating both
public and private activities, range from $360 billion to $415 billion. The current
state of commercial activities still consists primarily of satellites and satellite
services with industry revenues of nearly $270 billion as of 2019, making up 74
percent of the space economy (table 8-1). NASA’s activities and procurements
still drive a large amount of economic activities across the Nation, however,
230 |

Chapter 8

Figure 8-2. Nongovernmental Equity Investment in Commercial
Space Companies, 2010–19
Dollars (billions)
14

U.S.

China

12
10
8
6
4

France

2
India
0
2010

2011

2012

2013

2014

Indonesia

U.K.

2015

Singapore

2016

2017

2018

2019

Sources: Space Capital; CEA calculations.

with NASA’s overall economic impact estimated at over $64 billion for financial
year 2019 based on IMPLAN analysis. Recent technological developments will
allow new industries to mature. For example, dramatic reductions in the cost of
space launches are increasing the economic viability of private space activities
such as tourism and mining. Technologies enabling the Moon, Mars, and asteroid surface operations will mark critical milestones for the next generation of
space exploration. The reduced cost of access to space will broaden the set of
countries that are able to take advantage of the opportunities in outer space
and ensure benefit to people around the globe. Furthermore, in situ resource
utilization of space resources derived from celestial bodies themselves for
potable water, breathable air, and spacecraft propellant will allow longer-term
survival away from Earth’s surface. Once long-term survival in cislunar space
is viable, further explorations deeper into space will be possible. And once
technologies advance to make long-term survival viable, government policies
clarifying property rights will provide the needed framework for a flourishing
space economy.
Figure 8-2 shows nongovernmental equity investment in space companies from 2010 through 2019. Most private investment in space companies
has occurred in the United States and China, with smaller levels of investment
occurring in European and Asian economies. Investment in commercial space
companies is only a small percentage of the total space economy, but it reflects

Exploring New Frontiers in Space Policy and Property Rights

| 231

Box 8-1. Public-Private Partnerships for Human Spaceflight
The year 2020 has been historic for NASA, as the SpaceX Crew Dragon Demo-2
mission marked the first commercially developed crewed mission to the
International Space Station as part of NASA’s Commercial Crew Program.
NASA has emphasized the implementation of public-private partnerships
to advance space exploration through collaborations with the developing
commercial space sector. After NASA’s Space Shuttle program ended in 2011,
the United States relied on the Russian-designed and operated Soyuz spacecraft to send American astronauts into space. However, the development of
domestic commercial alternatives has allowed the U.S. government to regain
its domestic human launch capability while supporting U.S. commercial
companies.
The Commercial Crew Program, which supported the Crew Dragon
mission by providing SpaceX with development funds, also used fixed-price
contracts, with NASA working as a partner rather than supervisor. Cost
reimbursement or cost-plus contracts had been more commonly used by
NASA in the past, because technically complex and novel projects prevented
it from receiving accurate advance estimates of risk and cost. However, these
types of contracts provide weak incentives for innovation, given that any cost
savings innovation undertaken by the firm leads to lower revenues and often
incentivizes companies to increase the costs and lengths of their contracts.
Fixed-price contracts, conversely, provide strong incentives for innovation
and delivery of products or services on time and under budget. Per NASA’s
2021 fiscal year budget request, fixed-price contracting is now considered the
“first choice whenever possible” due to the incentives produced by placing
increased responsibility on contractors.
Public-private partnerships have been shown to lower the costs of
space products and services for taxpayers and to speed the growth of the
space economy. The Commercial Crew Program’s investment in the private
sector has driven innovation, efficiency, and effective manufacturing and
business techniques, and NASA has projected that it will save between $20 billion and $30 billion relative to the cost to develop its own crewed spacecraft.
After the Space Shuttle program ended, the cost to fly an American astronaut
on a Russian Soyuz rocket rose from $40 million in 2011 to about $90 million
in 2020, given that the Russians held a monopoly on crewed launch vehicles. A
SpaceX launch, by comparison, costs about $65 million per astronaut. SpaceX
is able to reduce costs through new approaches to recover and reuse its
spacecraft and launch vehicles.
In 2011, there were zero commercial launches in the United States,
because the market was dominated by international competitors that were
largely subsidized by their governments. Today, as a direct result of U.S.
Government investments in the U.S. commercial space sector, most commercial space launches are conducted in the United States by companies such as
SpaceX, which employs over 6,000 people throughout the country.

232 |

Chapter 8

Looking to the future of public-private partnerships, NASA has been
increasingly making the ISS available to commercial research and manufacturing activities, as supported by the ISS National Laboratory. In addition,
NASA is allowing visits to the ISS by commercial astronauts aboard SpaceX’s
Dragon 2 and Boeing’s Starliner. Companies are expected to purchase seats
on private sector rockets for missions in low-Earth orbit and to the ISS in early
2022. These missions include opportunities for space tourism and commercial
enterprise, and represent the next step in the space economy.
In addition, NASA will rely heavily on the private sector for the Artemis
program in its mission to accomplish the next chapter in U.S. exploration of
deep space: returning humans to the lunar surface by 2024. These include
contracts for the Human Landing System to take astronauts to and from the
Moon with stays lasting as long as two weeks.

the growing excitement about space companies and optimism for future
returns on investment.
Although there has been large growth in economic activity in the space
sector as a whole, a significant portion of space industry revenue is still made
up of satellite services. As illustrated in table 8-1, over 75 percent of global
spending in space is for the satellite industry, and the majority of the remainder
is government spending. The only other category that is large enough to break
out is the commercial human spaceflight industry.

Space Policy Developments
As investment and innovation grow in the space economy and we surpass new
milestones in space exploration, the United States will continue to work to
ensure the international and domestic framework for property rights in outer
space resources develops in a manner that provides certainty and predictability for industry. Doing so will reinforce the progress the United States has made
in the space sector that, based on CEA estimates, could double investment in
space and accelerate new space technologies. Here, we first describe the main
international treaties and domestic laws that have developed since the 1950s
and provide a legal framework that supports the space economy. We then
describe the efforts of the Trump Administration to advance and execute these
agreements.
The United States is a party to four United Nations treaties on space.
The United Nations Outer Space Treaty of 1967 laid the foundation for international space law, establishing outer space as a peaceful territory, designating astronauts as envoys of humankind, and declaring that each State bears
responsibility for activities in space, “whether such activities are carried on
by governmental agencies or by non-governmental entities.” Whereas private

Exploring New Frontiers in Space Policy and Property Rights

| 233

entities are usually responsible for damages they impose, the Outer Space
Treaty explicitly states that the country from which the object launches or the
country that procures the launch bears responsibility for damages on Earth or
in space.
The United States also approved the 1968 Rescue Agreement, which
outlined the rescue provisions in the Outer Space Treaty requiring countries to
assist personnel when landing within national borders or in places not under
any jurisdiction, such as space. The Liability Convention, which entered into
force in 1972, clarified the meaning of “launching State” to be the country
“which launches or procures the launching of a space object” or “from whose
territory or facility a space object is launched.” The convention also defined
what “damage” consists of and outlined a diplomatic process for resolving
claims for compensation.
Finally, the United States agreed to be a party to the 1976 Registration
Convention that instructs nations to register space objects launched into orbit
or space. While the United States was a party to these four early United Nations
treaties and resolutions establishing international space law, it did not ratify
the United Nations Moon Agreement in 1979, which effectively banned private
ownership of extraterrestrial property. Many other major spacefaring nations,
including Russia and the People’s Republic of China, are not parties to the
Moon Agreement.
Domestically, the United States has gradually developed a framework of
private property rights in space through legislative and executive action. U.S.
space law was first codified in the 1958 National Aeronautics and Space Act,
which created NASA, although military space activities were already under
way within the Department of Defense. The Commercial Space Launch Act of
1984 created the process for licensing U.S. commercial space launches. The
subsequent Commercial Space Launch Amendments of 1988 encouraged commercial space launches by providing Federal Government indemnification for
damages exceeding $500 million to more than $2 billion.
Moving into the 21st century, three concrete policy achievements helped
further codify property rights in space. First, the U.S. Commercial Space
Launch Competitiveness Act of 2015 established the statutory framework for
the Federal Government to permit domestic private entities to extract and use
resources in space:
A United States citizen engaged in commercial recovery of an asteroid
resource or a space resource . . . shall be entitled to any asteroid resource or
space resource obtained, including to possess, own, transport, use, and sell
the asteroid resource or space resource obtained in accordance with applicable law, including the international obligations of the United States.

234 |

Chapter 8

The U.S. Commercial Space Launch Competitiveness Act designates how
the United States licenses and approves attempts to utilize space resources in
line with authority granted to national governments in the Outer Space Treaty.
Article VI of the Outer Space Treaty states that “activities of non-governmental
entities in outer space . . . shall require authorization and continuing supervision by the appropriate State Party to the Treaty.”
In 2020, the Trump Administration further clarified expectations and
responsibilities for commercial activities in space by enumerating the U.S.
position on property rights and laying out principles for international bilateral
agreements. In April 2020, the Trump Administration announced Executive
Order 13914, “Encouraging International Support for the Recovery and Use of
Space Resources.” This executive order announced the United States’ intention to work with international partners to ensure that commercial exploration
and the use of space resources is consistent with applicable laws. It also explicitly rejected the Moon Agreement, which the United States had not signed,
because it was perceived to have prevented the application of private property
rights to resources in space.
On October 13, 2020, the United States and seven partner spacefaring
nations signed the Artemis Accords, a set of principles grounded in the Outer
Space Treaty to ensure safety and avoid conflict. The principles of the Artemis
Accords are peaceful exploration, transparency, interoperability, emergency
assistance, registration of space objects, release of scientific data, preserving
heritage, space resources, deconfliction of activities, and orbital debris. The
Artemis Accords uphold that resource extraction and utilization must comply
with the Outer Space Treaty, while also affirming that “extraction of space
resources does not inherently constitute national appropriation under Article II
of the Outer Space Treaty.” The accords provide investors with more certainty
when considering other countries’ positions on resource extraction. Eight
founding member nations signed the Artemis Accords: Australia, Canada, Italy,
Japan, Luxembourg, the United Arab Emirates, the United Kingdom, and the
United States. NASA anticipates that additional countries will join the Artemis
Accords in the months and years ahead.
Taken together, these three policy developments built on past treaties
and laws to further clarify outer space property rights. The increased security
of property rights should lead to increased investment and economic activity,
as individuals are able to form expectations and plan for future returns on that
investment. As is discussed further below, the ability to make long-term plans
has many direct and indirect positive effects.

The Economics of Property Rights
A large body of economic literature demonstrates the positive effects on
investment from the initiation of policies that are similar to the space policy
Exploring New Frontiers in Space Policy and Property Rights

| 235

developments discussed above. The examples come from a wide range of
geographies, natural resources, and time periods. This section provides an
overview of the economic theory of property rights as well as several examples
of the theory in practice, including how it applies to the space economy.
North (1991) considers the importance of institutions for shaping and
constraining political, economic, and social interactions. Institutions guide
economic change toward more growth, decline, or stagnation, depending on
the incentive structure they enforce. Tangibly, government institutions determine and enforce property rights as rules governing the economy that shape
the competitiveness and efficiency of markets. As rules for property rights are
further specified, market participants interact with more certainty about the
benefits and costs of potential activities.
The seminal work of Demsetz (1967) outlines the economics behind the
evolution of property rights. Property rights bring clarity to people when they
are weighing potential decisions. Accordingly, the benefits to setting and further clarifying property rights allow individuals to form more accurate expectations of how the rest of society will interact and respond to their actions.
Property rights encourage an individual to undertake investments with the
understanding of which benefits will accrue to that individual.
Establishing and enforcing property rights impose costs on society, as
resources are devoted to monitoring and ensuring compliance. An individual’s
expectations are based on the understanding that the rest of society will
comply with the rights specified, but if other parties are allowed to violate
an individual’s property right without recourse, then it will be difficult to set
expectations.
As figure 8-3 shows, the optimal specification of property rights changes
as the benefits and costs change. The figure depicts the optimal specification
of property rights at two different points in time. In 1967, when the Outer
Space Treaty was signed, there were only two entities engaged in outer space
activity: the United States and Soviet Union. As access to space and other
space technologies have increased, the benefits that companies can expect
from engaging in economic activity in space have grown. These increase the
benefits of property right specification, as ensuring investors have clear expectations about how benefits accrue across society will lead to higher gains from
investment.
Advances in technology that improve monitoring and enforcement will
lower the cost of further specifying property rights or adding more “sticks” to
the bundle. This decrease in the cost of enforcement, along with the increase
in the benefits from setting investment expectations, implies that the optimal
level of property rights specification should increase (as shown in figure 8-3).
The Artemis Accords, for example, are giving investors clearer guidance for how
civil space activities will be conducted and the principles that will guide government decisionmaking. Although the Artemis Accords do not apply directly
236 |

Chapter 8

Figure 8-3. Marginal Cost and Benefit of Property Rights
Specification
Cost of enforcement

1
0.9

Demand for
property rights,
2020

0.8

Marginal cost of
enforcement, 1967

0.7
0.6
0.5

Greater benefits from
internalizing payoffs

Lower cost of
enforcement
Marginal cost of
enforcement, 2020

0.4
0.3

Demand for
property rights,
1967

0.2
0.1
0

Xold

1-Jan

Xnew
Level of property right specification
2-Jan
3-Jan
4-Jan
5-Jan

6-Jan

7-Jan

to the private sector, the United States is responsible, via Article VI of the 1967
Outer Space Treaty, for all individuals subject to its jurisdiction or control. In
this regard, the principles of the Artemis Accords provide clarification to companies about the role of governments in space and eliminate uncertainty about
public-private interactions.

Historical Examples of Property Rights Evolution
Historical examples of the development of property rights establish that
without these extra sticks in the property rights bundle, we should expect to
see higher costs and lower benefits from investments in the space economy,
potentially hindering future developments in outer space.
The early history of oil drilling provides an example of how resources
are likely to be wasted if property rights are not established in a timely manner. Until the early 20th century, oil was not considered property until it was
extracted. This led to what Libecap and Smith (2002) call extractive anarchy.
Companies drilled wells without concern for maximizing the amount of oil produced from a well, but instead sought to be the first to extract and claim ownership of the oil. Oil flows from a well because of the pressure inside the reservoir;
if too many wells are drilled into one reservoir, then the pressure escapes
too quickly to push the oil in the reservoir up the well. As a result, less oil is
extracted. By 1914, the director of the Federal Bureau of Mines estimated that

Exploring New Frontiers in Space Policy and Property Rights

| 237

a quarter of the value of all petroleum production was being wasted due to the
race to extract oil. Further, due to oil and natural gas being found together in a
reservoir, the lower-valued natural gas was often vented into the atmosphere
to ensure that the oil was extracted and thus ownership was secured. As time
went on, the structure of property rights for oil and gas has changed to allow
for increased value to be created from investments in resource extraction.
Without clear in situ property rights for subsurface resources, space
could see a repeat of this behavior for its natural resources. Many elements
that are common in space are frequently used in important technologies.
Iron, aluminum, and titanium are elements critical to the production of electrical components. Silicon is a raw material for solar panels and computers.
Extracted water can be broken down into hydrogen and oxygen to meet a
variety of needs—oxygen is breathable, recombining hydrogen and oxygen
generates electrical power, and liquid hydrogen and liquid oxygen can serve as
propellants (Butow et al. 2020). Though it may sound futuristic, we can imagine
a situation where mining expeditions recklessly extract resources from various
celestial bodies, severely depleting the deposit of resources and diminishing
the returns on future investment in mining. Therefore, defining property rights
now to ensure the responsible use of resources in space could lead to future
higher levels of demand and investment in exploration and a more sustainable
space economy.
A similar story emerges for mineral rights in Nevada during the 19th century (Libecap 1978). As new deposits of minerals were found, especially those
deposits further underground requiring increased investment for extraction,
the specification and enforcement of property rights increased. One of the largest deposits in Nevada, the Comstock Lode, was discovered while Nevada was
still a Federal territory. Property rights for discoveries on Federal lands were
lacking at the time, so citizens created a series of local laws and eventually
founded the State of Nevada to ensure these property rights. Libecap (1978)
shows that as deposits increased in value, local property rights specification
also increased. It may seem difficult to imagine how local property rights would
be formed in space as in territorial Nevada, given the lack of settlements in
space. However, this history implies that it is important to set these rules as
economic actors spend extended time in space in order to maximize the future
investment in the space economy.

Investment Responses to Property Right Enhancement
All the space policy developments discussed above have improved the ability
of investors to set expectations for the manner in which benefits flow from
investments in space. The historical examples given argue that further specifying property rights will bolster investment in the space economy. Increased
investments in the space economy will lead to advances in space technology.
In this subsection, we discuss the economics literature that addresses the
238 |

Chapter 8

effects of setting and strengthening property rights on both investment and
economic growth. The research presented here aims to convey that the benefits for economic activity from improved setting of expectations that clarifies
property rights is universal and not just due to specific circumstances of time
and/or place.
Losses from short-term decisionmaking. A growing concern for future
space exploration activities arises from a lack of property rights security leading to short-term decisionmaking, which may inhibit long-term human activity.
Many empirical studies show that insecure property rights lead to investment
decisions with lower values. Many of these studies have come from analyses
of water rights in the western United States. In what is known as the Prior
Appropriation Doctrine, water rights are handed out based on a “first in time,
first in right” principle. Given that the amount of water available changes each
year due to precipitation patterns, water rights holders that were, earlier in
time, known as senior rights holders are more likely to receive their water
allocation each year than those that were later in time, known as junior rights
holders.
Leonard and Libecap (2019) argue that the Prior Appropriation Doctrine,
with its clear rights for senior rights holders, allowed for investment in irrigation technologies. Given the climate of the western United States, large-scale
investment in irrigation is required to maximize the productivity of large
swaths of land. Leonard and Libecap estimate that 16 percent of western
States’ income in 1930 is attributable to investments made in irrigation that
would not have occurred without secure property rights.
Another concern with insecure property rights is that owners of natural
resources rush to extract them to ensure that they accrue the benefits of their
investments. This rush to extract resources has a detrimental effect on the
value obtained from those resources and other negative spillover effects on
society. One example is the increase in the rate of deforestation that occurs
when property rights for the land are insecure (Bohn and Deacon 2000).
Ferreira (2004) finds that those countries with clearly defined property rights
experience less deforestation than those with weaker protections. Kemal and
Lange (2018) find that a reduced chance of oil well expropriation in Indonesia
lowered the rate of extraction by up to 40 percent.
If short-term decisionmaking prevails in the initial incursions into space,
the future of the space economy could be seriously harmed. Depleting the
resources necessary to sustain life in space would mean having to transport
these resources from Earth at a prohibitive cost and complexity. Therefore,
protecting and responsibly using the resources available in space is more efficient in the long term. If done prudently, establishing property rights in space
could diminish the risk of short-term decisionmaking and strengthen the ability of humans to receive benefits from space.

Exploring New Frontiers in Space Policy and Property Rights

| 239

Enhanced investment and asset value. Frameworks such as the U.S.
Commercial Space Launch Competitiveness Act and the Artemis Accords
enhance property rights by providing clear expectations of the benefits one
can receive from their investment and providing a list of principles that partner
nations will follow as a way to encourage economic activity in space. One
branch of the economics literature uses legal or legislative decisions that
enhance or diminish property rights to determine how investment and asset
values respond to a change in property rights specification. We discuss this
literature here. Later in the chapter, we apply the conclusions of these studies
to estimate the value of enhancing property rights in space.
Alston and Smith (2020) measure the effect of uncertain property rights
resulting from the manner in which Northern Pacific Railroad’s land grants
were structured. The Federal Government provided generous land grants
to railroad companies in hopes of ensuring the quick buildout of rail infrastructure. Northern Pacific was granted almost 16 percent of the land area in
Montana, a State that requires coordination among its farmers and ranchers
to irrigate any tract of land for productive use. Delays in the completion of the
rail line in the 1870s led to uncertainty as to whether Northern Pacific owned
(and could sell) land in its land grant or whether the land was the property of
the Federal Government.
As a result of this uncertainty, completed irrigation projects averaged
delays of four years, while investment in irrigation projects decreased by 28
percent. Insecure property rights affected the landowners whose rights were
secure, because irrigation projects often require coordination among many
parcels due to their high capital costs. The delay in undertaking irrigation
investments led to these landowners being more junior water rights holders and, subsequently, holding less secure water rights. In total, Montana’s
economic activity was 6 percent lower in 1930 as a result of these insecure
property rights.
Grainger and Costello (2014) compare the value of more secure property rights for fisheries in the United States, Canada, and New Zealand. New
Zealand’s regulations on quotas to operate in a given fishery explicitly state
that these quotas are a property right, yet similar quota systems in the United
States and Canada have regulations that explicitly state that the quotas are not
property rights. The fact that the United States’ and Canada’s fishery quotas
are not as secure as New Zealand’s quotas leads to a lower perpetuity value of
the quotas relative to their current annual value. Because U.S. and Canadian
firms have the potential for their quotas to be taken away without recourse,
their assets have lower values relative to New Zealand’s firms. In an additional
analysis, Grainger and Costello (2014) show that the increased security of
property rights with the settling of an ownership dispute between native New
Zealanders, known as the Maori, and New Zealanders of European descent
improved the perpetuity value of fishing quotas by 50 percent. Ensuring that
240 |

Chapter 8

property rights will be honored is very important for market participants in
understanding the value of their asset.
Galiani and Schargodsky (2010) use a court case in Argentina to estimate the effect of secure property rights for one’s home on household decisions. Their results show that households that gained secure property rights
increased their investments in the home structure. Investment in walls and
roofs increased by 40 percent and 47 percent, respectively, as a result of
households being granted title to the home. Though not directly related to
space assets, the available evidence demonstrates that more secure property
rights lead to other spillover benefits that are not directly related to the assets
on which a property rights are granted. Galiani and Schargodsky (2010) find
that when households had increased property rights security, they increased
investment in their children’s education. Children in households who obtained
the secure property rights on their land achieved an extra 0.7 year of schooling
on average. This is an important spillover effect given the large individual and
societal benefits of extra years of education (see chapter 7 of this Report).
Telecommunications satellites orbiting Earth provide an example of positive spillovers from ensuring secure property rights in space. The International
Telecommunication Union (ITU) is an organization that standardizes rules and
regulations for a wide range of communications. Through the ITU, the United
States was able to operate satellites that used specific frequencies to transmit
information to Earth, thereby allowing companies to invest in utilizing those
signals for commercial purposes. Communications satellites in geosynchronous orbit rely on the ITU to secure access to specific orbital slots as well as
specific frequencies.
Protection against expropriation. A number of nongovernmental organizations produce indices that measure property rights protections or general
institutional quality. The indices attempt to quantify the relative level of
property rights characteristics, such as the rule of law or protection against
expropriation risk, that are consistent across countries and time. A large body
of economics literature uses these country-level indices of institutional quality
to determine the extent to which improvements in property rights enforcement
affect economic outcomes. Policies initiated under the Trump Administration
would likely alter these indices in a measurable way if there were a property
rights index for space.
Seminal work by Acemoglu, Johnson, and Robinson (2001) shows that
improving the enforcement of property rights, in this case property rights that
protect against expropriation risk, has large effects on gross domestic product
(GDP). In their analysis, the authors show that a one-unit improvement in the
protection against expropriation risk would lead to more than doubling GDP
per capita 10 years later.
Similar results are found when researchers examine specific industries.
For example, Cust and Harding (2020) show that firms drill for oil twice as
Exploring New Frontiers in Space Policy and Property Rights

| 241

often in countries with stronger property rights enforcement relative to their
neighbors with weaker property rights. They also show that the effect of the
enforcement of rights is most important for private international oil companies
relative to national oil companies, highlighting the important role of stronger
rights for harnessing private investment. Bohn and Deacon (2000) find a similar
pattern for the effect on oil drilling as property rights security improves, with a
30 percent increase in security leading to a 60 percent increase in drilling per
year.
Some changes in property rights enforcement come through improvements in technology. Hornbeck (2010) uses the invention and widespread use
of barbed wire as a technology advancement that reduced the costs of enforcing property rights in agriculture. Importantly, Hornbeck compares areas that
had access to timber for wooden fences with those that did not and finds a 23
percent relative improvement in crop productivity when barbed wire came
into use, as barbed wire lowered the relative cost of fencing. Most of the gain
came from farmers altering the type of crop that they planted once they were
confident that livestock would not destroy the crop. This increased ability to
effectively enforce property rights led to investments that increased the total
area of farmland that had been improved by 19 percentage points, while also
increasing land values. In many ways, this example of marking off territory is
similar to the Artemis Accords’ “Deconfliction of Activities” Principle. This principle prescribes setting “safety zones” to limit harmful interference and keep
the probability of accidental loss to a minimum.

The Effects of Policies on Investment
in Space Industries
The previous section detailed the expansive literature showing that more
secure property rights increase both investment and economic activity. The
examples discussed varied across time and space, leaving little doubt that the
results are not driven by random chance; the studies as a whole reveal that the
findings hold outside specific examples. Because the examples are numerous
and varied, determining an average effect of more secure property rights on
investment is difficult. Each study concerns a particular improvement in the
security of property rights that is difficult to quantify. However, it is still a goal
of this chapter to estimate the effect of the last year’s space policy developments on future investment, given the available evidence.
Table 8-2 summarizes the effects of most of the studies discussed in the
previous section. All these effects are large in magnitude. Another data point
is the increase of investment in the space economy in the United States with
the passage of the U.S. Commercial Space Launch Competitiveness Act in 2015
relative to investments in other countries. Using the Space Capital data discussed in the second section, and the historical examples given above, the CEA
242 |

Chapter 8

Table 8-2. Summary of Effects of Property Rights Improvement
Study
Acemoglu,
Johnson, and
Robinson
(2001)
Alston and
Smith (2020)
Bohn and
Deacon
(2000)
Cust and
Harding
(2020)
Galiani and
Schargodsky
(2010)
Grainger and
Costello
(2014)
Hornbeck
(2010)
Leonard and
Libecap
(2019)

Industry

Cause

All

Expropriation
risk

Land

Tenure
uncertainty

Oil

Expropriation
risk

Oil

Expropriation
risk

Housing

Tenure
uncertainty

Fisheries

Tenure
uncertainty

Agriculture

Enforcement

Water

Tenure
uncertainty

Effect
GDP per
capita
increased
100%
Investment
delayed
Investment
increased
100%
Investment
increased
200%
Investment
increased
40%
Asset value
increased
50%
Productivity
increased
23%
Income 16%
higher

Timing of
Impact
10 years

5–10 years
Immediate

Immediate

15 years

Immediate

5–10 years

40 years

Note: This table summarizes the main findings of the papers discussed in the previous
section of the main text. Each study has a different issue with property rights and the impact
on the outcomes of interest.

estimates the increase in investment in the United States due to the improved
property rights specification in 2015. Controlling for country and time period
effects, the data show a statistically significant increase in investment of 92
percent—or roughly double—in the United States since passage of the U.S.
Commercial Space Launch Competitiveness Act relative to countries that did
not improve property specification. Together, these small improvements in
the security of property rights have the potential to lead to large increases in
investment. As an approximation, the CEA assumes that these improvements
in property rights security will double the amount of investment in space. This
number is in line with the evidence that has been discussed here.
To project the effect of the enhancements of property rights security that
the Trump Administration’s policies have achieved, the CEA starts with data
from Space Capital on total private investment in space activities. Figure 8-4
illustrates the increasing rate of private investment in space activities.
The review of the literature discussed above shows that further property rights specification leads to increased investment and further economic

Exploring New Frontiers in Space Policy and Property Rights

| 243

Figure 8-4. Nongovernmental Equity Investment in Commercial
Space Companies in the United States, 2010–28
Dollars (billions)
50
45
Enhanced
property
rights

40
35
30
25
20

Current
trend

15
10
5
0
2010

2012

2014

2016

2018

2020

2022

2024

2026

2028

Sources: Space Capital; CEA calculations.
Note: Data before 2020 reflect actual investment; subsequent years are forecasts.

activity. In figure 8-4, the diverging lines from 2020 to 2028 project the expected
path of private investment as a result of policy developments in 2020.
The Space Capital data suggest that a linear projection of private investment in space would reach $23 billion in 2028, which is illustrated by the blue
dashed line in figure 8-4. However, this does not take into account property
rights enhancements that occurred in 2020 or will be occurring in the future.
Therefore, the CEA projects that private investment in space will reach $46
billion by 2028. This projection is based on a doubling of investment over the
eight-year period, which is in line with empirical estimates in the academic
literature discussed above.
Establishing rights to distant resources with the goals of incentivizing
economic development and investment has not always produced the desired
results. The above-mentioned examples demonstrate how property rights
specification and security can lead to increased investment. However, aligning incentives is a necessary but not sufficient condition in the short term. For
example, the leading asteroid mining companies that were supporting the
space resources language in the Commercial Space Launch Amendments Act
of 2004 have both failed, despite the benefit of positive Federal legislation. In
addition, the Deep Seabed Hard Mineral Resources Act, which was passed in
1980, established a legal system for extracting resources from the deep seabed with hopes of achieving economic viability before 2000. Forty years after
the law’s passage, the deep seabed mineral extraction industry still lacks the
244 |

Chapter 8

technology for economical extraction and does not bolster the argument that
enhanced property rights typically unlock commercial value. Certain similarities do exist with the space industry, such as the need for technological innovation, the considerable distance to the resources, and some uncertainty about
the types of resources for extraction.
Moreover, the space resource extraction industry currently lacks a customer base other than national governments, and even government demand
will not become substantive until robust human and robotic operations on
the lunar surface and elsewhere can be established. However, several key
differences would support a space resource extraction industry. First, the
commercial space industry benefits from public investment in civil space
exploration, which might result in a decreased amount of investment necessary for the development of basic technologies. In addition, space exploration
and research remain a national priority for many countries, which may drive
further development of the industrial base. Moreover, space resource extraction potentially offers more valuable resources than deep sea mining (Barton
and Recht 2018).

Looking Ahead
Increased investment, flowing from the enhancement of property rights,
expands the possibilities of economic activity in space and transforms abstract
issues into real considerations for national economies, companies, and
individuals.

Flag of Choice
The origins of spacecraft and the settlement of international disputes beyond
Earth’s surface remain critical issues for space policy. The flag of choice in
commercial space activity will depend on a nation’s ability to provide the
domestic infrastructure and international support needed to spur investment
while mitigating risk. The development of a healthy space economy built on a
strong industrial base, sensible regulatory environment, and the enforcement
of property rights, along with national support in international disputes, will
ensure that the United States becomes and remains the flag of choice for
private space ventures.
Space vehicles, similar to naval vessels, are required to operate under
the laws, or “flag,” of a particular country. The process of flagging occurs when
a company incorporates itself in a country or launches from that country.
Once flagged, the vessel must abide by the flag state’s laws, which include tax
liabilities as well as labor and environmental regulations (Taghdiri 2013). The
process of selecting a flag leads companies to seek flag countries with a legal,
policy, and regulatory environment that is most favorable for their business
activities.

Exploring New Frontiers in Space Policy and Property Rights

| 245

The practice of finding a “flag of convenience” is one threat to maintaining a functional system of space travel, because companies could opt for flags
of countries with little oversight, as is seen with waterborne vessels (Llinás
2016). Panama, for example, has become the flag of choice for ships, with more
than double the number of ships of any other country due to an easy registration process and low-cost labor. In contrast to maritime law, which places the
responsibility for redressing damages on private actors, the 1967 UN Outer
Space Treaty established that countries assume the full responsibility and risk
of spacecraft launching from their territory. This forces countries to weigh the
costs and benefits of flagging spacecraft before allowing them to launch from
their territories. Inevitably, accidents in space will occur, such as with the 2009
satellite collision between Iridium 33 and Cosmos 2251. The incident occurred
when the Cosmos 2251, a derelict satellite from Russia, collided with Iridium
33, a commercial U.S. communications satellite, and both parties placed the
blame on the other for not avoiding the collision. The United States and Russia
were able to settle the potential dispute outside the Liability Convention, but
this event highlighted the need for a predictable system for resolving disputes
in space to provide the certainty needed for long-term investment in space
ventures. Flying under the flag of the United States will provide companies
with the backing of a sovereign state with substantial diplomatic capital that is
willing to engage on their behalf, supporting a growing space economy in the
United States (box 8-2).

Incentivizing the Private Sector
The Department of Defense continues to foster partnerships with the private
sector through design competitions that award contracts to both large and
small space technology companies, and through consulting programs that
mentor small companies in competing for these contracts. These events and
programs include the Space Enterprise Consortium; the Space Pitch Day,
which awards grants to accelerate new technology; and the National Security
Space Launch, which is helping to create new engines and launch vehicles.
These partnerships help break down barriers to entry for smaller firms in this
industry, which will drive competition and innovation, while decreasing the
cost of operating within the space economy. To ensure that the United States
maintains its leadership in space innovation and remains the flag of choice for
space commerce, it must maintain a business-friendly regulatory environment
that offers streamlined permitting, encourages innovation and risk-taking, and
safeguards workers, the public, and property.
The Trump Administration has prioritized regulatory reform over the
past four years, and it continues to focus on cutting red tape in the space
sector. With regulatory authorities distributed among the Federal Aviation
Administration, Federal Communications Commission, and National Oceanic
Atmospheric Administration, the Trump Administration has made efforts to

246 |

Chapter 8

Box 8-2. National Security and Space
Space-based capabilities are crucial for the United States’ security. Space has
become a primary component of U.S. military operations, including missile
warning, geolocation and navigation, target identification, and activities to
track adversaries. Remote-sensing satellites have greatly improved military
and intelligence collection capabilities, thereby reducing other countries’
ability to carry out covert military exercises and operations.
As advancements in the space sector occur, such as technological
improvements and lower barriers to entry, foreign governments are developing capabilities that could threaten the United States’ freedom to operate in
space. In a 2020 report, the Defense Intelligence Agency points out how China
and Russia, in particular, are trying to undermine the United States’ advantage in space (DIA 2019). For example, Chinese and Russian military doctrines
present a view that counterspace capabilities serve as a tool to reduce the
effectiveness of U.S. and allied military forces. Both countries have developed
extensive space surveillance networks that enable them to monitor, track,
and target American and allied forces. Additionally, both China and Russia are
working on their cyberspace and jamming capabilities.
The Trump Administration recognizes the importance of establishing
and maintaining influence in space and providing space security for U.S.
interests and the American people. In March 2018, the White House unveiled
a new National Space Strategy that places an emphasis on “peace through
strength in the space domain.” Though adversaries are attempting to use
space as a weapon, the United States’ stance is to protect the space domain
from conflict and secure the United States’ vital interests in space—namely,
the freedom of operation in space to advance security, economic prosperity,
and scientific knowledge.
Although peace in the space domain is a top priority, the National Space
Strategy affirms that the United States needs to be vigilant about any harmful
interference within the space domain that negatively affects America’s or its
allies’ vital interests and must “deter, counter, and defeat” any such threats.
Space systems are vital to the U.S. economy and national security,
and they enable key functions such as global communications; positioning, navigation, and timing; scientific observation; exploration; weather
monitoring; and multiple vital national defense applications. In September
2020, President Trump issued Space Policy Directive (SPD)–5, “Cybersecurity
Principles for Space Systems,” which provides guidance on the protection of
space assets and supporting infrastructure from evolving cyber threats.
The National Space Strategy also emphasizes the importance of better
leveraging and supporting the commercial sector to ensure that American
companies are leaders in space technology. This is discussed more throughout this chapter.
To strengthen the United States’ military position in the space domain,
President Trump established the United States Space Force (USSF) as the

Exploring New Frontiers in Space Policy and Property Rights

| 247

sixth branch of the U.S. Armed Forces by signing the National Defense
Authorization Act for fiscal year 2020. Vice President Pence has stated that the
mission of the Space Force is to “develop and implement the unique strategy,
doctrine, tactics, techniques and procedures our armed forces need to deter
and defeat a new generation of threats in space” (Pence 2019). Its responsibilities include “developing military space professionals, acquiring military
space systems, maturing the military doctrine for space power, and organizing space forces to present to our Combatant Commands” (USSF 2020).

modernize the authorization process for new space missions, as directed
in Space Policy Directive-2. In addition, Federal Government procurement
regulations are often complex and burdensome for the private sector. In fact,
government-procured space systems were historically characterized by high
costs, long program schedules, and frequent delays due to these regulations
(Butow et al. 2020). This discouraged efficiency, innovation, and the entrance
of new actors into the market. In the interest of increasing competition and
innovation while reducing costs and bureaucracy, the Administration continues to remove undue regulatory barriers and increase the efficiency of existing
processes. Doing so will foster a free and prosperous space economy, enable
commercial space companies to operate more efficiently, and allow new firms
to participate in the private space industry.
Furthermore, the Administration has recognized the important role the
Federal Government plays in promoting an environment that encourages
investment in the space economy. This starts with outlining clear and coordinated policy goals and stimulating public and private activity to achieve them.
By increasingly shifting the role of the government in the space domain from
that of owner and operator of technology to that of customer of private products and services, the United States increases demand for commercial activity
and supports the growth of a viable space economy.
For example, NASA can use commercial service contracts within the
Artemis Program, including those governing transportation, communications,
and power systems to facilitate the return of manned missions to the lunar
surface and to encourage their permanent operation there. The Department
of Defense also serves a critical role in creating demand within the private
sector because this Administration has prioritized the protection of national
security in space. Applying the same concepts to space resources, the Federal
Government can reduce risk to the private sector for new technologies such
as space mining and manufacturing. By acting as an initial, substantial, and
dependable customer for early entrants into space resource markets, the
Federal Government can encourage private investment by offering to purchase
products on forward contracts. With the assured revenue that comes from

248 | Chapter 8

these contracts, private firms can use increased economies of scale to further
reduce the costs of these new technologies, which opens the market to new
customers.
Prioritizing regulatory reform and investment in the space sector builds
a strong foundation for a thriving space economy. The Trump Administration
has taken action to make this future a reality, and it will continue to foster the
environment that spurs investment in the private space sector.

Conclusion
Secure property rights are a fundamental tenet of the U.S. economy. Property
rights help individuals and firms set expectations for how the outcome of their
investments will be distributed. However, there are costs for setting up and
further specifying property rights. The literature on the economics of property
rights discusses how to balance the benefits from improved expectation setting for individuals’ investment decisions against the costs of enforcement.
Although applications like space mining and space solar power satellites
might be decades away from being profitable enterprises, it is worth laying the
foundation for the emergence of future space industries now.
Economic activity in space will benefit from further property rights
enhancement and specification, which is advantageous when net enforcement
costs are exceeded by net benefits. To this end, the Trump Administration has
initiated policies to enhance property rights and thus to encourage further
investment in space. The Executive Order “Encouraging International Support
for the Recovery and Use of Space Resources” and the Artemis Accords help
to further property rights specification by rejecting an ineffective treaty that
suggests communal property and by motivating other economies to follow the
United States’ lead in developing safe and sustainable best practices in space.
Recent policies to improve the ability of firms to gain certainty regarding their investments lay the foundation for further development of the space
economy. The academic literature provides many examples across time,
geographic range, and resource application of the large effects on investment
and economic activity driven by enhanced property rights security. Based on
these previous experiences with improvements in property rights security, the
CEA estimates that recent Trump Administration policies will add an additional
$23 billion to private investment in the space sector by 2028. Property rights
enhancement, coupled with public-private partnerships, can solidify the longterm health of the commercial space economy.

Exploring New Frontiers in Space Policy and Property Rights

| 249

x
Chapter 9

Pursuing Free, Fair, and
Balanced Trade
As documented in the 2018 Economic Report of the President, the Trump
Administration inherited a legacy of asymmetric trading arrangements that
had imposed steep costs on U.S. manufacturing and segments of the labor
market. Indeed, some recent academic literature finds that import displacement after the establishment of Permanent Normal Trade Relations with the
People’s Republic of China was the single biggest factor in the decline of the
U.S. employment-to-population rate after 1999 (Abraham and Kearney 2020).
Also, this shock was associated not only with a precipitous decline in previously relatively-stable U.S. manufacturing employment but also with increased
mortality from drug overdoses in adversely affected communities (Autor, Dorn,
and Hanson 2019; Case and Deaton 2017; Pierce and Schott 2020).
The Trump Administration has worked over the past four years to renegotiate
unfair trading arrangements that have harmed, in particular, U.S. manufacturing and manufacturing employment. Much of this work came to fruition in 2020,
with several historic trade agreements entering into force this year. This chapter
details the benefits of the trade accomplishments of 2020 and shows that
further work in renegotiating trade agreements to safeguard American workers
will play a key part in returning the U.S. economy to the economic prosperity
of the Great Expansion. In addition, we describe the changing global economic
environment, punctuated by the COVID-19 pandemic, that has caused firms
and governments to rethink existing configurations of global supply chains and,
in some cases, bring production closer to home.

251

On January 15, 2020, President Trump joined Chinese Vice Premier Liu
He in signing the U.S.-China Economic and Trade Agreement—the Phase
One Agreement—a landmark deal that requires structural reforms and other
changes to China’s economic and trade systems in the areas of intellectual
property, technology transfer, agriculture, financial services, and currency
and foreign exchange. The centerpieces of the agreement include not only
the numerous, specific commitments that China made in these areas but also
China’s agreement to expand trade by importing an additional $200 billion in
U.S. goods and services on top of 2017 import levels. The United States maintains significant tariffs on $370 billion worth of Chinese imports because China
has not ended all of its unfair trade practices, setting the stage for a Phase Two
agreement in the future.
In addition to achieving this historic trade deal with China, the Trump
Administration has followed through on its pledge to restore U.S. manufacturing, farming, and business by signing into law the United States–Mexico–
Canada Agreement (USMCA) on January 29, 2020, overhauling the North
American Free Trade Agreement. With the rules of the agreement entering
into force on July 1, 2020, USMCA promises to better balance trade with our
neighbors to the north and south. USMCA establishes requirements for digital
trade, environmental standards, and standards for workers’ rights between
the three countries. Also, American agricultural exports will increase by $2.2
billion under USMCA thanks to better access to Canadian markets secured
through the new agreement. Overall, pre-COVID estimates by the nonpartisan
U.S. International Trade Commission found that under moderate assumptions,
USMCA will increase U.S. gross domestic product by $68.2 billion (0.35 percent)
and create 176,000 U.S. jobs.
Prepandemic analyses indicated that both the Phase One Agreement and
USMCA would grow the U.S. economy. However, with the COVID-19 pandemic
spreading from China around the globe, worldwide lockdowns temporarily disrupted the projected benefits from negotiating these trade deals and

252 |

Chapter 9

highlighted the substantial risk in centralizing supply chains in China. As China
shut down factories in the city of Wuhan—a major manufacturing hub, and also
the origin of the coronavirus—supply shocks rippled throughout the global
economy and reached the United States long before the virus spread within
the United States. This chapter examines COVID-19’s effects on global supply
chains and discusses evidence that companies are already reducing their reliance on Chinese manufacturing.
As the process of U.S. economic recovery continues, fair and reciprocal trade
agreements will continue to be a critical component of returning to the economic prosperity of the Great Expansion. The Phase One Agreement, USMCA,
an additional trade agreement with Japan, and a renegotiated trade agreement
with South Korea, among others, underscore the Trump Administration’s commitments to securing trade deals that drive growth in the economy and the
labor market. These agreements are the culmination of the Administration’s
commitment to the American worker—addressing unfair trade practices that
have adversely affected U.S. employment, while also incentivizing domestic
hiring and capital formation.

nternational trade plays a critical role in driving economic growth, and
thus in helping countries around the world achieve unprecedented prosperity in the 21st century. Institutions in which the United States plays a
key leadership role, such as the World Trade Organization and its predecessor
the General Agreement on Tariffs and Trade, established the institutional
framework of the global trading system and facilitated growth in global trade.
But the benefits of the global trading system have at times come at the cost
of America’s own national interest. The Trump Administration has actively
pursued actions to make trade with the United States’ international partners
fairer and more sustainable. The Administration has imposed tariffs to protect
national security and other U.S. economic interests. And by forging new trade
agreements with China, our North American neighbors, and Japan, as well as
a renegotiated agreement with South Korea and narrower agreements with
several other countries, the Administration has ushered in a new international
trade environment that promises to enhance U.S. economic growth and broadbased economic prosperity. The first three sections of this chapter examine
these agreements.

Pursuing Free, Fair, and Balanced Trade | 253

Although growth in global trade has increasingly taken the form of
trade in intermediate goods and the globalization of supply chains, events in
recent years may slow that trend. The 2008 global recession, trade, and other
geopolitical tensions (especially between the United States and China), and
most recently the 2020 COVID-19 pandemic, have all underscored risks associated with trade that have prompted firms and governments around the world
to reconsider the benefits and costs of their existing configurations of global
supply chains. The Trump Administration’s recent and ongoing activities focus
on reaping the benefits of trade while advancing the interests of American
industry and safeguarding national security. The last section of the chapter
considers these issues related to global supply chains.

The Phase One Agreement with China
The U.S.-China Phase One Agreement is a first step in the resolution of U.S.China trade disputes. These tensions came to the forefront in August 2017 with
the announcement in the Federal Register of an investigation by the Office of
the United States Trade Representative (USTR) into China’s technology transfer and intellectual property (IP) protection policies under Section 301 of the
Trade Act of 1974. This investigation led to a determination by the USTR that
China engaged in certain unreasonable and discriminatory trade practices. The
USTR took appropriate action by imposing tariffs on U.S. imports from China,
triggering tariff increases by China and responsive tariff increases by the United
States, from July 2018 to September 2019.
On December 13, 2019, the two countries agreed to the Phase One
Agreement, which President Trump signed on January 15, 2020. This agreement has seven main chapters addressing structural reforms in the areas of
IP, technology transfer, agricultural nontariff barriers, financial services, currency, and Chinese purchases of U.S. exports; and establishing a strong dispute
resolution system. An eighth chapter provides details on amending the agreement, effective dates and termination, further negotiations, and “notice and
comment” on implementing measures (USTR 2020b). The Chinese purchases
of U.S. exports are an important component of this agreement, given that they
were expected to provide immediate positive effects for U.S. producers. With
the COVID-19 pandemic, Chinese purchases of U.S. exports this year started
slowly but have increased significantly in recent months. The structural provisions are the important first steps in creating much-needed reform in the
Chinese economy.

Background
In August 2017, the USTR opened an investigation into Chinese policies and
practices regarding technology and IP under Section 301 of the Trade Act of
1974. The USTR issued a report in March 2018 that detailed a variety of unfair

254 | Chapter 9

' шҊрѵѵѵ-$!!/$*)."$)./#$)Ѷспрч–рш
-)# җ(*)/#)
4 -/# 4/**& !! /Ҙ

'0 *!I(+*-/.
җbillion *''-.Ҙ

-$!!R/
җ+ - )/Ҙ

рҗ 0'4спрчҘ

ту

сф

сҗ0"0./спрчҘ

рх

сф

тҗ +/ ( -спрчҘ

спп

уҗ +/ ( -спршҘ

рсп

рп; rose to 25 in
June спрш

/'$/*-4#$)T-$!!s
*)ѵѵE3+*-/.
сф‫ڦ)*ڿ‬ту$''$*)
in"**.
сф‫ڦ)*ڿ‬рх$''$*)
in"**.
Range up to 25% on $53
in goods
billion
ф–рп‫ڦ)*ڿ‬сф billion
in goods

рф

*0- ѷ!!$ *!/# ѵѵ- +- . )//$1 ѵ
Note: This table reflects the tariff rates when they went into effect, not necessarily the current
state of play. Tranche 4B was set to take effect December 15, 2019, but due to the U.S.–China
Phase 1 agreement, these tariffs were never imposed as noted in the text. Dollar values are in
nominal terms.

Table 9-2. U.S. Bilateral Trade Deficit with China in Goods and Services
Type of Deficit
Bilateral trade
deficit in goods
Bilateral trade
deficit in goods
and services

2018:Q2
Dollars
Share of
(billions)
GDP

2019:Q4
Dollars
Share of
(billions)
GDP

2020:Q3
Dollars
Share of
(billions)
GDP

–98.4

1.9

–77.4

1.4

–79.0

1.5

–88.7

1.7

–68.1

1.3

–74.6

1.4

Sources: Census Bureau; CEA calculations.
Note: The trade deficit is annualized to calculate the share of GDP. Dollar values are in nominal terms.

Chinese policies and practices: (1) forced technology transfer from U.S. inventors and companies to Chinese firms for market access in China; (2) nonmarketbased terms for technology licenses; (3) Chinese state-directed and facilitated
acquisition of strategic U.S. assets; and (4) cyber-enabled intrusions into U.S.
commercial networks to steal trade secrets for commercial gain (CEA 2019).
Table 9-1 describes the four tranches of tariffs that the United States imposed
to bring China to the negotiating table to reform these costly policies. China
retaliated with its own tariff actions against the United States in each tranche.
As part of the conclusion of the Phase One negotiations, the United States
suspended a tariff rate increase from 25 percent to 30 percent on tranches
1 through 3 and reduced the tariff rate on tranche 4A from 15 percent to 7.5
percent. China also cut its 4A tranche tariff rates by half.
Since the Section 301 tariffs went into effect starting in July 2018, we
have observed a decline in the bilateral trade deficit between the U.S. and
China, from $88.7 billion ($354.8 at an annualized rate, or 1.7 percent of GDP)
to $68.1 billion ($272.4 billion at an annualized rate, or 1.3 percent of GDP) at

Pursuing Free, Fair, and Balanced Trade

| 255

the end of 2019 (table 9-2). The bilateral trade deficit has increased slightly
during the COVID-19 pandemic, particularly as international trade in goods
has recovered faster than international trade in services, but remains below
pre–Section 301 levels.

Major Provisions
On December 13, 2019, the United States and the People’s Republic of China
announced that they had reached the Phase One Trade Agreement. This agreement came just two days before the United States was set to impose tranche
4B, which would have imposed 15 percent tariff rates on an additional $160
billion worth of U.S. imports from China. As a part of the agreement, the United
States suspended the tariffs set for December 15 and agreed to halve tranche
4A tariffs on $120 billion of Chinese goods to 7.5 percent (USTR 2019b). On
January 15, 2020, President Trump signed the Phase One Agreement with the
People’s Republic of China, establishing a foundation for a fair and reciprocal trade relationship between the two countries. This agreement requires
structural reforms and other changes to China’s economic and trade policies
in these seven areas, each of which corresponds to a chapter in the agreement:
1. Addressing concerns related to intellectual property,
2. Ending China’s practice of forced foreign technology transfer,
3. Lowering structural barriers to agricultural trade,
4. Expanding market access for U.S. financial service companies in China,
5. Addressing unfair currency practices,
6. Expanding trade through Chinese purchase commitments, and
7. Introducing a dispute resolution mechanism to effectively implement
and resolve issues arising under the agreement.
Chapters 1 and 2 of the Phase One Agreement address U.S. concerns
relating to intellectual property theft and forced foreign technology transfer
and should help create a fair market and protect U.S. companies operating in
China. Chapters 4 and 5 require China to lower financial service barriers and
end unfair currency practices. Chapter 6 sets forth purchase commitments,
including purchase commitments for agricultural commodities, which China
must meet to help create a more balanced and fairer trading relationship that
benefits the United States. The removal of structural barriers to agricultural
trade in Chapter 3 should help achieve these purchase commitments. Finally,
Chapter 7, which addresses dispute resolution, creates a process for discussing implementation of, and resolving issues arising under, the Phase One
Agreement.

Intellectual Property
Chapter 1 of the Phase One Agreement includes specific commitments to
strengthen protection and enforcement of IP in China and reduce IP theft,
including with respect to trade secrets, pharmaceutical-related IP, and
256 |

Chapter 9

enforcement against counterfeit and pirated goods. The Section 301 report
found a $50 billion annual cost to the United States from IP theft (USTR 2018a).
Several of the provisions in the Phase One Agreement are novel and require
significant changes in China’s practices. A large component of this addresses
specific concerns regarding adequate and effective IP protection and enforcement in China. As part of China’s implementation of the Phase One Agreement,
China has published numerous draft measures for public comment and has
issued final measures in areas including criminal prosecution standards for
trade secret theft, civil enforcement of trade secrets, destruction of counterfeit
and pirated goods, and online infringement on e-commerce platforms.
Reports from the Organization for Economic Cooperation and
Development (OECD 2019) show China as the top source of counterfeit and
pirated goods, endangering the public with goods that pose potential health
and safety threats. The Phase One Agreement contains provisions on the
expeditious takedown and destruction of counterfeit goods. The agreement
also includes obligations for China to take effective action against e-commerce
platforms that fail to take necessary measures against infringement and
ensures that government agencies and state-owned enterprises use only
licensed software.

Technology Transfer
In Chapter 2 of the Phase One Agreement, China agreed to end the practice of
requiring or pressuring U.S. companies to transfer technology to Chinese entities, including in relation to joint ventures, acquisitions, or obtaining business
licenses. These commitments extend to any informal, unwritten measures that
China takes to force or pressure foreign companies to transfer their technology to Chinese entities, which is a key concern identified in the Section 301
investigation. China also committed to provide transparency, fairness, and
due process in administrative proceedings and to ensure that any technology
transfer and licensing take place on market terms. Moreover, China agreed not
to support or direct outbound foreign direct investment activities aimed at
acquiring foreign technology with respect to sectors and industries targeted
by market-distorting industrial plans. Though Phase One negotiations were
ongoing, China enacted its new Foreign Investment Law and amended its
existing Administrative Licensing Law in order to address the use of “administrative means” to force technology transfer, and the disclosure of trade secrets
and confidential business information submitted by administrative license
applicants.

Agriculture
Chapter 3 of the Phase One Agreement lowers structural barriers to agriculture
trade. As for nontariff agricultural barriers, China has removed many restrictive
and burdensome import requirements, including lifting its effective ban on

Pursuing Free, Fair, and Balanced Trade

| 257

258 |

Chapter 9

4#)" ./* (+' ( )/

/- ('$) ++-*1'+-* .. .!*-ѵѵ!$'$/$ .)+-*0/.Ѷ)China’s - "0'/$*).2$''
.$(+'$!$ *- '$($)/ ѵ
''*2ѵѵ 3+*-/.*!+*//* .Ѷ) /-$) .Ѷ'0 --$ .Ѷ)1**.ѵ

3+)(-& / ..!*-ѵѵ !)+*0'/-4ѵ+ 0+)$(+-*1 ++-*1'+-* .. .
)# '/# -/$!$/$*).ѵ *') ..-4/ #)$'$.0..$*).!*-+-*/**'/*''*2ѵѵ'$1
- $)"//' $(+*-/.ѵ
(*1 )./#/- ./-$/ѵѵ+ /!** 3+*-/.ѵ/- ('$) - "$./-/$*)!*-ѵѵ!$'$/$ .ѵ

''*2/# $(+*-//$*)*!ѵѵ-$ +-*0/.2$/#$)сп4.*!/# $-++-*1'4/# U.S.
Department of Agricultureѵ#)" #*2$/./-$!!-/ ,0*/$.++'$ .*/# ,0*/$.++'$
#4 -Ѷ)/# +-* ..!*-''*/$*)$.)*/$.-$($)/*-4and transparent forѵѵ!$-(.ѵ

()!*-$(+*-/ $-4)$)!)/!*-(0'+-*0/.#.+*.$/$1
relationship with China’s growing wealth, but strict regulations limit U.S.
3+*-/.ѵ

'/#*0"##$) . #*-/$0'/0-'$(+*-/.$)- . 4чп‫ ڿ‬/2 )спрт
and 2018, the U.S. market share of imports is only 10%.

!/ -ртҊ4 -)Ѷ$)спрцѶ#$)+-/$''4- *+ ) $/. !(-& /./*
ѵѵ 3+*-/.Ѷ'/#*0"#- "0'/*-4- ./-$/$*).- ($)ѵ )спрфѶ#$)
)) ѵѵ+*0'/-4Ѷ0//#$.2.+-/$''4- (*1 $)*1 ( -спршѵ

#$) . +*0'/-4)*1$) - ./-$/$*).'*)"2$/#$..0 .- "$./ -$)"
!$'$/$ .#1 '$($/ ѵѵ+ /!** 3+*-/.$)#$)Ѷ 1 ).#$) .
()#."-*2)ѵ

China is the world’s lar" ./$(+*-/ -*!-$ ѵ /./-$!!-/ ,0*/)
+#4/*.)$/-4+-*/**'#0-/ѵѵ 3+*-/.ѵ

$-4)i)!)/
f*-(0'

*-/$0'/0-'
p-*0/.

/Ѷp*0'/-4Ѷ
)l$1 b- $)"
c//'

/f**

*0- ѷ.. epartment of griculture (2020)ѵ

"'$5 /# $(+*-//$*)*!сх) 2. !**+-*0/.ѵ#)" . !**- "0'/$*)..* goods
approved bythe Food and Drug Administration)the /$*)' )$)/(*.+# -$
($)$./-/$*)) $(+*-/ 4#$)ѵ

.#$) . . !** ()-$. .Ѷ- "0'/$*).#1 # '&ѵѵ
. !** 3+*-/.ѵ

!**

ст+-*0/./*$/.'$./*!''*2' ! $/$1 +-*0/$(+*-/.ѵ/- ('$) /#
- "$./-/$*)-)-'$ ).$)"+-* ..!*-ѵѵ! 3+*-/.ѵ

#$) . ()!*-! #."-*2)2$/#$/.'$1 ./*&.0++'4Ѷ0/ѵѵ
! 3+*-/.#ve./-0""' 0 /*- ./-$/$*).$)- "$./ -$)")
'$ ).$)"!$'$/$ .)+-*0/.ѵ

)$('f

(+' ( )//-).+- )/Ѷ+- $/' Ѷ !!$$ )/Ѷ.$ ) Ҋ)-$.&Ҋ. - "0'/*-4++-*1'
+-* ..!*-"-$0'/0-'$*/ #)*'*"4+-*0/./#/*)/$)../-*)"($)$./-/$1
$(+-*1 ( )/.ѵ $($/++-*1'2$//$( ./*)1 -" *!су(*)/#.ѵ

#$) . ()!*-$*/ #)*'*"4Ѷ2#$#$.0. !*-!**)!
+-*0/.Ѷ#.-./$''4"-*2).Chinese 2 '/##."-*2)ѵBut
e3$./$)"- "0'/$*).#$) -ѵѵ 3+*-/..

"-$0'/0-'
$*/ #)*'*"4

' шҊ3ѵ"-$0'/0- #. ) "- ( )/-*1$.$*).
-*0/
-& /1 -1$ 2

<BILL: PLEASE MAKE THIS A TURNED, FULL PAGE TABLE; PHOTO-REDUCE/ENLARGE IF NEEDED:>

U.S. poultry products, which had been in place since 2015. Although not all the
structural changes and commitments listed in table 9-3 have been completed,
certain key changes have been made, resulting in improved market access for
a number of U.S. agricultural products. The market overview for each product
reveals that Chinese demand has continued to rise for these products, but
before the Phase One Agreement, U.S. exports had been restricted. Changes
made by China include the listing of additional U.S. food manufacturing and
feed additive facilities eligible for export to China, recognition of the Food and
Drug Administration’s oversight of dairy food safety, and the removal of other
sanitary and phytosanitary barriers. The lowering of barriers to agricultural
trade, described in table 9-3, is necessary for the agricultural purchases China
has promised.

Financial Services
Chapter 4 of the Phase One Agreement addresses long-standing barriers faced
by a wide variety of U.S. financial services companies, including those in banking, insurance, securities, credit rating, and electronic payment services. These
barriers include joint venture requirements, foreign equity limitations, and
various discriminatory regulatory requirements. As one key example, China
committed to allow U.S. securities, fund management, futures, and insurance
companies to establish wholly foreign-owned companies in China, thereby
providing the potential for U.S. companies to fully control and generate profits
from their businesses. Removal of these barriers will allow U.S. financial services companies to compete on a more level playing field in China.

Currency
Chapter 5 of the Phase One Agreement includes policy and transparency
commitments related to currency issues. The chapter addresses unfair currency practices by requiring strong commitments to refrain from competitive
devaluations and targeting of exchange rates, while promoting transparency
and providing mechanisms for accountability and enforcement. The enforcement mechanism enables either the U.S. Department of the Treasury or the
People’s Bank of China to refer exchange rate policy or transparency issues to
the Bilateral Evaluation and Dispute Resolution Arrangement established in
Chapter 7 of the agreement, which we discuss below.
China has a long history of pursuing a variety of economic and regulatory
policies that provide their economy with a competitive advantage in international trade. This includes intervention in foreign exchange markets in concert
with the maintenance of capital controls that together harm U.S. export competitiveness by facilitating the undervaluation of the renminbi. In August 2019,
the U.S. Treasury ruled that China was manipulating its currency under the
Omnibus Trade and Competitiveness Act of 1988.

Pursuing Free, Fair, and Balanced Trade

| 259

After this determination, the U.S. Treasury and the People’s Bank of China
engaged in negotiations over currency issues to eliminate unfair Chinese practices that gain competitive advantages. More broadly, China made commitments in the Phase One Agreement to refrain from competitive devaluations
and to not target its exchange rate for competitive purposes, and it agreed to
publish relevant information related to exchange rates and external balances.
In this context, in January 2020, the U.S. Treasury determined in its “Report to
Congress on Macroeconomic and Foreign Exchange Policies of Major Trading
Partners of the United States” that China would not be designated as a currency manipulator at that time.

Chinese Purchase Commitments
Chapter 6 of the Phase One Agreement specifies commitments for Chinese purchases of a selected group of U.S. goods and services. For the first year of the
agreement, China has committed to purchase an additional $76.7 billion of U.S.
goods and services over the 2017 baseline, followed by an additional $123.3
billion of purchases in the second year (table 9-4). Purchases are broken down
into four sectors: manufactured goods (38.9 percent), energy (26.2 percent),
services (19.0 percent), and agriculture (16.0 percent). From 2022 to 2025, the
agreement states that the countries “project that the trajectory of increases
in the amounts of manufactured goods, agricultural goods, energy products,
and services purchased and imported into China from the United States will
continue in calendar years 2022 through 2025.”
As part of the trade deal, China will also reduce its structural trade barriers, which should result in an expansion of market-based access for U.S. goods
and services aiding these purchase commitments, as described above for U.S.
exports of agriculture. Because barriers to China’s markets primarily take
the form of nontariff barriers, specific purchase commitments help promote
China’s adherence to its structural reforms in the agreement. As a result of the
Phase One Agreement, China has already begun taking actions to reduce its

' шҊуѵ$/$*)'ѵѵ3+*-/./*#$)*)*+*!спрц. '$)
-*0/
C/ "*-4
)0!/0-
"**.
) -"4
-1$ .
"-$0'/0-
*/'

спсп
җ*''-.Ѷ
$''$*).Ҙ

спср
җ*''-.Ѷ
$''$*).Ҙ

тсѵш
рчѵф
рсѵч
рсѵф
цхѵц

ууѵч
ттѵш
сфѵр
ршѵф
рстѵт

спсп–ср*/'
җ*''-.Ѷ
$''$*).Ҙ

спсп–ср
- )/age*!
/*/'

ццѵц

тчѵш

фсѵу
тцѵш
тсѵп
сппѵп

схѵс
ршѵп
рхѵп
рппѵп

Source: “Trade Agreement between the United States of America and the People’s Republic of China”
җ-!/ҘѶ)) 3хѵрѵ
*/ ѷthe twoҊ4 -+ - )/age*!of the /*/'* .)*/ 3/'40+/*рпп0 /*-*0)$)"ѵ

260 |

Chapter 9

retaliatory tariffs, enabling greater market access and increased purchases. In
two separate waves, China announced tariff reductions, or chose not to impose
tariffs, on $75 billion worth of U.S. goods and available tariff exemptions on
goods classified under 696 tariff codes (MOF 2020).
Due to the COVID-19 outbreak, international trade fell dramatically as
economies around the world contracted. As a result, purchases from China
early in the year were lower than anticipated. However, purchases have
increased significantly in the past few months. The USTR (2020c) estimates
that as of mid-October, China had purchased over $23 billion in U.S. agricultural products, about 71 percent of its target under the Phase One Agreement.
Record or near-record U.S. exports to China are expected in 2020 for corn, pork,
beef, pet food, alfalfa hay, pecans, peanuts, and prepared foods.

Dispute Resolution
Chapter 7 creates the Trade Framework Group, to be led by the United States
Trade Representative with a designated Vice Premier of China. This group’s
purpose is to ensure implementation of the Phase One Agreement and resolve
disputes in a fair and expeditious manner. The United States and China created the Bilateral Evaluation and Dispute Resolution Office to deal with dayto-day matters. Each party can file a complaint to the other for not acting in
accordance with the agreement at the working level, and then escalate to the
deputy and principal levels if no resolution is achieved. Each party may also
raise matters of urgency directly at the principal level. Regular consultations
are also set up to ensure compliance. Regular consultations are also set up to
ensure compliance.
Unlike other trade agreements, the Phase One Agreement does not provide for independent third-party dispute resolution. If the two parties cannot
resolve a dispute, then the complaining party is authorized under the agreement to take proportionate responsive action that it considers appropriate
against the offending party. This can take the form of tariffs on goods imported
from the other party or the suspension of a provision in the agreement benefiting that other party, among other actions. The responsive action may remain
in effect until the resolution of the dispute. If the respondent party finds that
the complaining party’s action was taken in bad faith, its only recourse is withdrawal from the agreement upon 60 days’ notice.
Chapter 7 of the Phase One Agreement provides for a series of meetings
or telephone calls to ensure the success of the agreement, including monthly
Designated Official calls, quarterly deputy calls, and twice-yearly principallevel calls. In addition, technical groups from the United States and China
confer on a regular basis. During a principal-level call in August 2020, the parties noted sustained commitment to the success of the agreement as well as
noticeable progress toward reaching the agreement’s goals. They discussed
China’s steps to engender “greater protection for intellectual property rights,

Pursuing Free, Fair, and Balanced Trade

| 261

remove impediments to American companies in the areas of financial services
and agriculture, and eliminate forced technology transfer” (USTR 2020d).
This call also served as a discussion on the significant increases of Chinese
purchases of U.S. products and any additional action needed to implement the
Phase One Agreement. Both parties remain committed to the success of the
agreement, with progress already made and a shared commitment to future
necessary steps.

What Is Not in Phase One?
The Phase One Agreement with China is designed to lay the groundwork for a
Phase Two negotiation. The Phase One Agreement does not address certain
underlying structural issues, creating interest for a Phase Two Agreement in
the future (White House 2020a). The two countries still have very different
economic systems as China asserts the state as the principal actor in the
economy, creating friction between the two countries. Despite some tariff
concessions, the United States maintains most of the Section 301 tariffs still in
effect because China has not addressed all the issues identified in the Section
301 investigation.
Some of the main issues remaining include China’s massive government
subsidies to and its preferential treatment of state-owned enterprises (SOEs).
Many Chinese SOEs depend on government subsidies and loans from stateowned banks to compete with more efficient private firms. Important sectors
in China’s economy are state-directed, leading to policies and practices that
provide significant artificial advantages to domestic companies while discriminating against or otherwise disadvantaging foreign competitors. China’s
“Made in China 2025” program is a plan announced in 2015 that aims to make
China a global high-technology manufacturing power (McBride and Chatzky
2019). Many see this as a threat to U.S. leadership in high-technology industries, citing China’s subsidies, the setting of market share targets for Chinese
companies, and the use of policies designed to substitute Chinese products for
other countries’ products in the Chinese market and eventually abroad. China
has released a list of 33 areas in which investment by foreign firms is extremely
limited or not allowed. These areas include infrastructure, the media, agriculture, and some types of scientific research.
Another issue of concern is China’s engagement in cyber-enabled theft
and its intrusions into U.S. commercial computer networks for commercial
purposes. According to the U.S. Department of Justice, China is involved in
more than 90 percent of economic espionage cases and more than 67 percent
of the trade secret theft cases the department has overseen since 2011 (DOJ
2018). National security concerns regarding technology and cyberspace are
also significant issues, including concerns related to the Chinese companies
Huawei and ZTE. The United States has banned the use of Huawei and ZTE
5G equipment by U.S. companies and citizens, and has been urging its allies
262 |

Chapter 9

to follow suit by not investing in Huawei’s 5G technology services due to
the potential for China to spy on customers (Vaswani 2020). Australia, New
Zealand, and the United Kingdom are other major countries that have banned
Huawei from their 5G networks.

The United States–Mexico–Canada Agreement
The United States–Mexico–Canada Agreement (USMCA), which took effect on
July 1, 2020, is a new agreement that replaces the North American Free Trade
Agreement (NAFTA). Free trade agreements establish areas between two or
more countries “in which the duties and other restrictive regulations of commerce . . . are eliminated on substantially all the trade between the constituent
territories in products originating in such territories” (WTO 2020). NAFTA established free trade in goods and services between the United States, Canada, and
Mexico. The ability to maintain free trade in goods and services in the NAFTA
area depends on rules pertaining to everything from customs administration
to identification of the scope of covered services to dispute settlement. The
original NAFTA rules were more than 25 years old.
USMCA updated these rules in order to reflect the lessons learned under
NAFTA and other trade agreements, as well as economic and technological
developments. Among these, USMCA ensures the free movement of data
across borders, improves trade facilitation, strengthens intellectual property
protection for U.S. firms, limits access to international arbitration in investment disputes (thus steering such disputes to the courts of the country hosting
the investment), and modifies the requirements for an automobile to be eligible for duty-free treatment. In particular, the agreement requires that a higher
percentage of an automobile’s parts be sourced within the USMCA region for
production if the vehicle is to qualify for duty-free treatment. Altogether, over
the next five years, USMCA is projected to increase U.S. gross domestic product
(GDP) by $68.2 billion (0.35 percent) and create 176,000 U.S. jobs, according to
pre-COVID estimates by the independent, nonpartisan U.S. International Trade
Commission (USITC).

Rules of Origin for Automobile Production
For certain products, USMCA revises the rules of origin, which establish the
value that must be added, or processes that must occur, in the territory of one
or more USMCA parties for a good to be considered a USMCA-originating good.
If considered a USMCA-originating good, it will be entitled to preferential dutyfree treatment upon importation into the territory of a USMCA party (USTR
2020a). In particular, USMCA increases the regional content requirements for
automobiles traded under the agreement. NAFTA required 62.5 percent (by
value) of an automobile’s parts to be sourced in North America, and USMCA
raises the requirement to 75 percent, depending on the vehicle type (USTIC

Pursuing Free, Fair, and Balanced Trade

| 263

2019). In addition, under USMCA, North American auto manufacturers that
trade under the agreement must purchase at least 70 percent of their steel and
aluminum from the United States and its territories, Canada, or Mexico. Finally,
annually 40 to 45 percent of the manufacturing costs of imported automobiles
must be produced by workers who earn at least $16 per hour.
Although the USITC estimates that these provisions will create on net
28,000 jobs in the U.S. automobile industry, they will likely reduce growth in
other sectors of the economy. And though these negative effects are more than
offset by other USMCA provisions, these provisions will collectively raise the
costs of producing automobiles in the United States, and thus increasing the
consumer prices of cars and reducing real incomes (USITC 2019).

Digital Trade
USMCA contains the most comprehensive set of provisions for digital trade in a
U.S. trade agreement. The agreement prohibits discriminatory restrictions on
trade in digital products and services between USMCA partners and ensures
the free movement of data across borders. Moreover, a U.S. company is not
required to disclose proprietary source codes and algorithms to USMCA partners or to locate its computing facilities in their territory as a condition of doing
business (USITC 2019). Altogether, these provisions are designed to reduce
barriers to U.S. investment in Mexico and Canada, including in the financial
services sector. The data transfer provisions are estimated to reduce U.S. trade
costs by between 0.6 and 4.5 percentage points for a broad class of sectors,
ranging from agriculture to manufacturing to business services (USITC 2019).
Many of the economic benefits of USMCA are generated through these
digital trade provisions. Specifically, the USITC estimates that these provisions
will reduce “trade policy uncertainty” (USITC 2019). Although the agreement
prevents USMCA partners from establishing restrictions on trade in digital services, data flowed freely among the United States, Canada, and Mexico before
the signing of the agreement. Nonetheless, research suggests that uncertainty
about whether such regulations will eventually be imposed can reduce trade
and investment between countries, and a reduction in this uncertainty can
yield economic benefits (USITC 2019).

Intellectual Property Protection
USMCA introduces many provisions to enhance the protection and enforcement of the IP rights of U.S. firms. First, USMCA requires that countries grant
patent extensions in response to unreasonable delays in their patent-granting
offices (USITC 2019). USMCA also provides procedural safeguards for the
recognition of new geographical indications, including strong standards for
protection against issuances of geographical indications that would prevent
U.S. producers from using common names, such as mozzarella, as well
as establish a mechanism for consultation between the parties on future

264 |

Chapter 9

geographical indications pursuant to international agreements (CRS 2020a;
USTR 2020a). In addition, the agreement calls for a minimum copyright term of
life of the author plus 70 years, and for those works with a copyright term that
is not based on the life of a person, a minimum of 75 years after first authorized
publication (USTR 2020a).
The USMCA provides for the most comprehensive protection for trade
secrets of any prior U.S. trade agreement. It requires countries to provide,
including with respect to trade secret theft by SOEs, civil procedures and
remedies, criminal procedures and penalties, prohibitions against impeding
licensing of trade secrets, judicial procedures to prevent disclosure of trade
secrets during the litigation process, and penalties for government officials for
the unauthorized disclosure of trade secrets (USTR 2018c). USMCA confirms
that the enforcement of IP also applies to the digital environment, ensuring
that firms relying on digital trade receive adequate protection (USITC 2019).

Labor
USMCA enshrines core worker rights, which will have the most notable effect in
Mexico. The Mexican labor market is hampered by a largely informal economy
and a lack of protections for workers (ILO 2014). Many workers are part of
undemocratic unions, also known as “ghost unions,” that are not supported
by the majority of the workers they allegedly represent. These unions form
illegitimate collective bargaining agreements, alternately known as protection
contracts, with an employer-dominated union, without workers seeing or
ratifying the agreement. Such protective contracts are estimated to represent
a significant percentage of collective bargaining agreements in Mexico. When
workers try to form their own union, their employer tells them that they were
already a part of one and subject to the parameters stipulated in the agreement they never had a chance to see (Mojtehedzadeh 2016). USMCA addresses
each of these issues by supporting Mexico’s creation of independent bodies to
resolve labor disputes—guaranteeing democratic worker representation and
collective bargaining rights—and providing enforcement tools to ensure that
Mexico meets the USMCA’s labor obligations (USTR 2020a).
Although these provisions will primarily affect Mexico by promoting
higher wages and improving labor market conditions for workers, the U.S.
economy could also benefit. To the extent that these provisions reduce this
disparity between Mexican and U.S. wages, U.S. firms will be less likely to offshore production to Mexico, increasing the bargaining power of U.S. workers.
In addition, U.S. export markets may benefit from the increased purchasing
power of the Mexican consumer (USITC 2019).

Reform of the Investor-State Dispute Mechanism
NAFTA contained provisions allowing an investor of one NAFTA party to submit
to international arbitration claims that another NAFTA party (the host party)

Pursuing Free, Fair, and Balanced Trade | 265

' шҊфѵThe (+/*!)) 3$* )- .$)" $)$($.#- .#*'.
* ' .0'/.
0-- )/1'0 *!'' Ҋ*(( - .#$+( )/.
0-- )/1'0 *!ѵѵ Ҋ*(( - .#$+( )/.
3+ / #)" $)1'0 *!ѵѵ Ҋ*(( - .#$+( )/.

)
җ*''-.Ѷ
($''$*).Ҙ
ссѶппп
цѶсхп
ттс

3$*
җ*''-.Ѷ
($''$*).Ҙ
чѶцрф
сѶфсц
шр

*0- ѷѵѵ )/ -)/$*)'- *(($..$*)ѵ

had breached investment-related obligations in NAFTA and thereby harmed
the investor or its investment in the territory of the host party. This arbitration
mechanism, known as the Investor-State Dispute Settlement (ISDS), raised
concerns about its effect on the ability of host state regulators to exercise their
prerogative to regulate in the public interest. Under USMCA, the ISDS will no
longer be available between the United States and Canada after June 30, 2023.
Instead, investment-related claims by U.S. investors against Canada and by
Canadian investors against the United States will need to be filed in local courts
(USITC 2019).
For disputes between U.S. and Mexican investors, USMCA limits the scope
of claims that can be submitted to ISDS, except where investors have certain
government contracts in specific economic sectors. When such conditions are
met, it is the view of the United States and Mexico that the risk of breach of
investment-related obligations and consequent harm warrant maintaining
the availability of ISDS. When these conditions are not met, aggrieved U.S.
and Mexican investors must first attempt to resolve their disputes in domestic
courts. Only if these efforts are unsuccessful after a period of 30 months may
they have recourse to ISDS. The USITC estimates that these changes to ISDS
for Mexico will reduce both domestic and foreign capital investment in Mexico
by $2.9 billion (0.44 percent) while slightly increasing investment in the United
States (USITC 2019).

Agricultural Provisions
Agricultural trade among member countries that was already duty-free under
NAFTA will continue to be duty-free under USMCA. Moreover USMCA expands
market access opportunities for U.S. dairy, poultry, and egg exports to Canada.
Canada imposes tariff rate quotas (TRQs) on dairy products, which restrict the
amount of dairy products it can import from other countries. Although Canada
will be permitted to retain its TRQs, USMCA will boost exports of U.S. dairy
products to Canada. Similarly, the agreement will preserve U.S. TRQs for sugar,
while slightly increasing U.S. imports of sugar from Canada (USITC 2019).
USMCA also provides a mechanism for biotechnology cooperation and fair
treatment in quality grading for American wheat and nondiscrimination and
transparency commitments regarding the sale and distribution of alcoholic
beverages. In total, USMCA is expected to increase U.S. agricultural exports by
$2.2 billion per year (1.1 percent) (USITC 2019).
266 |

Chapter 9

Table 9-6. U.S. Employment Sector Effects of the
United States–Mexico–Canada Agreement
Sector
Employment, overall
Agriculture
Manufacturing and mining
Services

Source: U.S. International Trade Commission.

Value
(thousands)
175.7
1.7
49.7
124.3

Percent Increase
over Five Years
0.12
0.12
0.37
0.09

Table 9-7. Effects of the United States–Mexico–Canada Agreement on
U.S. Trade (percent changes relative to the baseline in 2017)
Aspect of Trade
U.S. trade with
Canada
U.S. trade with
Mexico

Exports
(percent)

Exports
(dollars, billions)

Imports
(percent)

Imports
(dollars, billions)

5.9

19.1

4.8

19.1

6.7

14.2

3.8

12.4

Source: U.S. International Trade Commission.

Trade Facilitation
Several USMCA provisions will improve “trade facilitation”—that is, the administrative procedures that enable traded goods to be processed quickly and
efficiently at the border. USMCA changes the threshold below which goods
are exempt from formal customs procedures (the de minimis threshold) for
goods entering Canada and Mexico, while the United States will retain its current thresholds for imports from these countries (USITC 2019). This will help
expedite the customs process for low- and moderate-value packages (i.e., with
a value under $2,500), which will benefit U.S. e-commerce firms in particular
by lowering processing costs. Other measures will boost trade facilitation for
e-commerce firms, for example by permitting electronic authentication for
transactions, e-signatures, and paperless trading. Altogether, these provisions
will boost U.S. e-commerce exports by $424 million (USITC 2019).

Overall Economic Effects
USMCA will have a positive effect on many industries in the U.S. economy,
creating jobs and increasing wages for U.S. workers in the agricultural, manufacturing, and services sectors. Over the next five years, USMCA will increase
U.S. GDP by $68.2 billion (0.35 percent) and will create 176,000 jobs across a
broad range of sectors, according to pre-COVID estimates by the USITC (table
9-6). It will also boost exports to and imports from USMCA partners by $64.8
billion (table 9-7).

Pursuing Free, Fair, and Balanced Trade

| 267

Box 9-1. Bahrain’s and the United Arab
Emirates’ Agreements with Israel
On September 15, 2020, the United States hosted representatives from
Bahrain and the United Arab Emirates as they signed agreements to normalize
relations with Israel. The President of the United Arab Emirates subsequently
issued a decree stating that the United Arab Emirates’ law requiring a boycott
of Israel was repealed. Since the founding of Israel in 1948, the Arab League
has maintained a boycott of the country. (The Arab League consists of 22
Middle Eastern and African countries and entities: Algeria, Bahrain, Comoros,
Djibouti, Egypt, Iraq, Jordan, Kuwait, Lebanon, Libya, Mauritania, Morocco,
Oman, the Palestinian Authority, Qatar, Saudi Arabia, Somalia, Sudan, Syria
(suspended since 2011), Tunisia, the United Arab Emirates, and Yemen.)
Due to the boycott, both Israel and the boycotting countries miss
opportunities for increased trade relations. Before 2020, only two Arab states
had normalized relations with Israel: Egypt (1979) and Jordan (1994). (There
was also a secondary boycott of companies that did business with Israel. It
ended with the Oslo Accords in 1993.) The Palestinian Authority also maintains relations with Israel. A literature review of the Arab boycott found that it
imposed trade costs on Israel of roughly $1 billion a year (Anthony et al. 2015).
If trade between Israel and Bahrain and the United Arab Emirates grows
in line with other countries that have normalized relations with Israel, it would
increase by an estimated $537 million annually. In order to estimate the trade
effect of Bahrain and the United Arab Emirates normalizing relations with
Israel, we use the examples of Egypt and Jordan. In 2019, Israel imported $195
million worth of goods from Egypt and Jordan, and exported $209 million of
goods to them. This is roughly 0.12 percent of Egypt and Jordan’s combined
GDP. The CEA estimates that if Israeli imports and exports to Bahrain and the
United Arab Emirates increased to this same percentage of Bahrain’s and
the United Arab Emirates’ GDP, annual Israeli imports and exports with them
would increase to $258 million and $278 million, respectively. The vast majority of this increase will occur with the United Arab Emirates, as its economy is
much larger than Bahrain’s.

Other Trade Agreements
Beyond USMCA and the Phase One agreement, the Trump Administration has
sought to improve the United States’ terms of trade with other countries. In
2018, the United States renegotiated parts of its trade deal with South Korea to
ensure fair trade for the U.S. auto industry. In 2019, the United States reached a
“Stage One” trade agreement with Japan to reduce or eliminate tariffs on many
U.S. food and agricultural goods exports, and then in 2020 opened talks with
the United Kingdom to pursue a free trade agreement. In addition, the United

268 |

Chapter 9

States hosted representatives from Bahrain and the United Arab Emirates as
they signed agreements to normalize relations with Israel (see box 9-1).

U.S.-Japan Trade Agreements
On October 7, 2019, the United States and Japan signed a “Stage One” outcome
from bilateral negotiations consisting of two individual agreements: the U.S.Japan Trade Agreement (USJTA), and the U.S.-Japan Digital Trade Agreement
(USTR 2019c, 2019d). The USJTA provides for Japan cutting or eliminating
agricultural tariffs and quota restrictions for scores of U.S. products, such as
beef and nuts, in return for limited cuts and the elimination of import tariffs
on U.S. industrial and agricultural goods (CRS 2019; Schott 2019). Each side
respectively agreed to remove or reduce restrictions on about $7.2 billion in
imports ($14.4 billion total), covering roughly 5 percent of all trade between
the countries (CRS 2019).
The Digital Trade Agreement covers $40 billion in digital trade, and negotiators modeled many of the provisions in the deal after those in the USMCA
(USTR 2020e). The agreement ensures barrier-free data flows, prohibits data
localization laws that mandate having domestic computing facilities, and
prohibits arbitrary disclosures of imported source codes and algorithms. Other
provisions include prohibitions on customs duties on electronic transmissions
(CRS 2019).
Several topics were not included in negotiations and were left for a
larger, future deal (CRS 2019). A key Japanese trade objective revolves around
automobiles, Japan’s top export to the United States. Japan hopes to reduce
current U.S. tariffs on its auto exports, as was originally negotiated in the TransPacific Partnership, and to ensure that new tariffs are not imposed (Goodman
et al. 2019). Although Japan imposes no tariffs on U.S. auto exports, the United
States maintains that nontariff barriers, such as certain testing protocols for
automobiles, limit U.S. exports with the result that Japan’s automotive exports
to the U.S. are 23 times higher than its imports from the United States (USTR
2019a; CRS 2019). Consistent with recent practice, the United States will also
seek provisions on exchange rate issues (USTR 2018b).

The U.S.-South Korea Free Trade Agreement
On September 24, 2018, the United States and South Korea signed an agreement to revise the U.S.–South Korea Free Trade Agreement (KORUS). The
agreement mainly focuses on automobile trade, considering that the largest
share of the $9 billion trade deficit that the U.S. has with South Korea is
concentrated in the automobile sector (Tankersley 2018; Overby et al. 2020).
For example, a key provision centers on U.S. auto exports to South Korea that
adhere to U.S. safety regulations. Under the renegotiated agreement, South
Korea’s allowance for U.S. exports of automobiles meeting U.S. safety standards doubled, from 25,000 to 50,000 per manufacturer per year, allowing U.S.

Pursuing Free, Fair, and Balanced Trade

| 269

exporters to ramp up sales and marketing for future exports and thus avoid
concerns about reaching the 25,000 limit. The United States will maintain its 25
percent tariff on Korean trucks through 2041 (USTR 2018e).
Beyond automobiles, KORUS and accompanying side letters also address
several other issues. These include updating certain trade remedy provisions
and improving customs procedures in South Korea related to verification of
U.S. origin, something that is necessary for U.S. exporters to claim tariff benefits under the agreement without unnecessary delays (CRS 2018; USTR 2018d).

Limited Trade Agreements with Brazil and Ecuador
On October 19, 2020, the United States and Brazil reached an agreement that
focuses on trade rules and transparency, contributing to the elimination of
nontariff barriers between the two countries. The deal will facilitate the processing of goods at the countries’ borders, enhance regulatory transparency,
promote good regulatory practices, and strengthen rules addressing corruption. Other trade issues are expected to be addressed in a more comprehensive
agreement, including those related to IP and agricultural issues (CRS 2020b).
On December 8, 2020, the United States and Ecuador signed an agreement that builds on the U.S.-Ecuador Trade and Investment Council Agreement
that has been in effect since 1990. Like the agreement with Brazil, the new
agreement with Ecuador will promote bilateral trade by updating trade facilitation between the two countries, enhancing regulatory transparency, and
strengthening anticorruption efforts. The new agreement also seeks to foster
trade and investment opportunities for small and medium-sized enterprises in
the two countries.

U.S.-U.K. Negotiations
The United Kingdom formally exited the European Union on January 31, 2020,
but has remained in a “status quo” transition period through December 31,
2020 (Henley, Rankin, and O’Carroll 2020). Consequently, the U.K. began negotiating its own trade agreements in 2020, which could enter into force from
January 1, 2021. In May 2020, the U.S. and the U.K. launched negotiations on
a comprehensive free trade agreement and conducted intensive negotiations
throughout 2020. The United States’ objectives for the agreement are to reduce
or eliminate market access barriers to U.S. industrial, agricultural, services,
and digital products to the U.K., including regulatory differences that impede
bilateral trade, and to deepen the already-extensive economic relationship
between the U.S. and the U.K. to support employment and economic growth
(USTR 2020f). Though the U.S. and the U.K. share the goal of an ambitious
agreement, the U.K. has political sensitivities in areas such as agricultural
market access and regulations governing product standards and food safety
(Packard 2020).

270 |

Chapter 9

Separately, after the transition period between the United Kingdom
and the European Union ends on January 1, 2021, the U.K. will no longer be
covered by existing EU agreements with other countries, including the United
States. Therefore, the U.S. and the U.K. completed new bilateral agreements
and mechanisms to ensure that there is no disruption to trade in certain
products—such as wine, distilled spirits, and marine and telecommunications
equipment—all of which are covered by existing U.S.-EU agreements (USTR
2020f). Finally, the United States and the United Kingdom have other bilateral
trade differences, including the U.K.’s implementation of digital services taxes
targeting U.S. multinational firms, unresolved World Trade Organization disputes on large civil aircraft that have resulted in retaliatory tariffs on both U.S.
and U.K. exports to each other, and U.S. tariffs on steel and aluminum imports
and the resulting EU (and U.K.) retaliatory tariffs (CEA 2019; Elliott and Mason
2020; Isaac 2020).

The Rise of Global Supply Chains
In early 2020, the global economic outlook changed dramatically as the
coronavirus responsible for the COVID-19 spread first through china and then
through much of the rest of the world. The COVID-19 pandemic disrupted
economic activity everywhere it spread, as people restricted their movement
to avoid health risks, and governments closed schools and nonessential businesses in order to mitigate the public health threat. Moreover, in an increasingly globalized economy, localized outbreaks of the virus created effects
that rippled beyond local borders to the rest of the world. The emergence of
global supply chains over the past decades meant, for example, that covid-19
disrupted production in auto assembly plants in north america even before the
pandemic spread to the United States, as plants in North America assembled
parts produced in wuhan, china. The risks posed by disease, natural disasters,
and trade wars have caused firms and governments to rethink global supply
chains and, in some cases, bring production closer to home (Schlesinger 2020).

China and the Emergence of Global Supply Chains
The past three decades have seen a rapid expansion of international trade,
and in particular, the use of global supply chains, as firms in the United States
and around the world have relocated production off shore to take advantage
of lower costs of labor and other inputs. Global supply chains have allowed for
specialization and net gains from trade, resulting in increased productivity and
lower costs for consumers (Grossman and Rossi-Hansberg 2008). Firms seeking
lower manufacturing costs often found them in China, with its large supply of
labor and resulting low labor costs (Cui, Meng, and Lu 2018).
Although U.S. consumers and importing firms, as well as competitive exporters, have benefited from globalized supply chains, some recent

Pursuing Free, Fair, and Balanced Trade

| 271

academic literature indicates that the establishment of Permanent Normal
Trade Relations (PNTR) with China at the end of 2001 imposed steep costs
on U.S. manufacturing employment and innovation (Pierce and Schott 2016;
Autor, Dorn, and Hanson 2019). PNTR status meant that the United States
extends permanently nondiscriminatory treatment to the products of China
(GPO 2000). Import competition from countries with low labor costs including China has exacerbated a reduction in U.S. manufacturing employment,
accounting for a quarter of the total decline in U.S. manufacturing employment
between 1990 and 2007 and causing lower labor force participation, higher
unemployment, and lower wages in affected communities (Autor, Dorn, and
Hanson 2013). Abraham and Kearney (2020) find that import displacement
after the establishment of PNTR with China was the single biggest factor in the
decline of the overall U.S. employment-to-population rate after 1999, although
automation also played a role.
Displaced American workers have struggled to transition to new opportunities, resulting in higher utilization of government safety net programs,
including unemployment, disability, and healthcare benefits (Autor, Dorn,
and Hanson 2013). The lack of valuable work reduced young males’ marriage
prospects, which was a factor in there being a greater share of single-headed
households in communities affected by the China trade shock (Autor, Dorn,
and Hanson 2019). Moreover, the worsening labor market conditions exacerbated socioeconomic distress, leading to substance abuse and increased
mortality from drug overdoses, suicides, and liver diseases (Autor, Dorn, and
Hanson 2019; Case and Deaton 2017; Pierce and Schott 2020).

U.S. Firms Begin to Hedge the Risks of Global Supply Chains
There are, however, some indications that the globalization of American supply chains has begun to partially reverse. First, the 2008 financial crisis was an
unprecedented shock to the global economy, from which the expanding use of
global supply chains has never quite recovered. The expansion of global supply
chains, measured as a share of trade, slowed after the crisis and even reversed
in 2015, the most recent year for which data are available, due to slowing economic growth and trade reforms (World Bank 2020). More recently, tensions
between the United States and China and the global COVID-19 pandemic have
brought into focus some of the risks of global supply chains, causing firms and
governments to look for ways to reduce exposure to these risks (Lund et al.
2020).

The Trade Slowdown in Response to COVID-19
As the COVID-19 pandemic spread around the globe, it disrupted economic
activity through private and public responses to quell its transmission. Real
GDP for all the OECD countries fell 12.2 percent in the first half of 2020, while
U.S. real GDP fell 10.2 percent (not at an annualized rate). Reductions in trade
272 |

Chapter 9

$"0- шҊрѵѵѵ (+*-/.Ѷспсп
Year-over-year change (percent)
рп

/. сп

п
Goods
–рп
**.)
s -1$ .
–сп

-1$ .

–тп

–уп
). сп

-. сп

4 сп

0'. сп

+. сп

*0- .ѷ ).0.0- 0Ѹ'0'/$*).ѵ

$"0- шҊсѵѵѵ3+*-/.Ѷспсп
Year-over-year change (percent)
рп

/. сп

–рп

**.)
s -1$ .

**.

–сп

-1$ .

–тп

–уп
). сп

-. сп

4 сп

0'. сп

+. сп

*0- .ѷ ).0.0- 0Ѹ'0'/$*).ѵ

Pursuing Free, Fair, and Balanced Trade | 273

' шҊ8ѵ (+*-/.)3+*-/.*! -1$ .4 /*-Ѷ/#-*0"# +/ ( 2020

/*-

(+*-/.
спсп
спсп
#)"
1 '
!-*(
җ*''-.Ѷ
спрш
$''$*).Ҙ
җ+ - )/Ҙ
ттѵч
–хшѵч
фчѵц
–туѵх
уѵц
–сцѵр
пѵч
–схѵх

-1 '
-).+*-/ation
$)/ )) )r +$-s -1$ .
*)./-0/$*)
' *((0)$/$*).Ѷc*(+0/ -sѶ)
information services
тпѵф

тсѵч
$))$'s -1$ .
тфѵш
#-" .!*-0. *!i)/ '' /0'p-*+ -/4
спѵф
*1 -)( )/g**.)s -1$ .
/# -b0.$) ..s -1$ .
шфѵх
-.*)'Ѷc0'/0-',)r - /$*)'s -1$ . ршѵч
).0-) s -1$ .
ушѵц
Total

383.0

–рхѵп
–сѵу
пѵц
сѵу
рѵп
13.3

17.8

–21.8

3+*-/.
спсп
спсп
#)"
1 '
!-*(
җ*''-.Ѷ
спрш
$''$*).Ҙ
җ+ - )/Ҙ
хцѵр
–фчѵу
уцѵц
–тцѵс
рсѵц
–ууѵц
сѵп
–счѵр
уфѵр
рртѵп
шфѵт
ршѵс
рффѵр

–2.1
пѵс
–сѵр
тѵт
–рѵц

рфѵр
рсѵу

–ссѵр
–рпѵу

584.5

–19.7

*0- .ѷ ).0.0- 0Ѹ'0'/$*).ѵ

exacerbated declines in GDP. Global merchandise exports as reported by the
World Trade Organization fell 14.0 percent in the first half of 2020 compared
with the first half of 2019, and U.S. merchandise trade (exports plus imports)
fell by 13.6 percent over the same period.
Total U.S. trade of goods and services (exports plus imports) through
October was $645 billion (13.7 percent) below 2019 levels, with the second
quarter falling $355 billion (25.1 percent). The drop in trade was largest in May,
with imports down 24.4 percent year-over-year and exports down 32.3 percent
(figures 9-1 and 9-2). Separating U.S. trade by goods and services reveals that
the pandemic had a particularly large effect on trade in services. Moreover,
though monthly trade in goods has shown signs of recovering, trade in services
has remained over 20 percent below the previous year.
Trade in services, a major sector of the U.S. economy, was hit particularly hard by the international travel restrictions implemented by the United
States and many other countries around the globe to prevent or slow the
spread of COVID-19. The impact, however, has been concentrated in travel
and transportation, with other service sectors seeing relatively lower declines
in trade. On January 31, 2020, President Trump issued a travel ban for most
non-U.S. citizens coming from China (White House 2020b). Then, on March 11,
President Trump issued a travel ban for most non-U.S. citizens coming from the
Schengen Area, which consists of 26 European countries with a common visa
policy (White House 2020c). Also, effective March 14, the Centers for Disease
274 |

Chapter 9

$"0- шҊтѵѵѵ (+*-/.*!**. - )/#)" !-*( ' /
*0)/-$ .Ѷспсп
Year-over-year change (percent)

рф

/. спсп
# )" )-

./*!w*-'

–рф

–тп

#$)
) and 3$*

–уф

–хп
). сп

-. сп

4 сп

0'. сп

+. сп

*0- .ѷ ).0.0- 0Ѹ'0'/$*).ѵ

Control and Prevention (CDC) issued a no sail order for all cruise ships (CDC
2020). International tourism and passenger travel services have taken the largest hit from COVID-19.
The effect of the travel restrictions and the private sector’s response to
COVID-19 is apparent when looking at the level of service trade broken down
by the type of service. As discussed, the largest decreases are in travel and
transportation services (table 9-8). Imports and exports of travel have fallen
69.8 percent and 58.4 percent, respectively, accounting for over two-thirds of
the total drop in service trade. The European Union, much of which was covered by the Schengen ban, has seen the largest decline in service trade with the
United States, with imports from the EU down 47.4 percent and exports to the
EU down 37.8 percent in 2020 through September.
Through October 2020, imports of goods to the United States have
decreased by 9.2 percent year-over-year. Most of the drop came in the second
quarter, when nominal U.S. imports of goods fell by 20.0 percent. However,
imports to the United States have fallen at different times for different countries (figure 9-3).
In the first quarter of 2020, imports from China plummeted reaching
their lowest point in March at 36.5 percent below their March 2019 value. This
quarter was also when COVID-19 was widely present in China, and China shut

Pursuing Free, Fair, and Balanced Trade

| 275

down its economy to address the health crisis. By comparison, in March, U.S.
imports of goods from the rest of the world were only 1.3 percent below their
March 2019 value, indicating that the economic effect of COVID-19 was mainly
concentrated in China but had begun spreading to other countries.
In the second quarter of 2020, U.S. imports of goods saw their steepest
declines in total. In April and May, imports were down 23.4 percent compared with the same period a year earlier. This drop was primarily driven by
a 48.0 percent drop in imports from Canada and Mexico. Imports from the
Schengen Area and the rest of the world also dropped, while imports from
China rebounded, although they remained below 2019 levels. This pattern of
imports may be explained by the fact that China was reopening its economy in
the second quarter, while many other countries, including the United States,
were shutting theirs down beginning in March. As governments around the
world imposed shelter-in-place orders shuttering many businesses, imports
fell. Imports of goods began to rebound in June and July as governments lifted
these orders.

The Decline in Imports of Intermediate Goods
As discussed above, the emergence of global supply chains has driven a
dramatic increase in international trade in recent years. Global supply chains
drive trade in intermediate goods, which are goods used in the production of
finished goods. The COVID-19 pandemic has had a particularly disruptive effect
on supply chains through its effects on trade in intermediate goods.
Because global supply chains intricately connect distant locations, localized shutdowns due to COVID-19 disrupted economic activity around the world
through “supply chain contagion” (Baldwin and Tomiura 2020). On January
23, 2020, China closed the city of Wuhan, a major manufacturing hub in Hubei
Province, and locked down additional cities shortly thereafter, leading to supply chain disruptions in the United States as Chinese businesses ceased production of intermediate goods intended for the U.S. (Xie 2020). Luo and Tsang
(2020) use a network model to estimate that about 40 percent of the impact of
the Hubei lockdown on global output occurred through the effect on supply
chains both inside and outside China.
The pandemic’s effect on supply chains can be seen by analyzing data on
international trade flows. These data are segmented into categories of goods—
consumption, capital, and intermediate goods—using the broad economic
categories defined by UN Comtrade (2016). Consumption goods are finished
goods that are durable (long lasting) or nondurable (one-time use) goods that
are readily available to a consumer to purchase directly or through a retailer or
wholesaler. Capital goods are durable goods used in the production of other
goods. Finally, as discussed, intermediate goods are goods used as inputs in
the production of finished goods, and accordingly are important for supply
chains.
276 |

Chapter 9

' шҊ9ѵѵѵ (+*-/.ѷ#)" !-*(спрш 1 '.!*- ' / *0)/-$ .
) "$*).Ѷ44+ *!**Ѷ/#-*0"#/* -спсп
*0)/-4*-
"$*)

)

)/ -( $/

*).0(+/$*)

+$/'
*''-.
җ$''$*).Ҙ

- )/

*''-.
җ$''$*).Ҙ

- )/

*''-.
җ$''$*).Ҙ

- )/

–тѵс

–рхѵц

–пѵч

–тѵр

–сцѵп

Ҋрчѵт

#$)

–ррѵр

–цѵш

–рчѵш

–рхѵч

–руѵш

Ҋруѵс

3$*
# )" )
-
./*!2*-'
*-'

–рпѵх

–ррѵф

пѵт

пѵц

–рцѵс

Ҋруѵч

хѵт

рпѵп

–цѵц

–чѵч

уѵц

тѵс

шѵф
–шѵп

чѵп
–сѵр

–уѵх
–трѵц

–сѵх
–цѵр

–тшѵх
–шуѵп

Ҋррѵч
Ҋррѵр

*0- .ѷ ).0.0- 0Ѹ'0'/$*).ѵ

As discussed above, early in 2020, imports of goods to the United States
declined, primarily as a result of China’s lockdowns. In April and May, as the
pandemic spread and governments around the world adopted shelter-in-place
orders, imports of goods slowed down more dramatically. Imports began to
recover in June and July as governments lifted their shelter-in-place orders.
However, while imports of consumption goods rebounded sharply in June,
imports of intermediate goods remained at about 15 percent below 2019
levels. By October, imports of consumption goods were 3 percent above 2019
levels; however, imports of intermediate goods continued to lag behind, at 3
percent below 2019 levels.
Through October 2020, U.S. imports of intermediate goods from the
world decreased 11.1 percent year-over-year, primarily driven by the declines
in imports from Canada, Mexico, and China (table 9-9). Imports of consumption
goods are down only $31.7 billion from 2019 levels, while intermediate goods
are down $94.0 billion.
Imports of consumption goods may have rebounded faster than imports
of intermediate goods because as governments lifted shelter-in-place orders,
consumers were able to increase their spending faster than firms were able
to ramp up their production. As production continues to rebound, it is not yet
clear whether U.S. imports of intermediate goods will return to pre-pandemic
levels, or whether some supply chains will relocate to the United States.

Evidence That Firms Are Reducing Their Exposure to China
With the COVID-19 pandemic and trade policy uncertainty highlighting risks in
global supply chains, U.S. businesses are considering moving production away
from China to either other Asian countries, closer to the U.S. (near-shoring),
or back to the U.S. (reshoring). Although a lag in the data commonly used to
measure global supply chains prevents the CEA from directly observing recent
changes in supply chains, surveys can provide a leading indicator of firms’

Pursuing Free, Fair, and Balanced Trade | 277

$"0- шҊуѵѵѵ (+*-/.*! **.!-*(/# *-'44+ *!
**Ѷспсп
Year-over-year change (percent)
сп

/. спсп

рп

+$/'

–рп
*).0(+/$*)
)/ -( $/

–сп

–тп
). сп

-. сп

4 сп

0'. сп

+. сп

*0- .ѷ ).0.0- 0Ѹ '0'/$*).ѵ

plans to locate facilities. In the past year, several surveys have attempted to
evaluate the extent to which firms in the United States and elsewhere plan to
change supply chains through the relocation of production or the diversification of input sourcing.
In January 2020, before the full extent of the COVID-19 pandemic was
known, the Bank of America surveyed analysts covering 3,000 firms ($67 trillion
in market capitalization) across North America, Europe, the Asia-Pacific region
(excluding China), and China, finding “clear evidence” of movement in global
supply chains. Across 12 sectors, 80 percent of firms with global supply chains
($22 trillion in market capitalization) are expected to shift “at least a portion
of their supply chains” from current locations. The report concludes that “the
trend is clear: global supply chains are on course to be uprooted and brought
home, or transplanted to strategic allies.” In a July update, the Bank of America
found that three quarters of that 80 percent were expanding their reshoring
plans (Bank of America 2020a, 2020b).
A March UBS survey of chief financial officers (CFOs) in the U.S., North
Asia, and China suggests that 20 to 30 percent of capacity represented by
these executives will relocate from China (UBS 2020). If actuated, this would
move between $500 billion and $750 billion in current Chinese exports out
of China. A June 2020 UBS survey shows that, among U.S. firms with manufacturing in China, 76 percent have moved or are planning to move some of

278 | Chapter 9

their manufacturing capacity out of China. Leading candidate destinations for
relocation are the United States, Canada, Japan, and Mexico.
Among North Asian (excluding Chinese) firms responding to the UBS
survey, 85 percent of CFOs have moved or are planning to move capacity from
China to home markets in Japan, Taiwan, and South Korea (UBS 2020). This
represents over 30 percent of the Chinese production among the firms in the
survey. The CFOs identified Vietnam, Thailand, and India as potential locations.
Even Chinese firms have relocated or are planning to do so—60 percent of CFOs
indicated that they would relocate a combined 30 percent of their Chinese
production. Further, respondents reported plans to establish or expand supply
chains closer to customers.
Among U.S. firms, 34 percent of CFOs reported that they had manufacturing in China, of whom 76 percent have moved or are planning to move capacity
out of China. For those planning to move, they indicated that they would shift
a combined 46 percent of their Chinese production. Plans to relocate are most
common among healthcare firms (92 percent), consumer staples (89 percent),
and technology (80 percent), followed by consumer discretionary (76 percent),
industrials (69 percent), and materials (57 percent). The Bank of America
(2020a) finds similar results for the Asia-Pacific region (excluding Chinese)
firms. Half of the sectors have firms that have already moved or intend to move,
largely to Southeast Asia and India. The survey finds that firms in 83 percent
of U.S. sectors representing $3.8 trillion in market capitalization have already
moved or are intending to do so.
Recent data on manufactured imports from low-cost Asian producers
support the survey evidence that firms are moving supply chains out of China.
The management consulting company Kearney reports the ratio of the value
of manufactured goods imported to the U.S. from 14 low-cost Asian countries
relative to the value of U.S. domestic gross output of manufactured goods
(Kearney 2020). A decline in this manufacturing import ratio does not necessarily indicate that production is reshoring to the U.S., but it does indicate
substitution away from supply chains running through these 14 countries.
In 2019, the manufacturing import ratio fell for the first time since 2011
(figure 9-5). The decline, from 13.1 percent in 2018 to 12.1 percent in 2019, was
driven by a 7 percent contraction in imports from the low-cost Asian countries
(the numerator), with U.S. domestic gross output of manufactured goods (the
denominator) essentially unchanged. The decline in imports was led by a 17
percent contraction in trade with China. Whereas 65 percent of goods imported
from the 14 Asian low-cost countries came from China in 2018:Q4, only 56
percent did so as of 2019:Q4. Though transshipment, where goods are altered
slightly to change their originating status, is likely a partial factor, this cannot
explain the entirety of the shift.
Another Kearney measure, the “near-to-far” ratio, measures the value
of U.S. imports from Mexico divided by the value of imports from the same
Pursuing Free, Fair, and Balanced Trade

| 279

$"0- шҊфѵѵѵ )0!/0-$)" (+*-//$*Ѷсппч–рш
Ratio (percent)
ру
рс
рп
ч
х
у
с

п
сппч сппш спрп спрр спрс спрт спру спрф спрх спрц спрч спрш
*0- .ѷ Ѹ0- 0*!*)*($)'4.$.Ѹ

-) 4җспспҘѵ

$"0- шҊхѵѵѵ -Ҋ/*Ҋ!-/$*Ѷсппш–рш
Percent (imports from Mexico / imports from Asian low-cost countries)
уф
уп
тф
тп
сф
сп
рф
рп
ф

п
сппш спрп спрр спрс спрт спру спрф спрх спрц спрч спрш
*0- .ѷ Ѹ 0- 0*!*)*($)'4.$.Ѹ

280 | Chapter 9

-) 4җспспҘѵ

14 Asian low-cost countries. An increase in the ratio indicates greater nearsourcing from Mexico relative to Asia (figure 9-6). Much like the reshoring index,
the near-to-far ratio jumped in 2019 as U.S. importers substituted away from
Asia and toward imports from Mexico.

Drivers of Shifting Supply Chains
Although no single factor is responsible for global supply chain dynamics, the
confluence of higher wages in China, technology and automation, trade policy
tensions and tariffs, and the COVID-19 pandemic have all been factors in a
change in how firms evaluate supply chains. In this section, we briefly discuss
these factors and interactions between them.
Wages in China have risen relative to many other countries in recent
years. In the 1970s, observers noted the seemingly “unlimited” labor supply in
China as workers migrated from rural China to urban areas (Cui, Meng, and Lu
2018). Whereas 80 percent of the population lived in rural areas in the 1970s,
the rural share of the population had shrunk to 43.9 percent by 2015. Owing
partially to China’s “one child” policy, China’s population is aging, exacerbating a restriction in labor supply now and in the future. The resulting rising
wages are eroding Chinese cost competitiveness (figure 9-7) and incentivizing
firms to consider manufacturing in Southeast Asia or closer to home with

$"0- шҊцѵ1 -" *)/#'4*($)'" .$)ѵѵ*''-.Ѷ
сппп–спрш
Wages (nominal U.S. dollars)
шпп

спрш

чпп

#$)

цпп
'4.$

хпп

#$')

фпп
упп
тпп

3$*
#$'$++$) .
$ /)(

спп
рпп

)*) .$

п
сппп сппс сппу сппх сппч спрп
*0- ѷ )/ -)/$*)' *--")$5/$*)ѵ

спрс

спру

спрх

спрч

Pursuing Free, Fair, and Balanced Trade

| 281

greater automation. Higher Chinese wages in comparison with other manufacturing centers persist, even when controlling for productivity.
Businesses are also increasingly aware of growing geopolitical tension
between the U.S. and China. Even if the U.S. and China remove tariffs in the
short term, businesses are adjusting for long-term tensions. COVID-19 has compounded these concerns, as firms recognize the need to increase resiliency. As
firms consider moving out of China, they may look to automation to make up
differences in labor productivity or to offset wage costs from relocating supply
chains to countries with higher wages. The Bank of America estimates that
by 2025, global robot installations will increase by 2.5 million, doubling 2019
levels (Bank of America 2020a).
U.S. tariffs on Chinese goods have been factors in the relocation of supply
chains. In addition to tariffs, businesses are also aware of the broader geopolitical context of the U.S.-China relationship, and believe that regardless of nearterm tariff policy, the two countries will face frictions. For this reason, Bank of
America analysts cite both tariffs and national security as primary reasons for
anticipated reconfigurations of global supply chains, with U.S.-China tariffs
expected to persist regardless of the Phase One trade deal (Bank of America
2020b). Since the Bank of America published its survey, the COVID-19 pandemic
appears to have further exacerbated U.S.-China tensions. As evidence of the
role of national security concerns in driving supply chain relocation, Intel and
TSMC (manufacturers of computer chips) have located plants in the United
States, despite higher labor and capital costs (Wu 2020).

Conclusion
The Trump Administration has reasserted U.S. interests in international trade
policy by forging new bilateral trade agreements with China and Japan, renegotiating the U.S. trade deal with South Korea, and reshaping regional trade
by modernizing the United States’ trade agreement with its most important
trading partners, Canada and Mexico. The Phase One Agreement with China,
when fully implemented, promises to achieve an unprecedented expansion of
U.S. exports to China, and commits China to internal reforms that will make
the country an improved trading partner for the United States. USMCA achieves
new protections for U.S. interests across a range of areas, including digital services, intellectual property, and labor. These agreements go well beyond the
lower tariffs that have been the focus of past trade agreements by addressing
structural and technical barriers to free and fair trade. Along with progress on
other trade agreements, these two major milestones will continue to drive U.S.
economic growth and create American jobs.
The COVID-19 pandemic reduced overall international trade and has
brought into sharp focus some risks of existing global supply chain configurations that previously may not have been fully priced in. These supply chains

282 |

Chapter 9

have the potential to tap into countries’ comparative advantages to create
mutual gains from trade. But global supply chains are susceptible to disruptions from pandemics, along with natural disasters and geopolitical tensions.
As the private sector considers the relocation of supply chains, governments
must weigh the benefits for some consumers and firms against the emerging
understanding of the full costs to sectors facing import competition, as well as
the costs associated with the risks of supply chain disruption.

Pursuing Free, Fair, and Balanced Trade

| 283

Part III

An Effort to Rebuild
Our Country

285

x
Chapter 10

The Year in Review and
the Years Ahead
In 2020, the U.S. economy experienced the single largest adverse economic
shock since the Great Depression due to COVID-19. The Business Cycle Dating
Committee of the National Bureau of Economic Research determined that the
economy peaked in February 2020, bringing to an end the economic expansion
that began in June 2009—the longest such expansion in U.S. history. There were
record declines in real gross domestic product (GDP) and payroll employment
in the second quarter of 2020. Despite the sharpest rebound in real GDP and
payroll employment since the Great Depression during third quarter, the U.S.
economy has only partially recovered from its April 2020 nadir.
Declines in payroll employment were concentrated among low-wage workers. Overall, inflation in 2020 was similar to that in previous years, given
that several months of substantial deflation early in the year were offset by
higher-than-average inflation in later months. Housing markets and interest
rates were affected by the pandemic, but not to the same extent as real GDP or
employment.
As of the writing of this Report, fourth-quarter data for most indicators are not
yet available. Blue Chip forecasts anticipate strong compensatory growth in
2021. However, GDP forecasts and the slowing pace of the recovery of labor
force participation show that many of these issues will persist through at least
2021.
The first three chapters of this Report provide deep analysis of the major macroeconomic developments in the U.S. economy during 2020. It is the purpose
of this chapter to provide a more succinct and summary review of the main

287

macroeconomic indicators for the U.S. economy during 2020. The chapter then
discusses the U.S. economy’s future outlook, including potential economic
gains in the event of full implementation of the President’s complete economic
policy agenda, as well as potential downside risks, particularly near-term risks
from the ongoing COVID-19 pandemic.

The Year in Review
This section summarizes the main U.S. macroeconomic indicators during 2020,
with a focus on total economic output, the labor market, inflation, the housing
market, financial markets, and oil markets.

Components of Economic Output
Real GDP fell by 4.6 percent at an annualized rate during the first three quarters
of 2020 (3.5 percent nonannualized). This decline was driven by an unprecedented contraction as a result of COVID-19 and measures taken to control the
virus in the second quarter, which saw real GDP fall at an annualized rate of
31.4 percent (9.0 percent nonannualized)—the largest quarterly decline since
the series began in 1947. The second quarter’s record decline followed a 5.0
percent annualized contraction in the first quarter of 2020. In the third quarter,
real GDP grew 33.1 percent at an annualized rate (7.4 percent nonannualized),
the largest single quarter of economic growth on record and roughly twice the
prior record of 16.7 percent at an annualized rate (3.9 percent nonannualized)
set in the first quarter of 1950. With growth in the third quarter, the United
States has recovered two-thirds of the economic output lost in the first half of
the year due to the pandemic.
The decline in real GDP was widespread, touching nearly every facet
of the economy and component of output (figure 10-1). Consumer spending, which accounts for roughly 70 percent of the U.S. economy, contributed
most to the decline, accounting for 2.2 percentage points of the 3.5 percent
(nonannualized) decline. Third quarter consumption recovered 71 percent of
its decline during the first half of 2020. Investment contributed 0.5 percentage point to the 3.5 percent (nonannualized) decline in real GDP, as increased
residential investment and inventories were more than offset by a drop-off in
nonresidential investment during the first three quarters. In the third quarter,
the level of investment recovered 82 percent of its decline from the first half
of 2020. Net exports made up 0.5 percentage point (nonannualized) of the 3.5
percent decline in real GDP, and government spending made up 0.06 percentage point (nonannualized) of this decline. Government spending (which rose in
the second quarter) and net exports both fell in the third quarter of 2020.

288 | Chapter 10

$"0- рпҊрѵ*(+*) )/.*! 'Ѷршшп–спсп
Dollars, chained 2012 (trillions)

спспѷт

сп

*/'

рф

рп

*).0(+/$*)

*1 -)( )/
)1 ./( )/

п
/e3+*-/.
Ҋф
ршшп

ршшф

сппп

сппф

спрп

спрф

спсп

*0- ѷ0- 0*!*)*($)'4.$.Ѹ'0'/$*).ѵ
*/ ѷ#$)" )*/ .- ..$*)ѵ

Consumer spending. Consumer spending fell markedly during 2020. Over
the course of the first three quarters, personal consumption expenditures
fell by 3.3 percent (nonannualized). Throughout the last 50 years, personal
consumption expenditures have remained between 80 to 95 percent of disposable personal income, but fell to 74 percent of disposable personal income
in the second quarter—its lowest level on record (figure 10-2). As a result, the
personal saving rate—personal saving as a percentage of disposable personal
income—surged from 7.3 percent in the fourth quarter of 2019 to 26.0 percent
in the second quarter of 2020, before declining to 16.1 percent in the third
quarter. On a monthly basis, the personal saving rate peaked in April 2020, setting a record high of 33.7 percent. Total net wealth (consisting of stock market
wealth, housing wealth, and other wealth, less liabilities) also fell in the second
quarter. Wealth data for the third quarter were not available as of the writing
of this Report.
Consumer spending, which as noted above accounts for roughly 70 percent of GDP, provided the largest contribution to the GDP decline in the first
two quarters of 2020 as well as to the GDP expansion in the third quarter. The
patterns in consumption reveal the uneven way that COVID-19 and measures
taken to control the virus affected economic activity. Declines in some components of services were particularly affected: travel industries, physician, and
dental services. In addition, motor vehicle purchases also fell drastically. Real

The Year in Review and the Years Ahead

| 289

Figure 10-2. Consumption and Wealth Relative Share of
Disposable Personal Income (DPI), 1952–2020

Total net wealth share of DPI
8
2020:Q3

Consumption share of DPI
1.0
Consumption
(left axis)
0.9

0.8

0.7

Total net
wealth
(right axis)

0.6
1950

4
1960

1970

1980

1990

2000

2010

2020

Sources: Federal Reserve Board; Bureau of Economic Analysis; CEA Calculations.
Note: Shading denotes a recession.

personal consumption expenditures accounted for 76.5 percent of the decline
in real GDP during the second quarter, having declined by 9.0 percent (nonannualized). While consumer spending on goods contracted 2.8 percent (nonannualized) in the second quarter, spending on services plummeted 12.7 percent
(nonannualized). A resurgence in personal consumption expenditures in the
third quarter of 2020 reflected the partial reopening of businesses impacted by
closures in the first and second quarters. Real personal consumption expenditures increased 7.4 percent (nonannualized) in the third quarter, accounting
for 76.2 percent of real GDP growth in that quarter. Consumer spending in the
services sector alone accounted for 47.5 percent of GDP growth in the third
quarter.
Investment. During the first three quarters of 2020, private investment
fell by 2.9 percent (nonannualized). The declines in the first and second quarters of 2.3 and 14.7 percent (nonannualized), respectively, were followed by a
strong rebound (16.6 percent, nonannualized) in the third. The drop in total
investment was mirrored across each of the three main types of investment:
nonresidential, residential, and inventories (figure 10-3). Notably, the decline
in investment was not as steep as during the Great Recession, and investment
rebounded more quickly, though generally remained below prepandemic levels as of the third quarter.

290 |

Chapter 10

Figure 10-3. Components of Investment, 1990–2020
Dollars (trillions)
4

2020:Q3
Total
investment

Nonresidential
investment

Residential
investment

Inventories
0

–1
1990

1995

2000

2005

2010

2015

2020

Sources: Bureau of Economic Analysis; CEA calculations.
Note: Shading denotes a recession.

Nonresidential investment contracted 1.7 and 7.6 percent (nonannualized) in the first and second quarters, respectively, and then rebounded 5.1
percent (nonannualized) in the third quarter so that the third quarter was 4.6
percent below the 2019:Q4 level. Within nonresidential investment, investment
in structures declined in the first three quarters of 2020, and as of the third
quarter was 14.3 percent below its 2019 year-end level. Declining investment
in oil and gas exploration and production weighed heavily on investment in
structures, with investment in mining exploration, shafts, and wells declining
49.7 percent (nonannualized) in the first three quarters of 2020.
Nonresidential investment in equipment contracted by 14.1 percent
(nonannualized) in the first two quarters of 2020, though unlike investment
in structures, it began to rebound in the third quarter, increasing 13.6 percent
(nonannualized). As of the third quarter, nonresidential investment in equipment was 2.5 percent below its 2019:Q4 level with investment in information
processing equipment surging to 13.4 percent above the 2019:Q4 level, though
industrial and transportation equipment investment were 3.8 and 22.8 percent,
respectively, below their 2019:Q4 level as of the third quarter of 2020. After edging up in the first quarter and declining in the second quarter, nonresidential
investment in intellectual property products increased in the third quarter and
was 1.0 percent below its 2019:Q4 level in the third quarter.

The Year in Review and the Years Ahead

| 291

In the first quarter of 2020, residential investment increased 4.4 percent
(nonannualized), the largest single-quarter increase since 2012:Q4. After contracting by 10.4 percent (nonannualized) in the second quarter, residential
investment surged by 12.9 percent (nonannualized) in the third quarter to a
level 5.6 percent above its 2019:Q4 level.
Inventory investment, or the change between goods produced (or
imported) and goods sold (or exported), contributed negatively to real GDP
growth in the first and second quarters of 2020. As firms invested to rebuild
inventories, private inventory investment accounted for 6.6 percentage points,
or 19.8 percent, of annualized real GDP growth in the third quarter.
Government purchases. Compared with other components of GDP, there
was very little change in government purchases in 2020. During the first three
quarters of 2020, government purchases fell by 0.2 percent (nonannualized). As
a share of GDP, government purchases grew by 3 percent over this period, with
all three categories of government purchases experiencing increases as a share
of GDP (figure 10-4). Federal Government purchases increased substantially in
the second quarter, supported by the Coronavirus Aid, Relief, and Economic
Security (CARES) Act and other types of emergency coronavirus funding (see
chapters 1 through 3 of this Report).
Although Federal spending rose during the first three quarters of 2020,
State and local spending fell. Quarterly Federal Government nondefense consumption rose by $49 billion (in chained 2012 dollars) (13 percent, nonannualized) from the fourth quarter of 2019 to a peak in the second quarter of 2020.
Similarly, during the first three quarters of 2020, quarterly Federal Government
nondefense gross investment rose by $2.9 billion (in chained 2012 dollars) (2
percent, nonannualized). By comparison, during the first three quarters of
2020, quarterly state and local government consumption fell by $47 billion
(in chained 2012 dollars) (3 percent, nonannualized), which outweighed the
increase of $4 billion (in chained 2012 dollars) (1 percent, nonannualized) in
quarterly State and local government gross investment.
Net exports. The first three quarters of 2020 saw a large drop in imports
and an even larger drop in exports. Real net exports (exports minus imports)
increased in the second quarter but fell in the third quarter. Overall, net exports
fell for the first three quarters because the decline in exports was larger than
the decline in imports. During the first three quarters of 2020, real exports of
goods and services fell by $391 billion (in chained 2012 dollars) (15 percent,
nonannualized), while imports fell by $242 billion (7 percent, nonannualized).
As a result, net exports fell by $149 billion over this period.
Trade in goods recovered faster than trade in services. During the first
three quarters of 2020, real exports of goods fell by $176 billion (in chained
2012 dollars) (25 percent, nonannualized) and real imports of goods fell by $46
billion (in chained 2012 dollars) (18 percent, nonannualized). During the same
period, real exports of services fell by $191 billion (in chained 2012 dollars) (25
292 |

Chapter 10

Figure 10-4. Government Purchases as a Share of GDP, 1948–2020
Percentage of GDP
30

2020:Q3

Total
government

20
15
State and local
10
5
0
1948

Federal
defense

Federal
nondefense

1960

1972

1984

1996

2008

2020

Sources: Bureau of Economic Analysis; CEA calculations.
Note: Shading denotes a recession.

percent, nonannualized) and real imports of services fell by $152 billion (in
chained 2012 dollars) (32 percent, nonannualized). By the third quarter of 2020,
real exports and imports of goods had both rebounded by about 20 percent
from their pandemic lows. By comparison, during the same period, real exports
of services had rebounded by less than 1 percent of pandemic lows, and real
imports of services had rebounded by roughly six percent.

The Labor Market
The U.S. labor market experienced historically unprecedented declines in
employment in March and April before posting a strong but partial recovery in
the months immediately thereafter, accentuated by record employment gains
in May and June. The labor force participation rate also fell dramatically before
retracing part of its earlier decline.
Unemployment. The scale and speed of the increase and decrease in
unemployment in 2020 were unprecedented. In February 2020, before the
COVID-19 pandemic struck, the unemployment rate stood at 3.5 percent. It
surged to 14.7 percent in April 2020 before falling sharply in the following
months. As of November 2020, the unemployment rate (U-3) had fallen to
6.7 percent (figure 10-5). During the Great Recession, the unemployment rate
peaked at 10.0 percent in October 2009, but it took more than four years to
fall to 6.7 percent (December 2013). The data for unemployment insurance

The Year in Review and the Years Ahead

| 293

Figure 10-5. Unemployment Rate, 1990–2020
Percentage of labor force
16

Oct. 2020

14
12
10
8
6
4
2
0
1990

1995

2000

2005

2010

2015

2020

Source: Bureau of Labor Statistics.
Note: Shading denotes a recession.

claims during 2020 are also historic: the week of March 21, 2020, saw a tenfold
increase in unemployment claims, from 282,000 to 3,307,000, the largest
increase on record.
The Bureau of Labor Statistics publishes several measures of the unemployment rate. U-3, the official unemployment rate, measures the share of
people in the labor force actively looking for a job who are unable to find one.
U-6 includes all these people, but also those who (1) want a job and are available for work and have looked for a job in the prior 12 months but not in the
past 4 weeks, or (2) have given up looking for a job in the past 4 weeks because
they are discouraged by job prospects, or (3) want a full-time job but are forced
to work part time for economic reasons. As a result, U-6 is a much broader measure of unemployment and labor underutilization. In February 2020, before the
pandemic struck, the U-6 rate stood at 7 percent. It rocketed to 22.8 percent in
April 2020 before falling sharply in the immediately subsequent months. As of
November 2020, U-6 had fallen to 12.0 percent, almost half its pandemic high
but still higher than at any prepandemic point since August 2014—more than
five years into the preceding expansion. Whereas the gap between U-3 and U-6
was small in February 2020, at 3.5 percentage points, it has widened over the
course of the pandemic and stood at 5.2 percentage points in October.
Labor force participation. The labor force participation rate—the fraction
of people who are either working or actively looking for work—fell in 2020,

294 |

Chapter 10

Figure 10-6. Labor Force Participation Rate, 1990–2020
Percent
68

Oct. 2020

66
64
62
60
58
56
1990

1995

2000

2005

Source: Bureau of Labor Statistics.
Note: Shading denotes a recession.

2010

2015

2020

Figure 10-7. Nominal Compensation and Earnings for Private
Industry Workers, 2006–20
Percent change (12-month)
7

2020:Q3

6
Average hourly
earnings

5
4
3
2
Hourly compensation
for private industry, ECI

1
0
2006

2008

2010

2012

2014

2016

2018

2020

Source: Bureau of Labor Statistics.
Note: Shading denotes a recession. ECI = Employment Cost Index.

The Year in Review and the Years Ahead

| 295

after rising in 2018 and 2019 and during the first two months of 2020, reversing
a previous downward trend. From February to October 2020, the rate fell by
1.7 percentage points. It fell to 60.2 percent in April 2020, but had recovered
1.5 percentage points by October (figure 10-6). Notably, most of the recovery
occurred in June. The rate did not recover much further between June and
October 2020. This suggests the possibility that the 1.7-percentage-point drop
in participation may be persistent. Consistent with this hypothesis, Coibion,
Gorodnichenko, and Weber (2020) find that a wave of early retirements
explains much of the drop in participation.
Wages. Average hourly real wages rose by 3 percent during the first three
quarters of 2020, primarily because layoffs were concentrated among lowwage workers (Crust, Daly, and Hobijn 2020) (figure 10-7). While this pattern is
not unique to the current recession, the change in average hourly real wages is
more pronounced now than in previous recessions given the magnitude of the
employment losses in March and April. That the rise in average hourly earnings
is due to the composition of the workforce can be seen in the contrast with the
measure of wages from the Employment Cost Index, which measures wages
directly for a fixed sample of job categories. This fixed-weighted measure,
increased only 2.7 percent during the 12 months through September.

Inflation
Overall, inflation in 2020 was below the average for 2019. The Federal Reserve
has a target of 2 percent inflation for the Personal Consumption Expenditures
Chain-Type Price Index (the PCE Price Index). But this index rose by only 1.4
percent during the 12 months through October, which was little changed from
the year-earlier rate (figure 10-8). The total PCE Price Index includes volatile
food and energy components, and if these are excluded (yielding the “core”
PCE Price Index), inflation rose by only 1.5 percent during those 12 months,
also little changed from the year earlier rate.
Looking at 2020 in detail, some months had negative inflation. In particular, month-over-month inflation was negative in March (–0.1 percent) and April
(–0.4 percent). This deflation was driven primarily by changes in nondurable
prices, which fell 1.1 percent in March and April. However, inflation rose at
above-trend rates in June, July, and August, leading the overall 12-month
change to rise back to the year-earlier rate of about 1.4 percent.

The Housing Market
Home construction and sales were substantially disrupted during the early
part of the year as a result of COVID-19. This disruption did not translate into
a decline in house prices, however, because tight housing supply—partly a
consequence of the pandemic—and strong demand from low mortgage rates
stabilized market conditions in the spring (Gascon and Haas 2020). Indeed,
house prices have actually risen 6.8 percent through the first nine months of
296 |

Chapter 10

$"0- рпҊчѵ*).0( --$ )!'/$*)җP-$ I) 3ҘѶспрс–сп
Percent change (12-month)
тѵп

+ѵсп

сѵф

-' . -1 ’.
'*)" -Ҋ-0)*% /$1

сѵп

рѵф

*-
P-$ I) 3

*/'
P-$ I) 3

рѵп

пѵф

пѵп

спрс

спрт

спру

спрф

спрх

спрц

спрч

спрш

спсп

*0- .ѷ0- 0*!*)*($)'4.$.Ѹ '0'/$*).ѵ
*/ ѷ #$)" )*/ .- ..$*)ѵۙ -.*)'*).0(+/$*)3+ )$/0- .ѵ

$"0- рпҊшѵѵѵ *0.$)"-$ ) 3Ѷршшп–спсп
Index (Jan. 2000 = 100; seasonally adjusted)

0"ѵ спсп

250
спп

рфп

рпп

фп

п
ршшп

ршшф

сппп

сппф

спрп

спрф

спсп

Source: Standard & Poor’s—S&P CoreLogic Case-Shiller National Home Price Index.
Note: Shading denotes a recession.

The Year in Review and the Years Ahead

| 297

Figure 10-10. U.S. New Housing Formation and Single-Family
Home Sales, 1990–2020
Thousands of units (seasonally adjusted)
9,000

Sep. 2020

8,000
7,000
6,000

New and
existing sales

5,000
4,000
3,000
2,000

Housing starts

1,000

Permits

0
1990

1995

2000

2005

2010

2015

2020

Sources: Census Bureau; National Association of Realtors; CEA calculations.

the year, according to the S&P Corelogic Case-Shiller Home Price Index (figure
10-9). The disruption caused a severe but short-lived drop in housing sales,
housing starts, and permits, followed by a complete recovery (figure 10-10).
New housing starts peaked in January 2020 at 1.617 million units (seasonally adjusted, annualized) and fell by 683,000 units (seasonally adjusted,
annualized) or 42 percent by April before recovering 87 percent of its loss to
1.530 million units (seasonally adjusted, annualized) by October. New housing permits reached a prepandemic peak in January 2020 at 1.536 million
units (seasonally adjusted, annualized) and fell by 470,000 units (seasonally
adjusted, annualized) or about 30 percent by April before recovering to 1.545
million units (seasonally adjusted, annualized) by October, slightly above
the prepandemic peak. Total housing starts reached a prepandemic peak in
January 2020 at 1.305 million units (seasonally adjusted, annualized) and fell
by 125,000 units (seasonally adjusted, annualized) or about 10 percent by May
before recovering to 1.343 million units (seasonally adjusted, annualized) by
October, roughly 3 percent above the prepandemic peak.
Existing home sales reached a prepandemic peak in February 2020 at
5.760 million units (seasonally adjusted, annualized) and fell by 1.850 million
units (seasonally adjusted, annualized) or about 32 percent by May before
recovering to 6.850 million units (seasonally adjusted, annualized) by October,
roughly 19 percent above the prepandemic peak. New home sales reached

298 | Chapter 10

Figure 10-11. U.S. Homeownership Rate, 1990–2020
Percent (seasonally adjusted)
70

2020:Q3

60
1990

1995

2000

Source: Census Bureau.
Note: Shading denotes a recession.

2005

2010

2015

2020

a prepandemic peak in January 2020 at 774,000 units (seasonally adjusted,
annualized) and fell by 204,000 units (seasonally adjusted, annualized) or about
26 percent by April before recovering to and stabilizing at about 1 million units
(seasonally adjusted, annualized) by August, roughly 30 percent above the
prepandemic peak. Brokers’ commissions and other ownership transfer costs
for real residential investment contracted 22.7 percent in the second quarter
of 2020, the largest single quarter contraction on record. However, in the third
quarter commissions jumped up 45.3 percent, boosting commissions above
prepandemic levels. The third quarter jump was the largest single-quarter
expansion on record. Brokers’ commissions for real nonresidential investment
fell 6.8 percent in the first three quarters of 2020. This drop is in the same range
as other three-quarter declines experienced during the past three years.
Evictions fell during the 2020 pandemic (see figure 2-6 in chapter 2 of
this Report), due to the CARES Act and the President’s Executive Order 13945
(August 8, 2020), which placed a moratorium on evictions until the end of
2020. The homeownership rate appears to have increased dramatically (figure
10-11), though there have been questions about data reliability because the
U.S. Census Bureau temporarily suspended (though June) personal visits for
the survey, which reduced response rates. The percentage of rental units making rent payments in 2020 remained relatively stable compared with the same

The Year in Review and the Years Ahead

| 299

$"0- рпҊрс. - )/" *! )/4( )/. Ѷспрш–сп
спрш

Rental units (percent)
рпп
шп

спсп

шхѵх шфѵр

шх шфѵш

шхѵх шфѵц

шфѵч шуѵф

шфѵф шуѵх

шхѵх шуѵч

шфѵс штѵх

0)ѵ

0'ѵ

0"ѵ

+ѵ

/ѵ

*1ѵ

чп
цп
хп
фп
уп
тп
сп
рп
п
*0- sѷ National Multifamily Housing CouncilѸ'0'/$*).ѵ

month in 2019, ranging from a drop of 0.1 to 1.8 percentage points (figure
10-12).

Financial Markets
In 2020, U.S. equity markets experienced substantial volatility but ultimately
recovered from losses experienced during the pandemic. On February 19, the
Standard & Poor’s 500 index closed at 3,386, a prepandemic peak in 2020.
However, by March 23, 2020, the S&P 500 index had fallen by 31 percent for the
year. Yet, by August 18, it closed at a higher level than on February 19 and, following brief downswings in September and October, achieved several all-time
highs starting in mid-November and continuing into at least early December.
The Dow Jones Industrial Average index, which measures performance of
shares of the 30 largest U.S. corporations, followed a similar trend, closing at
a prepandemic peak of 29,551 on February 12, falling 37 percent by March 23,
and achieving all-time highs starting in mid-November and continuing through
at least early December. The NASDAQ index, which is heavily weighted with
shares of technology firms, experienced shallower losses and a larger recovery,
closing at a prepandemic peak of 9,817 on February 19, falling roughly 30
percent by March 20, and achieving an all-time high of 12,056 on September 2,
roughly 23 percent above the prepandemic high.

300 |

Chapter 10

$"0- рпҊртѵ-& / *'/$'$/4 ) 3җ ҘѶсппц–сп
Index

шп

/ѵ тпѶ спсп

чп
цп
хп
фп
уп
тп
сп
рп
п
2007

2009

2011

2013

2015

2017

2019

*0- ѷWall Street Journalѵ
*/ ѷ#$)" )*/ . - ..$*); VIX = Chicago Board Options Exchange’s Volatility Index.

Stock market volatility increased during the pandemic and remained
elevated through at least October 30. The Chicago Board Options Exchange’s
Market Volatility Index measures the market’s assessment of the volatility of
the stock market (derived from options prices). This measure of volatility rose
from 12.5 on January 2, to reach a peak of 82.7 on March 16, before dropping
to 38.0 by October 30 (figure 10-13). As concerns over COVID-19 escalated in
February and March, enormous selling pressures led to a precipitous deterioration in corporate bond market liquidity conditions. In the two weeks prior to
the Federal Reserve’s announcement of numerous credit and liquidity facilities
on 23 March, bond transaction costs soared in both high-yield and investment
grade bonds. Over those two weeks, the average cost for investment-grade
bond transactions tripled, jumping from 30 basis points in February to a peak
of almost 90 basis points in mid-March. Similarly, transaction costs among
high-yield bonds jumped up from around 50 basis points in February to nearly
110 basis points in mid-March (Sharpe and Zhou 2020).
The spread between corporate bond yields and comparable Treasury
yields took a similar path. High-yield corporate bond spreads rose from just
below 4 percentage points in early February to just below 11 percentage points
in Mid-march. Investment grade spreads quadrupled to 4 percent in mid-March
from 1 percent in February. After the announcement and implementation

The Year in Review and the Years Ahead

| 301

of several credit facilities, corporate bond spreads substantially eased and
transaction costs saw initial declines. As the Federal Reserve’s lending facilities
continued to offer relief, corporate bond yield spreads generally continued
to fall throughout the rest of the year, reaching pre-pandemic levels in both
investment-grade and high-yield bonds. Chapter 3 of this report discusses the
specific Federal Reserve lending facilities implemented to address the corporate bond market crises.

Interest Rates
U.S. Treasury notes are the main form of debt issued by the Federal Government,
and their interest rates are relevant to Federal interest expenses. Because corporate debt usually move roughly in parallel with government debt, these rates
affect the cost of business borrowing as well.
The Federal Reserve’s Open Market Committee (FOMC) lowered its target
for the Federal funds rate by 150 basis points to 0.125 percent at unscheduled
meetings on March 3 and March 15. Most short-term rates dropped by similar
amounts. For example, the yield on 3-month Treasury bills fell by 145 basis
points during March. The yield on 10-year Treasury notes (which averages the
expected value of short-term rates during the next 10 years), fell by 48 basis
points in March and another 20 basis points in April.

$"0- рпҊруѵTenҊY -$)0. ThreeҊ*)/#- .0-4
*)./)//0-$/4/ Ѷршшп–спсп
Rate (basis points)

/ѵ тпѶспсп

фпп
упп
тпп
спп
рпп
п
–рпп
–спп
ршшп

ршшф

сппп

сппф

*0- sѷ -' . -1 *-Ѹ'0'/$*).ѵ
*/ ѷ#$)" )*/ .- ..$*)ѵ

302 |

Chapter 10

спрп

спрф

спсп

The spread between yields on long- and short-term Treasury notes is
useful as a forecasting tool. The yield spread between 10-year and 3-month
Treasury notes began 2020 at low positive levels. Negative yield spreads have
often preceded recessions, so they are generally thought of as a leading indicator of recessions. In February the spread narrowed to zero (foreshadowing a
recession), but has since rebounded to 80 basis points, a positive signal (figure
10-14).
Interest rates are also important because the Federal Reserve often
adjusts interest rates as one of its main methods to support its dual mandate
of price stability and maximum sustainable employment. However, short-term
nominal interest rates in 2020 were near zero. The Federal Reserve may not be
able to lower nominal interest rates below this “zero lower bound.” If it cannot, then interest rates near zero could take away one of its primary tools for
stimulating economic growth. See chapter 3 of this Report for a discussion of
methods used by the Federal Reserve to combat the current recession at the
zero lower bound.

Oil Markets
Worldwide oil consumption fell 8 percent during the first three quarters of
2020, following the typical pattern of energy demand falling during recessions
(EIA 2020). As a result, Brent crude oil prices fell from $66 per barrel on January
1, 2020, to a low of $19 per barrel on April 21, before recovering to $38 as of
October 30. The price of West Texas Intermediate, an important U.S. oil benchmark, actually went negative for the first time in history on April 20, driven by
fears of insufficient storage capacity (BBC 2020). In response, world production
of crude oil and liquid fuels fell by 10 percent between 2019 Q4 and 2020 Q3
(EIA 2020).

The Global Macroeconomic Situation
The global economy contracted in 2020 as a consequence of the COVID-19
pandemic. In its October 2020 World Economic Outlook, the International
Monetary Fund forecast that global output would contract at a 4.4 percent
rate (year-over-year) (IMF 2020a). Due in part to China’s rapid return to growth
and faster than expected growth among developed countries in the third
quarter, the outlook improved slightly from June when the IMF expected
global output to fall 4.9 percent (IMF 2020b). A separate forecast published by
the Organization for Economic Cooperation and Development (OECD 2020) in
December expects the global economy to contract 4.2 percent in 2020. With
this economic contraction, the World Bank (2020) anticipates an additional 88
to 115 million people worldwide will fall into extreme poverty. These forecasts
are highly uncertain and depend critically on any resurgence of the virus, intensity of social distancing policies, and the efficiency and efficacy of vaccination
programs.

The Year in Review and the Years Ahead

| 303

Unprecedented fiscal and monetary policy undertaken by governments
and central banks helped to avoid or dampen the adverse financial transmission mechanisms of the Great Recession. Fiscal measures in advanced economies were equivalent to 9 percent of output, while liquidity supports alone
were equivalent to 11 percent (IMF 2020a). Though smaller as a share of output,
fiscal and monetary support among emerging and developing economies was
sizeable as well, with fiscal measures equaling 3.5 percent of output and liquidity measures equaling 2 percent.
Despite the vast spread of COVID-19, there were notable differences in
the timing and size of economic contractions. China, which experienced the
first outbreak of the virus, experienced a 10.7 percent (nonannualized) contraction in the first quarter of 2020 before officially rebounding to its 2019 level
in the second quarter of 2020. Other countries, including the United States,
experienced their largest contractions in the second quarter as the virus spread
from China. Differences in output in the second quarter among these countries
can be explained, in part, by changes in the stringency of measures undertaken
to contain the virus and subsequent changes in mobility (OECD 2020).
To consider the cumulative loss in real GDP across countries, the CEA
calculated the percent of one year’s real GDP lost during 2020 assuming a
baseline with no growth during the first three quarters (table 10-1). This calculation represents the integral of real GDP losses during the first, second, and
third quarters, as proposed by Fernández-Villaverde and Jones (2020). Using
this approach, a country that experiences a 9 percent decline in output during
the first quarter, and then experiences no growth in subsequent quarters, has
suffered a loss three times greater than a country that experiences no decline
in real GDP during the first two quarters but a 9 percent decline during the third
quarter. This measure applies greater weight to contractions in growth at the
beginning, as this results in a longer period of lower economic activity. By this
measure, the United States has lost 3.7 percent of a year’s real GDP through
the first three quarters of 2020. China lost the lowest share of a year’s GDP (1.7
percent), while Spain lost the largest share (9.1 percent), among countries for
which data are available.
Advanced economies. Economic growth in advanced economies—such as
Germany, France, Japan, the United Kingdom, and other European Union countries—are expected to contract by 5.8 percent in 2020, reflecting an upward
revision of 2.3 percentage points from June to October. This change reflects
less severe than expected contractions in the second quarter and growth that
exceeded expectations in the third quarter. In 2021, economic growth among
advanced economies is expected to reach 3.9 percent, leaving the group 2
percent below 2019 levels. However, there is substantial heterogeneity within
the advanced economies. Asian countries that were better at containing the
spread of COVID-19 are expected to see smaller declines in growth. Japan, for
example, is forecast to contract 5.3 percent before rebounding with 7.2 percent
304 |

Chapter 10

Table 10-1. Cumulative Losses of Real Gross Domestic
Product, Selected Countries, through 2020:Q3
Country (or Group)
China
South Korea
Norway
Sweden
Japan
Israel
United States
Germany
European Union (27 members)
Mexico
Italy
France
India
United Kingdom
Iceland
Spain

Decline (percentage of annual GDP)
–1.7
–2.0
–2.3
–2.6
–3.4
–3.6
–3.7
–4.4
–5.5
–6.8
–6.9
–7.1
–8.4
–8.5
–8.5
–9.1

Source: CEA calculations.
Note: Countries are shown for which data are available.

growth in 2021. Euro area countries that were comparatively worse at containing the virus, like Spain, are expected to see growth contract by 9.8 percent in
2020 before growing at 5.9 percent in 2021, according to the IMF (2020a).
Emerging markets and developing economies. The IMF anticipates that
emerging markets and developing economies will contract 3.3 percent in
2020 before growing at 6 percent in 2021. However, this forecast is buoyed by
China, which experienced a strong rebound in the second quarter, according to
official statistics. The IMF forecasts that China will grow at 1.9 percent in 2020.
When China is excluded from emerging markets and developing countries, the
forecast contraction in 2020 is 5.7 percent while the forecast growth in 2021 is
5.5 percent.
India experienced a high volume of COVID-19 cases and undertook severe
measures to control the spread of the virus. Subsequent real GDP growth was
worse than expected in the second quarter of 2020, leading the IMF to revise
downward its forecast for 2020. India’s economy is expected to contract 10.3
percent in 2020, before returning to positive growth in 2021 at an 8.8 percent
pace (IMF 2020a).

The Future Economic Outlook
The United States is in the midst of a recovery from what has been a very severe
recession triggered by the exogenous shock of the COVID-19 pandemic. Strong
compensating growth is anticipated in 2021, buoyed by complementary fiscal
The Year in Review and the Years Ahead

| 305

and monetary policy, as well as reductions in the disease burden as vaccine
candidates become more widely available through Operation Warp Speed.
This section reviews several economic forecasts, as detailed in table 10-2, and
discusses upside and downside risks to the economic outlook.

306 | Chapter 10

2.2

2.1

2.3

2.4

2.3

2.0

1.8

2022

2023

2024

2025

2026

2027

2028

2029

2030

2.2

2.3

2.4

2.9

3.8

1.9

2.0

2.1

2.3

2.9

3.6

–2.6

2.3

1.7

1.6

1.7

1.9

2.1

1.5

–3.3

2.3

Consensus Bottom

Blue Chip*

Real GDP
(chain-type)

1.9

2.5

3.0

4.0

–3.7

2.3

4.0

3.8

3.9

4.1

4.4

4.5

4.4

4.1

6.2

4.1

4.0

4.1

4.2

4.4

4.8

5.5

–5.7 –1.4

4.0

2.0

2.1

1.9

1.8

1.3

0.2

1.6

FOMC CBO BC* CBO

2.1

2.0

1.8

1.2

1.6

BC*

2.2

2.3

2.2

2.0

1.6

0.4

2.0

CBO

2.2

2.0

1.2

2.0

BC*

4.4

4.5

4.8

5.2

5.6

6.0

6.5

7.1

8.4

10.6

3.7

CBO

4.7

4.9

5.1

5.5

6.4

6.7

8.2

3.7

Top

4.3

4.4

4.3

4.5

4.8

5.5

6.1

8.1

3.7

3.9

3.8

3.9

4.1

4.6

5.5

8.0

3.7

Consensus Bottom

Blue Chip

Unemployment
Rate (percent)

4.1

4.6

5.5

7.6

3.5

FOMC**

2.1

1.6

1.1

0.6

0.3

0.2

0.4

2.1

CBO

1.7

1.3

1.1

0.9

0.6

0.4

0.1

0.3

2.1

Interest Rate,
91-Day Treasury
Bills (percent)

Level (calendar year)

3.2

3.0

2.8

2.6

2.2

1.9

1.6

1.4

1.1

0.9

2.1

CBO

2.5

2.3

2.2

2.0

1.7

1.4

1.1

0.9

2.1

Interest Rate,
10-Year Treasury
Notes (percent)

Note: CBO = the Congressional Budget Office’s July 2020 forecast; BC = Blue Chip Economic Indicators’ December 2020 forecast, combined with its October 2020 forecast for long-term
projections; FOMC = the Federal Open Market Committee’s September 2020 forecast; Top = the average forecast of the 10 highest forecasters; Bottom = the average forecast for the 10 lowest
forecasters.*The Blue Chip Forecasts for 2022 and beyond use October’s 2020 year-over-year projections; **the FOMC’s forecasted unemployment rate for the fourth quarter of that year.
Sources: Congressional Budget Office; Blue Chip Economic Indicators; Federal Reserve Board.

2031

4.8

2021

6.0

–5.9 –2.0

2020

2.3

Top

2.3

CBO

2019
(actual)

Year

GDP Price
Nominal
Consumer
Index
GDP
Price Index
(chain-type)

Percent Change (Q4-over-Q4)

Table 10–2. Economic Forecasts, 2019–31

<BILL: PLEASE TURN THIS TABLE, AND PHOTO-REDUCE/ENLARGE AS NEEDED>

Forecasts from the Blue Chip Consensus, the Congressional
Budget Office, and the Federal Reserve Open Market
Committee
Private and official forecasts anticipate GDP to bounce back strongly from
the 2020 recession, with 4 percent or higher growth expected during the four
quarters of 2021 by the July 2020 Congressional Budget Office’s projection,
the November Blue Chip consensus, and the September FOMC projection.
The Blue Chip forecasters range from 1.7 percent, projected by the bottom 10
forecasters, to 5.3 percent, projected by the top 10, potentially indicating different assessments regarding the upside and downside risks discussed below,
and the magnitude and composition of additional support from fiscal and
monetary policy. After a strong short-term recovery, all forecasters predict that
long-term growth rates will gradually fall back to averages of about 2 percent
per year, although the Congressional Budget Office predicts a second growth
rate peak in 2025, with a gradual decline to a long run average afterward.
The Federal Reserve expects increases in the Consumer Price Index to
remain near target, though interest rates are projected to remain below their
long-term levels until the second half of the decade. In August 2020, the FOMC
announced that it would target an inflation rate that averages 2 percent,
thereby allowing periods of higher than 2 percent inflation to compensate for
periods when inflation fails to reach 2 percent. As a result, the FOMC may aim
for inflation moderately above 2 percent if prior rates of inflation have persisted
below 2 percent. This policy change means it will likely not preempt projected
inflationary pressures with interest rate hikes, as was done between 2015 and
2019. This shift will give greater space for the labor market to strengthen before
the FOMC considers raising interest rates relative to the FOMC’s previous policy
stance.
The unemployment rate is expected to continue to fall throughout the
upcoming years, before settling at a long-run rate. There are substantial differences in estimates of what this long-run rate will be, with the top 10 Blue
Chip estimates averaging 4.7 percent and the bottom 10 averaging 3.7 percent.
These estimates are above the February 2020 observed unemployment rate
of 3.5 percent, which was associated with a 12-month change in the Personal
Consumption Expenditures Price Index of just 1.8 percent, below the Federal
Reserve’s target. As discussed below, the interaction between labor force participation and labor market slack will have large effects on the unemployment
rate, wages, and inflationary pressures.

Economic Objectives and Policy
Economic prospects in the coming years depend critically on the Nation’s economic policies. The Employment Act of 1946 called for the Federal Government
to pursue the goal of maximum employment, production, and purchasing

The Year in Review and the Years Ahead

| 307

power, and it established the Council of Economic Advisers to support the
President in meeting this goal. Building on this, the Full Employment and
Balanced Growth Act of 1978 called for the President to set forth annual
numerical goals for several key economic indicators over a multiyear period,
as well as a program of policies for achieving the prescribed objectives, regardless of the probability of that program being administratively or legislatively
implemented.
The projections reported in table 10-3 reflect the Trump Administration’s
goal of achieving maximum employment, production, and purchasing power

308 | Chapter 10
уѵш
уѵш

сптп

сптр

фѵт

сѵч

сѵш

тѵп

тѵт

тѵш

сѵп

рѵт

рѵх

сѵт

рѵр

сѵп

P-$
*).0( -
I) 3
P-$ I) 3
җ#$)Ҋ/4+ Ҙ

уѵп

уѵр

уѵт

фѵс

чѵр

тѵц

сѵр

рѵц

рѵт

пѵш

пѵф

пѵс

пѵр

пѵу

сѵр

сѵч

сѵц

сѵх

сѵу

сѵс

сѵп

рѵц

рѵф

рѵс

рѵп

пѵч

пѵш

сѵр

)/ - ./R/ Ѷ
)/ - ./R/ Ѷ
) (+'*4( )/
шрҊD4
10-Y -- .0-y
R/ җ+ - )/Ҙ - .0-4B$''.
N*/ .җ+ - )/Ҙ
җ+ - )/Ҙ

1 'җ' )-4 -Ҙ

*0- .ѷ0- 0*!*)*($)'4.$.Ѹ0- 0*! *-//$./$.Ѹ +-/( )/*!/# - .0-4Ѹ!!$ *!)" ( )/)0" /Ѹ
*0)$'*!*)*($1$. -.ѵ
*/ ѷ*- ./2.. *)/1$'' .*!*1 ( -хѶспспѵ# $)/ - ./-/ *)шрҊ4Ҋ$''.$.( .0- *). *)-4Ҋ
(-& /$.*0)/.$.ѵ GDP = gross domestic product.

уѵш
уѵш

спсч

уѵш

спсц

спсш

фѵп
уѵш

спсф

спсх

фѵу
фѵр

спст

спсу

цѵу
хѵп

спср

–сѵс

–рѵп

спсп

спсс

сѵт

'
җ#$)Ҋ/4+ Ҙ

уѵп

*($)'

спрш
җ/0'Ҙ

- )/C#)" җуҊ/*ҊуҘ

Table 10-3. Policy-Inclusive Economic Projections, 2019-31

within the Federal budget window of ten years, consistent with the objectives
of the 1946 and 1978 employment acts, and accordingly include the estimated
impacts of the Administration’s full economic policy agenda. Considering
the economic challenges discussed in this Report and recent editions of the
Economic Report of the President, these are very ambitious economic projections. Achieving these targets will require full implementation of the
Administration’s complete economic policy agenda, most of which requires
Congressional legislation.
As discussed in the chapters of this Report—as well as in the 2018, 2019,
and 2020 editions of the Economic Report of the President—the ambitious
economic projections reflect the fact that the Administration’s economic
policy agenda has been similarly ambitious. In the very near-term, the agenda
includes full implementation of further legislation to support economic recovery from the COVID-19 pandemic. Specifically, the projections reported in table
10-3 assume passage of additional legislation to provide for reauthorization
and expansion of the Paycheck Protection Program to support small business
employment, an expanded employee retention tax credit, a reemployment
bonus, and a temporary extension of targeted fiscal support to State and local
governments, schools, and low- and middle-income households and households with unemployed workers.
In the near term, the economic targets reported here also assume enactment of the President’s $1.5 trillion infrastructure proposal, as analyzed in the
2018 Economic Report of the President. They also assume that all provisions of
the 2017 Tax Cuts and Jobs Act that are currently scheduled to sunset or expire
are instead made permanent. These include, but are not limited to, full expensing of new equipment investment, the near-doubling of the standard deduction, reductions of personal income tax rates across multiple brackets, doubling of and expanded eligibility for the Child Tax Credit, and a 20 percent small
business tax deduction. In addition, the numeric targets assume enactment of
a new middle-class tax cut, as discussed in chapter 11 of this Report, including
elimination of the second-earner penalty, further raising the standard deduction, and reducing income tax liability in the lowest personal income tax rate
brackets, offset at higher incomes by rate and threshold adjustments to ensure
targeted tax cuts with no net tax increases. Such a design would target tax relief
to lower- and middle-income taxpayers, who face some of the highest effective
marginal personal income tax rates, thereby substantially raising labor force
participation rates, particularly among women and low-income workers. This
is reflected in the supply-side components reported in table 10-4.
The economic targets also assume enactment of skills-based immigration
reform, a continuation of the Administration’s comprehensive deregulatory
agenda, improved bilateral trade agreements with major trading partners, and
longer-run fiscal consolidation, as discussed in the 2020 Economic Report of the
President. They further assume additional labor market policies to incentivize
The Year in Review and the Years Ahead

| 309

higher labor force participation, including expanding work requirements for
nondisabled, working-age recipients in noncash welfare programs; increasing
childcare assistance for low-income families; and enhancing assistance for
reskilling programs through the National Council for the American Worker.
By any measure, this is a very ambitious economic policy agenda.
However, it reflects the bold requirements of the 1946 and 1978 acts: to set forth
a program for achieving, as rapidly as possible, the goals of full employment,

310 |

Chapter 10
–0.3
3.0

Average weekly hours (nonfarm business)

Output per hour (productivity, nonfarm business)

Ratio of real GDO to nonfarm business output

Sum: actual real GDOa

Output per worker differential: GDO vs. nonfarm**

–0.3

2.0

–0.2

–0.3

3.5

–0.6

2.4

–0.1

0.3

0.1

1.2

–0.6

2.4

–0.2

2.5

–0.2

0.4

0.1

–0.3

1.1

2001:Q1 to
2007:Q4

–0.3

1.7

–0.2

1.4

–0.1

0.1

–0.4

1.0

2007:Q4 to
2019:Q4

Growth Rate (percentage points)
1990:Q3 to
2001:Q1

–0.4

2.7

–0.5

2.7

0.0

0.1

–0.0

–0.2

0.7

2019:Q4 to
2031:Q4

GDP and gross domestic income. Population, labor force, and household employment have been adjusted for discontinuities in the population series.

due to rounding; 1953:Q2, 1990:Q3, 2001:Q1, 2007:Q4, and 2019:Q4 are all quarterly business-cycle peaks. Gross domestic output (GDO) is the average of

Note: All contributions are in percentage points at an annual rate. The forecast is based on data available on November 9, 2020. The total may not add up

in the nonfarm business sector, and is also equal to row 4 + row 7.

Real GDO and real nonfarm business output are measured as the average of income- and product-side measures.
**The output-per-worker differential (row 9) is the difference between output-per-worker growth in the economy as a whole and output-per-worker growth

Sources: Bureau of Labor Statistics; Bureau of Economic Analysis; Department of the Treasury; Office of Management and Budget; CEA calculations.

0.0

Ratio of nonfarm business employment to household
employment

Memo:

0.0

Employed share of the labor force

0.1

Labor force participation rate

1.3

Civilian noninstitutional population age 16+

1953:Q2 to
2019:Q4

Component

Table 10-4. Supply-Side Components of Actual and Potential Real Output Growth, 1953–2031

<BILL: PLEASE TURN THIS TABLE, AND PHOTO-REDUCE/ENLARGE AS NEEDED

full production, and rising real incomes. Achieving these projected outcomes
is therefore contingent on full implementation of the complete range of economic policies articulated here and in the 2018, 2019, and 2020 editions of the
Economic Report of the President. In the absence of full implementation, not
only does the CEA anticipate that economic growth in the coming years will
be lower than the numeric targets reported in table 10-3, but also perhaps
substantially lower, in line with the economic projections summarized in table
10-2.

Near-Term Upside and Downside Risks
As discussed throughout this Report, the emergence of COVID-19 in late 2019
has burdened the economic outlook and continues to present near-term
downside risks. The burden of the virus is heaviest for elderly people and for
individuals suffering from-co-morbidities, but extends to all segments of the
population. The outlook is complicated further by the long-term health effects
of COVID-19, which are not fully understood.
The COVID-19 pandemic dominated economic developments during
2020 and will continue to do so in 2021. In the near term, the biggest downside risk to the economic outlook is that policy and behavioral responses to a
resurgence of COVID-19 disrupt the considerable recovery in output and labor
markets observed to date. For this reason, in late 2020 the Administration continued to articulate support for additional fiscal measures, discussed above,
to provide a bridge to the widespread availability of vaccine candidates developed under Operation Warp Speed (Goodspeed and Navarro 2020).
Upside risk to economic activity includes the possibility that an effective vaccine or vaccines will be rapidly distributed and administered to a
high percentage of the population, which, thanks in part to Operation Warp
Speed, looks highly probable. Indeed, multiple candidates have had successful trials, evincing high effectiveness. Preliminary results have exceeded
expectations, and an Emergency Use Authorization (EUA) from the Food and
Drug Administration has been issued for the Pfizer/BioNTech vaccine, with
an EUA for the Moderna vaccine expected by the end of the year. In addition,
treatment for COVID-19 has improved throughout 2020 and will likely continue
to improve, thanks in part to EUAs of advanced therapeutics that reduce the
illness severity and fatality rate for those affected. However, there remains the
possibility of viral mutation, especially if the virus recrosses transspecies borders, removing the resistance afforded by immune system responses built by
infection or vaccination. For example, Denmark has been forced to cull its mink
populations to avoid such a result, and other animal populations may follow.
Substantial challenges to distributing vaccines remain. Several of the
promising candidates require cold temperatures during transportation and
storage to maintain effectiveness, an issue that will create challenges in many
developing countries and some areas of the developed world. The distinct
The Year in Review and the Years Ahead | 311

threat posed by misinformation regarding the safety and effectiveness of vaccination will also need to be addressed. Creating a vaccine is only useful if a
high enough percentage of individuals use it to protect themselves and others.
Dramatic pandemic-related shocks were felt in the labor market during
early 2020. The U.S. employment level fell from a record-high of 158.8 million in
February 2020 to 133.4 million in April, a decline of over 25 million in 2 months.
For comparison, during the Great Recession employment fell by 8.6 million in
25 months. However, since April many temporary layoffs and furloughed workers have been recalled. American entrepreneurship has met the challenge,
as new high-propensity business applications (i.e., applications with a high
probability of turning into businesses with a payroll) were 29,000 by week 47
of 2020 (a 23 percent increase from 2019) according to the Business Formation
Statistics from the Census Bureau. These positive developments increased
employment by 16.3 million from April to November and reduced the official
unemployment rate from 14.7 to 6.7 percent.
In the near term, there is a risk that these trends temporarily reverse.
Rising cases have prompted the reimposition of lockdown restrictions and
endogenous individual social distancing, leading to a curtailment of expenditures on in-person consumer services. Even if case levels fall, further recovery
may still be characterized by a slower pace. Many temporary employment separations have now been restored, increasing the proportion of the unemployed
who will not be returning to their previous employment. Permanent separations require new search and matching as well as more structural adjustments,
which become more difficult the longer they are without employment. For
example, some workers may need to retrain for new industries that develop in
response to permanent changes in consumer preferences.
Hall and Kudlyak (2020) observe that the labor market’s recoveries from
recessions have consistently been measured as roughly a reduction of 0.55
percentage point a year in the unemployment rate. As the labor market regains
the temporary layoffs and begins to reallocate permanent layoffs, the rate of
recovery will likely converge toward this rate. However, looking at output,
Bordo and Haubrich (2017) find that typically the amplitude of a recovery is
strongly correlated with the amplitude of the preceding contraction, with the
recovery following the Great Recession constituting a notable exception of the
past 140 years. As the long-term unemployed experience skill deterioration and
potentially permanent income losses (Hamermesh 1989; Ruhm 1991), rapid
action to reemploy the most people possible is needed to minimize lasting
harm to the labor market. One area where layoffs may continue is State and
local governments, which have seen reduced revenue in 2020.
Another variable that has shown partial recovery is labor force participation. The 12-month moving average of participation had risen to 63.1 percent
in February 2020 from a nadir of 62.6 percent in January 2016. In recent years,
the U.S. labor force participation has faced demographic headwinds from the
312 | Chapter 10

retirement of the baby boom population, with rising participation on the eve
of the pandemic driven by a 1.9-percentage-point increase in prime-age labor
force participation. Even in February, participation remained well below (4.2
percentage points) its early 2000 peak of 67.3 percent, though it was only 1
percentage point lower for prime-age workers.
During the pandemic, participation fell 3.2 percentage points, and subsequently rose 1.5 percentage points during the recovery. Further recovery could
be imperiled if the pandemic continues to encourage individuals near the age
of retirement to retire early or encourages individuals to delay school or labor
force entry. However, there could be greater gains in participation if individuals choose to work later in life due to a desire to accumulate more savings for
retirement, if higher female participation rises during prime-age relative to
previous cohorts, and if workplaces offer greater availability of physically
accommodating occupations, including remote work. An additional upside risk
is that some pandemic-induced investment in teleworking facilitates greater
labor force participation from individuals who otherwise might face binding
childcare constraints. Similarly, the decentralization of work could cause
individuals to move to more affordable and less restrictive areas, which might
increase economic activity in areas with lower costs of living, such as rural
areas, and thereby improve family finances.
A tight labor market benefits workers by leading to higher wages and
shorter spells of unemployment. A slack labor market does the opposite,
leading to substantial downside risks for the American worker from a slower
recovery and to upside risks from a faster recovery. The lack of consensus in
the U.S. Congress to implement the President’s economic policy objectives for
additional fiscal support in response to the pandemic, particularly in the form
of an additional round of the Paycheck Protection Program to help maintain
employer-employee matches and organizational capital, is therefore a substantial downside risk in the near term.
Relatedly, in the absence of Congressional support for the Administration’s
near-term legislative priorities, there is a risk of mounting business insolvencies, particularly among small- and medium-sized firms adversely affected by
the reimposition of lockdown restrictions. As losses incurred as a result of the
pandemic and associated lockdowns are realized, there is a nontrivial risk that
defaults and insolvencies may impair collateral assets in commercial credit
markets, and therefore trigger downgrades of securities collateralized by those
assets, which would elevate the risk of broader credit disintermediation, of the
type discussed by Bernanke (1983). Tax code changes introduced by the CARES
Act were designed to mitigate this risk by modifying the treatment of business
tax assets, specifically by introducing a five-year carryback for net operating
losses (NOLs) in 2018, 2019, and 2020; suspending the NOL limit of 80 percent
of taxable income; and allowing pass-through business owners to use NOLs to
offset non-business income above the prior limit in 2018, 2019, and 2020. These
The Year in Review and the Years Ahead

| 313

modifications were designed to mitigate the adverse shock to business cash
flow in 2020 by allowing firms to spread losses over time.
Internationally, however, there is the risk that insolvency issues generated by severe and protracted lockdown restrictions abroad introduce new
strains on fiscally weak sovereign governments, most notably in emerging
markets and Europe. Such strains would elevate the risk of a replay of the
sovereign debt concerns that emerged in the aftermath of the Great Recession.

Long-Term Upside and Downside Risks
A crucial variable for the long-run outlook for growth is the productivity of the
workforce. This area has substantial upside risk for growth to outperform what
would otherwise be forecasted based on the experiences of the post–Great
Recession economy.
Productivity measures how much economic output is generated from
a given amount of inputs. As shown in table 10-4, output per hour averaged
annual growth of 2 percent between 1953:Q2 and 2019:Q4. By contrast, productivity growth averaged only 1.32 percent between 2007:Q4 and 2019:Q4.
After the passage of the 2017 Tax Cuts and Jobs Act (TCJA), productivity growth
averaged 1.52 percent annually between 2017:Q4 and 2019:Q4. Improving this
rate of increase is of paramount importance for meeting goals of raising real
incomes across the income distribution, as was observed in 2018-19. Chapter
11 of this Report discusses several possible prescriptions for achieving this
result, such as incentivizing higher education institutions to better prepare
students for the workforce, immigration reform, and infrastructure investment,
as well as making permanent some of the provisions of TCJA that are currently
legislated to phase out.
As mentioned above, demographic shifts continue to constitute a challenge to the supply-side potential of the U.S. labor market. The 65-and-older
population grew by over a third (34.2 percent, or 13,787,044) during the past
decade. The first cohort of the baby boom generation turned 65 in 2011, and
the last cohort will turn 65 in 2029. These demographic shifts have generated
downward pressure on the aggregate participation rate over the past decade,
and will continue to generate downward pressure over the next decade.
Whether these individuals retire early or continue to actively participate in the
labor market will have a major impact on the economic trajectory over the
next decade. Policies such as those described in chapter 11 of this Report can
have a positive effect on participation, alleviating the demographic drag. As
discussed in the 2019 and 2020 editions of the Economic Report of the President,
making permanent the marginal personal income tax rate reductions in the
TCJA can also incentivize continued participation among retirement and nearretirement age workers, who theoretical and empirical research indicate are
more responsive to changes in marginal personal income tax rates.

314 | Chapter 10

The COVID-19 pandemic has highlighted shortcomings in the healthcare
system in the United States that lead to both upside and downside risks for
long-term growth. As the pandemic has dramatically illustrated, individuals
with poor health are much more likely to suffer serious illness in the event of
contracting the disease, and consequently face a higher mortality rate. Life
expectancy among particular segments of the population in the United States
was falling even before the pandemic, a trend that is the result in large part of
drug overdoses (particularly from opioids), suicides, and liver diseases, which
recent research suggests may be related to displacement effects of increased
exposure to import competition from China following the establishment of permanent normal trade relations in 2000 (Pierce and Schott 2020). Addressing the
underlying reasons for these disturbing trends provides an upside risk that the
loss of social cohesion, mental health deterioration, and poor diet and exercise
can be successfully reversed. However, there is also a downside risk that the
situation gets worse, with losses in productivity and participation resulting
from more sickness and death.
There is also an upside potential that pandemic-induced investments in
healthcare research and the rapid deployment of new therapies and vaccines,
as well as deregulatory actions to increase choice and access in the medical
system, can lead to better health in the future. Examples of deregulatory
actions include expanded access to telehealth services, increased scope of
practice, and further mRNA-based interventions. Observations by individuals
may also result in greater precautionary personal measures in future influenza
seasons, reducing the annual burden of endemic disease.
An additional long-run risk is that the pandemic and associated lockdowns generate long-term economic scarring and amplify issues of economic
inequality. Whereas—in a stark reversal of trends under way during the 2009–16
expansion—wage, income, and wealth inequality, including among races, were
declining in the three years immediately preceding the pandemic, the extreme
regressivity of lockdown restrictions and consequent loss of employment and
disruption to human capital acquisition may exacerbate economic inequality
for years to come. Although the CARES Act attenuated income inequality in
the near term, over the long run, school closures, disparate access to remote
learning, and losses of on-the-job training and skills acquisition may introduce human capital deficits that compound over time and have a particularly
adverse impact on the lower end of the skills and income distributions. Longterm scarring that depresses future supply-side potential could also complicate the task of monetary policy if inflation expectations become unanchored
(Kozlowski, Veldkamp, and Venkateswaran 2020).
Finally, to ensure robust long-term growth, the United States must also
address its rising debt burden. Structural and taxation incentives for debtfinancing investments have created a large increase in nonfinancial corporate
debt, growing from $6.1 trillion at the end of 2010 to $11 trillion in 2020. The
The Year in Review and the Years Ahead

| 315

TCJA codified several improvements in capital allocation mechanisms, including limiting the tax deductibility of interest payments on debt, precipitating
upside risk that this will lead to improved productivity growth. Lower interest
rates due to the pandemic are unlikely to substantially increase during much of
the budget window, which may continue to incentivize debt-financing of business activity. The debt burden on individuals is also a cause of concern, especially loans used to finance higher education. Since 2003, inflation-adjusted
student debt balances have more than doubled in nearly every State, and in
parts of the Southeast they have nearly quadrupled, with the student loan
delinquency rate rising commensurately (Hedlund 2019). This debt burden has
been found to constrain occupational choice, reduce marriage prospects and
homeownership, and increase the risk of bankruptcy (Rothstein and Rouse
2011; Gicheva 2016; Mezza et al. 2020; Gicheva and Thompson 2015). As students who have put off entry into higher education decide whether to return,
the question of whether higher education will act as an expensive signaling
device or as a skill-formulating institution has substantial upside and downside
risks.

Conclusion
The events of the past 12 months have created a historically unprecedented
year for the U.S. economy. Record declines in GDP and employment in the
second quarter were followed by record increases in both of these economic
indicators in the third quarter. Inflation, housing markets, and financial and
energy markets were also affected, although to a lesser extent than output and
labor markets. Strong compensatory growth is anticipated in 2021. However,
GDP forecasts and the slowing pace of the recovery of labor force participation
show that many of these issues will persist through at least 2021.
Despite the historic pace of the economic recovery observed to date,
there remain risks to both the near- and long-term outlooks. In the near term,
particularly in the face of viral resurgence, the Administration recognizes the
need for further fiscal support to maintain attachments between employees
and employers until the widespread availability of vaccines through Operation
Warp Speed allows the resumption of normal levels of economic activity. Over
the longer term, building on the historic economic gains observed in 2017,
2018, and especially 2019, a program of economic policies that continues to
incentivize domestic capital formation and increased labor force participation will be essential for ensuring a rapid return to the economic conditions
prevailing on the eve of the pandemic. As discussed throughout this Report,
this program includes but is not limited to extending the provisions of the
TCJA, investing in infrastructure, lowering high effective marginal income tax
rates on lower-income workers, further regulatory reform, and continuing to
upgrade bilateral trading arrangements. Such a program was instrumental in

316 |

Chapter 10

generating a historically tight labor market and broad-based real income gains
in 2017-19, following what had been the weakest economic recovery in postwar
U.S. history between 2009 and 2016.

The Year in Review and the Years Ahead

| 317

x
Chapter 11

Policies to Secure
Enduring Prosperity
This Report analyzes the unprecedented health and economic shock of the
COVID-19 pandemic, and the historic policy responses to mitigate its impact
on the Nation. The United States is making progress toward emerging from
this crisis; however, our country continues to contend with an adverse shock of
historic magnitude. The purpose of this final chapter is to review a collection
of policy areas highlighted by the COVID-19 pandemic and to analyze potential
reforms to meet the ongoing challenges facing the U.S. economy. We introduce
these areas in this prefatory section, and then the full chapter presents them
in detail.
Strengthening connections to the labor force. The U.S. labor market’s recovery since the initial effect of COVID-19 has been unprecedented, with the
unemployment rate falling by 8 percentage points in seven months. However,
workers with a weaker prior connection to the labor force have experienced a
slower recovery. This chapter discusses two important ways in which the tax
code discourages lasting connections to the workforce: high effective taxes
both on nonprimary earners in families and on low-income earners navigating
the various Federal assistance programs.
Supporting balance between work and family. The recent suppression of
economic activity has posed particular challenges for families. Parents of
children whose schools were closed faced challenges in obtaining childcare
while working. Parents who needed time off due to illness or to care for a sick
relative faced difficult decisions regarding work and family responsibilities. This
chapter discusses how, even in normal times, a lack of accessible paid leave

319

and childcare for parents can lead to wider detrimental effects on society, and
how this challenge could be addressed.
Enhancing international coordination to address 21st-century challenges. Both
the health and economic consequences of COVID-19 cross national boundaries,
given that disease transmission and supply chain disruptions in one country
can have large effects on other countries. This chapter discusses how strong
reciprocal trade relationships between the United States and other countries
can preserve U.S. consumer access to foreign products and U.S. producer
access to global supply chains, while ensuring that American entrepreneurs
face an even playing field that protects U.S. economic interests.
Creating a more effective healthcare system. COVID-19 caused a public health
crisis that exposed strains on the U.S. healthcare system. This chapter reviews
mechanisms that inefficiently drive up costs and reduce access to quality
health care. These include restrictions on the supply of healthcare professionals, information problems inherent in balance billing, and a disconnect
between Medicare prices and competitive prices for some medical services.
Building a dynamic economy through infrastructure improvement. Continued
adjustment to the potential reallocation of economic activity and factors
of production in response to the pandemic requires strong and versatile
infrastructure. The Federal Government can target investment to increase the
productivity of American industry. This chapter discusses the structural factors
that inhibit improvements in infrastructure along with mechanisms to resolve
them.
Generating a more skilled and resilient workforce. COVID-19 is imposing a large
reallocation shock on the U.S. economy because temporarily suppressed
output and changes in consumer preferences may weaken some firms and
industries and strengthen others. Highly skilled workers will be needed not
only to take advantage of these new opportunities but also to create them. This
chapter discusses two ways to expand the skilled workforce: moving toward a
more transparent and merit-based immigration system, and improving human
320 |

Chapter 11

capital formation for Americans attending institutions of higher education.
This chapter also highlights the success of Historically Black Colleges and
Universities.

he American economy faces challenges that not only were exacerbated
by the COVID-19 pandemic but also extend into the postpandemic
future, as outlined above and as explained in detail below. Meeting
these challenges will ensure that the United States not only recovers to its
prepandemic levels of prosperity but also builds a more dynamic and resilient
economy that will benefit all Americans.

Strengthening Connections to the Labor Force
The COVID-19 pandemic and subsequent economic shock decreased prime-age
labor force participation by 3.2 percentage points, erasing the unanticipated
gains of the preceding three years and reaching its lowest value in April 2020
since the early 1980s, before partially recovering. Increasing the labor force
participation rate will require action addressing the elements of the Federal
tax code that disproportionately deter labor force entry and skill upgrading.
This section identifies two areas in which Federal policy changes can remove
barriers to participation in the workforce. Taken together, these tax reforms
would constitute a momentous middle class tax cut.
From the early 1960s until the turn of the century, the United States
experienced a sustained and pronounced rise in the employment-to-population ratio—which measures the percentage of the civilian, noninstitutional
population that is working—from 55 percent to nearly 65 percent. This trend
was driven by the participation of females, many of whom were in two-earner
households. However, over the past 20 years, this trend has been eroded
by two recessions—in 2001 and 2008-09—and subsequent slow recoveries,
coupled with an aging of the population. Even among the prime-age labor force
of 25- to 54-year-olds, the employment-to-population ratio fell from a peak of
over 80 percent in 2000 to only 75 percent in the immediate aftermath of the
Great Recession. Only by 2019 did the U.S. economy nearly return to its 2000s
peak under the historically strong labor market conditions that existed before
the arrival of the COVID-19 shock. Figure 11-1 summarizes these dynamics.
Abraham and Kearney (2020) discuss several factors behind the stagnation and decline in the employment-to-population ratio between 1999 and
2018. These include the effects of import competition from China, automation,
disability insurance programs, childcare costs, and shifting social norms reducing the stigma of not working (especially among men) on labor supply.

Policies to Secure Enduring Prosperity | 321

$"0- ррҊрѵ(+'*4( )/Ҋ/*Ҋ*+0'/$*)/$*Ѷршхп–спсп
Ratio (percent)

спршѷт

чф

-$( "
җсф–фу yearsҘ

чп
цф
цп
хф
хп

''" "-*0+.

фф
фп
ршхп

ршцп

ршчп

ршшп

сппп

спрп

спсп

*0- .ѷ0- 0 *! *-//$./$.Ѹ (2020)ѵ

In addition to these factors, the Federal Government’s income tax code
is an important impediment to the growth of the labor force due to both the
way that second-earners are penalized by the system of joint taxation and the
high combined effective marginal rates of Federal and State taxes faced by
low-income earners. While the 2017 Tax Cuts and Jobs Act reduced hindrances
to investment and brought more low-income earners out of Federal income tax
liability altogether by nearly doubling the Standard Deduction, many filers in
these two groups continue to face high effective marginal tax rates under the
current code. This section discusses the negative effects of family taxation and
the causes of the high effective marginal tax rates faced by many low-income
earners (estimated to be as high as 70 percent by Altig et al. 2020, when taking into account Federal and State taxes along with phase-outs of credits and
deductions). This section also provides a broad outline of possible tax reforms
that could spur economic growth by stimulating labor market participation
by second-earners and low-income earners, two groups with relatively high
responsiveness to labor market incentives.

Dual-Earner Couples
Among married couples, the prevalence of dual-earners increased steadily
during the postwar years (figure 11-2). This trend demonstrates the growing

322 |

Chapter 11

$"0- ррҊсѵ$./-$0/$*) *! *$)/$' -.Ѷршхс–спрш
Share (percent)

спрш

чп
цп

2*Ҋ -) -
*0+' .

хп
фп
уп

$)"' Ҋ -) -
*0+' .

тп
сп
рп
п
ршхс

ршцс

ршчс

ршшс

сппс

спрс

*0- ѷ0-- )/*+0'/$*)0-1 4ѵ

importance of two-earner couples, but also how their growth has stagnated
since the 1990s. The female labor force participation rate in countries that
do not differentiate between single-earner and dual-earners families, such
as Sweden, also stagnated during this period, but remained at a higher level.
Guner, Kaygusuz, and Ventura (2012a) find that the participation rate of married women in Sweden is nearly 15 percentage points higher than in the United
States. Even though Sweden’s overall tax burden on labor earnings is considerably higher, its system of separate taxation, which taxes individuals based on
their own earnings instead of penalizing them for the earnings of their spouse,
leads to noticeably lower marginal tax rates on second-earners—the individual
in a dual-earner couple that has lower earnings. In some cases Sweden has a
rate that is nearly 10 percent lower than in the United States, according to Bick
and Fuchs-Schundeln (2017).
Before 1948, the United States levied income taxes at the individual level,
although couples living in States with community property laws were taxed
as if each spouse earned half of household income. As the tax system became
much more progressive, concerns began to arise that wealthy husbands could
engage in income-shifting to avoid heavy taxation in upper brackets. By transferring some of their assets to their wives (who generally had lower incomes),
their transferred asset income being might be taxed in a lower tax bracket. The
shift to joint taxation meant that couples added their income together when

Policies to Secure Enduring Prosperity | 323

filing taxes. This switch greatly increased marginal tax rates on second-earners
because the first dollar of the second-earner is effectively taxed at the marginal
rate of the last dollar earned by the primary-earner. A 2008 study suggests that
this switch depressed married female labor force participation by 2 percentage
points in the postwar period among women most likely to be affected by the
law, despite going into effect before the widespread acceptance of women in
the workplace (LaLumia 2008). If the move from individual to joint taxation had
occurred after the shift in norms, the effect would likely have been considerably larger.

The Marriage Penalty and the Second-Earner Penalty
The second-earner penalty is distinct from the marriage penalty that is more
often discussed in that the former deals with marginal taxation and the distribution of work incentives within couples, whereas the latter is related to
changes in the total tax burden a couple faces before and after they get married. For example, in 2016, before the 2017 TCJA, the increase in the tax rate
from 25 to 28 percent occurred at $91,150 for single persons but at $151,900 for
couples. Thus, if two individuals in a relationship each had $90,000 in taxable
income, they would each fall into the 25 percent tax bracket before marriage
but would be pushed well into the 28 percent bracket after marriage because
of their combined $180,000 in taxable income. As a result, they would face a
higher total tax bill as a married couple than they faced as individuals in a relationship before marriage (because each new tax bracket for married couples
began at an income level at less than twice the income level for single persons).
In other cases, a couple may have a marriage bonus if their tax burden under
joint filing is lower than their combined tax burden when they filed two separate returns as unmarried individuals. The Office of Tax Analysis at the Treasury
Department estimated that before the TCJA, roughly 40 percent of nonelderly
married tax filers faced a marriage penalty while 51 percent enjoyed a marriage
bonus. The TCJA greatly reduced this tax penalty for the vast majority of married couples by ensuring that the size of the standard deduction and the location of tax bracket thresholds for joint filers were double those for single filers.
In contrast, the second-earner penalty refers to the fact that, under a
progressive tax code, joint filing imposes higher tax rates on second-earners
than if they were filing taxes as a single person. Figures 11-3 and 11-4 plot
the average combined income and payroll tax rate faced by single filers and
second-earners without and with children based on current law. Single filers face an average tax rate—defined as total tax obligation divided by total
income—that starts near 10 percent. If, however, that person gets married to
someone earning $40,000, his or her average second-earner tax rate—defined
as the added tax the joint household faces from the second-earner’s decision
to work divided by the amount of those second-earnings—starts at over 25
percent. If the single filer were to marry someone earning $120,000, he or she

324 |

Chapter 11

$"0- ррҊтѵAverage Tax Rate on the Pretax Wages of Single
Persons and Second-Earners (without Children), 2020
Tax rate (percent)
уп
тф
тп
сф
сп
рф
$)"'
--$
--$
--$

рп
ф

persons
/*‫ڦ‬уп e-) /*‫ڦ‬чп e-) /*‫ڦ‬рсп e-) -

п
рп

сп

тп

уп

фп

хп

цп

чп

шп

рпп

Pretax wages (thousands of dollars)
*0- .ѷ/$*)'0- 0*!*)*($ . -# Ѹ'0'/$*).ѵ

would be subjected to average second-earner tax rates starting at nearly 40
percent, with State income taxes further pushing up this rate. This does not
take into account governmental aid programs, which impose a high effective
tax rate on individuals who are in the phase-out range for governmental programs, as discussed below.
Figure 11-4 reveals that the second-earner penalty is even starker for
people with children because of means-tested provisions in the tax code, such
as the Earned Income Tax Credit (EITC), a refundable tax credit that subsidizes
the wages of low-income households, especially those with children. The credit
gradually rises with income in the phase-in region before eventually leveling
off and then phasing out as household income continues to grow. The design
of the EITC therefore incentivizes labor force participation. The net result for
single filers with less than $35,000 is a negative total tax obligation, with average rates for some below –30 percent (i.e., a subsidy rate of over 30 percent,
not including other governmental assistance programs). However, if the single
filer gets married to a person earning $40,000, their joint income causes the
EITC to shrink in addition to pushing the second-earner into a higher tax
bracket—resulting in an average tax rate of about 35 percent, which represents
an increase of nearly 70 percentage points for low-income second-earners.

Policies to Secure Enduring Prosperity | 325

Figure 11-4. Average Tax Rate on the Pretax Wages of Single
Persons and Second-Earners (with Children), 2020
Tax rate (percent)
40
30
20
10
0
–10
Single persons
Married to $40K earner
Married to $80K earner
Married to $120K earner

–20
–30
–40
10

Pretax wages (thousands of dollars)
Sources: National Bureau of Economic Research TAXSIM; CEA calculations.

The joint nature of the income tax code introduces a bias toward singleearner families, encouraging them to specialize, with one spouse at work in
the market and the other engaging in tax-free home production. This may not
be an optimal allocation for the individual family or the overall labor force
absent such a skewed taxation system. For example, both individuals may
wish to work outside the home, but the tax penalty for doing so discourages
them. The existence of a second-earner penalty is intrinsic to any progressive
income tax code with joint filing, though the magnitude of the penalty can
vary—the steeper the rate structure, the larger the second-earner penalty. For
this reason, past tax reforms in the United States that lowered marginal rates
also mitigated—but did not eliminate—the second-earner penalty.
For example, the 1981 and 1986 tax reforms, which brought the top marginal tax rate down from 70 percent to 31 percent and eliminated loopholes to
broaden the taxable base, were responsible for at least one-fifth to one-quarter
of the 13-percentage-point rise in labor force participation by married females
during the 1980s, according to research by Kaygusuz (2010). This estimate is
based only on the direct effects of the tax code change, but after taking into
account the contribution of the tax cuts to higher wages, the effect may very
well have been much larger. This same research attributes 62 to 64 percent
of the rise in participation to rising female wages during the 1980s. Bronson
and Mazzocco (2018) also conclude that the primary effect of the Reagan and

326 |

Chapter 11

George W. Bush Administrations’ tax cuts was to increase married female participation. Malkov (2020) finds that—along with the 1986, 2001, and 2003 tax
reforms—the 2017 TCJA created welfare gains for married couples and reduced
the second-earner penalty because of the overall lowering of the marginal tax
rate schedule.

Tax Reforms to Mitigate the Second-Earner Penalty and Boost
the Labor Supply
There are two ways to eliminate the second-earner penalty: reduce progressivity—which the Administration does not recommend—or move toward
individual taxation. Guner, Kaygusuz, and Ventura (2012a) find large gains
in economic output, welfare, and female labor supply (because women are
more likely to be second earners) from moving to a proportional income tax.
However, Bick and Fuchs-Schundeln (2017) point out that taxing two-earner
labor income jointly acts as a greater impediment to female labor supply than
does the progressivity of the tax code. They also find that moving completely
to a system of individual instead of joint taxation—that is, replacing the current
single, head of household, and joint filing statuses with one individual status
that features a revamped system of deductions and tax brackets—would boost
female labor supply by 7.8 percent. Guner, Kaygusuz, and Ventura (2012b) find
a 10.4 percent increase in the supply of married women and 18.1 percent rise
for married women with children in response to a shift from joint to individual
taxation that eliminates the second-earner penalty. Similarly, Borella, De
Nardi, and Yang (2019a, 2019b) estimate that shifting away from joint taxation
completely would raise the labor force participation rate of married women
by 20 percentage points for women under the age of 35 and by 10 percentage
points for women between the ages of 45 and 60. These numbers are high, but
research by Crossley and Jeon (2007) indicates that when Canada reformed its
tax code in 1988 in a way that reduced the marginal tax rate for certain married women, that group’s participation rate increased by nearly 10 percentage
points.
Such a complete shift toward individual taxation would mark a dramatic
reform for the United States. Moreover, Fruttero and others (2020) point out
that eliminating the current joint tax rate schedule entirely could have a negative effect on single-earner households. This finding assumes that the tax rate
schedule for the new, unified individual filing status that replaces it would have
income brackets between those of the current single and joint brackets (if the
new schedule instead adopted the current joint brackets, the static drop in
income tax revenues would be larger).
As an alternative to universal individual taxation, the Federal Government
could allow second-earners to directly protect their earned income through
segmentation, whereby married couples filing jointly have the option of
applying the joint rate schedule to the primary earners’ income and the rate

Policies to Secure Enduring Prosperity | 327

schedule for single persons to the earned income of the secondary (lower)
earner. Other proposals include allowing a second-earner deduction or credit.
Under segmentation, all deductions, credits, and dependents enter into the
joint tax calculation based on the primary-earners’ income (and any income
not derived from wages, salary, or self-employment of the second-earner). The
Federal Government could also use means-testing for the EITC to exclude the
earnings of the second-earner, reducing the implicit tax in the phase-out region
of the EITC for dual-earning couples, because each $1 in higher second-earner
wages and salary income has no effect on the EITC amount received by the
household.
Under this reform, the earnings of the secondary earner would be taxed
as if earned by a single person having no children with only the standard
deduction applicable. As a result, this tax reform option would allow families to
protect the second-earner from tax penalties associated with the income of the
primary earner. In other words, second-earners would owe the same amount of
tax based on their paycheck income regardless of the earnings of their spouse
or other sources of family income, thereby directly eliminating much of the
second-earner penalty currently embedded in the tax code. Correcting this
disincentive would create a situation wherein second-earners would be able to
participate in the labor market on a basis similar to single persons.

High Marginal Rates for Low Earners
Perversely, some of the highest effective marginal tax rates on labor income
fall upon low-income earners, individuals making at or slightly above the
poverty line. Altig and others (2020) calculate that one in four low-wage workers face lifetime marginal tax rates above 70 percent, taking into account the
combination of Federal, State, and local taxation and benefits programs. Over
half of low-wage workers face lifetime marginal rates over 45 percent. Chien
and Macartney (2019) find that among households just above the poverty line
and that have children, the median marginal tax rate is 51 percent, as shown
in figure 11-5. Some households face a marginal tax rate above 100 percent.
As a result, a low-wage household that increases its earnings by $1 will lose
more than $1 to combined explicit and implicit taxation. This mechanism locks
households in a cycle of poverty and impedes their ability to climb into the
middle class.
This situation is the result of a combination of the structure of benefit
programs and Federal and State income taxes. U.S. Federal individual income
taxes are progressive and the first 2 bracket rates (10 and 12 percent) are relatively low. States collect most of their revenue from sales and property taxes;
however, 41 States also tax individuals’ labor income, accounting for 24 percent of State and local government tax revenue. The lowest-bracket State-level
income tax is as high as 5 percent in Illinois, Kentucky, Massachusetts, Oregon,
and Utah; 5.3 percent in North Carolina; and 5.4 percent in Minnesota—thus

328 |

Chapter 11

Figure 11-5. The Marginal Effective Rate on Low Earners
Median marginal tax rate (percent)
60
50
40
30
20
10
0
–10
–20
–30
0–24

50–74

100–124
150–74
200–224
Poverty status (percent)

250–74

Sources: Chien and Macartney (2019); U.S. Department of Health and Human Services; CEA
calculations.
Note: This figure shows the marginal tax rates on households below the poverty line after a
$2,000 earnings increase. Because households with children are recipients of more
government aid programs and, consequently, see a greater reduction in benefits, they pay
a higher effective marginal tax than households without children. The most common
combination of aid programs is SNAP, EITC, Child Tax Credits, and Medicaid / Children's
Health Insurance Program. For a household of two, the dollar value of 100 percent poverty
is $17,200 and of 200 percent poverty is $34,400.

increasing the tax burden on low-wage labor income. In addition, complex benefit programs often include phase-outs that can jointly create extreme losses
in benefits as a result of gains in income. This reduction in benefits functions
similarly to a tax on income. There are also programs in which earning above a
certain threshold can result in a sudden large loss in benefits with no gradual
phase-out.
Programs such as the EITC, Medicaid, Temporary Assistance to Needy
Families (TANF), Supplemental Nutrition Assistance Program, Child Care
Assistance, Section 8 Housing Vouchers, Energy Assistance, and Children’s
Health Insurance Program can provide valuable assistance but at the cost of
a large administrative burden to the government and a complex web of procedures that families must navigate to receive aid. In combination, they also
impose high costs on the acquisition of earnings-enhancing human capital,
effectively punishing families for augmenting their human capital by rapidly
withdrawing government assistance.
Altig and others (2020) illustrate the benefit cliff faced by a hypothetical
mother with two children. She loses access to benefits as her income rises, with
notable drop-offs in total benefits after $44,000 in annual earnings. In terms
of net resources, she is nearly as well off financially earning $53,000 a year as
when she is earning only $11,000 a year. This constitutes a severe impediment
to the acquisition of new human capital through labor market advancement.

Policies to Secure Enduring Prosperity | 329

Box 11-1. Limiting Tax Expenditures to Facilitate
Pro-Growth Reform: the SALT+MID Deduction
The 2017 TCJA combined lower taxation on investment, individual rate
reductions, an increase in the Child Tax Credit, and a dramatic expansion in
the standard deduction with the imposition of tighter caps on the State and
local tax and mortgage interest deductions. Specifically, the TCJA increased
the standard deduction from $6,500 to $12,000 for single filers and from
$13,000 to $24,000 for joint filers while capping the State and local tax (SALT)
deduction at $10,000 and reducing the maximum mortgage principal eligible
for the mortgage interest deduction (MID) from $1 million to $750,000. These
reforms weakened the MID and SALT deduction by both reducing the incentive to claim them relative to the larger standard deduction, and by reducing
the maximum MID and SALT deductions that can be claimed.
One reason for limiting these tax expenditures is that they are skewed
to high-income households, as shown in figure 11-i. In addition, they each
create economic distortions. Specifically, the SALT deduction makes it easier
for State and local governments to increase their revenue at the expense of
taxpayers in other jurisdictions by diverting taxes that would otherwise be
paid to the Federal Government into local receipts. This forces taxpayers in
other locales to shoulder a greater share of the burden. Because local taxes
are capitalized in local home prices, particularly in supply-inelastic markets,
the partial defraying of tax increases causes the SALT deduction to artificially
inflate housing prices in high-tax areas. The MID also fuels price increases

$"0- ррҊ$.- Ҋ $./-$0/$*)*! s4%0./ -*.. )*(

330 | Chapter 11

while encouraging homeowners to finance their home purchases with debt
instead of equity.
During the crafting and passage of the TCJA, some outside groups
(e.g., National Association of Realtors n.d.) expressed concerns that the
changes outlined above would diminish the tax benefits of homeownership
by inducing people to switch from itemization to claiming the standard
deduction. Indeed, the share of individual returns that claimed itemized
deductions fell from 31 percent in 2017 to only 11.4 percent in 2018. Notably,
the individuals who switched to claiming the standard deduction generally
benefited, because they chose this option over claiming the still-existent MID.
However, the housing market has proven incredibly strong and resilient in the
years since passage of the TCJA. Homeownership has increased since 2017
after nearly a decade of consistent declines during and in the aftermath of the
2007–9 Great Recession.
As predicted, home price growth did weaken in some areas due to the
TCJA reforms. Li and Yu (2020) find that the $10,000 SALT cap caused the
growth rate of home values to decline by 0.8 percentage point per year in
high-tax areas, with the effects felt most strongly within the medium range of
properties on the market. Rappoport (2019) measures the response of house
prices to all the deduction provisions mentioned above and estimates a 3
percent average reduction across 269 metropolitan areas. Martin (2018), in
turn, finds an even larger average decline, of 5.7 percent, but with variation
across zip codes and income classes. Each of these research papers comports
with the assertion above that the SALT deduction and MID prop up home
values, and thus their removal should create the opposite effect and make
homeownership more affordable for Americans.
Although slowed home price growth reduces equity increases for
incumbent homeowners in high-tax areas, first-time buyers gain easier admission into homeownership by facing more affordable housing choices and
being able to make smaller down payments. In fact, 2017 marked the beginning of the turnaround in the U.S. homeownership rate, which had been on a
stubborn downward path, from 68.2 percent in 2007 to 63.4 percent in 2016.
By 2020:Q1, the homeownership rate had recovered to 67.4 percent. In fact,
research by Hilber and Turner (2014) find that the MID has had no discernible
effect on the overall level of U.S. homeownership. Sommer and Sullivan
(2018) go even further, demonstrating that limiting the MID actually improves
homeownership by making housing more affordable, which is particularly
relevant to young prospective buyers who lack the accrued savings to make
large down payments. Consistent with this finding, the data reveal that households under the age of 35 have experienced the largest homeownership gains.
Looking across states, the CEA finds that the period after the TCJA’s
enactment evinced relative homeownership gains—not declines—in states
with high mortgage income plus State and local tax (MID + SALT) deduction
intensity compared with States with lower MID + SALT deduction intensity.

Policies to Secure Enduring Prosperity | 331

Here, intensity is defined as the ratio of MID + SALT deductions to adjusted
gross income in 2016. States with an above-median ratio (which equaled 7
percent in 2016) are considered to have high MID + SALT intensity, and those
below the median are categorized as having low MID + SALT intensity.
Using State-level homeownership data from the Census Bureau covering the 2014:Q1 through 2020:Q2 period, the CEA employs regression analysis
to measure changes in comparative homeownership dynamics between
these two groups of States in the years after the TCJA compared with the
years before. This analysis controls for permanent State differences as well as
seasonality. The CEA finds that homeownership rates in States with high MID +
SALT deduction intensity increased by an average of 0.9 percentage point per
quarter in the period after the TCJA’s enactment relative to States with lower
MID + SALT deduction intensity, with the difference growing over time, as
shown in figures 11-ii and 11-iii. States with high and lower MID + SALT deduction intensity had statistically indistinguishable homeownership rates in 2018
and the first half of 2019. Elevated homeownership in States with high MID +
SALT deduction intensity began in 2019:Q3 (1 percentage point higher than in
lower-tax States)—one and a half years after the TCJA’s enactment—and was
over 3.5 percentage points higher in 2020:Q2. The average 0.9-percentagepoint increase in States with high MID + SALT deduction intensity translates
into a 1.4 percent gain per quarter relative to the average homeownership
rate of 66.2 percent across all States during the analysis period.

$"0- ррҊ$$.- Ҋ $./-$0/$*)*! 4%0./ -*.. )*(

332 |

Chapter 11

$"0- ррҊ$$$ѵ *( *wner.#$+/ .$) $"#-Tax - .*(+-
with *2-3- .Ѷспрч–сп
Percentage points

2020:Q2

уѵп
тѵф
тѵп
сѵф
сѵп
рѵф
рѵп
пѵф
пѵп
–пѵф
–рѵп
спрчѷр

спрчѷт

спршѷр

спршѷт

спспѷр

*0- .ѷ ).0.0- 0Ѹ )/ -)' 1 )0 -1$ Ѹ'0'/$*).ѵ
*/ .ѷ// $.*).$ - /*#1 #$"#+- Ҋ ‫ ۔‬0/$*)$)/ ).$/4$!Ѷ$)
спрхѶ/# -/$* ‫ ۔‬0/$*)/*%0./ "-*..$)*( җ Ҙ2."- / -/#)ц
+ - )/Ѷ2#$#2./# ( $)-/$**!/*/' ‫ ۔‬0/$*)(*0)/./*/*/'
(*0)/.-*..''// .$)спрхѵ# # $"#/*! #-- +- . )/./# - "- ..$*)
./$(/ *!/# 1 -" #*( *2) -.#$+-/ $)#$"#+- Ҋ Dҝ 0/$*)
$)/ ).$/4// .- '/$1 /*'*2 -+- Ҋ Dҝ 0/$*)$)/ ).$/4// .!*-"$1 )
,0-/ -Ҋ4 -ѵ)./ -$.&$)$/ ./#//# - "- ..$*) ./$(/ $..//$./$''4.$")$!$)/ѵ
/)- --*-.*0)/!*-/# +*..$' *-- '/$*)(*)"*. -1/$*).2$/#$)/# .(
// *1 -/$( ѵ# - "- ..$*)$)'0 .*)/-*'.!*-// 0((41-$' .)(*)/#
0((41-$' .ѵ

Reducing the benefit cliffs faced by families receiving assistance can be
accomplished by reforming these programs to ensure that there is a low or no
penalty for improving wage income. Removing provisions that create nonconvexity and nonlinearity in benefit schedules and consolidating the patchwork
of benefits into a more user-friendly system would be a substantial improvement. Allowing a grace period during which an individual maintains benefits
after commencing a new job or receiving a raise can smooth the transition to
a higher income level. When phase-outs do happen, starting them sooner and
having them progress more slowly will reduce the disincentives they create.
Progress toward the goal of skill accrual and independence can be made
by reducing the Federal labor tax rate on the lowest income tax brackets.
Although it may appear at first glance that the burden of taxation is light on
the lowest earners in the U.S. economy, the structure of benefit programs and
the income tax system impose a high tax rate on low-earners’ wage income.
Removing impediments to increasing productivity and earning higher wages
is of critical importance for the long-term recovery of the U.S. economy. In
the spring of 2020, labor force participation dropped 3.2 percentage points,
and has to date only partially recovered, by 1.3 percentage points. Increasing
participation among marginalized groups can assist in reversing this trend.

Policies to Secure Enduring Prosperity | 333

The middle class tax cut discussed above would remove impediments to
higher labor force participation and economic growth. However, it would likely
reduce Federal tax revenues even when dynamic growth effects are taken into
account. In the past, the U.S. has successfully increased fiscal capacity for progrowth tax reform by coupling rate reductions and other broad-based relief
provisions with the elimination or limitation of tax benefits. Such benefits act
effectively as a form of spending, even if they are disguised as a broad reduction in tax liabilities. Proposals for reducing these tax expenditures have been
subject to claims about pernicious results in the past. However, as detailed in
box 11-1, this prediction did not come to pass after the 2017 limitation of the
State and local tax (SALT) and mortgage interest deductions (MIDs).

Supporting a Balance between Work and Family
The COVID-19 crisis has had divergent effects on families with children. In April
and May 2020, when most schools did not provide in-person learning opportunities, employed workers with children under 13 years of age were more likely
than employed workers without children to work fewer hours (figure 11-6). The
crisis has also illuminated an underlying issue with the lack of high-quality,
affordable childcare and paid family leave. This absence not only hurts the
labor market prospects of the parents or family members in question, but also
affects the entire U.S. economy.
Family demographic changes and increased participation of women in
the workforce have caused the lack of paid family and medical leave to generate costs, not only for workers and their families but also for society. From 1979
to 2019, the labor force participation rate increased for mothers with children
younger than three years (+ 21.9 percentage points), younger than six (+ 19.8
percentage points), and younger than 18 (+ 15.3 percentage points). Families
are facing increased pressure to balance caregiving needs at home with work
demands. Paid family leave (PFL) policies attempt to ease this pressure by
allowing families to take time off from work when a baby is born or adopted,
or when someone in the family is ill and needs care. The lack of PFL is a serious
issue that affects the most vulnerable workers, reducing their ability to engage
in the workforce and meet family responsibilities.
This Administration helped to address these issues by offering tax
credits to employers that voluntarily offer paid family and medical leave to
employees earning below $75,000. Although this provision of the TCJA will
sunset at the end of 2020, the COVID-19 pandemic and containment measures
have expanded the need for such leave by altering the ways in which many
Americans work and attend school. Individuals must look after their children
more than before because schools and daycare centers are closed or have
limited hours. In addition, the ability to take time off to recover from illness
or help others recuperate is critical in containing the virus. The Families First

334 | Chapter 11

$"0- ррҊхѵ#)" $)(+'*4 *-& -.2$/#/0'
*0-. *2 - T#).0' *0-.,4($'44+ Ѷспсп
$/##$'0) -рт

$/#*0/#$'0) -рт

Year-over-year change (percent)
тп

Jul. 2020

сф
сп

рф
рп
ф
п
–ф
–рп
–рф
–сп

+-.

0).

0'.

Sources: Current Population Survey; Integrated Public Use Microdata Series database; CEA
calculations.

Coronavirus Response Act extended paid family and medical leave to employees of businesses with fewer than 500 employees, which allowed workers to
take time off when they were ill or needed to care for family members. The act
was a temporary action—set to expire at the end of 2020—funded by a refundable tax credit and advanced funds not already on deposit with the Internal
Revenue Service. Paid leave remains an important policy issue as Americans
continue to navigate the pandemic and look toward the future. However,
access to such leave policies is often underprovided by private markets.
The market failure that any Federal paid leave program addresses centers on the positive externalities that paid leave programs generate. A paid
leave program accrues some benefit to both the employer and employee, in
the form of higher efficiency and productivity. This increase is often not large
enough for low-wage workers to receive such benefits from their employers.
However, there are additional benefits to provision that spill over and result
in a positive externality for society. When workers are unable either to take
leave or work while ill (which creates additional problems), they drop out of the
workforce, lose income, contribute less in tax revenues and economic growth,
become more dependent on the government’s safety net, and may even live

Policies to Secure Enduring Prosperity | 335

shorter lives.1 Budig and England (2001) estimate that, of the 7 percent wage
penalty mothers endure per child, about one-third can be explained by a loss
of job experience due to time off or part-time work as a result of childrearing.
Staff and Mortimer (2012) similarly conclude that loss in time spent either
at work or in school is the greatest factor in explaining the motherhood pay
gap. At the same time, these families suffer, and there may be adverse consequences for maternal and family health. Even though workers may realize the
cost that such a lack of leave may impose on them, they may not account for
external costs to the public healthcare system. Similarly, the costs for society
of not giving workers access to paid leave are not internalized by businesses,
which are focused on minimizing their own private costs of production. It may
also be impossible for small businesses with liquidity and capacity constraints
to offer paid leave, even if leave would generate a direct net benefit for their
operations.
Aguirre and others (2012) find that if women’s labor force participation
rates increased to equal those of their male counterparts, U.S. gross domestic product (GDP) could increase by 5 percent. Houser and Vartanian (2012)
estimate that women who take paid leave are 39 percent less likely to receive
public assistance and 40 percent less likely to receive food stamps in the year
after a child’s birth, when compared with those who do not take any leave. Not
only is paid leave associated with fewer dollars in public assistance spending,
it reduces the chance that a family receiving public assistance will increase its
use of public assistance after a child’s birth.

Unequal Access to Paid Family and Medical Leave
The Family and Medical Leave Act of 1993 (FMLA) guarantees unpaid family and
medical leave to 56 percent of American workers (U.S. Department of Labor
2020). The FMLA grants employees the right to take 12 weeks of unpaid leave
to care for newborn children, seriously ill close family members, or themselves.
Employers are not required by the Federal Government to provide paid leave
for employees.
Figure 11-7 shows how access to PFL, regardless of whether provided
through the FMLA, varies with wages, as of 2019. Generally, higher-income
workers are more likely to have access to PFL; 30 percent of workers in the
highest wage quartile have access to PFL, while only 9 percent of workers in
the lowest quartile do.
Although the majority of workers were eligible for leave through the FMLA
in 2018, access to leave through the FMLA is not uniformly distributed throughout the population. In 2018 those who worked for large employers were more
likely to have access to FMLA leave because employers with fewer than 50
1 Sullivan and von Wachter (2009) find a sharp increase in mortality rates for male workers as a
result of work displacement, even 20 years after the displacement takes place. Displaced workers
therefore have a lower life expectancy, by about 1 to 1.5 years.

336 |

Chapter 11

$"0- ррҊцѵ ../*$
*+,0-/$'

1 41 -" "

*)Ҋ#$"# ./,0-/$'

*//*(,0-/$'

Type of Leave
($'4' 1 җ0)+$Ҙ
($'4' 1 җ+$Ҙ
$+ -.*)'' 1
$1/$*).
$.$&' 1
$#*'$4.
п

сп

уп
хп
Percentage of workers

чп

рпп

*0- ѷ0- 0*! *-//$./$.Ѷспрш(+'*4 ) !$/.0-1 4ѵ
*/ ѷ*-& -.- "-*0+ $)/*,0-/$' .41 -" 2" .ѵ*- 3(+' Ѷтп+ - )/*!
2*-& -.2#*. 1 -" .2" .!''$)/# /*+сф+ - )/*!''2*-& -.#1 ../*+$
!($'4' 1 ѵ

employees are not required to offer it; 59 percent of private sector workers
were eligible for FMLA, but they worked at only 10 percent of worksites. Lowwage workers were more likely to have an unmet need for FMLA leave. Nearly
1 in 10 (9 percent) workers who made less than $15 an hour reported that they
needed to take leave but did not qualify for FMLA, compared with 6 percent of
workers who made more than $15 an hour. Several States have supplemented
unpaid leave through FMLA with paid leave programs of their own. As of
January 2020, eight States have enacted PFL with divergent requirements and
benefits.2 This creates a patchwork system that generates a complex burden
on employees and employers that could be alleviated with a nationwide paid
leave policy. Worryingly, Sarin (2016) finds that employers may discriminate
against female job candidates if paid leave is offered; prohibiting firms from
firing employees for taking State-sanctioned paid family leave reduced the
female share of new hires at large firms by 0.6 percentage point, or 1.1 percent.
A paid leave program that is not directly paid for by the employer could reduce
the incentive for such discrimination.

2 These States are California, Connecticut, Massachusetts, New Jersey, New York, Oregon, Rhode
Island, and Washington. A paid family leave law will be effective in the District of Columbia in July
2020, in Massachusetts in January 2021, in Connecticut in January 2022, and in Oregon in January
2023.

Policies to Secure Enduring Prosperity | 337

Effects of Paid Leave on Employment and Earnings
Because very few States in the United States have experience offering paid
leave programs, the research on paid leave has either relied on household
surveys like the Current Population Survey or State-level administrative data
on actual take-up of leave. The empirical literature provides evidence that
paid leave promotes employment, more hours worked, increased income, and
breastfeeding. These factors often disproportionately benefit disadvantaged
populations.
Many studies of paid family leave programs find that the programs do
increase labor force participation, though some find no or negative effect.
Jones and Wilcher (2019) study the effects of State-family leave policies in
California and New Jersey and find that access to PFL increases maternal labor
market participation by over 5 percent in the year of a birth, an effect that
remained significant even five years later. However, Bailey and others (2019)
study the short- and long-term effects of PFL in California and find that for
first-time mothers who elected to take the paid leave, there was a negative
effect on their employment of between 2.8 percent and 3.7 percent in the short
term and between 5.4 and 6.9 percent in the long term. Rossin-Slater, Ruhm,
and Waldfogel (2012) find that California’s PFL initiative doubled the use of
maternity leave, from three to six weeks on average. In addition, it increased
working hours and wages for mothers of young children by between 10 and 17
percent. This effect was particularly pronounced in disadvantaged groups, a
finding backed up by Bartel and others (2019), who find a disproportionately
large difference between White and Hispanic access to paid leave and that
under California’s PFL, fathers of infants were 46 percent more likely to take
leave, an effect particularly pronounced for fathers of first-born children.
Finally, Bartel and others (2019) note that PFL increases breastfeeding by an
average of 18 days, which might lead to long-term health benefits, particularly
for disadvantaged families.
The effects of paid family leave on incomes and earnings are mixed. Even
with PFL, families may suffer from lower earnings in the long run, although
some lower-income women may benefit from a short-run wage boost. Bailey
and others (2019) find that the earnings of first-time mothers with access to
paid family leave were reduced by between $346 and $549 in the short term
and between $541 and $791 in the long term relative to their mean level of prebirth earnings. For first-time mothers who elected to take paid leave, the negative effect on their earnings was between $1,613 and $2,559 in the short term
and $2,522 and $3,685 in the long term. Timpe (2019) similarly finds that expansion of disability insurance programs to cover pregnant women and mothers
of infants caused women’s wages to fall by 5 percent and led to decreases in
family income for families in the middle of the income distribution. In contrast,
Campbell and others (2017) study the effect of expanding temporary disability

338 |

Chapter 11

insurance to new mothers in Rhode Island and find no wage effects for women
in households making less than $50,000 as a whole, but positive wage effects
in the three years after giving birth for women for households making less than
$20,000. For women in households making between $20,000 and $40,000, the
wage effect was positive in the year after birth and indistinguishable from zero
thereafter. Kleven and others (2020) find no long-term effect on female labor
market outcomes, but that leave of longer duration can have a negative effect
on the labor force penalty imposed by children.
Although the empirical evidence on paid leave shows varying effects,
this could be a consequence of different empirical approaches, data used, and
years covered. Analysis on this topic is often hindered by a lack of high-quality
data on access to and take-up of paid leave, and the fact that few States currently offer a State paid leave plan. At the same time, while employers are starting to offer paid leave voluntarily, such programs are more common among
larger employers in certain industries. Finally, while labor market outcomes
are important for measuring the efficacy of paid-leave programs, gains in alternative metrics such as child health quality can be persuasive in determining
whether net societal benefit is generated as a result of a program.

Implementation of Paid Leave
The 2017 TCJA incentivized private provision of paid leave by offering a tax
credit to employers. Several members of Congress have proposed possible
reforms to give more American workers additional access to paid leave. The
Federal Employee Paid Leave Act (FEPLA), signed into law December 2019,
expands the FMLA’s 12-week paid leave benefit for the civil service to cover
all FMLA leave, and to allow the Office of Personnel Management to grant an
additional four weeks of leave. The FAMILY Act proposal would create a new
payroll-tax financed wage insurance program that would pay cash to those
caring for a new child or close family member. The New Parents and CRADLE
Acts would instead allow those caring for newborn or newly adopted children
to receive a portion of their Social Security benefits while on leave. Members
' ррҊрѵ*)"- ..$*)' 3-*+*.'.
"$.'/$*)T$/'
1)$)"0++*-/!*-*-&$)"($'$ .
/
# *-&$)"- )/.' 3$$'$/4/*!спрш
# - *(!*-($'$ ./
# 0++*-/*-&$)"($'$ ./

1 -1$ 2
+/*‫ڦ‬фѶппп1) +4( )/$)/# 4 -
of the child’s birth or adoption
- / ./3Ҋ1)/" +- )/'' 1
.1$)".*0)/.
3+). .!*-!($'4)( $'
' 1 0.
+/*‫ڦ‬хѶппп/3- $/!*-+- )/'' 1

*0- ѷ*)"- ..$*)' . -# -1$ ѵ
Note: FMLA = Family and Medical Leave Act of 1993 ; HSA = Health Savings Account.

Policies to Secure Enduring Prosperity | 339

' ррҊсѵ))0'*./*!$- )/') $'
+Scenario(b$''$*).*!d*''-.)
Type of Leave
$'

1 4& Ҋ

)-$*р

)-$*с

)-$*т

)-$*у

шѵш

фѵп

чѵт

уѵс

- )/'

чѵх

цѵц

цѵф

хѵч

*/'

рчѵф

рсѵц

рфѵч

ррѵп

*0- .ѷ( -$))/ -+-$. )./$/0/ Ѹ-**&$)". )./$/0/$*)Ѹ'0'/$*).ѵ
*/ ѷThe four s )-$*.- ! -/*..0(+/$*).*0/+-*"-(0. ѵ )-$*р’s+-/$$+/$*)-/ .-
.$($'-/*/#*. *. -1 0) -p-*"-(. of the Family and Medical Leave Actѵ )-$*’sс
+-/$$+/$*)-/ .- (*- *).$./ )/2$/# 3$./$)"S// Ҋ' 1 '+$' 1 +-*"-(.ѵ )-$*т
..0( ./#/#'!*!the 2*-& -.are *).$&' 1 )ш*0/*!рп2*-& -.*)+- )/'' 1 2*0'
+-/$$+/ ѵ And s )-$*у..0( .that х$)рп2*-& -.- $1$)"+$' 1 2*0''$(/#
) !$/.!/ -!*0-2 &.*!+$ (+'*4 -Ҋ+-*1$ ' 1 ѵ

of Congress have proposed four tax policy changes to support new parents, as
seen in table 11-1.
The difference between the proposals hinges on the method of financing
any new program of paid leave, as well as the scope of that program. In general,
some proposals have favored using existing programs, such as Social Security
or the Child Tax Credit, and allowing workers to access funds early. In contrast,
other proposals have favored new types of financing, such as a new payroll
tax on employers and employees, to finance paid leave. The proposal on paid
leave sponsored by Senator Bill Cassidy, the Advancing Support for Working
Families Act, would allow families to claim an advance payment, paid back
over 10 years through lower child credits. This act would provide no additional
money to families beyond the value of bringing a future payment forward in
time (or qualified delayed or nonrepayment due to unfortunate circumstances
faced by the family).
An additional policy to expand paid leave could be financed and distributed by State Unemployment Insurance (UI) programs, as proposed in the
President’s Budget, and would be scored at $21 billion for a UI-based proposal
offering 6 weeks of leave. This estimate depends on the assumptions of take-up
rates (table 11-2).3 The cost ranges from $32 million in Wyoming to $2.3 billion
in California, with a median cost of $241 million in Louisiana, using take-up
assumptions based on the FMLA experience. The cost as a share of State wages
ranges from 0.08 percent of annual 2018 wages in the District of Columbia to
0.32 percent in Idaho and Mississippi, with a median of 0.25 in Georgia, Maine,
Missouri, Nebraska, Tennessee, and Vermont (figure 11-8).
3 Estimates made using the American Enterprise Institute–Brookings Working Group on Paid
Family Leave Calculator, available at https://www.aei.org/spotlight-panels/paid-family-andmedical-leave-cost-model/. The CEA further assumed a wage replacement rate of 70 percent, with
maximum weekly benefits of $600, a one-week waiting period, and work requirements in line with
those from FMLA.

340 |

Chapter 11

$"0- ррҊчѵ*./*!$($'4)$&
// " .Ѷспрш

*0- .ѷ0- 0*! *-//$./$.Ѹ( -$)
)/ -+-$. )./$/0/ Ѹ'0'/$*).ѵ

1 .#- *!

../#)пѵрп‫ڿ‬
пѵрп‫ڿ‬/*пѵрф‫ڿ‬
пѵрф‫ڿ‬/*пѵсп‫ڿ‬
пѵсп‫ڿ‬/*пѵсф‫ڿ‬
пѵсф‫ڿ‬/*пѵтп‫ڿ‬
пѵтп‫*ڿ‬-#$"# -

One potential issue with funding paid leave through State UI systems
is that typically State programs vary significantly when it comes to eligibility, wage replacement rates, and the duration of benefits. This can lead to
uncertainty and confusion for workers regarding their eligibility for paid leave.
It can also be problematic for employers to understand if their employees
have access to the State program or not, and can lead to different compliance requirements across states. Therefore, at a very minimum, to guarantee
uniformity across states, it would be important to ensure that any leave taken
for the explicit purposes of birth, adoption, fostering a child, or medical leave
be subject to the same rules across State programs. Tenure eligibility could be
similar to the FMLA program, so that workers need to work with an employer
for about a year. Finally, as the current COVID-19 crisis has shown, State UI
systems can come under significant pressure if additional programs are tacked
onto them. Therefore, adopting this approach to paid leave provision would
require additional planning and administration, and also an investment in
State UI programs.
Paid leave could alternatively be offered through the EITC or the Child
Tax Credit (CTC). Allowing parents to access the portion of the EITC and the CTC
they normally receive as a tax refund at the time of birth instead of after filing
taxes could help parents finance the costs of parental leave. Married parents
making between $10,000 and $40,000 a year with two children could receive

Policies to Secure Enduring Prosperity | 341

between $5,000 and $8,600 in advanced funds to fund parental leave. An additional $2,000 flat payment upon the birth of a child also could give parents of
all income levels funds to cover their expenses and to take time off work, at an
estimated annual cost of $7.6 billion.
This policy can be illustrated with a hypothetical two-parent, married
family with two children under age 13. Both parents work, and it is assumed
each parent makes the same amount of income and has only earned income.
As noted above, the family can receive payments only from the refundable
portion of each tax credit, since the nonrefundable portion can be used only to
reduce actual tax liabilities and is not available as a cash transfer. Withholdings
can be reduced to take into account these credits, thereby increasing takehome pay.
In 2019, for a family with annual income of $20,000, the EITC contributed
$5,828 to their income, while the CTC contributed roughly half that. The combination of these credits could thus provide the family an advanced payment
of $8,453, which could be used to meet childcare expenses associated with
the birth of a child. In combination with the $2,000 bonus, the family receives
over $10,000 toward meeting childcare expenses at the time of a birth or the
adoption of a child. For a family earning $20,000, this is a significant means of
financial support that enables them to provide care for the child for several
weeks, even if they do not have paid leave from their employer.
According to the Centers for Disease Control and Prevention, nearly 3.8
million babies were born in the United States in 2018.4 Offering a $2,000 baby
bonus would cost an estimated $7.6 billion each year. Though this may not
involve the need for new funding, advancing the CTC and EITC could increase
improper payments and would increase administrative costs. In the past, the
take-up rate of the advanced EITC was low, leading to an end to that program.
However, that program was not specifically targeted at new parents.

The Lack of Childcare
Although paid leave is important for working parents who need time off
immediately after the birth or adoption of a child, affordable childcare is often
needed for parents to transition back into the workplace. In an earlier report,
the CEA estimated that as of 2016, the high cost of childcare was preventing up
to 3.8 million parents from joining the labor force (CEA 2019). Over 71 percent
of these parents were married mothers, 21 percent were single mothers, 6
percent were married fathers, and 2 percent were single fathers. In addition
to these 3.8 million parents, the CEA estimated that another 6.6 million nondisabled, working-age parents were working only part time and may require
childcare to increase their working hours. Each of these 6.6 million parents had
4 According to the U.S. Department of State, only 4,058 babies were adopted internationally in
fiscal year 2018.

342 |

Chapter 11

a child under age 13 and had no other potential nonworking caretaker in the
household.
There are societal benefits to high-quality, affordable childcare that do
not accrue solely to the parent and their employer. The positive externalities
generated by higher labor force participation at the intensive and extensive
margins, such as increased tax payments and reduced enrollment in assistance
programs, provide a basis for government interventions that support childcare.
Though substantial government assistance for childcare is currently offered
through the tax code and transfer programs, the benefits are spread over
multiple programs and may not necessarily reflect current childcare costs. The
high cost of childcare is also a result of government regulation of childcare
centers and providers. And though implementing a high standard of safety for
caretakers is of paramount importance, excessive regulation and credentialing
reduces the supply of childcare available, raising the costs above what some
Americans can afford.
A large body of literature in economics has studied the effect of the high
costs of childcare on labor force participation. One Federal program, discussed
further below, is the Child Care and Development Fund (CCDF). A recent study
by the U.S. Department of Health and Human Services finds that a 10 percent
increase in the CCDF leads to a nearly 0.7 percent increase in maternal employment; this conclusion tracks with a meta-analysis (Morrissey 2017) finding that
a 10 percent increase in the price of childcare reduces maternal employment
by 0.5 to 2.5 percent. The effects were strongest for single mothers, mothers
with young children under the age of four, and mothers with low incomes. A
tripling of CCDF funds, for example, could bring an additional 300,000 mothers with young children into the labor force. Blau and Kahn (2013) show that
the gap in labor force participation between women in the United States and
other countries belonging to the Organization for Economic Cooperation and
Development (OECD) could be explained by the lack of paid leave laws and
childcare availability in the U.S.
An earlier report from the Pew Research Center (2014) showed that for
families with working mothers, average weekly childcare expenses rose by 70
percent between 1985 and 2011, and that costs as a fraction of family income
were much higher for lower-income families. In addition to its effect on labor
force participation, the high cost of formal childcare is a possible reason why
the U.S. has a higher reliance on informal care compared with other OECD
countries.

Increasing Access to Childcare
As discussed in the last subsection, spending on childcare can be helpful for
enabling work and improving labor force participation, especially for women.
Today, families receive some support for meeting childcare expenses through
the CCDF, as well as through various tax credit programs.
Policies to Secure Enduring Prosperity | 343

The CCDF is a consolidated block grant to States funded by both discretionary and mandatory Federal dollars that generally funds childcare by
providing vouchers to families for use in childcare centers, family childcare
homes, before- and after-school care, and in some informal settings. In total,
the CCDF provided $8.7 billion in childcare assistance in 2016, with 75 percent
of funds coming from the Federal government and 25 percent coming from
the States. States additionally subsidize childcare directly through the TANF
program. States provide additional funds to eligible families based on TANF
rules, which vary by State but generally include only low-income families that
meet program requirements. In 2018, $3.8 billion in TANF Federal block grant
and State maintenance-of-effort funds were spent on childcare.
In addition to these subsidies, two tax benefits for households specifically subsidize childcare while enabling parental work or educational activities.
The larger of these tax benefits is the Child and Dependent Care Tax Credit
(CDCTC), which allows taxpayers to take a credit of up to $3,000 per child
under age 13 for qualified childcare expenses for up to two children, for a total
of $6,000.5 This credit is worth a fixed proportion, which ranges from 20 to 35
percent of these qualified expenses and depends upon the taxpayer’s adjusted
gross income, with the higher percentages applying to lower incomes. The second tax benefit specifically tied to childcare is a provision whereby employers
may allow employees to contribute up to $5,000 in pretax earnings to flexible
spending arrangements, which can then be used to pay for childcare expenses.6 However, expenses claimed for the exclusion may not be included among
the childcare expenses claimed for the CDCTC. In combination, the CDCTC and
flexible spending arrangements for childcare expenses benefited 6.9 million
families in 2016, for an average benefit of $769 per family. The combined cost
of the CDCTC and flexible spending arrangements was $5.3 billion in 2016.
Many of the policy ideas discussed above for funding paid leave could
also be used to fund family childcare needs. For instance, tax credits like the
EITC and the CTC added over $8,000 for two-parent, two-child families with
annual incomes of $20,000 in 2019. If these credits were further expanded so
that families could claim them in advance, this would allow families to pull forward money in a time of need. Of these, the EITC is the best targeted at lowerincome households and is the most beneficial for covering their childcare costs
because it is fully refundable. The CTC is only partly refundable and is not as
targeted to lower-income households as the EITC. Its benefits extend well up
the income ladder. Modifications of the CDCTC could increase its capacity to
5 The CDCTC is nonrefundable and thus only kicks in once the taxpayer begins to pay income tax.
Crucial to this credit, both spouses (if filing jointly) must earn income or be enrolled in school, and
the childcare provider cannot be a spouse, parent, or other dependent. The CDCTC is also available
for the care of disabled dependents.
6 Flexible spending arrangements use the same qualifications for eligible expenses as the CDCTC.
Employers may also fund the childcare flexible spending arrangement directly, up to the statutory
limit.

344 |

Chapter 11

cover childcare costs. Currently, the cap on the size of the credit has not kept
pace with inflation, which means that while childcare costs have increased,
the maximum benefit has not kept pace with these changes. In addition, the
dependent care credit is nonrefundable, which means the lowest-income
families cannot take advantage of it.
Federal policy could encourage an increase in the supply of high-quality
childcare to reduce its cost to American families. Policy changes could allow
more individuals to provide noninstitutional childcare to friends and neighbors, either by relaxing regulatory requirements or by ensuring that the burden
of credentialing requirements for childcare providers is efficiently achieving
the goals of safety and quality. Potential policies could include lowering the
educational requirements for caretakers and increasing the ratio of students to
teachers when it is consistent with safety and educational benefit to do so. As
discussed in chapter 6 of this Report, the benefits of deregulation tend to favor
households in the lower-income quintiles, and improving access to affordable
childcare while maintaining a high standard of quality has the potential to
greatly benefit lower-income parents.

Enhancing International Coordination
to Meet 21st-Century Challenges
The COVID-19 pandemic caused a historically unprecedented simultaneous
global supply and demand shock, reducing the output of each of the Group of
Seven economies by about 10 to 20 percent. This section discusses issues with
existing international bodies that were underscored during the response to the
pandemic. Rather than a one-size-fits-all approach, this section analyzes the
benefits of a narrow-deep relationship with the United States’ trusted allies
and friends, and a broad-shallow relationship with nations that do not share
the same value systems as the United States. Incongruity in values takes on
profound economic significance because the current global economy is driven
by interwoven networks, a prominent example of which is the Internet itself.
Although networks possess great economic potential, they also introduce
vulnerabilities, and, as detailed below, tend to work best and most securely
between trusted participants. These profound differences in fundamental
values create distrust, and limit the extent to which it is beneficial to share systems. It is U.S. policy that multilateral institutions are useful, and they continue
to be effective for implementing U.S. policy priorities. This section discusses
the benefits of supplementing existing institutions with stronger bilateral ties
with allies that share U.S. values.
The global economy is in the midst of the most profound technological revolution in history—the information revolution—which continues to
transform communication, production, commerce, and conflict. Social and
economic transformations continue to accelerate. The continued growth in

Policies to Secure Enduring Prosperity | 345

connectivity and computing power is driving 5G, artificial intelligence, nanotechnology, three-dimensional printing, and the Internet of Things—each a
major revolution. Quantum computing, rapidly advancing biotechnology, and
profound innovations in energy—all facilitated by, or a part of, the information
revolution—loom on the near horizon. The international order is at a historical
inflection point, fraught with both great opportunities and dangers. To ensure
success, global and domestic strategies must be grounded in the economic
realities of the 21st century. The choices made now will reverberate for a very
long time.

The Economics of Networks, Coordination, and Standard
Setting
Today’s world is driven in large part by the economics of networks, which can
be characterized as any economic or institutional relationship in which the
greater the number of participants, the more valuable the network’s function
becomes to each participant. A classic example is a telephone system. If only
a few people have access to a telephone system, its usefulness is obviously
limited to calls between those few. If, in contrast, the vast bulk of the population of a region has access to the system, then its usefulness to any one user
(and to all of them jointly) is vastly increased. Human languages themselves
are networks—the more people who use a language, the more beneficial for
all users it is to know that language. Any interconnected system of transportation, communication, or technologies constitutes a network, because the more
linkages and connections there are, the more useful it is to all users. As noted
above, the Internet is a network. Railroads are networks, as is the highway
system; if a new, hardtop road more effectively connects a rural village to a
superhighway, it also simultaneously more effectively connects the rest of the
world to that rural village.
In their foundational work, Katz and Shapiro (1985, 1986) and Farrell and
Saloner (1985, 1986) define network effects as positive consumption externalities, such that the benefit that a user derives from consuming a good is increasing in the number of other consumers that use the good. They distinguish
between direct network effects, whereby a user’s utility is directly dependent
on the number of other users such as arise in a telephone system, and indirect
or market-mediated network effects, such as arise in markets for operating
systems where complementary goods such as software will be in better supply
the more users adopt the system. They explain how the benefits of standardization and interoperability are rooted in such network effects. In this subsection,
we use a broad definition of network effects to encompass any coordinated
network of standards. When countries share the same standards, business is
simpler to conduct between these separate jurisdictions, leading to benefits
for all participants. The more members are sharing standards, the greater the
gains for each participant, leading to a network effect working through the

346 |

Chapter 11

supply side. This definition of network effects includes any kind of widespread
standard that generates benefits from interoperability.
For example, standardizing the manner in which computers can communicate with one another constitutes a network (the more users of the
standard, the better for all concerned). If equipment, hardware, machinery,
parts, or tools of any sort are standardized, then they constitute a network; for
instance, the metric system is a network, and the more tools and machines that
are built using the calibrations inherent in this system, the more valuable each
of them is, because they are more interoperable. Networks mesh and overlap
with one another, and they are used in various complementary combinations.
One might use the English language (a communications network) to convey a
message within an email message (part of a communications network) about
purchasing a train ticket (employing a commercial and transportation network) and an intent to hire a ride-share car upon arrival (coordinated through
a communications and commercial network).
Modern market economies are vast networks of networks that pull in
entrepreneurs, labor, capital, materials, and legal infrastructure (among other
factors) to meet the evolving demands of consumers and governments around
the world in real time. What makes this broad network system tick are the
many networks enmeshed within it. For example, what facilitates an exchange
at arm’s length—such as a credit card purchase of a pizza—is a set of trusted
networks, including not only the confidence that the money will be moved
among the pertinent accounts to render payment but also the legal remedies
implied if the ingredients of the pizza are harmful due to negligence.
Similarly, international institutions and standards constitute networks
of international trade, investment, communications, regulation, and procedures to resolve disputes. The Organization for Economic Cooperation and
Development, which originated in 1948 as the Organization for European
Economic Cooperation, was established in its original form to coordinate the
administration of the U.S. Marshall Plan, and today serves as a network by
which countries identify and discuss common problems (OECD 2011, n.d.).
The Society for Worldwide Interbank Financial Transactions network allows
financial institutions, including central banks, a secure means by which to communicate and transact (Cook and Soramäki 2014). More recently, the United
States–Mexico–Canada Agreement (USMCA) is a network, setting standards for
economic activity between the three joining countries (USITC 2019).
Whereas the purchaser of tainted pizza has recourse in U.S. courts,
international networks can often fail to enforce standards. This is especially
problematic when participants in international networks do not share the values of free societies and exploit these same networks to their advantage, disadvantaging the free societies. As a result, the effectiveness of networks at the
international level is reduced. Economic activity will recede when international
coordination and standard-setting create uncertainty. This Administration in
Policies to Secure Enduring Prosperity | 347

particular has been focused on the failure of global networks to enforce intellectual property protection. It is of vital national interest that global networks—
such as international trade, capital markets, and anti-money-laundering or
antiterrorist financing—are aligned with the values of the United States. This
necessitates working alongside like-minded allies and partners individually
and within existing international frameworks to ensure that global standards
do not disadvantage the United States.

The Current Paradigm for Coordination and Standard Setting
This Administration has both recognized the potential benefits of international
coordination yet also highlighted the real limitations faced by international
institutions in their coordination and standard-setting efforts. Recent successes include the update to the U.S.-Korea Trade Agreement (KORUS), the
U.S.-Japan Trade Agreements, and USMCA. As discussed in chapter 9, trade
agreements enhance U.S. firms’ access to supply chains and foreign markets,
allow U.S. consumers to enjoy a wider variety of goods and services, and generate gains for the U.S. economy.
At the same time, it has become increasingly difficult to work with some
existing international institutions. Institutions made up of a broad membership with disparate goals, value sets, and trust structures are most vulnerable
to suffer ossification and become ineffective. Although these institutions can
and do provide broad value, they often fail to produce deep gains through
enhanced cooperation between members, and are unable to allocate gains
among competing interests. And though several international institutions
are aptly characterized by these circumstances, the trade space provides an
appropriate example. Rather than working through the WTO’s Doha Round or
multilaterally, balancing the interests of many parties, this Administration has
focused on achieving gains through narrower agreements, as discussed above.
In the case of the WTO, the Office of the U.S. Trade Representative (USTR) has
noted in the most recent annual report on the WTO that the Appellate Body
“has added to U.S. obligations and diminished U.S. rights” while “several
of [the Appellate Body’s] interpretations have directly harmed the ability of
the United States to counteract economic distortions caused by nonmarket
practices of countries like China” (USTR 2020a, 2020b). This is the predictable
outcome of a network that includes countries with fundamentally different
values and limited enforcement capabilities. More broadly, international organizations have faltered due to a confluence of factors, including the size of the
organizations, the emerging multipolarity of international affairs, and the fact
that existing organizations largely addressed “low-hanging fruit” early in their
tenure. Large institutions contain members with wildly divergent situations
and goals, increasing the frictions and transaction costs of achieving gains
through coordination.

348 |

Chapter 11

' ррҊт.0-/$*)s*!ҝ*0).
)$/$/

*(+' /

-/$$+)/.

0-/$*)
җ(*)/#.Ҙ

) 1

+-. 19уц

/. 19уц

ст

)) 4

+-. 19уш

0". 19уш

рт

*-,04

+. 19фп

+-. 19фр

тч

) 1

). 19фф

4 19фх

сх

рх

*0)

+. 19хп

0'. 19хс

сх

сс

4 19ху

0). 19хц

хс

тц

+. 19цт

*1. 19цш

рпс

цу

-0"04

+. 19чх

+-. 19шу

рст

*1. 20пр

$''*)
)) 4
*&4*

рфт

шр
ۛ ссш

*0- .ѷ*. -)*. җспрсҘѸ'0'/$*).ѵ
Note: GATT/WTO = General Agreement on Tariffs and Trade / World Trade Organization.

To illustrate this first point, consider the durations of the negotiation rounds of the General Agreement on Tariffs and Trade / World Trade
Organization (GATT/WTO), shown in table 11-3. These rounds have increased
steadily over time, alongside the number of participants—with the Doha round,
initiated in November 2001, still outstanding (Moser and Rose 2012). Increased
participation, an indicator of broader multipolarity in the WTO, is associated
with longer negotiation durations. Measures of productivity, such as average
tariff cut per year of negotiations, show a relative stability through the Uruguay
round (Martin and Messerlin 2007) though there will need to be large cuts as
part of the Doha round for this trend to continue.
This relationship makes clear the trade-off the United States faces in working through some broad international organizations. Though the potential for
benefits rises with organization size, so do heterogeneity costs from increased
diversity among these states (Posner and Sykes 2013; Bradford 2014). On the
margin, a new member must be valued against the cost of reduced cohesion
and ability to make decisions. It is important that international organizations
reach optimal membership decisions, with failures of judgment resulting in a
free-rider problem where countries are unwilling to contribute and unlikely
to engage in voluntary arrangements (Buchanan 1965). This is not to say that
the WTO or other international institutions would necessarily be better off as
a result of a U.S. withdrawal. Aside from working to generate new gains for
the U.S. through narrow cooperation, U.S. participation in broad international
institutions also serves as an institutional safeguard against the possibility of
those institutions taking actions that contradict American values and priorities,
or becoming dominated by America’s rivals. The United States’ participation
leverages heterogeneity costs to its advantage, making it difficult for countries

Policies to Secure Enduring Prosperity | 349

with different values to co-opt existing institutions. The Trump Administration
has recognized this, and thus has worked both within existing institutions and
outside them to generate gains from deeper coordination.
Evaluating the costs and benefits of admitting new members to an organization or evaluating an agreement is complicated, because of changes in
the global landscape and difficulty in enforcing international agreements. For
example, the situation in 1947 was quite different than today. Many of the trade
agreements of that period were made with a desire to support countries as a
strategic counterbalance against totalitarian economies (Martin and Messerlin
2007). At that time, the United States was willing to accept nonreciprocal and
unfavorable trade deals to benefit American allies (Baldwin 2006). However,
this has proved harmful when international institutions have been unable to
enforce compliance. The USTR has noted that “China’s entry into the WTO
[was] on terms that have proven to be ineffective in securing China’s embrace
of an open, market oriented trade regime” (USTR 2018, 2019, 2020a, 2020b).
As detailed in the 2018 Economic Report of the President, the United States has
very little negotiating power today within the existing WTO structure because
the current U.S. trade barriers are so low. This makes the negotiating process
exceptionally difficult for the United States.
Another difficulty is that enforcing agreements is made conditional on
a country’s accession to an international organization. Consider the case of
China’s accession to the WTO. As part of this agreement, the WTO engaged
in consent tailoring, which involved requiring China to engage in economic
reforms as part of the accession process. At the time, it was thought that
through accession, China would also engage in economic reform. But this
has not come to fruition. Neomercantilist nations, such as China, that engage
in industrial policy create dangers in markets characterized by network economics. A government that is heavily subsidizing a champion company for a
network niche, as China has been doing with Huawei and 5G, might exploit
the dominance of the company to pursue geostrategic interests and might not
be a trustworthy steward of a network upon which so many and so much will
rely. These issues beg the question of how to proceed toward the goal of free,
fair, and reciprocal trade within the broader network of trade agreements and
international organizations.
Adam Smith stated in The Theory of Moral Sentiments (1759) that to
make a market economy work, trust is an essential component. Successfully
navigating the challenges inherent in this rivalrous global environment will
necessitate economic partnerships with other like-minded and trusted nations
in a deeper and more integrated manner than what has been done in the past.
As Evensky (2011, 261) states: “When trust is shaken, individuals pull back and
the market system contracts. Where trust grows, individual energy and creativity are unleashed and the system grows.” This is the great geostrategic and
economic challenge that confronts international economic structures today:
350 |

Chapter 11

how to build, govern, and maintain extensive cross-jurisdictional networks
while ensuring that they are secure, reliable, and based on well-founded trust.

Opportunities for Advancing Coordination
To remain nimble, future deep international partnerships must be based on
economic and geostrategic interests as well as on shared values. Economic
theory suggests a way forward. As opposed to seeking shallow agreements
between countries with differing belief systems, the United States can generate
gains through deep integration with countries with a similar economic situation and regulatory system (Buchanan 1965). One approach to this integration
is to develop narrow collectives like the European Union and African Union
that work in parallel to and in support of broad international organizations.
These collectives rebalance the cost-benefit analysis by attempting to reduce
heterogeneity costs and allowing for gains through deeper integration. This
can be done by adopting global rules or creating rules from scratch to advance
the goal of borderless markets (Davies and Green 2008).
Another approach is to create bilateral agreements between countries
with similar economic values that can later be extended into multilateral agreements such as USMCA. Under this framework, political and monetary action
would remain the prerogative of each country. Moreover, regulatory uniformity
would not be enforced by an extranational government, but alignments could
be achieved through mutual recognition and acceptance of equivalency in the
outcome of each system. Nations may pursue two different methods of regulating industry, but can still reach regulatory convergence on key issues such as
safe products, fair work environments, and well-stewarded natural resources.
For example, in 2008 the United States and Australia entered a limited mutual
recognition arrangement for regulatory exemptions that would permit U.S.
and eligible Australian stock exchanges and broker-dealers to operate in both
jurisdictions, without the need for these entities to be separately regulated in
both countries (SEC 2008; Jackson 2015). This foundation of trust provides a
model for international cooperation that could provide economic returns not
captured by current multilateral efforts, while still constituting a laboratory for
eventual broader, multinational efforts (Buchanan 1965).
These approaches have the potential to benefit all the countries involved,
by allowing deep integration with aligned nations while maintaining an economic relationship with countries that are unable or unwilling to couple their
economies under a framework of shared values and trust. As the importance of
these collectives grow, they will offer greater benefits to membership, creating
an incentive for countries to meet the criteria for joining (Bradford 2014).
The U.S. has the potential to generate particularly large gains through
coordination with like-minded countries to help limit negative externalities
from countries with different goals. The enforcement of intellectual property rights provides an example. The U.S. used Section 301 tariffs to enforce

Policies to Secure Enduring Prosperity | 351

intellectual property rights against China. Intellectual property theft is quite
costly: the OECD estimates in a 2019 study that international trade in counterfeit goods amounted to $509 billion in 2016. The IP Commission (2017) estimated in 2017 that the cost to the U.S. economy of counterfeit goods, pirated
software, and theft of trade secrets is more than $225 billion a year. Assuming
a discount rate of 3 percent, the cumulative cost of inaction is $7.5 trillion.
The Office of the Director of National Intelligence estimated in November 2015
that economic espionage through hacking costs $400 billion a year. Although
the U.S. has achieved enhanced protection of intellectual property through
bilateral negotiations with China (discussed in chapter 9), additional gains
could be achieved by working alongside like-minded countries that share the
United States’ perspective on intellectual property protection. By creating a
collective to better enforce intellectual property rights against China, the U.S.
can increase the gains and lower the costs.
Reorienting American policy from a broad-shallow framework to include
narrow-deep coordination will allow for greater benefits and flexibility in making trade agreements. This will allow for the stagnation of the past decades
to be overcome and create a higher-income and more closely linked world.
As recovery from the COVID-19 pandemic continues, building international
structures for a return to prosperity and prevention of future reoccurrences
becomes of paramount importance.

Prospects for U.S.-U.K. Coordination
As an example of this new approach to economic partnerships, the U.S. could
explore the potential for an explicit economic and geostrategic deep partnership with the United Kingdom. There are many possible nations with which the
U.S. could partner in this way, and this is just one possibility. Although trade
agreements like KORUS and USMCA provide excellent templates for future
partnerships and highlight important areas of coordination for contemporary economies (see chapter 9 for a discussion of these agreements), future
agreements could go beyond a trade deal. Blueprints for such cross-border
arrangements have been outlined in depth by Tafara and Peterson (2007) in the
context of financial market regulation.
The optimal economic result would be economic integration that
increases gross trade flows and innovation. Trade in goods, services, labor,
and ideas would be as free as possible, yet consistent with maintaining full
sovereign independence. Facilitating this goal would involve intensive bilateral
processes, such as a free flow of labor through streamlined processes facilitating citizens of the U.S. and U.K. to work and live in the other’s jurisdictions
or mutual recognition of financial institutions that would enhance the global
reach of both markets. Another area of potential benefit is the security of
devices, software, and networks. The U.S. and U.K. would benefit from coordination on determining what types of devices are not secure and pose a risk to

352 |

Chapter 11

the development of safe networks. Yet another area for enhanced coordination
is institutions of higher learning, which provide the foundation for future innovation. A U.S.-U.K. consortium of universities might act as a catalyst for centers
of innovation and further collaboration. The exchange and flow of faculty, students, and ideas would build tremendous technological momentum in itself.
The U.S. and U.K. could vigorously and jointly prosecute the theft of intellectual
property and coordinate remedies—like tariffs, sanctions, and prohibiting the
operation of certain firms—to form a truly deep partnership.
The economic benefits of such a partnership could be substantial. As a
narrow example, consider the 1958 agreement on automotive industry standards between the European Union, Japan, South Korea, and other countries
(though not the United States), which increased automotive trade between
partner countries by more than 20 percent through regulatory harmonization.
In exploring the elimination of tariffs between the U.S. and the U.K., a 2000
report by the USITC found that U.S. imports from the U.K. would increase 7 to
12 percent, and U.K. imports from the U.S. by 11 to 16 percent, although the
aggregate output effects were not substantial (USITC 2000). A broader review
of the literature on mutual recognition agreements by the OECD found that, in
almost all cases, such agreements boosted international trade flows between
partner countries (Correia de Brito, Kauffmann, and Pelkmans 2016). The direct
benefits of any agreement will be sensitive to the provisions therein. However,
what is both more valuable, and more difficult to value, are the early and future
network effects of such an agreement.
The U.S. is positioned to lead in the development of a new generation
of flexible, bilateral economic partnerships that protect national sovereignty
and interests while seeking to produce gains from cooperation. This stands in
contrast to the expansion and sharing of networks with countries with which
the U.S. has fundamental disagreements, and with which it does not share
sufficient trust to warrant generating the vulnerabilities that are inherent in
shared processes. Pursuing deep integration with allied nations will facilitate
economic recovery from the COVID-19 pandemic and create a safer world
where such a crisis can be addressed in a more coordinated fashion.

Creating a More Effective Healthcare System
The U.S. healthcare system faces several interwoven challenges. The current
COVID-19 pandemic is focusing attention on the importance of a resilient and
efficient healthcare system for maintaining a strong and vibrant economy. This
section discusses several of these challenges and potential reforms that would
increase the efficiency of the American healthcare system. Increasing transparency in healthcare markets and increasing the supply of healthcare will help
individuals access treatment, both for any direct COVID-19 health effect and
also for other diseases and injuries.

Policies to Secure Enduring Prosperity | 353

Rationalizing the Provision of Healthcare Professionals
Several elements of the current healthcare market structure impose large distortions on the mobility and training of healthcare professionals. These distortions limit the supply of medical practitioners, especially to underserved rural
and low-income populations, artificially increasing medical costs. Alleviating
these distortions could generate gains for many Americans.
Medical professionals currently face large hurdles to labor mobility due
to restrictive licensing requirements. Because medical licensing is performed
at the State level, providers cannot easily move between States as they face
both monetary and temporal costs in the form of additional examinations,
interviews, fees, and paperwork. This creates a strong distortionary effect on
the labor market, which is particularly pronounced in metropolitan areas that
cross State lines, where healthcare workers can face major bureaucratic and
monetary barriers to taking a job only a few miles away. Such regulatory burdens on employment are associated with lower levels of job-switching, which
may decrease upward economic mobility, result in higher rates of unemployment, and ultimately lead to higher prices for consumers.
Efforts are being made to limit the distortionary effects of these regulations within the health sector through interstate licensure compacts. These
compacts aim to either provide portability or streamline the acquisition of
licensure in other signatory states. However, the effectiveness of such plans
is limited by incomplete adoption across states. Figure 11-9 demonstrates the
patchwork nature of three of the largest licensing compacts for healthcare
providers: the Nurse Licensure Compact (NLC), the Advanced Practice Nurse
Compact (APRNC), and the Interstate Medical Licensure Compact (IMLC). The
IMLC is further limited because it provides only a streamlined process for
physicians to apply for licensure in other states, which decreases some of the
bureaucratic obstacles but retains the negative effect of licensure fees on provider mobility. Further complicating the regulatory landscape, there are similar
compacts for other healthcare professions, including social workers, mental
health professionals, physical and occupational therapists, pharmacists, and
dentists.
Efforts to combat the inefficiencies of individual state licensing have
been ongoing for decades. When the Federal Government has taken targeted
action to remove barriers, it has been successful. The Department of Veterans
Affairs (VA) allows licensed physicians to practice in any state to increase the
quality and decrease the cost of care, and the Health Resources and Services
Administration award grants to State-licensing boards to encourage cooperation, which has resulted in the landscape of interstate compacts that can
somewhat ameliorate the issue. More recently, the Centers for Medicare &
Medicaid Services (CMS) has taken deregulatory actions spurred by the COVID19 pandemic that allow licensed providers to care for Medicare patients across

354 |

Chapter 11

$"0- ррҊшѵThe -

*1 ( )/*!

*0- ѷ$.&) ) )/ -ѵ
*/ .ѷ# 2$' *( /$1
!/ - $)" )/ 4/*/'*!рп
.// .ѵспрш-$5*)'2 ).0- .
3+ $/ '$ ).0- ++-*1'!*-''
# '/#- +-*! ..$*)'.ѵ

'/#- *-Ѷспсп

./-$/ //
3+ $/ #4.$$)++-*1'
җ Ҙ
0''*$'$/4!*-ҝ
0-.$)"
җ
Ҙ
0''*$'$/4!*-ҝ
о3+ $/
#4.$$)++-*1'җ о Ҙ
0''*$'$/4!*-ҝ ҝо3+ $/
#4.$$)++-*1'җ оо Ҙ

State boundaries to facilitate increased care in hotspot areas and enable a
nationwide expansion of telemedicine. The Federal Government may be able
to increase the availability of care and decrease the cost of healthcare by playing a coordinating role. Creating incentives for States to either adopt stronger
portability of licensure between States or encourage the usage of a Federal
licensure system modeled on the approach of the VA are strong steps toward
increasing access to care and curbing rising medical prices.
Other supply-side problems include the limited number of medical
schools and accredited residency slots for medical school graduates. Of 53,030
medical school applicants, only 22,239 (42 percent) matriculated into a medical school in 2020. Though U.S. medical school enrollment has increased by 31
percent between 2002 and 2018, residency training positions have expanded at
a rate of just 1 percent a year.
Reducing the barriers to entry for aspiring doctors and improving the
educational process for individuals pursuing a medical profession would
ensure that the United States has superior healthcare provision in the years to
come. There would be positive feedback effects for patients from adding more
doctors in historically underserved areas, because it would enable a lighter
burden to be placed on each doctor, reducing burnout and encouraging more

Policies to Secure Enduring Prosperity | 355

individuals to join the medical field as doctors, nurses, and other healthcare
professionals. According to the Agency for Healthcare Research and Quality
(2017), in recent years the rising prevalence of burnout among clinicians (over
50 percent in some studies) has led to concerns on negative effects to access
care, patient safety, and care quality. Doctors suffering from burnout are more
likely to leave their practice, which reduces patients’ access to care. Burnout
can also threaten patient safety and care quality when depersonalization leads
to poor interactions with patients and when affected physicians suffer from
impaired attention, memory, and executive function.
To address concerns that funding for graduate medical education (GME)
is poorly allocated, this Administration has proposed, since fiscal year 2019,
to consolidate all GME spending in the Medicare, Medicaid, and the Children’s
Hospital GME Payment Program into a new mandatory, capped Federal grant
program. The distribution of funding to hospitals through this new grant program would depend on the proportion of residents training in priority specialties as well as other criteria identified by the Secretary of Health and Human
Services. Such an improvement in the distribution of GME funds would serve
to achieve a better distribution of healthcare specialties, address shortages in
healthcare professionals nationally and especially in medically underserved
communities, and incentivize better training of healthcare professionals.

Balance Billing
When a patient sees an out-of-network provider, they may be liable for the
difference between what the provider charges and the amount their insurer
would have paid an in-network provider. This difference, known as the “balance billed” amount, is owed in addition to any other out-of-pocket amounts
such as deductibles and copayments. In some cases, patients actively choose
to pay this additional amount in order to seek care from an out-of-network
provider. When a consumer lacks key information or choice in what they are
purchasing, such as when a patient unknowingly receives care from an out-ofnetwork provider or when a patient does not have the ability to select an innetwork provider, there is a market failure. This situation can arise even when
a patient receives care at an in-network hospital, because different providers
within a given hospital independently make decisions on which types of insurance to accept. Adopting network matching at the Federal level would require
any provider that takes care of patients at a hospital to bill as in-network any
patient who the hospital also considers to be in-network.
The Trump Administration has taken direct actions to address the issue
of surprise billing. In June 2019, Executive Order 13877 directed agencies to
ensure that patients have access to meaningful price and quality information before the delivery of care. Beginning in 2021, hospitals will be required
to publish their real price for every service, and to publicly display—in a
consumer-friendly, easy-to-understand format—the prices of at least 300

356 | Chapter 11

different common services that are able to be purchased in advance. In April
2020, the Administration began requiring providers to certify, as a condition for
receiving supplemental COVID-19 funding, that they would not seek to collect
out-of-pocket expenses from a patient for treatment related to COVID-19 in an
amount greater than what the patient would have otherwise been required to
pay an in-network provider. In May 2020, the Department of Health and Human
Services released the Health Quality Roadmap to empower patients to make
fully informed decisions about their healthcare by facilitating the availability
of appropriate and meaningful price and quality information.
Several States have also taken action on balance billing, though many
resort to price-setting or arbitration, which can alter the negotiating power of
hospitals, insurers, and physicians. For example, California has attempted to
limit patients’ cost sharing for all nonemergency physician services at in-network hospitals from out-of-network physicians at the greater of the insurer’s
local average contracted rate or 125 percent of the Medicare rate for the given
service. As a result, physicians have criticized the law for giving insurers the
upper hand in negotiations and for decreasing patients’ access to care. In addition, New York’s arbitration system has been criticized for granting excessive
bargaining power to providers in their rate negotiations with insurers, resulting in higher reimbursements and premiums. By contrast, an analysis by the
Congressional Budget Office (CBO) found that network matching could actually
lower costs by reducing the ability of healthcare providers to negotiate higher
rates from insurers, avoiding New York and California’s pitfalls by harnessing
market forces to address the balance billing issue.
The CEA finds that protecting patients from balance billing could provide
an economic benefit of $2.8 billion a year by creating greater predictability in
healthcare expenses. A total of 11.1 percent of privately insured patients in a
given year will seek emergency room care, and 6.2 percent will be admitted to
the hospital. Data from a recent study indicates that, of these patients, about
42 percent can expect to receive a surprise balance bill with an average amount
of $628 for emergency room care and $2,040 for inpatient admissions. The
elimination of balance billing lowers uncertainty and increases transparency.
Based on the statistics above, the actuarial value of this reduction in risk is
$82.40 per patient a year, and patients value the elimination of uncertainty at
25 percent of this amount.7 Thus, 25 percent of $82.40 per patient multiplied by
the 137 million adults covered by private insurance yields an aggregate annual
economic benefit of $2.8 billion from eliminating balance billing.

7 This is calculated from the statistic that households willingly pay $1.25 in health insurance
premiums for each $1 in average payouts.

Policies to Secure Enduring Prosperity | 357

Medicare Inpatient Rates
According to data on national health expenditures from CMS, Medicare
hospital payment is currently one of the most regulated price mechanisms
in the U.S. health economy, accounting for about $300 billion in government
expenditures in 2019 alone. A recent proposal for a new Federal rule seeks to
better calibrate the price mechanism to market prices. The existing pricing
systems rely on estimates of growth rates of costs based on assumptions that
often have little bearing on market prices outside Medicare Fee for Service. One
proposed solution is to rely on data from Medicare Advantage pricing, which is
partly based on private sector negotiations. However, there is a concern that
Medicare Advantage pricing is closely linked to Medicare Fee for Service pricing. If the private sector negotiations are using prices set by the government
as an anchor, then the price discovery is curtailed and the usefulness of the
negotiated price is limited.
The CEA completed an analysis to compare private and government
prices for the top 25 inpatient Diagnosis Related Groups (DRGs), based on the
inpatient payment system used for Medicare fee for service (FFS) payment. A
DRG is a patient classification system that standardizes prospective payment
to hospitals and encourages cost containment initiatives. In general, a DRG
payment covers all charges associated with an inpatient stay from the time
of admission to discharge. Specifically, Medicare FFS average payments are
compared with Medicare Advantage and Private Insurer data from 2015 using a
publicly available data source (Parente 2018). Figure 11-10 displays the percent
change in the relative price of Medicare FFS when compared with Medicare
Advantage and Private Insurers. The relationship between Medicare FFS and
private insurance payments and Medicare Advantage does not support using
Medicare Advantage as a close substitute for competitive market pricing (i.e.,
prices set by private insurers). If they were close substitutes, the observations
would cluster narrowly around the 45-degree line shown in the figure. This
analysis supports furthering the development of the proposed Federal rule to
integrate pricing that more closely matches competitive market prices.
Future policy analyses will be able to take advantage of the new all-payer
insurer synthetic database created by the price transparency Executive Order
138777 of June 2019. This database could easily confirm and extend this initial
analysis and provide strong evidence of the need to revisit the economic price
control mechanism used to set Medicare FFS inpatient payment prices.
The health reforms discussed above would be complemented by a continuation of the health information technology reforms that took place this
year (see box 11-2). Better storage and sharing of information about health
care will increase transparency, reduce cost, and improve the experience of
patients.

358 | Chapter 11

$"0- ррҊрпѵ - )/#)" $) $- 4( )/.
*(+- 2$/# $- 1)/" )-$1/ ).0- -.Ѷ*+
сф$")*.$.Ҋ '/ -*0+.Ѷспрф
Percent change compared with private insurer

рсп
рпп
чп
хп
уп
сп
п

Ҋхп

Ҋуп

Ҋсп

сп

уп

хп

Ҋуп
Ҋхп
Ҋчп
Percent change compared with private Medicare Advantage

*0- .ѷ4)спрчѸ'0'/$*).ѵ

Building a Dynamic Economy through
Infrastructure Improvement
The infrastructure of the United States is made up of physical elements like
roads and ports, but also of less obvious components, like digital infrastructure. The COVID-19 pandemic highlighted the importance of high-quality
infrastructure in responding to the crisis, as well as increasing the productivity
of the American workforce. Locally planned and led infrastructure projects
are desirable because of their ability to be responsive to the specific needs of
local communities. However, when projects are too large for local financing,
require coordination between multiple States, or are instrumental in achieving
national goals, the Federal Government has a role to play. This section details
the infrastructure investment channels in which the Federal Government’s
intervention would be beneficial.

The Federal Government’s Role in Infrastructure Investment
Federal involvement in infrastructure investment can have a beneficial effect
as a countervailing force in local politics. Local politicians often face strong
incentives to prefer new projects over maintaining existing ones, and they may

Policies to Secure Enduring Prosperity | 359

Box 11-2. Continuing the Historic Modernization of Health
Information Technology Begun during COVID-19
The decentralized nature of the U.S. healthcare system can create challenges
for coordination in a national public health crisis. In March 2020, it became
evident that more precise health data were needed to coordinate the COVID19 response. Given the number of agencies needing these data, the Federal
Government embarked on an ambitious plan to modernize the national
health information technology (IT) infrastructure to facilitate the seamless, secure reporting of critical and sensitive health data from healthcare
organizations and IT vendors. This effort led to the creation of the Federal
data platform named HHS Protect that has allowed Federal agencies, along
with State and local partners, to coordinate using shared data. This platform
streamlines processes and connections to ensure data quality while eliminating time-intensive duplicative IT work.
HHS Protect contains, within a single portal, over 3.5 billion data
elements across 200 different data sets. This information is available in real
time to drive the Federal and State government responses to the COVID-19
pandemic. Having access to this real-time data allows the government to
more accurately pinpoint patients in need, to identify regions and healthcare
systems that are under strain, and to allocate treatments and resources
more rapidly and efficiently. Having granular, hospital-level data in near real
time has been critical for developing an understanding of the severity of the
disease, the status of capacity and staffing constraints, and the supply and
demand of personal protective equipment. The HHS Protect platform has also
played a central role in the COVID-19 vaccine trials. Specifically, using data
from claims and other sources, it allows analysts to algorithmically identify
at-risk populations and locations on the verge of a COVID-19-surge to target
data gathering so as to quickly and efficiently meet the necessary benchmarks
for trial validity.
No such infrastructure existed before this massive data mobilization
and IT modernization. In contrast to the prepandemic status quo, currently
99 percent of hospitals report data consistently and 94 percent report every
single day. Similarly, automated connections now exist for case and laboratory reporting from State and local jurisdictions. In addition to the data and
platform, the Federal Government has built out new and existing teams of
experienced, well-trained analysts from HHS, the CDC, the U.S. Digital Service,
and other agencies, as well as leveraging the expertise of trusted external
partners to ensure that these data are available in an actionable, interpretable form to inform decisionmaking.
This groundbreaking effort will also prove useful for future pandemics and other health crises. For example, it can facilitate more in-depth
exploration of the recent declines in American life expectancy and can
target resources to combat opioid overdoses, suicides, and heart disease. One
important innovation that has emerged is the creation of Health Information

360 | Chapter 11

Exchanges in multiple States that link real-time electronic health records data
to track intensive care unit surge capacity for COVID-19 patients. This platform
could combine data from public and private insurance transactions to greatly
enhance the planning, execution, and response to future pandemics. As mentioned above, one policy proposal advocates the allocation of GME funding to
areas with greater medical need. This platform could be adapted to identify
those areas and ensure that they receive the healthcare professionals needed
to provide a high quality of life for residents. The innovations developed during the fight against the COVID-19 pandemic will continue to serve Americans
in the years to come.

also avoid imposing user fees that could be unpopular among frequent users
(Kahn and Levinson 2011; Glaeser and Ponzetto 2017).
Over a short time horizon, maintenance is often a more effective use of
funds than new capital investment. Keeping existing infrastructure in good
repair is likely to have a larger economic effect than building new infrastructure, given that existing infrastructure is already woven into the fabric of highoutput economic environments and generates a higher marginal return than
new construction. Nadiri and Mamuneas (1996) find no evidence of overinvestment or underinvestment in highway capital by the end of the 1980s, indicating
that maintaining the existing stock would ensure the correct amount of infrastructure intensity after that point, with new infrastructure only needing to be
built at a rate commensurate with the growth of the population and economic
needs. Every $1 spent to keep a road in good condition prevents $7 in costs
when it has fallen into a poor condition (AASHTO and TRIP 2009). Given that
estimated output multipliers for transfer payments to State and local governments for infrastructure range between 0.4 and 2.2, focusing funding on repair
could have a positive effect on GDP (CBO 2015).
According to the CBO (2018), in 2017 the share of Federal spending on
maintenance was just 27 percent of total Federal spending on transportation
and water infrastructure, and the real dollar amount spent has remained flat
since the 1980s, even as the stock of infrastructure has increased. State and
local spending have not increased their growth rates to compensate. Currently,
the Federal Government primarily funds road infrastructure through the
Highway Trust Fund. But this fund is now facing insolvency, with a projected
deficit of over $6 billion as soon as 2022. Zhao, Fonseca-Sarmiento, and Tan
(2019) estimate that the current cost of deferred repairs might be as large as
$873 billion, or 4.2 percent of GDP. Nongovernmental estimates find that $110
billion to $150 billion per year would be needed to cover the infrastructure
investment gap through 2025 (McBride and Moss 2020; American Society of
Civil Engineers 2016).

Policies to Secure Enduring Prosperity | 361

Simple and transparent metrics for discretionary grants would allow
fulfillment of projects that have difficulty getting local funding because of
their size or cross-jurisdictional nature. This would include projects across
multiple States and that fulfill significant national goals. The process could
expand upon the existing TIGER/BUILD model, which entails federally funded
discretionary grants that attempt to achieve national objectives, but could
improve on them by emphasizing numerical metrics for economic, safety, and
environmental impact by using cost-benefit analysis that follows a consistent
and clear evaluation process. The Federal Government can provide technical
assistance to avoid biasing the process against smaller applicants and ensure
adherence to best practices (U.S. Department of Transportation 2020).
As discussed in chapter 8 of this Report, public-private partnerships
(PPPs) can enable provision of high value infrastructure at a low cost to the
government. Well-designed PPPs are structured to ensure that the private partner has strong performance incentives at the same time that the public interest
is protected (Istrate and Puentes 2011). For example, if the same entity builds
and manages the project, this can align incentives to minimize operational
costs in the design and implementation of the project. Best practices for PPPs
include robust competition between private vendors to win the partnership
and contractual terms that optimally divide the risk burden between the vendor and the government. If the private entity is allowed to earn a return to the
investment through user fees, the partnership contract must carefully consider
what, if any, role the public retains in terms of approving or setting the fees. If
the infrastructure will be a natural monopoly—as a road on public land with
few competing roads nearby—this issue is particularly important to address.
Finally, efforts can be made to avoid deterring private financing of infrastructure investment funded by user fees. The cost of capital of PPPs relative
to public risk assumption and funding is high (Arezki et al. 2017). As detailed
by Makovsek (2018), the government can take several actions to offset the risk
and uncertainty private investors face when contemplating publically beneficial investment. Such hurdles include failure of environmental review, change
in political situation, or other regulatory hurdles that introduce additional
risk to the process of providing public infrastructure. In addition, the interest
on State and local government debt instruments is exempted from taxation
and government-owned entities providing infrastructure services are given
preferential tax treatment. Offering guarantees against the unavoidable risk
of infrastructure provision and equalizing taxation treatment could increase
private investment.

Infrastructure and Productivity
Productivity, a measure of the ratio of outputs to inputs, is the result of a complex interaction between infrastructure, education, research, investment, and

362 |

Chapter 11

$"0- ррҊррѵѵѵ-*0/$1$/4Ѷршуш–спрш
Percent change at annual rate
ч

спрш
-'41 -"

FiveҊ4 1 -"

у
с
п
–с

–у
ршуш ршфф ршхс ршхш ршцф ршчс ршчш ршшф сппс сппш спрф
*0- .ѷ (2020)Ѹ'0'/$*).ѵ

implementation. Especially important for productivity growth and innovation
are public infrastructure investments, because the choice of where and what
to build has tremendous implications for the enterprises that rely on them.
If productivity increases, an economy can create more with less. This leads
to a higher-income economy with more leisure time and less environmental
degradation for the same level of economic output.
For decades, there has been concern within the United States regarding
a decline in labor productivity growth (Munnell 1990). As shown in figure 11-11,
the late 1990s saw a rise in productivity growth as industry implemented new
information technologies; but since the Great Recession, productivity growth
has remained depressed. This decline in productivity growth does not appear
to be due to mismeasurement (Byrne, Fernald, and Reinsdorf 2016; Syverson
2016). Research productivity has been found to be declining as a result of
requiring more resources to advance the frontiers of scientific knowledge
in existing fields (Bloom and others 2020). New fields such as artificial intelligence, quantum computing, and autonomous vehicles can reverse this trend
when their benefits are diffused to the broader economy. However, there is a
concern that dynamism is falling for cutting-edge fields. Astebro, Braguinsky,
and Ding (2020) find that start-up formation by doctorate recipients in science
and engineering has fallen, partly due to increased complexity and administrative costs.

Policies to Secure Enduring Prosperity | 363

Box 11-3. 5G Infrastructure
The United States has begun the transition to 5G technology for wireless
communications. Far from a minor improvement on 4G, the next generation
of wireless communications features vastly expanded data capacity and
speed unleashing new opportunities for innovation and economic growth.
As a simple example, movies will be downloaded over the Internet at speeds
more than 10 times faster than is possible today. However, the technology
may also enable transformative applications such as self-driving cars, remote
surgery, and increasingly intelligent manufacturing. In a 2018 report, the
Federal Communications Commission (FCC) describes the country as “at the
brink of another technological revolution,” in which 5G networks “will make
possible once-unimaginable advances” (FCC 2018). The FCC further reports
that the wireless industry is expected to invest more than $275 billion over 10
years to deploy infrastructure for 5G, that 3 million jobs will be created, and
that the boost to GDP will be half a trillion dollars.
The integration of digital technologies with artificial intelligence and
machine learning has been called the fourth industrial revolution (Schwab
2016). This integration will be enabled by 5G technology. Over time, 5G is
expected to enable transformative applications through two novel facilities: massive, machine type communications (mMTC); and ultrareliable,
low-latency communications (URLCC). The abbreviation mMTC refers to the
capacity of the 5G infrastructure to support a very large number of devices in
a network of sensors, known as the Internet of Things. For example, “smart
cities” may deploy dense monitoring systems that reduce the cost of public
services from city lighting to garbage collection by allowing for more efficient
provision. URLLC refers to 5G performance standards that are designed to
support “mission critical” communications. URLLC provides for data delivery
at latencies as low as 1 millisecond with 99.9999 percent reliability. Use cases
include automated energy distribution in a factory or energy grid, intelligent
transportation systems, and bioelectronic medicine (Rysavy Research and 5G
Americans 2020).
Although private providers—such as ATT, Verizon, and T-Mobile—are
investing in and building out the 5G infrastructure, the Federal Government
is playing a key role in organizing auctions for commercial licenses to the
electromagnetic spectrum. This is a complex undertaking because 5G is
designed to use different parts of the electromagnetic spectrum in combination. 5G providers will use high-band spectrum (frequencies above 24
gigahertz) to transmit vast amounts of data at high speeds with low latency.
However, high-band spectrum does not travel far and cannot, for example,
penetrate walls. Providers will use mid-band spectrum (frequencies between
1 and 6 gigahertz) to augment the high band spectrum for broader coverage
at somewhat reduced speeds. Finally, providers will use low-band spectrum
(frequencies below 1 gigahertz) to efficiently transmit data across very broad
geographic areas at lower speeds.

364 |

Chapter 11

The FCC has made significant progress in implementing its strategy
named Facilitate America’s Superiority in 5G Technology. In 2019 and 2020,
the FCC ran three auctions (auctions 101, 102, and 103) that released almost
5 gigahertz of high-band spectrum to the market (FCC 2020a, 2020b, n.d.).
These auctions generated about $10.3 billion in gross bids for just over 20,000
licenses. Winning bidders included ATT, Verizon, T-Mobile, and US Cellular,
along with smaller wireless providers. On August 25, 2020, the FCC concluded
its first auction (auction 105) of mid-band spectrum for 5G, releasing about
70 megahertz of spectrum in the range of 3.55 to 3.65 gigahertz. The auction
generated $4.6 billion in gross bids for 20,625 licenses. Winning bidders
included Verizon, the Dish Network, and several large cable companies. The
FCC is currently organizing a spectrum auction to repurpose 280 megahertz
of mid-band spectrum for use with 5G. The spectrum, which is in the range
of 3.7 to 4.2 gigahertz, is currently used by fixed satellite service companies,
primarily to deliver audio and video content to cable systems. As part of the
auction process, the satellite companies will vacate the spectrum, allowing 5G
providers to use it instead.
Despite the disruptions caused by the COVID-19 pandemic, the FCC has
pursued an expedited schedule for the auction, finalizing bidding procedures
in August, with bidding under way as of December 2020. The FCC is also working on targeted changes to facilitate 5G usage of low-band spectrum. Finally,
the Department of Defense has contributed to the government’s 5G efforts; in
August, it announced that it will release 100 megahertz of mid-band spectrum
in the range of 3.45 to 3.55 gigahertz, which was hitherto reserved for military
use, to be auctioned in late 2021 with commercial use starting in 2022.

To the extent that the Federal Government can support the infrastructure necessary to ensure that these emerging fields can flourish, this will
enhance productivity. However, government support for business structures
that are no longer viable will decrease the rate of productivity by preventing
“creative destruction,” a process whereby innovative firms enter the market
and stagnant firms exit (Acemoglu et al. 2018). Consequently, government
operates optimally not when it invests in specific enterprises, favoring certain
companies over others, but when it provides rules and transparency that allow
companies to compete on an even playing field (see box 11-3 for a discussion
of how the government allocates spectrum, allowing companies to fairly compete). Judicious government investments in infrastructure can fulfill this goal.

The Costs of Building Infrastructure
Foerster and others (2019) find that the annual rate of GDP growth has fallen
more than 2 percentage points since 1950. As they argue, part of the reason
for this is a decline in the trend in total factor productivity of the construction

Policies to Secure Enduring Prosperity | 365

sector of 0.15 percentage point. Although this decline in productivity was
reversed after 1999, the trend value of construction sector labor fell by 0.07
percentage point between 1999 and 2016. The cost of building highway infrastructure in the United States has risen 94 percent between 2003 and 2020,
according to the National Highway Cost Construction Index. This increase
began long before this period, having increased threefold from the 1960s to the
1980s (Brooks and Liscow 2019); these increases are partly driven by increased
regulation (see box 11-4).
Labor regulations can also increase the overall cost of building infrastructure. For example, the 1931 Davis-Bacon Act requires the Federal Government
to pay wages to construction workers that are no less than what they would
earn working on similar projects in that area, known as “locally prevailing
wages.” This regulation can artificially increase labor costs, which can result in
a nonoptimal allocation of capital. Although this was a Great Depression–era
attempt to raise wages, Davis-Bacon also increases the burden of regulation, which discriminates against small firms. Reform could ensure that such
regulations are achieving an appropriate balance between competing policy
priorities.
In addition to the National Environmental Act in 1970, other environmentally focused legislation has increased the difficulty of developing on public
lands, such as the 1973 Endangered Species Act and the 1972 Clean Water Act.
Citizen organizations and environmental nonprofits have also increased since
the 1970s, often opposing some development projects. Environmental impact
reviews can delay projects or force developers to take expensive routes to fulfill
the project, increasing overall costs (Brooks and Liscow 2019). While ensuring
that new investment does not “steal” from the public by despoiling public
goods is crucial, a rapid and transparent approval process could ensure that
the capital is not tied up, but, if rejected, can be repurposed for alternate, more
environmentally friendly projects.
Another factor in rising infrastructure costs is a lack of transparency and
competitive processes. This limits public oversight and leads to wasteful and
corrupt spending, which reduces the return on investment. For example, the
Long Island Rail Road project has paid $3.5 billion for each mile of track, a rate
seven times the world average. Schwartz and others (2020) estimate that 15
percent of advanced economy infrastructure investment is lost to waste.

The Critical Importance of User Fees
Certain goods have characteristics that are not amenable to pricing, and therefore must be financed by general revenues. Clean air, for example, cannot be
priced for use because there is no practical way to exclude those who do not
pay from breathing it. Moreover, there is no cost from additional consumption,
as one’s breathing it does not reduce another’s ability to do so by a notable
amount. Economists refer to such goods as “nonexcludable and nonrivalrous.”

366 |

Chapter 11

Box 11-4. Reforming the NEPA Process
The National Environmental Policy Act (NEPA), which was signed into law in
1970, requires Federal agencies to assess the environmental effects of their
proposed actions and prepare an environmental impact statement (EIS) for
actions that will significantly affect the quality of the human environment.
For such actions, agencies must consider ways to minimize significant effects
through reasonable alternatives or mitigation. The lead Federal agency must
also solicit and consider public comments on potential environmental effects
and alternatives. If it is completed, the entire EIS process takes about four
and half years to complete on average, and averages over 600 pages in length.
The Council on Environmental Quality (CEQ) finalized reforms in 2020
to modernize its NEPA regulations and reduce delays in the environmental
review and decisionmaking process. Changes include establishing a presumptive two-year time limit for the process, clarifying definitions and procedural
requirements, codifying efficient agency practices to reduce unnecessary
paperwork and delays, and updating the regulations to reflect current
technologies. CEQ’s updated NEPA regulations also include aspects of the
Administration’s One Federal Decision policy, established by Executive Order
13807, which addresses major infrastructure projects that require multiple
agencies to approve permits. The One Federal Decision Executive Order,
which is codified in the updated regulations, requires agencies issuing multiple permits for a project to develop a joint permitting schedule, develop one
EIS, and then issue a joint record of decision for the project.
By reducing the time for completing NEPA reviews from 4 to 2 years,
the CEA estimates that these policies will lead to $739 billion in benefits from
infrastructure projects over the next 10 years. The benefits come from earlier
completion and lower costs of financing projects, leading to improved infrastructure and amenities. The CEA bases the estimate on the $2.35 trillion in
roads, airports, waterways, pipelines, and utility investments that are needed
to modernize infrastructure in the next 10 years (American Society of Civil
Engineers 2016). Though it is not clear that all these projects would require
review under NEPA or preparation of an EIS, the CEA estimate is conservative
in assuming that delays under the current NEPA permitting process are only
4 years rather than 4.5 years. Research by the CEA has shown that public
infrastructure investments provide a marginal product to society at a rate of
12.9 percent a year.
Accordingly, the value of moving the benefits of $2.35 trillion in investments forward two years is $479 billion. An additional benefit of the reduced
delay in the permitting process is that developers of this infrastructure do
not need to hold loans for as many years, incurring interest on the principal
loaned to undertake the project. The estimated reduction in financing costs
is $260 billion for loans made on a principal of $2.35 trillion, or the difference
in interest payments on a 4-year versus a 2-year loan. If the decision is made

Policies to Secure Enduring Prosperity | 367

to not approve a proposed project, those resources can more quickly be reallocated to a more beneficial use.
The benefits from reform could be even larger than we have estimated
because some permits are not just delayed by some years but instead are
never issued at all, because the company requesting them moves into financial hardship while awaiting a response. Bear Lodge mine in Wyoming is one
example. Rare Element Resources attempted to open a mine for rare earth
minerals, which have been designated as a critical mineral, on U.S. Forest
Service (USFS) land. The company submitted its plan of operations to the
USFS in November 2012. In September 2013, the USFS accepted the initial
plan and started the process of finding a contractor to undertake the EIS. In
January 2016, Rare Element Resources suspended permitting efforts because
the company had run out of money waiting for the project to be approved.
With so much attention being paid to reshoring critical industries
(see chapter 9), simply capitalizing domestic natural resources responsibly
would ensure access to many key commodities, improving the U.S. trade
balance, mitigating U.S. reliance on vulnerable commodity supply chains,
and ensuring that resources are extracted and used in a sustainable way.
This would allow for more sustainable trading partnerships without imposing
distortionary trade barriers that disrupt supply chains and impose costs on
American consumers. It would also allow commodities to be produced in a
more sustainable way than is often done in other countries.

However, when it is possible to exclude users and when additional consumption imposes costs on others, setting a price on a good can efficiently internalize the costs associated with its use. Roads, canals, and bridges are examples
of goods that are both excludable and rivalrous, and therefore would benefit
from pricing plans.
As detailed in the 2018 Economic Report of the President, it is optimal
when the users of a public good are those who pay for it. This prevents overconsumption of public goods by ensuring that the costs of using them are borne by
the users, and provides a source of funding for the maintenance and upkeep
of these goods. Without a clear and sustainable funding stream, infrastructure
can become a burden on future generations.
Although Federal gasoline taxes partly fund Federal infrastructure projects, and so partly align the users with the costs of use, most of these projects
are financed through general revenues. Because drivers are not bearing the
costs of driving on public roads and bridges, they do not have any incentive
to economize their use of them, leading to congestion and high maintenance
costs. User fees, such as tolls or fees based on vehicle miles traveled (with
both scaled to the damage the use does to the infrastructure), reduce congestion and help provide a stable source of funding for infrastructure. Expanding

368 |

Chapter 11

their use, and other forms of congestion pricing, would yield further economic
benefits.
The Federal Government has implemented several kinds of user fees
over many decades to finance public infrastructure, although these have not
adequately addressed the problems of depreciation and congestion. The
Federal Government passed a gasoline tax in 1931, initially set at 3 cents a
gallon, which then was roughly 10 percent of the price of gasoline. The Federal
Highway Act and Highway Revenue Act of 1956 attached user fees explicitly to
the Interstate Highway System. These included taxes on gasoline, diesel fuel,
tires, and heavy vehicle use, though the vast majority of the Highway Trust
Fund revenues depend on fuel taxes. However, these taxes are not pegged
to inflation, and the Federal gasoline tax in particular has not been raised
since October 1993, although the general price level (price index for GDP) has
increased by a multiple of 1.65. The gasoline tax is now 18.4 cents a gallon,
which was 17 percent of the cost of gasoline in 1993, and is currently roughly
half that.
In addition, higher-mileage vehicles, including electric vehicles, render
the gasoline tax an incomplete and flawed user fee. Higher-mileage vehicles
depreciate physical assets, such as roads and bridges, as much as lowermileage vehicles of equivalent mass, but pay less per mile when fuel taxes are
utilized as a user fee. Though a gasoline tax may generate ancillary benefits by
reducing pollution and greenhouse gas emissions, dependence on this tax to
fund the maintenance of Federal roads and bridges is inadvisable.
Alternatively, Federal, State, and local governments could consider
increasing their use of toll roads to finance public infrastructure and reduce
congestion. These toll roads could vary their charges based on the vehicle type.
One example of these is high-occupancy toll (HOT lanes). HOT lanes charge
low occupancy vehicles a fee, while buses and emergency vehicles can use
the lanes free of change. Currently, there are 10 HOT lanes operating across
8 States. Some academics and government officials have advocated converting high-occupancy vehicle (HOV) lanes, which restrict use to only qualifying
vehicles, to HOT lanes to increase usage and reduce congestion in other lanes.
Research on the effectiveness of HOT lanes has been mixed, however. If
the toll is set too low, a HOT lane may actually reduce the incentive to carpool
and therefore generate more congestion, given that single occupants may be
content to simply pay the toll to access the lane (Burris et al. 2014; Konishi and
Mun 2010). In HOV lanes, these single-occupant vehicles would typically have
been excluded. However, toll prices that are set optimally can reduce congestion and finance the maintenance of the public asset.
Governments could also consider varying toll prices based on the time
of day. Variable tolls, as opposed to flat-rate tolls, charge drivers more during
peak travel hours to reduce congestion. In Fort Myers, Florida, a 50 percent
discount on the toll was offered on the Midpoint and Cape Coral bridges for a
Policies to Secure Enduring Prosperity | 369

short period before and after the rush hours. Survey data revealed that, among
those eligible for the discount, there was an increase in traffic of as much as
20 percent during the discount period before the morning rush hour, with corresponding drops in the rush hour itself.
A handful of cities and countries have embraced “cordon pricing,” which
charges drivers for entering certain areas. Instead of traditional tollbooths,
vehicles are charged through transponders that are scanned by overhead
antennas to detect entry. Currently, 70 to 80 percent of toll fees are collected
this way in the United States. In Germany, highway authorities use Global
Positioning System technology to administer truck tolls on its autobahns. An
in-vehicle device records all the charges based on the location of the vehicle,
and then the owner of the vehicle uploads the charge to a processing center.
The costs of such systems are as much as $500 per vehicle in Germany, but their
presence reduces the need for roadside equipment and labor for toll collection.
Cordon pricing has had considerable effects on congestion. Table 11-4
details the cities and countries that have embraced this form of congestion
pricing, and summarizes the economic effects. In the year after implementation, traffic congestion fell in London by 30 percent. Bus service increased by 23
percent due to improved reliability and reduced travel times. Of the thousands
of car trips that no longer traveled into the cordon zone, 50 percent shifted to
public transit, 25 percent were diverted to other parts outside the cordon area,
and the remainder shifted to carpooling, walking, biking, and traveling outside
peak hours. These results have been sustained over time, despite nearly 20
percent population growth in London since 2000. The city has also achieved
concomitant public health benefits, as carbon dioxide emissions declined 16
percent from 2002 to 2003. In Stockholm, traffic in the cordon area has fallen
20 percent and carbon dioxide emissions have fallen 14 percent.
Singapore’s road pricing plan reduced congestion by 20 percent from
1975 to 1998, and generated revenues that were nearly nine times the costs
of investment. When Singapore switched to its current electronic road pricing
system in 1998, this reduced congestion even further, despite strong population growth since then. Congestion in the inner city has fallen by 24 percent,
while average speeds have increased by more than 30 percent, and bus and
train usage have increased by 15 percent. Revenues have been used to support
public transit; street safety; and bus, rail, and bicycle infrastructure projects.
However, Singapore also introduced stringent measures to restrict car ownership, requiring the purchase of a certificate that can cost as much as $37,000
that must be recertified every 10 years. Since 2018, Singapore has required
new drivers to bid on existing certificates, as it will not allow any increase in
car ownership.
Although most economists agree that congestion pricing would make
the average person better off, many States and localities are reluctant to adopt
such pricing mechanisms because of the system’s perceived inequities. As of
370 |

Chapter 11

Policies to Secure Enduring Prosperity | 371

сппч

ршцф

спсрҡ

$')

$)"+*-

2*-&$/4

0/*(/$/*''$)"/($) )/-4+*$)/./*$)) -$/4
.' 1$ /2 )хѷтпѵ(ѵ)хѷтп+ѵ(ѵMonday–FridayѶ
varying based *) ()
'/ $'4 #-" ++'$ . /* -$1 -. )/ -$)" /# $/4 /2 )
цѷтпѵ(ѵ)цѷтп+ѵ(ѵ
.$ )/. 3 (+/!*-!$-./уп )/-$ .Ѷ/# )+4- 0 #-"
' /-$)#4-$-. 3 (+/
-$' +-$$)".# ( %0././**)" ./$*)' 1 '.$)- '/$(
!-*(цѵ(ѵҊч+ѵ(ѵѶMonday–/urday
#$' . ,0$++ 2$/#.#*- 1$ ../*-$)"!-
-*2) -.#$+ -/$!$/ .җ'$($/ Ҙ)*./‫ڦ‬фпѶппп
Ҋ.. .4./ (—'- 4 0. !*- -$" ) /0)) ' /*''s—
#-" .1 #$' . )/ -$)")#//).*0/#*!хп/#/- /
3- $/.!*-- .$ )/. -)$)"' ../#)‫ڦ‬хпѶппп))0''4
)$/$*)/*+- 3$./$)"*)" ./$*)! .!*-/3$. and rideshares

8 square miles inside the city’s Inner Ring Road
-!!$( -.+/0- '$ ). +'/ )0( -.0+*) )/-4ҝ 3$/Ѹ
+42$/#$)) 3/4
'/$'4! !-*(цѵ(ѵ–х+ѵ(ѵѶonday–riday
0/*(/$/*''$)"/($) )/-4+*$)/./*$)) -$/4
.' 1$ /2 )хѷтпѵ(ѵ)хѷтп+ѵ(ѵMonday–FridayѶ
1-4$)". *) ()

*'$$ .

*/ ѷ 2*-&$/4*)" ./$*)+-$$)"+').- ./$''/ )//$1 ѵ
*0- .ѷ-$Ҋ// -).+*-//$*)(+$")Ѹ0-*+ )*(($..$*)ѸNew York TimesѸBloomberg CityLabѵ

спрт

сппх

/*&#*'(

*/# )0-",2 )

сппт

-
(+' ( )/

*)*)

$/4

' ррҊ4ѵ*)" ./$*)-$$)" )$/$/$1 .*-'2$

))0'*+ -/$)"*./ ҝ ) /- 1 )0 ѷ‫ڦ‬сф($''$*) ҝ ‫ڦ‬рфп($''$*)
$) ршшчѶ/-!!$$)$)) -$/4have -*++ су+ - )/
- )#*0. ". ($..$*).$)*-*)- - 0 рп–рф+ - )/

*- /#)тп+ - )/- 0/$*)$)/-!!$
тт+ - )/-*+$)'&-*) ($..$*).

))0'*+ -/$)"*./ ҝ ) /- 1 )0 ѷрпп($''$*)&-*) ҝ рѵт
$''$*)&-*)
-!!$ '4. - . тп–фп+ - )/
чп + - )/ - .$ )/ *++*.$/$*) - 1 -. !/ - .0 ..!0' +$'*/Ѹ
$/$5 ).1*/ /*& ++ -() )/
-!!$ '$) рс+ - )/0-$)"#-" #*0-.
0'$/-).$/0. %0(+ су+ - )/!*-*((0/ -.

))0'*+ -/$)"*./ҝ) /- 1 )0 ѷ‫ڧ‬ртп($''$*) ҝ ‫ڧ‬ртц($''$*)
тп+ - )/- 0/$*)$)/-!!$*)" ./$*)
0.-$ -.#$+- # фпҊ4 -#$"#$)спрр

(+/

Box 11-5. Digital Infrastructure
The COVID-19 crisis has revealed inadequacies in medical information systems, with some officials relying on fax machines to relay critically important
health data (Kliff and Sanger-Katz 2020). Box 11-2 above details the digital
infrastructure investment made in response to this need. As the American
public suffered job losses of unprecedented scope, dated computer software
was overwhelmed by skyrocketing State Unemployment Insurance claims.
Government agencies could be better served by modernizing data
storage. For example, the Department of Veterans Affairs (VA) began creating
an in-house records system amid the advent of personal computing in the
1970s. Though state-of-the-art in its prime, the VA’s system continues to draw
criticism from lawmakers concerned about maintenance costs and security
issues related to commercially tested software alternatives. As currently
implemented, despite several modernization attempts and billions of dollars in annual spending, the VA’s information technology fails to adequately
support its services and protect against security threats. However, a recent
change in policy will allow veterans to have more access to private healthcare,
which will require the VA to coordinate with other healthcare providers and
reinforces the necessity of updating its outdated IT systems (Steinhauer 2019).
Without modernization, critical infrastructure like regional power grids
and hospital medical records remain at risk for potentially catastrophic cyberattacks (GAO 2018). In 2017 alone, Federal executive branch civilian agencies
reported more than 35,000 security incidents, ranging from email phishing to
malicious software installations. The Government Accountability Office finds
that legacy systems at the Department of Homeland Security pose 168 “highor critical-risk” network vulnerabilities (GAO 2019). In many cases, foreign
intelligence agencies spearhead such intrusions. As recently as October 2020,
the Department of Justice indicted six Russian military officials for disrupting
the 2017 French elections and the PyeongChang Winter Olympics, among
other attacks (DOJ 2020).
Standardized data storage and streamlined information-sharing processes would foster more efficient interagency collaboration on governmentwide initiatives like coronavirus relief. In addition, providing information for
public use once it has been scrubbed of identifying characteristics would
significantly benefit scientific research. Academics argue that providing
direct access to administrative data can further strengthen the position of the
United States on the forefront of academic progress (Card et al. 2011).
New digital infrastructure will also be needed in cutting edge sectors
such as the autonomous vehicles industry. The Federal Government can ease
the entry of new players in such sectors, thereby supporting competition, by
supporting infrastructure investment. For the autonomous vehicles industry,
this will involve ensuring that localities have the infrastructure in place to support autonomous vehicle technology for passenger cars, long-haul trucking,
and short-range drones in a consistent and universal manner. Investing in

372 |

Chapter 11

unmanned traffic management systems and streamlining regulatory approval
for engineers seeking airspace to test their inventions will be a major step
toward a future of half-hour package deliveries and decongested roadways.
Federal intervention can facilitate the development of robust national-scale
systems instead of a State-by-State patchwork.
Altogether, advances in digital infrastructure could generate tremendous economic gains for the United States. The autonomous vehicle industry
alone is expected to increase output by $1.2 trillion, or roughly $3,800 per
person (Clements and Kockelman 2017). A Virginia Tech study found in a
cross-sectional analysis of three large U.S. cities that drone delivery could
produce $583 million in total time savings for consumers each year in those
cities (Lyon-Hill et al. 2020).

2008, only 1 percent of roads in the United States used some type of congestion
pricing beyond traditional tollroads. However, the Federal Government could
incentivize State and local governments to adopt such policies by offering to
match infrastructure investments if such pricing is adopted, or offer Federal
support to help compensate low-income populations that might be adversely
affected by congestion pricing.
An additional area in which government and private interests alike could
benefit from increased Federal investment is in digital infrastructure. Relevant
projects involving improved data gathering, enhanced security, and revolutionized computing efficiency. These public goods will increase the productivity of private industry and the well-being of the American citizenry (box 11-5).

Investing in Port Infrastructure
America’s seaports serve as our economy’s gateways to the vast maritime
network that transports more than 90 percent of global trade tonnage. These
360 American ports facilitate more than 70 percent of America’s international
trade by weight and nearly 50 percent by value. They are absolutely vital for
American prosperity and economic security. However, America’s ports now lag
far behind the competitiveness and productivity of the world’s leading ports.
As a result, increased transportation costs place American companies at a significant disadvantage, effectively locking the U.S. economy out of the world’s
lowest-cost trade routes. Over time, the poor state of the Nation’s ports has
taken a significant toll on America’s position in global trade and manufacturing, and has adversely affected net exports. Because maritime commerce is
the dominant global trade mode, competing nations have prioritized enhancements to their seaports to improve the flow of trade. Increasing the efficiency
and productivity of ports decreases transportation costs and expands trade
opportunities. Capital investment to modernize American ports is long overdue
and would yield substantial returns for the Nation’s trade and manufacturing

Policies to Secure Enduring Prosperity | 373

competitiveness by supporting net exports, as high-quality port infrastructure
benefits exports more than imports through supply chain interactions.
Container ships spend less time making port in Taiwan (0.46 day), China
(0.62 day), Japan (0.35 day), Spain (0.66 day), and Norway (0.33 day) than in
the United States, where it takes a full day on average. U.S. ports take longer
than the world average in five out of six market segments (UNCTAD 2020). Deep
channels, round-the-clock automated cargo handling, and constant competitive improvement enable nearly frictionless trade. Ports with these features
serve, allowing tremendous economies of scale and complex supply chains to
flourish. Low transportation costs are pivotal for the production of goods that
require the movement of many players. For large parts and bulk goods—such
as those required in aerospace, automotive, and industrial production—waterborne commerce often offers a transportation mode for intermediate inputs.
Conversely, higher transportation costs resulting from poor maritime
infrastructure relegate host nation economies to favor less complex supply
chains, the exporting of primary goods, and the importing of intermediate
and finished goods. Over time, this can have the effect of deindustrializing a
nation. Poor maritime logistics, (including the absence of entire classes of bulk
transportation legs) leads to situations where hog farmers in North Carolina
import feed grain from Brazil instead of America. The United States also leads
the world in exporting scrap metal because the existing water transportation
modes cannot move this cheap and abundant domestic strategic commodity
to American minimills.
Half of all goods, by value, entered the U.S. in 2017 through maritime
ports, and the Federal Maritime Commission estimates that these ports generated nearly $4.6 trillion in economic activity in 2014. Increased investment
in infrastructure at U.S. seaports would speed the flow of goods, expand
trade, and increase U.S. GDP. For example, the World Bank (2020) gives the
United States a rating of 5.8 out of 7 on its port infrastructure and ranks the
Netherlands number 1, with a 6.8 rating. Extrapolating from an analysis by
Munim and Schramm (2018), the CEA finds that increasing the United States’
port infrastructure by 1 point (to 6.8 out of 7) would increase U.S. real GDP by
0.3 percent ($56 billion).
In 2018, 5 of the 50 busiest seaports in the world were located in the
United States (figure 11-12). By contrast, 16 were located in China, and 10 of
these were busier than the United States’ busiest port—Los Angeles. Whereas
U.S. ports processed an average of 7 million 20-foot equivalent units of container traffic annually, Chinese ports averaged 38 million units.
Infrastructure investment plays a role in this discrepancy. The average
service life of U.S. airports and seaports is 38 years, below the global average,
and the required port investment forecast through 2040 is 90 percent higher
than current port investment trends, according to a 2017 analysis by Oxford
Economics, which concluded that the United States’ gap between needed
374 |

Chapter 11

$"0- ррҊрсѵ0( -*! +*-/.4*0)/-4 a(*)"/# *+фп
0.$ ./ +*-/.$)/# *-'Ѷспрч
Quantity of ports
рч
16
рх
ру
рс
рп

у
с
п

#$)

European Union

*0- ѷ*-'#$++$)"*0)$'ѵ
*/ ѷ The b0.$ ./. +*-/.are .( .0- 4/*))" ѵ

and actual investment is one of the highest in the world. Over the last decade,
increased trade volume has led to busier maritime ports and increased congestion. In 2014, the top three container ports alone accounted for nearly half of
all containerized trade in the United States. The slower throughput caused
by congestion has started to affect the performance of U.S. ports relative to
other countries’ ports and has begun to divert trade away from American ports
through increasingly competitive Canadian and Mexican ports. Figure 11-13
shows the tonnage shipped out of the Port of Seattle against that shipped from
the Port of Prince Rupert (one of the main ports on the West Coast of Canada).
Although the Port of Seattle has seen increased traffic over the last four years,
the growth in the Port of Prince Rupert’s traffic is almost double that of Seattle.
The Federal Maritime Commission estimates that if trade volumes at maritime
ports continued to grow between 5 and 7 percent annually, the average growth
rates for the East Coast and Gulf Coast ports since 2009, port capacity would
need to double in the next decade to accommodate this growth.
The depth of shipping channels plays a significant role in the volumes
and rates of the trade a port can facilitate. If a waterway is too shallow, large
vessels benefiting from low unit transportation costs will be unable access the
foreland infrastructure through the adjacent port. Many American ports are
too shallow to facilitate trade with the large ships regularly used in the lowest
cost trade routes internationally. Overall, due to inadequate dredging and

Policies to Secure Enduring Prosperity | 375

$"0- ррҊртѵ -"**'0( .*!the *-/*! //' *(+-
with the*-/*!-$) 0+ -/Ѷспрф–рч
Index (2015 = 1)
рѵф

*-/*!-$) 0+ -/

*-/*! //'

пѵф

п
спрф

спрх

спрц

спрч

*0- .ѷ-$) 0+ -/*-/0/#*-$/4ѸU.S. +-/( )/*!-).+*-//$*)Ѹ '0'/$*).ѵ

insufficient port depth, the U.S. likely forgoes $376 billion in trade annually,
according to the U.S. Army Corps of Engineers (2017). This has become a major
challenge for American energy producers in particular. Increased oil production
in the United States has elevated the demand for oil transportation services via
very large crude carriers (VLCCs) across Gulf Coast ports. However, all but one
American oil export terminal have insufficient depth to handle VLCCs. As a
result, Gulf Coast ports need to use smaller tankers to carry out costly and inefficient ship-to-ship oil transfers, thus loading VLCCs in open water, many miles
offshore. Russian and Saudi producers are able to load VLCCs directly, reducing
their oil’s price per barrel and dramatically increasing their pace of exporting
to market. This inefficient method of loading American oil exports generally
adds several extra days and $1 million to shipping costs for each shipment of
American oil exported from the Gulf Coast. This inefficiency translates into an
additional $0.50 to $0.80 cost per barrel for American oil (Huchzermeyer 2018;
Miller 2019). Updating and upgrading U.S. maritime laws to ensure adequate
dredging capacity could therefore generate substantial economic gains.
Improving American trade competitiveness can be accelerated by leveraging the Nation’s unique comparative advantages. The United States has
rapidly emerged as the world’s leading producer of natural gas and is becoming one of the world’s top liquefied natural gas (LNG) exporters. Concurrently,
the global deep-sea shipping industry has begun a seismic shift toward LNG

376 |

Chapter 11

as a marine propulsion fuel. Because fuel costs are typically the leading driver
of operating expenses for vessel owners, the 15–30 percent reduction in costs
realized by shifting from fuel oil to LNG has become a superior alternative. In
addition to fuel conversions, dual-fuel- and LNG-powered vessels are becoming increasingly ubiquitous. This shift toward LNG as a marine fuel has not been
matched with investment in the bunkering infrastructure that is necessary to
fuel LNG-powered vessels except in the most modern competitive ports in
Asia and Europe. Investing in U.S. port infrastructure to enable LNG refueling
terminals would make American ports more attractive.
Other Federal regulations have exacerbated congestion at the ports. One
of the largest amplifiers of this situation is the shortage of trailer truck beds, or
chassis. Once the goods arrive at maritime ports, trucks are typically used to
transport the goods to other destinations. Stricter Federal safety requirements
introduced in 2009 were factors in creating a situation in which providing a
chassis is too cost-prohibitive for ocean carriers. Known as the “roadability
rule,” this regulation placed the burden of the safety inspection of a chassis
on drivers, increasing regulatory compliance costs. As a result, many of the
carriers sold their chassis to third-party leasing companies, which now provide
the majority of chassis at ports. Truckers must now make multiple trips to pick
up and return chassis, leading to delays in the availability of chassis at any one
time. This limits the amount of goods truckers can move in a timely manner.
For example, in January 2019, chassis shortages led to large backlogs at the Los
Angeles and Long Beach terminals.
Research demonstrates that improving port infrastructure increases
trade. Countries that enhance their seaport efficiency can decrease shipping
costs by up to 12 percent, leading to increased trade (Clark, Dollar, and Micco
2002). Cheaper shipping both increases firm sales and helps spread new technology (Lakshmanan 2011). These outcomes allow firms to scale up their activities and increase domestic output, which has positive spillover effects into the
labor market and the rest of the economy.
There is also evidence that port infrastructure investment not only
boosts total trade but also provides a larger boost to exports than imports.
Improved port infrastructure increases exports by $4 for every $1 increase in
imports (Wilson, Mann, and Otsuki 2004; Korinek and Sourdin 2011). Physical
infrastructure has an even higher positive effect on exports as income grows
(Portugal-Perez and Wilson 2012). Improved port efficiency has increased
incomes in communities around ports by up to 70 percent (Brooks, GendronCarrier, and Rua 2018).

Policies to Secure Enduring Prosperity | 377

Generating a More Skilled and
Resilient Workforce
As a result of the COVID-19 pandemic, an unprecedented fall in economic
activity occurred in 2020. Although this was met by a rapid and massive policy
response that focused on maintaining the social capital of employee-employer
relationships, there is still work to be done. Improving the productivity of the
U.S. workforce is more important than ever as the country recovers from the
pandemic-induced recession. This section outlines two ways in which this
can be done: increasing the skills of the immigrants who contribute to this
country, and improving the institutions of higher-learning that equip millions
of Americans for the labor market.

Points-Based Immigration
In contrast to other developed countries—such as Canada, Australia, and
Japan—the United States immigration system limits the ability of high-skilled
workers to immigrate to the United States if they do not have existing family
relationships. Under current U.S. law, immigrants obtain lawful permanent
resident (LPR) status (i.e., green cards) through immigration categories for
familial relations, employment, the diversity lottery, and the refugee and
asylum programs. Table 11-5 shows the distribution of those receiving LPR
status in fiscal year 2018. The U.S. distribution of immigrant visas diverges
significantly from countries with merit-based points systems where over half of
permanent immigration visas were granted based on the employment or skills
of the applicants (figure 11-14).
Points-based immigration systems select their employment-based immigrants by awarding points based on factors such as age, education, and
earnings that are associated with positive outcomes. For example, Canada’s
merit-based immigrants earn more than other Canadian immigrants. Abbott
and Beach (2011) find that the median 10-year average earnings of Canada’s
merit-based immigrants are 35 to 56 percent above the median 10-year average earnings of all immigrants in the most recent cohort they consider. The
CEA’s estimates, which are discussed below, and a review of the economic
literature suggest that there is strong evidence that shifting the U.S. immigration system toward a merit-based system would lead to benefits for the U.S.
economy—increasing growth, wages, and tax revenue.

Estimated Economic Benefits
In this subsection, the CEA estimates the economic and fiscal benefits if the
United States prioritized the highest-skilled workers within the applicant pool
and allocated 56 percent of green cards to high-skilled applicants, thereby
putting the U.S. in line with the average percentage of employment-based visas
offered by Japan, Australia, and Canada. In this modeling, it is assumed that
378 | Chapter 11

' ррҊфѵ 2!0' -() )/ .$ )/җ Ҙ//0./$) 4-*
'..*!($..$*)Ѷ$.' -спрч
-*'..*!($..$*)

0( -*! .

- )/" *! .

(( $/ - '/$1 .*!ѵѵ$/$5 ).

уцчѶшхр

уу

($'4Ҋ.+*).*- +- ! - ) .

срхѶфхт

сп

(+'*4( )/Ҋ. +- ! - ) .

ртчѶрцр

рт

$1 -.$/4

уфѶтфп

!0" .

рффѶцту

ру

тпѶрцф

.4' .
''o/# -

трѶхфц

*/'

рѶпшхѶхрр

рпп

*0- .ѷѵѵ +-/( )/*! *( ') 0-$/4Ѹ'0'/$*).ѵ

Figure 11-14. Share of Permanent Legal Immigration Based on
Employment, 2017
South Korea
Switzerland
Finland
United States
Germany
Mexico
Spain
Luxembourg
Netherlands
Japan
New Zealand
Ireland
0

Percent
Sources: OECD International Migration Outlook 2019; CEA calculations.
Note: Data are for calendar year 2017. Employment base includes accompanying
family members on work visas. Totals exclude permanent immigration from free
movements.

immigrants receiving new high-skilled green cards would bring their spouse
and dependent children and that these dependents would count against the
total number of green cards. The total amount of immigration would remain
consistent with the current flow.

Policies to Secure Enduring Prosperity | 379

The characteristics of recent immigrants to the United States are identified using data from the three most recent years of the American Community
Survey from the Census Bureau from 2016 through 2018, as provided by the
Integrated Public Use Microdata Series (known as IPUMS) database (Ruggles
et al. 2019).
Figure 11-15 shows the educational characteristics of recent immigrants
under the current system, among those admitted as new high-skilled workers,
and among the anticipated composition resulting from a policy shift. Two
assumptions are made. First, it is assumed that immigrants below the 85th
percentile of wage earners continue to have the same characteristics despite
their percentage of the overall immigration flow being smaller. Second, it is
assumed that new immigrants on merit-based green cards match the characteristics of immigrants whose earnings are above 85th percentile in the
American Community Survey.
Although 51 percent of recent immigrants have less than a bachelor’s
degree, increasing the share of green cards awarded to high-skilled immigrants
suggests that the share with less than a bachelor’s degree would fall to 33 percent (figure 11-15). By allocating a larger share of visas based on employment
and skill, the employment rates of recent immigrants would also increase. Even
including those arriving on non-skills-based visas, along with the spouses of
new employment-based visas who may not be working, the employment rate
of new immigrants rises by 8 percentage points from current rates (from 60
percent to 68 percent) and the average wage of employed recent immigrants
increases from $49,000 to $94,000.
To estimate the effect of increased high-skilled immigration on national
income and national income per capita over the next 10 years, the CEA
approximated the contribution to national income of immigrant workers to
the economy as their total compensation. For the subset of visas converted to
high-skilled green cards, the existing employment and earnings rates are compared with those projected. Among new high-skilled recipients and their families, additional increases in employment rates in future years is not assumed.
Having determined the total wages of recent immigrants under the current system and in the new system, total compensation is estimated based on
the Bureau of Labor Statistics’ Employer Costs for Employee Compensation
survey. The employer share of payroll taxes and fringe benefits (insurance,
retirement benefits, and legally required benefits) represent over 20 percent of
compensation. Thus, dividing wages by 0.8 leads to an estimate of total compensation. The change in total compensation from the introduction of additional high-skilled workers represents our estimate of national income growth
in a single year. To project the contributions of immigrants in future years, it is
assumed that the nominal compensation of new and recent immigrants grows
by 3 percent a year, which is consistent with recent overall wage growth trends
from the Bureau of Labor Statistics’ Current Employment Statistics.
380 | Chapter 11

Figure 11-15. Estimated Educational Attainment of Recent
Immigrants Age 25 Years and Older, 2016–18
High school degree or less

Some college or associate degree

Bachelor’s degree

Advanced degree

Current system

High skilled

New
composition

10
20

34
60

100

Percent
Sources: American Community Survey, 2016–18; Integrated Public Use Microdata Series
database ; CEA calculations.
Note: Estimates assume that the total visas issued remain at current levels. Estimates for
recent immigrants are based on noncitizens who have been in the United States between
one and three years.

Given that shifting toward high-skilled immigrants increases both innovation and the productivity of the existing U.S. workforce and capital, additional increases in national income are expected. While recognizing that there
is substantial uncertainty about the magnitude of these productivity gains, the
same 0.3 percent long-run increase in the productivity of domestic workers
(about 0.03 percent increase a year) is included in the analysis of wage gains
given above. This effect is slightly more than half the long-term productivity
gains estimated by the CBO for the Border Security, Economic Opportunity,
and Immigration Modernization Act.
Using this approach, the total increase in nominal national income in
2029 would be about $570 billion above the baseline (pre-COVID-19) forecast
under the current immigration system. Relative to the baseline national
income growth projections, this reflects an increase in national income growth
of about 0.20 percentage point a year. In addition, using the baseline population forecast, and recognizing that the exercise does not change the total number of immigrants arriving legally in the United States, this increases nominal
national income per capita by about $1,600 in 2029 (table 11-6). If, instead,
these new immigrants have average characteristics that match those of the top
30 percent of recent immigrants, as opposed to the top 15 percent, national
income per capita would still be nearly $1,200 above the baseline forecast. The

Policies to Secure Enduring Prosperity | 381

' ррҊхѵ/$*)' )*( )- . !-*( -$/Ҋ.
(($"-/$*)4./ (

)- . $)рпҊ -1 -" )- . $)спсш
/$*)' )*( -*2/#/$*)' )*( + -
җ+ - )/" +*$)/.Ҙ

-*% / #)"

пѵрф–пѵсп

+$/җспсш*''-.Ҙ
рѶспп–рѶхпп

)- . $)спсш/$*)'
)*( җ$''$*).*!
спсш*''-.Ҙ
утп–фцп

*0- .ѷ( -$)*((0)$/40-1 4Ѷспрх–рчѸ )/ "-/ 0'$. $-*/ -$ ./. Ѹ
'0'/$*).ѵ
*/ .ѷ# . ./# . ./$(/ .*)+- Ҋ Ҋрш)/$*)'$)*( +-*% /$*).ѵ# ./$(/ *!
)/$*)'$)*( + -+$/$.-*0) /*/# ) - ./‫ڦ‬рппѵ""- "/ )/$*)'$)*( ./$(/ .-
-*0) /*/# ) - ./‫ڦ‬рп$''$*)ѵ

average growth of national income per year would be 0.15 percentage point,
and the total increase in nominal national income in 2029 would be $430 billion
above the baseline national income growth projection.

Estimated Effects on the Wages of Domestic Workers
Through this shift to a merit-based immigration system, low- and middleskilled existing U.S. workers would face less competition from substitutable
foreign workers for employment. As of December 2019, there were 42 million
people age 25 and older in the labor force with a high school degree or less,
37 million with some college or an associate degree, and 62 million with a
bachelor’s degree. The shift in the makeup of the immigrant population, along
with the emigration of 2 to 3 percent of recently arrived immigrants each year
(based on estimates from Schwabish 2009), means that after 10 years, there
would be roughly 600,000 fewer permanent adult immigrants in the labor force
with a high school degree or less. There would be 200,000 fewer immigrants
with some college or an associate degree and 1.4 million more immigrants with
at least a bachelor’s degree. These changes represent a decline in the size of
the labor force with less than a high school degree and with some college of
about 1.5 percent and 0.4 percent, respectively, whereas the size of the workforce with at least a bachelor’s degree would increase by about 2.3 percent.
Using de Brauw and Russell’s (2014) updated elasticity from Borjas (2003)
as the upper end of the likely wage elasticity from immigration—and using
Longhi, Nijkamp, and Poot’s (2010) estimate of the lower end of the likely
range—wage effects before capital adjustments and any long-run productivity
gains are estimated. The result is that the revised immigrant flows from this
exercise would increase wages for those with some college by up to 0.1 percent
and for those with a high-school degree or less by up to 0.3 percent. The reform
would reduce the wages of existing U.S. workers with at least a bachelor’s
degree by between 0.1 and 0.5 percent after a decade, indicating that there

382 |

Chapter 11

would be some redistribution from highly educated workers to less educated
workers as a result.
An alternative approach to estimate the change in relative wages
between workers at different levels of education is to consider the effects on
relative wages from a shift in relative supplies of labor. Based on the estimates
by Katz and Murphy (1992) that log relative wages increase by about 0.1 percent for a 1.5 percent decline in relative supply, there would be a 0.15 percent
decline in wages for workers with at least a bachelor’s degree, a 0.1 percent
increase in wages for workers with a high school degree or less, and a 0.03 percent increase in wages for workers with some college or an associate degree.
These estimates are within the range of short-run relative wage changes using
the wage elasticities from the immigration literature.
The CEA anticipates productivity gains once capital has adjusted, resulting in additional wage gains for individuals at all education levels. This is
consistent with the estimates of the CBO (2013) finding that wage gains from
changes in immigrant flows increase once capital adjusts to the number of
workers. In its analysis of the Border Security, Economic Opportunity, and
Immigration Modernization Act, which substantially increased both highskilled and low-skilled immigration, the CBO (2013) projected that the proposal
would increase productivity and result in a long-run increase in wages of 0.5
percentage point. These productivity gains would, consequently, mitigate and
possibly reverse any shorter-run wage declines among high-skilled domestic
workers, while further increasing the wage gains for low- and middle-skilled
domestic workers. The CEA estimates that the productivity gains would be just
over half of those estimated by the CBO for the larger changes in immigration
flows in the 2013 Border Security, Economic Opportunity, and Immigration
Modernization Act, and it is assumed that these productivity gains are distributed equally throughout the distribution, causing overall wages to increase by
0.3 percent.
The CEA expects the overall effect on wages for those with a bachelor’s
degree or above to be between a 0.2 percent decline and a 0.2 percent increase.
Wages of those with a high school degree or less would increase by 0.3 to 0.6
percent, and wages of those with some college would increase by 0.3 to 0.4
percent. For a typical high school graduate working full time, this would result
in an additional $130 to $230 per year of earnings. Furthermore, consistent
with the observations by Moretti (2013) and Borjas (2017) that shifting toward
high-skilled immigration would likely lead to a reduction in income inequality,
this distribution of wage gains would reduce wage inequality among current
U.S. workers.
These estimates are also broadly consistent with the long-run wage
effects found by Chassamboulli and Peri (2018) from a shift in the composition of immigration flows toward high-skilled employment-based immigrants.
In addition to these wage effects, they also find that the shift toward skilled
Policies to Secure Enduring Prosperity | 383

' ррҊцѵцфҊ - /- . )/'0 *!$.' ) !$/.!-*(-*+*.
#)" ./*/# -() )/ "' (($"-/$*)4./ (җ$''$*).*!*''-.Ҙ
-*% / #)"

-' ) !$/.
чуп–рѶтсп

// ) *' ) !$/.
схп–тпп

*/'
рѶпрп–рѶхсп

*0- .ѷ/$*)' (y*!$ ) .Ѷ)"$) -$)"Ѷ) $$) җспрцҘѸ( -$)*((0)$/4
0-1 4Ѷспрх–рчѸ )/ "-/ 0'$. $-*/ -$ ./. Ѹ'0'/$*).ѵ
*/ ѷ./$(/ ...0( /#//# /*/'1$..$..0 - ($)/0-- )/' 1 '.ѵ./$(/ .- -*0) /*
/# ) - ./‫ڦ‬рп$''$*)ѵ

immigration will decrease the unemployment rate for native workers at all skill
levels. This suggests an even more positive effect on the outcomes for U.S. citizens than is seen through wages alone, and one that is especially important in
light of high unemployment rates resulting from the economic consequences
of COVID-19.

Estimated Effects on Government Revenue and Expenditures
The CEA estimates that tax revenues would rise and outlays for social welfare
programs would fall as a result of these wage, productivity, and employment
gains. Using average tax rates (JCT 2019), the proposed changes could increase
tax revenues by $470 billion in the CEA’s primary forecast, and by $340 billion
in the conservative forecast. These estimates do not include potential gains for
domestic worker productivity, and they are broadly consistent with those that
would be derived by multiplying the total increase in national income over the
10-year budget window from the higher-skilled immigrants by the 16 percent
revenue-to-GDP ratio in 2018. However, the actual revenue gains are larger
once the progressivity of the U.S. tax system is accounted for, which means
that the higher-income immigrants would pay a higher share of their income in
taxes than the lower-income immigrants who arrive in the United States under
the current system.
To estimate the long-run fiscal effects of this new composition, the CEA
uses the average 75-year fiscal benefits of immigrants by education level,
as estimated by the National Academy of Sciences (2017). This approach is
similar to that taken by Borjas (2019) to illustrate the substantial increase
in U.S. wealth that would result from increasing the education profile of the
immigrant population. Although changes in the education levels of parents
arriving in the United States will likely also increase the education levels of
children arriving with them, only the long-run fiscal costs and benefits of adult
immigrants age 25 to 64 and those 65+ are included. Based on these estimates,
shifting the composition of permanent immigration toward more highly educated and younger individuals will, after 10 years, result in a net present value
of between $840 billion and $1.3 trillion of fiscal benefits from immigration for
the Federal Government (table 11-7). In addition, it will result in a further $260
billion to $300 billion in fiscal benefits for State and local governments.

384 |

Chapter 11

$"0- ррҊрхѵ(+'*4( )/40/$*)' //$)( )/
Composition (percent)
рпп

чп

хп

# '*-’. "- *-#$"# *( *'' " ҝ ' ../#)# '*-’. "$"#.#**'"-0/
../#)#$"#.#**'

уп
35

сп

ршшс

сппп

спрп

спрш

*0- .ѷ 0- 0*! *-//$./$.Ѹ'0'/$*).ѵ

Improving Postsecondary Education and Skill Development
Postsecondary education and skill development are integral to the health of
the U.S. economy. As shown in figure 11-16, an estimated 68 percent of all
jobs require a postsecondary education, of which 42 percent require at least a
bachelor’s degree and an additional 26 percent require an associate degree or
some form of higher education less than a bachelor’s degree. In comparison, in
1992 only 53 percent of all jobs required a postsecondary education.
Wage premiums and job security often accompany education and skills
attainment; however, the rising cost of college and increases in student loan
balances erode the overall return that accompanies a college degree. Through
the Higher Education Act of 1965, and subsequent reauthorizations since that
time, the Federal Government has taken action to address the costs of higher
education by subsidizing both students and educational institutions. Figure
11-17 shows the growth in the average level of student aid from Federal and
other sources awarded to undergraduate students since 2010. As shown in
figure 11-11 above, this took place during a time of stagnant productivity.
Improved allocation of human capital investments could generate an increase
in productivity and may be hindered due to distorted decisionmaking at higher
education institutions.

Policies to Secure Enduring Prosperity | 385

$"0- ррҊрцѵ1 -" -'Ѷ// Ѷ *'Ѷ )./$/0/$*)',*-
/# -*0- .*!-)/$2- /*) -"-0/ /0 )/.Ѷ
спрп–рч
Dollars (current)
рсѶппп
ррѶфпп
ррѶппп
рпѶфпп
рпѶппп
шѶфпп
шѶппп
чѶфпп
чѶппп

спрп–рр спрр–рс спрс–рт спрт–ру спру–рф спрф–рх спрх–рц спрц–рч

*0- .ѷ/$*)' )/ -!*-0/$*)//$./$.Ѹ'0'/$*).ѵ
*/ ѷ#$.$)'0 .fourҊ4 - +0'$)+-$1/ )*n+-*!$/0)$1 -.$/$ .ѵ

Federal regulatory reforms of higher education could better hold institutions accountable for the economic return that they provide to students, as
well as assist students and families to make more informed decisions regarding their educational options. This section explains how this could be done by
increasing incentives for schools to improve the economic return to students
and by improving Federal support for educational programs that directly help
more Americans secure well-paying jobs. This section also highlights the success of Historically Black Colleges and Universities (HBCUs) and illustrates the
lessons of their experiences for the higher education system as a whole.

Increasing Incentives for Schools to Improve the Economic
Gains of Students
Increased institutional accountability could improve the economic return to
students. Investing in higher education generally provides substantial value for
students and taxpayers. However, when an institution fails to deliver the type
of high-quality education that enables students to repay their Federal student
loans, this institution is not held responsible for losses. Instead, taxpayers are
left to foot the bill. Institutions that lack a focus on generating positive value for
their students exacerbate an increased rate of student loan default and stress
throughout a student’s career.

386 |

Chapter 11

As currently configured, the credit risk associated with student loans is
not efficiently distributed between all parties related to the transaction—that
is, borrowers (students and parents), taxpayers, and the higher education
institutions. The burden of repayment currently rests solely with the borrowers, who may face daunting loan payments if the expected education premium
underdelivers, and with taxpayers, who foot the bill when the borrowers default
or the loan is forgiven. Institutions of higher education bear none of the direct
expenses of such failed outcomes and thus have limited incentive to assist students in optimizing educational skill development and career paths. The U.S.
Department of Education (2020) provides useful institution-level data—such as
annual costs, average earnings, and graduation rates—to help students avoid
making poor investments in education. However, better accountability by the
institutions themselves could further limit failed outcomes.
A reformed system could require postsecondary institutions that accept
taxpayer funds to share in the financial responsibility associated with student
loans. Such a risk-sharing arrangement could require postsecondary institutions to pay a small percentage of the value of the loans on which their former
students have defaulted, or alternatively require institutions with worse repayment outcomes to pay fees. Such fees could be adjusted to account for variation in the composition of student intake so as to align institutions’ interests
with their students and incentivize them to improve repayment outcomes, but
without disproportionately penalizing institutions serving higher-risk students.
There have been three major pieces of Federal legislation pertaining
to risk-sharing on Federal student loans in recent years. A bipartisan bill, the
Student Protection and Success Act, first introduced in 2015, would create
a program where institutions are responsible for paying a percentage of the
cohort nonrepayment balance, 2 percent in the 2019 version of the bill, for
loans that had not paid down at least $1 of principal in three years. The legislation factors in the national unemployment rate and includes a list of exceptions
for loans in deferment and mandatory forbearance. A Republican-sponsored
bill, the PROSPER Act, introduced in 2017, would make changes to provisions
for repayment of Federal student aid if a student withdraws from an institution
of higher education, shifting 90 percent of the repayment responsibility to the
institution (U.S. House of Representatives 2018). A Democratic-sponsored bill,
the Protect Student Borrowers Act, first introduced in 2013, would require covered institutions to make risk-sharing payments on defaulted loans on a sliding
scale based on the default rates of their students.
As shown in figure 11-18, education expenses have grown faster than
inflation, with expenses for academic and institutional support (which includes
expenses associated with noninstructional activities, such as admissions,
student activities, libraries, administrative activities, and executive activities)
growing faster than expenses for instruction and research. This trend may be
driven by instances of mismanagement such as those uncovered in a 2017
Policies to Secure Enduring Prosperity | 387

$"0- ррҊрчѵ-*2/# $)FourҊ -0'$)-$1/
Universities’ 3+ ). .4 ' / / "*-4Ѷршшц–спрч
Index (1997 = 100)

упп
спрч
тфп

($)
$)./$/0/$*)'.0++*-/

тпп
сфп

)./-0/$*))
- . -#

спп
*- Ҋ
рфп

рпп
ршшц ршшш сппр сппт сппф сппц сппш спрр спрт спрф спрц
*0- .ѷ/$*)' )/ - !*-0/$*)//$./$.Ѹ0- 0*! *-//$./$.Ѹ
'0'/$*).ѵ
Note: Core CPI-U = Consumer Price Index for All Urban Consumers.

State audit of the University of California (2017). The audit found that the
university’s Office of the President had spent over $2 million on meeting and
entertainment costs over five years and had awarded salaries and benefits to
personnel far higher than salaries awarded for other comparable positions. As
measured by the 2019–20 AAUP Faculty Compensation Survey, the salary for
the average category I chief academic officer was $383,000, compared with
$160,000 for a full-time professor.8 The quantity of these hires has increased as
well. Administrative hires increased 50 percent faster than classroom instructors between 2001 and 2011.

Improving Support for Educational Programs That Promote
Skill Development
The Federal Government could also improve outcomes for students by better
aligning education with the needs of today’s workforce. The higher education
system has been slow to adapt to the changing nature of work. In recent years,
millions of jobs have remained unfilled, in part due to a lack of Americans with
appropriate skills. Federal policy could better align higher education with the
needs of today’s workforce in multiple ways.
8 This includes institutions that grant 30 or more doctoral-level degrees annually from at least three
distinct programs. See AAUP (2020, n.d.).

388 |

Chapter 11

Box 11-6. Historically Black Colleges and Universities
One example of higher education institutions delivering a high return for their
students is that of Historically Black Colleges and Universities. HBCUs have
played a crucial role in expanding educational opportunity for all students,
especially for the African American students who make up 76 percent of their
student populations. As of 2019, there were 101 accredited HBCUs across
the United States. HBCUs enroll over 300,000 students, including around
80,000 non–African Americans (National Center for Education Statistics 2020).
According to a 2017 economic impact report produced by the United Negro
College Fund, HBCUs generated an employment contribution of 134,090 jobs,
work-life earnings of $130 billion for HBCU students, and a total economic
contribution to the U.S. economy of $14.8 billion.
HBCUs historically have served distinct student populations. HBCU
students are largely low-income, first-generation-college students (nearly
three in five students), and over a quarter of HBCUs are open admission.
Open admission enrollment implies that all qualifying students (students
with a high school degree or general education development certificate) are
welcome to apply and enter the program without additional qualifications
or performance benchmarks. This appeal to low- and moderate-income,
first-generation students suggests that HBCUs have lower barriers to entry
(e.g., costs of attendance, required test scores) than many comparable non-

$"0- ррҊivѵ *(+*.$/$*)*!/0 )/*4*+0'/$*)4
)*( 0$)/$'
HBCUs

Composition (percent)

Non-HBCUs

уп
тф
тп
сф
сп
рф
рп
ф
п
р

т
Income quintile

*0- ѷ/# ).*)Ѷ(4*Ѷ).()җспршҘѵ
Note: HBCUs = Historically Black Colleges and Universities.

Policies to Secure Enduring Prosperity | 389

Figure 11-v. Comparative Rates of Return for a Four-Year Degree
Annual rate of return on degree (percent )

HBCUs

Non-HBCUs

20
15
10
5
0
–5
–10
10

Years
Sources: Integrated Postsecondary Education Data System; CEA calculations.
Note: Based on the most recent institution-level data as of June 1, 2020. HBCUs =
Historically Black Colleges and Universities.

HBCU institutions. HBCUs draw 24 percent of their student body’s population
from the lowest 20 percent of incomes. By this measure, HBCUs serve more
economically disadvantaged populations than non-HBCU institutions, which
are composed of only 8 percent of students in the bottom 20 percent.
Although HBCUs account for a mere 10 percent of the African American
college student population, in 2014 they represented 17 percent of bachelor’s
degrees and 24 percent of STEM (science, technology, engineering, and
mathematics) degrees earned by African Americans. From 2002 to 2011, the
top eight institutions where African Americans earned PhDs in science and
engineering were HBCUs.
Here, the CEA estimates the rates of return on an education from the
HBCU system. The estimates use traditional approaches and are in keeping
with the work of Mincer (1958), Schultz (1961), and Becker (1962) on differences in earnings across persons resulting from levels of human capital,
accumulated primarily through education and training. Using institutionallevel data obtained from the U.S. Department of Education’s (2020) College
Scorecard and the Federal Reserve’s 2019 Survey of Consumer Finance, the
CEA estimates the comparative rates of return over 40 years for graduates
of four-year HBCUs and comparable non-HBCU institutions (figure 11-iv).
Comparable institutions are located within the same commuting zone of at

390 | Chapter 11

least one of these HBCUs and are of similar institutional selectivity, according
to the Barron’s Selectivity Index. The long-term rates of return for graduates
receiving a college education at an HBCU are significantly positive and track
those of graduates from a non-HBCU school. Short-term rates of return for
students of HBCUs and non-HBCUs are significantly negative and vary largely
to the extent that forgone income for non-HBCU graduates tends to be larger
than it is for HBCU graduates. However, as time passes, both cohorts experience income growth and thus see an increase in their rates of return.
Although non-HBCU graduates initially benefit, on average, from higher
incomes than do HBCU students, HBCU graduates tend on average to experience greater annual growth in income than non-HBCU graduates (figure
11-v). Thus, over the long run, alumni of HBCUs will tend to experience rates
of return comparable to those for non-HBCU alumni. This shows that from a
productivity standpoint, HBCUs can deliver comparable returns at a lower
cost. HBCUs have a slightly lower level of earnings, which is attributable to the
different student composition (e.g., the presence of first-generation students,
and the selection of college majors). Figure 11-vi shows that graduates of
HBCUs also track closely with graduates of non-HBCUs in cumulative earnings
over time.

Figure 11-vi. Comparison of Real Cumulative Median Net Earnings
Dollars (thousands)

HBCUs

Non-HBCUs

1,800
1,600
1,400
1,200
1,000
800
600
400
200
0
–200

Years

Sources: Integrated Postsecondary Education Data System; CEA calculations.
Note: Based on the most recent institution-level data as of June 1, 2020. HBCUs =
Historically Black Colleges and Universities.

Policies to Secure Enduring Prosperity | 391

One approach would be for Congress to expand Pell Grant eligibility to
include high-quality, short-term programs that provide students with a credential, certification, or license in a high-demand field and that demonstrate
strong employment and earnings outcomes. Pell Grants are typically used to
support students in traditional two- or four-year degree programs. Though
some certificate programs are eligible for Pell Grants, programs must cover
at least 15 weeks of instruction. Expanding support to shorter-term programs
designed to teach skills specific to well-paying jobs could better meet the
needs of students with near-term employment goals.
Federal program requirements could also encourage, rather than limit,
partnerships between higher education providers and employers. Employers
are most aware of the skills needed to succeed in the workplace. Congress
could reform the Federal Work Study program to support workforce- and
career-oriented opportunities for low-income undergraduate students. Workbased learning improves students’ chances of developing important workplace
skills, yet the Federal Work Study rules favor campus-based jobs.
Building on the successes of the National Council for the American Worker
(NCAW) could promote multiple pathways to career success. Since January
2017, the NCAW has enrolled more than 750,000 apprentices; has modernized
the Perkins Career and Technical Education Act to increase dual enrollment,
work-based learning, and employer engagement; and has encouraged more
than 400 companies to commit to providing 16 million employees with the
training, reskilling, and career opportunities needed to increase productivity.
Improving the four-year degree to generate greater skill increases for
students, as well as providing alternative paths for human capital accumulation, can avoid a one-size-fits-all approach that leaves individuals and groups
behind. Apprenticeships, training programs, and four-year degrees are all
paths to a more productive workforce and a higher quality of life for millions
of Americans. Historically Black Colleges and Universities (HBCU) demonstrate
how large gains from a high-quality education can be successfully provided to
underserved groups (see box 11-6).

Conclusion
In this final chapter of the 2021 Economic Report of the President, the Council
has identified various challenges for the American economy that not only were
exacerbated by the pandemic, but also extend beyond the COVID-19 crisis
into the postpandemic future. The policy ideas discussed here to potentially
address these challenges can lead to a more resilient and prosperous economy
for all Americans.
We have examined workers’ connections to the labor force—relationships that COVID-19 strained and in millions of instances severed during the
economic crisis—and how provisions of the tax code may discourage and

392 |

Chapter 11

disrupt these connections and their reconstitution. Reforming the provisions
that disproportionally place the tax burden on second-earners and low-income
earners could rebuild employees’ relationships with their employers and therefore strengthen the current economic recovery.
In addition, we have discussed the importance of paid leave and childcare, not only in the context of the global health crisis, but also as a way to
support a stronger workforce after the pandemic abates. These provisions
have become especially relevant during the COVID-19 crisis as sick family
members have required care at home and children have attended virtual
school from home, creating barriers for working parents to return to their jobs.
Although the Federal Government has passed temporary measures to mitigate
the lack of access to paid leave and childcare during this crisis, this chapter
demonstrates that permanent policies to provide such access would increase
labor force participation and earnings beyond the current pandemic recession.
Furthermore, we have analyzed the effects of negotiating reciprocal trade
agreements on consumers’ and manufacturers’ access to the international
market and how deepening trade integration with like-minded U.S. allies can
allow for future flexibility in negotiations and can assist the Nation in achieving its trade goals. In working to preserve America’s economic interests on the
world stage, the economy can achieve real gains from trade while protecting
the economic security of American enterprise.
We have illustrated the strain COVID-19 has placed on the U.S. healthcare
system in this Report. Historic financial relief from the Federal Government
alleviated the worst of the crisis for hospitals and patients. Nevertheless,
the COVID-19 pandemic has revealed persisting issues within the healthcare
system that create distortions in the healthcare market. This chapter has
examined how to increase the supply of healthcare and remove opaque pricing
structures, which will provide patients and doctors alike with benefits.
We have also highlighted the Federal Government’s role in participating in the creation and maintenance of a world-class infrastructure system.
Fostering public-private partnerships to lower taxpayer costs, targeting funds
to high-productivity areas such as ports, and reforming the country’s evermore-important digital infrastructure have all been explored in this chapter.
Investing in infrastructure projects would not only increase the productivity of
the American economy in and of itself, but would also have spillover benefits
for all sectors of the economy.
Finally, we have showed that the American economy could support a
more resilient workforce by shifting the U.S. immigration system toward a
merit-based system for higher-skilled immigrants and by realigning the goals
of higher education institutions to better equip students seeking nontraditional career paths. This would lead to increases in economic growth, wages,
and tax revenue, and thus to prosperity for all Americans.

Policies to Secure Enduring Prosperity | 393

The policy reforms discussed in this chapter are designed to support the
American economy and the American people long after the COVID-19 pandemic
subsides. These policies would boost productivity for manufacturers, increase
investment in workers, enhance labor force participation, and grow families’
earnings. In accordance with its mandate to “recommend national economic
policy to promote employment, production, and purchasing power under free
competitive enterprise,” the Council of Economic Advisers has used this chapter to analyze reforms that could provide substantial economic benefits for
Americans in every walk of life. Solving the issues and challenges articulated in
this chapter would aid in restoring the American economy to its prepandemic
levels of prosperity and would offer a solid foundation upon which to build an
even greater and more resilient economy for all Americans.

394 |

Chapter 11

References
Chapter 1
Angrist, J., and A. Krueger. 1992, Estimating the Payoff to Schooling Using the
Vietnam-Era Draft Lottery. NBER Working Paper 4067. Cambridge, MA:
National Bureau of Economic Research.
Baqaee, D., and E. Farhi. 2020. Nonlinear Production Networks with an Application to the
COVID-19 Crisis. CEPR Discussion Paper DP14742. London: Centre for
Economic Policy Research.
BEA (Bureau of Economic Analysis). 2020. “Gross Domestic Product (Third Estimate),
Corporate Profits (Revised), and GDP by Industry, Second Quarter 2020.”
https://www.bea.gov/news/2020/
gross-domestic-product-third-estimate-corporate-profits-revised-and-gdpindustry-annual.
Bernanke, B., T. Geithner, and H. Paulson Jr., with J. Liang. 2020. First Responders:
Inside the U.S. Strategy for Fighting the 2007–2009 Global Financial Crisis. New
Haven, CT: Yale University Press.
Bhuller, M., M. Mogstad, and K. Salvanes. 2017. “Life-Cycle Earnings, Education
Premiums, and Internal Rates of Return.” Journal of Labor Economics 35, no 4:
993–1030.
Bhutta, N., J. Bricker, L. Dettling, J. Kelliher, and S. Laufer. 2019. Stress Testing
Household Debt. Finance and Economics Discussion Series Working Paper
2019-008. Washington: Board of Governors of the Federal Reserve System.
BLS (Bureau of Labor Statistics). 2014. “Public Workforce Programs during the Great
Recession.” https://www.bls.gov/opub/mlr/2014/article/public-workforceprograms-during-the-greatrecession.htm.
Burns, A., D. van der Mensbrugghe, and H. Timmer. 2006. “Evaluating the Economic
Consequences of Avian Influenza.” World Bank. https://web.worldbank.org/
archive/website01003/WEB/IMAGES/EVALUATI.PDF.
CDC (Centers for Disease Control and Prevention). 2020a. “First Travel-Related Case of
2019 Novel Coronavirus Detected in United States.” https://www.cdc.gov/
media/releases/2020/p0121-novel-coronavirus-travel-case.html.
———. 2020b. “CDC Confirms Possible Instance of Community Spread of COVID-19 in
U.S.” https://www.cdc.gov/media/releases/2020/s0226-Covid-19-spread.html.
———. 2020c. “Nonpharmaceutical Interventions (NPIs).” https://www.cdc.gov/
nonpharmaceutical-interventions/index.html.

395

CEA (Council of Economic Advisers). 2019. Economic Report of the President.
Washington: U.S. Government Publishing Office.
———. 2020a. Economic Report of the President. Washington: U.S. Government
Publishing Office.
———. 2020b. “Evaluating the Effect of the Economic Response to COVID-19.” https://
www.whitehouse.gov/wp-content/uploads/2020/08/Evaluating-the-Effectsof-the-Economic-Response-to-COVID-19.pdf.
Cronin, C., and W. Evans. 2020. Private Precaution and Public Restrictions: What Drives
Social Distancing and Industry Foot Traffic in the COVID-19 Era? NBER Working
Paper 27531. Cambridge, MA: National Bureau of Economic Research.
Crouzet, N., and F. Gourio. 2020. “Financial Positions of U.S. Public Corporations: Part
1, Before the Pandemic.” Federal Reserve Bank of Chicago.
Davis, M., W. Larson, S. Oliner, and B. Smith. 2019. A Quarter Century of Mortgage Risk.
FHFA Working Paper 19-02. Washington: Federal Housing Finance Agency.
Dharmasankar, S., and B. Mazumder. 2016. “Have Borrowers Recovered from
Foreclosures during the Great Recession?” Chicago Fed Letter 370: 1.
El-Sibaie, A., E. York, G. Watson, and T. LaJoie. 2020. “FAQ on Federal Coronavirus Relief
Law (CARES Act).” Tax Foundation. https://taxfoundation.org/federalcoronavirus-relief-bill-caresact/#:~:text=Individuals%20with%20a%20
Social%20Security,)%2C%20or%20%24112%2C500 %20(heads%20of.
Emsellem, M., and M. Evermore. 2020. “Understanding the Unemployment Provisions
of the Families First Coronavirus Response Act.” National Employment Law
Project. https://www.nelp.org/publication/
understanding-the-unemployment-provisions-of-thefamilies-firstcoronavirus-response-act/.
Federal Reserve. 2020. “Financial Stability Report, May 2020.” https://www.
federalreserve.gov/publications/2020-may-financial-stability-report-purpose.
htm.
Goger A., T. Loh, and C. George. 2020. “Unemployment Insurance Is Failing Workers
during COVID-19. Here’s How to Strengthen it.” Brookings Institution. https://
www.brookings.edu/research/
unemployment-insurance-is-failing-workers-duringCOVID-19-heres-how-tostrengthen-it/.
Goolsbee, A., and C. Syverson. 2020. Fear, Lockdown, and Diversion: Comparing Drivers
of Pandemic Economic Decline 2020. NBER Working Paper 27432. Cambridge,
MA: National Bureau of Economic Research.
IMF (International Monetary Fund). 2019. Global Financial Stability Report. Washington:
IMF. https://www.imf.org/en/Publications/GFSR/Issues/2019/03/27/
Global-Financial-StabilityReport-April-2019.
———. 2020. Global Financial Stability Report. Washington: IMF. https://www.imf.org/
en/Publications/GFSR/Issues/2020/04/14/
Global-Financial-StabilityReport-April-2020-49020.

396 |

References

Jonas, O. 2013. “Pandemic Risk.” Background paper, World Bank. https://www.
worldbank.org/content/dam/Worldbank/document/HDN/Health/WDR14_bp_
Pa ndemic_Risk_Jonas.pdf.
Kilbourne, E. 2006. “Influenza Pandemics of the 20th Century.” Emerging Infectious
Diseases 12, no. 1: 9–14.
Krueger, A. 1992. Government Failures in Development. NBER Working Paper 3340.
Cambridge, MA: National Bureau of Economic Research.
LCD News. 2020. “U.S. Leveraged Loan Defaults Total $23B in Q2, the Most Since 2009.”
S&P Global, July 6.. https://www.spglobal.com/marketintelligence/en/newsinsights/latest-newsheadlines/
us-leveraged-loan-defaults-total-23b-in-q2-the-most-since-2009-59301227.
Marr, C., S. Jacoby, C. Huang, S. Hingtgen, A. Sherman, and J. Beltrán. 2020. “Future
Stimulus Should Include Immigrants and Dependents Previously Left Out,
Mandate Automatic Payments.” Center on Budget and Policy Priorities.
https://www.cbpp.org/research/economy/
future-stimulus-should-include-immigrants-anddependents-previouslyleftout.
McKibbin, W. 2009. “The Swine Flu Outbreak and Its Global Economic Impact.” https://
www.brookings.edu/on-the-record/
the-swine-flu-outbreak-and-its-global-economicimpact/.
McKibbin, W., and A. Sidorenko. 2006. “Global Macroeconomic Consequences of
Pandemic Influenza.” Lowy Institute for International Policy. https://www.
lowyinstitute.org/sites/default/files/pubfiles/McKibbin_Sidorenko%2C_
Global_ macroeconomic_1.pdf.
Nunn, R., and J. Shambaugh. 2020. “Whose Wages Are Rising and Why?” Brookings
Institution. https://www.brookings.edu/policy2020/votervital/
whose-wages-are-rising-and-why/.
Patel, A., D. Jernigan, and 2019-nCov CDC Response Team. 2020. “Initial Public Health
Response and Interim Clinical Guidance for the 2019 Novel Coronavirus
Outbreak—United States, December 31, 2019–February 4, 2020.” Morbidity
and Mortality Weekly Report 69, no. 5: 140–46.
Robertson, J. 2019. “Faster Wage Growth for the Lowest-Paid Workers.” Federal Reserve
Bank of Atlanta. https://www.frbatlanta.org/blogs/macroblog/2019/12/16/
faster-wage-growth-for-the-lowest-paid-workers.
Tedeschi, E. 2020. “Pay Is Rising Fastest for Low Earners. One Reason? Minimum
Wages.” New York Times, January 1. https://www.nytimes.com/2020/01/03/
upshot/minimum-wage-boost-bottom-earners.html.
Tung, I. 2020. “Minimum Wage Increases Reverse Post-Recession Wage Declines for
Workers in Lowest-Paid Jobs.” National Employment Law Project. https://
www.nelp.org/publication/
minimum-wage-increases-reverse-post-recession-wage-declines-workerslowest-paid-jobs/.

References | 397

Van Dam, A., and R. Siegel. 2020. “Minimum Wage Increases Fueling Faster Wage
Growth for Those at the Bottom.” Washington Post, January 2. https://www.
washingtonpost.com/business/2020/01/02 /
minimum-wage-increases-fueling-faster-wage-growth-those-bottom/.
Verikios, G., M. Sullivan, P. Stojanovski, J. Giesecke, and G. Woo. 2011. “The Global
Economic Effects of Pandemic Influenza.” https://static.rms.com/email/
documents/liferisks/papers/the-globaleconomic-effects-of-pandemicinfluenza.pdf.
White House. 2020a. “Proclamation on Suspension of Entry as Immigrants and
Nonimmigrants of Persons who Pose a Risk of Transmitting 2019 Novel
Coronavirus.” https://www.whitehouse.gov/presidential-actions/
proclamation-suspension-entry-immigrants-nonimmigrants-persons-poserisk-transmitting-2019-novel-coronavirus/.
———. 2020b. “Proclamation: Suspension of Entry as Immigrants and Nonimmigrants
of Certain Additional Persons Who Pose a Risk of Transmitting 2019 Novel
Coronavirus.” https://www.whitehouse.gov/presidential-actions/
proclamation-suspension-entry-immigrants-nonimmigrants-certainadditional-persons-pose-risk-transmitting-2019-novel-coronavirus/.
———. 2020c. “Proclamation on Declaring a National Emergency Concerning the Novel
Coronavirus Disease (COVID-19) Outbreak.” https://www.whitehouse.gov/
presidential-actions/
proclamation-declaring-national-emergency-concerning-novel-coronavirusdisease-covid-19-outbreak/.

Chapter 2
Altonji, J., Z. Contractor, L. Finamor, R. Haygood, I. Lindenlaub, C. Meghir, C. O’Dea, D.
Scott, L. Wang, and E. Washington. 2020. “Employment Effects of
Unemployment Insurance Generosity during the Pandemic.” Manuscript, Yale
University, July 14.
Autor, D., D. Cho, L. Crane, M. Goldar, B. Lutz, J. Montes, W. Peterman, D. Ratner, D.
Villar, and A. Yildirmaz. 2020. “An Evaluation of the Paycheck Protection
Program Using Administrative Payroll Microdata.” Working paper,
Massachusetts Institute of Technology.
Baker, S., R. Farrokhnia, S. Meyer, M. Pagel, and C. Yannelis. 2020. Income, Liquidity, and
the Consumption Response to the 2020 Economic Stimulus Payments. NBER
Working Paper 27097. Cambridge, MA: National Bureau of Economic
Research.
Bartik, A., M. Bertrand, F. Lin, J. Rothstein, and M. Unrath. 2020. Measuring the Labor
Market at the Onset of the COVID-19 Crisis. NBER Working Paper 27613.
Cambridge, MA: National Bureau of Economic Research.
CBO (Congressional Budget Office). 2020. “Interim Economic Projections for 2020 and
2021.” https://www.cbo.gov/publication/56368#_idTextAnchor027.

398 |

References

CEA (Council of Economic Advisers). 2020. “Evaluating the Effects of the Economic
Response to COVID-19.” August. https://www.whitehouse.gov/wp-content/
uploads/2020/08/Evaluating-the-Effects-of-the-Economic-Response-toCOVID-19.pdf.
Chetty R., J. Friedman, N. Hendren, M. Stepner, and Opportunity Insights Team. 2020.
How Did COVID-19 and Stabilization Policies Affect Spending and Employment?
A New Real-Time Economic Tracker Based on Private Sector Data. NBER
Working Paper 27431. Cambridge, MA: National Bureau of Economic
Research.
Federal Reserve Board of Governors. 2020a. “Net Worth by All Families.” Survey of
Consumer Finances, 1989–2019. https://www.federalreserve.gov/econres/scf/
dataviz/scf/chart/#series:Net_Worth;demographic:all;population:all;units:me
dian.
―――. 2020b. “Net Worth by Percentile of Net Worth.” Survey of Consumer Finances,
1989–2019. https://www.federalreserve.gov/econres/scf/dataviz/scf/
chart/#series:Net_Worth;demographic:nwcat;population:1,2,3,4,5;units:
mean.
―――. 2020c. “Net Worth by Race or Ethnicity.” Survey of Consumer Finances, 1989–
2019. https://www.federalreserve.gov/econres/scf/dataviz/scf/
chart/#series:Net_Worth;demographic:racecl4;population:all;units:median.
―――. 2020d. “Primary Residence by All Families.” Survey of Consumer Finances,
1989–2019. https://www.federalreserve.gov/econres/scf/dataviz/scf/
chart/#series:Primary_Residence;demographic:all;population:1;units:have.
Forsythe, E., L. Kahn, F. Lange, and D. Wiczer. 2020. Labor Demand in the Time of COVID19: Evidence from Vacancy Postings and UI Claims. NBER Working Paper
27061. Cambridge, MA: National Bureau of Economic Research.
Fox, Z., B. Yang, A. Sikander, and B. Scheid. 2020. “As Virus Crisis Persists, PPP
recipients Lay Off Thousands.” S&P Global Market Intelligence, July 30.
https://www.spglobal.com/marketintelligence/en/news-insights/latest-newsheadlines/as-virus-crisis-persists-ppp-recipients-lay-off-thousands-59602815.
Ganong, P., P. Noel, and J. Vavra. 2020. U.S. Unemployment Insurance Replacement
Rates During the Pandemic. NBER Working Paper 27216. Cambridge, MA:
National Bureau of Economic Research.
Goodspeed, T., and P. Navarro. 2020. “The White House Favors a Bridge to Recovery.”
Wall Street Journal, December 2. https://www.wsj.com/articles/
the-white-house-favors-a-bridge-to-recovery-11606950989.
Granja, J., C. Makridis, C. Yannelis, and E. Zwick. 2020. Did the Paycheck Protection
Program Hit the Target? NBER Working Paper 27095. Cambridge, MA: National
Bureau of Economic Research.
Green, D. and E. Loualiche. 2020. State and Local Government Employment in the
COVID-19 Crisis. HBS Working Paper 21-023. Cambridge, MA: Harvard Business
School. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3651605.

References | 399

Han, J., B. Meyer, and J. Sullivan. 2020. Income and Poverty in the COVID-19 Pandemic.
NBER Working Paper 27729. Cambridge, MA: National Bureau of Economic
Research. https://www.nber.org/papers/w27729.
Hepburn, P., R. Louis, and M. Desmond. 2020. “Eviction Tracking System: Version 1.0.”
Princeton University. www.evictionlab.org.
JCT (Joint Committee on Taxation). 2020. “Description of the Tax Provisions of Public
Law 116-136, the Coronavirus Aid, Relief, and Economic Security (“CARES”)
Act.” JCX-12R-20, April 23. https://www.jct.gov/publications/2020/jcx-12r-20/.
Marinescu, I., D. Skandalis, and D. Zhao. 2020. “Job Search, Job Posting and
Unemployment Insurance During the COVID-19 Crisis.” Working paper.
Meyer, B., and J. Sullivan. 2020. “Percent Below Federal Poverty Line or Multiple of the
Federal Poverty Line, Basic Monthly CPS, January 2019 to Date.” Near Real
Time COVID-19 Income and Poverty Dashboard, Poverty Measurement. http://
povertymeasurement.org/covid-19-poverty-dashboard/.
National Multifamily Housing Council. 2020. “NMHC Rent Payment Tracker Finds 79.4
Percent of Apartment Households Paid Rent as of October 6.” https://www.
nmhc.org/research-insight/nmhc-rent-payment-tracker/.
Parolin, Z., M. Curran, and C. Wimer. 2020. “The CARES Act and Poverty in the COVID-19
Crisis.” Poverty and Social Policy Brief 4, no. 8.
Ruhle, S., L. Miranda, and M. Capetta. 2020. “PPP Likely Saved 35 Million Jobs, Says
JPMorgan Chase CEO Jamie Dimon.” NBC News, August 11. https://www.
nbcnews.com/business/economy/
ppp-likely-saved-35-million-jobs-says-jpmorgan-chase-ceo-n1236341.
White House. 2020a. “Executive Order on Fighting the Spread of COVID-19 by Providing
Assistance to Renters and Homeowners.” https://www.whitehouse.gov/
presidential-actions/
executive-order-fighting-spread-covid-19-providing-assistance-rentershomeowners/.
―――. 2020b. “Memorandum on Authorizing the Other Needs Assistance Program for
Major Disaster Declarations Related to Coronavirus Disease 2019.” https://
www.whitehouse.gov/presidential-actions/
memorandum-authorizing-needs-assistance-program-major-disasterdeclarations-related-coronavirus-disease-2019/.
―――. 2020c. “Memorandum on Continued Student Loan Payment Relief During the
COVID-19 Pandemic.” https://www.whitehouse.gov/presidential-actions/
memorandum-continued-student-loan-payment-relief-covid-19-pandemic/.
―――. 2020d. “Memorandum on Deferring Payroll Tax Obligations in Light of the
Ongoing COVID-19 Disaster.” https://www.whitehouse.gov/presidentialactions/
memorandum-deferring-payroll-tax-obligations-light-ongoing-covid-19disaster/.

400 | References

Chapter 3
Autor, D., D. Cho, L. Crane, M. Goldar, B. Lutz, J. Montes, W. Peterman, D. Ratner, D.
Villar, and A. Yildirmaz. 2020. “An Evaluation of the Paycheck Protection
Program Using Administrative Payroll Microdata.” Working paper,
Massachusetts Institute of Technology.
Baker, S., R. Farrokhnia, S. Meyer, M. Pagel, and C. Yannelis. 2020. Income Liquidity, and
the Consumption Response to the 2020 Economic Stimulus Payments. NBER
Working Paper 29097. Cambridge, MA: National Bureau of Economic
Research.
Bartik, A., Z. Cullen, E. Glaeser, M. Luca, C. Stanton, and A. Sunderam. 2020. The
Targeting and Impact of Paycheck Protection Program Loans to Small
Businesses. NBER Working Paper 27623. Cambridge, MA: National Bureau of
Economic Research.
Chetty, R., J. Friedman, N. Hendren, and M. Stepner. 2020. How Did COVID-19 and
Stabilization Policies Affect Spending and Employment? A New Real-Time
Economic Tracker Based on Private Sector Data. Working Paper 2020-05.
Cambridge, MA: Opportunity Insights.
Cox, J., D. Greenwald, and S. Ludvigson. 2020. What Explains the COVID-19 Stock
Market? NBER Working Paper 27784. Cambridge, MA: National Bureau of
Economic Research.
Crane, L., R. Decker, A. Flaaen, A. Hamins-Puertolas, and C. Kurz. 2020. Business Exit
during the COVID-19 Pandemic: Non-Traditional Measures in Historical Context.
Finance and Economics Discussion Series Working Paper 2020-089.
Washington: Board of Governors of the Federal Reserve System.
Elenev, V., T. Landvoigt, and S. Van Nieuwerburgh. 2020. Can the Covid Bailouts Save
the Economy? NBER Working Paper 27207. Cambridge, MA: National Bureau
of Economic Research.
Federal Reserve. 2020. “Financial Stability Report—May 2020.” Washington: Board of
Governors of the Federal Reserve System. https://www.federalreserve.gov/
publications/files/financial-stability-report-20200515.pdf.
Fox, Z., B. Yung, A. Sikander, and B. Scheid. 2020. “As Virus Crisis Persists, PPP
Recipients Lay Off Thousands.” S&P Global Market Intelligence, July 30.
https://www.spglobal.com/marketintelligence/en/news-insights/latest-newsheadlines/as-virus-crisis-persists-ppp-recipients-lay-off-thousands-59602815.
Granja, J, C. Makridis, C. Yannelis, and E. Zwick. 2020. Did the Paycheck Protection
Program Hit the Target? NBER Working Paper 27095. Cambridge, MA: National
Bureau of Economic Research.
Gilchrist, S., B. Wei, V. Yue, and E. Zakrajsek. 2020. The Fed Takes on Corporate Credit
Risk: An Analysis of the Efficacy of the SMCCF. NBER Working Paper 27809.
Cambridge, MA: National Bureau of Economic Research.

References | 401

Haddad, V., A. Moreira, and T. Muir. 2020. When Selling Becomes Viral: Disruptions in
Debt Markets in the COVID-19 Crisis and the Fed’s Response. NBER Working
Paper 27168. Cambridge, MA: National Bureau of Economic Research.
Hamilton, S. 2020. “From Survival to Revival: How to Help Small Businesses through
the COVID-19 Crisis.” Hamilton Project. https://www.hamiltonproject.org/
assets/files/PP_Hamilton_Final.pdf.
Keshner, A. 2020. “Coronavirus Has Rocked America’s Economy—and It’s Had a
Surprising Effect on Bankruptcy Filings.” MarketWatch, May 6. https://www.
marketwatch.com/story/
the-coronavirus-has-rocked-americas-economy-but-its-had-a-surprisingeffect-on-bankruptcy-filings-2020-05-05.
Li, L., P. Strahan, and S. Zhang. 2020. Banks as Lenders of First Resort: Evidence from the
COVID-19 Crisis. NBER Working Paper 27256. Cambridge, MA: National Bureau
of Economic Research.
Ruhle, S., L. Miranda, and M. Cappetta. 2020. “PPP Likely Saved 35 Million Jobs, Says
JPMorgan Chase CEO Jamie Dimon.” NBC News, August 11. https://www.
nbcnews.com/business/economy/
ppp-likely-saved-35-million-jobs-says-jpmorgan-chase-ceo-n1236341.
Tett, G. 2020. “Swamped Bankruptcy Courts Threaten U.S. Recovery.” Financial Times,
May 14. https://www.ft.com/
content/14b07c0e-95e3-11ea-af4b-499244625ac4.

Chapter 4
André, F. 2002. “How the Research-Based Industry Approaches Vaccine Development
and Establishes Priorities.” Developmental Biology 110: 25–29.
Berndt, E., I. Cockburn, and Z. Griliches. 1996. “Pharmaceutical Innovations and Market
Dynamics: Tracking Effects on Price Indices for Antidepressant Drugs.”
Brookings Papers on Economic Activity, Microeconomics, no. 2: 133–99. https://
www.brookings.edu/wp-content/uploads/1996/01/1996_bpeamicro_berndt.
pdf.
Bestsennyy, O., G. Gilbert, A. Harris, and J. Rost. 2020. “Telehealth: A Quarter-TrillionDollar Post COVID-19 Reality?” McKinsey & Company. https://www.mckinsey.
com/industries/healthcare-systems-and-services/our-insights/
telehealth-a-quarter-trillion-dollar-post-covid-19-reality.
Bynum, A., C. Irwin, C. Cranford, and G. Denny. 2003. “The Impact of Telemedicine on
Patients’ Cost Savings: Some Preliminary Findings.” Telemedicine Journal and
e-Health 9 no. 4: 361–67. https://doi.org/10.1089/153056203772744680.
Caves, R., M. Whinston, and M. Hurwitz. 1991. “Patent Expiration, Entry, and
Competition in the U.S. Pharmaceutical Industry.” Brookings Papers on
Economic Activity, Microeconomics, no. 1: 1–66. https://www.brookings.edu/
wp-content/uploads/1991/01/1991_bpeamicro_caves.pdf.

402 |

References

CBO (Congressional Budget Office). 2006. “Research and Development in the
Pharmaceutical Industry.” https://www.cbo.gov/sites/default/files/109thcongress-2005-2006/reports/10-02-drugr-d.pdf.
———. 2020. “Federal Subsidies for Health Insurance Coverage for People Under 65:
2020 to 2030.” https://www.cbo.gov/publication/56650.
CEA (Council of Economic Advisers). 2019. “Mitigating the Impact of a Pandemic
Influenza through Vaccine Innovation.” https://www.whitehouse.gov/
wp-content/uploads/2019/09/Mitigating-the-Impact-of-Pandemic-Influenzathrough-Vaccine-Innovation.pdf.
Chauhan, V., S. Galwankar, B. Arquilla, M. Garg, S. Di Somma, A. El-Menyar, V. Krishnan,
J. Gerber, R. Holland, and S. Stawicki. 2020. “Novel Coronavirus (COVID-19):
Leveraging Telemedicine to Optimize Care While Minimizing Exposures and
Viral Transmission.” Journal of Emergencies, Trauma, and Shock 13, no. 1:
20–24. https://doi.org/10.4103/JETS.JETS_32_20.
DiMasi, J., H. Grabowski, and R. Hansen. 2016. “Innovation in the Pharmaceutical
industry: New Estimates of R&D Costs.” Journal of Health Economics 47:
20–33. https://www.sciencedirect.com/science/article/abs/pii/
S0167629616000291.
FDA (U.S. Food and Drug Administration). 2020. “In Vitro Diagnostic EUAs.” https://
www.fda.gov/medical-devices/coronavirus-disease-2019-covid-19emergency-use-authorizations-medical-devices/vitro-diagnostics-euas.
Garfield, R., G. Claxton, A. Damico, and L. Levitt. 2020. “Eligibility for ACA Health
Coverage Following Job Loss.” Kaiser Family Foundation. https://www.kff.
org/coronavirus-covid-19/issue-brief/
eligibility-for-aca-health-coverage-following-job-loss/.
Grabowski, H., and J. Vernon. 1992. “Brand Loyalty, Entry, and Price Competition in
Pharmaceuticals after the 1984 Drug Act.” Journal of Law and Economics 32:
331–50. https://www.jstor.org/stable/725543.
Grady, C., S. Shah, F. Miller, M. Danis, M. Nicolini, J. Ochoa, Holly T., Dave W., and A. Rid.
2020. “So Much at Stake: Ethical Tradeoffs in Accelerating SARSCoV-2 Vaccine
Development.” Vaccine 8, no. 41: 6381–87. https://doi.org/10.1016/j.
vaccine.2020.08.017.
Ivanov, A. 2013. “Barriers to the Introduction of New Medical Diagnostic Tests.”
Laboratory Medicine 44, no. 4: e132–e136. https://doi.org/10.1309/
LMMHGYKY7LIUEEQ6.
Kennedy, J. 2018. “How to Ensure That America’s Life-Sciences Sector Remains
Globally Competitive.” Information Technology and Innovation Foundation.
https://itif.org/publications/2018/03/26/
how-ensure-americas-life-sciences-sector-remains-globally-competitive.
Kleiner, M., A. Marier, K. Park, and C. Wing. 2016. “Relaxing Occupational Licensing
Requirements: Analyzing Wages and Prices for a Medical Service.” Journal of
Law and Economics 59, no. 2. https://doi.org/10.1086/688093.

References | 403

Lenz, E., M. Mundinger, R. Kane, S. Hopkins, and S. Lin. 2004. “Primary Care Outcomes
in Patients Treated by Nurse Practitioners or Physicians: Two Year Follow Up.”
Medical Care Research and Review 61, no. 3: 332–51. https://journals.sagepub.
com/doi/abs/10.1177/1077558704266821.
Lichtenberg, F. 2014. “The Impact of Pharmaceutical Innovation on Disability Days and
the Use of Medical Services in the United States, 1997–2010.” Journal of
Human Capital 8, no. 4: 432–80. https://www.jstor.org/stable/10.1086/679110.
Mann, D., J. Chen, R. Chunara, P. Testa, and O. Nov. 2020. “COVID-19 Transforms Health
Care through Telemedicine: Evidence from the Field.” Journal of the
American Medical Informatics Association 27, no. 7: 1132–35. https://doi.
org/10.1093/jamia/ocaa072.
Martin-Misener, R., P. Harbman, F. Donald, K. Reid, K. Kilpatrick, N. Carter, D. BryantLukosius, et al. 2015. “Cost-Effectiveness of Nurse Practitioners in Primary
and Specialized Ambulatory Care: Systematic Review.” BMJ Open. https://
bmjopen.bmj.com/content/5/6/e007167.
Milken Institute. 2020. “COVID 19 Tracker.” https://airtable.com/shrSAi6t5WFwqo3GM/
tblEzPQS5fnc0FHYR/viweyymxOAtNvo7yH.
Mullard, A. 2020. “COVID-19 Vaccine Development Pipeline Gears Up.” The Lancet 395,
no. 10239: 1751–52. https://doi.org/10.1016/S0140-6736(20)31252-6.
Mundinger, M., R. Kane, E. Lenz, A. Totten, W. Tsai, P. Cleary, W. Friedewald, A. Siu, and
M. Shelanski. 2000. “Primary Care Outcomes in Patients Treated by Nurse
Practitioners or Physicians: A Randomized Trial.” JAMA 283, no. 1: 59–68.
https://doi.org/10.1001/jama.283.1.59.
Oliver, G., L. Pennington, S. Revelle, and M. Rantz. 2014. “Impact of Nurse Practitioners
on Health Outcomes of Medicare and Medicaid Patients.” Nursing Outlook 62,
no. 6: 440-447. https://www.sciencedirect.com/science/article/abs/pii/
S002965541400150X.
Ortiz, J., R. Hofler, A. Bushy, Y. Lin, A. Khanijahani, and A. Bitney. 2018. “Impact of Nurse
Practitioner Practice Regulations on Rural Population Health Outcomes.”
Healthcare 6, no. 2: 65. https://doi.org/10.3390/healthcare6020065.
Perloff, J., S. Clarke, C. DesRoches, M. O’Reilly-Jacob, and P. Buerhaus. 2017.
“Association of State-Level Restrictions in Nurse Practitioner Scope of
Practice with the Quality of Primary Care Provided to Medicare Beneficiaries.”
Medical Care Research and Review 76, no. 5: 597–626. https://doi.
org/10.1177/1077558717732402.
Philipson, T., E. Berndt, A. Gottschalk, and E. Sun. 2008. “Cost-Benefit Analysis of the
FDA: The Case
of the Prescription Drug User Fee Acts.” Journal of Public Economics 92, nos.
5–6: 1306–25. https://doi.org/10.1016/j.jpubeco.2007.09.010.
Philipson, T., and A. Jena. 2006. “Who Benefits from New Medical Technologies?
Estimates of Consumer and Producer Surpluses for HIV/AIDS Drugs.” Forum
for Health Economics & Policy 9, no. 2. https://doi.org/10.3386/w11810.

404 | References

Philipson, T., E. Sun, D. Goldman, and A. Jena. 2012. “A Reexamination of the Costs of
Medical R&D Regulation.” Forum for Health Economics & Policy 15, no. 3:
132–159. https://doi.org/10.1515/fhep-2012-0020.
Plotkin, S., J. Robinson, G. Cunningham, R. Iqbal, and S. Larsen. 2017. “The Complexity
and Cost of Vaccine Manufacturing: An Overview.” Vaccine 35, no. 33: 4064–
71. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5518734/.
Poghosyan, L., E. Timmons, C. Abraham, and G. Martsolf. 2019. “The Economic Impact
of Nurse Practitioner Scope of Practice for Medicaid.” Journal of Nursing
Regulation 10, no. 1: 15–20. https://doi.org/10.1016/S2155-8256(19)30078-X.
Pronker, E., T. Weenen, H. Commandeur, E. Claassen, and A. Osterhaus. 2013. “Risk in
Vaccine Research and Development Quantified.” Plos One 8, no. 3: e57755.
https://doi.org/10.1371/journal.pone.0057755.
Qureshi, Z., E. Seoane-Vazquez, R. Rodriguez-Monguio, K. Stevenson, and S. Szeinbach.
2011. “Market Withdrawal of New Molecular Entities Approved in the United
States from 1980 to 2009.” Pharmacoepidemiology and Drug Safety 20, no. 7:
772–77. https://doi.org/10.1002/pds.2155.
Roebuck, M., J. Liberman, M. Gemmill-Toyama, and T. Brennan. 2011. “Medication
Adherence Leads to Lower Health Care Use and Costs Despite Increased Drug
Spending.” Health Affairs 30, no. 1. https://doi.org/10.1377.hlthaff.2009.1087.
Rosenblatt, R., and L. Hart. 2000. “Physicians and Rural America.” West Journal of
Medicine 173, no. 5: 348–51. https://doi.org/10.1136/ewjm.173.5.348.
Shi, L., and M. Samuels. 1997. “Practice Environment and the Employment of Nurse
Practitioners, Physician Assistants, and Certified Nurse Midwives by
Community Health Centers.” Journal of Allied Health 26, no. 3: 105–11. PMID:
9358300. https://pubmed.ncbi.nlm.nih.gov/9358300/.
Spetz, J., S. Parente, R. Town, and D. Bazarko. 2013. “Scope-of-Practice Laws Far Nurse
Practitioners Limit Cost Savings That Can Be Achieved in Retail Clinics.”
Health Affairs 32, no. 11. https://www.healthaffairs.org/doi/full/10.1377/
hlthaff.2013.0544.
Spoont, M., N. Greer, J. Su, P. Fitzgerald, I. Rutks, and T. Wilt. 2011. Rural vs. Urban
Ambulatory Health Care: A Systematic Review. Washington: U.S. Department
of Veterans Affairs. https://www.ncbi.nlm.nih.gov/books/NBK56147/.
Stanik-Hutt, J., R. Newhouse, K. White, M. Johantgen, E. Bass, G. Zangaro, R. Wilson, et
al. 2013. “The Quality and Effectiveness of Care Provided by Nurse
Practitioners.” Journal of Nurse Practitioners 9, no. 8: 492–500. https://doi.
org/10.1016/j.nurpra.2013.07.004.
Verma, S. 2020. “Early Impact of CMS Expansion of Medicare Telehealth during COVID19.” Health Affairs Blog. https://www.healthaffairs.org/do/10.1377/
hblog20200715.454789/full/.

References | 405

Chapter 5
Abravanel, M., N. Pindus, B. Theodos, K. Bertumen, R. Brash, and Z. McDade. 2013.
“New Markets Tax Credit NNMTC Program Evaluation Final Report.” Urban
Institute. https://www.urban.org/sites/default/files/
publication/24211/412958-New-Markets-Tax-Credit-NMTC-ProgramEvaluation.PDF.
Arefeva, A., M. Davis, A. Ghent, and M. Park. 2020. “Job Growth from Opportunity
Zones.” Working paper.
Bauguess, S., R. Gullapalli, and V. Ivanov. 2018. “Capital Raising in the U.S.: An Analysis
of the Market for Unregistered Securities Offerings, 2009-2017.” Securities
Exchange Commission White Paper. https://www.sec.gov/files/DERA%20
white%20paper_Regulation%20D_082018.pdf.
Bernstein, J., and K. Hassett. 2015. “Unlocking Private Capital to Facilitate Economic
Growth in Distressed Areas.” Economic Innovation Group. https://eig.org/
wp-content/uploads/2015/04/Unlocking-Private-Capital-to-Facilitate-Growth.
pdf.
Bloom, D., and C. Michalopoulos. 2001. “How Welfare and Work Policies Affect
Employment and Income: A Synthesis of Research.” MDRC. https://www.
mdrc.org/publication/
how-welfare-and-work-policies-affect-employment-and-income.
Brummet, Q., and D. Reed. 2019. The Effects of Gentrification on the Well-Being and
Opportunity of Original Resident Adults and Children. Working Paper WP 19-30.
Philadelphia: Federal Reserve Bank of Philadelphia. https://www.
philadelphiafed.org/-/media/research-and-data/publications/workingpapers/2019/wp19-30.pdf.
Busso, M., J. Gregory, and P. Kline. 2013. “Assessing the Incidence and Efficiency of a
Prominent Place Based Policy.” American Economic Review 103, no. 2:
897–947.
CEA (Council of Economic Advisers). 2017. “The Growth Effects of Corporate Tax Reform
and Implications for Wages.” https://www.whitehouse.gov/sites/whitehouse.
gov/files/images/Corporate%20Tax%20Reform%20and%20Growth%20Final.
pdf.
———. 2019. Economic Report of the President. https://www.whitehouse.gov/
wp-content/uploads/2019/03/ERP-2019.pdf.
Chen, J., E. Glaeser, and D. Wessel. 2019. The (Non-)Effect of Opportunity Zones on
Housing Prices. NBER Working Paper 26587. Cambridge, MA: National Bureau
of Economic Research.
CRS (Congressional Research Service). 2011. “Empowerment Zones, Enterprise
Communities, and Renewal Communities: Comparative Overview and
Analysis.” https://www.everycrsreport.com/files/20110214_R41639_
b18ae5bf0fbe93505d7b6c2b13b744b76124b9ed.pdf.

406 | References

Dahlby, B. 2008. The Marginal Cost of Public Funds: Theory and Applications. Cambridge,
MA: MIT Press.
Dragan, K., Ellen, I. and S. Glied. 2019. Does Gentrification Displace Poor Children? New
Evidence from New York City Medicaid Data. NBER Working Paper 25809.
Cambridge, MA: National Bureau of Economic Research.
Freedman, M. 2012. “Teaching New Markets Old Tricks: The Effects of Subsidized
Investment on Low-Income Neighborhoods.” Journal of Public Economics 96,
nos. 11–12: 1000–1014.
Gamper-Rabindran, S., and C. Timmins. 2013. “Does Cleanup of Hazardous Waste Sites
Raise Housing Values? Evidence of Spatially Localized Benefits.” Journal of
Environmental Economics and Management 65, no. 3: 345–60.
GAO (Government Accountability Office). 2014. “New Market Tax Credit.” GAO-14-500.
https://www.gao.gov/assets/670/664717.pdf.
Grainger, C. 2012. “The Distributional Effects of Pollution Regulations: Do Renters Fully
Pay for Cleaner Air?” Journal of Public Economics 96, nos. 9–10: 840–52.
Ham, J., C. Swenson, A. Imrohoroglu, and H. Song. 2011. “Government Programs Can
Improve Local Labor Markets: Evidence from State Enterprise Zones, Federal
Empowerment Zones and Federal Enterprise Community.” Journal of Public
Economics 95, no. 7: 779–97.
Harger, K., and A. Ross. 2016. “Do Capital Tax Incentives Attract New Businesses?
Evidence Across Industries from the New Markets Tax Credit.” Journal of
Regional Science 56, no. 5: 733–53.
Harger, K., A. Ross, and H. Stephens. 2019. “What Matters the Most For Economic
Development? Evidence from the Community Development Financial
Institutions Fund.” Papers in Regional Science 98, no. 2: 883–904.
Hoynes, H., and D. Schanzenbach. 2012. “Work Incentives and the Food Stamp
Program.” Journal of Public Economics 96, nos. 1–2: 151–62.
Hula, R., and M. Jordan. 2018. “Private Investment and Public Redevelopment: The
Case of New Markets Tax Credits.” Poverty & Public Policy 10 no. 1: 11–38.
Jacob, B., and J. Ludwig. 2012. “The Effects of Housing Assistance on Labor Supply:
Evidence from a Voucher Lottery.” American Economic Review 102, no. 1:
272–304.
Koby, Y., and K. Wolf. 2019. “Aggregation in Heterogeneous-Firm Models: Theory and
Measurement.” https://scholar.princeton.edu/sites/default/files/ckwolf/files/
hetfirms_agg.pdf.
Lowry, S., and D. Marples. 2019. “Tax Incentives for Opportunity Zones: In Brief.”
Congressional Research Service, R45152.
Markusen, A., and A. Glasmeier, 2008. “Overhauling and Revitalizing Federal Economic
Development Programs.” Economic Development Quarterly 22, no. 2: 83–91.
Nichols, A., and Rothstein, J., 2015. “The Earned Income Tax Credit.” In Economics of
Means-Tested Transfer Programs in the United States, Volume 1. Chicago:
University of Chicago Press.

References | 407

Ohrn, E. 2019. “The Effect of Tax Incentives on U.S. Manufacturing: Evidence from State
Accelerated Depreciation Policies.” Journal of Public Economics 180, article
104804.
Sage, A., M. Langen, and A. Van de Minne. 2019. “Where Is the Opportunity in
Opportunity Zones? Early Indicators of the Opportunity Zone Program’s
Impact on Commercial Property Prices.” http://dx.doi.org/10.2139/
ssrn.3385502.
Tax Policy Center. 2020. “What Is the New Market Tax Credit, and How Does It Work?”
https://www.taxpolicycenter.org/briefing-book/
what-new-markets-tax-credit-and-how-does-it-work.
Theodos, B., C. Stacy, and H. Ho. 2017. “Taking Stock of the Community Development
Block Grant.” Urban Institute. http://www.urban.org/sites/default/files/
publication/89551/cdbg_brief.pdf.
Vardell, R. 2019. “The Land of Opportunity Zones: Deferring Taxable Capital Gains
Through Investments in Low-Income Communities.” Missouri Law Review 84,
no 3: article 13.

Chapter 6
Anderson, S., and J. Sallee. 2011. “Using Loopholes to Reveal the Marginal Cost of
Regulation: The Case of Fuel-Economy Standards.” American Economic Review
101, no. 4: 1375–1409.
Berry, S., J. Levinsohn, and A. Pakes. 2004. “Differentiated Products Demand Systems
from a Combination of Micro and Macro Data: The New Car Market.” Journal
of Political Economy 112, no. 1 (2004): 68–105.
Bento, A., M. Freedman, and C. Lang. 2015. “Who Benefits from Environmental
Regulation? Evidence from the Clean Air Act Amendments.” Review of
Economics and Statistics 97, no. 3: 610–22.
Bento, A., et al. 2018. “Flawed Analyses of U.S. Auto Fuel Economy Standards.” Science
Magazine 362, no. 6419: 1119–21.
CEA (Council of Economic Advisers). 2019. “The Economic Effects of Federal
Deregulation since January 2017: An Interim Report.” https://www.
whitehouse.gov/wp-content/uploads/2019/06/The-Economic-Effects-ofFederal-Deregulation-Interim-Report.pdf.
———. 2020. Economic Report of the President. Washington: U.S. Government
Publishing Office.
Chambers, D., C. Collins, and A. Krause. 2019. “How Do Federal Regulations Affect
Consumer Prices? An Analysis of the Regressive Effects of Regulation.” Public
Choice 180, no. 1: 57–90.
Coffey, B., P. McLaughlin, and P. Peretto. 2020. “The Cumulative Cost of Regulations.”
Review of Economic Dynamics 38: 1–21.
Crain, M., and N. Crain. 2014. “The Cost of Federal Regulation to the U.S. Economy,
Manufacturing and Small Business.” National Association of Manufacturers.

408 | References

Crews, W. 2017. “Mapping Washington’s Lawlessness; An Inventory of Regulatory Dark
Matter.” Competitive Enterprise Institute. https://cei.org/sites/default/files/
Wayne%20Crews%20-%20Mapping%20Washington%27s%20
Lawlessness%202017.pdf.
———. 2020. “The E.O. 13891 Guidance Document Portal: An Exercise in Utility.” https://
cei.org/blog/eo-13891-guidance-document-portal-exercise-utility.
Duclos, J. 2008 “Horizontal and Vertical Equity.” In The New Palgrave Dictionary of
Economics. https://link.springer.com/referenceworkentry/10.105
7%2F978-1-349-95121-5_1961-1.
Dynan, K., J. Skinner, and S. Zeldes. 2004. “Do the Rich Save More?” Journal of Political
Economy 112, no. 2: 397–444.
EPA (Environmental Protection Agency). 2016. “Proposed Determination on the
Appropriateness of the Model Year 2022–2025 Light-Duty Vehicle Greenhouse
Gas Emissions Standards under the Midterm Evaluation: Technical Support
Document.” https://nepis.epa.gov/Exe/ZyPDF.cgi?Dockey=P100Q3L4.pdf.
———. 2019. “The 2018 EPA Automotive Trends Report: Greenhouse Gas Emissions,
Fuel Economy, and Technology since 1975.” https://nepis.epa.gov/Exe/ZyPDF.
cgi/P100W5C2.PDF?Dockey=P100W5C2.PDF.
EPA/DOT (U.S. Environmental Protection Agency / U.S. Department of Transportation).
2012. “Final Rulemaking for 2017–2025 Light-Duty Vehicle Greenhouse Gas
Emission Standards and Corporate Average Fuel Economy Standards.”
https://nepis.epa.gov/Exe/ZyPDF.cgi/P100EZI1.PDF?Dockey=P100EZI1.PDF.
———. 2018. “The Safer Affordable Fuel-Efficient (SAFE) Vehicles Rule for Model Years
2021–2026 Passenger Cars and Light Trucks.” Preliminary Regulatory Impact
Analysis. https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/documents/ld_
cafe_my2021-26_pria_0.pdf.
———. 2020. “The Safer Affordable Fuel-Efficient (SAFE) Vehicles Rule for Model Years
2021–2026 Passenger Cars and Light Trucks.” Final Regulatory Impact
Analysis. https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/documents/final_
safe_fria_web_version_200701.pdf.
Figueroa, E., and S. Waxman. 2017. “Which States Tax the Sale of Food for Home
Consumption in 2017?” Center on Budget and Policy Priorities. http://www.
cbpp.org/research/state-budget-and-tax/
which-states-tax-the-sale-of-food-for-home-consumption-in-2017.
Forbes. 2017. “Tesla’s Lucrative ZEV Credits May Not Be Sustainable.” https://www.
forbes.com/sites/greatspeculations/2017/09/01/
teslas-lucrative-zev-credits-may-not-be-sustainable/?sh=16b653416ed5.
Garicano, L., C. LeLarge, and J. Van Reenen. 2016. “Firm Size Distortions and the
Productivity Distribution: Evidence from France.” American Economic Review
106, no. 11: 3439–79.
Jacobsen, M. 2013. “Evaluating U.S. Fuel Economy Standards in a Model with Producer
and Household Heterogeneity.” American Economic Journal: Economic Policy
5, no. 2: 148–87.

References | 409

Leard, B., and V. McConnell. 2017. “New Markets for Credit Trading under U.S.
Automobile Greenhouse Gas and Fuel Economy Standards.” Review of
Environmental Economics and Policy 11, no. 2: 207–26.
Levinson, A. 2019. “Energy Efficiency Standards Are More Regressive Than Energy
Taxes: Theory and Evidence.” Journal of the Association of Environmental and
Resource Economists 6, no. S1: S7–S36.
Mulligan, C., and T. Philipson. 2000. Merit Motives and Government Intervention: Public
Finance in Reverse. NBER Working Paper 7698. Cambridge, MA: National
Bureau of Economic Research.
Stigler, G. 1971. “The Theory of Economic Regulation.” Bell Journal of Economics and
Management Science 2, no. 1: 3–21.
Thomas, D. 2012. Regressive Effects of Regulation. Arlington, VA: Mercatus Center at
George Mason University.

Chapter 7
Abdulkadiroğlu, A., P. Pathak, A. Roth, and T. Sönmez. 2006. Changing the Boston
School Choice Mechanism. NBER Working Paper 11965. Cambridge, MA:
National Bureau of Economic Research.
Abdulkadiroğlu, A., P. Pathak, and C. Walters. 2018. “Free to Choose: Can School Choice
Reduce Student Achievement?” American Economic Journal: Applied
Economics 10, no. 1: 175–206.
Anderson, K. 2017. “Evidence on Charter School Practices Related to Student
Enrollment and Retention.” Journal of School Choice 11, no. 4: 527–45.
Angrist, J., S. Cohodes, S. Dynarski, P. Pathak, and C. Walters. 2016. “Stand and Deliver:
Effects of Boston’s Charter High Schools on College Preparation, Entry, and
Choice.” Journal of Labor Economics 34, no. 2.
Ayscue, J., R. Levy, G. Siegel-Hawley, and B. Woodward. 2015. “Choices Worth Making:
Creating, Sustaining, and Expanding Diverse Magnet Schools.” Civil Rights
Project / Proyecto Derechos Civiles.
Baude, P., M. Casey, E. Hanushek, G. Phelan, and S. Rivkin. 2020. “The Evolution of
Charter School Quality.” Economica 87, no. 345: 158–89.
Bifulco, R., and R. Reback. 2014. “Fiscal Impacts of Charter Schools: Lessons from New
York.” Education Finance and Policy 9, no. 1: 86–107.
Booker, K., T. Sass, and R. Zimmer. 2011. “The Effects of Charter High Schools on
Educational Attainment.” Journal of Labor Economics 29, no. 2: 377–415.
Boston School Finder. 2020. “What Is the Home-Based Assignment Policy (HBAP)?”
https://www.bostonschoolfinder.org/home-based-assignment-policy-hbap.
Buerger, C., and R. Bifulco. 2019. “The Effect of Charter Schools on Districts’ Student
Composition, Costs, and Efficiency: The Case of New York State.” Economics
of Education Review 69: 61–72.

410 |

References

Butler, J., D. Carr, E. Toma, and R. Zimmer. 2013. “Choice in a World of New School
Types.” Journal of Policy Analysis and Management 32, no. 4: 785–806.
CEA (Council of Economic Advisers). 2020. “The Impact of Opportunity Zones: An Initial
Assessment.” https://www.whitehouse.gov/wp-content/uploads/2020/08/
The-Impact-of-Opportunity-Zones-An-Initial-Assessment.pdf.
Chakrabarti, R. 2008. “Can Increasing Private School Participation and Monetary Loss
in a Voucher Program Affect Public School Performance? Evidence from
Milwaukee.” Journal of Public Economics 92, nos. 5–6: 1371–93.
Chakrabarti, R., and J. Roy. 2010. “The Economics of Parental Choice.” Economics of
Education 3: 336–42.
Chingos, M., and P. Peterson. 2015. “Experimentally Estimated Impacts of School
Vouchers on College Enrollment and Degree Attainment.” Journal of Public
Economics 122: 1–12.
Cohodes, S., E. Setren, and C. Walters. 2019. Can Successful Schools Replicate? Scaling
Up Boston’s Charter School Sector. NBER Working Paper 25796. Cambridge,
MA: National Bureau of Economic Research.
CREDO (Center for Research on Education Outcomes). 2009. “Multiple Choice: Charter
School Performance in 16 States.” Stanford University.
———. 2013. “National Charter School Study.” Stanford University. https://credo.
stanford.edu/sites/g/files/sbiybj6481/f/ncss_2013_final_draft.pdf.
CRS (Congressional Research Service). 2018. “Higher Education Tax Benefits: Brief
Overview and Budgetary Effects.” https://files.eric.ed.gov/fulltext/ED593609.
pdf.
———. 2019. “District of Columbia Opportunity Scholarship Program (DC OSP):
Overview, Implementation, and Issues.” https://fas.org/sgp/crs/misc/R45581.
pdf.
David, R. 2018. “National Charter School Management Overview.” National Alliance for
Public Charter Schools. https://www.publiccharters.org/sites/default/files/
documents/2019-06/napcs_management_report_web_06172019.pdf.
DeAngelis, C., and P. Wolf. 2019. “Will Democracy Endure Private School Choice? The
Effect of the Milwaukee Parental Choice Program on Adult Voting Behavior.”
Journal of Private Enterprise.
———. 2020. “Private School Choice and Character: More Evidence from Milwaukee.”
Journal of Private Enterprise.
DeAngelis, C., P. Wolf, L. Maloney, and J. May. 2018. “Charter School Funding: (More)
Inequity in the City.” November. School Choice Demonstration Project.
Dobbie, W., and R. Fryer. 2015. “The Medium-Term Impacts of High-Achieving Charter
Schools.” Journal of Political Economy 123, no. 5: 985–1037.
DCPCSB (D.C. Public Charter School Board). 2020. “Student Enrollment.” https://
dcpcsb.org/student-enrollment.
ECS (Education Commission of the States). 2018. “Open Enrollment Quick Guide.”
https://www.ecs.org/wp-content/uploads/Open-Enrollment-Quick-Guide.pdf.

References | 411

EdChoice. 2019. “Fast Facts on School Choice.” https://www.edchoice.org/engage/
fast-facts/.
———. 2020a. “The ABCs of School Choice.” https://www.edchoice.org/wp-content/
uploads/2020/01/2020-ABCs-of-School-Choice-WEB-OPTIMIZED-REVISED.pdf.
———. 2020b. “Types of School Choice.” https://www.edchoice.org/school-choice/
types-of-school-choice/.
———. 2020c. “Wisconsin—Milwaukee Parental Choice Program.” https://www.
edchoice.org/school-choice/programs/
wisconsin-milwaukee-parental-choice-program/.
EdSurge. 2020. “What’s the Plan? How K-12 School Districts Are Preparing to Resume
and Reopen.” https://www.edsurge.com/research/guides/
what-s-the-plan-how-k-12-school-districts-are-preparing-to-resume-andreopen.
Egalite, A., J. Mills, and P. Wolf. 2017. “The Impact of Targeted School Vouchers on
Racial Stratification in Louisiana Schools.” Education and Urban Society 49,
no. 3: 271–96.
Epple, D., and R. Romano. 2008. “Educational Vouchers and Cream Skimming.”
International Economic Review 49, no. 4: 1395–1435.
Epple, D., R. Romano, and M. Urquiola. 2017. “School Vouchers: A Survey of the
Economic Literature.” Journal of Economic Literature 55, no. 2: 441–91.
Epple, D., R. Romano, and R. Zimmer. 2016. “Chapter 3: Charter Schools—A Survey of
Research on Their Characteristics and Effectiveness.” In Handbook of the
Economics of Education, ed. G. Bulman and R.W. Fairlie. Amsterdam: Elsevier.
Vol. 5, 139–208.
Figlio, D., and C. Hart. 2014. “Competitive Effects of Means-Tested School Vouchers.”
American Economic Journal: Applied Economics 6, no. 1: 133–56.
Figlio, D., C. Hart, and K. Karbownik. 2020. Effects of Scaling Up Private School Choice
Programs on Public School Students. NBER Working Paper 26758. Cambridge,
MA: National Bureau of Economic Research.
Florida Department of Education. 2019a. “School Choice.” http://www.fldoe.org/
schools/school-choice/.
———. 2019b. “Florida’s Charter Schools.” http://www.fldoe.org/core/fileparse.
php/7696/urlt/Charter-Sept-2019.pdf.
———. 2019c. “Fact Sheet: Florida Tax Credit Scholarship Program.” http://www.fldoe.
org/core/fileparse.php/5606/urlt/FTC-Sept-2019-line.pdf.
Friedman, M. 1955. “The Role of Government in Education.” In Economics and the
Public Interest. New Brunswick, NJ: Rutgers University Press.
Gilraine, M., U. Petronijevic, and J. Singleton. 2019. Horizontal Differentiation and the
Policy Effect of Charter Schools. EdWorkingPaper 19-80. https://
edworkingpapers.com/sites/default/files/ai19-80.pdf.
Grube, L., and D. Anderson. 2018. “School Choice and Charter Schools in Review: What
Have We Learned?” Journal of Private Enterprise 33, no. 4: 21–44.

412 |

References

Harris, D., and M. Larsen. 2018. “What Effect Did the New Orleans School Reforms Have
on Student Achievement, High School Graduation, and College Outcomes?”
Education Research Alliance for New Orleans. July 15. https://
educationresearchalliancenola.org/files/publications/071518-Harris-LarsenWhat-Effect-Did-the-New-Orleans-School-Reforms-Have-on-StudentAchievement-High-School-Graduation-and-College-Outcomes.pdf.
Hitt, C., M. McShane, and P. Wolf. 2018. “Do Impacts on Test Scores Even Matter?
Lessons from Long-Run Outcomes in School Choice Research.” American
Enterprise Institute. https://www.aei.org/research-products/report/
do-impacts-on-test-scores-even-matter-lessons-from-long-run-outcomes-inschool-choice-research/
Hoxby, C. 2003. “Chapter 8: School Choice and School Productivity: Could School
Choice Be a Tide That Lifts All Boats?” In The Economics of School Choice, ed.
C. Hoxby. Cambridge, MA: National Bureau of Economic Research.
JEC (Joint Economic Committee). 2019. Zoned Out: How School and Residential Zoning
Limit Educational Opportunity. SCP Report 6-19.
Kaplan, L., and W. Owings. 2018. “Funding School Choice: Implications for American
Education.” Journal of Education Finance 44, no. 2: 199–217.
Ladd, H., and J. Singleton. 2020. “The Fiscal Externalities of Charter Schools: Evidence
from North Carolina.” Education Finance and Policy 15, no. 1: 191–208.
Leuken, M. 2018. Fiscal Effects of School Vouchers: Examining the Savings and Costs of
America’s Private School Voucher Programs. Indianapolis: EdChoice.
Massachusetts Department of Elementary and Secondary Education. 2016.
“Enrollment Policy and Practice Frequently Asked Questions.” http://www.
doe.mass.edu/charter/guidance/2016-3-faq.html.
———. 2019. “Massachusetts Charter School Waitlist Initial Report for 2019–2020
(FY20).” March 15. http://www.doe.mass.edu/charter/enrollment/fy2020/
waitlist.html#:~:text=73%20out%20of%2081%20charter,waitlists%2C%20
representing%2027%2C743%20unique%20students.
Mayer, D., P. Peterson, D. Myers, C. Tuttle, and W. Howell. 2002. “School Choice in New
York City after Three Years: An Evaluation of the School Choice Scholarships
Program.” Mathematica Policy Research.
McShane, M. 2020. “The School Choice Now Act and the Fate of Private Schools.”
Forbes, July 23. https://www.forbes.com/sites/mikemcshane/2020/07/23/
the-school-choice-now-act-and-the-fate-of-private-schools/#2db492ef7278.
National Association of Realtors. 2018. “2018 Profile of Home Buyers and Sellers.”
https://nationalmortgageprofessional.com/sites/default/files/NAR_
HBS_2018_10_29_18.pdf.
NCES (National Center for Education Statistics). 2019a. “Table 216.20. Number and
Enrollment of Public Elementary and Secondary Schools by School Level,
Type, and Charter, Magnet, and Virtual Status: Selected Years, 1990–91
through 2017–2018.” https://nces.ed.gov/programs/digest/d19/tables/
dt19_216.20.asp?current=yes.

References | 413

———. 2019b. “School Choice in the United States: 2019.” https://nces.ed.gov/
programs/schoolchoice/ind_02.asp.
———. 2019c. “Table 235.20: Revenues for Public Elementary and Secondary Schools,
by Source of Funds and State or Jurisdiction: 2016–2017.” https://nces.
ed.gov/programs/digest/d19/tables/dt19_235.20.asp?current=yes.
———. 2019d. “Table 216.30: Number and Percentage Distribution of Public Elementary
and Secondary Students and Schools, by Traditional or Charter School Status
and Selected Characteristics: Selected Years, 1999–2000 through 2017–18.”
https://nces.ed.gov/programs/digest/d19/tables/dt19_216.30.
asp?current=yes
———. 2020a. “Public School Revenue Sources.” https://nces.ed.gov/programs/coe/
indicator_cma.asp.
———. 2020b.” Revenues and Expenditures for Public Elementary and Secondary
Education: FY 17.” https://nces.ed.gov/pubs2020/2020303.pdf.
Office of Senator Ted Cruz. 2019. “Sen. Cruz Introduces Education Freedom
Scholarships and Opportunity Act.” February 28. https://www.cruz.senate.
gov/?p=press_release&id=4343.
OII (Office of Innovation and Improvement). 2004. “Creating Successful Magnet School
Programs.” U.S. Department of Education. https://www2.ed.gov/admins/
comm/choice/magnet/report.pdf.
Ridley, M., and C. Terrier. 2018. Fiscal and Education Spillovers from Charter School
Expansion. NBER Working Paper 25070. Cambridge, MA: National Bureau of
Economic Research.
Rothstein, J. 2006. “Good Principals or Good Peers? Parental Valuation of School
Characteristics, Tiebout Equilibrium, and the Incentive Effects of Competition
among Jurisdictions.” American Economic Review 96, no. 4: 1333–49.
Sass, T., R. Zimmer, B. Gill, and T. Booker. 2016. “Charter High School’s Effect on LongTerm Attainment and Earnings.” Journal of Policy Analysis and Management
35, no. 3: 683–706.
Supreme Court of the United States. 2019. Espinoza et al. v. Montana Department of
Revenue. https://www.supremecourt.gov/opinions/19pdf/18-1195_g314.
pdf#page=23.
Teach New Orleans. 2020. “Nola by the Numbers.” https://teachneworleans.net/nolaby-the-numbers/#:~:text=98%25%20of%20students%20attend%20
charter,create%20its%20own%20unique%20culture.
Tiebout, C. 1956. “A Pure Theory of Local Expenditures.” Journal of Political Economy 5:
416–24.
Urquiola, M. 2005. “Does School Choice Lead to Sorting? Evidence from Tiebout
Variation.” American Economic Review 95, no. 4: 1310–26.
U.S. Department of Education. 2015. “Welcome to ED’s Charter Schools Program.”
https://www2.ed.gov/about/offices/list/oii/csp/index.html.

414 | References

———. 2017. “The Federal Role in Education.” https://www2.ed.gov/about/overview/
fed/role.html.
———. 2019a. “School Choice Improvement Programs.” https://oese.ed.gov/offices/
office-of-discretionary-grants-support-services/
school-choice-improvement-programs/.
———. 2019b. “The U.S. Department of Education’s Charter Schools Program
Overview.” https://oese.ed.gov/files/2019/12/CSP-Data-OverviewWestEd-7.22.2019.pdf.
———. 2019c. “How Education Freedom Scholarships Can Expand Private and Home
Education Options.” https://sites.ed.gov/freedom/2019/09/26/
how-education-freedom-scholarships-can-expand-private-and-homeeducation-options/
———. 2020a. “Magnet Schools Assistance Program.” https://oese.ed.gov/offices/officeof-discretionary-grants-support-services/
school-choice-improvement-programs-assistance-program-msap/.
———. 2020b. “Secretary DeVos Awards $65 Million to Create and Expand Public
Charter Schools in Areas of Greatest Need.” https://www.ed.gov/news/pressreleases/
secretary-devos-awards-65-million-create-and-expand-public-charterschools-areas-greatest-need.
Walters, C. 2018. “The Demand for Effective Charter Schools.” Journal of Political
Economy 126, no. 6: 2179–2223.
Wisconsin Department of Public Instruction. 2020. “2019–2020 Charter School
Enrollment.” https://dpi.wi.gov/sms/charter-schools/current.
Witte, J. 1998. “The Milwaukee Voucher Experiment.” Educational Evaluation and Policy
Analysis 20, no. 4: 229–51.
Wolf, P., B. Gutmann, M. Puma, B. Kisida, L. Rizzo, N. Eissa., and M. Carr. 2010.
“Evaluation of the DC Opportunity Scholarship Program.” National Center for
Education Evaluation and Regional Assistance. https://ies.ed.gov/ncee/
pubs/20104018/pdf/20104018.pdf.
Wolf, P., B. Kisida, B. Gutmann, M. Puma, N. Eissa, and L. Rizzo. 2013. “School Vouchers
and Student Outcomes: Experimental Evidence from Washington, DC.”
Journal of Policy Analysis and Management 32, no. 2: 246–70.
Zimmer, R., B. Gill, K. Booker, S. Lavertu, T. Sass, and J. Witte. 2009. “Charter Schools in
Eight States: Effects on Achievement, Attainment, Integration, and
Competition.” RAND Corporation.

Chapter 8
Acemoglu, D., S. Johnson, and J. Robinson. 2001. “The Colonial Origins of Comparative
Development: An Empirical Investigation.” American Economic Review 91, no.
5: 1369–1401.

References | 415

Alston, E., and S. Smith. 2020. “Development Derailed: Uncertain Property Rights and
Asset-Specific Investment.” Mountain West Economic History Conference.
Barton, S., and H. Recht. 2018. “The Massive Prize Luring Miners to the Stars.”
Bloomberg, March 8. https://www.bloomberg.com/
graphics/2018-asteroid-mining/.
Barzel, Y. 1997. Economic Analysis of Property Rights. New York: Cambridge University
Press.
Bohn, H., and R. Deacon. 2000. “Ownership Risk, Investment, and the Use of Natural
Resources.” American Economic Review 90, no. 3: 526–49.
Butow, S., T. Cooley, E. Felt, and J. Mozer. 2020. “State of the Space Industrial Base
2020: A Time for Action to Sustain US Economic & Military Leadership in
Space.” http://aerospace.csis.org/wp-content/uploads/2020/07/State-of-theSpace-Industrial-Base-2020-Report_July-2020_FINAL.pdf.
CEA (Council of Economic Advisers). 2020. “Expanding Educational Opportunity
through Choice and Competition.” https://www.whitehouse.gov/wp-content/
uploads/2020/10/Expanding-Education-through-Choice.pdf.
Cust, J., and T. Harding. 2020. “Institutions and the Location of Oil Exploration.” Journal
of the European Economic Association 18, no. 3: 1321–50.
Demsetz, H. 1967. “Toward a Theory of Property Rights.” American Economic Review 57,
no. 2: 347–59.
DIA (Defense Intelligence Agency). 2019. “Challenges to Security in Space.” https://
www.dia.mil/Portals/27/Documents/News/Military%20Power%20
Publications/Space_Threat_V14_020119_sm.pdf.
Ferreira, S. 2004. “Deforestation, Property Rights, and International Trade.” Land
Economics 80, no. 2: 174–93.
Galiani, S., and E. Schargrodsky. 2010. “Property Rights for the Poor: Effects of Land
Titling.” Journal of Public Economics 94, nos. 9–10: 700–729.
Grainger, C., and C. Costello. 2014. “Capitalizing Property Rights Insecurity in Natural
Resource Assets.” Journal of Environmental Economics and Management 67,
no. 2: 224–40.
Hertzfeld, H., and F. von der Dunk. 2005. “Bringing Space Law into the Commercial
World: Property Rights without Sovereignty.” Chicago Journal of International
Law 6, no. 1: 81–99.
Hornbeck, R. 2010. “Barbed Wire: Property Rights and Agricultural Development.”
Quarterly Journal of Economics 125, no. 2: 767–810.
Kemal, M., and I. Lange. 2018. “Changes in Institutional Design and Extraction Paths.”
Environmental and Development Economics 23, no. 4: 478–94.
Leonard, B., and G. Libecap. 2019. “Collective Action by Contract: Prior Appropriation
and the Development of Irrigation in the Western United States.” Journal of
Law and Economics 62, no. 1: 67–115.
Libecap, G. 1978. “Economic Variables and the Development of Law: The Case of
Western Mineral Rights.” Journal of Economic History 38, no. 2: 338–62.

416 |

References

Libecap, G., and J. Smith. 2002. “The Economic Evolution of Petroleum Property Rights
in the United States.” Journal of Legal Studies 31, no. S2: S589–S608.
Llinás, C. 2016. “Pretending to Be Liberian and Panamanian; Flags of Convenience and
the Weakening of the Nation-State on the High Seas.” Journal of Maritime Law
& Commerce 47, no. 1: 1–28.
North, D. 1991. “Institutions.” Journal of Economic Perspectives 5, no. 1: 97–112.
O’Connor, A., M. Gallaher, K. Clark-Sutton, D. Lapidus, Z. Oliver, T. Scott, D. Wood, M.
Gonzalez, E. Brown, and J. Fletcher. 2019. Economic Benefits of the Global
Positioning System (GPS). Research Triangle Park, NC: RTI International.
https://www.space.commerce.gov/wp-content/uploads/2019-08-gpspresentation.pdf.
Pace, S., G. Frost, I. Lachow, D. Frelinger, D. Fossum, D. Wassem, and M. Pinto. 1995.
The Global Positioning System: Assessing National Priorities. Santa Monica, CA:
RAND Corporation.
Pence, M. 2019. “Mike Pence: It’s Time for Congress to Establish the Space Force.”
Washington Post, March 1. https://www.washingtonpost.com/opinions/mikepence-its-time-for-congress-to-establish-the-spaceforce/2019/03/01/50820a58-3c4e-11e9-a06c-3ec8ed509d15_story.html.
Taghdiri, A. 2013. “Flags of Convenience and the Commercial Space Flight Industry:
The Inadequacy of Current International Law to Address the Opportune
Registration of Space Vehicles in Flag States.” Journal of Science and
Technology Law 19, no. 2: 405–31.
USSF (U.S. Space Force). 2020. “About the United States Space Force.” https://www.
spaceforce.mil/About-Us/About-Space-Force.
Weinzierl, M. 2018. “Space, the Final Economic Frontier.” Journal of Economic
Perspectives 32, no. 2: 173–92.

Chapter 9
Abraham, K., and M. Kearney. 2020. “Explaining the Decline in the US Employment-toPopulation Ratio: A Review of the Evidence.” Journal of Economic Literature
58, no. 3: 585–643.
Anthony, C., et al. 2015. The Costs of the Israeli-Palestinian Conflict. Santa Monica, CA:
RAND Corporation.
Autor, D., D. Dorn, and G. Hanson. 2013. “The China Syndrome: Local Labor Market
Effects of Import Competition in the United States.” American Economic
Review 103, no. 6: 2121–68.
———. 2019. “When Work Disappears: Manufacturing Decline and the Falling Marriage
Market Value of Young Men.” American Economic Review: Insights 1, no. 2:
161–78.
Baldwin, R., and E. Tomiura. 2020. “Thinking Ahead About the Trade Impact of COVID19.” In Economics in the Time of COVID-19, ed. R. Baldwin and B. di Mauro.
London: Centre for Economic Policy Research.

References | 417

Bank of America. 2020a. “Tectonic Shifts in Global Supply Chains.” February 4. Bank of
America Global Research. https://www.bofaml.com/content/dam/
boamlimages/documents/articles/ID20_0147/Tectonic_Shifts_in_Global_
Supply_Chains.pdf.
———. 2020b. “The Cost of Remaking Supply Chains: Significant but not Prohibitive.”
July 23. Bank of America Global Research. https://www.bofaml.com/content/
dam/boamlimages/documents/articles/ID20_0734/cost_of_remaking_
supply_chains.pdf.
Case, A., and A. Deaton. 2017. “Mortality and Morbidity in the 21st Century.” Brookings
Papers on Economic Activity. casetextsp17bpea.pdf.
CDC (U.S. Centers for Disease Control and Prevention). 2020. “No Sail Order and
Suspension of Further Embarkation.” Federal Register. https://www.
federalregister.gov/documents/2020/03/24/2020-06166/
no-sail-order-and-suspension-of-further-embarkation.
CEA (Council of Economic Advisers). 2019. Economic Report of the President.
Washington: U.S. Government Publishing Office.
CRS (Congressional Research Service). 2018. “U.S.-South Korea (KORUS) FTA.” https://
fas.org/sgp/crs/row/IF10733.pdf.
———. 2019. “‘Stage One’ U.S.-Japan Trade Agreements.” https://fas.org/sgp/crs/row/
R46140.pdf.
———. 2020a. “USMCA: Intellectual Property Rights (IPR).” https://crsreports.congress.
gov/product/pdf/IF/IF11314#:~:text=Geographical%20Indications%20
(GIs),-GIs%20are%20geographical&text=USMCA%20contains%20due%20
process%20procedures,GI%20protection%20in%20international%20
agreements.
———. 2020b. “U.S.-Brazil Trade Relations.” https://crsreports.congress.gov/product/
pdf/IF/IF10447.
Cui, Y., J. Meng, and C. Lu. 2018. “Recent Developments in China’s Labor Market: Labor
Shortage, Rising Wages, and Their Implications.” Review of Development
Economics 22, no. 3: 1217–38.
DOJ (U.S. Department of Justice). 2018. “Statement of John C. Demers Before the
Committee on the Judiciary, United States Senate.” https://www.justice.gov/
sites/default/files/testimonies/witnesses/
attachments/2018/12/18/12-05-2018_john_c._demers_testimony_re_china_
non-traditional_espionage_against_the_united_states_the_threat_and_
potential_policy_responses.pdf.
Elliott, L., and R. Mason. 2020. “U.K. to Impose Tax on Tech Giants but Risks U.S. Tariffs
on Car Exports.” Guardian, January 22. https://www.theguardian.com/
business/2020/jan/22/
uk-to-impose-tax-on-tech-giants-but-risks-us-tariffs-on-car-exports.
Goodman, M., D. Gerstel, N. Szechenyi, and M. Green. 2019. “The U.S.-Japan Trade
Deal.” Center for Strategic and International Studies. https://www.csis.org/
analysis/us-japan-trade-deal.

418 |

References

GPO (U.S. Government Publishing Office). 2000. “Public Law 206–286—Oct. 20, 2000:
Normal Trade Relations for the People’s Republic of China.” https://www.
congress.gov/106/plaws/publ286/PLAW-106publ286.pdf.
Grossman, G., and E. Rossi-Hansberg. 2008. “Trading Tasks: A Simple Theory of
Offshoring.” American Economic Review 98, no. 5: 1978–97.
Henley, J., J. Rankin, and L. O’Carroll. 2020. “Brexit Explained: How It Happened and
What Comes Next.” Guardian, January 27. https://ustr.gov/countries-regions/
europe-middle-east/europe/united-kingdom/
us-uk-trade-agreement-negotiations.
ILO (International Labor Organization). 2014. “Informal Employment in Mexico: Current
Situation, Policies, and Challenges.” https://www.ilo.org/wcmsp5/groups/
public/---americas/---ro-lima/documents/publication/wcms_245889.pdf.
Isaac, A. 2020. “U.K. Scrambles for Legal Route to Hit Back on U.S. Airbus Tariffs.”
Politico, December 1. https://www.politico.eu/article/
uk-scrambles-for-legal-route-to-hit-back-on-us-airbus-tariffs/.
Kearney. 2020. “Trade War Spurs Sharp Reversal in 2019 Reshoring Index,
Foreshadowing COVID-19 Test of Supply Chain Relience.” https://www.
kearney.com/operations-performance-transformation/us-reshoring-index/
full-report.
Luo, S., and K. Tsang. 2020. “China and World Output Impact of the Hubei Lockdown
During the Coronavirus Outbreak.” Contemporary Economic Policy,
forthcoming.
Lund, S., et al. 2020. “Risk, Resilience, and Rebalancing in Global Value Chains.”
McKinsey Global Institute, August 6. https://www.mckinsey.com/businessfunctions/operations/our-insights/
risk-resilience-and-rebalancing-in-global-value-chains.
McBride, J., and A. Chatzky. 2019. “Is ‘Made in China 2025’ a Threat to Global Trade?”
Council on Foreign Relations. https://www.cfr.org/backgrounder/
made-china-2025-threat-global-trade.
MOF (Ministry of Finance of the People’s Republic of China). 2020. “Announcement of
the Customs Tariff Commission of the State Council on Adjusting Measures to
Impose Tariffs on Certain Imported Commodities Originating in the United
States.” http://gss.mof.gov.cn/gzdt/zhengcefabu/202002/t20200206_3466540.
htm.
Mojtehedzadeh, S. 2016. “Inside Mexico’s ‘Ghost’ Unions.” Toronto Star, May 22. https://
www.thestar.com/news/gta/2016/05/22/inside-mexicos-ghost-unions.html.
OECD (Organization for Economic Cooperation and Development). 2019. “Trade in Fake
Goods Is Now 3.3% of World Trade and Rising.” https://www.oecd.org/
newsroom/trade-in-fake-goods-is-now-33-of-world-trade-and-rising.htm.
Overby, T., S. Snyder, T. Stangarone, and E. Swicord. 2020. Stabilizing the U.S.-Korea
Trade Agenda Under Trump and Moon. Washington: Council on Foreign
Relations. https://efile.fara.gov/docs/3327-InformationalMaterials-20200318-108.pdf.

References | 419

Packard, C. 2020. “Trump and Johnson Can Quickly Strike a Trade Deal—If They Avoid
the Pitfalls.” Foreign Policy, March 11. https://foreignpolicy.com/2020/03/11/
trump-johnson-us-britain-trade-agreement-fta/.
Pierce, J., and P. Schott. 2016. “The Surprisingly Swift Decline of U.S. Manufacturing
Employment.” American Economic Review 106, no. 7: 1632–62.
———. 2020. “Trade Liberalization and Mortality: Evidence from U.S. Counties.”
American Economic Review: Insights 2, no. 1: 47–64.
Schlesinger, J. 2020. “How the Coronavirus Will Reshape World Trade.” Wall Street
Journal, June 19. https://www.wsj.com/articles/
how-the-coronavirus-will-reshape-world-trade-11592592995.
Schott, J. 2019. “Reinventing the Wheel: Phase One of the U.S.-Japan Trade Pact.”
Peterson Institute for International Economics. https://www.piie.com/blogs/
trade-and-investment-policy-watch/
reinventing-wheel-phase-one-us-japan-trade-pact.
Tankersley, J. 2018. “Trump Signs Revised Korean Trade Deal.” New York Times,
September 24. https://www.nytimes.com/2018/09/24/us/politics/southkorea-trump-trade-deal.html.
UBS. 2020. “Supply Chains Are Shifting: How Much and Where?” UBS Evidence Lab.
https://www.ubs.com/global/en/investment-bank/in-focus/covid-19/2020/
supply-chains-are-shifting.html.
UN Comtrade. 2016. “Classification by Broad Economic Categories Rev.5.” https://
unstats.un.org/unsd/trade/classifications/Manual%20of%20the%20Fifth%20
Revision%20of%20the%20BEC%20(Unedited).pdf.
USITC (U.S. International Trade Commission). 2019. “U.S.-Mexico-Canada Trade
Agreement: Likely Impact on the U.S. Economy and on Specific Industry
Sectors.” https://www.usitc.gov/publications/332/pub4889.pdf.
USTR (Office of the United States Trade Representative). 2018a. “Section 301
Investigation Fact Sheet.” https://ustr.gov/about-us/policy-offices/pressoffice/fact-sheets/2018/june/section-301-investigation-fact-sheet.
———. 2018b. “United States–Japan Trade Agreement (USJTA) Negotiations: Summary
of Specific Negotiating Objectives.” https://ustr.gov/sites/default/
files/2018.12.21_Summary_of_U.S.- Japan_Negotiating_Objectives.pdf.
———. 2018c. “United States–Mexico–Canada Trade Fact Sheet: Modernizing NAFTA
into a 21st Century Trade Agreement.” https://ustr.gov/about-us/policyoffices/press-office/fact-sheets/2018/october/united-states%E2%80%93mexi
co%E2%80%93canada-trade-fa-1.
———. 2018d. “New U.S. Trade Policy and National Security Outcomes with the
Republic of Korea.” https://ustr.gov/about-us/policy-offices/press-office/factsheets/2018/march/new-us-trade-policy-and-national.
———. 2018e. “Fact Sheet on U.S.-Korea Free Trade Agreement Outcomes.” https://ustr.
gov/about-us/policy-offices/press-office/fact-sheets/2018/september/
fact-sheet-us-korea-free-trade.

420 |

References

———. 2019a. “2019 National Trade Estimate Report on Foreign Trade Barriers.” https://
ustr.gov/sites/default/files/2019_National_Trade_Estimate_Report.pdf.
———. 2019b. “United States and China Reach Phase One Trade Agreement.” https://
ustr.gov/about-us/policy-offices/press-office/press-releases/2019/december/
united-states-and-china-reach.
———. 2019c. “U.S.-Japan Trade Agreement Text.” https://ustr.gov/countries-regions/
japan-korea-apec/japan/us-japan-trade-agreement-negotiations/
us-japan-trade-agreement-text.
———. 2019d. “U.S.-Japan Digital Trade Agreement Text.” https://ustr.gov/countriesregions/japan-korea-apec/japan/us-japan-trade-agreement-negotiations/
us-japan-digital-trade-agreement-text.
———. 2020a. “United States–Mexico–Canada Agreement.” https://ustr.gov/tradeagreements/free-trade-agreements/united-states-mexico-canada-agreement.
———. 2020b. “Economic and Trade Agreement between the United States of America
and the People’s Republic of China.” https://ustr.gov/sites/default/files/files/
agreements/phase%20one%20agreement/Economic_And_Trade_
Agreement_Between_The_United_States_And_China_Text.pdf.
———. 2020c. “USTR and USDA Release Report on Agricultural Trade between the
United States and China.” https://ustr.gov/about-us/policy-offices/pressoffice/press-releases/2020/october/
ustr-and-usda-release-report-agricultural-trade-between-united-states-andchina.
———. 2020d. “Statement on Call between the United States and China.” https://ustr.
gov/about-us/policy-offices/press-office/press-releases/2020/august/
statement-call-between-united-states-and-china.
———. 2020e. “Fact Sheet on the 2020 National Trade Estimate: Removing Barriers to
U.S. Exports Worldwide.” https://ustr.gov/about-us/policy-offices/pressoffice/fact-sheets/2020/march/
fact-sheet-2020-national-trade-estimate-removing-barriers-us-exportsworldwide.
———. 2020f. “U.S.-U.K. Trade Agreement Negotiations.” https://ustr.gov/countriesregions/europe-middle-east/europe/united-kingdom/
us-uk-trade-agreement-negotiations.
Vaswani, K. 2020. “U.S.-China Trade Deal: Five Things That Aren’t in It.” BBC News,
January 16. https://www.bbc.com/news/business-51130434.
White House. 2020a. “Remarks by President Trump at Signing of the U.S.-China Phase
One Trade Agreement.” https://www.whitehouse.gov/briefings-statements/
remarks-president-trump-signing-u-s-china-phase-one-trade-agreement-2/.
———. 2020b. “Proclamation on Suspension of Entry as Immigrants and
Nonimmigrants of Persons Who Pose a Risk of Transmitting 2019 Novel
Coronavirus.” https://www.whitehouse.gov/presidential-actions/
proclamation-suspension-entry-immigrants-nonimmigrants-persons-poserisk-transmitting-2019-novel-coronavirus/.

References | 421

———. 2020c. “Proclamation—Suspension of Entry as Immigrants and Nonimmigrants
of Certain Additional Persons Who Pose a Risk of Transmitting 2019 Novel
Coronavirus.” https://www.whitehouse.gov/presidential-actions/
proclamation-suspension-entry-immigrants-nonimmigrants-certainadditional-persons-pose-risk-transmitting-2019-novel-coronavirus/.
World Bank. 2020. World Development Report 2020: Trading for Development in the Age
of Global Value Chains. Washington: World Bank. https://www.worldbank.org/
en/publication/wdr2020.
WTO (World Trade Organization). 2020. “The General Agreement on Tariffs and Trade
(GATT 1947).” https://www.wto.org/english/docs_e/legal_e/gatt47_01_e.htm.
Wu, D. 2020. “TSMC Scores Subsidies and Picks Site for $12 Billion U.S. Plant.”
Bloomberg, June 8. https://www.bloomberg.com/news/articles/2020-06-09/
tsmc-confident-of-replacing-any-huawei-orders-lost-to-u-s-curbs
Xie, E. 2020. “Build-up to Coronavirus Lockdown: Inside China’s Decision to Close
Wuhan.” South China Morning Post, April 2. https://www.scmp.com/news/
china/society/article/3078189/
build-coronavirus-lockdown-inside-chinas-decision-close-wuhan.

Chapter 10
BBC (British Broadcasting Corporation). 2020. “U.S. Oil Prices Turn Negative as Demand
Dries Up.” https://www.bbc.com/news/business-52350082.
Bernanke, B. 1983. “Nonmonetary Effects of the Financial Crisis in the Propagation of
the Great Depression.” American Economic Review 73, no. 3: 257–76.
Bordo, M., and J. Haubrich. 2017. “Deep Recessions, Fast Recoveries, and Financial
Crises: Evidence from the American Record.” Economic Inquiry 55: 527–41.
CEA (Council of Economic Advisers). 2018. Economic Report of the President.
Washington: U.S. Government Publishing Office.
———. 2019. Economic Report of the President. Washington: U.S. Government
Publishing Office.
———. 2020. Economic Report of the President. Washington: U.S. Government
Publishing Office.
Chen, Y., A. Ebenstein, M. Greenstone, and H. Li. 2013. “Evidence on the Impact of
Sustained Exposure to Air Pollution on Life Expectancy from China’s Huai
River Policy.” Proceedings of the National Academy of Sciences 110, no. 32:
12936–41.
Coibion, O., Y. Gorodnichenko, and M. Weber. 2020. Labor Markets during the COVID-19
Crisis: A Preliminary View. NBER Working Paper 27017. Cambridge, MA:
National Bureau of Economic Research.
Crust, E., M. Daly, and B. Hobijn. 2020. “The Illusion of Wage Growth.” Federal Reserve
Bank of San Francisco Economic Letter.

422 |

References

Deschenes, O., M. Greenstone, and J. Shapiro. 2017. “Defensive Investments and the
Demand for Air Quality: Evidence from the NOx Budget Program.” American
Economic Review 107, no. 10: 2958–89.
EIA (Energy Information Administration). 2020. “Short-Term Energy Outlook.”
December 8. https://www.eia.gov/outlooks/steo/.
Fernández-Villaverde, J., and C. Jones. 2020. Macroeconomic Outcomes and COVID-19: A
Progress Report. NBER Working Paper 28004. Cambridge, MA: National
Bureau of Economic Research.
Gascon, C., and J. Haas. 2020. “The Impact of COVID-19 on the Residential Real Estate
Market.” Federal Reserve Bank of Saint Louis.
Gicheva, D. 2016. “Student Loans or Marriage? A Look at the Highly Educated.”
Economics of Education Review 53: 207–16.
Gicheva, D., and J. Thompson. 2015. “The Effects of Student Loans on Long-Term
Household Financial Stability.” In Student Loans and the Dynamics of Debt,
edited by B. Hershbein and K. Hollenbeck. Kalamazoo: W. E. Upjohn Institute
for Employment Research.
Goodspeed, T., and P. Navarro. “The White House Favors a Bridge to Recovery.” Wall
Street Journal, December 2. https://www.wsj.com/articles/
the-white-house-favors-a-bridge-to-recovery-11606950989.
Hall, R., and M., Kudlyak. 2020. Why Has the U.S. Economy Recovered So Consistently
from Every Recession in the Past 70 Years? NBER Working Paper 27234.
Cambridge, MA: National Bureau of Economic Research.
Hamermesh, D. 1989. “What Do We Know About Worker Displacement in the U.S.?”
Industrial Relations: A Journal of Economy and Society 28, no. 1: 51–59.
He, J., H. Liu, and A. Salvo. 2019. “Severe Air Pollution and Labor Productivity: Evidence
from Industrial Towns in China.” American Economic Journal: Applied
Economics 11, no. 1: 173–201.
Hedlund, A. 2019. “Failure to Launch: Housing, Debt Overhang, and the Inflation
Option.” American Economic Journal: Macroeconomics 11, no. 2: 228–74.
IMF (International Monetary Fund). 2020a. World Economic Outlook: A Long and Difficult
Ascent. Washington: IMF. https://www.imf.org/-/media/Files/Publications/
WEO/2020/October/English/text.ashx.
———. 2020b. World Economic Outlook Update: A Crisis Like No Other, An Uncertain
Recovery. Washington: IMF. https://www.imf.org/-/media/Files/Publications/
WEO/2020/Update/June/English/WEOENG202006.ashx.
Kozlowski, J., L. Veldkamp, and V. Venkateswaran. 2020. Scarring Body and Mind: The
Long-Term Belief-Scarring Effects of COVID-19. NBER Working Paper 27439.
Cambridge, MA: National Bureau of Economic Research.
Mezza, A., D. Ringo, S. Sherlund, and K. Sommer. 2020. “Student Loans and
Homeownership.” Journal of Labor Economics 38, no. 1: 215–60.
OECD (Organization for Economic Cooperation and Development). 2020. OECD
Economic Outlook. Vol. 2020, Issue 2. https://www.oecd-ilibrary.org/

References | 423

deliver/39a88ab1-en.pdf?itemId=%2Fcontent%2Fpublication%2F39a
88ab1-en&mimeType=pdf.
O’Hara, M., and A. Zhou. 2020. “Anatomy of a Liquidity Crisis: Corporate Bonds in the
COVID-19 Crisis.” Journal of Financial Economics, forthcoming.
Pierce, J., and P. Schott. 2020. “Trade Liberalization and Mortality: Evidence from U.S.
Counties.” American Economic Review: Insights 2, no. 1: 47–64.
Rothstein, J., and C. Rouse. 2011. “Constrained after College: Student Loans and EarlyCareer Occupational Choices.” Journal of Public Economics 95, nos. 1–2:
149–63.
Ruhm, C. 1991. “Are Workers Permanently Scarred By Job Displacements?” American
Economic Review 81, no. 1: 319–24.
Sharpe, S., and X. Zhou, 2020. “The Corporate Bond Market Crises and the Government
Response.” FEDS Notes 2020-10-07-2. Board of Governors of the Federal
Reserve System.
World Bank. 2020. “Reversals of Fortune: Poverty and Shared Prosperity 2020.” https://
openknowledge.worldbank.org/bitstream/handle/10986/34496/9781464816
024.pdf.

Chapter 11
AASHTO (American Association of State Highway and Transportation Officials) and
TRIP. 2009. “Rough Roads Ahead: Fix Them Now or Pay For It Later.” Road
Information Program. https://t2.unh.edu/sites/t2.unh.edu/files/documents/
publications/RoughRoads_FullReport.pdf
AAUP (American Association of University Professors). 2020. “2019–20 Faculty
Compensation Survey Results.” https://www.aaup.
org/2019-20-faculty-compensation-survey-results.
―――. No date. “Explanation of Statistical Data.” https://www.aaup.org/
explanation-statistical-data-6.
Abbott, M., and C. Beach. 2011. “Do Admission Criteria and Economic Recessions Affect
Immigrant Earnings?” Institute for Research on Public Policy Working Paper.
https://irpp.org/research-studies/
do-admission-criteria-and-economic-recessions-affect-immigrant-earnings/
Abraham, K., and M. Kearney. 2020. “Explaining the Decline in the U.S. Employment-toPopulation Ratio: A Review of the Evidence.” Journal of Economic Literature
58, no. 3: 585–643.
Acemoglu, D., U. Akcigit, H. Alp, N. Bloom, and W. Kerr. 2018. “Innovation, Realocation,
and Growth.” American Economic Review 108, no. 11: 3450–91.
Agency for Healthcare Research and Quality. 2017. “Physician Burnout.”
https://www.ahrq.gov/prevention/clinician/ahrq-works/burnout/index.htm.
Aguirre, D., L. Hoteit, C. Rupp, and K. Sabbagh. 2012. “Empowering the Third Billion.
Women and the World of Work in 2012.” Booz and Company. https://www.

424 |

References

hrtoday.ch/sites/hrtoday.ch/files/article_inline_images/BoozCo_
Empowering-the-Third-Billion_Full-Report.pdf.
Altig, D, A. Auerbach, L. Kotlikoff, E. Ilin, and V. Ye. 2020. Marginal Net Taxation of
Americans’ Labor Supply. NBER Working Paper 27164. Cambridge, MA:
National Bureau of Economic Research.
American Society of Civil Engineers. 2016. Failure to Act: Closing the Infrastructure
Investment Gap for America’s Economic Future. Boston: Economic
Development Research Group.
Arezki, R, P. Bolton, S. Peters, F. Samama, and J. Stiglitz. 2017. “From Global Savings
Glut to Financing Infrastructure.” Economic Policy 32, no. 90: 221–61.
Astebro, T., S. Braguinsky, and Y. Ding. 2020. Declining Business Dynamism among Our
Best Opportunities: The Role of the Burden of Knowledge. NBER Working Paper
27787. Cambridge, MA: National Bureau of Economic Research. https://www.
nber.org/system/files/working_papers/w27787/w27787.pdf.
Bailey, M., T. Byker, E. Patel, and S. Ramnath. 2019. The Long-Term Effects of California’s
2004 Paid Family Leave Act on Women’s Careers: Evidence from U.S. Tax Data.
NBER Working Paper 26416. Cambridge, MA: National Bureau of Economic
Research.
Baldwin, R. 2006. “Failure of the WTO Ministerial Conference at Cancun: Reasons and
Remedies.” World Economy 29, no. 6: 677–96.
Bartel, A., S. Kim, J. Nam, M. Rossin-Slater, C. Ruhm, and J. Waldfogel. 2019. “Racial
and Ethnic Disparities in Access to and Use of Paid Family and Medical Leave:
Evidence from Four Nationally Representative Datasets.” Monthly Labor
Review, January.
Becker, G. 1962. “Investment in Human Capital: A Theoretical Analysis.” Journal of
Political Economy 70, no. 5: 9–49.
Bick, A., and N. Fuchs-Schündeln. 2017. “Quantifying the Disincentive Effects of Joint
Taxation on Married Women’s Labor Supply.” American Economic Review 107,
no. 5: 100–104.
Blau, F., and L. Kahn. 2013. “Female Labor Supply: Why Is the United States Falling
Behind?” American Economic Review: Papers and Proceedings 103, no. 3:
251–56.
Bloom, N., C. Jones, J. Van Reenen, and M. Webb. 2020. “Are Idea Getting Harder to
Find?” American Economic Review 110, no. 4: 1104–44.
Borella, M., M. De Nardi, and F. Yang. 2019a. Are Marriage-Related Taxes and Social
Security Benefits Holding Back Female Labor Supply? NBER Working Paper
26097. Cambridge, MA: National Bureau of Economic Research. https://www.
nber.org/papers/w26097.
―――. 2019b. “Marriage-Related Taxes and Social Security Benefits Are Holding Back
Women’s Labour Supply in the U.S.” Vox EU and Centre for Economic Policy
Research. https://voxeu.org/article/
marriage-related-taxes-social-security-benefits-and-women-s-laboursupply-us.

References | 425

Borjas, G. 2003. “The Labor Demand Curve Is Downward Sloping: Reexamining the
Impact of Immigration on the Labor Market.” Quarterly Journal of Economics
118, no. 4: 1335–74.
―――. 2017. “Why Trump’s New Immigration Bill Makes Sense; Hey, It Works in
Canada.” Politico, August 4. https://www.politico.com/magazine/
story/2017/08/04/why-trumps-new-immigration-bill-makes-sense-215457
―――. 2019. Immigration and Economic Growth. NBER Working Paper 25836.
Cambridge, MA: National Bureau of Economic Research.
Bradford, A. 2014. “How International Institutions Evolve.” Chicago Journal of
International Law 15, no. 5: 47–83.
Bronson, M., and M. Mazzocco. 2018. “Taxation and Household Decisions: An
Intertemporal Analysis.” https://wpcarey.asu.edu/sites/default/files/taxation_
sep_2018.pdf.
Brooks, L., N. Gendron-Carrier, and G. Rua. 2018. “The Local Impact of
Containerization.” Finance and Economics Discussion Series 2018-045.

Full text of Economic Report of the President : 2021

FRASER