Full text of Economic Report of the President : 2020

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

– President Donald J. Trump

Economic Report
of the President

Though the American economy is stronger than
ever, my Administration’s work is not yet done.
With a continued focus on policies that increase
economic growth, promote opportunity, and
uplift our workers, there is no limit on how great
America can be.

Economic Report
of the President

February 2020

Together with
The Annual Report
of the
Council of Economic Advisers
February 2020

250-840_cover.indd 1

2/10/20 9:07 AM

Economic Report
of the President
Together with
The Annual Report
of the
Council of Economic Advisers
February 2020

250-840_text_.pdf 1

2/7/20 3:46 PM

250-840_text_.pdf 2

2/7/20 3:46 PM

x

Contents
Economic Report of the President. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
The Annual Report of the Council of Economic Advisers.. . . . . . . . . . . . . . . . . . . . 9
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Part I: The Longest Expansion on Record
Chapter 1:

The Great Expansion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Chapter 2:
		

Economic Growth Benefits Historically Disadvantaged
Americans. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

Chapter 3:

Regulatory Reform Unleashes the Economy. . . . . . . . . . . . . . . . . 105

Chapter 4:

Energy: Innovation and Independence. . . . . . . . . . . . . . . . . . . . . . 137

Chapter 5:

Free-Market Healthcare Promotes Choice and Competition. . . . 171
Part II: Evaluating and Addressing Threats to the Expansion

Chapter 6:

Evaluating the Risk of Declining Competition. . . . . . . . . . . . . . . . 199

Chapter 7:

Understanding the Opioid Crisis. . . . . . . . . . . . . . . . . . . . . . . . . . . 227

Chapter 8:

Expanding Affordable Housing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Part III: The Economic Outlook

Chapter 9:

The Outlook for a Continued Expansion. . . . . . . . . . . . . . . . . . . . . 295

References

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303

Appendix A	Report to the President on the Activities of the Council of
Economic Advisers During 2019. . . . . . . . . . . . . . . . . . . . . . . . . . . . 343
Appendix B	Statistical Tables Relating to Income, Employment, and
Production.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355

____________
*For a detailed table of contents of the Council’s Report, see page 19.

iii

250-840_text_.pdf 3

2/7/20 3:46 PM

250-840_text_.pdf 4

2/7/20 3:46 PM

Economic Report of the President

1

250-840_text_.pdf 5

2/7/20 3:46 PM

250-840_text_.pdf 6

2/7/20 3:46 PM

x
Economic Report of the President
To the Congress of the United States:
Over the past three years, my Administration has championed policies to
restore the United States’ economic strength, propelling growth to levels far
exceeding preelection expectations. These results did not come about by
accident. Instead, they were supported by our foundational pillars for economic growth that put Americans first, including tax cuts, deregulation, energy
independence, and trade renegotiation. Our success has created a historically
strong labor market and greater economic security for millions of American
families.

The Transformative Power of Work
My Administration’s focus on economic growth comes from a deep appreciation of the power of work to drive the economy and transform lives. The truth
is, jobs do not just provide paychecks; they give people meaning, allow them
to engage with their communities, and help them reach their true potential. As
we have shown, the right policies offer Americans paths to self-reliance rather
than trapping them in reliance on government programs.
The unemployment rate is 3.5 percent, the lowest it has been in 50 years.
Since I came into office, labor force participation is up and wages are growing
fastest for historically disadvantaged workers, reversing the trends seen under
the previous administration. Under my Administration, and for the first time
on record, job openings exceeded people looking for work, with 1 million more
open jobs than job seekers at the end of 2019. Because of record-low unemployment rates across demographic categories and continued job creation,
people from all backgrounds can more easily find work, build their skills, and
grow their incomes.
In today’s tight labor market, employers realize the vast potential of
many individuals whom they may have previously overlooked. This includes
those facing long-term unemployment, balancing family responsibilities,
thinking they lack necessary job skills, overcoming substance abuse, returning from the justice system, or living in poverty. It is also encouraging those
individuals to put themselves back in the workforce. My Administration has
placed a special focus on these forgotten Americans because every individual
deserves to experience the dignity that comes through work.
In the fourth quarter of 2019, three quarters of workers entering employment came from outside the labor force rather than from unemployment, the
highest share in the series’ history. As paid parental leave spreads across the
country, including the expansion to Federal workers that I signed in December,

Economic Report of the President | 3

250-840_text_.pdf 7

2/7/20 3:46 PM

parents will have more freedom to choose a balance between working outside
the home and caring for their children. And in another encouraging sign that
people previously on the sidelines will continue entering the workforce, more
than 420 companies have signed the Pledge to America’s Workers. These
companies have pledged to create upward of 14 million new job and training
opportunities for current and future employees over the next five years.
Apprenticeships are one way for these companies to deliver on their
pledges, and expanding apprenticeships has been a top priority since I took
office. During my presidency, more than 680,000 new apprenticeships have
been created. To have a labor market that works for everyone, the Federal
Government must encourage a variety of paths for people to get the skills they
need to build family-sustaining careers.
Although all sectors benefit from more apprenticeships, my Administration
knows that manufacturing is a pillar of the American economy. Manufacturing
spurs innovation and fuels economic growth, which is why I am so pleased that
more than 500,000 manufacturing jobs have been created since my election.
Rather than still shrinking, American manufacturing is now growing again.
Critically, wages for nonsupervisory and production workers are rising at an
even higher rate than managers’ wages.
Renegotiated or new trade deals with Canada and Mexico, China, South
Korea, and Japan will modernize international trade and create freer, fairer,
and more reciprocal trade between the United States and our largest trading
partners, allowing the manufacturing renaissance to continue. Trade deals are
in development with the United Kingdom and the European Union, among
other countries that need access to the coveted United States market. These
deals will both expand United States markets abroad and keep businesses here
in America, which means keeping jobs here in America.
I have the deepest respect for America’s workers and job creators who
have made this economic boom possible. That is why we are fighting back
against other nations that have exploited the pioneering spirit of our country’s
entrepreneurs. Through combating intellectual property theft and unfair trade
deals, along with leading the way on 5G development and deployment, my
Administration is standing up to countries around the world to give American
job creators the freedom to innovate and make life better for their fellow citizens. These proactive steps will benefit everyone, from large companies that
employ hundreds or thousands of Americans to budding entrepreneurs trying
to turn their ideas into reality.
The labor market experiences that people are gaining today will change
the trajectories of their lives—and those of their children—for years to come.
No matter their pasts, people deserve agency over their own lives, and my
Administration will never tell Americans that they cannot or do not deserve the
ability to work and earn a living for themselves and their families.

4 |

250-840_text_.pdf 8

Economic Report of the President

2/7/20 3:46 PM

Previously Forgotten Americans Are Forgotten No More
America’s labor market successes are also helping us defeat the opioid crisis.
While the causes of the crisis are multifaceted, work must play an integral role
in any solution. Research shows that holding a job is a key factor in helping
people overcome drug addiction. Over the rest of my presidency, I will continue
to promote policies that beat back this deadly crisis and encourage work for
Americans who are rebuilding their lives after struggling with addiction.
Because of my Administration’s aggressive efforts to end the overprescription of opioids, promote effective treatment, and secure the border,
the tide is finally turning on the opioid crisis. Overdose deaths and first-time
users are down, but that does not mean the crisis is over. Failure is not an
option when it comes to helping people avoid the pain and suffering caused
by addiction.
Unfortunately, the largest drug crisis in our history has left many people
with criminal records. After someone leaves the justice system, they face two
options: find honest work and successfully reenter society, or stay out of work
and face the increased likelihood of committing another crime. Finding work is
one of the top indicators of whether someone who commits a crime will turn
his or her life around and live crime-free. This is why work is not just essential
for reforming individuals; it is also necessary for promoting public safety.
Beyond signing the landmark First Step Act to promote public safety and make
America’s justice system fairer, my Administration is also putting substantial
resources behind programs that improve employment outcomes for the formerly incarcerated. Likewise, criminal justice reform that emphasizes work
helps break the cycle of generational poverty.
In 2018 alone, 1.4 million Americans were lifted out of poverty, and
the poverty rate fell to its lowest level since 2001. For African Americans and
Hispanic Americans, poverty rates are at historic lows, and the poverty rate for
single mothers and children is falling much faster than the average. Since I took
office, food insecurity has fallen and nearly 7 million people have been lifted
off food stamps. Beneficiaries entering the labor market or increasing their
incomes through work is likely driving falling enrollment in Medicaid, TANF,
and disability insurance.
These Americans are not simply rising out of poverty; they are building
careers of which they and their families can be proud. Wages are rising fastest for people with the lowest incomes, meaning people currently working in
lower-paying jobs will not have low incomes for long. Getting that first job is
critical, because it serves as a foundation for progressively better jobs over a
worker’s career.
A commitment to the transformative power of work is why I signed an
Executive Order instructing agencies to reduce dependence on welfare programs by encouraging work. Less than 3 percent of people who work full time

Economic Report of the President

250-840_text_.pdf 9

| 5

2/7/20 3:46 PM

live in poverty. Individuals will not be able to build the lives they want through
welfare alone: Work is a necessary condition for upward mobility.
While strengthening and expanding work requirements for public assistance programs lead people to reenter the workforce and increase their
household incomes, work requirements are most effective when employers
are hiring. This is one reason why my Administration emphasizes policies that
lead to job creation.

Pro-Growth Policies Are Pro-Worker Policies
One foundational policy that continues to drive job creation is tax reform. Since
the Tax Cuts and Jobs Act—the biggest package of tax cuts and tax reforms in
our country’s history—took effect, more than 4 million jobs have been created
and economic growth has beaten previous projections. America’s outdated tax
code drove away businesses and investment, but tax reform has brought rates
down and made the United States globally competitive again.
Many workers saw bonuses and raises immediately after tax reform, and
nearly 40 million American families received an average benefit of $2,200 in
2019 from doubling the child tax credit. Yet the biggest payoff is still to come.
Tax reform put an end to America’s counterproductive policy of punishing
business investments, which means that workers will see even greater benefits
once these investments pay off.
My Administration has also prioritized healthcare reforms that make
the system more competitive and, therefore, more affordable. We are giving
patients increased choice and control, and protecting the high-quality care
that Americans expect and deserve. Healthcare is a top priority because healthcare costs are among the top annual expenses for American families. Under my
Administration, the Food and Drug Administration has approved more generic
drugs than ever before in United States history and enhanced its approval
process for new, lifesaving drugs. This past year, prescription drug prices experienced the largest year-over-year decline in more than 50 years.
Whether it is through reforms that bring choice to Veterans Administration
care, promote Health Reimbursement Arrangements, or give terminally ill
patients access to potentially lifesaving drugs, among many other successes,
every healthcare reform that lowers costs and increases quality allows American
workers to live longer, healthier lives and keep more of their paychecks.
Tax cuts and healthcare reforms put more money in the hands of working
families and job creators, creating a virtuous cycle of even more jobs and even
higher paychecks. On the other hand, when regulations limit individuals’ ability
to experience the dignity that comes through work, those regulations deserve
additional scrutiny. Over the previous decades, the Federal Government has
disproportionately regulated sectors of the economy—like energy and manufacturing—that offer fulfilling, blue collar jobs for the majority of Americans
who do not have a college degree. These misguided policy decisions imposed
6 |

250-840_text_.pdf 10

Economic Report of the President

2/7/20 3:46 PM

real-world costs that created barriers to success and prosperity for hardworking Americans. Those days are over.
American energy powers our cities and towns, empowers innovators, and
ultimately drives our economy. Energy companies across the world are ready
to build in our Nation, and permitting reform that cuts red tape shows that
we welcome their investments. My Administration continues to support the
energy industry’s growth by removing unnecessary regulations and unleashing America’s vast natural and human resources. Through these actions, the
United States is now on track to be a net exporter of crude oil and natural gas
for all of 2020, a major milestone not achieved in at least 70 years. In addition
to being the world’s largest natural gas producer, we also became the world’s
top crude oil producer in 2018.
The positive records of our energy boom are widespread. Energy production has created jobs in areas of the United States where job opportunities
were scarce. It also provides enormous benefits to families across the Nation
by lowering energy prices. And it further distances us from geopolitical foes
who wish to cause us harm. More jobs, lower costs, and American dominance—
these are the predictable results of our pro-growth policies.
Many pundits and Washington insiders laughed when I promised to cut
two regulations for every new regulation. They were correct that two-for-one
was the wrong goal. Instead, the Federal Government has cut more than
seven regulations for every significant new regulation. After only three years,
my Administration has already cut more regulations than any other in United
States history, and we have put the brakes on an endless assault of new, costly
actions by Federal agencies.
Our commitment to regulatory reform stems from the simple truth that
the vast majority of business owners want to do the right thing, comply with
the law, and treat their workers fairly. The Federal Government ignored this
reality for far too long and abused its authority to go after businesses, especially small businesses and entrepreneurs, in ways that can only be described
as arbitrary and abusive.
To promote regulatory fairness, I signed two Executive Orders that will
improve Federal agencies’ transparency and fairness while holding them
accountable for their actions. Agencies will now need to give people fair
notice and a chance to respond to any Federal complaint filed against them.
Furthermore, the rules agencies enforce will no longer be secret, because all
agencies’ interpretations of rules will need to be made publicly accessible.
Additionally, significant interpretations of rules will need to go through the
public review process that is central to a flourishing democracy. Deregulation
and increased transparency will save job creators money, leading to more hiring and higher paychecks.
Every American, no matter his or her background, can share in the dignity
of work. The era of putting American workers second and doubling down on
Economic Report of the President | 7

250-840_text_.pdf 11

2/7/20 3:46 PM

the failed Federal policies of the past is over. While job creation during my
Presidency has surpassed expectations, the credit belongs to the job creators
and workers who risk everything and devote themselves to building a better
future for themselves, their families, and their Nation. The Federal Government
does not create jobs; hardworking Americans create jobs. My Administration’s
role is to follow our foundational policy pillars and allow our job creators and
workers to do what they do best.
As the following Report shows, because of the strength, resiliency, and
determination of the United States workforce, which is the envy of the world,
my pro-growth policies continue producing unquestionably positive results
for the economy. The Report also makes it clear that, though the American
economy is stronger than ever, my Administration’s work is not yet done. With
a continued focus on policies that increase economic growth, promote opportunity, and uplift our workers, there is no limit on how great America can be.

The White House
February 2020

8 |

250-840_text_.pdf 12

Economic Report of the President

2/7/20 3:46 PM

The Annual Report
of the

Council of Economic Advisers

9

250-840_text_.pdf 13

2/7/20 3:46 PM

250-840_text_.pdf 14

2/7/20 3:46 PM

x

Letter of Transmittal
Council of Economic Advisers
Washington, February 20, 2020
Mr. President:
The Council of Economic Advisers herewith submits its 2020 Annual
Report in accordance with the Employment Act of 1946, as amended by the Full
Employment and Balanced Growth Act of 1978.
Sincerely yours,

Tomas J. Philipson
Acting Chairman

Tyler B. Goodspeed
Member

Economic Report of the President | 11

250-840_text_.pdf 15

2/7/20 3:46 PM

250-840_text_.pdf 16

2/7/20 3:46 PM

Introduction
Three years into the Trump Administration, the U.S. economy continues to
outperform expectations across numerous metrics, with growth in output,
employment, and employee compensation all exceeding pre-2017 forecasts.
The evident success of the Administration’s economic policy agenda demonstrates that its foundational policy pillars are enabling the U.S. economy to
overcome structural trends that were previously suppressing growth.
During the four quarters of 2019, real gross domestic product grew
0.7 percentage point faster than had been projected by the independent
Congressional Budget Office’s (CBO) August 2016 projections. As shown in
figures I-1 and I-2, the U.S. labor market added 2.1 million new jobs—2.0 million
more than projected in 2016—bringing the civilian unemployment rate down
to 3.5 percent, which is its lowest level since 1969 (and 1.4 percentage points
below 2016 CBO projections).1 Higher pay accompanied abundant job vacancies, as employee compensation rose to 1.4 percent above the 2016 forecast,
implying an additional $1,800 in compensation per household.
In July 2019, the current expansion of the U.S. economy became the
longest on record. Contrary to expectations that the expansion would slow
as it matured, economic output has accelerated over the past 3 years relative
to the preceding 7½ years, with output growth rising from 2.2 to 2.5 percent
at a compound annual rate. In the first three quarters of 2019, U.S. economic
growth was the highest among the Group of Seven countries.
Reflecting this outperformance of expectations, in the first five chapters
of this Report we present evidence that the Trump Administration’s foundational policy pillars are continuing to deliver economic results. In particular, we
highlight the role of the Administration’s prioritization of economic efficiency
and pro-market reforms in the realms of tax, labor, regulation, energy, and
healthcare in elevating the growth potential of the U.S. economy and increasing the well-being of those previously left behind during the current expansion.
In the subsequent three chapters, we then identify several challenges to
continued growth. Efforts to address these obstacles include ensuring that U.S.
markets remain economically fair and competitive, combating the ongoing
threat of widespread opioid addiction, and addressing the overregulation of
housing markets. We conclude by setting forth the Administration’s long-run,
policy-inclusive economic projections, and highlighting potential risks to the
outlook.
We begin in chapter 1 by documenting that, despite strong headwinds from the global economy and several idiosyncratic adverse shocks,
Administration policies have helped to keep the U.S. economy resilient. As
a result, output has grown at the fastest rate among the Group of Seven
1 In preparing this Economic Report of the President, data available as of January 30, 2020, were
incorporated as publicly reported and are reflected in the chapters that follow.

Economic Report of the President

250-840_text_.pdf 17

| 13

2/7/20 3:46 PM

$"0- 
 Ҋрѵ
# 
/0'
) (+'*4( )/
/ 
i)
-$*0.
0-/ -.
1 -.u.
/# 
0"0./ спрх
/ ,
+ -
/# 
 
)
Ѷ
спрс–рш
спрх
 ' /$*)

спрсѷу–спрхѷс, /0'

0"0./
спрхѶ


спрхѷт–спршѷу,
/0'

Percent

ч

спршѷу

ц
х

ф
у
т
спрсѷу спртѷу спруѷу спрфѷу спрхѷу спрцѷу спрчѷу спршѷу
*0- .ѷ
0- 0 *!
 *-
//$./$.
җ ҘѸ
*)"- ..$*)'
0" /
!!$ 
җҘѵ

Figure I-2. Actual Nonfarm Payrolls versus the August 2016 Payroll,
per the CBO, 2012–19
спрх
 ' /$*)

спрсѷу–спрхѷс,
/0'

0"0./
спрхѶ


спрхѷт–спршѷу,
/0'

Hundreds of thousands (seasonally adjusted)

рфф

спршѷу

рфп

руф
руп
ртф

ртп
спрсѷу спртѷу спруѷу спрфѷу спрхѷу спрцѷу спрчѷу спршѷу
*0- .ѷ
0- 0 *!
 *-
//$./$.
җ ҘѸ
*)"- ..$*)'
0" /
!!$ 
җҘѸ


'0'/$*).ѵ

14 |

250-840_text_.pdf 18

Economic Report of the President

2/7/20 3:46 PM

economies in the past year. During 2019, several macroeconomic indicators—
including consumer spending, productivity, and labor share of income—grew
at faster rates than preelection projections. The labor market also tightened
further, even after strong gains during the previous two years. During this
Administration, the unemployment rate hit a 50-year low, and the number of
job openings exceeded job seekers for the first time in recorded U.S. history,
which has helped to pull potential workers into the labor force and boost real
wages. The stabilization of labor force participation after years of decline,
particularly among prime-age workers, has also boosted long-term potential
output.
We continue to evaluate the performance of the U.S. labor market in
chapter 2, paying particular attention to how the Administration’s pro-growth
agenda has disproportionately benefited those previously left behind during
the current expansion. We document how, in stark contrast to the expansion
through 2016, policies that both raised labor demand and incentivized employers to invest more in their workers have resulted in wage gains for historically
disadvantaged Americans. Average wage growth for workers now outpaces
wage growth for supervisors; wage growth for individuals at the 10th percentile
of the income distribution now outpaces wage growth for individuals at the
90th percentile; wage growth for those without a college degree now outpaces
wage growth for those with a college degree; and wage growth for African
Americans now outpaces wage growth for white Americans. With monthly
payroll employment growth outpacing that required to maintain a stable
employment-to-population ratio, we also document the extent to which the
U.S. economy is pulling millions back into the labor force and out of poverty.
Looking ahead, we outline the Administration’s continued prioritization
of initiatives aimed at promoting alternative paths to work, supporting onthe-job training and reskilling, reducing recidivism, combating opioid abuse,
expanding access to affordable childcare, and enabling economic growth that
provides expanded employment opportunities for every American who seeks
work.
In chapter 3, we analyze the effects of the Administration’s regulatory
reform agenda. We estimate that after 5 to 10 years, the Administration’s
approach to Federal regulation will have raised real incomes by $3,100 per
household per year, with 20 notable Federal deregulatory actions alone saving
American consumers and businesses about $220 billion per year once they go
into full effect, which will raise real incomes by about 1.3 percent. We further
calculate that the ongoing introduction of costly regulations had previously
been subtracting 0.2 percent a year from real incomes. By increasing competition, productivity, and wages, and reducing the prices of consumer goods,
the Administration’s approach to regulation is raising real incomes while
maintaining regulatory protections for workers, public health, safety, and the
environment.
Economic Report of the President

250-840_text_.pdf 19

| 15

2/7/20 3:46 PM

Continuing the focus on regulation, in chapter 4 we focus specifically on
U.S. energy markets. By lowering prices, the CEA estimates that the shale revolution saves the average family of four $2,500 annually. Because low-income
households spend a larger share of their income on energy bills, they benefit
disproportionately from lower energy prices: shale-driven savings represent a
much larger percentage of income for the poorest fifth of households than for
the richest fifth. At the same time, shale-driven production growth has affected
U.S. energy independence. This goal, initiated by President Nixon and pursued
by every subsequent Administration, was finally achieved under the Trump
Administration. In September 2019, the United States became a net exporter of
petroleum, and the United States is projected to remain a net exporter for all of
2020, for the first time since at least 1949. We estimate that from 2005 to 2018,
the shale revolution in particular was responsible for reducing carbon dioxide
emissions in the electric power sector by 21 percent. Finally, we demonstrate
how, by limiting unnecessary constraints on private innovation and investment, the Administration’s approach to eliminating excessive regulation of
energy markets supports further unleashing of the country’s abundant human
and energy resources.
In chapter 5, we identify government barriers to market competition in
healthcare that increase prices, reduce innovation, and hinder improvements
in quality. We also summarize the achievements and expected effects of the
Administration’s health policy initiatives to reduce these impediments and
facilitate greater competition in healthcare markets. The Administration’s
reforms aim to foster a healthcare system that delivers high-quality services at
affordable prices through greater choice, competition, and consumer-directed
spending, in contrast to government mandates that too often reduce consumer
choice in healthcare markets and increase premiums. The Administration has
addressed many of these problems through a series of Executive Orders, regulatory reforms, and legislation.
Turning to potential obstacles, in chapter 6, we analyze concerns about
possible trends in market competition, recognizing the vital role that competition plays in economic growth, promoting innovation and entrepreneurship,
and serving consumers. We find that the best available evidence suggests there
is no need to rewrite the Federal Government’s antitrust rules. Because Federal
enforcement agencies are already empowered with a flexible legal framework,
they possess the necessary tools to promote economic dynamism. Ongoing
investigations and resolved cases show that these agencies are well equipped
to handle the competition challenges posed by the changing U.S. economy. We
conclude that in addition to vigorously combating anticompetitive behavior
from companies using existing tools, the Administration will focus on changing government policies that create an unfair playing field. As the recent
historic regulatory reform across American industries has shown, eliminating

16 |

250-840_text_.pdf 20

Economic Report of the President

2/7/20 3:46 PM

government-imposed barriers to innovation leads to increased competition,
stronger economic growth, and a revitalized private sector.
In chapter 7, we analyze the ongoing threat of widespread opioid addiction that, since 2000, has been responsible for more than 400,000 deaths. We
find that actions taken by the Administration to lower the supply of opioids,
reduce new demand for opioids, and treat those with current opioid use
disorder may have contributed to a flattening in overdose deaths involving
opioids. Recognizing that understanding the origins of the crisis is essential
to effectively combating it, we find that a first wave of the crisis, from 2001 to
2010, was driven in large part by steep declines in out-of-pocket prescription
opioid prices. Prices fell due to expanded government healthcare coverage,
as well as to the increased availability of prescription opioids due to pain
management practices that encouraged liberalized dispensing practices by
doctors. We then find that a second wave of the opioid crisis, starting in 2010,
likely began because of efforts to limit the supply of the powerful prescription
opioid OxyContin, an unintended consequence of which was the creation of a
large illicit market for the development and sale of cheaper illegal substitutes.
In chapter 8, we study the challenges posed by rising housing unaffordability in some U.S. real estate markets. We find that a key driver of the housing
unaffordability problem is the overregulation of housing markets by State and
local governments, which limits supply. By driving up home prices, overregulation adversely affects low-income Americans in particular, who spend
the largest share of their income on housing. Among 11 particularly supplyconstrained metropolitan areas, we estimate that regulatory reform would
increase the housing supply and decrease rents enough to reduce homelessness by 31 percent on average. In addition, we find that overregulation of housing markets has broader negative effects on all Americans by reducing labor
mobility and thus productivity growth, amplifying inequality across regions
and workers, and harming the environment by forcing longer commutes. We
conclude by documenting the Administration’s actions to address the housing unaffordability challenge by incentivizing State and local governments to
increase housing supply in supply-constrained areas and by establishing the
White House Council on Eliminating Regulatory Barriers to Affordable Housing.
Finally, in chapter 9, we present the Trump Administration’s full, policyinclusive economic forecast for the next 11 years, including risks to the economic outlook. Overall, assuming full implementation of the Administration’s
economic policy agenda, we project that real U.S. economic output will grow
at an average annual rate of 2.9 percent between 2019 and 2030. We expect
growth to moderate, from 3.0 percent in 2020 to 2.8 percent in the latter half of
the budget window, as the capital-to-output ratio asymptotically approaches
its new, postcorporate tax reform steady state and as the near-term effects of
the Tax Cuts and Jobs Act’s individual provisions on the rate of growth dissipate into a permanent-level effect. Partially offsetting this moderation are the
Economic Report of the President | 17

250-840_text_.pdf 21

2/7/20 3:46 PM

expected positive contributions to growth from enacting the Administration’s
infrastructure plan, making permanent the individual provisions of the Tax
Cuts and Jobs Act, reforming the U.S. immigration system, continuing deregulatory actions, improving trade deals with international trading partners, and
incentivizing higher labor force participation through additional labor market
reforms.

18 |

250-840_text_.pdf 22

Economic Report of the President

2/7/20 3:46 PM

x

Contents
Part I: The Longest Expansion on Record. . . . . . . . . . . . . . . . . . . . . . . . . . 29
Chapter 1: The Great Expansion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Productivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Wages and Income. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Consumer Spending. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Investment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Inflation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Price Inflation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Different Measures of Inflation: The CPI, Chained CPI, and PCE Price
Index and Their Cores . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Hourly Compensation Inflation, Productivity Growth, and Stable Inflation.
Deregulation and Inflation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

51
51
53
55
57

The Global Macroeconomic Situation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
The U.S. Dollar and Monetary Policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Domestic Headwinds. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Chapter 2: Economic Growth Benefits Historically Disadvantaged
Americans. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
Shared Prosperity from Strong Economic Growth.. . . . . . . . . . . . . . . . . . . . . . .
The Current State of the Labor Market. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Demographic Change and Labor Force Statistics. . . . . . . . . . . . . . . . . . . . .
Wage and Income Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Poverty and Inequality. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Health Insurance and Medicaid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Full-Income Measures of Poverty. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

73
73
77
80
81
85
87

Supporting Further Economic Gains.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Making Sure That Workers Have the Skills to Succeed.. . . . . . . . . . . . . . . . . 90
Limiting Geographic Frictions in the Labor Market. . . . . . . . . . . . . . . . . . . . 93
Reforming Occupational Licensing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Opportunity Zones: Matching People, Communities, and Capital.. . . . . . . . 96
Opportunity Zones: Evidence of Investor Interest and Activity. . . . . . . . . . . 97
Expanding Opportunities for Ex-Offenders.. . . . . . . . . . . . . . . . . . . . . . . . . . 98
Supporting Working Families . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Combating the Opioid Crisis.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

Economic Report of the President | 19

250-840_text_.pdf 23

2/7/20 3:46 PM

Chapter 3: Regulatory Reform Unleashes the Economy. . . . . . . . . . . . . . . . . . 105
Reversing the Regulatory Trend. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
Analyzing Regulatory Reform. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Deregulatory Actions Considered. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
Consumer Savings on Internet Access. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
Consumer and Small Business Savings on Healthcare. . . . . . . . . . . . . . . . . . . 123
Employment Regulations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Financial Regulations.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Additional Regulations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
The Doubling Effect of Shifting from a Growing Regulatory State to a
Deregulatory One. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Regulations Before 2017 with Disproportionate Costs. . . . . . . . . . . . . . . . . . . 133
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

Chapter 4: Energy: Innovation and Independence. . . . . . . . . . . . . . . . . . . . . . . 137
Market Pricing, Resource Access, and Freedom to Innovate. . . . . . . . . . . . . . . 140
The Effects of Innovation on Productivity, Prices, and Production.. . . . . . . . .
The Impact on Productivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Impact on Prices and Production. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
The Impact of the Shale-Induced Decline in Energy Prices. . . . . . . . . . . . .
Innovation-Driven Consumer Savings, Energy Independence, and
Environmental Benefits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Consumer Savings—Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Consumer Savings—Findings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Energy Independence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Environmental Benefits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

144
144
147
150
152
152
154
157
160

The Value of Deregulatory Energy Policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
Allowing Innovation to Flourish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
The Critical Role of Energy Infrastructure. . . . . . . . . . . . . . . . . . . . . . . . . . 166
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170

Chapter 5: Free-Market Healthcare Promotes Choice and Competition.. . . . 171
Building a High-Quality Healthcare System. . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Creating More Choice in Health Insurance Markets . . . . . . . . . . . . . . . . . . 173
Creating More Competition among Healthcare Providers.. . . . . . . . . . . . . 178
Healthcare Accomplishments under the Trump Administration.. . . . . . . . . . .
Increasing Choice. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Increasing Competition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Increasing Access to Valuable Therapies. . . . . . . . . . . . . . . . . . . . . . . . . . .

183
183
190
194

Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196

20 | Economic Report of the President

250-840_text_.pdf 24

2/7/20 3:46 PM

Part II: Evaluating and Addressing Threats to the Expansion. . . . . 197
Chapter 6: Evaluating the Risk of Declining Competition. . . . . . . . . . . . . . . . . 199
The Origin and Principles of Antitrust Policy. . . . . . . . . . . . . . . . . . . . . . . . . . . 202
Economic Analysis at the Agencies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
A Renewed Interest in Concentration and the State 0f Competition. . . . . . . .
Problems with the CEA’s 2016 Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Problems with Related Research .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Connecting Concentration and Markups with Antitrust Law. . . . . . . . . . . .

209
209
211
215

Calls for a Broader Interpretation of Antitrust Policy.. . . . . . . . . . . . . . . . . . . . 216
Antitrust Enforcement for the Digital Economy. . . . . . . . . . . . . . . . . . . . . . . . . 217
Background.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
Proposals for Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
Competition Policy to Reduce Entry Barriers. . . . . . . . . . . . . . . . . . . . . . . . . . . 222
Other Government-Created Barriers to Entry.. . . . . . . . . . . . . . . . . . . . . . . 224
Promoting Innovation through Sound Enforcement of Competition Law . 225
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

Chapter 7: Understanding the Opioid Crisis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
The Supply-and-Demand Framework.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
The First Wave of the Crisis: Prescription Opioids. . . . . . . . . . . . . . . . . . . . . . . 240
Public Subsidies for Opioids. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
The Second Wave of the Crisis: Illicit Opioids.. . . . . . . . . . . . . . . . . . . . . . . . . . 258
Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262

Chapter 8: Expanding Affordable Housing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
The Housing Affordability Problem.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
The Role of Overregulation in the Housing Affordability Problem. . . . . . . . . . 278
Consequences of Overregulation of Housing. . . . . . . . . . . . . . . . . . . . . . . . . . .
The Increased Cost of Attaining Homeownership and Higher Rents. . . . . .
Increased Homelessness.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fewer People Are Served by Housing Assistance Programs.. . . . . . . . . . . .
Weakened Labor Mobility and Economic Growth. . . . . . . . . . . . . . . . . . . .
Reduced Opportunity for Children. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Increased Traffic Congestion and Harm to the Environment.. . . . . . . . . . .

283
284
284
286
287
289
289

Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291

Part III: The Economic Outlook. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Chapter 9: The Outlook for a Continued Expansion. . . . . . . . . . . . . . . . . . . . . . 295
GDP Growth during the Next Three Years. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
GDP Growth over the Longer Term. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Upside and Downside Forecast Risks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301

References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
Economic Report of the President | 21

250-840_text_.pdf 25

2/7/20 3:46 PM

Appendixes
A.

Report to the President on the Activities of the Council of
Economic Advisers During 2019. . . . . . . . . . . . . . . . . .  343

B.

Statistical Tables Relating to Income, Employment, and
Production. . . . . . . . . . . . . . . . . . . . . . . . . . .  355

I-1

The Actual Unemployment Rate in Various Quarters versus the August
2106 Rate, per the BLS and CBO, 2012–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

I-2

Actual Nonfarm Payrolls versus the August 2016 Payroll, per the CBO,
2012–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

1-1

Real GDP per Working-Age Population by Expansion, 1960–2019 . . . . . . . . . . 33

1-2

Real GDP Growth Relative to Pre–November 2016 Projections,
2017–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

1-3

Actual versus Consensus Projections of Real Gross Domestic Product,
2014–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

1-4

Length and Depth of U.S. Expansions and Contractions, 1949–2019. . . . . . . . 36

1-5

Average of Absolute Troika Forecasting Errors by Horizon and
Administration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

1-6

Nonfarm Business Sector Labor Productivity Growth, 2009–19. . . . . . . . . . . . 38

1-7

Growth in Real GDP per Employed Person Among the Advanced
Economies, 2009–19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

1-8

Actual versus Consensus Projections for Real Disposable Personal
Income, 2014–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

1-9

Growth of Real Disposable Personal Income per Household, 2009–19. . . . . . 40

1-10

Labor Share of Income, 1947–2019 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

1-11

Cumulative Change in Nominal Household and Nonprofit
Wealth, 2014–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

1-12

Main Contributors to Real GDP Growth, 2017–19. . . . . . . . . . . . . . . . . . . . . . . . . 43

1-13

Consumption and Wealth Relative to Disposable Personal Income,
1952–2019 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

1-14

Personal Saving Rate, 2000–2019. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

1-15

Actual versus Preelection Projections for Nonresidential Private
Fixed Investment, 2014–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46

1-16

Cumulative Change in Gross Fixed Private Capital Formation among
the Group of Seven Member Countries, 2017:Q4–2019:Q4. . . . . . . . . . . . . . . . . 46

1-17

The User Cost of Capital, 2011–19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

1-18

Average Annual Growth in Real Business Fixed Investment and
Component Contributions, 2010–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

1-19

Real Mining and Drilling Structures Investment versus Oil Rigs
Operating in the United States, 2007–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

1-20

Brent Crude Oil Prices versus Oil Rigs Operating in the United States,
2007–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Figures

22 |

250-840_text_.pdf 26

Economic Report of the President

2/7/20 3:46 PM

1-21

Predictions of an Investment Accelerator Model, 2014–19. . . . . . . . . . . . . . . . . 50

1-22

The Growth in Number of Private Establishments versus Small
Business Optimism, 2000–2019. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

1-23

Inflation: GDP Price Index versus the PCE Price Index, 2009–19. . . . . . . . . . . . 52

1-24

Import Prices versus GDP Price Index, 1955–2019. . . . . . . . . . . . . . . . . . . . . . . . 52

1-25

Consumer Price Inflation, 2012–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

1-26

Core CPI Inflation and Inflation Expectations, 1960–2019 . . . . . . . . . . . . . . . . . 54

1-27

Price-Price Phillips Curve Scatter Diagram, 1960–2018. . . . . . . . . . . . . . . . . . . . 56

1-28

IMF Five-Year Real GDP Growth Forecasts for the World, 2012–24. . . . . . . . . . 58

1-29

Forecast of 2019 Real GDP Growth. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58

1-30

Composite Output Purchasing Manager’s Index (PMI), 2015–19. . . . . . . . . . . . 59

1-31

China’s Change in Automobile Sales, 2014–19 . . . . . . . . . . . . . . . . . . . . . . . . . . . 61

1-32

German Vehicle and Car Engines Exported, 2016–19. . . . . . . . . . . . . . . . . . . . . . 62

1-33

Central Bank Policy Rates, 2010–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63

1-34

Federal Reserve Trade-Weighted Broad Nominal versus Real
Dollar, 1973–2019. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

2-1

Total Jobs versus Preelection Forecast, 2012–19. . . . . . . . . . . . . . . . . . . . . . . . . 70

2-2

Unemployment Rate versus Preelection Forecasts, 2011–19. . . . . . . . . . . . . . 71

2-3

Number of Unemployed People versus Number of Job Openings,
2001–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

2-4

Unemployment Rate by Race, 2003–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

2-5

Multiple Jobholders as a Percentage of All Employed, 1994–2019. . . . . . . . . . 76

2-6

Demographically Adjusted Labor Force Participation for African
Americans, 1973–2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

2-7

Demographically Adjusted Labor Force Participation Rate for Hispanics,
1994–2018 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

2-8

Nominal Weekly Wage Growth among All Adult Full-Time Wage and
Salary Workers, 2010–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

2-9

Average Hourly Earnings for Production and Nonsupervisory Workers
and the Personal Consumption Expenditures Price Index, 2007–19 . . . . . . . . 81

2-i

Consumer Savings on Prescription Drugs and Internet Access by
Household Income Quintile. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

2-10

Poverty Rates by Race and Ethnicity, 1966–2018 . . . . . . . . . . . . . . . . . . . . . . . . . 84

2-11

Number of Medicaid and CHIP Enrollees by Month in Expansion and
Nonexpansion States, 2014–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

2-12

Change in the Official Poverty Measure versus Other Poverty Measures,
2016 and 2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

2-13

Number of Unemployed Workers per 100 Job Openings, Q2:2019 . . . . . . . . . 94

2-14

Share of U.S. Residents Who Moved, 1948–2018. . . . . . . . . . . . . . . . . . . . . . . . . . 94

2-15

Labor Force Participation Rate among Parents by Age of Youngest
Child in Household and Sex of Adult, 1968–2019. . . . . . . . . . . . . . . . . . . . . . .  100

Economic Report of the President | 23

250-840_text_.pdf 27

2/7/20 3:46 PM

2-ii

Female Labor Force Participation Rate, by Selected OECD
Country, 2018 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  102

3-1

Significant Final Rules by Presidential Year, Excluding Deregulatory
Actions, 2000–2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  109

3-2

Regulatory Restrictions by All Agencies, 1970–2019. . . . . . . . . . . . . . . . . . . . . 110

3-3

Regulatory Restrictions on Manufacturing, 1970–2018. . . . . . . . . . . . . . . . . .  112

3-4

Wireless and Wired Internet Service Provider Price Cuts Close to
Congressional Review Act Nullification of Federal Communications
Commission Rule, 2016–17 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  121

3-5

Inflation-Adjusted CPI for Prescription Drugs, 2009–19. . . . . . . . . . . . . . . . . . 124

3-6

Deregulation Creates More Growth Than a Regulatory Freeze, 2001–21. . .  131

4-i

U.S. Cellulosic Biofuel Statute and Final Volumes, 2010–19. . . . . . . . . . . . . .  143

4-1

Innovation in Natural Gas Production. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  145

4-2

Productivity Gains: New-Well Production per Rig, Oil and Natural Gas,
2007–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  145

4-3

Gains in Productivity Lower Breakeven Prices Across Key Shale
Formations, 2014–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  146

4-4

Natural Gas Actual Production versus Projected Production, 2005–18. . . .  147

4-5

Natural Gas Actual Prices versus Projected Prices, 2005–18 . . . . . . . . . . . . .  148

4-6

U.S. Monthly Wholesale Electricity Price and Natural Gas . . . . . . . . . . . . . . .  148

4-7

U.S. Crude Oil Production, 2005–18. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  149

4-8

Imported Oil Prices, 2005–18. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  150

4-9

Shale Oil and Gas Consumer Savings per Year by Sector. . . . . . . . . . . . . . . .  155

4-10

Total Consumer Savings as a Share of Income by Quintile. . . . . . . . . . . . . . .  155

4-11

U.S. Monthly Net Exports of Natural Gas, 1999–2020. . . . . . . . . . . . . . . . . . . .  158

4-12

U.S. Monthly Net Exports of Crude Oil and Petroleum Products,
1990–2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  158

4-13

Changes in Price of Oil (Prior Month) and Changes in the Goods Trade
Balance, 2000–2010 and 2011–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  159

4-14

Actual versus Projected Carbon Dioxide Emissions, 2005–18 . . . . . . . . . . . .  161

4-15

Annual GHG Emission Reductions from Shale Innovation and Major
Environmental Policies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  162

4-16

U.S. versus EU GDP-Adjusted Carbon Dioxide Emissions, 2005–17 . . . . . . .  163

4-17

U.S. versus EU GDP Adjusted Particulate Emission, 2005–17. . . . . . . . . . . . . .
164

4-ii

Methane Production and Leakage Rates, 1990–2017. . . . . . . . . . . . . . . . . . . .  165

4-18

Citygate Natural Gas Prices in Michigan and New York Relative to
National Average Prices, 2005–18. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  169

5-1

Health Insurance Coverage by Type of Insurance, 2018 . . . . . . . . . . . . . . . . .  174

5-2

Annual Change in Average Family Premium Including Employee and
Employer Contributions, 2000–2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  175

24 |

250-840_text_.pdf 28

Economic Report of the President

2/7/20 3:46 PM

5-3

Percentage of Covered Workers Enrolled in a Plan with a General
Annual Deductible of $2,000 or More for Single Coverage, 2009–19 . . . . . .  176

5-4

Average Annual Worker and Employer Premium Contributions for
Single Coverage, 2019. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  177

5-5

Average Annual Worker and Employer Premium Contributions for
Family Coverage, 2019. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  177

5-6

Ratio of State Average Price to National Average Price of Cataract
Removal, 2015. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  180

5-7

Distribution of Average State Price Relative to Average National Price
of Care Bundles in Four States, 2015. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  181

5-8

Health Insurance Enrollment across Employer Scenarios, 2019–29. . . . . . .  189

5-9

Revenue Impact from Expanding High-Deductible Health Plans,
2019–29. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  189

6-1

Summary of Transactions by Fiscal Year, 2009–18. . . . . . . . . . . . . . . . . . . . . .  204

6-i

New Entry of Hard Disk Drive Firms, 1976–2012. . . . . . . . . . . . . . . . . . . . . . . .  213

6-ii

Exit and Merger of Hard Disk Drive Firms, 1976–2012. . . . . . . . . . . . . . . . . . .  214

7-1

Opioid-Involved Overdose Deaths, 1999–2019 . . . . . . . . . . . . . . . . . . . . . . . . .  231

7-2

Share of Potency-Adjusted Prescription Opioids, by Primary Payer,
2001–15. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  232

7-3

Opioid-Involved Overdose Death Rate by Presence of Prescription
Opioids, 2001–16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  235

7-4

Effect of Supply Expansions and Government Subsidies on the Price and
Quantity of Prescription Opioid Misuse, Primary Market. . . . . . . . . . . . . . . .  237

7-5

Effect of Supply Expansions and Government Subsidies on the Price and
Quantity of Prescription Opioid Misuse, Secondary Market . . . . . . . . . . . . .  237

7-6

Effect of Demand Expansions on the Quantity and Price of Illicit Opioids.  239

7-7

Effect of Supply Expansions on the Quantity and Price of Illicit Opioids . .  239

7-8

Opioid Overdose Death and Unemployment Rate, 1999–2016. . . . . . . . . . .  240

7-9

Real Supply Price and Real Out-of-Pocket Price Index of Potency-Adjusted
Prescription Opioids, 2001–15. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  243

7-10

Brand Share of Potency-Adjusted Prescription Opioids and Supply Price,
2001–16. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  243

7-11

Potency-Adjusted Quantity (MGEs) of Prescription Opioids per Capita in
the United States, 2001–15. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  246

7-12

Proportion of Users Obtaining Misused Prescription Opioids by Most
Recent Source, 2013–14 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  247

7-13

Actual and Predicted Rates of Overdose Deaths Involving Prescription
Opioids, by the Price Elasticity of Demand for Misuse, 2001–15. . . . . . . . . .  251

7-14

Share of Potency-Adjusted Prescription Opioids, by Primary Payer,
2001–15. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  253

7-15

Potency-Adjusted Prescription Opioids per Capita, by Primary Payer,
2001–15. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  254

Economic Report of the President | 25

250-840_text_.pdf 29

2/7/20 3:46 PM

7-16

Adults Receiving Social Security Disability Insurance and Supplemental
Security Income, and Opioid-Involved Drug Overdose Deaths, per
100,000 People, 1980–2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  257

7-17

Rate of Overdose Deaths Involving Synthetic Opioids Other Than
Methadone, and Fentanyl Reports in Forensic Labs per 100,000
Population, 2001–16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  259

7-18

The Opioid-Involved Overdose Death Rate by the Presence of Illicit
Opioids, 2001–16 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  259

7-19

Real Price Index of Potency-Adjusted Illicit Opioids, 2001–16. . . . . . . . . . . .  261

8-1

Homeownership Rates by Race and Ethnicity, 2000–2019. . . . . . . . . . . . . . .  270

8-i

The Carpenter Index by CBSA, 2018. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  277

8-2

The Effect of Regulation on Supply and Demand for Housing. . . . . . . . . . . .  279

8-3

Wharton Land Use Index by Metropolitan Statistical Area, 2008. . . . . . . . . .  281

8-4

Ratio of Home Prices to Production Costs by CBSA, 2013 . . . . . . . . . . . . . . .  281

8-5

Home Price Premium Resulting from Excessive Housing Regulation. . . . . . 285

8-6

Percentage Reduction in Homelessness by CBSA from Deregulating
Housing Markets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  285

9-1

Forecast of Growth Rate in Real GDP, 2019–30 . . . . . . . . . . . . . . . . . . . . . . . . .  298

Tables
1-1

Effects of Deregulation on Relative Price Increases on the Core CPI,
2006–19. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

2-1

Unemployment Rates by Demographic Group. . . . . . . . . . . . . . . . . . . . . . . . . . . 75

2-2

Change in the Number of People Age 18–64 Years Old with Different
Types of Insurance by Family Income Level, 2016–18. . . . . . . . . . . . . . . . . . . . . 87

2-3

Change in the Number of Children with Different Types of Insurance by
Family Income Level, 2016–18. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87

3-1

Regulatory and Statutory Actions’ Annual Impact on Real Income
Relative to a Regulatory Freeze, by Sampling Strata. . . . . . . . . . . . . . . . . . . .  119

5-1

Variation in Knee Replacement Prices across MSAs within States, 2015. . .  182

6-1

Change in Market Concentration by Sector, 1997–2012 . . . . . . . . . . . . . . . . .  210

7-1

Estimates of the Price Elasticity of Demand for Heroin. . . . . . . . . . . . . . . . . .  249

8-1

Percentage of Renter Households Paying More Than 30 Percent of
Income on Housing by Income, 2009 versus 2017 . . . . . . . . . . . . . . . . . . . . . .  276

9-1

The Administration’s Economic Forecast, 2018–30 . . . . . . . . . . . . . . . . . . . . .  299

9-2

Supply-Side Components of Actual and Potential Real Output Growth,
1953–2030 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  300

2-1

Who Bears the Burden of Regulatory Costs? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

2-2

The Federation of Advanced Manufacturing Education . . . . . . . . . . . . . . . . . . . 92

Boxes

26 |

250-840_text_.pdf 30

Economic Report of the President

2/7/20 3:46 PM

2-3

The Women’s Global Development and Prosperity Initiative and
Female Labor Force Participation Globally . . . . . . . . . . . . . . . . . . . . . . . . . . . .  101

3-1

Looking Forward and Backward to Study Regulatory Reform. . . . . . . . . . . .  115

3-2

How Old Are Midnight Regulations?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  132

4-1

The Limits of Energy Mandates to Induce Innovation. . . . . . . . . . . . . . . . . . .  143

4-2

Economic Effects Linked to Drilling and Production. . . . . . . . . . . . . . . . . . . .  156

4-3

Innovation in Pipeline Leak Detection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  165

4-4

Shale Development and Local Communities. . . . . . . . . . . . . . . . . . . . . . . . . . .  167

5-1

The Consumer Price Index for Prescription Drugs . . . . . . . . . . . . . . . . . . . . . .  192

6-1

Antitrust and Monopsony: George’s Foods and Tyson Foods . . . . . . . . . . . .  205

6-2

Measuring Concentration and the HHI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  208

6-3

Concentration, Innovation, and Competition . . . . . . . . . . . . . . . . . . . . . . . . . .  213

6-4

The Effects of Deregulation within the Pharmaceutical Drug Market . . . . .  223

7-1

Opioid Crisis Lawsuits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  245

8-1

Measuring the Housing Affordability Problem with the
Carpenter Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  277

8-2

Poor Substitutes for Regulatory Reform. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  290

Economic Report of the President | 27

250-840_text_.pdf 31

2/7/20 3:46 PM

250-840_text_.pdf 32

2/7/20 3:46 PM

x
Part I

The Longest Expansion
on Record

29

250-840_text_.pdf 33

2/7/20 3:46 PM

250-840_text_.pdf 34

2/7/20 3:46 PM

x
Chapter 1

The Great Expansion
Two years since the Tax Cuts and Jobs Act (TCJA) was signed into law, and buttressed by the Administration’s probusiness deregulation policy and support
for innovative energy infrastructure, the U.S. economy continues expanding
at a healthy pace, as predicted by the 2018 and 2019 volumes of the Economic
Report of the President. As of December 2019, the U.S. economic expansion
reached its 127th month, the longest in the Nation’s history.
This chapter shows that, despite headwinds from the global economy and
the maturing length of the expansion, the U.S. economy remains resilient.
As a result, it grew at the fastest rate among the Group of Seven countries in
the first three quarters of 2019. During 2019, several macroeconomic indicators—including consumer spending, productivity, and labor shares of income—
continued to grow at faster rates than pre-TCJA projections. The labor market
also tightened further, even after strong gains in the previous two years. During
2019, the unemployment rate hit a 50-year low and, for the first time on record,
job openings exceeded job seekers, which have helped pull potential workers
from the sidelines and into the labor force. Wages rose faster than inflation,
which ultimately boosted real middle-class incomes. After years of decline,
the labor force participation rate stabilized because of increased prime-age
participation, which also boosts long-term potential output.
The tepid recovery from the Great Recession prompted economic forecasters
in 2016 to project historically modest growth into the future. Many observers
concluded that low growth would persist indefinitely. However, the experience
of the first three years of the current Administration proves that a prolonged
period of low growth was in fact far from inevitable. This increased growth

31

250-840_text_.pdf 35

2/7/20 3:46 PM

has coincided with Administration policies favoring lower taxes, substantial
deregulation, and pro-innovation energy policy. The CEA forecasts that there
is substantial additional room to grow—given the historically strong labor
market, the potential for further deregulation, and the supply-side impact of
TCJA on long-term growth.

A

fter growing briskly in 2017 and 2018, the U.S. economy continued to
expand at a healthy pace in 2019. During the year’s four quarters, real
gross domestic product (GDP) moderated to 2.3 percent at an annual
rate, from its 2.5 percent pace in 2018. This growth rate is notable considering
the maturing length of the current expansion and that it was achieved despite
headwinds from a slowing global economy. As of December, the U.S. economy
marked the 127th month and the 42nd consecutive quarter of expansion (figure 1-1), surpassing the longest U.S. expansion, which ended in March 2001
after 120 months or 40 quarters.
The U.S. economy is currently operating with a strong labor market and
subdued inflationary pressure. Evidence of the strength of the labor market
can be observed across many indicators. The U.S. unemployment rate was 3.5
percent as of December 2019, a 50-year low previously hit in September and
November 2019. Nominal average hourly earnings increased 2.9 percent during the 12 months of 2019, but had been at or above 3 percent for the prior 16
consecutive months. The tightness of the labor market and rising demand for
workers have continued to pull people from outside of the labor force into the
labor market, increasing the labor force participation rate to 63.1 percent for
the year as a whole, up 0.2 percentage point from a year earlier. Specifically,
the prime-age adult (25–54 years) participation rate increased to 82.5 percent
during these 12 months, the fourth year of increases after years of decline
since 2008. During the 12 months of 2019, the U.S. economy added 2.1 million
nonfarm jobs, averaging 176,000 jobs per month.
Despite the strong labor market, core consumer price inflation was
subdued, at 1.6 percent in 2019 (as measured by the price index for core personal consumption expenditures, PCE). Because nominal disposable personal
income grew faster than inflation, real disposable personal income grew at
a 2.6 percent annual rate during the four quarters of 2019. For the median
household, real income rose by $1,834 in the first 10 months of 2019, reaching
the highest level on record, at about $66,500 in 2019 dollars (Green and Coder
2019). In addition to rising real income, household wealth surged as stock
market valuations rose to new heights in 2019.
An increase in real household income and wealth has supported consumer spending, which constitutes 70 percent of GDP. In the four quarters of

32 | Chapter 1

250-840_text_.pdf 36

2/7/20 3:46 PM

Figure 1-1. Real GDP per Working-Age Population by Expansion
Period, 1960–2019
Index (100 = real GDP per working-age population at
the quarterly business-cycle trough)
145

1961–69

140
135
130
125

1982–90

120
115

1991–2001

2009–19

1975–80

110

2001–7

105

1980–81

100
0

6

12

1970–73

18
24
30
Quarters after start of expansion

36

42

Sources: Bureau of Economic Analysis; National Bureau of Economic Research; Census Bureau; CEA
calculations.
Note: The working-age population refers to those age 25–64 years. Series are smoothed using a
four-quarter, centered moving average. Quarterly population estimates are interpolated from
annual data.

2019, real consumer spending maintained the 2.6 percent pace of 2018, and
accounted for nearly 80 percent of real GDP growth. Government purchases
have also supported aggregate demand, rising 3.0 percent during 2019, compared with 1.5 percent in 2018.
Although American consumers have sustained the U.S. expansion, a general slowdown in the global economy has restrained U.S. growth. The Group of
Seven (G7) countries’ economies slowed sharply in the past year; in particular,
real GDP growth in Germany and the United Kingdom contracted in 2019:Q2.
Major emerging market economies such as China and India also experienced
slowdowns. These countries’ slowdowns reduced global aggregate demand,
which dampened U.S. economic growth.
Despite the headwinds from abroad, the U.S. economy was the fastestgrowing in the G7 in the first three quarters of 2019. The United States was one
of only two G7 countries (the other being Japan, where projected growth was
a moribund 0.9 percent) that did not require the International Monetary Fund
to make large downward revisions to its one-year-ahead growth projections
for 2019 (IMF 2018, 2019c), whereas the other advanced countries saw large
downward revisions.
Moreover, growth in the U.S. economy, for the third consecutive year,
exceeded the consensus real GDP growth projection made before the 2016
election, as well as projections made before the 2017 TCJA. Three years ago, a
widespread belief among economic forecasters was that subpar growth in the

The Great Expansion

250-840_text_.pdf 37

| 33

2/7/20 3:46 PM

Figure 1-2. Real GDP Growth Relative to Pre–November 2016
Projections, 2017–19
2016 FOMC

2016 CBO

Actual

Percent
3.0
2.5
2.0
1.5
1.0
0.5
0.0
2017

2018

2019

Sources: Congressional Budget Office, August 2016 Baseline Forecast; Federal Open Market
Committee, September 2016; Bureau of Economic Analysis; CEA calculations.
Note: FOMC = Federal Open Market Committee; CBO = Congressional Budget Office.
Q4-over-Q4 growth rates are used.

U.S. economy will be permanent, with one of the more prominent explanations
being secular stagnation.1 This pessimism was reflected in the modest growth
projections by outside forecasters at the time. In 2016, the Federal Open
Market Committee (FOMC) forecast real GDP over the four quarters of 2019 to
be 1.8 percent, while the Congressional Budget Office (CBO) forecast real GDP
growth of just 1.6 percent over the same period (see figure 1-2). The 2.3 percent
real GDP growth during 2019 surpassed these forecasts. Similarly, actual real
GDP growth in 2017 and 2018 surpassed preelection projections from the FOMC
and the CBO. Relative to the 2016 real GDP projections by the Blue Chip panel
of private professional forecasters, the annual level of U.S. real GDP in 2019 was
1.2 percent higher (figure 1-3).
Although the strong growth was a surprise relative to pre-2017 forecasts
by the FOMC, the CBO, and the Blue Chip consensus panel, it was largely
anticipated by the current Administration. In May 2017, the Administration
forecasted average annualized growth over the three years 2017–19 to be 2.5
percent; subsequently the Administration revised 2018 and 2019 forecasts
up to 3.1 percent, which was deemed optimistic and unrealistic compared
with external forecasts. The optimism of the CEA’s forecasts was grounded
1 Hansen (1939) was the first to put forward this concept, which was popularized by Summers
(2013, 2014, 2016) and more recently by Rachel and Summers (2019). Specifically, Summers
argued that when neutral real interest rates fall to an abnormally low level because of decreasing
propensity to invest but increasing propensity to save, and are below nominal interest rates, the
resultant excessive savings would act as a persistent drag on demand and growth.

34 | Chapter 1

250-840_text_.pdf 38

2/7/20 3:46 PM

Figure 1-3. Actual versus Consensus Projections of Real Gross
Domestic Product, 2014–19
Index (2016 = 100)

Election

109

2019

107

Actual, postelection

105
103
101

Pre-TCJA projection
consensus (March 17)

Actual, preelection

Preelection consensus
projection (Oct. 16)

99
97
95
2014

2015

2016

2017

2018

2019

Sources: Bureau of Economic Analysis; CEA calculations.
Note: Consensus forecasts from the October 2016 and March 2017 issues of Blue Chip Economic
Indicators begin with 2017 growth for levels implied by year-over-year forecasts.

in the expectation that the Administration’s tax policies and deregulatory
policies would have a more positive effect than projected by others. In the
2018 Economic Report of the President, the CEA drew on an extensive body of
academic literature to predict that tax reform would raise real capital investment and the growth rate of output. In the 2019 Report, we reviewed data
through 2018:Q3 showing that the U.S. economy’s responses along multiple
margins were consistent with predictions from that academic literature. Over
the 12 quarters through 2019:Q4, the actual average annual growth rate of real
GDP was 2.5 percent, slightly outpacing the May 2017 forecast, and an increase
compared with the 2.2 percent average annual growth rate over the 26-quarter
expansion period from 2009:Q3 through 2016:Q4 (see figure 1-4). As figure 1-5
shows, the average absolute errors of the ex-ante Administration forecasts
under the current Administration were the lowest among those of the last five
administrations.
The Trump Administration adopted structural reforms and policies that
were designed to support continued U.S. economic growth. The TCJA, which
was enacted on December 22, 2017, permanently reduced the statutory corporate tax rate from 35 to 21 percent, sharply lowering the user cost of capital. It
also enabled 100 percent expensing of new equipment investment, retroactive
to September 27, 2017 (the date of the first draft of the proposed tax legislation that included the 100 percent expensing provision from the House Ways
and Means Committee). The international provisions of the TCJA, specifically

The Great Expansion

250-840_text_.pdf 39

| 35

2/7/20 3:46 PM

Figure 1-4. Length and Depth of U.S. Expansions and Contractions,
1949–2019
Annual growth rate (percent)
10

2019:Q4

7.6

8

5.5

6

4.0

4

5.1

4.9

4.3 4.3

4.3

3.6

2.9

2.2 2.5

2
0

–2

–0.2
–1.5

–4

–2.4

–2.0

–2.5
–3.9

–6
1949

1959

Current
Administration,
so far

0.6

–0.2

–2.7

–2.7

–4.3
1969

1979

1989

1999

2009

2019

Sources: Bureau of Economic Analysis; National Bureau of Economic Research; CEA calculations.
Note: Values represent the change in real GDP as an annual growth rate for each quarterly
expansion and contraction period, as defined by the National Bureau of Economic Research.

Figure 1-5. Average of Absolute Troika Forecasting
Errors, by Horizon and Administration
Current-year error
Two-year-ahead error

One-year-ahead error
Average error across horizons

Forecasting errors (percentage points)
2.0

1.4
1.1

1.0

1.6
0.5

1.6

1.6

1.5

1.2

1.4

1.5

1.8

1.5

1.5 1.6 1.6
0.8 1.3

0.7

1.3
0.7

0.5 0.5

0.0

George H. W. Bush

Clinton

George W. Bush Obama
expansion,
2010–16

Trump

Sources: Federal Reserve Bank of Saint Louis (FRED); CEA calculations.
Note: Budget forecasts and Q4-over-Q4 growth rates were used to evaluate errors.

36 | Chapter 1

250-840_text_.pdf 40

2/7/20 3:46 PM

the change in the tax treatment of earnings from foreign affiliates (CEA 2019b),
led to repatriation of past overseas earnings of U.S. multinationals in low-tax
jurisdictions, as evidenced by the $1.04 trillion capital inflows from direct
investment income on equity from dividends and withdrawals since 2017:Q4.
The alterations in the tax treatment of foreign affiliates came in two parts: one
for past earnings (a one-time transition tax at a low rate on past earnings held
overseas), and one for future foreign-subsidiary earnings (eliminating the tax
on normal repatriated dividends).
Businesses responded to the lower user cost of capital and geographical incentives under the TCJA with an increase in domestic investment. This
investment led to capital deepening, increasing capital services per unit of
labor input, which raised labor productivity, real wages, and U.S. real output. In addition, as discussed in more detail in chapter 3 of this Report, the
Administration’s deregulatory agenda also helped lower prices, from Internet
prices to drug prices, and increased real income for American households.
The 2018 Bipartisan Budget Act also increased government spending, raising
aggregate demand. The combination of these factors lays the foundation for
continued prosperity in the future.
As the current record expansion matures beyond the 42nd quarter, some
worry that the expansion will “die of old age.” But evidence suggests that
expansions do not end simply because of their length. A study by Diebold and
Rudebusch (1990) was among the first to find that in the postwar period, the
probability of an expansion coming to an end was not increasing in the age
of the expansion. In a follow-up study, Rudebusch (2016) provided empirical
evidence that long expansions during the past 70 years are “no more likely to
end than short ones.” Australia’s economy, which has experienced the longest
expansion of any advanced economy in modern history, at 28 years, exemplifies how expansions can continue for decades. Old age does not kill expansions, though bad policies and adverse shocks can lead to recessions.
The remainder of this chapter provides evidence on the strength of different areas of the U.S. economy in the recent past, including: productivity, wages
and income, consumer spending, employment, investment, and subdued inflation. The chapter also discusses the impact of the global economic downturn,
monetary policy, and domestic factors slowing U.S. growth.

Productivity
Productivity growth is a key driver of long-term real output growth. Labor
productivity in the post-TCJA period, 2018:Q1–2019:Q3, increased at an average annual pace of 1.4 percent—in particular, it picked up to 1.9 percent in the
three quarters through 2019:Q3, a faster pace than the average growth rate

The Great Expansion

250-840_text_.pdf 41

| 37

2/7/20 3:46 PM

Figure 1-6. Nonfarm Business Sector Labor Productivity Growth,
2009–19
2009:Q3–2016:Q4

1.1

2013:Q1–2016:Q4

0.9

President Trump pre-TCJA
2017:Q1–2017:Q4

1.3

President Trump post-TCJA
2018:Q1–2019:Q3

1.4

0.0

0.5
1.0
Annual growth rate (percent)

1.5

Sources: Bureau of Labor Statistics; CEA calculations.
Note: The annual growth rate is calculated for real output per hour of all persons in the
nonfarm business sector.

of 1.1 percent in the pre-TCJA economic expansion period 2009:Q3–2016:Q4
(figure 1-6).2
Academic research suggests at least two channels through which the
current Administration’s policies can increase labor productivity. The first
is through deregulatory actions pursued since the end of 2016 that have
increased competition and productivity (CEA 2019a). The second channel is
through capital deepening in response to a lower cost of capital under the
TCJA. By raising investment, capital services per worker rises and, as a result,
so does labor productivity (CEA 2019b). Since the passage of the TCJA, capital
services have grown faster than projected by outside forecasters.3
Comparing the performance of the U.S. economy with other advanced
economies provides another instructive benchmark. Since the start of the current Administration and through 2019:Q3 (the latest quarter available for all G7
countries as of the date of writing), U.S. productivity growth, as measured by
output per worker, notably outperformed that of other countries (figure 1-7).
2 Comparisons can be made with other subperiods in the past. Excluding the contractionary
periods during the Great Recession, labor productivity grew at just a 1.1 percent compound annual
rate during the period 2009:Q3–2016:Q4.
3 Actual capital services grew at an annual rate of 3.2 percent over the two years after passage
of the TCJA, compared with 2.9 percent as projected by Macroeconomic Advisers in October
2017, and 3.1 percent projected by Blue Chip Econometric Detail in February 2018. With a slightly
different accounting method, the CBO also expected overall capital services to grow at 2.3 percent,
compared with the actual annual growth rate of 2.7 percent.

38 |

250-840_text_.pdf 42

Chapter 1

2/7/20 3:46 PM

Figure 1-7. Growth in Real GDP per Employed Person among
the Advanced Economies, 2009–19
2009:Q3–2016:Q4 (30 quarters)
Annual growth rate (percent)
1.5
1.2
1.2
1.0
1.0
0.8 0.8
0.5

0.5

0.4

1.1 1.2

0.9

0.8
0.4

0.4
0.0

0.0
–0.5

2017:Q1–2019:Q3 (11 quarters)

0.1

–0.4

–1.0

Australia Canada

France Germany

Italy

Japan

United United
Kingdom States

Sources: Australian Bureau of Statistics; Statistics Canada; Institut national de la statistique et des
études économiques; Deutsche Bundesbank; Istituto Nazionale di Statistica; Japan Cabinet Office;
U.K. Office for National Statistics; Bureau of Economic Analysis; Bureau of Labor Statistics; Haver
Analytics; CEA calculations.
Note: Values represent an annual growth rate calculated over the given quarters. Growth rates are
based on real GDP divided by seasonally adjusted employment. Employment includes goverment
employees.

While U.S. labor productivity, as measured by output per employed person for
cross-country consistency, grew at a compound annual rate of 1.2 percent during this period, the average growth rate among non-U.S. G7 member countries
and Australia was just 0.3 percent.
Another striking observation is that the United States is the only economy among this group of advanced economies to experience an acceleration in
labor productivity. As noted in the 2017 Economic Report of the President, from
2005 to 2015 all G7 countries experienced a sharp decline in labor productivity
growth from the 10 earlier years, due to slowdowns in both capital deepening
and total factor productivity (CEA 2017). Figure 1-7 shows the later of these
periods, with the inclusion of 2016, when labor productivity growth in the
United States was similar to that in the other G7 countries (plus Australia). In
the 11 quarters since that period, productivity growth has been flat or falling
in all these advanced economies, while productivity growth has risen in the
United States.

Wages and Income
In traditional economic models, equilibrium in the labor market requires that
nominal hourly compensation equals the marginal product of labor. Although
real output per unit of labor is a measure of the average instead of the marginal product, the measure is a convenient proxy for the marginal product.

The Great Expansion

250-840_text_.pdf 43

| 39

2/7/20 3:46 PM

Figure 1-8. Actual versus Consensus Projections for Real Disposable
Personal Income, 2014–19
Index (2016 = 100)

Election

112

2019

109

Actual, postelection
106
103

Pre-TCJA projection
consensus (March 2017)
Preelection consensus
projection (Oct. 2016)

Actual, preelection
100
97
94
2014

2015

2016

2017

2018

2019

Sources: Bureau of Economic Analysis; CEA calculations.
Note: Consensus forecasts from the October 2016 and March 2017 issues of Blue Chip Economic
Indicators and begin with 2017 growth for levels implied by year-over-year forecasts.

Figure 1-9. Growth of Real Disposable Personal Income per Household,
2009–19
Annual growth rate (percent)
сѵп

рѵф

рѵц

рѵт

рѵп

пѵф

пѵп
сппшѷт–спрхѷу

*./Ҋ ѷ
спрцѷу–спршѷу

*0- .ѷ 0- 0
*!
*)*($
)'4.$.Ѹ
 ).0.
0- 0Ѹ

'0'/$*).ѵ
*/ ѷ
'0 .
- +- . )/
"-*2/#
/
)
))0'
-/ 
*1 -
/# 
"$1 )
,0-/ -.ѵ
 *0. #*'.
- 

( .0- 
!-*(
/# 
 ).0.
0- 0’.
#*0.$)"
/. 
.
/# 
- &Ҋ%0./ 
/*/'
)0( -
*!

#*0. #*'.ѵ

40 | Chapter 1

250-840_text_.pdf 44

2/7/20 3:46 PM

$"0- 
рҊрпѵ
 *-
#- 
*!
 )*( Ѷ
ршуц–спрш
ршуц–спрх

спрц–рш

Labor share (percent)
хп

спршѷт

фч

фх

фу

фс

фп
ршуц

ршфф

ршхт

ршцр

ршцш

ршчц

ршшф

сппт

спрр

*0- .ѷ 0- 0
*!
*)*($
)'4.$.Ѹ

'0'/$*).ѵ
*/ ѷ
#$)"
 )*/ .

-  ..$*)ѵ
The l*-
.#- 
$.
'0'/ 
.
the /*/'
*(+ )./$*)
*!

(+'*4 .
.

+ - )/" 
*!
"-*.. *( ./$
$)*( ѵ

спрш

Coincident with the increase in labor productivity growth has been an increase
in real average hourly earnings growth, particularly for many disadvantaged
groups (see chapter 2 of this Report). Real average hourly earnings grew at an
annual rate of 1.1 percent during the post-TCJA period and 1.3 percent for nonsupervisory workers, compared with 0.4 percent and 0.5 percent, respectively,
in the first seven and a half years of the expansion through 2016:Q4. Real wage
growth further picked up for nonsupervisory workers, to 1.4 percent in the four
quarters of 2019, as the labor market continued to heat up.
The net tax savings from the TCJA—from a combination of increasing
standard deductions, lowering marginal rates, and doubling the child tax
credit—is also expected to boost real disposable income. In its pre-TCJA
projections (March 2017), the Blue Chip consensus panel forecasted that real
disposable personal income would grow at an average of 2.65 percent during
2018 and 2019; in actuality, it grew at a 3.5 percent rate (figure 1-8), well above
the consensus forecast and well above the 2.1 percent average annual growth
rate over the period 2009:Q3–2016:Q4. A similar pattern is observed on a perhousehold basis, where real disposable personal income per household grew
in the post-TCJA period at an annual average rate of 1.7 percent, outpacing the
1.3 percent of the earlier period (figure 1-9).
As income accelerates, labor’s share of gross domestic income (GDI) also
continues on an upward trajectory. Measuring labor’s share as total employee

The Great Expansion | 41

250-840_text_.pdf 45

2/7/20 3:46 PM

Figure 1-11. Cumulative Change in Nominal Household and Nonprofit
Wealth, 2014–19
Stock market wealth

Net housing wealth

Other wealth

Dollars (trillions)
20
Total change: +$17.3 trillion
15
Total change: +$12.8 trillion
10

8.3

5.8
3.5

5

4.0
5.4

3.0
0
2014:Q1–2016:Q4

2017:Q1–2019:Q3

Sources: Federal Reserve Board (Financial Accounts of the United States); CEA calculations.

compensation as a percentage of GDI, the series partially retraced a multidecade trend decline through 2014. During the 11 quarters through 2019:Q3, it
rose a further 0.5 percentage point, to 53.6 percent (figure 1-10).
While labor’s share of GDI and real disposable income growth has
increased, total household wealth has also increased. The cumulative change
in nominal household and nonprofit-sector wealth, as reported by the Federal
Reserve’s Financial Accounts of the United States, in the first 11 quarters
through 2019:Q3 exceeds the cumulative change in the preceding 11 quarters
by over $4 trillion (figure 1-11).

Consumer Spending
A more productive workforce with greater disposable income has bolstered
overall economic growth. Consumer spending as a share of nominal gross
domestic product averaged 67.9 percent during the 10 years through 2018.
Given this sizable share of GDP, changes in consumer spending carry substantial contributions to overall real GDP growth. In 2019, real consumer spending
grew by 2.6 percent, maintaining the same pace as in 2018. Since the TCJA’s
passage, real consumer spending has grown 2.6 percent at an annual rate,
higher than the 2.3 percent pace during the 7½ years from 2009:Q3 through
2016:Q4, when real consumer spending contributed 1.6 percentage points to
real GDP growth. In the 12 quarters through 2019:Q4, real consumer spending

42 |

250-840_text_.pdf 46

Chapter 1

2/7/20 3:46 PM

Figure 1-12. Main Contributors to Real GDP Growth, 2017–19
Contributor
Government spending
Gross private domestic investment
Real GDP growth
Contribution (percentage points)
6.0
3.2
3.0

2.3

2019:Q4

3.5

3.5

2.2

Net exports
Personal consumption expenditures

3.1

2.1
2.1

2.9

2.5

2.0
1.1
0.0

-3.0
2017:Q1

2017:Q3

2018:Q1

2018:Q3

2019:Q1

2019:Q3

Source: Bureau of Economic Analysis.

Figure 1-13. Consumption and Wealth Relative to Disposable
Personal Income, 1952–2019
Ratio to annual DPI
1.15

Years of DPI

2019:Q4

8

Total net wealth
(right axis)
1.05

0.95

6

Consumption
(left axis)
Stock market wealth
(right axis)
Net housing wealth

0.85

0.75
1950

4

(right axis)

2

0
1960

1970

1980

1990

2000

2010

Sources: Federal Reserve; Bureau of Economic Analysis; CEA calculations.
Note: DPI = disposable personal income. Data for 2019:Q4 values are estimated from the
latest daily or monthly data. Shading denotes a recession.

The Great Expansion

250-840_text_.pdf 47

| 43

2/7/20 3:46 PM

$"0- 
рҊруѵ
 -.*)'
1$)"
/ Ѷ
сппп–спрш

Percent
рс

спршѷу

рп
/0'
ч

х
- – 0'4
спрш

))0'
- 1$.$*)
у

с
сппп

сппу

сппч

*0- .ѷ 0- 0
*!
*)*($
)'4.$.Ѹ

'0'/$*).ѵ
*/ ѷ
#$)"
 )*/ .

-  ..$*)ѵ

спрс

спрх

contributed on average 1.9 percentage points to the quarterly real GDP growth
rate (figure 1-12).
Gains in household wealth (also known as net worth) have supported
the solid growth of real consumer spending during the past three years (figure
1-13), with gains in stock-market wealth and other housing wealth accounting
for the increase. Over long-periods, gains in the wealth-to-income ratio are correlated with consumer spending (Poterba 2000; Lettau and Ludvigson 2004).
From that point of view, the gains in the wealth-to-income ratio could have
supported an even larger increase in consumer spending.
The prospect of future consumer spending supporting overall output
growth is strong, given the elevated levels of consumer confidence. The
University of Michigan’s Index of Consumer Sentiment rose to 97.2 in 2019:Q4—
in the middle of the range in which it has fluctuated in the past three years—and
is currently 5.4 points above its 2016 level. The Conference Board’s version of
consumer sentiment fell to 126.5 in 2019:Q4, toward the lower end of the range
in which it has fluctuated in the past three years, but is still 26.7 points above
2016. These persistently strong readings for both measures indicate resilient
consumer demand, which represents a sizable portion of the U.S. economy,
and thus point to its continued support of growth.
Further, personal saving as a share of disposable personal income
remains elevated. After notable upward revisions by the Bureau of Economic
Analysis in July 2018, as reported in chapter 10 of the 2019 Economic Report of

44 | Chapter 1

250-840_text_.pdf 48

2/7/20 3:46 PM

the President, the saving rate was further revised upward in the Bureau’s July
2019 annual revision. The personal saving rate during 2019 of 8.0 percent far
exceeds the average of the last two decades (figure 1-14). The saving rate has
been increasing in the past three years due to the faster increase in personal
disposable income relative to the already robust growth in personal outlays.
The high saving rate together with elevated levels of household wealth, leave
some room for saving to buffer consumer spending against temporary adverse
developments in income.

Investment
In the past volumes of the Economic Report of the President, the CEA projected
that the Tax Cuts and Jobs Act would raise real capital investment on the basis
that lowering the user cost of capital would increase the target steady-state
flow of capital services; and this projection was based on a substantial body
of academic research. Chapter 1 of the 2019 Economic Report of the President
confirmed these anticipated positive effects with the then–available data up
through 2018:Q3. The positive effect of the TCJA on investment was also corroborated by outside studies (Kopp et al. 2019).
During the 9-quarter post-TCJA period, the annual rate of real private
nonresidential fixed investment growth averaged 3.4 percent, with growth
being faster in the first 4 quarters (6.8 percent) than in the next 5 quarters (0.8
percent).4 Some moderation of the investment growth rate was anticipated by
most models, which predicted that the positive effects on investment and overall economic activity would be front-loaded in 2018 (CEA 2019b; Mertens 2018).
In particular, standard neoclassical growth models suggest that during the
transition to the new steady state, the rate of growth in fixed investment would
initially spike, and would subsequently return to its pre-TCJA trend. Absent
other, exogenous shocks, the level would then remain at a higher, post-TCJA
level, with the capital-to-output ratio thereby asymptotically approaching its
new, higher steady-state level (CEA 2019b).
Figure 1-15 shows that the level of investment has been higher throughout the post-TCJA period than the consensus pre-TCJA projections (the March
2017 Blue Chip consensus). In 2018 as a whole, investment was 2.3 percent
higher than the consensus projection. In 2019, even with the recent investment slowdown, private nonresidential fixed investment was still 0.8 percent
higher than the pre-TCJA consensus projection. Also, compared with other G7
countries, the cumulative increase in investment, or the cumulative addition

4 Nine quarters are included in the post-TCJA period because the TCJA’s allowance for full
expensing of new equipment investment was retroactive to September 27, 2017 (the date of the
first draft of the proposed tax legislation that included the full expensing provision from the House
Ways and Means Committee).

The Great Expansion

250-840_text_.pdf 49

| 45

2/7/20 3:46 PM

Figure 1-15. Actual versus Preelection Projections for Nonresidential
Private Fixed Investment, 2014–19
Index (2016 = 100)

Election

115
112

2019

Actual, postelection

109
106
103

Pre-TCJA projection
consensus (March 2017)

Actual, preelection
100

Preelection consensus
projection (Oct. 2016)

97
2014

2015

2016

2017

2018

2019

Sources: Bureau of Economic Analysis; CEA calculations.
Note: Consensus forecasts from the October 2016 and March 2017 issues of Blue Chip Economic
Indicators begin with 2017 growth for levels implied by year-over-year forecasts.

Figure 1-16. Cumulative Change in Gross Fixed Private Capital
Formation among the Group of Seven Member Countries,
2017:Q4–2019:Q3
Change (percent)
8
6

6.5

5.9

5.7
4.4

4

3.7

2

0.6

1.6

0
-2
-4
-6
United
States

France Germany

Italy

Japan

–5.1
United Canada Australia
Kingdom

Sources: Australian Bureau of Statistics; Statistics Canada; Institut national de la statistique et des
études économiques; Deutsche Bundesbank; Istituto Nazionale di Statistica; Cabinet Office of
Japan; U.K. Office for National Statistics; Bureau of Economic Analysis; CEA calculations.

46 | Chapter 1

250-840_text_.pdf 50

2/7/20 3:46 PM

Figure 1-17. The User Cost of Capital, 2011–19
Percent

Index (2017:Q4 = 100)
112

2019:Q3

60

48

108

36

104
Top statutory corporate tax rate
(right axis)

24

100
User cost of capital
(left axis)

96

92
2011

12

0
2012

2013

2014

2015

2016

2017

2018

2019

Source: CEA calculations.

to the capital stock, since the TCJA’s enactment has been one of the highest
(figure 1-16).
Outside the expected slowdown in investment growth, other forces suppressed investment in 2019. One is the increase in the user cost of capital since
2018:Q3. From the CEA’s calculations, the user cost of capital is measured by
the Shiller cyclically adjusted Standard & Poor’s price/earnings ratio, in addition to a function of corporate tax rates and depreciation allowances. As seen
in figure 1-17, the user cost of capital fell sharply in 2018:Q1, when the TCJA
lowered the top statutory corporate tax rate from 35 percent to 21 percent, but
increased over the period 2018:Q4–2019:Q3. A confluence of factors—tighter
domestic monetary policy and lower stock market valuations, possibly due to
a global growth slowdown—all ultimately led to a tightening of financial conditions in 2018:Q4 and thereafter raised the user cost of capital.
The imprints of weaker global factors on investment can be seen in a
decomposition of nonresidential investment growth (figure 1-18). The slowdown in nonresidential investment in 2019 was mainly accounted for by business structures, which shrank 7.0 percent in 2019, and by equipment, which
decreased 1.5 percent. Intellectual property products investment, which is
less exposed to fluctuations in global conditions, grew at a robust pace of 6.2
percent in 2019.
The decline in structures investment was primarily because of a pullback in energy investment. Mining and wells investment fell 16.7 percent in
2019, and were a factor in about 45 percent of the slowdown in structures

The Great Expansion

250-840_text_.pdf 51

| 47

2/7/20 3:46 PM

Figure 1-18. Average Annual Growth in Real Business Fixed Investment
and Component Contributions, 2010–19
Structures

Equipment

Intellectual property

Total

Percent (annualized)
12
10
8
6
4
2
0
-2
-4
2010–11

2012–14

2015–16

2017

2018

2019

Sources: Bureau of Economic Analysis; CEA calculations.
Note: Average annual growth is measured on a Q4-over-Q4 basis for each year or multiyear period.

Figure 1-19. Real Mining and Drilling Structures Investment versus Oil
Rigs Operating in the United States, 2007–19
Percent change (annual rate)

Rigs
1,800

Total rigs (left axis)

1,600

2019:Q4

120
100
80

1,400

60

1,200

40

1,000

20

800

0

600

‐20

Business
investment in
mining and gas
(right axis)

400
200

‐40
‐60
‐80

0
2007

2009

2011

2013

2015

2017

2019

Sources: Bureau of Economic Analysis; Baker-Hughes; CEA calculations.

48 |

250-840_text_.pdf 52

Chapter 1

2/7/20 3:46 PM

Figure 1-20. Brent Crude Oil Prices versus Oil Rigs Operating in the
United States, 2007–19
Dollars per barrel

Rigs
1,800

Total rigs (left axis)

2019:Q4

1,600

140
120

1,400

100

1,200
1,000

80

Brent crude oil
(right axis)

800

60

600

40

400

20

200

0

0
2007

2009

2011

2013

2015

2017

2019

Sources: Bureau of Economic Analysis; Baker-Hughes; CEA calculations.

investment. As seen in figure 1-19, investment in mining and wells started
contracting in 2018:Q3, when market concerns about global growth escalated
and as oil prices fell to near the breakeven price for shale producers, which is
about $50 a barrel. As oil prices approached or fell below the breakeven price
for some producers, they responded by slowing drilling or deciding to reduce
the large inventory of drilled but not completed wells (figure 1-20). Indeed, the
U.S. rig count fell by 236 in December compared with a year earlier.
Equipment investment also contracted by 1.5 percent in 2019, compared
with 5.0 percent growth in 2018. Investment in equipment turned negative in
the first quarter, briefly bounced back in the second quarter, and returned to
negative in the third quarter. The two main equipment categories that most
exacerbated the slowdown are information processing and transportation. As
is discussed in more detail in the “Global Macroeconomic Situation” section of
this chapter, the transportation sector experienced a series of negative supply
and demand shocks from economies abroad, but by far the largest drag was
the decrease in domestic sales at the aircraft supplier Boeing. Confirming the
importance of global factors, the CEA finds that an investment accelerator
model augmented with foreign growth (proxied by a weighted average of
non-U.S. G7 growth) can explain a sizable portion of the recent slowdown in
equipment investment (see figure 1-21), compared with a fundamental version
of the neoclassical model.

The Great Expansion

250-840_text_.pdf 53

| 49

2/7/20 3:46 PM

Figure 1-21. Predictions of an Investment Accelerator Model, 2014–19
Four-quarter change (percent)
1.5
1.0

Out-of-sample forecast
2019:Q3
Neoclassical
model

Model with
foreign growth

0.5
0.0
-0.5

Actual

-1.0
-1.5
2014

2015

2016

2017

2018

2019

Sources: Macroeconomic Advisers; Robert Shiller; Bureau of Economic Analysis; Internal Revenue
Service; various national statistical offices; CEA calculations.
Note: Foreign growth is a weighted average of Group of Seven country growth, excluding the
United States.

Figure 1-22. The Growth in Number of Private Establishments versus
Small Business Optimism, 2000–2019
Small business optimism index
110

Four-quarter percent (change)
4
3

2019:Q3

105

Private establishments
(left axis)

2

100
95

1
0
-1

90

NFIB small business optimism
index (right axis)

85
80

-2
2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

Sources: Bureau of Labor Statistics; National Federation of Independent Business; CEA
calculations.
Note: A three-month moving average is used for the index from the National Federation of
Independent Business (NFIB). Data for private establishments are only available through 2019:Q2.

50 |

250-840_text_.pdf 54

Chapter 1

2/7/20 3:46 PM

The decreases in both structures and equipment investment suggest that
the slowdown in growth in the rest of world has constituted a strong headwind
to U.S. investment. Indeed, as figure 1-18 shows, the current slowdown in
investment is similar to the slowdown in 2015–16, a period that also experienced an investment slowdown precipitated by weakening conditions abroad.
A later section of this chapter further explores the international economic
developments that are weighing on U.S. growth.
To the extent that changes in business fixed investment predominantly
reflect actions of large multinational firms that were responding to fluctuations
in global demand conditions, this situation could conceal the developments
among smaller firms that are more domestically oriented.5 One of the TCJA’s
aims is lowering the business costs of small firms, which tend to be more
credit-constrained than large multinational firms. As figure 1-22 shows, this
predicted effect of the TCJA is supported by survey data, with 2018 level small
business optimism rising to the highest level in almost two decades, and the
number of private establishments surging in 2019.

Inflation
Despite a tight labor market, price inflation remains low and stable. Measures
of inflation expectations have also been stable. The stability of price inflation and of inflation expectations indicate the economy is not facing supply
constraints and has been a key factor in extending the duration of the current
expansion.
What is different about the structure of the recent economy that accounts
for the coexistence of a tight labor market and low and stable inflation—that
is, the flattening of the Phillips curve? Partial explanations include the falling relative price of imports, a different monetary policy regime, and recent
deregulatory actions.

Price Inflation
Key measures of price inflation are essentially flat, and are all roughly in the
range of 2 percent at an annual rate. The price index for GDP, the aggregate
price for everything that is produced in the United States, rose 1.7 percent during the four quarters of 2019, down from 2.0 and 2.3 percent in 2017 and 2018,
respectively. Consumer price inflation—as measured by the price of personal
consumption expenditures in the National Income and Product Accounts
(known as the PCE Price Index)—was only 1.5 percent during the four quarters
of 2019. With the exception of the third quarter in 2016, consumer price inflation has generally been below (or equal to) GDP price inflation for each of the
past eight years, as shown in figure 1-23.
5 A well-documented stylized fact in the international economics literature is that larger firms have
a higher propensity to export and import (WTO 2016).

The Great Expansion

250-840_text_.pdf 55

| 51

2/7/20 3:46 PM

$"0- 
рҊстѵ
 )!'/$*)ѷ
The 
-$ 
 ) 3
1 -.0.

-$ 
 ) 3Ѷ
сппш–рш
Four-quarter change (percent)
тѵп

спршѷу

сѵф


+-$ 

$) 3

сѵп
рѵф

рѵп


P-$ 
I) 3

пѵф
пѵп
Ҋпѵф
Ҋрѵп
сппш

спрс

спрф

*0- .ѷ
0- 0
*!
*)*($
)'4.$.Ѹ

'0'/$*).ѵ
*/ ѷ

ۙ
p -.*)'
c*).0(+/$*)
e3+ )$/0- .ѵ
#$)"
 )*/ . 
-  ..$*)ѵ

спрч

$"0- 
рҊсуѵ
 (+*-/
-$ .
1 -.0.

-$ 
 ) 3Ѷ
ршфф–спрш
Log levels
рѵт

спршѷу

ѷ
#$)
+-$ 
$) 3

рѵс
рѵр
рѵп

(+*-/.
*!
"**.
)
. -1$ .ѷ

#$)
+-$ 
$) 3

пѵш
пѵч
пѵц
пѵх
ршфф

ршхр

ршхц

ршцт

ршцш

ршчф

ршшр

ршшц

сппт

сппш

спрф

*0- .ѷ
0- 0
*!
*)*($
)'4.$.Ѹ

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

)
-*0/
*0)/.ѵ
# 
$)$ .
- 
'*"" 
)
- )*-('$5 ѵ
#$)"
 )*/ .

-  ..$*)ѵ

52 |

250-840_text_.pdf 56

Chapter 1

2/7/20 3:46 PM

One reason that consumer price inflation has been below the pace of GDP
price inflation has been the persistent decline in the relative price of imports.
During the eight quarters through 2019:Q4, import prices did not increase,
while GDP prices (i.e., goods and services produced in the United States)
increased at a much faster rate of 2.0 percent, so that the relative price of
imports fell at a 2.0 percent annual rate. The declining relative price of imports
has held down consumer price inflation (1.7 percent over eight quarters) by
more than it has held down GDP price inflation because imported goods and
services are included directly in consumer prices, but influence GDP prices only
indirectly through competition.
A situation of declining relative prices of imports has not always been the
case, as can be seen in figure 1-24, which shows the log levels of GDP prices
and the log levels of import prices. In particular, import prices increased 1.6
percentage points per year faster than GDP prices from 1955 to 1981, increased
1.7 percentage points more slowly from 1981 through 2011, and increased 3.1
percentage points more slowly during the eight years since 2011. As can be
seen in figure 1-24, the separation between the log levels of GDP and import
prices is currently the largest recorded in the 1955–2019 period.

Different Measures of Inflation: The CPI, Chained CPI, and PCE
Price Index and Their Cores
The Consumer Price Index (CPI) tends to increase slightly faster—by about
0.29 percentage point a year, on average—than the PCE Price Index.6 These
two commonly used measures of consumer prices are both important. The CPI
tends to overstate a cost-of-living price index, however, largely because it uses
a fixed market basket updated every two years, which means that it does not
capture real-time substitution by consumers toward goods and services with
declining relative prices. Another version of the CPI, known as the chained CPI,
corrects for this substitution bias, and as a result also rises about 0.28 percentage point per year less than the official CPI. The chained CPI is now used to
index the notches in the new TCJA tax schedules. The PCE Price Index also
begins with most of the same CPI components and aggregates with a formula
that allows for substitution.
Price indices that exclude the volatile components of food and energy
provide a smoother signal of inflation trends than the overall index. The core
CPI (which excludes food and energy) increased 2.3 percent during the 12
months of 2019, up only slightly from the 2.2 percent year-earlier pace. The PCE
Price Index version of core inflation rose 1.6 percent in 2019, down from the
year-earlier pace of 1.9 percent. The 2019 rate of core PCE inflation was below
the Federal Reserve’s target of 2.0 percent, as was the rate of overall PCE inflation, as shown in figure 1-25.
6 Computed from 2002:Q4 to 2018:Q4.

The Great Expansion

250-840_text_.pdf 57

| 53

2/7/20 3:46 PM

Figure 1-25. Consumer Price Inflation, 2012–19
Percent change (12-month)
3.0

Nov-19

2.5

Federal Reserve’s long-run objective
2.0

Core PCE Price
Index

1.5

1.0

0.5

0.0

Total PCE Price Index

2012

2013

2014

2015

2016

2017

2018

2019

Sources: Bureau of Economic Analysis; CEA calculations.
Note: PCE = personal consumption expenditures.

$"0- 
рҊсхѵ
*- 
 
 )!'/$*) )
 )!'/$*)
3+ //$*).Ѷ

ршхп–спрш
Percent
рх

 Ҋрш

ру
рс
рп
ч

*- 
 Ҋ

х

)$1 -.$/4
*!

$#$")

.0-1 4ѷ

3+ / 
фҊ
4 -
$)!'/$*)

 
фҊ4 -

!*-2-

3+ / 

$)!'/$*)

*(+ )./$*)

у

с
п
ршхп ршхф ршцп ршцф ршчп ршчф ршшп ршшф сппп сппф спрп спрф

сѵт

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

)'4/$.ѵ
*/ ѷ

# 
рсҊ(*)/#
+ - )/
#)" 
$.
/& )
!*-
*- 
 Ҋѵ

тҊ(*)/#
 )/ - 
(*1$)"
1 -" 

$.
/& )
!*-
/# 
)$1 -.$/4
*!
$#$")
.0-1 4
)
- .0-4
 )!'/$*)
-*/ / 
 0-$/$ .
җ Ҙѵ

#$)"
 )*/ .

-  ..$*)ѵ

54 |

250-840_text_.pdf 58

Chapter 1

2/7/20 3:46 PM

Measures of inflation expectations have also been stable at a rate close to
the 2.0 percent Federal Reserve target, as shown in figure 1-26, which graphs
two measures: one from the University of Michigan’s Survey of Consumers, and
one extracted from the market for the Treasury’s Inflation Protected Securities.
Buttressed by the stability of core inflation, and of expectations of core
inflation, the Administration forecasts rates of increase in the CPI at 2.3 percent
and the GDP price index at 2.0 percent during the 11-year Budget forecasting
interval.

Hourly Compensation Inflation, Productivity Growth, and
Stable Inflation
Nominal hourly compensation inflation—as measured by the Employment
Cost Index for the private sector—increased by 2.7 percent at an annual rate
during the 12 months of 2019, down slightly from the 3.0 percent 2018 pace.
This 2.7 percent pace edged up from the annual pace of 2.1 percent during the
four years through 2016.
Over long periods, wage inflation can exceed price inflation by the rate
of labor productivity growth. And over the seven quarters through 2019:Q3,
nonfarm labor productivity grew at a 1.4 percent annual rate. As a result, the
roughly 3.0 percent rate of annual hourly compensation growth (which suggests unit labor costs rising at 1.6 percent) is compatible with price inflation
of 2 percent (or slightly less), without putting upward pressure on the price
structure.
The sensitivity of inflation to fluctuations in the unemployment rate has
decreased during the past two decades, as shown in the scatter diagram given
in figure 1-27, which illustrates a version of the Phillips curve. The vertical axis
shows the difference in core PCE inflation relative to a year-earlier survey of
inflation expectations. The horizontal axis shows a version of the unemployment rate, one that is demographically adjusted to control for the major
fluctuations in the share of young people in the labor force during these past
60 years. (The share of young people in the labor force was exceptionally high
in the 1970s, when the baby boom cohorts entered the labor market.)
As can be seen in figure 1-27 by the blue regression line fitted through
the early years 1960–2000, an extra percentage point of unemployment lowered the rate of inflation by 0.36 percentage point a year. In contrast, the red
regression line fitted on the last 19 years (2000–2018) indicates that an extra
percentage point of unemployment lowered the rate of inflation by only 0.08
percentage point. One could argue that this shallow slope estimated during
the past 20 years provides the best guide to the future. Or one might argue that
the best estimate of the slope is the one covering the entire 60-year sample
(0.27 percentage point of inflation per 1 percentage point of unemployment;
not shown).

The Great Expansion

250-840_text_.pdf 59

| 55

2/7/20 3:46 PM

Figure 1–27. Price-Price Phillips Curve Scatter Diagram,
1960–2018
Percent change in core PCE
6

5
4
3

1960–2000
Slope = –0.36

2

2000–2018
Slope = –0.08

1
0
-1
-2
-3
2

3

4

5

6

7

8

9

10

Fixed-weighted unemployment rate (2017 weights)
Sources: Federal Reserve Bank of Philadelphia; Bureau of Economic Analysis; Bureau of Labor
Statistics; CEA calculations.
Note: PCE = Personal consumption expenditures. Inflation expectations are measured by the
Livingston Survey for 1960–70; by the Survey of Professional Forecasters’ (SPF) 10-year Consumer
Price Index for 1970–90; and by the SPF expectation for 10-year PCE inflation for 1990–2018.

Table 1-1. Effects of Deregulation on Relative Price Increases on the Core CPI,
2006–19
Priced good/service

Ten-year % change
in relative prices,
Dec. 2006–Dec.
2016, AR

(1)

34-month %
change
Change
since Dec. in trend,
2016, AR
p.p.
(2)

Relative
importance
weight in Core
CPI

(3)
= (2) – (1)

Effect on
Core CPI
inflation

(4)

(5)
= (3) * (4)

Prescription drugs

1.62

–0.96

–2.58

1.711

–0.044

Internet services

–1.83

–2.28

–0.44

0.952

–0.004

Sources: Bureau of Labor Statistics; CEA calculations.
Note: AR = annualized rate; p.p. = percentage point; CPI = Consumer Price Index.

Explanations for the declining slope of the Phillips curve include the
influence of import prices in holding down the rate of inflation in recent years
(as argued above), the wage and price rigidity that kept inflation from falling
below zero during the early years of this recovery (2009–13), the diminishment
of the Phillips curve coefficient in a monetary policy regime that effectively
targets inflation (Hooper, Mishkin, and Sufi 2019), and the evolution of the
input-output structure of the economy toward increasing intermediate inputs
(Rubbo 2020). Another possible explanation is the deregulation efforts of the
current Administration.

56 | Chapter 1

250-840_text_.pdf 60

2/7/20 3:46 PM

Deregulation and Inflation
As discussed in chapter 3 of this Report, estimates suggest that deregulation
has lowered the relative price of prescription drugs and Internet services. We
calculate that these effects lower total inflation by about 0.05 percentage point
a year. The relative price of prescription drugs, in particular, is increasing by 2.6
percentage points a year less that during the 10 years through 2016; see table
1-1. To summarize this analysis, inflation remains low and stable, inflation
expectations are well anchored at this low level, and recent estimates of the
Phillips curve suggest a diminishing sensitivity of inflation to unemployment
rates.

The Global Macroeconomic Situation
As alluded to in previous sections, a major headwind to growth in 2019 was a
synchronized slowdown in global growth. In its latest semiannual economic
outlook, the International Monetary Fund (IMF 2019c) revised down global
growth sharply, by 0.7 percentage point, to what would be the lowest growth
rate since the Global Recession, 3 percent—one of the largest one-year
downward-revisions in recent years (figure 1-28). Among advanced economies, growth was revised down by 0.4 percentage point, with growth disappointments concentrated in Europe, especially Germany. Emerging market
economies also saw a downward revision, of 0.8 percentage point. Amid this
global slowdown, the U.S. economy has performed largely as projected by the
IMF in October 2018, growing faster than any other G7 country in the first three
quarters of 2019 (figure 1-29).
At the heart of the current global slowdown has been a manufacturing
downturn. Uncertainty about trade policy is one often-cited culprit in the manufacturing slowdown, particularly uncertainty surrounding the Administration’s
negotiations toward a bilateral trade agreement with the People’s Republic of
China on enforceable commitments to remove or lower structural barriers in
China (BIS 2019a, 2019b; IMF 2019a, 2019b; OECD 2019a; World Bank 2019a,
2019b). However, other reasons for the global manufacturing slowdown also
preceded, or were contemporaneous with, trade policy developments. These
reasons make it difficult to isolate the effects of trade policy uncertainty, and
possibly result in an upward bias of its effects on the global economy. Other
factors weighing on manufacturing include a change in European automobile
emission standards in September 2018 that caused a production bottleneck in
Europe, especially Germany, and a growth slowdown in China caused by the
government’s efforts to deleverage the financial system beginning in 2017. The
manufacturing sectors of these two countries—two of the world’s preeminent
manufacturing powerhouses—had begun slowing down before or around the
time of the imposition of tariffs on Chinese goods by the current Administration
(figure 1-30).

The Great Expansion

250-840_text_.pdf 61

| 57

2/7/20 3:46 PM

Figure 1-28. IMF Five-Year Real GDP Growth Forecasts for the World,
2012–24
Year-over-year percent change
4.5

Forecast
Oct. 14

Forecast
Oct. 15
Forecast
Oct. 16 Forecast
Oct. 17

4.0

Actual
growth
3.5

Forecast
Oct. 19

3.0

2.5
2012

2014

2016

2018

2020

Forecast
Oct. 18

2022

2024

Source: International Monetary Fund.
Note: Each forecast is taken from World Economic Outlook, which is published by the IMF in
October of each year.

Figure 1-29. Forecast of 2019 Real GDP Growth
Oct. 2018 WEO

Oct. 2019 WEO

Year-over-year percent change
3.0
2.5

2.0
1.5

1.0
0.5
0
United
States

Germany

Italy

France

United
Kingdom

Japan

Canada

Source: International Monetary Fund.
Note: WEO = World Economic Outlook, published annually by the IMF.

58 |

250-840_text_.pdf 62

Chapter 1

2/7/20 3:46 PM

Figure 1-30. Composite Output Purchasing Manager’s Index
(PMI), 2015–19
Index ( > 50 = expansion)
58

2019:Q4

57

Germany’s PMI

56
55

Global PMI

54
53
52

China’s PMI

51
50
49
48
2015

2016

2017

2018

2019

Sources: Caixin; IHS Markit; JPMorgan Chase.
Note: Index levels over 50 represent an expansion.

The Administration’s efforts to create a more reciprocal environment
and rebalance the trading relationship between the United States and China
required negotiation over how this new relationship should be shaped.
Negotiations have covered a wide range of critical issues, including the ways
that U.S. companies are required to transfer proprietary technology as a condition of market access; the numerous tariff and nontariff barriers faced by U.S.
businesses in China; and China’s other market-distorting practices and policies
that have weighed on U.S. and global economic growth, such as industrial
subsidies and support for state-owned enterprises.
China’s weak protection and enforcement of intellectual property rights
is symptomatic of a broader challenge. Chinese firms engage in systematic
theft of U.S. intellectual property because the costs are insufficient to incentivize them to do otherwise.7 Instead of pursuing an enforceable bilateral trade
agreement through targeted tariffs, prior Administrations took a multilateral
approach that imposed no costs on the offenders and failed to resolve these
issues. The Administration first imposed tariffs on imports from China based on
7 There is a common misconception that the grievances against China relate exclusively to
intellectual property. Although Chinese forced technology transfer and intellectual property theft
(discussed at length in the Section 301 investigation) are important, the actions are also designed
to address a number of other long-standing trade issues with China: expanding the Chinese market
access for services and agriculture, implementing an agreement like the United States–Mexico–
Canada Agreement’s provision on currency, addressing the many nontariff barriers on U.S. exports
to China, and increasing Chinese purchases of U.S. products (White House 2018).

The Great Expansion

250-840_text_.pdf 63

| 59

2/7/20 3:46 PM

the findings of the Section 301 investigation of China’s acts, policies, and practices related to technology transfer, intellectual property, and innovation. The
Administration then took supplemental action in 2018 and 2019 in response
to China’s imposition of retaliatory tariffs and failure to eliminate these unfair
acts, policies, and practices.
These Administration actions have prompted a renegotiation of the trading relationship between the two countries. Studies that examined the effect
of the tariffs point out that tariffs impose near-term costs on the United States
(Amiti, Redding, and Weinstein 2019a, 2019b; Caldara et al. 2019; Fajgelbaum
et al. 2019).8 Negotiations over a new agreement necessitate a degree of
uncertainty over how that agreement will be shaped, exacerbating near-term
costs. However, achieving a new trade relationship with China that is balanced
and reciprocal will deliver long-term economic benefits for the United States,
including a reduction in near-term costs.
In January 2020, the Administration finalized a historic and enforceable
agreement on phase one of the trade deal. The trade deal requires structural
reforms and other changes to China’s economic and trade policies in the areas
of intellectual property, technology transfer, agriculture, financial services, and
currency and foreign exchange. The ultimate goal is that, with lower market
barriers and further market orientation in China, the global trading system will
operate in a more balanced, reciprocal environment. Global growth, as a result,
would benefit from the increase in trade liberalization.
While trade policy uncertainty has held the spotlight, another underappreciated reason for the global manufacturing slump was both supply and
demand problems in the global motor vehicle industry. Supply problems in
the European motor vehicle industry were precipitated by a change in the
European Union’s emissions regulations in September 2018, which led to
bottlenecks at testing agencies and production cuts from automobile manufacturers to avoid unwanted inventory accumulation. Germany, a global hub
for automobile production, particularly felt the impact of the supply disruption
(Deutsche Bundesbank 2019; IMF 2019b). German automobile production fell
10 percent in 2018 as a whole, and shrank another 9 percent in 2019. Given its
long global value chains and sizable share in global output and global exports,
weaknesses in the automobile sector extend well beyond the industry in
Europe, propagating the shock through upstream industries around the world
like steel, metal, and automobile parts, as well as downstream industries like
services (OECD 2019b).9
8 Caldara et al. (2019) look at the costs imposed by this trade policy uncertainty and find cumulative
costs of up to 1 percent of GDP after two years. Amiti, Redding, and Weinstein (2019b) examine the
direct impact of implemented tariffs in 2018 and 2019 and find that they impose a net deadweight
loss of 0.4 percent of GDP per year. Fajgelbaum et al. (2019) find that the additional tariffs in 2018
imposed a cost of 0.04 percent on GDP after accounting for tariff revenues and gains to domestic
producers.
9 The automobile sector accounts for 5 percent of global output and 8 percent of global exports.

60 |

250-840_text_.pdf 64

Chapter 1

2/7/20 3:46 PM

Figure 1-31. China’s Change in Automobile Sales, 2014–19
Twelve-month change (percent)
25

Dec-19

20
15
10
5
0
-5
-10
-15
-20
2014
2015
2016
2017
2018
2019
Sources: China Association of Automobile Manufacturers; CEA calculations.

These adverse shocks to the motor vehicle industry were further compounded by a cyclical downturn in automobile demand in China. Efforts by
China’s authorities to deleverage the shadow-banking sector since 2017 have
led to a protracted slowdown in credit growth, including consumer credit.
Increasing difficulty in accessing credit, heightened risk aversion among households in a slowing economy, and the termination—in 2019—of consumer tax
breaks for automobile purchases in 2017–18 all led to a substantial pullback
in Chinese automobile consumption. As a result, China’s automobile consumption has contracted in consecutive quarters since mid-2018 (figure 1-31),
and has accounted for over half the global contraction of automobile sales.
Accordingly, the quantity of German automobile exports, for which China is an
important market, have plunged since early 2018, and were 14 percent below
the mid-2018 level, as of November 2019 (figure 1-32).
Beyond the problems in the automobile industry and the slowdown in
China, country-specific shocks have also exacerbated the global slowdown. In
the United Kingdom, uncertainty over Brexit has continued to weigh on growth.
After the U.K. Parliament failed to ratify a deal negotiated between Prime
Minister Boris Johnson’s government and the EU, his government secured an
extension of the Brexit deadline to January 2020. With the December 2019 elections in the U.K. securing a large majority for Johnson’s party in Parliament,
Parliament passed legislation for Britain to leave the European Union with a

The Great Expansion

250-840_text_.pdf 65

| 61

2/7/20 3:46 PM

Figure 1-32. German Vehicle and Car Engines Exported, 2016–19
Index (June 2018 = 100)
110

Nov-19

105

100

Value
95

90

85
2016

Quantity

2017

2018

2019

Sources: Federal Statistical Office; Kraftfahrtbundesamt; CEA calculations.

withdrawal agreement on January 31, 2020, after which the U.K. will enter a
transitional period and adhere to EU rules until end of 2020.
Japan, after experiencing surprisingly positive growth of 2.3 percent
at annual rate in the first half of 2019, saw its growth edge down to a 1.8
percent annual rate in the third quarter, as exports slumped amid weakening
global demand, mainly due to a drop in demand from China and a boycott of
Japanese goods in South Korea. The long-planned sales tax increase from 8
to 10 percent also came into effect in October, causing consumer spending to
plummet.
Emerging market economies, which until 2018 had been an engine of
global growth, became a drag in 2019. After months of antigovernment protests, Hong Kong entered its first recession since the global financial crisis.10
In India, increasing defaults in the shadow-banking sector have resulted in a
large pullback of domestic credit growth, causing GDP growth to slow sharply.
In Mexico, uncertainty over domestic policies, reinforced by the sudden resignation of Mexico’s financial minister, and the slowdown in global trade have
impeded growth. Meanwhile, growth remains weak in Brazil, as high public
debt levels have constrained the government from using fiscal stimulus to
further support the economy in the face of subdued domestic and external
demand.
10 Hong Kong’s real GDP contracted by 1.9 percent at an annual rate in 2019:Q2 and by 12.1 percent
in 2019:Q3.

62 |

250-840_text_.pdf 66

Chapter 1

2/7/20 3:46 PM

Figure 1-33. Central Bank Policy Rates, 2010–19
Percent

Dec-19

2.5
2.0

United States

1.5
1.0
0.5

Japan
Sweden
Euro area
Denmark
Switzerland

0.0
-0.5
-1.0
2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

Sources: Swiss National Bank; Sveriges Riksbank; Danmarks Nationalbank; Federal Reserve Board;
European Central Bank; Bank of Japan.
Note: For Japan, the effective uncollateralized overnight call rate was used.

The U.S. Dollar and Monetary Policy
Because of the weak international economic outlook, several non-U.S. major
economies eased monetary policies throughout 2019. In particular, the
European Central Bank announced in September that it would resume its
asset purchase program at a pace of €20 billion a month, and it lowered its
policy rate by 10 basis points to –0.5 percent. The National Bank of Denmark
(a non-euro country) also followed the European Central Bank in lowering its
policy rate further into negative territory. Global negative-yielding sovereign
debt—mostly issued by European countries—has recently reached a record
amount of about $15 trillion.
In contrast, in response to an improved outlook for the U.S. economy,
the Federal Reserve began to normalize its balance sheet in December 2015.
During the years 2016–18, the Federal Reserve raised its policy rate eight
times, while several central banks across Europe (Denmark, the European
Central Bank, Sweden, and Switzerland) kept their policy rates negative (figure
1-33). Though the Federal Reserve subsequently reduced rates on three occasions in 2019, U.S. policy rates continued to exceed those of other advanced
economies, which induced capital inflows into the United States, and in turn
contributed to an appreciation of the dollar through September 2019, before it
edging lower during the final three months of the year.
Looking through the fluctuations of 2019, the real and nominal tradeweighted broad dollar was little changed from December to December.

The Great Expansion

250-840_text_.pdf 67

| 63

2/7/20 3:46 PM

$"0- рҊтуѵ
  -'
 . -1 
- Ҋ $"#/  -*
*($)'
1 -.0.

 '
*''-Ѷ
ршцт–спрш
Index (January 2006 = 100)
руп
-#
ршчфѶ
ртрѵф

 Ҋрш
 ѵ
сппсѶ
ррчѵс

рсп

 '

рпп

 ѵ
сппсѶ
ррхѵх

чп

-#
ршчфѶ
хсѵц

 '

#$./*-$'

1 -" Ѷ
шшѵу

*($)'

хп

уп
сп
ршцт

ршцч

ршчт

ршчч

ршшт

*0- .ѷ
  -'
 . -1 
*-Ѹ

'0'/$*).ѵ
*/ ѷ
#$)"  )*/ .

-  ..$*)ѵ

ршшч

сппт

сппч

спрт

спрч

Relative to other major advanced country currencies, the dollar edged up 0.6
percent over the same period in real terms. Curcuru (2017) finds that for every
divergence of 1 percentage point in interest rates between the United States
and other advanced economies, the real advanced dollar index appreciates
3.4 percent. Applying this elasticity, one finds that the interest rate differential
between the United States and the other G7 countries would have predicted
a depreciation of 2.6 percent in the advanced dollar.11 As of December, the
real level of the broad dollar is 7.8 percent higher than its historical average
calculated from 1973 January to the present, though most of the appreciation
occurred from the summer of 2014 to 2015 (figure 1-34). The real broad dollar
is, however, still below the record highs of 1985 and 2002.
Although higher U.S. interest rates than in other advanced countries
would, ceteris paribus, cause some dollar appreciation and reduce U.S.
exports, monetary spillovers from abroad also have an offsetting positive economic effect by lowering the longer end of the Treasury yield curve. This effect
could be observed in August 2019, when data in Germany and China that were
weaker than expected triggered global growth concerns that caused an immediate influx of safe haven flows to the U.S. Treasury market. Market expectations of future easing actions by the European Central Bank then caused
an immediate decrease in U.S. 10-year Treasury yields, contributing to the
11 Collins and Truman (2019) employed the same methodology for the period July 2014–September
2019, and found that 4.1 percentage points of the 21 percent appreciation in the major dollar over
this period was due to the United States / G7 interest rate differential.

64 | Chapter 1

250-840_text_.pdf 68

2/7/20 3:46 PM

inversion of the yield curve at that time. As a result, U.S. mortgage rates came
down, which on the whole supported the U.S. housing market and allowed U.S.
households to refinance their mortgages, unlocking more disposable income
for consumption.

Domestic Headwinds
In addition to international headwinds, four other idiosyncratic domestic factors impeded U.S. growth by almost 0.3 percentage point in 2019: (1) the partial
government shutdown for 25 days in January, (2) the grounding of Boeing 737
MAX jets, (3) industrial action at General Motors, and (4) the Midwest’s spring
flooding.12
Boeing. After two fatal accidents of the Boeing 737 MAX in 2018 and
2019, civil aviation authorities around the world (including the United States)
grounded the aircraft. The accidents and eventual grounding caused Boeing
737 deliveries to collapse to nearly zero, and production to fall. This drop in
production and deliveries lowered GDP because fewer planes were produced,
and those produced were placed into inventory instead of being delivered. The
CEA estimates that these effects depressed real GDP growth during the four
quarters of 2019 by 0.14 percentage point.
GM strikes. In mid-September, the United Auto Workers began a work
stoppage that halted production at General Motors for six weeks. The CEA
estimates that the strike subtracted at most 0.08 percentage point from GDP
growth in the four quarters of 2019; but the effects will be reversed by an equal
amount in 2020.
Midwest flooding. Production of corn and soybeans (the Nation’s most
valuable crops, at about $51 billion and $39 billion in 2018, respectively) fell
in 2019 by 4.4 percent and 19.8 percent. Spring flooding—due to excessive
rain and snowmelt, which damaged production in the Upper Midwest—may
be partly responsible for the decline in production. We estimate that these
declines reduced the value of corn and soybean crops (the major crops
throughout the Midwest) by $10 billion in 2019, or 0.04 percent of GDP.

Conclusion
This chapter has shown that despite strong headwinds from the global
economy and expectations of growth moderating as the current expansion
matures, the U.S. economy continued expanding at a healthy pace in the
past year. During 2019, consumer spending continued to grow strongly, while
the labor share of income continued to increase. The labor market tightened
further, even after strong gains in the previous two years. Wages rose faster
12 The partial government shutdown affected the 2019 level of real GDP, as well as the 2019 annual
average-to-annual average growth rate, but not the 2019 fourth quarter–to–fourth quarter growth
rate.

The Great Expansion

250-840_text_.pdf 69

| 65

2/7/20 3:46 PM

than inflation, which ultimately boosted real middle-class incomes. After
years of decline, the stabilization of labor force participation, due to increased
prime-age participation, combined with capital deepening to boost potential
long-term output.
The tepid recovery from the Great Recession in the years before the
Trump Administration prompted economic forecasters to project pessimistic
growth into the future, reflecting a widespread belief that the U.S. economy
is in the midst of a period of secular stagnation. But the first three years of the
current Administration have demonstrated that stagnation is not inevitable.
And the Administration’s structural reforms—including lower taxes, deregulation, and pro-innovation energy policies—can overcome secular stagnation
and have set the stage for continued economic strength.
As the current record expansion matures beyond the 42nd quarter, some
worry that the expansion will “die of old age.” But academic evidence indicates
that expansions do not end simply because of their length. Old age does not
kill expansions, though bad policies and exogenous shocks can and do lead to
recessions. The United States’ historically strong labor market, the potential
for further deregulation, and the capital deepening that is having a positive
impact on productivity suggest that there is still substantial room to grow in
the present U.S. expansion.

66 |

250-840_text_.pdf 70

Chapter 1

2/7/20 3:46 PM

x
Chapter 2

Economic Growth Benefits
Historically Disadvantaged
Americans
The U.S. labor market is the strongest it has been in the last half century, as President Trump’s pro-growth economic policies continue boosting
labor demand and lowering structural barriers to entering the labor market.
Economic data show that recent labor market gains disproportionately benefit
Americans who were previously left behind. These groups are becoming more
and more self-reliant through their economic activity, rather than remaining
inactive in the labor market to qualify for means-tested government programs.
Under the Trump Administration, and for the first time on record, there are
more job openings than unemployed people. In 2019, the U.S. unemployment rate has reached 3.5 percent, the lowest rate in five decades. Falling
unemployment has reduced the share of the population on unemployment
insurance to the lowest level since recording started in 1967. Importantly, the
African American unemployment rate has hit the lowest level on record, and
series lows have also been achieved for Asians, Hispanics, American Indians
or Alaskan Natives, veterans, those without a high school degree, and persons
with disabilities, among others.
Since the 2016 election, the economy has added more than 7 million jobs,
far exceeding the 1.9 million predicted by the Congressional Budget Office
in its final preelection forecast. These gains have brought people from the
sidelines into employment. In parts of 2019, nearly three quarters of people
entering employment came from out of the labor force—the highest rate on
record. And the prime-age labor force is growing, reversing losses under the

67

250-840_text_.pdf 71

2/7/20 3:46 PM

prior administration’s expansion period. This evidence suggests that the labor
market’s revival over the past three years is not a continuation of past trends
but instead is the result of President Trump’s pro-growth policies.
The Trump Administration’s policies are not only leading to more jobs but
also to higher pay. While nominal wage growth for all private-sector workers
has been at or above 3 percent for all but one month in 2019, wage growth for
many historically disadvantaged groups is now higher than wage growth for
more advantaged groups, as is the case for lower-income workers compared
with higher-income ones, for workers compared with managers, and for African
Americans compared with whites. These income gains mark a fundamental
change relative to those opposite trends observed over the expansion before
President Trump’s inauguration, contributing to reduced income inequality.
Employment and earnings gains continue pulling people out of poverty and
off of means-tested welfare programs. The number of people living in poverty
decreased by 1.4 million from 2017 to 2018, and the poverty rates for blacks and
Hispanics reached record lows. Food insecurity has fallen, and there are nearly
7 million fewer people participating in the Supplemental Nutrition Assistance
Program (SNAP, formerly known as the Food Stamp Program) than at the time
of the 2016 election. The caseload for Temporary Assistance for Needy Families
(TANF) has fallen by almost 700,000 individuals, and the number of individuals
on Social Security Disability Insurance has fallen by almost 380,000 since the
2016 election. Similarly, due primarily to rising incomes, Medicaid rolls are
decreasing.
Today’s strong labor market helps all Americans, but the largest benefits are
going to people who were previously left behind during the economic recovery.
Additional deregulatory actions targeted at remaining barriers in the labor
market will allow the economy to add to its record-length expansion and lead
to further employment and income gains, particularly for these historically
disadvantaged groups.

68 |

250-840_text_.pdf 72

Chapter 2

2/7/20 3:46 PM

T

he U.S. labor market is the strongest it has been in the last half century,
as shown by economic data across various metrics. President Trump’s
pro-growth economic policies are contributing to this strength. While
the economic gains realized over the past three years are widespread, this
chapter shows that they are disproportionately benefiting Americans who
were previously left behind during the recovery. The Administration’s policies increase labor demand and decrease structural barriers to entering labor
markets. This approach has contributed to reduced inequality through an economic boom that is greatly benefiting historically disadvantaged groups. These
groups are becoming more and more self-reliant through economic activity
rather than by remaining economically inactive to qualify for means-tested
government programs.
Today’s tighter labor market and the resulting wage growth are predictable outcomes of the Administration’s historic tax cuts and deregulatory
actions, which have delivered continued economic expansion. Eliminating
unnecessary regulatory burdens and lowering taxes spur labor demand and
incentivize firms to make productivity-enhancing investments (see chapter 3).
As a result, worker productivity, wages, and employment all increase.
Ultimately, these policies help boost the job market’s continued expansion, as increased demand with unchanged supply raises quantity (employment) and prices (wages) in labor markets.1 The United States has experienced 111 consecutive months of positive job growth, continuing the longest
positive job growth streak on record. The civilian unemployment rate, which
in December 2019 remained at its 50-year low of 3.5 percent, has been at or
below 4 percent for 22 consecutive months. Today’s historically low level of
unemployment makes rapid job creation more difficult as it becomes harder
for companies to find available workers. Since the Bureau of Labor Statistics
(BLS) started collecting data on job openings in 2000, the number of unemployed people exceeded the number of recorded available jobs until March
2018. Since then, there have been more job openings than unemployed people
for a remarkable 20 consecutive months.
In total, since the 2016 election, the economy has added 7 million jobs,
more than the population of Massachusetts.2 These job gains are impressive,
given that the economic recovery since the Great Recession became the longest in United States history during the summer of 2019. Figure 2-1 shows the
total number of jobs by quarter. Before the 2016 election, the Congressional
Budget Office (CBO) expected job growth to slow and the total number of jobs
to level off, as workers who were out of the labor force were largely expected to
remain on the sidelines (CBO 2016). Instead, job growth under President Trump

1 Tax cuts also increase the supply of labor, as after-tax wages increase for a given pretax wage.
Because supply and demand both increase, quantity will increase and the effect on price (wage)
will depend on the relative magnitude of the increases.
2 The most recent jobs data are preliminary and are subject to revision.

Economic Growth Benefits Historically Disadvantaged Americans | 69

250-840_text_.pdf 73

2/7/20 3:46 PM

*/ ѷ 
ۙ
*)"- ..$*)'
0" /
!!$ ѵ

!*- ./
$.
!-*(
0"0./
спрхѵ


$"0- 
сҊрѵ
*/'
 *.
1 -.0.
спрс–рш

' /$*)
*- ./Ѷ


Total jobs (millions)
рфф

рфп

руф

спршѷу
*/'
%*.



!*- ./

руп

ртф

ртп
спрсѷр спртѷр спруѷр спрфѷр спрхѷр спрцѷр спрчѷр спршѷр
*0- .ѷ
*)"- ..$*)' 0" /
!!$ Ѹ
0- 0
*!
 *-
//$./$.Ѹ

'0'/$*).ѵ
*/ ѷ
 ۙ
*)"- ..$*)'
0" /
!!$ ѵ

!*- ./
$.
!-*(
0"0./
спрхѵ

has far exceeded the 1.9 million predicted by this point in the recovery by the
CBO in its final preelection forecast. Americans coming from the sidelines to
get jobs have led to employment growth at a similar rate as before the election,
even as the unemployment rate has fallen to historic lows. Similarly, before the
election, the CBO and the Federal Reserve forecasted that the unemployment
rate, which had been declining steadily for many years, would level off at about
4.5 percent, as seen in figure 2-2 (FOMC 2016).
As it becomes more difficult for employers to find available workers,
employers will offer higher pay or expand the pool of workers whom they
consider. Annual nominal wage growth reached 3 percent in 2019 for the first
time since the Great Recession, and nominal wage growth has been at or above
3 percent for all but one month in 2019. Importantly, wage growth for many
disadvantaged groups is now higher than wage growth for more advantaged
groups. And the lowest wage earners have seen the fastest nominal wage
growth (10.6 percent) of any income group since the Tax Cuts and Jobs Act
was signed into law. Beyond this pay increase for low-income workers, from
the start of the current expansion to December 2016, average wage growth for
workers lagged that for managers, and that for African Americans lagged that
for white Americans. Since President Trump took office, each of these trends
has been reversed, contributing to reduced income inequality. When measured
as the share of income held by the top 20 percent, income inequality fell in 2018
by the largest amount in over a decade. The Gini coefficient, an overall measure
of inequality in the population, also fell in 2018 (U.S. Census 2019).
70 | Chapter 2

250-840_text_.pdf 74

2/7/20 3:46 PM

$"0- 
сҊсѵ
) (+'*4( )/
/ 
1 -.0.
спрр–рш

' /$*)
*- ./.Ѷ


Unemployment rate (percent)

10

спршѷу

ш
ч
) (+'*4( )/
-/

ц
х


( $)

!*- ./


!*- ./

ф
у
т
спррѷр

спртѷр

спрфѷр

спрцѷр

спршѷр

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

ۙ
*)"- ..$*)'
0" /
!!$ ѵ

ۙ
  -'
+ )
-& /
*(($// ѵ 
!*- ./
$.

!-*(
0"0./
спрхѵ

!*- ./
$.
!-*(
 +/ ( -
спрхѵ

These employment and income gains have brought people from the
sidelines into employment. In the fourth quarter of 2019, 74.2 percent of workers entering employment came from out of the labor force rather than from
unemployment, which is the highest share since the series began in 1990.3
Additionally, the prime-age labor force is growing, reversing losses under the
prior administration’s expansion period until the 2016 election. Under the
prior administration’s expansion period, the prime-age labor force shrank
by roughly 1.6 million; in contrast, under the current Administration it has
expanded by 2.3 million people so far. Importantly, a strong market for jobs
creates work opportunities for those with less education or training, prior
criminal convictions, or a disability.
This movement from the sidelines into the labor market also pulls people
out of poverty and off of means-tested welfare programs, increasing their selfreliance through economic activity while decreasing their reliance on government programs that incentivize people to limit their hours or stop working to
qualify. The number of people living in poverty decreased by 1.4 million from
2017 to 2018, and the poverty rates for blacks and Hispanics reached record
lows. Furthermore, the number of working-age adults without health insurance who are below the Federal poverty line fell by 359,000 between 2016
and 2018. Because of the strong job market and sustained wage gains, food
insecurity has fallen and, as of August 2019, there are nearly 7 million fewer
3 This CEA calculation is from labor force transition data reported by the BLS.

Economic Growth Benefits Historically Disadvantaged Americans | 71

250-840_text_.pdf 75

2/7/20 3:46 PM

people participating in the Supplemental Nutrition Assistance Program (SNAP,
formerly known as the Food Stamp Program) than at the time of the 2016
election. The caseload for Temporary Assistance for Needy Families (TANF)
has fallen by almost 700,000 individuals, and the number of individuals on
Social Security Disability Insurance has fallen by almost 380,000 since the 2016
election. Similarly, Medicaid rolls are decreasing even as the U.S. population
increases. Our analysis shows that this decrease is predominantly due to a
reduction in the number of Medicaid-eligible individuals because of income
growth, not eligibility restrictions.
In addition to having encouraged these unprecedented gains for disadvantaged groups, the Trump Administration is launching several new initiatives
to increase economic opportunity by removing barriers to work. One of the
most significant barriers is that available workers do not always have the skills
and training required to fill available jobs. Additionally, available workers may
not be located near available jobs. The increase in prevalence in occupational
licensing has made it more difficult for individuals to find and take jobs in different States. Individuals’ labor market participation can also be limited by
a struggling local economy, childcare responsibilities, opioid addiction, and
prior criminal convictions. The Administration is addressing these barriers
with initiatives like the National Council for the American Worker, the Pledge to
America’s Workers, the Initiative to Stop Opioid Abuse, and the Second Chance
Hiring Initiative.
The Trump Administration continues its relentless focus on reducing
poverty by expanding self-sufficiency. The CEA (2019a) accounted for the value
of government subsidies for goods (in-kind transfers) like healthcare, food, and
housing, and we found that—contrary to claims from the policy community
and the media—poverty has decreased dramatically since the War on Poverty
began in the 1960s. However, the war was largely “won” through increasing
government dependency (demand side) rather than through promoting selfsufficiency (supply side), meaning that there is still more progress to be made.
This is where Opportunity Zones come in.
Opportunity Zones, which were created by the 2017 Tax Cuts and Jobs
Act, are best understood as supply-side economic policies. These zones entail
tax cuts, analogous to the corporate tax cut, designed to spur investment and
drive up labor demand, and thus directly help the disadvantaged achieve selfsufficiency through increased economic activity. Supply-side tax cuts are the
opposite of the traditional, failed approach to fighting poverty, which entails
higher taxes to fund demand-side subsidies for healthcare, food, and other
goods or services that incentivize people to limit their hours or stop working
to qualify.
Although the economic benefits of the Trump Administration’s policies
are widespread, this chapter’s main finding is that a stronger U.S. economy
over the past three years has especially helped racial and ethnic minorities,
72 |

250-840_text_.pdf 76

Chapter 2

2/7/20 3:46 PM

less-educated individuals, people living in poverty, and those who had been
out of the labor force. As the Administration continues to implement a progrowth agenda, the benefits to these historically disadvantaged groups are
likely to persist and intensify.
This chapter is organized in two main sections. In the first, we outline
how today’s strong labor market is benefiting lower-income individuals and
individuals in historically disadvantaged groups. In the second section, we
discuss barriers that continue keeping some individuals from benefiting from a
strong national economy, along with the actions the Administration is taking to
address these barriers and add to historically disadvantaged groups’ employment and income gains.4

Shared Prosperity from Strong Economic Growth
The Trump Administration’s tax and deregulatory policies increase labor
demand of firms. The continued economic expansion enabled by these policies
has predictably been accompanied by a very strong labor market. As additional
workers became more difficult to find, firms started considering a broader pool
of potential workers. Low unemployment and strong wage growth have drawn
workers into the labor force from the sidelines, increasing the quantity of labor
supplied.

The Current State of the Labor Market
In December 2019, the national unemployment rate was 3.5 percent—matching the lowest rate in 50 years.5 The unemployment rate has been at or below 4
percent for 22 consecutive months. This consistently low unemployment rate
is an indication of a relatively tight labor market.
Just as a low unemployment rate signals a strong labor market, a high
number of job openings—as measured by the BLS’s Job Opening and Labor
Turnover Survey (JOLTS)—indicates strong labor demand. Compared with
the time of the 2016 election, there were over 1.4 million more job openings
in October 2019. In total, there were 7.3 million job openings in October—1.4
million more than the number of unemployed persons. October was the 20th
consecutive month in which there were more job openings than unemployed.
Figure 2-3 shows the number of unemployed workers and job openings over
time. Since the JOLTS data began being collected by the BLS in 2000, the current period beginning under the Trump Administration is the first time when
there have been more job openings than unemployed people.
4 A version of this chapter was previously released as “The Impact of the Trump Labor Market on
Historically Disadvantaged Americans” (CEA 2019b).
5 Unemployment statistics are produced by the BLS and are calculated from data collected in
the monthly Current Population Survey (CPS). Unless otherwise stated, the data are seasonally
adjusted.

Economic Growth Benefits Historically Disadvantaged Americans | 73

250-840_text_.pdf 77

2/7/20 3:46 PM

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

0( -
*!
 *
+ )$)".Ѷ
сппр–рш


Thousands

спѶппп

спршѷт

рхѶппп

рсѶппп

) (+'*4 
чѶппп
*
*+ )$)".
уѶппп

п
сппр

сппу

сппц

спрп

спрт

спрх

спрш

*0- ѷ
0- 0
*!
 *-
//$./$.ѵ
*/ ѷ
#$)"
 )*/ .

-  ..$*)ѵ
# 
  
. -$ .
 ") $)
  ( -
спппѵ


As a result of a more robust U.S. economy, many groups that historically have had a tougher time getting ahead are now gaining ground. Under
the Trump Administration, many of these groups have reached notable lows
in their unemployment rates (see table 2-1). In August 2019, the unemployment rate for African Americans fell to 5.4 percent—the lowest rate on record
since the series began in 1972. Meanwhile, the unemployment rate for African
American women also reached its series low in August 2019. For Hispanics, the
September 2019 unemployment rate achieved its series low of 3.9 percent (the
series began in 1973). In 2019 the unemployment rate for American Indians
or Alaska Natives fell to 6.1 percent—the lowest rate since the series began in
2000. Figure 2-4 shows the unemployment rates for different racial and ethnic
groups compared with their prerecession lows. The decline in unemployment
after the recession and before the start of the Trump Administration was
largely the result of a recovery from the losses during the recession. During the
last two years, the black and Hispanic unemployment rates have fallen below
their prerecession lows and Asian unemployment has fallen to its prerecession
low.
Among various levels of educational attainment, those with less education typically face tougher labor market prospects. The Administration’s tax
and regulatory policies, however, are stimulating labor demand and are helping to provide labor market opportunities for those with less education and

74 |

250-840_text_.pdf 78

Chapter 2

2/7/20 3:46 PM

Table 2-1. Unemployment Rates by Demographic Group
December
2019
(percent)

Characteristic

Series low
(percent)

Low of the
Trump
Administration
(date)

The Trump low is
lowest since

Education
Less than high school

5.2

4.8 (Sept. 2019) 4.8 (Sept. 2019)

High school diploma

3.7

3.2 (Nov. 1999)

3.4 (April 2019)

Series began
(Jan. 1992)
April 2000

Some college

2.7

2.4 (Oct. 2000)

2.7 (Dec. 2019)

Nov. 2000

Bachelor's or higher

1.9

1.5 (Dec. 2000)

1.9 (Dec. 2019)

Mar. 2007

African American

5.9

5.4 (Aug. 2019)

5.4 (Aug. 2019)

Hispanic

4.2

3.9 (Sept. 2019) 3.9 (Sept. 2019)

White

3.2

3.0 (May 1969)

3.1 (April 2019)

Asian

2.5

2.1 (June 2019)

2.1 (June 2019)

Race and ethnicity
Series began
(Jan. 1972)
Series began
(Mar. 1973)
May 1969
Series began
(Jan. 2003)

Age and gender
Adult women (age 20+)

3.2

2.4 (May 1953)

3.1 (Sept. 2019)

Adult men (age 20+)

3.1

1.9 (Mar. 1969)

3.1 (Dec. 2019)

Aug. 1953
Oct. 1973

Teenagers (age 16–19)

12.6

6.4 (May 1953)

12.0 (Nov. 2019)

Dec. 1969

Sources: Bureau of Labor Statistics, Current Population Survey; CEA calculations.
Note: The series for “high school diploma,” “some college,” and “bachelor's or higher” began in 1992.
The series for "white" began in 1954. The series for “adult women,” “adult men,” and “teenagers”
began in 1948.

$"0- 
сҊуѵ
) (+'*4( )/ / 
4
 Ѷ
сппт–рш
Unemployment rate (percent)

рч
рх

 Ҋрш

'&

ру
$.+)$

рс
рп
ч
х

.$)

у

с
п
сппт

сппф

сппц

сппш

спрр

спрт

спрф

спрц

спрш

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


-  ..$*)ѵ

Economic Growth Benefits Historically Disadvantaged Americans | 75

250-840_text_.pdf 79

2/7/20 3:46 PM

$"0- 
сҊфѵ
0'/$+' 
 *#*' -. .

 - )/" 
*!
''
(+'*4 Ѷ

ршшу–спрш
Percent
цѵп

 Ҋрш

хѵф

хѵп

фѵф

фѵп

уѵф
ршшу

ршшч

сппс

сппх

спрп

спру

спрч

*0- ѷ
0- 0 *!
 *-
//$./$.ѵ

training. In September 2019, the unemployment rate for individuals without a
high school degree fell to 4.8 percent, achieving a series low (the series began
in 1992). Since the President’s election, the unemployment rate for those
without a high school degree has fallen at a faster rate than the rate for those
with a bachelor’s degree or higher. The gap between the two rates reached
a series low under the Trump Administration. For people with a high school
degree but not a college education, the unemployment rate fell to 3.4 percent
in April 2019, the lowest it has been in over 18 years. And for individuals with
some college experience but no bachelor’s degree, the rate fell to 2.7 percent
in December 2019, the lowest since 2001.
Persons with disabilities can have a harder time finding work, as can veterans. However, President Trump’s policies are translating into economic gains
for these populations as well. In September 2019, the unemployment rate for
persons with a disability dropped to 6.1 percent, the lowest it has been since
the series began in 2008.6 In April 2019, the unemployment rate for American
veterans fell to 2.3 percent, matching the series low previously achieved in
2000.7

6 The unemployment rate by disability status is not seasonally adjusted.
7 The unemployment rate for veterans is not seasonally adjusted.

76 |

250-840_text_.pdf 80

Chapter 2

2/7/20 3:46 PM

Working multiple jobs can be a negative labor market indicator if individuals must work multiple part time jobs due to the lack of available full time
work. However, having multiple jobs is not necessarily a negative economic
indicator as the opportunities to supplement one’s main source of income may
be greater during expansions. The share of people with multiple jobs has been
around 5 percent since the end of the Great Recession (figure 2-5). It reached a
high of 6.5 percent in 1996 and has been decreasing since that year. The data
does not exhibit a strong cyclical trend, as the share of people working multiple
jobs has declined during the last two recessions. It has declined by 0.2 percentage point since the election; but the average under the Trump Administration
has been 5 percent, and the annual average has been between 4.9 and 5.1
percent since 2010.

Demographic Change and Labor Force Statistics
In this subsection, we construct labor force participation rates that control
for changing demographics over time. The demographically adjusted participation rates are near prerecession levels for Hispanics and have exceeded
prerecession levels for blacks. The adjusted participation rates show that due
to the strong labor market in recent years, many workers are coming from the
sidelines and are reentering the labor force.
Various measures of the labor market such as job growth and the unemployment rate indicate a strong labor market, but the labor force participation rate has not recovered to its prerecession level. Before the recession, in
December 2007, the participation rate was 66.0 percent. The participation rate
fell during the recession and continued to fall during the recovery, reaching a
low of 62.4 percent in September 2015, before rebounding slightly to its current
level of 63.2 percent (in December 2019). In past recoveries, workers reentering
the labor force due to the stronger economy caused the participation rate to
increase. However, comparing participation rates over time can be complicated by demographic changes. To get a clearer picture of the labor market, we
construct demographically adjusted participation rates by race and ethnicity,
using 2007 as the reference period.8
Adjusting the labor force participation rate for changing demographics
is necessary because participation varies predictably over a person’s lifetime.
The overall participation rate will depend on participation at each age and on
the share of people in each age group. For example, as the overall population
ages, a larger share of people are in the older age groups, where participation
is lower due to retirement. The aging of the population therefore will likely
cause a decrease in the participation rate, even if participation at each age is
unchanged. The baby boom generation, which is currently leaving the labor
force through retirement, is a relatively large generation. Even though workers
8 The choice of reference year is arbitrary; 2007 is chosen to facilitate comparison between current
rates and precrisis rates.

Economic Growth Benefits Historically Disadvantaged Americans | 77

250-840_text_.pdf 81

2/7/20 3:46 PM

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

!-$)
( -$).Ѷ
ршцт–спрч
Labor force participation rate (percent)

хх

спрш

хф

 (*"-+#$''4

%0./ 
'*-
!*- 

+-/$$+/$*)
-/

ху
хт
хс
хр

)%0./ 
'*-
!*- 

+-/$$+/$*)
-/

хп
фш
фч
ршцп

ршцф

ршчп

ршчф

ршшп

ршшф

сппп

сппф

спрп

спрф

*0- .ѷ 0- 0
*!
 *-
//$./$.Ѹ

'0'/$*).ѵ

are coming from the sidelines and reentering the labor force due to the strong
labor market, the positive effect on the participation rate is largely offset by
retiring baby boomers, even as some boomers are working longer.
Narrower measures such as the prime-age labor force participation rate
(i.e., those age 25–54 years) offer one alternative to mitigate the effects of
demographic changes on labor market measures across time. But this is only
a partial solution, because there is still heterogeneity among groups of primeage individuals, so prime-age participation is still subject to demographic shifts
among the different age groups within the larger prime-age category. There can
also be important participation trends among both older and younger workers
that will affect the overall participation rate. Demographically adjusted participation rates are a single measure of participation that separates changes
in participation from changes in demographics by holding demographics constant (Szafran 2002). To find this adjusted rate, the age and sex distribution of
the population is first held fixed at a given reference period. The demographically adjusted participation rate for each period is constructed by using that
period’s age- and gender-specific participation rates and the population of the
reference period.9
Keeping in mind that the demographically adjusted labor force participation rate holds the age, race, and sex population distribution constant
at 2007 levels, figure 2-6 presents the demographically adjusted labor force
9 We use the following age groups: 16–19, 20–24, 25–34, 35–44, 45–54, 55–64, and 65 and over.

78 |

250-840_text_.pdf 82

Chapter 2

2/7/20 3:46 PM

$"0- 
сҊцѵ
 (*"-+#$''4 %0./ 
 *- *- 
-/$$+/$*)
/ 

!*- $.+)$.Ѷ ршшу–спрч
Labor force participation rate (percent)

цр

спрш

цп

 (*"-+#$''4

%0./ 
'*-
!*- 

+-/$$+/$*)
-/

хш
хч
хц
хх

)%0./ 
'*-
!*- 

+-/$$+/$*)
-/

хф
ху
хт
ршшу

ршшц

сппп

сппт

сппх

сппш

спрс

спрф

спрч

*0- .ѷ
0- 0
*!
 *-
//$./$.Ѹ

'0'/$*).ѵ

participation rate for blacks. The data are aggregated to the annual level
due to the relatively small sample size at the level of race by gender by age
group.10 The overall participation rate for blacks has fallen since the global
financial crisis of 2007–8, although the decline during the recession was the
continuation of a longer-term, downward trend starting in the late 1990s. The
adjusted participation rate shows that much of this decline can be explained
by demographic changes. The participation rate for blacks was higher in 2018
than it was before the Great Recession, and it is slightly below the peak in 2000
once the effects of an aging population are removed. For comparison, the
adjusted participation rate for the entire population age 16 and above fell from
66 percent in 2007 to a low of 64.5 percent in 2015, before recovering to 65.9
percent in 2019.
Adjusting for demographic change has a large impact on the labor force
participation rate for Hispanics in recent years. Figure 2-7 shows the demographically adjusted participation rate for Hispanics. From 1994 to the start of
the Great Recession, demographic changes had a minimal effect on the overall
participation rate for this group, as there tends to be little difference between
the adjusted and unadjusted rates. However, the adjusted and unadjusted participation rates have diverged since the Great Recession. The unadjusted rate
10 The BLS does not produce seasonally adjusted monthly or quarterly labor force participation
data by race for the finer-grained age groups needed to produce the demographically adjusted
participation rate.

Economic Growth Benefits Historically Disadvantaged Americans | 79

250-840_text_.pdf 83

2/7/20 3:46 PM

$"0- 
сҊчѵ
*($)'
 &'4
" 
-*2/#
(*)"
''
0'/

0''Ҋ$( 
" 
)
'-4
*-& -.Ѷ
спрп–рш
Year-over-year change (percentҘ
ч

спршѷт

ц
рп/#
+ - )/$'

х
ф

 $)

у
т
с

р
п
Ҋр
спрпѷр

спрсѷр

спруѷр

спрхѷр

спрчѷр

*0- .ѷ
0- 0
*!
 *-
//$./$.Ѹ
0-- )/
*+0'/$*)
0-1 4Ѹ

'0'/$*).ѵ
*/ ѷ
/
- 
)*)–. .*)''4
%0./ ѵ

initially fell by a relatively large amount and has only increased slightly during
the recovery. The demographically adjusted rate has fully recovered and now
exceeds its preelection level.

Wage and Income Growth
Over the past three years, the higher demand for labor and the tighter job
market have been leading to larger wage gains, especially for the lowestincome workers. In the third quarter of 2019, the 12-month change in nominal
weekly wages for the 10th percentile of full-time workers was up 7.0 percent
(see figure 2-8).11 This is higher than the year-over-year change in the nominal
weekly wage for the median worker (3.6 percent), and well above inflation.
Furthermore, in 2019:Q3, median weekly wages for full-time workers without a
high school degree were up 9.0 percent over the year.
Figure 2-9 shows that, as of November 2019, nominal average hourly
earnings of production and nonsupervisory workers grew at 3.4 percent year
over year.12 Inflation, as measured by the Personal Consumption Expenditures
(PCE) Price Index, remains modest, at 1.5 percent year over year in November.13
Therefore, the real wages of private sector production and nonsupervisory
workers increased by 1.9 percent during the year ending in November 2019.

11 Weekly earnings data are released by the BLS and are from the CPS.
12 Average hourly earnings are measured by the BLS in the Current Employment Statistics.
13 December inflation data are not yet available at the time of writing.

80 |

250-840_text_.pdf 84

Chapter 2

2/7/20 3:46 PM

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

*).0+ -1$.*-4
*-& -.
)
/# 
 -.*)'
*).0(+/$*)
3+ )$/0- .

-$ 
 ) 3Ѷ
сппц–рш
Year-over-year change (percent)

ф

 –рш

у

*($)'
1 -" 

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

т
с


-$ 

) 3

р
п
Ҋр
Ҋс
сппц

сппш

спрр

спрт

спрф

спрц

*0- .ѷ
0- 0
*!
 *-
//$./$.Ѹ
0- 0
*!
*)*($
)'4.$.ѵ
*/ ѷ
 ۙ
 -.*)'
*).0(+/$*)
3+ )$/0- .Ѹ
.#$)"  )*/ .

-  ..$*)ѵ

спрш

Minorities are experiencing some of the fastest increases in pay. In
2019:Q3, African Americans saw their weekly earnings grow by 6.0 percent
over the year, while Hispanics’ weekly earnings grew by 4.2 percent. For comparison, the 12-month change in weekly earnings for all Americans rose by 3.6
percent. In addition to faster earnings growth, lower-income households are
seeing the largest benefits from deregulatory actions that lower the costs of
goods and services. Box 2-1 shows an example of the beneficial impact of the
Administration’s deregulatory agenda on lower-income households.

Poverty and Inequality
The gains in employment and wages for those who had previously been left
behind are lifting many out of poverty. In September 2019, the Census Bureau
released its official measures of the economic well-being of Americans for 2018
using data from the Annual Social and Economic Supplement (ASEC) to the
Current Population Survey (CPS). While Americans across the board generally
saw improvements, the data show that there were larger gains among historically disadvantaged groups.
In 2018, the official poverty rate fell by 0.5 percentage point, to 11.8
percent, the lowest level since 2001, lifting 1.4 million Americans out of poverty. This decline follows a decline of 0.4 percentage point in 2017, meaning
that the U.S. poverty rate fell almost a full percentage point over the first two
years of the Trump Administration. In the CPS-ASEC, income is defined as

Economic Growth Benefits Historically Disadvantaged Americans | 81

250-840_text_.pdf 85

2/7/20 3:46 PM

Box 2-1. Who Bears the Burden of Regulatory Costs?
Well-designed regulations promote important social purposes, but at a
cost. The question of who bears the burden of regulatory costs is like the
question of who bears the burden of the taxes needed to fund government
spending programs. The Federal income tax has a progressive structure;
thus, compared with lower-income households, higher-income households
bear a greater share of the burden of taxation. Unfortunately, however,
lower-income households can bear a disproportionate share of the burden of
regulatory costs. We estimate that the cost savings from deregulatory actions
in two sectors—Internet access and prescription drugs (see figure 2-i)—especially helped lower-income households. These are two of the regulations
whose benefits were estimated by the CEA (2019c). The lower burden of
regulatory costs reinforces the gains in employment and wages from today’s
strong labor market.

$"0- 
сҊ$ѵ
*).0( -
1$)".
*)
- .-$+/$*) -0".
)
 )/ -) /

 ..
4
 *0. #*'
 )*( 
0$)/$'
Share of income (percent)
тѵп
сѵф

2.4

сѵп
1.3

рѵф

0.9

рѵп

0.6
0.3

пѵф
пѵп
*2 ./
сп
+ - )/

 *)
сп
+ - )/

#$-
сп
+ - )/

*0-/#
сп
+ - )/

$"# ./
сп
+ - )/

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

'0'/$*).ѵ

Note: Values represent the CEA’s estimates of consumers’ savings as a share of their income,
which applied the Consumer Expenditure Survey’s quintile and expenditure data to national
$)*( 
/ѵ


Costly regulations hurt lower-income households because they spend
a larger share of their budgets on goods and services produced by regulated
sectors of the economy. For example, in data from the 2018 Consumer
Expenditure Survey, the poorest fifth of households spend 2.7 percent of their
incomes out-of-pocket on prescription drugs, while the richest fifth of households spend only 0.3 percent. The poorest fifth of households also spend a

82 | Chapter 2

250-840_text_.pdf 86

2/7/20 3:46 PM

higher percentage of their incomes on Internet access. As a result, the costs
savings from deregulatory actions in these two sectors represent 2.4 percent
of the income for the poorest fifth of households, compared with 0.3 percent
for the richest fifth.
Many regulations also hurt lower-income households because they
impose standards that tend to increase the price of those goods that are disproportionately purchased by lower-income households. For example, after
controlling for other differences, Levinson (2019) finds that higher-income
households purchase more fuel-efficient cars. As a result, he estimates that
the corporate average fuel economy (CAFE) standards are regressive and
disproportionately burden lower-income households. The CAFE standards
matter less to higher-income households because they prefer to purchase
more fuel-efficient cars anyway. The 20 notable actions analyzed by the CEA
(2019c) include other deregulations of standards that restricted the ability
of lower-income households to choose the products that best suited their
preferences and budgets.

money income before taxes. It includes cash assistance but not the value of
in-kind benefits for government assistance programs or refundable tax credits
targeted at low-income working families. Including the value of these benefits
raises the total resources available to households at the bottom of the income
distribution. We conduct an analysis later in this chapter that examines the
effect of using after-tax and after-transfer income (including the value of inkind transfers) on the changes in poverty during the Administration.
Disadvantaged groups experienced the largest poverty reductions in
2018. The poverty rate fell by 0.9 percentage point for black Americans and by
0.8 percentage point for Hispanic Americans, with both groups reaching historic lows (see figure 2-10). The poverty rates for black and Hispanic Americans
in 2018 were never closer to the overall poverty rate in the United States.
Children fared especially well in 2018, with a decrease in poverty of 1.2 percentage points for those under 18. Poverty among single mothers with children fell
by 2.5 percentage points.
Although real income at the bottom of the income distribution increased
and the percentage of people in poverty fell, it can also be informative to examine how these gains compare with gains elsewhere in the income distribution,
which will be reflected in the changes in various measures of income inequality. Inequality fell in 2018, as the share of income held by the top 20 percent fell
by the largest amount in over a decade, as did the Gini index (an overall measure of inequality in the population). In fact, households between the 20th and

Economic Growth Benefits Historically Disadvantaged Americans | 83

250-840_text_.pdf 87

2/7/20 3:46 PM

$"0- 
сҊрпѵ
*1 -/4 / .
4
 
)
/#)$$/4Ѷ
ршхх–спрч
Poverty rate (percent)

уф

!-$)

( -$)

+*1 -/4
-/

уп
тф

спрч

тп
сф

$.+)$

+*1 -/4
-/

сп
рф
рп

*/'
+*1 -/4

-/

ф
п

ршхх ршцр ршцх ршчр ршчх ршшр ршшх сппр сппх спрр спрх

*0- ѷ 0-- )/
*+0'/$*)
0-1 4
җҘ
))0'
*$'
)
*)*($
0++' ( )/ѵ
*/ ѷ
# 
/
!*-
спрт
)
 4*)
- !' /
/# 
$(+' ( )//$*)
*!
/# 
-  .$") 
$)*( 

,0 ./$*).ѵ
# 
/
!*-
спрц
)
 4*)
- !' /
/# 
$(+' ( )//$*)
*!
)
0+/ 

))0'

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

40th percentiles of the distribution experienced the largest increase in average
household income among all quintiles in 2018, with a gain of 2.5 percent.14
Low unemployment, rising incomes, and declining poverty mean that
more Americans are becoming self-sufficient. The caseload for Temporary
Assistance for Needy Families (TANF) is on the decline, falling by almost
700,000 individuals since the election, as of March 2019. Meanwhile, the number of individuals on Social Security Disability Insurance (SSDI) has fallen by
almost 380,000 since the 2016 election. The decline in the official poverty rate
mirrors a decline of 0.7 percentage point in food insecurity in 2018.15 Since the
2016 election, nearly 7 million Americans have moved off the SNAP rolls. These
substantial declines in enrollment suggest that a growing economy may lead
to positive outcomes in moving families toward self-sufficiency. While some
of the enrollment decline in welfare programs could be due to administrative
or policy changes designed to prevent ineligible individuals from receiving

14 Data from the American Community Survey (ACS), which is a separate data source also released
by the Census Bureau, showed that inequality increased from 2017 to 2018. The ACS has a much
larger sample size than the CPS-ASEC, but it measures income less accurately. For this reason, the
Census recommends using the CPS-ASEC for national income statistics, like inequality.
15 U.S. Department of Agriculture, Economic Research Service, using data from the December 2018
Current Population Survey Food Security Supplement (https://www.ers.usda.gov/topics/foodnutrition-assistance/food-security-in-the-us/key-statistics-graphics.aspx).

84 |

250-840_text_.pdf 88

Chapter 2

2/7/20 3:46 PM

benefits, it is possible that some otherwise-eligible individuals would be affected.16 However, the decline in food insecurity combined with the decline in
poverty suggests that the net effect of any administrative changes and the
strong economy has been to reduce hardship, in turn reducing reliance on
public benefits.

Health Insurance and Medicaid
Strong job growth is the key to expanding and improving access to health
insurance. Employer-sponsored health insurance is by far the largest source
of health insurance coverage in the United States. The employment and earnings gains that are reducing poverty are also driving a decrease in the number
of people on Medicaid. Medicaid rolls are decreasing in both expansion and
nonexpansion States, even though the U.S. population is increasing (see figure
2-11). Our analysis of the data indicates that the reduction in the number of
people on Medicaid is due predominantly to a reduction in the number of
Medicaid-eligible individuals because of income growth as opposed to eligibility restrictions.
The Census Bureau asks about health insurance coverage during the
previous year in the CPS-ASEC. Individuals are classified as being uninsured
if they lack coverage for the entire year. For each of the insurance types, individuals are asked if they were covered by that type of insurance at any point
during the year. Comparisons of insurance coverage in recent years have been
complicated by changes in the CPS-ASEC data. In 2014, the CPS-ASEC revised
its questionnaire to better measure health insurance coverage. Starting with
the release of the 2019 data, the Census Bureau implemented improvements
in data processing to fully take advantage of the revised questionnaire. Data
for 2017 and 2018 have been released with the updated data processing, so
consistent comparisons can be made for health insurance coverage in 2016,
2017, and 2018 using CPS-ASEC data.17
Table 2-2 shows the change from 2016 to 2018 in the number of people
between age 18 and 64 with different types of health insurance coverage at
different levels of income in the CPS-ASEC. For all individuals, the number of
uninsured increased by about 2 million and the number covered by employer
provided coverage increased by about 1.4 million. Directly purchased individual coverage fell by 2.35 million people and Medicaid fell by 1.6 million people.
The distribution of income relative to the Federal poverty line for the overall
population of those age 18–64 shows that income relative to the poverty level
16 Administrative costs of program participation can prevent eligible individuals from enrolling
in public programs (Aizer 2007). Administrative changes that increase the nonmonetary cost of
enrollment could lead to an increase in the number of eligible individuals choosing not to enroll.
17 The updated files are the 2018 ASEC bridge files and the 2017 ASEC research files. Note that the
updated data processing will cause the health insurance estimates for these years to differ from the
results using the production files that were published by the Census Bureau in the works by Barnett
and Berchick (2017) and Berchick, Hood, and Barnett (2018).

Economic Growth Benefits Historically Disadvantaged Americans | 85

250-840_text_.pdf 89

2/7/20 3:46 PM

$"0- 
сҊррѵ
0( -
*!
 $$
)
 
)-*''
3+).$*)
)
*) 3+).$*)
// .Ѷ
спру–рш

. 4
*)/#
$)


Enrollment (millions)

Enrollment (millions)
фц

0"Ҋрш

3+).$*)
.// .

җ' !/
3$.Ҙ

фф

су
ст

фт

сс

фр

ср

уш

сп

уц

рш

*) 3+).$*)

.// .
җ-$"#/
3$.Ҙ

уф
ут
спру

рч
рц

спрф

спрх

спрц

спрч

спрш

*0- .ѷ
 )/ -. !*-
 $- 
о
 $$
 -1$ .Ѹ
 $. -
($'4
*0)/$*)Ѹ


'0'/$*).ѵ
Note: “Expansion” - ! -.
/*
.// .
/#/
#1 
 3+) 
 $$
*1 -" 
!*''*2$)"
/# 

!!*-' 
- 
/ѵ

increased, and the number of people living below the Federal poverty line fell
by 1.6 million. Of the 2 million increase in the number of uninsured, 1.35 million have a family income 300 percent or more of the Federal poverty line. The
number of people without insurance who are below the Federal poverty line
fell by 359,000 between 2016 and 2018. These results indicate that from 2016
to 2018, the income gains for working age adults led to reduced participation
in Medicaid.
A particularly vulnerable population is children living in poverty. Table
2-3 presents the change in the number of people under the age of 18 years with
different types of insurance by family income level. The number of uninsured
children increased by 340,000 between 2016 and 2018, even as the total number of children fell. Almost half the increase in the number of uninsured children is due to children in families that earn at least 300 percent of the Federal
poverty line. The number of children on Medicaid (includes the Children’s
Health Insurance Program, CHIP) fell by 1.45 million, which is largely due to a
decline in the number of children living in poverty. Some have argued that the
decrease in the number of children enrolled in Medicaid and the increase in
the number of uninsured is due to administrative changes that exclude eligible
children and discourage otherwise-eligible children from being enrolled.18
The small increase in the number of children below the poverty line who are
18 For example, see Goodnough and Sanger-Katz (2019).

86 |

250-840_text_.pdf 90

Chapter 2

2/7/20 3:46 PM

' 
сҊсѵ
#)" 
$)
/# 
0( -
*!
 *+' 
" 
рч–ху
 -.
'
2$/#
$!! - )/
4+ .
*!
 ).0-) 
4
($'4
 )*( 

1 'Ѷ
спрх–рч

)*( 
' 1 '
''

*+0'/$*)

җ/#*0.).Ҙ
цтх

)$).0- 

җ/#*0.).Ҙ
рѶшхр

(+'*4 -
+-*1$ 

җ/#*0.).Ҙ
рѶтхш

$- /
+0-#. 

җ/#*0.).Ҙ
–сѶтуц

 $$

җ/#*0.).Ҙ
–рѶхрт

–рѶфцх

–тфш

–счт

–рчс

–рѶпус

рпп–ршш
+ - )/
*!


рс

хпч

–рср

–ушу

–фпц

спп–сшш
+ - )/
*!


–хпч

тхс

–ухп

–ххц

сх

тпп
+ - )/
*!
 
)
*1 -


сѶпхх

рѶтфп

сѶстт

–рѶппу

–шр

 '*2


*0- .ѷ
0-- )/
*+0'/$*)
0-1 4Ѷ
))0'
*$'
)
*)*($
0++' ( )/Ѹ

'0'/$*).ѵ
*/ ѷ
 
ۙ
!  -'
+*1 -/4
'$) ѵ

' 
сҊтѵ
#)" 
$)
/# 
0( -
*!
#$'- )
2$/#
$!! - )/
4+ .
*!
 ).0-) 
4
($'4
 )*( 


1 'Ѷ
спрх–рч


*+0'/$*)

җ/#*0.).Ҙ

)$).0- 

җ/#*0.).Ҙ

(+'*4 -
+-*1$ 

җ/#*0.).Ҙ

$- /
+0-#. 

җ/#*0.).Ҙ

 $$

–уст

туп

стр

–тчш

–рѶууф

–рѶтфр

сф

–сцп

–ртр

–рѶсст

рпп–ршш
+ - )/
*!


стр

хч

цт

–ррт

–сч

спп–сшш
+ - )/
*!


спс

чф

рсп

–фт

–рфу

тпп
+ - )/
*!
 
)
*1 -


ушф

рхс

тпш

–шу

–уп

)*( 
' 1 '
''
 '*2


*0- .ѷ

0-- )/
*+0'/$*)
0-1 4Ѷ
))0'
*$'
)
*)*($
0++' ( )/Ѹ

'0'/$*).ѵ
*/ ѷ
 
ۙ
!  -'
+*1 -/4
'$) ѵ

uninsured suggests that administrative changes may be playing a small role.
However, the data indicate that income gains and the reduction in the number
of children living in poverty are primarily responsible for the large decline in the
number of children on Medicaid.
The number of people without health insurance can increase for a number of reasons. Two factors behind the increase in the number of uninsured
over the past couple of years are the elimination of the Affordable Care Act’s
(ACA) individual mandate penalty and a decline in the number of people who
qualify for Medicaid and ACA exchange subsidies. One consequence of higher
household incomes is that households will lose eligibility for public assistance
programs. Because households have a choice to remain eligible by working
less, revealed preference shows that the higher income more than offsets the
loss of Medicaid or ACA subsidies in terms of their overall level of utility. The
other reason why a lack of insurance is increasing is that some individuals
thought the elimination of the mandate penalty applied to 2018, while the
Tax Cuts and Jobs Act set the mandate penalty to $0 starting in 2019. The CBO
estimates that about 1 million people opted out of insurance coverage in 2018
due to the mistaken belief about the timing of the elimination of the mandate
penalty (CBO 2019). For individuals who were only buying insurance to avoid
the mandate penalty, the elimination of the penalty makes them better off
(CEA 2018b).

Full-Income Measures of Poverty
Income at the bottom of the distribution is rising, and poverty, based on the
Official Poverty Measure (OPM), is falling. As people move out of poverty,
Economic Growth Benefits Historically Disadvantaged Americans | 87

250-840_text_.pdf 91

2/7/20 3:46 PM

their benefits under various public assistance programs are phased out. The
potential to lose government benefits acts as a disincentive to participate at
all in the labor market for those who are out of the labor market or to increase
participation for those who are in the labor market, as the loss of benefits acts
as a tax on increasing engagement with the labor market. Because of the level
of wages and the available jobs, the labor market gains that are pulling people
out of poverty on average more than offset the loss in government benefits in
terms of total available resources.
The OPM, which is based on pretax money income, has many limitations as a measure of the total resources available to a family, which leads
it to understate resources for low-income families. The Full-Income Poverty
Measure (FPM) overcomes these limitations by considering a broader resourcesharing unit—the household instead of the family—and by including a comprehensive set of income sources.
The FPM estimates the share of people living in poverty using a posttax,
posttransfer definition of income. It subtracts Federal income and payroll taxes
and adds tax credits (e.g., the Earned Income Tax Credit and Child Tax Credit)
and cash transfers. It also includes the market value of SNAP, subsidized school
lunches, rental housing assistance, employer-provided health insurance, and
public health insurance (Medicare and Medicaid).19 It is important to note,
however, that despite using a comprehensive set of income sources, the FPM
may still understate income due to the underreporting of income sources and
especially transfers in survey data (Meyer, Mok, and Sullivan 2015). For more
details on the FPM, see Burkhauser and others (2019) and chapter 9 of the CEA’s
2019 Economic Report of the President.
The OPM and FPM differ in how they define the unit that shares resources.
Because there are economies of scale in consumption, the cost per person of
achieving a given standard of living falls as the number of people in the unit
increases. The FPM treats the household as the resource-sharing unit and
adjusts the thresholds proportionally based on the square root of the number
of people in the household. In contrast, the OPM restricts the sharing unit to
those in the same household who have family ties. By using the household as
the resource-sharing unit (which is standard in studies of income distribution),
the FPM reflects the increasing rates of cohabitation among non–family members in the United States.
Figure 2-12 shows the change in the poverty rate under the OPM from
2016 to 2018 compared with poverty measures that incorporate progressively
broader measures of income. All measures are anchored to equal the official
19 We calculate the market value of public health insurance based on the cost of its provision, and it
is adjusted for risk based on age, disability status, and State of residence (for additional details, see
Elwell, Corinth, and Burkhauser 2019). The market value of employer-provided health insurance is
included as well, and is imputed for 2018 because employer contributions are no longer reported
in the CPS-ASEC. The CBO has used a similar method for valuing health insurance since 2013 in its
reports on the distribution of income.

88 |

250-840_text_.pdf 92

Chapter 2

2/7/20 3:46 PM

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

*1 -/4
 .0- .Ѷ
спрх )
спрч
спрх

спрч


‫
۔‬%0./ 
 ,0$1' ) 
.' Ѷ
#*0. #*'
.#-$)"
0)$/Ѷ


‫*
۔‬.//3
‫*
۔‬.//-).! -

‫
۔‬

'/#
$).0-)
рпѵф

ррѵп

ррѵф
рсѵп
рсѵф
Poverty rate (percent)

ртѵп

*0- .ѷ
0-- )/
*+0'/$*)
0-1 4Ѷ
))0'
*$'
)
*)*($
0++' ( )/Ѹ
0-&#0. -
 /
'ѵ

җспршҘѸ

'0'/$*).ѵ
*/ ѷ

ۙ
!!$$'
*1 -/4
 .0- ѵ

ۙ
 -.*)'
*).0(+/$*)
3+ )$/0- .ѵ

poverty rate in 2016 of 12.7 percent. The official poverty rate fell by 0.9 percentage point from 2016 to 2018. Using the adjusted equivalence scale, making
the sharing unit the household, and using the PCE as the preferred measure of
inflation instead of the Consumer Price Index for All Urban Consumers (called
the CPI-U) caused the poverty rate to fall by 1.1 percentage points from 2016 to
2018. Moving to posttax and posttransfer income causes the reduction in poverty to be smaller. This reflects the fact that as individuals gain labor income
(which is included in the OPM poverty measure), they receive less in tax credits
and transfer income (including the value of in-kind transfers).The effective
tax rate of individuals on public assistance can be very high, which can be a
disincentive to increasing labor market participation. Given that the posttax
and posttransfer poverty rate still fell by 0.6 percentage point, we can conclude
that, overall, the increase in labor income more than offset the decrease in tax
credits and transfers. Finally, including the value of employer-provided and
publicly provided health insurance leads to an even larger decline in poverty,
of 1.4 percentage points. This occurred even as enrollment in Medicaid fell,
because the individuals losing coverage tended to be living above the poverty
threshold. The decline is partially due to the value of public health insurance
increasing over this period, which raised the full incomes of those who remain
enrolled.
The choice of income measure also affects the measurement of income
inequality. When taxes and transfers are progressive, using pretax income

Economic Growth Benefits Historically Disadvantaged Americans | 89

250-840_text_.pdf 93

2/7/20 3:46 PM

will tend to overstate the level of inequality. The United Nations’ handbook
on income statistics notes that the preferred measure of income is posttax
and posttransfer (including in-kind transfers), as that allows for an evaluation
of the effectiveness of redistributive policies as well as for meaningful comparisons between countries with different degrees of redistribution (Canberra
Group 2011). Elwell, Corinth, and Burkhauser (2019) calculate income growth
by decile from 1959 to 2016. Using a posttax and posttransfer measure of
income that includes government health insurance and the value of employersponsored health insurance, they calculate that the Gini coefficient was 0.341
in 2016, but it was 0.502 for the same year using pretax and pretransfer market
income adjusted for household size.20 Furthermore, the posttax and posttransfer income Gini coefficient was lower in 2016 than it was in 1959.

Supporting Further Economic Gains
The strong U.S. labor market has led to historic labor market successes, including higher incomes, lower poverty, and a reduced reliance on government
programs for many groups of people who had been previously left behind
during the economic recovery. In this section, we discuss some of the remaining barriers that are preventing people from fully benefiting from the strong
labor market. The skills of the available workers may not match those needed
by employers. There can also be a geographic mismatch between workers and
jobs. Childcare costs, a criminal record, or drug addiction can also prevent
certain individuals from fully participating in the labor market. Continuing the
current rate of job growth, with the unemployment rate at a historically low
level, will likely require drawing even more workers from the sidelines. This
will require targeted policies, which the Trump Administration is pursuing, to
address the barriers that have prevented these individuals from entering the
labor force despite a very strong labor market.

Making Sure That Workers Have the Skills to Succeed
In a previous report, “Addressing America’s Reskilling Challenge” (CEA 2018a),
we outlined the emerging issue of the skills gap in the ever-changing U.S.
economy. The skills gap refers to the situation whereby the skills of available
workers are not matching the skills needed by employers. Even in a booming
economy, the lack of necessary skills can prevent some individuals from enjoying the benefits of a robust labor market. Our previous report highlighted the
importance of addressing this issue, as well as the challenges facing workers
and firms that seek to do so.
The CEA also examined the existing infrastructure of Federal worker
training programs and reviewed the evidence regarding their effectiveness
20 The Gini coefficient measures inequality on a scale from 0 to 1, where values closer to 0 indicate
greater equality.

90 |

250-840_text_.pdf 94

Chapter 2

2/7/20 3:46 PM

(CEA 2019d). Overall, we found mixed evidence that these programs improve
labor market outcomes. The programs may have small positive effects overall,
but they may be more effective for particular groups of people and at certain
times in the business cycle. The large number of these programs and their
heterogeneity make it difficult to reach a single, general conclusion, but rather
suggest that some programs are effective whereas others are failing to live up
to their hoped-for potential.
To help close the skills gap, the Trump Administration has taken action to
address the limitations of these existing Federal worker training/reskilling programs. The United States needs innovative solutions for worker training given
the mixed effectiveness of the existing Federal programs. Addressing this problem is necessary in response to employers’ struggles to find skilled workers and
to enable more people on the sidelines to benefit from the booming economy.
In this context, to develop a national strategy for workforce development, the Administration has created the National Council for the American
Worker (NCAW). The NCAW is addressing issues related to improving skillstraining programs, focusing on private-sector-led approaches and promoting
multiple education and training pathways for individuals to enable them to
achieve family-sustaining careers. The NCAW is also focusing on enhancing
transparency in the outcomes of Federal and State workforce programs to
allow job seekers, policymakers, and program administrators to better understand which programs are effective. Additionally, with better data, there are
opportunities to learn from the successes and failures across public programs
and to shift resources to the types of programs that show the greatest returns.
In the previous CEA (2019d) report, we did not determine an optimal level
of government spending on employment and training programs, but we did
argue that Federal efforts should shift their spending, depending on what the
evidence says is the most effective. Among the current Federal worker training
programs, Registered Apprenticeships have shown strong improvements in
labor market outcomes, and the Administration has already increased spending on these types of “learn while you earn” models. Additionally, job search
assistance provided through the Workforce Innovation and Opportunity Act is
more effective in improving job outcomes than is access to training funded by
this act. Job search assistance aims to reduce the time an individual is unemployed and helps individuals assess their skill sets and address other barriers
that may be preventing them from entering the workforce.
Along with existing dedicated Federal programs, industry-led and nonprofit-led sectoral training programs have shown significant promise in randomized studies. Sectoral training programs are industry-specific programs
that seek to provide training for skilled, entry-level positions within a given
industry. Currently, these programs tend to be small, focusing on a particular
industry in a particular city, and are run by nonprofit groups in cooperation with
State and local governments. A randomized study of three sectoral training
Economic Growth Benefits Historically Disadvantaged Americans | 91

250-840_text_.pdf 95

2/7/20 3:46 PM

Box 2-2. The Federation of Advanced Manufacturing Education
Industry collaboration is one solution to the shortage of skilled workers in
a given area. An example of a program built on this model is the Federation
for Advanced Manufacturing Education (FAME), which is a cooperative organization of employers that seeks to build advanced manufacturing career
pathways. Businesses form partnerships with local community colleges to
provide a specialized degree program whereby students can work at the businesses while completing their associate degrees. FAME began as a successful
partnership between Toyota and Bluegrass Community and Technical College
in Lexington, Kentucky. A company sponsors a student in the Advanced
Manufacturing Technician (AMT) program. The student goes to classes two
days a week, and works at the sponsoring company three days a week. Once
the student completes the associate degree, they have the option to continue
full time at the company or to continue on to pursue a four-year engineering
degree.
The first class completed the AMT program in 2010, and FAME has
expanded rapidly to additional sites. There are currently FAME operations in
eight States, with multiple operations in the original state, Kentucky, where
FAME now coordinates directly with and receives support from the State
government.

programs found that they were effective at increasing participants’ earnings
(Maguire et al. 2010). A follow-up study of one of these programs found that the
gains persisted and may have grown over time (Roder and Elliot 2019). Other
randomized studies of sectoral training programs have also shown evidence of
effectiveness (Hendra et al. 2016; Fein and Hamadyk 2018).
The sector-based approach guides the Administration’s proposed
Industry Recognized Apprenticeship Program, which seeks to expand the
apprenticeship model into sectors that have not traditionally used it. The
private sector has taken note of the success of the sector-based approach and
has launched similar programs to address industry-level worker shortages (see
box 2-2). One option is to further scale up these existing industry-led sectoral
training programs through Federal support.
Finally, it could be beneficial to incentivize the private sector to invest
in training. Private firms generally have a disincentive to provide training in
general human capital because trained workers can be poached by other firms
before the firm has recovered the cost of training. Yet even with this risk of
employee poaching, firms will provide training in general skills when the labor
market is tight and new workers are difficult to find. Firms also provide general
training as a fringe benefit in order to improve employee retention. Financial
incentives, in the form of subsidies for private sector training, are less likely
to be effective if they end up subsidizing training that the firms would have
92 | Chapter 2

250-840_text_.pdf 96

2/7/20 3:46 PM

provided even in the absence of the subsidy. The difficulty is to design incentives to encourage more private sector training without subsidizing training
that would otherwise occur in any case.
The Administration is working to better highlight the efforts of the private
sector and to show the return on those investments to a company’s bottom
line as well as to a worker’s increased wages and career opportunities. Through
the Administration’s Pledge to America’s Workers, companies commit to provide a given number of training or reskilling opportunities for their current and
future workforces over a five-year period. To date, more than 350 companies
have pledged to provide over 14 million new opportunities for American students and workers.

Limiting Geographic Frictions in the Labor Market
Although labor market data are often presented for the Nation as a whole,
the national labor market is a collection of local labor markets. Available jobs
and available workers do not always match geographically. Economic theory
predicts that wages will rise in areas with worker shortages and fall in areas
with surpluses of workers, causing workers to move to the areas with worker
shortages. Yet moving itself can be very costly, which limits the degree to
which migration can alleviate local labor market imbalances; but government
policies and regulations can impose additional barriers and costs to moving to
a different labor market.
For over a year, monthly JOLTS data have illustrated the strong job market for people looking for work. The JOLTS data show that at a national level,
there are more job openings than unemployed workers. For the first time, the
BLS is producing experimental State JOLTS estimates that also allow for an
analysis of job openings at the State level. These new data demonstrate that
not only are there more job openings than unemployed workers nationwide,
but this is true in most States as well (see figure 2-13). Comparing the number
of unemployed people in each state from BLS data on State-level employment
and unemployment to the number of job openings shows that, as of the second
quarter of 2019, there were more job openings than people looking for work in
43 States and the District of Columbia.21 Although State-level labor markets
appear to generally be strong, some are in greater need of additional workers
than others. The very best States in which to be looking for work, where there
were fewer than 60 unemployed workers per 100 job openings, include many
States in the Midwest and the Great Plains. The States where there are as many
or more unemployed workers as job openings are Alaska, Arizona, Connecticut,
Kentucky, Louisiana, Mississippi, and New Mexico.
21 The experimental JOLTS data are monthly. However, due to the limited sample size, they are
calculated as three-month moving averages. The analysis here uses the June 2019 experimental
State JOLTS data, which correspond to the average of the months in the second quarter.

Economic Growth Benefits Historically Disadvantaged Americans | 93

250-840_text_.pdf 97

2/7/20 3:46 PM

$"0- 
сҊртѵ
0( -
*!
) (+'*4 
*-& -.
+ -
рпп

*
+ )$)"., Q2:2019 

(U.S. total = 79)

фш
*-
! 2 -

хп
/*
хш
цп
/*
цш
чп
/*
чш
шп
/*
шш
рпп
*-
#$"# -

*0- .ѷ
0- 0
*!
 *-
//$./$.Ѹ


'0'/$*).ѵ

$"0- 
сҊруѵ
#- *!
ѵѵ
 .$ )/.
#*
*1 Ѷ
ршуч–спрч
Percentage share
сс

спрч

сп

рч
рх

ру
рс

рп
ч
ршуф

ршфф

ршхф

ршцф

ршчф

ршшф

сппф

спрф

*0- ѷ
0-- )/
*+0'/$*)
0-1 4ѵ
*/ ѷ
# 
*) Ҋ4 -
" *"-+#$
(*$'$/4
,0 ./$*)
2. )*/
.& 
 /2 )
ршцс
)
ршцф
*-

 /2 )
ршцц
)
ршчпѶ
.*
/# 
1'0 
$.
$)/ -+*'/ Ѷ
.
.#*2)
4
/# 
*// 
'$) ѵ

94 |

250-840_text_.pdf 98

Chapter 2

2/7/20 3:46 PM

In addition to booming job markets in many States, geographic mobility
has reached the lowest rate in at least 70 years, declining by 0.8 percentage
point over the year, to 9.8 percent in 2018 (see figure 2-14). This decline in
mobility, which could be exacerbated by government policies that limit worker
mobility, is one reason for the persistence of geographic disparities in the labor
market. Although not discussed in this chapter, unnecessary regulations that
drive up housing costs can also limit mobility into certain metropolitan areas
with strong labor markets (see chapter 8).

Reforming Occupational Licensing
Occupational licensing requirements impose an additional cost on entering
a given occupation. There is a wide range of licensed occupations, including
plumbers, electricians, florists, and barbers (Meyer 2017). Some occupational
licensing restrictions can be justified to protect the public, but the existing
requirements for many occupations in many States include jobs that pose no
physical or financial risk to the public. Instead, licensing is being used as a barrier to entry into a profession to artificially inflate wages for those already in the
profession. A 2018 report from the Federal Trade Commission found that the
share of American workers holding an occupational license has increased fivefold, from less than 5 percent in the 1950s to 25–30 percent in 2018 (FTC 2018).
Obtaining the needed license and paying the necessary fees is a barrier
that can be particularly prohibitive for those with low incomes, negatively
affecting these workers by preventing them from entering professions where
they would earn more even if they have the skill set to do the job. A 2015 report
from the Obama Administration supports this claim, finding that the licensing
landscape in the United States generates substantial costs for workers (White
House 2015).
One such cost is how licensing adversely affects worker mobility. Workers
in licensed occupations see the largest reductions in interstate migration rates
(Johnson and Kleiner 2017). Absent State agreements to recognize outside
licenses, State-by-State occupational licensing laws prevent workers from
being able to provide their services across State lines, or move to another State
to work in a licensed profession.
Johnson and Kleiner (2017) find that the relative interstate migration rate
of workers in occupations with State-specific licensing requirements is 36 percent lower than that of workers in other occupations. There are substantial differences in relative interstate migration rates across occupations, particularly
for jobs frequently held by middle- to low-income people. Teachers have one of
the lowest relative interstate migration rates (about –39 percent). Electricians
have a reduced relative interstate migration rate of –13 percent, while barbers
and cosmetologists have such a rate of –7.5 percent. Occupational licensing can
also serve as a barrier to upward economic mobility for low- to middle-income

Economic Growth Benefits Historically Disadvantaged Americans | 95

250-840_text_.pdf 99

2/7/20 3:46 PM

workers because it is associated with hefty administrative charges, test fees,
tuition payments, and education and time requirements.
Occupational licensing also affects the employment of military spouses.
Military spouses had an unemployment rate of 18 percent in 2015, more than
four times greater than the U.S. overall employment rate at that time (Meyer
2017). This is partially because military spouses regularly move across State
lines, and those in licensed occupations are required to renew or reissue their
licenses after moving to a new State. Additionally, military spouses are more
likely to be licensed than the civilian population, and they are 10 times more
likely to move across State lines in a given year. (For more details, see chapter
3 of the 2018 Economic Report to the President.) Overall, the evidence indicates
that occupational licensing limits workers’ ability to enter professions or move
to new areas with greater opportunity.
The regulation of occupational licenses is primarily at the State level, so
there are limited options at the Federal level to reform occupational licensing, other than recognizing and supporting best practices at the State level.
The Administration is currently evaluating these options. States can enter
reciprocal agreements to recognize out-of-State licenses, work to standardize
the licensing requirements for a given occupation across States, and expedite
license applications for military spouses and others who hold an out-of-State
license (FTC 2018).

Opportunity Zones: Matching People, Communities, and
Capital
Historically, areas with less income grew faster than areas with more income,
leading to convergence in income per capita. Since the late 20th century,
however, this convergence has stopped or has possibly been reversed (Nunn,
Parsons, and Shambaugh 2016). There are many explanations for this change,
such as a slowdown in individuals with lower incomes moving to higher-income
areas for better-paying jobs or businesses moving to lower-wage regions that
have lower input costs (Ganong and Shoag 2017).
The Opportunity Zone provision of the 2017 Tax Cuts and Jobs Act seeks
to counter the solidification of geographic economic inequality by bringing
capital to low-income communities through tax cuts on capital gains. It contrasts with antipoverty policies that increase taxes to fund transfers to lowincome households, giving them income but not necessarily spurring opportunity in their communities. Under the Opportunity Zone provision, an investor
who realizes a capital gain can defer and lower taxes on the gain if he or she
invests it in an Opportunity Zone Fund. The fund, in turn, invests in businesses
or properties in census tracts that have been selected as Opportunity Zones. If
the investor keeps his or her money in the fund for at least 10 years, they receive
the additional benefit of paying no taxes on the gains earned while invested in
the fund. In doing so, the provision acts like a means-tested reduction in the
96 |

250-840_text_.pdf 100

Chapter 2

2/7/20 3:46 PM

cost of capital, where the cost reduction only occurs in selected communities
that meet the provision’s eligibility requirements.
The design of the Opportunity Zone provision improves upon that of
the Federal New Markets Tax Credit (New Markets), which has arguably been
the most significant Federal place-based incentive in recent years. Investors
must complete an extensive application to the Department of the Treasury for
approval before receiving these tax credits. In the 2018 allocation round, only
34 percent of applicants received credits (CRS 2019). This highlights another
limitation of New Markets—it has a cap. In 2018, the Treasury only awarded $3.5
billion in credits. In addition, recipients of credits must adhere to substantial
compliance and reporting requirements (CDFI Fund 2017, 2019). The complexity of participating in New Markets and the limit on total allocations have led
some to conclude that New Markets is unable to induce large-scale investment
that can revitalize entire communities (Bernstein and Hassett 2015).
The Opportunity Zone incentive, in contrast, has no application process
or limitation on scale (CRS 2019). Within broad guidelines, the incentive lets
investors act upon their insights about where to invest, in what to invest, and
how much to invest. The Opportunity Zone statute also carves out roles for
State and local governments and communities. States nominated tracts to
become Opportunity Zones, and the Department of the Treasury made the
final designation and ensured that the tracts met the income or poverty criteria
in the statute. Many areas have incorporated the incentive into their broader
development initiatives. Alabama, for example, adopted a new law to align its
development incentives with the Opportunity Zone incentive.
Today, there are 8,764 Opportunity Zones across all 50 States, the District
of Columbia, and five U.S. possessions (CDFI Fund 2018). The zones are home
to nearly 35 million Americans, and on average they have a poverty rate nearly
twice as high as the average census tract.

Opportunity Zones: Evidence of Investor Interest and Activity
Early evidence indicates considerable investor interest in Opportunity Zones.
The National Council of State Housing Agencies maintains an Opportunity Zone
Fund Directory. As of July 2019, the directory listed 163 funds seeking to a raise
a total of $43 billion (NCSHA 2019). The funds are diverse, with two-thirds having a regional focus and the rest a national focus. Most funds plan to invest in
commercial development, such as multifamily residential or in hospitality, but
more than half also plan to invest in economic or small business development.
Evidence from real estate markets also suggests that the Opportunity
Zone incentive is getting attention from investors. Data from Real Capital
Analytics, which tracks commercial real estate properties and portfolios valued at $2.5 million or more, show that year-over-year growth in development
site acquisitions in zones surged by more than 25 percent late in 2018 after
the Department of the Treasury had designated the zones, greatly exceeding
Economic Growth Benefits Historically Disadvantaged Americans | 97

250-840_text_.pdf 101

2/7/20 3:46 PM

growth in the rest of the United States. Similarly, Sage, Langen, and Van de
Minne (2019), using the same data, find that a zone 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.
Sage, Langen, and Van de Minne (2019) only find appreciation effects for
particular property types, and they conclude that the Opportunity Zone incentive is having limited economic spillovers in communities. Their data, however,
only include very particular types of properties—commercial properties valued
at less than $2.5 million. An analysis by Zillow, which uses many more properties and transactions, suggests that the zone incentive is bringing a broader
economic stimulus. The year-over-year change in the average sales price for
properties in zones reached over 20 percent in late 2018, compared with about
10 percent in tracts that met the zone eligibility criteria but that were not
selected (Casey 2019). The greater appreciation in zones suggests that buyers
expect zone tracts to become more economically-vibrant in years to come.

Expanding Opportunities for Ex-Offenders
Another barrier to employment is a prior criminal conviction, and not only
because incarceration lowers the available labor force. Having a job can
help someone just released from prison reenter society, and it reduces the
likelihood of recidivism. There is evidence that strong job growth, particularly
in manufacturing and construction, can reduce recidivism (Schnepel 2016).
Guo, Seshadri, and Taber (2019) estimate that an increase of 0.01 percent in
county-level construction employment decreases the county’s working age
population’s recidivism rate by 1 percent.
In December 2018, President Trump signed into law the historic First
Step Act, which is aimed at establishing a fairer justice system for all, reducing
recidivism, and making communities across America safer. Since this reform
was signed into law, 90 percent of the individuals who have had their sentences
reduced have been African American.
Also since then, the Trump Administration has taken steps to provide
individuals leaving prison with the opportunities and resources needed to
obtain employment. This Second Chance hiring initiative is an effort coordinated across the Federal government, States, the private sector, and the
nonprofit sector. Nonprofits serve a crucial role in assisting former prisoners
to obtain transitional housing, counseling, and education. Across the Federal
government, the Department of Justice and Bureau of Prisons have launched
the Ready to Work Initiative, which links employers to former prisoners; the
Department of Education is expanding an initiative that will help people in
prison receive Pell Grants; the Department of Labor has issued grants to support comprehensive reentry programs that promote work as well as grants to
expand fidelity bonds to employers to assist formerly incarcerated individuals
with job placement; and the Office of Personnel Management has made the
98 |

250-840_text_.pdf 102

Chapter 2

2/7/20 3:46 PM

Federal government’s job posting website accessible to people serving in and
released from Federal prisons.
Americans are reaping the benefits of the First Step Act. Data in this area
are scarce, but a number of positive anecdotes have been reported in the news.
For instance, Troy Powell, a former prisoner and guest at the White House,
had served 16 years in prison. When he was released in February 2019 under
the First Step Act, he found a job at a lumber company in less than 10 days. A
Cleveland native, Andre Badley, was released from a Federal prison in February
2019, and within three months was hired as a driver for Amazon. The number
of such success stories will continue to grow as more inmates who have served
their time and pose no danger to society are released and as more is done to
prepare them for employment and a second chance.
The Administration’s initiatives in this area, like the First Step Act and
Second Chance hiring, can help assist former prisoners seeking to reenter society as productive members of the community, meet the needs of businesses
that may be struggling to find workers, and reduce crime across American
communities.

Supporting Working Families
Since the start of the Trump Administration, supporting working families has
been a top priority. In December 2017, the President signed into law the Tax
Cuts and Jobs Act, which increased the reward for working by doubling the
Child Tax Credit and increasing its refundability. The President signed into law
the largest-ever increase in funding for the Child Care and Development Block
Grants—expanding access to high-quality childcare for nearly 800,000 families
across the country. In addition, President Trump was the first president to
include nationwide paid parental leave in his annual budget.
The President has continued to support pro-growth, pro-family policies,
including those that address obstacles that mothers of young children may
face in entering the labor force. Figure 2-15 shows the labor force participation
rate of mothers and fathers with young children. For fathers with a youngest
child age 5 or under, the participation rate fell from 98 percent in 1968 to 94
percent in 2018. A similar decline occurred among fathers of older children.
Though participation rates have fallen, the vast majority of fathers continue to
either work or look for work. This high level of participation contrasts with participation among mothers with young children. For mothers with a child under
age 6, participation increased from 30 percent to 66 percent between 1968 and
2000. This increase was driven largely by shifting cultural norms, as well as
welfare reforms that rewarded and required work for those receiving welfare
benefits and tax credits. However, participation rates stopped growing in 2000.
Today, the participation rate of mothers with a child under 6 is 67 percent—just
1 percentage point higher than their rate 19 years earlier. Moreover, the gender

Economic Growth Benefits Historically Disadvantaged Americans | 99

250-840_text_.pdf 103

2/7/20 3:46 PM

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

*0)" ./
#$'
$)
 *0. #*'
)
 3
*!
0'/Ѷ
ршхч–спрш
Labor force participation rate (percent)
рпп

' 
җп–фҘ

шп

спрш

Male 6–12

чп
 (' 
җх–рсҘ

цп
хп

 (' 
җп–фҘ

фп
уп

тп
сп
ршхф

ршцф

ршчф

ршшф

сппф

спрф

*0- .ѷ
0- 0
*!
 *-
//$./$.Ѹ

'0'/$*).ѵ
*/ ѷ
# 
" 
*!

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

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

 /2 )
/# 
" .
*!
рч
)
ху
- 
$)'0 ѵ



gap in participation rates stands at 29 percentage points for parents of children
under age 6 and at 17 percentage points for parents of children age 6 to 12.
Some parents opt out of the labor force on the basis of personal preference. For others—especially mothers with young children—the inefficiently
high cost of childcare may play a role in their decision to remain out of the
labor force. Thus, addressing this barrier to work by reducing inefficiently high
childcare costs could potentially bring more parents into the formal labor force
and increase economic efficiency.
As documented in a recent CEA report (2019e), regulations that do not
improve the health and safety of the children increase childcare costs, and
these inefficiently high costs can weaken incentives to work. For the average
State, as of 2017, the average hourly price of center-based childcare for a child
age 4 represented 24 percent of the hourly median wage. Evidence on the
responsiveness of work status and hours to childcare costs suggests that some
of these parents would enter the labor force or increase their work hours in
response to a reduction in the cost of childcare. The Administration is focused
on ensuring that more parents have safe options for their children while simultaneously giving parents more opportunities to work.
Globally, the Administration is working to expand female labor opportunities as discussed in box 2-3.

100 |

250-840_text_.pdf 104

Chapter 2

2/7/20 3:46 PM

Box 2-3. The Women’s Global Development and Prosperity
Initiative and Female Labor Force Participation Globally
A wide range of circumstances can have an effect on a woman’s decision
about whether to participate in the labor force. For example, some women
desire to partake in productive activities outside the formal labor market,
such as taking care of children or family members. At the same time, increasing female labor force participation by offering opportunities to women not
in the labor force who might otherwise elect to participate could have a
substantial effect on a country’s economy.
Among the developed countries that belong to the Organization for
Economic Cooperation and Development (OECD), in 2018, the United States
had a female labor force participation rate higher than 22 of 36 OECD countries (the most recently available data for OECD-wide comparisons are from
2018). The lowest rate within the OECD was 34.2 percent (Turkey)—a full 22.9
percentage points below the United States. Iceland had the highest female
participation rate of all OECD countries—about 21 percentage points higher
than the United States. Although the United States has a relatively high
female participation rate compared with other OECD nations, there may yet
be opportunities for additional growth, given the higher rates in some peer
countries (figure 2-ii).
A number of factors can likely explain the differences in female labor
force participation rates among developed countries in the OECD, including
policy differences, cultural factors, and demographics. For example, Blau
and Kahn (2013) estimate that almost 30 percent of the decrease in women’s
prime-age participation in the United States relative to other OECD countries
between 1990 and 2010 can be attributed to differences in family-related
policies such as those relating to childcare.
For developing countries, too, there could be a range of reasons that
women may opt against, or be prevented from, pursuing formal employment
opportunities, including but not limited to discriminatory laws and practices,
a failure to enforce relevant laws, and social and cultural practices that limit
female employment opportunities or in other instances, a desire to participate in other productive activities that are outside the formal labor market.
Nevertheless, research has found that increasing opportunities for women
to participate in the workforce has several potential positive outcomes. For
example, the World Bank has suggested that increasing opportunities for
women’s workforce participation increases political stability and reduces the
likelihood of violent conflict (Crespo-Sancho 2018).
For low-income countries, increasing female labor force participation rates also creates an opportunity for countries to increase the size of
their workforce and achieve additional economic growth. When women are
empowered economically, they reinvest back into their families and communities, producing a multiplier effect that spurs economic growth and can
potentially create societies that are more peaceful.

Economic Growth Benefits Historically Disadvantaged Americans | 101

250-840_text_.pdf 105

2/7/20 3:46 PM

$"0- 
сҊ$$ѵ
 (' 
 *-
*- 
-/$$+/$*)
/ Ѷ
4
 ' / 


*0)/-4Ѷ
спрч
Percent

77.8

чп

70.7

хп

48.5

52.5 51.8

59.1 60.5 61.3
56.2 55.9 57.1 58.1

41.1
уп

сп

п

*0- ѷ

.//$./$.ѵ
*/ ѷ
$"0- - +- . )/.
" .
рф‫۔‬ѵ

Accelerating women’s economic empowerment is critical to ensuring
that developing countries can achieve economic self-reliance, and transition from being aid partners to trade partners. To this end, the Trump
Administration established the Women’s Global Development and Prosperity
(W-GDP) initiative, which seeks to spur growth in developing countries by
promoting economic empowerment among women. W-GDP aims to economically empower 50 million women in the developing world by 2025 through
U.S. government activities, private-public partnerships, and a new, innovative
fund.
W-GDP focuses on three pillars: vocational education for women,
empowering women to succeed as entrepreneurs, and eliminating barriers
that prevent women from fully participating in the economy. W-GDP’s third
pillar addresses legal and cultural, employer practices, and social and cultural
barriers that preclude women’s economic empowerment in developing
countries. On legal barriers specifically, W-GDP focuses on five foundational
factors: economic empowerment on the basis of five principles: (1) accessing institutions, (2) building credit, (3) owning and managing property, (4)
traveling freely, and (5) working in the same jobs and sectors as males. There
is much evidence showing that amending or passing laws in these categories
results in measurable economic benefits—both on an individual level and also
on a global scale.

102 |

250-840_text_.pdf 106

Chapter 2

2/7/20 3:46 PM

One estimate shows that eliminating discriminatory laws and practices (both formal and informal) could have added $12 trillion to the global
economy, 16 percent of global gross domestic product (GDP)in 2011 (Ferrant
and Thim 2019). In terms of gender parity in the workforce, a McKinsey &
Company report estimates that if barriers to participation in the workforce
were removed and women chose to participate in the economy identically to
men, up to $28 trillion would be added to global GDP (or 26 percent) in 2025
(Woetzel et al. 2015). This includes adding $2.9 trillion to India, $2.7 trillion
to the Middle East and North Africa, $2.6 trillion to Latin America, and $721
billion to Sub-Saharan Africa.
Additionally, a World Bank (2014) report found that strengthening land
rights has a positive impact on female farmer productivity. Evidence using
data on women’s property rights spanning 100 countries over a period of 50
years shows that legal reforms was correlated with higher female labor force
participation and higher rates of women in formal (wage-earning) labor, in
addition to higher educational enrollment.
Overall, the W-GDP initiative is backed by economic research and
evidence-based policy recommendations that would help empower women
around the globe and boost global GDP.

Combating the Opioid Crisis
Another barrier to labor market success for many are the high rates of drug
addiction and overdoses. Beyond deaths from opioids, research suggest that
the abuse of prescription opioids decreases labor force participation (Krueger
2017). The CEA estimates that the full cost of the opioid crisis was $2.5 trillion
over the four-year period from 2015 to 2018 (CEA 2019f). This cost estimate
includes the value of lives lost and also higher criminal justice costs, lost labor
productivity, and higher healthcare and treatment costs. See chapter 7 for a
discussion of the trends in opioid overdose deaths and steps the Administration
has taken to address the opioid crisis.

Conclusion
The U.S. labor market is strong, even as the economy continues its record
expansion. The Trump Administration’s agenda of tax cuts and deregulation
has contributed to a strong demand for labor and an increasing labor supply.
We would expect to find the largest increases in labor demand in the industries that benefit the most from deregulatory actions, but further research is
required to confirm this. As unemployment falls to record low rates, groups
that were previously left behind in the economy’s recovery are beginning to
see substantial benefits in job opportunities and income growth. The increase
in labor market earnings is pulling millions of families out of poverty and off

Economic Growth Benefits Historically Disadvantaged Americans | 103

250-840_text_.pdf 107

2/7/20 3:46 PM

public assistance, showing how economic growth likely benefits historically
disadvantaged Americans more than expanded government programs.
However, there are still barriers that prevent lower-income workers from
realizing the full benefits of the strong labor market—such as skill mismatches,
geographic mismatches, occupational licensing, distressed communities, prior
criminal convictions, childcare affordability, and drug addiction. These barriers
prevent many from finding jobs. The Administration is seeking to reduce these
barriers to both labor demand and supply by focusing on improving worker
training, reforming occupational licensing, incentivizing private investment
in disadvantaged areas, facilitating the successful reentry of ex-offenders,
assisting working families with access to high-quality and affordable childcare,
and reducing the impact of the opioid crisis. Successful reforms in these areas
will help to grow the economy by increasing the number and productivity of
workers. The Administration’s current and future economic agenda will focus
on ensuring that all American households can benefit from strong, sustained
economic growth.

104 |

250-840_text_.pdf 108

Chapter 2

2/7/20 3:46 PM

x
Chapter 3

Regulatory Reform
Unleashes the Economy
The Trump Administration’s focus on deregulation has led to historic reductions in costly regulation. The Administration has cut more than two significant
regulations for each new significant regulation it has finalized, while maintaining critical protections for workers, public health, safety, and the environment.
This fundamental shift in how the Federal government views regulation breaks
with the decades-long accumulation of regulatory mandates that place high
costs on the U.S. economy.
The Council of Economic Advisers estimates that after 5 to 10 years, this
new approach to Federal regulation will have raised real incomes by $3,100
per household per year by increasing choice, productivity, and competition.
Twenty notable Federal deregulatory actions alone will be saving American
consumers and businesses about $1,900 per household per year after they go
into full effect. These results show that the Trump Administration’s deregulatory actions across a vast array of American industries are among the most
significant in U.S. history.
Beyond eliminating outdated or costly regulations established by prior administrations, the Trump Administration has also sharply reduced the rate at
which new Federal regulations are introduced. The ongoing introduction of
these costly regulations had previously been subtracting an additional 0.2
percent per year from real incomes, thereby giving the false impression that
the American economy was fundamentally incapable of anything better than
slow growth in real incomes and gross domestic product. Now, consumers and

105

250-840_text_.pdf 109

2/7/20 3:46 PM

small businesses no longer need to dread the steadily accumulating costs of
new Federal regulations.
Concurrently with the 2017 Presidential inauguration, real growth in gross
domestic product began outperforming experts’ forecasts where it was previously underperforming them. This should not come as a surprise, because
studies that evaluate regulation across countries show that, all else being
equal, countries that deregulated experienced more economic growth.
The new regulatory approach also significantly reduces consumer prices in
many markets—such as those for prescription drugs, health insurance, and
telecommunications—while it prevents price increases in other markets.
Furthermore, deregulation removes mandates from employers, which especially benefits smaller businesses that, unlike their large companies, do not
typically have a team of in-house lawyers and regulatory compliance staff to
help them understand and comply with onerous regulations.
By increasing choice, productivity, and competition, the Trump Administration’s
regulatory reforms have cut red tape for American businesses and have
extended them greater freedom to create jobs. Given the Administration’s
ambitious plans for this year, deregulatory benefits for consumers, job creators,
and the economy are bound to grow further in 2020.

T

he Trump Administration’s focus on deregulation has led to historic
reductions in costly regulation, while protecting workers, public health,
safety, and the environment. In January 2017, President Trump signed
Executive Order 13771, “Reducing Regulations and Controlling Regulatory
Costs,” which is the cornerstone of the Administration’s regulatory reform success. Executive Order 13771 requires Federal agencies to eliminate two regulations for every new regulation issued (2-for-1), and has created incremental
regulatory cost caps. After Executive Order 13771 was issued in fiscal year
(FY) 2017, there were 13 significant deregulatory actions and only 3 significant
regulatory actions (4-for-1). In FY 2018, there were 57 significant deregulatory
actions and only 14 significant regulatory actions (4-for-1). In FY 2019, there
were 61 significant deregulatory actions and only 35 significant regulatory

106 |

250-840_text_.pdf 110

Chapter 3

2/7/20 3:46 PM

actions (2-for-1). In total, the Trump Administration has exceeded its 2-for-1
goal, though many critics thought that even 2-for-1 would not happen.
The Council of Economic Advisers (CEA) previously looked at regulation
across countries, finding that, all else being equal, countries that deregulated
experienced more economic growth (CEA 2018a). We then related crosscountry regulatory indices to potential regulatory developments in the United
States and estimated that regulatory reform had the potential to increase U.S.
gross domestic product (GDP) by at least 1.0 to 2.2 percent over a decade.
This chapter reexamines the impact of the Administration’s regulatory
reform agenda now that it has been more completely implemented. It also
takes an alternative approach to the CEA’s earlier analysis and estimates the
aggregate economic effects of deregulation by examining specific Federal rules
and by accounting for the unique circumstances of the industries targeted by
the rules, in addition to the rules and industries similarly analyzed in previous CEA reports.1 Our analysis utilizes an economic framework that situates
each industry in a larger economy that includes market distortions from
taxes, imperfect competition, and other sources. To date, we have conducted
industry-specific analyses for 20 deregulatory actions.
The primary subject of this chapter is the impact of regulation and
deregulation on nationwide real income. In contrast, guided by the Office of
Management and Budget (OMB 2003), Federal agencies and OMB’s Office of
Information and Regulatory Affairs (OIRA) prepare and discuss related calculations of the benefits and costs of Federal regulations that do not typically
calculate effects on GDP or nationwide real incomes. GDP and real income
are of independent interest because they are important aspects of national
accounting, and they are included in the budget forecasts made by OMB, the
Social Security and Medicare Trustees, and the Congressional Budget Office, to
name a few.2 Moreover, economists and journalists routinely use GDP and real
income as familiar metrics of the performance of the economy (Brynjolfsson,
Eggers, and Gannamaneni 2018).
The CEA estimates that after 5 to 10 years, regulatory reform will have
raised real incomes by $3,100 per household per year.3 Twenty notable Federal
deregulatory actions alone will be saving American consumers and businesses
about $220 billion per year after they go into full effect. They will increase
real (after-inflation) incomes by about 1.3 percent. Many of the most notable
deregulatory efforts in American history, such as the deregulation of airlines

1 The CEA previously released research on some of the topics covered in this chapter; the text that
follows builds on these reports (CEA 2019a, 2019b, 2019c).
2 Estimates of the welfare effects of deregulation are therefore not enough by themselves to know,
among other things, how GDP forecasts should be revised to account for the economic impact of
deregulation.
3 Throughout this chapter, all dollar amounts are in 2018 dollars unless noted otherwise.

Regulatory Reform Unleashes the Economy | 107

250-840_text_.pdf 111

2/7/20 3:46 PM

and trucking that began during the Carter Administration, did not have such
large aggregate effects.
Regulatory reform not only reduces or eliminates costly regulations
established by prior Administrations, but also sharply reduces the rate at which
costly new Federal regulations are introduced. The ongoing introduction of
costly regulations had previously been subtracting an additional 0.2 percent a
year from real incomes, thereby giving the false impression that the American
economy was fundamentally incapable of anything better than slow growth.
Now, new regulations are budgeted and kept to a minimum.
In the first section of this chapter, we review the trends in Federal regulation before and after regulatory reform. We next turn to describing our general
analytical approach and how we selected 20 deregulatory actions for analysis.
The subsequent sections discuss the industry-specific deregulatory actions
with the largest aggregate effects. We estimate large reductions in regulatory
costs in the market for Internet access, healthcare markets, labor markets, and
financial markets. Next, we estimate the additional cost-savings from reversing
the trend of adding new regulations and regulatory costs each year. We also
explain why some pre-2017 regulations carried disproportionate costs, and we
offer a brief conclusion.

Reversing the Regulatory Trend
Before turning to industry-specific analyses, we provide an overview of the
recent history of Federal regulation. This history is one of rapid growth until
2017, when the growth was halted by regulatory reform. Between 2000 and
2016, Federal agencies added an average of 53 economically significant regulatory actions each year (figure 3-1). In 2017 and 2018, the average dropped to
less than 30. Figure 3-1 excludes rules that were deregulatory actions. As in
previous years, in 2017 and 2018 a subset of the economically significant rules
included in figure 3-1 are considered “transfer rules” and are not considered by
OMB/OIRA to be either regulatory or deregulatory actions. When the transfer
rules are excluded, in 2017 and 2018 the average number of economically significant regulatory actions falls to 10. The economically significant rules shown
in figure 3-1 are those the Federal agencies and OMB/OIRA expected to have
an aggregate impact on the economy of at least $100 million or to adversely
affect the economy in a material way (Executive Order 12866). Figure 3-1 also
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
Executive Order 12866. Including economically significant and other significant
rules, Federal agencies added an average of 279 significant regulatory actions
per year between 2000 and 2016; the average fell to 61 in 2017 and 2018 after
regulatory reform.

108 |

250-840_text_.pdf 112

Chapter 3

2/7/20 3:46 PM

Figure 3-1. Significant Final Rules by Presidential Year, Excluding
Deregulatory Actions, 2000–2018
Economically significant rules

Number of final rules
400

Significant rules

350
300
250
200
150
100
50
0
2000

2004

2008

2012

2016

Sources: George Washington University Regulatory Studies Center; Office of Information and
Regulatory Affairs; CEA calculations.
Note: Presidential years begin in February and end in January of the following year. Rule counts for
2017 and 2018 exclude rules considered economically significant deregulatory actions. Before
2017, we estimate one economically significant deregulatory action per year.

Last year, the CEA discussed in depth the cumulative economic impact
of regulatory actions on the U.S. economy and explained why the regulatory
whole is greater than the sum of its parts (CEA 2019b). Based on the annual
accounting of rules published in OMB’s annual Reports to Congress, we found
that regulatory costs grew by an average of $8.2 billion each year from 2000
through 2016. However, OMB’s annual Reports for 2000–2016 only included 200
rules with fully quantified cost-benefit analyses. Over this same period, there
were just over 900 economically significant rules; including other significant
rules increases the count to almost 5,000. By definition, the regulatory actions
expected to have the largest effects on the economy are included in the count
of economically significant rules. However, this focus misses the sheer bulk of
Federal regulation.
This year, we use textual analysis of the Code of Federal Regulations (CFR)
to provide a broader and longer perspective on the cumulative regulatory burden. The CFR lists all regulations issued by Federal agencies and departments
that are currently in force at the time of its publication; it is updated annually.
RegData is a database applying textual analysis to the CFR that measures the
restrictions imposed by the regulations based on the number of times words
such as “shall” and “must” appear (Al-Ubaydli and McLaughlin 2014). Figure
3-2 shows the RegData index of regulatory restrictions from 1970 through 2019.
The total number of regulatory restrictions in the CFR nearly tripled
between 1970 (the earliest available data) and 2016, increasing from 400,000

Regulatory Reform Unleashes the Economy | 109

250-840_text_.pdf 113

2/7/20 3:46 PM

Figure 3-2. Regulatory Restrictions by All Agencies, 1970–2019
Restrictions in the Code of Federal Regulations (thousands)
1,200

2019

1,000

800
600

400

200

0
1970

1975

1980

1985

1990

1995

2000

2005

2010

2015

Sources: Code of Federal Regulations; Mercatus Center RegData.

to almost 1.1 million. Aside from a few isolated year-to-year declines, the trend
was steadily upward through 2016. From 2017 through 2019 the trend flattened
and began to reverse, showing the first declines in regulatory restrictions that
have been sustained for more than a single year. The turnaround in the growth
of regulatory restrictions parallels the turnaround in the growth of regulatory
costs that the CEA documented last year (CEA 2019b). Last year we reviewed
estimates of the total regulatory costs in the United States that ranged from
almost half a trillion to over a trillion dollars. Putting those estimates together
with the total number of regulatory restrictions implies that each restriction
is on average associated with somewhere between $380,000 and $1 million of
regulatory costs.
Because deregulatory actions might involve words like “shall” and
“must,” the RegData index of restrictions shown in figure 3-2 cannot distinguish
between the impact of regulatory and deregulatory actions. To explore this, we
searched the text of two Final Rules published in the Federal Register—the 2016
regulatory action and the 2018 deregulatory action on short-term health insurance (discussed in more detail below and in CEA 2019a). The Federal Register
text of the 2018 deregulatory action was longer and included 97 restrictions,
compared to only 30 regulatory restrictions in the text of the 2016 regulatory
action. It is not known to what extent this pattern generalizes to the RegData
index of restrictions in the CFR. It seems likely that if it were possible to adjust

110 |

250-840_text_.pdf 114

Chapter 3

2/7/20 3:46 PM

for restrictions included in the deregulatory actions taken since 2017, the index
in figure 3-2 would show an even steeper decline beginning in 2017.
Figure 3-2 includes restrictions due to Federal agencies covered by
Executive Order 13771 as well as restrictions due to independent Federal
agencies that are not subject to Executive Order 13771 accounting. In recent
years restrictions due to independent agencies account for about 15 percent
of all restrictions. Since 1990, the number of restrictions due to independent
agencies has grown by about 75 percent. Even though the independent agencies were not subject to Executive Order 13771 accounting, starting in 2017 the
growth in their regulatory restrictions began to decline.
In addition to regulations, Federal agencies also issue guidance documents that advise the public about the agency’s approach to adjudication or
enforcement. Figure 3-2 does not include regulatory restrictions stemming
from guidance documents because they are not part of the CFR. Moreover,
guidance documents are non-binding, so in principle they cannot impose binding restrictions. However, a common concern is that agencies can treat guidance documents as binding in practice. Estimates suggest that some agencies
issue anywhere from twenty to two-hundred pages of guidance documents for
every page of regulations they issue (Parrillo 2019). To the extent those guidance documents impose regulatory restrictions that are binding in practice,
the restrictions should ideally be added to the count of regulatory restrictions
in figure 3-2. Although not reflected in figure 3-2, Federal agencies’ guidance
documents are subject to Executive Order 13771 accounting of the 2-for-1
requirement and regulatory cost caps. Significant guidance documents that
increase costs are defined to be regulatory actions; guidance documents that
yield cost savings are defined to be deregulatory actions.
Figure 3-3 shows how CFR regulatory restrictions on the manufacturing
industry has grown over time, until regulatory reform. RegData uses further
text analysis to determine the applicability of the regulatory restrictions to specific industries. The method uses search strings to identify phrases related to
each industry (Al-Ubaydli and McLaughlin 2014). The resulting measure shows
that regulatory restrictions on manufacturing remained roughly constant from
the late 1970s until 1986. From 1986 through 2016, the number of regulatory
restrictions almost quadrupled, from a little over 50,000 to more than 200,000.
Again, starting in 2017, the upward trend reverses; the index shows sustained
declines in regulatory restrictions on manufacturing from 2017 and 2018.
The regulatory reform results to date are notable accomplishments,
given that it is difficult and time-consuming to identify opportunities for
appropriate deregulatory actions. In a follow-up to Executive Order 13771, in
February 2017 President Trump signed Executive Order 13777, “Enforcing the
Regulatory Reform Agenda,” which requires each Federal agency to designate
a regulatory reform officer to oversee deregulatory initiatives and policies. In
an innovative response to meet this challenge, the Department of Health and
Regulatory Reform Unleashes the Economy | 111

250-840_text_.pdf 115

2/7/20 3:46 PM

Figure 3-3. Regulatory Restrictions on Manufacturing, 1970–2018
Restrictions in the Code of Federal Regulations (thousands)
250
2018
200

150

100

50

0
1970

1975

1980

1985

1990

1995

2000

2005

2010

2015

Sources: Code of Federal Regulations; Mercatus Center RegData.

Human Services began exploring the use of artificial intelligence and machinelearning algorithms to identify opportunities for regulatory reform. As an
example of the project’s potential, the department discovered that 85 percent
of its regulations created before 1990 have never been updated.
Because regulatory reform takes time, Federal agencies’ efforts that
began in 2017 are continuing to unfold. As a result, important pending and inprogress deregulatory actions cannot be included in this chapter. For example,
our analysis does not include the deregulatory actions related to emission and
fuel economy standards for automobiles; once finalized, the SAFE rule might
be the largest deregulatory effort to date. Other important deregulatory efforts
include the Department of Energy’s reforms of regulatory restrictions on residential dishwashers and lightbulbs.

Analyzing Regulatory Reform
The Trump Administration uses regulatory cost caps to reduce the cumulative
burden of Federal regulation. In addition to regulation-specific cost-benefit
tests, the cost caps induce agencies to view all their regulations as a portfolio,
which is more congruent with the experiences of the households and businesses subject to them. While pursuing their agency-specific missions, the
regulatory cost caps provide the framework for agencies to evaluate regulatory

112 |

250-840_text_.pdf 116

Chapter 3

2/7/20 3:46 PM

costs, to consider deregulatory actions, and to set priorities among new regulatory actions.
The CEA uses a pragmatic, streamlined approach to analyze the costs
that regulatory actions impose on consumers, small businesses, and other
economic actors. This approach requires making estimates of a small set
of key parameters that describe the market that is primarily affected by the
regulatory action in question. We follow a standard approach in cost-benefit
analysis and rely on revealed preferences in markets (OMB 2003). For example,
the price-elasticity of demand—which shows how consumers change their consumption in response to a price change—reflects the value consumers place
on the good or service, relative to their next-best alternatives. For this reason,
the price-elasticity of demand serves as one of the “sufficient statistics” to analyze the impact of a policy change on consumer welfare within the regulated
industry (Chetty 2009).4 Detailed applications, and a sensitivity analysis, of our
approach are given in our earlier reports (CEA 2019a, 2019b, 2019c).
To account for effects outside the regulated industry, the analysis again
takes a streamlined approach that does not require a fully detailed model of
the economy (known as a structural general equilibrium model), but instead
relies on an implementable formula that provides a good approximation of the
excess burden that a regulatory action imposes on the markets for labor and
capital (Goulder, Parry, and Williams 1999; Parry, Williams, and Goulder 1999;
Goulder and Williams 2003; Dahlby 2008; CEA 2019b). For example, anticompetitive regulation reduces the demand for labor and capital in the regulated
industry and thereby reduces the aggregate quantities of those production
factors. Marginal excess burdens in labor and capital markets are translated
into an additional increment to aggregate output by dividing them by our
48 percent estimate of the marginal tax wedge, which is broadly interpreted

4 Our analysis is not as detailed as the regulatory impact analyses that Federal agencies conduct
to comply with Executive Order 12866 (OMB 2003).This chapter is independent of the rulemaking
process. Instead, this chapter contributes to the CEA’s mission, as established by Congress in the
Employment Act of 1946, to offer objective economic advice based on economic research and
empirical evidence. Our analysis is consistent with the economic principles that guide cost-benefit
analysis, including our focus on the key concepts of willingness to pay and opportunity cost.
Another report (CEA 2019b) provides an additional discussion of our approach; and still another
report (CEA 2019a) provides a detailed discussion of the methods used to conduct prospective
cost-benefit analyses of three of the deregulatory actions considered in this chapter. Our approach
complements agencies’ completed analyses and fills in gaps, for example, when a regulatory
impact analysis was not able to quantify costs or benefits, or when a regulatory impact analysis
was not required. Note that, consistent with standard practice, shifts of resources between
industries are not counted as a cost or a benefit or a real income effect, except to the extent that
market prices indicate that the industries put different values on those resources.

Regulatory Reform Unleashes the Economy | 113

250-840_text_.pdf 117

2/7/20 3:46 PM

to include implicit taxes and imperfect competition.5 This formula captures
general equilibrium interactions that would be left out of an analysis that only
considered the impact of the regulatory action in the primary market. OMB’s
guidance on cost-benefit analysis of federal programs (Circular A-94) recommends analysis of the marginal excess tax burden. To date, however, for practical reasons the guidelines for regulatory cost accounting for the Executive
Order 13771 regulatory budget have not required agencies to include the costs
imposed on the private sector by excess tax burdens induced by regulatory
actions. The analysis in this chapter demonstrates the feasibility and importance of a more complete accounting of regulatory costs, including marginal
excess tax burdens.
The economic effects of regulation can be summarized in several ways,
such as the costs to businesses, nationwide costs, nationwide benefits, or
national incomes. The CEA employs three nationwide outcome concepts in this
chapter: costs savings, net benefits, and real income. The distinction between
the first two arises because a single regulation can create costs for one segment of the population while it creates a benefit for other segments. We refer
to the aggregate of these as the “net cost” of the regulation, which (aside from
sunk startup costs) is equal to the “net benefit” of overturning the regulation. We refer to the “cost savings” of overturning the regulation as the costs
imposed on the segment of the population that was harmed by the regulation.6
Real income is similar to GDP, except that real income subtracts depreciation
and reflects the effects of international terms of trade on the purchasing power
of U.S. residents, which is an important result of one of the larger deregulatory actions. GDP and real income, which can differ from welfare or “utility,”
subtract the opportunity costs of the Nation’s labor and capital as well as
environmental and other nonpecuniary costs. As used in this chapter, all these
concepts refer only to domestic benefits, costs, and incomes.
The primary subject of this chapter is the impact of regulation and
deregulation on nationwide real income; we estimate that over time, the
impact of regulatory reform will be worth $3,100 per household each year.
This chapter also estimates the net benefits of deregulatory actions. Some
regulatory actions trade private goods for public goods, such as environmental
quality. With public goods, and in other situations where private markets may
fail, it is necessary to carefully consider the benefits and costs of regulatory
actions. Even if the original regulatory action addressed a private market
5 An aggregate increase in a factor of production by 1 unit increases output by its marginal product
(MP), but the entire output exceeds the net benefit (i.e., marginal excess burden) because the
production factor has a marginal opportunity cost of supply. The net aggregate benefit of that 1
unit is 0.48*MP, where 0.48 is the marginal tax wedge. The additional output is therefore the net
aggregate benefit divided by 0.48.
6 The CEA’s concept of cost savings is analogous to the revenue savings from eliminating a Federal
program, whereas the net benefit would be the difference between revenue savings and the
forgone benefits of the program’s expenditures.

114 |

250-840_text_.pdf 118

Chapter 3

2/7/20 3:46 PM

Box 3-1. Looking Forward and Backward
to Study Regulatory Reform
Federal agencies conduct forward-looking, or prospective, cost-benefit analyses of proposed regulatory and deregulatory actions. In contrast, academic
policy analysts typically conduct backward-looking, or retrospective, analyses of past public policies. For example, the definitive academic studies of
the Airline Deregulation Act of 1978 were conducted in the 1980s and early
1990s (Winston 1993). The retrospective studies took advantage of data that
reflected what actually happened in the deregulated airline market.
However, analysts conducting either prospective or retrospective studies face the challenging task of predicting market outcomes in a world that
they cannot observe. Analysts in Federal agencies observe current market
outcomes that, in many cases, reasonably approximate the “no action” baseline of “what the world will be like [in the future] if the proposed rule is not
adopted” (OMB 2003, 2). But the agency analysts cannot look into the future
and observe how the proposed rule would change market outcomes. In their
prospective studies, the agency analysts use economic reasoning and empirical evidence to predict what an unobserved, counterfactual world would be
like if the proposed rule were adopted. Academics who conduct retrospective
analyses of past policies face the opposite challenge. They observe market
outcomes in the real world, where the policy was implemented, but they
cannot observe the counterfactual world without the policy. The academic
policy analysts must also rely on economic reasoning and empirical evidence
to predict outcomes in a counterfactual world.
Academic studies of airline deregulation illustrate the difficulty of doing
an accurate retrospective analysis. Although the analysts observed airline
market outcomes both before and after deregulation, they had to disentangle
the effects of deregulation from other changes that affected the airline industry. In particular, airline deregulation in 1978 happened to coincide with an
energy crisis that increased fuel prices and led to higher air fares and lower
airline profits. Analysts took a counterfactual approach to isolate the effects
of the energy crisis and to estimate the causal effects of deregulation—lower
air fares and higher profits (Winston 1993).
When done well, prospective and retrospective analyses contribute
valuable evidence about regulatory reform. Federal agencies, by necessity,
must conduct prospective analysis of proposed actions. Likewise, in this
chapter we mainly rely on prospective analysis in order to predict outcomes
of the Trump Administration’s regulatory reform agenda. Future academic
research will undoubtedly conduct retrospective analysis and provide more
evidence and new insights about the effects of the regulatory reforms that
began in 2017. Research on the deregulations of the 1970s and 1980s provides
reasons to be both optimistic and cautious about prospective analysis. When
Winston compared predictions that deregulation would lead to lower prices
to retrospective assessments, he described them as “surprisingly close,”

Regulatory Reform Unleashes the Economy | 115

250-840_text_.pdf 119

2/7/20 3:46 PM

even though they were “often made more than a decade apart by different
researchers” (Winston 1993, 1272). At the same time, he noted that the economics profession’s predictions failed to quantify the value of reducing the
inconvenience costs of airline travel restrictions and “grossly underestimated
the benefits from deregulation” (Winston 1993, 1276).

failure, a deregulatory action is still warranted when the regulatory cost savings outweigh the forgone regulatory benefits.7 GDP and real income capture
the value of private goods production, but these measures do not capture the
value of public goods or other important nonpecuniary effects. However, when
including nonpecuniary costs and benefits that are not part of real income, we
estimate that the deregulatory actions have a net benefit of more than $2,500
per household each year, compared with the previous trend of growing regulatory costs. This gain stems from the implementation of the regulatory reform
agenda and from achieving a better balance between the cost of regulations
and their societal benefits.
Because the preparation of this chapter occurred long enough after some
of the regulatory or deregulatory actions to enable us to adequately measure
relevant market outcomes, the CEA could also deviate from the regulatory
impact analyses that accompany economically significant rulemaking by relying more heavily on retrospective analysis (see box 3-1).

Deregulatory Actions Considered
We sampled deregulatory actions for industry-specific analyses. When applicable, we also examined the corresponding regulatory action taken by the
previous Administration. The actions were sampled from four broad categories.8 The first category consists of the statutes passed by Congress and signed
by President Trump. The second category consists of the 16 Federal rules or
guidance overturned under the Congressional Review Act (CRA) since January
2017.9 The third category consists of the rules in FY 2018 Regulatory Budget (i.e.,
the rules covered by Executive Order 13771 and finalized during that fiscal year,
of which there are 261), as well as the rules in the FY 2019 Regulatory Budget
7 The concept of market failure plays a central role in cost-benefit analysis, but the existence of
a market failure does not guarantee that the original regulatory action’s benefits outweighed its
costs. Market failure is a necessary but not sufficient condition for this conclusion. In practice,
it is not clear that many of the 20 deregulatory actions we consider overturned regulations that
addressed market failures.
8 In statistical terms, the categories are strata, and the overall population of interest consists of all
economically important Federal regulatory actions taken since January 2017. Also see CEA (2019b,
appendix I).
9 For each rule, Congress passed a resolution of disapproval that was signed by President Trump,
thereby overturning the rule.

116 |

250-840_text_.pdf 120

Chapter 3

2/7/20 3:46 PM

(OMB 2018).10 The fourth category consists of agency guidance documents and
rulemaking by independent agencies.
Because the purpose of this chapter is to estimate the aggregate economic effect of all new regulatory and deregulatory actions, as opposed to the
effect of an “average” deregulatory action, we designed a sampling procedure
to identify the likely largest actions in terms of economic impact. The average
effect of the sampled actions is not necessarily a good estimate of the effect
of the average unsampled action, but that is not our purpose. Rather, if the
unsampled actions have an average effect that is in the same direction (but
not necessarily magnitude) as the sampled actions, then the total effect of the
sampled actions is a conservative estimate of the total effect of all the actions.
Moreover, sampling the potentially larger effects yields a more accurate
estimate of the total effect than sampling randomly. The omitted regulatory
actions are those with few (most often, zero) comments from the public and
little attention from Congress. These are the regulations where we have more
confidence that the effects are comparatively small, so that excluding them
from the total is less likely to have a large effect on our estimate of the total.11
Our sampling procedure is not perfect. Some regulations attract attention from the public or Congress for various reasons unrelated to their regulatory costs. Our sample includes a few actions that we estimate have comparatively small aggregate effects, even though they received many comments
from the public. At the same time, there might be regulatory actions that will
have large aggregate effects but are excluded from our sample because they
did not receive many public comments.
From the first category/stratum, we selected sections of two important
new Federal laws enacted during the Trump Administration: the 2017 Tax Cuts
and Jobs Act; and the 2018 Economic Growth, Regulatory Relief, and Consumer
Protection Act. From the second category, we selected three employment rules
that affect a large number of workers as well as the top four economic regulatory actions, in terms of number of comments received from the public. From
the third category, we selected the top six regulatory actions from FY 2018, in
terms of the number of comments received from the public.
We selected four regulatory actions from the FY 2019 Regulatory Budget
that we expected to be among the comment leaders. Three of these contribute
to both our estimate of the cost savings from deregulation since 2017 and to
10 A number of the 16 rules disapproved under the CRA were part of the FY 2017 Regulatory Budget.
11 To analogize, suppose that you wanted to measure the number of automobiles in a house. It
would be unnecessarily inaccurate to take a random sample of rooms, because most of the time
the garage would not be sampled and therefore most of the time the conclusion would be zero
automobiles. Looking exclusively in the garage is the obviously superior alternative to a random
sample. That is what the CEA has done with regulations: we looked exclusively at those with a
significant chance of having a large economic effect. The formal statistical proof of this conclusion
is provided above.

Regulatory Reform Unleashes the Economy | 117

250-840_text_.pdf 121

2/7/20 3:46 PM

our estimate of the costs of the growing regulatory state before that.12 A fourth
regulatory area with heavy commenting, and potentially large costs imposed
by the previous Administration, relates to emission and fuel economy standards for automobiles. To be conservative, we do not include any cost savings
from deregulatory actions in this area.13
Finally, our sample of regulatory actions includes important guidance
at the Food and Drug Administration (FDA) regarding the approval of generic
drugs, as well as a rule from the Federal Communications Commission (FCC)
that received millions of comments from the public. All the comment leaders
for FY 2017 and FY 2018 were deregulations rather than regulations, and most
of them have had an economically significant nationwide impact.14 And though
we have not measured the economic impact of hundreds of other FY 2017 and
FY 2018 Federal rules, the aggregate cost savings for the other rules reported in
the Federal Register are in the direction of additional cost savings.15
Table 3-1 lists the regulations and our estimates, with 2 of the 18 rows
(“Savings arrangements” and “Joint Employer”) each showing the combined
effect of a pair of deregulatory actions, so the table represents a total of 20
deregulatory actions.16
Although numbers of pages of regulations are not part of our quantitative
analysis, it is interesting to note that the regulatory actions and their deregulatory companions in our sample were promulgated with more than 6,000 pages
of Federal statutes, the Federal Register, or separate agency impact analyses.

12 These are the Joint-Employer proposed rule (RIN 3142-AA13) from the National Labor Relations
Board (NLRB), and the Joint Employer proposed rule (RIN 1235-AA26) from the Department of
Labor (DOL). Because our analysis does not separate the effects of the DOL guidance and the NLRB
proposed rule on joint employers, technically we have also selected the NLRB rule, even though
it is not part of any year’s Regulatory Budget. The Fiduciary Rule (RIN 1210-AB82) is in the FY 2019
budget, but its temporary predecessor rule (82 FR 31278) also appears in the FY 2018 Regulatory
Budget, with many comments.
13 The Trump Administration has not yet finalized a rule establishing fuel economy or emissions
standards for automobiles. The CEA plans to estimate its economic effects after such a rule is
finalized.
14 The top 10 commented rules from each of the FY 2017 and FY 2018 budgets were all deregulatory
actions. Most rules in the Regulatory Budget receive no comments.
15 Some analysts have concluded that many regulatory impact analyses reported in the Federal
Register omit important resource and opportunity costs of regulation (Harrington, Morgenstern,
and Nelson 2000; Belfield, Bowden, and Rodriguez 2018), which holds on average in our sample. An
example is the 2016 rule restricting short-term, limited duration health insurance while asserting
that “this regulatory action is not likely to have economic impacts of $100 million or more in any
one year” (81 FR 75322), whereas the CEA (2019a) found the annual costs to exceed $10 billion
(100 times the upper bound cited by the rule). This suggests that estimates of the costs savings
from deregulation based on the Federal Register would be understated, although not necessarily
relative to the cost additions of regulations.
16 As is explained in more detail below, the pre-2017 regulatory actions that made table 3-1’s
deregulatory actions necessary are used to estimate the economic effects of a regulatory freeze.

118 |

250-840_text_.pdf 122

Chapter 3

2/7/20 3:46 PM

' 
тҊрѵ
 "0'/*-4
)
//0/*-4
/$*).ҁ
))0'
 (+/
*)

 '
 )*( 
 '/$1 
/*

 "0'/*-4
- 5 Ѷ
4
(+'$)"
/-/


(+'$)"
/-/



0''$!$/$*)ѷ

*)*($

 "0'/$*)
2$/#

$"#
*(( )/

*'0( 



0''$!$/$*)ѷ

-*
(+'*4( )/

 "0'/$*)


( ҝ .-$+/$*)

-*/ /$)"
/# 
-$14
*!
0./*( -.
*!

-*)
)
/# -

 ' *((0)$/$*).
 -1$ .
җ+/Ҋ )Ҙ

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

3/-/$*)
 ..0 -.


) + ) )/

" )4
)

0$) 

*0( )/.

*/' 
//0/ .


‫ڦ‬сс


‫ڦ‬т



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



‫ڦ‬с



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


‫ڦ‬р



1$)".
--)" ( )/.
./'$.# 
4

// .
!*-
*)Ҋ*1 -)( )/'

(+'*4 .
о
0'$!$ 
// 
*'$/$'

0$1$.$*).
!*-
*)Ҋ*1 -)( )/'

(+'*4 .

  -'
,0$.$/$*)
 "0'/$*)Ѹ
$-
4

)
! 
*-&+' .

 
0$) ҝ0' 
)
 
0' 

- "-$)"
/# 
/)-
!*-
 / -($)$)"

*$)/Ҋ(+'*4 -
//0.


 !$)$/$*)
*!
҂(+'*4 -҃
) -
 /$*)

тҗфҘ
*!
 Ҋ..*$/$*)
 '/#
').

җ 
0' Ҙ



спрч
*-

спрш

 "0'/*-4
0" /ѷ

*)*($

 "0'/$*)
2$/#

$"#
*(( )/

*'0( 



 (+/
*)
 '

)*( 



җ$)
‫
ڦ‬$''$*).
+ -

 -Ҙ


 .$..$*)
*!
0' 
 )/ -+- /$)"
҂1$ ҃

3 (+/$*)
$)
 /$*)
сптҗҘ
*!
/# 

ҡ
җ -.0 -
0' Ҙ

#*-/Ҋ -(Ѷ
 $($/ Ҋ0-/$*)

).0-) ҡ
җ  Ҙ

44Ѷ
 #$' 
$/' Ѷ
)
 -/$)
 $"#Ҋ
*./
 )./''( )/
 *).

рчҊ*)/#
3/ ).$*)
*!
-).$/$*)

 -$*
)
 '4
*!
++'$$'$/4
/ .ҡ

җ$0$-4
0' Ҙ
*+ 
*!
 /$*).
спсҗҘ
)
җҘ
*!

& -.
)
/*&4-.
/

./ 
- 1 )/$*)Ѷ
-*0/$*)
0% /

/*
*4'/$ .Ѷ
)
 .*0- 

*). -1/$*)Ѹ
 .$..$*)
*-
 1$.$*)ҡ


‫ڦ‬рт



‫ڦ‬п


‫ڦ‬рцa

‫ڦ‬рцb



‫ڦ‬рф


‫ڦ‬рт


‫ڦ‬ц


‫ڦ‬ф


‫ڦ‬п


‫ڦ‬п



 + '
*!
-*/ /$)"
)
-*(*/$)"
/# 

+ )
 )/ -) /
)
 ..0) 
*!

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


‫ڦ‬фу




)


‫ڦ‬тс




* -)$5/$*)
!!*-/.


# 
3
0/.
)
 *.
/Ҋ
 0 
/# 

)$1$0'
)/ 
 )'/4
/*
 -*

*)*($
-*2/#Ѷ
 "0'/*-4
 '$ !Ѷ

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


‫ڦ‬сч


‫ڦ‬х



5 


‫ڦ‬стф



*/'
i(+/
r '/$1 
/*
сппрҌрх
r "0'/*-4
t- )



‫ڦ‬тцц



0(
ۙ
t*/'
i(+/
r '/$1 
/*

r "0'/*-4
f-

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

0' (&$)"
-*"-(
)" ( )/
!!$ Ѹ
 $--4
*!
*)"- ..Ѹ

'0'/$*).ѵ

*/ ѷ

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

)
./ -$.&
җҡҘ
.$")$!$ .
/# 
0. 
*!

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

- '
$)*( 
- 
-*0) 
/*
/# 
) - ./
$''$*)ѵ
# 
 ./$(/ 
!*-
%*$)/
 (+'*4 -
-0' .
$)'0 .

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

a. The estimate for joint employer rules includes anticompetitive effects of other DOL and
NLRB regulations.
bѵ
# 
'0'/$*)
!*-
 .
..0( .
/#/
/# 
 3+).$*).
*!
/# 
 !$)$/$*)
*!
)
 (+'*4 -
!*-

 .
2$''
 
!*0)
'2!0'
2# )
%0$/$*)
$.
*(+' / ѵ

Regulatory Reform Unleashes the Economy | 119

250-840_text_.pdf 123

2/7/20 3:46 PM

Consumer Savings on Internet Access
Deregulation frequently reduces consumer prices by enhancing competition
and productivity. To show how this happens, we begin our analysis of specific
Federal rules with two examples from the broadband or Internet service
provider (ISP) industry, which includes wireless smartphone service as well as
home Internet service over cables, telephone lines, fiber-optics, and satellites.
Before 2016, ISPs were permitted to, and often did, use and share customer personal data, such as Internet browsing history, unless the consumer
“opted out” of data sharing. With so many consumers staying with the default
sharing option, ISPs could earn revenue both from subscriber fees, which are
tracked by the industry’s consumer price index (CPI), and from using or sharing
customer data. Equivalently, the receipt of customer data allowed ISPs to earn
the same profits with a lower subscriber fee. In effect, consumers paid for their
subscription part with money and part by providing personal data.
In 2016 the FCC proposed and finalized a broadband privacy rule requiring ISPs to have consumers to pay by default with only money, thus prohibiting
the opt-out system and instead requiring the opt-in system. This rule, which
was likely anticipated well before 2016 as the FCC was moving ISPs under
the stricter “Title II” regulation (see below), was to go into effect on January
3, 2017. However, in 2017, Congress passed and President Trump signed a
resolution of disapproval under the Congressional Review Act to overturn the
2016 FCC rule and prevent future Administrations from adopting similar rules.
This 2017 deregulatory action assured market participants that the ISP market
would proceed with low subscriber fees. By overturning the 2016 rule, the 2017
action restored the FCC’s pre-2016 regulatory approach to protecting customer
privacy. Consumers with privacy concerns may opt out and request that their
ISP not share their data.17
Overturning the FCC’s opt-in rule resulted in lower prices for wired and
wireless Internet service, as shown by the CPIs graphed in figure 3-4. Wireless
service prices fell at the same time that Congress was considering the resolution of disapproval, and wired Internet prices fell a couple of months later.
Both these declines are about $40 per subscriber over the life of the subscription, which is similar to independent estimates of the per-subscriber cost of

17 In 2013, AT&T introduced its Internet Preferences Program, which gave consumers the choice to
opt out of data sharing. If consumers opted in and allowed data sharing, they received the lowest
available subscription rate, which was at least $29 per month lower. Media reports suggest that the
vast majority of consumers opted in; i.e., they were willing to allow data sharing in order to qualify
for the lower subscription rate.

120 |

250-840_text_.pdf 124

Chapter 3

2/7/20 3:46 PM

Figure 3-4. Wireless and Wired Internet Service Provider Price Cuts
Close to Congressional Review Act Nullification of Federal
Communications Commission Rule, 2016–17
Congress using CRA
to nullify FCC rule

Wireless price index (2016 = 100)

110

Wired price index
(2016 = 100)

Dec-17
Wireless (left
axis)

105

Wired broadband
(right axis)

102
101

100

100

95

99

90

98

85

97
Jan. 2016

July 2016

Jan. 2017

July 2017

Sources: Bureau of Labor Statistics; CEA calculations.
Note: CRA = Congressional Review Act; FCC = Federal Communications Commission.

obtaining personal data consent from retail customers that are the basis for
our quantitative analysis.18
At the aggregate level, we estimate the effect of overturning the opt-in
rule to be a net savings (including a subtraction for the cost to consumers of
providing personal data and an addition for producer surplus) of about $11
billion per year.19 Overturning the rule also encourages the aggregate supplies
of capital and labor (CEA 2019b), as well as competition in online advertising and other markets where consumer data are valuable. We estimate that
these effects would create additional net benefits of $5 billion per year and
18 Staten and Cate (2003) report results from a credit card issuer that tried an opt-in program
for personal customer information, and found that it cost an average of about $37 (converted
to 2018 prices) per customer in terms of mailings and phone calls to obtain opt-in from their
customers. Amortized over a 24-month wireless contract and over a wired Internet contract lasting
60 months—i.e., about 4.0 percent and 1.0 percent of the retail price, respectively. Assuming
that costs are passed through retail price according to the 60 percent markup rate measured by
Goolsbee (2006) for the broadband industry, we predict retail price effects of 6.5 percent and 1.6
percent, respectively. The actual price drops shown in figure 3-4 are 7.0 percent and 1.6 percent,
respectively.
19 We estimate that broadband industry revenue (wired and wireless combined) would be $202
billion per year under the FCC rule. We estimate that the consumers providing personal data as a
result of the overturning of the FCC rule do so at an aggregate annual cost of $1.5 billion, offsetting
an aggregate annual savings in subscription fees of $11 billion as well as an addition to producer
surplus.

Regulatory Reform Unleashes the Economy | 121

250-840_text_.pdf 125

2/7/20 3:46 PM

corresponding additional real income of about $11 billion per year, which is
small compared with total activity in those other markets but significant compared with the regulated market.20 After 5 to 10 years when these effects are
fully realized, the total impact on real incomes is estimated to be $22 billion
(see table 3-1).
Before the Trump Administration, another FCC rule adopted in 2015
restricted the vertical pricing arrangements of ISPs—that is, monetary transactions between ISPs and the providers of Internet content such as Netflix and
Yahoo.21 The 2015 rule also imposed government oversight on communication
services, making it difficult for these companies to quickly respond to competition and provide new goods and services on the market. These vertical pricing
and other restrictions are being removed by the FCC through its “Restoring
Internet Freedom” order, returning to regulating ISPs under Title I of the
Communications Act.
Previous research shows that vertical pricing restrictions in broadband
significantly reduce the quantity and quality of services received by broadband
consumers.22 Hazlett and Caliskan (2008), for example, looked at “open access”
restrictions that were applied to U.S. Digital Subscriber Line service (DSL) but
not Cable Modem (CM) access. They found that three years after restrictions on
DSL services were relaxed, in 2003 and 2005, U.S. DSL subscriptions grew by
about 31 percent relative to the trend, while U.S. CM subscriptions increased
slightly relative to the trend. Average revenue per DSL subscriber fell, while
average revenue per CM subscriber was constant (although quality increased).
At the same time, DSL and CM subscriptions in Canada, which was not experiencing the regulatory changes, did not increase relative to the trend. Applying
these findings to ISPs in the years 2017–27, we find that, by removing vertical
pricing regulations, the Trump Administration’s “Restoring Internet Freedom”

20 See also Goulder and Williams (2003) and Dahlby (2008). Throughout this chapter, as in our other
reports (CEA 2019a, 2019b), we use a 0.5 marginal cost of public funds to approximate the extraindustry net costs of an industry’s regulation, except when we estimate those costs to be primarily
outside the United States (see especially figure 3-4 and the associated discussion).
21 Both the vertical pricing restrictions and the opt-in requirement are linked to the alternative
regulatory frameworks that the FCC has variously proposed for ISPs—Title I versus Title II of
the Communications Act. However, vertical pricing restrictions and the opt-in requirement are
economically distinct and were also implemented by separate rulemaking (see, respectively, 81 FR
8067 and 81 FR 87274).
22 See also Becker, Carlton, and Sider (2010, 499), who conclude that regulating vertical pricing
in broadband “interfere[s] with the development of business models and network management
practices that may be efficient responses to the large, ongoing, and unpredictable changes
in Internet demand and technology, . . .[which] is likely to harm investment, innovation, and
consumer welfare.” Flexible contracting between customer and supplier is generally expected to
increase productivity because of the complementary relationship between the two, in contrast
to contracts between two suppliers of the same good that have the potential to increase market
power.

122 |

250-840_text_.pdf 126

Chapter 3

2/7/20 3:46 PM

order will increase real incomes by more than $50 billion per year and consumer welfare by almost $40 billion per year.

Consumer and Small Business
Savings on Healthcare
Deregulation is also reducing prices for healthcare. Figure 3-5 shows an
inflation-adjusted index of retail prescription drug prices compared with its
previous trend growth. Prescription drug prices outpaced general inflation
for decades; but in the past two years, they have fallen more than 11 percent
below the previous trend as of May 2019, and below general inflation. In 2018,
prescription drug prices even declined in nominal terms over the calendar
year for the first time since 1972. Much of this is the result of the Trump
Administration’s efforts at the FDA, such as its 2017 Drug Competition Action
Plan and 2018 Strategic Policy Roadmap, to enhance choice and price competition in the biopharmaceutical markets. Under these policies, the FDA has
approved a record number of generic and new brand name drugs to compete
against existing drugs (CEA 2018b).23 We estimate that the results of these
actions will save consumers almost 10 percent on retail prescription drugs,
which results in an increase of $32 billion per year in the purchasing power of
the incomes of Americans (including both consumers and producers).24
The Trump Administration has also taken deregulatory actions in other
healthcare markets, such as insurance. Previous CEA reports provided analyses of four healthcare deregulatory actions: the process improvements at the
FDA reflected in figure 3-5, and three actions deregulating health insurance
for individuals and small groups (CEA 2019a, 2019b).25 These four actions,
which remove restrictions and alleviate some of the costs of Federal policies
introduced during the years 2010–16, are by themselves expected to increase
average real incomes by about 0.5 percent, or an average of about $700 per

23 Another indicator of the quantitative importance of new FDA procedures is the July 2017 crash
of the stock price of at least one foreign generic drug maker, which analysts attributed to “greater
competition as a result of an increase in generic drug approvals by the U.S. FDA.” See Sheetz (2017).
24 The 10 percent assumes that 1 standard deviation below the pre-2017 trend is due to factors
other than deregulation. Retail prescription drug expenditures of $326 billion per year were
measured by Roehrig (2018). Note that prices may have fallen even more than shown in figure 3-5,
because in 2016 the Bureau of Labor Statistics changed its formula from geometric to Laspeyres,
which increases the measured rate of inflation (CEA 2018b).
25 The three health insurance actions are (1) reducing, through the Tax Cuts and Jobs Act of 2017,
the individual mandate penalty to zero owed by consumers who did not have federally approved
coverage or an exemption; (2) permitting, via a June 2018 rule, more small businesses to form
Association Health Plans (AHPs) to provide lower-cost group health insurance to their employees;
and (3) expanding, through an August 2018 rule, short-term, limited-duration insurance plans.

Regulatory Reform Unleashes the Economy | 123

250-840_text_.pdf 127

2/7/20 3:46 PM

Figure 3-5. Inflation-Adjusted CPI for Prescription Drugs, 2009–19
Prescription drug CPI / all items CPI (ratio)
2.5

2.3

Sep-19

Preinauguration
trend

12.7%
below trend

2.1
Postinauguration

Preinauguration

1.9

1.7
July 2009

July 2011

July 2013

July 2015

July 2017

July 2019

Sources: Bureau of Labor Statistics; CEA calculations.
Note: The Consumer Price Index (CPI) covers retail transactions, which are about three-fourths of all
prescription drug sales. Inflation adjustments are calculated using the ratio of the CPI of
prescription drugs relative to the CPI-U for all items. The preinauguration expansion trend in annual
growth rates is estimated over a sample period July 2009–December 2016, with 2017–19 projected
levels then reconstructed from projected growth rates.

household each year.26 Among those who benefit from these deregulatory
actions are an estimated 1 million consumers who will save on their individual
health insurance policy premiums by switching to less-regulated short-term
plans, with savings that may exceed 50 percent.27 Also included are small businesses, which may see substantial premium savings from obtaining access to
cheaper large-group health insurance coverage.

Employment Regulations
Unlike large companies, small businesses do not typically have a team of
in-house lawyers and regulatory compliance staff, making understanding
26 This average includes zeros for households not affected by the four deregulatory actions. For
the purposes of calculating real income effects, we do not count parts of the net benefit that are
consumer hassle costs because those costs are traditionally excluded from GDP, even though they
are genuine costs from a consumer’s point of view. Similarly, we treat the revealed preference
value of public health insurance as part of “net benefits” but not GDP or real income, which
traditionally are assigned those values according to cost rather than revealed preference value. As
a result, the GDP effect of the health insurance deregulations is less than the net benefit, while the
opposite tends to occur for other deregulations.
27 Part of the premium savings comes from the fact that the short-term plans restricted by
the Obama Administration have different characteristics than the individual plans regulated
by the Affordable Care Act. The CEA’s (2019a) analysis shows how the Trump Administration’s
deregulatory actions reduced health insurance prices significantly, even after adjusting for
differences in plan characteristics. See also our report (CEA 2019a) for sources on short-term plan
premiums.

124 |

250-840_text_.pdf 128

Chapter 3

2/7/20 3:46 PM

and complying with regulations particularly onerous. Of the small businesses
surveyed monthly by the National Federation of Independent Business (NFIB)
between 2012 and the election of President Trump, a plurality of surveyed
businesses selected “government requirements and red tape”—that is, regulations—as their single most important problem 45 percent of the time they were
asked. Though a plurality of small businesses have never selected regulations
as their single most important problem since President Trump’s election,
regulations remain an important issue.
During President Trump’s Administration, the Department of Labor (DOL)
and National Labor Relations Board (NLRB) have been working to eliminate
a number of regulations that disproportionately burden small businesses,
reduce worker productivity and real wages, and distort competition in the
labor market. The NLRB, under the Obama Administration, expanded the
definitions of both joint employer and independent contractor, which, among
other things, would have categorized some franchisers as joint employers of
their franchisee employees. DOL had also changed its guidance under certain
statutes regarding joint employers and independent contractors.
Without the Trump Administration’s proposed deregulatory actions,
thousands of small businesses, including franchisees and subcontractors,
would no longer be able to compete against larger corporations, and millions
of workers’ wages would have fallen due to the effect of these labor regulations. The CEA (2019b) estimates that, together, the Obama Administration’s
DOL guidance and the NLRB standard related to joint employers would have
created more than $5 billion in annual net costs and reduced real incomes by
about $11 billion.
Federal rulemaking also plays a role in maintaining a level playing field
for small businesses that are subject to State regulations. In 2015, DOL determined that Federal rulemaking was likely required in order to permit States to
mandate private employers to administer payroll deductions, with proceeds
to be invested in State-managed individual retirement accounts (IRAs), and
automatically enroll their employees in those accounts. In the revealed preference framework, the fact that a number of small businesses did not voluntarily
offer these plans strongly suggests that the costs of administering these plans
exceeded the value they create for employees.28 Nevertheless, a number of
States are requiring all employers to automatically enroll employees, and
legislation is pending before other State legislatures to require the same.29
If employers are forced to comply, the administrative costs, or the penalty
for noncompliance, reduce what can be paid out in employee compensa28 Between 39 million and 72 million people work for an employer that does not offer a retirement
plan (AARP 2014; Panis and Brien 2015; and the final rule). Following the standard approach
in labor economics (Lazear 1979, 1981; Mortensen 2010), we assume that the composition of
employee compensation maximizes the joint surplus of employer and employee.
29 See State of Oregon (2015).

Regulatory Reform Unleashes the Economy | 125

250-840_text_.pdf 129

2/7/20 3:46 PM

tion, which is why Congress and President Trump overturned previous DOL
rulemaking designed to facilitate the State-level employer mandates.
The CEA uses the same economic framework for analyzing the IRA mandate that it used for health insurance mandates (CEA 2019a).30 We assume
that Federal rulemaking is relevant and will be affecting 10 million workers
with an average annual IRA contribution of $1,571 per year.31 We estimate that
each $1,571 deposited in an IRA is, in present value terms, a transfer from the
Federal Treasury to the worker of $526. Because employers need to be forced
to provide the accounts, we infer that there is some combination of marginal
employer and employee costs of providing a retirement plan that equals or
exceeds $526 per worker each year. Conversely, this cost is bounded above
by $526, plus the annual per-worker fine for noncompliance, which we take to
be $250 per employee each year.32 Following Harberger (1964), this makes the
aggregate of the employer and employee costs $6.5 billion per year.33 Adding
in the deadweight cost of taxes, that is a net cost of $10 billion per year, most of
which is borne outside the State implementing the program. As a real income
loss (i.e., ignoring factor-supply costs in the net cost calculation), it is $13 billion per year.
In 2011, DOL proposed costly “Persuader Rule” amendments to the
Labor-Management Reporting and Disclosure Act that would potentially have
generated reporting requirements for consultants (including attorneys) when
the employer posed labor law questions, even if the attorney or consultant
did not communicate directly with employees.34 These amendments were
30 One difference is that the IRA mandates allow individuals to opt out without penalty. Our
analysis assumes that some, but not all, workers affected by the rule will opt out. Research has
found that automatic enrollment in retirement plans generates substantial inertia, so workers
remain in plans that they would not have voluntarily chosen (Madrian and Shea 2001; Bernheim,
Fradkin, and Popov 2015).
31 “Since 2012, 40 States have studied proposals for State-facilitated savings programs or
considered or adopted legislation to create them. At least 10 States enacted legislation to
expand access to retirement savings for nongovernmental workers. California, Connecticut,
Illinois, Maryland, and Oregon have all adopted auto IRA models” (NCSL 2018). As to the average
contribution, the CEA notes that the Illinois pilot had 196 employees investing an average of
$392.86 per employee per quarter (about $1,571 a year) (Hayden 2018).
32 The Illinois fine is $250 per employee a year (Hopkins 2015). California has a $250 penalty 90
days after receiving a noncompliance notice and a $500 penalty after 180 days (UC Berkeley Labor
Center 2017). It is unclear whether and how often the State will send notices. It does not appear
that Oregon has yet established its penalty.
33 It is often the case in cost-benefit analysis that a reduction in subsidy payments is merely a
transfer that leaves social benefits unchanged; the benefits to taxpayers are exactly offset by the
costs to the recipients who lose the subsidy. The tax subsidy to IRA deposits is properly treated as a
transfer when the task is evaluating the effects of the subsidy—i.e., when comparing current policy
with a hypothetical policy that has no tax subsidy for IRAs. But the purpose of this chapter is to
evaluate the effect of relaxing restrictions on choices by employers and employees, not changing
the tax subsidy rules for IRAs. See also CEA (2019a).
34 Cummings (2016) and 81 FR 15924.

126 |

250-840_text_.pdf 130

Chapter 3

2/7/20 3:46 PM

finalized and set to take effect in 2016, but were delayed due to ongoing litigation.35 The Persuader Rule amendments were rescinded by DOL in 2018.36
Under the Persuader Rule, consultants (including attorneys) would
have needed to file with DOL a Form LM-20, which becomes publicly available, reporting the amount of their fee and the type of advice provided.37 As
another example, persons attending an invited talk at their local Chamber of
Commerce related to employment law would have had their names “likely
disclosed to DOL and made [publicly] available.” In order to comply with the
Persuader Rule, a practitioner of labor law might have had to “identify and
segregate every increment of time billed to each of [their] clients for ‘labor relations advice or services’ even if the firm was not doing any ‘persuader’ consulting under the New Rule for that client currently.” The American Bar Association
understood the Persuader Rule to require labor lawyers to violate their ethical
duties to their clients (Brown 2016, 8–10), while some labor law firms refused
to take on any work that would fall under the Persuader Rule’s new reporting
requirements.38
Due to the large number of employers subject to the rule, the midpoint
of Furchtgott-Roth’s (2016) estimates shows the rule to have ongoing compliance costs of $5.4 billion per year combined for employers, attorneys, and
consultants. Initial costs of the rule were estimated as $3.6 billion. The CEA
determined that 1 of the 18 components of the estimates may be overstated,
and therefore we adjusted the ongoing costs downward to $4.9 billion per year
in 2018 prices. The compliance costs come out of productivity and thereby
have additional net annual costs of $2.4 billion, as they reduce aggregate supplies of capital and labor.
These and other rules introduced by DOL and the NLRB during the Obama
Administration had anticompetitive effects on the labor market.39 We do not
attempt to parse the combined effects among the various rules and guidance,
but instead allocate it entirely to the rules regarding joint employers, and we
then avoid double-counting by omitting any competition costs of other NLRB
and/or DOL regulations. The combination of regulations cited in this section
would have reduced real incomes by about $45 billion per year, or an average
of almost $400 per household each year.

35 See NFIB v. Perez (2016). Also see Eilperin (2017).
36 See DOL (2017).
37 This paragraph quotes or paraphrases Cummings (2016).
38 See page 79 of the June 20, 2016, testimony in NFIB v. Perez (Federal case number 5:16-cv-66).
39 See the CEA’s (2019b) analysis (as well as 81 FR 15929) of how a broader definition of “joint
employer” would reduce competition among employers in some industries.

Regulatory Reform Unleashes the Economy | 127

250-840_text_.pdf 131

2/7/20 3:46 PM

Financial Regulations
In the wake of the 2007–9 global financial crisis, banking reforms attempted
to address the systemic risk created by large financial institutions. Congress
and regulators raised banks’ capital standards, imposed new stress tests, and
bestowed new regulatory powers on bank regulators. Though these reforms
were intended to reduce the risks created by large financial institutions, the
Dodd-Frank Act’s regulations imposed costly new regulatory requirements on
small and mid-sized banks that did not pose a systemic risk.
Ultimately, Dodd-Frank’s overly broad regulations hurt lending to small
businesses by unnecessarily burdening community and regional banks, which
play an outsized role in supporting small businesses and local economies
across the Nation. Per the Federal Deposit Insurance Corporation’s definition,
community banks make up 92 percent of federally insured banks and thrifts,
and they are responsible for 16 percent of total loans and leases. Community
banks also hold 42 percent of small loans to farms and businesses. Also, in
2014 there were 646 United States counties in which the only banking offices
belonged to community banks, and another 598 counties where community
banks held at least 75 percent of deposits. Together, these counties made up
almost 40 percent of all U.S. counties.
The 2018 Economic Growth, Regulatory Relief, and Consumer Protection
Act, also known as the “Crapo Bill,” signed by President Trump, removes the
restrictions from smaller banks that were misapplied to them as part of earlier
efforts to alleviate the “too big to fail” banking problem. The CEA (2019b) posits
that this act “recognizes the vital importance of small and midsized banks, as
well as the high costs and negligible benefits of subjecting them to regulatory
requirements better suited for the largest financial institutions. [It] is expected
to reduce regulatory burdens and help to expand the credit made available to
small businesses that are the lifeblood of local communities across the nation.”
Heightened consolidation among small banks (those with assets less
than $1 billion) followed the enactment of Dodd-Frank, with the number of
institutions declining by more than 2,000 (–31.0 percent) since 2011. Bank
consolidation is not inherently uncompetitive, but consolidation that is driven
by regulations reflects the distortionary burden of regulatory costs. After DoddFrank, the total loans by small banks has declined from $889 billion to $815
billion (–8.3 percent) since 2011. If these small banks had instead grown their
loan portfolios by 1.55 percent—the average of the past three expansions—
during this period, there would have been about 20 percent more small bank
loans now than there actually are. These missing loans are associated with
about $6.3 billion in additional annual value added in small banking, which we

128 |

250-840_text_.pdf 132

Chapter 3

2/7/20 3:46 PM

estimate to produce about $3 billion in annual surplus for lenders and borrowers.40 Including effects on the entire economy due to additional employment
and investment, the Crapo Bill has annual net benefits of almost $5 billion and
raises real annual incomes by about $6 billion by removing regulatory burdens
from small bank lenders.
The CEA has also conducted industry-specific analyses of the effects of
several other regulations that were introduced during the years 2010–16 and
have been removed (or are in the process of being removed) during the Trump
Administration. One of these was the attempt by the Consumer Financial
Protection Bureau (CFPB) to largely eliminate the small dollar lending industry,
which had revenues of about $7 billion per year in 2015 (82 FR 54479). Small
dollar lending is a valuable service that provides consumers with important
resources and flexibility to better manage their finances. The CFPB’s analysis acknowledged that consumers found the loans helpful for paying “rent,
childcare, food, vacation, school supplies, car payments, power/utility bills,
cell phone bills, credit card bills, groceries, medical bills, insurance premiums,
student educational costs, daily living costs,” and other pressing expenses (82
FR 54515). The CFPB predicted that its rule would reduce activity in the small
dollar lending industry by 91 percent. The lost flexibility to use small dollar
lending to help pay for pressing expenses is indicative of the opportunity costs
of sharply contracting the industry. Using revealed preference methods, the
CEA estimates a corresponding loss of consumer and producer surplus of $3
billion, and a reduction of real incomes by about $7 billion.41

Additional Regulations
Among our sample of 20 rules, we find that 6 have comparatively small aggregate effects: DOL’s Fiduciary Rule, the Security and Exchange Commission’s
Disclosure of Foreign Payments by Resource Extraction Issuers, the Department
of the Interior’s Stream Protection Rule, the CFPB’s prohibition of arbitration agreements in financial contracts, the Waste Prevention Rule, and a
U.S. Department of Agriculture (USDA) rule implementing the Packers and

40 Our estimate of lender surplus uses the Lerner-index estimates from Koetter, Kolari, and
Spierdijk (2012) and assumes a unit price-elasticity of loan demand with respect to net interest
margin.
41 Assuming that the industry demand for small dollar lending is linear in the fees charged and has
a point elasticity of –1, the lost consumer surplus alone is $2.7 billion. The lost consumer surplus
is even more if the demand for small dollar lending has a constant elasticity, even if this elasticity
were as far from zero, as is the firm-level elasticity of -4.28 estimated by McDevitt and Sojourner
(2016).

Regulatory Reform Unleashes the Economy | 129

250-840_text_.pdf 133

2/7/20 3:46 PM

Stockyards Act.42 We estimated that eliminating these 6 rules, as the Trump
Administration has done, increases real incomes by about 0.06 percent in total,
which is about $11 billion per year. A 7th rule that has also been eliminated,
the Fair Pay and Safe Workplaces Rule, may technically have zero effect on
GDP and real incomes because it raises the costs of Federal contractors whose
contribution to GDP is by definition its costs.43 Although the effects of these 7
rules are likely large compared with many of the rules not in our sample, $11
billion per year is a small fraction of the combined effects of the other 13 rules
in our sample.
We have not measured the economic impact of hundreds of FY 2017 and
FY 2018 Federal rules, including a few regulations. However, the aggregate cost
savings reported for the other rules as recorded in the Federal Register are in
the direction of additional cost savings, suggesting that the cost savings from
our sample of 20 deregulatory actions may be a conservative estimate of the
cost savings from all regulatory and deregulatory actions since January 2017.

The Doubling Effect of Shifting from a Growing
Regulatory State to a Deregulatory One
Before 2017, the Federal regulatory norm was the perennial addition of new
regulations. As shown above, in figure 3-1, between 2000 and 2016, the Federal
government added an average of 53 economically significant regulations each
year. During the Trump Administration, the average has been only 10 (not
counting deregulatory actions or transfer rules).
Even if no old regulations were removed, freezing costly regulation would
allow real incomes to grow more than they did in the past, when regulations
were perennially added (shown by the dark blue line in figure 3-6), as with the
42 The Fiduciary Rule added to the costs of saving for retirement by further expanding the
circumstances under which a financial adviser is considered to be fiduciary. DOL estimated at the
time the rule was published in 2016 that it would benefit investors on net. The rule was vacated in
toto by the Fifth Circuit Court of Appeals in Chamber of Commerce v. Department of Labor, 885 F.3d
360 (5th Cir. 2018). The Disclosure of Foreign Payments by Resource Extraction Issuers Rule raised
costs for U.S. extraction companies. “Hydrological balance” provisions of the Stream Protection
Rule would shut down much of the U.S. longwall mining industry (Murray Energy Corporation v.
U.S. Department of the Interior, 2016). The CFPB “prohibit[ed] consumers and providers of financial
products and services from agreeing to resolve future disputes through arbitration rather than
class-action litigation,” which would have raised the prices of consumer financial products (U.S.
Department of the Treasury 2017). The Waste Prevention Rule added additional restrictions on
“oil and gas drilling and extraction operations on Federal and tribal lands” (CEA 2019b, 287). The
USDA rule interfered with vertical contracts in the production of poultry and pork, raising costs
throughout the supply chains (8th circuit 2018).
43 In contrast, raising the costs of private enterprises typically does reduce GDP and real incomes
because their contribution to GDP depends on the value those enterprises create for their
customers, as measured by what customers pay. The CEA notes that the production of some of
the Federal contractors may be measured like those of private enterprises, in which case zero is a
conservative estimate of the real income effect of overturning the rule.

130 |

250-840_text_.pdf 134

Chapter 3

2/7/20 3:46 PM

Figure 3-6. Deregulation Creates More Growth Than a Regulatory
Freeze, 2001–21
Real income

Additional
gain from
deregulation
(1.3%)
Cumulative
gain from a
regulatory
freeze (0.8%)

Previous
growth path

2001

2006

2011

2016

2021

Source: CEA calculations.

yellow line in figure 3-6. The amount of extra income from a regulatory freeze
depends on (1) the length of time that the freeze lasts and (2) the average
annual cost of the new regulations that would have been added along the
previous growth path. For the sake of illustration, figure 3-6 shows a freeze
through 2021. We also have a conservative estimate of the average annual cost
of regulatory additions during the years 2010–16, namely, the cost of 20 of the
rules created during those years and identified in our sampling. At 1.3 percent
of real income spread over those 7 years, that is an annual cost addition of
about 0.19 percent a year (i.e., about $1,900 per household after 7 years). Those
years are somewhat unusual in terms of numbers of new economically significant regulations, so we take the previous trend (for 2001–16) to be 0.16 percent
a year. In other words, by the fifth year of a regulatory freeze, real incomes
would be 0.8 percent (about $1,200 per household in the fifth year) above the
previous growth path.
As well as restraining the addition of new regulations, the Trump
Administration has removed previous ones. As shown by the red line in figure
3-6, removing costly regulations allows for even more growth than freezing
them. As explained above, the effect, relative to a regulatory freeze, of removing 20 costly Federal regulations has been to increase real incomes by 1.3
percent. In total, this is 2.1 percent more income—about $3,100 per household

Regulatory Reform Unleashes the Economy | 131

250-840_text_.pdf 135

2/7/20 3:46 PM

Box 3-2. How Old Are Midnight Regulations?
A number of the regulations reversed by the Trump Administration have
been called “midnight regulations,” which are final rules published between
Election Day and the inauguration of a new President. (Thus, midnight regulations refer to regulations finalized at the end of a Presidential term and before
the change to a President of the other political party.)
A new President can reverse the midnight regulations by using the
standard rulemaking process to refuse to defend the regulations in court, or
by (together with Congress) overturning them with procedures established by
the 1996 Congressional Review Act (CRA). In theory, the publishing of a costly
midnight regulation, along with its reversal soon afterward, could have little
or no effect on industry or the wider economy if market participants recognize
that the midnight rules would not last long enough to constrain economic
activity. (If market participants anticipate use of the Congressional Review
Act, a costly midnight regulation could have the opposite effect, because
the CRA would prohibit all future administrations from promulgating the
same or a similar rule imposing those costs, until a future Congress expressly
approved that type of regulation.) However, the most costly of the 2016 midnight regulations cannot be characterized this way because (1) they had been
in the rulemaking process for years before the 2017 inauguration, (2) most of
the 2016 polls and media predicted a different election outcome, and (3) the
CRA had been used only once before 2017.
Sixteen Obama-era regulations were ultimately nullified by the CRA.
The more economically important of these are the Federal rule allowing
States to mandate employers to provide retirement accounts (the “IRAmandate rule”), the FCC rule regarding broadband privacy, and the Securities
and Exchange Commission’s rule requiring public disclosure of foreign payments (RIN 1210-AB71; see also 1210-AB76, document FCC-2016-0376-0001,
and RIN 3235-AL53, respectively). They date back as far as 2010 but became
eligible for CRA nullification in the 115th Congress because challenges from
courts and the public extended the rulemaking process until late 2016, or
later. (See also Public Citizen 2016, which found that midnight regulations
“of Presidents Bill Clinton and George W. Bush took longer [3.6 years], and
underwent more days of OIRA review than the average rule over the past 17
years.”) The IRA-mandate rule dates back to at least 2015. The proposed FCC
privacy rule was released April 1, 2016, although arguably it was anticipated
by the FCC’s actions on “net neutrality” dating back to 2010.
The CEA therefore sees the Obama-era economic regulations as part of
a normal rulemaking process rather than an economically irrelevant signaling
of a political platform. Although final rules follow their notices of proposed
rulemakings with a time lag, and a new Administration may decline to finalize
notices of proposed rulemaking from a previous Administration, the length
of the time lag should not affect estimation of the medium- to long-term
economic effects of deregulation or of a regulatory freeze. The length of the
time lag does affect the timing of the economic effects.
132 | Chapter 3

250-840_text_.pdf 136

2/7/20 3:46 PM

each year—relative to the previous growth path.44 (Also see box 3-2 on socalled midnight regulations.)

Regulations Before 2017 with
Disproportionate Costs
The analysis thus far has primarily considered the effects of regulation on
income, but regulation—or the lack of it—can affect well-being in nonpecuniary
ways not captured by income. However, even when including nonpecuniary
costs and benefits, we estimate that deregulatory actions have a net benefit of
more than $2,500 per household each year, compared with the previous trend
of growing regulatory costs. The gain stems from the fact that the new level
of regulation strikes a better balance between the costs of regulations and
their societal benefits, where benefits include things valued by people but not
necessarily bought or sold in the marketplace (and that thus are not included in
the National Income and Product Accounts or in the usual income measures).
The Trump Administration requires Federal agencies to conduct cost-benefit
analyses of significant regulatory actions, including deregulatory actions, and
that they only be issued “upon a reasoned determination that benefits justify
costs” (OMB 2017).
An example from health policy illustrates how regulations before 2017
created disproportionate incremental costs and benefits. The Affordable Care
Act created an individual mandate in order to reduce the costs of uncompensated care.45 But the average annual costs of uncompensated care are about
$1,000 per uninsured person (including zeros in the average for those who are
uninsured who do not use uncompensated care during the year), whereas the
annual economic costs of the individual mandate are over $3,000 per uninsured person induced to purchase coverage (CEA 2019a).
One economic reason that regulations before 2017 were so costly is that
some of them were implemented with only a little “safety valve” in terms of
an option for regulated businesses to pay a moderate fine in instances when
compliance is especially costly. For example, whereas automobile manufacturers had the option of paying a penalty to the National Highway Traffic Safety
Administration (NHTSA) for falling short of Federal fuel economy standards,
the EPA is prohibited by the Clean Air Act from adopting the NHTSA’s penalty
structure to enforce the greenhouse gas standard that began with model year
2012 (75 FR 25482). As another example, a consultant incorrectly filling out
DOL Form LM-21 (one of the requirements under the rescinded Persuader
Rule) would be exposed to criminal penalties. Another reason is that the labor
44 The red line’s path in figure 3-6 is drawn as linear for illustration purposes only. The 1.3 percent
effect (relative to a freeze) of deregulation is likely nonlinear over time, and it may take more than
five years to be fully realized.
45 Section 1501(a)(2)(F) of the Patient Protection and Affordable Care Act.

Regulatory Reform Unleashes the Economy

250-840_text_.pdf 137

| 133

2/7/20 3:46 PM

market is arguably the largest market of all, with annual revenues of more
than $10 trillion, and it was the object of active rulemaking by DOL during the
Obama Administration.

Conclusion
Coincidentally with the 2017 Presidential inauguration, real GDP growth
changed from underperforming experts’ forecasts to outperforming them
(Tankersley 2019). The CEA’s findings on the aggregate effects of regulations
and deregulations may help explain this turnaround. Regulatory actions and
their aggregate effects may be easily overlooked and underestimated because
the actions are numerous and, if not seen through the lens of economic analysis, may appear cryptic to the general public. This chapter helps to narrow this
information gap by showing the importance of the deregulatory agenda for
everyday Americans as well as the national economy.
Since 2017, consumers and small businesses have been able to live and
work with more choice and less Federal government interference. They can
purchase health insurance in groups or as individuals without paying for categories of coverage that they do not want or need. Small businesses can design
compensation packages that meet the needs of their employees, enter into a
genuine franchise relationship with a larger corporation, or seek confidential
professional advice on how to organize their workplaces. Consumers have a
variety of choices for less expensive wireless and wired Internet access. Small
banks are no longer treated as “too big to fail” (which they never actually were)
and as subject to the costly regulatory scrutiny that goes with this designation.
In addition to regaining freedoms that they once had, consumers and
small businesses no longer need to dread the steady accumulation of costly
new Federal regulations. In a time frame of 5 to 10 years, these landmark
changes to regulatory policy are anticipated to increase annual incomes by
about $3,100 per household ($380 billion in the aggregate), by increasing
choice, productivity, and competition. This chapter arrives at its aggregate
total by building estimates from the industry level. In doing so, it closely examines specific Federal rules, accounts for the unique circumstances of the industries targeted by these rules, and quantifies benefits of regulation—such as
consumer data privacy, environmental protection, fuel savings, and reductions
in uncompensated healthcare. The analysis employs an economic framework
that situates each industry in a larger economy that includes market distortions caused by taxes, imperfect competition, and other factors.
The benefits of the newest wave of deregulation compare favorably with
those during the most significant deregulatory waves of American history. Take
the deregulation of airlines and trucking that occurred four decades ago, as the
major parts of a deregulation wave described as “one of the most important
experiments in economic policy of our time” (Winston 1993). Combined, the

134 | Chapter 3

250-840_text_.pdf 138

2/7/20 3:46 PM

Carter Administration’s deregulation of these two industries provided net
aggregate benefits of about 0.5 percent of national income. Although no 2 of
the 20 deregulatory actions analyzed in this chapter have had (according to
our estimates) such a large net benefit, their combined net aggregate benefits
exceed 0.6 percent of national income.46
Other notable historical deregulations were of natural gas markets
between 1985 and 1993, which had benefits estimated at about 0.2 percent
of national income (Davis and Kilian 2011). This is hardly more than the combined net benefit of the three health insurance rules analyzed in this chapter.
Moreover, the totals reported in this chapter reflect only deregulatory actions
occurring during less than three years, whereas the full effects of the deregulation of airlines, trucking, and natural gas each reflect actions taken over almost
a decade.47 .
There is room for additional deregulation to further grow the economy,
increasing benefits to American consumers, workers, and businesses. According
to the accounting for Executive Order 13771, the projected cost savings from
planned deregulatory actions in FY 2020 exceed the combined cost savings
achieved in 2017, 2018, and 2019. The Administration has also taken further
steps to promote regulatory reform. On October 9, 2019, President Trump
signed two regulatory reform Executive Orders. The first is titled “Promoting
the Rule of Law Through Improved Agency Guidance Documents.” Many discussions of Federal regulatory and deregulatory actions, including most of this
chapter, focus on rules adopted through the Administrative Procedure Act’s
notice-and-comment rulemaking process. In addition to such rules, Federal
agencies issue nonbinding guidance documents. Although guidance documents are not subject to the notice-and-comment requirements, some impose
substantial regulatory costs. The new Executive Order’s improvements to
guidance documents include requirements that clarify their nonbinding status.
Significant guidance documents are also now subject to cost-benefit analysis.
The second Executive Order, signed on October 9, is titled “Promoting the Rule
of Law Through Transparency and Fairness in Civil Administrative Enforcement
and Adjudication.” In an economic framework, agencies’ enforcement strategies can have important implications for regulatory costs (Fenn and Veljanovski
1988). Perhaps more important, the enforcement of regulations should be fair
to the public. The new Executive Order “prohibits agencies from enforcing rules
they have not made publicly known in advance.” Finally, in parallel with the
46 Winston (1993, table 6) reports net benefits accruing in the airline and trucking industries that
hold aggregate factor supplies constant. In calculating the 0.6 percent for comparison, we also held
aggregate factor supplies constant.
47 Murphy (2018, 76) cites “U.S. Federal intervention into the petroleum industry in the 1970s [as]
arguably the largest peacetime government interference with the economy in the nation’s history.”
Arrow and Kalt (1979) estimate the cost of this intervention to be 0.2 percent of national income.
Moreover, the 1979–81 deregulation did not realize this full amount in cost savings because price
controls were replaced with a windfall profits tax.

Regulatory Reform Unleashes the Economy | 135

250-840_text_.pdf 139

2/7/20 3:46 PM

reforms of Federal regulations, the Administration has created the Governors’
Initiative on Regulatory Innovation to encourage States to adopt regulatory
reforms. The initiative will help governors and the White House work with leaders in local and tribal governments to cut regulatory costs, advance reforms to
occupational licensing, and align regulations across levels of government.

136 |

250-840_text_.pdf 140

Chapter 3

2/7/20 3:46 PM

x
Chapter 4

Energy: Innovation and
Independence
U.S. energy innovation has continued to flourish under the Trump
Administration. Innovation—and the policies that support it—lowers costs
and prices, and increases production. This is illustrated by the American shale
revolution and its dramatic rise in oil and gas drilling productivity in shale and
similar geologic formations. Gains in shale drilling productivity have led to
lower prices for natural gas, electricity, and oil, saving the average family of
four $2,500 annually. Shale-driven savings represent a much larger percentage
of income for the lowest fifth of households than for the highest fifth.
Production growth due to shale innovation has also brought energy independence to the United States, a goal first set by President Richard M. Nixon, and
pursued by subsequent Administrations, but accomplished under the Trump
Administration. In 2017, the United States became a net exporter of natural gas
for the first time since 1958; and in September 2019, the United States became
a net exporter of crude oil and petroleum products and is projected to remain
a net exporter for all of 2020 for the first time since at least 1949. Historically,
a rise in energy prices increased the trade deficit and costs for firms and
households, sometimes pushing the U.S. economy into a recession. The
innovation-driven surge in production and exports has made the U.S. economy
more resilient to global oil price spikes. It has also improved the country’s
geopolitical flexibility and influence, as evidenced by concurrent sanctions on
two major oil-producing countries, Iran and Venezuela.
In addition to consumer savings and energy independence benefits, the shale
revolution has reduced carbon dioxide and particulate emissions through

137

250-840_text_.pdf 141

2/7/20 3:46 PM

changes in the composition of electricity generation sources. We estimate that
from 2005 to 2018, the shale revolution in particular lowered carbon dioxide
emissions in the electric power sector by 21 percent. This contributed to a
greater decline in carbon dioxide and particulate emissions (relative to the size
of the economy) in the United States than in the European Union, according to
the most recent data.
The Trump Administration’s deregulatory energy policy follows earlier Federal
deregulatory policies that helped to spur the shale revolution. By limiting unnecessary constraints on private innovation and investment, the Administration
supports further unleashing of the country’s abundant human and energy
resources. In contrast, the State of New York has banned shale production and
stymied new pipeline construction, leading to falling natural gas production in
the State, greater reliance on energy produced elsewhere, and higher energy
prices. Similarly, evidence on renewable energy mandates at the State and
Federal levels shows their costs and limitations. More broadly, predicting the
evolution of energy markets and technologies remains difficult—few anticipated the shale revolution’s effect on lower prices for natural gas, electricity,
and oil or the current economic challenges in the nuclear power sector. This
difficulty highlights the value of policies that avoid picking winners and losers
and instead provides a broad platform upon which innovation will flourish.

T

he classic effects of innovation are improvements in productivity,
which lower costs and prices and increase production.1 Energy sector
innovations—and the policies that support them—have similar effects
and ultimately reduce prices for American households and businesses. This
chapter describes the causes and consequences of growth in oil and natural gas
production from shale and similar geologic formations, while also highlighting
broader energy sector innovations and policy questions. We first discuss the
dramatic rise in productivity and its effects on cost, production, and price.
Second, we estimate the consumer savings brought by shale-driven declines in
energy prices. Third, we document how the surge in shale production has led to

1 The CEA previously released research on topics covered in this chapter. The text that follows
builds on the report “The Value of U.S. Energy Innovation and Policies Supporting the Shale
Revolution” (CEA 2019).

138 |

250-840_text_.pdf 142

Chapter 4

2/7/20 3:46 PM

U.S. energy independence, as measured by positive net exports of both oil and
natural gas. Fourth, we assess total and shale-related changes in emissions in
the United States. Finally, we consider the implications of deregulatory versus
government-directed energy policies.
From 2007 to 2019, innovation in shale production brought an 8-fold
increase in extraction productivity (new well production per rig) for natural gas
and a 19-fold increase for oil. These productivity gains have reduced costs and
spurred production to record-breaking levels. As a result, the United States
has become the world’s largest producer of both commodities, surpassing
Russia in 2011 (for natural gas) and Saudi Arabia and Russia in 2018 (for oil).
The Council of Economic Advisers (CEA) estimates that greater productivity
reduced the domestic price of natural gas by 63 percent as of 2018 and led to a
45 percent decrease in the wholesale price of electricity. The increase in U.S. oil
production linked to shale oil development helped not only moderate but also
reduce the global price of oil over the same period in the face of “peak oil” forecasts. By lowering energy prices, we estimate that the shale revolution saves
U.S. consumers $203 billion annually, or $2,500 for a family of four. Nearly 80
percent of the total savings stem from a substantially lower price for natural
gas, of which more than half comes from lower electricity prices. Because lowincome households spend a larger share of their income on energy bills, lower
energy prices disproportionately benefit them; shale-driven savings represent
6.8 percent of income for the lowest fifth of households, compared with 1.3
percent for the highest fifth. These consumer savings are in addition to economic benefits linked to greater employment in the sector.
At the same time, shale-driven production growth has fulfilled the nearly
50-year goal of U.S. energy independence. In 2017, the United States became a
net exporter of natural gas for the first time since 1958; and in September 2019,
the United States became a net exporter of crude oil and petroleum products
and is projected to remain a net exporter for all of 2020 for the first time since
at least 1949. The long-standing goal of energy independence was motivated
by the historic vulnerability of the U.S. economy to oil price spikes. Historically,
a rise in energy prices increased the trade deficit and costs for firms and households, potentially pushing the U.S. economy into a recession. In fact, a sudden
rise in the price of oil preceded 10 of the 11 postwar recessions in the United
States (Hamilton 2011). With energy independence, spikes in global energy
prices continue to affect U.S. households and businesses, but they now have
a more muted effect on gross domestic product (GDP) because they do not
inflate the trade deficit as they did when net imports were high. From 2000
to 2010, a $1 increase in oil prices reduced the U.S. trade balance in goods by
$0.83 billion; from 2011 to 2019, it reduced it by only $0.17 billion. Higher prices
could even increase GDP if they cause a large enough increase in investment
by U.S. energy producers. Greater exports and resilience to price shocks have

Energy: Innovation and Independence | 139

250-840_text_.pdf 143

2/7/20 3:46 PM

also improved the country’s geopolitical flexibility and influence, as evidenced
by concurrent sanctions on two major oil producers.
In addition to consumer savings and energy independence benefits,
the shale revolution has reduced carbon dioxide and particulate emissions
through changes in the composition of electricity generation sources. The
CEA estimates that from 2005 to 2018, the shale revolution in particular was
responsible for reducing electric power sector carbon dioxide emissions by 21
percent. This contributed to a greater decline in carbon dioxide emissions and
particulate emissions (relative to the size of the economy) in the United States
than in the European Union from 2005 to 2017, the most recent year for data
in both areas.
The Trump Administration’s deregulatory energy policy follows earlier Federal deregulatory policies that helped to spur the shale revolution.
By limiting unnecessary constraints on private innovation and investment,
the Administration’s deregulatory policy supports further unleashing of the
country’s abundant human and energy resources. In contrast, the State of
New York has banned shale production and stymied new pipeline construction, leading to falling natural gas production in the State, greater reliance on
energy produced elsewhere, and higher energy prices. Similarly, evidence on
renewable energy mandates at the State and Federal levels shows their costs
and limitations. More broadly, predicting the evolution of energy markets and
technologies remains difficult—few anticipated the shale revolution’s effect on
lower prices for natural gas, electricity, and oil or the current economic challenges in the nuclear power sector. This highlights the value of policies that
avoid picking winners and losers and instead provides a broad platform upon
which innovation will flourish.

Market Pricing, Resource Access,
and Freedom to Innovate
Growth in the extraction of oil and natural gas from shale and similar geologic
formations—often referred to as the shale revolution—is arguably the most
consequential energy development in the last half century. Its far-reaching
consequences are in part because fossil fuels account for 80 percent of U.S.
energy consumption (EIA 2019b). Most oil goes to fuel the planes, trains, and
automobiles of the transportation sector, while most natural gas generates
electric power or heat for industry and households.
Since at least the late 1970s, geologists knew that shale and other lowpermeability formations contained prodigious amounts of natural gas. For
decades, methods to profitably extract the gas eluded the industry, much of
which pursued easier-to-access resources in the United States and abroad.
Although various countries have abundant shale resources, entrepreneurs
and engineers working in the United States’ innovation-friendly context first
140 |

250-840_text_.pdf 144

Chapter 4

2/7/20 3:46 PM

unlocked the potential of shale, which would eventually bring large savings
to consumers and environmental benefits relative to a scenario without shale
development.
The shale revolution came after major deregulatory changes in the
governance of natural gas pricing and distribution. Three major deregulatory actions—the 1978 Natural Gas Policy Act, the Federal Energy Regulatory
Commission’s 1985 Open Access Order, and the 1989 Natural Gas Wellhead
Decontrol Act—liberalized access to pipelines and increased the role of market
forces in determining prices paid to natural gas producers. Earlier price controls
discouraged production and exploration, leading to supply shortages. Once
freed to move with supply and demand, wellhead prices increased, encouraging more innovation, which eventually lowered prices (MacAvoy 2008). Prices,
however, would begin to increase again in the late 1990s and early 2000s.
Higher wellhead prices justified taking innovative risks on new methods
and geologic formations, and private ownership of underground resources
made it easy for firms to access these resources and experiment in diverse
locations. The United States is unique in that the private sector—homeowners,
farmers, and businesses—owns the majority of subsurface mineral rights. This
system allows private owners to grant access to energy firms through lease
contracts, which can be for one-tenth of an acre or 10,000 acres (Fitzgerald
2014). As a result, energy firms do not need to navigate a cumbersome central
government bureaucracy to begin accessing subsurface resources. Although
firms must still abide by Federal and State regulations, gaining the right to
access resources is straightforward—they just need to adequately compensate
the owner of the relevant acreage.
The role of the Federal government in unlocking the shale revolution is
often overstated. Certainly, the U.S. Department of Energy’s (DOE) investment
of about $130 million from 1978 to 1992 in Federal funding for research on
drill bit technology, directional drilling, modeling for shale basin reservoirs,
and microseismic monitoring of multistage hydraulic fracturing treatment
helped spur sector innovation. A more detailed analysis shows that primary
credit belongs to the private sector. Federally subsidized research to aid the
development of shale gas in the East carried limited transferability to the early
breakthroughs in Barnett shale formation. Moreover, an early tax credit aimed
at stimulating the production of natural gas from unconventional sources
expired in 1992, well before important breakthroughs in the early 2000s.2
Among firms pioneering in shale extraction, the most important is
arguably Mitchell Energy. In the 1980s and 1990s, Mitchell Energy, which had
long-term contracts to sell its natural gas, experimented with methods to
coax natural gas from a Texas geologic formation known as the Barnett Shale.
2 Wang and Krupnick (2015) discuss Federal government policies that may have aided Mitchell
Energy as it experimented in the Barnett and generally conclude that subsidies, tax credits, taxpreferred business structure, and research and development played a secondary role.

Energy: Innovation and Independence | 141

250-840_text_.pdf 145

2/7/20 3:46 PM

Consistent commercial success emerged in the early 2000s, when Devon
Energy acquired Mitchell Energy. This acquisition accelerated the merger
of two complementary technologies. Devon had considerable experience
with horizontal drilling, which involves drilling a conventional vertical well,
and at the bottom of the vertical leg, transitioning to a horizontal leg, which
can extend for several miles. Mitchell Energy had more experience pumping
liquids and sand under high pressure into wells to fracture low-permeability
formations, thereby releasing gas and/or oil trapped in the rock. This stimulation technique is known as hydraulic fracturing (Wang and Krupnick 2015).
Promising results from Devon’s wells, coupled with rising natural gas prices,
spurred a drilling boom in the Barnett Shale. Thus, the number of well permits
issued in the Barnett grew from less than 300 in 2000 to more than 4,000 in
2008. The revolution had begun.
The shale revolution may not have been sustained if it had not been for
continued innovation by scores of engineers, geologists, and entrepreneurs,
who refined and adapted methods to draw oil from western North Dakota and
southern Texas as well as natural gas from Appalachia in the Eastern United
States. Persistent innovation and opportunity for its diffusion has transformed energy markets, with considerable implications for consumers and the
environment.
Important innovations have also occurred elsewhere in the energy sector. Advances in the design of combined-cycle turbines in natural gas plants
have allowed the plants to generate more electricity from each unit of heat.
From 2008 to 2017, the amount of heat needed to generate a kilowatt-hour of
electricity declined by 10 percent. In addition, the cost of turbines, measured in
dollars per unit of capacity, has fallen by 11 percent since 2014. Alongside more
efficient and less costly natural gas turbines, the cost of wind power projects
has also fallen recently, causing wind power prices to fall by more than 50
percent from 2010 to 2017. These gains stem from various factors, including
larger turbines and lower manufacturing costs. Solar power generation has
made similar gains.
Innovations in these sectors proved complementary. Electricity from
wind and solar technologies remain variable and present challenges for grid
management because generation may not align with the demands of the
electric grid in any given hour. Relative to most other sources of electricity,
natural-gas-fired generators can quickly ramp up and ramp down generation
to assist with grid integration and systems balancing requirements. Gains from
innovation, however, have not occurred everywhere. Cellulosic biofuel production has grown slowly and is well below levels prescribed by a Federal mandate
(see box 4-1).

142 |

250-840_text_.pdf 146

Chapter 4

2/7/20 3:46 PM

Box 4-1. The Limits of Energy Mandates to Induce Innovation
The directness of government mandates can have great appeal. Commands
that the market conform to government targets, however, have limits in what
they can achieve, as illustrated by the Federal Renewable Fuel Standard. Even
when targets are met, they can come at a much higher cost than projected.
$"0- 
уҊ$ѵ
ѵѵ
 ''0'*.$
$*!0 '
//0/ 
)
$)'
*'0( .Ѷ
спрп–рш
Gallons (billions)
ш

спрш

ч

//0/

ц
х
ф

у
т

с
р

п
спрп

$)'
спрр

спрс

спрт

спру

спрф

спрх

спрц

спрч

спрш

*0- ѷ *)"- ..$*)'
 . -#
 -1$ ѵ

To further U.S. energy independence and provide additional revenue
sources to U.S. farmers, the Federal Renewable Fuel Standard, which was set
in 2005 and expanded in 2007, mandated increases in the domestic production and consumption of renewable fuels. The standard mandated the use of
different categories of renewable fuels, with type-specific targets increasing
over time for most categories. Technology to produce ethanol from corn was
well established by the mid-2000s, and corn-based ethanol production and
consumption quickly increased and have generally kept in line with the targets set in the 2007 statute. In contrast, technology to convert cellulosic plant
material, such as corn fodder, into renewable fuels was not well established
when the standard went into effect, and progress has been slow despite the
mandate. As a result, the EPA has utilized its waiver authority, authorized in
the 2007 statute, when setting targets for cellulosic biofuel (figure 4-i). The
cellulosic mandate has been waived every year since its establishment in
2010, resulting in no significant production of cellulosic biofuel. By 2019, the
industry was to have produced 8.5 billion gallons of cellulosic biofuel.

Energy: Innovation and Independence | 143

250-840_text_.pdf 147

2/7/20 3:46 PM

The Effects of Innovation on
Productivity, Prices, and Production
Innovation raises productivity and lowers production costs, allowing firms to
offer lower prices. This dynamic corresponds to the textbook case of an outward shift in the domestic supply curve, as shown in figure 4-1, for the case of
natural gas. The shift means that firms produce more at every price level than
they did before innovation, which lowers the market equilibrium price, which
is shown on the vertical axis in figure 4-1 as a change in P, while increasing the
quantity produced, as shown on the horizontal axis as the change in Qp. The
lower price stimulates an increase in consumption, as shown on the horizontal
axis as the change in Qc.
Because of imports and exports of natural gas, the market price is
affected by the global price and does not occur at the intersection of domestic supply and domestic demand. Before shale gas development, domestic
consumption exceeded domestic production, leading to imports, as shown in
figure 4-1 as the difference between domestic production and consumption
before shale. After shale, domestic production exceeds domestic consumption,
leading to exports.

The Impact on Productivity
Horizontal drilling and hydraulic fracturing made the development of shale
and other low-permeability formations economical. In the last decade, all
growth in onshore oil and gas production has come from the development of
these formations. One measure of innovation and productivity gains by energy
producers is the quantity that new wells are producing relative to the number
of rigs in use, which the DOE’s Energy Information Administration (EIA) tracks
for all major shale formations. This measure, known as new-well production
per drilling rig, is defined as the total production of wells recently brought into
production divided by the number of drilling rigs recently in operation.
New–well production per rig increased by more than 8-fold between 2007
and 2019 for key shale gas regions and by more than 19-fold for key shale oil
regions. Particularly strong growth has occurred in the last five years for both
oil and gas (figure 4-2).3 The recent growth highlights how energy firms have
continued to improve upon the earlier breakthroughs of shale pioneers.
The productivity gains in production per rig stem from several factors
that allow firms to generate more production from each rig per unit time. For
example, across regions and over time, the number of days needed to drill a
3 The sharp rise in productivity in 2016 largely reflects firms deciding to operate fewer drilling rigs
(because of very low prices) and focus on bringing wells already drilled into production. This can
be seen by a sharp decline in drilled but uncompleted wells in 2016. Similarly, a rise in drilled but
uncompleted wells in 2017 helps explain the apparent slowdown in productivity in that year. See
EIA (2019) for estimates of drilled but uncompleted wells.

144 | Chapter 4

250-840_text_.pdf 148

2/7/20 3:46 PM

Figure 4‐1. Innovation in Natural Gas Production
Domestic supply,
no shale

Price

Domestic
supply, shale

The effect of
innovation
Imports
P, no shale
Δ Price

A
P, shale

Exports
B

C

D

Domestic
demand
Qc, no shale Qc, shale
Qp, shale
Qp, no shale

Quantity

Δ Production

$"0- 
уҊсѵ
-*0/$1$/4
$).ѷ
 2Ҋ ''
-*0/$*)
+ -
$"Ѷ
$'
)

/0-' .Ѷ
сппц–рш

Cubic feet per day (thousands)
цѶппп
 +Ҋрш

Barrels per day
шпп
чпп

хѶппп

цпп

фѶппп

хпп

уѶппп

фпп
упп

тѶппп

/0-'
".
1 -" 
җ-$"#/ 3$.Ҙ

тпп

сѶппп

спп

$'
1 -" 
җ' !/ 3$.Ҙ

рпп

рѶппп

п

п
сппц

сппш

спрр

спрт

спрф

спрц

спрш

*0- .ѷ
) -"4
 )!*-(/$*)
($)$./-/$*)Ѹ

'0'/$*).ѵ
*/ ѷ
 2Ҋ2 ''
+-*0/$*)
$.
/# 
)0( -
*!

*$'
җ*-
".Ҙ
2 ''.Ѷ
)
/# $-
*0/+0/Ѷ
/#/
- 
$)
/# $-

!$-./
(*)/#
*!
+-*0/$*)ѵ
# 
-$"
*0)/
$.
/# 
)0( -
*!
/$1 
*$'
җ*-
".Ҙ
-$''$)"
-$".
/2*
(*)/#.

+-$*-ѵ


Energy: Innovation and Independence | 145

250-840_text_.pdf 149

2/7/20 3:46 PM

$"0- 
уҊтѵ
$).
$)
-*0/$1$/4
 *2 -
- & 1 )
-$ .
-*..

#' 
*-(/$*).Ѷ
спру–рш

4


Dollars (2018) per thousand cubic feet
уѵп
0'Ҋрш

Dollars (2018) per barrel
чф

цф

тѵф
/0-' ". 1 -" 

җ-$"#/ 3$.Ҙ

хф

тѵп

$'
1 -"
җ' !/ 3$.Ҙ

фф

сѵф
сѵп

уф
тф
спру

рѵф
спрф

спрх

спрц

спрч

спрш

*0- .ѷ
'**( -"Ѹ

)'4/$.Ѹ

'0'/$*).ѵ
*/ ѷ
- & 1 )
+-$ .
$)'0 
/# 
*./
*!
-$''$)"
)
*+ -/$)"

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

/*
(-& /Ѷ
$)'0$)"
-*4'/$ .Ѷ
/3 .Ѷ
)
"/# -$)"
)
*(+- ..$*)
*./.ѵ
The o$'
1 -" 
$.
/#
1 -" 
+-$ 
 /2 )
&& )
*-(/$*)Ѷ
 )1 -
.$)Ѷ
"' 
*-Ѷ
)
 -($)
.$)ѵ
The
n/0-'
".
1 -" 
$.
/# 
1 -" 
+-$ 
 /2 )
- ''0.Ҋ/$
)
 4) .1$'' ѵ
/,
%0./ 
/*
спрч
*''-.
0.$)"
/# 
*).0( -
-$ 
 ) 3
җ ҊҘ,
are 
.$3Ҋ(*)/#
(*1$)"
1 -" ѵ

well has fallen (EIA 2016), and the average production from a well’s first month
has grown (EIA 2018b). The improvements come partly from firms drilling wells
with longer horizontal portions, and from placing more wells per pad—both of
which allow each well and pad to access more oil and gas.
Greater productivity reduces the cost of producing each barrel of oil or
cubic foot of natural gas. Lower unit costs lead to a lower breakeven price,
which is the price needed to cover the costs of drilling and operating an oil or
gas well. Figure 4-3 shows an estimated breakeven price based on modeling of
production costs in different regions.4 From 2014 to 2019, the breakeven price
for natural gas (averaged across key shale formations) fell by 45 percent; for oil,
it fell by 38 percent. The link between productivity—as measured by new-well
production per rig in operation—and breakeven prices is direct. Well operators
typically lease drilling rigs, paying as much as $26,000 per day, so finishing a
well in half the time yields considerable savings. Similarly, higher volumes of
initial production return cash more quickly to the firm and can mean greater
lifetime production from the well.
4 The breakeven price, calculated by BTU Analytics, is best interpreted as the price needed to
justify drilling another well, assuming that the energy firm already holds the necessary acreage.
The price for a given period is calculated based on historical production data and projections of
future production to model revenue and costs for every well brought into production in the period.
This analysis assumes a discount rate of 10 percent and a well life of 240 months. It is not based
on energy firm calculations of their own breakeven costs and excludes potential costs that energy
firms may incur, such as interest payments on debt and costs to acquire their acreage.

146 |

250-840_text_.pdf 150

Chapter 4

2/7/20 3:46 PM

The Impact on Prices and Production
In its Annual Energy Outlook, the EIA projects energy-related outcomes for
the coming decades. The projections incorporate detailed information and
assumptions on resource reserves, emerging technologies, new policies, and
numerous other relevant trends. The difference between projected and actual
outcomes provides one measure of the surprise and disruption brought by
the shale revolution. This difference does not necessarily isolate the shale
revolution’s contribution because markets may have evolved differently than
expected for reasons other than shale.
The 2006 Annual Energy Outlook, which made projections for 2005 and
later, projected that natural gas production in the lower 48 States would rise
gradually and reach 19 trillion cubic feet by 2018. Actual dry gas production for
the lower 48 states reached more than 30 trillion cubic feet in 2018, 58 percent
higher than projected, and now greatly exceeds that of any other country
(figure 4-4). The production growth was not because of higher-than-expected
prices. To the contrary, prices in 2018 were 46 percent lower than projected
(figure 4-5).

$"0- 
уҊуѵ
/0-'
.
/0'
-*0/$*) 1 -.0. -*% / 

-*0/$*)Ѷ
сппф–рч
Cubic feet (trillions)
тс

спрч

тп
сч

/0'
+-*0/$*)

сх
су
сс
-*% / 
+-*0/$*)

сп
рч
рх
сппф

сппш

спрт

спрц

*0- .ѷ
) -"4
 )!*-(/$*)
($)$./-/$*)
җ ҘѸ

'0'/$*).ѵ
*/ ѷ
-*% /$*).
- 
!-*(
/# 
 
2006 Annual Energy Outlookѵ
-*0/$*)
$.
!*-
/# 

'*2 -
уч
// .Ѷ
2#$#
 3'0 
'.&
)
 2$$ѵ
-4
".
- ! -.
/*
".
/#/
$.

+-$(-$'4
( /#) Ѷ
-/# -
/#)
#4-*-*)
*(+*0).ѵ

Energy: Innovation and Independence

250-840_text_.pdf 151

| 147

2/7/20 3:46 PM

$"0- 
уҊфѵ
/0-'
.
/0'
-$ .
1 -.0.
-*% / 

-$ .Ѷ
сппф–рч
Dollars (2018) per million Btu
рс

спрч

рр

рп
ш

ч

-*% / 
+-$

ц

х
ф

/0'
+-$

у
т
с
сппф

сппш

спрт

спрц

*0- .ѷ
) -"4
 )!*-(/$*)
($)$./-/$*)
җ ҘѸ

'0'/$*).ѵ
*/ ѷ
/0
ۙ
-$/$.#
/# -('
0)$/ѵ
-*% /$*).
- 
!-*(
/# 
 
2006 Annual Energy
Outlookѵ
-$ .
- 
%0./ 
/*
спрч
*''-.
0.$)"
/# 
*).0( -
-$ 
 ) 3
җ Ҋ
Ҙѵ
-4
".
- ! -.
/*
".
/#/
$.
+-$(-$'4
( /#) Ѷ
-/# -
/#)
#4-*-*)
compounds.

Figure 4-6. U.S. Monthly Wholesale Electricity Price and Natural Gas
Dollars per megawatt-hour (2018)

Dollars per million Btu (2018)

140

Dec-2018

15

120
100

12

Henry hub
(right axis)

80

9
6

60
40
20

18

Wholesale
electricity
(left axis)
2005

2007

3
0
2009

2011

2013

2015

2017

Sources: Energy Information Administration; Intercontinental Exchange; CEA calculations.
Note: Btu = British thermal unit. Wholesale electricity prices were weighted by volume across
weeks and eight wholesale electricity hubs. Wholesale natural gas prices are the Henry Hub
spot price. Prices are adjusted to 2018 dollars using the Consumer Price Index (CPI-U).

148 |

250-840_text_.pdf 152

Chapter 4

2/7/20 3:46 PM

Figure 4-7. U.S. Crude Oil Production, 2005–18
Barrels per day (millions)
11

2018

10
Actual production
9
8
7
Projected production

6
5
4
2010

2012

2014

2016

2018

Sources: Energy Information Administration; CEA calculations.
Note: Projections are from the EIA 2011 Annual Energy Outlook. Production is for the
lower 48 states, which excludes Alaska and Hawaii. Production includes both
onshore and offshore production.

The unexpected production growth and price decline of natural gas
spilled over to electricity markets. Wholesale electricity prices oscillated
around $80 per megawatt–hour from 2005 to 2008, but then dropped markedly as the price of natural gas fell. Although natural gas-fired generators have
accounted for less than one-third of electricity generating in recent years, they
play an outsized role in influencing prices in competitive wholesale electricity
markets. This is because such generators are often the marginal generator of
electricity, and their operators can adjust output quickly in response to the
market with relative ease, making their costs and bid prices an important
determinant of the market price of electricity. Figure 4-6 shows the close tracking of wholesale natural gas and electricity prices, and several studies have
documented a strong causal effect of natural gas prices on wholesale electricity prices (Linn, Muehlenbachs, and Wang 2014; Borenstein and Bushnell 2015).
Turning to oil, the difference between projected and actual oil production is even starker than the case of natural gas. Actual production in the lower
48 States in 2018 exceeded the production projected by the EIA in 2011 by
85 percent, leading the United States to surpass Russia and Saudi Arabia to
become the top global oil producer. Some of the difference between actual

Energy: Innovation and Independence | 149

250-840_text_.pdf 153

2/7/20 3:46 PM

Figure 4-8. Imported Oil Prices, 2005–18
Dollars per barrel (2018)
115

2018

105
Projected price

95
85
75
65

Actual price

55
45
35
2010

2012

2014

2016

2018

Sources: Energy Information Administration; CEA calculations.
Note: Projections are from the EIA 2011 Annual Energy Outlook. Prices are adjusted to
2018 dollars using the Consumer Price Index (CPI-U). Imported crude prices are the
refiners’ average acquisition cost for imported crude oil.

and projected production stems from greater-than-expected oil prices in the
first half of the 2010–18 period. The benefit of oil sector innovation, however, is
still evident; since 2015, actual prices have been below projected prices, while
production has greatly exceeded projections (figures 4-7 and 4-8).

The Impact of the Shale-Induced Decline in Energy Prices
A simple supply-and-demand framework permits estimating how much energy
prices have fallen because of the shale revolution as opposed to other factors
that have changed over time. For natural gas, we draw from Hausman and
Kellogg (2015), who look at the market effects of shale gas from 2007 to 2013.
Their analysis focuses on estimating the price of natural gas in a world without
the shale revolution, noting that the actual change in price before and after
the emergence of shale is not necessarily the causal effect of shale because
the demand curve could have shifted. As a result, they estimate supply and
demand curves for natural gas for 2007 and for 2013. The price of natural gas
in the no-shale scenario is then estimated as the price at the intersection of

150 |

250-840_text_.pdf 154

Chapter 4

2/7/20 3:46 PM

not because of higher-than-expected prices. To the contrary, prices in 2018 were 46
percent
lower than
projected
the 2007 supply
(pre-shale)
curve
and the(figure
2013 4-5).
demand curve.5 (For details
The unexpected
growth and
price
decline
of natural gas spilled
on estimating the shale-driven
price production
effect, see Hausman
and
Kellogg
2015).
to electricity
markets.
electricity
pricesthe
oscillated
Our primary over
modifications
to their
price Wholesale
analysis are
to use 2018,
most around $80 per
megawatt–hour
from
to 2008,
then
dropped
markedly
as the price of natural
recent year for
annual data, as
the 2005
end year,
notbut
2013;
and
to use more
recent
gas supply
fell. Although
natural
gas-fired
generators
have
accounted
for less than one-third
estimates of the
elasticity
of natural
gas from
Newell,
Prest,
and Vissing
of electricity generating in recent years, they play an outsized role in influencing prices
(2019).
competitive
electricity
markets.
This on
is because
such generators are
We alsoinestimate
the wholesale
effect of lower
natural
gas prices
wholesale
often
the
marginal
generator
of
electricity,
and
their
operators
electricity prices. Natural gas plays a unique role in the electricity sector. In can adjust output
quickly
in response
the market
with relative
ease, making
their costs and bid prices
many parts of
the United
Statestothat
have competitive
wholesale
electricity
an important determinant
of the market
price ofunit
electricity.
Figure 4-6 shows the close
markets, natural-gas-fired
plants generated
the marginal
of electricity
tracking
of
wholesale
natural
gas
and
electricity
prices,
and several studies have
sold. As a result, a decline in their costs lowers the market price of electricity,
documented
a
strong
causal
effect
of
natural
gas
prices
on
wholesale
meaning that all electricity generators, regardless of their fuel source, receive electricity prices
(Linn, Muehlenbachs, and Wang 2014; Borenstein and Bushnell 2015).
a lower price. Likewise, all buyers, regardless of who provides their electricity,
Turning to oil, the difference between projected and actual oil production is
pay a lower price. Linn, Muehlenbachs, and Wang (2014) studied the effect of
even starker than the case of natural gas. Actual production in the lower 48 States in
the shale-driven decline in natural gas prices on electricity prices and found
2018 exceeded the production projected by the EIA in 2011 by 85 percent, leading the
that across wholesale market hubs, a 1 percent decrease in the price of natural
United States to surpass Russia and Saudi Arabia to become the top global oil
gas lowers the price of electricity by 0.72 percent. To estimate the shale-driven
producer. Some of the difference between actual and projected production stems
change in thefrom
wholesale
price of electricity,
therefore
shalegreater-than-expected
oil we
prices
in the multiply
first halfthe
of the
2010–18 period. The
driven percentage
change
in
the
price
of
natural
gas
(described
in
the
benefit of oil sector innovation, however, is still evident; sinceprior
2015, actual prices have
paragraph) bybeen
0.72.below projected prices, while production has greatly exceeded projections
For estimating
effect
of shale oil on prices, we consider two surges in
(figures the
4-7 and
4-8).
shale oil production, with the second surge associated with production cuts by
The of
Impact
of the Shale-Induced
Decline
in Energy
Prices
the Organization
the Petroleum
Exporting Countries
(OPEC).
The first
wave
is defined by Kilian (November 2008–August 2015), and the second we define
A simple 2019.
supply-and-demand
framework
estimating
as January 2017–May
For the first wave,
we drawpermits
from Kilian
(2017),how
whomuch energy prices
fallen
because
shale
revolution
as opposed
to other factors that have
estimates thehave
monthly
Brent
crudeofoilthe
price
absent
U.S. shale
oil development.
changed
over
time.
For
natural
gas,
we
draw
from
Hausman
and Kellogg (2015), who
For the second wave, we take the Killian effect from the end of the first wave
look at the market effects of shale gas from 2007 to 2013. Their analysis focuses on
and apply it to the change in U.S. shale oil production in the second wave, after
estimating the price of natural gas in a world without the shale revolution, noting that
taking into account the production cuts among OPEC countries since 2016.
the actual change in price before and after the emergence of shale is not necessarily
Kilian (2017) estimates the first shale oil wave reduced the global oil
the causal effect of shale because the demand curve could have shifted. As a result,
price by roughly $5.00 per barrel by August 2015. Extending his analysis to the
they estimate supply and demand curves for natural gas for 2007 and for 2013. The
second wave of production growth from shale, we estimate that the additional
price of natural gas in the no-shale scenario is then estimated as the price at the
production further
cut $1.29 per barrel by May 2019, resulting in a total price
intersection of the 2007 supply (pre-shale) curve and the 2013 demand curve.5 (For
drop of $6.29details
per barrel.
This represents
a 10 percent
decline
in the
price and Kellogg 2015.)
on estimating
the shale-driven
price
effect,
see 2018
Hausman
of oil relativeOur
to primary
what it modifications
would have been
if
the
shale
revolution
had
never
to their price analysis are to use 2018, the most recent year
occurred. for annual data, as the end year, not 2013; and to use more recent estimates of the
Turningsupply
to natural
gas, we
estimate
that
in aNewell,
no-shale
scenario,
the price
elasticity
of natural
gas
from
Prest,
and Vissing
(2019).
of natural gas would be $7.79 per thousand cubic feet, which is given by the

5
Both
prices by
arefinding
estimated
finding
price
that solves
similar basic
equation: Quantity Supplied
5 Both prices are
estimated
the by
price
that the
solves
a similar
basica equation:
Quantity
(P) Imports
+ Net Imports
(P) = Residential
(P) + Commercial
(P) + Industrial Demand (P) +
Supplied (P) + Net
(P) = Residential
DemandDemand
(P) + Commercial
DemandDemand
(P) + Industrial
ElectricPower
Power
Demand
where
is the
price
of natural
demand
Demand (P) + Electric
Demand
(P),(P),
where
P isP the
price
of natural
gas.gas.
The The
demand
and and supply curves are
where 𝜂𝜂 is
elasticity.The
The net import function is
supply curves areassumed
assumedto
totake
takethe
the form 𝑄𝑄 = 𝐴𝐴 ∙ (𝑃𝑃 + 𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚)/ , where
is an elasticity.
assumed
to be to
linear
in price
and is
estimated
using
datadata
fromfrom
2000
to 2018.
net import function
is assumed
be linear
in price
and
is estimated
using
2000
to 2018.

Energy: Innovation and Independence | 151

250-840_text_.pdf 155

2/7/20 3:46 PM

intersection of the 2007 natural gas supply curve and the 2018 demand curve.
With the shale-driven outward shift in the supply curve, the price falls to $2.87
per thousand cubic feet, a 63 percent decrease. Put differently, natural gas
prices in 2018 were 63 percent lower than they would have been if the shale
revolution had never occurred, and they were far less variable. This is roughly
the same percentage change in the Henry Hub price of natural gas over the
2007–18 period.
Based on the estimates by Linn, Muehlenbachs, and Wang (2014), the
lower price of natural gas implies that shale gas led to a 45 percent decrease
in the wholesale price of electricity as of 2018. This estimated decline is also
consistent with the wholesale futures price data listed by the EIA from the
Intercontinental Exchange. In real terms, the weighted-average wholesale
price across market hubs fell by 44 percent from 2007 to 2018.
We note that retail electricity prices did not decline during the same
period, in part because of State renewable portfolio standards mandating that
a certain percentage of a State’s electricity must come from renewable sources
like wind or solar. At least 29 States have adopted such standards, with the first
being Iowa in 1983. The most recent study of these standards finds that even
modest renewable electricity targets bring considerable retail price increases
(Greenstone and Nath 2019). They find that 12 years after a State adopted a
renewable portfolio standard, retail electricity prices increased by an average
of 17 percent. Over the same period, the standards raised the proportion of
renewable electricity generation by at most 7 percentage points.6

Innovation-Driven Consumer Savings, Energy
Independence, and Environmental Benefits
This section first explores methods of estimating consumer savings from
lower energy prices. Then it examines the salient findings related to these
consumer savings. Next, it delineates the United States’ path toward energy
independence. And finally, it discusses the environmental benefits of the shale
revolution.

Consumer Savings—Methods
Lower energy prices can benefit consumers in diverse ways—through lower
bills for heating or lighting, less spending at the gas pump, and lower prices for
goods or services that require considerable energy inputs such as airline travel
or building materials. The standard approach to estimating the total consumer
6 This assumes that the state started with zero renewable electricity generation, which is why it is a
generous estimate of the increase in renewable generation caused by the standard. The 7 percent
is based on the finding by Greenstone and Nath (2019) that the gross renewable requirement
increased to roughly 11 percent 12 years after adopting a standard and that the actual level of
renewable generation was about 4 percentage points below the grow requirement.

152 |

250-840_text_.pdf 156

Chapter 4

2/7/20 3:46 PM

benefit from a price decline is to calculate the savings for those consuming
before the price decline, whose value is represented in figure 4-1 above by
the rectangle formed by areas A, B, and C, and the savings on additional consumption spurred by the price decline, represented by area D.7 We take this
approach for oil, multiplying the shale-induced change in the price of oil ($6.29
per barrel) with the pre-shale quantity consumed (about 7.0 billion barrels
annually), and adding it to one-half the product of the price change and the
price-induced change in consumption (0.1 billion barrels).
We modify this approach for natural gas to account for the spillover
effects in the electricity market. First, we estimate savings using the standard
approach described above and following Hausman and Kellogg (2015), who
break total demand into its sectoral components, including the electricity sector. We first estimate savings for the electric power sector in the same manner
as Hausman and Kellogg (2015); call this SHK. Their approach assumes that each
$1 saved because of cheaper natural gas translates into $1 saved for electricity
consumers. This is a reasonable approach for the share of the power sector
with cost-of-service regulation, in which case regulators would only reduce
compensation to natural-gas-fired generators, not to other generators, and
only by as much as such generators had cost reductions.
For the share of the sector without cost-of-service regulation, however,
we translate the lower natural gas prices into lower wholesale electricity
prices, following Linn, Muehlenbachs, and Wang (2014). The price-setting effect
of natural-gas-fired electricity generators magnifies the effect of lower natural
gas prices because the gas-driven decline in wholesale electricity prices applies
to all electricity consumed in deregulated markets, not just the electricity
generated by natural gas. We then assume that wholesale market savings pass
through to retail savings, dollar for dollar, which is consistent with the research
of Borenstein and Bushnell (2015), who find high rates of pass-through in
deregulated markets.
One-third of the electricity generated in the United States in 2018 was
generated in States without cost-of-service regulation of generators.8 Based
on this share, we estimate total electric power sector savings to be the sum of
the savings in regulated markets (= 0.67 x SHK) and the savings from unregulated
markets (= 0.33 x SWholesale).

7 The supply shift and price change will also affect producer surplus (not shown in figure 4-1),
which is the difference between revenue and cost across all units produced and all producers.
Whether producers benefit from innovation (as measured by producer surplus) depends in large
part on how much prices fall and quantities increase. It is likely that there is a net loss in producer
surplus for natural gas producers (Hausman and Kellogg 2015) but a gain for oil producers, whose
production has increased greatly with only a modest price decline.
8 The EIA provided the CEA with an analysis of data from EIA Form 923, which collects detailed
information from the electric power sector. The analysis showed that in 2018, 33 percent of electric
power supply occurred in regional transmission organizations in unregulated states.

Energy: Innovation and Independence | 153

250-840_text_.pdf 157

2/7/20 3:46 PM

The approach to estimating natural gas savings, which involves sectorspecific consumption amounts and demand curves, permits calculating savings for the residential, commercial, industrial, and electric sectors, which we
collapse into two sectors: the nonelectric sector and the electric sector. For
oil, we break savings into transportation and nontransportation sector savings, allocating savings to the transportation sector based on its share of total
petroleum consumption in the United States (70 percent) as reported by the
EIA for 2018.
Regarding the pass-through of energy savings to household income
groups, we first allocate residential natural gas and residential electricity
savings based on each income group’s share of spending on natural gas and
electricity, as reported in the 2018 Consumer Expenditure Survey of the Bureau
of Labor Statistics. We then estimate the oil-related transportation sector savings associated with direct household consumption by multiplying the total
oil savings by the share of transportation sector energy use accounted for by
light-duty vehicles such as cars and sport utility vehicles. These direct household savings are then distributed to household income groups based on each
group’s spending on “gasoline, other fuels, and motor oil,” as reported in the
2018 Consumer Expenditure Survey.
Finally, we allocate the natural gas, electricity, and oil-related savings
that initially occur in the commercial and industrial sectors. We assume that
the savings are eventually passed through to households in the form of lower
product prices, with savings allocated to each household income group
according to its share of total household expenditures, as reported in the 2018
Consumer Expenditure Survey. This is a common approach in the literature
on the incidence of carbon taxes, which increase energy prices (Mathur and
Morris 2014). It also has empirical support in important product markets (e.g.,
Muehlegger and Sweeney 2017). The exporting of some of the industrial sectors’ output to global markets would suggest that the approach overstates
savings to U.S. consumers. The shale revolution, however, has also reduced
global energy prices, which would lower the costs of foreign producers, some
of whom serve the U.S. market. We assume that these competing effects offset
each other.

Consumer Savings—Findings
By lowering energy prices, the shale revolution is saving U.S. consumers $203
billion annually, or an average of $2,500 for a family of four. Nearly 80 percent of
the savings stem from a substantially lower price for natural gas, of which more
than half comes through lower electricity prices (figure 4-9). The large decline
in the price of natural gas, and therefore large savings, is because domestic
supply has overwhelmed domestic demand, and the capacity to liquefy and
export natural gas to global markets has expanded too slowly to absorb the
supply growth. Oil, in contrast, is economical to transport and is traded on a
154 |

250-840_text_.pdf 158

Chapter 4

2/7/20 3:46 PM

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

Dollars (billions)
рчп

*/'ѷ
‫ڦ‬рфш
$''$*)
рфп

*/'
))0'
.1$)".ѷ
‫ڦ‬спт
$''$*)
*-

‫ڦ‬сѶфппѶ
!*-
)
1 -" 
!($'4
*!
!*0-



*)Ҋ
' /-$$/4
рсп
. /*-

җ‫ڦ‬цт
$''$*)Ҙ
шп
хп

*)/-).+*-//$*) */'ѷ
‫ڦ‬уу
$''$*)
. /*-
җ‫ڦ‬рт
$''$*)Ҙ

' /-$$/4
. /*-

тп
җ‫ڦ‬чф
$''$*)Ҙ

-).+*-//$*)
. /*-
җ‫ڦ‬тр
$''$*)Ҙ

п

/0-'
g.
$'
*0- .ѷ ) -"4
 )!*-(/$*)
($)$./-/$*)Ѹ
 $'$)
җспрхҘѸ
'a-
 /
'ѵ
җспршҘѸ

'0'/$*).ѵ

Figure 4-10. Total Consumer Savings as a Share of Income by Quintile
Percent
8

6.8
6

3.6

4

2.5
1.9

2

1.3

0
Lowest 20
percent

Second 20
percent

Third 20 percent

Fourth 20
percent

Highest 20
percent

Sources: Bureau of Labor Statistics; CEA calculations.
Note: Values represent the CEA’s estimates of consumer savings as a share of pretax
income in 2018.

Energy: Innovation and Independence | 155

250-840_text_.pdf 159

2/7/20 3:46 PM

Box 4-2. Economic Effects Linked to Drilling and Production
Although much of this chapter focuses on the shale revolution’s effect on
consumers, growth in drilling and production has also brought employment,
income, and public revenues to producing regions and beyond. Relative to
the State of New York’s border counties, which have not had shale development, Komarek (2016) found that counties in the Marcellus region that were
developed had a 6.6 percent increase in earnings. Across the United States,
Feyrer, Mansur, and Sacerdote (2017) estimate that new extraction increased
aggregate employment by as much as 640,000 jobs. In addition to creating
wage-earning opportunities, expanded drilling in places like North Dakota
and Pennsylvania has also brought large payments to landowners holding
rights to subsurface resources. Energy firms typically compensate resource
owners by paying them a share of the value of production from their land. In
2014, production from major shale formations generated nearly $40 billion in
payments to resource owners (Brown, Fitzgerald, and Weber 2016).
Drilling and production can also generate revenue for some State and
local governments and local school districts. Between 2004 and 2013, State
revenues from taxes on oil and gas production in the lower 48 states nearly
doubled, reaching $10.3 billion in real terms (Weber, Wang, and Chomas
2016). At the local level, increases in revenues have largely outweighed costs
for local governments in most producing states (Newell and Raimi 2018). In
certain states, such as Texas, oil and gas wells are also taxed as property and
can therefore provide revenues to local school districts. For example, shale
development in Texas’s oil formations increased the property tax base by over
$1 million per student in the average shale district, leading to 20 percent more
spending per student (Marchand and Weber 2019).

massive global market, which domestic oil production has influenced but not
overwhelmed. As a result, oil accounts for the other 20 percent of the savings,
most of which are transportation sector savings on fuel.
Because lower-income households spend a larger share of their income
on energy bills, the savings have greater relative importance for them. Energy
savings represent 6.8 percent of income for the lowest fifth of households,
compared with 1.3 percent for the highest fifth (figure 4-10). In other words,
lower energy prices are like a progressive tax cut that helps the lowest households the most. The variation in savings stems heavily from differences in
spending on electricity; according to the 2018 Consumer Expenditure Survey,
the bottom 20 percent of households account for 8.6 percent of expenditures
in general but for 14.1 percent of electricity expenditures. We also considered
the economic benefits of increased drilling and production on employment,
income, and public revenues in differing regions as well (box 4-2).

156 |

250-840_text_.pdf 160

Chapter 4

2/7/20 3:46 PM

Energy Independence
Historically, a rise in energy prices increases the trade deficit and costs for
firms and households, sometimes pushing the U.S. economy into a recession.
For example, a sudden rise in the price of oil preceded 10 of the 11 postwar
recessions in the United States (Hamilton 2011). The vulnerability of the U.S.
economy to price shocks motivated a long-standing goal of U.S. Presidents:
U.S. energy independence.
President Richard M. Nixon began the push for energy independence,
announcing Project Independence in 1973 when the Organization of Arab
Petroleum Exporting Countries halted oil shipments to the United States. In
the ensuing years, Congress and the executive branch directed much attention
and resources to pursue energy independence, including the Energy Policy and
Conservation Act (1975), the establishment of the Department of Energy (1977),
the Energy Policy Act (2005), and the Energy Independence and Security Act
(2007).
By a common measure of independence—net exports (Greene 2010)—the
United States essentially achieved independence in both natural gas and oil
at the end of 2019, and net exports are projected to grow in 2020 and beyond.
Today’s achievement, however, does not stem primarily from these government efforts but rather from private sector innovation that few expected. The
shale-driven growth in domestic production documented earlier in this chapter
reduced imports and, most recently, led to a surge in exports of both oil and
gas. Fewer imports and more exports caused U.S. net imports of natural gas to
fall below zero in 2017, making the United States a net exporter of natural gas
for the first time since 1957 (figure 4-11). And, in September 2019, net imports
of crude oil and petroleum products fell below zero on a monthly basis (figure
4-12). The United States is projected to remain a net exporter of crude oil and
petroleum products for all of 2020 for the first time since at least 1949.
Energy independence—as measured by positive net exports, and by
increased sectoral diversification of the U.S. economy, especially in places like
Texas—means that higher global energy prices have a negligible or perhaps
positive effect on the U.S. economy in the aggregate. With a large domestic
energy sector, increases in investment by the domestic energy sector offset the
effect of higher prices on consumers (Baumeister and Kilian 2016). If, for example, higher oil prices induce substantial new investment in drilling wells, with
its associated demands for steel and equipment, GDP would likely increase
as long as the reduced disposable income of consumers has a small effect on
their overall spending (see box 4-2 for an in-depth explanation of the economic
impact of increased drilling and production). This does not mean that the
typical U.S. consumer is unaffected by higher oil prices or benefits from them.
Rather, it means that the country’s total output may expand as prices rise.

Energy: Innovation and Independence | 157

250-840_text_.pdf 161

2/7/20 3:46 PM

$"0- 
уҊррѵ
ѵѵ
*)/#'4
 /
3+*-/. *!
/0-'
.Ѷ
ршшш–спсп
- ' /$*)
) /
 3+*-/.

*./ ' /$*)
) /
 3+*-/.

-*% / 
) /
 3+*-/.
Cubic feet per day (billions)

рп

 Ҋсп

ф
 /
 3+*-/ -

п

 /
$(+*-/ -

Ҋф

Ҋрп

Ҋрф
ршшш

сппт

сппц

спрр

спрф

спрш

*0- .ѷ ) -"4
 )!*-(/$*)
($)$./-/$*)Ѹ

'0'/$*).ѵ

$"0- 
уҊрсѵ
ѵѵ
*)/#'4  /
3+*-/.
*!
-0 
$'
)
 /-*' 0(

-*0/.Ѷ
ршшп–спсп
- ' /$*)
) /
 3+*-/.

*./ ' /$*)
 /
3+*-/.

-*% / 
) /
 3+*-/.

Barrels per day (millions)
у

 Ҋсп
 /
 3+*-/ -

п

 /
$(+*-/ -

Ҋу

Ҋч

Ҋрс

Ҋрх
ршшп

ршшу

ршшч

сппс

сппх

спрп

спру

спрч

*0- .ѷ ) -"4
 )!*-(/$*)
($)$./-/$*)Ѹ

'0'/$*).ѵ

158 |

250-840_text_.pdf 162

Chapter 4

2/7/20 3:46 PM

Figure 4-13. Changes in Price of Oil (Prior Month) and Changes in the
Goods Trade Balance, 2000–2010 and 2011–19
Change in goods deficit (dollars, billions)
30

2000–2010 $1 increase in oil
prices reduces the trade
balance by $0.83 billion

20

10

0

2011–19
$1 increase in oil prices
reduces the trade balance by
$0.17 billion

-10

-20

-30
-30

-20

-10

0

10

20

30

Sources: Energy Information Administration; Wall Street Journal; Census Bureau; CEA calculations.

In addition, if net imports are near zero, large changes in the global price
of oil will have negligible effects on the U.S. trade balance, which directly
affects the country’s GDP (Cavallo 2006). Figure 4-13 shows that over the
2000–10 period, when the United States imported record amounts of oil and
petroleum products, a $1 per barrel increase in the price of oil reduced the
trade balance in goods by $0.83 billion. In the 2011–19 period, which saw falling
net imports, the same price increase reduced the trade balance by only $0.17
billion. As U.S. net exports increase, higher prices should eventually increase
the trade balance, reflecting greater transfers from foreign consumers to
domestic producers.
Energy independence also brings geopolitical benefits, such as more
influence abroad and fewer constraints on foreign policy. The rise of the
United States as a net contributor to the global oil market has reduced oil
prices (Kilian 2016), and has also reduced the dependence of the global market
on particular producers. Currently, the United States has sanctions on two
major oil-producing countries, Iran and Venezuela. These sanctions, combined
with internal factors in the case of Venezuela, have taken millions of barrels
of oil per day off the market. Since the United States announced sanctions
in November 2018, Iranian exports have declined by 1.4 million barrels per

Energy: Innovation and Independence | 159

250-840_text_.pdf 163

2/7/20 3:46 PM

day, an 89 percent decrease from their pre-sanction level; since sanctions on
Venezuela took effect in January 2019, exports have fallen by 0.7 million barrels
per day, a 60 percent decrease. Energy independence increases the feasibility
of such sanctions. In addition, it reduces the incentive to expend foreign policy
resources on efforts to lower global energy prices.
Geopolitical gains also stem from net exports of U.S. natural gas. For
example, exports of U.S. LNG to Europe have and will continue to provide a
diversified source of competitively priced natural gas to reduce the continent’s
dependence on Russian gas supplies. The U.S. share of Europe’s total natural
gas imports increased from 0.1 percent in the first five months of 2018 to 1.3
percent in the first five months of 2019. The potential for greater exports of U.S.
natural gas to Europe gives U.S. leaders greater influence when discouraging
them from supporting the controversial new Nord Stream 2 pipeline project
from Russia to Germany. Poland’s and Lithuania’s leaders are the most recent
heads of state to denounce the project as a threat to energy security that would
increase European dependence on Russian natural gas supplies.

Environmental Benefits
In addition to bringing energy independence and saving the average family of
four $2,500, the shale revolution has brought several environmental benefits.
The shift to generating more electricity from natural gas and renewable energy
sources reduced energy-related carbon dioxide emissions at the national level
to a degree that was not predicted before these innovations. In its 2006 Annual
Energy Outlook, the EIA projected a 16.5 percent increase in carbon dioxide
emissions from 2005 to 2018 (figure 4-14). Actual emissions decreased by about
12 percent.
Actual energy-related carbon emissions for 2018 were 24 percent lower
than projected in 2006. Some of the decline is because projections assumed
greater GDP growth and therefore greater electricity demand than what actually occurred, in part because of the Great Recession and slow recovery. An
important part of the decline, however, stems from lower natural gas prices
reducing reliance on electricity generated from coal. Over the period, the
proportion of generation from coal-fired power plants fell from 50 percent to
28 percent, while the share from natural gas increased from 19 percent to 35
percent.
Low natural gas prices also aided growth in the generation of wind power,
which expanded from less than 1 percent of generation to 7 percent. Although
Federal and State policies, such as renewable portfolio standards and tax
credits, contributed to the increase in wind power generation, Fell and Kaffine
(2018) document the important role of lower natural gas prices in spurring
greater market penetration by wind generation. The complementarity stems
from the ability of natural gas generators to quickly ramp up or slow down in
response to the intermittent wind generation from gusts or lulls in wind.
160 |

250-840_text_.pdf 164

Chapter 4

2/7/20 3:46 PM

Figure 4-14. Actual versus Projected Carbon Dioxide Emissions, 2005–18
Metric tons (millions)
7,500

2018

7,000

Projected

6,500
6,000

Actual

5,500
5,000
2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

Source: Energy Information Administration (EIA).
Note: Carbon dioxide emissions represent total emissions from the consumption of energy as reported by the EIA. Projections are
from the EIA 2006 Annual Energy Outlook.

We estimate that from 2005 to 2018, the shale revolution lowered annual
electric power carbon dioxide emissions by 506 million metric tons, a 21 percent decline relative to electric power sector emissions in 2005 (figure 4-15).
For the estimate, we assume that coal emissions in the electricity sector would
have otherwise remained constant, and we calculate the observed decline in
coal emissions, which is 833 million metric tons. We assume that 92 percent of
the decline is from shale-driven decreases in natural gas prices. This percentage is from Coglianese, Gerarden, and Stock (2019), who estimate the share
of the decline in coal use attributable to the decline in the price of natural gas
relative to the price of coal apart from other factors such as environmental
regulations, which accounted for another 6 percent of the decline.9 Finally, we
subtract the increase in emissions from greater use of natural gas in electricity
generation (506 million metric tons = 833 x 0.92 – 260).10
The shale-driven reduction in electric power emissions is larger than
what the U.S. Environmental Protection Agency (EPA) projected its 2012
9 Note that the decline in coal use and coal emissions is linked to the decline in the price of
natural gas relative to the price of coal, not to the number of coal plants that are replaced with
natural gas plants. Natural-gas-driven changes in electricity prices have caused coal plants to
close, and the retired generation capacity has been replaced with a mix of natural gas plants and
renewable sources. Also, we note that Coglianese, Gerarden, and Stock (2018) look explicitly at
coal production, not consumption, but the two are similar. Over most of their study period, more
than 90 percent of production was consumed domestically.
10 A more detailed analysis could be done to estimate the net greenhouse gas (GHG) effects from
shale gas. For example, the CEA estimate does not include leaks from natural gas wells or pipelines.
According the EPA’s emissions inventory, total GHG emissions from natural gas systems declined
from 2005 to 2017. Alvarez et al. (2018) estimate that emissions are 60 percent greater than what
the EPA reports. Even if this were true for the 2005 and 2017 EPA measurements, emissions from
natural gas systems would have still declined over the period. If emissions were understated in
2017 but not in 2005, the shale-driven declines in emissions would still be larger than those from
the policies mentioned in figure 4-15. In general, innovation in leak detection has lowered leak
rates over time (see box 4-3).

Energy: Innovation and Independence | 161

250-840_text_.pdf 165

2/7/20 3:46 PM

Figure 4-15. Annual GHG Emission Reductions from Shale
Innovation and Major Environmental Policies
Metric tons (millions)
600
500

506
380

400
300

240

200
100
0
Shale innovation

2012 Fuel standards
(projected for 2025)

Clean power plan
(projected for 2025)

Sources: Environmental Protection Agency; Stock (2017); CEA calculations.
Note: The Fuel Standards refer to the 2012 Light-Duty Vehicle Greenhouse Gas Emissions and Corporate
Average Fuel Economy Standards, which applied to the 2017–25 period.

Light-Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel
Economy Standards would achieve in 2025 (380 million metric tons) following
a considerable increase in stringency. The shale reduction is also more than
double what the EPA initially projected that the now-rescinded Clean Power
Plan would achieve by 2025 (240 million metric tons).
The shale-driven decline in emissions allowed the United States to
have a greater rate of decline in total greenhouse gas (GHG) emissions than
the European Union, holding constant the size of the two economies (figure
4-16). From 2005 to 2019, the European Union has developed and expanded an
increasingly stringent cap-and-trade system for GHG emissions across its member countries. Although it substantially raised electricity prices for consumers
(Martin, Muuls, and Wagner 2015), the system helped the European Union
achieve a 20 percent decline in GDP-adjusted emissions from 2005 to 2017, the
most recent year of data. Over the same period, emissions fell by 28 percent in
the United States, which did not implement a national cap-and-trade system,
although various States have pursued policies to cap emissions.
If policymakers had averted the shale revolution through a ban on
hydraulic fracturing or other integral components of shale development,
energy sector GHG emissions would most likely be higher today. Absent low
natural gas prices, renewable electricity sources are unlikely to have enabled
similar emissions reductions. A megawatt-hour of coal-fired electricity generates about 1 metric ton of GHG emissions. Achieving the 506 million metric ton
decline in GHG emissions is roughly equivalent to reducing coal-fired electricity
generation by about 506 million megawatt-hours and replacing it with renewable power generation. This amounts to a 150 percent increase in wind and
162 |

250-840_text_.pdf 166

Chapter 4

2/7/20 3:46 PM

Figure 4-16. U.S. versus EU GDP-Adjusted Carbon
Dioxide Emissions, 2005–17
Metric tons of C𝑂𝑂� per billion dollars of GDP (2005 = 100)
100

United States

2017

90

80

European Union
70

60
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017

Sources: Environmental Protection Agency; Bureau of Economic Analysis; European
Environment Agency; Statistical Office of the European Communities; CEA calculations.
Note: Data are total CO2 emissions per $1 billion (2017) of each region’s GDP.

solar generation above their 2018 level, an increase that is not projected to
happen until the 2040s.11
During the shale era, the percentage decline in coal-fired generation has
roughly equaled the percentage decline in the wholesale price of electricity,
suggesting that prices would need to fall 25 percent below their pre-shale
level to reduce coal generation by 506 million megawatt-hours (25 percent).
This decline would leave wholesale electricity prices about one-third above
their 2018 level. This higher price is unlikely to have supported a 150 percent
increase in wind and solar generation over their 2018 level (and an even larger
percentage increase over their pre-shale level). It implies an elasticity of supply
close to 5, roughly twice as large as the empirical estimate by Johnson (2014).
Shale-driven declines in emissions have been large as well as economical. Many policies seek to reduce emissions. Most of them, however, impose a
cost on the economy. Gillingham and Stock (2018) summarize research on the
cost of reducing a ton of carbon emissions by various methods. They report
that renewable fuel subsidies cost $100 per ton of carbon abated, Renewable
Portfolio Standards cost up to $190 per ton, and vehicle fuel economy standards cost up to $310 per ton. By comparison, shale innovation brings emissions savings without requiring greater public spending (e.g. subsidies) or
costly regulations or mandates.
11 The year 2046 is estimated using the EIA’s 2019 Annual Energy Outlook forecast of wind and solar
generation in the electric power sector through 2050 (EIA 2019c).

Energy: Innovation and Independence | 163

250-840_text_.pdf 167

2/7/20 3:46 PM

Figure 4-17. U.S. versus EU GDP Adjusted Particulate Emissions,
2005–17
Particulate tons per billion dollars of GDP (2005 = 100)
100

2017

90
European
Union

80
70
United States
60
50
2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

2017

Sources: Environmental Protection Agency; Bureau of Economic Analysis; European Environment Agency;
Statistical Office of the European Communities; CEA calculations.
Note: Values are total particulate matter emissions that are 2.5 microns or less in size per billion 2017 U.S.
dollars of each respective region’s GDP. Values are normalized such that 2005 is equal to 100. U.S. emissions
exclude miscellaneous sources.

Lower natural gas prices have also affected emissions of particulates
such as soot, which can affect heart and lung health, especially for those with
asthma or heart or lung disease. As with GHG emissions, GDP-adjusted particulate emissions have declined faster in the United States than in the European
Union over the 2005–17 period (figure 4-17). The difference in the rate of reduction is considerable, with U.S. particulate emissions per $1 of GDP declining by
57 percent and EU emissions declining by 41 percent. The decline has brought
health benefits. Johnsen, LaRiviere, and Wolff (2019) estimate that, as of 2013,
the shale-driven decline in particulate and related emissions had $17 billion in
annual health benefits (see box 4-3).

The Value of Deregulatory Energy Policy
This section explores the value of deregulatory energy policy. First, it shows
how deregulation allows innovation to flourish. Then it explains the private
sector’s part in the critical responsibilities of building and maintaining energy
infrastructure.

Allowing Innovation to Flourish
Government deregulation of natural gas markets—including the 1978 Natural
Gas Policy Act, the Federal Energy Regulatory Commission’s 1985 Open Access
Order, and the 1989 Natural Gas Wellhead Decontrol Act—helped encourage
164 |

250-840_text_.pdf 168

Chapter 4

2/7/20 3:46 PM

Box 4-3. Innovation in Pipeline Leak Detection
Pipelines are one of the most effective methods of transporting oil and gas,
but they require monitoring and maintenance. Traditionally, monitoring has
required that people travel along pipelines by foot, automobile, plane, or
all-terrain vehicle. Innovation in technologies such as drones and advanced
acoustics has allowed the industry to prevent leaks and more quickly find and
stop them when they occur. For example, a Shell pilot drone program illustrates how well-equipped drones can identify pipeline corrosion, abnormal
heat signatures, and any effects on wildlife. This helps the company identify
leaks, but also reveals areas where preventive maintenance is most needed.
With improvements to technology for monitoring pipeline leaks and other
improvements across the supply chain, the leak rate for natural gas and
petroleum systems fell 31 percent from 2005 to 2017 (figure 4-ii).
Figure 4-ii. Methane Production and Leakage Rates, 1990–2017
Metric tons (millions)
650

Leakage rate (percent)

2017

3.0

600
Methane production
(left axis)

550

2.5

500
2.0
450

Leakage rate
(right axis)

400

1.5

350
300
1990

1.0
1997

2005

2013

2014

2015

2016

2017

Sources: Environmental Protection Agency; Energy Information Administration; CEA calculations.
Note: The leakage rate was calculated by assuming that wellhead gas is about 85 percent methane by
volume and assuming that the methane density is 0.0447 pounds per cubic foot.

the innovation that brought the shale revolution. In the same vein, the Trump
Administration has sought to identify and remove regulations that unduly
stifle energy development. This is seen in the Presidential Executive Order on
Promoting Energy Independence and Economic Growth and the Executive
Order on Promoting Energy Infrastructure and Economic Growth. It is also
seen in actions such as permitting for the Keystone XL Pipeline and the DOE’s

Energy: Innovation and Independence | 165

250-840_text_.pdf 169

2/7/20 3:46 PM

approval of a record amount of Liquefied Natural Gas export capacity to non–
free trade agreement countries.
The laboratory of State policy experiments provides examples of contrasting policy approaches and their effects. State governments have the primary responsibility to regulate oil and gas development on non-Federal lands,
specifying where wells can be drilled, how they must be drilled and monitored,
and how they are to be reclaimed at the end of their useful life. Subject to such
regulations, most States allow shale development. Maryland, Vermont, and
New York, however, have banned hydraulic fracturing, a practice integral to
shale development. Of the three States, the New York ban is most consequential because the Marcellus Shale formation, which is the most prolific shale gas
formation in the United States, extends into much of Southern New York. Since
New York’s initial 2010 moratorium on fracking, which morphed into a ban in
2014, energy firms have drilled more than 2,500 wells in Pennsylvania counties
adjacent to the New York border (see box 4-4 for further discussion on the risks
and benefits of shale development).
The difference in energy-related outcomes in the two States is stark.
Development of the Marcellus and Utica Shale in Pennsylvania caused natural
gas production to increase 10-fold from 2010 to 2017. Over the same period,
New York’s production fell by nearly 70 percent. Pennsylvania leads the
country in net exports of electricity to other States and produces more than
twice the amount of energy it consumes. New York, in contrast, has grown
more dependent on electricity generated elsewhere; and in 2017, the State
consumed four times as much energy as it produced.
Despite the growth in energy production in Pennsylvania, total energyrelated carbon dioxide emissions fell 15 percent from 2010 to 2016, the most
recent year of data, twice as much as in New York (7 percent). The greater
decline in Pennsylvania stems from larger reductions in the electric power
sector.
Innovation, however, can create challenges for particular sectors. Despite
substantial and sustained Federal support, including a mid-2000s expectation
of a nuclear renaissance, low wholesale electricity prices have reduced the
profitability of the nuclear power sector. As a result, a wave of early retirements
from existing nuclear power plants has occurred, with more closures planned
in coming years (CRS 2018). Given that changes in the market are impossible
to predict, a diversified research-and-development portfolio for new energy
technologies will best prepare the economy for tomorrow’s market realities.

The Critical Role of Energy Infrastructure
Pipelines, electric transmission lines, and export facilities allow energy
resources to flow from resource-rich places to resource-scarce ones. The
growth in oil and gas supply documented above increases demand for pipelines. For example, with a dramatic rise in production over the last decade,
166 |

250-840_text_.pdf 170

Chapter 4

2/7/20 3:46 PM

Box 4-4. Shale Development and Local Communities
Many academic studies have explored the effects of shale oil and gas development on nearby communities. Two studies estimate measures of local net
benefits across all major shale regions and reach a similar conclusion: on
average, local wage and income effects from development exceed increases
in living costs or deterioration in local amenities (Bartik et al. 2019; Jacobsen
2019). Jacobsen (2019) finds that wages across all occupations increased
in response to the growth in drilling, regardless of whether they had direct
links to the oil and gas industry. Similarly, Bartik and others (2019) estimate
that shale development generated $2,500 in net benefits to households in
surrounding communities.
It is also evident that local effects can vary greatly, which is illustrated in
the diverse effects of development on housing values. Housing values reflect
an area’s standard of living, including earnings opportunities and amenities,
such as good roads. Shale development affects both, creating jobs but also
truck traffic and associated disamenities, particularly during times of drilling
(Litovitz et al. 2013; Graham et al. 2015). In addition, development, when
poorly managed, can pose a risk to groundwater and health, and improper
disposal of wastewater can induce earthquakes when best management practices are not followed (Darrah et al. 2014; Keranen et al. 2014; Wrenn, Klaiber,
and Jaenicke 2016; Hill and Ma 2017; Currie, Greenstone, and Meckel 2017).
Development has had large, positive effects on average housing values over
time in many places (Boslett, Guilfoos, and Lang 2016; Weber, Burnett, and
Xiarchos 2016; Bartik et al. 2019; Jacobsen 2019). Drilling itself, however, has
depressed property values, at least temporarily, for groundwater-dependent
homes in Pennsylvania or properties without mineral rights in Colorado
(Muehlenbachs, Spiller, and Timmins 2015; Boslett, Guilfoos, and Lang 2016).
Welfare effects can also vary across households in shale areas based on the
value that households place on greater earning opportunities relative to
disamenities, such as noise and congestion.
The nuisances and risks that can come with drilling and fracturing
wells highlight the value of prudent State and local policies that match local
realities, safeguard the environment and human health, and allow private
landowners to contract with energy firms to bring valuable energy resources
to market. Almost all major producing States have revised oil and gas laws to
address hydraulic fracturing and shale development more generally. North
Dakota, for example, adopted rules limiting the flaring of natural gas in 2014,
a practice that is especially common in the State because oil producers there
have limited infrastructure to deliver to market the natural gas that accompanies oil production. Similarly, as shale development grew in Pennsylvania,
the State adopted a policy that effectively ended the treatment of fracking
wastewater at publicly owned treatment plants, which were shown to be
poorly equipped to properly treat the water.

Energy: Innovation and Independence | 167

250-840_text_.pdf 171

2/7/20 3:46 PM

Pennsylvania has switched from being a major importer of natural gas to being
a major exporter. Acquiring regulatory approval and building the necessary
pipelines has taken time, progressing to completion in some places but not
others.
In 2017 and 2018, private firms finished two major pipeline projects, the
Rover and Nexus pipelines, to take Appalachian gas into Michigan and beyond,
with the projects adding nearly 1,000 miles of pipeline and 3.2 billion cubic feet
of gas per day of capacity. The first phase of the Rover pipeline was finished in
August 2017 and ran from Southeastern Ohio (near the Pennsylvania border) to
Northwestern Ohio (near the Michigan border). The second phase was finished
in May 2018 and extended the pipeline through Michigan and into Canada. The
Nexus pipeline was also completed in 2018 and follows a similar route, eventually connecting with existing pipelines near Detroit.
No new interstate pipelines were built from Pennsylvania into New York
(and therefore into New England) over the same period. Total expansions
or extensions of existing pipelines that transit New York totaled 21 miles in
length and 0.46 billion cubic feet per day in additional capacity. The 125-mile
Constitution Pipeline, which would take Pennsylvania gas to New York and
beyond, has been repeatedly delayed since the project’s inception in 2012,
with a major source of delay being the refusal of the New York Department of
Environmental Conservation to grant a necessary certification.
Natural gas price differences across States and over time illustrate the
implications of new investments in pipelines. As natural gas production grew
in Pennsylvania, Ohio, and West Virginia, citygate prices in Michigan fell relative
to the national average price, plausibly reflecting the benefit of being closer to
a place of burgeoning supply growth. (The citygate price measures local wholesale natural gas prices). From 2016 to 2018, when two main pipeline projects
were being completed, the Michigan price relative to the national average price
fell 14 percent. The New York price went in the opposite direction, increasing
by 16 percent, potentially reflecting the interaction between high demand
(from an above-average number of cooling-degree days in 2018) and pipeline
constraints (figure 4-18).
The 14 percent decline in the Michigan citygate price relative to the
national price provides a credible estimate of the price effect of expanded
pipeline capacity. It is similar to estimates of the effect of major capacity
expansions (Oliver, Mason, and Finnoff 2014) or the price premium associated
with insufficient capacity (Avalos, Fitzgerald, and Rucker 2016).
A 14 percent decline in the New York and New England citygate price
would save consumers in the region an estimated $2.0 billion annually, or
$233 for a family of four. Some of the savings would be from residential, commercial, and industrial consumers paying less for the natural gas that they
consume, but the bulk of savings would be from lower electricity prices. New
York and most of New England have deregulated electricity markets, where
168 |

250-840_text_.pdf 172

Chapter 4

2/7/20 3:46 PM

$"0- 
уҊрчѵ
$/4"/ 
/0-'
.
-$ . $)
$#$")
)
 2
*-&
 '/$1 

/*
/$*)'
1 -" 
-$ .Ѷ
сппф–рч
*1 +#. 
с
State price / national average price (dollars)
*(+' /
рѵс

*1 +#. 
р
*(+' /

 2
*-&

спрч

рѵр

рѵп

$#$")

пѵш

пѵч
сппф сппх сппц сппч сппш спрп спрр спрс спрт спру спрф спрх спрц спрч
*0- .ѷ
) -"4
 )!*-(/$*)
($)$./-/$*)Ѹ

'0'/$*).ѵ
*/ ѷ
# *1 -
$+ '$) 
+#. 
р
2.
*(+' / 
$)
0"0./
спрц,
)
+#. 
с
2.
*(+' / 

$)
4
спрчѵ
 -/$'
-.
- +- . )/ /# 
 "$))$)"
*!
/# 
4 -
2#en
/# 
+$+ '$) 
2.

*(+' / Ѷ
"$1 )
))0'
/ѵ

electricity-generating firms sell into competitive markets. Linn, Muehlenbachs,
and Wang (2014) find that for New York and New England, a 1 percent decrease
in the price of natural gas lowers the price of electricity by 0.8 percent. Applying
this gas-driven decline in wholesale prices to the region’s consumption of electricity in 2018 provides $1.2 billion of the total $2.0 billion in savings.
Other infrastructure investments could provide similar value. The Atlantic
Coast Pipeline, for example, would take natural gas from West Virginia to North
Carolina, where citygate prices have been about 10 percent higher than in West
Virginia in 2019. We also note that pipelines are not the only means of transporting natural gas domestically. The Pipeline and Hazardous Materials Safety
Administration recently approved a permit request to transport LNG by rail.
Just as pipelines allow producers to reach high-price markets in other
states, facilities for exporting LNG allow U.S. producers‒whose production
now exceeds domestic consumption‒to reach high-price markets abroad. In
response, export volumes have surged, averaging 4.7 billion cubic feet per day
(Bcf/d) in the first 10 months of 2019, compared with less than 2 Bcf/d in the
first 10 months of 2017. Under the Natural Gas Act, exports of LNG must be
approved by the DOE on the basis of whether the exports are consistent with
the public interest. Under the Trump Administration, the DOE has doubled the
volume of LNG approved for export, increasing capacity from 17 Bcf/d to more
than 34 Bcf/d as of October 2019.

Energy: Innovation and Independence | 169

250-840_text_.pdf 173

2/7/20 3:46 PM

Conclusion
The shale revolution provides a striking example of the potential of private
sector energy innovation and the resulting implications for consumers and the
environment. In less than a decade, productivity in oil and gas extraction has
increased several-fold. As a result, production costs have fallen, making energy
goods and services more affordable for consumers, especially lower-income
households. By several measures, the shale revolution has led to greater environmental progress in the United States than in the European Union, which
exercises more government control and has more stringent emissions policies.
The Trump Administration’s deregulatory policies aim to support private
sector innovation and initiative by reducing excessively prescriptive government regulation. In doing so, the Administration seeks to further unleash the
country’s abundant human and energy resources. This policy stance is consistent with the approach taken by most States, which have allowed shale production to flourish as long as companies meet updated State policies that limit
risks to human health and the environment. However, some States have taken
a more command-and-control approach, which has had predictable effects.
In particular, New York State has taken an alternative, unsafe-at-any-speed
approach to shale development. As it has done so, its natural gas production
has fallen, its imports of electricity have increased, and its rate of GHG emissions reduction has been less than that of neighboring Pennsylvania.
State and Federal policy questions related to shale will persist in debates
about environmental and energy policy. The shale revolution will continue
to influence energy prices because the private sector has shown that large
amounts of oil and gas can be extracted from shale and similar formations at
moderate prices. The knowledge and capability gained from innovation will
remain through periods of low energy prices that drive overleveraged firms
into bankruptcy. In addition, policies that would severely constrain use of
this capability come with large, forgone benefits—in large part the consumer
savings and environmental gains documented in this chapter. The Trump
Administration’s deregulatory energy agenda, in contrast, seeks to overcome
government barriers to private sector innovation that lowers energy prices and
benefits the environment.

170 |

250-840_text_.pdf 174

Chapter 4

2/7/20 3:46 PM

x
Chapter 5

Free-Market Healthcare Promotes
Choice and Competition
Driven by unparalleled medical innovation, the American healthcare system
remains the envy of the world. However, its past success does not mean that
healthcare in the United States always delivers the value that it should. Costs
for many procedures and medications are too high, access to the healthcare
that patients demand is limited, and competition is lacking. But these challenges do not mean that the only solution is increased government intervention. These improvements can be accomplished by enhancing healthcare
choice and competition in ways that embrace the value of the market while
focusing on patients’ needs.
The Trump Administration has already made major progress in delivering
high-quality, lower-cost healthcare by creating more choice in health insurance
markets and more competition among healthcare providers. In other words,
it is possible to keep what works and fix what is broken. For example, the
Administration has sought to make healthcare more affordable by lowering
out-of-control prescription drug prices and expanding access to more affordable healthcare options. Additional policy changes put patients in control of
their healthcare by ensuring price transparency and allowing Americans to pick
the care that fits their needs. At the same time, accelerating medical innovation
has provided new treatment options for patients living with disease.
Under the Trump Administration, the Food and Drug Administration approved
more generic drugs than ever before in U.S. history and updated its approval
process for new, lifesaving drugs. This past year, prescription drug prices
experienced the largest year-over-year decline in more than 50 years. Whether

171

250-840_text_.pdf 175

2/7/20 3:46 PM

it is through reforms that seek to expand association health plans, promote
health reimbursement arrangements, or give terminally-ill patients access to
potentially lifesaving drugs, among many other successes, every healthcare
reform that lowers costs and increases quality allows American workers to live
longer, healthier lives and keep more of their paychecks.
The Administration’s focus on consumer-centric health policies will make the
healthcare marketplace more competitive and protect as well as enable consumers to obtain life-enhancing technologies. For example, the Administration’s
recent policy change to permit insurers to offer policies with additional benefits
covered before a deductible is met and allow enrollees to maintain health
savings accounts are real changes already helping those with preexisting conditions. And with future changes under way to enable patients using the real price
for major medical services, the effect of the free market to lower health care
costs for all consumers has just begun.
Healthcare regulations at all levels of government can increase price, limit
choice, and stifle competition—which, in combination, lead U.S. healthcare
to fail to provide its full value. These regulations can also harm the broader
economy. For example, the Affordable Care Act has impeded economic recovery by introducing disincentives to work. The Trump Administration’s successes
in addressing these policies over the past three years show the value of empowering the market to deliver the affordable healthcare options that Americans
rightly expect. Further patient-centered reforms will provide Americans with
improved healthcare through enhanced choice and competition.

T

he United States’ healthcare system relies more on private markets to
provide health insurance and medical care than do those of other countries. And the U.S. system is supplemented by public sector programs
to finance the care of vulnerable populations, which include low-income and
senior populations. Most Americans are in employer-sponsored group health
plans and are often satisfied with the insurance coverage and medical care they
receive. However, the U.S. system does not always deliver the value it should.
Market competition leads to an efficient allocation of resources that should

172 |

250-840_text_.pdf 176

Chapter 5

2/7/20 3:46 PM

lower prices and increase quality. But every market has features that deviate
from optimal conditions, and healthcare is no exception. Last year (CEA 2019),
we discussed obstacles in healthcare markets and concluded that they are not
insurmountable problems that mandate the government’s intervention.
This chapter identifies government barriers on the Federal and State
levels to healthcare market competition that lead to higher prices, reduce
innovation, and hinder quality improvements. The chapter proceeds with a
review of barriers to competition and choice, and then it provides a summary
of the accomplishments and expected effects of Administration health policy
in reducing these impediments and creating competitive innovation in the
healthcare markets for all Americans. The Administration’s reforms aim to foster healthcare markets that create value for consumers through the financing
and delivery of high-quality and affordable care. Government mandates can
reduce competitive insurance choices and raise premiums.
By focusing on choice and competition, the Administration is encouraging States to provide flexibility to develop policies that accommodate
numerous consumer preferences for healthcare financing and delivery. The
Administration has addressed these problems through a series of Executive
Orders, deregulatory measures, and signed legislation. By 2023, we estimate
that 13 million Americans will have new insurance coverage that was previously unavailable due to high prices and overregulation.1

Building a High-Quality Healthcare System
A key goal for the healthcare marketplace is to provide effective, high-value
care to all Americans. Achieving this goal requires careful consideration and
revision of specific Federal and State regulations and policies that inhibit
choice and competition. This section identifies two ways to increase choice
and competition: creating more choice in health insurance markets, and creating more competition among healthcare providers.

Creating More Choice in Health Insurance Markets
The majority of Americans obtain health insurance coverage through private
sector, employer-sponsored group plans and other private (individual or nongroup) plans (see figure 5-1). The public sector Medicaid program provides coverage to people with low incomes, while Medicare provides coverage to older
Americans. Figure 5-1 shows the percentages of Americans that have various
1 The CEA previously released research on topics covered in this chapter. The text of this chapter
builds on the 2019 Economic Report of the President; the CEA report “Measuring Prescription Drug
Prices: A Primer on the CPI Prescription Drug Index” (CEA 2019c); the CEA report Mitigating the
Impact of Pandemic Influenza through Vaccine Innovation (CEA 2019d); the report “Reforming
America’s Healthcare System through Choice and Competition,” from the Department of Health
and Human Services (HHS 2018); and policy announcements from the Executive Office of the
President.

Free-Market Healthcare Promotes Choice and Competition | 173

250-840_text_.pdf 177

2/7/20 3:46 PM

Figure 5-1. Health Insurance Coverage by Type of Insurance, 2018
Employment based

55.1

Marketplace

3.3

Other private

10.1

Medicare

17.8

Medicaid

17.9

Other public

1.0

Uninsured

8.5
0

10

20

30

40

50

60

Percentage of Americans covered
Sources: Census Bureau; CEA calculations.
Note: Numbers do not sum to 100 percent due to overlap for individuals with multiple health
insurance plans. Other private plans include nongroup, direct-purchase plans, and TRICARE. Other
public plans include veterans health insurance. Blue indicates private health insurance plan types,
and red indicates public health insurance plan types.

types of health insurance coverage, but many people have multiple coverage
sources; for instance, many older adults on Medicare purchase private supplemental insurance plans. In 2018, more than 67 percent of all Americans were
covered by private health insurance plans, while just over 34 percent were
covered by public plans. Among the insured population, 12.2 percent had more
than one type for all of 2018 (Census Bureau 2019). Employer-sponsored insurance dominates most of the private health insurance market. The individual
insurance market accounts for a smaller share of the insured population. In
the individual market, consumers buy their insurance through the insurance
exchanges established by the Affordable Care Act (ACA) or through ACAcompliant individual policies.
Since earlier in the 2000s, when private health insurance premiums grew
rapidly, growth rates have moderated, especially since 2017 (Claxton et al.
2019). Figure 5-2 shows the inflation-adjusted growth in the average premium
for family coverage through employer-sponsored group plans. The total premium is paid partly through the employer contribution and partly through the
employee contribution. We focus on the total premium because health economists agree that, ultimately, employees also pay the employer-contribution
in the form of reduced wages. In the individual insurance market, after the
Affordable Care Act established health insurance exchanges, the premiums
almost doubled in the first few years. From 2018 to 2019, the benchmark ACA
premiums dropped by 1.5 percent. From 2019 to 2020, the benchmark ACA
premiums dropped by an additional 4 percent (CMS 2018, 2019).

174 |

250-840_text_.pdf 178

Chapter 5

2/7/20 3:46 PM

Figure 5-2. Annual Change in Average Family Premium Including
Employee and Employer Contributions, 2000–2018
Percent change (year-over-year)
12

2018

10
8
6
4
2
0
2000

2002

2004

2006

2008

2010

2012

2014

2016

2018

Sources: Kaiser Family Foundation’s Employer Benefits Survey; CEA calculations.

Recent health policy changes at the Federal and State levels have sought
to give consumers more control over their medical expenditures so they can
seek greater value for their health investment. Two of the best illustrations
of these consumer-focused policies are health saving accounts (HSAs) and
health reimbursement arrangements (HRAs). As described in the Department
of Health and Human Services’ (HHS) report “Reforming America’s Healthcare
System through Choice and Competition,” the promotion and expansion of
these policies, combined with price and quality transparency initiatives, will
encourage consumers to make better and more informed care choices to
enhance their health (HHS 2018).
“Consumer-directed health plans” (CDHPs) is an all-encompassing term
for HRAs, HSAs, and similar medical accounts that allow patients to have
greater control over their health budgets and spending. The growth of CDHPs
has been substantial, especially by large employers that offer these highdeductible plans, HRAs, and HSAs in a larger strategy to introduce consumerism in employer-sponsored health insurance. HRAs allow employees to shop
in the individual market for their preferred plans. Expanding consumer choice
in health plans decreases the deadweight loss associated with poor plan
matching and leads to gains in consumer surplus (Dafny, Ho, and Varela 2013).
HSAs may be especially attractive to consumers because they may be used for
nonmedical healthcare expenses and are portable (Greene et al. 2006). In an
analysis of firms that completely replaced traditional managed care plans with

Free-Market Healthcare Promotes Choice and Competition | 175

250-840_text_.pdf 179

2/7/20 3:46 PM

Figure 5-3. Percentage of Covered Workers Enrolled in a Plan with a
General Annual Deductible of $2,000 or More for Single Coverage,
2009–19
Percentage of covered workers
50

40

2019
All small firms

30

All firms
20

All large firms

10

0
2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

Source: Kaiser Family Foundation’s Employer Benefits Survey.
Note: Small firms have 3 to 199 workers, and large firms have 200 or more workers.

CDHPs for their employees, Parente, Feldman, and Yu (2010) saw significant
decreases in total healthcare costs, though they were inconsistent among firms
that offered different mixes of HRAs and HSAs. CDHPs may also be beneficial for
low-income families and high-risk families, where total health spending significantly decreased for vulnerable (low-income or high-risk) families with CDHPs
(Haviland et al. 2011). Healthcare costs are also lower for employers offering
CDHPs, whose costs in the first three years after a CDHP is offered are significantly lower relative to firms that do not offer a CDHP (Haviland et al. 2016).
As seen in figure 5-3, the share of individuals enrolled in high-deductible
health plans in the employer-sponsored health insurance market has risen
substantially. This has led consumers to have greater incentives to shop for
medical services that are not reimbursed before their deductible is met.
Although the growth of CDHPs has increased out-of-pocket medical
expenses on average, the plans are available with significantly lower premiums than other health insurance choices, as seen in figures 5-4 and 5-5.
Furthermore, with the Administration’s new options to cover predeductible
care for the chronically ill with little to no out-of-pocket expense, as discussed
later is this chapter, more choices are available for more vulnerable populations than before 2016.

176 |

250-840_text_.pdf 180

Chapter 5

2/7/20 3:46 PM

Figure 5-4. Average Annual Worker and Employer Premium
Contributions for Single Coverage, 2019
Employer contribution

Worker contribution

Premium contributions (dollars)
8,000

7,000

1,454

1,058

1,072

6,000

5,000

1,242
1,071

6,222

6,180

6,112
5,341

5,946

4,000
HMO

PPO

POS
HDHP/SO
Type of health insurance

All plans

Source: Kaiser Family Foundation’s Employer Health Benefits Survey.
Note: HMO = health maintenance organization; PPO = preferred provider organization; POS = pointof-service plan; HDHP/SO = high-deductible health plan with savings option.

Figure 5-5. Average Annual Worker and Employer Premium
Contributions for Family Coverage, 2019
Employer contribution

Worker contribution

Premium contributions (dollars)
25,000
21,000
17,000

6,009

6,638
6,945

4,866

6,015

13,000
9,000

14,688

15,045

HMO

PPO

12,894

14,114

14,561

5,000
POS
HDHP/SO
Type of health insurance

All plans

Source: Kaiser Family Foundation’s Employer Health Benefits Survey.
Note: HMO = health maintenance organization; PPO = preferred provider organization; POS = pointof-service plan; HDHP/SO = high-deductible health plan with savings option.

Free-Market Healthcare Promotes Choice and Competition | 177

250-840_text_.pdf 181

2/7/20 3:46 PM

Creating More Competition among Healthcare Providers
Recent studies of variation in health service pricing suggest that the market
lacks needed competition. If competition is reduced among providers (e.g.,
physicians or hospitals), and in addition there is no change in patient demand,
then higher prices and fewer choices are likely to result. These can also lower
overall healthcare quality and limit the efficient allocation of resources.
Government policies can diminish competition by adversely limiting the supply
of providers and the scope of services they offer.
Choice and competition can be limited by State policies that restrict
entry into provider markets. This, in turn, can stifle innovation that could
lead to more cost-effective care provision. Higher healthcare prices and fewer
incentives for quality improvement by providers can be the results of these
market-stifling State policies. In particular, state-specific certificate-of-need
laws could reduce provider access and create unnecessary monopoly pricing
where there is limited competition. In chapter 6 of this Report, we discuss
advocacy efforts by the Trump Administration to limit the harmful effects of
certificate-of-need regulation.
Since the 1990s, markets for a variety of healthcare services have become
more consolidated (NCCI 2018). Some consolidation involves cross-market
mergers—as, for example, when hospitals operating in different regions form
a system—but there is also evidence of increasing concentration in local
markets. As discussed in chapter 6, the Federal Trade Commission (FTC) and
the Department of Justice’s (DOJ) Antitrust Division classify markets using the
Herfindahl-Hirschman Index (HHI). Between 1990 and 2006, the proportion
of metropolitan statistical areas (MSAs) with hospital market HHIs classified
as “highly concentrated” (i.e., with an HHI above 2,500) rose from 65 percent
to more than 77 percent (Gaynor, Ho, and Town 2015). Concentration has
also risen significantly in health insurance markets. Even when consolidation
occurs between close competitors, consumers can benefit from substantial
efficiency gains.
However, the trends of rising concentration have properly drawn attention to the question of how consumers are affected. A recent but growing
body of literature has linked increasing concentration in hospital markets to
rising prices, markups, and falling quality. A number of studies have found that
mergers between hospitals that are close competitors leads to significantly
higher prices without improving quality (Vogt and Town 2006; Gaynor and
Town 2012), or in settings with regulated prices, to lower quality (Kessler and
McClellan 2000; Cooper et al. 2011). This literature is still young, and more
needs to be done, particularly to assess what is driving the consolidations.
Fuchs (1997) argued that the rise of health maintenance organizations is a
contributing factor, as hospitals seek to offset the bargaining power of large

178 |

250-840_text_.pdf 182

Chapter 5

2/7/20 3:46 PM

insurers by becoming large themselves; but as discussed by Gaynor, Ho, and
Town (2015), the empirical evidence for this is mixed.
More generally, it is important to understand if rising concentration is
associated with factors, such as rising fixed-cost investments or economies of
scale, that may benefit consumers. This causality issue is discussed in chapter
6. At a minimum, however, these results suggest that market structure is an
important aspect of healthcare markets.
Consolidation is also seen in the prescription drug market. The growth in
importance of pharmacy benefit managers (PBMs) to serve as intermediaries
between drug manufacturers and health insurers also increased the size of
the largest PBMs, their purchasing power, and their ability to obtain rebates
and discounts from manufacturers (Aitken et al. 2016). PBMs are resistant to
list drug price increases, as their profits are usually a percentage of drug list
prices—thus, there is little incentive to reduce the amount charged to insurers.
As discussed later in this chapter, the three largest PBMs hold 85 percent of
market share.
One way to gauge the uneven competition among healthcare providers
is to examine the degree of competition (or lack thereof) in major metropolitan
markets. Data made available by the Health Care Cost Institute (HCCI 2016)
used negotiated provider price data to illustrate the degree of lack of competition present in the market at the national and regional levels. Using data from
HCCI, Newman and others (2016) examined variations in the negotiated rates
of providers from 242 possible medical services. They calculated the ratio of
the average price paid in each State to the average national price for a given
medical service by ratio categories for each of the 242 services. Figure 5-6 presents a map depicting variation in cataract surgery prices by state.
The map illuminates both regional patterns and variations among Statelevel average cataract removal prices. For example, Iowa, Illinois, and Indiana
all have prices between 125 and 150 percent of the national average price.
Alternatively, across four States in the Southeast, the ratio of State average
price to national average price decreases from 150 through 175 percent in the
Carolinas to a ratio of less than 75 percent in Florida.
Kansas and New York have prices close to the national average price
for cataract surgery, at $3,382 and $3,678, respectively, compared with
$3,541 (HCCI 2016). However, the average prices in the neighboring States of
Nebraska and Connecticut are $957 and $1,181 more. With respect to knee
replacements, New Jersey and Kansas have the lowest average prices; and
Washington, Oregon, and South Carolina have the highest average prices.
Prices in Connecticut and Iowa are about the same as the national average
price of roughly $36,000. The data show that Arizona, Texas, Rhode Island, and
West Virginia have the lowest average prices for a pregnancy ultrasound, while
Oregon, Wisconsin, and Alaska have some of the highest average prices.

Free-Market Healthcare Promotes Choice and Competition | 179

250-840_text_.pdf 183

2/7/20 3:46 PM

Figure 5-6. Ratio of State Average Price to National Average Price of
Cataract Removal, 2015

Sources: Newman et al. (2016); Health Care Cost
Institute.

Insufficient data
Less than 75%
75% to 100%
101% to 125%
126% to 150%
151% to 175%
Greater than 175%

Although the national average price for a knee replacement is more than
100 times larger than a pregnancy ultrasound, there is greater variation in average prices for ultrasounds. For example, in South Carolina, the average knee
replacement price is more than 30 percent higher than the national average,
while in Wisconsin the average pregnancy ultrasound is more than 220 percent
greater than the national average. This suggests that relative to the average
price, there are higher high prices and lower low prices among the pregnancy
ultrasound prices. Much of this variation could be due to the lack of transparency in shoppable services to create a truly competitive market.
There is also variation within regions or States in price trends. HCCI
(2016) also calculated the ratio of each State’s average price relative to the
national average price for each medical service. The percentages of services
within eight ranges of ratios were then graphed for each state (Newman et al.
2016). Figure 5-7 provides a visual representation of the distribution of all care
medical services and can be compared across States.
Figure 5-7 shows the distribution of prices for four States: Florida, Ohio,
Connecticut, and Minnesota. Of the 241 care bundles calculated for Florida,
the prices for 95 percent of them were at or below the national averages. Ohio,
with 240 care bundles, had higher prices on average than Florida; but roughly
75 percent of all prices were at or below the national averages. Connecticut,
with 232 care bundles estimated, on average had higher prices than Florida
and Ohio, with 30 percent of its care bundle prices being at least 20 percent

180 |

250-840_text_.pdf 184

Chapter 5

2/7/20 3:46 PM

Figure 5-7. Distribution of Average State Price Relative to Average
National Price of Care Bundles in Four States, 2015
Florida

Ohio

Connecticut

Minnesota

Percentage of care bundles
80
70
60
50
40
30
20
10
0

≤ 60

61–80

81–100
101–20 121–40 141–60
Percentage of average national price

161–80

> 181

Source: Health Care Cost Institute.
Note: Price data for Florida include 241 care bundles; for Ohio, 240 bundles; for
Connecticut, 232 bundles; and for Minnesota, 221 bundles.

higher than the respective national averages. Minnesota, with 221 estimated
care bundles, had the highest prices on average, with more than 45 percent of
the care bundles having prices 50 percent or more above the national average.
Table 5-1 presents the highest average and lowest average price for a
knee replacement reported for a metropolitan statistical areas in 12 States.2
Sacramento has the highest average price ($57,504)—more than twice as high
as Tucson, Miami, Saint Louis, Syracuse, Toledo, Allentown, Knoxville, and
Lubbock. California also has the largest within-State difference in average
price ($27,243) across any paired set of MSAs in the State. Though the two
California markets are 440 miles apart, it is worth noting that a three-hour
drive from Palm Bay, Florida, to Miami could potentially save $17,122 on knee
replacement surgery—a difference of roughly $100 per mile driven—assuming
one’s insurance plan design covered the individual in both locations. Absolute
dollar differences across MSAs were small in Connecticut, South Carolina, and
Virginia for the MSAs for which we had sufficient data to calculate prices.
These findings demonstrate that there is wide geographic variation in
prices within the privately insured population. Although some of the variation
may be a result of the differences in the costs of doing business (e.g., supplies,
2 These are indicative differences because prices could not be calculated for every MSA in a State.
There could have been higher or lower prices in an unreported MSA in a State. These reported
prices should drive inquiries into why these differences exist and whether any differences are
justified by local differences or other evidence.

Free-Market Healthcare Promotes Choice and Competition | 181

250-840_text_.pdf 185

2/7/20 3:46 PM

Table 5-1. Variation in Knee Replacement Prices across MSAs within
States, 2015
Number
of MSAs

State

Difference between Distance
Highest MSA- Lowest MSAhighest and lowest between
level average level average
MSA-level average MSA cities
price (dollars) price (dollars)
(miles)
price (dollars)

Arizona

2

28,264

21,976

6,288

116

California

6

57,504

30,261

27,243

440

Connecticut

3

37,417

33,594

3,823

39

Florida

8

44,237

27,115

17,122

173

Missouri

2

26,601

23,114

3,487

248

New York

4

36,584

24,131

12,453

247

Ohio

7

34,573

24,491

10,082

203

Pennsylvania

3

33,338

27,188

6,150

62

South Carolina

2

46,591

43,635

2,956

103

Tennessee

2

34,895

26,291

8,604

180

Texas

5

45,275

28,456

16,819

345

Virginia

2

39,298

39,292

6

107

Source: Health Care Cost Institute.
Note: MSA = metropolitan statistical area.

wages, and rent), the remaining variation could be attributable to other factors, such as a lack of transparency, market power, or alternative treatments.
A patient-centered healthcare policy’s goal would be the least unjustified price difference as possible and a low average price for a service. For
example, Arizona has the sixth-largest price difference ($123) in the pregnancy
ultrasound prices—a service that should be similar in scope and quality across
providers, care settings, cities, and States. The average of the average prices
paid in Tucson and Phoenix is the lowest ([$320 + $197] / 2 = $258.5).
To address how competition can lower prices more broadly, the
Administration’s report “Reforming America’s Healthcare System through
Choice and Competition” outlined many other important measures to increase
competition for the entire healthcare sector, including hospitals and doctors,
which make up the bulk of total spending. For example, a recent Executive
Order set the way for increasing price transparency in healthcare, which allows
competition to more effectively operate.

182 |

250-840_text_.pdf 186

Chapter 5

2/7/20 3:46 PM

Healthcare Accomplishments under
the Trump Administration
Since the beginning of his Administration, President Trump has sought to make
healthcare more affordable by lowering prescription drug prices and making
new, affordable healthcare options available. Policies have been advanced to
provide transparency and choice so patients can choose the care that fits their
needs. In addition, pathways have been sought to unleashing American innovation that will provide new treatment options for patients living with disease.
To increase choice, the Administration has increased insurance options and
reduced the regulatory burden. To increase competition, the Administration
has focused on three major areas: (1) accelerating innovation, (2) increasing
access to valuable therapies, and (3) making the health market stronger with
greater transparency. Efforts in each of these areas are discussed in this section, with the goal of setting out how to keep what works and fix what is broken.

Increasing Choice
This subsection addresses a number of key aspects of how to increase choice.
These include reducing regulatory burdens, stabilizing health insurance
exchanges, lowering the individual mandate penalty to zero, encouraging
State innovation in insurance design, expanding association health plans and
short-term limited-duration insurance, strengthening Medicare, expanding
health reimbursement arrangements, and modernizing high-deductible health
plans.
Reducing regulatory burdens. In our 2019 Report, we estimated the impact
of deregulated health insurance markets to provide more plan competition
and choice for small businesses and American consumers through expanding
association health plans and short-term, limited duration plans. These deregulations, in addition to eliminating the individual mandate, were estimated to
generate $450 billion in benefits over the next decade. We estimated that the
reforms will benefit lower- and middle-income consumers and all taxpayers
but will impose costs on some middle- and higher-income consumers, who will
pay higher insurance premiums. The benefits of giving a large set of consumers more insurance options will far outweigh the projected costs imposed on
the smaller set who will pay higher premiums. In 2019, we provided estimates
supporting the claim that these reforms do not “sabotage” the ACA but rather
provide a more efficient focus of tax-funded care for those in need.
Stabilizing health insurance exchanges. In April 2017, HHS issued a final
rule aimed at stabilizing the exchanges. Among other provisions, this rule made
it more difficult for consumers to wait until they needed medical services to
enter the exchanges. This limits gaming of the program and the driving up of
premiums for those who maintain continuous coverage.

Free-Market Healthcare Promotes Choice and Competition | 183

250-840_text_.pdf 187

2/7/20 3:46 PM

The 2019 HRA rule is expected to cause a significant increase in individual
market enrollment in the early 2020s. The rule is projected to do so through
additional choice and market competition and without any new government
mandates. Younger and healthier employees may be more likely to prefer the
typical individual market coverage of relatively high deductibles and more limited provider networks due to their lower premiums, so it is possible that the
HRA rule could lead to an improved individual market risk pool (Effros 2009).
This would occur if the HRA rule generates greater demand in the individual
market and from younger and older workers, given the relative attractiveness
of lower premium cost generated by the HRA contribution to the employee
when they purchase insurance.
Lowering the individual mandate penalty to zero. In December 2017,
President Trump signed the Tax Cuts and Jobs Act, which set the ACA’s individual mandate penalty to zero. This benefits society by allowing people to choose
not to have ACA-compliant health coverage without facing a tax penalty, and
by saving taxpayers money if fewer consumers purchase subsidized ACA coverage. As we discussed last year, the CEA estimates that from 2019 through 2029,
setting the mandate penalty to zero will yield $204 billion in net benefits for
consumers (CEA 2019).
Encouraging State innovation in insurance design. As of 2019, seven States
operated State Innovation waivers under Section 1332 of the ACA that utilized a
reinsurance component. As a way to lower risk, the State establishes a fund to
subsidize insurers for a certain amount of the expenses from people with costly
claims. These waivers lead to lower ACA plan premiums and thus lower associated premium tax credit costs. These seven States had a median premium
decline of 7.5 percent, compared with an increase in nonwaiver states of 3.0
percent (Badger 2019). Compared with what would have occurred if the States
had not passed waivers, the decrease in premiums has likely caused increased
enrollment in these States. By the end of 2019, States received back roughly 60
percent of savings of their initial contribution in Federal pass-through funding
(Blase 2019a).
Expanding association health plans and short-term limited-duration insurance. In June 2018, the Department of Labor (DOL) finalized a rule to expand
the ability of employers, including sole proprietors, to join together and
purchase health coverage through association health plans (AHPs).3 For many
employers, employees, and their families, AHPs offer more affordable premiums by reducing the administrative costs of coverage through economies of
scale. The AHP rule also gave small businesses more flexibility to offer their
employees health coverage that is more tailored to their needs.
In August 2018, HHS, the Department of the Treasury, and DOL finalized
a rule to expand Americans’ ability to purchase short-term, limited-duration
3 The revised definition of an employer for bona fide AHPs established under this rule is being
adjudicated.

184 |

250-840_text_.pdf 188

Chapter 5

2/7/20 3:46 PM

insurance (STLDI). STDLI premiums generally cost less than premiums for
individual insurance on the ACA exchanges. Because of lower costs, additional
choice, and increased competition, millions of Americans, including middleclass families that cannot afford ACA plans, stand to benefit from this reform.
Recently, the Congressional Budget Office (CBO 2019) stated that is will count
some short-term plans as health coverage, just as it did with pre-ACA plans
with benefit exclusions or annual and lifetime limits (Aron-Dine 2019). Though
these plans are more limited in coverage than the ACA-compliant insurance
plans, they are priced at up to 60 percent less than the unsubsidized premium
cost of ACA exchange plans and give consumers more insurance protection
than being uninsured.
As a result of STDLI and AHP rules, the CBO and the U.S. Congress’s Joint
Committee on Taxation estimates that over the next decade, roughly 5 million
more people are projected to be enrolled in AHPs or short-term plans. Of this
increase, almost 80 percent constitute individuals who would otherwise have
purchased coverage in the small-group or nongroup markets. The remaining
20 percent (roughly 1 million people) are made up of individuals who are projected to be newly insured as a result of the rules (CBO 2019).
Strengthening Medicare. The Administration’s reforms to Medicare
include payment policies that align with patients’ clinical needs rather than the
site of care, simplified processes for physicians’ documentation of evaluation
and management visits, new consumer-transparency measures, and increased
flexibility for insurers so that they can offer more options and benefits through
Medicare Advantage.
In 2019, President Trump signed an Executive Order to improve seniors’
healthcare outcomes by providing patients with more plan options, additional
time with providers, greater access to telehealth and new therapies, and
greater alignment between payment models and efficient healthcare delivery
(White House 2019b). In addition, a priority will be streamlining the approval,
coverage, and payment of new therapies while reducing obstacles to improved
patient care. Finally, the effort improves the fiscal sustainability of Medicare by
eliminating waste, fraud, and abuse.
Expanding health reimbursement arrangements. In June 2019, HHS, the
Treasury Department, and DOL issued a final rule expanding the flexibility and
use of health reimbursement arrangements to employers (84 FR 28888). The
rule issued two new types of tax-advantaged HRA plans—excepted benefit
HRAs (EBHRAs) and individual coverage HRAs (ICHRAs)—to be offered as early
as January 2020. EBHRAs may be offered to employees with traditional group
plans to receive an excepted benefit HRA of up to $1,800 a year in 2020 (indexed
to inflation afterward) for the purchase of certain qualified medical expenses,
such as short-term, limited duration, vision, and dental plans. ICHRAs allow
employers to reimburse employees who purchase their own health plans and

Free-Market Healthcare Promotes Choice and Competition | 185

250-840_text_.pdf 189

2/7/20 3:46 PM

equalizes the tax treatment of a traditional employer-sponsored insurance
plan and an individual market plan paid by employer contributions.
The Treasury Department performed microsimulation modeling to evaluate the coverage changes and transfers that are likely to be induced by the
final rules. The Treasury’s model of health insurance coverage assumes that
workers are paid the marginal product of their labor. Employers are assumed
to be indifferent between paying wages and payroll taxes and paying compensation in the form of benefits. The Treasury model therefore assumes that total
compensation paid by a given firm is fixed, and the employer allocates this
compensation between wages and benefits based on the aggregated preferences of their employees. As a result, employees bear the full cost of employersponsored health coverage (net of the value of any tax exclusion) in the form of
reduced wages and the employee share of premiums.
The Treasury Department’s model assumes that employees’ preferences
regarding the type of health coverage (or no coverage) are determined by their
expected healthcare expenses and the after-tax cost of employer-sponsored
insurance, exchange coverage with the premium tax credit (PTC), or exchange
or other individual health insurance coverage integrated with an individual
coverage HRA, and the quality of different types of coverage (including actuarial value).
When evaluating the choice between an individual coverage HRA and
the PTC for exchange coverage, the available coverage is assumed to be the
same, but the tax preferences are different. Hence, an employee will prefer the
individual coverage HRA if the value of the income and payroll tax exclusion
(including both the employee and employer portion of payroll tax) is greater
than the value of the PTC. In modeling this decision, the Federal departments
assume that premiums paid by the employee are tax-preferred through the
reimbursement of premiums from the individual coverage HRA, with any
additional premiums (up to the amount that would have been paid under a
traditional group health plan) paid through a salary reduction arrangement.
In the Treasury Department’s model, employees are aggregated into
firms, based on tax data. The expected health expenses of employees in
the firm determine the cost of employer-sponsored insurance for the firm.
Employees effectively vote for their preferred coverage, and each employer’s
offered benefit is determined by the preferences of the majority of employees.
Employees then decide whether to accept any offered coverage, and the
resulting enrollment in traditional or individual health insurance coverage
determines the risk pools and therefore premiums for both employer coverage
and individual health insurance coverage.
Based on microsimulation modeling, the Federal departments expect
that the final rules will cause some participants (and their dependents) to
move from traditional group health plans to individual coverage HRAs. As
noted above, the estimates assume that for this group of firms and employees,
186 |

250-840_text_.pdf 190

Chapter 5

2/7/20 3:46 PM

employer contributions to individual coverage HRAs are the same as contributions to traditional group health plans would have been, and the estimates
assume that tax-preferred salary reductions for individual health insurance
coverage are the same as salary reductions for traditional group health plan
coverage. Thus, by modeling construction, there is no change in income or payroll tax revenues for this group of firms and employees (other than the changes
in the PTC discussed below).
Although the tax preference is assumed to be unchanged for this group,
after-tax, out-of-pocket costs could increase for some employees (whose premiums or cost sharing are higher in the individual market than in a traditional
group health plan) and could decrease for others. A small number of employees who are currently offered a traditional group health plan nonetheless
obtain individual health insurance coverage and the PTC, because they cannot
afford a traditional group health plan or such a plan does not provide minimum
value. Some of these employees would no longer be eligible for the PTC for
their exchange coverage when the employer switches from a traditional group
health plan to an individual coverage HRA because the HRA is determined to be
affordable under the final PTC rules.
The regulatory impact analysis conducted by the Treasury Department
concluded that the benefits of the HRA rule substantially outweigh its costs.
The Treasury Department estimated that 800,000 employers are expected to
provide HRAs after being fully ramped up. In addition, it is estimated that there
will be a reduction in the number of uninsured by 800,000 by 2029. From these
employers’ HRA contributions, it is expected that firms will cover more than 11
million employees with individual health insurance by 2029.
Modernizing high-deductible health plans. A major component of the
Trump Administration’s health policy has been a focus on consumer-directed
health plans, in particular modernizing high-deductible health plans (HDHPs)
and their accompanying HSAs. As directed by the President, the Treasury
released a new Internal Revenue Service (IRS) guidance (Notice 2019-45) on
July 17, 2019, that allows high-deductible health plan issuers to permit coverage of prevention therapies for those with certain chronic conditions, including
diabetes, asthma, heart disease, and major depression. The impact could be
profound. For example, these plans could now cover all or nearly all the cost of
insulin for diabetic patients before the deductible being met.
HSA-eligible plans are a growing proportion of the overall HDHP market.
In 2018, about 21.8 million Americans were enrolled in HSA-eligible HDHPs, up
from an estimated 15.5 million in 2013 (AHIP 2017). In 2018, nearly 29 percent
of all firms offered an HDHP with a savings option, such as an HSA (KFF 2018).
Among companies studied in 2018 by a survey of the National Business Group
on Health, 30 percent offered a full replacement HSA-type plan to employees in
2019 (NBGH 2018). HSA market growth is expected to continue.

Free-Market Healthcare Promotes Choice and Competition | 187

250-840_text_.pdf 191

2/7/20 3:46 PM

According to the Centers for Disease Control and Prevention (CDC 2019),
about 60 percent of Americans have a chronic disease such as heart disease
or diabetes. The economic burden of chronic diseases in the United States is
estimated to be about $1 trillion per year (Waters and Graf 2018). Decreasing
financial barriers to evidence-based care for chronic conditions provides
opportunities to enhance clinical outcomes and reduce the long-term growth
rate of healthcare spending. Because about 75 percent of total U.S. health
spending is due to chronic diseases, appropriate chronic disease management
is key to lowering long-term healthcare cost growth (NACDD n.d.). The IRS
guidance allows for the creation of an enhanced HSA-eligible plan to provide
predeductible coverage for targeted, evidence-based, secondary preventive
services that prevent chronic disease progression and related complications.
This can improve patient outcomes, enhance HDHP attractiveness, and add
efficiency to medical spending.
The creation of these new high-deductible health plans plus secondary
prevention coverage (HDHP+) will give patients with certain conditions better
access. VBID Health (2019) estimated that it could increase tax revenue in a
variety of scenarios dependent on the updating of the new plan. Note that VBID
Health’s analysis was performed before Congress repealed the Cadillac tax in
December 2019.
The authors of this report (VBID Health 2019) used the ARCOLA microsimulation model to gauge the Federal tax revenue and insurance take-up
impact of an HDHP+ among those under 65 and not in the Medicare market.
The model assumes bronze plans in health insurance exchanges migrate into
the new HDHP+ design. That said, it is challenging for HSA-eligible plans in the
exchanges to meet bronze level actuarial value given their lower out-of-pocket
maximum required in statute compared with the out-of-pocket maximum
limits for the individual market. Providing more predeductible coverage will
make this more challenging. The model also assumes that everyone in the
individual market has the option of an out-of-exchange HSA-eligible plan that
does not switch to the HDHP+ design. The results are split into four scenarios
for firms that offer an HSA-HDHP: all firms additionally offer HDHP+, half of all
firms additionally offer HDHP+, all firms replacing current plans with HDHP+,
and half of all firms replacing current plans with HDHP+. Differences across
employer scenarios illustrate a range of possibilities that may play out.
Across all employer scenarios, the initial uptake and forecasted growth
of the novel HDHP+ are positive as people switch plan types. What varies by
employer scenario, however, are the magnitude and growth of uptake over
time. The HDHP+ generally has high initial uptake across employer scenarios.
The lowest uptake is in the scenario where half of employers additionally offer
the HDHP+ with other HDHP options. Because of the higher HDHP+ premiums,
due to selection, this result is expected (figure 5-8).

188 |

250-840_text_.pdf 192

Chapter 5

2/7/20 3:46 PM

Figure 5-8. Health Insurance Enrollment across Employer Scenarios,
2019–29
All PPOs

HDHP+

Change in enrollment (millions of people)
45

HDHP

36

30

18
10

15
0

–4

–6

5

–2

–3

1

–38

0

–20

-15
-30
-45
Full addition

Half addition
Full replace
Employer scenario

Half replace

Sources: VBID Health (2019); CEA calculations.
Note: PPOs = preferred provider organizations; HDHP+ = enhanced high-deductible health plan;
HDHP = high-deductible health plan.

Figure 5-9. The Net Revenue Impact of Expanding High-Deductible
Health Plans, 2019–29
Dollars (billions)
25
23

22

20
15
10

7

5
0
-5

–6

-10
Full addition

Half addition
Full replace
Employer scenarios

Half replace

Source: VBID Health (2019).
Note: Scenarios apply to the 7 percent premium for enhanced high-deductible health plans.

Free-Market Healthcare Promotes Choice and Competition | 189

250-840_text_.pdf 193

2/7/20 3:46 PM

Net revenue effects can be seen in three of the four scenarios modeled
after introducing HDHP+ to employer and individual markets and the migration
of people across plan types (figure 5-9).
Different employer decisions regarding plan offerings, as seen in the scenarios modeled, may lead one scenario to have a larger effect than another one
(VBID Health 2019). More than the magnitudes of the different budget effects is
the clustering of each scenario around budget neutrality. The one scenario that
shows a small net reduction in tax revenue (full replacement) was modeled
as an extreme case. The net effects of each scenario are small relative to the
net impact of tax subsidy of the entire employer-sponsored insurance market.
Thus, the net impact of expanding the secondary prevention safe harbor is
likely close to zero, if not modestly positive.

Increasing Competition
This subsection explores how to increase competition in providing healthcare.
The topics it covers include enforcing antitrust laws, accelerating generic drug
approvals, creating price and quality transparency, promoting new vaccine
manufacturing, and clarifying the Physician Self-Referral Law and the Federal
Anti-Kickback Statute.
Enforcing antitrust laws. Chapter 6 discusses the importance of sound
antitrust policy, which protects consumers from anticompetitive mergers. As
discussed there, the Antitrust Division of the DOJ and the FTC—collectively,
the Agencies—share responsibility for enforcing the Nation’s antitrust laws.
Although the vast majority of mergers do not raise competitive concerns, the
Agencies use their investigative powers to identify those that do by obtaining
and analyzing the detailed evidence that is needed to make this distinction.
Challenging a merger is often risky, as evidenced by the fact that between
1994 and 2000, the Agencies lost all seven lawsuits that they filed to block
hospital mergers (Moriya et al. 2010). In response to this, the FTC engaged in
a retrospective study of hospital mergers that advocated against the outdated
methodology that the courts had been using to evaluate these mergers. Joseph
Simons, the FTC chairman, recently reported to Congress that the FTC has
successfully defended in blocking a merger between healthcare providers (FTC
v. Sanford Health). This was the FTC’s fifth straight appellate victory involving
health provider mergers.
The DOJ has worked to stop anticompetitive mergers among health
insurers. In 2016, the DOJ successfully blocked two proposed mergers that
would have combined four of the largest health insurers (Anthem, Cigna,
Aetna, and Humana) into two companies. More recently, the DOJ reached a
settlement with CVS in its bid to acquire Aetna. The DOJ raised concerns relating to the sale of individual prescription drug plans (PDPs) under Medicare’s
Part D program. CVS and Aetna competed head-to-head in U.S. regions covering 9.3 million PDPs, of which 3.5 million had coverage from CVS or Aetna.
190 |

250-840_text_.pdf 194

Chapter 5

2/7/20 3:46 PM

The DOJ alleged that this competition had led to lower premiums and lower
out-of-pocket-expenses, and had improved formularies and service in many
regional markets. To preserve competition, the DOJ required Aetna to divest
its individual prescription drug plan. As discussed in an earlier report (CEA
2018), CVS, Express Scripts, and OptumRx are the three largest pharmacy benefit managers in the United States. The American Medical Association (2018)
expressed concern to the DOJ that but for the CVS-Aetna merger, Aetna might
become a disruptive competitor in PBM markets. At the time, Aetna engaged in
some PBM activities while outsourcing other activities to CVS. The DOJ did not
raise concerns along these lines.
The DOJ also recently reached a settlement in a conduct case against
Atrium Health (formerly the Carolinas HealthCare System). The DOJ was concerned about provisions in Atrium’s contracts with health insurers that were
preventing insurers from offering financial incentives to their customers to
choose providers that offer better value than Atrium, in terms of lower prices,
better service, or both. The restrictions undercut the efforts of health insurers
to induce competition between providers by creating health plans that provide
incentives for consumers to use providers that qualify for preferred tiers or
in-network status. As discussed by Gee, Peters, and Wilder (2019), the DOJ’s
economic analysis was consistent with academic research suggesting that
these plans help to reduce premiums.
Accelerating generic drug approvals. HHS has taken a number of actions
to empower consumers and promote competition, building on accomplishments such as the Food and Drug Administration’s (FDA’s) record pace of
generic drug approvals (CEA 2018). Initiatives to clarify regulatory expectations
for drug developers, coupled with internal review process enhancements,
improved the speed and predictability of the generic drug review process at
the FDA, resulting in a record number of generic drug approvals in the first
three years of the Trump Administration. In fiscal year 2019, the FDA approved
a record 1,171 generic drugs, after record approvals from the previous two
years (HHS 2019c). These actions contributed to the recent decrease (see box
5-1) in prescription drug prices; in June 2019, these prices saw their largest
year-over-year decrease in 51 years (see chapter 2 for more discussion of the
Administration’s deregulatory actions).
Creating price and quality transparency. On June 24, 2019, the President
signed an Executive Order to promote price and quality transparency through
a set of new initiatives (White House 2019b). A major problem in the healthcare
market is that patients often do not know the price or quality of healthcare
services. This lack of transparency denies patients the vital information they
need to make informed choices and exacerbates increased costs, suppressed
competition, and lower quality. As a result, there are wide variations in prices
across healthcare markets, even for the same services, as was described earlier
in this chapter. Accurate, accessible price and quality information will allow
Free-Market Healthcare Promotes Choice and Competition | 191

250-840_text_.pdf 195

2/7/20 3:46 PM

Box 5-1. The Consumer Price Index for Prescription Drugs
Despite arguments that prescription drug prices have increased in 2019, drug
prices according to the Consumer Price Index for prescription drugs (CPI-Rx)
have declined (year-over-year) in 9 of the past 11 months, as of the October
2019 release of CPI. The CPI is designed to provide an empirical measure of
the impact of price changes on the cost of living. As a component of the general CPI, the CPI-Rx measures how prices are changing in the prescription drug
market by indexing the weighted average of the price changes in a random
sample of prescription drugs (see figure 3-5).
The CPI-Rx has several strengths (CEA 2019c). First, it includes a random sample of prescription drugs and provides a summary measure that is
representative of the entire market of prescription drugs. Even if prices are
increasing for a large number of rarely prescribed drugs, the CPI-Rx can show
an average decrease if the prices of the most commonly prescribed drugs
are decreasing. A second strength of the CPI-Rx is that it accounts for generic
drugs. Lower-cost generic bioequivalents of many prescription drugs are
widely available and are often purchased over name brands, and the CPI-Rx
captures price decreases from new generic entries. The CPI-Rx also measures
transaction prices instead of list prices. The transaction price includes all payments received by the pharmacy, including out-of-pocket payments and payments from insurance companies, and it corresponds to the negotiated price
and reflects discounts—though not rebates. The list price does not include
discounts and rebates and is less representative of what the customer pays.
Though the CPI-Rx is the best measure of overall prescription drug
inflation, it is not a perfect measure. One of its main limitations is that it
does not account for the improvement in consumer value that occurs with
the entry of new goods, particularly when they are of a higher quality than
existing goods. This bias is believed to cause the CPI-Rx to overstate the true
level of prescription drug inflation and has been estimated to be as high as
2 percentage points a year (Boskin et al. 1996). A comparison between the
CPI-Rx and a separately constructed large alternative data set of drug prices
from the research firm IQVIA showed larger price increases in the IQVIA index,
indicating that the CPI-Rx may not be fully representative of a larger sample
(Bosworth et al. 2018). Additionally, even though the CPI-Rx for drug prices
indicates reasonable increases or declines, there may be some drug products
for which price changes can appear extreme.

patients to identify savings by “shopping” for healthcare services and make
choices that fit their healthcare needs and financial situations. Additionally,
transparency in healthcare prices and quality will lead to better value and more
innovations by facilitating increased competition among healthcare providers.
One of the first results of this initiative is a rule requiring hospitals to publish
their negotiated hospital charges (84 FR 61142). The new Executive Order
192 |

250-840_text_.pdf 196

Chapter 5

2/7/20 3:46 PM

directs providers as well as insurers to reveal negotiated prices on a routine
basis to aid consumers in their purchase of competitively priced medical care
and treatments.
The Executive Order also includes the development of the Health Quality
Roadmap (HHS 2019a). The Roadmap will align and improve reporting on
data and quality measures across Medicare, Medicaid, the Children’s Health
Insurance Program, the Health Insurance Marketplace, the Military Health
System, and the Veterans Affairs Health System. To accomplish this goal, the
Roadmap will provide a strategy for advancing common quality measures,
aligning inpatient and outpatient measures; and eliminating low-value or
counterproductive measures.
The Executive Order also calls for increased access to de-identified claims
data from taxpayer-funded healthcare programs and group health plans.
Healthcare researchers, innovators, providers, and entrepreneurs can use
these de-identified claims, which will still ensure patient privacy and security,
to develop tools that enable patients to access information that helps with
decisions about healthcare goods and services. Increased data access can
reveal inefficiencies and opportunities for improvement, including performance patterns for medical procedures that are outside the recommended
standards of care.
The 2019 Price and Quality Transparency Executive Order seeks to make
all healthcare prices negotiated between payers and providers non-opaque
and to help those shopping for healthcare to get the best value and lowest
price, as they do in other markets outside healthcare. The policy execution of
revealing negotiated prices between payers is currently under way, and the
impact will be able to be assessed in future analyses. One estimate places the
potential savings from common medical procedures to be nearly 40 percent on
a nationwide basis (Blase 2019b).
Promoting new vaccine manufacturing. In September 2019, the President
signed an Executive Order promoting new influenza vaccine manufacturing
technologies to reduce production times and increase vaccine effectiveness.
Millions of Americans suffer from seasonal influenza every year, and new vaccines are formulated each year to decrease infections from the most prevalent
influenza viruses. Vaccines are incredibly effective against influenza, with one
study finding that vaccines prevented over 40,000 influenza-related deaths
between 2005 and 2014 (Foppa et al. 2015). Despite their effectiveness, current
methods of vaccine production are often very slow and can diminish vaccines’
efficacy in protecting against seasonal influenza infection. Production delays
could be even more important in the event of a pandemic influenza outbreak.
The CEA (2019d) found that the cost of delay in vaccine availability in the case
of a pandemic is $41 billion per week for the first 12 weeks and $20 billion per
week for the next 12 weeks.

Free-Market Healthcare Promotes Choice and Competition | 193

250-840_text_.pdf 197

2/7/20 3:46 PM

The new Executive Order identifies the weaknesses in current methods
of vaccine production and promotes new technologies, such as cell-based and
recombinant vaccine manufacturing, to speed vaccines’ development and
improve their efficacy. Additionally, the new initiative establishes a task force
to increase Americans’ access to vaccines. If sufficient doses of vaccines are
delivered at the outset of an influenza pandemic, the CEA (2019c) estimates
that $730 billion in economic benefits could be gained by Americans, primarily
due to the prevention of loss of life and health.
Clarifying the Physician Self-Referral Law and the Federal Anti-Kickback
Statute. The Administration proposed two rules in 2019 to provide coordinated
care for patients (84 FR 55766) and to ensure that there are safeguards and flexibility for healthcare providers in value-based arrangements (84 FR 55694). The
first rule proposed by CMS is part of the Administration’s efforts to promote
value-based care by lifting Federal restrictions on healthcare providers so that
they have greater ability to work together on delivering coordinated patient
care.
The second proposed rule issued by the HHS Office of the Inspector
General focuses on the Federal Anti-Kickback Statute and the Civil Monetary
Penalties Law. This proposal addresses the concern that these laws needlessly
limit how healthcare providers can coordinate patient care. Expanding flexibility could, for example, encourage outcome-based payment arrangements that
reward improved health outcomes. The changes would also offer specific safe
harbors to make it easier for healthcare providers to ensure they are complying
with the law (HHS 2019b).

Increasing Access to Valuable Therapies
This section covers a number of key topics on how to increase access to valuable therapies. These include ending the HIV epidemic, expanding kidney
disease treatment options, combating the opioid crisis, and expanding the
right to try clinical trials.
Ending the HIV epidemic. For the last four decades, the Human
Immunodeficiency Virus (HIV) has been one of the most prominent health risks
confronting people in our country and around the world. In 2019, President
Trump announced a plan to end the HIV epidemic within 10 years. This epidemic has claimed the lives of about 700,000 Americans since 1981. The new
initiative is designed to reduce the number of new HIV infections in the United
States by 75 percent over the next five years, and by at least 90 percent over the
next decade. Through efforts across HHS, an estimated 250,000 HIV infections
could be averted over the next 10 years. The Administration also facilitated a
large private donation of pre-exposure prophylaxis (PrEP) medication, which
will help reduce the risk of HIV infection for up to 200,000 patients per year for
up to 11 years to provide critical PrEP medication to uninsured individuals who
might otherwise be unable to access or afford it.
194 |

250-840_text_.pdf 198

Chapter 5

2/7/20 3:46 PM

Expanding kidney disease treatment options. In July 2019, the President
signed an Executive Order to enable better diagnosis, treatment, and preventive care for Americans suffering from chronic kidney disease. In line with the
Administration’s broader deregulatory agenda, a key focus of the Executive
Order is an effort to remove regulatory barriers to the supply of kidneys.
Currently, the Federal Government bears most of the cost paying for chronic
kidney disease and end-stage renal disease care, which affect more than 37
million Americans (White House 2019d). More than 100,000 Americans begin
dialysis each year to treat end-stage renal disease, half of whom die within five
years. The Executive Order seeks to modernize and increase patient choice
through affordable treatment options that are too expensive and fail to provide
a high quality of life.
As directed by the Executive Order, the Centers for Medicare and
Medicaid Services issued a proposed rule to hold organ procurement organizations more accountable for their performance (84 FR 70628). More than 113,000
Americans are currently on the waiting list for an organ transplants, a number
that far exceeds the number of organs available. The rule raises performance
standards for organ procurement organizations to reduce discarding viable
organs, encourage higher donation rates, and shorten transplant waiting lists
(CMS 2019a). Additionally, the Health Resources and Services Administration
issued a proposed rule to alleviate financial barriers of organ donations (84 FR
70139). This rule would allow for reimbursement of lost wages and childcare
and eldercare expenses for living donors lacking other means of financial support, potentially increasing the number of transplant recipients over a shorter
time period.
Combating the opioid crisis. The Trump Administration is using Federal
resources to fight against the opioid crisis in U.S. communities. Actions are
focused on supporting those with substance use disorders and involving the
criminal justice system to crack down on illicit opioid suppliers, both foreign
and domestic. Over $6 billion in funding was secured in fiscal years 2018 and
2019 for preventing drug abuse, treating use disorders, and disrupting the supply of illicit drugs (OMB 2019). Investments include funding for programs supporting treatment and recovery, drug diversion, and State and local assistance.
Chapter 7 outlines in more detail many of the Administration’s accomplishments in combating the opioid crisis.
Expanding the right to try. The Administration has made increased access
to new and critical therapies a priority. One of the new bold programs in
2018 was the passage of “Right-to-Try” legislation for patients with terminal
illnesses, such as cancer. The National Cancer Institute (n.d.) estimates that
1.76 million new Americans will be diagnosed with cancer and 606,880 will
die from cancer in 2019. Currently, only 2 to 3 percent of adult cancer patients
are enrolled in clinical trials—an indication of the limited options for patients
with life-threatening diseases (Unger et al. 2019). For these patients who are
Free-Market Healthcare Promotes Choice and Competition | 195

250-840_text_.pdf 199

2/7/20 3:46 PM

ineligible to participate in clinical trials and have exhausted all approved treatment options, this bill amended Federal law to provide a new option, in addition to the FDA’s long-standing expanded access program, for unapproved,
experimental drugs (including biologics) to potentially extend their lives. To
ensure safety and transparency, manufacturers or sponsors of an eligible drug
that has undergone the FDA Phase I (safety) testing are required to provide
annual summary reports to the FDA on any use of the drug under Right-to-Try
provisions.

Conclusion
This chapter has identified Federal and State barriers to healthcare that
increase prices, reduce innovation, and hinder improvements in quality. It
also provided a summary of the accomplishments and expected effects of
the Trump Administration’s policies to address these barriers and deliver a
healthcare system that offers high-quality care at affordable prices. By 2023,
we estimate that 13 million Americans will have new insurance coverage that
was previously unavailable due to high prices and overregulation.
In contrast to the Administration’s focus on improving consumerdirected healthcare spending, government mandates often reduce consumer
choice. At all levels of government, healthcare regulations that limit choice,
stifle competition, and increase prices should be updated so that the U.S.
healthcare system can provide greater value. These regulations can also harm
the broader economy. For example, the Affordable Care Act has impeded economic recovery by introducing disincentives to work (Mulligan 2015). Though
market competition leads to an efficient allocation of resources that should
lower prices and increase quality, every market has features that deviate
from optimal conditions, and healthcare is no exception. Although the U.S.
healthcare system has challenges, they are not insurmountable problems that
mandate greater government intervention. The healthcare policy successes
over the past three years show the value of empowering the market to deliver
the affordable healthcare options that Americans rightly expect, and further
reform will provide Americans with improved healthcare through enhanced
choice and competition.

196 |

250-840_text_.pdf 200

Chapter 5

2/7/20 3:46 PM

x
Part II

Evaluating and Addressing
Threats to the Expansion

197

250-840_text_.pdf 201

2/7/20 3:46 PM

250-840_text_.pdf 202

2/7/20 3:46 PM

x
Chapter 6

Evaluating the Risk of
Declining Competition
America’s economic strength has always been driven by private sector competition. When large corporations, small businesses, and entrepreneurs all
must innovate to compete for market share on a level playing field, American
consumers win and the economy grows stronger.
Yet even with the economic expansion becoming the longest in U.S. history,
wage growth consistently meeting or exceeding 3 percent, unemployment
falling to a 50-year low, and small business optimism within the top 20 percent
of historical results, there is growing concern that the playing field is no longer
level, harming innovation and thus the American economy. The increasing
size of many of the Nation’s largest companies and the growing importance
of economies of scale has led some to hold the mistaken, simplistic view that
“Big Is Bad.” Though anticompetitive behavior by companies of any size should
lead to investigations and specific enforcement actions against offenders, an
across-the-board backlash against large companies simply because of their size
is unwarranted. Antitrust enforcers should continue to be particularly vigilant
where firms have significant market power, given the harm they can cause if
they engage in anticompetitive conduct. Moreover, under U.S. antitrust law,
conduct that may be procompetitive for a small firm can become problematic
if undertaken by a monopolist. However, the focus must be on the conduct and
not on size alone. Successful companies benefit the economy and consumers,
and they are not necessarily the threat to competition and economic growth
that they are sometimes perceived to be. Instead, companies that achieve scale

199

250-840_text_.pdf 203

2/7/20 3:46 PM

and large market share by innovating and providing their customers with value
are a welcome result of healthy competition.
As this chapter explains, the Trump Administration understands the vital role
competition plays in growing the economy, promoting new business, and
serving consumers. This understanding is underpinned by a deep appreciation
of economic evidence, and the best available evidence shows that there is
no need to hastily rewrite the Federal Government’s antitrust rules. Federal
enforcement agencies, which are already empowered with a flexible legal
framework, have the tools they need to promote economic dynamism; as ongoing investigations and resolved cases show, they are well equipped to handle
the competition challenges posed by the changing U.S. economy.
This does not mean that the Trump Administration’s work promoting competition is finished. In addition to vigorously combating anticompetitive behavior
from companies, the Administration is especially focusing on government
policies that distort and limit competition. As historic regulatory reform across
American industries has shown, cutting government-imposed barriers to
innovation leads to increased competition, strong economic growth, and a
revitalized private sector.

V

igorous competition is essential for well-functioning markets and a
dynamic economy. Therefore, the Trump Administration has championed policies that promote competition, such as reforming the tax
code and removing costly and burdensome regulations. The Administration
also promotes competition through sound antitrust policy, which protects
consumers from anticompetitive mergers and business practices. Effective
antitrust enforcement supports the Administration’s deregulatory agenda by
fostering self-regulating, competitive free markets. The Antitrust Division of
the Department of Justice (DOJ) and the Federal Trade Commission (FTC)—collectively, the Agencies—share responsibility for enforcing the Nation’s antitrust
laws. This chapter evaluates antitrust policy and the Agencies’ roles in light of
recent trends in the U.S. economy and pressing debates about competition.
In recent years, new technologies and business models have revolutionized the relationships between firms and consumers. Some of these changes,
such as rapidly improving information technology, have enabled firms to

200 | Chapter 6

250-840_text_.pdf 204

2/7/20 3:46 PM

grow, expanding their offerings from local markets to national ones, and from
national markets to international ones.
These changes have exacerbated concerns about rising concentration.
That is, in some parts of the economy, the largest firms appear to account for an
increasing share of revenues. An influential Obama-era CEA report, “Benefits of
Competition and Indicators of Market Power” (CEA 2016), argued that competition may be decreasing. This report is part of a broader debate—currently taking place in government, academia, and policy circles—about the state of competition in the economy. Proponents of the view that competition is declining
(e.g., Faccio and Zingales 2018; Gutiérrez and Philippon 2019; Philippon 2019)
argue that big businesses face little competition and are earning profits at the
expense of consumers and suppliers. Advocates such as Furman (2018) and the
Stigler Committee on Digital Platforms (2019) have called for changes to competition policy that would broaden the scope of antitrust enforcement. Others
have cautioned that these proposals are not supported by the economic evidence (Syverson 2019), or that current antitrust rules are adequate to address
legitimate concerns about anticompetitive behavior (Yun 2019).
Calls for changing the goals of the antitrust laws are based on empirical
research that misinterprets high concentration as necessarily harmful to consumers and reflective of underenforcement. That argument was discredited
long ago, when economists such as Demsetz (1973) and Bresnahan (1989)
articulated the fundamental reasons why high concentration is not in and of
itself an indicator of a lack of competition. The main point is that concentration
may result from market features that are benign or even benefit consumers.
For example, concentration may be driven by economies of scale and scope
that can lower costs for consumers. Also, successful firms tend to grow, and
it is important that antitrust enforcement and competition policy not be used
to punish firms for their competitive success. Finally, antitrust remedies may
not be required, even when firms exercise market power, because monopoly
profits create incentives for new competitors to enter the market—unless substantive entry barriers or anticompetitive behavior stand in their way.
Moreover, recent empirical arguments that competition is in decline have
been based on broad, cross-industry studies. The findings from these studies
are both problematic and incomplete, and their implications for competition
remain speculative. In contrast, the methods that the Agencies use to analyze
competition are rooted in microeconomic, empirical evidence and involve
detailed analyses of competitive conditions in specific industries. Any conclusions about the state of competition should be made on the basis of this type
of careful research.
In addition, criticisms about the capabilities of antitrust enforcement to
address novel enforcement challenges in dynamic markets fail to account for
the flexibility of antitrust rules to accommodate a range of market conditions.
Effective antitrust enforcement takes account of the evidence and economics
Evaluating the Risk of Declining Competition | 201

250-840_text_.pdf 205

2/7/20 3:46 PM

appropriate to particular markets, and in turn adapts to innovation and development in the markets over time.
In short, we argue that major policy initiatives to completely rewrite antitrust rules and to create a new regulator for the digital economy are premature.
In this chapter, we discuss and critique proposals for such initiatives advanced
by proponents in the debate. As we explain, because these proposals are likely
to impose significant costs, they should not be undertaken on the basis of current evidence.
Finally, we discuss competition policy beyond antitrust law and the
Administration’s efforts to combat the negative impact of overly burdensome
regulation on competition. We highlight the Trump Administration’s successful
efforts to streamline the process by which new drugs are brought to market,
particularly generic drugs. We also discuss the Agencies’ efforts to advocate
for the removal of unnecessary occupational licensing requirements that limit
entry into professions, certificate-of-need laws that limit entry by new hospitals, and automobile franchising laws that limit the ability of car manufacturers
to sell cars directly to consumers. Here, we also discuss the Agencies’ work at
the intersection of intellectual property law and antitrust law.
The structure of the chapter is as follows. We first provide an overview
of antitrust policy and the economic analyses that the Agencies do to evaluate
whether there is a need for the Federal Government to be involved to prevent
anticompetitive mergers or other similar conduct. We then discuss the claims
of rising concentration and the evidence on which they are based, contrasting
this to the type of analysis that the Agencies do. Next, we discuss the proposals for regulation, with a focus on the digital economy. In the last section, we
discuss the Trump Administration’s policies to spur competition outside the
context of antitrust rules.

The Origin and Principles of Antitrust Policy
The Agencies follow the guiding principle that the role of antitrust law is to
protect the competitive environment and the process of competition. The
Agencies use their given authority for robust enforcement of antitrust law to
prevent anticompetitive behavior by firms. They also seek to avoid undue
interference by the Federal Government in the competitive process.
The main antitrust statutes are the Sherman Antitrust Act of 1890, the
Clayton Act of 1914, and the Federal Trade Commission Act of 1914. Together,
these laws address three categories of conduct: mergers, monopolization, and
anticompetitive agreements. First, under the Clayton Act, both Agencies challenge mergers that have a reasonable likelihood of reducing competition. They
also challenge acts of monopolization under Section 2 of the Sherman Act or
the equivalent provision of the Federal Trade Commission Act. Finally, both
Agencies challenge agreements among separate economic actors that place

202 | Chapter 6

250-840_text_.pdf 206

2/7/20 3:46 PM

unreasonable restraints on trade under Section 1 of the Sherman Act or the
Federal Trade Commission Act (FTC 2019d).
Certain types of conduct, such as collusion among competitors to fix
prices or rig bids, are considered so harmful to competition that they are categorized as criminal violations of the Sherman Act. The DOJ has long prioritized
criminal enforcement of the antitrust laws, and violations carry significant
financial fines and, for culpable individuals, jail time.
For noncriminal conduct, whether for mergers or monopolization, a central challenge facing the Agencies is determining when conduct is procompetitive and when it is anticompetitive. It can be difficult to distinguish between
the two, and optimal enforcement is often a balancing act. The Agencies and
the Courts want to avoid mistakenly prohibiting conduct that is procompetitive, and they also want to avoid allowing conduct that is anticompetitive.
To understand these challenges, consider a merger between direct competitors (i.e., a horizontal merger). The reduction in competition could encourage the merged firm—and also, perhaps, its competitors—to raise prices. If
higher prices or other competitive types of harm to consumers are the likely
outcome of a merger, then the Agencies may file a lawsuit to seek to block the
transaction. Conversely, a merger, even one between close competitors, can
enhance competition by creating a stronger competitor. Mergers often allow
firms to combine complementary assets to realize a variety of efficiencies.
For example, they may realize cost reductions, improve the quality of their
products, or develop new products. Cost reductions, in particular, create an
incentive to reduce prices that can offset or even reverse any incentives to raise
prices. As a result, horizontal mergers may in some cases lead to lower prices,
not higher ones. As we discuss in the next section, when the Agencies review
mergers, they conduct a detailed economic analysis to assess these complex
issues.
Most mergers do not raise competition issues. For example, the merging
firms may not operate in the same or even related markets. Antitrust concerns
are usually greatest when the merging parties are direct competitors. In rarer
cases, antitrust concerns can arise when the merging firms are vertically
related, such as when one firm sells inputs to the other. This was the case in the
DOJ’s challenge of the merger between AT&T and Time Warner, as is discussed
by Gee, Peters, and Wilder (2019).
When mergers are large enough, the merging parties must notify the
Agencies in advance of merging. In 2018, the most recent year for which
data are available, the Agencies received notice of 2,028 mergers that were
potentially subject to review (DOJ and FTC 2019a). Most deals were allowed to
proceed after an initial review that takes place within 30 days of the notification. In 45 matters, the reviewing agency identified competition issues and
sought additional discovery from the parties to allow an in-depth investigation,
in what is referred to as a “Second Request.” As figure 6-1 shows, the number
Evaluating the Risk of Declining Competition | 203

250-840_text_.pdf 207

2/7/20 3:46 PM

Figure 6-1. Summary of Transactions by Fiscal Year, 2009–18
Mergers receiving second-stage review
Total mergers potentially subject to second-stage review

Number of firms
2,500

2,000

1,500

1,000

500

0
2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

Source: Federal Trade Commission and Department of Justice (2019a).

of second requests conducted by the Agencies has remained relatively stable
over time.

Economic Analysis at the Agencies
To aid in distinguishing between procompetitive and anticompetitive conduct,
the Agencies employ Ph.D. economists who specialize in the analysis of competition. The Agencies also hire outside economic experts to examine evidence
in particular cases. Here, we provide an overview of how economic analysis
is used in merger enforcement. Similar methods are used in other areas of
antitrust enforcement.
The central question in any merger review is whether the merger may
substantially lessen competition. As explained in the “Horizontal Merger
Guidelines” (DOJ and FTC 2010), this means that one or more firms affected
by the merger are reasonably likely to raise prices, reduce output, decrease
quality, reduce consumer choice, diminish innovation, or otherwise harm consumers. This is sometimes referred to as a consumer welfare standard, because
the focus is on economic harm to consumers. Usually, this means harm to
downstream customers of the merging firms, but the Agencies may also
evaluate harm to upstream suppliers if there is a concern that the merger will
enhance monopsony power, leading to lower prices or other types of economic
harm for the suppliers deprived of competition for the sale of their goods or
services; see box 6-1. Importantly for the digital age, the consumer welfare
204 | Chapter 6

250-840_text_.pdf 208

2/7/20 3:46 PM

Box 6-1. Antitrust and Monopsony:
George’s Foods and Tyson Foods
Although most merger reviews focus on types of harm to downstream
consumers, the Agencies may also investigate antitrust concerns relating
to monopsony. In 2011, the DOJ challenged George’s Foods’ acquisition of
a chicken-processing complex in Harrisonburg, Virginia, that was owned by
Tyson Foods. Both companies provide chicken-processing services for birds
that are raised by the surrounding area’s farmers. The processors own the
birds, provide the chicks and feed, and transport the birds between the farm
and processing plants. The farmers (“growers”) work under contract with the
processors, providing chicken houses, equipment, and labor for raising the
chickens.
Before the merger, George’s Foods and Tyson Foods competed directly
with each other for purchasing the services of growers in the Shenandoah
Valley. The merger reduced the number of competitors from three to two,
leaving George’s Foods with about 40 percent of local processing capacity.
The DOJ raised concerns that the merger would allow George’s Foods to
decrease prices or degrade contract terms to growers in the region. The other
competing processor lacked the capacity to take on significant numbers
of growers if George’s were to depress prices. To remedy these concerns,
George’s agreed to invest in improvements in Tyson’s chicken processing
facilities, giving it an incentive to operate at a greater scale than before
the merger. With an increased demand for chickens, George’s also had an
increased demand for the local growers (DOJ 2011a, 2011b).

standard considers harm beyond price effects, including harm to innovation,
quality, and choice. The consumer welfare standard is also different from a
total welfare standard, which would focus on overall efficiency, or outcomes
that maximize the joint surplus of consumers and firms.1
To evaluate the likelihood of consumer harm, the Agencies analyze a
variety of evidence. They may seek documents, testimony, and data from the
merging parties. They may also seek information from other affected parties
including customers, suppliers, and rival firms.
An important part of the analysis is to determine the nature of competition. Competition takes a variety of forms, and the effect of a merger depends
on how competition works in the affected markets. For example, firms set
prices in a variety of ways. They may be posted, as is common in the retail sector, or they may be negotiated, as is common in business-to-business services.
In some cases, negotiations between buyers and sellers are structured with a
formal auction process. These and other differences shape the nature of competition. In some markets, competition is so fierce that two competing firms
1 Wilson (2019) has a discussion of the pros and cons of alternative antitrust standards.

Evaluating the Risk of Declining Competition | 205

250-840_text_.pdf 209

2/7/20 3:46 PM

are enough to drive prices down to the marginal cost. In other markets, many
firms can profitably set prices significantly above the marginal cost.
The strength of competition between any firms depends on the extent to
which consumers view their products as substitutes. Firms often sell differentiated products. This means that their products are similar, but not identical,
and consumers may have strong (or weak) preferences between them. An
important part of the economic analysis is assessing how close the merging
firms’ products are to each other in the view of consumers. Concerns about a
lessening of competition will usually be greatest if many consumers view the
firms’ products as each other’s closest substitutes. For example, some brands
of breakfast cereal are so different in flavor, nutrition, and other attributes that
few consumers regard them as substitutes, and competition between them is
weak. Other brands of breakfast cereal probably compete head-to-head. To
assess the closeness of products, economists at the Agencies review evidence
such as win/loss reports, discount approval processes, customer switching
patterns, and consumer surveys.
Based on such evidence, the Agencies identify relevant markets where
competition is likely to be harmed. This analysis is based on demand substitution, or how consumers would respond to the increase in the price of a product.
For example, if the evidence were to show that few people would switch to
eating sugary breakfast cereals if the price of “heart-healthy” breakfast cereals
were to rise, the Agencies might define a market for “heart-healthy” breakfast
cereals that excludes the sugary alternatives. How broadly or narrowly to
define markets can be a source of contention, as the shares of the merging
firms will appear lower in broader markets. If markets are defined too broadly,
they will contain products that do not significantly constrain the prices of the
merging firms. The lower shares of the merging firms may then wrongly suggest that there is more competition than actually exists.
The Agencies also identify the relevant geography for a market. Markets
may have a limited geography based either on consumers’ preferences or on
sellers’ ability to serve them. For example, for most people, restaurants in Los
Angeles and New York are probably not close substitutes. Nor would a flight
from Los Angeles to New York be a good substitute for a flight from New York
to Washington. In mergers of airlines, the DOJ often defines markets consisting
of origin and destination pairs. A relevant market might include nonstop flights
from San Francisco to Los Angeles if the merging parties both offer such flights.
The Agencies use a methodological tool, known as the hypothetical
monopolist test, to delineate relevant markets. The test imagines that a single
profit-maximizing firm monopolizes the candidate market and then analyzes
whether the monopolist would “impose at least a small but significant and
non-transitory increase in price” (DOJ and FTC 2010, 9). The Agencies usually
define markets to be the smallest ones that satisfy the test. When a market is

206 |

250-840_text_.pdf 210

Chapter 6

2/7/20 3:46 PM

defined this way, products in the market significantly constrain each other’s
prices, but products outside the market do not.
After defining a relevant market, the Agencies calculate shares for all
firms in the market and assess the level of concentration. Markets are classified as unconcentrated, moderately concentrated, or highly concentrated,
based on thresholds of the HHI; see box 6-2. Markets with HHIs above 2,500
are considered highly concentrated. In such markets, the Agencies presume
that mergers that increase the HHI by more than 200 points are likely to be
anticompetitive. However, the merging parties can rebut this presumption
with persuasive evidence.
To illustrate the role of market definition, consider the recent merger of
the Walt Disney Company and Twentieth-Century Fox. The DOJ was concerned
about competition between ESPN, which was owned by Disney, and the Fox
Regional Sports networks. A key question was how much competition these
cable sports networks faced from the sports programming shown on the major
broadcast networks. The DOJ alleged that the licensing of cable sports programming to multichannel video programming distributors, such as Comcast
and FIOS, was a relevant market, and one in which the merging parties had
high shares. In excluding broadcast programming from the market, the DOJ
alleged that the broadcast networks did not provide sufficiently close competition to prevent competitive harm. As stated in the complaint, multichannel
video programming distributors do not typically consider broadcast network
programming as a replacement for cable sports programming because broadcast networks offer limited airtime to sports programming and are focused on
marquee events with broad appeal. The DOJ approved the merger only after
the parties agreed to divest Fox’s interests in its regional sports networks (DOJ
2018a, 2018b).
The inquiry into market share is a starting point for economic analysis,
but the ultimate goal is to assess whether the merger is likely to have adverse
competitive effects. A merger may harm competition because there are fewer
competitors competing (unilateral effects), or it could harm competition by
encouraging explicit or tacit coordination between rivals (coordinated effects).
As noted above, mergers may harm competition in prices, or they may harm
competition in nonprice dimensions, such as quality or innovation.
To evaluate competitive effects, the Agencies use a variety of evidence.
Market shares are one type of evidence, but other evidence is also considered.
For example, the Agencies may analyze how a recent merger in the same market affected competition. Or, if the merging firms compete in some local markets, but not others, the Agencies may compare prices across regions where
the firms do and do not compete. In markets with differentiated products, such
as breakfast cereal, the Agencies may estimate diversion ratios. A diversion
ratio is a measure of how closely two products compete. For a first product sold
by one of the merging firms and a second product sold by the other merging
Evaluating the Risk of Declining Competition | 207

250-840_text_.pdf 211

2/7/20 3:46 PM

Box 6-2. Measuring Concentration and the HHI
Concentration is a measure of the number and size of firms competing in a
market. When markets are delineated around competition, concentration can
be a useful reflection of competitive conditions. In highly concentrated markets—those markets with a small number of large firms—mergers between
large firms are relatively likely to enhance market power, leading the merged
firm to raise prices, reduce quality, reduce innovation, or otherwise harm
consumers.
The Agencies usually measure concentration in terms of a firm’s
share of market revenues, but concentration can be defined around other
measures, such as unit sales. The Agencies use the measure that best reflects
the competitive significance of firms in the market. For example, if physical
capacity limits the ability of firms to expand their production, market shares
may be measured in terms of physical capacity. A firm that is poised to enter a
market, but is not yet selling anything, may be assigned a market share based
on projected revenues.
The Agencies measure concentration using the Herfindahl-Hirschman
Index (HHI), which is calculated as the sum of the squares of the individual
firms’ market shares in a relevant market. In a monopolized market with only
one firm, the firm’s share is 100 percent, and the HHI is 100^2, or 10,000. In
a market with 100 firms each with 1 percent share, the HHI is much lower,
at 100. A higher HHI corresponds to a more concentrated market. A merger
between two firms combines their shares, so the HHI increases. For example,
if a market has four equal-sized firms and two of the firms merge, the HHI
increases from 2,500 to 3,750.

firm, the diversion ratio is the percentage of sales that the first product would
lose to the second product, if the price of the first product increases. The higher
the diversion ratio, the closer the competition. The Agencies sometimes use
diversion ratios in the context of economic models that simulate how firms
would change their prices after a merger. The Agencies also consider whether
efficiencies or entry are likely to offset or reverse adverse competitive effects.
The analysis of competitive effects has become more important over
time. As discussed by Shapiro (2010), the Agencies revised the Horizontal
Merger Guidelines in 1982 to downplay the emphasis on market shares and
to increase the emphasis on competitive effects.2 With this change in emphasis, antitrust enforcement also became less interventionist. Shapiro (2010)
observes that the 1968 Horizontal Merger Guidelines stated that the Agencies
“ordinarily challenge” mergers between an acquiring firm with at least 15 percent market share and an acquired firm with at least 1 percent market share.
2 Shapiro (2010, 51–52). See also Lamoreaux (2019); Berry, Gaynor, and Morton (2019); and
Peltzman (2014).

208 | Chapter 6

250-840_text_.pdf 212

2/7/20 3:46 PM

Mergers of this sort would be unlikely to be challenged today, because the
analysis of competitive effects is rarely supportive of antitrust enforcement in
such cases.
However, many people argue that the Agencies intervene too rarely in
the modern era. Opponents of this view argue that antitrust overenforcement
is more harmful than antitrust underenforcement. This is because if markets
become overly concentrated to the point that profits are excessive, new firms
are likely to enter to take up the slack. Proponents of more aggressive enforcement argue that new firm entry is often not guaranteed. In markets where entry
is difficult (i.e., there are high barriers to entry), established firms may reap
excessive profits for long periods of time (Baker 2015). In the next section, we
turn to this debate.

A Renewed Interest in Concentration
and the State 0f Competition
Some observers of the U.S. economy have raised concerns that it is becoming less competitive. As noted above, in 2016, an influential CEA policy brief
(CEA 2016) argued that competition may be decreasing in many sectors, and
President Obama issued an executive order directing Federal Government
agencies to promote competition (White House 2016). Similar diagnoses and
calls to regulatory action have been sounded by pundits and economists alike.3
In this section, we first discuss problems with the evidence presented in
the 2016 CEA report, and then we explain how similar issues are manifested
in other research on this topic. We explain why drawing inferences about the
state of competition or antitrust enforcement from this weak evidence is problematic. Finally, we discuss alternative approaches to assessing if there is in
fact a competition problem in the United States.

Problems with the CEA’s 2016 Report
A central argument made in the 2016 CEA report, “Benefits of Competition
and Indicators of Market Power,” is that the rising market shares of the largest
firms in many industries constitute evidence of declining competition. This
argument is flawed both in terms of the evidence on market shares and the
inference about competition.
Table 6-1, which is taken from the 2016 CEA report, examines trends in
the revenue share of the 50 largest firms—known as the CR50—in different
industry segments. For background, the U.S. Census Bureau classifies firms
using the North American Industry Classification System (NAICS), which divides
the entire economy into 24 sectors classified with two-digit numerical codes, or
3 Examples include Furman (2018); Grullon, Larkin, and Michaely (2019); Krugman (2016); Kwoka
(2015); Lamoreoux (2019); Wessel (2018); Wu (2018); and the Economist (2016).

Evaluating the Risk of Declining Competition | 209

250-840_text_.pdf 213

2/7/20 3:46 PM

Table 6-1. Change in Market Concentration by Sector, 1997–2012
Industry

Transportation and warehousing
Retail trade
Finance and insurance
Wholesale trade
Real estate rental and leasing
Utilities
Educational services
Professional, scientific, and
technical services
Administrative and support
Accommodation and food
services
Other services
Arts, entertainment and
recreation
Healthcare and assistance

Change in revenue
Revenue earned by
Revenue share earned share earned by 50
50 largest
by 50 largest firms in largest firms from 1997
firms in 2012 (dollars,
2012 (percent)
to 2012 (percentage
billions)
points)
307.9

42.1

11.4

1555.8
1762.7
2183.1

36.9
48.5
27.6

11.2
9.9
7.3

121.6

24.9

5.4

367.7
12.1

69.1
22.7

4.6
3.1

278.2

18.8

2.6

159.2

23.7

1.6

149.8

21.2

0.1

46.7

10.9

-1.9

39.5

19.6

-2.2

350.2

17.2

-1.6

Source: Census Bureau.
Note: Data represent all North American Industry Classification System sectors for which data were available from 1997 to
2012.

two-digit sectors. These sectors are further divided into three-, four-, five-, and
six-digit subsectors. The CEA (2016) and Furman (2018) examine concentration
in 13 of the two-digit NAICS sectors. Table 6-1 shows that 10 sectors became
concentrated by this measure over the 15-year period from 1997 to 2012.
A key problem with table 6-1 is that the two-digit sectors are aggregations
of overly broad geographic and product markets that shed little light on the
state of competition. For example, retail trade includes all grocery stores, hardware stores, and gasoline stations, among many others, across the Nation. But
grocery stores in Florida and Wisconsin do not compete for the same customers, and hardware stores and gas stations, even those in the same local area,
largely sell products that are unrelated in demand. Concentration measures
defined by national segments also miss the dimension of local competition.
Rossi-Hansberg, Sarte, and Trachter (2019) find that the expansion of national
firms into local markets has been a factor both in increasing concentration at
the national level and in decreasing concentration at the local level.
This approach contrasts with how the Agencies define relevant markets
for antitrust analysis. As discussed above, the Agencies, and antitrust economists more generally, analyze data on demand that reveal the extent to which
consumers regard products as substitutes. In this way, markets are defined to
include products that are in competition with each other in the local product
markets where they compete. Even the finest six-digit NAICS sectors are far
broader than typical antitrust markets. Werden and Froeb (2018) calculate the
volume of commerce of the relevant markets alleged in DOJ merger complaints

210 |

250-840_text_.pdf 214

Chapter 6

2/7/20 3:46 PM

between 2013 and 2015 as a share of industry shipments in the six-digit NAICS
sector. They find that in most cases, the antitrust markets accounted for less
than 0.5 percent of the six-digit NAICS sector. In many cases, this is because
the antitrust markets where the DOJ identified a competition problem involved
single localities such as a city, State, or region, whereas the NAICS sectors are
national. Although studies of broad swaths of the economy, such as the 2016
CEA report, are necessarily limited by the data that are publicly available, the
coarseness of the data limits what they can say about competition.
A second problem with table 6-1 is the use of the CR50. The Agencies
and other economists often find evidence of robust competition in markets
with only a few firms engaged in head-to-head competition. Either the HHI
(discussed above) or a four-firm concentration ratio (known as the CR4) would
be more appropriate for a competition study. Note that in table 6-1, the CR50
are also usually much less than 100, meaning that there are more than 50 firms
operating in the segment.
Because of the overly broad market definition and the use of the CR50,
the data presented in table 6-1 tell us nothing about competition in specific
markets, let alone across the entire economy. Carl Shapiro, a former CEA member and Deputy Assistant Attorney General for Economics under the Obama
Administration, concluded that table 6-1 “is not informative regarding overall
trends in concentration in well-defined relevant markets that are used by antitrust economists to assess market power, much less trends in competition in
the U.S. economy” (Shapiro 2018, 722).

Problems with Related Research
The CEA’s 2016 report, “Benefits of Competition and Indicators of Market
Power,” is part of a larger body of recent research arguing that competition
may be in decline. Much of this literature tries to infer the state of competition
from correlations between flawed concentration measures, such as those
presented in table 6-1, and market outcomes, such as prices, profits, and
markups. This methodology rests on a problematic assumption that increases
in concentration create conditions of softer competition. That is, if undesirable
outcomes—such as higher prices, profits, and markups—are correlated with
concentration, then the cause of these outcomes is assumed to be weaker
competition. Recent papers in this vein include the 2016 CEA report; and those
by Furman (2018); Furman and Orszag (2018); Gutiérrez and Philippon (2017a,
2017b); and Grullon, Larkin, and Michaely (2019).
Problems with this assumption have been understood since at least
the 1970s (Demsetz 1973; Bresnahan 1989).4 The most fundamental problem
is that there are alternative explanations for why a market might demonstrate both high concentration and high markups that are consistent with
4 For a recent, in-depth discussion, see Berry, Gaynor, and Morton (2019); and Syverson (2019).

Evaluating the Risk of Declining Competition | 211

250-840_text_.pdf 215

2/7/20 3:46 PM

procompetitive behavior by firms. These include fixed costs, scale economies,
and globalization.
To see that this is true, consider the issue of fixed costs. In many markets,
firms make upfront investments in assets such as physical plant, equipment,
research and product development, and information technology. Firms will
make these investments only if they anticipate earning sufficient profit margins to recoup them. In terms of basic economics, if a firm has substantial
fixed costs, then its average cost may be substantially higher than its marginal
cost. A firm may earn a profit close to zero when fixed costs are accounted
for, but still maintain a positive margin between price and marginal cost.
The Agencies do not regard this as inherently problematic. As the Horizontal
Merger Guidelines state, “High margins commonly arise for products that are
significantly differentiated. Products involving substantial fixed costs typically
will be developed only if suppliers expect there to be enough differentiation
to support margins sufficient to cover those fixed costs. High margins can be
consistent with incumbent [established] firms earning competitive returns”
(DOJ and FTC 2010, 4); see box 6-3.
Even if high concentration and high markups are not inherently problematic, what about rising concentration and rising markups? This depends on why
the markups and concentration are rising. Suppose that fixed costs are rising.
If they are rising for anticompetitive reasons, such as if new and unnecessary
government regulations are raising the cost of entry, then the trend may be
associated with higher prices and consumer harm. But fixed costs could also be
rising because firms are making increasingly expensive investments to become
more competitive. Information technology in particular can involve upfront
investments in business systems that help to reduce a firm’s marginal cost of
production or improve product quality. A firm that makes such investments
may outcompete less efficient firms and grow its market share. Through such
a process, information technology could transform a market to one with fewer,
more efficient firms. Because the surviving firms have lower marginal costs,
their prices may fall even as their markups rise. This scenario is procompetitive
because consumers derive benefits from the lower prices or improved quality.
Berry, Gaynor, and Morton (2019) review recent research, providing
evidence that investments in intangible assets such as software and business
processes are becoming more important. Crouzet and Eberly (2019), in particular, find a positive correlation between firms’ market shares (in broad industry
segments) and their investments in intangible assets. In the view of Berry,
Gaynor, and Morton (2019), the broad category of “increasing investments
in fixed and sunk costs” may be the most important source of rising global
markups. Autor and others (2019) find evidence that increases in concentration
reflect a reallocation of output toward large, productive firms. They argue that
this could be the result of globalization and technological change, and further
observe that their explanation for rising concentration has “starkly different
212 |

250-840_text_.pdf 216

Chapter 6

2/7/20 3:46 PM

Box 6-3. Concentration, Innovation, and Competition
Industries that rely on innovation often provide dramatic examples of high
fixed costs. Consistent with this situation, concentration is often high. The
relationship between concentration, competition, efficiency, and consumer
welfare is complex. Competition can spur firms to innovate, but it can also
weaken their incentives to innovate by making it difficult for them to recoup
their investments. In research spanning decades, economists have found that
different models give different answers about whether higher concentration
increases or decreases innovation, and results about the optimal level of concentration are often sensitive to market conditions (Marshall and Parra 2019).
To illustrate, Igami and Uetake (2019) study these trade-offs in the hard
disk drive industry. As shown in figures 6-i and 6-ii, the period had waves of
entry and exit as the industry matured and consolidated. Innovation was of
central importance, as the industry followed Kryder’s law, that the storage
capacity of hard disk drives doubles roughly every 12 months. After estimating
a model of dynamic oligopoly, Igami and Uetake (2019) simulate the effect of
alternative merger policies on expected social welfare. They conclude that
a policy to block mergers if there are three or fewer firms would have found
“approximately the right balance between pro-competitive effects and valuedestruction side effects.” Although such a policy might not be optimal in

Evaluating the Risk of Declining Competition | 213

250-840_text_.pdf 217

2/7/20 3:46 PM

other industries or for any particular merger, this study helps to illustrate why
competition can be robust in markets with relatively few firms.
The proposed acquisition of Baker Hughes by Halliburton provides
an example of when innovation was central to a merger review (DOJ 2016).
Halliburton, Baker-Hughes, and Schlumberger were the three leading firms
in the oilfield services industry, providing sophisticated drilling technology
and related services for drilling oil wells. Each invested hundreds of millions
of dollars annually in research and development; for products where innovation was most important, there were few other competitors. The DOJ sued
to block Halliburton’s proposed acquisition of Baker-Hughes, delineating 23
relevant products and services where the proposed merger would result in
markets dominated by the merged firm and Schlumberger. The DOJ was not
satisfied that Halliburton’s proposed divestitures would remedy the potential
harm, and the parties ultimately abandoned their plans (Chugh et al. 2016).

implications” for welfare than explanations based on weakened competition
or antitrust enforcement. That is, if rising concentration and markups are
driven by conduct that benefits consumers, such as can be the case for investments in intangible assets, then there may be no competition problem and no
antitrust implications.
In addition to the fundamental error of equating concentration with a
lack of competition, there are also other problems with the recent literature on

214 |

250-840_text_.pdf 218

Chapter 6

2/7/20 3:46 PM

concentration. Similar to the CEA’s 2016 report, these studies’ use of Census
and other macroeconomic data limits them to examining concentration in
NAICS industry segments that are too broad to shed light on competitive conditions in properly defined antitrust markets. Many of the studies use data for
three-digit or four-digit NAICS segments (e.g., Gutiérrez and Philippon 2017a,
2017b, 2019; Grullon, Larkin, and Michaely 2019); but as discussed above, even
the finest six-digit NAICS segments are far broader than antitrust markets.
Another problem is that many of the studies explore links between
concentration and financial measures, such as markups and profits, that are
difficult to measure—especially across broad industry segments. Price-cost
markups, in particular, are a basic measure of market power, but firm-level
data on markups are rarely available. Accounting data are sometimes informative about the markup of price over average variable cost, but they do not
accurately measure the economic profit margins that are relevant to economic
analysis. Basu (2019) reviews different approaches to estimating markups used
in the recent research discussed above. He discusses problems with the methods, including that most of the estimates of markups are implausibly large.

Connecting Concentration and Markups with Antitrust Law
The assessment of the competitive health of the economy should be based
on studies of properly defined markets, together with conceptual and empirical methods and data that are sufficient to distinguish between alternative
explanations for rising concentration and markups. This continues to be the
approach of the Agencies.
In line with this, Berry, Gaynor, and Morton (2019, 63) call for a wave of
“industry-level econometric studies . . . to help us understand shifts in markups,
the underlying causes, and more broadly how markets in our modern economy
are functioning and evolving.” In their view, regressions of market outcomes on
measures of concentration should carry little weight in policy debates because
they do not and cannot illuminate causal relationships. Syverson (2019) is
more optimistic that economy-wide studies can be helpful to identify patterns
of increasing concentration for further research, but he concludes that the
evidence does not yet support conclusions that rising aggregate market power
exists and is causing problematic trends in the economy. Like Berry, Gaynor,
and Morton (2019), Syverson (2019) calls for more careful research.
The airline industry provides an example where detailed, publicly available data have enabled insightful research. Werden and Froeb (2018) review
this literature to conclude that since deregulation in the late 1970s, studies
have not found systematic increases in concentration at the route level.
Berry, Carnall, and Spiller (2006) note that investments in hub-and-spokes
networks enabled airlines to earn high markups, but also benefited consumers.
Moreover, Berry, Gaynor, and Morton (2019) cite Borenstein (2011) to observe
that for many years, the large fixed costs associated with hub-and-spokes
Evaluating the Risk of Declining Competition | 215

250-840_text_.pdf 219

2/7/20 3:46 PM

networks were just offset by high markups, leaving the major airlines with
near-zero profits.
Other useful studies focus on how consummated mergers have affected
market outcomes. In these studies, the increase in concentration is explicitly
caused by a merging of competitors, so there is no question about why concentration has increased. For example, Ashenfelter, Hosken, and Weinberg (2015)
study the 2008 joint venture between the beer giants Miller and Coors. The
DOJ approved the deal, in part because it was expected to significantly reduce
the costs of shipping and distribution (Heyer, Shapiro, and Wilder, 2009).
Ashenfelter, Hosken, and Weinberg (2015) find little effect on prices, because
the efficiencies created by the merger nearly exactly offset the realized price
increases in the average market. However, in an analysis of the same market,
Miller and Weinberg (2017) find evidence that the joint venture may have
facilitated price coordination between competitors. These conflicting results
illustrate some of the important nuances related to competition that broad
industry studies cannot assess.
At this point, the evidence that the United States has a broad competition problem is inconclusive. However, the CEA’s 2016 report and the related
literature discussed above have spurred debate in government, academia, and
policy circles about ways to strengthen antitrust enforcement to deal with the
perceived competition problem. We now turn to this debate.

Calls for a Broader Interpretation
of Antitrust Policy
The 2016 CEA report, “Benefits of Competition and Indicators of Market
Power,” and the related literature discussed above are part of a broader movement that is concerned with the growth of large firms across the U.S. economy.
Lamoreaux (2019) provides a useful overview. Some of these observers want
to amend or rewrite the antitrust laws to expand the Federal Government’s
involvement beyond its traditional scope to consider outcomes unrelated to
market competition, including the political influence of large corporations,
control of advertising and news media, and rising income inequality. For
example, Furman and Orszag (2018) raise the question of whether a rising
share of firms earning “supernormal returns on capital” might increase wage
inequality due to workers at these firms sharing in the supernormal returns.
Also, as we discuss in the next section, some observers are calling for regulations specifically for the digital economy.
Other observers are focused on traditional antitrust law, but would
like enforcement to be expanded by lowering the threshold for an act to be
considered anticompetitive. For example, one Senate bill would change the
language of the Clayton Act, which prohibits mergers where the effect “may
be substantially to lessen competition.” The bill would change the standard of

216 |

250-840_text_.pdf 220

Chapter 6

2/7/20 3:46 PM

“substantially” to a standard of “materially.” This would mean that the Federal
Government could block a merger that has a smaller effect on competition
(U.S. Congress 2019a).
As we have discussed, the argument that the U.S. economy is suffering
from insufficient competition is built on a weak empirical foundation and
questionable assumptions. Antitrust law has evolved through careful development of its case law, based on the legal system’s accumulated experience
with enforcement actions and the effects of specific types of acts on industries
characterized by specific competitive dynamics. Throughout its development,
antitrust law has consistently proven flexible to the evolving market conditions presented by new industries and business models in the ever-changing
American economy. Before making radical changes to the law, the case for
such change should be better grounded.
Moreover, the antitrust laws are a poor tool for addressing issues that
go beyond questions of anticompetitive market conduct. Using antitrust law
to regulate markets in the absence of competition problems will exact costs
on the economy by preventing efficient market organization. If society wants
to pursue goals such as rising income inequality or the political power of large
firms, there are better policy tools to deal with these issues (Shapiro 2018).
We next turn to the related debate about whether more expansive antitrust enforcement is needed for the digital economy.

Antitrust Enforcement for the Digital Economy
In this section, we focus on the rapidly evolving issue of antitrust enforcement
and competition in the digital economy. In recent years, digital platforms
have come under increasing scrutiny. In the United Kingdom, the government
commissioned an expert panel to review competition policy for the digital
economy (Digital Competition Expert Panel 2019c). Since the panel made its
recommendations, the U.K. has been working to create its Digital Markets
Unit. The European Union has also commissioned an expert report (Crémer,
Montjoye, and Schweitzer 2019), and has introduced several regulations for
digital platforms.5
In the United States, the FTC has conducted hearings to examine whether
new technologies and business practices, including those associated with
digital platforms, require adjustments to competition policy (FTC 2019b).The
House and Senate Judiciary Committees have also held hearings related to
competition policy for digital platforms (U.S. House 2019a, 2019b, 2019c; U.S.
5 The U.K. Digital Markets Unit would develop and enforce regulations related to data
interoperability, data mobility, and data openness. It would have the authority to designate
certain platforms as having “strategic market status.” Such platforms would be subject to stronger
regulations. In July 2019, the European Union issued new regulations governing how platforms
interact with businesses (European Commission 2019). Rules on data portability and privacy,
known as the General Data Protection Regulation (GDPR), went into effect in 2018.

Evaluating the Risk of Declining Competition | 217

250-840_text_.pdf 221

2/7/20 3:46 PM

Senate 2019). Independently, the Stigler Center at the University of Chicago
has organized a committee on digital platforms that has developed recommendations for stronger antitrust enforcement and a digital regulator (Stigler
Committee on Digital Platforms 2019). The Agencies have also opened reviews
into market-leading online platforms, focusing on antitrust and related issues
(Bloomberg 2019; DOJ 2019).
Although this chapter focuses on competition concerns, we note that
some of these reviews also consider whether consumer protection regulations
are warranted for issues such as data privacy and the moderation of media
content.

Background
Digital platforms are intermediaries that enable interactions between users.
They include search engines, online market places, social networks, communication and media platforms, and home-sharing and ride-sharing services,
among other examples. Many of these platforms have been enormously successful and have reshaped the economy over the last 20 years.
Some concerns about digital platforms rest on the idea that they often
operate in markets with economic features that naturally tend toward high
concentration. One such feature is network effects, which arise when consumers place more value in a platform because many other people use it. For
example, the more people one can reach with a messaging service, the more
valuable that service is to users. When network effects are important, the largest platforms enjoy an advantage over their rivals simply because they have
more users, regardless of the quality of their services. In some cases, the advantage may be so great that other firms are unable to compete. For example,
in the videocassette recording industry, the Betamax technology essentially
disappeared after VHS technology pulled ahead (Werden 2001).
In markets with network effects or other types of economies of scale,
firms may compete for the entire market, rather than for shares in the market.
The resulting monopolies may not be permanent. Bourne (2019) gives many
examples of firms that achieved dominance through network effects or production economies of scale, only to eventually lose out to competition from
innovative rivals. His examples range from the Great Atlantic & Pacific Tea
Company in the 1920s to MySpace and Nokia in the early part of this century.
One of the current debates is about the extent to which digital platform
industries are characterized by high barriers to entry. A barrier to entry is an
obstacle that puts new firms at a disadvantage relative to firms already in the
market.6 Network effects can be a barrier to entry, particularly if an entrant
must simultaneously attract two groups of users. For example, in the payments
6 The formal definition of a barrier to entry has a long history of debate among economists. For a
discussion, see Werden (2001).

218 |

250-840_text_.pdf 222

Chapter 6

2/7/20 3:46 PM

industry, a new payment system might need to sign up thousands of merchants
before consumers see it as valuable, and vice versa. However, network effects
are not always sufficient to deter entry. If an entrant has an offsetting advantage, it may be able to overcome the advantage enjoyed by the established
platform. For example, when Microsoft introduced the Xbox platform for video
gaming, it was able to overcome the network effects enjoyed by the Sony
PlayStation 2 by focusing on a few blockbuster games (Lee 2013).
There is also a debate about the extent to which access to data can be a
barrier to entry. Mahnke (2015) discusses the issue in the context of the DOJ’s
2008 investigation of the merger of the media firms Thomson and Reuters. The
DOJ alleged that the merger would lead to higher prices for data sets related to
company fundamentals, earnings, and aftermarket research, and that entrants
would not be able to replicate the high quality of these data sets. The DOJ
approved the merger, but only after the parties agreed to divest copies of the
data sets along with supporting assets (DOJ 2008).
Data can also be a barrier to entry in the digital economy. Because
dominant platforms have more users, they often have access to much more
data than new entrants, and this can give them an insurmountable advantage
(Rubinfeld and Gal 2017). For example, dominant platforms may be better able
to target advertising at their users and so earn more revenues from advertising. However, a lack of access to data does not always deter entry. Lambrecht
and Tucker (2015) observe that Airbnb, Uber, and Tinder entered markets
where established firms (e.g., Expedia) had better data. They were able to succeed because of their innovative products. Lambrecht and Tucker (2015) also
observe that data are nonrivalrous, in the sense that data can be shared and
consumed by many users, in contrast to rivalrous goods such as food, which
are consumed only once. Because of this, entrants can sometimes buy data as
a substitute for collecting them internally from their users. However, this is not
always the case, and the role of data as a barrier to entry depends on the facts
and context of each market.
Finally, another debate asks whether dominant platforms are harming
competition by buying too many smaller firms, such as start-ups funded with
venture capital. It is common for large platforms to acquire smaller firms.
The digital economy relies heavily on innovation, and being acquired by an
established firm can be an important exit path for initial investors. Acquisition
can also be important for a start-up’s success. The acquiring firm may bring
marketing, financing, and other business assets that enable the start-up to
grow. However, if a start-up is not acquired, it might instead grow into an
independent, full-fledged competitor. Some acquisitions may occur precisely
to prevent such competition, as Cunningham, Ederer, and Ma (2019) find to

Evaluating the Risk of Declining Competition | 219

250-840_text_.pdf 223

2/7/20 3:46 PM

be the case in the pharmaceutical industry.7 However, as we discuss further
below, it can be challenging for the Agencies to assess whether acquisitions
of nascent competitors are procompetitive or anticompetitive in light of the
benefits associated with them.
In summary, many digital platform markets have demand and supply features, suggesting that high concentration is efficient. The concentration has led
to concerns about market dominance, anticompetitive behavior, and a lack of
competition. But concentration can also be efficient, and there may be robust
competition for the market, even in the face of high concentration.

Proposals for Intervention
Advocates of stronger regulation for digital platforms recommend a range of
measures encompassing both antitrust reform and regulation—see, for example, the Stigler Committee on Digital Platforms (2019); the Digital Competition
Expert Panel (2019c); and Crémer, Montjoye, and Schweitzer (2019). Here, we
consider proposals related to data portability and interoperability, acquisitions of nascent competitors, and the creation of a digital regulatory authority.
Data portability and interoperability. Proposals to increase data portability and interoperability involve new regulations and legislation. Portability
regulations would require digital platforms to enable customers to access their
data from different platforms on request. Interoperability legislation would
require digital platforms to enable their customers to switch their data from
one platform to another. For example, a bill recently proposed in the Senate
would require large communication platforms that generate income from
their users’ data to enable data portability and interoperability with other
communication platforms. The goal is to reduce entry barriers for competitors
to these platforms by making it less costly for customers to switch from one
platform to another, and also by allowing customers of dominant platforms to
communicate easily with customers of rival platforms (U.S. Congress 2019b).
As with any regulation, however, this would impose costs on the regulated platforms. Jia, Jin, and Wagman (2019) study the effect of the recent
rollout of rules on data privacy and portability in Europe, known as the General
Data Protection Regulation (GDPR), on venture capital funding. They find negative effects on European firms relative to their U.S. counterparts in terms of
total funding, the number of deals and the amount raised per deal, with more
pronounced effects for newer and data-related firms.
Acquisitions of nascent competitors. As discussed above, proponents of
stronger antitrust enforcement raise concerns that dominant platforms are
protecting themselves by acquiring small firms that would otherwise develop
7 In a study of the pharmaceutical industry, Cunningham, Ederer, and Ma (2019) conclude that
about 6 percent of acquisitions in their sample were “killer acquisitions” that forestalled the
development of new drugs that would otherwise have competed with the acquirer’s existing
products.

220 | Chapter 6

250-840_text_.pdf 224

2/7/20 3:46 PM

into future competitors. Antitrust law has an existing framework to challenge
such mergers under theories of potential competition and disruptive entrants
(DOJ and FTC 2010). In 2018, the FTC challenged a merger between CDK Global
and Auto/Mate. The acquiring firm, CDK, was a market leader in specialized
business software for franchise automotive dealers. Auto/Mate was a much
smaller competitor with an innovative business model that was an emergent
threat. Although Auto/Mate was already competing, the FTC was largely concerned about the competition it would likely provide in the future (FTC 2018b;
Ohlhausen 2019).
Predicting future competition can be difficult in the digital economy
because products and services evolve rapidly. Dominant platforms may
acquire start-ups that offer no competing products, but that could compete
with them in the future through expansion into adjacent markets. To address
this issue, some proposals for revising the antitrust laws would weaken the evidentiary standards when a dominant firm seeks to acquire a firm in a separate
but adjacent market. For example, the Agencies might meet their initial burden
of proof by showing that there is a reasonable likelihood that the target firm
would compete with the acquiring firm in the future, even if the target firm has
no specific plans to do so (Shapiro 2019).
Such policies could have important downsides. More aggressive standards for blocking mergers of nascent competitors would raise the likelihood
that procompetitive mergers would be blocked. As discussed above, the digital
economy relies heavily on innovation. If dominant platforms were routinely
deterred from acquiring start-ups, such a policy could reduce venture capital
funding in this segment. During the U.K. panel review, a variety of organizations
and individuals raised these concerns (Digital Competition Expert Panel 2019a,
2019b). At a minimum, the potential effect of any new policy on venture capital
deserves study. More research, including merger retrospectives focused on
acquisitions in the digital economy, would also be helpful.
Creation of a digital regulatory authority. The Stigler Committee on Digital
Platforms (2019) found that “the strongest indication emerging from the four
reports is the importance of having a single powerful regulator capable of
overseeing all aspects of [digital platforms].” In terms of competition goals,
the digital regulator would have a mandate to design and enforce regulations
aimed to enhance competition, such as standards for data portability and
interoperability. The authority would be able to designate dominant platforms
as “bottlenecks” and subject them to stronger regulations. For example, such
platforms might need to obtain approval from the authority for any acquisition,
no matter how small, and the digital authority would be able to challenge these
acquisitions under a legal standard that imposes a lower burden of proof on
the Agencies than does current antitrust law.
The Stigler Committee on Digital Platforms (2019) also makes recommendations that fall outside antitrust and competition policy. A subcommittee
Evaluating the Risk of Declining Competition | 221

250-840_text_.pdf 225

2/7/20 3:46 PM

on politics, in particular, recommends that a digital authority have the power
to take actions to limit concentration, not due to concerns about economic
harm to consumers, but due to concerns about the political power of large
platforms. A subcommittee on data privacy and security recommends that a
digital authority oversee consumer protection regulation that would develop,
among other regulations, rules similar to the GDPR in Europe.
Proposals to establish a new digital authority raise a host of issues. A
basic concern is that the breadth of the mandate is far from obvious. As noted
above, digital platforms provide a wide-ranging set of goods and services,
from search engines, to operating systems, to ride-sharing services. The Stigler
Committee on Digital Platforms (2019) points to the Federal Communications
Commission (FCC) as a model for a digital regulator, but the scope of the FCC’s
authority is the telecommunications sector. The scope of a digital authority
would likely be harder to delineate, and firms in some of the most innovative
sectors of the economy would face uncertainty as to whether they fall under
its regulations.
Perhaps the most serious concern is for the possibility of regulatory capture. In a speech, FCC chair Ajit Pai (2013) relays a cautionary tale of FCC regulatory capture, describing how AT&T made commitments to the FCC in 1913 that
effectively allowed it to divide up territories with independent local telephone
companies. These commitments tamed competition that had emerged after
the patents of Alexander Graham Bell began to expire. The Stigler Committee
on Digital Platforms (2019) discusses the need to deter regulatory capture and
cites Pai’s speech. It also cites the foundational work on regulatory capture
by the Nobel laureate economist George Stigler, for whom the Stigler Center
is named. Though there is some irony here, the point is that the downsides of
new, far-reaching regulation need to be taken seriously.
Although today’s digital economy warrants further study—and, where
necessary, vigilant antitrust enforcement—a cautious approach to regulation
is clearly warranted. As we have discussed, there is a fundamental problem
in inferring that high concentration is indicative of a lack of competition. The
nature of competition also varies across markets, so one-size-fits-all policies
may not work well. Instead, fact-specific investigations along the lines of what
the Agencies already do are more sensible.

Competition Policy to Reduce Entry Barriers
In the preceding sections, we have argued for caution in responding to calls
for Federal Government intervention to address increasing concentration in
the U.S. economy. However, it is true that entry barriers can protect firms from
competition. Sometimes, these entry barriers are structural, in that they are
associated with the nature of the market itself, such as products that require
large investments in research and development. In other cases, entry barriers

222 | Chapter 6

250-840_text_.pdf 226

2/7/20 3:46 PM

Box 6-4. The Effects of Deregulation within
the Pharmaceutical Drug Market
As noted, some barriers to entry are purposefully constructed. To illustrate, consider the pharmaceutical drug industry, where the Food and Drug
Administration (FDA) plays a crucial role in supplying drugs through the
management of drug application reviews. The FDA ultimately determines
if and when a drug will be available on the market. Although the stringent
evaluations conducted by the FDA are necessary to ensure the safety and
efficacy of drugs, they are also partly responsible for raising entry barriers for
many generic and new drugs. This has led to a higher concentration of brand
name drugs in some markets, along with higher prices that reduce consumer
welfare.
The Trump Administration realizes the significance of improving competition in markets for pharmaceutical drugs, and it has implemented a series
of deregulatory reforms with the hope of reducing costs for consumers. One of
its proposals highlights the need for the transparency of negotiated discount
rates with insurers, requiring hospitals to disclose this information to their
patients (CEA 2018a). The Administration also signed the Food and Drug
Administration Reauthorization Act in 2017, which reauthorized the FDA to
collect user fees from generic drug applications and to process applications
efficiently for another five years. Since the start of the Trump Administration,
aspects of the FDA’s drug application process, most prominently that for
generic drugs, have been streamlined to encourage quick market entry. In the
first 20 months of the Administration, an average of 17 percent more generic
drugs were approved each month than were approved during the previous
20-month period (CEA 2018b).
In 2018, the FDA expanded its Strategic Policy Roadmap in efforts to
not only increase efficiencies in the drug review process but also reduce anticompetitive behavior from brand name drug makers that try to inhibit generic
market entry. The FDA is also taking steps to address scientific and regulatory
barriers that are obstacles to entry of some complex generic medicines. The
FDA’s efforts to lower barriers and have a more predictable and efficient
development process may enable new and innovative drug makers to enter
the market. Consumers would benefit both from the development of new
classes of drugs and from new therapies for conditions treated by existing
drugs. Such new therapies could discipline the prices of existing drugs. This
was the case for drugs such as simvastatin, which held a large portion of the
market for lowering cholesterol in the 1990s. However, starting in 1996, after
the introduction of the therapy drug atorvastatin, competition flourished, and
cholesterol-lowering drugs are now affordable (CEA 2018b).

are purposefully constructed by governments in situations where private markets may fail; see box 6-4. However, as discussed in chapter 3 of this Report,

Evaluating the Risk of Declining Competition | 223

250-840_text_.pdf 227

2/7/20 3:46 PM

even if a regulatory action addresses a private market failure, a deregulatory
action is still warranted if the costs of the regulation outweigh the regulatory
benefits. This section describes the Agencies’ efforts to call attention to regulations that harm consumers by creating entry barriers that limit competition.
It also discusses how the Agencies apply the antitrust laws to intellectual
property rights to promote sound competition.

Other Government-Created Barriers to Entry
As we discuss in chapter 2 of this Report, occupational licensing requirements
impose an additional cost on a person entering a given occupation. Some
licensing requirements may be justified on public safety grounds; but in many
professions, they also function as barriers to entry that artificially inflate wages
by protecting those already in the profession from competition. To support
the claim that the majority of State occupational licensing requirements are
unnecessary to protect public safety, the FTC points out that 1,100 occupations
require a license in at least one State but only 60 occupations are licensed by
every State. If an occupation poses a substantiated threat to public safety,
the argument goes, then that occupation would be universally licensed (FTC
2018a, 2019c).
The Agencies have long advocated measures to limit the competitive
harm associated with occupational licensing. In 2017, the FTC established a
task force on the issues, and in 2018, it released a report outlining options to
mitigate the harm. These options include interstate pacts that allow groups of
States to recognize a common license, as well as other portability and mutual
recognition measures (FTC 2018a).
Certificate-of-need (CON) laws were originally designed in the 1970s to
discourage overinvestment in healthcare markets (e.g., building too many
hospitals) in an attempt to limit costs. A CON law requires a firm to convince
a State regulator that there is an unmet need for the new services. Over years
of review, the Agencies have found that these laws often harm competition,
and they regularly advocate for their removal. In 2019, for example, staff at
the Agencies sent letters to legislatures in Alaska and Tennessee in support
of their plans to revise these laws (DOJ and FTC 2019b, 2019d). The Agencies’
analysis of evidence, accumulated over decades, finds that instead of reducing
healthcare costs, CON laws tend to create inefficiencies by suppressing healthcare supply to the benefit of established suppliers, preventing investment that
would stimulate competition and lower consumer prices.
Many States require car manufacturers to distribute vehicles through
independent, franchised dealerships. The Agencies have long advocated
against such automobile franchising laws. They argue that when manufacturers are free to choose their method of distribution, the competitive process
aligns their interests with those of consumers, so the products and services are
brought to market as efficiently as possible. In 2019, Nebraska took up a bill

224 | Chapter 6

250-840_text_.pdf 228

2/7/20 3:46 PM

that would remove restrictions on direct vehicle sales to consumers, but only
for vehicle manufacturers that had not used independent, franchised dealers
in the State before. The Agencies sent a joint letter to the Nebraska Legislature
encouraging it to remove the restrictions for all vehicle manufacturers (DOJ
and FTC 2019c).

Promoting Innovation through Sound Enforcement of
Competition Law
As we have discussed, consumers often benefit most from dynamic competition, as driven by investment and innovation in new products, inventions, and
technologies. Intellectual property rights—such as patents, trademarks, and
copyrights—limit competition from infringing products in order to encourage
this dynamic competition. However, in certain circumstances, intellectual
property rights, like any asset, may be used in a manner that unlawfully limits
competition. To prevent this, the Agencies apply the same antitrust principles
to conduct involving intellectual property as they do to conduct involving other
forms of property (DOJ and FTC 2017). They apply an effects-based economic
analysis to conduct involving intellectual property that considers its efficiencies and weighs procompetitive benefits of the conduct against any competitive harm. The Agencies also engage in advocacy for the correct application of
antitrust law to intellectual property rights.
The DOJ has emphasized the need to avoid rigid presumptions in the
intellectual property area that could deter innovation. In particular, it has
cautioned against the misapplication of antitrust laws, which carry the specter
of treble damages, to commercial disputes involving the exercise of patent
rights. In December 2017, the DOJ withdrew its support from its 2013 joint
policy statement with the Patent and Trademark Office on remedies associated
with standard essential patents, because the statement had been construed to
suggest that the antitrust laws should limit patent holders from seeking injunctions or exclusionary remedies to defend their intellectual property rights. The
DOJ’s work in this area ensures that there are strong incentives to invest in
developing technologies, and thus fostering dynamic competition.
A top priority of the FTC is to oppose “pay-for-delay” patent settlements,
whereby branded drug manufacturers pay generic drug producers to stay out
of the market. In 2013, in FTC v. Actavis, Inc., the Supreme Court held that,
in certain circumstances, the FTC can challenge such settlements under the
antitrust law, provided that courts weigh anticompetitive effects against the
procompetitive benefits of such conduct. Since that year, the FTC has regularly
reported on these settlements. In its most recent report, the FTC found that the
number of pay-for-delay payments of the type that are likely to be anticompetitive has been decreasing (FTC 2019a).

Evaluating the Risk of Declining Competition | 225

250-840_text_.pdf 229

2/7/20 3:46 PM

Conclusion
The Trump Administration understands the vital role that competition plays in
the economy, promoting new businesses and serving consumers. Timely antitrust enforcement is an important tool for protecting the competitive process.
By contrast, confusion surrounding the effects of rising concentration appears
to be driven by questionable evidence and an overly simple narrative that “Big
Is Bad.” When companies achieve scale and large market share by innovating
and providing their customers with value, this is a welcome result of healthy
competition.
This chapter has explained why recent calls for changing the goals
of the antitrust laws and expanding the scope of regulations are based on
inconclusive evidence that competition is in decline. These calls also ignore
the flexibility of the existing legal system to accommodate changing market
circumstances. Research purporting to document a pattern of increasing
concentration and increasing markups uses data on segments of the economy
that are far too broad to offer any insights about competition, either in specific
markets or in the economy at large. Where data do accurately identify issues
of concentration or supercompetitive profits, additional analysis is needed
to distinguish between alternative explanations, rather than equating these
market indicators with harmful market power.
Antitrust actions and any major changes to competition policy should be
based on sound economic evidence, including evidence on consumer harm.
Research based on broad industry studies may be helpful for indicating trends
in concentration, but is unable to diagnose the underlying causes or determine whether consumers in relevant antitrust markets have been harmed.
Ultimately, today’s detailed, evidence-based approach to antitrust remains the
most powerful lens available to protect consumers and suppliers by accurately
diagnosing and responding to anticompetitive behavior.
For these reasons, this chapter argues that the DOJ’s Antitrust Division
and the FTC are well-equipped to protect consumers from anticompetitive
behavior. The Agencies have maintained their focus on illegal or anticompetitive actions by businesses, while expanding their scope to advocate
against government policies that harm competition. Vigorous competition is
essential for building upon the economy’s record expansion, and the Trump
Administration will continue following the economic evidence and using the
Federal Government’s authority to promote competition in ways that lead to
greater consumer benefits.

226 |

250-840_text_.pdf 230

Chapter 6

2/7/20 3:46 PM

x
Chapter 7

Understanding the Opioid Crisis
The opioid crisis poses a major threat to the U.S. economy and America’s public
health. Since 2000, more than 400,000 people have lost their lives because
of opioids. This staggering number of deaths has pushed drug overdoses to
the top of the list of leading causes of death for Americans under the age of
50 years, and has cut 2.5 months from U.S. life expectancy. The Council of
Economic Advisers (CEA) has previously estimated that the annual economic
cost of the opioid crisis is substantially higher than previously thought, at over
half a trillion dollars in 2015. Using a similar methodology, the CEA estimates
that the crisis cost $665 billion in 2018, or 3.2 percent of gross domestic product.
There are signs that the opioid crisis is past its peak because the growth in
opioid overdose deaths has stopped during the Trump Administration, stopping the upward trend that has persisted since at least 1999. From January
2017 through May 2019, the CEA estimates that there were 37,750 fewer opioid
overdose deaths—representing an economic cost savings of over $397 billion—
relative to the number of deaths expected based on previous trends. Actions
taken by the Trump Administration to reduce the supply of opioids, reduce new
demand for opioids, and treat those with current opioid use disorder may have
contributed to the flattening in overdose deaths involving opioids.
The Trump Administration understands that the crisis is ongoing and that
there is much more work to do to combat this threat to American lives and the
American economy. In order to continue mitigating the cost of the opioid crisis,
it is crucial to understand all its underlying factors. We describe and analyze
two separate waves of the crisis—the first wave, from 2001 to 2010, which was
characterized by growing overdose deaths involving the misuse of prescription

227

250-840_text_.pdf 231

2/7/20 3:46 PM

opioids; and the second wave, from 2010 to 2016, which was characterized by
growing overdose deaths involving illicitly manufactured opioids (heroin and
fentanyl).
We find that in the first wave, between 2001 and 2010, out-of-pocket prices
for prescription opioids declined by an estimated 81 percent. This dramatic
drop in prices was a consequence of the expansion of government healthcare
coverage, which increased access to all prescription drugs—including opioids.
We argue that these falling out-of-pocket prices effectively reduced the price
of opioid use in the primary market and in the secondary (black) market for
diverted opioids, from which most people who misuse prescription opioids
obtain their drugs. We estimate that the decline in observed out-of-pocket
prices is capable of explaining between 31 and 83 percent of the growth in the
death rate involving prescription opioids from 2001 to 2010.
However, falling out-of-pocket prices could not have led to a major rise in opioid
misuse and overdose deaths without the increased availability of prescription
opioids resulting from the new specialty of pain management, the creation of
pain management practices that encouraged liberalized dispensing practices
by doctors, illicit “pill mills,” increased marketing and promotion efforts from
industry, and inadequate monitoring or controls against drug diversion. The
subsidization of opioids is in stark contrast to the taxation of other addictive
substances such as tobacco and alcohol. The dilemma this poses is how to
make available the appropriate medical use of opioids for pain relief while
preventing nonmedical use of subsidized products.
We find that the second wave of the opioid crisis likely started in 2010 because
of efforts to limit the misuse of prescription OxyContin, enabling a large market
for the sale and innovation of illegal opioids. Although these efforts eventually
successfully reduced prescription opioid-involved overdose deaths, they had
the unintended consequence of raising demand for cheaper substitutes in the
illicit market among misusers of prescription drugs. An expansion in foreignsourced supply was also important for the growth of illicitly manufactured

228 |

250-840_text_.pdf 232

Chapter 7

2/7/20 3:46 PM

opioids, as evidenced by falling quality-adjusted prices, largely due to expanded
heroin trafficking from Mexico and relatively inexpensive synthetic opioids
from both Mexico and China, specifically fentanyl and its analogues, which can
be many times more potent than heroin.1

T

he Trump Administration has undertaken serious efforts to tackle
the ongoing opioid crisis that continues to threaten the American
economy and American lives. This is demonstrated by the declaration
of the opioid epidemic as a public health emergency, the establishment of the
President’s Commission on Combating Drug Addiction and the Opioid Crisis,
the highest expenditures in history directed toward the opioid epidemic, and
ongoing efforts throughout the Federal government to address the crisis. The
damage resulting from the opioid crisis is dramatic in its proportions compared
with other health crises. For example, in 2017, the number of people who died
of an opioid-involved drug overdose (47,600) exceeded the number of deaths
from the HIV/AIDS epidemic at its peak in 1995 (CDC 2019).2 Additionally,
since 2000, the United States has lost as much of its population to the opioid
crisis as it lost to World War II—with both causing more than 400,000 fatalities
(DeBruyne 2017). This staggering number of deaths has pushed drug overdoses
to the top of the list of leading causes of death for Americans under the age of
50 years, and has cut 2.5 months from U.S. life expectancy (Dowell et al. 2017).
To assess the full damage caused by this crisis, the CEA has previously
assessed its full economic cost. In 2015 alone, the CEA estimated that the total
cost of the opioid crisis was $504 billion, several times larger than previous cost
estimates (CEA 2017). The CEA’s approach constituted a more complete assessment of the costs because it incorporated the full cost of increased morbidity
and mortality from the crisis. We also adjusted opioid-involved deaths—which
had been underreported—upward and incorporated nonfatal costs. Using
similar methods as in the earlier CEA assessment, the annual cost of the opioid
crisis has only risen since 2015, amounting to $665 billion in 2018. The annual
number of reported opioid-involved overdose deaths increased from 33,091 in
2015 to 47,600 in 2017, a 44 percent increase. According to preliminary data,
deaths have since decreased slightly in 2018, an indication of a flattening in

1The CEA previously released research on topics covered in this chapter. The text that follows
builds on this research paper produced by the CEA: “The Role of Opioid Prices in the Evolving
Opioid Crisis” (CEA 2019b).
2 We identify overdose deaths throughout the report using the 10th revision of the International
Statistical Classification of Diseases and Related Health Problems (ICD-10) underlying cause-ofdeath classification codes: X40–X44 (unintentional), X60–X64 (suicide), X85 (assault), and Y10–Y14
(undetermined). Deaths involving opioids are identified using ICD–10 multiple cause-of-death
classification codes: T40.0–T40.4 and T40.6.

Understanding the Opioid Crisis | 229

250-840_text_.pdf 233

2/7/20 3:46 PM

the trend of increasing annual deaths that has persisted since 1999 (see figure
7-1).3
When President Trump took office in January 2017, monthly overdose
deaths involving opioids had reached an all-time record high, a 41 percent
increase from the number of deaths 12 months earlier, in January 2016.
Since then, the growth in opioid deaths may have finally stopped. Monthly
overdose deaths fell by 9.6 percent between January 2017 and May 2019, the
latest month for which provisional data are available (see figure 7-1). If the
growth rate in opioid overdose deaths from 1999 through 2016 had continued,
37,750 additional lives would have been lost due to opioid overdoses between
January 2017 and May 2019, a 33 percent increase over the actual number of
deaths that occurred over this period. The economic cost savings since January
2017 from reduced mortality compared with the preexisting trend was over
$397 billion.4
In order to continue mitigating the large costs imposed by the opioid crisis through appropriate policy measures, it is crucial to understand the forces
that underlie it. We separate our analysis into two sections: The first one analyzes the first wave of the crisis, lasting through 2010, which was characterized
by growth in prescription opioid-involved overdose deaths; and the second
analyzes the period since 2010, which has been characterized by growth in
illicit opioid-involved overdose deaths.5
During the first wave, between 2001 and 2010, the annual populationbased rate of overdose deaths involving prescription opioids increased by 182
percent (CDC WONDER n.d.). Throughout this period, opioid manufacturers
aggressively promoted the safety and effectiveness of opioids, and guidelines
for the treatment of pain were liberalized to encourage physicians to prescribe
3 Official estimates of opioid-involved overdose deaths are extracted from the CDC’s WONDER
Multiple Cause of Death Database (https://wonder.cdc.gov/mcd.html). As of December 31, 2019,
official data were available through December 2017. Preliminary estimates of opioid-involved
overdose deaths are extracted from Ahmad et al. (2019). The provisional data include deaths of
foreign residents and include approximately 500 additional drug overdose records compared with
data from CDC WONDER that is limited to residents of the United States.
4 The number of lives saved is calculated from the difference between the projected trend in
deaths from January 2017 to May 2019, the most recent month of preliminary data as of December
31, 2019 (see figure 7-1). The calculated number of lives saved is sensitive to the assumption that
the projected trend is nonlinear. We use the value of a statistical life to estimate the value of lives
saved, adjusting the Department of Transportation’s value of a statistical life to about $10.5 million
in 2018 dollars (DOT 2016).
5 We use “illicit opioids” throughout the chapter to refer to illicitly produced opioids such as heroin
and fentanyl, which excludes the misuse of prescription opioids such as OxyContin. It is important
to note that data on overdose deaths do not distinguish between illicitly manufactured synthetic
opioids, such as illicitly manufactured fentanyl, and synthetic prescription opioids, such as
prescription fentanyl. This analysis includes this broader category of synthetic opioids other than
methadone in the illicit opioid category, given that illicitly manufactured fentanyl is commonly
believed to have dominated this category in recent years, and that the category was much less
important in the earlier years of the crisis.

230 | Chapter 7

250-840_text_.pdf 234

2/7/20 3:46 PM

Figure 7-1. Opioid-Involved Overdose Deaths, 1999–2019
Monthly number of deaths

Inauguration
May-19

7,000
6,000
5,000

Trend

4,000
3,000
2,000
1,000
0
1999

2003

2007

2011

2015

2019

Sources: Centers for Disease Control and Prevention (CDC); CEA calculations.
Note: Data from before January 2018 are compiled from the CDC WONDER database, and monthly
data beginning in January 2018 are calculated using the provisional reported number of deaths
from the CDC. The preinauguration trend is calculated for January 1999 to January 2017. Shading
denotes a recession.

more opioids (Van Zee 2009). Over the same period, we estimate that the
out-of-pocket price of prescription opioids fell by 81 percent (see also Zhou,
Florence, and Dowell 2016). We argue that the falling out-of-pocket price translated into a lower price of misuse not only for those who obtain prescriptions in
the primary market but also for the majority of misusers who obtain prescription opioids from the secondary (black) market.
The decline in out-of-pocket prices between 2001 and 2010 occurred
in conjunction with a rising share of generic opioids in the market as well
as increased public subsidies. Though we do not attempt to apportion their
respective roles, these two factors may have contributed significantly to the
out-of-pocket price decline. With regard to a rising generic share in the prescription opioid market, we note that supply prices paid to pharmacies fell
by 45 percent between 2001 and 2010, fueled by an increase in the cheaper
generic opioid share, from 53 percent to 81 percent.
In addition, we document a large increase in the share of prescription
opioids funded by public programs. As shown in figure 7-2, the share of prescribed opioids purchased with public subsidies increased from 17 percent in
2001 to 60 percent in 2010, rising further to 63 percent in 2015. Public programs
accounted for three-fourths of the growth in total prescription opioids between
2001 and 2010 (data from the Medical Expenditure Panel Survey, MEPS). The
introduction of the Medicare Part D prescription drug benefit in January 2006

Understanding the Opioid Crisis | 231

250-840_text_.pdf 235

2/7/20 3:46 PM

Figure 7-2. Share of Potency-Adjusted Prescription Opioids, by
Primary Payer, 2001–15
Share (percent)
100

2015

Self
80

Private

60
40

Public

20
0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: Medical Expendtiure Panel Survey; National Drug Code Database; CEA calculations.
Note: The primary payer is the third-party payer with the highest payment for a given prescription.
In addition to Medicare, Medicaid, and private insurers, the other possible primary payers include
veterans’ benefits, workers’ compensation, other Federal government insurance, other State or
local goverment insurance, or other public insurance. All prescriptions are converted into
morphine gram equivalents based on the quantity of pills prescribed and their potency.

coincided with a growing share of prescriptions reimbursed by the program,
including for many opioids. Additionally, Social Security Disability Insurance
(SSDI) enrollment has rapidly increased since the late 1990s (see figure 7-16).
More than half of SSDI recipients received drug coverage before the 2006 start
of Medicare Part D through Medicaid and other programs. After 2006, SSDI
recipients, along with the general Medicare population, were for the most part
eligible for prescription drug coverage through Medicare Part D.
Expansions in insurance coverage that reduce out-of-pocket prices make
misused prescription opioids more affordable for patients with prescriptions
and users who purchase the drugs on the secondary market. Before generics were as widely available, it was very costly for the average American with
opioid use disorder to afford prescription opioids, if not subsidized through
insurance. In 2007, Americans could buy 1 gram of OxyContin—one of the
most common brand name opioids prescribed—for an average of $144 without
health insurance. Some individuals on opioids may require up to a gram or
more per day of OxyContin for pain relief (Schneider, Anderson, and Tennant
2009). Without insurance, a person with an opioid use disorder consuming
between 0.5 gram and 1 gram of OxyContin every day for a year would have
spent between $26,280 and $52,560 in 2007—which could be more than the
median household income of about $50,000 in 2007 (in 2007 dollars) (Fontenot,

232 | Chapter 7

250-840_text_.pdf 236

2/7/20 3:46 PM

Semega, and Kollar 2018).6 To put this in perspective, a person on Medicare
would only pay $9.78 per gram, or between $1,785 and $3,570 per year (in 2007
dollars), to support an opioid use disorder in the same year.
The subsidization of opioids is in stark contrast to the taxation of other
addictive substances such as tobacco and alcohol. The challenge this poses is
how to ensure access to opioids for legitimate medical needs, such as for pain
relief, when other substances are contraindicated or insufficient, while not
subsidizing nonmedical uses.
Given the role the government played in subsidizing the purchase of
prescription opioids through the expansion of health insurance, we examine
the possible roles of specific public programs. We find that the number of
potency-adjusted opioids per capita subsidized by Medicare increased by
2,400 percent between 2001 and 2010, the largest increase among all thirdparty payers. SSDI rolls also expanded over this period. We estimate that SSDI
recipients, who are generally eligible for Medicare (including prescription
coverage in Part D, starting in 2006), were prescribed a disproportionate share
of 26 to 30 percent of total potency-adjusted opioids in 2011 across all payer
types (while representing under 3 percent of the U.S. population). Of course,
any role of SSDI expansion in the opioid crisis would be attributable to the
design of the program rather than program recipients. SSDI recipients generally have debilitating conditions that prevent them from working, and these
conditions are often associated with high levels of pain. These conditions are
the primary reason SSDI recipients are prescribed a disproportionate share of
opioids; indeed, SSDI benefits, in conjunction with Medicare coverage, provide
vital protection for these disabled workers. Additionally, the majority of SSDI
recipients prescribed opioids use them appropriately and do not contribute to
opioid misuse directly or indirectly.
As a calibration exercise, we take published estimates of the price elasticity of prescription opioid sales to estimate the increase in sales resulting
from an 81 percent price decline. This exercise suggests that, without the
price decline, per capita opioid sales would have increased by half as much
or less than the actual increase between 2001 and 2010. In order to estimate
the size of the price decline as a factor in the increase in the number of deaths
involving prescription opioids, we assume that (1) secondary market prices are
proportional to out-of-pocket prices in the primary market, and (2) the price
elasticity of opioid use ranges from the elasticity of prescriptions at the low end
to the own-price elasticity of heroin use at the high end. This second calibration
6 Due to heightened risk to patients, the CDC recommends that physicians avoid prescriptions at
or above 90 morphine milligram equivalents per day, equivalent to 60 milligrams of oxycodone or
0.06 gram, or carefully justify a decision to titrate dosage to 90 or more milligram equivalents per
day (CDC n.d.). Schneider, Anderson, and Tennant (2009) observe that some chronic pain patients
require doses that may range from 1,000 to 2,000 or more milligram equivalents per day. These
doses would be equivalent to 667 to 1,333 milligrams (0.7 to 1.3 grams) of oxycodone per day.

Understanding the Opioid Crisis | 233

250-840_text_.pdf 237

2/7/20 3:46 PM

exercise suggests that the observed decline in out-of-pocket prices for prescription opioids, which makes physicians’ prescriptions more affordable for
beneficiaries to fill, was a factor in between 31 and 83 percent of the increase in
overdose deaths involving prescription opioids between 2001 and 2010.
However, falling out-of-pocket prices could not have led to a major rise
in opioid misuse and deaths without the increased availability of prescription
opioids resulting from changes in pain management practice guidelines that
encouraged liberalized dispensing practices by doctors, illicit “pill mills,”
increased marketing and promotion efforts from industry, and inadequate
monitoring or controls against diversion. Without these factors, patients would
have been unable to respond to lower prices by obtaining prescription opioids
and diverting them to the secondary market. In other words, the change in
the environment for obtaining prescription opioids was a precondition for the
effect of falling out-of-pocket prices on opioid misuse. In addition, it is important to emphasize that the falling price of the medical use of opioids—due to
expanded insurance coverage and generic entry—benefited patients because
they could access needed drugs at a lower out-of-pocket cost. By contrast,
the falling price of the nonmedical use of opioids, enabled by a lax prescribing
environment in conjunction with lower out-of-pocket prices, may have played
an important role in fueling the opioid crisis.
More generally, these findings of increased opioid misuse associated with
the growth of public programs do not imply that these programs lack social
value, but rather show the importance of instituting safeguards to ensure the
appropriate prescribing and use of opioids, and measures to reduce the misuse
of opioids.7 Government policy for other addictive products, such as cigarettes,
deliberately discourages consumption by raising prices through sales taxes
and placing restrictions on purchase and sales; most analysts agree that such
policies successfully reduced cigarette use and made new addiction cases
less likely (HHS 2014). Unlike cigarettes, which are not safe or beneficial for
anyone in any quantity, opioids have legitimate medical uses. The challenge of
prescription opioids is balancing the goal of subsidizing opioids when they are
prescribed for appropriate use with the need to discourage overprescription
and misuse.
Next, we analyze the second wave of the opioid crisis, which was characterized by the growth of illicit, opioid-involved overdose deaths between
2010 and 2016. In this case, demand-side expansions due to efforts to curtail
prescription opioid use disorder along with supply-side expansions appear to
have been important. Most notably on the demand side, an abuse-deterrent
formulation of the widely abused prescription opioid OxyContin was released
in 2010, and the original formulation was no longer made available from the
manufacturer. Research has found that although the reformulation stemmed
7 See HHS (2016) for further discussion.

234 | Chapter 7

250-840_text_.pdf 238

2/7/20 3:46 PM

Figure 7-3. Opioid-Involved Overdose Death Rate by the
Presence of Prescription Opioids, 2001–16
Deaths (per 100,000)
8

Second wave
2016

7
6

Involved prescription
opioids

5
4
3

Did not involve
prescription opioids

2
1
0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: CDC WONDER; CEA calculations.
Note: Prescription opioids include both natural and semisynthetic opioids (T40.2) and also
methadone (T40.3).

the rise of overdose deaths involving prescription opioids, it led opioid misusers to substitute toward cheaper, more available heroin, resulting in increased
heroin-involved deaths (Alpert, Powell, and Pacula 2018; Evans, Lieber, and
Power 2019). Thus, the buildup of a pool of people with addictions to prescription opioids during the first wave ultimately facilitated the increase in demand
for illicit opioids in the second wave. This large pool of new demand created
additional profit opportunities for illegal sellers entering the market. Supply
increased as Mexican heroin traffickers increased shipments to the United
States in response to shrinking markets for cocaine, and other foreign manufacturers—especially in China—introduced cheaper and more potent synthetic
opioids like fentanyl. Figure 7-3 illustrates how overdose deaths involving
prescription opioids leveled off after 2010, while other opioid deaths (those
only involving illicit opioids and possibly nonopioid drugs) escalated rapidly.
In an attempt to assess the relative importance of demand and supply
expansions in driving the second wave of the opioid crisis, we estimate the
price of illicit opioids over time. Though these estimates are subject to a number of highly imperfect assumptions, we find that the price of illicit opioids was
roughly constant between 2010 and 2013, before falling by about half by 2016,
due to the increased supply of illicit fentanyl (see figure 7-17) starting in about
2013 (increasingly available via shipment from China and from other foreign
sources). Given the extreme potency and low cost of fentanyl, it dramatically
reduced the “cost of a high” for users. It is notable that even though demand for

Understanding the Opioid Crisis | 235

250-840_text_.pdf 239

2/7/20 3:46 PM

illicit opioids increased beginning in 2010, the price of illicit opioids remained
constant until about 2013, implying that in these first years of the illicit wave,
the heroin supply must have also expanded to keep prices steady; if supply
had remained constant, prices would have risen. Falling prices between 2013
and 2016 imply that supply expansions of illicit opioids were more important
drivers of the crisis in these later years.
Due to constraints on data availability for prices of both prescription
and illicit opioids, this analysis focuses on the period ending in 2016. However,
provisional mortality data are available through part of 2019.
The rest of the chapter proceeds as follows. The next section presents our
basic methodology in assessing how demand, supply, and government policies
can affect quantities and prices of opioids. The subsequent section analyzes
the first wave of the crisis based on prescription opioids, and the section after
that analyzes the substantial growth in public subsidies for opioids during this
period. The last section turns to the second wave, which spawned the rise of
illicit opioids.

The Supply-and-Demand Framework
Although we cannot quantify the extent to which government-subsidized drugs
are diverted and resold for nonmedical use, a simple supply-and-demand
framework can provide powerful insights into how changing prices and quantities reflect the underlying forces driving the opioid crisis. Figures 7-4 and 7-5
consider the case of prescription opioids, showing how market dynamics and
government subsidies in the primary market ultimately affect market prices
and quantities in the secondary market. First, a supply expansion (e.g., due
to generic entry) in the primary market for patients obtaining opioids via prescription reduces the price of prescription opioids (from P0 to P1) and increases
the quantity prescribed (from Q0 to Q1)—assuming, of course, that prescribers
are willing to provide additional pills to patients as their demand rises. This
expansion has the effect of reducing the price of prescription opioids in the
secondary market because individuals purchasing prescription opioids in the
primary market now face a lower acquisition cost if pills are diverted to family
members, friends, and others. On top of a supply expansion, the introduction
of a government subsidy for prescription opioids in the primary market drives
a wedge between the price consumers pay (the demand price, P2,D) and the
price prescription drug suppliers receive (the supply price, P2,S), with the difference made up by the amount of the subsidy. The demand price is lower than
the price paid by patients before the introduction of the subsidy (P1), which
further reduces the price of prescription opioids in the secondary market. Thus,
both supply expansions and government subsidies in the primary market for
prescription opioids decrease the price and increase the quantity of opioid
misuse in the secondary market, especially in an environment where there is
overprescribing. As noted above, however, whether secondary market prices
236 |

250-840_text_.pdf 240

Chapter 7

2/7/20 3:46 PM

Figure 7-4. Effect of Supply Expansions and Government Subsidies on
the Price and Quantity of Prescription Opioid Misuse, Primary Market
Prescription opioid price

Supply
Demand
Supply
(new)
P2,S
P0

Government
subsidy

P1

P2,D

Q0

Q2

Q1

Prescription opioid quantity

Note: This figure shows the impact on prices and quantities of an outward supply shift and

government subsidy in the primary market for prescription opioids.

Figure 7-5. Effect of Supply Expansions and Government Subsidies on
the Price and Quantity of Prescription Opioid Misuse, Secondary
Market
Prescription opioid price

Demand

P0

Supply

P1

Supply (new)

P2

Supply (postsubsidy)
Q0

Q1

Q2

Prescription opioid quantity
Note: This figure shows the corresponding impact of an outward supply shift and government
subsidy in the primary market (shown in figure 7-4) on prices and quantities in the secondary
market.

Understanding the Opioid Crisis | 237

250-840_text_.pdf 241

2/7/20 3:46 PM

can actually respond to changes in the primary market depends on an environment in which obtaining prescriptions is relatively easy.
Figures 7-6 and 7-7 consider the case of illicit opioids (i.e., heroin and
illicitly manufactured fentanyl), for which a legal market does not exist.
Because the quantity of illicit opioid use increased substantially between 2010
and 2016, it stands to reason that demand or supply expanded, or both did.
However, whether it was demand or supply that drove the increase in illicit opioid misuse has a testable implication. If demand expansions dominate, then
the price of illicit opioids must rise, whereas if supply expansions dominate,
then the price must fall.8 In fact, we find that illicit opioid prices were relatively
stable between 2010 and 2013, suggesting that both demand—itself fueled in
part by efforts to curtail the prescription opioid wave of the crisis—and supply
expansions were important during this period. Then, between 2013 and 2016,
the price of illicit opioids fell markedly with the influx of illicitly manufactured
fentanyl, suggesting that supply expansions were most important during this
later period.
Our findings suggest that subsidies and supply expansions, in combination with changes in prescribing behavior, can account for much of the rise in
opioid overdose deaths. Some have argued that demand-side factors, such as
economic stagnation in past years, was an important driver of increasing mortality from drug use and other causes (Stiglitz 2015). However, there is direct
evidence that demand growth due to worsening economic conditions was not
the primary factor driving the growth of the opioid crisis.
First, the hypothesis that lower incomes raise demand does not explain
the aggregate time series within the United States. If worsening economic
conditions increase demand, then one would expect that the Great Recession
would have fueled a substantial increase in opioid-involved overdose fatalities.
However, figure 7-8 suggests that the growth rate of opioid-involved overdose
deaths was unaffected by the Great Recession. The crisis grew at roughly the
same pace straight through one of the greatest recessions experienced in
the last century, and in fact picked up growth well after the recession ended.
More important, two of the four lowest growth rates in opioid deaths occurred
between 2008 and 2010, in the midst of the Great Recession. It was not until
2014, 2015, and 2016 that growth rates again rose significantly—but that was
in a period of lower unemployment, the opposite prediction of demand growth
of opioids being fueled by lower incomes unless effects are lagged by several
years.
Despite this lack of association between aggregate economic conditions and opioid deaths, Hollingsworth, Ruhm, and Simon (2017) do report a
positive association between county-level unemployment and opioid-involved
overdose deaths—a 1-percentage-point increase in a county’s unemployment
8 The relative price elasticities of demand and supply also affect which expansion dominates.

238 | Chapter 7

250-840_text_.pdf 242

2/7/20 3:46 PM

Figure 7-6. Effect of Demand Expansions on the Quantity and Price of
Illicit Opioids
Illicit opioid price

Supply
P1

Demand
(new)
P0

Demand
Q0
Q1
Illicit opioid quantity

Note: This figure shows the impact of demand shifting outward while the supply curve remains in
place; in this case, the price must rise.

Figure 7-7. Effect of Supply Expansions on the Quantity and Price of
Illicit Opioids
Illicit opioid price

Supply

P0

Supply
(new)
P1

Demand
Q0
Illicit opioid quantity

Q1

Note: This figure shows the impact of supply shifting outward while the demand curve remains in
place; in this case, the price must fall. If the price falls while the quantity increases, then the supply
must have expanded.

Understanding the Opioid Crisis | 239

250-840_text_.pdf 243

2/7/20 3:46 PM

Figure 7-8. Opioid Overdose Death and Unemployment Rate, 1999–
2016
Unemployment rate (percent)

Deaths (per 100,000)
14

2016

12

12
10

Unemployment
rate (right axis)

10

8

8
6

Opioid
overdose rate
(left axis)

6
4

4
2

2
0
1999

0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: CDC WONDER; Bureau of Labor Statistics; CEA calculations.

rate is associated with a 0.19-person increase in the rate of opioid-involved
overdose deaths per 100,000. However, this association does not appear quantitatively large enough to be a primary driver of the massive growth in opioid
deaths. It would take a 54-percentage-point increase in the unemployment
rate between 1999 and 2016 to explain the 10.2-person increase in the rate of
opioid-involved overdose deaths during this period. However, the unemployment rate increased by a net 0.7 percentage point (from 4.2 to 4.9 percent)
between 1999 and 2016.
In addition, Ruhm (2019) formally tests whether a number of demandside factors that reflect changing economic conditions can explain the growing
crisis during this period. He finds that very little of the rise in opioid overdose
deaths during this period can be explained by economic conditions. Instead, he
points to changes in the drug environment, reflective of supply conditions, as
being central. Consistent with Ruhm’s findings, Currie, Yin, and Schnell (2018)
find no clear evidence of a substantial overall effect of the employment-topopulation ratio on the amount of opioids prescribed in a county.

The First Wave of the Crisis: Prescription Opioids
The opioid crisis unfolded in two waves. The first wave, beginning in about
2001 and lasting until about 2010, was characterized by a rising misuse of

240 | Chapter 7

250-840_text_.pdf 244

2/7/20 3:46 PM

prescription opioids.9 The second wave began in about 2010, when, prescription opioids were made more difficult to abuse and illicit opioids—including
heroin and, more recently, illicitly manufactured fentanyl—grew in the market.
This and the next sections focus on the first wave, and the subsequent section
focuses on the second wave.
Between 2001 and 2010, the rate of overdose deaths involving prescription opioids (which we define as natural and semisynthetic opioids and methadone) increased by 182 percent, while other opioid-involved deaths grew much
more slowly (figure 7-3).10 In order to analyze the potential roles of expanded
supply of prescription opioids, we first estimate the out-of-pocket price of prescription opioids. We then conduct a calibration exercise, in which we assume
that secondary market prices for prescription opioids are proportional to outof-pocket prices, and that prescription opioid misusers respond to these prices
of misuse in the same way that heroin users respond to heroin prices. We also
assume that prescription opioid deaths are proportional to prescription opioid
misuse. If falling prices suggest a large quantity response relative to the magnitude of the observed increase in prescription opioid-involved overdose deaths,
then this would suggest that these price declines, when combined with other
factors, may have played a role in the first wave of the opioid crisis.
An environment in which opioid prescriptions were promoted and easier
to obtain and fill is a necessary precondition for falling out-of-pocket prices to
have played a substantial role—otherwise, it is unlikely that secondary market
prices could have responded to falling out-of-pocket prices. This environment
was created by a campaign to persuade doctors that pain was being undertreated and that opioids were the solution. Pain-alleviation societies, patient
advocacy groups, and professional medical organizations urged physicians to
treat pain more aggressively (Max et al. 1995). Pain was labeled “the 5th Vital
Sign,” which should be regularly assessed and treated (VA 2000). Starting in
2001, the Joint Commission, an accrediting body for hospitals and other health
facilities, instituted new standards requiring facilities to establish procedures
to assess the existence and intensity of pain and to treat it with “effective
pain medicines.” At the same time, multiple medical organizations promoted
opioids as a safe and effective treatment for chronic, noncancer pain (DuPont,
Bezaitis, and Ross 2015). This coincided with aggressive marketing efforts
by opioid manufacturers starting in the late 1990s to assure physicians that
their products were safe with little abuse potential (Van Zee 2009; President’s
9 We focus on the 2001–10 period throughout the chapter, due to the unavailability of consistent
overdose data before 1999, the unavailability of illicit drug seizure data before 2001 used for
estimating the illicit opioid price series, and the substantial volatility in the out-of-pocket price
series before 2001.
10 Some opioid-involved deaths include both prescription and other opioids. Figure 7-3
distinguishes between opioid-involved overdose deaths with prescription opioids present versus
those without prescription opioids present. Similarly, figure 7-18 distinguishes between opioidinvolved overdose deaths with illicit opioids present versus those without illicit opioids present.

Understanding the Opioid Crisis | 241

250-840_text_.pdf 245

2/7/20 3:46 PM

Commission 2017). Because of space limitations, this chapter does not provide
a comprehensive review of either the change in medical guidance regarding
the appropriate use of opioids or the marketing and promotion efforts by
opioid manufacturers.
We use the Medical Expenditure Panel Survey to construct a time series
of the out-of-pocket price per potency-adjusted unit of prescription opioids.
The MEPS asks respondents to report all prescription drugs they obtain and
how much they pay out of pocket for each drug. Opioid prescriptions are converted into morphine gram equivalents (MGEs), and then prices are estimated
by dividing expenditures by the total number of MGEs. We use the terms MGEs
and potency-adjusted units interchangeably throughout. Prices are converted
into real dollars, and then a real price index is shown. Figure 7-9 shows the real
supply and out-of-pocket price index for prescription opioids. The supply price
is calculated as the ratio of total expenditures to total MGEs, and the out-ofpocket price is calculated as the ratio of self (out-of-pocket) expenditures to
total MGEs. Note that out-of-pocket expenditures include individual payments
made for prescriptions without third-party coverage as well as individual
copayments made for prescriptions that are only partially covered by third
parties.
Between 2001 and 2010, the out-of-pocket price fell by 81 percent before
stabilizing. One potential factor in this decline, which is analyzed in depth in
the next section, was the inception of Medicare Part D in 2006, which introduced subsidies for prescription drugs, including opioids, and lowered the
out-of-pocket price for enrolled consumers. Another potential factor was the
rapid expansion of disability (SSDI) enrollment, which before 2006 provided
drug coverage for many enrollees through Medicaid or other programs, and
after 2006 provided coverage through Medicare Part D. Finally, between 2001
and 2010, supply prices fell by 45 percent in conjunction with the expansion of
generic opioids. A recent analysis by the Food and Drug Administration (FDA)
similarly finds that potency-adjusted opioid acquisition prices for pharmacies
fell by about 28 percent during this same period, although it also finds that
prices substantially increased during the 1990s before the crisis took off (FDA
2018a). Figure 7-10 shows the decline in the brand market share of potencyadjusted opioids as the generic market share rose from about 55 to 81 percent
between 2001 and 2010 (FDA 2018a).
The law of demand says that, all else remaining the same, consumers
engage in more of an activity when the activity becomes cheaper. However,
the law by itself does not tell us the magnitude of the effect of an 81 percent
reduction in the potency-adjusted price of prescription opioids on either the
quantity of prescriptions or the number of deaths involving prescription opioids. Previous econometric studies that have related opioid prescriptions and
other prescriptions to out-of-pocket prices suggest a range of likely quantitative effects of the price changes shown in figure 7-9 on the number of opioid
242 | Chapter 7

250-840_text_.pdf 246

2/7/20 3:46 PM

Figure 7-9. Real Supply Price and Real Out-of-Pocket Price Index of
Potency-Adjusted Prescription Opioids, 2001–15
Index (2001 = 1)
1.2

2015

1
0.8

Supply price

0.6
0.4

Out-of-pocket
price

0.2
0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: Medical Expenditure Panel Survey; National Drug Code Database; Bureau of Labor
Statistics; CEA calculations.
Note: Prices are calculated by dividing real total spending in a given year by the total number of
morphine gram equivalents prescribed in that year. All prescriptions are converted into morphine
gram equivalents based on the quantity of pills prescribed and their potency, using the National
Drug Code database.

Figure 7-10. Brand Share of Potency-Adjusted Prescription Opioids
and Supply Price, 2001–16
Percent

Index (2001 = 100)

120

2016

100

100

Real supply price
index (right axis)

80

80
60

60

Brand share of
prescription opioids
(left axis)

40
20
0
2001

120

40
20
0

2003

2005

2007

2009

2011

2013

2015

Sources: Food and Drug Administration (2018a); Medical Expenditure Panel Survey; CEA
calculations.
Note: Price data are available up to 2015. Brand share data are provided up to 2016.

Understanding the Opioid Crisis | 243

250-840_text_.pdf 247

2/7/20 3:46 PM

prescriptions. Predicting the effect on the number of deaths requires additional information because the deaths derive from misuse. Only a fraction of
opioid prescriptions is given to people with opioid use disorder, and their price
sensitivity of demand may differ from the sensitivity of average consumers.
We begin with the effect of reduced prescription opioid prices on the
number of opioid prescriptions. A number of studies look at the effects of drug
prices and insurance coverage on the sales of all prescription drugs as well as
the sales of opioid prescriptions specifically. The more responsive drug users
are to prices, the more they consume as prices decline. This price responsiveness is typically measured by the price elasticity of demand—the percentage change in quantity demanded when the price increases by 1 percent.11
Because elasticity studies typically make cross-sectional comparisons, they are
holding constant physician prescribing norms and marketing efforts by sellers
that are changing over time. In other words, the effects of changing prescribing
norms and marketing efforts need to be added to the price effects measured by
the cross-sectional studies of the price elasticity of demand. Box 7-1 offers an
overview of the ongoing opioid settlements between governments and opioid
manufacturers over misleading marketing efforts by the manufacturers.
Soni (2018) found that the introduction of Medicare Part D increased
opioid prescriptions for the population age 65 to 74 (relative to the population
age 55 to 64 and not on Medicare) over a four-year period by a factor of 1.5. At
the same time and for the same population, Soni (2018) found that the out-ofpocket price was reduced by a factor of 0.44 from the introduction of Part D,
which is less than the price change for the entire U.S. population from 2001 to
2010, as shown in figure 7-9. These estimated effects of Part D are economically
significant and do not support the hypothesis that the changes shown in figure
7-9 have a minimal effect on the number of prescriptions. Indeed, they show
an arc elasticity (calculated with the natural logarithm) of –0.49 which suggests
that the price change shown in figure 7-9 would increase potency-adjusted
prescriptions per capita by a factor of 2.3 between 2001 and 2010. A factor of
2.3 is close to the actual change as estimated with data from the Automation
of Reports and Consolidated Orders System (ARCOS) and shown in figure 7-11
(DOJ n.d.).
Insurance plans should have coinsurance rates varying across drugs
to the extent that the sensitivity of consumer demand to the out-of-pocket
price varies across drugs (Feldstein 1973; Besley 1988). Health insurance
plans behave that way in practice (Einav, Finkelstein, and Polyakova 2018).
Coinsurance rates for opioids (43 percent) are higher than for other common
therapeutic classes (39 percent). Similarly, coinsurance rates for hydrocodone
11 When sales effects are estimated from small price changes, the result is sometimes called
“point elasticity.” “Arc elasticity” refers to an estimate from large price changes and typically
uses midpoints for calculating percentage changes or uses logarithm changes so that the same
elasticity can be applied to price increases as to price decreases.

244 | Chapter 7

250-840_text_.pdf 248

2/7/20 3:46 PM

Box 7-1. Opioid Crisis Lawsuits
Thousands of municipal governments nationwide and nearly two dozen
states have sued the pharmaceutical industry in an effort to hold opioid
manufacturers and distributers accountable for the opioid crisis. These
lawsuits argue that opioid manufacturers launched misleading marketing
campaigns underplaying the risks and exaggerating the benefits of opioids.
Additionally, these lawsuits allege that opioid distributors unlawfully allowed
the drugs to proliferate.
These civil litigation cases have resulted in the conclusion of multiple
settlement agreements, at least one large trial, and the promise of more
settlements to come. OxyContin maker Purdue Pharma, as well as its owners,
the Sackler family, announced a tentative settlement expected to be worth
more than $10 billion in September 2019. Under the proposed agreement, the
company will be restructured into a public corporation, with profits from drug
sales going toward the plaintiffs. The settlement would be the largest payout
from any company involved in the opioid crisis. Purdue Pharma previously
agreed to pay a total of $270 million to Oklahoma to settle a lawsuit in March
2019. Purdue’s Oklahoma settlement set the stage for subsequent settlements
with the State, including Teva Pharmaceutical’s $85 million settlement in May
2019. Johnson & Johnson refused to settle, and the landmark trial resulted in
an order to pay $572 million to Oklahoma in August 2019. Both the State and
Johnson & Johnson are contesting this verdict—alleging, respectively, that
the award is too small or too large.
The three largest drug distributors—McKesson, Cardinal Health,
and AmerisourceBergen—and the generic opioid manufacturer Teva
Pharmaceuticals reached a settlement worth about $260 million in October
2019. These settlements are the early conclusions to nearly two years of legal
battles and may serve as a benchmark for resolution in other opioid cases.
The first of a new series of Federal trials began on October 21, 2019, after talks
dissolved of a deal worth $48 billion to resolve all opioid lawsuits filed against
the three drug distributors, Teva, and Johnson & Johnson.
The settlements include a combination of donations to substance
use disorder treatment program research, and cash payouts and will likely
provide a benchmark for thousands of similar cases brought before the courts
in an attempt to hold pharmaceutical companies accountable for an opioid
crisis that has killed hundreds of thousands and cost trillions.

(50 percent) are higher than for other common nonopioid drugs (40 percent).
The observed coinsurance rates thus suggest that opioid prescriptions are not
less price sensitive than the average prescription drug over the annual time
frame (or longer) that is of interest to the sponsors of insurance plans.12 If
12 The coinsurance rates are inferred from the estimates by Einav, Finkelstein, and Polyakova
(2018) and are for Part D participants who have not yet reached the “donut hole.”

Understanding the Opioid Crisis | 245

250-840_text_.pdf 249

2/7/20 3:46 PM

Figure 7-11. Potency-Adjusted Quantity (MGEs) of Prescription Opioids
per Capita in the United States, 2001–15
MGEs per capita
0.7

2015

0.6

Other

0.5
0.4

Hydrocodone

0.3

Oxycodone

0.2
0.1
0.0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: Automation of Reports and Consolidated Orders System; National Drug Code
database; CEA calculations.
Note: MGEs = morphine gram equivalents. Quantities are converted into MGEs and divided by
the total U.S. population in a given year to calculate MGEs per capita.

Einav, Finkelstein, and Polyakova (2018)’s one-month arc elasticity of –0.27 for
therapeutic drug classes were applied to the price change from 2001 to 2010
shown in figure 7-9, it suggests that opioid prescriptions would have increased
by a factor of 1.6 due to price changes alone.13
A factor of 1.6 is economically significant, but is still only a minority of the
actual change in opioid prescriptions between 2001 and 2010. The discrepancy
between the findings of Soni (2018) and Einav, Finkelstein, and Polyakova
(2018) could be that behavior is more sensitive to a price change that lasts
more than one month, or that applies to a larger population of people.14 But
this discrepancy may also reflect the imprecision of estimating price effects,
which is why our data are consistent with the view that the increase in prescriptions cannot be explained by price reductions alone but also reflect changes in
physicians’ prescribing norms and marketing efforts by opioid sellers.
13 Einav, Finkelstein, and Polyakova (2018) report a point elasticity for a linear demand curve,
but their reports of price and quantity changes are sufficient for their readers to calculate the
corresponding arc elasticity. We also note that the authors’ elasticity is estimated for a selected
group of Part D participants who have high drug costs.
14 The demand for habit-forming products responds more to price changes that last longer (Pollak
1970; Becker and Murphy 1988; Gallet 2014), which is why it would be especially problematic to
apply the approach of Einav, Finkelstein, and Polyakova (2018) specifically to opioids because it
refers to price changes lasting only a month. The estimates by Einav, Finkelstein, and Polyakova
(2018) also exclude “social multiplier” price effects that may occur when the entire population
experiences a price change, rather than a selected few who are at a special spot in their
prescription-benefit formula (Glaeser, Sacerdote, and Scheinkman 2003).

246 |

250-840_text_.pdf 250

Chapter 7

2/7/20 3:46 PM

Figure 7-12. Proportion of Users Obtaining Misused Prescription
Opioids by Most Recent Source, 2013–14
Stole

Other

Bought from
friend or relative

Bought from drug dealer

Free from friend
or relative

Doctor

Source: Lipari and Hughes (2017).

One reason that falling opioid prices may increase opioid deaths at a
different rate than they increase opioid prescriptions is that opioid prices for
medical purposes might follow a different trend than the prices paid by opioid
misusers. In fact, only 25 percent of people who misuse prescription opioids
most recently obtained the drugs from a doctor, while the remaining 75 percent obtained them from friends or relatives, via theft, from a drug dealer, or
from some other source (figure 7-12). But even when the drugs are obtained
on the secondary market, the price is likely positively correlated with the outof-pocket price. A lower out-of-pocket price decreases the acquisition cost for
those selling the drugs in the secondary market. It also should decrease the
implicit price for those giving the drugs away with no expected reciprocal gifts,
and it should reduce the precautions taken by individuals to safeguard their
drugs against theft.15 Of course, the out-of-pocket price is only one component
of the total price of obtaining prescription opioids for misuse. The ease of finding a doctor to prescribe the opioids and a pharmacy that receives a supply and
is willing to fill the prescription is also important.
As a calibration exercise for contextualizing whether falling out-of-pocket
prices could have played a role in the first wave of the opioid crisis, we assume
that the price of prescription opioid misuse is proportional to the out-ofpocket price. For example, a 10 percent decline in the out-of-pocket price of
15 This does not mean that the amount of theft varies with the price because thieves can be
expected to put more effort toward stealing more valuable items. We only assume that thieves
experience greater cost of theft for high-priced items, due to owners’ precautions.

Understanding the Opioid Crisis | 247

250-840_text_.pdf 251

2/7/20 3:46 PM

prescription opioids is assumed to reduce the price of pills in the secondary
market (and for misusers obtaining pills in the primary market) by 10 percent.
This assumption is clearly reasonable for the 25 percent of prescription opioid
misusers who obtain their pills directly from drugs prescribed by medical providers in the primary market because they only face the out-of-pocket price.
Wemay
mayalso
alsoexpect
expect
secondary
market
price
to proportional
be proportional
to outWe
thethe
secondary
market
price
to be
to the
We
may also expect
the
secondary
market
price to who
be proportional
to the
outthe
out-of-pocket
price.
Consider,
first,
the
misusers
purchase
their
pills
of-pocket price. Consider, first, the misusers who purchase their pills in the secondary
of-pocket
price.
Consider,
first,
the
misusers
who
purchase
their
pills
in
the
secondary
in the(as
secondary
(as opposed
to receiving them
complimentarily).
Theseek
market
opposedmarket
to receiving
them complimentarily).
The
sellers of these pills
market
(as opposed
to receiving
them complimentarily).
The sellers
of these
pills
seek
sellers
of
these
pills
seek
to
maximize
their
profits,
which
are
equal
to
the
price
to maximize their profits, which are equal to the price of each pill P minus the cost of
to maximize
their
profits,the
which
equal to the
price
of the
each
pill P minus
the
cost of
of each each
pill
Ppill
minus
cost are
of market
obtaining
each
pill in
primary
C (the
obtaining
in the primary
C (the
out-of-pocket
price),market
multiplied
by the
obtaining
each pillprice),
in the primary
market
C (the
out-of-pocket
price),
multiplied by the
out-of-pocket
multiplied
by
the
number
of
pills
sold,
Q:
number of pills sold, Q:
number of pills sold, Q:
(𝑃𝑃
𝜋𝜋 =
− 𝐶𝐶)𝑄𝑄
𝜋𝜋 = (𝑃𝑃 − 𝐶𝐶)𝑄𝑄

competitivemarket,
market, profits are competed
InInaacompetitive
competed down
downto
tozero
zerofor
forallallsellers,
sellers, so
In a competitive market, profits are competed down to zero for all sellers, so
sothe
thatprice
the price
charged
onsecondary
the secondary
market
is equal
toout-of-pocket
the out-of-pocket
that
charged
on the
market
is equal
to the
price. In
that the price charged on the secondary market is equal to the out-of-pocket price. In
a noncompetitive
market, each
seller
hasseller
the power
influence
the secondary
price. In a noncompetitive
market,
each
has theto
power
to influence
the
a noncompetitive market, each seller has the power to influence the secondary
market
price market
based on
how
many
sells. pills
In terms
of In
the
equation
secondary
price
based
onpills
howitmany
it sells.
terms
of theabove,
equa- this
market price based on how many pills it sells. In terms of the equation above, this
means
that the
is athat
function
of is
quantity.
It can
be shown
thatbeashown
necessary
tion above,
thisprice
means
the price
a function
of quantity.
It can
means that the price is a function of quantity. It can be shown that a necessary
condition
for maximizing
profits
that a necessary
condition
for is
maximizing profits is
condition for maximizing profits is
1
𝑃𝑃 = 1
𝐶𝐶
𝑟𝑟
𝑃𝑃 = 1 + 𝐶𝐶
1 + 𝑟𝑟

where r is the responsiveness, in percentage terms, of the market price to the quantity
where
r is the
in percentage
terms, of theofmarket
price to theto
quantity
r isresponsiveness,
the responsiveness,
in percentage
of where
pills provided
by a particular seller.
Thus, anterms,
increasethe
in market
the costprice
(or thethe
out-ofof pills
provided
by aprovided
particular
seller.
Thus, seller.
an increase
in
the
cost (or
out-ofquantity
of
pills
by
a
particular
Thus,
an
increase
in
the
cost
pocket price) C leads to a proportional increase in the secondary market price P,
pocket
price)
C leads to price)
a proportional
in theincrease
secondary
market
price P,
(or the
out-of-pocket
C leads to increase
a proportional
in the
secondary
assuming that r remains constant.
assuming
that
r
remains
constant.
market price P, assuming that r remains constant.
Assuming that the share of prescription opioids obtained via various segments
Assuming
thatthat
the the
share
of prescription
opioids obtained
via variousvarious
segments
Assuming
share
of prescription
of the secondary
market
with
different
markups opioids
remainsobtained
constantvia
over time, the
of the
secondary
market
with different
markups
remains constant
over
time, the
segments
of the
secondary
market
with
remains
constant
average
secondary
market price
across
all different
segmentsmarkups
would change
proportionally
average
secondary
market price
acrossmarket
all segments
would all
change
proportionally
over
time,
the
average
secondary
price
across
segments
with changes in the out-of-pocket price. It is important to emphasizewould
that this
withchange
changes
in the out-of-pocket
price.
It isout-of-pocket
important to
emphasize
that this
proportionally
with
changes
in
the
price.
It
is
important
assumption would be plausible only if suppliers to the secondary market face
assumption
wouldthat
bethis
plausible
only would
if suppliers
to theonly
secondary
marketthe
face
to emphasize
assumption
beprescriptions
plausible
if suppliers
relatively
low transaction
costs for obtaining
from
doctors to
and filling
relatively
low
transaction
costs
for
obtaining
prescriptions
from
doctors
and
filling
secondary market
face relatively
lowreason,
transaction
costsinfor
obtainingguidelines
prescrip- and
prescriptions
from pharmacies.
For this
changes
prescribing
prescriptions
from
pharmacies.
Forprescriptions
this reason, changes
in prescribing
guidelines
and
tions
from
doctors
and
filling
from
pharmacies.
For
this
practices, a greater emphasis on pain management, and the expansion ofreason,
“pill mills”
practices,
a greater
emphasis
on pain management,
and
the expansion
of “pillpain
mills”
changes
into
prescribing
guidelines
and practices,
a greater
and
supplies
pharmacies
are preconditions
for falling
pricesemphasis
to have aon
potentially
and management,
supplies to pharmacies
are
preconditions
for
falling
prices
to
have
a
potentially
the expansion
significant effect and
on opioid
misuse. of “pill mills” and supplies to pharmacies are
significant
effect onfor
opioid
misuse.
preconditions
falling
prices
to have
a potentially
opioid at a
Another reason that falling
opioid
prices cansignificant
increase effect
opioidondeaths
Another reason that falling opioid prices can increase opioid deaths at a
misuse.
different rate than they increase opioid prescriptions is that most opioid prescriptions
different rate than they increase opioid prescriptions is that most opioid prescriptions
reason
that purposes,
falling opioid
increase
deaths
a a
are likelyAnother
used for
medical
andprices
thosecan
who
misuseopioid
opioids
mayathave
are likely used for medical purposes, and those who misuse opioids may have a
different
rate thantothey
increase
opioidofprescriptions
is that most
opioid
pre-price
different
sensitivity
prices.
One point
view is that medical
users
are less
different sensitivity to prices. One point of view is that medical users are less price
scriptions
are likely
forfollowing
medical purposes,
and those
who
misusemisusers
opioids are
sensitive
because
they used
are just
their providers’
orders,
whereas
sensitive because they are just following their providers’ orders, whereas misusers are
may haveprice
a different
sensitivity
to prices.
of view
is that
medical by
necessarily
sensitive
to the extent
that One
mostpoint
of their
income
is exhausted
necessarily price sensitive
to the extent that most of their income is exhausted by
16
Another
is that
who misuse
are less
purchasing
users areopioids.
less price
sensitiveperspective
because they
are those
just following
theiropioids
providers’
purchasing opioids.16 Another perspective is that those who misuse opioids are less
price
sensitive
because
they
less interested
in saving
orders,
whereas
misusers
areare
necessarily
price sensitive
to themoney
extent on
thattheir
mostdrug
price sensitive because they are less interested in saving money on their drug
acquisitions.
acquisitions.
Unfortunately, we are not aware of studies estimating price elasticities for the
Unfortunately, we are not aware of studies estimating price elasticities for the
misuse of prescription opioids distinctly from price elasticities for the overall number
misuse
distinctly from price elasticities for the overall number
248of|prescription
Chapter
7opioids
of prescription
opioids
(regardless of their use). Thus, we use estimates of the price
of prescription opioids (regardless of their use). Thus, we use estimates of the price
16

People who misuse opioids—who, for example, spend all disposable income on opioids—have a price

People
who
opioids—who,
forpurchased
example, spend
disposable
income
on opioids—have
250-840_text_.pdf
252
2/7/20 3:46 PM
elasticity
of misuse
–1 because
the quantity
is theall
ratio
of disposable
income
to price. Seea price
Becker
16

elasticity of –1 because the quantity purchased is the ratio of disposable income to price. See Becker
(1962) for a more general analysis.

Table 7-1. Estimates of the Price Elasticity of Demand for Heroin
Study type and
outcomes

Elasticity estimates

Silverman and Spruill (1977);
Caulkins (1995); Dave (2008);
Olmstead et al. (2015)

Outcomes related to
heroin use (crime,
emergency room visits,
etc.)

–0.27; –1.50; –0.10; –0.80

Saffer and Chaloupka (1999)

National household
surveys

–0.94

Government historical
records

–0.7 to –1.0;
–0.48 to –1.38

Studies

van Ours (1995); Liu et al. (1999)

Bretteville-Jensen and Biorn (2003);
Interviews with heroin
Bretteville-Jensen (2006); Roddy and
users
Greenwald (2009)

Petry and Bickel (1998); Jofre-Bonet
and Petry (2008); Chalmers et al.
(2010)

Laboratory studies

–0.71 to –0.91;
–0.33 to –0.77; –0.64

–0.87 to –1.3;
–0.82 to –0.92;
–1.54 to –1.73

Source: Olmstead et al. (2015).

of their income is exhausted by purchasing opioids.16 Another perspective is
that those who misuse opioids are less price sensitive because they are less
interested in saving money on their drug acquisitions.
Unfortunately, we are not aware of studies estimating price elasticities
for the misuse of prescription opioids distinctly from price elasticities for the
overall number of prescription opioids (regardless of their use). Thus, we use
estimates of the price elasticity of heroin, a substitute for prescription opioids,
for which a large body of academic literature is available. Olmstead and others (2015) provide an extensive review of the literature and categorize studies
based on the methods used—table 7-1 summarizes their work. Although the literature contains a broad range of estimates, studies generally find that higher
prices reduce demand. For our calibration exercise, we rely on a meta-analysis
of the literature on illicit drug price elasticities by Gallet (2014), who synthesizes 462 price elasticities from 42 studies, mostly based on U.S. data. He finds
that the price elasticity of heroin falls in the range of –0.47 to –0.56, which coincides with the arc elasticity of –0.49 calculated from Soni’s (2018) results for
16 People who misuse opioids—who, for example, spend all disposable income on opioids—have a
price elasticity of –1 because the quantity purchased is the ratio of disposable income to price. See
Becker (1962) for a more general analysis.

Understanding the Opioid Crisis | 249

250-840_text_.pdf 253

2/7/20 3:46 PM

elasticity
of heroin,
a substitute
prescription
opioids,
which
a large
body
of ofof
elasticity
of heroin,
a substitute
for for
prescription
opioids,
for for
which
awhich
large
ofbody
elasticity
elasticity
ofofheroin,
heroin,
aasubstitute
substitute
for
forprescription
prescription
opioids,
opioids,
for
forwhich
abody
alarge
large
body
prescription
opioids
but
is available.
furtherOlmstead
from
zeroand
than
the
short-run
estimates
for
or which a large body
ofacademic
literature
is
Olmstead
and
others
(2015)
provide
an
extensive
academic
literature
is
available.
others
(2015)
provide
an
extensive
academic
academicliterature
literatureisisavailable.
available.Olmstead
Olmsteadand
andothers
others(2015)
(2015)provide
providean
anextensive
extensive
allreview
prescription
drugs
reported
by Einav,
Finkelstein,
and
(2018).17
5) provide an extensive
review
of the
literature
categorize
studies
based
onPolyakova
the
methods
used—table
of
the
literature
andand
categorize
studies
based
on
the
methods
used—table
7-17-1 7-1
review
review
ofofthe
theliterature
literature
and
andcategorize
categorize
studies
studies
based
based
on
onthe
themethods
methods
used—table
used—table
7-1
Because
previous
studies
show
aliterature
range
ofcontains
price
elasticities,
we
can
only
methods used—table summarizes
7-1summarizes
their
work.
Although
literature
contains
a broad
range
of
estimates,
their
work.
Although
thethe
a broad
of
estimates,
summarizes
summarizes
their
their
work.
work.
Although
Although
the
theliterature
literature
contains
contains
aarange
broad
broad
range
range
ofofestimates,
estimates,
provide
agenerally
range
offind
estimates
of higher
the
role
of
price
changes
asFor
acalibration
factor
in exercise,
theexercise,
road range of estimates,
studies
generally
find
that
higher
prices
reduce
demand.
our
calibration
studies
that
higher
prices
reduce
demand.
ForFor
our
studies
studies
generally
generally
find
find
that
that
higher
prices
prices
reduce
reduce
demand.
demand.
For
our
ourcalibration
calibration
exercise,
exercise,
our calibration exercise,
growth
ofwe
misuse
and
the
number
of
deaths
involving
prescription
rely
on
a
meta-analysis
of
the
literature
on
illicit
drug
price
elasticities
by Gallet
we we
relywe
onopioid
a
meta-analysis
of
the
literature
on
illicit
drug
price
elasticities
by
Gallet
rely
relyon
onaameta-analysis
meta-analysisofofthe
theliterature
literatureon
onillicit
illicitdrug
drugprice
priceelasticities
elasticities
by
byGallet
Gallet
rice elasticities by Gallet
opioids.
As
a synthesizes
low
value,
we
take
one
interpretation
of
findings
(2014),
who
synthesizes
price
elasticities
from
42the
studies,
mostly
based
on
(2014),
who
462462
price
elasticities
from
42
studies,
mostly
based
on
U.S.U.S.
(2014),
(2014),
who
who
synthesizes
synthesizes
462
462
price
price
elasticities
elasticities
from
from
42
42short-run
studies,
studies,
mostly
mostly
based
based
on
onU.S.
U.S.
es, mostly based onof
U.S.
data.
He
finds
that
the
price
elasticity
offor
heroin
falls
in the
range
ofnamely,
–0.47
to –0.56,
data.
He
finds
that
the
price
elasticity
of heroin
falls
infalls
the
range
–0.47
–0.56,
Einav,
Finkelstein,
and
Polyakova
(2018)
all
prescription
drugs,
data.
data.
He
He
finds
finds
that
that
the
the
price
price
elasticity
elasticity
ofof
heroin
heroin
falls
ininthe
theof
range
range
ofofto
–0.47
–0.47
toto–0.56,
–0.56,
range of –0.47 to –0.56,
which
coincides
with
the
arc
elasticity
of –0.49
calculated
from
Soni’s
(2018)
results
which
with
the
arc
elasticity
of –0.49
calculated
from
Soni’s
(2018)
results
which
which
coincides
coincides
with
with
the
thearc
arc
elasticity
ofof
–0.49
–0.49
calculated
calculated
from
from
Soni’s
(2018)
(2018)
results
results
that
thecoincides
price
elasticity
of
demand
iselasticity
constant
and
equal
to
–0.27.
As aSoni’s
middle
om Soni’s (2018) results
for
prescription
opioids
but
is
further
from
zero
than
the
short-run
estimates
for
all
for
prescription
opioids
but
is
further
from
zero
than
the
short-run
estimates
for
all for
for
for
prescription
prescription
opioids
but
butisisfurther
further
from
from
zero
zerothan
thanthe
theshort-run
short-run
estimates
estimates
forall
all
value, we
take
the otheropioids
interpretation
of their
results:
that
demand
curve
17 17
1717
ort-run estimates forprescription
allprescription
drugs
reported
by Einav,
Finkelstein,
and
Polyakova
(2018).
drugs
reported
by Einav,
and
Polyakova
(2018).
prescription
prescription
drugs
drugs
reported
reported
by
byFinkelstein,
Einav,
Einav,
Finkelstein,
Finkelstein,
and
andPolyakova
Polyakova
(2018).
(2018).
is
linear
in
price.18
As
a
high
value,
we
take
Gallet’s
high-end
elasticity
of
–0.56.
kova (2018).17
Because
previous
show
a range
ofshown
price
elasticities,
we
can
onlyonly
Because
previous
studies
show
ashow
range
of
price
we
can
only
Because
Because
previous
previous
studies
studies
show
aarange
range
ofofelasticities,
price
price
elasticities,
elasticities,
we
we
can
can
only
The corresponding
results
forstudies
predicted
deaths
are
in
figure
7-13
as
lasticities, we can only
provide
a
range
of
estimates
of
the
role
of
price
changes
as
a
factor
in
the
growth
of ofof
provide
a
range
of
estimates
of
the
role
of
price
changes
as
a
factor
in
the
growth
of
provide
provide
a
a
range
range
of
of
estimates
estimates
of
of
the
the
role
role
of
of
price
price
changes
changes
as
as
a
a
factor
factor
in
in
the
the
growth
growth
“low constant elasticity,” “low linear demand,” and “high constant elasticity,”
a factor in the growthopioid
ofopioid
misuse
and
the
number
of deaths
involving
prescription
opioids.
aAs
low
value,
misuse
and
theand
number
of deaths
involving
prescription
opioids.
As aAs
low
value,
opioid
opioid
misuse
misuse
and
the
thenumber
number
ofofdeaths
deaths
involving
prescription
prescription
opioids.
opioids.
As
aalow
lowvalue,
value,
respectively.19
For
reference,
figure 7-13
alsoinvolving
shows
the
actual rate
of overn opioids. As a low value,
take
one
interpretation
of
the
short-run
findings
of
Einav,
Finkelstein,
andand
we we
take
one
interpretation
of
the
short-run
findings
of
Einav,
Finkelstein,
and
we
wetake
takeone
oneinterpretation
interpretationofofthe
theshort-run
short-runfindings
findingsofofEinav,
Einav,Finkelstein,
Finkelstein,
and
deaths(2018)
involving
prescription
opioids.
Price
changes
would
be capable
Einav, Finkelstein, dose
and
Polyakova
(2018)
for
all
prescription
drugs,
namely,
the
price
elasticity
of demand
Polyakova
for
all prescription
drugs,
namely,
thatthat
thethat
price
elasticity
of demand
Polyakova
Polyakova
(2018)
(2018)
for
for
all
allprescription
prescription
drugs,
drugs,
namely,
namely,
that
the
the
price
price
elasticity
elasticity
ofofdemand
demand
ofisexplaining
between
31–0.27.
83aAs
percent
of value,
thewegrowth
between
2001
and of of
rice elasticity of demand
is constant
and
equal
toand
–0.27.
aAsmiddle
we
take
the
other
interpretation
constant
and
equal
middle
value,
take
the
other
interpretation
isisconstant
constant
and
andto
equal
equal
totoAs
–0.27.
–0.27.
Asaamiddle
middle
value,
value,
we
we
take
take
the
the
other
other
interpretation
interpretation
ofof
18
18 assuming
2010
inresults:
the
death
rate
involving
prescription
opioids,
that
the
rise
intake
18a
18 high
e other interpretation
oftheir
results:
that
the
demand
curve
is linear
in price.
value,
we
take
Gallet’s
As
aAshigh
we
take
Gallet’s
their
that
thethat
demand
curve
iscurve
linear
in
price.
their
their
results:
results:
that
the
thedemand
demand
curve
isis
linear
linear
inin
price.
price.
AsAsvalue,
aahigh
high
value,
value,
we
we
take
Gallet’s
Gallet’s
overdose
deaths
iselasticity
proportional
toThe
thecorresponding
rise in misuse.
Infor
other
words,
without
gh value, we take Gallet’s
high-end
elasticity
of –0.56.
The
corresponding
results
predicted
deaths
are
shown
high-end
elasticity
of –0.56.
corresponding
results
for
predicted
deaths
are
shown
high-end
high-end
elasticity
ofofThe
–0.56.
–0.56.
The
corresponding
results
results
for
forpredicted
predicted
deaths
deaths
are
areshown
shown
dicted deaths are shown
the
changes,
the
estimates
suggest
that
there
would
have
been
between
in figure
7-13
as
“low
constant
elasticity,”
“low
linear
demand,”
and
“high
constant
in price
figure
7-13
as
“low
constant
elasticity,”
“low
linear
demand,”
and
“high
constant
ininfigure
figure7-13
7-13asas“low
“lowconstant
constantelasticity,”
elasticity,”“low
“lowlinear
lineardemand,”
demand,”and
and“high
“highconstant
constant
19 19
1919 involving
nd,” and “high constant
11,500
and
22,800
fewer
deaths
prescription
opioids
during
those
elasticity,”
respectively.
reference,
figure
7-13
also
shows
the
actual
rate
of ofof
elasticity,”
respectively.
ForFor
reference,
figure
7-13
also
shows
the
actual
rate
ofrate
elasticity,”
elasticity,”
respectively.
respectively.
For
Forreference,
reference,
figure
figure
7-13
7-13
also
also
shows
shows
the
the
actual
actual
rate
hows the actual rate
of
overdose
deaths
involving
prescription
opioids.
Price
changes
would
be
capable
of
overdose
deathsdeaths
involving
prescription
opioids.
Price changes
would would
be
capable
of
years.20
overdose
overdose
deathsinvolving
involving
prescription
prescription
opioids.
opioids.
Price
Pricechanges
changes
would
be
becapable
capable
ofof
ges would be capableexplaining
ofexplaining
between
31that
and
83
percent
of the
growth
between
and
2010
in the
between
31 and
83
offraction
the
growth
between
2001
and
2010
in2010
the
explaining
explaining
between
between
31
31and
and
83
83percent
percent
ofofthe
the
growth
growth
between
between
2001
2001
and
and
2010
ininthe
the
Figure
7-13
suggests
apercent
greater
of
the
increase
in2001
actual
overen 2001 and 2010 indose
the
death
rate
involving
prescription
opioids,
assuming
the
rise
in
overdose
deaths
is isis
death
rate
involving
prescription
opioids,
assuming
thatthat
the
rise
in
overdose
deaths
isdeaths
death
death
rateinvolving
involving
prescription
prescription
opioids,
opioids,
assuming
assuming
that
that
the
the
rise
rise
ininoverdose
overdose
deaths
deaths
israte
explained
with
constant
elasticity
models
(the
red
and
gold
lines
rise in overdose deathsproportional
isproportional
to the
rise
in
misuse.
In other
words,
without
the
price
changes,
to2010
the
rise
inin
misuse.
Inyears,
other
words,
without
thepattern
price
thethe the
proportional
proportional
to
to
the
the
rise
rise
ininmisuse.
misuse.
InIn
other
other
words,
without
without
the
thechanges,
price
price
changes,
changes,
the
in the figure)
in
than
earlier
such
aswords,
2005.
This
occurs
t the price changes, the
estimates
suggest
that
there
would
have
been
between
11,500
and
22,800
fewer
estimates
suggest
that
there
would
have
been
between
11,500
and
22,800
fewer
estimates
estimates
suggest
suggest
that
that
there
there
would
would
have
have
been
been
between
between
11,500
11,500
and
and
22,800
22,800
fewer
fewer
because our price measure shows proportionally fewer20 price
declines in the
20
2020
1,500 and 22,800 fewer
deaths
involving
prescription
opioids
during
those
years.
deaths
involving
prescription
opioids
during
those
years.
deaths
deaths
involving
prescription
prescription
opioids
during
during
those
those
years.
years.
early years
thaninvolving
in the later
ones andopioids
likely
reflects
the
substantial
influences of
Figure
7-13
suggests
a greater
fraction
of the
increase
in actual
overdose
Figure
7-13
suggests
thatthat
a greater
fraction
of the
increase
in actual
overdose
Figure
Figure
7-13
7-13
suggests
suggests
that
that
aagreater
greater
fraction
fraction
ofofthe
theincrease
increase
ininactual
actual
overdose
overdose
nonprice
factors
(e.g.,
prescribing
norms
and
marketing
efforts)
in
addition to
rease in actual overdose
deaths
is explained
with
constant
elasticity
models
red
and
gold
lines
in the
deaths
is
explained
with
constant
elasticity
models
(the(the
red
and
gold
lines
inlines
the
deaths
deaths
isisexplained
explained
with
with
constant
constant
elasticity
elasticity
models
models
(the
(the
red
red
and
and
gold
gold
lines
ininthe
the
factors.
the
demand
specification
shows
a time
pattern
of ourour
ed and gold lines inprice
the
figure)
inHowever,
2010
than
in linear
earlier
years,
such
as 2005.
This
pattern
occurs
because
figure)
in
2010
than
inthan
earlier
years,
such
assuch
2005.
pattern
occurs
because
figure)
figure)
inin
2010
2010
than
ininearlier
earlier
years,
years,
such
asasThis
2005.
2005.
This
This
pattern
pattern
occurs
occurs
because
becauseour
our
deaths
(as
opposed
to a total
increase)
that
is closer
to
actual
deaths,
tern occurs becausepredicted
our
price
measure
shows
proportionally
fewer
price
declines
in the
early
years
than
in the
price
measure
shows
proportionally
fewer
price
declines
in the
early
years
than
inthan
the
price
price
measure
measure
shows
shows
proportionally
proportionally
fewer
fewer
price
price
declines
declines
ininthe
the
early
early
years
years
than
ininthe
the
he early years than in the
later
ones
and
likely
reflects
substantial
influences
of nonprice
factors
(e.g.,
later
ones
and
likely
reflects
thethe
substantial
influences
of nonprice
factors
(e.g.,
later
later
ones
ones
and
and
likely
likely
reflects
reflects
the
thesubstantial
substantial
influences
influences
ofofnonprice
nonprice
factors
factors
(e.g.,
(e.g.,
f nonprice factors (e.g.,

17 17Gallet
finds
that
demand
forfor
drugs
(1)(1)
is is
more
responsive
to to
price
thethe
extensive
margin
17 (2014)
17
17 (2014)
Gallet
finds
that
demand
for drugs
(1)more
is(1)
more
responsive
toat
price
at
the
margin
(in (in
Gallet
(2014)
finds
that
demand
drugs
responsive
price
at
extensive
margin
(in
Gallet
Gallet
(2014)
(2014)
finds
finds
that
thatdemand
demand
for
fordrugs
drugs
(1)is
ismore
more
responsive
responsive
to
toprice
price
atatextensive
the
theextensive
extensive
margin
margin
(in
(in
decisions
about
whether
to
use
drugs)
than
at
the
intensive
margin
(how
much
of
the
drug
to use),
about
whether
to use
than
at
the
intensive
margin
(how
much
of drug
the
drug
to
and
about
whether
to
use
drugs)
than
at the
intensive
margin
(how
much
ofmuch
the
to
use),
ce at the extensive margindecisions
(indecisions
decisions
decisions
about
about
whether
whether
totodrugs)
use
use
drugs)
drugs)
than
than
atat
the
theintensive
intensive
margin
margin
(how
(how
much
ofof
the
the
drug
drugand
totouse),
use),and
and
use),
and
is
responsive
inin
the
long
run
than
inin
the
short
run.
(2)
is(2)
more
responsive
in long
the
long
run
in
the
short
run. run.
(2) is
more
in the
run
than
in
the
short
run.
much of the drug to use),
and
(2)
(2)responsive
isismore
more
more
responsive
responsive
in
the
thelong
longthan
run
run
than
than
in
the
the
short
short
run.
18
18 18Einav,
Finkelstein,
and
Polyakova
(2018)
calculate
anelasticity
elasticity
–0.15
based
on
18
18 Finkelstein,
Einav,
Polyakova
(2018)
calculate
an
elasticity
of
–0.15
based
onpercentage
percentage
changes
Einav,
Finkelstein,
andand
Polyakova
(2018)
calculate
an
ofof–0.15
based
onbased
percentage
changes
Einav,
Einav,Finkelstein,
Finkelstein,
and
andPolyakova
Polyakova
(2018)
(2018)
calculate
calculate
ananelasticity
elasticity
of
of–0.15
–0.15
based
on
onpercentage
percentage
changes
changes
changes
from
the
low
price
to
the
high
price,
which
is
a
valid
point
elasticity
only
if
the
demand
the
low
price
to
the
high
price,
which
is
a
valid
point
elasticity
only
if
the
demand
curve
is linear
in inin
fromfrom
the
low
price
to
the
high
price,
which
is
a
valid
point
elasticity
only
if
the
demand
curve
is curve
linear
inislinear
based on percentage changes
from
fromthe
thelow
lowprice
pricetotothe
thehigh
highprice,
price,which
whichisisa avalid
validpoint
pointelasticity
elasticityonly
onlyif ifthe
the
demand
demand
curve
is
linear
curve
iswith
linear
in
price,
withelasticity
aofpoint
elasticity
of
–0.15
at the
average
out-of-pocket
price
paid
with
awith
point
elasticity
of –0.15
at average
the
average
out-of-pocket
price
paid
by
low-cost
Medicare
price,
a point
elasticity
–0.15
atof–0.15
the
out-of-pocket
price
paid
by low-cost
Medicare
PartPartPart
the demand curve is linear
inprice,
price,
price,
with
a apoint
point
elasticity
of
–0.15
atat
the
the
average
average
out-of-pocket
out-of-pocket
price
price
paid
paid
byby
low-cost
low-cost
Medicare
Medicare
Part
Medicare
Part
Dand
recipients
2007
and
2011.
aonly
valid
arc
only
ifcan
D recipients
between
2007
and
2011.
It
is aItvalid
arc
elasticity
ifonly
converted
to –0.27
so that
itthat
can
Dlow-cost
recipients
between
2007
2011.
It between
is
a valid
elasticity
onlyItonly
ifisconverted
toelasticity
–0.27
so
that
itso
beititbe
aid by low-cost MedicarebyPart
DDrecipients
recipients
between
between
2007
2007
and
and
2011.
2011.
Itisarc
isa avalid
valid
arc
arc
elasticity
elasticity
if ifconverted
converted
to
to–0.27
–0.27
sothat
can
canbebe
converted
to
–0.27
so
thatincreases
itincreases
can
bewell
applied
to
price
increases as well as decreases.
to price
increases
as
as
decreases.
applied
to
price
increases
as
well
as
decreases.
erted to –0.27 so that it can
beapplied
applied
applied
to
toprice
price
asas
well
well
as
asdecreases.
decreases.
19
&
19
&𝐴𝐴
For19
the
constant
elasticity
predictions,
weause
demand
function
ofform
the
form
𝑄𝑄
=𝑄𝑄&𝑄𝑄
𝐴𝐴𝑃𝑃
, where
𝐴𝐴
For
the
constant
elasticity
predictions,
weuse
use
awe
demand
function
of
the
form
𝑄𝑄
=
a is a𝐴𝐴𝐴𝐴isisa a
19 19For
the
constant
elasticity
predictions,
we
demand
function
of
the
, where
For
For
the
theconstant
constant
elasticity
elasticity
predictions,
predictions,
weause
use
a ademand
demand
function
function
ofof
the
the
form
𝐴𝐴𝑃𝑃
,&where
,iswhere
"𝐴𝐴𝑃𝑃
"form
" "==𝐴𝐴𝑃𝑃
𝑄𝑄"the
is
quantity
determined
based
on
initial
quantity
price
as
ofasas
is
quantity
andand
determined
based
on
the
initial
quantity
andand
price
asprice
of
e form 𝑄𝑄" = 𝐴𝐴𝑃𝑃& , where 𝐴𝐴parameter
isisaaparameter
parameter
and
determined
based
onthe
the
initial
quantity
and
price
as2001,
2001,
the
𝑄𝑄𝑄𝑄
is
is
the
the
quantity
quantity
parameter
parameter
and
and
determined
determined
based
based
on
onthe
the
initial
initial
quantity
quantity
and
and
price
of𝑄𝑄
of"2001,
2001,
""
demanded,
is price,
the
price,
and
𝜖𝜖0and
<
is<constant
the
constant
elasticity
of demand
with
respect
to price.
the
price.
demanded,
𝑃𝑃 is 𝑃𝑃
the
and
𝜖𝜖 <and
is 𝜖𝜖0
the
elasticity
ofelasticity
demand
with
respect
to
the
of 2001, 𝑄𝑄" is the quantity
quantity
demanded,
the
price,
is
the
constant
of
demand
with
respect
demanded,
demanded,
𝑃𝑃𝑃𝑃isisis
the
the
price,
price,
and
𝜖𝜖<
00isis
the
the
constant
constant
elasticity
elasticity
ofofdemand
demand
with
with
respect
respect
toto
the
theprice.
price.
20
2020Pacula,
Powell,
Pacula,
and
Taylor
(2017,
1) directly
link
the
of Medicare
D—a
source
of ofof
Powell,
and
Taylor
(2017,
1)(2017,
directly
link
the link
introduction
of Medicare
PartPart
D—a
source
of
with respect to the price.to20the
Powell,
Powell,
Pacula,
Pacula,
and
and
Taylor
Taylor
(2017,
1)1)directly
directly
linkintroduction
the
theintroduction
introduction
ofofMedicare
Medicare
Part
Part
D—a
D—a
source
source
price.
some
of the
price
reduction
between
and
2010—to
deaths
involving
prescription
opioids,
the
price
reduction
between
20012001
and
2010—to
deaths
involving
prescription
opioids,
Medicare Part D—a source
of ofsome
some
ofof
the
the
price
price
reduction
reduction
between
between
2001
2001
and
and2010—to
2010—to
deaths
deaths
involving
involving
prescription
prescription
opioids,
opioids,
20 some
Powell,
Pacula,
and
Taylor
(2017,
1)
directly
link
the
introduction
of Medicare
Part
D—a
source
including
“deaths
among
the
Medicare-ineligible
population,
suggesting
substantial
diversion
“deaths
among
the
Medicare-ineligible
population,
suggesting
substantial
diversion
fromfromfrom
volving prescription opioids,
including
“deaths
“deaths
among
among
the
the2001
Medicare-ineligible
Medicare-ineligible
population,
population,
suggesting
suggesting
substantial
substantial
diversion
diversion
from
ofincluding
some
ofincluding
the
price
reduction
between
and 2010—to
deaths
involving
prescription
opioids,
medical
markets.”
medical
markets.”
ng substantial diversionincluding
from
medical
medical
markets.”
markets.”
“deaths
among
the Medicare-ineligible population, suggesting substantial diversion
from medical markets.”

250 | Chapter 7

250-840_text_.pdf 254

2/7/20 3:46 PM

Figure 7-13. Actual and Predicted Rates of Overdose Deaths
Involving Prescription Opioids, by the Price Elasticity of Demand for
Misuse, 2001–15
Deaths (per 100,000)
6

2015
Actual

5
4
3
2

Low elasticity

High elasticity

1
0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: CDC WONDER; Bureau of Labor Statistics; Medical Expenditure Panel Survey; National
Drug Code database; CEA calculations.
Note: Predicted deaths are calculated by holding the demand curve constant and moving down
the demand curve based on the amount of the price decrease. The functional form of the demand
function is provided in the text. The low elasticity is 0.47; the high elasticity is 0.56.

which suggests that constant elasticity might not be the correct model of the
effects of price changes.21
Again, it is important to emphasize that the potential role of prices in
explaining the rise of overdose deaths depends on the ability of consumers in
the primary market to obtain more pills as prices decline. This was facilitated
by an environment in which prescribers were encouraged and even required
to aggressively treat pain with opioids (President’s Commission 2017).22 As a
result, physicians wrote more opioid prescriptions for more patients, lowering
the amount of time and effort needed to acquire the drugs. In some places, the
rise of pill mills further increased the convenience of acquiring these drugs by
combining prescription writing with dispensing.
We further note that the death rate involving prescription opioids
increased by a factor of 2.8 between 2001 and 2010 (figure 7-13), at the same
time that the per capita quantity of prescription opioids increased by a factor
of 2.6 (figure 7-11). This suggests that whatever factor was increasing prescriptions over this period was also increasing opioid use, with only somewhat
21 Given that the research of price effects on drug sales finds most of them to be on the
“extensive margin,” the market demand curve largely reflects the inverse distribution of consumer
heterogeneity. Distribution functions can generate convex demand functions like the constantelasticity function, concave demand functions, or a combination of both, such as with the normal
distribution.
22 In technical terms, prescribing norms affect both the number of prescriptions at a given price
and the sensitivity of that number to price changes.

Understanding the Opioid Crisis | 251

250-840_text_.pdf 255

2/7/20 3:46 PM

greater proportional effects on misuse. One possible explanation for this result
is that the price elasticity of misuse is similar to—but somewhat further from
zero than—the price elasticity of medical use, so price declines increase both
types of use but proportionally somewhat more for misuse.

Public Subsidies for Opioids
A potentially relevant factor for the 81 percent decline in out-of-pocket prices
for prescription opioids between 2001 and 2010 is the expansion of public
health insurance programs that subsidize access to and the purchasing of
prescription drugs, including opioids. These subsidies lower out-of-pocket
prices in the legal market, thereby lowering prices directly for the 25 percent
of prescription opioid misusers who obtain their drugs from a physician and
indirectly for the 75 percent of misusers (see figure 7-12) who receive them on
the secondary market from friends, family, and dealers who first obtained the
drugs in the primary market.23
The share of potency-adjusted prescription opioids funded by government programs grew from 17 percent in 2001 to 60 percent in 2010 (figure
7-14). However, this may understate the share of diverted opioids that were
obtained with the assistance of funding from public programs. The diversion
of opioids to the secondary market is more profitable when out-of-pocket
prices are lower, and drugs purchased with government subsidies cost less on
average than drugs purchased out of pocket or with private insurance (MEPS).
Thus, government subsidies that cut out-of-pocket prices the most may lead
to opioids obtained with the assistance of funding from these programs to be
the most likely to be diverted. In fact, government programs funded 74 percent
of all opioids that were covered at least in part by a third-party payer in 2010
(MEPS).
Figure 7-14 shows the shares of potency-adjusted opioids covered by
public programs, private insurers, and no third-party payer. Public programs
have become much more important sources for funding opioids over time, and
Medicare coverage expansions appear to have largely driven this growth. The
share of opioids covered by Medicare spiked in 2006, coinciding with the implementation that year of Medicare Part D, which offers prescription drug benefits
to Medicare beneficiaries.24 It is important to note that the vast majority of
Medicare Part D enrollees dispensed opioids do not misuse them. Carey, Jena,
23 See Schnell (2017), who analyzes the linkages between the primary and secondary markets.
24 In a similar calculation, Zhou, Florence, and Dowell (2016) find that the share of expenditures on
prescription opioids accounted for by Medicare increased from 3 percent in 2001 to 26 percent in
2012. As shown in figure 7-14, we find that the number of prescriptions for which Medicare was the
primary payer increased from 5 percent in 2001 to 29 percent in 2012. The slight differences may
be because the Medicare share of expenditures (as reported by Zhou, Florence, and Dowell 2016)
does not include out-of-pocket copayments made by Medicare enrollees for prescriptions where
Medicare was the primary payer (figure 7-14).

252 |

250-840_text_.pdf 256

Chapter 7

2/7/20 3:46 PM

Figure 7-14. Share of Potency-Adjusted Prescription Opioids, by
Primary Payer, 2001–15
Percent
100
90
80
70
60
50
40
30
20
10
0
2001

2015

No third-party payer
Private
Other public
Medicaid
Medicare

2003

2005

2007

2009

2011

2013

2015

Sources: Medical Expenditure Panel Survey; National Drug Code Database; CEA calculations.
Note: The primary payer is defined as the third-party payer with the highest payment for a given
prescription. In addition to Medicare, Medicaid, and private insurers, the other possible primary
payers include veterans’ benefits, workers’ compensation, other Federal government insurance,
other State or local government insurance, or other public insurance. All prescriptions are
converted into morphine gram equivalents based on the quantity of pills prescribed and their
potency, using the National Drug Code database.

and Barnett (2018) studied a sample of more than 600,000 Medicare beneficiaries who had an opioid prescription. Using several different measures, only 0.6
to 8.5 percent of the beneficiaries fulfilled a misuse measure.
The implementation of Medicare Part D and the resulting growth in the
share of opioids funded by Medicare do not appear to have simply displaced
opioids covered by other sources. Figure 7-15 shows the quantity of opioids per
capita funded by each source. Though the number of potency-adjusted opioids
covered by Medicaid fell between 2005 and 2006, the increase in the number
of opioids covered by Medicare was over three times larger than this decline.25
The number of potency-adjusted opioids covered by private insurance also
increased between 2005 and 2006. Furthermore, between 2005 and 2008, the
MEPS data suggest that the total quantity of potency-adjusted opioids that

25 An estimated 6.2 million Medicaid beneficiaries became eligible for Medicare Part D prescription
drug coverage on January 1, 2006 (KFF 2006). These “full dual eligibles” included low-income
seniors and low-income disabled individuals under age 65. Nonelderly disabled dual eligibles,
including both full and partial, made up about one-third of all duals (2.5 million out of almost 7.5
million—per Holahan and Ghosh 2005, 3). Applying this one-third ratio to 6.2 million means that
about 2.0 to 2.1 million nonelderly disabled Medicaid participants transitioned from Medicaid to
Medicare prescription drug coverage in 2006. For comparison, the SSDI rolls grew from 6.5 million
to 6.8 million individuals between 2005 and 2006.

Understanding the Opioid Crisis | 253

250-840_text_.pdf 257

2/7/20 3:46 PM

Figure 7-15. Potency-Adjusted Prescription Opioids per Capita, by
Primary Payer, 2001–15
MGEs per capita
0.6

2015

0.5

No third-party payer

0.4

Private

0.3

Other public

0.2

Medicaid
Medicare

0.1
0.0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: Medical Expenditure Panel Survey; National Drug Code Database; CEA calculations.
Note: MGEs = morphine gram equivalents. The primary payer is the third-party payer with the
highest payment for a given prescription. In addition to Medicardi
re, Medicaid,
and private insurers,
ca
the other possible primary payers include veterans’ benefits, workers’ compensation, other Federal
government insurance, other State or local government insurance, or other public insurance. All
prescriptions are converted into MGEs based on the quantity of pills prescribed and their potency,
using the National Drug Code Database.

were dispensed increased by 73 percent, with almost three-fourths of this
growth coming from opioids paid for by Medicare.26
Between 2001 and 2010, Medicare-covered opioids increased by over
2,400 percent, Medicaid-covered opioids increased by over 360 percent, and
total publicly covered opioids increased by over 1,200 percent (MEPS). Given
that Medicare covers the elderly and SSDI recipients who tend to have greater
needs related to pain relief, it is not surprising that Medicare is the largest payer
of prescription drugs as well as the largest public payer of prescription opioids.
Previous research has studied the implications of the rise in public
funding for opioids fueling the opioid crisis and, in particular, the diversion of
pills to the secondary market. Powell, Pacula, and Taylor (2017) found that a
Medicare Part D–driven 10 percent increase in opioid prescriptions results in 7.4
percent more opioid-involved overdose deaths among the Medicare-ineligible
population. The authors use the fact that Medicare Part D was plausibly more
important in driving prescription drug benefits in States with a greater share of
the population over age 65 to estimate the impact of drug benefits on opioidinvolved overdose deaths.
26 As shown in a comparison of figures 7-11 and 7-15, the MEPS data undercount the total
number of prescription opioids. See also Hill, Zuvekas, and Zodet (2011, 242), which looks more
systematically at the propensity of MEPS respondents “to underreport the number of different
drugs taken.” MEPS underreporting presents greater challenges for measuring total quantities
rather than average prices, which is why the CEA measures the former from ARCOS and the latter
from MEPS.

254 | Chapter 7

250-840_text_.pdf 258

2/7/20 3:46 PM

Moreover, because the elderly—the major population that is eligible for
Medicare benefits—are a disproportionately small fraction of those reported
to die of drug overdoses, these results suggest that the impact of Medicare
expansion on opioid-involved death rates may have been due to an increased
supply of prescription opioids in the secondary market. Others have examined opioid prescriptions covered by Medicaid.27 In a recent report, the U.S.
Senate Committee on Homeland Security and Governmental Affairs (2018)
notes numerous examples of Medicaid fraud that fuel abuse of prescription
opioids—for example, with drug dealers paying Medicaid recipients to obtain
taxpayer-funded pills.
Similarly, Eberstadt (2017) suggests that Medicaid has helped finance
increasing nonwork by prime-age adults by subsidizing prescription opioids
that could be sold on the secondary market. Goodman-Bacon and Sandoe
(2017), Venkataramani and Chatterjee (2019), and Cher, Morden and Meara
(2019), however, find little evidence for Medicaid expansion fueling the opioid
crisis. These findings are not necessarily inconsistent with other evidence that
public programs worsened the opioid crisis. It is possible that Medicaid expansion did not increase opioid misuse because the expansion population is less
likely to be prescribed opioids. Before State expansions, Medicaid already covered all disabled adults receiving Supplemental Security Income (SSI), as well
as elderly adults not eligible for Medicare. Medicaid expansion only covered
nondisabled, nonelderly adults with low incomes, a population less likely to
be prescribed opioids. In fact, figure 7-15 shows that the per capita quantity
of opioids covered by Medicaid decreased between 2013 and 2015, despite
the fact that Medicaid enrollment grew from 60 million to 70 million people
over this same period, as the majority of States expanded Medicaid coverage.
In addition, the Medicaid expansions studied by Goodman-Bacon and Sandoe
(2017) occurred in 2014, after measures had been taken to reduce the ability of
people to misuse prescription opioids (e.g., the reformulation of OxyContin in
2010 and the introduction of other medicines along with the rescheduling of
certain opioids to higher schedules with more restrictions).
Public subsidies for prescription opioids have also been fueled by the
growing number of Americans claiming disability insurance. SSDI is a Federal
disability assistance program that offers a maximum possible benefit of $2,687
a month, with an average monthly benefit of $1,173. Only adults who have
significant work experience are eligible to receive SSDI, and the amount of

27 In 2017, 15.6 percent of the total U.S. population was age 65 or older, but only 3.6 percent of all
opioid-involved overdose deaths were age 65 or older (CDC WONDER).

Understanding the Opioid Crisis | 255

250-840_text_.pdf 259

2/7/20 3:46 PM

benefits is higher for those who had higher lifetime earnings before becoming
disabled.28
SSDI disabled workers are generally eligible for Medicare after 24 months
of enrollment in the program. SSDI rolls have increased dramatically since
1990. The growth in SSDI rolls can be attributed to several factors, including
the aging of the population, the increased labor force participation of women,
and more lenient disability determinations (Autor 2015). Another disability program, SSI, provides more modest benefits to Americans without sufficient work
experience to qualify for SSDI, and provides automatic eligibility for Medicaid
in most States. Figure 7-16 shows the rise in SSDI and SSI rolls per 100,000
people over time. Notably, SSDI rolls and opioid overdose deaths, especially
those involving prescription opioids, have risen in tandem. It is also important
to note SSDI growth occurred over the same period as increased treatment of
pain conditions with opioids.
The 8.6 million SSDI disabled workers in 2011 represent less than 3 percent of the total U.S. population, and thus are overrepresented as a source of
prescription opioids given disabilities (increasingly related to pain) that lead to
a greater use of prescription opioids. The CEA estimates the total market share
of SSDI recipients in two ways, each suggesting that SSDI recipients make up
about 26 to 30 percent of the prescription opioid market. First, we use data
from Morden and others (2014), who estimate the average potency-adjusted
opioid prescriptions for SSDI recipients across the United States in 2011 (6.9
MGEs per SSDI recipient). We multiply this average rate by the total number of
SSDI recipients in 2011 (8.6 million recipients). And finally, we divide the total
opioids prescribed to SSDI recipients (59.2 million MGEs) by the total opioids
distributed in the United States according to ARCOS data (196.9 million MGEs).
The result is that 30 percent of potency-adjusted opioid prescriptions in the
U.S. are filled by SSDI recipients, which is over 10 times their proportion of the
U.S. population.
Second, the CEA uses MEPS data that report opioid prescriptions for a
random sample of Americans each year. We identify SSDI recipients as individuals between age 18 and 64 who receive Medicare. This may slightly overstate
the SSDI population, given that a small number of non-SSDI recipients under
age 65 are eligible for Medicare as well, including people with end-stage renal
disease and amyotrophic lateral sclerosis.29 Nonetheless, dividing the potencyadjusted opioids prescribed to these recipients by the total in the population
28 Qualification for SSDI requires a sufficient number of work credits that were earned recently
enough. Up to 4 credits can be earned in one year and are accrued based on sufficient annual
earnings. Applicants generally require 40 credits to qualify for SSDI, although standards are
different for younger workers.
29 There were just under 273,000 Medicare recipients under age 65 with end-stage renal disease
in 2013 (HHS 2014). The prevalence of amyotrophic lateral sclerosis is just 5 per 100,000, implying
that in 2013, there were just under 16,000 Americans with the disease (Stanford Medicine n.d.).

256 |

250-840_text_.pdf 260

Chapter 7

2/7/20 3:46 PM

Figure 7-16. Adults Receiving Social Security Disability Insurance and
Supplemental Security Income, and Opioid-Involved Drug Overdose
Deaths, per 100,000 People, 1980–2016
Adult recipients (per 100,000)

Deaths (per 100,000)
14

3,000

12

2,500

SSI (left axis)

2,000

8

1,500
1,000

SSDI (left axis)

500
0
1980

10

Other opioid
overdose (right
axis)

6
4

Rx opioid overdose (right
axis)
1984

1988

1992

1996

2
0
2000

2004

2008

2012

2016

Sources: Social Security Administration; CDC WONDER; CEA calculations.
Note: SSDI = Social Security Disability Insurance. SSI = Supplemental Security Income. Prescription
opioids include natural and semisynthetic opioids as well as methadone. Data for opioid overdose
deaths were accessed in CDC WONDER beginning in 1999.

results in an estimated SSDI market share of 26 percent for the period 2010–
12.30 The somewhat lower share estimated using MEPS data may be due to the
exclusion of SSDI recipients who have been on the program for less than two
years.31 These SSDI recipients would not yet be eligible for Medicare and may
instead receive coverage via Medicaid or other programs.32
It is important to emphasize that the disproportionate market share of
SSDI recipients receiving prescription opioids is a result of their higher levels
of conditions that prevent them from working and that may also cause pain.
SSDI benefit payments, in conjunction with Medicare coverage, provide a vital
means of support for disabled workers with major healthcare needs. Thus,
reforms that seek to reduce nonmedical use of opioids should be careful to
preserve access to needed pain relief through the medical use of opioids for
SSDI recipients.
30 Based on a five-year average centered on 2011, we similarly estimate a market share of 26
percent.
31 MEPS excludes the institutionalized population, so if SSDI recipients are overrepresented in this
population, this could further affect our estimate.
32 We note that Finkelstein, Gentzkow, and Williams (2018) estimate that SSDI recipients account
for about 13 percent of opioid prescriptions. However, they do not appear to analyze potencyadjusted opioids, as we do. Indeed, when we use the MEPS data to estimate the market share
of non-potency-adjusted opioid prescriptions for the same 2006–14 period that Finkelstein,
Gentzkow, and Williams (2018) appear to consider, we estimate a similar 15.5 percent market
share.

Understanding the Opioid Crisis | 257

250-840_text_.pdf 261

2/7/20 3:46 PM

The Second Wave of the Crisis: Illicit Opioids
The second wave of the opioid crisis began in about 2010, when prescription opioids became more difficult to access due to efforts to rein in abuse.
However, the buildup of a pool of people misusing prescription opioids that
they could no longer access provided a large pool of new demand and a profit
opportunity for sellers entering the illicit opioid market. Because, for people
suffering from addiction, legal and illicit opioids can function as substitutes,
raising the price (in terms of both money and time) of legal opioids raises the
demand for illicit ones.
The reformulation of OxyContin in 2010 made it more physically difficult
to use. States have implemented prescription drug monitoring programs that
require doctors to consult patient prescription histories before prescribing opioids (Dowell et al. 2016; Buchmueller and Carey 2018; Dave, Grecu, and Saffer
2018). Professional societies and accrediting organizations have reconsidered
their pain treatment guidelines. These changes have reduced the overall quantity of prescription opioids distributed, with potency-adjusted quantities of
opioids peaking in 2011 (DOJ n.d.). Unfortunately, recent research has shown
that overdose deaths averted from prescription opioid overdoses, at least
those resulting from the reformulation of OxyContin, have been replaced by
overdose deaths from heroin (Alpert, Powell, and Pacula 2018; Evans, Lieber,
and Power 2019).
As users have substituted toward heroin, it has increasingly been made
even more potent—suppliers and drug dealers now frequently lace heroin with
illicitly manufactured fentanyl. Fentanyl is 30 to 50 times more potent in its
analgesic properties than heroin, so even small amounts can vastly increase
the potency of the drugs with which it is mixed. Illicitly manufactured fentanyl
can also be obtained independently of heroin. Figure 7-17 documents the
rise of fentanyl, showing both the rate of overdose deaths involving synthetic
opioids other than methadone (a category dominated by fentanyl, although
whether the product is illicit or by prescription is not determinable), and the
rate of fentanyl reports in forensic labs acquired by law enforcement during
drug seizures.
Figure 7-18 shows the rise in overdose deaths involving heroin and fully
synthetic opioids (mostly fentanyl), along with opioid deaths not involving heroin and synthetic opioids. As a reminder, we refer to overdose-related opioid
deaths from heroin and fentanyl as “illicit deaths,” even though fentanyl can
also be prescribed.33 From 2010 through 2016, the rate of illicit opioid deaths
has increased by 364 percent, while the rate of opioid deaths not involving
illicit opioids has fallen by 17 percent. Importantly, fentanyl also tends to be
combined with nonopioids, and deaths in which fentanyl and nonopioids are
factors are included in the illicit opioid series shown in figure 7-18.
33 We use ICD-10 codes T40.1 and T40.4 to identify deaths involving heroin and fentanyl.

258 |

250-840_text_.pdf 262

Chapter 7

2/7/20 3:46 PM

Figure 7-17. Rate of Overdose Deaths Involving Synthetic Opioids
Other Than Methadone, and Fentanyl Reports in Forensic Labs per
100,000 Population, 2001–16
Rate (per 100,000)
12

2016

10

Fentanyl reports in
forensic labs

8
6
4

Deaths involving synthetic
opioids other than
methadone (e.g., fentanyl)

2
0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: National Forensic Laboratory Information System; DOJ and DEA (2017b); CDC WONDER;
CEA calculations.

Figure 7-18. The Opioid-Involved Overdose Death Rate by the
Presence of Illicit Opioids, 2001–16
Deaths (per 100,000)

Second wave

10

2016

8

Did not involve
illicit opioids

6
4

Involved illicit
opioids

2
0
2001

2003

2005

2007

2009

2011

2013

2015

Sources: CDC WONDER; CEA calculations.
Note: Illicit opioids include both heroin (T40.1) and the category “synthetic opioids other than
methadone” (T40.4) in the CDC data, which is primarily composed of illicitly produced fentanyl.

Understanding the Opioid Crisis | 259

250-840_text_.pdf 263

2/7/20 3:46 PM

Given their illegal nature, the price of illicit opioids is more difficult
to measure than the price of prescription opioids. Accurate data cannot be
reliably obtained from dealers or users, who may fear criminal sanctions for
truthful reporting. In recent years, the influx from Mexico and China of cheap
but highly potent fentanyl, which can vastly increase the potency of drugs with
which it is mixed, has complicated matters (U.S. Department of State n.d.).
Market quantities of heroin and fentanyl also cannot be directly observed, so
the extent to which added fentanyl reduces the price per potency-adjusted
unit of opioids is difficult to determine. Subject to these limitations, the CEA
has assembled data from several sources to create a time series for the price
of illicit opioids.
The Drug Enforcement Administration’s (DEA’s) System to Retrieve
Information from Drug Evidence (STRIDE) and STARLIMS databases collect
heroin price data. Heroin prices in these data sets are obtained by government agents, who pay informants to purchase heroin on the street. The price
is recorded, and the heroin sample is analyzed in a laboratory to determine its
potency so that prices can be adjusted for quality. Between 2010 and 2016, the
potency-adjusted real price of heroin increased by 10 percent.
However, any fentanyl contained within heroin is not considered when
determining the price per pure gram of heroin in the DEA data. Thus, the true
price per potency-adjusted unit of heroin purchases has likely increased by less
than 10 percent or has even declined. In addition, fentanyl can be consumed on
its own outside heroin, which, if cheaper on a potency-adjusted basis, would
lead overall illicit opioid prices to fall even more. Moreover, increased heroin
purity and product modifications have increasingly allowed for heroin use by
means other than injection. These changes lower the nonmonetary costs of
using heroin, and although nonmonetary costs are not estimated here, these
changes would have further reduced the cost of illicit opioid use.
The CEA uses data from several sources to estimate the quantity of
fentanyl mixed with heroin and available on its own, along with the potencyadjusted price of heroin (including the fentanyl with which it is mixed) and
the potency-adjusted price of fentanyl when consumed alone or with other
drugs. Quantity data are based on seizures of heroin and fentanyl recorded in
the National Seizure System, along with exhibits of each drug recorded in the
National Forensic Laboratory Information System. Price data are based on the
DEA heroin price series and on DEA reports on the cost of fentanyl relative to
heroin, along with the quantity data in order to adjust heroin prices based on
fentanyl with which it is mixed. A detailed methodology for estimating illicit
opioid prices is provided in the appendix of a previously published CEA report
(CEA 2019b). We acknowledge that seizure data are a highly imperfect proxy of
the relative presence of heroin and fentanyl. Seized products reflect a combination of market shares and law enforcement priorities rather than market shares

260 |

250-840_text_.pdf 264

Chapter 7

2/7/20 3:46 PM

Figure 7-19. Real Price Index of Potency-Adjusted Illicit Opioids,
2001–16
Index (2001 = 1)
1.2

2016

1.0

0.8

Price per MGE of
illicit opioids

0.6

0.4

0.2
2001

2003

2005

2007

2009

2011

2013

2015

Sources: National Seizure System; National Forensic Laboratory Information System; U.S. Bureau of
Labor Statistics; ONDCP (2017); DOJ and DEA (2017a, 2017b); CEA calculations.
Note: MGE = Morphine gram equivalent.

alone. Still, absent an alternative data source, and without a clear direction for
the bias in this proxy for market shares, we use the seizure data as reported.
Figure 7-19 shows a real price index for illicit opioids between 2001 and
2016, which, given the data limitations involved, should be used only to draw
qualitative conclusions. The price of illicit opioids is relatively stable before falling temporarily in 2006, and then quickly recovering, and then falls by over half
(58 percent) between 2013 and 2016. Each of these declines is due to surges
in fentanyl that is either mixed with heroin or sold on its own or with other
drugs. The 2006 price decline was due to a laboratory in Mexico that dramatically increased the supply of fentanyl to the United States but was quickly shut
down through cooperative action between the United States and Mexico. The
price decline between 2013 and 2016 is attributed to the widely documented
influx of fentanyl into the United States, including from China and Mexico
(NIDA 2017). The price series shown in figure 7-19 is the outcome of a series
of assumptions documented more completely in the appendix of the CEA’s
previously published report and is necessarily only a highly imperfect estimate
of the real price from which only qualitative conclusions should be drawn (CEA
2019b). If data on the illicit opioid market in this period improve, revisions to
this series may be possible.
It is clear from figure 7-19 that supply expansions were important for driving the growth in overdose deaths involving illicit opioids. Between 2010 and
2013, the price of illicit opioids was relatively stable. This implies that both supply and demand expansions were important during the first three years of the
Understanding the Opioid Crisis | 261

250-840_text_.pdf 265

2/7/20 3:46 PM

illicit wave of the opioid crisis. If only demand had expanded, then prices would
have increased; and if only supply had expanded, then prices would have
decreased. Demand expansions can likely be traced at least in part to efforts
to clamp down on abuse that grew during the first wave of the crisis without
providing additional access to quality treatment services. Expanded supply
is likely due to increased supply from source countries, including Mexico and
Colombia, and it may reflect a substitution of drug production from marijuana
(which has been legalized or decriminalized in some U.S. States) to heroin
(ONDCP 2019). Meanwhile, supply expansions are likely more important than
demand expansions for the 2013–16 period, given that the price of illicit opioids fell by more than half during these three years. The shift toward fentanyl
produced in China and distributed through the mail has increased the potency
of drugs without significantly increasing their prices, and may have increased
competition in the illicit opioid market, thereby also putting downward pressure on the price of heroin.
To the extent that monetary price declines have been accompanied by
an increased ease of obtaining illicit opioids (given the proliferation of drug
dealers in more locations and the increased availability of online markets),
supply expansions may have been even more important than the falling illicit
price series suggests. For instance, Quinones (2015) notes that Mexican heroin
dealers who illegally cross the border have become much more efficient in
delivering heroin to users rather than forcing users to find them. These drug
dealers communicate with users via cell phones to establish a place to meet, at
which point the user enters the dealer’s car to receive their heroin.

Conclusion
The opioid crisis poses a major threat to the U.S. economy and American lives,
and many factors have exacerbated this threat. In addition to taking more
than 400,000 lives since 2000, the opioid crisis cost $665 billion in 2018, or
3.2 percent of U.S. gros domestic product. In this chapter, the CEA presents
evidence that falling prices may have played a role in increasing opioid misuse
and opioid-involved overdose deaths.
During the first wave of the opioid crisis, which was characterized by
growing overdose deaths involving prescription opioids between 2001 and
2010, the out-of-pocket price of prescription opioids fell by 81 percent. This
likely reduced the price of prescription opioids in the secondary market, from
which most people who misuse prescription opioids obtain their drugs. Using
the proportional price assumption and given elasticities from the academic literature, we find that the decline in observed out-of-pocket prices is capable of
explaining between 31 and 83 percent of the growth in the number of overdose
deaths involving prescription opioids between 2001 and 2010. At the same
time that out-of-pocket prices were falling, government subsidies and the

262 |

250-840_text_.pdf 266

Chapter 7

2/7/20 3:46 PM

market share of generic opioids were expanding. We estimate that the share of
prescribed opioids funded by government programs increased from 17 percent
in 2001 to 60 percent in 2010 (and to 63 percent in 2015). The share of publicly
funded opioids diverted to the secondary market may be even higher, given the
relatively low acquisition cost for suppliers of diverted opioids.
Falling prices could not elicit a change in the quantity of opioids misused
and the resulting opioid deaths unless providers were encouraged to prescribe
the opioids, health plans were paying for the prescriptions, and pharmacies
were filling these prescriptions. We describe the change in the environment
resulting from changing pain management guidelines and aggressive marketing tactics that reduced barriers to obtaining larger quantities of opioids.
The CEA finds that the second wave of the opioid crisis—characterized
by growing deaths from illicit opioids between 2010 and 2016—was driven
by a combination of supply and demand expansions. Efforts to restrict the
supply and misuse of prescription opioids led an increased number of users
from the first wave to substitute illicit opioids in place of prescription opioids.
At the same time, the supply of illicit opioids expanded, and this substitution
decreased quality-adjusted prices to reduce the “cost of a high.” Despite the
importance of demand through a substitution effect in the initial years of the
second wave, the CEA finds that the evidence supports the idea that supply
expansions have been more important causes of the crisis’s growth than
demand increases.
The Trump Administration has taken significant steps to stem the tide
of the opioid crisis. In October 2017, the Administration declared a nationwide
Public Health Emergency. President Trump later established his Initiative to
Stop Opioid Abuse and Reduce Drug Supply and Demand in March 2018 (White
House 2018). These and other measures taken by the government include
securing more than $6 billion in new funding in 2018 and 2019 to address the
opioid crisis by reducing the supply of opioids, reducing new demand for opioids, and treating those with opioid use disorder.
To restrict the supply of illicitly produced opioids, there have been
increased efforts to prevent the flow of illicit drugs into the U.S. through
ports of entry and international shipments. The President also signed into
law the International Narcotics Trafficking Emergency Response by Detecting
Incoming Contraband with Technology (INTERDICT) Act, which funds U.S.
Customs and Border Protection (CBP) to expand technologies to help interdict illicit substances including opioids. The CBP is also training all narcotic
detector dogs at ports of entry to detect fentanyl. These efforts have seen
success—during fiscal year 2019, the CBP seized almost 2,800 pounds of fentanyl and over 6,200 pounds of heroin (CBP 2019). The Administration has also
increased enforcement against illicit drug producers and traffickers. In 2018,
the Department of Justice (DOJ) indicted two Chinese nationals accused of
manufacturing and shipping fentanyl analogues, synthetic opioids, and 250
Understanding the Opioid Crisis | 263

250-840_text_.pdf 267

2/7/20 3:46 PM

other drugs to at least 37 U.S. States and 25 other countries (DOJ 2018). In addition, the Department of the Treasury has levied kingpin designations against
fentanyl traffickers that operate in China, India, the United Arab Emirates, and
Mexico, and also throughout Southeast Asia, including Vietnam, Thailand, and
Singapore. To stop the flow of this deadly drug before it reaches Americans,
the Administration is working with more than 130 nations that signed onto
President Trump’s Call to Action on this issue. The Federal government is also
engaging private sector partners to help secure U.S. supply chains against traffickers attempting to exploit those platforms (ONDCP 2019). One example is
the promotion of increased private sector self-policing of products entering the
U.S. via third-party marketplaces, and other intermediaries to an e-commerce
transaction (via the Department of Homeland Security).
Immigrations and Customs Enforcement’s Homeland Security
Investigations (HSI) organization has also increased its efforts targeting
transnational criminal organizations (TCO) involved with the opioid epidemic.
HSI has increased its partnerships—such as the Border Enforcement Security
Taskforce (BEST) platforms—with other Federal, international, tribal, State,
and local law enforcement agencies to increase information and resource sharing within U.S. communities. BESTs eliminate the barriers between Federal and
local investigations (access to both Federal and State prosecutors), close the
gap with international partners in multinational criminal investigations, and
create an environment that minimizes the vulnerabilities in our operations
that TCOs have traditionally capitalized on to exploit the Nation’s land and sea
borders.
To better combat 21st-century crime exploiting ecommerce, HSI has
increased its presence at international mail facilities and express consignment centers by establishing BESTs at John F. Kennedy International Airport
in New York, Los Angeles International Airport, Memphis International
Airport, Cincinnati–Northern Kentucky International Airport, and Louisville
International Airport as part of HSI’s comprehensive and multilayered strategy
to combat TCOs and their smuggling activities. This strategy facilitates the
immediate application of investigative techniques on seized parcels, which aid
in establishing probable cause needed to effect enforcement actions on individuals associated with narcotics laden parcels. Consequently, these seizures
and arrests disrupt the movement of narcotics transiting through the mail and
express consignment shipments, and aid in the dismantling of distribution
networks. BEST partners with the CBP, the United States Postal Inspection
Service, and DEA at these facilities. As of September 2019, BESTs are located at
69 locations throughout the nation, including Puerto Rico.
Along with reducing the supply of opioids, Federal and State governments are also playing a key role in curtailing the demand for prescription and
illicit opioids. Prescription drug monitoring programs that track controlled
substance prescriptions are operational in 49 states, the District of Columbia,
264 |

250-840_text_.pdf 268

Chapter 7

2/7/20 3:46 PM

and Guam, and they can provide timely information about prescribing and
patient behaviors that exacerbate the crisis and enable response (CRS 2018).
In 2017, the number of high-dose opioid prescriptions dispensed monthly
declined by over 16 percent, and the prescribing rate of opioids fell to its lowest
rate in more than 10 years. The Administration has also invested over $1 billion
in innovative research to develop effective nonopioid options for pain management. In addition to reducing opioid prescriptions to decrease new initiates
to opioid misuse, the Administration has launched information campaigns to
create awareness and inform the public about opioid use disorder to prevent
new drug users. In June 2018, the White House’s Office of National Drug Control
Policy, the Ad Council, and the Truth Initiative announced a public education
campaign over digital platforms, social media, and television targeting youth
and young adults. Importantly, nearly 60 percent fewer young adults between
the age of 18 and 25 began using heroin in 2018 than in 2016.
Improved guidelines are also being established to target the vulnerable
veteran population, who are twice as likely as the average American to die
from an opioid drug overdose (Wilkie 2018). The Department of Veterans Affairs
(VA) and the Department of Defense updated their Opioid Safety Initiative in
2017 to provide prescribers with a framework to evaluate, treat, and manage patients with chronic pain, including ways to better aggregate electronic
medical records and track opioid prescriptions. In the first six years of the
program, from 2012 to 2018, the number of veteran patients receiving opioids
was reduced by 45 percent. Over the same period, the number of veterans on
long-term opioid therapies declined by 51 percent and the number of veterans
on high-dose opioid therapies declined by 66 percent (Wilkie 2018).
Finally, the Administration is also focusing on treating and saving the
lives of those currently struggling with opioid addictions by expanding access
to the life-saving drug naloxone and other evidence-based interventions,
such as medication-assisted treatment and other recovery support services.
Prevention of drug use is important, but in addition, the Trump Administration
has invested in increased treatment and recovery support for people who
suffer from opioid use disorder. The Surgeon General has promoted access
and carrying naloxone, the lifesaving reversal agent of an opioid overdose.
In October 2018, President Trump signed into law the bipartisan Substance
Use Disorder Prevention That Promotes Opioid Recovery and Treatment
(SUPPORT) for Patients and Communities Act, which includes provisions to
improve substance use disorder treatments for Medicaid patients and to
expand Medicare coverage of opioid use disorder treatment services. In fiscal
years 2018 and 2019, a total of $3 billion was appropriated for State grants to
fund opioid use disorder prevention and treatment. Many States—including
West Virginia, Indiana, Wyoming, Tennessee, Florida, and Virginia—have implemented legislation to expand the availability of naloxone, and inpatient and
outpatient use of the life-saving treatment is increasing (ASTHO 2018).
Understanding the Opioid Crisis | 265

250-840_text_.pdf 269

2/7/20 3:46 PM

Many of the measures taken by the Trump Administration to cut the
supply of opioids, prevent new demand, and save the lives of those currently
struggling with opioid use disorders may have contributed to the flattening
growth of overdose deaths involving opioids. Between January 2017 and May
2019, monthly overdose deaths fell by 9.6 percent. If the growth rate in opioid
overdose deaths from January 1999 through December 2016 had continued,
the CEA estimates that 37,750 additional lives would have been lost due to
opioid overdoses between January 2017 and May 2019. The CEA estimates the
economic cost savings since January 2017 from reduced mortality compared
with the preexisting trend was over $397 billion. The opioid crisis remains at an
emergency level, but its dramatic growth has been halted. Despite successful
efforts to curb the opioid crisis and stop the increase in overdose deaths, there
has been an increase in psychostimulant-related overdose deaths, primarily
driven by methamphetamine use, that is a cause of concern. Psychostimulantrelated deaths now outnumber fentanyl deaths in 12 States (Wilner 2019).
The economic and human costs of drug misuse continue to pose a threat
to the United States. The Trump Administration is working to determine the
underlying causes of the opioid crisis so that it can implement effective solutions. Lower drug prices clearly played a role in the opioid crisis’s growth, and
understanding this dynamic will help policymakers successfully respond to
this threat and avoid mistakes that could lead to another costly, deadly crisis.

266 |

250-840_text_.pdf 270

Chapter 7

2/7/20 3:46 PM

x
Chapter 8

Expanding Affordable Housing
Incomes in the United States are rising, but home prices are rising much faster
in some highly regulated markets. While overall homeownership rates have
increased since 2016, some disadvantaged groups lag behind. As households
turn to the rental market, moderate-income households are dedicating a large
share of their incomes to rent. The housing affordability problem shows no
signs of subsiding in certain markets, as housing construction fails to keep up
with demand, putting upward pressure on home prices and rents.
Fortunately, the majority of areas in the United States have relatively wellfunctioning housing markets in which regulations do not significantly drive
up prices. Indeed, smart regulations that balance the need to build enough
housing to meet growing demand while reflecting the reasonable concerns of
neighborhood residents are achieved by many growing areas in the country.
While areas with relatively moderate home prices may still suffer from some
issues, such as delays for building permits, regulations do not necessarily make
homes substantially less affordable.
However, research has shown that there are 11 metropolitan areas where the
inability to build enough housing to meet demand has driven home prices far
higher than the cost to produce a home. These 11 metropolitan areas include
San Francisco, Honolulu, Oxnard, Los Angeles, San Diego, Washington, Boston,
Denver, New York City, Seattle, and Baltimore.
Housing is particularly difficult to build in these 11 metropolitan areas due to
excessive regulatory barriers imposed by State and local governments. Such
overly restrictive regulations include zoning and growth management controls,
rent controls, building and rehabilitation codes, energy and water efficiency

267

250-840_text_.pdf 271

2/7/20 3:46 PM

mandates, maximum-density allowances, historic preservation requirements,
wetland or environmental regulations, manufactured-housing regulations
and restrictions, parking requirements, permitting and review procedures,
investment or reinvestment tax policies, labor requirements, and impact or
developer fees. Research has linked higher home prices and lower housing
supply to many of these regulations.
Resulting higher housing prices in these 11 metropolitan areas make homeownership less attainable for otherwise-qualified borrowers, thereby constraining
their ability to achieve sustainable homeownership and putting additional
pressure on rental markets for lower- and middle-income households. The
lowest-income households are especially burdened. Among these 11 metropolitan areas, homelessness would fall by an estimated 31 percent on average if
overly burdensome regulations were relaxed. Higher rents resulting from these
regulations also deprive families of Federal rental housing assistance, because
higher government expenditures on households in high-rent areas, through
higher Fair Market Rents, reduce the amount of funds available to serve other
needy families. For example, housing a family in a three-bedroom apartment
can cost the Federal Government more than $4,000 per month in San Francisco
County, California, compared with about $1,500 per month in Harris County,
Texas.
Excessive regulatory barriers to building more housing in these specific areas
also have broader negative effects beyond those imposed on lower-income
Americans. State and local housing regulations reduce labor mobility by
pricing workers out of several of the Nation’s most productive cities, which
stunts aggregate economic growth and increases inequality across regions and
workers. Excessive regulatory barriers also reduce parents’ ability to access
neighborhoods that best advance their children’s economic opportunity. And
by incentivizing families to venture further from their places of work to find
affordable housing, overregulation can increase commuting times to work,

268 |

250-840_text_.pdf 272

Chapter 8

2/7/20 3:46 PM

thus harming the environment, straining local budgets, and decreasing worker
productivity.
Removing government-imposed barriers to more affordable housing is a priority for the Trump Administration. Beyond establishing the White House Council
on Eliminating Regulatory Barriers to Affordable Housing, the Department of
Housing and Urban Development is encouraging State and local governments
to focus on increasing housing supply in areas where supply is constrained.
Increasing housing choice for all Americans requires taking on regulatory
barriers that place housing in large swaths of specific areas out of reach for
lower-income families.

S

ince 2000, real median (posttax/posttransfer) household income has
grown by 20 percent, while real home prices have grown by almost 50
percent, according to the Standard & Poor’s / Case-Shiller Index (CBO
2019). With rising home prices outpacing income gains in some areas, households are spending larger portions of their incomes on housing, and fewer
people can afford to purchase their own homes.
Although the overall homeownership rate has increased since 2016, some
groups lag behind. Based on the four-quarter moving average, the black homeownership rate remains 31.5 percentage points below that of non-Hispanic
white households (see figure 8-1). The Hispanic homeownership rate remains
26.2 percentage points lower than that of non-Hispanic white households,
despite increasing by 1.3 percentage points since the fourth quarter of 2016,
when President Trump was elected. Differences in homeownership between
races exacerbate the wealth gap. In 2016, white families had a median wealth
of $171,000, while black families had a median wealth of $17,600, resulting in
part from their lower homeownership rate (Dettling et al. 2017).
Many American households, particularly low-income households, spend
a large portion of their income on rent. According to the American Community
Survey, out of 43 million renter households in the United States in 2017, 46
percent paid more than 30 percent of their income on housing, 31 percent
paid more than 40 percent, and 23 percent paid more than 50 percent. Among
renters with incomes of less than $20,000 in 2017, about 74 percent paid more
than 30 percent of their income in rent. For those renters with income between
$20,000 and $50,000, about 61 percent paid more than 30 percent of their
income in rent.
Meanwhile, a significant number of Americans go without housing
altogether, sleeping instead on the streets or in homeless shelters. Just over

Expanding Affordable Housing | 269

250-840_text_.pdf 273

2/7/20 3:46 PM

Figure 8-1. Homeownership Rates by Race and Ethnicity, 2000–2019
Homeownership rate (percent)
90

2019:Q3

80

Non-Hispanic white

70

Total

60
50

Hispanic
Black

40
30
2000

2005

2010

2015

Sources: Census Bureau; CEA calculations.
Note: Data represent a four-quarter moving average.

half a million people were homeless on a single night in January 2018, with
35 percent of those found in unsheltered locations not intended for human
habitation, such as sidewalks and public parks (HUD 2018). Research has
linked higher rents to higher rates of homelessness (e.g., Quigley, Raphael, and
Smolensky 2001; Corinth 2017; Hanratty 2017; Nisar et al. 2019).
The housing affordability problem shows no signs of subsiding, given
that home construction fails to keep up with demand in some places, putting
upward pressure on home prices and rents. Indeed, from 2010 to 2016, housing
construction failed to keep pace with household formation, according to the
Census Bureau. Home construction per capita has declined every decade since
the 1970s. While an average of 8.2 homes were built for every 1,000 residents
between 1970 and 1979, annual average construction fell to 3.0 homes per
1,000 residents between 2010 and 2018. Across States, there is large variation in housing construction, according to State-level data from the Bank of
Tokyo–Mitsubishi. For example, from 2010 to 2018, Texas built 5.3 homes and
Florida built 4.3 homes per 1,000 residents, on average. Meanwhile, over the
same period, California built 2.0 homes and New York built 1.7 homes per 1,000
residents.
A key driver of the housing affordability problem is excessive regulatory
barriers to building (single and multifamily) housing in a selected number
of areas in the United States. In a competitive market, developers will build
homes until (economic) profits fall to zero or, in other words, until the price
the developer receives for the home equals the cost to produce the home.
270 |

250-840_text_.pdf 274

Chapter 8

2/7/20 3:46 PM

However, overly burdensome regulations in some areas restrict housing supply
and drive the price of a home above its minimum profitable production cost:
the cost of construction plus the price of land to build on in a free market and
a normal profit margin. In terms of the standard model of supply and demand,
regulations make supply less elastic, causing prices to increase and quantity
to decrease. In this way, Glaeser and Gyourko (2018) note that regulation that
drives home prices above production costs acts as a “regulatory tax” on housing. Regulations that can potentially drive up home prices include, for example,
overly burdensome permitting and review procedures, overly restrictive zoning
and growth management controls, unreasonable maximum-density allowances, historic preservation requirements, overly burdensome environmental
regulations, and undue parking requirements.
It is important to emphasize that an adequate amount of smart regulation is important to address market failures and reflect the reasonable concerns of current neighborhood residents regarding new housing development.
In chapter 1 of this Report, we review evidence that gains in housing wealth
contributed to the growth of total household wealth from 2016 through 2019.
Many growing areas are highly successful in balancing neighborhood concerns
with the need to expand housing supply to meet growing demand. In fact,
housing prices are near or below the cost to produce a home in most areas of
the United States, suggesting that low income levels (despite incomes rising
in recent years) rather than high home prices are the reason some households
struggle to cover housing costs in those areas. However, research has shown
that as a result of excessive local regulatory barriers to building housing,
there are 11 metropolitan areas where the inability to build enough housing
to meet demand has driven home prices far higher than the cost to produce
a home (Glaeser and Gyourko 2018). These 11 metropolitan areas include San
Francisco, Honolulu, Oxnard, Los Angeles, San Diego, Washington, Boston,
Denver, New York City, Seattle, and Baltimore. Even in these areas, it is not
necessary to build high-rise apartments throughout neighborhoods currently
zoned for single-family homes or to eliminate all regulations. Rather, steps to
remove excessive regulatory barriers must be taken so that housing supply can
expand to meet demand and alleviate extreme housing cost burdens placed on
low- and middle-income families.
The excessive regulatory barriers placed on building housing in these
11 metropolitan areas cause economic distress to their current and potential
residents. In addition to restricting the ability of property owners to use their
property in reasonable ways, these regulations increase costs for both renters
and those trying to buy a home. Based on estimates from Glaeser and Gyourko
(2018), excessive regulatory barriers (defined as regulations that drive up home
prices at least 25 percent above home production costs) drive up home prices
by between 36 and 184 percent in each of these 11 metropolitan areas, which

Expanding Affordable Housing | 271

250-840_text_.pdf 275

2/7/20 3:46 PM

also leads to higher rents. These cost burdens are especially problematic for
low-income Americans, who pay the largest share of their income on housing.
By increasing rents, overly burdensome regulatory barriers to building
housing increase homelessness. As estimated by the CEA (2019), relaxing
excessive regulatory barriers in these 11 metropolitan areas where housing
supply is significantly constrained would reduce homelessness by an average of 31 percent in these areas. For example, homelessness would fall by 54
percent in San Francisco, 40 percent in Los Angeles, and 23 percent in New
York. Because these areas contain 42 percent of the U.S. homeless population,
homelessness would fall by 13 percent in the United States overall if each area
adequately addressed its regulatory barriers.
Overregulation of these selected housing markets also reduces the
efficiency of government housing assistance because fewer American families receive assistance for a given budget outlay. In 2019, the Department of
Housing and Urban Development (HUD) was provided $42 billion for its largest
rental housing assistance programs: Section 8 Housing Choice Vouchers ($23
billion), Section 8 Project-Based Rental Assistance ($12 billion), and Public
Housing ($7 billion). Because HUD rental assistance is tied to market rents in
an area, regulations that drive up rents also increase the costs of serving a fixed
number of families. Deregulation that reduces rents in supply-constrained
areas could produce savings for HUD that could be used to serve more families.
For example, Federal taxpayers can pay more than $4,000 per month in rental
assistance toward a three-bedroom unit in San Francisco County, California,
compared with about $1,500 per month in Harris County, Texas.
In addition to specific harmful effects on low-income Americans, excessive regulatory barriers in selected markets have other negative consequences
for all Americans. First, they reduce labor mobility across areas, which stunts
aggregate economic growth and increases inequality across regions and workers. When it is more expensive for workers to live in areas where they are most
productive, they are less likely to do so and their productivity falls. Hsieh and
Moretti (2019), for example, estimate that gross domestic product would have
been 3.7 percent higher by 2009 if housing supply restrictions in the New York,
San Jose, and San Francisco areas were relaxed beginning in 1964.
Second, excessive regulatory barriers to building housing in selected
markets reduces parents’ ability to access neighborhoods that advance their
children’s economic opportunity. A series of papers by Raj Chetty and his colleagues have identified neighborhoods that are most likely to improve longterm outcomes of children (Chetty et al. 2018). High home prices are a common
characteristic of such neighborhoods, suggesting that excessive regulation
that artificially increases home prices may reduce in-migration and diminish opportunity for children. A report from the U.S. Senate Joint Economic
Committee similarly found that the average U.S. zip code with the highestquality public elementary schools has a median home price that is four times
272 | Chapter 8

250-840_text_.pdf 276

2/7/20 3:46 PM

as high as those zip codes with the lowest-quality public schools (JEC 2019).
This is partly due to the willingness of some parents to pay more for homes
located in high-quality school districts. Many of these areas have excessive
regulatory barriers, however.
Third, excessive regulatory barriers to building housing increase commuting times because housing cannot be built near where people work,
increasing driving time and traffic congestion, which harm the environment.
The average commuter spent 54 hours in traffic congestion in 2017, up from
20 hours in 1982 (Schrank, Eisele, and Lomax 2019). The aggregate travel delay
increased from 1.8 billion hours to 8.8 billion hours over this period, and the
total cost associated with congestion rose from $15 billion to $179 billion. As
a result of this rise in average commuting times, an extra 3.3 billion gallons of
fuel were consumed.
Fortunately, growing evidence of the importance of addressing excessive
regulatory barriers to building housing has led to increased bipartisan focus
on the issue. The CEA under the previous Administration released a “Housing
Development Toolkit” in 2016 for State and local regulators. While some of
the proposed reforms could be problematic, the toolkit called for a number of
productive steps to reduce local government barriers to housing development.
These reforms include establishing by-right development to streamline the
process for approving projects, permitting multifamily zoning to boost urban
density, and shortening the process for obtaining building permits (CEA 2016).
Some counterproductive reforms were also suggested, including requirements
that developers build certain types of units with regulated rents in exchange
for building more market-rate units, a policy that can potentially hinder overall
supply expansions and increase prices in some areas (Schuetz, Meltzer, and
Been 2011). The CEA (2016) connected regulatory barriers to a number of problems, including stunted economic growth, increased inequality, harm to the
environment, and increased homelessness.
To more successfully address the overregulation of housing markets,
President Trump signed an Executive Order on June 25, 2019, establishing
the White House Council on Eliminating Regulatory Barriers to Affordable
Housing. Recognizing the harmful impact of these regulations on economic
growth, opportunities for children, homelessness, and the cost of government
programs, the council is tasked with identifying the most burdensome Federal,
State, and local regulatory barriers to housing supply as well as actions that
can best counter them. The Executive Order requires the council to determine
how each Federal agency can curtail impediments to housing development,
including in ways that “align, support, and encourage” State and local authorities to address local regulatory barriers.
HUD has also taken action under the Trump Administration to counter
regulatory barriers to building affordable housing. The Affirmatively Furthering
Fair Housing rule, which was finalized during the previous Administration,
Expanding Affordable Housing | 273

250-840_text_.pdf 277

2/7/20 3:46 PM

is being revised to focus more clearly on increasing housing supply in areas
where supply is constrained, rather than encouraging localities to subsidize
housing in more affluent areas. This rule recognizes that increasing housing
choice for disadvantaged groups requires taking on regulatory barriers that
place housing in large swaths of specific areas out of reach for lower-income
families.
This chapter proceeds by first documenting the housing affordability
problem in the United States. It then identifies the key role that excessive regulatory barriers play in the problem in a selected number of metropolitan areas.
Next, it provides evidence of the many harmful consequences of these barriers,
especially harm to low-income Americans. Finally, it concludes by discussing
actions the Administration has taken to encourage the relaxation of excessive
regulatory barriers in local housing markets.1

The Housing Affordability Problem
When home prices rise faster than incomes, fewer households can afford to
purchase a home. Those still able to qualify for a loan and purchase a home
may do so in neighborhoods or regions with fewer opportunities, and they may
commit larger shares of their income to mortgage payments and savings to a
down payment. Renter households may pay a greater portion of their income
in rent, leaving less income available for other needs. The burden is especially
severe for lower-income households. By these definitions, the “housing affordability” problem in America is worsening, a result of home prices that have
outpaced income gains and home construction that has not kept up with
demand in certain areas.
Based on a four-quarter moving average, as of the third quarter of
2019, 64.5 percent of households owned their own homes (figure 8-1). This
represents an increase of 1.1 percentage points since reaching its low point
in 2016:Q3. However, the current homeownership rate is still 4.6 percentage
points below its 69.1 percent peak in 2005:Q1.
Some groups have particularly low homeownership rates. The black
homeownership rate was 41.8 percent in 2019:Q3, 31.5 percentage points
below the non-Hispanic white homeownership rate (figure 8-1). While the
Hispanic homeownership rate increased by 1.3 percentage points since
2016:Q4, when President Trump was elected, it was still at 47.2 percent in
2019:Q3, 26.2 percentage points lower than that of non-Hispanic white households (figure 8-1).
For those who are homeowners, owned homes are an important source
of wealth. Thus, gaps in homeownership rates have direct implications for
1The CEA previously released research on topics covered in this chapter. The text that follows
builds on the research paper produced by the CEA titled “The State of Homelessness in America”
(CEA 2019).

274 | Chapter 8

250-840_text_.pdf 278

2/7/20 3:46 PM

wealth gaps. According to the Federal Reserve Board’s Survey of Consumer
Finances, in 2016, white families had a median wealth of $171,000, while black
families had a median wealth of $17,600 and Hispanic families had a median
wealth of $20,700, partly as a result of their much lower homeownership rates
(Dettling et al. 2017).
Among those who own a home, mortgages can take up a large share
of income, especially for lower-income families. In 2017, housing costs represented 67.5 percent of household income for homeowners with less than
$20,000 in annual income, and 40.6 percent of income for homeowners with
between $20,000 and $50,000 in annual income (Dumont 2019). Thus, housing
affordability can be a problem even for those able to purchase their own home.
In chapter 1 of this Report, we discuss how current low mortgage rates on the
whole should support the housing market. However, other factors, such as high
mortgage underwriting costs, hurt mortgage affordability.
As homeownership rates have fallen, the number of renter households
has grown. The Federal Reserve Board estimates that of the 6.2 million
households formed between 2009 and 2017, 5.7 million (92 percent) were
new renter households (Dumont 2019). Renter households pay large shares of
their income on rent—without building equity—which can make it difficult for
low- and moderate-income households to address other needs. From 1970 to
2010, the share of renter households spending more than half of their income
on housing increased from 16 percent to 28 percent; over the same period,
the share spending at least 30 percent on housing increased from 31 percent
to 52 percent (Albouy, Ehrlich, and Liu 2016). According to the 2017 American
Community Survey, out of 43 million renter households in the United States,
46 percent pay more than 30 percent of their income on housing, 31 percent
pay more than 40 percent, and 23 percent pay more than 50 percent. As shown
in table 8-1, among renters with incomes of less than $20,000 in 2017, about
74 percent paid more than 30 percent of their income in rent, a smaller share
than in 2009. For those renters with incomes between $20,000 and $50,000, 61
percent paid more than 30 percent of their income in rent, rising from about 50
percent in 2009.
Meanwhile, a significant number of Americans go without housing
altogether, sleeping instead on the streets or in homeless shelters. Just over
half a million people were homeless on a given night in January 2018, with
35 percent of those found in unsheltered locations not intended for human
habitation, such as sidewalks and public parks (HUD 2018). Research has linked
higher rents to higher rates of homelessness (e.g., Quigley 2001; Corinth 2017;
Hanratty 2017; Nisar et al. 2019).
The growing housing affordability problem is not driven by falling
incomes (with the exception of the Great Recession, which led to severe housing problems, including widespread foreclosures; see Steffen et al. 2013). Since
2000, real median (posttax, posttransfer) household income increased by 20
Expanding Affordable Housing | 275

250-840_text_.pdf 279

2/7/20 3:46 PM

Table 8-1. Percentage of Renter Households Paying More Than 30
Percent of Income on Housing by Income, 2009 versus 2017
(percent)

Percentage
point change

Percent
change

76.6

74.3

-2.3

-3.0

$20,000 to $49,999

50.2

61.0

10.8

21.5

$50,000 to $74,999

15.2

23.5

8.3

54.4

$75,000 to $99,999

6.8

10.3

3.5

51.3

$100,000 or more
2.1
3.5
1.3
All renter
47.7
46.0
-1.7
households
Sources: American Community Survey; CEA calculations.

61.8

2009

2017

(percent)

Less than $20,000

Household income

-3.6

percent (CBO 2019). Real income gains were even larger for the bottom fifth
of households (CBO 2019). The driver of growing unaffordability is rising home
prices. According to the Standard & Poor’s / Case-Shiller U.S. National Home
Price Index, real home prices have increased by 49 percent since 2000, outpacing real median income gains. Home prices have increased the fastest for entrylevel homes—according to the American Enterprise Institute National Home
Price Appreciation Index, home prices in the lowest price tier have increased
more than 50 percent more than home prices in the highest price tier since
2012 (Pinto and Peter 2019). As shown in box 8-1, the housing affordability
problem is concentrated in a selected number of areas in the United States,
where the people who build houses are unable to afford them.
Although home prices are rising, home construction has been slow to
respond, implying that supply is not keeping up with the demand for homes in
certain places. Home construction per capita has declined every decade since
the 1970s, according to the Census Bureau. While an average of 8.2 homes were
built for every 1,000 residents between 1970 and 1979, annual average construction fell to 3.0 homes per 1,000 residents between 2010 and 2018. Across
States, there is large variation in housing construction. For example, from 2010
to 2018, Texas built 5.2 homes and Florida built 4.3 homes per 1,000 residents,
on average. Meanwhile, over the same period, California built 2.0 homes and
New York built 1.7 homes per 1,000 residents. This represents a large decline
for California, which built more than 7.0 homes per 1,000 residents in the 1970s
and 1980s before falling to less than 4.0 per 1,000 residents in every decade
since then. Meanwhile, New York is one of only two States in the country (along

276 |

250-840_text_.pdf 280

Chapter 8

2/7/20 3:46 PM

Box 8-1. Measuring the Housing Affordability
Problem with the Carpenter Index
One way to assess the affordability of housing is to ask whether the people
who build homes can afford to buy them. The American Enterprise Institute’s
Carpenter Index compares the average income of households headed by
carpenters to home prices in a given area. If the price of a home is less than
three times the carpenter’s household income, then that home is deemed
“affordable.” For each metropolitan area, the index calculates the share of
entry-level homes that are affordable to the carpenter.
Figure 8-i shows the share of the entry-level housing stock that is
affordable for the 100 largest CBSAs, with the darker shades illustrating
areas where housing is less affordable to the average carpenter. The average carpenter can afford only 6.5 percent of entry-level homes built in the
San Diego–Carlsbad, California, CBSA; 8.2 percent in the Oxnard–Thousand
Oaks–Ventura, California, CBSA; 10.3 percent in the Los Angeles–Long Beach–
Anaheim, California, CBSA; 10.7 percent in the San Jose–Sunnyvale–Santa
Clara, California, CBSA; and 11.8 percent in the San Francisco–Oakland–
Hayward, California, CBSA—the five least affordable areas in the country.
By contrast, the average carpenter can afford 100 percent of entry-level
homes in the Chicago–Naperville–Elgin, Illinois–Indiana–Wisconsin, CBSA;
the Pittsburgh, Pennsylvania, CBSA; the Saint Louis, Missouri–Illinois, CBSA;

Figure 8-i. The Carpenter Index by CBSA, 2018

Source: American Enterprise Institute.
Note: CBSA = core-based statistical area.

Expanding Affordable Housing | 277

250-840_text_.pdf 281

2/7/20 3:46 PM

and a number of other areas in the Midwest. The index signals that the most
expensive metropolitan areas are located in California and to a lesser extent
the rest of the West Coast and the Northeast, while most of the affordable
metropolitan areas are located in the Midwest.

with West Virginia) that has never built more than 3.0 homes per 1,000 residents in an average year across every decade since the 1970s.

The Role of Overregulation in the
Housing Affordability Problem
When the housing affordability problem is defined as housing expenditures
that constitute a sufficiently large share of income, there are three potential
causes: (1) rising home prices, (2) falling household incomes, and (3) choices
among households to consume higher-quality homes (with either high physical
quality or in closer proximity to desirable amenities). As reported in the previous section, real home prices have risen 49 percent since 2000. Meanwhile,
household incomes are rising rather than falling, and consumer decisions to
choose higher-quality homes should not be considered an affordability problem. Thus, the fundamental problem with housing affordability in the United
States today is excessively high home prices in certain areas.
Overly stringent housing regulations play a key role in driving up home
prices in the face of growing demand. Figure 8-2 shows how excessive regulatory barriers to building housing in some areas constrain supply and thus
increase home prices. In a market unconstrained by excessive regulation,
developers can build new homes at a constant cost when demand shifts outward (for example, because higher wages increase the desirability of living in
an area), and thus, price remains constant at P1 while quantity increases to Q2.
By contrast, new home construction cannot keep up with growing demand in
a market constrained by excessive regulations, such as lengthy permitting processes and unreasonable land use regulations. Excessive regulations lead to an
upward sloping, relatively more inelastic housing supply curve, which drives
home prices above the cost to produce a home in a market without excessive
regulatory barriers. Prices rise to P2 and quantity falls to Q1. In this way, Glaeser
and Gyourko (2018) note that excessive regulation that drives home prices
above production costs acts as a “regulatory tax” on housing. This regulatory
tax is represented in figure 8-2 as the gap between P1 and P2.
Some regulations add additional costs to the development process, driving up the total cost of housing development and reducing supply. For example,
environmental reviews can delay construction, imposing additional costs on
developers. An unintended consequence of these regulations is that housing is

278 | Chapter 8

250-840_text_.pdf 282

2/7/20 3:46 PM

Figure 8-2. The Effect of Regulation on Supply and Demand for
Housing
Price
Supply (regulated)

P2
Supply (deregulated)

P1

Demand (new)
Demand (old)

Q0

Q1
Quantity

Q2

Sources: Glaeser and Gyourko (2018); CEA calculations.

instead built in less central areas where regulations do less to drive up home
prices, which can increase commuting times and ultimately cause even greater
environmental harm. More generally, approval processes for new development can be lengthy and uncertain, thus increasing the price and reducing
the supply of housing by, for example, forcing developers to carry high-cost
construction loans for a longer period of time, or having to spend additional
money on extending options to purchase land. Gyourko, Hartley, and Krimmel
(2019) formulate an Approval Delay Index and find that the review time for
housing construction projects is more than twice as long in highly regulated
areas compared with relatively lightly regulated areas, with an average review
time of 8.4 months. Environmental reviews alone can add substantial costs to a
housing project. For example, the California Environmental Quality Act, which
requires certain construction in California to undergo an environmental impact
assessment, can add an estimated $1 million in costs to completing a housing
development (Jackson 2018).
Other regulations that can potentially constrain supply are focused
explicitly on reducing density. Building permit caps, population caps, and
density restrictions limit the amount of new housing that can be built in an
area. Similarly, urban growth boundaries prevent urban expansion beyond
designated areas. Other kinds of regulations reduce density by regulating the
type and size of housing that can be constructed in a locality. Minimum lot size

Expanding Affordable Housing | 279

250-840_text_.pdf 283

2/7/20 3:46 PM

requirements prevent homebuilders from subdividing a lot in order to build
more homes. Height restrictions prevent taller buildings with more floors and
more housing units. Maximum floor area ratios (which are calculated by dividing floor area by lot area) limit the amount of living space, potentially across
multiple units, that can be built on a given lot. Zoning regulations also may
prevent certain types of housing, such as multifamily buildings, from being
constructed.
Of course, when these types of regulations are not excessive, they can
be beneficial—for example, by maintaining standards that promote safety, or
by providing information about housing characteristics—without significantly
constraining supply. In addition, certain types of land use may generate pollution or congestion externalities, and some amount of regulation, such as
impact fees, can help developers internalize these costs of construction. Local
citizens may also wish to preserve certain land for public use or conservation
purposes, such as parks. However, in a selected number of places, excessive
regulations prevent supply from expanding to meet housing demand, substantially driving up home prices.
It is generally believed among economists that the overall effect of
excessive regulatory barriers that constrain housing supply is to reduce overall
well-being. For example, Albouy and Ehrlich (2018, 117) not only find that
stringent housing regulation increases home prices, but also that any benefits
of these regulations for improving quality of life are outweighed by their cost.
They note: “On net, the typical land-use regulation in the United States reduces
well-being by making housing production less efficient and housing consumption less affordable.” Glaeser and Gyourko (2018, 14) summarize the literature
and state: “Empirical investigations of the local costs and benefits of restricting
building generally conclude that the negative externalities are not nearly large
enough to justify the costs of regulation.”
The stringency of housing regulations and their impact on housing supply vary across the country. One way to measure the stringency of regulations
is to analyze the regulations themselves. One measure that is heavily relied
upon is the Wharton Residential Land Use Regulatory Index. Gyourko, Saiz, and
Summers (2008) constructed the index from a national survey of municipalities
regarding their regulatory process and land use regulations. The resulting index
is shown by metropolitan statistical area in figure 8-3, with a darker shade of
blue indicating cities that have more stringent land use regulations. The South
and the Midwest have the least restrictive regulations, while California and the
Northeast have the most.
Areas with higher regulatory burdens tend to have higher home prices.
Figure 8-4 shows metropolitan areas by the ratio of their median home prices
to the cost to produce a home, as constructed by Glaeser and Gyourko (2018).
Where regulations are lax, the ratio of home prices to production costs should
be near or below 1. Where regulations are more stringent and demand is strong,
280 | Chapter 8

250-840_text_.pdf 284

2/7/20 3:46 PM

Figure 8-3. Wharton Land Use Index by Metropolitan Statistical Area,
2008

Source: Gyourko, Saiz, and Summers (2008).

Figure 8-4. Ratio of Home Prices to Production Costs by CBSA, 2013

Sources: Glaeser and Gyourko (2018); CEA calculations.
Note: CBSA = core-based statistical area.

Expanding Affordable Housing | 281

250-840_text_.pdf 285

2/7/20 3:46 PM

ratios may exceed 1. It is important to note that production costs include not
only the construction cost of the home but also a normal profit margin and
a small cost of land on which to build the home that would be achieved in a
market without overly stringent regulations.
It is certainly the case that, even in an unconstrained market, land prices
for a fixed size plot (i.e., an acre) of land will be higher in more desirable locations. Davis and others (2019) document large variation in land prices per acre
across the United States—much of this variation would remain even if all areas
relaxed overly stringent housing regulations. However, the price of a parcel of
land used for each housing unit may be similar across areas absent excessive
regulation. In dense areas, each housing unit would require a smaller plot of
land, and so, though the price of an acre of land is likely to be higher in denser
areas, the cost of the smaller piece of land used for each two-bedroom housing unit may be similar to the cost of the larger piece of land used for a twobedroom unit in less dense areas. Of course, this will only roughly be true, and
other factors, such as differences in property taxes, may drive some remaining
differences. Partly for this reason, Glaeser and Gyourko (2018) focus on areas
where home prices significantly exceed production costs.
Figure 8-4 shows that the places where ratios of home price to production cost significantly exceed 1 (i.e., where home prices are at least 25 percent
higher than home production costs) are largely the same places with high
regulatory indices. Though correlational, this provides suggestive evidence
that housing regulations help determine home prices. Figure 8-4 also indicates
that excessive regulation is currently a major problem in a selected number of
places, indicated by the darker shade of blue. As noted earlier in this chapter,
these 11 metropolitan areas include San Francisco, Honolulu, Oxnard, Los
Angeles, San Diego, Washington, Boston, Denver, New York City, Seattle, and
Baltimore.
Examples of overly burdensome regulations abound in these 11 CBSAs.
Four of the 11 are located in California, where multifamily homes may be built
on less than a quarter of the land in Los Angeles, Long Beach, Anaheim, and San
Diego and less than half of the land in San Francisco and Oakland (Mawhorter
and Reid 2018). In the cities of Los Angeles and San Diego, two parking spaces
are required for every typical two-bedroom apartment, one and a half parking
spaces are required for every typical one-bedroom apartment, and one parking
space is required for every studio apartment, increasing costs for multifamily
housing developers and, ultimately, renters (San Francisco eliminated its parking requirements in early 2019). Across Hawaii, only 4 percent of land may be
developed due to its network of local and State zoning regulations.
Although overly burdensome permitting processes and other barriers
may still be a problem and put some degree of upward pressure on home
prices in the rest of the country, the major problem with excessive regulation
is currently limited to these 11 areas. Nonetheless, future demand growth in
282 |

250-840_text_.pdf 286

Chapter 8

2/7/20 3:46 PM

additional areas with excessive regulatory barriers could increase the number
of areas with artificially inflated home prices.
Consistent with figures 8-3 and 8-4, a number of academic studies find
that stringent regulation increases housing prices. In a review of much of the
earlier literature, Ihlanfeldt (2004) concludes that growth controls and minimum lot size restrictions reduce the supply of housing and increase its price.
Quigley and Raphael (2005) find that cities in California with more stringent
regulations have higher levels and growth in home prices and rents, and that
housing supply is much less responsive to price increases in more regulated
areas. Glaeser, Gyourko, and Saks (2005) argue that land-use restrictions
explain why prices for high-rise apartments in Manhattan far exceed the cost
to construct them. Ihlanfeldt (2007) finds that more stringent land-use regulation increases home prices in Florida. Glaeser and Ward (2009) find that more
stringent regulations, especially minimum lot sizes, are associated with higher
home prices and less construction in Massachusetts. Saiz (2010) finds that
land-use regulations, in addition to geographical constraints, are important
determinants of the responsiveness of housing supply to price increases.
Summarizing the literature, Glaeser and Gyourko (2018, 8) state: “The general
conclusion of existing research is that local land use regulation reduces the
elasticity of housing supply, and that this results in a smaller stock of housing,
higher house prices, greater volatility of house prices, and less volatility of new
construction.”
Some might argue that there are reasons other than regulation that might
be driving higher home prices. One reason could be that construction costs are
rising. However, Gyourko and Molloy (2015) find that real construction costs
(including the cost of labor and materials) remained relatively constant from
1980 to 2013. Another potential cause is geographical constraints on building. For example, Saiz (2010) argues that many areas with supply constraints
have steep-sloped terrain that prevents the development of new housing.
Nonetheless, even in areas that appear to have land constraints, developers
could build more densely and with fewer permitting delays, which would exert
downward pressure on housing prices. Finally, though we focus on supply,
housing regulations may also increase prices through increased demand for
housing if land use restrictions increase the appeal of living in a certain community. Empirically, however, Albouy and Ehrlich (2018) find that supply effects
dominate demand effects.

Consequences of Overregulation of Housing
The overregulation of housing markets in selected metropolitan areas has
several negative consequences. By increasing home prices well above home
production costs, it increases the cost of attaining homeownership and
increases the rent for renter households. It hurts low-income Americans in

Expanding Affordable Housing | 283

250-840_text_.pdf 287

2/7/20 3:46 PM

particular by increasing homelessness and by reducing the number of people
government housing assistance programs can serve. More generally, it reduces
labor mobility across areas and thus weakens economic growth, reduces the
ability of children to access high-opportunity neighborhoods, and harms the
environment.

The Increased Cost of Attaining Homeownership and Higher
Rents
In most areas in the United States, reasonable regulations do not substantially
drive up home prices. But in a selected number of metropolitan areas, excessive regulatory barriers to building housing substantially increase the price of
purchasing a home above the cost to produce it.
Figure 8-5 shows the extent to which excessive regulations drive up home
prices in these 11 metropolitan areas, according to data published by Glaeser
and Gyourko (2018) and shown above in figure 8-4. Home prices are more
than 150 percent higher in the San Francisco–Oakland–Hayward, California,
CBSA, and the Urban Honolulu, Hawaii, CBSA; are about 100 percent higher in
the Oxnard–Thousand Oaks–Ventura, California, CBSA; the Los Angeles–Long
Beach–Anaheim, California, CBSA; and the San Diego–Carlsbad, California,
CBSA—and are 36 percent higher in the Baltimore–Columbia–Towson,
Maryland, CBSA, the smallest price premium of the 11 supply-constrained
metropolitan areas.
The higher home prices resulting from excessive regulations make it
more difficult for households to purchase their own homes and build wealth. As
HUD Secretary Ben Carson recently stated, “As a result [of the shortage in the
housing supply], Americans have fewer housing opportunities, including the
opportunity to achieve sustainable homeownership, which is the No. 1 builder
of wealth for most U.S. families” (Carson 2019). Excessive regulation also
increases rents in these 11 metropolitan areas, because higher home prices
increase the amount property owners need to receive in revenue each year to
maintain a normal profit margin. Higher rents are especially burdensome for
lower- and moderate-income Americans—and, for some, may make it prohibitively expensive to live in these excessively regulated areas.

Increased Homelessness
Another harmful effect of overregulation of housing markets is its impact on
homelessness. Several studies that rely on data on homelessness over time
in various communities find that a 1 percent increase in rent is associated
with about a 1 percent increase in homelessness. Because housing regulations generally drive up rents, they should thus be expected to also increase
homelessness.
The CEA (2019) estimates the extent to which removing excessive regulatory barriers that reduced home prices to their production costs would reduce
284 | Chapter 8

250-840_text_.pdf 288

2/7/20 3:46 PM

Figure 8-5. Home Price Premium Resulting from Excessive Housing
Regulation
Baltimore–Columbia–Towson, MD
Boston–Cambridge–Newton, MA-NH
Denver–Aurora–Lakewood, CO
Los Angeles–Long Beach–Anaheim, CA
New York–Newark–Jersey City, NY-NJ-PA
Oxnard–Thousand Oaks–Ventura, CA
San Diego–Carlsbad, CA
San Francisco–Oakland–Hayward, CA
Seattle–Tacoma–Bellevue, WA
Urban Honolulu, HI
Washington–Arlington–Alexandria, DC-VA-MD-WV

0

50

100
Percent

150

200

Sources: Glaeser and Gyourko (2018); CEA calculations.

Figure 8-6. Percentage Reduction in Homelessness by CBSA from
Deregulating Housing Markets
Baltimore–Columbia–Towson, MD
Boston–Cambridge–Newton, MA-NH
Denver–Aurora–Lakewood, CO
Urban Honolulu, HI
Los Angeles–Long Beach–Anaheim, CA
New York–Newark–Jersey City, NY-NJ-PA
Oxnard–Thousand Oaks–Ventura, CA
San Diego–Carlsbad, CA
San Francisco–Oakland–Hayward, CA
Seattle–Tacoma–Bellevue, WA
Washington–Arlington–Alexandria, DC-VA-MD-WV

0

20
40
60
Reduction in homelessness (percent)

Sources: Department of Housing and Urban Development, Point-in-Time Counts, 2018; Census
Bureau; Corinth (2017); Glaeser and Gyourko (2018); Goodman (2004); CEA calculations.
Note: CBSA = core-based statistical area. Each continuum of care is merged into the metropolitan
area where the majority of its overall population lives. This simulation assumes that deregulation
reduces the ratio of home value to production cost to 1 for all metropolitan areas with a ratio of
at least 1.25; see the text for further details about the simulation.

Expanding Affordable Housing | 285

250-840_text_.pdf 289

2/7/20 3:46 PM

homelessness. The results are summarized in figure 8-6. Homelessness would
fall by 54 percent in the San Francisco–Oakland–Hayward, California, CBSA;
by 50 percent in the Urban Honolulu, Hawaii, CBSA; by 40 percent in the Los
Angeles–Long Beach–Anaheim, California, CBSA; by 38 percent in the San
Diego–Carlsbad, California, CBSA; by 36 percent in the Washington–Arlington–
Alexandria, D.C.–Virginia–Maryland–West Virginia, CBSA; and by between
19 and 26 percent in the Boston–Cambridge–Newton, Massachusetts–New
Hampshire, CBSA; the Denver–Aurora–Lakewood, Colorado, CBSA; the New
York–Newark–Jersey City, New York–New Jersey–Pennsylvania, CBSA; the
Seattle–Tacoma–Bellevue, Washington, CBSA; and the Baltimore–Columbia–
Towson, Maryland, CBSA.
The aggregate reduction in homelessness in these 11 metropolitan
areas, which contain 42 percent of the U.S. homeless population, would have
important effects for the United States as a whole, with total U.S. homelessness falling by just under 72,000 people, or 13 percent. These findings are
also broadly consistent with results from Raphael (2010), who uses a different
methodology to assess how housing market regulation drives up homelessness
rates. Using an index of housing market regulation by metropolitan area, he
finds that deregulation could reduce overall United States homelessness by
7 to 22 percent. He does not show how homelessness reductions would vary
across specific areas. It is important to note that the housing supply responses
resulting from deregulation would take many years to translate into the types
of price reductions, and thus homelessness reductions, shown here. Still, these
results suggest that the severe homelessness problems in a selected number
of metropolitan areas are substantially driven by city-created regulations on
housing.

Fewer People Are Served by Housing Assistance Programs
By driving up rents, overly stringent housing regulations in selected metropolitan areas increase the government’s cost of providing rental housing
assistance, resulting in fewer assisted families. The Federal Government
provides rental housing assistance across a number of programs that are
administered by different agencies. Three major programs are administered
by HUD—these include (1) Section 8 Housing Choice Vouchers, (2) Section 8
Project-Based Rental Assistance, and (3) public housing. The largest of these
three HUD programs is the Housing Choice Voucher program, which served 2.3
million families at a cost of $23 billion in fiscal year (FY) 2019 (42 percent of the
overall HUD budget). Under the voucher program, qualified tenants receive
Federal subsidies that cover a portion of their rent in private rental apartments
of their choosing. The second-largest HUD program is Section 8 Project-Based
Rental Assistance, which served 1.2 million families at a cost of $12 billion in
FY 2019. Under Project-Based Rental Assistance, apartment owners receive
government subsidies to lease units to low-income families. The third-largest

286 |

250-840_text_.pdf 290

Chapter 8

2/7/20 3:46 PM

HUD program is public housing, which served 1.0 million families in FY 2019, at
a Federal operating cost of $7 billion (excluding the opportunity cost of holding
the property). Public housing is built and managed by government authorities. Unlike with Housing Choice Vouchers, tenants living in units covered by
Project-Based Rental Assistance and in public housing do not maintain their
subsidy if they move.
Eligibility for these programs is based on a family’s income relative to
median income in their area. However, only about one in four eligible families
actually receives assistance, because housing costs are too high to serve every
family that meets the income requirements for the programs, especially in
high-cost areas. For example, the maximum payment standard for a three-bedroom unit is more than $4,500 per month in San Francisco County, California,
compared with about $1,500 per month in Harris County, Texas. Many areas
have waiting lists for assistance that extend multiple years, and in some cases,
waiting lists are not reopened for long periods of time.
Housing deregulation that removes excessive barriers and reduces market rents could extend assistance to many eligible families not currently being
served in expensive markets. Under each of the three major HUD programs, the
government generally covers the difference between 30 percent of a household’s adjusted income and the allowable rent or operating cost for housing
units. For the voucher program, if market rents decrease, Public Housing
Authorities would pay less for contract rent, assuming the tenants’ payments
remain mostly constant at 30 percent of adjusted income. HUD would also
need to pay private property owners less to house people under Project-Based
Rental Assistance. These savings from deregulation could be used to serve
additional families under current funding amounts.
Removing excessive regulatory barriers could also improve the effectiveness of the Low-Income Housing Tax Credit (LIHTC), a program that subsidizes the developers of affordable housing units. The Federal Government is
estimated to spend about $9 billion per year on LIHTC (JCT 2017). Given the
budgetary restrictions on how much can be spent on this program, excessive
housing regulation increases the costs of building subsidized housing and
reduces the amount of it that can built.

Weakened Labor Mobility and Economic Growth
Aside from its specific harm to low-income Americans, excessive regulation in
selected housing markets also has negative consequences for the general population. One important example is the reduction in labor mobility across areas
because higher home prices in certain areas reduce the incentive to move to
places where wages may be higher. This reduces the productivity of workers
and shrinks aggregate economic output. Hsieh and Moretti (2019) estimate that
reducing housing regulations in New York City, San Jose, and San Francisco to
that of the median U.S. city would have substantially increased growth from

Expanding Affordable Housing | 287

250-840_text_.pdf 291

2/7/20 3:46 PM

1964 to 2009, leading to 3.7 percent higher gross domestic product in 2009.
Hsieh and Moretti argue that this missing growth is the result of spatial misallocation of workers, as high-productivity cities construct barriers to increasing
housing supply to meet demand from workers. Glaeser and Gyourko (2018) find
that restrictive land use regulations reduce national output by a smaller but
still important 2 percent. Herkenhoff and others (2018) similarly find significant
economic growth effects from relaxing land use restrictions.
Reducing labor mobility has important regional effects in addition to
aggregate ones. When home prices are higher due to overregulation, workers
are less able to migrate to areas with higher wages. This results in a persistent
gap in wages between high-productivity and low-productivity areas that cannot be reduced through migration that would expand the supply of workers
in high-wage areas. Zabel (2012) finds that housing prices increase more in
response to an increase in labor demand in cities with an inelastic housing
supply than in those with a more elastic housing supply, thus reducing the
incentive for in-migration to areas with an inelastic housing supply. Saks (2008)
similarly finds that more heavily regulated housing markets are less responsive
to changes in demand for housing, lowering employment growth in areas
with relatively more extensive land use regulations. Saks estimated that the
employment response to an increase in labor demand in an area in the 75th
percentile of her State regulatory index is 11 percentage points smaller than
the response in an area in the 25th percentile.
Ganong and Shoag (2017) find that higher home prices resulting from
stringent land use regulation can help explain why disparities between economic regions have grown since 1980, breaking from the previous pattern of
regional economic convergence. Hämäläinen and Böckerman (2004) examine
migration in Finland and come to a similar conclusion as Ganong and Shoag:
high housing prices discourage in-migration.
Even within cities, high levels of land use regulations can increase socioeconomic segregation. Owens (2019) examines segregation between neighborhoods, between places (municipalities, cities, and towns), and between cities
and their suburbs and finds that most housing segregation occurs between
neighborhoods, rather than between places or between cities and their suburbs, which suggests that zoning regulations could play an important role.
Rothwell and Massey (2010) find that restrictive zoning laws lead to greater
socioeconomic segregation and reduce interaction between the poor and the
affluent. Lens and Monkkonen (2016) find that land-use regulation and income
segregation are positively related, with density restrictions leading to a concentration of more affluent households, although not necessarily a concentration of poor households.

288 | Chapter 8

250-840_text_.pdf 292

2/7/20 3:46 PM

Reduced Opportunity for Children
Overregulation of housing markets can also potentially reduce the ability of
children to access neighborhoods that advance opportunity. A series of papers
by Raj Chetty and his colleagues have identified neighborhoods that are most
likely to improve long-term outcomes of children. A child that moves from a
neighborhood at the 25th percentile to the 75th percentile of the opportunity
index increases his or her lifetime earnings by $206,000. Chetty and others
(2018) calculate the “cost of opportunity,” and find that an additional $1,000
in children’s future annual income costs $190 each year for rent for every year
of childhood. The cost of opportunity varies considerably across the United
States, however, and much of the variance is due to differences in land use
regulatory regimes. An additional $1,000 in future annual income for a child
costs only $47 in Wichita but $260 in Boston or Baltimore. Thus, relaxing excessive regulatory barriers to building housing could reduce the cost for families of
accessing higher-opportunity neighborhoods for their children and potentially
improve their long-term prospects.
Similarly, a report from the U.S. Senate Joint Economic Committee
finds that U.S. zip codes with the highest-quality public elementary schools
have a median home price that is four times as large as those zip codes with
the lowest-quality public schools (JEC 2019). Many of these areas have highly
restrictive zoning. Although expanded school choice weakens the association
between home prices and the quality of public schools, housing deregulation
could potentially promote greater access to high-quality schools for students
(JEC 2019).

Increased Traffic Congestion and Harm to the Environment
Finally, excessive regulatory barriers to building housing in certain areas
increases commuting times and traffic congestion because sufficient housing
cannot be built near where people work. The average commuter spent 54
hours sitting in traffic in 2017, up from 20 hours in 1982 (Schrank, Eisele, and
Lomax 2019). The aggregate travel delay increased from 1.8 billion hours to 8.8
billion hours during this period, and the total cost associated with congestion
rose from $15 billion to $179 billion.
As a result of this rise in average commuting times, an extra 3.3 billion
gallons of fuel were consumed, increasing carbon emissions and harming the
environment. Moreover, as Glaeser notes, “when environmentalists resist new
construction in their dense but environmentally friendly cities, they inadvertently ensure that it will take place somewhere else—somewhere with higher
carbon emissions” (Glaeser 2009). Indeed, Glaeser (2009) finds that households
in urban areas emit less carbon than those in the suburbs, even after adjusting
for differences in climate and environmental regulation across these areas.
Factors contributing to fewer emissions in cities include smaller housing units
and that people are less likely to drive or would drive shorter distances than
Expanding Affordable Housing | 289

250-840_text_.pdf 293

2/7/20 3:46 PM

Box 8-2. Poor Substitutes for Regulatory Reform
Policymakers have proposed a litany of policies aside from regulatory reform
to lower rents or incentivize affordable housing construction in high-cost
areas. However, these proposals alone—such as rent control, increases in
rental housing assistance, and so-called inclusionary zoning—are unlikely to
have their intended effects on rents or construction, and in some cases may
be counterproductive.
Rent controls, or policies that limit rent increases for certain rental
units, are sometimes offered as a means of addressing high housing costs.
Though existing tenants in rent-controlled units may benefit from smaller
rent increases, supply is reduced for new potential tenants and the incentive
for developers to build more units is diminished. There are few issues where
economists are in as much as agreement as they are regarding the outcomes
of rent control. In a 2012 University of Chicago Booth poll of economists
across the political spectrum, 95 percent disagreed that rent control ordinances, such as those imposed in New York and San Francisco, had boosted
affordable housing or improved the quality of rental units (IGM 2012).
The economists’ consensus is supported not only by economic theory
but also by the empirical literature. In a recent paper examining the effect
of a 1994 rent control law on housing supply and prices in San Francisco,
Diamond, McQuade, and Qian (2019) find that the law had the opposite of its
intended effect on rents. While those living in rent-controlled units benefit
from lower rents, and remain in these units longer than they would without
rent control, those who do not have access to these units are substantially
harmed in the long run. Landlords responded to the law by converting existing buildings into condominiums and by taking other steps to avoid being
subject to rent control laws. This lowered the supply of rental housing by 15
percent and incentivized the creation of housing that served the preferences
of high-income households. As a result, this rent control law likely raised rents
in the long run rather than lowering them. Moreover, even existing tenants
who benefit from rent control may suffer from unintended consequences.
Jiang (2019) finds that rent control increases unemployment among tenants in New York City, potentially because they can sustain longer bouts of
joblessness given their lower housing costs, or because tenants are tied to a
particular housing unit and restrict their job search to opportunities nearby.
Expansions of government housing programs to combat rising rents are
also unlikely to provide much relief to the general population of residents in
supply-constrained areas. When the supply of housing is inelastic, expanding
demand by increasing government subsidies increases prices rather than
quantities. As a result, government rental subsidies to low income-renters
will likely increase rents in markets with overly restrictive housing regulations.
Eriksen and Ross (2015) find that housing vouchers increased rents for housing within 20 percent of the Fair Market Rent threshold in supply-constrained
communities. They estimate that a 10 percent increase in the number of

290 |

250-840_text_.pdf 294

Chapter 8

2/7/20 3:46 PM

vouchers increased rents by 0.39 percent for these units. LIHTC, a program
that subsidizes developers of below market-rate rental housing units, may
also be ineffective at addressing the underlying supply problem according to
some evidence. Eriksen and Rosenthal (2010) find that new LIHTC development largely crowds out private development, leaving total housing supply
unchanged. Glaeser and Gyourko (2008) note that the credit tends to increase
the profits of subsidized builders, while pushing unsubsidized builders out of
the housing market.
Regulations that require a certain share of housing units to be set aside
for low-income residents, often referred to as “inclusionary zoning,” also
fail to solve the affordable housing problem. For example, Schuetz, Meltzer,
and Been (2011) find that inclusionary zoning can increase home prices and
in some cases reduce housing development. Hamilton (2019), in a study of
Washington and Baltimore, similarly finds that inclusionary zoning increases
prices.

they would if they lived in the suburbs. As discussed in box 8-2, regulatory
reform—rather than rent control, expansion of government programs, or inclusionary zoning—offers the most effective solution to the problems posed by
high housing costs and overregulation.

Conclusion
How to increase housing affordability through regulatory reform is an issue
that has garnered bipartisan attention in recent years. In this chapter, we have
focused on excessive regulations that substantially drive up home prices in
a selected number of metropolitan areas. Relaxing these regulations would
greatly benefit Americans, especially those with lower incomes, by reducing
the cost of attaining homeownership and reducing rents in supply-constrained
areas. Falling rents resulting from relaxing excessive regulations would reduce
homelessness by 31 percent on average in these areas, and more families
could be served by Federal rental housing assistance programs. Broader benefits would include increased economic growth, reduced regional disparities,
expanded opportunities for children, and a cleaner environment.
We have also emphasized that addressing the problem of overregulation
with more regulation would be counterproductive. Rent control can increase
housing prices by reducing the incentive for developers to build new housing.
Similarly, expanded government subsidies for housing do not solve the problem of overregulation. When housing supply is constrained, housing subsidies
for tenants may increase market rents without increasing the quantity of housing, counteracting the goals of these programs.

Expanding Affordable Housing | 291

250-840_text_.pdf 295

2/7/20 3:46 PM

The Trump Administration has taken steps to address onerous housing
regulations. President Trump issued an Executive Order in 2019 to establish the
White House Council on Eliminating Regulatory Barriers to Affordable Housing,
which is tasked with reviewing housing regulations at all levels of government
and submitting a report to the President in 2020 with recommendations on
how to ameliorate these excessive regulatory burdens.
HUD has also taken action under the Trump Administration to counter
regulatory barriers to building affordable housing. The Affirmatively Furthering
Fair Housing rule, which was finalized during the previous Administration,
is being revised to focus more clearly on increasing housing supply in areas
where supply is constrained. This rule recognizes that increasing housing
choice for disadvantaged groups requires taking on regulatory barriers that
place housing in large swaths of specific areas out of reach for lower-income
families.

292 |

250-840_text_.pdf 296

Chapter 8

2/7/20 3:46 PM

x
Part III

The Economic Outlook

293

250-840_text_.pdf 297

2/7/20 3:46 PM

250-840_text_.pdf 298

2/7/20 3:46 PM

x
Chapter 9

The Outlook for a
Continued Expansion
As this Report has shown, under the Trump Administration, economic growth
and the labor market gains it enables have exceeded pre-2017 expectations.
The U.S. economy’s performance has withstood strong headwinds from a weak
global economy and several idiosyncratic domestic shocks, as pro-growth policies have kept the U.S. economy resilient.
By increasing competition, productivity, and wages, and reducing the prices
of consumer goods and services, the Administration’s approach to regulation
is raising real incomes while maintaining regulatory protections for workers,
public health, safety, and the environment. Specifically, the Administration’s
approach to eliminating excessive regulation of energy markets supports
further unleashing of the country’s abundant human and energy resources.
Furthermore, the Administration’s healthcare reforms are building a system
that delivers high-quality care at affordable prices through greater choice and
competition. Across the board, this pro-growth agenda has disproportionately
benefited those previously left behind during the current expansion.
To further expand the economy and extend the longest expansion in U.S. history, additional policy issues may need to be addressed. This challenge is why
the Trump Administration remains focused on promoting competitive markets,
combating the opioid crisis, promoting affordable housing, enacting a comprehensive infrastructure plan, rendering the individual provisions of the 2017 Tax
Cuts and Jobs Act permanent, updating the U.S. immigration system, continuing deregulatory actions, improving trade agreements and international trade

295

250-840_text_.pdf 299

2/7/20 3:46 PM

practices, and incentivizing higher labor force participation through additional
labor market reforms.
Overall, assuming full implementation of the Trump Administration’s economic
policy agenda, we project real U.S. economic output to grow at an average
annual rate of 2.9 percent over the budget window from 2019 to 2030. During
that time, inflation is expected to settle at a 2.0 percent fourth-quarter-overfourth-quarter rate, and the unemployment rate is expected to remain at or
below an annual average rate of 4.0 percent. Relative to the current-law baseline
projection, we estimate that full policy implementation of the Administration’s
economic agenda would cumulatively raise output by 4.3 percent over this
budget window.
The first three years of the Trump Administration show that long-lamented
structural trends that were constraining potential growth in the United States
are not policy-invariant. The right pro-growth policies attract greater investment, encourage more people to enter the labor market, and lead to higher
wages from businesses investing in and competing for workers. Even with
recent success, there is ample room for the U.S. economy to expand, especially
if the Administration’s approach to international trade produces results that are
greater than expected.

S

ince 1975, the Council of Economic Advisers, in collaboration with
the Office of Management and Budget and the U.S. Department of
the Treasury, has published a long-run forecast for the U.S. economy
that assumes full enactment and implementation of the Administration’s economic policy agenda. This reflects the Council’s mandate, as stipulated in the
Employment Act of 1946, to set forth in the Economic Report of the President
“current and foreseeable trends in the levels of employment, production, and
purchasing power,” and a program for carrying out the objective of “creating and maintaining . . . conditions under which there will be afforded useful
employment opportunities, including self-employment, for those able, willing,
and seeking to work, and to promote maximum employment, production, and
purchasing power.” Since 1996, execution of this mandate has involved providing an 11-year, policy-inclusive economic forecast.

296 |

250-840_text_.pdf 300

Chapter 9

2/7/20 3:46 PM

Because of this charge, the Administration’s forecast is historically
unique from other long-run economic forecasts, both official and private. The
Congressional Budget Office, for example, publishes a current-law forecast,
which assumes no change in economic policy (CBO 2019). The Blue Chip
panel of professional private sector forecasters often reveals substantial heterogeneity in expectations, reflecting both different estimates of economic
potential under current law, as well as objective and subjective estimations of
the probability of policy implementation. Although the assumptions underlying projections of the Federal Open Market Committee are ambiguous, those
forecasts presumably also reflect committee members’ differing views both on
potential growth under current law, as well as potential growth under possible
future law.
To better distinguish the estimated effects of the Administration’s economic policy objectives—the results of which may be contingent on legislative
support and other factors—from current-law projections, beginning with the
2018 Economic Report of the President and continuing through this Report, we
have decomposed this forecast into a current-law baseline and intermediate
and top lines that reflect estimated growth effects discussed in this Report,
as well as in the 2018 and 2019 Reports and the President’s Fiscal Year 2021
Budget. We then build up to our top-line, policy-inclusive forecast by successively adding to the current-law baseline the estimated effects of future
deregulatory actions, immigration reform, additional labor market reforms to
incentivize higher labor force participation, rendering the individual provisions
of the Tax Cuts and Jobs Act (TCJA) permanent, additional fiscal policy proposals, including the Administration’s infrastructure plan, and improved trade
deals with international trading partners. The top-line forecast constitutes the
Administration’s official “Troika” forecast of the Council of Economic Advisers,
Office of Management and Budget, and Department of Treasury. For comparison, we also report a pre-policy baseline consisting of the Congressional
Budget Office’s 2019–27 projection made in August 2016, extended by its
August 2019 current-law projection.

GDP Growth during the Next Three Years
As illustrated in figure 9-1 and reported in the third column (“real GDP”) of
table 9-1, the Administration anticipates economic growth to rise in 2020 from
its projected 2019 pace of 2.5 percent, and to remain at or above 3.0 percent
through 2022, assuming full implementation of the economic agenda detailed
in this Report, its two predecessors, and the President’s Fiscal Year 2021
Budget. We expect near-term growth to be supported by the continuing effects
of the TCJA, as well as new measures to promote increased labor force participation, deregulatory actions, immigration reform, reciprocal trade deals,

The Outlook for a Continued Expansion

250-840_text_.pdf 301

| 297

2/7/20 3:46 PM

$"0- 
шҊрѵ
*- ./
*!
-*2/#
/ 
$)
 '
Ѷ
спрш–тп
0".
спрх

f*- ./Ѷ
спрш–сх
0".
спрш

f*- ./Ѷ
спсц–сш
0-- )/-l2
f*- ./
‫* ۔‬-
m-& /
)
d - "0'/$*)
p*'$$ .
‫ ۔‬$.'
)
t- 
po'$$ .
Q4-over-Q4 change (percent)

тѵф
тѵп
сѵф
сѵп
рѵф
рѵп
пѵф
пѵп

спрш спсп спср спсс спст спсу спсф спсх спсц спсч спсш сптп
*0- .ѷ
0- 0
*!
*)*($
)'4.$.Ѹ 0- 0
*!
 *-
//$./$.Ѹ
*)"- ..$*)'

0" /
!!$ Ѹ
 +-/( )/
*!
/# 
- .0-4Ѹ
!!$ 
*!
)" ( )/
)
0" /Ѹ
EA
calculationsѵ
*/ ѷ
#e
current-law !*- ./
i.
. 
*)
/
1$'' 
.
*!
/* -
трѶ
спршѵ

and an infrastructure program, which we assume will commence in 2020 with
observable effects on output beginning in 2021.
The Administration also expects the labor market to continue to exhibit
strength in the near term, with the civilian unemployment rate remaining
below 4.0 percent through 2022, as reported in the sixth column, “unemployment rate,” of table 9-1. Despite low unemployment, inflation is expected to
remain low and close to the Federal Reserve Board’s 2.0 percent target for the
Personal Consumption Expenditures Price Index. The Administration expects
broad inflation beyond 2019 to remain stable at 2.0 percent through 2022, as
shown in the fourth column (“GDP price index”) of table 9-1.

GDP Growth over the Longer Term
As discussed in the 2018 and 2019 volumes of the Economic Report of the
President, over the longer term, the Administration’s current-law baseline
forecast is for output growth to moderate as the capital-to-output ratio asymptotically approaches a higher steady state level in response to corporate tax
reform, and as the near-term effects of the TCJA’s individual provisions on the
rate of growth dissipate into a permanent level effect. As reflected by our intermediate forecast, we expect the latter moderation would be partially offset in

298 | Chapter 9

250-840_text_.pdf 302

2/7/20 3:46 PM

' 
шҊрѵ
# 
($)$./-/$*)’s
*)*($
*- ./Ѷ
спрч–тп

 - )/
#)" 
җуҊ/*ҊуҘ


 -

*($)'



1 '
җ' )-
4 -
1 -" Ҙ


 '


 
+-$ 
$) 3

*).0( -
+-$ 

җ#$)Ҋ/4+ Ҙ

җ#$)Ҋ/4+ Ҙ

$) 3


)/ - ./
-/ Ѷ

)/ - ./
-/ Ѷ

) (+'*4( )/

рпҊ4 шрҊ4
- .0-4

- .0-4
)*/ .

-/ 
җ+ - )/Ҙ

$''.
җ+ - )/Ҙ
җ+ - )/Ҙ


спрч

җ/0'Ҙ


уѵш


сѵф


сѵт


сѵс


тѵш


рѵш


сѵш


спрш


уѵс


сѵф


рѵч


рѵш


тѵц


сѵр


сѵс


спсп


фѵс


тѵр


сѵп


сѵт


тѵф


рѵу


сѵп


спср


фѵр


тѵп


сѵп


сѵт


тѵх


рѵф


сѵс


спсс


фѵр


тѵп


сѵп


сѵт


тѵч


рѵф


сѵф


спст


фѵр


тѵп


сѵп


сѵт


уѵп


рѵх


сѵц


спсу


фѵр


тѵп


сѵп


сѵт


уѵп


рѵц


тѵп


спсф


фѵп


сѵш


сѵп


сѵт


уѵп


сѵп


тѵр


спсх


уѵш


сѵч


сѵп


сѵт


уѵп


сѵс


тѵр


спсц


уѵш


сѵч


сѵп


сѵт


уѵп


сѵу


тѵр


спсч


уѵш


сѵч


сѵп


сѵт


уѵп


сѵф


тѵр


спсш


уѵш


сѵч


сѵп


сѵт


уѵп


сѵф


тѵс


сптп


уѵш


сѵч


сѵп


сѵт


уѵп


сѵф


тѵс


*0- .ѷ
0- 0
*!
*)*($
)'4.$.Ѹ
0- 0
*!
 *-
//$./$.Ѹ
 +-/( )/
*!
/# 
- .0-4Ѹ
!!$ 
*!
)" ( )/
)
0" /Ѹ
EA calculations.
*/ ѷ
#$.
!*- ./
2.
. 
*)
/
1$'' 
.
*!
/* -
трѶ
спршѵ
# 
$)/ - ./
-/ 
*)
шрҊ4
Ҋ$''.
$.
( .0- 
*)

. *)-4Ҋ(-& /
$.*0)/
.$.ѵ

*($)'

)
/# 
.0(
*!
- '

)
/# 

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


2026 and 2027 if the individual provisions of the TCJA—currently legislated to
expire on December 31, 2025—were instead made permanent.
The Administration’s full policy-inclusive forecast is reported as the green
line in figure 9-1. In addition to successful implementation of the President’s
infrastructure plan and extension of the individual provisions of the TCJA, this
forecast assumes full achievement of the Administration’s agenda with respect
to deregulation, immigration, improved trade agreements, fiscal consolidation, and labor market policies designed to incentivize higher labor force participation. The latter includes expanding work requirements for nondisabled,
working-age welfare 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.
Though we anticipate growth moderating toward the end of the budget
window, to 2.8 percent on average between 2019 and 2030, the policy-inclusive
forecast is for output to grow at an average annual rate of 2.9 percent. Relative
to the current-law baseline, we estimate that full policy implementation would
cumulatively raise the level of output by 4.3 percent over the budget window.
Reflecting moderating growth in the latter half of the budget window, the
Administration expects unemployment to converge to 4.0 percent, consistent

The Outlook for a Continued Expansion

250-840_text_.pdf 303

| 299

2/7/20 3:46 PM

' 
шҊсѵ
0++'4Ҋ$ 
*(+*) )/.
*!
/0'
)
*/ )/$'
 '

0/+0/
-*2/#Ѷ
ршфт–сптп

-*2/#
-/ 
җ+ - )/" 

+*$)/.Ҙ

*(+*) )/


р

с


$1$'$)
)*)$)./$/0/$*)'
+*+0'/$*)
" 
рх‫
۔‬
*-
!*- 
+-/$$+/$*)
-/ 


ршфтѷс
/*

спршѷт


спрш
/*

сптп


рѵу


пѵш


пѵр


–пѵс


т


(+'*4 
.#- 
*!
/# 
'*-
!*- 



пѵп



пѵп


у


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

#*0. #*'
 (+'*4( )/


пѵп



пѵп


ф


1 -" 
2 &'4
#*0-.
җ)*)!-(
0.$) ..Ҙ


–пѵс


пѵр


х


0/+0/
+ -
#*0-
җ+-*0/$1$/4Ѷ
)*)!-(
0.$) ..Ҙ


сѵп


сѵх


ц


/$*
*!
- '

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


–пѵт


–пѵф


ч


0(ѷ
/0'
- '



тѵп


тѵп


 (*ѷ


ш




*/ )/$'
- '



тѵп



тѵп


рп




0/+0/
+ -
2*-& -
$!! - )/$'ѷ

1.ѵ
)*)!-(


–пѵт


–пѵу



 '

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

+-*0/Ҋ.$ 
( .0- .ѵ

*0- .ѷ
0- 0
*!
 *-
//$./$.Ѹ
0- 0
*!
*)*($
)'4.$.Ѹ
 +-/( )/
*!
/# 
- .0-4Ѹ

!!$ 
*!
)" ( )/
)
0" /Ѹ

'0'/$*).ѵ

*/ ѷ
''
*)/-$0/$*).
- 
$)
+ - )/" 
+*$)/.
/
)
))0'
-/ Ѷ
!*- ./
!$)'$5 
*1 ( -
рѶ

спршѵ
*/'
(4
)*/

0+
0 
/*
-*0)$)"ѵ
The quarter ршфтѷс
2.

0.$) ..Ҋ4' 
+ &;
спршѷт
$.
/# 

'/ ./
,0-/ -
2$/#
1$'' 
/ѵ
-*..
*( ./$
*0/+0/
җҘ
$.
/# 
1 -" 
*!

)
"-*..

*( ./$
$)*( ѵ
*+0'/$*)Ѷ
'*-
!*- Ѷ
)
#*0. #*'
 (+'*4( )/
#1 
 )
%0./ 
!*-

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




with the Federal Open Market Committee’s December 2019 “Summary of
Economic Projections,” which reports a range of participant estimates from
3.9 to 4.3 percent (Federal Reserve 2019). The unemployment rate rising to
4.0 percent is also expected to maintain a rate of inflation of 2.0 percent, as
measured by the GDP chained price index (see the fourth column of table 9-1).
As shown in table 9-2, the Administration anticipates that the primary
contributor to increased growth through 2029 will be higher output per hour
worked. During much of the current expansion, U.S. labor productivity growth
was disappointing by historical standards, partly due to low contributions of
capital deepening. By substantially raising the capital stock and consequent
flows of capital services, attracting increased net capital inflows—including
investment both by foreign firms and overseas affiliates of U.S. multinational
enterprises—and facilitating efficient capital reallocation from mature firms
to more dynamic enterprises, we expect enactment of corporate tax reform to
considerably increase capital per worker, and thus labor productivity. Already,
during the first seven quarters since the TCJA was enacted, labor productivity growth in the nonfarm business sector rose substantially relative to its
300 | Chapter 9

250-840_text_.pdf 304

2/7/20 3:46 PM

pre-TCJA, postrecession average, as reported in chapter 1 of this Report. If fully
implemented, we also expect the Administration’s labor market reforms to
partially offset the effects of demographic-related trends in labor force participation, as reflected in line 2 of table 9-2.

Upside and Downside Forecast Risks
Since the Administration’s forecast is a policy-inclusive one, a key downside
risk is the political contingency of full implementation of the President’s
economic agenda, particularly in light of the inherent unpredictability of
the legislative process. In addition, by definition the policy-inclusive forecast assumes that the Administration’s policies will be implemented and
remain in place throughout the forecast window. In scenarios where future
Administrations or Congress partially or fully reverse the TCJA, otherwise raise
taxes, or significantly expand the Federal regulatory state, economic growth
would be lower or even negative. For example, the 2019 Economic Report of
the President estimated that “Medicare for All” bills then discussed in Congress
would reduce real GDP by about 9 percent in the long run if financed by taxes
on labor income, while recent proposals to introduce a top marginal income
tax rate of 70 percent on personal income over $10 million would lower the
long-run level of GDP by 0.2 percent.
As observed in the 2019 Report and discussed in chapter 1 of this Report,
a sharp slowdown in the global economy also poses a significant downside
risk to the outlook, through both direct and indirect channels. In particular,
continued or worsening weakness in other advanced economies—particularly
Germany and Italy, but also Europe more broadly, in the event of Brexit-related
disruptions—would have an adverse impact on U.S. growth through both a
direct export channel and indirect exchange rate, financial market, and supply chain channels. A significant growth slowdown in the People’s Republic
of China, similar to that observed in the years 2015–16, would also introduce
substantial risks to the outlooks for advanced economies, including the United
States. High public debt levels in several advanced and emerging economies
may generate economic headwinds, while high corporate debt levels in the
United States could act as an accelerant to potential adverse financial shocks.
Idiosyncratic shocks also pose risks to the outlook. In 2019, these
included but were not limited to production cuts at Boeing—whose production accounts for 0.23 percent of U.S. GDP—a partial government shutdown
in the first quarter, and industrial action at General Motors. As this Report was
being finalized, Boeing announced plans to halt production of the 737 MAX, a
development that could subtract 0.5 percent from annualized real GDP growth
in the first quarter of 2020.
Perhaps the single biggest upside risk to the outlook is that the
Administration’s more robust approach to international trade achieves

The Outlook for a Continued Expansion

250-840_text_.pdf 305

| 301

2/7/20 3:46 PM

greater-than-expected success in its pursuit of freer, fairer trade, with zero
tariffs, zero nontariff barriers, and zero subsidies. Recent research by the
Organization for Economic Cooperation and Development (Cadot, Gourdon,
and van Tongeren 2018; Lamprecht and Miroudot 2018; OECD 2018) finds that
lowering international tariff and nontariff barriers to trade, as well as reducing
international restrictiveness on trade in services, would substantially raise U.S.
and global trade and output. With investment in intellectual property products
now accounting for about one-third of U.S. private nonresidential fixed investment, trade agreements that enhance international protection of intellectual
property—such as the United States–Mexico–Canada Agreement and Phase
I of U.S.-China negotiations—could also elevate the level of innovation and
productivity growth.
Additional upside risks to the forecast include, first, higher net capital
inflows due to international capital mobility exceeding estimates, which would
attenuate the potential crowding out of private fixed investment in response to
individual tax reform and public infrastructure investment. Second, academic
studies demonstrating that individual marginal income tax rates may have
differential effects across the age distribution suggest that estimated trends
in labor force participation may overstate the growth-detracting effect of
demography. Third, insofar as the growth estimates presented in this Report
and its predecessor have been derived from standard neoclassical growth
models, they may omit the positive externalities and spillover effects captured
by endogenous growth models, such as that of Ehrlich, Li, and Liu (2017). Tax
reform that incentivizes investment in human capital, regulatory reform that
eliminates prohibitive barriers to entry for more innovative and entrepreneurial firms, and health investments and labor market policies that facilitate
human capital accumulation may, therefore, yield higher-growth dividends
than those estimated here.

302 | Chapter 9

250-840_text_.pdf 306

2/7/20 3:46 PM

x

References
Chapter 1
Amiti, M., S. Redding, and D. Weinstein. 2019a. “The Impact of the 2018 Trade War on
U.S. Prices and Welfare.” CEP Discussion Paper dp1603.
———. 2019b. “New China Tariffs Increase Cost to U.S. Households.” Liberty Street
Economics. https://libertystreeteconomics.newyorkfed.org/2019/05/
new-china-tariffs-increase-costs-to-us-households.html.
BIS (Bank for International Settlements). 2019a. “Annual Economic Report.” https://
www.bis.org/publ/arpdf/ar2019e.htm.
———. 2019b. “BIS Quarterly Review: International Banking and Financial Market
Developments.” https://www.bis.org/publ/qtrpdf/r_qt1909.htm.
Caldara, D., M. Iacoviello, P. Molligo, A. Prestipino, and A. Raffo. 2019. “Does Trade
Policy Uncertainty Affect Global Economic Activity?” FEDS Notes. Washington:
Board of Governors of the Federal Reserve System.
CEA (Council of Economic Advisers). 2017. Economic Report of the President.
Washington: U.S. Government Publishing Office.
———. 2018. Economic Report of the President. Washington: U.S. Government
Publishing Office.
———. 2019a. “The Economic Effects of Federal Deregulation since January 2017: An
Interim Report.” https://www.whitehouse.gov/wp-content/uploads/2019/06/
The-Economic-Effects-of-Federal-Deregulation-Interim-Report.pdf.
———. 2019b. Economic Report of the President. Washington: U.S. Government
Publishing Office.
Collins, C., and E. Truman. 2019. “The U.S. Dollar’s Strength Is Not at Historical Highs.”
Peterson Institute for International Economics. https://www.piie.com/
research/piie-charts/us-dollars-strength-not-historical-highs.
Curcuru, S. 2017. “The Sensitivity of the U.S. Dollar Exchange Rate to Changes in
Monetary Policy Expectations.” IFDP Notes. Washington: Board of Governors
of the Federal Reserve System. https://doi.org/10.17016/2573-2129.36.
Deutsche Bundesbank. 2019. “Monthly Report: October 2019.” https://www.
bundesbank.de/resource/blob/811960/
a0bf7575c07b3754dda37ec6739d4b13/mL/2019-10-monatsbericht-data.pdf.

303

250-840_text_.pdf 307

2/7/20 3:46 PM

Diebold, F., and G. Rudebusch. 1990. “A Nonparametric Investigation of Duration
Dependence in the American Business Cycle.” Journal of Political Economy 98,
no. 3.
Fajgelbaum, P., P. Goldberg, P. Kennedy, and A. Khandelwal. 2019. The Return to
Protectionism. NBER Working Paper 25638. Cambridge, MA: National Bureau
of Economic Research.
Green, G., and J. Coder. 2019. “Household Income Trends.” Sentier Research LLC.
Hansen, A. 1939. “Economic Progress and Declining Population Growth.” American
Economic Review 29, no. 1: 1–15.
Hooper, P., F. Mishkin, and A. Sufi, 2019. Prospect for Inflation in a High Pressure
Economy: Is the Phillips Curve Dead or Is It Just Hibernating? NBER Working
Paper 25792. Cambridge, MA: National Bureau of Economic Research.
IMF (International Monetary Fund). 2018. World Economic Outlook: Challenges to Steady
Growth. Washington: IMF.
———. 2019a. “People’s Republic of China. 2019 Article IV Consultation—Press Release;
Staff Report; Staff Statement; and Statement by the Executive Director for
China.” IMF Country Report 19/266. Washington: IMF.
———. 2019b. “Germany: 2019 Article IV Consultation—Press Release; Staff Report; and
Statement by the Executive Director for Germany.” IMF Country Report
19/213. Washington: IMF.
———. 2019c. World Economic Outlook: Global Manufacturing Downturn, Rising Trade
Barriers. Washington: IMF.
———. 2019d. “United States: 2019 Article IV Consultation—Press Release; Staff Report;
and Statement by the Executive Director for the United States.” IMF Country
Report 19/174. Washington: IMF.
Kopp, E., D. Leigh, S. Mursula, and S. Tambunlertchai. 2019. U.S. Investment since the
Tax Cuts and Jobs Act of 2017. IMF Working Paper 19/120. Washington: IMF.
Lettau, M., and S. Ludvigson. 2003. Understanding Trend and Cycle in Asset Values:
Reevaluating the Wealth Effect on Consumption. NBER Working Paper 9848.
Cambridge, MA: National Bureau of Economic Research.
Londono, J., S. Ma, and B. Wilson. 2019. “Quantifying the Impact of Foreign Economic
Uncertainty on the U.S. Economy.” FEDS Notes. Washington: Board of
Governors of the Federal Reserve System. https://doi.
org/10.17016/2380-7172.2463.
Mertens, K. 2018. “The Near Term Growth Impact of the Tax Cuts and Jobs Act.” Federal
Reserve Bank of Dallas Working Paper 1803. https://doi.org/10.24149/
wp1803.
OECD (Organization for Economic Cooperation and Development). 2019a. OECD
Economic Outlook, Volume 2019, Issue 2. Paris: OECD Publishing.
———. 2019b. “Trade Simulations in the Car Sector.” Presentation given at STEP
Meeting, October 22.

304 | References

250-840_text_.pdf 308

2/7/20 3:46 PM

Poterba, J. 2000. “Stock Market Wealth and Consumption.” Journal of Economic
Perspectives 14, no. 2.
Rachel, L., and L. Summers. 2019. “On Falling Neutral Real Rates, Fiscal Policy, and the
Risk of Secular Stagnation.” Brookings Papers on Economic Activity,
conference draft, spring.
Rubbo, E. 2020. “Networks, Phillips Curves, and Monetary Policy.” Working paper.
https://scholar.harvard.edu/files/elisarubbo/files/rubbo_jmp.pdf.
Rudebusch, G. 2016. “Will the Economic Recovery Die of Old Age?” Economic Letter.
San Francisco: Federal Reserve Bank of San Francisco.
Summers, L. 2013. “IMF Fourteenth Annual Research Conference in Honor of Stanley
Fischer.” http://larrysummers.com/
imf-fourteenth-annual-research-conference-in-honor-of-stanley-fischer/.
———. 2014. “U.S. Economic Prospects: Secular Stagnation, Hysteresis, and the Zero
Lower Bound.” Business Economics 49, no. 2: 65–73.
———. 2016. “The Age of Secular Stagnation.” Foreign Affairs, February 17. http://
larrysummers.com/2016/02/17/the-age-of-secular-stagnation/.
White House. 2018. “Statement from the Press Secretary Regarding the President’s
Working Dinner with China” https://www.whitehouse.gov/briefingsstatements/
statement-press-secretary-regarding-presidents-working-dinner-china/.
World Bank. 2019a. Global Economic Prospects: Heightened Tensions, Subdued
Investment. Washington: World Bank. https://elibrary.worldbank.org/doi/
abs/10.1596/978-1-4648-1398-6.
———. 2019b. Global Monthly: October 2019. Washington: World Bank.
WTO (World Trade Organization). 2016. World Trade Report 2016: Levelling the Trading
Field for SMEs. Geneva: WTO Publishing.

Chapter 2
Aizer, A. 2007. “Public Health Insurance, Program Take-Up, and Child Health.” Review of
Economics and Statistics 89, no. 3: 400–415.
Barnett, J., and E. Berchick. 2017. “Health Insurance Coverage in the United States:
2016.” U.S. Census Bureau. https://www.census.gov/library/
publications/2017/demo/p60-260.html.
Berchick, E., E. Hood, and J. Barnett. 2018. “Health Insurance Coverage in the United
States: 2017.” U.S. Census Bureau. https://www.census.gov/library/
publications/2018/demo/p60-264.html.
Bernstein, J., and K. Hassett. 2015. Unlocking Private Capital to Facilitate Economic
Growth in Distressed Areas. Washington: Economic Innovation Group.
Blau, F., and L. Kahn. 2013. Female Labor Supply: Why is the U.S. Falling Behind? NBER
Working Paper 18702. Cambridge, MA: National Bureau of Economic
Research.

References | 305

250-840_text_.pdf 309

2/7/20 3:46 PM

Burkhauser, R., K. Corinth, J. Elwell, and J. Larrimore. 2019. “Evaluating the Success of
President Johnson’s War on Poverty: Revisiting the Historical Record Using a
Full-Income Poverty Measure.” Working paper. https://papers.ssrn.com/sol3/
papers.cfm?abstract_id=3353906.
Canberra Group. 2011. Handbook on Household on Household Income Statistics, 2nd ed.
Geneva: United Nations Economic Commission for Europe.
Casey, A. 2019. “Sale Prices Surge in Neighborhoods with New Tax Break.” Zillow.
https://www.zillow.com/research/prices-surge-opportunity-zones-23393/.
CBO (U.S. Congressional Budget Office). 2016. An Update to the Budget and Economic
Outlook: 2016 to 2026. https://www.cbo.gov/sites/default/files/
recurringdata/51135-2016-08-economicprojections-2.xlsx.
———. 2019. “Federal Subsidies for Health Insurance Coverage for People Under Age 65:
2019 to 2029.” https://www.cbo.gov/system/files/2019-05/55085HealthCoverageSubsidies_0.pdf.
CDFI Fund (U.S. Department of the Treasury, Community Development Financial
Institutions Fund). 2017. “Compliance Review of New Markets Tax Credit
Program.” https://www.cdfifund.gov/Documents/Summit%20-%20
Compliance%20Review%20of%20New%20Markets%20Tax%20Credit%20
Program%20-%20August%20Date%20-%20508%20Compliant.pdf.
———. 2018. “Opportunity Zones Resources.” https://www.cdfifund.gov/Pages/
Opportunity-Zones.aspx.
———. 2019. “New Markets Tax Credit Program Compliance Monitoring: Frequently
Asked Questions.” https://www.cdfifund.gov/programs-training/Programs/
new-markets-tax-credit/Pages/compliance-step.aspx#step5.
CEA (Council of Economic Advisers). 2018a. “Addressing America’s Reskilling
Challenge.” https://www.whitehouse.gov/wp-content/uploads/2018/07/
Addressing-Americas-Reskilling-Challenge.pdf.
———. 2018b. “Deregulating Health Insurance Markets: Value to Market Participants.”
https://www.whitehouse.gov/wp-content/uploads/2019/02/DeregulatingHealth-Insurance-Markets-FINAL.pdf.
———. 2018c. Economic Report of the President. Washington: U.S. Government
Publishing Office.
———. 2019a. Economic Report of the President. Washington, U.S. Government
Publishing Office.
———. 2019b. “President Trump’s Policies Continue to Benefit All Americans, Especially
the Disadvantaged.” https://www.whitehouse.gov/articles/
president-trumps-policies-continue-benefit-americans-especiallydisadvantaged/.
———. 2019c. “Economic Effects of Deregulation.” https://www.whitehouse.gov/
wp-content/uploads/2019/06/The-Economic-Effects-of-Federal-DeregulationInterim-Report.pdf.

306 | References

250-840_text_.pdf 310

2/7/20 3:46 PM

———. 2019d. “Government Employment and Training Programs: Assessing the
Evidence on Their Performance.” https://www.whitehouse.gov/wp-content/
uploads/2019/06/Government-Employment-and-Training-Programs.pdf.
———. 2019e. “The Role of Affordable Child Care in Promoting Work Outside the
Home.” https://www.whitehouse.gov/wp-content/uploads/2019/12/The-Roleof-Affordable-Child-Care-in-Promoting-Work-Outside-the-Home-1.pdf.
———. 2019f. “The Full Cost of the Opioid Crisis: $2.5 Trillion Over Four Years.” https://
www.whitehouse.gov/articles/full-cost-opioid-crisis-2-5-trillion-four-years/.
Crespo-Sancho, C. 2018. “Can Gender Equality Prevent Violent Conflict?” The World
Bank Blogs, March 28. http://blogs.worldbank.org/dev4peace/
can-gender-equality-prevent-violent-conflict.
CRS (Congressional Research Service). 2019. “New Market Tax Credit: An Introduction.”
https://fas.org/sgp/crs/misc/RL34402.pdf.
Dev Bhardwaj, R. 1992. “China’s Economic Reform: The Role and Significance of SEZs.”
Indian Journal of Political Science 53, no. 3: 332–73.
Elwell, J., K. Corinth, and R. Burkhauser. 2019. Income Growth and Its Distribution from
Eisenhower to Obama: The Growing Importance of In-Kind Transfers (1959–
2016). NBER Working Paper 26439. Cambridge, MA: National Bureau of
Economic Research.
Fein, D., and J. Hamadyk. 2018. Bridging the Opportunity Divide for Low-Income Youth:
Implementation and Early Impacts of the Year Up Program. OPRE Report 201865. Washington: Office of Planning, Research, and Evaluation, Administration
for Children and Families, U.S. Department of Health and Human Services.
https://www.yearup.org/wp-content/uploads/2018/06/Year-Up-PACE-FullReport-2018.pdf.
Ferrant, G., and A. Thim. 2019. Measuring Women’s Economic Empowerment: Time Use
Data and Gender Inequality. OECD Development Policy Paper 16. Paris: OECD
Publishing. http://www.oecd.org/dev/development-gender/MEASURINGWOMENS-ECONOMIC-EMPOWERMENT-Gender-Policy-Paper-No-16.pdf.
FOMC (Federal Open Market Committee). 2016. “Chair FOMC’s Press Conference
Projections Materials, September 21, 2016.” Press release, September 21.
https://www.federalreserve.gov/monetarypolicy/files/fomcprojtabl20160921.
pdf.
Food and Nutrition Service. 2019a. “Food Security in the U.S.: Key Statistics &
Graphics.” U.S. Department of Agriculture. https://www.ers.usda.gov/topics/
food-nutrition-assistance/food-security-in-the-us/key-statistics-graphics/.
———. 2019b. “SNAP Data Tables.” U.S. Department of Agriculture. https://www.fns.
usda.gov/pd/supplemental-nutrition-assistance-program-snap.
Fry, R. 2018. “Millennials Projected to Overtake Baby Boomers as America’s Largest
Generation.” Pew Research Service. https://www.pewresearch.org/facttank/2018/03/01/millennials-overtake-baby-boomers/.

References | 307

250-840_text_.pdf 311

2/7/20 3:46 PM

FTC (Federal Trade Commission). 2018. “Policy Perspectives: Options to Enhance
Occupational License Portability.” https://www.ftc.gov/reports/
options-enhance-occupational-license-portability.
Ganong, P., and D. Shoag. 2017. “Why Has Regional Income Convergence in the U.S.
Declined?” Journal of Urban Economics 102: 76–90.
Goodnough, A., and M. Sanger-Katz. 2019. “Medicaid Covers a Million Fewer Children.
Baby Elijah Was One of Them.” New York Times, Upshot Blog, October 22.
https://www.nytimes.com/2019/10/22/upshot/medicaid-uninsured-children.
html.
Guo, J., A. Seshadri, and C. Taber. 2019. Jobs, Skills, and the Prison-to-Work Transition.
Madison: Center for Research on the Wisconsin Economy.
Hendra, R., D. Greenberg, G. Hamilton, A. Oppenheim, A. Pennington, K. Schaberg, and
B. Tessler. 2016. “Encouraging Evidence on a Sector-Focused Advancement
Strategy.” MDRC. https://www.mdrc.org/sites/default/files/2016_
Workadvance_Final_Web.pdf.
Johnson, J., and M. Kleiner. 2017. Is Occupational Licensing a Barrier to Interstate
Migration? NBER Working Paper 24107. Cambridge, MA: National Bureau of
Economic Research.
Krueger, A. 2017. “Where Have All the Workers Gone? An Inquiry into the Decline of the
U.S. Labor Force Participation Rate.” Washington: Brookings Institution.
https://www.brookings.edu/wp-content/uploads/2017/09/1_krueger.pdf.
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.
Maguire S., J. Freely, C. Clymer, M. Conway, and D. Schwartz. 2010. “Tuning In to Local
Labor Markets: Findings from the Sectoral Employment Impact Study.”
Public/Private Ventures.
Meyer, B., W. Mok, and J. Sullivan. 2015. “Household Surveys in Crisis.” Journal of
Economic Perspectives 29, no. 4: 199–226.
Meyer, J. 2017. How Occupational Licensing Inhibits Economic Opportunity. Naples, FL:
Foundation for Government Accountability. https://thefga.org/wp-content/
uploads/2017/10/How-Occupational-Licensing-Inhibits-EconomicOpportunity-10-23-17.pdf.
NCSHA (National Council of State Housing Agencies). 2019. “Opportunity Zone Target
Investment Surges to Nearly $43 Billion Nationwide.” https://www.cdfifund.
gov/Documents/Summit%20-%20Compliance%20Review%20of%20New%20
Markets%20Tax%20Credit%20Program%20-%20August%20Date%20-%20
508%20Compliant.pdf.
Nunn, R., J. Parsons, and J. Shambaugh. 2018. “The Geography of Prosperity.” In PlaceBased Policies for Shared Economic Growth, edited by J. Shambaugh and R.
Nunn. Washington: Brookings Institution. Office of Family Assistance. 2019.
“TANF Caseload Data 2018.” U.S. Department of Health and Human Services.
https://www.acf.hhs.gov/ofa/resource/tanf-caseload-data-2018.

308 | References

250-840_text_.pdf 312

2/7/20 3:46 PM

Real Capital Analytics. “U.S. Opportunity Zones: A Baseline.” https://www.rcanalytics.
com/opportunity-zones-baseline/
Roder, A., and M. Elliot. 2019. “Nine Year Gains: Project QUEST’s Continuing Impact.”
Economic Mobility Corporation. https://economicmobilitycorp.org/
wp-content/uploads/2019/04/NineYearGains_web.pdf.
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.” https://papers.ssrn.com/sol3/
papers.cfm?abstract_id=3385502.
Schnepel, K. 2016. “Good Jobs and Recidivism.” Economic Journal 128: 447–69.
Szafran, R. 2002. “Age-Adjusted Labor Force Participation Rates, 1960–2045.” Monthly
Labor Review 125: 25–38.
U.S. Census Bureau. 2019. “Income and Poverty in the United States: 2018.” https://
www.census.gov/library/publications/2019/demo/p60-266.html.
White House. 2015. “Occupational Licensing: A Framework for Policymakers.” https://
obamawhitehouse.archives.gov/sites/default/files/docs/licensing_report_
final_nonembargo.pdf.
Woetzel, J., A. Madgavkar, K. Ellingrud, E. Labaye, S. Devillard, E. Kutcher, J. Manyika, R.
Dobbs, and M. Krishnan. 2015. “The Power of Parity: How Advancing Women’s
Equality Can Add $12 Trillion to Global Growth.” McKinsey Global Institute.
World Bank. 2014. “Levelling the Field: Improving Opportunities for Women Farmers in
Africa.” https://www.one.org/international/policy/
levelling-the-field-improving-opportunities-for-women-farmers-in-africa/.

Chapter 3
AARP (American Association of Retired Persons). 2014. “Workplace Retirement Plans
Will Help Workers Build Economic Security.” https://www.aarp.org/content/
dam/aarp/ppi/2014-10/aarp-workplace-retirement-plans-build-economicsecurity.pdf.
Al-Ubaydli, O., and P. McLaughlin. 2014. RegData: A Numerical Database on IndustrySpecific Regulations for All U.S. Industries and Federal Regulations, 1997–2012.
Working paper. Arlington, VA: Mercatus Center at George Mason University.
Arrow, K., and J. Kalt. 1979. Petroleum Price Regulation: Should We Decontrol? Studies
in Energy Policy. Washington: American Enterprise Institute.
Becker, G., D. Carlton, and H. Sider. 2010. “Net Neutrality and Consumer Welfare.”
Journal of Competition Law and Economics 497.
Belfield, C., A. Bowden, and V. Rodriguez. 2018. “Evaluating Regulatory Impact
Assessments in Education Policy.” American Journal of Evaluation, 1–19.
Bento, A., K. Gilingham, M. Jacobsen, C. Knittel, B. Leard, and J. Linn. 2018. “Flawed
Analyses of U.S. Auto Fuel Economy Standards.” Science 362, no. 6419:
1119–21.

References | 309

250-840_text_.pdf 313

2/7/20 3:46 PM

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: 68–105.
Bernheim, B., A. Fradkin, and I. Popov. 2015. “The Welfare Economics of Default
Options in 401(k) Plans.” American Economic Review 105, no. 9: 2798–2837.
Brown, P. 2016. “Statement Submitted to the House Committee on Education and the
Workforce.” April 27.
Brynjolfsson, E., F. Eggers, and A. Gannamaneni. 2018. Using Massive Online Choice
Experiments to Measure Changes in Well-Being. NBER Working Paper 24514.
Cambridge, MA: National Bureau of Economic Research.
Busse, M., C. Knittel, and F. Zettelmeyer. 2013. “Are Consumers Myopic? Evidence from
New and Used Car Purchases.” American Economic Review 103, no. 1: 220–56.
CEA (Council of Economic Advisers). 2018a. Economic Report of the President.
Washington: U.S. Government Publishing Office.
———. 2018b. “The Administration’s FDA Reforms and Reduced Biopharmaceutical
Drug Prices.” https://www.whitehouse.gov/wp-content/uploads/2018/10/
The-Administrations-FDA-Reforms-and-Reduced-Biopharmaceutical-DrugPrices.pdf.
———. 2019a. “Deregulating Health Insurance Markets: Value to Market Participants.”
https://www.whitehouse.gov/wp-content/uploads/2019/02/DeregulatingHealth-Insurance-Markets-FINAL.pdf.
———. 2019b. Economic Report of the President. Washington: U.S. Government
Publishing Office.
———. 2019c. “The Economic Effects of Federal Deregulation since January 2017: An
Interim Report.” https://www.whitehouse.gov/wp-content/uploads/2019/06/
The-Economic-Effects-of-Federal-Deregulation-Interim-Report.pdf.
Chetty, R. 2009. “Sufficient Statistics for Welfare Analysis: A Bridge Between Structural
and Reduced-Form Methods.” Annual Review of Economics 1: 451–88.
Congressional Budget Office. 2003. “The Economic Costs of Fuel Economy Standards
versus a Gasoline Tax.”
Covert, T., and R. Kellogg. 2017. Crude by Rail, Option Value, and Pipeline Investment.
NBER Working Paper 23855. Cambridge, MA: National Bureau of Economic
Research.
Cummings, S. 2016. “Preliminary Injunction Order in NFIB vs. Perez.”
Dahlby, B. 2008. The Marginal Cost of Public Funds: Theory and Applications.
Cambridge, MA: MIT Press.
Davis, L., and L. Kilian. 2011. “The Allocative Cost of Price Ceilings in the U.S.
Residential Market for Natural Gas.” Journal of Political Economy 119, no. 2:
212–41.
DOL (U.S. Department of Labor). 2017. “U.S. Department of Labor Rescinds 2016
Persuader Rule.” News release, July 17. https://www.dol.gov/newsroom/
releases/olms/olms20180717.

310 |

250-840_text_.pdf 314

References

2/7/20 3:46 PM

Eilperin, J. 2017. “In a Setback for Unions, Labor Department Moves to Revoke
Obama-Era ‘Persuader Rule.’” June 9. https://www.washingtonpost.com/
news/wonk/wp/2017/06/09/
in-a-setback-for-unions-labor-department-moves-to-revoke-obama-erapersuader-rule/.
Feldman, R., J. Schmidt, and K. Heinecke. 2013. Quantifying the Costs of Additional
Regulation on Community Banks. Economic Policy Paper 13-3. Minneapolis:
Federal Reserve Bank of Minneapolis.
Fenn, P., and C. Veljanovski. 1988. “A Positive Economic Theory of Regulatory
Enforcement.” Economic Journal, no. 393: 1055–70.
Furchtgott-Roth, D. 2016. “The High Costs of Proposed New Labor-Law Regulations.”
Manhattan Institute Issue Brief, updated.
Furth, S., and D. Kreutzer. 2016. Fuel Economy Standards Are a Costly Mistake.
Washington: Heritage Foundation.
Goolsbee, A. 2006. “The Value of Broadband and the Deadweight Loss of Taxing New
Technology.” Contributions to Economic Analysis and Policy 5, no. 1: article 8.
Goulder, L., I. Parry, and R. Williams III. 1999. “The Cost-Effectiveness of Alternative
Instruments for Environmental Protection in a Second-Best Setting.” Journal
of Public Economics 72, no 3: 329–60.
Goulder, L., and R. Williams III. 2003. “The Substantial Bias from Ignoring General
Equilibrium Effects in Estimating Excess Burden, and a Practical Solution.”
Journal of Political Economy 111, no. 4: 898–927.
Harberger, A. 1964. “The Measurement of Waste.” American Economic Review 54, no. 3:
58–76.
Harrington, W., R. Morgenstern, and P. Nelson. 2000. “On the Accuracy of Regulatory
Cost Estimates.” Journal of Policy Analysis and Management 19, no. 2:
297–322.
Haveman, R., and B. Weisbrod. 1975. “Defining Benefits of Public Programs: Some
Guidance for Policy Analysts.” Policy Analysis 1, no. 1: 169–96.
Hayden, K. 2018. “Illinois First State in U.S. to Pass Mandatory Secure Choice
Retirement Savings.” HR Source. https://www.hrsource.org/maimis/
Members/Articles/2018/11/November_27/Illinois_First_State_in_U.S._to_
Pass_Mandatory_Secure_Choice_Retirement_Savings_Via_AARP_Illinois.
aspx.
Hazlett, T., and A. Caliskan. 2008. “Natural Experiments in U.S. Broadband Regulation.”
Review of Network Economics 7, no. 4: 460–80.
Hopkins, J. 2015. “Illinois Enacts Automatic IRAs; Is Federal Mandate Coming?” Forbes,
January 7. https://www.forbes.com/sites/jamiehopkins/2015/01/07/
automatic-enrollment-iras-first-government-mandated-automatic-ira-isenacted-in-illinois/#4f37805e290c.
Koetter, M., J. Kolari, and L. Spierdijk. 2012. “Enjoying the Quiet Life Under
Deregulation? Evidence from Adjusted Lerner Indices for U.S. Banks.” Review
of Economics and Statistics 94, no. 2: 462–80.

References | 311

250-840_text_.pdf 315

2/7/20 3:46 PM

Lazaer, E. 1979. “Why Is there Mandatory Retirement?” Journal of Political Economy 87,
no. 6: 1261–84.
———. 1981. “Agency, Earnings Profiles, Productivity, and Hours Restrictions.” American
Economic Review 71, no. 4: 606–20.
Madrian, B., and D. Shea. 2001. “The Power of Suggestion: Inertia in 401(k)
Participation and Savings Behavior.” Quarterly Journal of Economics 116, no.
4: 1149–87.
McDevitt, R., and A. Sojourner. 2016. “Demand, Regulation, and Welfare on the Margin
of Alternative Financial Services.” Working paper, Duke University.
Mortensen, D. 2010. “Nobel Lecture: Markets with Search Friction and the DMP Model.”
https://www.nobelprize.org/uploads/2018/06/mortensen-lecture.pdf.
Murphy, R. 2018. “Removing the 1970s Crude Oil Price Controls: Lessons for FreeMarket Reform.” Journal of Private Enterprise 33, no. 1: 63–78.
Murray Energy Corporation v. U.S. Department of the Interior. 2016. https://www.
courthousenews.com/wp-content/uploads/2016/12/murraysuit.pdf.
NCSL (National Conference of State Legislatures). 2018. “State-Facilitated Retirement
Savings Programs for Private Sector Workers.” http://www.ncsl.org/research/
fiscal-policy/state-facilitated-retirement-savings-programs-for-private-sectorworkers.aspx.
NFIB v. Perez (National Federation of Independent Business v. Perez). 2016. http://www.
constangy.net/nr_images/nfib-v-perez-pI-order.pdf.
OMB (Office of Management and Budget). 1992. “Circular A-94: Guidelines and Discount
Rates for Benefit-Cost Analysis of Federal Programs.” https://www.
whitehouse.gov/sites/whitehouse.gov/files/omb/circulars/A94/a094.pdf.
———. 2003. “Circular A-4: Regulatory Impact Analysis—A Primer.” https://
obamawhitehouse.archives.gov/omb/circulars_a004_a-4/.
———. 2017. “Guidance Implementing Executive Order 13771, Titled ‘Reducing
Regulation and Controlling Regulatory Costs.’” https://www.whitehouse.gov/
sites/whitehouse.gov/files/omb/memoranda/2017/M-17-21-OMB.pdf.
———. 2018. “Regulatory Reform: Regulatory Budget for Fiscal Year 2019.” https://www.
reginfo.gov/public/pdf/eo13771/EO_13771_Regulatory_Budget_for_Fiscal_
Year_2019.pdf.
Panis, C., and M. Brien. 2015. “Target Populations of State-Level Automatic IRA
Initiatives.” Department of Labor. https://www.dol.gov/sites/default/files/
ebsa/researchers/analysis/retirement/target-populations-of-state-levelautomatic-ira-initiatives.pdf.
Parrillo, N. 2019. “Should the Public Get to Participate Before Federal Agencies Issue
Guidance? An Empirical Study.” Administrative Law Review 71, no. 1: 57–125.
Parry, I., R. Williams III, and L. Goulder. 1999. “When Can Carbon Abatement Policies
Increase Welfare? The Fundamental Role of Distorted Factor Markets.” Journal
of Environmental Economics and Management 37, no. 1: 52–84.

312 | References

250-840_text_.pdf 316

2/7/20 3:46 PM

Public Citizen. 2016. “Shining a Light of the ‘Midnight Rule’ Boogeyman: An Analysis of
Economically Significant Final Rules Reviewed by OIRA.” Washington: Public
Citizen.
Roehrig, C. 2018. The Impact of Prescription Drug Rebates on Health Plans and
Consumers. Ann Arbor, MI: Altarum Institute.
Sheetz, M. 2017. “Teva Stock Plummets 18% After Rough Quarter in U.S. Generic Drug
Market.” CNBC, August 3. https://finance.yahoo.com/news/teva-stockplummets-18-rough-135334052.html.
Staten, M., and F. Cate. 2003. “The Impact of Opt-In Privacy Rules on Retail Credit
Markets: A Case Study of MBNA.” Duke Law Journal 52: 745–86.
State of Oregon. 2015. “78th Oregon Legislative Assembly: 2015 Regular Session—
Enrolled House Bill 2960.” https://olis.leg.state.or.us/liz/2015R1/Downloads/
MeasureDocument/HB2960/Enrolled.
Tankersley, J. 2019. “Fed, Dimming Its Economic Outlook, Predicts No Rate Increases
This Year.” New York Times, March 20. https://www.nytimes.com/2019/03/20/
us/politics/fed-rates.html.
UC Berkeley Labor Center. 2017. “California Secure Choice Retirement Savings Program
(SB 1234): FAQ for Unions and Union Members.” http://laborcenter.berkeley.
edu/pdf/2017/SB-1234-Fact-Sheet.pdf.
U.S. Court of Appeals for the Eighth Circuit. 2018. “Decision in Organization for
Competitive Markets, et al. v. U.S.D.A.”
U.S. Department of the Treasury. 2017. Limiting Consumer Choice, Expanding Costly
Litigation: An Analysis of the CFPB Arbitration Rule.
Winston, C. 1993. “Economic Deregulation: Days of Reckoning for Microeconomists.”
Journal of Economic Literature 31: 1263–89.

Chapter 4
Alvarez, R., D. Zavala-Araiza, D. Lyon, D. Allen, Z. Barkley, A. Brandt, K. Davis, S.
Herndon, D. Jacob, A. Karion, and E. Kort. 2018. “Assessment of Methane
Emissions from the U.S. Oil and Gas Supply Chain.” Science 361, no. 6398:
186–88.
Avalos, R., T. Fitzgerald, and R. Rucker. 2016. “Measuring the Effects of Natural Gas
Pipeline Constraints on Regional Pricing and Market Integration.” Energy
Economics 60: 217–31.
Barteczko, A. 2019. “Poland and Lithuania see Nord Stream 2 as Threat to Energy
Security.” Reuters. Edited by Jason Neely. https://www.reuters.com/article/
us-poland-lithuania-nordstream2/
poland-and-lithuania-see-nord-stream-2-as-threat-to-energy-securityidUSKBN1W20TQ.
Bartik, A., J. Currie, M. Greenstone, and C. Knittel. 2019. “The Local Economic and
Welfare Consequences of Hydraulic Fracturing.” American Economic Journal:
Applied Economics 11, no. 4: 105–55.

References | 313

250-840_text_.pdf 317

2/7/20 3:46 PM

Baumeister, C., and L. Kilian. 2016. “Lower Oil Prices and the U.S. Economy: Is this Time
Different?” Brookings Papers on Economic Activity, no. 2: 287–357.
Borenstein, S., and J. Bushnell. 2015. “The U.S. Electricity Industry After 20 Years of
Restructuring.” Annual Review of Economics 7, no. 1: 437–63.
Boslett, A., T. Guilfoos, and C. Lang. 2016. “Valuation of Expectations: A Hedonic Study
of Shale Gas Development and New York’s Moratorium.” Journal of
Environmental Economics and Management 77: 14–30.
Brown, J., T. Fitzgerald, and J. Weber. 2016. “Capturing Rents from Natural Resource
Abundance: Private Royalties from U.S. Onshore Oil & Gas Production.”
Resource and Energy Economics 46: 23–38.
Caldara, D., M. Iacoviello, P. Molligo, A. Prestipino, and A. Raffo. 2019. “Does Trade
Policy Uncertainty Affect Global Economic Activity?” FEDS Notes. Washington:
Board of Governors of the Federal Reserve System.
Cavallo, M. 2006. “Oil Prices and the U.S. Trade Deficit.” Federal Reserve Bank of San
Francisco Economic Letter. http://www.frbsf.org/economic-research/
publications/economic-letter/2006/september/
oil-prices-and-the-us-trade-deficit/.
CEA (Council of Economic Advisers). 2019. “The Value of U.S. Energy Innovation and
Policies Supporting the Shale Revolution.” https://www.whitehouse.gov/
wp-content/uploads/2019/10/The-Value-of-U.S.-Energy-Innovation-andPolicies-Supporting-the-Shale-Revolution.pdf.
Coglianese, J., T. Gerarden, and J. Stock. 2019. “The Effects of Fuel Prices, Regulations,
and Other Factors on U.S. 2008–2016.” Energy Journal 41, no. 1, forthcoming.
https://scholar.harvard.edu/files/stock/files/coglianese_gerarden_stock.pdf.
CRS (Congressional Research Service). 2018. “Nuclear Energy: Overview of
Congressional Issues.” https://fas.org/sgp/crs/row/IF10715.pdf.
———. 2019. “Venezuela: Overview of U.S. Sanctions.” https://fas.org/sgp/crs/row/
IF10715.pdf.
Currie, J., M. Greenstone, and K. Meckel. 2017. “Hydraulic Fracturing and Infant Health:
New Evidence from Pennsylvania.” Science Advances 3, no. 12: 1–9.
Darrah, T., A. Vengosh, R. Jackson, N. Warner, and R. Poreda, 2014. “Noble Gases
Identify the Mechanisms of Fugitive Gas Contamination in Drinking-Water
Wells Overlying the Marcellus and Barnett Shales.” Proceedings of the
National Academy of Sciences 111, no. 39: 14076–81.
DOT (Department of the Treasury). 2019. “Iran Sanctions.” https://www.treasury.gov/
resource-center/sanctions/programs/pages/iran.aspx.
EIA (U.S. Energy Information Administration). 2016. “Trends in U.S. Oil and Natural Gas
Upstream Costs.” https://www.eia.gov/analysis/studies/drilling/pdf/
upstream.pdf.
———. 2019a. “EIA Estimates of Drilled but Uncompleted Wells (DUCs).” https://www.
eia.gov/petroleum/drilling/pdf/duc_supplement.pdf.

314 | References

250-840_text_.pdf 318

2/7/20 3:46 PM

———. 2019b. “Fossil Fuels Continue to Account for the Largest Share of U.S. Energy.”
Today in Energy. https://www.eia.gov/todayinenergy/detail.php?id=41353.
———. 2019c. Annual Energy Outlook 2019 with Projections to 2050. https://www.eia.
gov/outlooks/aeo/pdf/aeo2019.pdf.
———. 2019d. “The United States Is Expected to Export More Energy Than It Imports by
2020.” Today in Energy. https://www.eia.gov/todayinenergy/detail.
php?id=38152.
———. 2018a. “The United States Exported More Natural Gas Than It Imported in 2017.”
Today in Energy. https://www.eia.gov/todayinenergy/detail.php?id=35392.
———. 2018b. “U.S. Crude Oil Production Efficiency Continues to Improve.” Today in
Energy. https://www.eia.gov/todayinenergy/detail.php?id=36012.
Fell, H., and D. Kaffine. 2018. “The Fall of Coal: Joint Impacts of Fuel Prices and
Renewables on Generation and Emissions.” American Economic Journal:
Economic Policy 10, no. 2: 90–116.
Feyrer, J., E. Mansur, and B. Sacerdote. 2017. “Geographic Dispersion of Economic
Shocks: Evidence from the Fracking Revolution.” American Economic Review
107 no. 4: 1313–34.
Financial Times. 2017. “In Charts: Has the U.S. Shale Drilling Revolution Peaked?”
https://www.ft.com/content/e17930dc-b288-11e7-a398-73d59db9e399.
Fitzgerald, T. 2014. “Importance of Mineral Rights and Royalty Interests for Rural
Residents and Landowners.” Choices 29, no. 4: 1–7.
Gillingham, K., and J. Stock. 2018. “The Cost of Reducing Greenhouse Gas Emissions.”
Journal of Economic Perspectives 32, no. 4: 53–72.
Gopalakrishnan, S., and H. Klaiber. 2013. “Is the Shale Energy Boom a Bust for Nearby
Residents? Evidence from Housing Values in Pennsylvania.” American Journal
of Agricultural Economics 96, no. 1: 43–66.
Gordon, D. 2012. The Carbon Contained in Global Oils, vol. 18. Washington: Carnegie
Endowment for International Peace.
Graham, J., J. Irving, X. Tang, S. Sellers, J. Crisp, D. Horwitz, L. Muehlenbachs, A.
Krupnick, and D. Carey. 2015. “Increased Traffic Accident Rates Associated
with Shale Gas Drilling in Pennsylvania.” Accident Analysis & Prevention 74:
203–9.
Greene, D. 2010. “Measuring Energy Security: Can the United States Achieve Oil
Independence?” Energy Policy 38, no. 4: 1614–21
Greenstone, M., and I. Nath. 2019. “Do Renewable Portfolio Standards Deliver?” Energy
Policy Institute at the University of Chicago Working Paper 2019-62.
Hamilton, J. 2011. Historical Oil Shocks. NBER Working Paper 16790. Cambridge, MA:
National Bureau of Economic Research. https://www.nber.org/papers/
w16790.pdf.
Hausman, C., and R. Kellogg. 2015. “Welfare and Distributional Implications of Shale
Gas.” Brookings Papers on Economic Activity, Spring, 71–125.

References | 315

250-840_text_.pdf 319

2/7/20 3:46 PM

Homans, C. 2012. “Energy Independence: A Short History.” Foreign Policy, January 3.
https://foreignpolicy.com/2012/01/03/energy-independence-a-short-history/.
Hill, E., and L. Ma. 2017. “Shale Gas Development and Drinking Water Quality.”
American Economic Review 107, no. 5: 522–25.
IEA (International Energy Agency). 2019. “United States 2019 Review.” https://webstore.
iea.org/energy-policies-of-iea-countries-united-states-2019-review.
Jacobsen, G. 2019. “Who Wins in an Energy Boom? Evidence from Wage Rates and
Housing.” Economic Inquiry 57, no 1: 9–32.
Johnsen, R., J. LaRiviere, and H. Wolff. 2019. “Fracking, Coal, and Air Quality.” Journal
of the Association of Environmental and Resource Economists 6, no. 5:
1001–37.
Johnson, E. 2014. “The Cost of Carbon Dioxide Abatement from State Renewable
Portfolio Standards.” Resource and Energy Economics 36, no. 2: 332–50.
Keranen, K., M. Weingarten, G. Abers, B. Bekins, and S. Ge. 2014. “Sharp Increase in
Central Oklahoma Seismicity since 2008 Induced by Massive Wastewater
Injection.” Science 345, no. 6195: 448–51.
Kilian, L. 2017. “The Impact of the Fracking Boom on Arab Oil Producers.” The Energy
Journal, International Association of Energy Economics, no. 8.
———. 2016. The Impact of the Shale Oil Revolution on U.S. Oil and Gasoline Prices.
Working Paper 5723. Munich: CESifo.
Komarek, T. 2016. “Labor Market Dynamics and the Unconventional Natural Gas Boom:
Evidence from the Marcellus Region.” Resource and Energy Economics 45:
1–17.
Linn, J., L. Muehlenbachs, and Y. Wang. 2014. How Do Natural Gas Prices Affect
Electricity Consumers and the Environment? RFF Discussion Paper 14-19.
Washington: Resources for the Future.
Litovitz, A., A. Curtright, S. Abramzon, N. Burger, and C. Samaras. 2013. “Estimation of
Regional Air-quality Damages from Marcellus Shale Natural Gas Extraction in
Pennsylvania.” Environmental Research Letters 8, no. 1: 14–17.
MacAvoy, P. 2008. The Natural Gas Market: Sixty Years of Regulation and Deregulation.
New Haven, CT: Yale University Press.
Marchand, J., and J. Weber. 2019. “How Local Economic Conditions Affect School
Finances, Teacher Quality, and Student Achievement: Evidence from the
Texas Shale Boom.” Journal of Policy Analysis and Management, forthcoming.
Martin, R., M. Muûls, and U. Wagner. 2015. “The Impact of the European Union
Emissions Trading Scheme on Regulated Firms: What Is the Evidence after
Ten Years?” Review of Environmental Economics and Policy 10, no. 1: 129–48.
Mathur, A., and A. Morris. 2014. “Distributional Effects of a Carbon Tax in Broader U.S.
Fiscal Reform.” Energy Policy 66. 326–34.
Muehlegger, E., and R. Sweeney. 2017. Pass-Through of Input Cost Shocks Under
Imperfect Competition: Evidence from the U.S. Fracking Boom. NBER Working
Paper 24025. Cambridge, MA: National Bureau of Economic Research.

316 |

250-840_text_.pdf 320

References

2/7/20 3:46 PM

Muehlenbachs, L., E. Spiller, and C. Timmins. 2015. “The Housing Market Impacts of
Shale Gas Development.” American Economic Review 105, no. 12: 3633–59.
Newell, R., and D. Raimi. 2018. “The Fiscal Impacts of Increased U.S. Oil and Gas
Development on Local Governments.” Energy Policy 117: 14–24.
Newell, R., B. Prest, and A. Vissing. 2019. “Trophy Hunting versus Manufacturing
Energy: The Price-Responsiveness of Shale Gas.” Journal of the Association of
Environmental and Resource Economists 6, no. 2: 391–431.
Oliver, M., C. Mason, and D. Finnoff. 2014. “Pipeline Congestion and Basis Differentials.”
Journal of Regulatory Economics 46, no. 3: 261–91.
Wang, Z., and A. Krupnick. 2015. “A Retrospective Review of Shale Gas Development in
the United States: What Led to the Boom?” Economics of Energy &
Environmental Policy 4, no. 1: 5–18.
Weber, J., J. Burnett, and I. Xiarchos. 2016. “Broadening Benefits from Natural
Resource Extraction: Housing Values and Taxation of Natural Gas Wells as
Property.” Journal of Policy Analysis and Management 35, no. 3: 587–614.
Weber, J., Y. Wang, and M. Chomas. 2016. “A Quantitative Description of State-Level
Taxation of Oil and Gas Production in the Continental U.S.” Energy Policy 96:
289–301.
Wrenn, D., H. Klaiber, and E. Jaenicke. 2016. “Unconventional Shale Gas Development,
Risk Perceptions, and Averting Behavior: Evidence From Bottled Water
Purchases.” Journal of the Association of Environmental and Resource
Economists 3, no. 4: 779–817.

Chapter 5
AHIP (America’s Health Insurance Plans). 2017. “Health Savings Accounts and
Consumer-Directed Health Plans Grow as Valuable Financial Planning Tools.”
https://www.ahip.org/2017-survey-of-health-savings-accounts/.
Aitken, M., E. Berndt, D. Cutler, M. Kleinrock, and L. Maini. 2016. “Has the Era of Slow
Growth for Prescription Drug Spending Ended?” Health Affairs 35, no. 9:
1595–1603.
American Medical Association. 2018. “U.S. Justice Department Should Block CVS-Aetna
Merger.” https://www.ama-assn.org/delivering-care/patient-supportadvocacy/us-justice-department-should-block-cvs-aetna-merger.
Armour, S. 2019. “Premiums for ACA Health Plans Are Set to Drop in 2020.” Wall Street
Journal, October 22. https://www.wsj.com/articles/
premiums-for-aca-health-plans-are-set-to-drop-in-2020-11571749200.
Aron-Dine, A. 2019. “CBO: Administration’s Short-Term Plans Rule Means a Return to
Pre-ACA Practices.” Center on Budget and Policy Priorities. https://www.cbpp.
org/blog/
cbo-administrations-short-term-plans-rule-means-a-return-to-pre-acapractices.

References | 317

250-840_text_.pdf 321

2/7/20 3:46 PM

Azar, A. 2019. “Ending the HIV Epidemic: A Plan for America.” Department of Health &
Human Services. https://www.hhs.gov/blog/2019/02/05/ending-the-hivepidemic-a-plan-for-america.html.
Badger, D. 2019. “Obamacare Caused Premiums to Spike; Here’s How States Are
Lowering Them Again.” Heritage Foundation. https://www.heritage.org/
health-care-reform/commentary/
obamacare-caused-premiums-spike-heres-how-states-are-lowering-them.
Berchick, E., J. Barnett, and R. Upton. 2019. Health Insurance Coverage in the United
States: 2018. U.S. Census Bureau Report P60-267. Washington: U.S.
Government Publishing Office.
Berry, S., M. Gaynor, and F. Morton. 2019. “Do Increasing Markups Matter? Lessons from
Empirical Industrial Organization.” Journal of Economic Perspectives 33, no. 3:
44–68.
Blase, B. 2019a. Health Reform Progress: Beyond Repeal and Replace. Paeonian Springs,
VA: Galen Institute.
———. 2019b. “How Price Transparency Would Revolutionize Healthcare.” New York
Post, October 12. https://nypost.com/2019/10/12/
how-price-transparency-would-revolutionize-healthcare/.
Bosworth, B., J. Bieler, M. Kleinrock, E. Koepcke, and E. Berndt. 2018. An Evaluation of
the CPI Indexes for Prescription Drugs. NBER Working Paper 24210.
Cambridge, MA: National Bureau of Economic Research.
Boskin, M., J. Dulberger, R. Gordon, Z. Griliches, and D. Jorgenson. 1996. “Toward a
More Accurate Measure of the Cost of Living.” Final Report to the Senate
Finance Committee, December 4.
Burns, L., and M. Pauly. 2018. “Transformation of the Health Care Industry: Curb Your
Enthusiasm?” Milbank Quarterly 96, no. 1: 57–109.
Carpenter, E., and C. Sloan. 2018. “Health Plans with More Restrictive Provider
Networks Continue to Dominate the Exchange Market.” Press release, Avalere.
https://avalere.com/press-releases/
health-plans-with-more-restrictive-provider-networks-continue-to-dominatethe-exchange-market.
CBO (Congressional Budget Office). 2019. “How CBO and JCT Analyzed Coverage
Effects of New Rules for Association Health Plans and Short-Term Plans.”
https://www.cbo.gov/system/files/2019-01/54915-New_Rules_for_AHPs_
STPs.pdf.
CDC (Centers for Disease Control and Prevention). 2019. “About Chronic Diseases.”
National Center for Chronic Disease Prevention and Health Promotion.
https://www.cdc.gov/chronicdisease/index.htm.
CEA (Council of Economic Advisers). 2019a. Economic Report of the President.
Washington: U.S. Government Publishing Office.
———. 2019b. Deregulating Health Insurance Markets: Value to Market Participants.
Washington: White House. https://www.whitehouse.gov/wp-content/
uploads/2019/02/Deregulating-Health-Insurance-Markets-FINAL.pdf.

318 |

250-840_text_.pdf 322

References

2/7/20 3:46 PM

———. 2019c. “Measuring Prescription Drug Prices: A Primer on the CPI Prescription
Drug Index.” https://www.whitehouse.gov/wp-content/uploads/2019/10/
Measuring-Prescription-Drug-Prices-A-Primer-on-the-CPI-Prescription-DrugIndex.pdf.
———. 2019d. Mitigating the Impact of Pandemic Influenza through Vaccine Innovation.
Washington: White House. https://www.whitehouse.gov/wp-content/
uploads/2019/09/Mitigating-the-Impact-of-Pandemic-Influenza-throughVaccine-Innovation.pdf.
———. 2018. Reforming Biopharmaceutical Pricing at Home and Abroad. Washington:
White House. https://www.whitehouse.gov/wp-content/uploads/2017/11/
CEA-Rx-White-Paper-Final2.pdf.
Census Bureau. 2019. “Health Insurance Coverage in the United States: 2018.” https://
www.census.gov/content/dam/Census/library/publications/2019/demo/
p60-267.pdf.
Chernew, M., A. Rosen, and A. Fendrick. 2007. “Value-Based Insurance Design.” Health
Affairs 26, no. 2: 195-203.
Claxton, G., M. Rae, A. Damico, G. Young, D. McDermott, and H. Whitmore. 2019.
“Health Benefits in 2019: Premiums Inch Higher, Employers Respond to
Federal Policy.” Health Affairs 38, no 10.
CMS (Centers for Medicare and Medicaid Services). 2018. “Premiums on the Federallyfacilitated Exchanges drop in 2019.” https://www.cms.gov/newsroom/
press-releases/premiums-federally-facilitated-exchanges-drop-2019.
———. 2019a. “Organ Procurement Organization (OPO) Conditions for Coverage
Proposed Rule: Revisions to Outcome Measures for OPOs.” https://www.cms.
gov/newsroom/fact-sheets/
organ-procurement-organization-opo-conditions-coverage-proposed-rulerevisions-outcome-measures-opos.
———. 2019b. “Premiums for HealthCare.gov Plans are down 4 percent but remain
unaffordable to non-subsidized consumers.” https://www.cms.gov/
newsroom/press-releases/
premiums-healthcaregov-plans-are-down-4-percent-remain-unaffordablenon-subsidized-consumers.
Conrad, L. 2019. “HSAs March Ever Higher, Expected to Continue.” Entertainment
Benefit News. https://www.benefitnews.com/news/
number-of-hsa-expected-to-continue.
Cooper, Z., S. Gibbons, S. Jones, and A. McGuire. 2011. “Does Hospital Competition
Save Lives? Evidence from the English NHS Patient Choice Reforms.”
Economic Journal 121, no. 554: 228–60.
Dafny, L., K. Ho, and M. Varela. 2013. “Let Them Have Choice: Gains from Shifting Away
from Employer-Sponsored Health Insurance and Toward an Individual
Exchange.” American Economic Journal 5, no. 1: 32–58.

References | 319

250-840_text_.pdf 323

2/7/20 3:46 PM

eHealth Insurance. 2018. “How Much Does Obamacare Cost in 2018?” June 20. https://
www.ehealthinsurance.com/resources/affordable-care-act/
much-obamacare-cost-2018.
Effros, R. 2009. “Increase Cost-Participation by Employees (e.g., Through HighDeductible Health Plans).” RAND Corporation. https://www.rand.org/pubs/
technical_reports/TR562z4/analysis-of-high-deductible-health-plans.html.
Fehr, R., R. Kamal, and C. Cox. 2019. “Insurer Participation on ACA Marketplaces, 2014–
2020.” Kaiser Family Foundation. https://www.kff.org/private-insurance/
issue-brief/insurer-participation-on-aca-marketplaces-2014-2020/.
Feldman, R., S. Parente, J. Abraham, J. Christianson, and R. Taylor. 2005. “Health
Savings Accounts: Early Estimations on National Take-Up.” Health Affairs 24,
no. 6.
Foppa, I., P. Cheng, S. Reynolds, D. Shay, C. Carias, et al. 2015. “Deaths Averted by
Influenza Vaccination in the U.S. during the Seasons 2005/06 through
2013/14.” Vaccine 33, no. 26: 3003–9.
Frogner, B., and J. Spetz. 2017. “Health Policy Debates and the Outlook for Health Care
Jobs.” Health Affairs, Blog, November 3. https://www.healthaffairs.org/
do/10.1377/hblog20171102. 115215/full/.
Fuchs, V. 2007. “Managed Care and Merger Mania.” Journal of the American Medical
Association 277, no. 11: 920–21.
Gaynor, M., and R. Town. 2012. “The Impact of Hospital Consolidation—Update.”
Synthesis Project Policy Brief, no. 9.
Gaynor, M., K. Ho., and R. Town. 2015. “The Industrial Organization of Health-Care
Markets.” Journal of Economic Literature 53, no. 2: 235–84.
Gee, E., C. Peters, and J. Wilder. 2019. “The Year in Review: Economics at the Antitrust
Division, 2018–2019.” Review of Industrial Organization 55: 537–50.
Greene, J., J. Hibbard, A. Dixon, and M. Tusler. 2006. “Which Consumers Are Ready for
Consumer-Driven Health Plans?” Journal of Consumer Policy 29, no. 3:
247–62.
Haas-Wilson, D., and C. Garmon. 2011. “Hospital Mergers and Competitive Effects: Two
Retrospective Analyses.” International Journal of the Economics of Business
18, no. 1: 17–32.
Haefner, M. 2019. “Fitch Solutions: U.S. Economic Burden of Chronic Disease
‘Unparalleled.’” Beckers Hospital Review. https://www.beckershospitalreview.
com/finance/fitch-solutions-us-economic-burden-of-chronic-diseaseunparalleled.html.
Haviland, A., M. Eisenberg, A. Mehrotra, P. Huckfeldt, and N. Sood. 2016. “Do
‘Consumer-Directed’ Health Plans Bend the Cost Curve Over Time?” Journal
of Health Economics 46: 33–51.
Haviland, A., N. Sood, R. McDevitt, and S. Marquis. 2011. “How Do Consumer-Directed
Health Plans Affect Vulnerable Populations?” Forum for Health Economics &
Policy 14, no. 2 article 3.

320 | References

250-840_text_.pdf 324

2/7/20 3:46 PM

HCCI (Health Care Cost Institute). 2016. “National Chartbook of Health Care Prices—
2015.” https://www.healthcostinstitute.org/images/pdfs/
HCCI-National-Chartbook-of-Health-Care-Prices-2015.pdf.
HHS (Department of Health and Human Services). 2018. “Reforming America’s
Healthcare System through Choice and Competition.” https://www.hhs.gov/
sites/default/files/Reforming-Americas-Healthcare-System-Through-Choiceand-Competition.pdf.
———. 2019a. “HHS Announces Quality Summit to Streamline and Improve Quality
Programs across Government.” https://www.hhs.gov/about/
news/2019/07/09/hhs-announces-quality-summit-streamline-improvequality-programs-government.html.
———. 2019b. “HHS Proposes Stark Law and Anti-Kickback Statute Reforms to Support
Value-Based and Coordinated Care.” https://www.hhs.gov/about/
news/2019/10/09/hhs-proposes-stark-law-anti-kickback-statute-reforms.
html.
———. 2019c. “Statement on Continued Progress Enhancing Patient Access to HighQuality, Low-Cost Generic Drugs.” https://www.fda.gov/news-events/
press-announcements/
statement-continued-progress-enhancing-patient-access-high-quality-lowcost-generic-drugs.
IRS (Internal Revenue Service), Employee Benefits Security Administration, and HHS
(Department of Health and Human Services). 2019. “Health Reimbursement
Arrangements and Other Account-Based Group Health Plans.” Federal
Register 84, no. 12571: 28888–9027.
Kessler, D., and M. McClellan. 2000. “Is Hospital Competition Socially Wasteful?”
Quarterly Journal of Economics 115, no. 2: 577–615.
KFF (Kaiser Family Foundation). 2018. “Employer Health Benefits Survey.” https://www.
kff.org/
report-section/2018-employer-health-benefits-survey-section-8-highdeductible-health-plans-with-savings-option/.
Moriya, A., Vogt, W., and Gaynor, M. 2010. Hospital Prices and Market Structure in the
Hospital and Insurance Industries. Health Economics, Policy and Law, 5, no. 4:
459-479.
Mulligan, C. 2015. The Economic Consequences of Health-Care Reform. Chicago:
University of Chicago Press.
NACDD (National Association of Chronic Disease Directors). No date. “Why Public
Health Is Necessary to Improve Healthcare.” https://cdn.ymaws.com/www.
chronicdisease.org/resource/resmgr/white_papers/cd_white_paper_
hoffman.pdf.
National Cancer Institute. No date. “Cancer Stat Facts: Cancer of Any Site.” https://seer.
cancer.gov/statfacts/html/all.html.
NBGH (National Business Group on Health). 2018. “Large U.S. Employers Eye Changes
to Health Care Delivery System as Cost to Provide Health Benefits Nears

References | 321

250-840_text_.pdf 325

2/7/20 3:46 PM

$15,000 per Employee.” Press release. https://www.businessgrouphealth.org/
news/nbgh-news/press-releases/press-release-details/?ID=348.
NCCI (National Council on Compensation Insurance). 2018. “The Impact of Hospital
Consolidation on Medical Costs.” https://www.ncci.com/Articles/Pages/II_
Insights_QEB_Impact-of-Hospital-Consolidation-on-Medical-Costs.aspx.
Newman, D., S. Parente, E. Barette, and K. Kennedy. 2016. “Prices for Common Medical
Services Vary Substantially Among the Commercially Insured.” Health Affairs
35, no. 5: 923–27.
OMB (Office of Management and Budget). 2019. “Combatting the Opioid Epidemic 2020
Budget Fact Sheet.” https://www.whitehouse.gov/wp-content/
uploads/2019/03/FY20-Fact-Sheet_Combatting-the-Opioid-Epidemic_FINAL.
pdf.
Parente, S., and R. Feldman. 2013. “Micro-Simulation of Private Health Insurance and
Medicaid Take-Up Following the U.S. Supreme Court Decision Upholding the
Affordable Care Act.” Health Services Research 48, no. 2: 826–49.
Parente, S., R. Feldman, J. Spetz, B. Dowd, and E. Baggett. 2017. “Wage Growth for the
Health Care Workforce: Projecting the Affordable Care Act Impact.” Health
Services Research 52, no. 2: 741–62.
Parente, S., R. Feldman, and X. Yu. 2010. “Impact of Full Replacement with Consumer
Driven Health Plan on Health Care Cost and Use of Preventive Services.”
Insurance Markets and Companies: Analyses and Actuarial Computations 1, no.
1: 56–66.
Sood, N., T. Shih, K. Van Nuys, and D. Goldman. 2017. “The Flow of Money through the
Pharmaceutical Distribution System.” Leonard Schaeffer Center for Health
Policy and Economics, University of Southern California. https://healthpolicy.
usc.edu/wp-content/uploads/2017/06/USC_Flow-of-MoneyWhitePaper_
Final_Spreads.pdf.
Tenn, S. 2011. “The Price Effects of Hospital Mergers: A Case Study of the SutterSummit
Transaction.” International Journal of the Economics of Business 18, no. 1:
65–82.
Unger, J., R. Vaidya, D. Hershman, L. Minasian, and M. Fleury. 2019. “Systematic Review
and Meta-Analysis of the Magnitude of Structural, Clinical, and Physician and
Patient Barriers to Cancer Clinical Trial Participation.” Journal of the National
Cancer Institute 111, no. 3: 245–55.
VBID Health (Value-Based Insurance Design Health). 2019. “Uptake and Federal
Budgetary Impact of Allowing Health Savings Account-Eligible High
Deductible Health Plans to Cover Chronic Disease Drugs and Services
Pre-Deductible.” http://vbidhealth.com/docs/NPC%20HDHP+%20FINAL%20
05-30-2019.pdf.
Vogt, W., and R. Town. 2006. “How Has Hospital Consolidation Affected the Price and
Quality of Hospital Care?” Research Synthesis Report, no. 9.

322 | References

250-840_text_.pdf 326

2/7/20 3:46 PM

Waters, H., and M. Graf. 2018. “The Cost of Chronic Diseases in the U.S.” Milken
Institute. https://assets1b.milkeninstitute.org/assets/Publication/Viewpoint/
PDF/Chronic-Disease-Executive-Summary-r2.pdf.
White House. 2019a. “President Donald J. Trump’s Healthcare Agenda Puts Seniors and
American Patients First.” https://www.whitehouse.gov/briefings-statements/
president-donald-j-trumps-healthcare-agenda-puts-seniors-americanpatients-first/.
———. 2019b. “Executive Order on Improving Price and Quality Transparency in
American Healthcare to Put Patients First.” https://www.whitehouse.gov/
presidential-actions/
executive-order-improving-price-quality-transparency-american-healthcareput-patients-first/.
———. 2019c. “Presidential Message on National HIV Testing Day, 2019.” https://www.
whitehouse.gov/briefings-statements/
presidential-message-national-hiv-testing-day-2019/.
———. 2019d. “President Donald J. Trump Is Taking a Bold New Approach to Advance
Kidney Health.” https://www.whitehouse.gov/briefings-statements/
president-donald-j-trump-taking-bold-new-approach-advance-kidneyhealth/.
Willis Towers Watson. 2013. “NBGH Employer Survey on Value in Purchasing Health
Care.” https://www.towerswatson.com/en/Insights/IC-Types/SurveyResearch-Results/2013/03/
Towers-Watson-NBGH-Employer-Survey-on-Value-in-Purchasing-Health-Care.

Chapter 6
Ashenfelter, O., D. Hosken, and M. Weinberg. 2015. “Efficiencies Brewed: Pricing and
Consolidation in the US Beer Industry.” RAND Journal of Economics 46, no. 2:
328–61.
Autor, D., D. Dorn, L. Katz, C. Patterson, and J. Reenen. 2019. The Fall of the Labor Share
and the Rise of Superstar Firms. NBER Working Paper 23396. Cambridge, MA:
National Bureau of Economic Research. https://www.nber.org/papers/
w23396.pdf; forthcoming in Quarterly Journal of Economics, https://
economics.mit.edu/files/12979.
Baker, J. 2015. “Taking the Error Out of ‘Error Cost’ Analysis: What’s Wrong with
Antitrust’s Right.” Antitrust Law Journal 80, no. 1: 1–38.
Basu, S. 2019. “Are Price-Cost Markups Rising in the United States? A Discussion of the
Evidence.” Journal of Economic Perspectives 33, no. 3: 3–22.
Berry, S., M. Carnall, and P. Spiller. 2006. “Airline Hubs: Costs, Markups and the
Implications of Customer Heterogeneity.” In Competition Policy and Antitrust,
Advances in Airline Economics 1, edited by D. Lee. Bingley, U.K.: Emerald
Group.

References | 323

250-840_text_.pdf 327

2/7/20 3:46 PM

Berry, S., M. Gaynor, and F. Morton. 2019. “Do Increasing Markups Matter? Lessons from
Empirical Industrial Organization.” Journal of Economic Perspectives 33, no. 3:
44–68. https://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.33.3.44.
Bloomberg. 2019. “FTC Chief Says Willing to Break Up Companies Amid Big Tech
Probe.” https://www.bloomberg.com/news/articles/2019-08-13/
ftc-chief-says-willing-to-break-up-companies-amid-big-tech-probe.
Borenstein, S. 2011. “Why Can’t U.S. Airlines Make Money?” American Economic Review:
Papers & Proceedings 101, no. 3: 233–57. https://pubs.aeaweb.org/doi/
pdfplus/10.1257/aer.101.3.233.
Bourne, R. 2019. “Is This Time Different? Schumpeter, the Tech Giants, and Monopoly
Fatalism.” CATO Institute. https://www.cato.org/publications/policy-analysis/
time-different-schumpeter-tech-giants-monopoly-fatalism.
Bresnahan, T. 1989. “Empirical Studies of Industries with Market Power.” In Handbook
of Industrial Organization 2, edited by R. Schmalensee and R. Willig.
Amsterdam: Elsevier Science.
CEA (Council of Economic Advisers). 2016. “Benefits of Competition and Indicators of
Market Power.” https://obamawhitehouse.archives.gov/sites/default/files/
page/files/20160414_cea_competition_issue_brief.pdf.
———. 2018a. “Reforming Biopharmaceutical Pricing at Home and Abroad.” https://
www.whitehouse.gov/wp-content/uploads/2017/11/CEA-Rx-White-PaperFinal2.pdf.
———. 2018b. “The Administration’s FDA Reforms and Reduced Biopharmaceutical
Drug Prices.” https://www.whitehouse.gov/wp-content/uploads/2018/10/
The-Administrations-FDA-Reforms-and-Reduced-Biopharmaceutical-DrugPrices.pdf.
Chugh, R., N. Goldstein, E. Lewis, J. Lien, D. Minehart, and N. Rose. 2016. “Economics at
the Antitrust Division 2015–2016: Household Appliances, Oil Field Services,
and Airport Slots.” Review of Industrial Organization 49, no. 4: 535–56. https://
www.justice.gov/atr/public-documents/rio-annual-review-articles.
Crémer, J., Y. Montjoye, and H. Schweitzer. 2019. Competition Policy for the Digital Era.
Brussels: European Commission. https://ec.europa.eu/competition/
publications/reports/kd0419345enn.pdf.
Crouzet, N., and J. Eberly. 2019. “Understanding Weak Capital Investment: The Role of
Market Concentration and Intangibles.” Paper prepared for Jackson Hole
Economic Policy Symposium. https://www.kellogg.northwestern.edu/faculty/
crouzet/html/papers/IntangiblesInvestmentConcentration_latest.pdf.
Cunningham, C., F. Ederer, and S. Ma. 2019. “Killer Acquisitions.” Working paper.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3241707.
Demsetz, H. 1973. “Industry Structure, Market Rivalry, and Public Policy.” Journal of
Law and Economics 16, no. 1: 1–9.
Digital Competition Expert Panel. 2019a. “Public Responses to Call for Evidence from
Individuals.” Chaired by J. Furman. https://assets.publishing.service.gov.uk/

324 | References

250-840_text_.pdf 328

2/7/20 3:46 PM

government/uploads/system/uploads/attachment_data/file/785548/DCEP_
Public_responses_to_call_for_evidence_from_individuals.pdf.
———. 2019b. “Public Responses to Call for Evidence from Organisations.” https://
assets.publishing.service.gov.uk/government/uploads/system/uploads/
attachment_data/file/785549/DCEP_Public_responses_to_call_for_
evidence_from_organisations.pdf.
——— . 2019c. “Unlocking Digital Competition.” https://assets.publishing.service.gov.
uk/government/uploads/system/uploads/attachment_data/file/785547/
unlocking_digital_competition_furman_review_web.pdf.
DOJ (U.S. Department of Justice). 2008. United States v. The Thomson Corp. and Reuters
Group PLC. Competitive Impact Statement, case 1:08, civil action 00262.
https://www.justice.gov/atr/case-document/
competitive-impact-statement-207.
———. 2011a. United States v. George’s Foods, LLC; George’s Family Farms, LLC; and
George’s Inc. Competitive Impact Statement, case 5:11, civil action 00043.
https://www.justice.gov/atr/case-document/file/497376/download.
———. 2011b. United States v. George’s Foods, LLC; George’s Family Farms, LLC; and
George’s Inc. Complaint, case 5:11, civil action 00043. https://www.justice.
gov/atr/case-document/file/497411/download.
———. 2016. United States v. Halliburton Co, LLC; and Baker Hughes Inc. Complaint, case
1:16, civil action 00233. https://www.justice.gov/atr/file/838661/download.
———. 2018a. “The Walt Disney Company Required to Divest Twenty-Two Regional
Sports Networks in Order to Complete Acquisition of Certain Assets from
Twenty-First Century Fox.” https://www.justice.gov/opa/pr/
walt-disney-company-required-divest-twenty-two-regional-sports-networksorder-complete.
———. 2018b. United States v. The Walt Disney Company and Twenty-First Century Fox,
Inc. Case 1:18, civil action 5800. https://www.justice.gov/atr/case-document/
file/1075201/download.
———. 2019. “Justice Department Reviewing Practices of Market Leading Online
Platforms.” https://www.justice.gov/opa/pr/
justice-department-reviewing-practices-market-leading-online-platforms.
DOJ and FTC (Department of Justice and Federal Trade Commission). 2010. “Horizontal
Merger Guidelines.” https://www.justice.gov/sites/default/files/atr/
legacy/2010/08/19/hmg-2010.pdf.
———. 2017. “Antitrust Guidelines for the Licensing of Intellectual Property.” https://
www.justice.gov/atr/IPguidelines/download.
———. 2019a. “Hart-Scott Rodino Annual Report: Fiscal Year 2018.” Annual
Competition, 41. https://www.ftc.gov/system/files/documents/reports/
federal-trade-commission-bureau-competition-department-justice-antitrustdivision-hart-scott-rodino/fy18hsrreport.pdf.
———. 2019b. “Joint Letter from the FTC and DOJ to Alaska State Senate.” https://www.
justice.gov/atr/page/file/1146346/download.

References | 325

250-840_text_.pdf 329

2/7/20 3:46 PM

———. 2019c. “Joint Letter from the FTC and DOJ to Nebraska Legislature.” https://
www.justice.gov/atr/page/file/1146236/download.
———. 2019d. “Joint Letter from the FTC and DOJ to Tennessee House of
Representatives.” https://www.justice.gov/atr/page/file/1146241/download.
European Commission. 2019. “Platform-to-Business Trading Practices.” https://
ec.europa.eu/digital-single-market/en/
platforms-to-business-trading-practices.
Faccio, M., and L. Zingales. 2018. Political Determinants of Competition in the Mobile
Telecommunication Industry. Finance Working Paper 494/2016. Brussels:
European Corporate Governance Institute.
FTC (Federal Trade Commission). 2018a. “Options to Enhance Occupational License
Portability.” Policy Perspectives. https://www.ftc.gov/system/files/
documents/reports/options-enhance-occupational-license-portability/
license_portability_policy_paper.pdf.
———. 2018b. “United States of America before the Federal Trade Commission.” Docket
9382. https://www.ftc.gov/system/files/documents/cases/docket_no_9382_
cdk_automate_part_3_complaint_redacted_public_version_0.pdf.
———. 2019a. “FTC Staff Issues FY 2016 Report on Branded Drug Firms’ Patent
Settlements with Generic Competitors.” https://www.ftc.gov/news-events/
press-releases/2019/05/
ftc-staff-issues-fy-2016-report-branded-drug-firms-patent.
———. 2019b. “Hearings on Competition and Consumer Protection in the 21st Century.”
https://www.ftc.gov/policy/hearings-competition-consumer-protection.
———. 2019c. “State-Based Initiatives: Selected Examples.” https://www.ftc.gov/policy/
advocacy/economic-liberty/state-based-initiatives.
———. 2019d. “The Antitrust Laws.” https://www.ftc.gov/tips-advice/competitionguidance/guide-antitrust-laws/antitrust-laws.
Furman, J. 2018. “Market Concentration: Note by Jason Furman.” Hearing on Market
Concentration, Prepared Testimony, Organization for Economic Cooperation
and Development. https://one.oecd.org/document/DAF/COMP/WD(2018)67/
en/pdf.
Furman, J., and P. Orszag. 2018. “A Firm-Level Perspective on the Role of Rents in the
Rise of Inequality.” In Towards a Just Society: Joseph Stiglitz and Twenty-First
Century Economics, edited by Martin Guzman. New York: Columbia University
Press.
Gee, E., C. Peters, and J. Wilder. 2019. “The Year in Review: Economics at the Antitrust
Division, 2018–2019.” Review of Industrial Organization 55, no. 4: 537-550.
Grullon, G., Y. Larkin, and R. Michaely. 2019. “Are U.S. Industries Becoming More
Concentrated?” Review of Finance 23, no. 4: 697–743. https://academic.oup.
com/rof/article/23/4/697/5477414.
Gutiérrez, G., and T. Philippon. 2017a. Declining Competition and Investment in the U.S.
NBER Working Paper 23583. Cambridge, MA: National Bureau of Economic
Research. https://www.nber.org/papers/w23583.pdf.

326 |

250-840_text_.pdf 330

References

2/7/20 3:46 PM

———. 2017b. Investment-Less Growth: An Empirical Investigation. NBER Working Paper
22897. Cambridge, MA: National Bureau of Economic Research. https://www.
nber.org/papers/w22897.pdf.
———. 2019. How EU Markets Became More Competitive than U.S. Markets: A Study of
Institutional Drift. NBER Working Paper 24700. Cambridge, MA: National
Bureau of Economic Research. https://www.nber.org/papers/w24700.pdf.
Heyer, K., C. Shapiro, and J. Wilder. 2009. “The Year in Review: Economics at the
Antitrust Division, 2008–2009.” Review of Industrial Organization 35: 349.
https://doi.org/10.1007/s11151-009-9232-1.
Igami, M., and K. Uetake. 2019. “Mergers, Innovation, and Entry-Exit Dynamics:
Consolidation of the Hard Disk Drive Industry, 1996–2016.” Review of
Economic Studies, forthcoming.
Jia, J., G. Jin, and L. Wagman. 2019. “GDPR and the Localness of Venture Investment.”
Working paper. http://dx.doi.org/10.2139/ssrn.3436535.
Krugman, P. 2016. “Robber Baron Recessions.” New York Times, April 18. https://www.
nytimes.com/2016/04/18/opinion/robber-baron-recessions.html.
Kwoka, J. 2015. Mergers, Merger Control, and Remedies: A Retrospective Analysis of U.S.
Policy. Cambridge, MA: MIT Press.
Lambrecht, A., and C. Tucker. 2015. “Can Big Data Protect a Firm from Competition?”
http://ec.europa.eu/information_society/newsroom/image/
document/2016-6/computer_and_communications_industry_association_-_
can_big_data_protect_a_firm_from_competition_13846.pdf.
Lamoreaux, N. 2019. “The Problem of Bigness: From Standard Oil to Google.” Journal of
Economic Perspectives 33, no. 3: 94–117. https://pubs.aeaweb.org/doi/
pdfplus/10.1257/jep.33.3.94.
Lee, R. 2013. “Vertical Integration and Exclusivity in Platform and Two-Sided Markets.”
American Economic Review 103, no. 7: 2960–3000. http://www.people.fas.
harvard.edu/~robinlee/papers/VIExclusivity.pdf.
Mahnke, R. 2015. “Big Data as a Barrier to Entry.” Competition Policy International.
https://www.competitionpolicyinternational.com/
big-data-as-a-barrier-to-entry/.
Marshall, G., and A. Parra. 2019. “Innovation and Competition: The Role of the Product
Market.” International Journal of Industrial Organization 65: 221–47.
Miller, N., and M. Weinberg. 2017. “Understanding the Pricing Effects of the MillerCoors
Joint Venture.” Econometrica 85, no. 6: 1763–91.
Ohlhausen, M. 2019. “Written Statement of Proposed Testimony to U.S. House of
Representatives, Hearing on ‘Online Platforms and Market Power Part 2:
Innovation and Entrepreneurship.’” https://docs.house.gov/meetings/JU/
JU05/20190716/109793/HHRG-116-JU05-Wstate-OhlhausenM-20190716.pdf
https://docs.house.gov/meetings/JU/JU05/20190716/109793/HHRG-116JU05-Wstate-OhlhausenM-20190716.pdf.

References | 327

250-840_text_.pdf 331

2/7/20 3:46 PM

Pai, A. 2013. “Remarks of the FCC Commissioner Ajit Pai at TechFreedom’s Forum on
the 100th Anniversary of the Kingsbury Commitment.” https://www.fcc.gov/
document/pai-remarks-100th-anniversary-kingsbury-commitment.
Peltzman, S. 2014. “Industrial Concentration under the Rule of Reason.” Journal of Law
and Economics 57, no. S3: 101–20.
Philippon, T. 2019. The Great Reversal: How American Gave Up on Free Markets.
Cambridge, MA: Belknap Press of Harvard University Press.
Rossi-Hansberg, E., P. Sarte, and N. Trachter. 2019. Diverging Trends in National and
Local Concentration. NBER Working Paper 25066. https://www.princeton.
edu/~erossi/DTNLC.pdf.
Rubinfeld, D., and M. Gal. 2017. “Access Barriers to Big Data.” Arizona Law Review 59,
no. 2: 339–81. http://arizonalawreview.org/pdf/59-2/59arizlrev339.pdf.
Shapiro, C. 2010. “The 2010 Horizontal Merger Guidelines: From Hedgehog to Fox in
Forty Years.” Antitrust Law Journal 77, no. 1: 49–108. https://heinonline.org/
HOL/Page?handle=hein.journals/antil77&id=53&collection=journals&index=#.
———. 2018. “Antitrust in a Time of Populism.” International Journal of Industrial
Organization 61: 714–48.
———. 2019. “Protecting Competition in the American Economy: Merger Control, Tech
Titans, Labor Markets.” Journal of Economic Perspectives 33, no. 3: 69–93.
https://www.aeaweb.org/articles?id=10.1257/jep.33.3.69.
Stigler Committee on Digital Platforms. 2019. Final Report. Chicago: Booth School of
Business at University of Chicago. https://research.chicagobooth.edu/-/
media/research/stigler/pdfs/digital-platforms---committee-report---stiglercenter.pdf.
Syverson, C. 2019. “Macroeconomics and Market Power: Context, Implications, and
Open Questions.” Journal of Economic Perspectives 33, no. 3: 23–43. https://
faculty.chicagobooth.edu/chad.syverson/research/macromarketpowerjep.
pdf.
Economist. 2016. “Too Much of a Good Thing.” March 26. https://www.economist.com/
briefing/2016/03/26/too-much-of-a-good-thing.
U.S. Congress. 2019a. “S.307, 116th Congress (2019–2020).” https://www.congress.gov/
bill/116th-congress/senate-bill/307/text.
———. 2019b. “S.2658, 116th Congress (2019–2020).” https://www.congress.gov/
bill/116th-congress/senate-bill/2658/text?r=2&s=1.
U.S. House, Committee on the Judiciary. 2019a. “Online Platforms and Market Power,
Part 1: The Free and Diverse Press.” June 11. https://judiciary.house.gov/
legislation/hearings/
online-platforms-and-market-power-part-1-free-and-diverse-press.
———. 2019b. “Online Platforms and Market Power, Part 2: Innovation and
Entrepreneurship.” July 16. https://judiciary.house.gov/legislation/hearings/
online-platforms-and-market-power-part-2-innovation-andentrepreneurship.

328 | References

250-840_text_.pdf 332

2/7/20 3:46 PM

———. 2019c. “Online Platforms and Market Power, Part 3: The Role of Data and Privacy
in Competition.” September 12. https://judiciary.house.gov/legislation/
hearings/
online-platforms-and-market-power-part-3-role-data-and-privacycompetition-0.
U.S. Senate, Committee on the Judiciary, Subcommittee on Antitrust, Competition
Policy, and Consumer Rights. 2019. “Competition in Digital Technology
Markets: Examining Acquisitions of Nascent or Potential Competitors by
Digital Platforms.” September 24. https://www.judiciary.senate.gov/
meetings/
competition-in-digital-technology-markets-examining-acquisitions-ofnascent-or-potential-competitors-by-digital-platforms.
Werden, G. 2001. “Network Effects and Conditions of Entry: Lessons from the Microsoft
Case.” Antitrust Law Journal 69, no. 1: 87–112. https://heinonline.org/HOL/
Page?collection=journals&handle=hein.journals/
antil69&id=95&men_tab=srchresults#.
Werden, G., and L. Froeb. 2018. “Don’t Panic: A Guide to Claims of Increasing
Concentration.” Antitrust Magazine 33, no. 1: 74–79.
Wessel, D. 2018. “Is Lack of Competition Strangling the U.S. Economy?” Harvard
Business Review, March. https://hbr.org/2018/03/
is-lack-of-competition-strangling-the-u-s-economy.
White House. 2016. “Executive Order: Steps to Increase Competition and Better Inform
Consumers and Workers to Support Continued Growth of the American
Economy.” https://obamawhitehouse.archives.gov/the-pressoffice/2016/04/15/
executive-order-steps-increase-competition-and-better-inform-consumers.
Wilson, C. 2019. “Welfare Standards Underlying Antitrust Enforcement: What You
Measure is What You Get.” Luncheon keynote address at George Mason Law
Review 22nd annual Antitrust Symposium. https://www.ftc.gov/es/system/
files/documents/public_statements/1455663/welfare_standard_speech_-_
cmr-wilson.pdf.
Wu, T. 2018. The Curse of Bigness: Antitrust in the New Gilded Age. New York: Columbia
Global Reports.
Yun, J. 2019. “Competition in Digital Technology Markets: Examining Acquisitions of
Nascent or Potential Competitors by Digital Platforms.” Prepared statement
for U.S. Senate. https://www.judiciary.senate.gov/imo/media/doc/Yun%20
Testimony.pdf.

Chapter 7
Ahmad, F., L. Rossen, M. Spencer, M. Warner, and P. Sutton. 2019. “Provisional Drug
Overdose Death Counts.” Centers for Disease Control and Prevention,
National Center for Health Statistics. https://www.cdc.gov/nchs/nvss/vsrr/
drug-overdose-data.htm.

References | 329

250-840_text_.pdf 333

2/7/20 3:46 PM

Alpert, A., D. Powell, and R. Pacula. 2018. “Supply-Side Drug Policy in the Presence of
Substitutes: Evidence from the Introduction of Abuse-Deterrent Opioids.”
American Economic Journal: Economic Policy 10, no. 4: 1–35.
ASTHO (Association of State and Territorial Health Officials). 2018. “2018 State
Legislation Addressing Naloxone.” http://www.astho.org/
StatePublicHealth/2018-State-Legislation-Addressing-Naloxone/04-05-18/.
Autor, D. 2015. “The Unsustainable Rise of the Disability Rolls in the United States:
Causes, Consequences, and Policy Options.” In Social Policies in an Age of
Austerity: A Comparative Analysis of the U.S. and Korea, edited by J. Scholz, H.
Moon, and S. Lee. Northampton, MA: Edward Elgar.
Becker, G. 1962. “Irrational Behavior and Economic Theory.” Journal of Political
Economy 70, no. 1: 1–13.
Becker, G., and K. Murphy. 1988. “A Theory of Rational Addiction.” Journal of Political
Economy 96, no. 4: 675–700.
Besley, T. 1988. “Optimal Reimbursement Health Insurance and the Theory of Ramsey
Taxation.” Journal of Health Economics 7, no. 4: 321–36.
Bretteville-Jensen, A. 2006. “Drug Demand—Initiation, Continuation and Quitting.” De
Economist 154, no. 4: 491–516.
Bretteville-Jensen, A., and E. Biorn. 2003. “Heroin Consumption, Prices and Addiction:
Evidence from Self-Reported Panel Data.” Scandinavian Journal of Economics
105, no. 4: 661–79.
Buchmueller, T., and C. Carey. 2018. “The Effect of Prescription Drug Monitoring
Programs on Opioid Utilization in Medicare.” American Economic Journal:
Economic Policy 10, no. 1: 77–112.
Cardinal Health Inc. 2019. “Distributors Reach Settlement with Ohio Counties in
Opioids Case.” Press release, October 21. https://ir.cardinalhealth.com/news/
press-release-details/2019/Distributors-Reach-Settlement-With-OhioCounties-in-Opioids-Case/default.aspx.
Carey, C., A. Jena, and M. Barnett. 2018. “Patterns of Potential Opioid Misuse and
Subsequent Adverse Outcomes in Medicare, 2008 to 2012.” Annals of Internal
Medicine 168, no. 12: 837
Case, A., and A. Deaton, A. 2017. “Mortality and Morbidity in the 21st Century.”
Brookings Papers on Economic Activity, Spring.
Caulkins, J. 1995. “Estimating Elasticities of Demand for Cocaine and Heroin with Data
from the Drug Use Forecasting System.” Final Report, NCJ 173823. U.S.
Department of Justice, National Institute of Justice, Washington.
CBP (U.S. Customs and Border Protection). 2019. “CBP Enforcement Statistics FY2019.”
https://www.cbp.gov/newsroom/stats/cbp-enforcement-statistics-fy2019.
CDC (Centers for Disease Control and Prevention). 2018. “Opioid Overdose:
Understanding the Epidemic.” https://www.cdc.gov/drugoverdose/epidemic/
index.html.

330 | References

250-840_text_.pdf 334

2/7/20 3:46 PM

———. 2019. “Drug and Opioid-Involved Overdose Deaths—United States, 2013–2017.”
https://www.cdc.gov/mmwr/volumes/67/wr/mm675152e1.htm.
———. No date. “Calculating Total Daily Dose of Opioids for Safer Dosage.” U.S.
Department of Health and Human Services, Washington. https://www.cdc.
gov/drugoverdose/pdf/calculating_total_daily_dose-a.pdf.
CEA (Council of Economic Advisers). 2017. “The Underestimated Cost of the Opioid
Crisis.” Executive Office of the President, the White House.
———. 2019a. “The Full Cost of the Opioid Crisis: $2.5 Trillion Over Four Years.” https://
www.whitehouse.gov/articles/full-cost-opioid-crisis-2-5-trillion-four-years/.
———. 2019b. “The Role of Opioid Prices in the Evolving Opioid Crisis.” https://www.
whitehouse.gov/wp-content/uploads/2019/04/The-Role-of-Opioid-Prices-inthe-Evolving-Opioid-Crisis.pdf.
Chalmers J., D. Bradford, and C. Jones. 2010. “The Effect of Methamphetamine and
Heroin Price on Polydrug Use: A Behavioral Economics Analysis in Sydney,
Australia.” International Journal of Drug Policy 21, no. 5: 381–89.
Cher, B., N. Morden, and E. Meara. 2019. “Medicaid Expansion and Prescription Trends:
Opioids, Addiction Therapies, and Other Drugs.” Medical Care 57, no. 3:
208–12.
Chermak, J., K. Krause, D. Brookshire, and H. Burness. 2013. “Moving Forward by
Looking Back: Comparing Laboratory Results with Ex Ante Market Data.”
Economic Inquiry 51, no. 1: 1035–49.
Chicago Tribune. 2016. “Chemical Weapon for Sale: A Synthetic Opioid from China That
Will Kill Heroin Users.” October 7.
Ciccarone, D., G. Unick, and A. Kraus. 2009. “Impact of South American Heroin on the
U.S. Heroin Market 1993–2004.” International Journal of Drug Policy 20, no. 5:
392–401.
CRS (Congressional Research Service). 2018. “Prescription Drug Monitoring Programs.”
https://fas.org/sgp/crs/misc/R42593.pdf.
Currie, J., J. Yin, and M. Schnell. 2018. “U.S. Employment and Opioids: Is There a
Connection? NBER Working Paper 24440. Cambridge, MA: National Bureau of
Economic Research.
Dave, D. 2008. “Illicit Drug Use Among Arrestees, Prices and Policy.” Journal of Urban
Economics 63: 694–714.
Dave, D., A. Grecu, and H. Saffer. 2018. “Mandatory Access Prescription Drug Monitoring
Programs and Prescription Drug Abuse.” Journal of Policy Analysis and
Management 38, no. 1: 181–209.
DeBruyne, N. 2017. “American War and Military Operations Casualties: Lists and
Statistics.” Congressional Research Service, April 26.
DOJ (U.S. Department of Justice). 2018. “Two Chinese Nationals Charged with
Operating Global Opioid and Drug Manufacturing Conspiracy Resulting in
Deaths.” https://www.justice.gov/opa/pr/

References | 331

250-840_text_.pdf 335

2/7/20 3:46 PM

two-chinese-nationals-charged-operating-global-opioid-and-drugmanufacturing-conspiracy.
———. No date. “Automated Reports and Consolidated Ordering System.” https://www.
deadiversion.usdoj.gov/arcos/retail_drug_summary/.
DOJ and DEA (U.S. Department of Justice and Drug Enforcement Administration).
2017a. “2017 National Drug Threat Assessment.” https://www.dea.gov/
documents/2017/10/01/2017-national-drug-threat-assessment.
———. 2017b. “NFLIS Brief: Fentanyl, 2001–2015.” Diversion Control Division, National
Forensic Laboratory Information System. https://www.nflis.deadiversion.
usdoj.gov/DesktopModules/ReportDownloads/Reports/
NFLISFentanylBrief2017.pdf.
DOT (U.S. Department of Transportation). 2016. “Guidance on Treatment of the
Economic Value of a Statistical Life (VSL) in U.S. Department of
Transportation Analyses – 2016 Adjustment.” https://www.transportation.
gov/sites/dot.gov/files/docs/2016%20Revised%20Value%20of%20a%20
Statistical%20Life%20Guidance.pdf.
Dowell, D., E. Arias, K. Kochanek, R. Anderson, G. Guy, J. Losby, and G. Baldwin. 2017.
“Contribution of Opioid-Involved Poisonings to the Change in Life Expectancy
in the United States, 2000–2015.” Journal of the American Medical Associaton
318, no. 11: 1065–67.
Drug Enforcement Administration. 2017. “2017 National Drug Threat Assessment.” U.S.
Department of Justice, Washington. https://www.dea.gov/docs/DIR-04017_2017-NDTA.pdf.
DuPont, S., A. Bezaitis, and M. Ross. 2015. “Stemming the Tide of Prescription Opioid
Overuse, Misuse, and Abuse.” Health Affairs Blog, September 22. https://www.
healthaffairs.org/do/10.1377/hblog20150922.050693/full/.
Eberstadt, N. 2017. “Our Miserable 21st Century.” Commentary, February. https://www.
commentarymagazine.com/articles/our-miserable-21st-century/.
Einav, L., A. Finkelstein, and M. Polyakova. 2018. “Private Provision of Social Insurance:
Drug-Specific Price Elasticities and Cost Sharing in Medicare Part D.”
American Economic Journal: Economic Policy 10, no. 3: 122–53.
Evans, W., E. Lieber, and P. Power. 2019. “How the Reforumlation of OxyContin Ignited
the Heroin Epidemic.”Review of Economics and Statistics 101, no. 1: 1–15.
FDA (Food and Drug Administration). 2018a. “FDA Analysis of Long-Term Trends in
Opioid Analgesic Products: Quantity, Sales, and Price Trends.” https://www.
fda.gov/downloads/aboutfda/reportsmanualsforms/reports/ucm598899.pdf.
———. 2018b. “National Drug Code Directory.” https://www.fda.gov/Drugs/
InformationOnDrugs/ucm142438.htm.
Feldstein, M. 1973. “The Welfare Loss of Excess Health Insurance.” Journal of Political
Economy 81, no. 2: 251–80.
Finkelstein, A., M. Gentzkow, and H. Williams. 2018. “What Drives Prescription Opioid
Abuse? Evidence from Migration.” Stanford Institute for Economic Policy

332 | References

250-840_text_.pdf 336

2/7/20 3:46 PM

Research Working Paper. https://siepr.stanford.edu/sites/default/files/
publications/18-028.pdf.
Fontenot, K., J. Semega, and M. Kollar. 2018. “Income and Poverty in the United
States.” Current Population Reports, U.S. Census Bureau, September.
Gallet, C. 2014. “Can Price Get the Monkey Off Our Back? A Meta-analysis of Illicit Drug
Demand.” Health Economics 23, no. 1: 55–68.
Glaeser, E., B. Sacerdote, and J. Scheinkman. 2003. “The Social Multiplier.” Journal of
the European Economic Association 1, nos. 2–3: 345–53.
Goodman-Bacon, A., and E. Sandoe. 2017. “Did Medicaid Expansion Cause the Opioid
Epidemic? There’s Little Evidence That It Did.” Health Affairs Blog, August 23.
https://www.healthaffairs.org/action/showDoPubSecure?doi=10.1377%2Fhbl
og20170823.061640&format=full.
Guo, J. 2014. “The Postal Service Is Losing Millions a Year to Help You Buy Cheap Stuff
from China.” Washington Post, September 12.
HHS (Department of Health and Human Services). 2014. “Current Status of Tobacco
Control.” https://www.surgeongeneral.gov/library/reports/50-years-ofprogress/sgr50-chap-14.pdf.
———. 2016. “Testimony Before the United States Senate Special Committee on Aging:
Opioid Use Among Seniors—Issues and Emerging Trends.” https://oig.hhs.
gov/testimony/docs/2016/maxwell0216.pdf.
Hill, S., S. Zuvekas, and M. Zodet. 2011. “Implications of the Accuracy of MEPS
Prescription Drug Data for Health Services Research.” Inquiry 48: 242-59.
Hoffman, J. 2019. “$260 Million Opioid Settlement Reached at Last Minute with Big
Drug Companies.” New York Times, October 21. https://www.nytimes.
com/2019/10/21/health/opioid-settlement.html.
Holahan, J., and A. Golsh. 2005. “Dual Eligibles: Medicaid Enrollment and Spending for
Medicare Beneficiaries.” Kaiser Commission on Medicaid and the Uninsured,
July.
Hollingsworth, A., C. Ruhm, and K. Simon. 2017. “Macroeconomic Conditions and
Opioid Use Disorder.” Journal of Health Economics 56: 222–33.
Jofre-Bonet, M., and N. Petry. 2008. “Trading Apples for Oranges? Results of an
Experiment on the Effects of Heroin and Cocaine Price Changes on Addicts’
Polydrug Use.” Journal of Economic Behavior & Organization 66, no. 2:
281–311.
KFF (Kaiser Family Foundation). 2006. “Dual Eligibles and Medicare Part D.” Kaiser
Commission on Medicaid and the Uninsured, May. https://
kaiserfamilyfoundation.files.wordpress.com/2013/01/7454.pdf.
———. 2018. “What’s in the Administration’s 5-Part Plan for Medicare Part D and What
Would it Mean for Beneficiaries and Program Savings?” https://www.kff.org/
medicare/issue-brief/
whats-in-the-administrations-5-part-plan-for-medicare-part-d-and-whatwould-it-mean-for-beneficiaries-and-program-savings/.

References | 333

250-840_text_.pdf 337

2/7/20 3:46 PM

Kochanek, K., E. Arias, and B. Bastian. 2016. “The Effect of Changes in Selected
Age-Specific Causes of Death in Non-Hispanic White Life Expectancy between
2000 and 2014.” NCHS Data Brief, no. 250.
Lipari, R., and A. Hughes. 2017. “How People Obtain the Prescription Pain Relievers
They Misuse.” CBHSQ Report, January 12.
Liu, J., J. Liub, J. Hammittc, and S. Choud. 1999. “The Price Elasticity of Opium in
Taiwan, 1914–1942.” Journal of Health Economics 18, no. 6: 795–810.
Max, M., M. Donovan, C. Miaskowski, S. Ward, D. Gordon, M. Bookbinder, C. Cleeland, N.
Coyle, M. Kiss, H. Thaler, N. Janjan, A. Anderson, S. Weinstein, and W.
Edwards. 1995. “Quality Improvement Guidelines for the Treatment of Acute
Pain and Cancer Pain.” Journal of the American Medical Association 274, no. 3:
1874–80.
Miroff, N. 2015. “Losing Marijuana Business, Mexican Cartels Push Heroin and Meth.”
Washington Post, January 11.
Morden, N., J. Munson, C. Colla, J. Skinner, J. Bynum, W. Zhou, and E. Meara. 2014.
“Prescription Opioid Use among Disabled Medicare Beneficiaries: Intensity,
Trends and Regional Variation.” Medical Care 52, no. 9: 852.
NIDA (National Institute on Drug Abuse). 2017. “Research on the Use and Misuse of
Fentanyl and Other Synthetic Opioids.” https://www.drugabuse.gov/aboutnida/legislative-activities/testimony-to-congress/2017/
research-use-misuse-fentanyl-other-synthetic-opioids.
Olmstead, T., S. Alessi, B. Kline, R. Pacula, and N. Petry. 2015. “The Price Elasticity of
Demand for Heroin: Matched Longitudinal and Experimental Evidence.”
Journal of Health Economics 41: 59–71.
ONDCP (Office of National Drug Control Policy). 2017. “National Drug Control Strategy,
Data Supplement 2016.” Executive Office of the President, the White House.
https://obamawhitehouse.archives.gov/sites/default/files/ondcp/policy-andresearch/2016_ndcs_data_supplement_20170110.pdf.
———. 2019. “21st Century Drug Trafficking Advisories on Fentanyl and Other
Synthetics.” Executive Office of the President, White House. https://www.
whitehouse.gov/wp-content/uploads/2019/08/White-House-FentanylAdvisories-Summary.pdf.
———. 2019. “National Drug Control Strategy.” Executive Office of the President, White
House. https://www.whitehouse.gov/wp-content/uploads/2019/01/NDCSFinal.pdf.
Pacula, R., B. Kilmer, M. Grossman, and F. Chaloupka. 2010. “Risks and Prices: The Role
of User Sanctions in Marijuana Markets.” B.E. Journal of Economic Analysis
and Policy 10, no. 1.
Petry, N., and W. Bickel. 1998. “Polydrug Abuse in Heroin Addicts: A Behavioral
Economic Analysis.” Addiction 93, no. 3: 321–35.
Pollak, R. 1970. “Habit Formation and Dynamic Demand Functions.” Journal of Political
Economy 78, no. 4: 745-63.

334 | References

250-840_text_.pdf 338

2/7/20 3:46 PM

Popper, N. 2017. “Opioid Dealers Embrace the Dark Web to Send Deadly Drugs by Mail.”
New York Times, June 10.
Powell, D., R. Pacula, and E. Taylor. 2017. “How Increasing Medical Access to Opioids
Contributes to the Opioid Epidemic: Evidence from Medicare Part D.” Paper
presented at 2017 meeting of National Tax Association.
President’s Commission on Combating Drug Addiction and the Opioid Crisis. 2017.
“Final Report.”
Purdue Pharma. 2019a. “Purdue Pharma Announces Agreement in Principle on
Landmark Opioid Litigation Settlement.” Press release, September 16.
https://www.purduepharma.com/news/2019/09/16/
purdue-pharma-announces-agreement-in-principle-on-landmark-opioidlitigation-settlement/.
———. 2019b. “Purdue Pharma Announces Landmark Agreement with Oklahoma to
Advance the Treatment of Addiction.” Press release, March 22. https://www.
purduepharma.com/news/2019/03/22/
purdue-pharma-announces-landmark-agreement-with-oklahoma-toadvance-the-treatment-of-addiction/.
Quinones, S. 2015. Dreamland: The True Tale of America’s Opiate Epidemic. New York:
Bloomsbury Press.
Roberts, A., and A. Skinner. 2014. “Assessing the Present State and Potential of
Medicaid Controlled Substance Lock-In Programs.” Journal of Managed Care
Pharmacy 20, no. 5: 439–46.
Roddy, J., and M. Greenwald. 2009. “An Economic Analysis of Income and Expenditures
by Heroin-Using Research Volunteers.” Substance Use & Misuse 44, no. 11:
1503–18.
Rosenberg, T. “When Is a Pain Doctor a Drug Pusher?” New York Times Magazine, June
17. https://www.nytimes.com/2007/06/17/magazine/17pain-t.html.
Rosenblum, D., J. Unick, and D. Ciccarone. 2014. “The Entry of Columbian-Sourced
Heroin into the U.S. Market: The Relationship between Competition, Price
and Purity.” International Journal of Drug Policy 25, no. 1: 88–95.
———. 2017. “An Instrumental Variables Approach to Estimating the Effects of Changes
in the Heroin Market on Overdose in the U.S.” Canadian Centre for Health
Economics Working Paper 170011.
Ruhm, C. 2019. “Drivers of the Fatal Drug Epidemic.” Journal of Health Economics 64:
25–42.
Saffer, H., and F. Chaloupka. 1999. “The Demand for Illicit Drugs.” Economic Inquiry 37:
401–11.
SAMHSA (Substance Abuse and Mental Health Service Administration). 2019. “Key
Substance Use and Mental Health Indicators in the United States: Results
from the 2018 National Survey on Drug Use and Health.” https://www.
samhsa.gov/data/report/2018-nsduh-annual-national-report.
Sarpatwari, A., M. Sinha, and A. Kesselheim. 2017. “The Opioid Epidemic: Fixing a
Broken Pharmaceutical Market.” Harvard Law and Policy Review, July. http://

References | 335

250-840_text_.pdf 339

2/7/20 3:46 PM

harvardlpr.com/wp-content/uploads/2017/07/SarpatwariSinhaKesselheim.
pdf.
Schneider, J., A. Anderson, and F. Tennant. 2009. “Patients Who Require Ultra-High
Opioid Doses.” Practical Pain Management 9, no. 7.
Schnell, M. 2017. “Physician Behavior in the Presence of a Secondary Market: The Case
of Prescription Opioids.” Working paper, November 17.
Silverman, L., and N. Spruill. 1977. “Urban Crime and the Price of Heroin.” Journal of
Urban Economics 4: 80–103.
Social Security Administration. “SSA State Agency Monthly Workload Data.” https://
www.ssa.gov/disability/data/ssa-sa-mowl.htm.
Soni, A. 2018. “Health Insurance, Price Changes, and the Demand for Pain Relief Drugs:
Evidence from Medicare Part D.” Unpublished paper, Indiana University.
Stanford Medicine. No date. “Descriptive Epidemiology: Epidemiology of Sporadic
ALS.” ALS Consortium of Epidemiologic Studies. http://aces.stanford.edu/
acesmem2/EpiDescSporadic.html.
Stiglitz, J. 2015. “When Inequality Kills.” Project Syndicate, December 7.
State of Oklahoma v. Purdue Pharma et al. 2019. https://int.nyt.com/data/
documenthelper/ 1660-oklahoma-opioid-trial-johnson-andjohnson/79f3fe55f5fa1a75bd48/optimized /full.pdf#page=1.
Teva Pharmaceutical Industries. 2019a. “Company Statement: Teva Reaches
Agreement with State of Oklahoma to Resolve State’s Claims against the
Company.” Press release, May 26. https://ir.tevapharm.com/investors/pressreleases/press-release-details/2019/
Teva-Settles-Track-1-Opioid-Cases-and-Reaches-Agreement-on-Frameworkto-Settle-Remaining-Litigation/default.aspx.
———. 2019b. “Teva Settles Track 1 Opioid Cases and Reaches Agreement on
Framework to Settle Remaining Litigation.” Press Release, October 21.
https://ir.tevapharm.com/investors/press-releases/press-releasedetails/2019/
Teva-Settles-Track-1-Opioid-Cases-and-Reaches-Agreement-on-Frameworkto-Settle-Remaining-Litigation/default.aspx.
U.S. Census Bureau. Various years. “Current Population Survey.” https://www.census.
gov/programs-surveys/cps.html.
U.S. Department of Health and Human Services. 2019. “Pain Management Best
Practices Inter-Agency Task Force Report: Updates, Gaps, Inconsistencies,
and Recommendations.” https://www.hhs.gov/sites/default/files/pmtf-finalreport-2019-05-23.pdf.
U.S. Department of Health and Human Services. “2014 CMS Statistics.” https://www.
cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-andReports/CMS-Statistics-Reference-Booklet/Downloads/CMS_Stats_2014_final.
pdf.

336 | References

250-840_text_.pdf 340

2/7/20 3:46 PM

U.S. Department of Health and Human Services, Agency for Healthcare Research and
Quality. Various years. “Medical Expenditure Panel Survey.” https://meps.
ahrq.gov/mepsweb/.
U.S. Department of Health and Human Services, Centers for Disease Control and
Prevention. Various years. “Multiple Cause of Death, 1999–2017, on CDC
WONDER Online Database.” http://wonder.cdc.gov/ucd-icd10.html.
U.S. Department of Labor, Bureau of Labor Statistics. 2018. “CPI Price Index for All
Urban Consumers (CPI-U).” https://www.bls.gov/cpi/tables/supplementalfiles/historical-cpi-u-201801.pdf.
U.S. Department of State. No date. “Programs to Fight Opioids.” https://www.state.
gov/j/inl/opioid/programs/index.htm.
U.S. Department of Transportation. 2016. “Guidance on Treatment of the Economic
Value of a Statistical Life (VSL) in U.S. Department of Transportation
Analyses—2016 Adjustment.” https://www.transportation.gov/sites/dot.gov/
files/docs/2016%20Revised%20Value%20of%20a%20Statistical%20Life%20
Guidance.pdf.
U.S. General Accounting Office. 2003. “Prescription Drugs: OxyContin Abuse and
Diversion and Efforts to Address the Problem.” GAO-04-110.
U.S. Senate Committee on Homeland Security and Governmental Affairs. 2018. “Drugs
for Dollars: How Medicaid Helps Fuel the Opioid Epidemic.” Majority Staff
Report.
VA (U.S. Department of Veterans Affairs). 2000. “Pain as the 5th Vital Sign Toolkit.”
October.
———. 2019. “Opioid Safety.” August 9. https://www.va.gov/PAINMANAGEMENT/Opioid_
Safety/index.asp.
Van Ours, J. 1995. “The Price Elasticity of Hard Drugs: The Case of Opium in the Dutch
East Indies, 1923–1938.” Journal of Political Economy 103, no. 2: 261–79.
Van Zee, A. 2009. “The Promotion and Marketing of OxyContin: Commercial Triumph,
Public Health Tragedy.” American Journal of Public Health 99, no. 2: 221–27.
Venkataramani, A., and P. Chaterjee. 2019. “Early Medicaid Expansions and Drug
Overdose Mortality in the USA: A Quasi-Experimental Analysis.” Journal of
General Internal Medicine 34, no. 1: 23–25.
White House. 2018. “President Donald J. Trump’s Initiative to Stop Opioid Abuse and
Reduce Drug Supply and Demand.” https://www.whitehouse.gov/briefingsstatements/
president-donald-j-trumps-initiative-stop-opioid-abuse-reduce-drug-supplydemand/.
Wilkie, R. 2018. “Fighting Pain and Addiction for Veterans.” White House, October 26.
https://www.whitehouse.gov/articles/fighting-pain-addiction-veterans/.
Wilner, M. 2019. “The ‘Fourth Wave’ of the Opioid Crisis: Feds Warn of a Rise in Meth
Use.” McClatchy DC, August 20. https://www.mcclatchydc.com/news/politicsgovernment/white-house/article234149267.html.

References | 337

250-840_text_.pdf 341

2/7/20 3:46 PM

Zhou, C., C. Florence, and D. Dowell. 2016. “Payments for Opioids Shifted Substantially
to Public and Private Insurers While Consumer Spending Declined, 1999–
2012.” Health Affairs 35, no. 5: 824–31.

Chapter 8
Albouy, D., G. Ehrlich, and Y. Liu. 2016. Housing Demand, Cost-of-Living Inequality, and
the Affordability Crisis. NBER Working Paper 22816. Cambridge, MA: National
Bureau of Economic Research. https://www.nber.org/papers/w22816.pdf.
Albouy, D., and G. Ehrlich. 2018. “Housing Productivity and the Social Cost of Land-Use
Restrictions.” Journal of Urban Economics 107: 101–20.
Beyer, S. 2016. “Washington, D.C., Reformed Its Zoning Code; Now Time to Ditch the
Height Limits.” Forbes, January 29. https://www.forbes.com/sites/
scottbeyer/2016/01/29/
washington-dc-reformed-its-zoning-code-now-time-to-ditch-the-heightlimits/#228f85814db8.
Carson, B. 2019. “Ben Carson: Trump Is Getting Our Housing Market Back on Track.”
Washington Post, June 27. https://www.washingtonpost.com/opinions/
ben-carson-our-affordable-housing-policy-is-failing-we-need-a-newapproach/2019/06/27/c5641dec-98fc-11e9-8d0a-5edd7e2025b1_story.html.
CEA (Council of Economic Advisers). 2016. “Housing Development Toolkit.” https://
www.whitehouse.gov/sites/whitehouse.gov/files/images/Housing_
Development_Toolkit%20f.2.pdf.
———. 2019. “The State of Homelessness in America.” https://www.whitehouse.gov/
wp-content/uploads/2019/09/The-State-of-Homelessness-in-America.pdf.
Chetty, R., J. Friedman, N. Hendren, M. Jones, and S. Porter. 2018. The Opportunity
Atlas: Mapping the Childhood Roots of Social Mobility. NBER Working Paper
25147. Cambridge, MA: National Bureau of Economic Research. https://www.
nber.org/papers/w25147.
CBO (U.S. Congressional Budget Office). 2019. “The Distribution of Household Income,
2016.” https://www.cbo.gov/publication/55413.
Corinth, K. 2017. “The Impact of Permanent Supportive Housing on Homeless
Populations.” Journal of Housing Economics 35: 69–84.
Davis, M., W. Larson, S. Oliner, and J. Shui. 2019. “The Price of Residential Land for
Counties, ZIP codes, and Census Tracts in the United States.” Federal Housing
Finance Agency Working Paper. https://www.fhfa.gov/
PolicyProgramsResearch/Research/PaperDocuments/wp1901.pdf.
Dettling, L., J. Hsu, L. Jacobs, K. Moore, J. Thompson, and E. Llanes. 2017. “Recent
Trends in Wealth-Holding by Race and Ethnicity: Evidence from the Survey of
Consumer Finances.” FEDS Notes. Washington: Board of Governors of the
Federal Reserve System. https://www.federalreserve.gov/econres/notes/fedsnotes/
recent-trends-in-wealth-holding-by-race-and-ethnicity-evidence-from-thesurvey-of-consumer-finances-20170927.htm.
338 | References

250-840_text_.pdf 342

2/7/20 3:46 PM

Diamond, R., T. McQuade, and F. Qian. 2019. “The Effects of Rent Control Expansion on
Tenants, Landlords, and Inequality: Evidence from San Francisco.” American
Economic Review 109, no. 9: 3365–94.
Dumont, A. 2019. “Housing Affordability in the U.S.: Trends by Geography, Tenure, and
Household Income.” FEDS Notes. Washington: Board of Governors of the
Federal Reserve System. https://www.federalreserve.gov/econres/notes/fedsnotes/
housing-affordability-in-the-us-trends-by-geography-tenure-and-householdincome-20190927.htm.
Eriksen, M., and S. Rosenthal. 2010. “Crowd Out Effects of Place-Based Subsidized
Rental Housing: New Evidence from the LIHTC Program.” Journal of Public
Economics 94: 953–66.
Eriksen, M., and A. Ross. 2015. “Housing Vouchers and the Price of Rental Housing.”
American Economic Journal 7, no. 3: 154–76.
Ganong, P., and D. Shoag. 2017. “Why Has Regional Income Convergence in the U.S.
Declined?” Journal of Urban Economics 102: 76–90.
Glaeser, E. 2009. “Green Cities, Brown Suburbs.” City Journal, February 16. https://
www.city-journal.org/html/green-cities-brown-suburbs-13143.html.
Glaeser, E., and J. Gyourko. 2008. Rethinking Federal Housing Policy: How to Make
Housing Plentiful and Affordable. Washington: AEI Press.
———. 2018. “The Economic Implications of Housing Supply.” Journal of Economic
Perspectives 32, no. 1: 3–30.
Glaeser, E., J. Gyourko, and R. Saks. 2005. “Why Have Housing Prices Gone Up?”
American Economic Review 95, no. 2: 329–33.
Glaeser, E., and B. Ward. 2009. “The Causes and Consequences of Land Use Regulation:
Evidence from Greater Boston.” Journal of Urban Economics 65: 265–78.
Gray, N. 2019. “Do Minimum Lot Size Rules Matter?” Strong Towns, June 20. https://
www.strongtowns.org/journal/2019/6/19/do-minimum-lot-size-rules-matter.
Gyourko, J., and R. Molloy. 2015. “Chapter 19: Regulation and Housing Supply.” In
Handbook of Regional and Urban Economics, vol. 5, edited by G. Duranton, J.
Henderson, and W. Strange. Amsterdam: Elsevier.
Gyourko, J., A. Saiz, and A. Summers. 2008. “A New Measure of the Local Regulatory
Environment for Housing Markets: The Wharton Residential Land Use
Regulatory Index.” Urban Studies 45, no. 3: 693–729.
Gyourko, J., J. Hartley, and J. Krimmel. 2019. The Local Residential Land Use Regulatory
Environment across U.S. Housing Markets: Evidence from a New Wharton Index.
NBER Working Paper 26573. Cambridge, MA: National Bureau of Economic
Research. https://www.nber.org/papers/w26573.pdf.
Hämäläinen, K., and P. Böckerman. 2004. “Regional Labor Market Dynamics, Housing,
and Migration.” Journal of Regional Science 44, no. 3: 543–68.
Hamilton, E. 2019. “Inclusionary Zoning and Housing Market Outcomes.” Arlington, VA:
Mercatus Center at George Mason University. https://www.mercatus.org/
publications/urban-economics/
inclusionary-zoning-and-housing-market-outcomes.
References | 339

250-840_text_.pdf 343

2/7/20 3:46 PM

Hanratty, M. 2017. “Do Local Economic Conditions Affect Homelessness? Impact of
Area Housing Market Factors, Unemployment, and Poverty on Community
Homeless Rates.” Housing Policy Debate 27, no. 4: 640–55.
Herkenhoff, K., L. Ohanian, and E. Prescott. 2018. “Tarnishing the Golden and Empire
States: Land-Use Restrictions and the U.S. Economic Slowdown.” Journal of
Monetary Economics 93: 89–109.
Hsieh, C., and E. Moretti. 2019. “Housing Constraints and Spatial Misallocation.”
American Economic Journal: Macroeconomics 11, no. 2: 1–39.
HUD (U.S. Department of Housing and Urban Development). 2018. “The 2018 Annual
Homeless Assessment Report (AHAR) to Congress.” https://files.hudexchange.
info/resources/documents/2018-AHAR-Part-1.pdf.
———. 2015. “Chapter 6: Calculating Rent and HAP Payments.” https://www.hud.gov/
sites/documents/DOC_35616.PDF.
———. 2019a. “Fair Market Rents: Introductory Overview.” https://www.huduser.gov/
portal/sites/default/files/pdf/fmr-overview.pdf.
———. 2019b. “Payment Standards” In Housing Choice Voucher Program Guidebook.
https://www.hud.gov/sites/dfiles/PIH/documents/HCV_Guidebook_
Payment_Standards.pdf.
IGM (Initiative on Global Markets). 2012. “Rent Control.” http://www.igmchicago.org/
surveys/rent-control.
Ihlanfeldt, K. 2004. “Exclusionary Land-Use Regulations within Suburban Communities:
A Review of the Evidence and Policy Prescriptions.” Urban Studies 41, no. 2:
261–83.
———. 2007. “The Effect of Land Use Regulation on Housing and Land Prices.” Journal
of Urban Economics 61, no. 3: 420–35.
Ikeda, S., and E. Washington. 2015. “How Land-Use Regulation Undermines Affordable
Housing.” Mercatus Center, George Mason University. https://www.mercatus.
org/system/files/Ikeda-Land-Use-Regulation.pdf.
Jackson, K. 2018. “Why Is Liberal California the Poverty Capital of America?” Manhattan
Institute, January 14. https://www.manhattan-institute.org/html/why-liberalcalifornia-poverty-capital-america-10882.html.
JCT (Joint Committee on Taxation). 2017. “Present Law and Data Relating to Tax
Incentives for Rental Housing.” JCX-40-17. Washington: Joint Committee on
Taxation.
JEC (Joint Economic Committee of the U.S. Congress). 2019. “Zoned Out: How School
and Residential Zoning Limit Educational Opportunity.” https://www.jec.
senate.gov/public/_cache/files/e18ff012-908e-4521-b1ce-a8b7b7f28ee3/
jec-report-zoned-out.pdf.
Jiang, H. 2019. “Ignorance Is Bliss? Rent Regulation, Policy Awareness, and Labor
Market Outcomes: Evidence from New York City.” http://www.econ2.jhu.edu/
jobmarket/2019/JiangH/JobPaper/JobPaperJiangH.pdf.

340 | References

250-840_text_.pdf 344

2/7/20 3:46 PM

Lens, M., and P. Monkkonen. 2016. “Do Strict Land Use Regulations Make Metropolitan
Areas More Segregated by Income?” Journal of the American Planning
Association 82, no. 1: 6–21.
Mawhorter, S., and C. Reid. 2018. “Local Housing Policies Across California: Presenting
the Results of a New Statewide Survey.” Terner Center for Housing
Innovation. https://californialanduse.org/download/Terner_California_
Residential_Land_Use_Survey_Report.pdf.
McCarty, M. 2018. “HUD FY2018 Appropriations: In Brief.” Congressional Research
Service. https://fas.org/sgp/crs/misc/R44931.pdf.
Nisar, H., M. Vachon, C. Horseman, and J. Murdoch. 2019. “Market Predictors of
Homelessness: How Housing and Community Factors Shape Homelessness
Rates Within Continuums of Care.” U.S. Department of Housing and Urban
Development. https://www.huduser.gov/portal/sites/default/files/pdf/
Market-Predictors-of-Homelessness.pdf.
Owens, A. 2019. “Building Inequality: Housing Segregation and Income Segregation.”
Sociological Science 6: 497–525.
Pinto, E., and T. Peter. 2019. “Home Price Appreciation Index and Months’ Remaining
Inventory.” AEI Housing Center, American Enterprise Institute.
Quigley, J., S. Raphael, and E. Smolensky. 2001. “Homeless in America, Homeless in
California.” Review of Economics and Statistics 83, no. 1: 37–51.
Quigley, J., and L. Rosenthal. 2005. “The Effects of Land Use Regulation on the Price of
Housing: What Do We Know? What Can We Learn?” Cityscape 8, no. 1: 69–137.
http://www.jstor.org/stable/20868572.
Quigley, J., and S. Raphael. 2005. “Regulation and the High Cost of Housing in
California.” American Economic Review 95, no. 2: 323–28. https://www.jstor.
org/stable/pdf/4132841.pdf.
Raphael, S. 2010. “Housing Market Regulation and Homelessness.” In How to House the
Homeless, edited by I. Ellen and B. O’Flaherty. New York: Russell Sage
Foundation.
Rossi, A. and F. D’Acunto. 2017. “How Dodd-Frank Helped the Wealthy and Left the
Middle Class Behind.” The Hill, April 16. https://thehill.com/blogs/punditsblog/finance/328998-how-dodd-frank-hurt-the-middle-class.
Rothwell, J., and D. Massey. 2010. “Density Zoning and Class Segregation in U.S.
Metropolitan Areas.” Social Science Quarterly 91, no. 5: 1123–43. https://www.
ncbi.nlm.nih.gov/pmc/articles/PMC3632084/pdf/nihms453211.pdf.
Saiz, A. 2010. “The Geographic Determinants of Housing Supply.” Quarterly Journal of
Economics 125, no. 3: 1253–96.
Saks, R. 2008. “Job Creation and Housing Construction: Constraints on Metropolitan
Area Employment Growth.” Journal of Urban Economics 64, no. 1: 178–95.
Schrank, D., B. Eisele, and T. Lomax. 2019. “2019 Urban Mobility Report.” Texas A&M
Transportation Institute. https://static.tti.tamu.edu/tti.tamu.edu/documents/
mobility-report-2019.pdf.

References | 341

250-840_text_.pdf 345

2/7/20 3:46 PM

Schuetz, J., R. Meltzer, and V. Been. 2011. “Silver Bullet or Trojan Horse? The Effects of
Inclusionary Zoning on Local Housing Markets in the United States.” Urban
Studies 48, no. 2: 297–329.
Schwartz, M., and E. Wilson. 2007. “Who Can Afford To Live in a Home? A Look at Data
from the 2006 American Community Survey.” U.S. Census Bureau. https://
www.census.gov/housing/census/publications/who-can-afford.pdf.
Steffen, B., S. Bucholtz, M. Martin, D. Vandenbroucke, and Y. Yao. 2013. “Worst Case
Housing Needs 2011: Report to Congress.” U.S. Department of Housing and
Urban Development. https://www.huduser.gov/portal//Publications/pdf/
HUD-506_WorstCase2011_reportv3.pdf.
White House. 2019. “Executive Order Establishing a White House Council on Eliminating
Regulatory Barriers to Affordable Housing.” https://www.whitehouse.gov/
presidential-actions/
executive-order-establishing-white-house-council-eliminating-regulatorybarriers-affordable-housing/.
Zabel, J. 2012. “Migration, Housing Market, and Labor Market Responses to
Employment Shocks.” Journal of Urban Economics 72, nos. 2–3: 267–84.

Chapter 9
Cadot, O., J. Gourdon, and F. van Tongeren. 2018. Estimating Ad Valorem Equivalents of
Non-Tariff Measures: Combining Price-Based and Quantity-Based Approaches.
OECD Trade Policy Paper 215. Paris: OECD Publishing. https://doi.
org/10.1787/f3cd5bdc-en.
CBO (U.S. Congressional Budget Office). 2019. “The Budget and Economic Outlook:
2019 to 2029.” https://www.cbo.gov/publication/54918.
CEA (Council of Economic Advisers). 2019. Economic Report of the President.
Washington: U.S. Government Publishing Office.
Ehrlich, I., D. Li, and Z. Liu. 2017. The Role of Entrepreneurial Human Capital as a Driver
of Endogenous Economic Growth. NBER Working Paper 23728. Cambridge, MA:
National Bureau of Economic Research.
Federal Reserve. 2018. “Economic Projections of Federal Reserve Board Members and
Federal Reserve Bank Presidents under Their Individual Assessments of
Projected Appropriate Monetary Policy, December 2019.” December 11.
https://www.federalreserve.gov/monetarypolicy/files/fomcprojtabl20191211.
pdf.
Lamprecht, P., and S. Miroudot. 2018. The Value of Market Access and National
Treatment Commitments in Services Trade Agreements. OECD Trade Policy
Paper 213. Paris: OECD Publishing. https://doi.org/10.1787/d8bfc8d8-en.
OECD (Organization for Economic Cooperation and Development). 2018. OECD
Economic Surveys: United States. Paris: OECD Publishing. https://www.oecd.
org/economy/surveys/Overview-United-States-2018-OECD.pdf.

342 | References

250-840_text_.pdf 346

2/7/20 3:46 PM

x

Appendix A

Report to the President
on the Activities of the
Council of Economic Advisers
During 2019

343

250-840_text_.pdf 347

2/7/20 3:46 PM

250-840_text_.pdf 348

2/7/20 3:46 PM

x

Letter of Transmittal
Council of Economic Advisers
Washington, December 31, 2019
Mr. President:
The Council of Economic Advisers submits this report on its activities
during calendar year 2019 in accordance with the requirements of the Congress,
as set forth in section 10(d) of the Employment Act of 1946, as amended by the
Full Employment and Balanced Growth Act of 1978.
Sincerely yours,

Tomas J. Philipson
Acting Chairman

Tyler B. Goodspeed
Member

Activities of the Council of Economic Advisers During 2019 | 345

250-840_text_.pdf 349

2/7/20 3:46 PM

Council Members and Their Dates of Service
Name

Position

Oath of office date Separation date

Edwin G. Nourse
Leon H. Keyserling

Chairman
Vice Chairman
Acting Chairman
Chairman
Member
Vice Chairman
Member
Member
Chairman
Member
Member
Member
Chairman
Member
Member
Member
Member
Chairman
Member
Member
Member
Chairman
Member
Member
Member
Chairman
Member
Member
Member
Chairman
Member
Member
Chairman
Member
Member
Member
Member
Chairman
Member
Member
Chairman
Member
Member
Member
Member
Chairman
Member
Member
Chairman
Member
Chairman

August 9, 1946
August 9, 1946
November 2, 1949
May 10, 1950
August 9, 1946
May 10, 1950
June 29, 1950
September 8, 1952
March 19, 1953
September 15, 1953
December 2, 1953
April 4, 1955
December 3, 1956
May 2, 1955
December 3, 1956
November 1, 1958
May 7, 1959
January 29, 1961
January 29, 1961
January 29, 1961
August 3, 1962
November 16, 1964
May 17, 1963
September 2, 1964
November 16, 1964
February 15, 1968
February 2, 1966
February 15, 1968
July 1, 1968
February 4, 1969
February 4, 1969
February 4, 1969
January 1, 1972
September 9, 1971
March 13, 1972
July 23, 1973
October 31, 1973
September 4, 1974
June 13, 1975
July 22, 1975
January 22, 1977
March 18, 1977
March 18, 1977
June 6, 1979
August 20, 1980
February 27, 1981
June 12, 1981
July 14, 1981
October 14, 1982
December 10, 1982
April 18, 1985

John D. Clark
Roy Blough
Robert C. Turner
Arthur F. Burns
Neil H. Jacoby
Walter W. Stewart
Raymond J. Saulnier
Joseph S. Davis
Paul W. McCracken
Karl Brandt
Henry C. Wallich
Walter W. Heller
James Tobin
Kermit Gordon
Gardner Ackley
John P. Lewis
Otto Eckstein
Arthur M. Okun
James S. Duesenberry
Merton J. Peck
Warren L. Smith
Paul W. McCracken
Hendrik S. Houthakker
Herbert Stein
Ezra Solomon
Marina v.N. Whitman
Gary L. Seevers
William J. Fellner
Alan Greenspan
Paul W. MacAvoy
Burton G. Malkiel
Charles L. Schultze
William D. Nordhaus
Lyle E. Gramley
George C. Eads
Stephen M. Goldfeld
Murray L. Weidenbaum
William A. Niskanen
Jerry L. Jordan
Martin Feldstein
William Poole
Beryl W. Sprinkel

November 1, 1949
January 20, 1953
February 11, 1953
August 20, 1952
January 20, 1953
December 1, 1956
February 9, 1955
April 29, 1955
January 20, 1961
October 31, 1958
January 31, 1959
January 20, 1961
January 20, 1961
November 15, 1964
July 31, 1962
December 27, 1962
February 15, 1968
August 31, 1964
February 1, 1966
January 20, 1969
June 30, 1968
January 20, 1969
January 20, 1969
December 31, 1971
July 15, 1971
August 31, 1974
March 26, 1973
August 15, 1973
April 15, 1975
February 25, 1975
January 20, 1977
November 15, 1976
January 20, 1977
January 20, 1981
February 4, 1979
May 27, 1980
January 20, 1981
January 20, 1981
August 25, 1982
March 30, 1985
July 31, 1982
July 10, 1984
January 20, 1985
January 20, 1989

346 | Appendix A

250-840_text_.pdf 350

2/7/20 3:46 PM

Council Members and Their Dates of Service
Name

Position

Oath of office date Separation date

Thomas Gale Moore
Michael L. Mussa
Michael J. Boskin
John B. Taylor
Richard L. Schmalensee
David F. Bradford
Paul Wonnacott
Laura D’Andrea Tyson
Alan S. Blinder
Joseph E. Stiglitz

Member
Member
Chairman
Member
Member
Member
Member
Chair
Member
Member
Chairman
Member
Member
Chair
Member
Member
Chairman
Member
Member
Chairman
Member
Member
Chairman
Member
Member
Chairman
Chairman
Member
Member
Chairman
Member
Chair
Member
Chairman
Member
Member
Member
Chairman
Member
Chairman
Member
Member
Member
Member
Chairman
Member
Member
Acting Chairman
Vice Chairman
Member

July 1, 1985
August 18, 1986
February 2, 1989
June 9, 1989
October 3, 1989
November 13, 1991
November 13, 1991
February 5, 1993
July 27, 1993
July 27, 1993
June 28, 1995
June 30, 1995
January 29, 1996
February 18, 1997
April 23, 1997
October 22, 1998
August 12, 1999
August 12, 1999
May 31, 2000
May 11, 2001
July 25, 2001
November 30, 2001
May 29, 2003
November 21, 2003
November 21, 2003
February 23, 2005
June 21, 2005
November 18, 2005
November 18, 2005
February 27, 2006
July 17, 2008
January 29, 2009
March 11, 2009
September 10, 2010
March 11, 2009
April 19, 2011
April 19, 2011
November 7, 2011
February 7, 2013
August 4, 2013
August 6, 2013
July 21, 2014
August 10, 2015
August 31, 2015
September 13, 2017
September 28, 2017
August 31, 2017
July 1, 2019
July 24, 2019
May 22, 2019

Martin N. Baily
Alicia H. Munnell
Janet L. Yellen
Jeffrey A. Frankel
Rebecca M. Blank
Martin N. Baily
Robert Z. Lawrence
Kathryn L. Shaw
R. Glenn Hubbard
Mark B. McClellan
Randall S. Kroszner
N. Gregory Mankiw
Kristin J. Forbes
Harvey S. Rosen
Ben S. Bernanke
Katherine Baicker
Matthew J. Slaughter
Edward P. Lazear
Donald B. Marron
Christina D. Romer
Austan D. Goolsbee
Cecilia Elena Rouse
Katharine G. Abraham
Carl Shapiro
Alan B. Krueger
James H. Stock
Jason Furman
Betsey Stevenson
Maurice Obstfeld
Sandra E. Black
Jay C. Shambaugh
Kevin A. Hassett
Richard V. Burkhauser
Tomas J. Philipson
Tyler B. Goodspeed

May 1, 1989
September 19, 1988
January 12, 1993
August 2, 1991
June 21, 1991
January 20, 1993
January 20, 1993
April 22, 1995
June 26, 1994
February 10, 1997
August 30, 1996
August 1, 1997
August 3, 1999
March 2, 1999
July 9, 1999
January 19, 2001
January 12, 2001
January 19, 2001
February 28, 2003
November 13, 2002
July 1, 2003
February 18, 2005
June 3, 2005
June 10, 2005
January 31, 2006
July 11, 2007
March 1, 2007
January 20, 2009
January 20, 2009
September 3, 2010
August 5, 2011
February 28, 2011
April 19, 2013
May 4, 2012
August 2, 2013
May 19, 2014
January 20, 2017
August 7, 2015
August 28, 2015
January 20, 2017
January 20, 2017
June 30, 2019
May 18, 2019

Activities of the Council of Economic Advisers During 2019 | 347

250-840_text_.pdf 351

2/7/20 3:46 PM

250-840_text_.pdf 352

2/7/20 3:46 PM

Report to the President on the
Activities of the Council of
Economic Advisers During 2019
The Employment Act of 1946 established the Council of Economic Advisers to
provide the President with objective economic analysis on the development
and implementation of policy for the full range of domestic and international
economic issues that can affect the United States. Governed by a Chairman,
who is appointed by the President and confirmed by the United States Senate,
the Council has two additional Members who are also appointed by the
President.

The Chairman of the Council
On June 28, 2019, Kevin A. Hassett resigned as Chairman of the Council of
Economic Advisers. In accordance with the Employment Act of 1946, the duties
and responsibilities of the Chairman have been subsequently executed by
Tomas J. Philipson, who has served as a Member of the Council since 2017 and
was appointed Vice Chairman on July 24, 2019.

The Members of the Council
Tomas J. Philipson is the Vice Chairman of the White House Council of Economic
Advisers, and in this capacity serves as acting Chairman. He is on leave from
the University of Chicago, and has been a Member of the Council of Economic
Advisers since his appointment in 2017. Previously, he served in the George
W. Bush Administration, among other public sector positions. He received
his M.A. and Ph.D. in economics from the Wharton School at the University of
Pennsylvania. He has been a visiting faculty member at Yale University and a
visiting senior fellow at the World Bank. He previously served as a fellow, board
member, or associate with a number of other organizations, including the
University of Chicago, the National Bureau of Economic Research, the American
Enterprise Institute, the Manhattan Institute, the Heartland Institute, the Milken
Institute, the RAND Corporation, and the University of Southern California’s
Schaeffer Center for Health Policy & Economics.
Tyler Beck Goodspeed is a Member of the Council of Economic Advisers,
having previously served as Chief Economist for Macroeconomic Policy and
Senior Economist for Macroeconomics. Before joining the CEA, he was a
member of the Faculty of Economics at the University of Oxford and was a

Activities of the Council of Economic Advisers During 2019 | 349

250-840_text_.pdf 353

2/7/20 3:46 PM

lecturer in economics at King’s College London. He has published extensively
on financial regulation, banking, and monetary economics, with particular
attention to the role of contingent liability and access to credit in mitigating
the effects of adverse aggregate shocks in historical contexts. His research has
appeared in three full-length monographs from academic publishers, as well as
numerous articles in peer-reviewed and edited journals. He received his B.A.,
M.A., and Ph.D. from Harvard University; and he received his M.Phil from the
University of Cambridge, where he was a Gates Scholar. He is a current member
of the American Economic Association, and was previously a member of the
Economic History Association, Economic History Society, and Royal Economic
Society, as well as an adjunct scholar at the Cato Institute.

Areas of Activity
Macroeconomic Policies
Throughout 2019, fulfilling its mandate from the Employment Act of 1946, the
Council continued “to gather timely and authoritative information concerning
economic developments and economic trends, both current and prospective.” The Council appraises the President and the White House staff of new
economic data and their significance on an ongoing basis. As core products of
the Council, these regular appraisals include written memoranda. The Council
also prepares in-depth briefings on certain topics, as well as public reports that
address macroeconomic issues.
One of the Council’s public reports this year addressed the economic
effects of Federal deregulation. According to the report, this historic reduction
in costly Federal regulation will raise real household incomes by a large enough
magnitude to have macroeconomic implications.
On employment and the labor market, the Council actively disseminated
analyses to the public. One report addressed the effectiveness of public jobtraining programs in improving participants’ labor market outcomes. Another
report showed that economic growth is more effective in lifting Americans out
of poverty than expanded government assistance programs. The Council also
released a report on how lower market costs for childcare could affect parents’
labor force participation. Reports on employment policies complement the
Council’s regular blog posts on new releases of labor market data.
The Council also released a report that shows U.S. energy innovation,
epitomized by the shale revolution in oil and natural gas production, increases
household incomes by lowering consumers’ energy costs. Furthermore, the
report highlighted how the shale revolution led the United States to experience
a greater decline in energy-related emissions than European Union countries.
Working alongside the Department of the Treasury and the Office of
Management and Budget, the Council participates in the “troika” process that
350 | Appendix A

250-840_text_.pdf 354

2/7/20 3:46 PM

generates the macroeconomic forecasts that underlie the Administration’s
budget proposals. The Council, under the leadership of the acting Chairman
and the Members, continued to initiate and lead this forecasting process.
The acting Chairman and Members maintained the Council’s tradition of
meeting regularly with the Chairman and Members of the Board of Governors
of the Federal Reserve System to exchange views on the economy.

Microeconomic Policies
The Council participated in discussions, internal to the Federal Government
as well as external, on a range of issues in microeconomic policy. Publication
topics included healthcare deregulation, vaccines, prescription drug pricing,
the opioid crisis, and homelessness.
On healthcare, the Council published a paper on the Trump
Administration’s policies to expand healthcare choice and competition. This
paper finds that these policy changes—including reducing the individual
mandate penalty; permitting more association health plans; and expanding
short-term, limited-duration insurance plans—will keep costs down for
consumers and taxpayers. The Council also released a report that estimates
the potentially large health and economic losses associated with influenza
pandemics and discusses policy options to increase vaccine innovation and
moderate pandemics’ risk.
Additionally, the Council published a paper that shows average prescription drug prices are falling because of improved Food and Drug Administration
policies that, if continued, will benefit patients by further lowering drug prices.
The Council also released a report on how lower prices and easier access to
opioids exacerbated the crisis’s growth, which finally shows signs of leveling
off.
Another Council report documents the state of homelessness in America.
This report finds that the Administration’s actions to reduce regulatory barriers
in the housing market, combat the drug crisis, expand mental illness treatment, improve the chances of people leaving prison, promote self-sufficiency,
support effective policing, and increase incomes for people at the bottom of the
distribution will address the root causes of homelessness.

International Economics
The Council participated in the analysis of numerous issues in the area of
international economics. The Council engages with a number of international
organizations. The Council is a leading participant in the activities of the
Organization for Economic Cooperation and Development, a forum for facilitating economic coordination and cooperation among the world’s high-income
countries. Council Members and Council staff have also engaged with the organization’s working-party meetings on a range of issues and shaped its agenda.

Activities of the Council of Economic Advisers During 2019 | 351

250-840_text_.pdf 355

2/7/20 3:46 PM

In addition, the Council analyzed a number of proposals and scenarios
in the area of international trade and investment. These included generating
estimates of the benefits, as well as any trade-offs, of prospective trade agreements as well as revisions to existing agreements.
The Council continues to actively monitor the U.S. international trade
and investment position and to engage with emerging issues in international
economics, such as malicious cyber activity. The Council looks forward to
continuing to analyze the United States’ international economic position.

352 | Appendix A

250-840_text_.pdf 356

2/7/20 3:46 PM

The Staff of the Council of Economic Advisers
Executive Office
Rachael S. Slobodien	��������������������������������Chief of Staff
Paige E. Terryberry	������������������������������������Deputy Chief of Staff
Robert M. Fisher 	����������������������������������������General Counsel and Senior Economist
Cale A. Clingenpeel	������������������������������������Special Adviser to the Chairman and
Economist
Jared T. Meyer	��������������������������������������������Special Adviser to the Chairman on
Communications
David N. Grogan 	����������������������������������������Staff Assistant
Emily A. Tubb	����������������������������������������������Staff Assistant

Senior Research Staff
Joseph V. Balagtas 	������������������������������������Senior Economist; Agriculture,
International Trade, and Infrastructure
Andre J. Barbe	��������������������������������������������Senior Economist; International Trade
Steven N. Braun	������������������������������������������Director of Macroeconomic Forecasting
Kevin C. Corinth	������������������������������������������Chief Economist for Domestic Policy
Jason J. Galui	����������������������������������������������Senior Advisor to the Chairman;
National Security
LaVaughn M. Henry	������������������������������������Senior Economist; Education, Banking,
and Finance
Donald S. Kenkel	����������������������������������������Chief Economist
Ian A. Lange 	������������������������������������������������Senior Economist; Energy
Brett R. Matsumoto	������������������������������������Senior Economist; Labor and Health
Deborah F. Minehart	����������������������������������Senior Economist; Industrial
Organization
Stephen T. Parente	������������������������������������Senior Economist; Health
Joshua D. Rauh	������������������������������������������Principal Chief Economist
Eric C. Sun	����������������������������������������������������Senior Economist; Health
Jeremy G. Weber	����������������������������������������Senior Advisor to the Council and Chief
Energy Economist
Anna W. Wong 	��������������������������������������������Chief International Economist

Junior Research Staff
Jackson H. Bailey	��������������������������������������Research Assistant; Housing and
Education
Andrew M. Baxter 	��������������������������������������Staff Economist; Deregulation and
Macroeconomics
Adam D. Donoho	����������������������������������������Research Economist; Macroeconomics
and International Trade

Activities of the Council of Economic Advisers During 2019 | 353

250-840_text_.pdf 357

2/7/20 3:46 PM

Alex J. Durante	��������������������������������������������Staff Economist; International Trade
and Public Finance
Troy M. Durie 	����������������������������������������������Research Economist; International
Trade, Macroeconomics
William O. Ensor 	����������������������������������������Staff Economist
Amelia C. Irvine	������������������������������������������Research Assistant; Labor,
Macroeconomics
Gregory K. Kearney	������������������������������������Research Economist; Tax, Deregulation,
and Macroeconomics
Nicole P. Korkos	������������������������������������������Research Economist; National Security
David J. Laszcz	��������������������������������������������Staff Economist
Caroline J. Liang	����������������������������������������Research Economist; Deregulation,
Health, and Education
Julia A. Tavlas 	��������������������������������������������Economist; Education, Labor, and
Poverty
Grayson R. Wiles 	����������������������������������������Research Assistant; Macroeconomics,
Health, and Deregulation

Statistical Office
Brian A. Amorosi	����������������������������������������Director of Statistical Office

Administrative Office
Doris L. Searles 	������������������������������������������Operations Manager

Interns
Student interns provide invaluable help with research projects, day-to-day
operations, and fact-checking. Interns during the year were: Justin Arenas,
William Arnesen, Michelle Bai, Quinn Barry, Matthew Baumholtz, Michael
Bugay, John Camara, Blythe Carvajal, Cross Di Muro, Ayelet Drazen, Soleine
Fechter, Kiyanoush Forough, Jelena Goldstein, Caroline Hui, Jacob Kronman,
Meg Leatherwood, Andrew Liang, Eric Menser, Hailey Ordal, Raj Ramnani,
Jacqueline Sands, Cindy Shen, Matthew Style, Sharon Yen, Michael Yin, and
Chris Zhao.

ERP Production
Alfred F. Imhoff 	������������������������������������������Editor

354 | Appendix A

250-840_text_.pdf 358

2/7/20 3:46 PM

x

Appendix B

Statistical Tables Relating to Income,
Employment, and Production

355

250-840_text_.pdf 359

2/7/20 3:46 PM

250-840_text_.pdf 360

2/7/20 3:46 PM

Contents
National Income or Expenditure
B–1.

Percent changes in real gross domestic product, 1969–2019�����������

362

B–2.

Contributions to percent change in real gross domestic product,
1969–2019��������������������������������������������������������������������������������������������

364

B–3.

Gross domestic product, 2004–2019��������������������������������������������������

366

B–4.

Percentage shares of gross domestic product, 1969–2019���������������

368

B–5.

Chain-type price indexes for gross domestic product, 1969–2019���

370

B–6.

Gross value added by sector, 1969–2019��������������������������������������������

372

B–7.

Real gross value added by sector, 1969–2019������������������������������������

373

B–8.

Gross domestic product (GDP) by industry, value added, in current
dollars and as a percentage of GDP, 1997–2018���������������������������������

374

B–9.

Real gross domestic product by industry, value added, and
percent changes, 1997–2018��������������������������������������������������������������

376

B–10. Personal consumption expenditures, 1969–2019������������������������������

378

B–11. Real personal consumption expenditures, 2002–2019����������������������

379

B–12. Private fixed investment by type, 1969–2019�������������������������������������

380

B–13. Real private fixed investment by type, 2002–2019�����������������������������

381

B–14. Foreign transactions in the national income and product accounts,
1969–2019�������������������������������������������������������������������������������������������� 382
B–15. Real exports and imports of goods and services, 2002–2019������������

383

B–16. Sources of personal income, 1969–2019��������������������������������������������

384

B–17. Disposition of personal income, 1969–2019��������������������������������������

386

B–18. Total and per capita disposable personal income and personal
consumption expenditures, and per capita gross domestic
product, in current and real dollars, 1969–2019��������������������������������

387

B–19. Gross saving and investment, 1969–2019������������������������������������������

388

B–20. Median money income (in 2018 dollars) and poverty status
of families and people, by race, 2011-2018����������������������������������������

390

B–21. Real farm income, 1954–2019�������������������������������������������������������������

391

Contents | 357

250-840_text_.pdf 361

2/7/20 3:46 PM

Labor Market Indicators
B–22. Civilian labor force, 1929–2019�����������������������������������������������������������

392

B–23. Civilian employment by sex, age, and demographic characteristic,
1975–2019��������������������������������������������������������������������������������������������

394

B–24. Unemployment by sex, age, and demographic characteristic,
1975–2019��������������������������������������������������������������������������������������������

395

B–25. Civilian labor force participation rate, 1975–2019�����������������������������

396

B–26. Civilian employment/population ratio, 1975–2019���������������������������

397

B–27. Civilian unemployment rate, 1975–2019��������������������������������������������

398

B–28. Unemployment by duration and reason, 1975–2019������������������������

399

B–29. Employees on nonagricultural payrolls, by major industry,
1975–2019��������������������������������������������������������������������������������������������

400

B–30. Hours and earnings in private nonagricultural industries,
1975–2019��������������������������������������������������������������������������������������������

402

B–31. Employment cost index, private industry, 2002–2019�����������������������

403

B–32. Productivity and related data, business and nonfarm business
sectors, 1970–2019������������������������������������������������������������������������������

404

B–33. Changes in productivity and related data, business and nonfarm
business sectors, 1970–2019���������������������������������������������������������������

405

Production and Business Activity
B–34. Industrial production indexes, major industry divisions,
1975–2019��������������������������������������������������������������������������������������������

406

B–35. Capacity utilization rates, 1975–2019�������������������������������������������������

407

B–36. New private housing units started, authorized, and completed
and houses sold, 1975–2019���������������������������������������������������������������

408

B–37. Manufacturing and trade sales and inventories, 1979–2019�������������

409

Prices
B–38. Changes in consumer price indexes, 1977–2019��������������������������������

410

B–39. Price indexes for personal consumption expenditures, and percent
changes, 1972–2019���������������������������������������������������������������������������� 411

358 |

250-840_text_.pdf 362

Appendix B

2/7/20 3:46 PM

Money Stock, Credit, and Finance
B–40. Money stock and debt measures, 1980–2019������������������������������������

412

B–41. Consumer credit outstanding, 1970–2019�����������������������������������������

413

B–42. Bond yields and interest rates, 1949–2019�����������������������������������������

414

B–43. Mortgage debt outstanding by type of property and of financing,
1960–2019��������������������������������������������������������������������������������������������

416

B–44. Mortgage debt outstanding by holder, 1960–2019����������������������������

417

Government Finance
B–45. Federal receipts, outlays, surplus or deficit, and debt, fiscal years
1955–2021��������������������������������������������������������������������������������������������

418

B–46. Federal receipts, outlays, surplus or deficit, and debt, as percent of
gross domestic product, fiscal years 1949–2021�������������������������������� 419
B–47. Federal receipts and outlays, by major category, and surplus or
deficit, fiscal years 1955–2021������������������������������������������������������������

420

B–48. Federal receipts, outlays, surplus or deficit, and debt, fiscal years
2016–2021��������������������������������������������������������������������������������������������

421

B–49. Federal and State and local government current receipts and
expenditures, national income and product accounts (NIPA) basis,
1969–2019��������������������������������������������������������������������������������������������

422

B–50. State and local government revenues and expenditures,
fiscal years 1956–2017�������������������������������������������������������������������������

423

B–51. U.S. Treasury securities outstanding by kind of obligation,
1980–2019��������������������������������������������������������������������������������������������

424

B–52. Estimated ownership of U.S. Treasury securities, 2006–2019�����������

425

Corporate Profits and Finance
B–53. Corporate profits with inventory valuation and
capital consumption adjustments, 1969–2019����������������������������������

426

B–54. Corporate profits by industry, 1969–2019������������������������������������������

427

B–55. Historical stock prices and yields, 1949–2003������������������������������������

428

B–56. Common stock prices and yields, 2000–2019������������������������������������

429

Contents | 359

250-840_text_.pdf 363

2/7/20 3:46 PM

International Statistics
B–57. U.S. international transactions, 1969–2019���������������������������������������

430

B–58. U.S. international trade in goods on balance of payments (BOP)
and Census basis, and trade in services on BOP basis, 1991–2019���

432

B–59. U.S. international trade in goods and services by area and
country, 2000–2018�����������������������������������������������������������������������������

433

B–60. Foreign exchange rates, 2000–2019����������������������������������������������������

434

B–61. Growth rates in real gross domestic product by area and
country, 2001-2020������������������������������������������������������������������������������

435

360 |

250-840_text_.pdf 364

Appendix B

2/7/20 3:46 PM

General Notes
Detail in these tables may not add to totals due to rounding.
Because of the formula used for calculating real gross domestic product (GDP),
the chained (2012) dollar estimates for the detailed components do not add
to the chained-dollar value of GDP or to any intermediate aggregate. The
Department of Commerce (Bureau of Economic Analysis) no longer publishes
chained-dollar estimates prior to 2002, except for selected series.
Because of the method used for seasonal adjustment, the sum or average
of seasonally adjusted monthly values generally will not equal annual totals
based on unadjusted values.
Unless otherwise noted, all dollar figures are in current dollars.
Symbols used:
p Preliminary.
... Not available (also, not applicable).
NSA Not seasonally adjusted.
Data in these tables reflect revisions made by source agencies through
January 31, 2020.
Excel versions of these tables are available at www.gpo.gov/erp.

General Notes | 361

250-840_text_.pdf 365

2/7/20 3:46 PM

National Income or Expenditure
Table B–1. Percent changes in real gross domestic product, 1969–2019
[Percent change, fourth quarter over fourth quarter; quarterly changes at seasonally adjusted annual rates]
Personal consumption
expenditures

Year or quarter

1969 ����������������������
1970 ����������������������
1971 ����������������������
1972 ����������������������
1973 ����������������������
1974 ����������������������
1975 ����������������������
1976 ����������������������
1977 ����������������������
1978 ����������������������
1979 ����������������������
1980 ����������������������
1981 ����������������������
1982 ����������������������
1983 ����������������������
1984 ����������������������
1985 ����������������������
1986 ����������������������
1987 ����������������������
1988 ����������������������
1989 ����������������������
1990 ����������������������
1991 ����������������������
1992 ����������������������
1993 ����������������������
1994 ����������������������
1995 ����������������������
1996 ����������������������
1997 ����������������������
1998 ����������������������
1999 ����������������������
2000 ����������������������
2001 ����������������������
2002 ����������������������
2003 ����������������������
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

Gross
domestic
product

2.0
–.2
4.4
6.9
4.0
–1.9
2.6
4.3
5.0
6.7
1.3
.0
1.3
–1.4
7.9
5.6
4.2
2.9
4.5
3.8
2.7
.6
1.2
4.4
2.6
4.1
2.2
4.4
4.5
4.9
4.8
3.0
.2
2.1
4.3
3.3
3.1
2.6
2.0
–2.8
.2
2.6
1.6
1.5
2.6
2.9
1.9
2.0
2.8
2.5
2.3
2.0
1.9
2.2
2.0
2.3
2.2
3.2
3.5
2.5
3.5
2.9
1.1
3.1
2.0
2.1
2.1

Gross private domestic investment
Fixed investment
Nonresidential

Total

3.1
1.7
5.4
7.3
1.8
–1.6
5.1
5.4
4.2
4.0
1.7
.0
.1
3.5
6.6
4.3
4.8
4.4
2.8
4.6
2.4
.8
.9
4.9
3.3
3.8
2.8
3.4
4.5
5.6
5.1
4.4
2.5
2.1
3.8
3.8
3.0
3.2
1.6
–1.8
–.1
2.7
1.2
1.6
1.9
3.8
2.9
2.8
2.9
2.6
2.6
3.2
2.9
2.6
2.5
2.4
2.4
2.4
4.6
1.7
4.0
3.5
1.4
1.1
4.6
3.2
1.8

Goods

2.0
.0
6.6
8.5
.4
–5.6
6.1
6.4
4.9
3.5
.3
–2.5
–.2
3.6
8.3
5.3
4.6
6.5
.4
4.5
1.8
–1.6
–.8
5.3
4.4
5.5
2.3
4.8
5.3
8.1
6.6
4.0
4.9
1.7
6.6
4.3
3.0
4.6
1.8
–6.8
.6
4.3
.9
2.4
3.5
5.0
3.7
3.6
5.0
2.9
4.1
4.2
4.5
4.0
1.9
3.2
5.5
4.1
7.5
1.3
5.4
3.6
1.6
1.5
8.6
5.3
1.2

Services

4.2
3.4
4.3
6.2
3.2
2.4
4.1
4.5
3.7
4.4
2.9
2.2
.3
3.4
5.3
3.6
5.0
3.0
4.6
4.7
2.7
2.3
2.0
4.7
2.7
2.8
3.0
2.7
4.0
4.3
4.3
4.7
1.2
2.4
2.3
3.5
3.0
2.5
1.5
.9
–.4
1.9
1.4
1.2
1.1
3.2
2.5
2.4
2.0
2.5
2.0
2.7
2.2
1.9
2.8
2.0
1.0
1.6
3.4
1.9
3.4
3.4
1.4
1.0
2.8
2.2
2.0

Total

2.2
–6.4
13.1
15.0
10.2
–10.4
–9.8
15.2
14.9
14.3
–3.4
–7.2
6.7
–17.3
31.3
14.2
1.9
–4.1
9.8
–.5
.7
–6.5
2.1
7.7
7.6
11.5
.8
11.2
11.4
9.7
8.5
4.3
–11.1
4.4
8.7
8.0
6.1
–1.5
–1.8
–15.3
–9.2
12.1
10.4
4.0
9.3
5.3
1.5
1.5
4.8
5.1
–1.9
–1.6
–1.7
.5
9.3
3.4
3.6
7.4
4.7
6.2
–1.8
13.7
3.0
6.2
–6.3
–1.0
–6.1

Total

2.5
–.9
10.5
12.0
3.5
–9.9
–2.6
12.1
12.1
13.1
1.1
–4.8
1.5
–8.0
18.3
11.3
3.7
.6
1.5
3.7
1.5
–4.2
–1.9
8.7
8.4
6.6
5.5
9.9
8.3
11.5
7.2
5.9
–4.7
–1.5
8.6
6.5
5.8
.0
–1.1
–11.1
–10.5
6.1
9.2
7.2
5.7
7.0
1.0
2.8
5.1
3.5
.2
2.6
2.7
3.8
2.0
7.7
2.8
1.4
8.7
5.5
5.2
.7
2.7
3.2
–1.4
–.8
.1

Total
5.5
–4.4
4.7
11.5
10.6
–3.9
–5.9
7.8
11.9
16.0
5.5
–.9
9.0
–9.5
10.4
13.9
3.2
–3.2
2.2
5.1
4.5
–.9
–3.4
7.1
7.6
8.5
7.4
11.3
9.7
11.6
8.4
8.5
–6.8
–5.1
6.8
6.5
6.1
8.1
7.3
–7.0
–10.3
8.9
10.0
5.6
5.4
6.9
–.9
2.4
5.4
5.9
–.1
–.6
4.0
5.6
.7
6.6
4.4
2.4
8.4
8.8
7.9
2.1
4.8
4.4
–1.0
–2.3
–1.5

Structures
6.4
–2.6
–1.1
5.1
7.9
–6.4
–8.1
3.8
5.7
21.7
8.8
2.7
14.1
–13.5
–3.9
15.7
3.3
–14.3
4.9
–3.3
3.3
–3.2
–12.8
1.0
.2
1.6
4.7
10.9
4.4
4.3
–.1
10.8
–10.6
–15.7
1.9
.3
1.5
9.0
17.7
–.8
–27.1
–3.6
8.6
4.0
6.7
9.3
–10.9
4.3
1.5
2.6
–7.0
–11.4
10.0
18.4
2.4
7.3
2.0
–7.7
5.2
12.1
11.0
–2.1
–9.0
4.0
–11.1
–9.9
–10.1

Equipment
5.2
–5.8
8.5
17.0
13.5
–3.7
–6.7
9.0
17.2
14.5
2.7
–4.4
4.6
–10.0
19.9
13.4
1.7
.8
.1
8.2
2.5
–2.7
–3.2
11.3
13.1
12.5
8.1
11.1
10.7
14.8
9.5
8.5
–7.7
–3.7
9.6
9.8
8.7
7.1
3.9
–15.9
–8.4
22.6
12.7
7.8
5.4
5.6
1.9
–1.4
8.5
5.0
–1.5
–3.9
–2.3
.3
.4
6.3
8.9
6.2
12.9
6.6
3.4
2.9
7.4
–.1
.8
–3.8
–2.9

Intellectual
property
products
4.5
–3.4
4.8
6.2
5.1
1.6
2.8
11.8
4.8
10.3
9.4
4.7
12.1
3.4
13.0
12.6
7.7
5.4
4.2
9.8
11.3
6.2
7.2
4.8
2.9
5.8
8.3
12.1
12.4
11.5
13.3
6.6
–2.1
.9
5.8
5.7
5.1
9.3
4.0
.9
3.8
1.6
7.2
3.7
4.5
6.9
2.9
6.6
4.0
9.3
6.2
12.9
9.3
4.7
.0
6.3
.3
4.9
4.7
9.7
11.9
4.1
11.7
10.8
3.6
4.7
5.9

Residential

–5.4
9.4
25.2
12.9
–10.5
–24.6
7.8
23.8
12.6
6.8
–9.1
–15.3
–22.0
–1.7
49.7
3.7
5.2
11.8
–.5
.1
–6.5
–13.6
2.9
13.6
10.6
1.6
.1
5.6
4.0
11.3
3.5
–1.5
2.0
8.1
12.7
6.6
5.2
–15.2
–21.2
–24.7
–11.5
–5.7
5.3
15.4
7.1
7.7
9.1
3.9
4.2
–4.4
1.5
14.7
–2.0
–2.6
6.4
11.9
–2.2
–2.0
9.9
–5.3
–3.7
–4.0
–4.7
–1.0
–3.0
4.6
5.8

Change
in
private
inventories
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������
����������������

See next page for continuation of table.

362 |

250-840_text_.pdf 366

Appendix B

2/7/20 3:46 PM

Table B–1. Percent changes in real gross domestic product, 1969–2019—Continued
[Percent change, fourth quarter over fourth quarter; quarterly changes at seasonally adjusted annual rates]
Net exports of
goods and services
Year or quarter

1969 ����������������������
1970 ����������������������
1971 ����������������������
1972 ����������������������
1973 ����������������������
1974 ����������������������
1975 ����������������������
1976 ����������������������
1977 ����������������������
1978 ����������������������
1979 ����������������������
1980 ����������������������
1981 ����������������������
1982 ����������������������
1983 ����������������������
1984 ����������������������
1985 ����������������������
1986 ����������������������
1987 ����������������������
1988 ����������������������
1989 ����������������������
1990 ����������������������
1991 ����������������������
1992 ����������������������
1993 ����������������������
1994 ����������������������
1995 ����������������������
1996 ����������������������
1997 ����������������������
1998 ����������������������
1999 ����������������������
2000 ����������������������
2001 ����������������������
2002 ����������������������
2003 ����������������������
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

Government consumption expenditures
and gross investment
Federal

Net
exports

Exports

Imports

�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������
�������������

8.7
5.9
–4.5
19.5
18.4
3.1
1.5
4.3
–1.4
18.8
10.5
3.9
.7
–12.2
5.5
9.1
1.5
10.6
12.8
14.0
10.2
7.4
9.2
4.5
4.4
10.8
9.4
10.1
8.3
2.6
6.3
6.0
–12.2
3.9
7.2
7.4
7.4
10.3
9.2
–2.4
1.2
9.9
4.6
2.1
6.0
2.9
–1.5
1.1
5.5
.4
.2
–3.0
4.0
6.1
–2.5
6.1
1.6
4.4
10.1
.8
5.8
–6.2
1.5
4.1
–5.7
1.0
1.4

5.9
3.0
1.3
17.9
–.5
–1.0
–5.6
19.2
5.7
9.9
.9
–9.3
6.2
–3.9
24.6
18.9
5.6
7.9
6.3
3.8
2.6
–.2
5.7
6.5
9.9
12.2
4.8
11.1
14.2
11.0
12.0
10.9
–7.8
9.5
5.7
11.2
6.3
4.3
1.3
–5.5
–5.7
12.0
3.8
.6
3.0
6.5
3.2
3.4
5.6
3.2
–2.2
.9
.8
4.7
7.5
4.1
3.5
1.3
14.0
.6
.3
8.6
3.5
–1.5
.0
1.8
–8.7

Total
–1.2
–1.2
–2.4
–.1
–.3
3.0
3.0
–1.3
1.9
4.4
.9
.3
2.5
2.6
1.9
6.3
6.1
4.7
3.0
1.4
2.5
2.6
.0
1.3
–.7
.0
–.6
2.6
1.7
2.8
3.9
.4
4.9
3.9
1.9
.8
.9
1.9
2.3
2.5
3.0
–1.3
–3.4
–2.1
–2.4
.3
2.3
1.5
.8
1.5
3.0
3.8
–.7
1.7
1.1
–.2
1.4
–.1
2.4
1.9
2.6
2.1
–.4
2.9
4.8
1.7
2.7

Total
–3.6
–5.8
–7.3
–2.6
–3.6
3.7
.8
–1.0
2.3
3.5
1.2
4.0
6.0
4.5
2.7
7.1
6.7
5.3
3.6
–1.4
.5
1.5
–2.3
1.6
–4.5
–4.2
–4.8
1.1
.2
–.3
3.5
–2.0
5.5
8.1
6.5
2.6
1.8
2.4
3.6
6.3
6.2
1.9
–3.5
–2.6
–6.1
–1.1
1.1
.1
1.7
2.7
4.3
.7
–2.7
2.0
.6
–1.2
3.3
.1
4.6
2.8
3.9
2.9
1.1
2.2
8.3
3.3
3.6

National Nondefense defense
–4.6
–8.6
–11.5
–5.8
–5.0
1.2
.5
–2.1
.1
2.9
2.4
3.7
7.9
7.3
6.5
5.6
8.2
4.7
5.3
–.8
–1.3
.0
–4.9
–.4
–5.4
–6.7
–5.0
.3
–.8
–2.4
3.9
–3.3
4.7
8.1
8.9
2.8
1.8
3.1
3.9
7.4
4.9
1.3
–3.6
–4.7
–6.5
–3.4
–.4
–.8
1.9
4.0
4.5
–.4
–5.2
3.4
–1.0
–1.9
6.8
–1.6
4.5
.6
7.5
3.0
5.2
7.7
3.3
2.2
4.9

–0.2
3.9
5.6
6.1
–.3
9.5
1.4
1.3
6.8
4.8
–1.1
4.6
2.0
–1.6
–6.6
11.5
2.8
6.8
–1.0
–3.0
5.8
5.4
4.3
6.2
–2.5
1.1
–4.3
2.6
1.9
3.3
2.8
.1
6.7
8.2
2.5
2.4
1.9
1.3
3.1
4.2
8.6
3.0
–3.2
1.2
–5.5
2.7
3.4
1.5
1.4
.7
4.0
2.2
1.0
–.1
2.8
–.2
–1.6
2.6
4.8
6.0
–1.0
2.8
–4.5
–5.4
16.1
5.0
1.6

State
and
local
1.8
4.3
2.8
2.3
2.9
2.4
4.9
–1.6
1.7
5.2
.7
–2.9
–.7
.8
1.1
5.4
5.5
4.1
2.4
4.1
4.3
3.6
1.9
1.1
2.2
3.1
2.2
3.6
2.7
4.6
4.1
1.8
4.6
1.6
–.7
–.2
.3
1.6
1.5
.3
1.0
–3.5
–3.3
–1.7
.2
1.2
3.0
2.3
.4
.9
2.2
5.8
.5
1.6
1.4
.3
.3
–.2
1.1
1.4
1.8
1.6
–1.2
3.3
2.7
.7
2.2

Final
Final
Gross sales to Gross
sales of domestic private domestic Average
of GDP
domestic pur- domestic income and
GDI
pur(GDI) 3
product chases 1
chasers 2
2.1
.7
4.0
6.4
2.8
–1.7
3.9
3.8
4.5
6.4
2.2
.5
.3
.4
6.0
5.0
4.6
3.9
3.0
4.6
2.9
1.0
.5
4.5
2.7
3.3
3.0
4.2
3.9
5.2
4.5
3.3
1.4
1.0
4.3
3.0
3.0
2.9
2.1
–2.0
–.1
1.8
1.4
1.9
2.0
3.2
1.8
2.2
2.9
2.2
2.7
2.8
2.7
2.7
.8
3.0
2.0
2.2
4.2
2.4
4.8
.8
1.0
2.6
3.0
2.1
3.2

1.9
–.3
4.7
6.8
2.9
–2.3
2.0
5.4
5.6
6.1
.5
–1.4
1.8
–.7
9.5
6.5
4.5
2.9
4.1
3.0
2.1
–.1
.9
4.6
3.2
4.3
1.8
4.6
5.2
5.9
5.5
3.7
.3
2.8
4.3
4.0
3.2
2.1
1.1
–3.3
–.8
3.1
1.6
1.2
2.2
3.4
2.5
2.3
2.9
2.9
1.9
2.5
1.5
2.1
3.3
2.1
2.4
2.8
4.3
2.5
2.8
4.9
1.4
2.3
2.6
2.2
.6

2.9
2.1
2.1
1.1
–.8
–.5
6.5
4.8
4.6
8.3
7.1
7.0
2.2
3.8
3.9
–3.5
–2.9
–2.4
3.4
2.7
2.6
6.7
3.8
4.1
5.9
6.0
5.5
6.1
5.4
6.0
1.5
.8
1.0
–1.2
1.3
.6
.4
1.2
1.2
.8
–1.3
–1.3
9.1
6.6
7.3
5.9
6.7
6.1
4.6
3.4
3.8
3.5
2.7
2.8
2.5
5.5
5.0
4.4
4.7
4.2
2.2
1.0
1.9
–.3
1.0
.8
.3
.7
.9
5.6
3.9
4.1
4.3
3.0
2.8
4.4
4.3
4.2
3.3
2.9
2.6
4.8
4.8
4.6
5.3
5.5
5.0
6.9
4.9
4.9
5.6
4.6
4.7
4.7
3.3
3.1
.9
.1
.1
1.4
2.8
2.4
4.8
2.8
3.6
4.3
3.8
3.6
3.6
4.3
3.7
2.5
2.7
2.6
1.0
–.7
.6
–3.7
–2.7
–2.7
–2.1
.5
.3
3.3
3.5
3.0
2.6
2.1
1.9
2.6
2.9
2.2
2.6
1.5
2.1
4.5
4.2
3.5
2.5
1.3
1.6
2.8
.9
1.5
3.4
2.5
2.6
2.8
2.3
2.4
2.2 ������������� ���������������
3.0
2.1
2.1
2.9
–1.7
.1
2.8
2.0
2.1
2.4
1.4
1.7
3.4
3.8
3.1
2.5
2.6
2.4
2.2
.8
2.0
5.5
2.7
3.1
2.4
4.7
3.6
4.2
.7
2.1
2.9
3.3
3.1
1.7
.8
.9
1.6
3.2
3.2
3.3
.9
1.4
2.3
2.1
2.1
1.4 ������������� ���������������

1 Gross domestic product (GDP) less exports of goods and services plus imports of goods and services.
2 Personal consumption expenditures plus gross private fixed investment.
3 Gross domestic income is deflated by the implicit price deflator for GDP.

Note: Percent changes based on unrounded GDP quantity indexes.
Source: Department of Commerce (Bureau of Economic Analysis).

National Income or Expenditure

250-840_text_.pdf 367

| 363

2/7/20 3:46 PM

Table B–2. Contributions to percent change in real gross domestic product, 1969–2019
[Percentage points, except as noted; annual average to annual average, quarterly data at seasonally adjusted annual rates]
Personal consumption
expenditures

Year or quarter

1969 ����������������������
1970 ����������������������
1971 ����������������������
1972 ����������������������
1973 ����������������������
1974 ����������������������
1975 ����������������������
1976 ����������������������
1977 ����������������������
1978 ����������������������
1979 ����������������������
1980 ����������������������
1981 ����������������������
1982 ����������������������
1983 ����������������������
1984 ����������������������
1985 ����������������������
1986 ����������������������
1987 ����������������������
1988 ����������������������
1989 ����������������������
1990 ����������������������
1991 ����������������������
1992 ����������������������
1993 ����������������������
1994 ����������������������
1995 ����������������������
1996 ����������������������
1997 ����������������������
1998 ����������������������
1999 ����������������������
2000 ����������������������
2001 ����������������������
2002 ����������������������
2003 ����������������������
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

Gross
domestic
product
(percent
change)

3.1
.2
3.3
5.3
5.6
–.5
–.2
5.4
4.6
5.5
3.2
–.3
2.5
–1.8
4.6
7.2
4.2
3.5
3.5
4.2
3.7
1.9
–.1
3.5
2.8
4.0
2.7
3.8
4.4
4.5
4.8
4.1
1.0
1.7
2.9
3.8
3.5
2.9
1.9
–.1
–2.5
2.6
1.6
2.2
1.8
2.5
2.9
1.6
2.4
2.9
2.3
2.0
1.9
2.2
2.0
2.3
2.2
3.2
3.5
2.5
3.5
2.9
1.1
3.1
2.0
2.1
2.1

Gross private domestic investment
Fixed investment
Nonresidential

Total

2.20
1.39
2.29
3.66
2.97
–.50
1.36
3.41
2.59
2.68
1.44
–.19
.85
.88
3.51
3.30
3.20
2.58
2.15
2.65
1.86
1.28
.12
2.36
2.24
2.51
1.91
2.26
2.45
3.42
3.42
3.32
1.66
1.71
2.13
2.53
2.39
2.05
1.49
–.14
–.85
1.20
1.29
1.03
.99
1.99
2.48
1.85
1.78
2.05
1.76
2.11
1.95
1.74
1.70
1.63
1.63
1.61
3.12
1.15
2.70
2.34
.97
.78
3.03
2.12
1.20

Goods

0.92
.23
1.23
1.90
1.52
–1.08
.20
2.03
1.26
1.19
.45
–.72
.33
.19
1.69
1.91
1.38
1.45
.47
.96
.64
.16
–.49
.76
.99
1.26
.71
1.06
1.12
1.54
1.83
1.23
.72
.92
1.15
1.21
.98
.87
.65
–.71
–.70
.62
.49
.48
.70
.90
1.01
.77
.83
.86
.79
.88
.94
.84
.41
.68
1.14
.85
1.55
.27
1.13
.75
.33
.32
1.74
1.09
.26

Services

1.28
1.16
1.06
1.76
1.45
.58
1.16
1.38
1.33
1.49
.99
.53
.52
.69
1.82
1.39
1.83
1.13
1.67
1.69
1.21
1.12
.61
1.60
1.26
1.26
1.20
1.20
1.33
1.88
1.59
2.09
.94
.80
.98
1.32
1.41
1.19
.84
.56
–.15
.57
.80
.55
.29
1.10
1.46
1.08
.94
1.18
.97
1.23
1.01
.90
1.29
.95
.49
.76
1.57
.88
1.57
1.59
.65
.46
1.29
1.02
.94

Total

0.93
–1.03
1.63
1.90
1.95
–1.24
–2.91
2.91
2.47
2.22
.72
–2.07
1.64
–2.46
1.60
4.73
–.01
.03
.53
.45
.72
–.45
–1.09
1.11
1.24
1.90
.55
1.49
2.01
1.76
1.62
1.31
–1.11
–.16
.76
1.64
1.26
.60
–.48
–1.52
–3.52
1.86
.94
1.64
1.11
.95
.85
–.23
.75
.87
.32
–.26
–.28
.09
1.50
.57
.59
1.25
.80
1.07
–.30
2.27
.53
1.09
–1.16
–.17
–1.08

Total

0.93
–.33
1.08
1.85
1.47
–.98
–1.68
1.54
2.23
2.10
1.11
–1.18
.50
–1.16
1.32
2.83
1.02
.34
.11
.59
.55
–.25
–.84
.83
1.17
1.29
.99
1.48
1.49
1.82
1.65
1.34
–.27
–.64
.77
1.23
1.33
.50
–.24
–1.05
–2.70
.44
.99
1.47
.87
1.07
.58
.32
.70
.78
.23
.43
.44
.62
.33
1.27
.48
.25
1.45
.94
.89
.13
.46
.56
–.25
–.14
.01

Total
0.79
–.10
–.01
.97
1.51
.10
–1.13
.66
1.26
1.72
1.34
.00
.87
–.43
–.06
2.18
.91
–.24
.01
.63
.71
.14
–.48
.33
.84
.91
1.15
1.13
1.38
1.44
1.36
1.31
–.31
–.94
.30
.67
.92
1.00
.89
.08
–1.95
.52
1.00
1.16
.54
.95
.25
.09
.57
.84
.29
–.08
.52
.72
.09
.84
.57
.32
1.08
1.15
1.04
.29
.64
.60
–.14
–.31
–.20

Structures
0.19
.01
–.06
.12
.30
–.08
–.42
.09
.15
.52
.51
.26
.39
–.09
–.56
.58
.31
–.49
–.11
.02
.07
.05
–.38
–.18
–.01
.05
.16
.15
.21
.16
.01
.24
–.04
–.56
–.09
.00
.06
.22
.42
.23
–.72
–.50
.07
.34
.04
.33
–.10
–.16
.14
.12
–.14
–.35
.27
.50
.07
.21
.06
–.24
.15
.35
.33
–.07
–.29
.12
–.36
–.30
–.30

Equipment
0.51
–.11
.05
.75
1.12
.14
–.73
.39
1.01
1.08
.62
–.35
.28
–.47
.32
1.29
.39
.08
.03
.43
.35
–.14
–.28
.34
.73
.75
.78
.65
.76
.91
.89
.71
–.31
–.35
.26
.49
.60
.57
.25
–.29
–1.22
.92
.69
.62
.28
.42
.20
–.08
.27
.39
.08
–.24
–.14
.02
.02
.36
.50
.36
.72
.39
.20
.17
.42
.00
.05
–.22
–.17

Intellectual
property
products
0.09
.00
.01
.11
.08
.05
.01
.18
.11
.12
.20
.09
.21
.12
.17
.30
.21
.17
.10
.18
.29
.22
.18
.17
.12
.11
.20
.33
.41
.37
.45
.36
.04
–.03
.14
.18
.26
.21
.23
.14
–.02
.11
.24
.20
.22
.20
.15
.33
.16
.32
.35
.52
.39
.20
.00
.27
.01
.21
.20
.41
.51
.18
.51
.48
.17
.22
.27

Residential

0.14
–.23
1.08
.87
–.04
–1.08
–.54
.88
.97
.38
–.22
–1.19
–.37
–.72
1.38
.65
.11
.58
.10
–.05
–.16
–.38
–.35
.49
.32
.38
–.15
.35
.11
.38
.29
.03
.04
.29
.47
.57
.41
–.50
–1.13
–1.14
–.74
–.08
.00
.31
.34
.12
.33
.23
.13
–.06
–.06
.50
–.07
–.10
.24
.43
–.09
–.08
.37
–.21
–.15
–.16
–.18
–.04
–.11
.17
.21

Change
in
private
inventories
0.00
–.70
.56
.06
.48
–.26
–1.24
1.37
.24
.12
–.40
–.89
1.13
–1.31
.28
1.90
–1.03
–.31
.41
–.13
.17
–.21
–.26
.28
.07
.61
–.44
.02
.52
–.07
–.03
–.03
–.84
.48
–.02
.41
–.07
.10
–.25
–.46
–.83
1.42
–.05
.17
.23
–.12
.28
–.55
.04
.09
.09
–.68
–.72
–.53
1.18
–.70
.11
1.00
–.64
.13
–1.20
2.14
.07
.53
–.91
–.03
–1.09

See next page for continuation of table.

364 |

250-840_text_.pdf 368

Appendix B

2/7/20 3:46 PM

Table B–2. Contributions to percent change in real gross domestic product,
1969–2019—Continued

[Percentage points, except as noted; annual average to annual average, quarterly data at seasonally adjusted annual rates]
Government consumption expenditures
and gross investment

Net exports of goods and services
Year or quarter

1969 ����������������������
1970 ����������������������
1971 ����������������������
1972 ����������������������
1973 ����������������������
1974 ����������������������
1975 ����������������������
1976 ����������������������
1977 ����������������������
1978 ����������������������
1979 ����������������������
1980 ����������������������
1981 ����������������������
1982 ����������������������
1983 ����������������������
1984 ����������������������
1985 ����������������������
1986 ����������������������
1987 ����������������������
1988 ����������������������
1989 ����������������������
1990 ����������������������
1991 ����������������������
1992 ����������������������
1993 ����������������������
1994 ����������������������
1995 ����������������������
1996 ����������������������
1997 ����������������������
1998 ����������������������
1999 ����������������������
2000 ����������������������
2001 ����������������������
2002 ����������������������
2003 ����������������������
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

Net
exports
–0.03
.33
–.18
–.19
.80
.73
.86
–1.05
–.70
.05
.64
1.64
–.15
–.59
–1.32
–1.54
–.39
–.29
.17
.81
.51
.40
.62
–.04
–.56
–.41
.12
–.15
–.31
–1.14
–.87
–.83
–.22
–.64
–.45
–.67
–.29
–.10
.53
1.04
1.13
–.49
–.01
.00
.22
–.25
–.77
–.30
–.28
–.29
–.16
–.50
.35
.05
–1.36
.13
–.31
.35
–.80
.00
.67
–2.05
–.35
.73
–.68
–.14
1.48

Exports
Total
0.25
.54
.10
.42
1.08
.56
–.05
.36
.19
.80
.80
.95
.12
–.71
–.22
.61
.24
.53
.77
1.23
.97
.78
.61
.66
.31
.84
1.02
.86
1.26
.26
.52
.86
–.61
–.17
.20
.88
.69
.94
.93
.66
–1.01
1.35
.90
.46
.48
.57
.06
.00
.41
.37
.00
–.38
.45
.71
–.30
.72
.20
.54
1.19
.10
.71
–.78
.18
.49
–.69
.11
.17

Goods
0.20
.43
.00
.43
1.05
.49
–.14
.34
.12
.64
.69
.88
–.05
–.63
–.21
.41
.20
.27
.62
.99
.72
.56
.45
.52
.22
.65
.83
.68
1.10
.17
.31
.73
–.48
–.23
.19
.57
.52
.70
.53
.48
–1.00
1.12
.61
.36
.30
.42
–.03
.04
.30
.34
.01
.05
.20
.54
–.06
.46
.18
.18
1.03
.11
.94
–.78
.21
.36
–.48
.17
–.08

Imports
Services
0.05
.11
.10
–.01
.02
.08
.09
.02
.07
.17
.11
.07
.17
–.08
.00
.20
.05
.25
.15
.24
.26
.22
.16
.14
.09
.19
.19
.18
.16
.08
.20
.13
–.12
.06
.01
.31
.17
.23
.40
.18
–.01
.23
.28
.10
.18
.14
.09
–.05
.11
.03
–.02
–.43
.25
.17
–.24
.25
.01
.36
.16
.00
–.23
.00
–.03
.13
–.21
–.05
.25

Total
–0.28
–.21
–.28
–.61
–.28
.17
.91
–1.41
–.89
–.76
–.16
.69
–.26
.12
–1.10
–2.16
–.63
–.82
–.60
–.41
–.46
–.37
.01
–.70
–.87
–1.25
–.90
–1.01
–1.57
–1.39
–1.39
–1.69
.39
–.47
–.64
–1.55
–.97
–1.04
–.41
.38
2.14
–1.84
–.91
–.46
–.26
–.81
–.83
–.30
–.69
–.66
–.15
–.11
–.10
–.66
–1.06
–.58
–.51
–.18
–1.99
–.10
–.04
–1.27
–.53
.23
.01
–.26
1.32

Goods
–0.20
–.14
–.32
–.55
–.33
.17
.85
–1.31
–.82
–.66
–.13
.66
–.18
.20
–.98
–1.78
–.50
–.80
–.39
–.35
–.37
–.25
–.04
–.76
–.82
–1.15
–.84
–.91
–1.40
–1.18
–1.31
–1.44
.40
–.40
–.64
–1.30
–.88
–.82
–.28
.49
2.08
–1.74
–.82
–.38
–.25
–.75
–.73
–.18
–.57
–.61
–.04
.03
–.11
–.42
–.92
–.48
–.40
–.10
–1.86
–.18
–.10
–1.11
–.28
.36
–.02
–.13
1.44

Federal
Services
–0.08
–.07
.04
–.06
.05
.00
.06
–.10
–.07
–.10
–.02
.03
–.09
–.08
–.12
–.38
–.13
–.02
–.21
–.07
–.09
–.13
.05
.05
–.05
–.10
–.06
–.10
–.17
–.21
–.07
–.25
–.01
–.07
–.01
–.24
–.09
–.21
–.12
–.10
.06
–.10
–.09
–.09
–.01
–.06
–.10
–.12
–.12
–.05
–.12
–.15
.01
–.24
–.14
–.10
–.11
–.08
–.12
.08
.06
–.16
–.24
–.13
.02
–.13
–.12

Total
0.02
–.50
–.45
–.12
–.07
.47
.49
.12
.26
.60
.36
.36
.20
.37
.79
.74
1.37
1.14
.62
.26
.58
.65
.25
.10
–.17
.02
.10
.18
.30
.44
.58
.33
.67
.82
.41
.30
.15
.30
.34
.48
.70
.00
–.66
–.42
–.47
–.17
.35
.32
.12
.30
.41
.67
–.12
.31
.19
–.04
.24
–.02
.42
.33
.44
.36
–.07
.50
.82
.30
.47

Total
–0.34
–.80
–.80
–.37
–.39
.06
.05
.01
.21
.23
.20
.38
.43
.35
.65
.33
.78
.61
.38
–.15
.15
.20
.01
–.15
–.32
–.31
–.21
–.09
–.06
–.06
.13
.02
.24
.47
.45
.31
.15
.17
.14
.46
.47
.35
–.23
–.16
–.44
–.19
–.01
.03
.05
.19
.23
.05
–.18
.13
.04
–.08
.21
.01
.30
.18
.25
.19
.07
.14
.53
.22
.23

National Nondefense defense
–0.45
–.83
–.97
–.60
–.40
–.07
–.07
–.04
.06
.04
.15
.22
.40
.47
.51
.38
.62
.52
.38
–.04
–.02
.02
–.06
–.31
–.32
–.28
–.21
–.08
–.13
–.09
.06
–.04
.13
.30
.35
.26
.11
.07
.13
.33
.29
.16
–.12
–.18
–.34
–.19
–.09
–.02
.03
.13
.19
–.01
–.21
.13
–.04
–.07
.25
–.06
.17
.02
.28
.11
.20
.29
.13
.09
.19

0.11
.03
.17
.22
.01
.14
.13
.06
.15
.19
.05
.16
.03
–.11
.14
–.04
.16
.09
.01
–.12
.18
.18
.07
.16
.00
–.02
.00
–.01
.07
.03
.07
.06
.11
.18
.10
.05
.04
.10
.01
.13
.18
.19
–.11
.03
–.10
.00
.08
.05
.02
.07
.04
.06
.03
.00
.08
.00
–.04
.07
.13
.16
–.03
.07
–.12
–.15
.40
.13
.04

State
and
local
0.36
.30
.35
.25
.32
.41
.43
.10
.05
.37
.16
–.02
–.23
.01
.14
.41
.59
.53
.24
.42
.43
.45
.24
.25
.15
.32
.31
.27
.36
.50
.46
.31
.43
.35
–.03
–.01
.00
.13
.20
.02
.23
–.35
–.44
–.26
–.03
.02
.35
.29
.07
.11
.18
.63
.06
.18
.15
.03
.03
–.02
.12
.15
.19
.17
–.14
.36
.29
.08
.23

Final
sales of
domestic
product
3.12
.89
2.74
5.20
5.16
–.28
1.03
4.01
4.38
5.42
3.56
.63
1.41
–.50
4.31
5.34
5.20
3.77
3.05
4.31
3.51
2.09
.15
3.24
2.68
3.41
3.13
3.76
3.92
4.55
4.78
4.16
1.84
1.26
2.88
3.39
3.59
2.75
2.12
.33
–1.71
1.14
1.60
2.08
1.61
2.65
2.63
2.19
2.33
2.84
2.24
2.71
2.62
2.72
.85
2.99
2.04
2.20
4.19
2.42
4.71
.78
1.02
2.57
2.92
2.13
3.17

Source: Department of Commerce (Bureau of Economic Analysis).

National Income or Expenditure

250-840_text_.pdf 369

| 365

2/7/20 3:46 PM

Table B–3. Gross domestic product, 2004–2019
[Quarterly data at seasonally adjusted annual rates]

Personal consumption
expenditures

Year or quarter

Gross
domestic
product

Gross private domestic investment
Fixed investment
Nonresidential

Total

Goods

Services

Total

Total

Total

Structures

Equipment

Intellectual
property
products

Residential

Change
in
private
inventories

Billions of dollars
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

12,213.7
13,036.6
13,814.6
14,451.9
14,712.8
14,448.9
14,992.1
15,542.6
16,197.0
16,784.9
17,527.3
18,224.8
18,715.0
19,519.4
20,580.2
21,429.0
18,424.3
18,637.3
18,806.7
18,991.9
19,190.4
19,356.6
19,611.7
19,918.9
20,163.2
20,510.2
20,749.8
20,897.8
21,098.8
21,340.3
21,542.5
21,734.3

8,212.7
8,747.1
9,260.3
9,706.4
9,976.3
9,842.2
10,185.8
10,641.1
11,006.8
11,317.2
11,822.8
12,284.3
12,748.5
13,312.1
13,998.7
14,563.9
12,523.5
12,688.3
12,822.4
12,959.8
13,104.4
13,212.5
13,345.1
13,586.3
13,728.4
13,939.8
14,114.6
14,211.9
14,266.3
14,511.2
14,678.2
14,799.8

2,902.0
3,082.9
3,239.7
3,367.0
3,363.2
3,180.0
3,317.8
3,518.1
3,637.7
3,730.0
3,863.0
3,920.3
3,995.9
4,165.0
4,364.8
4,508.6
3,933.2
3,988.6
4,017.8
4,044.0
4,097.9
4,124.9
4,173.3
4,264.0
4,298.5
4,363.2
4,398.0
4,399.4
4,397.7
4,507.0
4,556.7
4,573.1

5,310.6
5,664.2
6,020.7
6,339.4
6,613.1
6,662.2
6,868.0
7,123.0
7,369.1
7,587.2
7,959.8
8,363.9
8,752.6
9,147.0
9,633.9
10,055.2
8,590.3
8,699.6
8,804.6
8,915.8
9,006.5
9,087.6
9,171.8
9,322.3
9,429.8
9,576.6
9,716.6
9,812.5
9,868.6
10,004.2
10,121.5
10,226.7

2,281.3
2,534.7
2,701.0
2,673.0
2,477.6
1,929.7
2,165.5
2,332.6
2,621.8
2,826.0
3,044.2
3,223.1
3,178.7
3,370.7
3,628.3
3,742.8
3,149.1
3,152.9
3,166.6
3,246.2
3,288.2
3,335.0
3,401.8
3,457.7
3,542.4
3,561.6
3,684.0
3,725.2
3,783.4
3,749.5
3,744.6
3,693.9

2004 ���������������������� 14,406.4 9,729.3
2005 ���������������������� 14,912.5 10,075.9
2006 ���������������������� 15,338.3 10,384.5
2007 ���������������������� 15,626.0 10,615.3
2008 ���������������������� 15,604.7 10,592.8
2009 ���������������������� 15,208.8 10,460.0
2010 ���������������������� 15,598.8 10,643.0
2011 ���������������������� 15,840.7 10,843.8
2012 ���������������������� 16,197.0 11,006.8
2013 ���������������������� 16,495.4 11,166.9
2014 ���������������������� 16,912.0 11,497.4
2015 ���������������������� 17,403.8 11,921.2
2016 ���������������������� 17,688.9 12,247.5
2017 ���������������������� 18,108.1 12,566.9
2018 ���������������������� 18,638.2 12,944.6
2019 p �������������������� 19,072.5 13,279.6
2016: I ������������������ 17,556.8 12,124.2
      II ����������������� 17,639.4 12,211.3
      III ���������������� 17,735.1 12,289.1
      IV ���������������� 17,824.2 12,365.3
2017: I ������������������ 17,925.3 12,438.9
      II ����������������� 18,021.0 12,512.9
      III ���������������� 18,163.6 12,586.3
      IV ���������������� 18,322.5 12,729.7
2018: I ������������������ 18,438.3 12,782.9
      II ����������������� 18,598.1 12,909.2
      III ���������������� 18,732.7 13,019.8
      IV ���������������� 18,783.5 13,066.3
2019: I ������������������ 18,927.3 13,103.3
      II ����������������� 19,021.9 13,250.0
      III ���������������� 19,121.1 13,353.1
      IV p ������������� 19,219.8 13,411.9
See next page for continuation of table.

3,250.0
3,384.7
3,509.7
3,607.6
3,498.9
3,389.8
3,485.7
3,561.8
3,637.7
3,752.2
3,905.1
4,088.6
4,236.6
4,403.4
4,583.3
4,756.6
4,176.2
4,222.4
4,263.8
4,284.2
4,318.2
4,375.9
4,419.7
4,499.8
4,513.9
4,573.5
4,614.0
4,631.8
4,649.2
4,746.4
4,808.0
4,822.8

6,479.2
6,689.5
6,871.7
7,003.6
7,093.0
7,070.1
7,157.4
7,282.1
7,369.1
7,415.5
7,594.9
7,838.5
8,021.1
8,182.2
8,388.1
8,560.8
7,955.8
7,998.9
8,037.2
8,092.2
8,133.0
8,154.1
8,186.6
8,254.9
8,293.5
8,362.9
8,433.6
8,462.6
8,483.1
8,541.4
8,587.9
8,630.9

2,502.6
2,670.6
2,752.4
2,684.1
2,462.9
1,942.0
2,216.5
2,362.1
2,621.8
2,801.5
2,959.2
3,104.3
3,064.0
3,198.9
3,360.5
3,421.2
3,054.7
3,041.6
3,045.5
3,114.0
3,140.3
3,167.9
3,225.2
3,262.1
3,311.8
3,296.6
3,404.2
3,429.5
3,481.1
3,424.7
3,416.2
3,363.0

2,217.2
2,477.2
2,632.0
2,639.1
2,506.9
2,080.4
2,111.6
2,286.3
2,550.5
2,721.5
2,960.2
3,091.2
3,151.6
3,340.5
3,573.6
3,676.1
3,102.2
3,133.8
3,169.3
3,201.3
3,274.8
3,316.1
3,345.0
3,426.0
3,500.9
3,571.6
3,596.7
3,625.2
3,670.1
3,674.7
3,677.6
3,682.0

1,467.4
1,621.0
1,793.8
1,948.6
1,990.9
1,690.4
1,735.0
1,907.5
2,118.5
2,211.5
2,400.1
2,457.4
2,453.1
2,584.7
2,786.9
2,878.7
2,415.6
2,441.8
2,471.6
2,483.5
2,531.1
2,567.4
2,591.6
2,648.9
2,717.3
2,782.0
2,807.7
2,840.7
2,882.7
2,890.0
2,877.2
2,864.9

307.7
353.0
425.2
510.3
571.1
455.8
379.8
404.5
479.4
492.5
577.6
572.6
545.8
586.8
633.2
625.8
520.5
537.1
559.6
566.0
580.2
589.0
583.7
594.4
615.9
640.0
641.7
635.2
645.8
633.2
619.4
604.7

721.9
794.9
862.3
893.4
845.4
670.3
777.0
881.3
983.4
1,027.0
1,091.9
1,121.5
1,093.6
1,143.7
1,222.6
1,240.9
1,101.4
1,092.7
1,091.2
1,088.9
1,108.8
1,132.9
1,149.5
1,183.6
1,201.8
1,214.3
1,227.9
1,246.4
1,249.0
1,252.9
1,237.4
1,224.4

437.8
473.1
506.3
544.8
574.4
564.4
578.2
621.7
655.7
691.9
730.5
763.3
813.8
854.2
931.1
1,012.0
793.8
812.1
820.9
828.6
842.1
845.5
858.4
870.9
899.6
927.7
938.1
959.1
987.9
1,003.9
1,020.5
1,035.8

749.8
856.2
838.2
690.5
516.0
390.0
376.6
378.8
432.0
510.0
560.2
633.8
698.5
755.7
786.7
797.4
686.6
692.0
697.7
717.8
743.7
748.8
753.4
777.1
783.7
789.5
789.0
784.4
787.4
784.7
800.3
817.1

64.1
57.5
69.0
34.0
–29.2
–150.8
53.9
46.3
71.2
104.5
84.0
131.9
27.1
30.2
54.7
66.8
46.9
19.1
–2.7
44.9
13.4
18.8
56.8
31.7
41.5
–10.0
87.3
100.1
113.3
74.8
67.0
11.9

456.3
466.1
501.7
568.6
605.4
492.2
412.8
424.1
479.4
485.5
538.8
522.4
496.4
519.5
540.9
516.8
476.4
487.9
509.0
512.1
521.1
523.7
513.3
519.9
534.9
549.1
546.2
533.4
538.6
523.0
509.6
496.2

688.6
760.0
832.6
865.8
824.4
649.7
781.2
886.2
983.4
1,029.2
1,101.1
1,136.6
1,122.3
1,175.6
1,255.3
1,272.4
1,126.5
1,120.0
1,120.9
1,122.0
1,139.3
1,163.8
1,181.4
1,217.8
1,237.5
1,247.8
1,256.7
1,279.2
1,278.9
1,281.5
1,269.3
1,259.9

459.2
493.1
521.5
554.3
575.3
572.4
588.1
624.8
655.7
691.4
724.8
750.7
810.0
839.6
901.6
971.1
792.0
809.8
819.2
819.2
831.8
832.3
842.3
852.0
872.0
896.9
905.9
931.3
955.6
964.2
975.2
989.3

830.9
885.4
818.9
665.8
504.6
395.3
383.0
382.5
432.0
485.5
504.1
555.3
591.2
611.9
602.9
593.5
593.0
590.1
586.2
595.5
612.4
608.9
605.9
620.4
612.1
606.3
600.1
593.0
591.4
587.0
593.7
602.1

82.6
63.7
87.1
40.6
–32.7
–177.3
57.3
46.7
71.2
108.7
86.3
132.4
23.0
31.7
48.1
65.3
51.1
10.8
–14.7
44.8
8.7
16.6
70.2
31.1
40.5
–28.0
87.2
93.0
116.0
69.4
69.4
6.5

Billions of chained (2012) dollars

366 |

250-840_text_.pdf 370

2,440.7
2,618.7
2,686.8
2,653.5
2,499.4
2,099.8
2,164.2
2,317.8
2,550.5
2,692.1
2,869.2
2,967.0
3,023.6
3,149.7
3,293.4
3,337.1
2,991.0
3,010.9
3,038.9
3,053.7
3,111.1
3,133.0
3,144.1
3,210.7
3,254.0
3,295.4
3,301.3
3,323.0
3,349.4
3,337.4
3,330.5
3,331.0

1,594.0
1,716.4
1,854.2
1,982.1
1,994.2
1,704.3
1,781.0
1,935.4
2,118.5
2,206.0
2,365.3
2,408.2
2,425.3
2,531.2
2,692.3
2,749.8
2,389.8
2,413.6
2,446.8
2,451.2
2,490.5
2,517.4
2,532.6
2,584.2
2,639.5
2,689.9
2,703.9
2,735.8
2,765.6
2,758.5
2,742.7
2,732.4

Appendix B

2/7/20 3:46 PM

Table B–3. Gross domestic product, 2004–2019—Continued
[Quarterly data at seasonally adjusted annual rates]

Net exports of
goods and services
Year or quarter

Government consumption expenditures
and gross investment
Federal

Net
exports

Exports

Imports

Total

2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

–619.1
–721.2
–770.9
–718.4
–723.1
–396.5
–513.9
–579.5
–568.6
–490.8
–507.7
–519.8
–518.8
–575.3
–638.2
–632.0
–522.2
–495.3
–499.7
–558.0
–570.9
–583.7
–550.6
–596.1
–629.0
–568.4
–671.4
–684.1
–633.8
–662.7
–653.0
–578.4

1,177.6
1,305.2
1,472.6
1,660.9
1,837.1
1,582.0
1,846.3
2,103.0
2,191.3
2,273.4
2,371.7
2,266.8
2,220.6
2,356.7
2,510.3
2,503.8
2,164.9
2,208.1
2,254.4
2,255.1
2,303.3
2,313.2
2,360.1
2,450.3
2,476.6
2,543.6
2,510.3
2,510.5
2,520.3
2,504.0
2,495.1
2,495.6

1,796.7
2,026.4
2,243.5
2,379.3
2,560.1
1,978.4
2,360.2
2,682.5
2,759.9
2,764.2
2,879.4
2,786.6
2,739.4
2,932.1
3,148.5
3,135.7
2,687.1
2,703.4
2,754.1
2,813.1
2,874.2
2,896.9
2,910.7
3,046.5
3,105.6
3,112.0
3,181.6
3,194.7
3,154.1
3,166.7
3,148.2
3,074.0

2,338.9
2,476.0
2,624.2
2,790.8
2,982.0
3,073.5
3,154.6
3,148.4
3,137.0
3,132.4
3,168.0
3,237.3
3,306.7
3,412.0
3,591.5
3,754.3
3,273.8
3,291.4
3,317.5
3,343.9
3,368.7
3,392.9
3,415.4
3,471.0
3,521.4
3,577.1
3,622.6
3,644.8
3,683.1
3,742.3
3,772.8
3,818.9

891.7
947.5
1,000.7
1,050.5
1,150.6
1,218.2
1,297.9
1,298.9
1,286.5
1,226.6
1,215.0
1,221.5
1,234.1
1,269.3
1,347.3
1,423.4
1,227.5
1,226.2
1,237.5
1,245.2
1,248.4
1,263.6
1,270.2
1,295.1
1,318.2
1,340.4
1,358.6
1,371.8
1,394.7
1,415.2
1,432.2
1,451.6

2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

–841.4
–887.8
–905.0
–823.6
–661.6
–484.8
–565.9
–568.1
–568.6
–532.8
–577.2
–721.6
–783.7
–849.8
–920.0
–954.2
–777.7
–760.9
–761.4
–834.6
–831.5
–850.0
–833.7
–883.8
–884.2
–850.5
–962.4
–983.0
–944.0
–980.7
–990.1
–902.0

1,431.2
1,533.2
1,676.4
1,822.3
1,925.4
1,763.8
1,977.9
2,119.0
2,191.3
2,269.6
2,365.3
2,376.5
2,376.1
2,458.8
2,532.9
2,531.9
2,345.1
2,367.9
2,403.4
2,388.1
2,423.5
2,432.9
2,459.5
2,519.2
2,524.0
2,559.9
2,519.3
2,528.5
2,554.4
2,517.5
2,523.4
2,532.4

2,272.6
2,421.0
2,581.5
2,646.0
2,587.1
2,248.6
2,543.8
2,687.1
2,759.9
2,802.4
2,942.5
3,098.1
3,159.8
3,308.5
3,453.0
3,486.1
3,122.7
3,128.9
3,164.9
3,222.7
3,255.0
3,282.9
3,293.2
3,403.0
3,408.2
3,410.4
3,481.8
3,511.6
3,498.3
3,498.2
3,513.6
3,434.4

2,992.7
3,015.5
3,063.5
3,118.6
3,195.6
3,307.3
3,307.2
3,203.3
3,137.0
3,061.0
3,033.4
3,091.8
3,147.7
3,169.6
3,223.9
3,299.4
3,143.0
3,137.5
3,151.0
3,159.3
3,157.3
3,168.0
3,167.1
3,186.1
3,201.1
3,221.4
3,238.0
3,234.9
3,258.1
3,296.6
3,310.4
3,332.4

1,077.5
1,099.1
1,125.0
1,147.0
1,218.8
1,293.0
1,346.1
1,311.1
1,286.5
1,215.3
1,183.8
1,182.7
1,187.8
1,197.0
1,232.2
1,275.7
1,190.6
1,182.5
1,188.2
1,189.9
1,186.4
1,195.9
1,196.1
1,209.8
1,218.1
1,229.9
1,238.7
1,242.1
1,248.8
1,273.9
1,284.4
1,295.7

Total

National Nondefense defense

State
and
local

Final
Final
Gross sales to Gross
sales of domestic private domestic Average
of GDP
domestic pur- domestic income and
GDI
pur(GDI) 3
product chases 1
chasers 2

Billions of dollars
569.9
609.4
640.8
679.3
750.3
787.6
828.0
834.0
814.2
764.2
743.4
730.1
728.4
746.2
793.6
846.6
727.6
722.3
731.3
732.3
732.1
746.2
746.2
760.4
769.9
789.5
800.6
814.4
831.8
841.6
849.3
863.9

321.9
338.0
359.9
371.2
400.2
430.6
469.9
465.0
472.4
462.4
471.6
491.4
505.7
523.1
553.7
576.8
500.0
503.9
506.1
512.9
516.3
517.4
524.0
534.8
548.3
550.9
558.0
557.4
562.9
573.5
583.0
587.7

1,447.1
1,528.5
1,623.5
1,740.3
1,831.4
1,855.3
1,856.7
1,849.4
1,850.5
1,905.8
1,953.0
2,015.7
2,072.6
2,142.7
2,244.2
2,330.8
2,046.3
2,065.2
2,080.0
2,098.7
2,120.3
2,129.3
2,145.2
2,175.9
2,203.2
2,236.7
2,263.9
2,273.0
2,288.4
2,327.1
2,340.5
2,367.3

12,149.7
12,979.1
13,745.6
14,417.9
14,742.1
14,599.7
14,938.1
15,496.3
16,125.8
16,680.3
17,443.3
18,092.9
18,688.0
19,489.2
20,525.5
21,362.2
18,377.4
18,618.1
18,809.5
18,946.9
19,177.0
19,337.8
19,554.9
19,887.2
20,121.7
20,520.1
20,662.4
20,797.7
20,985.5
21,265.5
21,475.5
21,722.4

12,832.8
13,757.8
14,585.5
15,170.3
15,435.9
14,845.4
15,506.0
16,122.0
16,765.6
17,275.6
18,034.9
18,744.6
19,233.8
20,094.8
21,218.4
22,061.0
18,946.5
19,132.6
19,306.5
19,549.8
19,761.4
19,940.4
20,162.3
20,515.0
20,792.1
21,078.6
21,421.1
21,582.0
21,732.7
22,002.9
22,195.6
22,312.7

10,429.8
11,224.3
11,892.3
12,345.5
12,483.2
11,922.6
12,297.4
12,927.4
13,557.4
14,038.7
14,783.0
15,375.5
15,900.1
16,652.6
17,572.2
18,239.9
15,625.7
15,822.0
15,991.7
16,161.0
16,379.2
16,528.6
16,690.0
17,012.3
17,229.3
17,511.4
17,711.2
17,837.1
17,936.3
18,185.9
18,355.8
18,481.8

12,235.8 12,224.8
13,091.7 13,064.2
14,022.5 13,918.6
14,434.2 14,443.0
14,530.0 14,621.4
14,256.8 14,352.9
14,931.0 14,961.5
15,595.8 15,569.2
16,438.4 16,317.7
16,945.2 16,865.0
17,816.4 17,671.8
18,479.7 18,352.2
18,827.0 18,771.0
19,587.0 19,553.2
20,569.4 20,574.8
������������� ���������������
18,673.5 18,548.9
18,718.3 18,677.8
18,880.6 18,843.7
19,035.5 19,013.7
19,307.0 19,248.7
19,496.9 19,426.8
19,638.4 19,625.0
19,905.6 19,912.3
20,252.2 20,207.7
20,460.1 20,485.1
20,716.9 20,733.3
20,848.6 20,873.2
21,056.7 21,077.8
21,237.8 21,289.0
21,440.4 21,491.5
������������� ���������������

14,335.7
14,852.3
15,263.0
15,588.7
15,639.7
15,373.0
15,546.6
15,796.5
16,125.8
16,386.2
16,822.3
17,267.1
17,647.6
18,058.4
18,571.3
18,988.7
17,492.6
17,607.5
17,726.7
17,763.5
17,895.1
17,985.3
18,082.5
18,270.7
18,380.4
18,595.6
18,630.9
18,678.3
18,797.5
18,935.2
19,035.7
19,186.4

15,254.1
15,804.5
16,246.7
16,454.6
16,270.7
15,698.9
16,164.7
16,408.8
16,765.6
17,028.6
17,487.7
18,114.2
18,455.9
18,931.2
19,523.2
19,994.4
18,318.9
18,387.3
18,482.5
18,635.1
18,732.7
18,844.8
18,974.1
19,173.1
19,290.7
19,422.1
19,656.0
19,724.2
19,836.1
19,965.4
20,073.7
20,102.2

12,194.2
12,725.8
13,102.6
13,293.8
13,108.0
12,557.6
12,805.7
13,161.2
13,557.4
13,858.9
14,366.5
14,888.0
15,270.8
15,716.4
16,237.8
16,616.3
15,114.9
15,221.9
15,327.6
15,418.7
15,549.7
15,645.6
15,730.1
15,940.2
16,036.7
16,204.4
16,320.9
16,389.2
16,452.7
16,587.1
16,683.1
16,742.4

14,432.4 14,419.4
14,975.5 14,944.0
15,569.1 15,453.7
15,606.9 15,616.5
15,410.8 15,507.7
15,006.6 15,107.7
15,535.2 15,567.0
15,894.9 15,867.8
16,438.4 16,317.7
16,652.9 16,574.1
17,191.1 17,051.5
17,647.3 17,525.6
17,794.7 17,741.8
18,170.8 18,139.4
18,628.4 18,633.3
������������� ���������������
17,794.3 17,675.6
17,716.2 17,677.8
17,804.7 17,769.9
17,865.2 17,844.7
18,034.1 17,979.7
18,151.7 18,086.3
18,188.3 18,175.9
18,310.2 18,316.3
18,519.7 18,479.0
18,552.7 18,575.4
18,703.1 18,717.9
18,739.3 18,761.4
18,889.5 18,908.4
18,930.5 18,976.2
19,030.5 19,075.8
������������� ���������������

Billions of chained (2012) dollars
692.7
708.6
719.8
740.3
791.5
836.7
861.3
842.9
814.2
759.6
728.4
713.0
708.7
714.0
737.5
773.6
713.2
703.8
709.8
708.1
704.7
716.4
713.4
721.4
722.5
735.7
741.2
750.6
764.5
770.8
775.0
784.3

384.8
390.6
405.3
406.7
427.3
456.3
484.8
468.3
472.4
455.6
455.2
469.3
478.5
482.4
494.2
501.9
476.8
478.0
477.8
481.1
480.9
479.0
482.0
487.7
494.9
493.6
497.0
491.3
484.5
502.9
509.1
511.1

1,920.1
1,920.1
1,941.6
1,974.7
1,978.7
2,015.6
1,961.3
1,892.2
1,850.5
1,845.3
1,848.6
1,907.5
1,957.9
1,970.6
1,990.0
2,022.5
1,950.5
1,953.0
1,960.8
1,967.4
1,968.9
1,970.1
1,969.0
1,974.5
1,981.2
1,989.9
1,997.7
1,991.4
2,007.9
2,021.4
2,024.9
2,035.8

1 Gross domestic product (GDP) less exports of goods and services plus imports of goods and services.
2 Personal consumption expenditures plus gross private fixed investment.
3 For chained dollar measures, gross domestic income is deflated by the implicit price deflator for GDP.

Source: Department of Commerce (Bureau of Economic Analysis).

National Income or Expenditure

250-840_text_.pdf 371

| 367

2/7/20 3:46 PM

Table B–4. Percentage shares of gross domestic product, 1969–2019
[Percent of nominal GDP]

Personal consumption
expenditures

Year or quarter

1969 ����������������������
1970 ����������������������
1971 ����������������������
1972 ����������������������
1973 ����������������������
1974 ����������������������
1975 ����������������������
1976 ����������������������
1977 ����������������������
1978 ����������������������
1979 ����������������������
1980 ����������������������
1981 ����������������������
1982 ����������������������
1983 ����������������������
1984 ����������������������
1985 ����������������������
1986 ����������������������
1987 ����������������������
1988 ����������������������
1989 ����������������������
1990 ����������������������
1991 ����������������������
1992 ����������������������
1993 ����������������������
1994 ����������������������
1995 ����������������������
1996 ����������������������
1997 ����������������������
1998 ����������������������
1999 ����������������������
2000 ����������������������
2001 ����������������������
2002 ����������������������
2003 ����������������������
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

Gross
domestic
product
(percent)

100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0

Gross private domestic investment
Fixed investment
Nonresidential

Total

59.3
60.3
60.1
60.1
59.6
60.2
61.2
61.3
61.2
60.5
60.3
61.3
60.3
61.9
62.8
61.7
62.5
63.0
63.4
63.6
63.4
63.9
64.0
64.4
64.9
64.8
65.0
65.0
64.5
64.9
65.2
66.0
66.8
67.1
67.4
67.2
67.1
67.0
67.2
67.8
68.1
67.9
68.5
68.0
67.4
67.5
67.4
68.1
68.2
68.0
68.0
68.0
68.1
68.2
68.2
68.3
68.3
68.0
68.2
68.1
68.0
68.0
68.0
67.6
68.0
68.1
68.1

Goods

29.9
29.7
29.4
29.2
29.2
29.2
29.2
29.2
28.8
28.2
28.1
28.0
27.1
26.9
26.8
26.3
26.2
26.1
25.9
25.5
25.2
25.0
24.3
24.0
23.9
24.0
23.8
23.8
23.4
23.3
23.7
23.9
23.9
23.8
23.8
23.8
23.6
23.5
23.3
22.9
22.0
22.1
22.6
22.5
22.2
22.0
21.5
21.4
21.3
21.2
21.0
21.3
21.4
21.4
21.3
21.4
21.3
21.3
21.4
21.3
21.3
21.2
21.1
20.8
21.1
21.2
21.0

Services

29.4
30.6
30.7
30.8
30.4
31.0
32.0
32.1
32.4
32.3
32.3
33.3
33.2
35.0
36.0
35.4
36.3
36.9
37.5
38.1
38.2
38.9
39.7
40.4
41.0
40.8
41.2
41.2
41.2
41.6
41.5
42.0
42.9
43.4
43.6
43.5
43.4
43.6
43.9
44.9
46.1
45.8
45.8
45.5
45.2
45.4
45.9
46.8
46.9
46.8
46.9
46.6
46.7
46.8
46.9
46.9
46.9
46.8
46.8
46.8
46.7
46.8
47.0
46.8
46.9
47.0
47.1

Total

17.1
15.8
16.9
17.8
18.7
17.8
15.3
17.3
19.1
20.3
20.5
18.6
19.7
17.4
17.5
20.3
19.1
18.5
18.4
17.9
17.7
16.7
15.3
15.5
16.1
17.2
17.2
17.7
18.6
19.2
19.6
19.9
18.3
17.7
17.7
18.7
19.4
19.6
18.5
16.8
13.4
14.4
15.0
16.2
16.8
17.4
17.7
17.0
17.3
17.6
17.5
17.1
16.9
16.8
17.1
17.1
17.2
17.3
17.4
17.6
17.4
17.8
17.8
17.9
17.6
17.4
17.0

Total

16.2
15.7
16.2
17.1
17.6
16.9
15.6
16.3
18.0
19.2
19.9
18.8
18.8
17.8
17.7
18.7
18.6
18.4
17.8
17.5
17.2
16.4
15.3
15.3
15.8
16.4
16.8
17.4
17.8
18.5
19.0
19.4
18.6
17.5
17.6
18.2
19.0
19.1
18.3
17.0
14.4
14.1
14.7
15.7
16.2
16.9
17.0
16.8
17.1
17.4
17.2
16.8
16.8
16.9
16.9
17.1
17.1
17.1
17.2
17.4
17.4
17.3
17.3
17.4
17.2
17.1
16.9

Total
11.8
11.6
11.2
11.5
12.1
12.4
11.7
11.7
12.4
13.4
14.2
14.2
14.7
14.5
13.3
14.0
14.0
13.3
12.7
12.6
12.7
12.4
11.8
11.4
11.7
11.9
12.6
12.9
13.4
13.8
14.2
14.6
13.8
12.4
12.0
12.0
12.4
13.0
13.5
13.5
11.7
11.6
12.3
13.1
13.2
13.7
13.5
13.1
13.2
13.5
13.4
13.1
13.1
13.1
13.1
13.2
13.3
13.2
13.3
13.5
13.6
13.5
13.6
13.7
13.5
13.4
13.2

Structures
3.7
3.8
3.7
3.7
3.9
4.0
3.6
3.5
3.6
4.0
4.5
4.8
5.2
5.3
4.2
4.4
4.5
3.9
3.6
3.5
3.4
3.4
3.0
2.6
2.6
2.6
2.7
2.8
2.9
3.0
3.0
3.1
3.2
2.6
2.5
2.5
2.7
3.1
3.5
3.9
3.2
2.5
2.6
3.0
2.9
3.3
3.1
2.9
3.0
3.1
2.9
2.8
2.9
3.0
3.0
3.0
3.0
3.0
3.0
3.1
3.1
3.1
3.0
3.1
3.0
2.9
2.8

Equipment
6.4
6.2
5.9
6.2
6.7
6.8
6.4
6.5
7.1
7.7
7.9
7.6
7.5
7.0
6.8
7.2
7.1
6.9
6.6
6.6
6.6
6.2
5.9
5.9
6.2
6.5
6.9
7.0
7.1
7.3
7.4
7.5
6.7
6.0
5.9
5.9
6.1
6.2
6.2
5.7
4.6
5.2
5.7
6.1
6.1
6.2
6.2
5.8
5.9
5.9
5.8
6.0
5.9
5.8
5.7
5.8
5.9
5.9
5.9
6.0
5.9
5.9
6.0
5.9
5.9
5.7
5.6

Intellectual
property
products
1.7
1.7
1.6
1.6
1.6
1.7
1.7
1.7
1.7
1.7
1.8
1.9
2.0
2.2
2.2
2.4
2.4
2.5
2.5
2.5
2.7
2.8
2.9
2.9
2.9
2.8
3.0
3.1
3.4
3.5
3.8
4.0
3.9
3.7
3.7
3.6
3.6
3.7
3.8
3.9
3.9
3.9
4.0
4.0
4.1
4.2
4.2
4.3
4.4
4.5
4.7
4.3
4.4
4.4
4.4
4.4
4.4
4.4
4.4
4.5
4.5
4.5
4.6
4.7
4.7
4.7
4.8

Residential

4.4
4.0
5.0
5.7
5.5
4.5
4.0
4.6
5.5
5.9
5.6
4.5
4.0
3.3
4.4
4.7
4.6
5.1
5.1
4.9
4.5
4.0
3.6
3.9
4.2
4.4
4.2
4.4
4.4
4.6
4.8
4.7
4.8
5.1
5.6
6.1
6.6
6.1
4.8
3.5
2.7
2.5
2.4
2.7
3.0
3.2
3.5
3.7
3.9
3.8
3.7
3.7
3.7
3.7
3.8
3.9
3.9
3.8
3.9
3.9
3.8
3.8
3.8
3.7
3.7
3.7
3.8

Change
in
private
inventories
0.9
.2
.7
.7
1.1
.9
–.4
.9
1.1
1.1
.7
–.2
.9
–.4
–.2
1.6
.5
.1
.6
.4
.5
.2
.0
.3
.3
.9
.4
.4
.8
.7
.6
.5
–.4
.2
.1
.5
.4
.5
.2
–.2
–1.0
.4
.3
.4
.6
.5
.7
.1
.2
.3
.3
.3
.1
.0
.2
.1
.1
.3
.2
.2
.0
.4
.5
.5
.4
.3
.1

See next page for continuation of table.

368 |

250-840_text_.pdf 372

Appendix B

2/7/20 3:46 PM

Table B–4. Percentage shares of gross domestic product, 1969–2019—Continued
[Percent of nominal GDP]

Government consumption expenditures
and gross investment

Net exports of goods and services
Year or quarter

1969 �����������������������
1970 �����������������������
1971 �����������������������
1972 �����������������������
1973 �����������������������
1974 �����������������������
1975 �����������������������
1976 �����������������������
1977 �����������������������
1978 �����������������������
1979 �����������������������
1980 �����������������������
1981 �����������������������
1982 �����������������������
1983 �����������������������
1984 �����������������������
1985 �����������������������
1986 �����������������������
1987 �����������������������
1988 �����������������������
1989 �����������������������
1990 �����������������������
1991 �����������������������
1992 �����������������������
1993 �����������������������
1994 �����������������������
1995 �����������������������
1996 �����������������������
1997 �����������������������
1998 �����������������������
1999 �����������������������
2000 �����������������������
2001 �����������������������
2002 �����������������������
2003 �����������������������
2004 �����������������������
2005 �����������������������
2006 �����������������������
2007 �����������������������
2008 �����������������������
2009 �����������������������
2010 �����������������������
2011 �����������������������
2012 �����������������������
2013 �����������������������
2014 �����������������������
2015 �����������������������
2016 �����������������������
2017 �����������������������
2018 �����������������������
2019 p ���������������������
2016: I �������������������
      II ������������������
      III �����������������
      IV �����������������
2017: I �������������������
      II ������������������
      III �����������������
      IV �����������������
2018: I �������������������
      II ������������������
      III �����������������
      IV �����������������
2019: I �������������������
      II ������������������
      III �����������������
      IV p ��������������

Net
exports
0.1
.4
.1
–.3
.3
–.1
.9
–.1
–1.1
–1.1
–.9
–.5
–.4
–.6
–1.4
–2.5
–2.6
–2.9
–3.0
–2.1
–1.5
–1.3
–.5
–.5
–1.0
–1.3
–1.2
–1.2
–1.2
–1.8
–2.7
–3.7
–3.5
–3.9
–4.4
–5.1
–5.5
–5.6
–5.0
–4.9
–2.7
–3.4
–3.7
–3.5
–2.9
–2.9
–2.9
–2.8
–2.9
–3.1
–2.9
–2.8
–2.7
–2.7
–2.9
–3.0
–3.0
–2.8
–3.0
–3.1
–2.8
–3.2
–3.3
–3.0
–3.1
–3.0
–2.7

Exports
Total
5.1
5.6
5.4
5.5
6.7
8.2
8.2
8.0
7.7
7.9
8.8
9.8
9.5
8.5
7.6
7.5
7.0
7.0
7.5
8.5
8.9
9.3
9.7
9.7
9.5
9.9
10.6
10.7
11.1
10.5
10.3
10.7
9.7
9.1
9.0
9.6
10.0
10.7
11.5
12.5
10.9
12.3
13.5
13.5
13.5
13.5
12.4
11.9
12.1
12.2
11.7
11.8
11.8
12.0
11.9
12.0
12.0
12.0
12.3
12.3
12.4
12.1
12.0
11.9
11.7
11.6
11.5

Goods
3.8
4.2
4.0
4.1
5.3
6.7
6.7
6.5
6.2
6.4
7.1
8.1
7.6
6.7
5.9
5.7
5.2
5.1
5.5
6.3
6.6
6.8
7.0
7.0
6.8
7.1
7.8
7.8
8.2
7.6
7.4
7.8
7.0
6.5
6.4
6.8
7.1
7.6
8.0
8.8
7.3
8.5
9.4
9.4
9.3
9.2
8.2
7.7
7.9
8.1
7.7
7.6
7.7
7.8
7.7
7.8
7.8
7.8
8.1
8.1
8.3
8.0
7.9
7.9
7.7
7.6
7.5

Imports
Services
1.3
1.4
1.4
1.4
1.4
1.5
1.6
1.5
1.5
1.6
1.6
1.8
1.9
1.8
1.7
1.8
1.7
2.0
2.0
2.1
2.3
2.5
2.7
2.7
2.7
2.8
2.9
3.0
3.0
2.9
2.9
2.9
2.7
2.6
2.6
2.8
2.9
3.1
3.5
3.7
3.6
3.8
4.1
4.1
4.3
4.3
4.2
4.1
4.2
4.1
4.0
4.1
4.2
4.2
4.1
4.2
4.2
4.2
4.2
4.2
4.1
4.1
4.1
4.1
4.0
4.0
4.0

Total

Goods

5.0
5.2
5.4
5.8
6.4
8.2
7.3
8.1
8.8
9.0
9.6
10.3
9.9
9.1
9.0
10.0
9.6
9.9
10.5
10.6
10.5
10.6
10.1
10.2
10.5
11.2
11.8
11.9
12.3
12.3
13.0
14.4
13.2
13.0
13.4
14.7
15.5
16.2
16.5
17.4
13.7
15.7
17.3
17.0
16.5
16.4
15.3
14.6
15.0
15.3
14.6
14.6
14.5
14.6
14.8
15.0
15.0
14.8
15.3
15.4
15.2
15.3
15.3
14.9
14.8
14.6
14.1

3.6
3.8
4.0
4.5
5.0
6.8
5.9
6.7
7.3
7.5
8.1
8.7
8.4
7.5
7.5
8.3
7.9
8.1
8.5
8.6
8.6
8.5
8.1
8.4
8.6
9.3
9.9
10.0
10.3
10.3
10.9
12.2
11.1
10.9
11.3
12.3
13.2
13.7
13.8
14.6
11.0
13.0
14.4
14.2
13.7
13.6
12.6
11.9
12.2
12.5
11.8
11.8
11.8
11.9
12.0
12.2
12.1
12.0
12.4
12.6
12.4
12.5
12.4
12.1
12.0
11.8
11.3

Federal
Services
1.3
1.4
1.4
1.4
1.4
1.5
1.4
1.4
1.4
1.5
1.5
1.6
1.6
1.6
1.5
1.7
1.7
1.8
1.9
1.9
1.9
2.0
2.0
1.9
1.9
1.9
1.9
1.9
2.0
2.0
2.0
2.2
2.1
2.1
2.2
2.4
2.4
2.5
2.6
2.8
2.7
2.8
2.8
2.8
2.8
2.8
2.7
2.8
2.8
2.8
2.8
2.8
2.7
2.8
2.8
2.8
2.8
2.8
2.8
2.8
2.8
2.8
2.8
2.8
2.8
2.8
2.8

Total
23.5
23.5
23.0
22.4
21.4
22.1
22.6
21.6
20.9
20.3
20.0
20.6
20.4
21.3
21.1
20.5
21.0
21.3
21.2
20.6
20.4
20.8
21.1
20.6
19.9
19.2
19.0
18.5
18.0
17.8
17.9
17.8
18.4
19.1
19.3
19.1
19.0
19.0
19.3
20.3
21.3
21.0
20.3
19.4
18.7
18.1
17.8
17.7
17.5
17.5
17.5
17.8
17.7
17.6
17.6
17.6
17.5
17.4
17.4
17.5
17.4
17.5
17.4
17.5
17.5
17.5
17.6

Total
12.9
12.4
11.5
11.1
10.3
10.3
10.3
9.9
9.6
9.3
9.2
9.6
9.8
10.4
10.5
10.2
10.4
10.5
10.4
9.8
9.5
9.4
9.5
9.0
8.5
7.9
7.5
7.2
6.8
6.5
6.3
6.2
6.3
6.8
7.2
7.3
7.3
7.2
7.3
7.8
8.4
8.7
8.4
7.9
7.3
6.9
6.7
6.6
6.5
6.5
6.6
6.7
6.6
6.6
6.6
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6.6
6.6
6.6
6.6
6.7

National
defense

Nondefense

10.0
9.4
8.4
7.9
7.2
7.1
7.0
6.7
6.5
6.2
6.1
6.4
6.7
7.3
7.5
7.4
7.6
7.7
7.7
7.3
6.9
6.8
6.7
6.2
5.7
5.2
4.9
4.7
4.3
4.1
4.0
3.8
3.9
4.2
4.5
4.7
4.7
4.6
4.7
5.1
5.5
5.5
5.4
5.0
4.6
4.2
4.0
3.9
3.8
3.9
4.0
3.9
3.9
3.9
3.9
3.8
3.9
3.8
3.8
3.8
3.8
3.9
3.9
3.9
3.9
3.9
4.0

2.9
3.0
3.1
3.2
3.1
3.2
3.3
3.2
3.2
3.1
3.0
3.2
3.1
3.1
3.0
2.8
2.8
2.8
2.7
2.5
2.5
2.6
2.7
2.8
2.7
2.6
2.6
2.5
2.5
2.4
2.4
2.4
2.4
2.6
2.7
2.6
2.6
2.6
2.6
2.7
3.0
3.1
3.0
2.9
2.8
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7

State
and
local
10.6
11.2
11.4
11.3
11.1
11.8
12.3
11.7
11.2
10.9
10.8
11.0
10.6
10.9
10.6
10.3
10.5
10.8
10.9
10.8
11.0
11.3
11.6
11.6
11.4
11.4
11.4
11.3
11.2
11.3
11.5
11.6
12.1
12.3
12.1
11.8
11.7
11.8
12.0
12.4
12.8
12.4
11.9
11.4
11.4
11.1
11.1
11.1
11.0
10.9
10.9
11.1
11.1
11.1
11.1
11.0
11.0
10.9
10.9
10.9
10.9
10.9
10.9
10.8
10.9
10.9
10.9

Source: Department of Commerce (Bureau of Economic Analysis).

National Income or Expenditure

250-840_text_.pdf 373

| 369

2/7/20 3:46 PM

Table B–5. Chain-type price indexes for gross domestic product, 1969–2019
[Index numbers, 2012=100, except as noted; quarterly data seasonally adjusted]

Personal consumption expenditures

Gross private domestic investment
Fixed investment

Year or quarter

1969 ����������������������
1970 ����������������������
1971 ����������������������
1972 ����������������������
1973 ����������������������
1974 ����������������������
1975 ����������������������
1976 ����������������������
1977 ����������������������
1978 ����������������������
1979 ����������������������
1980 ����������������������
1981 ����������������������
1982 ����������������������
1983 ����������������������
1984 ����������������������
1985 ����������������������
1986 ����������������������
1987 ����������������������
1988 ����������������������
1989 ����������������������
1990 ����������������������
1991 ����������������������
1992 ����������������������
1993 ����������������������
1994 ����������������������
1995 ����������������������
1996 ����������������������
1997 ����������������������
1998 ����������������������
1999 ����������������������
2000 ����������������������
2001 ����������������������
2002 ����������������������
2003 ����������������������
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

Gross
domestic
product

20.590
21.676
22.776
23.760
25.061
27.309
29.846
31.490
33.445
35.798
38.766
42.278
46.269
49.130
51.051
52.894
54.568
55.673
57.041
59.055
61.370
63.676
65.819
67.321
68.917
70.386
71.864
73.178
74.446
75.267
76.346
78.069
79.822
81.039
82.567
84.778
87.407
90.074
92.498
94.264
94.999
96.109
98.112
100.000
101.773
103.647
104.688
105.770
107.795
110.382
112.358
104.933
105.618
105.987
106.543
107.040
107.394
108.032
108.715
109.341
110.209
110.765
111.212
111.504
112.173
112.679
113.076

Nonresidential
Total

20.015
20.951
21.841
22.586
23.802
26.280
28.470
30.032
31.986
34.211
37.251
41.262
44.958
47.456
49.474
51.343
53.134
54.290
55.964
58.151
60.690
63.355
65.473
67.218
68.892
70.330
71.811
73.346
74.623
75.216
76.338
78.235
79.738
80.789
82.358
84.411
86.812
89.174
91.438
94.180
94.094
95.705
98.131
100.000
101.346
102.830
103.045
104.091
105.929
108.143
109.670
103.297
103.910
104.344
104.812
105.355
105.596
106.033
106.733
107.401
107.988
108.413
108.772
108.879
109.522
109.928
110.352

Goods

30.934
32.114
33.079
33.926
35.949
40.436
43.703
45.413
47.837
50.773
55.574
61.797
66.389
68.198
69.429
70.742
71.877
71.541
73.842
75.788
78.704
81.927
83.930
84.943
85.681
86.552
87.361
88.321
88.219
86.893
87.349
89.082
89.015
88.166
88.054
89.292
91.084
92.306
93.331
96.122
93.812
95.183
98.773
100.000
99.407
98.920
95.885
94.318
94.586
95.232
94.785
94.181
94.465
94.231
94.393
94.898
94.264
94.425
94.759
95.228
95.400
95.319
94.982
94.590
94.955
94.772
94.822

Services

15.078
15.913
16.781
17.491
18.336
19.890
21.595
23.093
24.841
26.750
28.994
32.009
35.288
38.058
40.396
42.498
44.577
46.408
47.796
50.082
52.443
54.846
56.992
59.018
61.059
62.719
64.471
66.240
68.107
69.549
70.970
72.938
75.171
77.123
79.506
81.965
84.673
87.616
90.516
93.235
94.231
95.957
97.814
100.000
102.316
104.804
106.704
109.120
111.793
114.851
117.458
107.979
108.765
109.553
110.182
110.745
111.452
112.038
112.935
113.707
114.520
115.220
115.958
116.339
117.133
117.865
118.497

Total

28.402
29.624
31.092
32.388
34.153
37.559
42.059
44.384
47.655
51.517
56.141
61.395
67.123
70.679
70.896
71.661
72.548
74.178
75.723
77.627
79.606
81.270
82.648
82.647
83.627
84.875
86.240
86.191
86.241
85.608
85.690
86.815
87.555
87.841
88.561
91.148
94.839
98.176
99.656
100.474
99.331
97.687
98.704
100.000
100.979
102.922
103.666
103.567
105.378
107.757
109.418
103.031
103.419
103.635
104.184
104.588
105.151
105.787
105.985
106.862
107.615
108.186
108.366
108.832
109.382
109.678
109.779

Total

27.498
28.699
30.134
31.420
33.169
36.449
40.874
43.232
46.550
50.444
54.977
60.105
65.624
69.311
69.575
70.253
71.277
73.021
74.506
76.586
78.561
80.278
81.683
81.728
82.711
83.983
85.378
85.450
85.599
85.133
85.277
86.486
87.241
87.500
88.265
90.843
94.597
97.958
99.456
100.296
99.076
97.568
98.641
100.000
101.091
103.172
104.187
104.234
106.057
108.507
110.164
103.720
104.082
104.297
104.837
105.269
105.852
106.395
106.714
107.595
108.386
108.951
109.096
109.577
110.110
110.426
110.543

Total
34.638
36.295
37.997
39.297
40.882
44.857
50.766
53.562
57.111
60.930
65.830
71.641
78.453
82.911
82.774
83.036
83.893
85.365
86.339
88.514
90.572
92.516
94.267
93.960
94.161
94.904
95.849
95.267
94.735
93.248
92.314
92.718
92.346
91.863
91.156
92.055
94.443
96.745
98.310
99.832
99.184
97.416
98.559
100.000
100.251
101.469
102.042
101.146
102.116
103.515
104.694
101.080
101.169
101.017
101.319
101.633
101.989
102.333
102.509
102.950
103.428
103.841
103.839
104.241
104.770
104.911
104.854

Structures Equipment
11.114
11.845
12.757
13.674
14.734
16.770
18.773
19.692
21.401
23.468
26.194
28.629
32.566
35.136
34.241
34.540
35.361
36.039
36.618
38.171
39.666
40.948
41.689
41.699
42.922
44.437
46.362
47.540
49.355
51.612
53.198
55.283
58.178
60.603
62.769
67.416
75.733
84.749
89.748
94.335
92.613
92.006
95.362
100.000
101.455
107.198
109.598
109.958
112.952
117.062
121.097
109.254
110.089
109.949
110.542
111.333
112.456
113.703
114.317
115.133
116.547
117.480
119.087
119.899
121.074
121.543
121.871

59.657
61.891
63.848
64.686
65.780
70.713
81.484
86.486
91.800
96.900
103.167
112.249
120.463
125.415
125.776
124.748
124.748
127.254
128.083
129.854
132.337
135.042
137.330
137.121
135.518
135.277
133.796
130.762
127.156
121.451
116.763
114.224
110.858
108.531
105.725
104.841
104.598
103.560
103.191
102.542
103.169
99.471
99.447
100.000
99.787
99.169
98.672
97.436
97.287
97.396
97.525
97.771
97.562
97.353
97.056
97.319
97.338
97.297
97.194
97.116
97.321
97.710
97.436
97.669
97.764
97.487
97.182

Intellectual
property
products
36.204
37.929
39.318
40.490
42.494
46.461
50.190
52.408
54.709
57.557
61.382
66.123
71.058
75.093
77.898
80.081
81.413
82.047
83.518
86.129
87.240
88.147
90.271
89.373
89.998
90.468
93.134
93.544
94.052
93.595
95.105
97.814
97.684
96.376
95.647
95.335
95.952
97.088
98.284
99.834
98.589
98.306
99.517
100.000
100.081
100.791
101.677
100.464
101.742
103.282
104.211
100.224
100.280
100.204
101.149
101.245
101.592
101.914
102.216
103.154
103.433
103.558
102.984
103.378
104.123
104.638
104.704

Residential

15.518
16.016
16.943
17.975
19.571
21.593
23.590
25.117
27.683
31.082
34.593
38.325
41.425
43.646
44.680
46.003
47.267
49.351
51.486
53.278
55.020
56.288
57.021
57.723
60.074
62.247
64.473
65.856
67.444
69.223
71.816
75.004
78.564
80.510
84.325
90.243
96.706
102.355
103.708
102.249
98.671
98.317
99.049
100.000
105.054
111.118
114.129
118.148
123.510
130.488
134.310
115.777
117.271
119.006
120.540
121.452
122.970
124.348
125.270
128.031
130.203
131.450
132.267
133.108
133.655
134.780
135.697

See next page for continuation of table.

370 |

250-840_text_.pdf 374

Appendix B

2/7/20 3:46 PM

Table B–5. Chain-type price indexes for gross domestic product, 1969–2019—Continued
[Index numbers, 2012=100, except as noted; quarterly data seasonally adjusted]

Exports and imports
of goods and
services

Government consumption
expenditures and
gross investment
Federal

Year or quarter
Exports

Imports

Total

Total

National Nondefense defense

State
and
local

1969 ����������������� 28.589 18.839 14.892 17.715 17.019 19.154 13.063
1970 ����������������� 29.711 19.954 16.078 19.109 18.294 20.906 14.117
1971 ����������������� 30.796 21.179 17.352 20.670 19.817 22.521 15.198
1972 ����������������� 32.145 22.662 18.662 22.485 21.883 23.579 16.163
1973 ����������������� 36.382 26.601 19.936 24.051 23.484 25.018 17.246
1974 ����������������� 44.807 38.058 21.852 25.971 25.404 26.904 19.157
1975 ����������������� 49.388 41.226 23.870 28.254 27.545 29.484 20.999
1976 ����������������� 51.009 42.467 25.181 30.012 29.345 31.124 22.024
1977 ����������������� 53.088 46.209 26.739 31.858 31.268 32.782 23.394
1978 ����������������� 56.317 49.466 28.507 34.008 33.561 34.612 24.914
1979 ����������������� 63.101 57.930 30.853 36.566 36.216 36.952 27.114
1980 ����������������� 69.503 72.166 34.045 40.099 39.919 40.106 30.081
1981 ����������������� 74.650 76.066 37.424 43.843 43.747 43.643 33.226
1982 ����������������� 75.006 73.506 39.969 46.943 47.039 46.289 35.401
1983 ����������������� 75.311 70.751 41.516 48.499 48.778 47.397 36.964
1984 ����������������� 76.016 70.139 43.317 50.637 51.013 49.279 38.544
1985 ����������������� 73.753 67.836 44.659 51.712 51.872 50.907 40.113
1986 ����������������� 72.523 67.834 45.409 51.957 51.894 51.748 41.269
1987 ����������������� 74.124 71.935 46.635 52.318 52.267 52.076 43.196
1988 ����������������� 77.920 75.377 48.177 54.025 53.904 53.974 44.640
1989 ����������������� 79.210 77.024 50.016 55.534 55.365 55.605 46.752
1990 ����������������� 79.657 79.233 52.113 57.250 57.162 57.093 49.153
1991 ����������������� 80.545 78.573 54.005 59.309 58.964 59.787 50.953
1992 ����������������� 80.153 78.636 55.642 60.824 60.678 60.825 52.690
1993 ����������������� 80.277 78.033 56.953 62.151 61.615 62.994 54.002
1994 ����������������� 81.210 78.766 58.463 63.861 63.229 64.898 55.394
1995 ����������������� 83.025 80.924 60.123 65.838 65.027 67.223 56.871
1996 ����������������� 81.923 79.514 61.355 66.937 66.114 68.344 58.177
1997 ����������������� 80.479 76.750 62.560 67.972 67.035 69.591 59.471
1998 ����������������� 78.574 72.618 63.624 68.841 67.871 70.518 60.630
1999 ����������������� 77.971 73.019 65.778 70.519 69.559 72.178 63.008
2000 ����������������� 79.467 76.221 68.601 72.886 71.908 74.578 66.032
2001 ����������������� 78.836 74.223 70.567 74.236 73.270 75.906 68.281
2002 ����������������� 78.201 73.242 72.393 76.631 75.714 78.222 69.815
2003 ����������������� 79.400 75.454 75.028 80.008 79.505 80.895 72.050
2004 ����������������� 82.284 79.060 78.153 82.760 82.263 83.637 75.369
2005 ����������������� 85.131 83.703 82.110 86.204 86.011 86.531 79.609
2006 ����������������� 87.842 86.909 85.661 88.949 89.022 88.799 83.617
2007 ����������������� 91.139 89.921 89.491 91.589 91.750 91.279 88.133
2008 ����������������� 95.410 98.960 93.308 94.381 94.801 93.597 92.558
2009 ����������������� 89.694 87.987 92.931 94.214 94.126 94.364 92.048
2010 ����������������� 93.348 92.783 95.386 96.421 96.128 96.942 94.669
2011 ����������������� 99.242 99.826 98.285 99.070 98.946 99.289 97.739
2012 ����������������� 100.000 100.000 100.000 100.000 100.000 100.000 100.000
2013 ����������������� 100.168 98.636 102.332 100.931 100.609 101.478 103.279
2014 ����������������� 100.272 97.854 104.435 102.632 102.056 103.593 105.645
2015 ����������������� 95.385 89.947 104.705 103.282 102.402 104.718 105.677
2016 ����������������� 93.455 86.696 105.050 103.900 102.776 105.701 105.854
2017 ����������������� 95.850 88.622 107.647 106.040 104.518 108.435 108.731
2018 ����������������� 99.104 91.181 111.403 109.336 107.609 112.040 112.772
2019 p ��������������� 98.886 89.945 113.787 111.587 109.441 114.931 115.244
2016: I ������������� 92.321 86.050 104.165 103.105 102.013 104.858 104.913
      II ������������ 93.253 86.407 104.906 103.697 102.631 105.412 105.746
      III ����������� 93.803 87.028 105.285 104.147 103.041 105.921 106.082
      IV ����������� 94.441 87.298 105.843 104.651 103.419 106.613 106.674
2017: I ������������� 95.054 88.312 106.697 105.230 103.893 107.347 107.694
      II ������������ 95.094 88.251 107.102 105.667 104.165 108.032 108.081
      III ����������� 95.974 88.394 107.843 106.201 104.601 108.710 108.949
      IV ����������� 97.277 89.529 108.946 107.063 105.411 109.651 110.200
2018: I ������������� 98.129 91.124 110.007 108.219 106.576 110.795 111.204
      II ������������ 99.364 91.250 111.047 108.992 107.317 111.617 112.408
      III ����������� 99.640 91.378 111.882 109.685 108.027 112.284 113.332
      IV ����������� 99.284 90.972 112.674 110.450 108.517 113.464 114.142
2019: I ������������� 98.663 90.158 113.046 111.691 108.804 116.187 113.973
      II ������������ 99.463 90.521 113.526 111.096 109.207 114.042 115.125
      III ����������� 98.876 89.597 113.973 111.517 109.595 114.513 115.589
      IV p �������� 98.544 89.503 114.605 112.043 110.158 114.980 116.290

Personal
consumption
Final
expenGross
sales of ditures domestic
Gross
domestic excludpurproduct
ing
chases 1 domestic
product
food
and
energy
20.465
21.547
22.642
23.624
24.923
27.154
29.680
31.326
33.284
35.637
38.591
42.084
46.046
48.921
50.836
52.671
54.371
55.492
56.851
58.890
61.205
63.519
65.663
67.169
68.765
70.239
71.722
73.055
74.344
75.200
76.296
78.037
79.793
81.004
82.541
84.751
87.388
90.058
92.489
94.259
94.970
96.086
98.100
100.000
101.795
103.692
104.782
105.895
107.923
110.523
112.499
105.061
105.743
106.112
106.666
107.168
107.525
108.147
108.853
109.478
110.354
110.908
111.351
111.644
112.311
112.821
113.222

21.136
22.126
23.167
23.912
24.823
26.788
29.026
30.791
32.771
34.943
37.490
40.936
44.523
47.417
49.844
51.911
54.019
55.883
57.683
60.134
62.630
65.168
67.495
69.547
71.436
73.034
74.625
76.040
77.382
78.366
79.425
80.804
82.258
83.639
84.837
86.515
88.373
90.392
92.378
94.225
95.315
96.608
98.139
100.000
101.526
103.122
104.407
106.070
107.795
109.897
111.670
105.322
105.848
106.363
106.746
107.189
107.540
107.934
108.516
109.131
109.707
110.136
110.612
110.902
111.414
111.997
112.366

20.010
21.087
22.185
23.175
24.499
26.986
29.452
31.071
33.119
35.474
38.585
42.602
46.532
49.214
50.926
52.649
54.214
55.345
56.908
58.921
61.240
63.663
65.662
67.190
68.706
70.147
71.661
72.908
73.983
74.476
75.632
77.575
79.039
80.125
81.776
84.126
87.037
89.783
92.206
94.849
94.559
95.923
98.246
100.000
101.468
103.138
103.453
104.185
106.148
108.647
110.339
103.418
104.016
104.405
104.902
105.474
105.797
106.319
107.001
107.770
108.461
108.978
109.378
109.591
110.192
110.585
110.990

4.9
5.3
5.1
4.3
5.5
9.0
9.3
5.5
6.2
7.0
8.3
9.1
9.4
6.2
3.9
3.6
3.2
2.0
2.5
3.5
3.9
3.8
3.4
2.3
2.4
2.1
2.1
1.8
1.7
1.1
1.4
2.3
2.2
1.5
1.9
2.7
3.1
3.1
2.7
1.9
.8
1.2
2.1
1.9
1.8
1.8
1.0
1.0
1.9
2.4
1.8
–.2
2.6
1.4
2.1
1.9
1.3
2.4
2.6
2.3
3.2
2.0
1.6
1.1
2.4
1.8
1.4

Percent change 2
Personal
consumption
expenditures

Total

Gross
domestic
Excludpuring chases 1
food
and
energy

4.5
4.7
4.2
3.4
5.4
10.4
8.3
5.5
6.5
7.0
8.9
10.8
9.0
5.6
4.3
3.8
3.5
2.2
3.1
3.9
4.4
4.4
3.3
2.7
2.5
2.1
2.1
2.1
1.7
.8
1.5
2.5
1.9
1.3
1.9
2.5
2.8
2.7
2.5
3.0
–.1
1.7
2.5
1.9
1.3
1.5
.2
1.0
1.8
2.1
1.4
.2
2.4
1.7
1.8
2.1
.9
1.7
2.7
2.5
2.2
1.6
1.3
.4
2.4
1.5
1.6

4.7
4.7
4.7
3.2
3.8
7.9
8.4
6.1
6.4
6.6
7.3
9.2
8.8
6.5
5.1
4.1
4.1
3.5
3.2
4.2
4.2
4.1
3.6
3.0
2.7
2.2
2.2
1.9
1.8
1.3
1.4
1.7
1.8
1.7
1.4
2.0
2.1
2.3
2.2
2.0
1.2
1.4
1.6
1.9
1.5
1.6
1.2
1.6
1.6
1.9
1.6
1.7
2.0
2.0
1.4
1.7
1.3
1.5
2.2
2.3
2.1
1.6
1.7
1.1
1.9
2.1
1.3

4.9
5.4
5.2
4.5
5.7
10.2
9.1
5.5
6.6
7.1
8.8
10.4
9.2
5.8
3.5
3.4
3.0
2.1
2.8
3.5
3.9
4.0
3.1
2.3
2.3
2.1
2.2
1.7
1.5
.7
1.6
2.6
1.9
1.4
2.1
2.9
3.5
3.2
2.7
2.9
–.3
1.4
2.4
1.8
1.5
1.6
.3
.7
1.9
2.4
1.6
–.6
2.3
1.5
1.9
2.2
1.2
2.0
2.6
2.9
2.6
1.9
1.5
.8
2.2
1.4
1.5

1 Gross domestic product (GDP) less exports of goods and services plus imports of goods and services.
2 Quarterly percent changes are at annual rates.

Source: Department of Commerce (Bureau of Economic Analysis).

National Income or Expenditure

250-840_text_.pdf 375

| 371

2/7/20 3:46 PM

Table B–6. Gross value added by sector, 1969–2019
[Billions of dollars; quarterly data at seasonally adjusted annual rates]
Business 1
Year or quarter

1969 ����������������������
1970 ����������������������
1971 ����������������������
1972 ����������������������
1973 ����������������������
1974 ����������������������
1975 ����������������������
1976 ����������������������
1977 ����������������������
1978 ����������������������
1979 ����������������������
1980 ����������������������
1981 ����������������������
1982 ����������������������
1983 ����������������������
1984 ����������������������
1985 ����������������������
1986 ����������������������
1987 ����������������������
1988 ����������������������
1989 ����������������������
1990 ����������������������
1991 ����������������������
1992 ����������������������
1993 ����������������������
1994 ����������������������
1995 ����������������������
1996 ����������������������
1997 ����������������������
1998 ����������������������
1999 ����������������������
2000 ����������������������
2001 ����������������������
2002 ����������������������
2003 ����������������������
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2017: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2018: I ������������������
      II �����������������
      III ����������������
      IV ����������������
2019: I ������������������
      II �����������������
      III ����������������
      IV p �������������

Gross
domestic
product

1,017.6
1,073.3
1,164.9
1,279.1
1,425.4
1,545.2
1,684.9
1,873.4
2,081.8
2,351.6
2,627.3
2,857.3
3,207.0
3,343.8
3,634.0
4,037.6
4,339.0
4,579.6
4,855.2
5,236.4
5,641.6
5,963.1
6,158.1
6,520.3
6,858.6
7,287.2
7,639.7
8,073.1
8,577.6
9,062.8
9,630.7
10,252.3
10,581.8
10,936.4
11,458.2
12,213.7
13,036.6
13,814.6
14,451.9
14,712.8
14,448.9
14,992.1
15,542.6
16,197.0
16,784.9
17,527.3
18,224.8
18,715.0
19,519.4
20,580.2
21,429.0
18,424.3
18,637.3
18,806.7
18,991.9
19,190.4
19,356.6
19,611.7
19,918.9
20,163.2
20,510.2
20,749.8
20,897.8
21,098.8
21,340.3
21,542.5
21,734.3

Total

782.7
815.9
882.5
972.5
1,094.0
1,182.8
1,284.8
1,443.3
1,616.2
1,838.2
2,062.8
2,225.8
2,502.0
2,568.6
2,801.9
3,136.7
3,369.6
3,539.3
3,735.2
4,019.3
4,326.7
4,542.0
4,645.0
4,920.2
5,177.4
5,523.7
5,795.1
6,159.5
6,578.8
6,959.2
7,400.1
7,876.1
8,062.0
8,264.4
8,642.4
9,240.6
9,898.0
10,509.1
10,994.6
11,054.9
10,669.9
11,140.5
11,612.9
12,189.5
12,670.5
13,280.5
13,826.3
14,180.6
14,830.7
15,680.8
16,329.9
13,942.9
14,120.5
14,255.5
14,403.6
14,557.5
14,690.6
14,910.7
15,163.9
15,345.7
15,633.5
15,823.3
15,920.7
16,070.6
16,271.9
16,417.6
16,559.4

Nonfarm 1

759.9
792.3
857.2
942.9
1,047.2
1,138.5
1,239.2
1,400.2
1,572.7
1,787.5
2,002.7
2,174.4
2,437.0
2,508.2
2,757.0
3,072.6
3,305.9
3,479.4
3,673.2
3,957.9
4,252.8
4,464.2
4,574.7
4,840.4
5,106.2
5,440.1
5,726.7
6,066.9
6,490.6
6,880.2
7,329.2
7,800.1
7,983.9
8,190.4
8,551.3
9,121.2
9,793.5
10,412.8
10,878.9
10,935.4
10,566.8
11,022.8
11,460.7
12,040.5
12,485.9
13,112.4
13,680.3
14,051.6
14,691.2
15,551.2
16,200.4
13,813.5
13,987.7
14,124.8
14,280.5
14,411.1
14,547.0
14,776.6
15,030.2
15,212.9
15,498.2
15,699.6
15,794.2
15,946.8
16,143.9
16,283.3
16,427.5

Households and institutions

Farm

22.8
23.7
25.4
29.7
46.8
44.2
45.6
43.0
43.5
50.7
60.1
51.4
65.0
60.4
44.9
64.2
63.7
59.9
62.0
61.4
73.9
77.8
70.4
79.9
71.3
83.6
68.4
92.6
88.1
79.0
70.9
76.0
78.1
74.0
91.1
119.4
104.5
96.3
115.7
119.5
103.1
117.6
152.2
148.9
184.6
168.1
146.0
129.0
139.4
129.6
129.5
129.4
132.8
130.7
123.1
146.4
143.6
134.1
133.7
132.8
135.3
123.7
126.5
123.8
128.0
134.4
131.9

Total

87.0
94.6
104.5
114.0
124.6
137.2
151.6
164.9
179.9
202.1
226.3
258.2
291.6
323.8
352.5
383.8
411.8
447.0
489.5
539.8
586.0
636.3
677.3
720.3
772.8
824.7
877.8
923.2
975.9
1,040.6
1,112.4
1,191.9
1,267.2
1,343.6
1,411.0
1,494.5
1,583.3
1,673.6
1,730.3
1,836.8
1,895.5
1,905.5
1,956.8
2,018.4
2,075.0
2,158.8
2,256.2
2,349.0
2,445.7
2,569.9
2,686.5
2,315.6
2,338.0
2,358.2
2,384.3
2,414.1
2,434.2
2,450.5
2,483.8
2,523.8
2,559.2
2,582.2
2,614.5
2,648.3
2,669.7
2,698.6
2,729.3

Households

57.1
61.2
67.2
72.7
78.5
85.5
93.7
101.7
110.7
124.8
139.5
158.8
179.2
198.2
213.6
230.9
248.2
268.4
289.8
316.4
341.4
367.6
386.6
407.1
437.6
472.7
506.9
534.6
565.7
601.6
645.2
693.5
744.7
780.7
816.6
868.4
933.4
991.2
1,016.9
1,075.2
1,097.0
1,091.0
1,108.0
1,128.0
1,157.0
1,203.3
1,250.9
1,301.8
1,363.0
1,437.4
1,503.4
1,282.7
1,296.4
1,306.9
1,321.2
1,342.4
1,355.7
1,366.3
1,387.5
1,409.2
1,432.5
1,446.0
1,462.0
1,480.6
1,497.5
1,510.6
1,525.0

Nonprofit
institutions
serving
households 2
30.0
33.4
37.4
41.4
46.1
51.7
58.0
63.2
69.2
77.3
86.9
99.3
112.4
125.6
138.9
152.8
163.6
178.6
199.7
223.4
244.6
268.8
290.7
313.2
335.1
352.0
370.9
388.7
410.2
439.0
467.3
498.5
522.6
562.9
594.4
626.1
649.8
682.4
713.4
761.6
798.5
814.5
848.8
890.3
918.0
955.4
1,005.4
1,047.2
1,082.7
1,132.5
1,183.0
1,032.8
1,041.5
1,051.3
1,063.0
1,071.7
1,078.5
1,084.3
1,096.4
1,114.6
1,126.7
1,136.2
1,152.5
1,167.7
1,172.2
1,188.0
1,204.3

General government 3

Total

147.9
162.8
177.8
192.6
206.8
225.3
248.4
265.3
285.7
311.3
338.2
373.4
413.5
451.4
479.7
517.1
557.5
593.3
630.4
677.4
728.8
784.9
835.8
879.8
908.3
938.8
966.9
990.3
1,022.9
1,063.0
1,118.1
1,184.3
1,252.6
1,328.4
1,404.8
1,478.7
1,555.4
1,631.9
1,726.9
1,821.2
1,883.5
1,946.1
1,972.9
1,989.1
2,039.3
2,088.0
2,142.2
2,185.4
2,243.1
2,329.5
2,412.6
2,165.8
2,178.8
2,193.0
2,204.0
2,218.9
2,231.8
2,250.4
2,271.2
2,293.6
2,317.5
2,344.3
2,362.6
2,379.9
2,398.7
2,426.3
2,445.6

Federal

76.9
82.5
87.5
92.4
96.4
102.5
110.5
117.3
125.2
135.8
145.4
159.8
178.3
195.7
207.1
225.3
240.0
250.6
261.0
278.5
292.8
306.7
323.5
329.6
331.5
332.6
333.0
331.8
333.5
336.8
345.0
360.3
370.3
397.8
434.7
459.4
488.4
509.9
535.7
569.1
603.0
640.0
659.8
663.7
658.4
666.8
674.8
686.3
701.7
729.0
754.7
680.3
684.6
688.5
691.9
695.4
698.0
703.3
710.1
718.2
725.7
733.4
738.7
745.3
750.5
758.4
764.7

State
and
local
70.9
80.3
90.3
100.2
110.4
122.8
138.0
148.0
160.6
175.5
192.8
213.5
235.2
255.6
272.6
291.9
317.6
342.7
369.4
398.8
436.1
478.2
512.2
550.2
576.9
606.2
633.9
658.6
689.3
726.2
773.1
824.0
882.3
930.6
970.1
1,019.3
1,067.0
1,122.1
1,191.2
1,252.1
1,280.5
1,306.1
1,313.1
1,325.5
1,380.9
1,421.1
1,467.4
1,499.1
1,541.4
1,600.5
1,657.9
1,485.5
1,494.2
1,504.5
1,512.1
1,523.5
1,533.8
1,547.2
1,561.1
1,575.5
1,591.8
1,610.9
1,624.0
1,634.6
1,648.1
1,668.0
1,680.8

Addendum:
Gross
housing
value
added
73.0
78.8
86.4
93.9
101.4
110.4
121.3
130.9
144.2
160.2
177.7
204.0
231.6
258.6
280.6
303.1
333.8
364.5
392.1
424.2
452.7
487.0
515.3
545.2
578.4
619.6
662.6
695.0
731.9
774.8
826.2
881.7
943.5
985.1
1,016.4
1,075.2
1,151.9
1,224.2
1,273.4
1,349.5
1,393.8
1,400.2
1,445.7
1,478.5
1,511.2
1,585.1
1,685.9
1,769.5
1,852.2
1,942.8
2,030.6
1,740.9
1,761.3
1,777.6
1,798.2
1,823.8
1,841.7
1,861.5
1,881.8
1,908.0
1,935.5
1,953.6
1,974.1
1,998.5
2,022.4
2,041.4
2,060.2

1 Gross domestic business value added equals gross domestic product excluding gross value added of households and institutions and of general
government. Nonfarm value added equals gross domestic business value added excluding gross farm value added.
2 Equals compensation of employees of nonprofit institutions, the rental value of nonresidential fixed assets owned and used by nonprofit institutions serving
households, and rental income of persons for tenant-occupied housing owned by nonprofit institutions.
3 Equals compensation of general government employees plus general government consumption of fixed capital.
Source: Department of Commerce (Bureau of Economic Analysis).

372 |

250-840_text_.pdf 376

Appendix B

2/7/20 3:46 PM

Table B–7. Real gross value added by sector, 1969–2019
[Billions of chained (2012) dollars; quarterly data at seasonally adjusted annual rates]
Business 1
Year or quarter

1969 ����������������������
1970 ����������������������
1971 ����������������������
1972 ����������������������
1973 ����������������������
1974 ����������������������
1975 ����������������������
1976 ����������������������
1977 ����������������������
1978 ����������������������
1979 ����������������������
1980 ����������������������
1981 ����������������������
1982 ����������������������
1983 ����������������������
1984 ����������������������
1985 ����������������������
1986 ����������������������
1987 ����������������������
1988 ����������������������
1989 ����������������������
1990 ����������������������
1991 ����������������������
1992 ����������������������
1993 ����������������������
1994 ����������������������
1995 ����������������������
1996 ����������������������
1997 ����������������������
1998 ����������������������
1999 ����������������������
2000 ����������������������
2001 ����������������������
2002 ����������������������
2003 ����������������������
2004 ����������������������
2005 ����������������������
2006 ����������������������
2007 ����������������������
2008 ����������������������
2009 ����������������������
2010 ����������������������
2011 ����������������������
2012 ����������������������
2013 ����������������������
2014 ����������������������
2015 ����������������������
2016 ����������������������
2017 ����������������������
2018 ����������������������
2019 p ��������������������
2016: I ������������������
      II