Rekindling Us Labor Productivity For A New Era Vfinal
Rekindling Us Labor Productivity For A New Era Vfinal
Rekindling Us Labor Productivity For A New Era Vfinal
Rekindling
US productivity
for a new era
Regaining historical rates of productivity
growth would add $10 trillion to US GDP—
a boost needed to confront workforce
shortages, debt, inflation, and the
energy transition.
Authors
Charles Atkins, San Francisco
Olivia White, San Francisco
Asutosh Padhi, Chicago
Kweilin Ellingrud, Minneapolis
Anu Madgavkar, New Jersey
Michael Neary, San Francisco
Editor
Mark Staples, New York
February 2023
McKinsey Global Institute
The McKinsey Global Institute was established in 1990. Our mission is to provide a fact base to
aid decision making on the economic and business issues most critical to the world’s companies
and policy leaders. We benefit from the full range of McKinsey’s regional, sectoral, and functional
knowledge, skills, and expertise, but editorial direction and decisions are solely the responsibility of
MGI directors and partners.
— Productivity and prosperity: Creating and harnessing the world’s assets most productively
— Resources of the world: Building, powering, and feeding the world sustainably
— Global connections: Exploring how flows of goods, people, and ideas shape economies
— Technologies and markets of the future: Discussing the next big arenas of value and competition
We aim for independent and fact-based research. None of our work is commissioned or funded by
any business, government, or other institution; we share our results publicly free of charge; and we
are entirely funded by the partners of McKinsey. While we engage multiple distinguished external
advisers to contribute to our work, the analyses presented in our publications are MGI’s alone, and
any errors are our own.
You can find out more about MGI and our research at www.mckinsey.com/mgi.
Introduction 1
Acknowledgments 22
30
2.2% productivity growth
Two projections (1948–2019 average)
25 Gap between
the 2 projections
is equivalent to:
$10 trillion
Cumulative GDP
added over 2021–30
20
$15,000
Additional output
per household in 2030
15
2005 2021 2030
Seattle Information
Washington
North Dakota
San Francisco Mining
San Jose
Wholesale trade
California
But if the US economy were a car, the engine has been sputtering for a while. In the past 15 years,
productivity growth has averaged just 1.4 percent, even as incredible advances in digital technology
put a supercomputer in every pocket. This state of affairs is well known. It is also not confined to the
United States. Most OECD countries have seen a drop in labor productivity growth since 2005.2
In the US economy, there’s also a problem with the transmission. The link between productivity
growth and real incomes has weakened. In the postwar boom from 1948–70, incomes grew at
3.0 percent annually, 3 while productivity growth averaged 2.8 percent. More recently, real incomes
have grown at 0.7 percent, well below the 1.4 percent gains in productivity (Exhibit 1). Worse, not
everyone has shared equally in the relatively low gains in income, and labor participation rates have
fallen from 67 percent in the 1990s to 63 percent in 2019, 4 as millions have become discouraged
about work. 5
Returning US productivity to its long-term trend of 2.2 percent annual growth would add $10 trillion in
cumulative GDP over the next ten years. This is equivalent to every US household seeing a cumulative
income gain of $15,000 over that period. Other things being equal, it would also boost stagnating
median incomes and encourage labor participation.6
It’s an aspirational prospect and one with historical precedent. In 1995, five-year productivity
growth averaged 1.6 percent, only for productivity to jump to 3 percent the following decade. Recent
technology advances, including in AI and biotechnology as well as productivity spikes since the global
COVID-19 pandemic, raised hopes (albeit briefly) that perhaps another boost may lie ahead.
1
In this study, we define labor productivity as the following: LP = the output of goods and services/working hours. Output is defined
as the real value-added output for the relevant sectors and states. For growth comparisons, output measures are deflated using
the appropriate sector- or state-level deflators. An extensive literature exists that documents the shortcomings of such deflators
in accounting for quality improvements in goods and services (see, for example, David M. Byrne, John G. Fernald, and Marshall
B. Reinsdorf, Does the United States have a productivity slowdown or a measurement problem?, Federal Reserve Bank of San
Francisco, working paper 2016-03, April 2016). Efforts to account for such errors find that they do not explain the aggregate
slowdown in productivity since 2005 in the United States. Throughout this research, we calculate labor productivity on a five-year
rolling average. For longer time period comparisons, we compare the level of labor productivity indexes versus the prior period. Due
to data limitations, we have occasionally used proxies or assumptions for the numerator or denominator in labor productivity based
on data availability. For national, sector-level, or state-level estimates, we use output, hours worked, and productivity data directly
from the US Bureau of Labor Statistics (BLS). For sector data at a state level, we estimate state-level productivity using state-level
GDP by sector data from the US Bureau of Economic Analysis (BEA) and employment by sector from the US Census Bureau. For
city-level estimates, we utilize GDP by metropolitan statistical area estimates from the BEA and employment-level estimates from
the Census Bureau.
2
Robert J. Gordon, Why has economic growth slowed when innovation appears to be accelerating?, National Bureau of Economic
Research, working paper 24554, April 2018; David Adler and Laurence B. Siegel, The Productivity Puzzle: Restoring Economic
Dynamism, New York, NY: CFA Institute Research Foundation Publications, 2019.
3
Real compensation includes employer costs for wages, salaries, and employee benefits; deflated against the consumer price index,
which is heavily influenced by rising housing and healthcare costs, to estimate worker purchasing power.
4
Labor participation rates fell to 63 percent from 2005 to 2019.
5
Working Economics Blog, “Wages for the top 1% skyrocketed 160% since 1979 while the share of wages for the bottom 90%
shrunk,” blog entry by Jori Kandra and Lawrence Mishel, December 1, 2020.
6
Economic Report of the President, White House and Council of Economic Advisors, February 2015.
Productivity is the measure of output Measures of productivity typically focus boom era in technology innovation over
relative to input. In macroeconomic terms, on the nonfarm business sector to exclude the past 15 years. Some have argued
it’s a measure of the value of the goods and government and farming activities3 that this paradox reflects measurement
services produced, divided by the amount of for which reliable productivity data is error, and with more accurate quality
labor, capital, and other resources required unavailable. The nonfarm business sector adjustments the productivity slowdown
for its production. For this paper, we focus represented 87 percent of US GDP in 2021. would disappear. For this to be true, there
on labor productivity, which is defined as the It encompasses 18 sectors, spread across not only needs to be mismeasurement but
economic output per hour worked.1 This is a services (78 percent) such as retail trade the scale of mismeasurement also needs
commonly used productivity measure, and is and goods production (22 percent). To to have increased over the past 15 years. A
of interest since it is the most consequential compare across periods, economists adjust number of researchers studying the issue
determinant of long-run economic growth. the value of output to account for changes in have found no evidence of an increase in
prices and quality. mismeasurement that would explain the
Rising prosperity—a higher standard of
aggregate slowdown in productivity since
living—can come only from productivity Any analysis of productivity is susceptible
2005 in the United States. 4
growth. And in the long run, labor to errors or blind spots. In the numerator,
productivity and real wages are closely, if rapid technological progress can complicate A note on our sources and methods: our
not perfectly, linked.2 Productivity growth measurement of output across time periods measures of output are primarily real value
will also be essential to address several by making it more challenging to account added, or sectoral output (in 2012 dollars)
looming challenges. As the US population for changes in the quality of goods and from data sets published by the US Bureau
ages and the ratio of nonworkers to services. In addition, reliable measures of of Labor Statistics (BLS) and the US Bureau
workers grows, productivity growth in the quality can be challenging to construct in of Economic Analysis (BEA). Our calculation
remaining workforce will be essential to areas such as healthcare where services, of working hours comes primarily from the
sustain output and meet the needs of the prices, and health outcomes are often BLS. We have also used proxies for inputs
population. Productivity will also ameliorate disconnected. In the denominator, too, there (such as total employment) where needed,
inflation, by reducing the need for supply- are complications. For instance, informal incorporating additional data from Moody’s
constrained labor, capital, and raw materials working hours are often not captured, Investors Service, the US Census Bureau,
to deliver a given output. And by delivering which can impede measurement in sectors the World Bank, the Conference Board,
a larger economic base, productivity will with high share of informal labor such and the Federal Reserve Board. The period
increase the affordability of the coming as construction. we studied in most cases stops in 2019
energy transition as well as growing debt or earlier, to stay focused on long-term
This esoteric topic has taken on new
and entitlements. patterns and avoid the gyrations of the
urgency, given the paradoxical productivity
COVID‑19 pandemic.
slowdown that has occurred during a
1
One productivity measure, total factor productivity (TFP), accounts for both capital and labor. Many of the trends and challenges that we’ve observed with labor productivity hold
true when measured with TFP.
2
Anna M. Stansbury and Lawrence H. Summers, Productivity and pay: Is the link broken?, National Bureau of Economic Research, working paper 24165, December 2017; Grant E.
Hayes, James M. Jedras, Edward Lazear, and Kathryn L. Shaw, Productivity and wages: What was the productivity-wage link in the digital revolution of the past, and what might
occur in the AI revolution of the future?, National Bureau of Economic Research, working paper 30734, December 2022.
3
Open Vault Blog, “Nonfarm payrolls: Why farmers aren’t included in jobs data,” blog entry by Maria Hasenstab, July 3, 2019.
4
See, for example, David M. Byrne, John G. Fernald, and Marshall B. Reinsdorf, Does the United States have a productivity slowdown or a measurement problem?, Federal Reserve
Bank of San Francisco, working paper 2016-03, April 2016.
Labor
5 productivity
5-year rolling
average¹
4 1949–2021
annual average
3 Real
compensation
5-year rolling
2 average¹
1948–2021
annual average
1
US recessions
0
1948 1971 1989 1995 2005 2019
Postwar boom Era of
(1948–71) contention Era of markets
1.8 1.9
1.5 1.4 Productivity and
1.1 compensation
0.7
0.3 averages for
each era, %
¹Data series for real compensation growth begins in 1948 and for labor productivity growth in 1949. Rolling averages in initial years are for fewer than 5 years,
reflecting that availability.
Source: US Bureau of Labor Statistics; US Federal Reserve economic data
It will not be easy, but some states, cities, sectors, and firms are growing productivity quickly and
set an example. For example, from 2007–19, Washington and North Dakota raised productivity at
a growth rate over 2.2 percent; California was close, at 2.1 percent. Seattle has powered growth in
Washington, and San Jose and San Francisco have led the way for California. Among sectors, mining
and information grew productivity faster than 2.2 percent, though together these represent only
10 percent of US output and 3 percent of working hours. Even in slower-growing sectors, such as
manufacturing, the most productive firms outpace laggards by a factor of five or more.
The $10 trillion goal poses difficult but not insurmountable challenges. In this research, we review
the patterns of productivity growth over the past 15 years, parse the challenges that lie ahead, and
identify implications for business leaders and policy makers.
US productivity is a story
of haves and have-nots
Averages obscure details that matter. Some states, sectors, firms, and cities have become
significantly more productive—but those gains have been far from universal. These patterns of
divergence highlight the unevenness with which the benefits of innovation, globalization, and other
shapers of productivity have spread across the United States. (A similar story has unfolded in other
nations; see sidebar “A global problem” for more.)
Seven states currently lead the way—California, Colorado, Massachusetts, New York, North Dakota,
Texas, and Washington. These leading states are both more productive and increasing productivity
faster than the US average. Together, they provide nearly one-third of the nation’s jobs and
40 percent of GDP.
7
Robert J. Barro and Xavier Sala-i-Martin, “Convergence,” Journal of Political Economy, April 1992, Volume 100, Number 2.
A global problem
The United States is not unique in its productivity challenges with declining productivity growth observed
across G7 nations since 2005 (exhibit). The United States starts from a position of strength, being the
most productive nation among the G7 in 2019. This has not always been the case: the United States
overtook France and Germany shortly after the 2007–08 financial crisis, a position it has held since.
Slowdowns in productivity growth can be seen across the G7 since 2005, with most countries
experiencing declines of 0.6 to 1.6 percentage points—equivalent to a halving of productivity growth.
Other countries confront more considerable challenges: Italy has barely sustained any productivity
growth at all, averaging only 0.1 percent per year since 2005. The United Kingdom has experienced the
most significant drop-off, falling from robust rates of 2.1 percent over 1995–2005 to 0.5 percent from
2005–19. These declines represent significant challenges to economic growth and individual prosperity
in these economies.
There is one area in which the United States is uniquely challenged—declining labor participation.
While each of the other G7 countries has experienced rising labor participation rates since the 1990s,
bolstering their total economic growth, labor participation in the United States has declined by 4 percent
since its peak in 1998. This adds a further headwind to economic growth, and one to be confronted in
concert with reviving productivity.
The US experience US
mirrors other
75
developed markets
with a 1-percentage-
point average decline
since 2005.
50
Labor productivity
by G-7 country,
output per hour worked,
2021 international dollars,
converted using purchasing 25
power parity, % 1989 1995 2005 2019
2.5
1.3
CAGR over the same 1.1
time periods, %
75 75 75
50 50 50
25 25 25
1989 1995 2005 2019 1989 1995 2005 2019 1989 1995 2005 2019
2.7 2.6 1.8
1.3 1.6 1.5
0.8 0.8 0.8
France UK Italy
US US US
75 75 75
50 50 50
25 25 25
1989 1995 2005 2019 1989 1995 2005 2019 1989 1995 2005 2019
1.9 1.7 1.9
2.3 2.1
0.7 0.5 0.6
0.1
US labor productivity in 2019 and productivity growth, 2007–19 Circle size = change in
total hours worked 2007–19
CAGR 2007–19, %
3.5
ND
Growing Leading
States moving up Top states
3 from a smaller base pulling away
Nonfarm
average $67
2.5
WA MA
CA
2
KS
OR NE MD
CO
MT SD
ID
KY NM MD
OK MA
1.5
UT GA NY
VT TX
WV NH VA
SC HI
1 PA
AL WI Nonfarm
TN MN IL
ME average 1.3%
MO RI OH NJ
FL NC
0.5
AR IN
AZ
MS IA MI DE CT
NV AK
0
LA WY
–0.5 Lagging Slowing
States falling States falling back
behind from a high level
–1.0
40 50 60 70 80 90 100
Real output 2019, $ per hour worked (2012 dollars)
NH
ME
States by category WA VT
MT ND
MN MA
OR NY
SD WI
ID MI
WY PA RI
AK IA OH
NE IL IN CT
NV NJ
UT WV VA
CO KS MO KY DE
CA NC MD
TN
OK AR SC
AZ NM
MS AL GA
LA
HI TX
FL
The remaining 18 states present mixed stories. Six states, representing 8 percent of employment and
9 percent of GDP, are slowing. They have above-average productivity—for now—but their average
productivity growth was near zero for the past decade. The story is reversed for 12 growing states
with below-average productivity but above-average productivity growth rate. Should these states
maintain their growth, they will continue to converge with the US average level of productivity.
Surprisingly, sector mix within states explains little of the variation in productivity levels and none of
the relative growth. High-productivity sectors have benefited some states, such as North Dakota,
which benefited from the boom in natural gas, and California, which reaped the benefits of the
technology boom. However, in aggregate, sector employment mix only explains a small share of the
productivity gap across states. 8 As an example, the eight most-productive sectors amount only to
28 percent of California employment—roughly the same share as in the United States overall.
Instead, both the absolute level of a state’s productivity and its growth rate seem to be driven by
differences within sectors. For example, in 2007, information firms in California were 18 percent more
productive than the US average, but this gap grew to 47 percent by 2017. In sectors such as retail,
accommodation, and transportation, leading states have double or more the productivity of states
that are lagging behind.
This means having a desirable sector mix is not enough—to secure productivity growth, states need
to provide an environment that enables firms to thrive across all sectors. That growth is necessary for
states (and cities) to fulfill their goal of delivering economic gains that are broadly shared.
Four sectors are leading, largely powered by adoption of digital technologies. Mining, information,
finance and insurance, and wholesale trade all enjoy above-average productivity and have grown
most rapidly since 2005. With the exception of mining, which has benefited from natural gas
innovation, all are among the most digitized sectors in the US economy. Information, which includes
businesses such as software, telecommunications, and internet publishing, has been a productivity
superstar, averaging 5.5 percent growth since 2005, propelled by internet services and software.
While these sectors represent 25 percent of the US economy, they only account for 15 percent of US
working hours.
8
State sector value-added statistics are available from the US Bureau of Economic Analysis. The US Census Bureau publishes
estimates of state employment level by sector.
Leading sectors in the United States are pulling away from the rest.
US labor productivity by sector,¹ sectoral output in 2012 dollars per hour worked
Productivity
2005 2019 Sizes of 2019 circles = working hours in 2019 CAGR
2005–19, %
0 100 200 300 400 500
Leading sectors
Mining 2.9
Information 5.5
Finance and insurance 1.5
Wholesale trade 2.1
Slowing sectors
Utilities 0.1
Real estate and
rental and leasing 1.0
Manufacturing 0.5
Management of companies
and enterprises 0.9
Growing sectors
Professional, scientific,
and technical services 1.3
Arts, entertainment,
and recreation 1.8
Retail trade 1.8
Administrative and waste
management services 2.2
Lagging sectors
Healthcare and
social assistance 0.5
Transportation
and warehousing 0.1
Construction –0.9
Other services,
except government 0.2
Accommodation
and food services 0.5
23.7 17.2
30.5 36.6
¹Private-education sector is excluded due to its small size. ²Figures do not sum to 100%, because they omit the private-education sector’s 0.8% share.
Source: US Bureau of Labor Statistics
Information
5
4
Wholesale trade
Administrative and waste management
~70%
Labor 3 Real estate and rental correlation between
productivity Retail trade productivity growth
CAGR Finance
Manufacturing and and digital adoption
1989–2019, % Arts and during the 1989–2019
insurance
entertainment Mining era of markets
2 Utilities
–1
0 1 2 3 4 5 6
¹Private-education sector is excluded due to its small size. ²The McKinsey Global Institute digitization index assesses digital assets, digital usage, and digital
workers using 27 indicators to capture the many possible ways in which companies are digitizing.
Source: US Bureau of Labor Statistics; “Digital America: A tale of the haves and have-mores,” McKinsey Global Institute, Dec 1, 2015
An additional four sectors representing 31 percent of the US economy are slowing: manufacturing,
real estate, utilities, and management of companies have experienced slower growth but continue to
outpace the US average. Employment in these sectors has shrunk, most notably in manufacturing,
which dropped from 15 to 12 percent of national employment between 2005–19. Globalization has
played an important role, shifting work in less productive subsectors such as textiles overseas, while
highly productive, R&D-intensive subsectors grew rapidly. It also meant US firms shifted their place in
global value chains, reducing their share in labor-intensive activities such as factory production while
retaining high-value activities such as design. The result was that subsectors such as semiconductor
manufacturing could shrink employment by 19 percent while expanding productivity by almost
50 percent.
In contrast, a group of services-led sectors (for example, accommodation and food services,
healthcare) are badly lagging on productivity growth while also being among the least-productive
industries. These sectors also provide a disproportionate number of jobs: while they contribute
only 24 percent of economic output, they are responsible for 37 percent of hours worked. Worse,
these sectors have accounted for more than two-thirds of employment growth since 2005, creating
headwinds for overall productivity as more workers shift into less-productive work. These sectors
tend to adopt technology more slowly than highly productive sectors do, attract lower-skilled
workers, and operate more prominently in local value chains, with limited exposure to global markets.
In sum, technology adoption and competition have proved central to sector-level productivity
improvement. Yet, the benefits of each have not been broadly shared to date.
The “frontier firms” in the productivity vanguard are accelerating away from their peers. These firms
tend to be larger, more connected to global value chains, and focus on technology-intensive aspects
of their sector. Research suggests these leading firms invest 2.6 times more in technology and other
intangibles such as research and intellectual property, and attract and invest in more skilled talent.11
As a result, the gap between frontier firms and laggards has grown over the past 30 years. In
manufacturing, the gap was 25 percent wider in 2019 than it was in 1989, with most of that change
happening before 2000. At the same time, industry dynamism has fallen, as seen in metrics such as
new firm entry rate (which has declined 29 percent from 1989 to 2019 in the United States) and labor
reallocation rates (which are down 31 percent across sectors).
Standard economic principles would suggest that less productive firms would be replaced or would
improve their performance. Researchers have offered multiple hypotheses for why this has not
happened. For example, there is evidence that firms within the same sector may coexist without fully
competing, by serving different customers, attracting different workers, or operating in different
geographic markets.12 Finally, some researchers have pointed to declining measures of competition
as a source of the divergence,13 which remains a matter of active debate.
Whatever the explanation for growing divergence, productivity gains must ultimately come from
firms. If laggards don’t catch up or get replaced by more productive firms, US productivity will
continue to splutter. For business leaders, the message is clear: improving your firm’s performance
matters much more than the productivity of the sectors in which you operate.
9
Leading firms are defined by those at the 90th percentile of productivity growth in their four-digit North American Industry
Classification System subsectors. Laggards are defined as those at the tenth percentile of productivity growth in their subsector.
10
See Chad Syverson, What determines productivity?, National Bureau of Economic Research, working paper 15712, January 2010;
Dan Andrews, Chiara Criscuolo, and Peter N. Gal, Frontier firms, technology diffusion and public policy: Micro evidence from OECD
countries, OECD Working Papers, 2015, No. 2.
11
“Getting tangible about intangibles: The future of growth and productivity?,” McKinsey Global Institute, June 16, 2021.
12
Mark Lijesen and Carlo Reggiani, “Specialization, generic firms and market competition,” SSRN Electronic Journal, January 1, 2019.
13
Germán Gutiérrez, Investigating global labor and profit shares, Society for Economic Dynamics, 2018.
A handful of coastal cities such as Houston, San Francisco, San Jose, and Seattle have particularly
benefited from these effects and jumped ahead of other areas of the United States. For example,
Seattle was 16 percent more productive than other cities in 2007 but had become 38 percent more
productive by 2019. These cities also account for much of the state-level performance—Seattle, for
example, drove productivity growth at three times the rate of surrounding areas within Washington
State while accounting for more than two-thirds of state GDP.
But being a city is no guarantee of growth; some cities have lost ground as they failed to spark
agglomeration effects or find renewal as once-productive sectors shrank. Even among the
40 fastest-growing metropolitan areas, several historically vibrant cities (Chicago, Detroit, Orlando)
have seen flat or negative productivity growth since 2005.
Increasing divergence among cities would matter less if talent moved freely across state and county
lines. However, over the past 15 years, interstate mobility has fallen by a shocking 49 percent; county-
to-county moves have fallen nearly as much (34 percent). Researchers point to many potential
causes such as the growth in occupational licensing, policies that tie benefits to state residency, and
high differentials in housing prices across regions.17 It remains uncertain whether pandemic-induced
moves or increases in remote work might reverse this trend.
Cities are the engine of US productivity, bringing together ecosystems of firms and talent to create
self-reinforcing agglomeration effects. Yet, productivity growth in cities is not predetermined; many
cities that are now lagging were once the standouts of their era.
14
“‘Superstars’: The dynamics of firms, sectors, and cities leading the global economy,” McKinsey Global Institute, October 24, 2018.
15
Enrico Moretti, The effect of high-tech clusters on the productivity of top inventors, National Bureau of Economic Research,
working paper 26270, September 2019.
16
Enrico Berkes and Ruben Gaetani, The geography of unconventional innovation, Rotman School of Management, working paper
3423143, July 2019.
17
See, for example, Benjamin Austin, Edward Glaeser, and Lawrence Summers, “Jobs for the heartland: Place-based policies in
21st-century America,” Brookings Papers on Economic Activity, 2018.
Looming challenges
make productivity
growth an imperative
Capturing the $10 trillion opportunity by returning US productivity to its long-term trend of
2.2 percent annual growth will require addressing existing challenges. Leaders will also need to
confront the future’s changing context; indeed, a recent perspective from McKinsey Global Institute
(MGI) suggests that we may be on the cusp of a new era. The most significant challenges today are
insufficient skilled labor supply, uneven technology adoption, and stalling investment. All will remain
relevant in the future as the workforce ages and the frontiers of technological possibility expand
while interest rates rise. At the same time, shifting geopolitical dynamics and the net-zero transition
introduce new complications that make productivity growth more needed than ever.
Historically, access to skilled talent has long been a prerequisite of productivity growth; the past
30 years have seen soaring returns to human capital for firms that invested in people.18 Those that
freed their top thinkers, engineers, and creatives to work on their toughest problems, and supported
them with technology and other intangibles, experienced outsize returns. Meanwhile, workers lacking
the skills the modern economy demands have been less fortunate, with many discouraged from the
labor force altogether, driving down labor participation.
Since the pandemic, three groups in particular have opted out of the labor force: aging baby boomers,
women with children,19 and men without college degrees. An additional source of human capital has
lately slowed. In the early years of this century, immigration bolstered the perpetual quest for skilled
labor. Immigrant numbers grew by 4.6 percent from 1990 to 2000. But growth slowed to 2.5 percent
from 2000–10 and just 1.1 percent from 2010–21. Another source is also under threat. As skilled
boomers retire, they will take important institutional or technical knowledge with them.
Reskilling workers to make up for the shortfalls and meet evolving job demands will be a massive
undertaking. To take a single sector as an example, MGI estimates 550,000 new energy transition
jobs will be available by 2030.20 Who will fill them?
Getting this right will not only unlock greater productivity growth but also address a second looming
challenge for the US economy: raising the labor force participation rate by bringing discouraged
workers back into the labor force.
Technology is advancing but even now its benefits aren’t broadly shared
Technology has lifted productivity for some sectors and firms, yet its benefits have not been fully
captured nor broadly shared.21 In a recent survey, companies report that digital transformations fail
five times more often than they succeed.22 A colossal opportunity awaits if the country can collect the
benefits of today’s technologies (never mind what’s to come) and ensure that its dividends are spread
economy-wide.
That “what’s to come” bit is even more tantalizing. New so-called transversal technologies such as
artificial intelligence and bioengineering are advancing by leaps and bounds. To unlock value from
truly new technology, firms must reconfigure how they work, often over sustained periods, as they
tinker with processes and workers adapt their skills. This is not a new phenomenon: historically,
revolutionary general-purpose technologies such as electricity took multiple decades before the
18
Łukasz Bryl, “Human capital orientation and financial performance. A comparative analysis of US corporations,” Journal of
Entrepreneurship, Management and Innovation, 2018, Volume 14, Issue 3.
19
“Labor force participation of mothers and fathers little changed in 2021, remains lower than in 2019,” US Bureau of Labor Statistics,
April 27, 2022.
20
“Toward a more orderly US energy transition: Six key action areas,” McKinsey, January 12, 2023.
21
Mohammed Aaser, Jonathan Woetzel, and Kevin Russell, “Five insights about harnessing data and AI from leaders at the frontier,”
McKinsey Global Institute, March 25, 2021.
22
Jacques Bughin, James Manyika, and Tanguy Catlin, “Twenty-five years of digitization: Ten insights into how to play it right,”
McKinsey Global Institute, May 21, 2019.
Firms’ readiness for these new technologies may be improving. The COVID-19 pandemic sped up
investment in AI and automation as firms responded to the needs of remote work.25 But productivity
measures since the pandemic have been mixed.
To be ready to capture value from digitization today and the new technologies of tomorrow, firms
will need to build their capabilities with the right investments in skilled talent, operating practices,
and platforms.26
Complicating matters, firms today are confronting a new capital markets reality. Interest rates
are higher, as is asset price volatility.27 More turbulence may be ahead as global balance sheets
deleverage. How this will impact investment is unclear: in the long term, real interest rates matter and
they remain low. But volatility can have a dampening effect on its own and is likely to have the most
effect on long-term investments.
Firms have certainly noticed that intangible investment (in R&D, software, and so on) has boosted
performance.28 Yet growth in these investments has been sluggish. This suggests that many
firms may have set their hurdle rate too high. There are several possible reasons: many intangible
investments may not have yielded a return, intangibles may be perceived as riskier given they are
difficult to collateralize and have limited recovery value, or firms may simply be taking a short-term
perspective and are unwilling to invest through the J-curve associated with intangibles investment.
Boosting productivity growth will require addressing this hesitancy to invest, even amid choppy
macroeconomic waters.
As firms now embark on the transition to net-zero emissions, the rising cost of energy could turn
into a headwind for productivity. Business leaders and policy makers will need to manage trade-offs
between the speed of transition and affordability. MGI estimates that for the United States to reach
net zero, investors will need to pump in an additional $1.7 trillion annually until 2050, much of which
is not currently planned.29 Freeing up those resources will require higher productivity growth. In the
short term, with energy markets in turmoil, investments in net zero could dampen productivity growth,
thereby also making subsequent investments needed for the transition less affordable. But the road
to net zero will also present some new opportunities to bolster productivity—for example, through
decreased volatility of energy costs—and for job creation—such as through new manufacturing or
construction jobs. 30
However, the productivity benefits of globalization were not uniform, with some states, sectors, firms,
and workers left behind. Sectors insulated from global forces (for example, healthcare) have seen
fewer benefits, while others—such as textiles, exposed to global competition but unable to respond—
have withered. And workers in many parts of the United States were not able to transition to new jobs
as their old ones moved overseas, fueling some of the observed geographic disparity. 32
In the new era taking shape now, trade policy and other economic institutions are likely to be revisited
as the world becomes more multipolar and as noneconomic strategic options are considered.
Policies that preserve the benefits of global interconnectedness while fostering resilience and better
distributing the gains of globalization could enhance productivity, as workers and value chains adjust
and build greater resilience to global shocks.
29
“The net-zero transition: What it would cost, what it could bring,” McKinsey Global Institute, January 2022.
30
Ibid.
31
“Global flows: The ties that bind in an interconnected world,” McKinsey Global Institute, November 15, 2022.
32
David Autor, David Dorn, and Gordon Hanson, “On the persistence of the China shock,” Brookings Institution, September 8, 2021.
Leaders can start by increasing the supply of newly trained workers. Improved access to education
through community colleges, apprenticeships, and vocational programs, as well as reforming the visa
process for skilled foreign workers, would provide a welcome boost. Private- and public-sector talent
reskilling programs must be scaled up to help more people acquire the skills they will likely need in
the future and match them with prospective employers. Skills acquired on the job explain a large
proportion of workers’ earnings and can be purposefully developed through on-the-job experience,
training, and mobility. 33
33
“Human capital at work: The value of experience,” McKinsey Global Institute, June 2, 2022.
We’ve argued throughout this report that the biggest improvements in US productivity will be the
result of the choices that business leaders make. The company will benefit from greater productivity,
as will the national economy. Here we provide a quick checklist of the ideas discussed in the
report. Most are steps that companies can take today. Others will require collaboration with other
institutions; we’ve put those in italics.
— Invest in on-the-job training and rotation programs that build your talent bench.
— Expand policies on childcare, elder care, and parental leave to retain top talent.
— Complement technology investments with R&D, brands, and other intangibles to reimagine core
business operations.
— Retool your organization to build the capacity and flexibility for reinvention.
— Invest in technology-ready talent at all levels to lead and deliver the change.
— Set long-term plans for the net-zero transitions to provide business units and employees time
to adjust.
— Engage suppliers and customers to form new ecosystems that can generate local agglomeration
benefits and co-innovation potential.
Finally, specific efforts will be needed to retain those leaving the workforce in disproportionate
numbers, such as women and retirement-age workers. Policies that improve access to affordable
childcare and make it easier to take parental leave would help parents (especially women) remain in
the workforce. Greater flexibility on the career path would make it more attractive for retirement-age
workers to stay in the workforce.
Government can also directly support innovation by coordinating across value chains, clarifying
regulatory standards, and streamlining policy to ease constraints on new investments. Agencies
such as NASA have done this successfully in the past, giving rise to new ecosystems of high-
tech manufacturing. Policies that ease land-use regulations could do the same for areas such as
renewables by making it easier to build solar, wind, and geothermal generation facilities and the
electrification infrastructure they require. Recently passed federal legislation (the CHIPS Act, 35
Bipartisan Infrastructure Law36) may help, containing potentially promising avenues to increase
research and development.
34
“Twenty-five years of insights,” May 21, 2019; Jacques Bughin and Tanguy Catlin, “3 digital strategies for companies that have fallen
behind,” McKinsey Global Institute, February 19, 2019.
35
“The CHIPS and Science Act: Here’s what’s in it, “McKinsey, October 4, 2022.
36
“The Inflation Reduction Act: Here’s what’s in it,” McKinsey, October 24, 2022.
Place-based policies attempt to address these “left behind” areas but have a checkered history,
with evidence of both success stories (for instance, the Tennessee Valley Authority) and failures.
Academic reviews of such policies have generally failed to see self-sustaining benefits. 40 Yet, people-
based policies such as the earned income tax credit have not been sufficient to address the scale of
the problem. Place-based policies should be explored further to identify specific interventions that
work and where such interventions are best targeted.
Returning labor productivity growth to its historical level is among the most important challenges
facing the US economy. Success will deliver not only greater prosperity, but also help finance the
net-zero transition, bring workers back to the job even as the population ages, and put the brakes on
rising inequality. Fortunately, the wellspring of American innovation remains strong even if unevenly
distributed. By taking action on several fronts, business leaders and policy makers can accelerate
productivity and create a more sustainable and inclusive economy in the new era.
37
Anne Case and Angus Deaton, Deaths of Despair and the Future of Capitalism, Princeton, NJ: Princeton University Press, 2020.
38
Benjamin Austin, Edward Glaeser, and Lawrence H. Summers, “Saving the heartland: Place-based policies in 21st century America,”
Brookings Institution, March 8, 2018.
39
Farid Farrokhi, Skill, agglomeration, and inequality in the spatial economy, Society for Economic Dynamics, 2019 Meeting Papers.
40
See, for example, David Neumark and Helen Simpson, “Place-based policies,” in Handbook of Regional and Urban Economics,
edited by Gilles J. Duranton, J. Vernon Henderson, and William C. Strange, Elsevier, 2015, Volume 5.
We are grateful to the academic advisers who challenged our thinking and added new insights:
Martin Baily, senior fellow in economic studies at Brookings, and Matthew Slaughter, Paul Danos
Dean and Earl C. Daum 1924 Professor of International Business at the Tuck School of Business,
Dartmouth College.
We thank McKinsey colleagues for their expertise, namely Jeffrey Condon, Ricardo Huapaya,
Max Klug, Javin Pombra, Maria Jesus Ramirez, Fernanda Reyes, Sven Smit, Bob Sternfels,
Sebastian Vargas, and Wei Su.
This report was edited and produced by editorial director Mark Staples, together with senior data
visualization editor Chuck Burke, and editorial operations manager Vasudha Gupta. We also thank
Rebeca Robboy for her support.
This research contributes to our mission to help business and policy leaders understand the forces
transforming the global economy. As with all MGI research, it is independent and has not been
commissioned or sponsored in any way by any business, government, or other institution. We
welcome your comments on this research at [email protected].