The US Economy is Slowing Down:Widespread Slowdown by IndustryOther Countries have also Slowed

Adapted from a presentation to the Business Leaders ForumJapan Productivity Center and The Conference Board

New York City, April 11-12, 2018

Martin Neil Baily

BrookingsResearch assistance from Siddhi Doshi. Research is part of a joint project with

Dany Bahar and Barry P. Bosworth

Productivity Growth Has Slowed in Advanced EconomiesAging Populations Mean Slow Labor Force Growth

Adverse Impacts on Living Standards, Government Budgets

• Almost all advanced economies have seen much slower growth in labor productivity (output per employee or output per hour worked).

• Labor productivity growth closely linked to average earnings– Over half of recent decline in median US real wage growth comes from slow productivity

growth. The remainder from the wider distribution of earnings and the decline in the share of income going to labor.

• Aging populations, lower share of population of working age– Japan and Europe reduced working hours in the 90s

– US long-term decline in labor force participation of older males (improved with recovery)

• Aging population and slower GDP growth: dangerous debt and deficits

• US growth picked up in 2018, but will slow to 2 percent growth or less.

Slower GDP Growth from Slower Productivity Growth;Aging Populations

Japan and Germany reduced hours worked per worker 1995-2004

-1

0

1

2

3

4

1995-2004 2004-2018

Log

per

cen

tage

po

int

1995-2004 2004-2018 1995-2004 2004-2018

USA Japan Germany

Contributions of Growth of Labor Productivity and Hours Worked to GDP Growth

Source: The Conference Board Total Economy Database, April 2019

Productivity Growth has Slowed in Almost All Mature Economies Productivity growth has even slowed recently in emerging economies

Trend growth of GDP per Person Employed using HP filter, Major Regions, 1970-2019

Source: The Conference Board Total Economy Database (adjusted version) April 2019

US Output per Worker Very High. Has Grown a little Faster than Europe and Japan

The underlying Gross Domestic Product levels are expressed in 2018 US$, converted using Purchasing Power Parities.Source: The Conference Board Total Economy Database (adjusted version) April 2019

Levels of GDP per Person Employed, 2018 and 2000.Size of the bubble represents per capita income relative to the US.

Brookings is Comparing Productivity in the US, Japan and Germany.Growth Slowdown is Widespread by Industry in these Economies

• Early results on US and Japan: Some common patterns in the slowdowns by industry

– Both countries saw slowdowns in manufacturing, finance, wholesale retail trade, transportation.

• Differences also

– US fracking boom. Japan big slowdown in mining

– Slowdown in Japan in information, utilities, real estate. Reasons to be determined

– US manufacturing slowdown much greater—computers and semicon

• Agriculture productivity slowed in both countries. Reasons to be determined

Which US Industries Have Slowed?Manufacturing, Wholesale Retail Trade, Finance, Transportation

Agriculture, forestry and fishingMining and quarryingManufacturingUtilitiesConstructionWholesale and retail tradeTransportation and storageInformation and communicationFinancial and insurance activitiesReal estatePrivate Business Sector

5.8 -0.3 6.1 0.5 -0.7 5.1 1.6 4.1 4.1 0.7 3.4

3.4 2.2 1.9 -0.7 -1.1 1.0 -0.1 3.9 1.3 2.0 1.1

Source: Calculations based on OECD Structural Analysis Statistics (STAN)

USA: Industry Labor Productivity Growth (Value-Added Per Hour Worked)

1995-2004 2004-2016

Which Japanese Industries Have Slowed?Mining, Utilities, Wholesale Retail Trade, Information, Finance,

Real Estate, Manufacturing

Agriculture, forestry and fishingMining and quarryingManufacturingUtilitiesConstructionWholesale and retail tradeTransportation and storageInformation and communicationFinancial and insurance activitiesReal estatePrivate Business Sector

2.0 4.0 3.3 2.5 -1.2 2.0 -0.3 5.5 1.1 1.6 2.2

1.4 -7.1 2.4 -3.5 0.9 0.4 -0.8 -0.2 0.3 0.3 1.0

Source: Calculations based on OECD Structural Analysis Statistics (STAN)

Japan: Industry Labor Productivity Growth (Value-Added Per Hour Worked)

1995-2004 2004-2016

9McKinsey & Company

Waning of the ICT boom played a role in the productivity growth decline

(tech sector example from McKinsey Global Institute, Japan data added)

US Sweden Germany France Japan UK

11.7

-1.1

3.2

10.5

3.02.8

5.1

2.6

8.6

6.1

1.0

0.4

2000-2004 2010-2014TFP growth in technology manufacturing

Percent

Reasons

▪ Exceptional boom in

tech productivity growth

at the turn of the century

▪ Complexity of innovation

▪ Change in setup of,

demands on, and

measurement of the

tech industry

10McKinsey & Company

Weakness in Capital Investment has Contributed to Slow Labor Productivity Growth

0.8

0.5

1995-2004 2004-2016

1.1

0.3

1995-2004 2004-2016

0.7

0.0

1995-2004 2004-2016

USA Japan Germany

Capital Deepening: Private Business Sector

Source: Calculations based on OECD Structural Analysis Dataset (STAN)

Note: Capital deepening contribution is the growth in the capital labor ratio times the share of capital income in value added.

11McKinsey & Company

Online sales are more productive than offline, but they still represent a small segment of total sales

(retail example from McKinsey Global Institute, Japan estimate added)

Online retail

~50

Store-based retail

~105

~2.0x

Share of online sales

% of total retail sales (excl. sales tax)

Productivity (Sales- COGS/employee)

In $K

This transformation also comes with lags and transition costs (e.g., initial duplication of structures and investment,

cannibalization of incumbent’s business)

3

7

7

8

8

10

14

OECD Study Finds Frontier Firms Pulling Away from the Rest in Labor Productivity(sample includes only small number of US and Japan firms)

Log-

po

int

dif

fere

nce

s fr

om

th

e st

arti

ng

year

Source: Andrews, Criscuolo, and Gal (2016)Note: global frontiers are the top 5% of companies with the highest productivity levels

In US, wider dispersion in labor productivity within industries, also suggests best practices not spreading as fast

Digital technologies harder to adoptInter-decile range of log labor productivity deviated from industry by year means

Source: Decker, Haltiwanger, Jarmin, and Miranda (2016)

What Can Governments and Companies Do?

• Artificial Intelligence, Machine Learning, Robots etc. Over-optimism about how fast these can improve productivity, but great potential– Managers must overcome internal resistance, be willing to invest.

– Figure out how to improve business models using new technologies.

• Competitive pressure forces companies to change.– McKinsey productivity studies showed importance of competing against global best practices.

– Governments must promote a competitive environment.

• Firms and Schools Cooperate to Train Workers in 21st Century Skills– Lack of skills a reason why best practices are not spreading.

– People fear the loss of their jobs from productivity growth, will need retraining.

What Can Governments and Companies Do? Continued

• Need a regulatory framework for the 21st century.– For example, health care is huge and efficiency held back by regulation.

– Efficient financial regulation to ensure safety but encourage business lending.

• Improve tax and budget policy– US provides investment incentives but federal budget is out of control

– Japan battles demand stagnation

• Demographic shifts– Both countries have aging populations. Encourage labor force participation

• Promote open trade and investment– Trans-Pacific Partnership was valuable, should not have been discarded

– Multilateral cooperation, protect intellectual property