Plenary Session on “Varieties of Stagnation? Europe, Japan and the US” Conference on “The Spectre of Stagnation? Europe in the World Economy”

FMM/IMK/Hans Böckler Foundation, Berlin, Germany October 23, 2015

Dr. Robert A. Blecker, Professor of Economics, American University Washington, DC, USA, blecker@american.edu

The US Economy Since the Crisis: Slow Recovery or Secular Stagnation?

Slow recovery and secular stagnation

The US economy is experiencing both the slowest and weakest recovery since World War II and a longer-term slowdown Sluggish recoveries from the last two recessions (2001 and

2008-9) are an indication of longer-term stagnation Some claim it’s not so bad (Ben Bernanke, Jason Furman)

The US has done better since 2008 than it did after 1929 True, but it’s not a high standard

The US is doing better than other advanced economies But most of the others (euro area, Japan) are also stagnating

The US economy is experiencing its slowest growth since World War II and the slowdown pre-dates the 2008-9 crisis and Great Recession Especially in terms of job growth/employment creation

(C+I)

Source: Jason Furman, “It Could Have Happened Here: The Policy Response That Helped Prevent a Second Great Depression,” Macroeconomic Advisers’ 25th Annual Washington Policy Seminar, September 9, 2015 (expanded version of remarks).

Comparisons with the Great Depression and the Euro Area

Average annual percentage growth rates Pre-crisis (1992-2007)

Crisis (2008-2009)

Recovery (2010-2015)

Advanced Economies United States 3.2 -1.5 2.3 Canada 3.0 -0.8 2.4 Euro area 2.1 -2.0 0.8 France 2.1 -1.4 1.0 Germany 1.5 -2.4 1.9 Italy 1.4 -3.3 -0.3 Spain 3.2 -1.2 0.1 Japan 1.1 -3.3 1.4 United Kingdom 2.8 -2.3 1.9

Emerging Market and Developing Economies ASEAN-5 4.9 3.9 5.4 China 10.7 9.4 8.2 India 6.7 6.2 7.3 Russia (data start in 1993) 2.0 -1.3 1.8 Latin America and the Caribbean 3.4 1.3 3.1 Middle East, North Africa, Afghanistan, and Pakistan 4.7 3.7 3.6 Sub-Saharan Africa 4.4 5.0 5.1

Source: International Monetary Fund, World Economic Outlook Database, April 2015 and July 2015 update, and author’s calculations. Data for 2015 are IMF staff projections from the July 2015 update.

The US only looks relatively good in comparison with the other advanced economies, which are doing even worse

Average annual GDP growth rates, US recoveries/expansions, last four business cycles

Cycle

(Peak-trough-peak)

First 16 quarters of recovery

Expansion (trough to peak)

Entire cycle (peak to peak)

1981Q3-1982Q4-1990Q3 5.2 4.3 3.4 1990Q3-1991Q1-2001Q1 3.4 3.6 3.3 2001Q1-2001Q4-2007Q4 3.2 2.8 2.6 2007Q4-2009Q2-2015Q2* 1.9 2.2 1.1

The current US recovery is the weakest in the last 35 years

Source: BEA, National Income and Product Accounts, release of August 27, 2015, www.bea.gov, and author’s calculations. Business cycles follow the NBER chronology. *2015Q2 is not a cycle peak, but is the most recent period available.

Each recovery and expansion has been weaker than the one before it since 1981.

Real GDP: Weakest and longest recovery in any business cycle since 1960

-5

-4

-3

-2

-1

0

1

2

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Perc

enta

ge c

hang

e re

lativ

e to

pea

k qu

arte

r

Quarters after GDP peak

Percent Real GDP losses in recessions since 19601960-Q2 1969-Q4 1973-Q4 1980-Q1 1981-Q3 1990-Q3 2001-Q1 2007-Q4

In the current recovery it took more than 6 years for employment to return to its previous peak.

Each of the previous three recoveries has been more

“jobless” than the previous one.

Employment: longer and more “jobless” recoveries since 1990

The 1990 and 2001 recessions were worse in terms of employment than GDP

80

90

100

110

120

130

140

150

1980 1985 1990 1995 2000 2005 2010 2015

Mill

ions

Total US Nonfarm Employment in Millions, Monthly, January 1980 to August 2015

Trend (peak-to-peak, March 1980 to February 2001)

Trend (peak-to-present, February 2001 to August 2015)

Source: US Bureau of Labor Statistics, www.bls.gov, Employment, Hours, and Earnings from the Current Employment Statistics survey (National), downloaded September 26, 2015, and author’s calculations.

Average Annual Growth of Employment (in millions)

Mar. 1980 – Feb. 2001 2.00

Feb. 2001 – Aug. 2015 0.66

The long-term slowdown in US employment growth (since 2001)

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

1947 1953 1959 1965 1971 1977 1983 1989 1995 2001 2007 2013

2013

US

dol

lars

Real median US family income, annually, 1947-2013

Source: Economic Policy Institute analysis of BLS, Current Population Survey Annual Social and Economic Supplement Historical Income Tables (Table F-5), in State of Working America, 2014, www.epi.org.

Demand-side causes of secular stagnation: increasing inequality

Inequality has worsened since the 1980s along several dimensions, including: Increased wage gaps between more and less educated

workers (so-called “skill premium”) Real wages have lagged behind labor productivity since

1980s Decreasing labor share of national income since 2000

Increasing shares of highest quintiles, top 1%, etc. in the distribution of household or family incomes

Rising remuneration of CEOs relative to employees This weakens most households’ ability to finance

consumption expenditures out of current disposable income

50

100

150

200

250

300

1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Inde

xes,

1960

-62

aver

age

= 10

0

Labor compensation (real wages and benefits per hour)

Labor productivity(output per hour)

Source: US Bureau of Labor Statistics, www.bls.gov, Major Sector Productivity and Costs, downloaded September 27, 2015, and author’s calculations.

Labor productivity and real hourly compensation, all employees, US nonfinancial corporate business sector, 1960Q1 to 2015Q2

85

90

95

100

105

1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015

Inde

x, 1

960-

62 a

vera

ge =

100

Source: US Bureau of Labor Statistics, www.bls.gov, Major Sector Productivity and Costs, downloaded September 12, 2013, and author’s calculations.

China joins WTO

Index of the labor share of value added, US nonfinancial corporate business sector, quarterly 1960Q1 to 2015Q2

Financial crisis

-2%

-1%

0%

1%

2%

3%

Bot

tom

fifth

Sec

ond

fifth

Mid

dle

fifth

Four

th fi

fth

Top

fifth

Top

5 pe

rcen

tAver

age

annu

al p

erce

ntag

e gr

owth

rate

Average annual US family income growth by quintileand top 5 percent, various time periods1947–1979 1979–2007 2007–2013

Source: Economic Policy Institute analysis of BLS, Current Population Survey Annual Social and Economic Supplement Historical Income Tables (Table F-3) , in State of Working America, 2014, www.epi.org.

0

2

4

6

8

10

0

5

10

15

20

25

1913 1923 1933 1943 1953 1963 1973 1983 1993 2003 2013

Share of top 1% (left scale)

Share of top 0.01% (right scale)

Shares of top 1% and top 0.01% in US income, 1913-2014

Source: World Top Incomes Database, http://topincomes.parisschoolofeconomics.eu/, last updated July 25,2015, downloaded September 27, 2015.

Household borrowing delayed the stagnation of demand

Rising inequality and stagnant earnings were counteracted by increasing household debt and asset price bubbles during the expansions of the late 1990s and early 2000s This was encouraged by a deregulated financial system that offloaded the

risk from the banks that originated the loans into securitized assets Financialization/money manager capitalism

But the debts and bubbles were unsustainable Godley (1999); Pollin (2003); Cynamon, Fazzari, and Setterfield, eds. (2013);

Lavoie and Stockhammer, eds. (2013); Hein (2012); and many others

Most of the increased debt was in mortgages, used to pay for housing

But some of the borrowed funds can be used directly or indirectly to pay for consumption

Rising debt service burdens, the collapse of house prices, rising unemployment, and tighter credit conditions after the bubble burst in 2007 put an end to this household debt-led consumption boom

0

20

40

60

80

100

120

140Pe

rcen

t of d

ispo

sabl

e pe

rson

al in

com

e

Consumer debt (credit cards etc.)

Mortgage debt (housing loans)

Sources: Federal Reserve, Financial Accounts of the United States (Z.1), http://www.federalreserve.gov/releases/Z1/default.htm; release date 9/18/15; BEA, NIPA Table 2.1, www.bea.gov, release date 9/25/15; all data downloaded 9/27/15; and author’s calculations.

US household debt as a percentage of disposable personal income, quarterly 1960Q1 to 2015Q2

90

100

110

120

130

140

150

160

170

1991 1994 1997 2000 2003 2006 2009 2012 2015

Inde

x, 1

991Q

1 =

100

Source: FHFA, www.fhfa.gov, data released 9/22/15; BEA, www.bea.gov, NIPA release of 9/25/15; all downloaded 9/27/15; and author’s calculations

Ratio of Federal Housing Finance Agency (FHFA) purchase-only house price index to the Bureau of Economic Analysis (BEA) chain-type price index for gross domestic product, both seasonally adjusted, and rebased to 1991Q1 = 100.

US Real Housing Price index, quarterly, 1991Q1 to 2015Q2 (SA)

Investment has not increased to take up the shortfall in aggregate demand

In spite of record high profits and low interest rates, business fixed investment as a share of GDP has been trending downward since 2000 Firms seem to be responding mainly to the lack of demand growth

(strong accelerator effect) Contrary to some studies that have found “profit-led growth” in the US This also shows the weak impact of monetary policies (conventional and

QE)

Both housing and business construction are still depressed Business investment in equipment and “intellectual property” has

recovered relatively more, but not enough to make up for weaker construction

Firms are diverting profits to stock buy-backs, mergers & acquisitions, FDI, and other uses that don’t augment domestic capital stocks

5

10

15

20

25

1980 Q1 1985 Q1 1990 Q1 1995 Q1 2000 Q1 2005 Q1 2010 Q1 2015 Q1

Per

cent

age

Business fixed investment share of GDP and profit share of corporate value added

Gross domestic business fixed investment share of GDPNet operating surplus share of value added, total corporations

Profit share

Investment rate

Source: US Bureau of Economic Analysis, www.bea.gov, NIPA Tables 1.1.5 and 1.14, data release of 25 September, 2015; downloaded 2 October, 2015; and author’s calculations.

0

1

2

3

4

5

6

7Pe

rcen

t per

yea

r

Federal funds rate 3-month Treasury bill rate10-year bond yield Moody's Aaa corporate bond rate30-year mortgage rate

US interest rates, monthly, January 2005 to August 2015

Source: Board of Governors of the Federal Reserve System, Statistical Release H.15 Selected Interest Rates, http://www.federalreserve.gov/datadownload/, downloaded September 25, 2015.

Source: US Census Bureau, www.census.gov, Business and Industry Surveys, New Residential Construction, downloaded September 25, 2015.

0

50

100

150

200

250

Thou

sand

s of

hou

sing

uni

ts (p

er m

onth

)US Housing Units Started, Monthly,

January 1970 to August 2015 (not seasonally adjusted)

Impact on growth of potential output

The slower accumulation of capital reduces the growth of potential output on the supply side The decrease in growth of potential output is endogenous and results

largely from the demand-side slowdown (L. Ball and others) Reduced labor force participation and slower human capital

accumulation also contribute “A reduction in capital deepening—which we view as mostly an

endogenous response to weak demand—caused almost half of the cumulative shortfall in potential output from its pre-crisis trend” Reifschneider, et al. (2013, p. 33)

Decreased labor force participation and depreciated human capital of long-term unemployment workers also contribute

Fiscal policy: Austerity US style

Tax cuts under Reagan (1980s) and George W. Bush (2000s) generated chronic budget deficits, which were later used as an excuse to slash social expenditures and public infrastructure investment

Stimulus policies during the crisis: too little, too late, too short A small tax cut under Bush, spring 2008 Combined tax cuts and spending increases under Obama, 2009-10

Government expenditures have been cut repeatedly since the Republicans retook control of the House of Representatives in 2011 and Senate in 2013 The debt ceiling compromise (2011) led to the fiscal cliff and sequestration

(2012-13) The Republicans have demanded spending limits as a condition to avert a

government shutdown or raise the (artificially imposed) debt ceiling

The result has been a serious “fiscal drag” on the recovery

Source: US Bureau of Economic Analysis, www.bea.gov, NIPA Table 1.1.6, data revised August 27, 2015, and author’s calculations. Note: Total government expenditures include federal, state, and local.

90

95

100

105

110

115

120

125

130

135

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

Inde

x, p

revi

ous

cycl

e pe

ak =

100

Quarters after previous cycle peak

US Real Government Expenditures in the Last Four Recessions and Recoveries (up to 30 quarters)

1981-Q3 to 1989-Q1 (Reagan)

2001-Q1 to 2007-Q4 (Bush II)

1990-Q3 to 1998-Q1 (Bush I - Clinton)

2007-Q4 to 2015-Q2 (Bush II - Obama)Republicans take over House of Representatives

Obama stimulus

Underlying structural causes

The story of greater inequality and a reduced wage share creating a tendency toward weaker consumer demand, which was offset by household borrowing and asset price bubbles until 2007, has been told many times

But what were the causes of increased inequality and a lower wage share in the US? There are many causes; I will focus on the ones most directly related

to secular stagnation of income and employment

The causes are structural and related to the deteriorating performance of the US labor market and the changing position of the US in the global economy

Changes in the composition of output and structure of industries

The share of manufactures and other goods in GDP is falling; the share of services is rising Driven by a rising trade deficit in manufactures and the vertical

disintegration of industries (offshoring of intermediate goods and assembly operations) as well as profound technological changes

What remains of US manufacturing needs relatively little labor as a result of technological innovation and vertical specialization The most labor-intensive operations are outsourced

This has a two-sided effect on labor markets and inequality: Downward pressure on wages

Most of the growth in manufactured imports comes from low-wage countries (Mexico, China, other developing and emerging economies)

Decline in high-wage employment As manufacturing shrinks and low-wage services expand

-40

-35

-30

-25

-20

-15

-10

-5

0

5

10

15

1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014

Per

cent

age

of v

alue

add

ed

US manufacturing trade balance as a percentage of value added in manufacturing, annually, 1978-2014

Source: Census, FT900, various years, www.census.gov; BEA, International transactions accounts (release of September 17, 2015) and GDP by industry (release of April 23, 2015), www.bea.gov, all downloaded October 3, 2015, and author’s calculations. Note: trade balance data for manufacturing in 1978-1991 are author’s extrapolations (backward forecasts). Details available on request.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1986 1990 1994 1998 2002 2006 2010 2014

Shares of US imports of goods, annually, 1986-2014

Major industrialized countries

Other developing and emerging countries

China + Mexico

Source: BEA, International Transactions Accounts (balance of payments basis), Table 2.1 and historical Table 2b, release of September 17, 2015, downloaded October 3, 2015, and author’s calculations.

6

8

10

12

14

16

18

20

1980 1985 1990 1995 2000 2005 2010 2015

Mill

ions

US Manufacturing Employment in Millions, Monthly, January 1980 to August 2015

Source: US Bureau of Labor Statistics, www.bls.gov, Employment, Hours, and Earnings from the Current Employment Statistics survey (National), downloaded September 26, 2015, and author’s calculations.

Changes in US manufacturing employment (in millions)

1980 to 2001 −2

2001 to 2015 −5

Cumulative 1980-2015 −7

The impact of a rising share of services in GDP on employment in cyclical recoveries

A growing proportion of services industries implies a slower recovery of employment after recessions (Olney & Pacitti, 2015) Service producers don’t need to restock inventories in

anticipation of increased demand Also few services are exported so exports don’t provide a

boost in the recovery Much of the recovery in manufactures is transmitted to the

exporting nations that supply US imports

These factors leader to longer recoveries for employment after a cycle trough, as we have seen following the past three recessions (1991, 2001, 2008-9)

The impact of a rising share of services in GDP on long-term employment trends

Services are not all equivalent (Basu & Foley, 2013) Some services have “measureable value added” (MVA) These sectors create jobs in proportion to value added

Other services don’t have measurable value added Their “output” is imputed based on income received, and not

closely related to job creation Especially “FIRE” (finance, insurance, real estate)

The growing proportion of non-MVA services in total GDP makes employment growth become delinked from (measured) output growth Thus employment growth has become weaker relative to GDP

growth in the last several business cycles and secularly since 2001

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1960 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008 2012

Per

cent

age

of G

DP

Composition of US Output (Value Added by Industry), Annually, 1960-2012

Services with measurable value added (wholesale and retail trade, transportation, information, arts and entertainment, accommodation and food, administrative)

Manufacturing

Other goods*

Finance, insurance, and real estate (FIRE) services

Other services with output imputed by income (professional and management, educational, health, government, and other)

Source: BEA, www.bea.gov, GDP by industry tables, release of April 25, 2013, and author’s calculations. Categories suggested by Basu and Foley (2013). *Other goods consist of agriculture, forestry, fishing, mining, and construction.

Goo

ds

Mea

sura

ble

valu

e ad

ded

Out

put i

mpu

ted

by in

com

e

Manufacturing exporters: China, Germany, Japan, East and South

Asia, Mexico

Deficit countries, demand

generators (US, UK, Southern

Europe)

Resource Exporters: South America, Africa,

Middle East, Australia, Canada

Principal Net Global Demand Flows: A Schematic View

A weakening of the US role in generating global demand is contributing to stagnation or slower growth in the two other groups

Global impact of US stagnation

In the 1990s and early 2000s, global growth was sustained by a triangular pattern of trade imbalances, financial flows, and demand transmission The US current account deficit and corresponding East Asian surpluses Intraregional imbalances and demand flows within the Euro Area, North America, and

East Asia

This pattern of imbalanced growth led to rising commodity prices for resource exporters in the early 2000s The commodity price boom has now ended, causing growth slowdowns in primary

commodity exporting nations During the boom, Dutch disease caused deindustrialization in mixed primary-

manufactures economies, e.g. Brazil and Canada

Manufacturing exporters (Germany, China, Mexico, etc.) ultimately relied on debt-driven household demand in the deficit countries (chiefly the US, also the European periphery) for their export-led growth The fallacy in this strategy is that it ultimately weakens the very source of its dynamic by

undermining employment and incomes in the deficit nations that generate the demand

-7

-6

-5

-4

-3

-2

-1

0

1

2

1980 1985 1990 1995 2000 2005 2010 2015

Per

cent

age

of G

DP

US current account balance as a percentage of GDP, quarterly, 1980Q1 to 2015Q2

Source: BEA, National Income and Product Accounts, Tables 1.1.5 and 4.1, revised September 25, 2015, www.bea.gov, and author’s calculations.

A smaller US current account deficit implies a weaker demand impulse to the rest of the world since 2008

Global economic prospects

The US is not likely to be as strong a generator of global demand in the foreseeable future as it was before 2008 The continued US large trade deficit in manufactures is only partly offset

by improved trade balances in services, and investment income Smaller US current account deficits imply less transmission of demand

stimulus to the rest of the world

To avoid sustained global deflationary pressures, the surplus nations and resource exporters need to generate more of their own demand, and not rely so much on the US or other deficit countries to be the locomotives of growth in the coming years Enhanced “South-South” trade (among developing and emerging

market countries) can play a pivotal role But the major surplus countries (Germany, China, Japan) also need to

do their part to prevent a deflationary adjustment

Summary

The US economy is currently locked into a trajectory that implies a tendency toward secular stagnation as a result of these interrelated factors: Underlying weakness of household demand due to stagnant real

wages and increasing inequality Once the artificial fillip of excessive borrowing and debt is removed

Structural changes leading to reduced employment generation in proportion to output growth and a shrinkage of high-wage manufactures

Weak private sector investment in spite of record profits and low interest rates

Political gridlock and the imposition of austerity in fiscal policy Reverberations from foreign slowdowns especially in the euro area

and resource exporters, which are key markets for US exports

Tendencies vs. predictions

The last time I predicted US stagnation was in a 1994 book chapter Very bad timing; I learned a lesson

Any prediction must be conditional on the absence of counteracting forces

Possible offsets include: A new financial bubble and debt-led spending boom

We should not underestimate the ingenuity of Wall Street or the short memories of policy makers and private agents

Of course it would not be sustainable

An enhanced role for the public sector For example, though public investment in infrastructure, clean energy (solar, wind), and

other social needs – or a new military build-up?

Restoring Hyman Minsky’s “big government” and possibly creating an “employer of last resort” for countercyclical purposes

None of this will make a dent in inequality unless it creates sustained high employment leading to a recovery of wages relative to productivity, and unless the underlying structural problems are also addressed