Transcript slides - Harvard Kennedy School

Understanding
the Global Slowdown
Jeffrey Frankel
Harpel Professor of Capital Formation & Growth
Program on
China Business Leadership and Innovation
HKS Executive Education and Cheung Kong School of Business (CKGSB)
Wednesday, September 14th, 2016
3:30 to 5:00 pm.
Outline
1. Current global economic slowdown, 2015-16
2. Long-term slowdown
–
–
–
–
versus 1950-2010
Population aging
Productivity
Explanations in advanced countries.
3. Emerging Markets / developing countries
– Slowed GDP growth since 2013
– Fall in EM capital inflows.
4. Macroeconomic policies in US, Europe & Japan
–
–
•
Limits to monetary policy
Fiscal policy mistakes
Appendices:
i.
ii.
iii.
iv.
v.
Commodity prices
Trade slowdown
More on the growth slowdown
More on Emerging Markets
Slowdown in China
1. Current global economic slowdown, 2015-16
Measures of global economic activity have slowed,
before recovery from 2007-09 Global Financial Crisis was complete.
World Trade, Industrial Production, and Manufacturing PMI
(3-month moving average)
}
annualized percentage change
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
I) Current global economic slowdown
The IMF in July 2016 again downgraded GDP forecasts,
particularly in euro-zone & UK.
http://www.imf.org/external/pubs/ft/weo/2016/update/02/pdf/0716.pdf
Now 3.1% global growth.
I) Current global economic slowdown
Possible reasons for recent slowdown
• Risks in Europe
– Euro crisis (2010- )
– Political instability associated with immigration (2015- )
– Brexit (2016-)
• Risks in the US
– Uncertainty from the 2016 presidential election
– Has US lost interest in global leadership?
• Risks in China
– Will the convergence to sustainable growth entail a hard landing?
– Excessive financial turmoil in response
to peaking of RMB & stock market in mid-2015.
• The fall in global commodity prices?
– Bad for exporters, but good for importers.
– Not a cause of overall slowdown, but a symptom.
I) Current global economic slowdown
Since Asia is now a large share of the global economy,
its growth rate matters more directly to all of us.
I) Current global economic slowdown
Asia’s contribution to growth last year was less than
in other recent years, but still more than other regions’.
2. Long-term slowdown
• (2.1) The postwar period 1950-2010
– Historic achievement
– Demographics & Productivity
– Forecasts for the long term
• (2.2) Population aging
• (2.3) Slowing productivity growth
• (2.4) Explanations in advanced countries.
• Diminished technological progress.
• Secular stagnation.
• “Hysteresis”.
• (2.5) Tentative conclusion.
(2.1) The world saw unprecedented levels of GDP growth
in the 20th century, driven by both strong population &
income growth (though peaking in 1950-1974).
Contributions to global GDP growth
Compound annual growth rate, %
Source: McKinsey
Global Institute
GDP is many times the level of 50 years ago,
due to strong growth in both (1) population & (2) GDP per capita.
GDP/capita
$ thousand
2012 purchasing
power parity
Country population, billion
Based on data for 99 countries, 1964–2014.
SOURCE: The Conference Board Total Economy Database; McKinsey Global Institute analysis
Global GDP growth peaked above 4% in 1964-1974.
Since then, it has been on a declining path.
G20 (G19+Nigeria)
1 Assuming historical productivity growth is sustained for
the next 50 years. GDP growth split by employment and
productivity growth—historical and future projection
1 Compound annual growth rate, %
SOURCE: The Conference Board Total Economy Database; UN Population Division;
International Labour Organisation; McKinsey Global Institute analysis. Exhibit 19.
II) Long-term slowdown
(2.2) Population aging
Demographic dividend: In the 1990s, baby-boomers in
advanced countries began to enter peak earning years.
Age distribution of people, 1964–1990, G20 (≡ G19 + Nigeria)
SOURCE: United Nations Population Division; McKinsey Global Institute analysis
II) Long-term slowdown
Advanced countries are now losing demographic dividends.
Soon EM countries will too.
Age distribution of people, G20 (≡ G19 + Nigeria)
2014 1/
2064 1/
1/ Estimates based on
medium fertility scenario.
SOURCE: UN Population Division;
McKinsey Global Institute analysis
Global increase in workers (Employee compound annual growth rate):
pop. gr. offset by 0.2%
1964–2014: 1.7%
2014-2064: 0.3% * * 0.4%
fall in working-age share.
II) Long-term slowdown
EU work forces have already peaked.
China & Korea forecast to peak in 2024.
Employees in G20 (G19+Nigeria) 1990 – 2064 estimated
Billions, at best activity and unemployment rates, 2007-12
SOURCE: McKinsey Global Institute analysis
(2.3) Productivity growth over the past 50 years,
slowed in advanced economies (to 0.8%),
accelerated in EMs (to 5.6%)
Productivity growth 1964–2014E (%)
Compound annual growth rate, G20 (G19+Nigeria)
SOURCE: The Conference Board Total Economy Database; McKinsey Global Institute analysis, exhibit 15.
Despite some convergence, productivity per worker
in advanced economics is still much higher than in EMs.
Productivity level, in $ thousand per employee,
2012 purchasing power parity
G20 (G19+Nigeria)
}
SOURCE: McKinsey Global Institute analysis
Productivity growth
varies widely:
not all countries
have narrowed the
gap to the frontier.
SOURCE: The Conference Board
Total Economy Database;
McKinsey Global Institute analysis
1 Russia data before 1989 approximated
using the Angus Maddison historical series.
(2.4) Proposed explanations in advanced countries.
• Estimated growth of US potential output has slowed
– from 3 ½ % in late 1990s to 1 ½ % today (CBO). Why?
• Explanations for long-term productivity slowdown:
– Diminished technological progress (Robert Gordon).
• Current technological innovations, as impressive as they appear,
do not compare with the great inventions that drove growth
1870-1970.
– Secular stagnation (Lawrence Summers):
• The equilibrium real interest rate is now close to zero.
– “Hysteresis” from the Global Recession of 2007-09:
– Permanent effect on capital stock,
– on labor force participation
– or on skills.
(2.5) Tentative conclusion: In retrospect,
1950-2008 was a period of unusually rapid growth.
We are likely reverting toward more normal averages.
Growth is slowing on a long-term basis, due to (1) Slower population
growth; (2) longer life expectancy; (3) slower productivity growth.
GDP growth
(compound annual % rate)
1700-1900
1900-1950
1950-2014
2014-64
(projected)
Source: McKinsey Global Institute analysis
Projection assumes 1.8% productivity growth rate, = average for 1964-2014.
http://www.mckinsey.com/global-themes/employment-and-growth/can-long-term-global-growth-be-saved
3) Slowdown in EMs /developing countries
• (3.1) Slower GDP growth since 2013
• (3.2) Fall in EM capital inflows
• (3.3) Slowdown in China
(3.1) EMs contributed more to 2010-13 global growth,
but then slowed.
https://www.weforum.org/agenda/2016/04/5-charts-explain-imf-global-growth-forecast/
Growth in EM & developing countries slowed,
from 2010 to 2014-16
(3.2) Capital inflows to EMs fell sharply in 2011-15,
as they had in two previous cycles
Net Capital Inflows to Emerging Markets 1980–2015Q3 (% of GDP)
International
debt crisis
East Asia crisis
Global financial crisis
IMF, April 2016,
WEO, Fig. 2.1 & 2.10
www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
Net EM capital inflows over the past four decades have exhibited cycles.
A slowdown phase of one such cycle has been taking place since 2010.
EM capital outflows accelerated in 2013-15,
after “taper tantrum.”
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
EMs have been losing FX reserves since mid-2014
-- particularly China.
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
Some determinants of EM capital flows
• Internal (“pull factors”)
– Debt sustainability
– Political and economic stability
– For more on EM policies, see appendix.
• External (“push factors”)
– US monetary policy
– Global risk perceptions, as measured by VIX.
After Fed “taper talk” in May 2013,
capital flows to Emerging Markets turned negative.
Jay Powell, 2013, “Advanced Economy Monetary Policy and Emerging Market Economies.”
Speech at the Federal Reserve Bank of San Francisco Asia Economic Policy Conference, Nov.
http://www.frbsf.org/economic-research/publications/economic-letter/2014/march/federal-reserve-tapering-emerging-markets/
Another push factor: “Risk on / risk off”
Capital flows to EMs fall when risk fears (VIX) are high
(↓ in graph)
2008
GFC
Kristin Forbes, 2014 http://www.voxeu.org/article/understanding-emerging-market-turmoil
Notes: Data on private capital flows from IMF's IFS database, Dec. 2013. Capital flows are private financial flows to emerging markets
and developing economies. Volatility index measured by the Chicago Board's VIX or VXO at end of period. 2013 data are estimates.
See K.Forbes & F.Warnock (2012), “Capital Flow Waves: Surges, Stops, Flight and Retrenchment”, J. Int.Ec.
EMs have had to pay higher interest rates
after 2010, and especially since 2014.
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
4) Macroeconomic policies
in advanced countries
• (4.1) Limits to monetary stimulus.
• (4.2) Fiscal policy mistakes.
– US
– EU
– Japan
Monetary Policy: Major central banks eased
in response to the 2008 recession,
but effectiveness of monetary stimulus ran into limits.
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
Inflation is almost as low as in 2009.
Inflation fell sharply in the 2009 global recession.
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
By 2014, some central banks even felt the need
to set negative interest rates
(in Japan, euro-zone & some other European countries).
Central banks cannot do a lot more.
https://www.weforum.org/agenda/2016/04/5-charts-explain-imf-global-growth-forecast/
But advanced countries tightened fiscal policy
prematurely, starting particularly 2012-13.
Fiscal Policies (% of GDP)
Consider: US, EU, Japan.
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG
Fig.1.14
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
US fiscal stimulus in 2009-2010
quickly helped end the 2007-09 recession,
but was abruptly reversed in 2011-14, slowing the recovery.
% of
GDP
Source: Congressional Budget Office, Department of Commerce, Joint Tax Committee, Goldman Sachs Global
Investment Research “US Economics Analyst: The Outlook for Post-Election Fiscal Policy ,” GS, Aug. 5, 2016
In Europe, the countries with sharpest fiscal contraction
have suffered the worst losses in GDP.
Source: P.Krugman
May 10, 2012.
What about the critique that fiscal policy responded endogenously to the magnitude
of the countries’ difficulties? See Blanchard & Leigh.
Japan’s PM Abe had the right idea with his 3 arrows.
– 1st arrow:
The monetary easing of 2013 (QQE) boosted the stock
market, depreciated the yen, and offered some stimulus.
– But the 2nd arrow was fired in the wrong direction:
• Japan went through with the scheduled increase
in the consumption tax taking effect April 1, 2014,
– from 5% to 8%.
– As many had warned, Japan went back into recession,
i.e., growth turned negative for next two quarters.
QQE raised Japan’s stock market & exchange rate in 2013.
HR dissolved,
Nov. 2012 =>
“Abenomics”
“Outlook 2014 - Recovery on a shaky footing,” Special , Economic Research Dept., Rabobank November 13, 2013, https://economics.rabobank.com/publications/2013/november/outlook-2014-recovery-on-a-shaky-footing/
But Japan’s growth turned sharply negative in 2014 Q2.
Consumption
tax raised
Appendices:
i. Commodity prices
ii. Trade slowdown
iii. More on the growth slowdown
iv. More on Emerging Market policies
v. China’s slowdown
Appendix (i): Commodity prices
Oil & other commodity prices have undergone
two complete boom-bust cycles over the last decade.
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
When commodity prices fell,
exporters lost and importers gained.
but the net effect was not a cause of the slowdown.
Terms-of-Trade Windfall Gains and Losses
IMF, April 2016, WORLD ECONOMIC OUTLOOK: TOO SLOW FOR TOO LONG. Fig. 1.11
http://www.imf.org/external/pubs/ft/weo/2016/01/pdf/text.pdf
Appendix (ii): The trade slowdown
Trade used to grow twice as fast as GDP.
Since the 2008-09 recession, trade has grown much more slowly.
Emine Boz,
Eugenio Cerutti,
& Sung Eun Jung
(IMF, 2016).
Since the rebound from the great trade collapse of 2008–09, when world trade fell by much more than GDP,
global trade growth has slowed notably, both in absolute terms and relative to world GDP growth. This
slowdown has been more pronounced in emerging market and developing economies, where it intensified in
2015. See Hoekman 2015 for a compilation of studies analyzing the drivers behind the recent trade slowdown.
Growth in trade was rapid
during most of the post-war period.
Trade always used to grow faster than GDP
-- about twice as rapidly.
World Trade & Real GDP, 1980 – 2014 (2010 = 100)
But the trend of economicintegration across
national borders is not inevitable or irreversible,
– even if technological progress in transport
and communication is one-directional.
– In the period 1914-1945, political forces worked to turn
the clock back on globalization:
• tariff protection,
• discriminatory economic blocs,
• and war.
– They had the effects one would expect: Trade fell sharply.
– It could happen again.
When the Global Financial Crisis hit in 2008…
• there was a fear that countries might revert to
protectionism, as in the 1930s, and with similar results.
– The first two meetings of the new G-20 Leaders Summit
pledged to refrain from imposing new protection,
• December 2008 in Washington
• and April 2009 in London.
– But many were skeptical of the rhetoric.
• There was in fact no return to protectionism.
• And yet both the economic downturn and the fall in trade
turned out to be as bad as feared, or even worse.
Since 2008, global trade has indeed slowed.
Sept. 2015. p.16, Fig.5.B
The 2008-09 collapse in global trade was bigger
than could be explained by the fall in GDP.
2009
Bussière, Callegari, Ghironi, Sestieri & Yamano, 2013,
"Estimating Trade Elasticities: Demand Composition and the Trade Collapse of 2008-2009."
Trade still lags GDP, in particular, in EM economies.
Why has trade slowed so much?
Protectionist measures did not rise in 2007-09 recession
as much as they had in past recessions.
Data source:
WTO
The WTO trade restrictiveness indicators – capturing border measures such as tariff increases, import licenses,
or new customs control s- show a modest increase in the share of world trade covered by new import
restricting measures since the Great Recession (Fig. 13)…. These findings suggest that protectionist trade
policies are playing a negligible (if any) role … in the current trade slowdown…
Cristina Constantinescu, Aaditya Mattoo, and Michele Ruta, 2015,
"The global trade slowdown: cyclical or structural?" IMF WP 15/6, January.
Why has trade slowed so much?
1. Global supply chains have matured
– Vertical specialization has largely run its course.
2. Physical investment spending has slowed
– which is trade-intensive.
3. The structure of China’s economy is shifting
– away from manufacturing, toward services;
– away from exports, toward domestic demand.
The slowdown in trade has been highly correlated
with a slowdown in physical investment
(which tends to be import-intensive).
IMF WEO, Apr 2016, Fig. 1.13
The structure of China’s economy is “rebalancing.”
• China long had great success with manufacturing;
– growth was led by exports and investment.
• But recently it has tried to move toward services,
– with growth led by consumer demand, appropriately.
– The leaders decided at the Third Plenum in 2013.
• Services are less trade-intensive than manufacturing.
China’s exports & imports had risen especially fast,
even relative to GDP, before 2008.
(Followed by the US. EU trade/GDP had been flat.)
1997=1
China is shifting into services,
judging by the available data.
% of GDP
Source: Nicholas Lardy, PIIE
It’s good to look also at other data in China.
Railway data show that freight traffic is declining
(especially relative to passenger traffic).
% change
(year-over-year)
Source: Nicholas Lardy
It’s good to look also at other data
Decline in China’s output of industrial products,
2010-2015
% change (year-over-year)
Source: Nicholas Lardy
Trade’s relative importance in China has peaked,
in part because services are less trade-intensive.
% of GDP
Source: Nicholas Lardy
Appendix (iii): More on the Growth Slowdown
GDP growth in developed countries is declining,
due to slower rates of growth in both productivity and employment.
Developed economies (in G20)
1 Assuming historical productivity growth is sustained for
the next 50 years. GDP growth split by employment and
productivity growth—historical and future projection
1 Compound annual growth rate, %
Exhibit 19 SOURCE: The Conference Board Total Economy Database; UN Population
Division; International Labour Organisation; McKinsey Global Institute analysis
In developing economies, growth will slow due to fewer workers.
Emerging Market economies (from G-19 + Nigeria)
1 Assuming historical productivity growth is sustained for
the next 50 years. GDP growth split by employment and
productivity growth—historical and future projection
1 Compound annual growth rate, %
Exhibit 19
SOURCE: The Conference Board Total Economy Database;
UN Population Division; ILO; McKinsey Global Institute analysis
Manufacturing’s share of total employment first rises and
then declines as countries develop and become more prosperous.
Share of
manufacturing
employment %
Exhibit 4
Addendum table: Slowdown forecast, country-by-country
McKinsey Global
Institute analysis
Appendix (iv): More on Emerging Markets
After the currency crises of the late 1990s,
many EM countries adopted stronger policies
• In particular:
–
–
–
–
–
More flexible exchange rates,
Higher holdings of foreign exchange reserves,
Smaller current account deficits,
Less $-denominated debt,
Less pro-cyclical fiscal policy.
• As a result, were less vulnerable to the 2008-09 shock.
• But there has been back-sliding since 2009:
– especially on debt.
Some EMEs reduced the pro-cyclicality of their
fiscal policies after 2000
• Fiscal policy was traditionally pro-cyclical:
– E.g., spending would rise in booms & fall in recessions.
– I.e., destabilizing.
– Especially among commodity-exporters.
• After 2000, some achieved counter-cyclical fiscal policy
– Notably: Botswana, Chile,
– China, India, Korea, Malaysia…
Correlations between government spending & GDP
1960-1999
procyclical
Adapted from Kaminsky, Reinhart & Vegh (2004)
countercyclical
G always used to be pro-cyclical
for most developing countries (yellow)
66
procyclical
Correlations between government spending & GDP
2000-2009.
Adapted from Frankel, Vegh & Vuletin
(JDE, 2013)
countercyclical
In the decade 2000-2009,
about 19 developing countries
switched to countercyclical fiscal policy:
Negative correlation of G & GDP.
DEVELOPING: 43% (= 32 /75) countercyclical. It was 17% (= 13/75) in 1960-1999.
INDUSTRIAL: 86% (= 18 / 21) countercyclical. It was 80% (= 16/20) in 1960-1999.
Update of Correlation (G, GDP): 2010-2014.
procyclical
countercyclical
Back-sliding among
some after 2009
Thanks to Guillermo Vuletin
DEVELOPING: 37% (= 29 out of 76) pursue counter-cyclical fiscal policy.
INDUSTRIAL: 63% (or= 12 out of 19) pursue counter-cyclical fiscal policy.
Currency composition in the post-2003 capital inflows
shifted away from $-denomination, toward Local Currency.
Share of External Debt in LC
(Mean of 14 sample countries)
Wenxin Du & Jesse Schreger, Harvard U., Dec. 2014, “Sovereign Risk, Currency Risk, & Corporate Balance Sheets,” Fig.2, p.19 .
But FX bond issuance by nonfinancial corporates
has increased since 2009, raising vulnerabilities.
Foreign Bond Issuance: Nonfinancial Corporates (US$ billion)
Peru
Colombia
Chile
Brazil
Mexico
Total
80
70
60
50
40
30
20
10
0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Sources: Dealogic; and IMF staff calculations.
Cubeddu, Iakova, & Sosa IMF, February 2015
Corporate debt post-2008 swung back to $-denomination,
away from Local Currency, in some EMs.
Wenxin Du & Jesse Schreger, Harvard U., Sept. 2014,
“Sovereign Risk, Currency Risk, & Corporate Balance Sheets” p.18
Appendix (v): China’s growth slowdown
from 10% average of 1980-2010, to just below 7% in 2015 (officially).
It began well before the fall in the stock market & RMB.
Source: Reuters
https://www.weforum.org/agenda/2016/04/5-charts-explain-imf-global-growth-forecast/
Financial markets got upset in the summer of 2015.
China’s stock market bubble peaked June 12, 2015
the date of the 3rd increase in margin requirements.
Chinese foreign exchange reserves fell after June 2014.
Through August 2016
DATA SOURCE: PEOPLE’S BANK OF CHINA, via TRADINGECONOMICS.COM
Longer-term Chinese transition to slower growth path
• Possible reasons for China’s longer-term slowdown
– Demography
– Convergence. Diminishing returns to:
• Capital/labor ratio
• Rural-urban migration
• Technical catch-up …
– Middle-income trap? e.g., Eichengreen, Park & Shin (2012)
– Regression to the mean: Pritchett & Summers (2014)
• Transition with hard-landing or soft-landing?
Demographic factors are reducing growth rate in China
even more than in other countries
“Is there a middle-income growth trap?”
“Avoiding middle-income
growth traps,” P. Agénor,
O. Canuto, M. Jelenic ,
VoxEU, 21 Dec., 2012
Eichengreen, B, D Park & K Shin
(2011), “When Fast Economies
Slow Down: International
Evidence and Implications
for China”, NBER WP 16919
“Formal evidence on growth slowdowns and middle-income traps has suggested that at per capita
incomes of about US$16,700 in 2005 constant international prices, the growth rate of per capita GDP
typically slows from 5.6 to 2.1%.
Using regression and standard growth accounting techniques, recent analysis (Eichengreen, Park, &
Shin 2011) suggests that growth slowdowns are essentially productivity growth slowdowns.”
Pritchett & Summers (2014): Regression to the mean
fits the data better than middle-income trap
“Asiaphoria Meets Regression to the Mean,” NBER WP No. 20573, Lant Pritchett and Lawrence Summers
Transition to slower growth path
• Hard landing or soft landing?
– High investment in heavy industry is unsustainable.
– Debt
• Leverage becomes unsustainable when growth slows.
• Bad loans in the shadow banking system.
• $-denominated loans especially problematic
– as in other Emerging Market countries.
• Need to carry out reforms
– as decided at the Third Plenum of 2013:
• Rural land rights and hukou system
• Market orientation
• Environment.
Understanding
the Global Slowdown
Jeffrey Frankel
Harpel Professor of Capital Formation & Growth