toward recoupling? - assessing the global impact of a ... · ludovic gauvin & cyril rebillard...

Toward Recoupling?

Assessing the Global Impact of a Chinese HardLanding through Trade and Commodity Price

Channels

Ludovic Gauvin & Cyril Rebillard

6th BoC/ECB Conference, 08/06/2015

The views expressed herein are those of the authors and do notnecessarily reflect those of the Banque de France, theEurosystem, the International Monetary Fund or its ExecutiveBoard. Any errors or omissions remain the sole responsibilityof the authors.

I China’s rapid growth over the past decade => driver of therise in energy and mineral commodity demand => benefitto commodity exporters (Jenkins et al., 2008).

I Slow down in the last few years and concerns about thesustainability of the growth model (Eichengreen et al.2012, IMF 2013b, RGE 2013).

I While a majority of analysts still view a soft landing astheir baseline scenario, several reasons can be put forwardto justify a more pronounced slowdown (overinvestment,debt and real estate trends).

Outline

China’s growth prospects and commodities

Methodology

Results

Towards Recoupling ?

Appendix

China’s growth prospects and commodities Methodology Results Towards Recoupling ? Appendix

Main arguments pointing to a hard landing :

1. Historical rebalancing precedents (followingoverinvestment)

2. Debt dynamics3. Probable real-estate bubble

L. Gauvin & C. Rebillard Impact of a Chinese Hard Landing on Commodity Exporters 6th BoC/ECB Conference, 08/06/2015 2 / 43


35

40

45

50

55% of GDP

China: Evolution of GDP components' shares

Sources: Datastream, CEIC,

National Bureau of Statistics

10

15

20

25

30

1978 1982 1986 1990 1994 1998 2002 2006 2010

Private consumption Public consumption InvestmentPrivate consumption Public consumption InvestmentExports Imports

FIGURE : China’s internal imbalances.



What does history tell us ?I RGE (2013) : The Myth of a Gradual Rebalancing and

Soft Landing for ChinaI 47 episodes of rebalancing after investment-led growthI Growth is 3.5 percent lower on average 5 years after the

investment peak (compared to 5 years before)I Imbalances in China much more acute

I Eichengreen et al (2012) : When fast-growing economiesslow down : International evidence and implications forChina

I China shares many of the characteristics of "slowdowneconomies"

I High investment-to-GDP ratio, undervalued currency,ageing population



60

80

100

120

160

180

200

220

% of GDP% of GDP China: Is credit growth excessive?

Source: People's Bank of China, National Bureau of Statistics

Note: Corporate bonds, other TSF, and TSF items of shadow banking

(Entrusted loans, Trust loans, Banks' acceptance bills) are computed

from flow data (starting point = 2002)

Note: Linear trend is computed using past data that was available at

the time of estimation

Note: According to Drehmann et al. (2011), the 12% threshold gives

11% of false positive crisis signals (when applied to private credit)

-20

0

20

40

80

100

120

140

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

11% of false positive crisis signals (when applied to private credit)

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Credit-to-GDP gap (RHS) Total Social Financing ex-Equity Truncated linear trend (1997-current)

FIGURE : China’s total debt surge



Korea

Israel

Qatar

Saudi Arabia

UAECyprus

Greece

Portugal

Spain

Lux.

1 000

1 200

1 400

1 600

1 800

Cement Production

per capita (kg/pers.)Cement Production & Level of Development (2000)

Brunei

China

Hong Kong

IndiaIndonesia

Japan

Malaysia

Taiwan

Thailand

Vietnam

Croatia

Czech Rep.

Hungary

Poland

Russia

Slovak Rep.

Slovenia

ArgentinaBrazil ChileEcuador

MexicoVenezuela

Bahrain

Egypt Iran

Jordan Kuwait

Lebanon

Libya

Morocco

Oman

TunisiaTurkey

AustraliaCanada

New ZealandUS

IrelandItaly

GabonSouth Africa

Austria

Belgium

Finland

France

Germany

Netherlands

Norway

Sweden

Switzerland

UK

0

200

400

600

800

0 5 000 10 000 15 000 20 000 25 000 30 000 35 000 40 000 45 000

GDP per capita (USD, PPP)

Asia Middle-East & North Africa Eastern Europe & ex-CIS North America & Pacific

Latin America Sub-Saharian Africa Peripherical Europe Western & Northern EuropeLatin America Sub-Saharian Africa Peripherical Europe Western & Northern Europe

Sources: US Geological Survey, IMF (WEO), calculations Banque de France - SERMI

FIGURE : Cement production and level of development (2000).



ChinaKorea

TaiwanCroatia

Lebanon

Libya

Oman

Qatar

Saudi Arabia

UAECyprus

Greece

IrelandPortugal

Spain

Belgium

Lux.

1 000

1 200

1 400

1 600

1 800

Cement Production


Brunei

Hong KongIndiaIndonesia

Japan

Malaysia

Taiwan

Thailand

Vietnam

Croatia

Czech Rep.

Hungary

PolandRussia

Slovak Rep.Slovenia

ArgentinaBrazil

ChileEcuador

MexicoVenezuela

Bahrain

EgyptIran

Iraq

IsraelJordan

Kuwait

Morocco

Tunisia

Turkey

AustraliaCanada

New Zealand

US

Italy

Gabon

South Africa

Austria

FinlandFrance

Germany

Netherlands

Norway

Sweden

Switzerland

UK

0

200

400

600

800

0 5 000 10 000 15 000 20 000 25 000 30 000 35 000 40 000 45 000 50 000 55 000


Asia Middle-East & North Africa Eastern Europe & ex-CIS North America & Pacific

Latin America Sub-Saharian Africa Peripherical Europe Western & Northern EuropeLatin America Sub-Saharian Africa Peripherical Europe Western & Northern Europe

Sources: US Geological Survey, IMF (WEO), calculations Banque de France - SERMI




China

Korea

Iran

Lebanon

Oman

Qatar

Saudi Arabia

Turkey

UAECyprus

Greece

Belgium

Lux.

1 000

1 200

1 400

1 600

1 800

Cement Production


Brunei

Hong KongIndia

Indonesia

Japan

MalaysiaTaiwan

Thailand

VietnamCroatia

Czech Rep.Hungary

Poland

Russia

Slovak Rep.Slovenia

ArgentinaBrazil

Chile

Ecuador

MexicoVenezuela

Bahrain

Egypt

Iran

Iraq

IsraelJordan

Kuwait

Libya

Morocco

Tunisia

Turkey

Australia

Canada

New ZealandUS

IrelandItaly

Portugal

Spain

Gabon

South Africa

Austria

Belgium

Finland

France

Germany

Netherlands

NorwaySweden

Switzerland

UK

0

200

400

600

800

0 5 000 10 000 15 000 20 000 25 000 30 000 35 000 40 000 45 000 50 000 55 000


Asia Middle-East & North Africa Eastern Europe & ex-CIS North America & PacificAsia Middle-East & North Africa Eastern Europe & ex-CIS North America & Pacific

Latin America Sub-Saharian Africa Peripherical Europe Western & Northern Europe

Sources: US Geological Survey, IMF (WEO), authors' calculations




Main arguments pointing to a hard landing :

1. Historical rebalancing precedents (followingoverinvestment)

2. Debt dynamics3. Probable real-estate bubble



60

70

80

90

100

Million barrels/day Oil: Apparent consumption by region

China

Japan & Korea

Other Asia Pacific

North America

0

10

20

30

40

50

North America

Latin America

Russian Federation

Other Europe

Western Europe

Middle East

Africa

Source: Bureau of

Resources and Energy

Economics (Australia)

0

2003 2004 2005 2006 2007 2008 2009 2010 2011

FIGURE : Oil consumption by region.



12 000

14 000

16 000

18 000

20 000

Thousand tons Copper: Apparent consumption by region

China

Japan

Rest of Asia

United States

0

2 000

4 000

6 000

8 000

10 000

United States

Rest of Americas

Europe

Africa

Source: Bureau of


Economics (Australia)

0

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

FIGURE : Copper consumption by region.



1 200

1 400

1 600

1 800

2 000

2 200

Millions of tons Iron ore: Apparent consumption by region

China

Asia-Pacific ex-China

0

200

400

600

800

1 000

1 200 Americas

Europe

Rest of the World

Sources: Bureau of


Economics (Australia),

authors' calculations

0

2003 2004 2005 2006 2007 2008 2009 2010 2011

FIGURE : Iron ore consumption by region.



Global VAR (first developped by Dees et al. 2007, Pesaran et al.2004).Widely used in the international macro literature :

I Useful to study international linkages with a limited timesample.

I Construct 1 VECMX by country.I For a given country, a given exogenous variable (for

example foreign GDP) is the weighted average of othercountries’ corresponding variables (other countries’ GDP).



VARX for each country :

xit = ai0 + ai1t +

p∑j=1

Φijxi,t−j +

q∑k=0

Γikx∗it−k + uit

By "pooling" estimated VARX one can rewrite the GVAR :

Xt = FXt−1 + Dt + Vt



Unconditional forecast mean and variance-covariance matrix :

µh = E1FhXT +h−1∑s=0

E1FsDT+h−s

Ωhh = E1

h−1∑s=0

FsΣF′sE′1

Conditional forecast is conceptually similar to counter-factualanalysis (Pesaran et al. 2007, Dubois et al. 2009).Conditional forecast mean :

µ∗h = µh + (s′hH ⊗ Ik)Ω(IH ⊗ Ψ′)[(IH ⊗ Ψ)Ω(IH(IH ⊗ Ψ′)]−1gH



Bootstrap R=1000 simulations :

1. For each r simulation we recalculate X(r)t :

X(r)t = FX(r)

t−1 + Dt + V(r)t

1.1 This allows to estimate F(r) (and intercept and trend).1.2 Compute µ

(r)h and µ

(r)∗h .

2. Calculate median and other quantiles for conditional andunconditional forecasts.



I 36 countries (88% of world GDP) + 1 Metal (MPI) block +1 oil block.

I From 1995 Q1 to 2014 Q3.I The impact of countries on "commodity blocks" is

weighted by the share of each countries in globalconsumption of metals and oil :

I MPI block : Copper and Iron ore consumption.I Oil block : Oil demand.

I Several "crisis dummies" for Asia, Russia, Brazil, Turkey,Argentina, GFC, Euro Area.



Countries Oil block Metal blockDom. For. Dom. For. Dom. For.

GDP X X X XInv. X X X XExp. XInf. X XREER X

Oil price X XOil surplus XOil prod. X

Metal price X XMetal invent. XMetal prod. X



The hard landing scenario

Some studies :I Pettis (2013) : China will grow at no more than 3% during

rebalancing.I Nabar & N’Daye 2013 (IMF WP 13-204) : downside

scenario in which China would grow at no more than 4%.



Our scenarios

Soft landing scenario :I GDP growth slowdown to 6% in 5 years.I Investment growth slowdown to 4% in 5 years.I Investment-to-GDP ratio around 42% after 5 years.

Hard landing scenario :I GDP growth slowdown to 3% within 2 years.I Investment growth slowdown to 1% within 2 years.I Investment-to-GDP ratio around 40% after 5 years.



−15

−10

−5

0

Asia

EM

EsAs

ia

Emer

ging

Cou

ntrie

sO

ther

EM

EsAS

EAN

Emer

ging

Cou

ntrie

s (e

x. C

hn)

Latin

Am

eric

a

Asia

EM

Es (e

x. C

hn)

Wor

ldAs

ia (e

x. C

hn)

Asia

AD

VsW

orld

(ex.

Chn

)Eu

ro A

rea

Adva

nced

Cum

ulat

ed g

row

th lo

st a

fter

5 ye

ars



3 possible transmission channels :

1. Commodity prices2. Real exports3. Investment

1 possible buffer :

1. Real effective exchange rate



Metal Price index Oil price

00Q1 03Q1 06Q1 09Q1 12Q1 15Q1 18Q1 21Q1

20

30

40

50

60

70

80

90

100

110

120

Met

al p

rice

ind

ex (

2011

=100

)

Soft Landing Hard Landing

00Q1 03Q1 06Q1 09Q1 12Q1 15Q1 18Q1 21Q120

40

60

80

100

120

140

160

Oil

pri

ce (

2011

=100

)




FIGURE : Cumulated export loss.

−35

−30

−25

−20

−15

−10

−5

0

Asia

EM

EsAs

iaAs

ia A

DVs

Asia

(ex.

Chn

)

Asia

EM

Es (e

x. C

hn)

ASEA

N

Emer

ging

Cou

ntrie

sW

orld

Emer

ging

Cou

ntrie

s (e

x. C

hn)

Adva

nced

Wor

ld (e

x. C

hn)

Euro

Are

aLa

tin A

mer

ica

Oth

er E

MEs

Cum

ulat

ed e

xpor

ts lo

st a

fter

5 ye

ars



FIGURE : Cumulated investment loss.

−40

−30

−20

−10

0AS

EAN

Oth

er E

MEs

Asia

EM

Es (e

x. C

hn)

Latin

Am

eric

a

Emer

ging

Cou

ntrie

s (e

x. C

hn)

Emer

ging

Cou

ntrie

sAs

ia E

MEs

Asia

(ex.

Chn

)As

iaW

orld

Wor

ld (e

x. C

hn)

Euro

Are

aAs

ia A

DVs

Adva

nced

Cum

ulat

ed In

vest

men

t los

t afte

r 5

year

s



Canada Peru

00Q1 03Q1 06Q1 09Q1 12Q1 15Q1 18Q1 21Q1

70

75

80

85

90

95

100

105

Rea

l Eff

ecti

ve E

xch

ang

e R

ate

(201

1=10

0)


can GDP growth rate yoy smoothed in %

00Q1 03Q1 06Q1 09Q1 12Q1 15Q1 18Q1 21Q1

90

95

100

105

110

115

Rea

l Eff

ecti

ve E

xch

ang

e R

ate

(201

1=10

0)


per GDP growth rate yoy smoothed in %

00Q1 03Q1 06Q1 09Q1 12Q1 15Q1 18Q1 21Q1

90

95

100

105

110

115

Rea

l Eff

ecti

ve E

xch

ang

e R

ate

(201

1=10

0)


per GDP growth rate yoy smoothed in %



00Q1 03Q1 06Q1 09Q1 12Q1 15Q1 18Q1 21Q1

−1

0

1

2

3

4

5

6

7

Gro

wth

rat

e di

ffere

nce

betw

een

EM

E a

nd A

DV




Limits

Financial :I Financial contagion=> Confidence=> InvestmentI Rise in risk aversion => Capital outflows from EMEsI Interaction with Fed tapering

Commodities :I Overestimation of the impact on metal pricesI Linear modelI Lack of supply considerations (inflexion of oil producers’

strategy)



Conclusion

I China : Imbalances, credit growth, real-estate bubble :towards a hard landing ?

I Large impact on both regional partners and commodityexporters

I Provides a possible interpretation for decoupling :I Decoupling = Emergence of China + Imbalanced growth +

Impact on commodity exporters ?I Hard landing in China may trigger partial recoupling



Thank you !



160

180

200

220

240

% GDP China: Total Debt (public & private)

Source: People's Bank of China, China Trustee Association,

National Bureau of Statistics

Note: Corporate bonds, other TSF, and TSF items of shadow

banking (Entrusted loans, Trust loans, Banks' acceptance bills)

are computed from flow data (starting point = 2002)

Note: Non-TSF items of shadow banking (Trusts' other AUM,

Informal lending) are extrapolated based on available data

60

80

100

120

140

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 20142002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Bank Loans Treasury Bonds Corporate Bonds Other TSF Shadow Banking Total Debt

FIGURE : China’s total social financing



160

180

200

220

240

index: start = 100China: Second-hand housing / Price-to-rent ratio

Beijing

Shanghai

Tianjin

Chongqing

Shenzhen

60

80

100

120

140

160 Guangzhou

Hangzhou

Nanjing

Wuhan

Chengdu

Source: SouFun

Holdings

60

janv.-06 janv.-07 janv.-08 janv.-09 janv.-10 janv.-11 janv.-12 janv.-13

FIGURE : Price-to-rent ratios in China’s ten largest cities.



200

220

240

260

280

index: start = 100China: Second-hand housing / Price-to-income ratio

Beijing

Shanghai

Tianjin

Chongqing

Shenzhen

60

80

100

120

140

160

180 Guangzhou

Hangzhou

Nanjing

Wuhan

Chengdu

Source: SouFun

Holdings,

National Bureau

of Statistics

60

janv.-05 janv.-06 janv.-07 janv.-08 janv.-09 janv.-10 janv.-11 janv.-12

FIGURE : Price-to-income ratios in China’s ten largest cities.



300

350

400

450

500

Jan 2001 = 100China: Real estate & Urbanization

Volume of

Annual Cement

ProductionSource: National Bureau of Statistics

0

50

100

150

200

250Floor Space

Completed:

Commodity

Building

Number of

New Urban

Residents per

year

0

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

FIGURE : Urbanization and real estate in China.



29,0

30,0

31,0

32,0

11,0

12,0

13,0

14,0

% of total% of total China: Repartition of Employees by sector

24,0

25,0

26,0

27,0

28,0

6,0

7,0

8,0

9,0

10,0

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Source: National Bureau of Statistics

Farming, Fishery & Mining Construction & Real Estate Manufacturing (Right Hand Scale)

FIGURE : China : Employment by sector.



150

200

250

300

350

USD Bn China: Decomposition of Trade Balance

Source: IMF DOTS

Asia includes: Developing Asia, Japan, Korea

Commodity exporters include: Africa, Middle-East, Commonwealth of

Independant States, Australia, South America

-150

-100

-50

0

50

100

1995 1997 1999 2001 2003 2005 2007 2009 2011

Total trade balance Bilateral: United States Bilateral: European UnionTotal trade balance Bilateral: United States Bilateral: European Union

Bilateral: Asia Bilateral: Commodity exporters Bilateral: Other



20

25

30

35

% of total exports Australia: Exports by destination

Source: IMF DOTS

0

5

10

15

1996 1998 2000 2002 2004 2006 2008 2010 2012

to US to EU to Japan to Brazil to China to India to Russia

20

25

30

35

% of total exports Brazil: Exports by destination

0

5

10

15

1996 1998 2000 2002 2004 2006 2008 2010 2012

Source: IMF DOTS

to US to EU to Japan to China to India to Russia

20

25

30

% of total exports Chile: Exports by destination

0

5

10

15

1996 1998 2000 2002 2004 2006 2008 2010 2012

Source: IMF DOTS


20

25

30

35

% of total exports Peru: Exports by destination

0

5

10

15

1996 1998 2000 2002 2004 2006 2008 2010 2012

Source: IMF DOTS




Global copper demand by end-use (2012)

Building Construction: Plumbing

Building Construction: Building Plant

Building Construction: Architecture

Building Construction: Communications

Building Construction: Electrical Power

Infrastructure: Power Utility

Source: International Wrought Copper Council

Infrastructure: Power Utility

Infrastructure: Telecommunications

Equipment Manufacture: Industrial: Electrical

Equipment Manufacture: Industrial: Non Electrical

Equipment Manufacture: Transport: Automotive

Equipment Manufacture: Transport: Other Vehicles

Equipment Manufacture: Other: Consumer durables

Equipment Manufacture: Other: Cooling equipment

Equipment Manufacture: Other: Electronic equipment

Equipment Manufacture: Other: DiverseEquipment Manufacture: Other: Diverse

FIGURE : Global demand by end-use : copper.



Global steel demand by end-use (2012)

Construction incl. structural & building work

Mechanical engineering

Automotive

Source: Metals Consulting International

Automotive

Other transport incl. shipbuilding & rail

Domestic appliances incl. electrical engineering

Metal goods & fabrication

Oil & gas exploration & transport

Other industries / miscellaneous

FIGURE : Global demand by end-use : steel.



Global oil demand by end-use (2010)

Transport

Source: International Energy Agency, Key

World Energy Statistics 2012

Industry

Non-Energy Use

Other (Includes agriculture, commercial

and public services, residential/heating,

and non-specified other)

FIGURE : Global demand by end-use : oil.



China: Energy sources by type (2009)

Coal

Source: US Energy Information Administration

Coal

Oil

Hydroelectric

Natural Gas

Nuclear

Other renewables

FIGURE : Sources of energy in China.



0,0

0,5

1,0

5 000

6 000

7 000

8 000

9 000

Million tons

Constant 2005 USD per

tonCopper: Prices & inventories

1,5

2,0

2,5

3,00

1 000

2 000

3 000

4 000

5 000

1980 1985 1990 1995 2000 2005 2010

Sources: International Copper Study Group, IMF (International

Financial Statistics), US Bureau of Economic Analysis

1980 1985 1990 1995 2000 2005 2010

Copper price Cumulated Refined Balance (proxy of inventories)

FIGURE : Copper : Prices and inventories.



FIGURE : Latin America : Negative correlation between commodityprices and sovereign spreads (Bastourre et al, 2013).


toward recoupling? - assessing the global impact of a ... · ludovic gauvin & cyril rebillard...

Documents