Prakash Loungani Oil Research

The Elusive Quest for Energy Security

Presentation by Prakash LounganiPresentation by Prakash Loungani(based on joint work with Gail Cohen & Fred Joutz and with Marianna Riggi)

OPEC (Research Department)February 28, 2011

The views expressed are those of the presenter and should not be attributed to the IMF or other institutions.

Motivation Concern of policymakers all over the globe U.S. Congress: 180+ bills mention ‘energy security’ Other countries

An Energy Policy for Europe, European Commission, 2007 Cebu Declaration on East Asian Energy Security, 2007

Energy independence is also a popular concern

Recent books on energy securityRecent books on energy security

See Loungani (2009), “The Elusive Quest for Energy Independence,” International Finance, for a critical review of these books

Organization of the talkSh i Short-run security Increased diversification in sources of supply

Medium-run security Lower macro sensitivity to energy shocks

i Long-run security Lower use of energy per unit of income

Overall framework similar to that laid out in LaCasse and Plourde, Energy Journal, 1995

Main findings on diversification Di ifi i i f il li h Diversification in sources of oil supplies has remained constant over past decade, but diversification of natural gas supplies has increased.diversification of natural gas supplies has increased. Given growing importance of natural gas in energy consumption, this points toward greater energy

itsecurity.

Transportation of supplies through “choke points” Transportation of supplies through choke points remains a risk.

Main findings on macro sensitivityS l f h l d l i i i Several factors have led to a lower macro sensitivity to energy shocks today than in the 1970s.

Better central bank response Lower wage rigidities Declining shares of energy in production & consumption

Th h l b h “ i ” There may nevertheless be short-run “uncertainty” effects from large supply disruptions

Main findings on energy efficiency

General trend is toward lower use of energy per unit of income; some question marks about natural gasof income; some question marks about natural gas intensities in some countries

Alternatives to oil use in transportation sector remains a key issue

Take-away messagesd i d i i i h h Trend is towards increasing energy security, though

vulnerability to shocks remains.

“security of demand is an integral part of energy security … Uncertainties in demand pose investment risks for the group and a waste of valuable resources.” (H. Qabazard, 2009)

“[an] unilateral approach to energy issues will not succeed [an] unilateral approach to energy issues will not succeed, but an inclusive approach to promote cooperation between producers and consumers to best serve the interests of all players and enhance global economic prosperity.” -- G. Bahgat, OPEC [Energy] Review, Sep. 2006

SHORT-RUN SECURITYINCREASED DIVERSIFICATION IN SOURCES OF SUPPLY

Oil Production and Net Exports, by Major Countries

Share of Oil Production in Total World Production, 1992

Share of Oil Production in Total World Production, 2009

Libya2.3%

Canada2.3% Nigeria

3.1%Norway3.5%

United Arab Emirates

3.6%

Venezuela4.2%

China4.8%

Rest of the World26.5%

,

Canada2.9% Norway

3.0% United Arab Emirates

3.1% Iraq3.3%

Kuwait3.4%Venezuela

3.4%

Mexico3.9%

Rest of the World33.1%

,

Mexico4.9%

Iran5.7%

United States

Saudi Arabia14.1%

Iran5.5%

China5.6%

United States7.5%

Russia11.9%

Russia13.0%

Saudi Arabia11.7%

13.6%

Indonesia2.1% Oman

2.5%Kuwait2.7%

Libya4.1%Saudi Arabia

24.0%

Share of Net Exporters in Total Net Exports of Oil, 1992

Libya3.7% Venezuela

4.2%Mexico

4 3%

Saudi Arabia19.1%

Share of Net Exporters in Total Net Exports of Oil , 2008

Venezuela4.9%

Mexico5.4%

Nigeria5.9%

Rest of the World16.2%

4.3%

Iraq4.8%

Kuwait4.8%

Norway4.9%

Angola5 0%

Rest of the World19.1%

Norway6.6%

United Arab Emirates

7.1%

Iran9.1%

Russia9.4%

5.0%

Nigeria5.2%

United Arab Emirates

5.8%Iran6.2%

Russia12.8%

Gas Production and Net Exports, by Major Countries

Share of Gas Production in Total World Production, 1992

Share of Gas Production in Total World Production, 2009

Algeria2.5%

United Kingdom

2.7% Indonesia2.9%

Netherlands3.6%

Canada5.9%

Russia29.9%

Total World Production, 1992Indonesia

2.6%Algeria2.9% China

3.0%Qatar3.1%

Norway3.5%

Iran4.8%

Canada5 1%

Rest of the World37.4%

,

United States24.2%

R t f th

5.1%

Russia18.6%

Rest of the World28.3%

United States19.1%

Malaysia2.0%

Rest of the world5.4%

Norway

Share of Net Exporters in Total Net Exports of Gas, 1992

Australia2.1% Egypt

2.2%Nigeria2.6%

Malaysia2.8%

Netherlands

Russia23.0%

Share of Net Exporters in Total Net Exports of Gas, 2008

6.0%

Indonesia7.4%

Netherlands7.7%

Russia40.7%

Netherlands3.8%

Indonesia4.9%

Turkmenistan6.6%

Algeria7.8%

Rest of the World12.9%

Algeria8.2%

Turkmenistan10.3%Canada

12.3%

Qatar7.8%

Canada10.9%

Norway12.5%

Oil Herfindahl index

7.0 6.0

5.0

5.5

6.0

6.5 Oil supply

4.0

4.5

5.0

5.5Net Exporters Oil supply

3.5

4.0

4.5

2.5

3.0

3.5

3.0

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2.0

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

6 0

6.5

7.0

Oil supply weighted by by political risk

5.5

6.0

Net exporters oil supply weighted by political risk

4.5

5.0

5.5

6.0

4.0

4.5

5.0

3.0

3.5

4.0

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

3.0

3.5

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

Gas Herfindahl index

17 11

11

13

15

Gas supply

7

8

9

10Net Exporters Gas Supply

7

9

4

5

6

5

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

3

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

15

17

Gas supply weighted by by political risk

9

10

11

Net exporters gas supply weighted by political risk

9

11

13

6

7

8

9

5

7

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

3

4

5

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

Literature review

Construction of diversification indices owes much to the work of: Blyth and Lefevre, IEA paper, 2004y , p p , Gupta, Energy Policy, 2008 LeCoq and Paltseva, Energy Policy, 2009q , gy y,

Short-run security: yindices of diversification in supply

2

( ) *100iNPICSI ( ) *100i

CSIC

max{0, }i ij ijNPI M X

Short-run security: ydiversification adjustment for political risk

2

[( ) * ]iNPICSI POL [( ) * ]ipol i

i

NPICSI POLC

[1 0 0 / 1 0 0 ]P O L IC R G[1 0 0 / 1 0 0 ]i iP O L IC R G

Short-run security: ydiversification adjustment for country size

SIZE = ratio of world imports of a fuel source to the country’s imports y p

Short-run security: ydiversification adjustment for distance

Distance (Di) = 1 if distance between capitals of importing and supplying countries is less than 1500 km;

Di = 2 if the distance is between 1500 and 4000 km; andDi = 2 if the distance is between 1500 and 4000 km; and Di = 3 if the distance exceeds 4000 km.

Diversification in supply sources: pp ymain findings Once political risk of suppliers is taken into

account, diversification in supply sources has increased since 1990 for most countries

Adjustment for size does not make a big difference, except for U.S. in the case of oildifference, except for U.S. in the case of oil

Diversification index for oil, 2008ve s cat o de o o , 00825.00

20.00

10.00

15.00

5.00

0.00

nmarkCan

adastr

alia

domStatesFran

ceala

ndSpainrtu

galIta

lylandsmanyKore

aAustr

iaIre

landJapan

SwedenlgiumTurkeyGree

cerla

ndublicFinland

Poland

ungaryublic

Denm CaAustr

United Kingd

United Sta Fr

New Zeal SPort

Netherl

aGerm

K Au Ire J SwBelg Tu G

Switzerl

Czech Repub Fin Po

HunSlovak Repub

Diversification Index for Oil, 2008, Adjusted for ve s c o de o O , 008, djus ed oPolitical Risk, Size and Distance

60

40

50

60

20

30

40

10

20

0

Diversification index for natural gas 2007Diversification index for natural gas, 200725.00

20.00

10.00

15.00

5.00

0.00

marklandsStates dom Fran

celgium Ita

lyPola

ndmany

Spainrla

ndngary

Austria

wedenblic

Greece

reland

rtugal

Japaninlandvakia

DenmNeth

erla

United Sta

United Kingdo Fr

Belg PoGerm

SpSwitz

erla

HungAu Swe

Czech Repub Gr Ire Port J Fin

Slova

Diversification Index for Natural Gas, 2007, ve s c o de o N u G s, 007,Adjusted for Political Risk, Size and Distance

70

50

60

70

20

30

40

0

10

20

Diversification index weightedDiversification index, weighted16

10

12

14

6

8

10

2

4

0

Denmarkd StatesingdomFran

ceherl

andsSpain

Swedentze

rland

Germany

Belgium

Austria

Italy

PortugalJapan

epublicPola

ndGree

ceIre

landHungary

FinlandRepublic

DeUnite

d Unite

d King

Nether S

Switze

Ge Be A PoCzec

h Rep I H FSlovak Rep

Energy DiversificationEnergy Diversification Based on CSI Values in 2007-08

Natural Gas Low

VulnerabilityMedium

VulnerabilityHigh

Vulnerability

Vulnerability CSI < 4

Vulnerability 4<CSI<10

Vulnerability CSI > 10

Low Vulnerability Denmark, UK Ireland

Crude Oil

yCSI < 4

,

Medium Vulnerability

France, Netherlands, Italy, Spain

Austria, Germany, Crude Oil 4<CSI<10 US Portugal, Sweden

High Vulnerability

Belgium, Poland,

Czech Rep., Finland, Greece, Hungary JapanVulnerability

CSI > 10 Switzerland Hungary, Japan,Slovak Rep.

CSI (weighted) (oil and gas sharesCSI (weighted) (oil and gas shares normalized to sum to 1)

50

35404550

202530

05

1015

0

CSI (weighted) weighted by oil andCSI (weighted) weighted by oil and gas shares

30

20

25

30

10

15

20

0

5

10

0

Diversification in supply sources: pp yrefinements underway

I l d h Include other energy sources

All f i t f l b tit ti Allow for interfuel substitution

Use energy reserves of the energy suppliers instead Use energy reserves of the energy suppliers instead of their imports

MEDIUM-RUN SECURITYLOWER MACRO SENSITIVITY TO ENERGY SHOCKS

A Slippery Relationship“The macroeconomic impacts of oil shocks are ignored [in the book]; this neglect is sensible given the wide varieties of prevailing views and the uncertainties about which results, if any, are valid.”

-- Richard L. Gordon(in a book review in the latest issue of The Energy Journal)

Two dominant viewsil i h k h l d k l i Exogenous oil price shocks have played a key role in

nearly every post-WWII U.S. recession and remain an important force even today (JPE, 1982; BPEA, 200x)p y ( , ; , )

The importance of oil price shocks in causing the 1970s stagflation has been overstated Oil price1970s stagflation has been overstated. Oil price increases today are driven by demand increases in emerging markets and are different from the oil shocks of the 1970s (Barsky and Kilian, 2001; KilianJEL 2008; see also Bohi 1991)

A middle path?il i h k did l i l i h Oil price shocks did play an important role in the

stagflation of the 1970s

But there have been changes since: Our luck may have changedy g Real wages are less rigid Monetary policy response is better Share of oil in production & consumption is lower

Net result: oil price shocks have smaller effects on output and Net result: oil price shocks have smaller effects on output and inflation in the 2000s than in the 1970s (Blanchard & Gali, 2009; Blanchard and Riggi, 2010)

Medium-run security: macro responseed u u secu ty: ac o espo se

5

Actual Lower bound estimate Upper bound estimate

Figure 2: Impact of Oil Shocks on U.S. GDP Growth

3

4 Actual pp

1

2

-1

0

-2

-1

-3 1974Q1-75Q1 1979Q2-80Q2 1990Q3-91Q3 2007Q4-08Q3

Notes: The lower and upper bound estimates show what real GDP growth would have been in the absence of oil shocks. Based on Hamilton (2009a)

Work in progressWork in progress (with Marianna Riggi)

Empirical evidence from 5 European countries (France, Germany, Italy, Spain, UK) on whether macroeconomic impact of oil is different post-1992 than pre-1992

New Keynesian model to sort out contributions of: New Keynesian model to sort out contributions of: change in oil share nominal wage rigidities changes in monetary policy response

New Keynesian modelil d i d i d i Oil used in production and consumption

Shares decline over time

Real wages and prices are rigid Degree of rigidity is estimatedDegree of rigidity is estimated

Monetary policy responds to output and inflation (Taylor rule), but with some smoothing/inertia

M t li t b f ll dibl Monetary policy may not be fully credible Degree of credibility is estimated

Key equations

Estimation techniquei i di i Minimum distance estimator

Parameter values that minimize distance between VAR based Parameter values that minimize distance between VAR-based IRFs and model-based IRFs (over the first 20 quarters)

Some parameter values chosen a priori; others based on data (shares of oil in production and consumption); others estimated includingestimated, including Degree of price and wage rigidity Monetary policy response and credibility

D li i h f il i d tiDeclining share of oil in production

D li i h f il i tiDeclining share of oil in consumption

Main resultsComparing post-92 period with pre-92 period: Lower real wage rigidity in all countriesg g y Lower price rigidity in all countries With the exception of the UK central bank With the exception of the UK, central bank

credibility higher St t li ( i t t t ) Stronger monetary policy (or interest rate)

response to inflation (but also a stronger t GDP )response to GDP …)

Empirical evidencei bl il i d fl Seven variable VAR: Oil price, CPI, GDP deflator,

nominal wage, real GDP, employment, nominal interest rate

Identification assumption: innovations in oil prices are exogenous to other variables included in the VAR

In slides that follow:I l f i (IRF) UK d S i Impulse response functions (IRF): UK and Spain

IRFs for all 5 countries Changing contribution of oil shocks to fluctuations: ratio Changing contribution of oil shocks to fluctuations: ratio

of conditional standard deviation, pre-92 to post-92

UK: CPI response to oil prices stronger pre-1992UK: CPI response to oil prices stronger pre 1992

UK: GDP response to oil prices stronger pre-1992UK: GDP response to oil prices stronger pre-1992

UK: Interest rate response to oil prices stronger pre-'92UK: Interest rate response to oil prices stronger pre- 92

Spain: CPI response to oil prices stronger pre-1992Spain: CPI response to oil prices stronger pre 1992

Spain: GDP response to oil prices stronger pre-1992Spain: GDP response to oil prices stronger pre-1992

Spain: Interest rate response to oil stronger pre-1992Spain: Interest rate response to oil stronger pre-1992

0 8

1

1.2CPI

1

1.2

Nominal wage

0.2

0.4

0.6

0.8

0.2

0.4

0.6

0.8

0Germany France Spain Italy UK 0

Spain Italy Germany France UK

GDP Interest rate

0 6

0.8

1

1.2

0 6

0.8

1

1.2

0

0.2

0.4

0.6

Germany France Italy Spain UK0

0.2

0.4

0.6

Germany France Italy Spain UKGermany France Italy Spain UK Germany France Italy Spain UK

Diminished contribution of oil price shocks to macro fluctuationsRatio of conditional standard deviation, 1970-1992 to 1992-2010

LONG-RUN SECURITYLOWER USE OF ENERGY PER UNIT OF INCOME

Long-run security: oil intensityLong run security: oil intensity1.00

08

-1.00

0.00

e, 2

000-

0

3 00

-2.00

% c

hang

e

-4.00

-3.00

annu

al %

-5.00

Slovakiaustr

ali aGreece

Korea

TurkeyCan adagdomHungary

IrelandJapanPola

ndSpa in

Zealand

d Sta teserla

ndsSweden

France

FinlandBelg

iumenmark

Germany

Austria Ital y

zerl an

dPort

uga lpu blic

a

Slo Aus G T CUnited

King Hu I

New Ze

United S

Netherl S F Be Den Ger A

Switze PoCzec

h Repu

Long-run security: natural gas intensityo g u secu ty: atu a gas te s ty12.00

07

4 00

8.00

e, 2

000-

0.00

4.00

% c

hang

e

-4.00

annu

al %

-8.00

Slovakiapu blic

gdomd Sta tesDenmark

erlands

Ireland

HungaryFinlandAustr

iaPola

ndFran

cezer

l and

SwedenBelg

iumGerm

any Ital y

Greece

Spa inJapanTurkey

Portuga l

a

SloCzec

h RepuUnited

KingUnited

S DenNeth

er I Hu F A

SwitzeS Be Ger Po

Long-run security: primary energy intensityo g u secu ty: p a y e e gy te s ty0.00

06

-1.00

e, 2

000-

0

-3.00

-2.00

% c

hang

e

-4.00

annu

al %

-5.00

SlovakiaIre

landSweden

Zealand

Poland

Greece

gdomHungarypu blicd Sta tes

erland

Belgium

enmarkKore

aFinlandCan adaTurkey

JapanSpainFran

cePort

uga lustr

ali aerm

anyAustr

iaerla

nds Ital y

a

Slo I SNew

Zea GUnited

KingdHuCzec

h RepuUnited

SSwitze

r Be Den F C T Po AusGer A

Netherl

Future work Refinements to diversification indices, as discussed Country-by-country estimates of short-run and

di f l GDP h kmedium-run responses of real GDP to energy shocks – work underway

R ki f i t iti Rankings of energy intensities

►C bi h l h i►Combine these elements to get comprehensive measures of short-, medium-, and long-run security

►I di id l t it t i it►Individual country security vs. systemic security

Thank you!Thank you!C t lComments welcome.