UNIVERSITY RICE EMF Some Results from the Rice World Natural

Document Sample
UNIVERSITY RICE EMF Some Results from the Rice World Natural Powered By Docstoc
					UNIVERSITY

RICE

EMF 23: Some Results from the Rice World Natural Gas Trade Model

Peter Hartley Kenneth B Medlock III

James A. Baker III Institute for Public Policy RICE UNIVERSITY
1

UNIVERSITY

RICE

The Rice World Natural Gas Trade Model
The model uses the MarketBuilder software platform from Altos Partners to calculate equilibrium prices and quantities for fixed locations and periods Dynamic spatial general equilibrium linked through time by Hotelling-type optimization of resource extraction
“Economic theory of the mine”
Producers consider the costs of development and extraction against all current and future prices when determining the profitability of developing a unit of reserves. Producers, therefore, maximize the net present value of resource extraction for the life of the investment. The shadow value of the resource determines marginal profit (or rent).

No opportunity for either spatial or temporal arbitrage at equilibrium.

Predicts
Regional prices Regional flows

Non-stochastic framework that allows analysis of a number of different scenarios
2

The Rice World Gas Trade Model (cont.)
UNIVERSITY RICE

Supply
associated and unassociated natural gas resources, conventional gas deposits (CBM gas deposits in North America and Australia) assessed in three categories:
proved reserves (updated 2003 Oil & Gas Journal estimates) growth in known reserves (P-50 USGS estimates) undiscovered resource (P-50 USGS estimates)

Demand
Econometric model using EIA, IEA and World Bank data
Level of economic development (GDP/capita) Population Prices of natural gas, oil and coal Country-specific factors
Energy demand increases with GDP/capita but at a decreasing rate Natural gas share in primary energy demand increases with development

From 2020, allow demand to be lost to new technologies at prices above $5 with up to 2.5% lost at $5.50 and 5% lost at $10
Each year, the proportion of demand vulnerable to the backstop at each price above $5 increases. By 2075, if real prices are above $5 the backstop is capable of meeting all base case demand.
3

UNIVERSITY

RICE

Demand sinks and supply sources are regionally disconnected…

Source: USGS

4

Proved and Potential Natural Gas
UNIVERSITY RICE

Oil and Gas Journal USGS World Resource Assessment 2000

Proved: 55.6 (35.7%) Proved: 7.2 (4.6%) Potential: 12.8 (9.8%) Proved: 5.4 (3.5%) Potential: 7.3 (5.6%) Potential: 40.7 (31.4%)

Proved: 56.1 (36.0%) Potential: 34.6 (26.7%) Proved: 11.8 (7.6%) Potential: 9.4 (7.2%) Proved: 12.6 (8.1%) Proved: 7.1 (4.5%) Potential: 11.9 (9.2%) Potential: 12.9 (10.0%)

Units: Trillion Cubic Meters 5

UNIVERSITY

RICE

Natural Gas Transportation
The model chooses new or expanded transport capacity from supply sources to demand sinks based on:
capital costs of expansion, and operating and maintenance costs of new and existing capacity

To facilitate calculations:
supplies and demands are aggregated into discrete “nodes”, parallel pipes are aggregated into a single link, and LNG routes are represented by hubs and spokes

Pipeline costs: regression analysis of EIA cost data (annual cost per unit of capacity) for 52 pipeline projects
Pipeline length and capacity - higher throughput capacity reduces per unit costs as a result of scale economies Indicator variables for whether the pipeline - crosses mountains, moves offshore or crosses a lake or sea, or crosses more populous areas

LNG costs from 2003 EIA report and industry sources
Shipping costs split into a fixed capital cost for ship development plus operating costs Liquefaction costs are a fixed cost plus a variable feed gas cost (model calculated) Regasification costs vary by location Allowed for technological change to reduce LNG costs at rates of change based on a statistical fit to the IEA World Energy Investment Outlook

6

LNG transportation network
UNIVERSITY RICE

We allow many potential pipeline links, and use a hub-and-spoke representation for LNG, which allows for many potential trading partners. Implicitly assumes contracts do not dictate flows. Note, this does not mean contracts are not important for project development.

7

UNIVERSITY
Middle East and Russia become largest producers of natural gas
2020: 15.2% and 37.6% of global gas consumption, respectively 2030: 22.8% and 35.4% of global gas consumption, respectively

RICE

Reference Case Results

North American production declines. Expanding demand results in large increases in LNG imports. South America is largely self-sufficient, with relatively small exports from Venezuela, Trinidad and Tobago, and Peru. European demand growth leads to increases in both pipeline and LNG imports. Africa supplies Europe via pipeline and increases its LNG volumes to the global market. Australian LNG exports expand significantly. Asia
Indonesian supplies expand from Greenfield projects… Arun becomes obsolete. Japan - LNG and pipeline gas from Sakhalin. Biggest barrier to larger Sakhalin pipeline is difficulty in developing a national grid. South Korea - Economics favor pipelines through North Korea. China - All three options… domestic supply, Russian gas and LNG fills gaps. India - All three options… domestic supply, east & west pipeline gas and LNG.

8

UNIVERSITY

RICE

Supply Projections
200 180 160 140 120 100

Europe and US shrink, Russia, Middle East and Australia expand

Middle East

Russia
80 60 40 20 0 2002
Australia

} Europe
US Asean Australia
2006
Asean

2010
US

2014

2018

2022
Other Latin America

2026

2030

2034
North Africa

2038
Qatar

Other Europe

China

Other Asia Pacific

Trinidad & Tobago

United Kingdom

Russia

Other Africa

Saudi Arabia

India

Canada

Venezuela

Norway

Other FSU

Iran

Other Middle East

Source: RWGTM, Hartley and Medlock (2005)

9

Demand projections
UNIVERSITY RICE
200 180 160 140 Russia 120 100 80 60 40 20 0 2002
China Japan India

Backstop constrains demand growth

Europe

US

2006
Australia China India Japan

2010

2014
Asean importers Other Asia Pacific Canada US Brazil

2018

2022
Mexico Venezuela Other Latin America France Germany

2026
Italy Spain

2030

2034
Other FSU Africa Iran Saudi Arabia

2038

United Kingdom Other Europe Russia

Asean exporters

Other Middle East

Source: RWGTM, Hartley and Medlock (2005)

10

Natural gas trades
UNIVERSITY
60
Exports
Saudi Arabia Qatar Iran

RICE

80

40
Russia

20 0 -20
US Australia China India Japan

-40
Imports

-60 -80 2002 2006 2010
Iran Other Africa North Africa Russia

Europe

2014

2018
Norway Venezuela Canada Asean exporters

2022
Australia

2026

2030
Italy Germany France Mexico

2034
US

2038
Japan India China

Other Latin America Other Middle East Saudi Arabia Qatar

Other Europe/FSU United Kingdom Spain

Other Asia Pacific Asean importers

Source: RWGTM, Hartley and Medlock (2005)

11

UNIVERSITY

RICE

Producer Export Shares (pipe and LNG)
Russia remains the single largest supplier of exports globally. Current OPEC member nations share of total exports is fairly steady through 2020 at around 24%, but rises to 39.5% by 2040.
Russia Kazakhstan Turkm e nistan Canada Alge ria N ige ria Ango la Indo ne sia Malaysia Brune i Australia Iran Q atar Saudi Arabia N o rway Gre e nland Trinidad & To bago Ve ne zue la Re st o f W o rld 2010 15.71% 8.11% 3.17% 7.56% 11.03% 2.99% 0.17% 5.03% 3.98% 1.32% 1.88% --3.47% 0.00% 7.05% --2.22% --26.31% 2020 16.18% 5.17% 3.11% 3.14% 6.72% 6.64% 0.55% 5.40% 2.69% 1.28% 4.10% --4.68% 1.65% 5.40% --1.55% 1.63% 30.10% 2030 17.35% 2.21% 3.65% --1.89% 7.04% 1.04% 5.55% 1.03% 1.12% 8.46% 3.17% 7.87% 5.28% 5.30% --0.65% 3.68% 24.69% 2040 19.54% 0.71% 4.55% --0.20% 5.33% 0.79% 4.50% --0.56% 8.09% 8.69% 9.50% 8.13% 5.11% 2.12% --3.15% 19.00%

12

UNIVERSITY

RICE

LNG export projections
60
Venezuela

50

Russia Atlantic

40

Dominant feature: Growth in the Middle East

Nigeria Saudi Arabia

Atlantic

30

UAE Iran

Mid-East

20
Qatar

10

Indonesia

Pacific
Australia

0 2002

2006
Other Atlantic Basin Venezuela Russia Atlantic

2010
Greenland Angola

2014

2018
Nigeria North Africa

2022

2026
Saudia Arabia UAE Iran

2030
Qatar

2034

2038
Malaysia Indonesia Australia

Other Pacific Basin Papua New Guinea

Equitorial Guinea

Other Middle East

Source: RWGTM, Hartley and Medlock (2005)

13

UNIVERSITY

RICE

LNG import projections
60
Mexico Atlantic

50

40

Dominant feature: Growth in US Imports
Atlantic

30
India

US Atlantic

20
S. Korea

Mexico Pacific

UK

Indian
US Pacific China Japan

10

Pacific

0 2002

2006

2010

2014

2018

2022

2026

2030

2034

2038

Other Atlantic Basin Mexico Atlantic Canada US Atlantic

Other Europe UK Spain

France Italy Other Indian Ocean

India Other Pacific Basin Mexico Pacific

US Pacific Vietnam Taiwan

China South Korea Japan

Source: RWGTM, Hartley and Medlock (2005)

14

UNIVERSITY
1.

RICE

Scenarios for Russian Gas in Northeast Asia
Pipelines from Nahodka & NE China through North Korea are blocked
Political relations with North Korea prevent them An undersea pipeline to South Korea from China can still be built Sakhalin pipeline to Japan still is possible
Connections between South Korea and Japan are also permitted, but these are too expensive to use
2.

Russia to China pipelines also don’t get built
Political difficulties may also prevent this development We also rule out the pipeline from Uzbekistan to China
It otherwise provides an indirect route for gas sales from the VolgaUrals region in Russia to China

Sakhalin pipeline to Japan with an extension to Korea still is possible

15

Reference Case: NE Asia Sources of Supply
UNIVERSITY RICE China
10 9 8 7 6 5 4 3 2 1 0 2002 2006 2010 2014 2018 2022 2026 2030 2034
Backstop

Japan
5 4.5 4 3.5
2038

3 2.5

Home production

Pipeline Imports

LNG Imports

Korea
3 2.5

2 1.5 1 0.5

2

0 2002 2006 2010 2014 2018 2022 2026 2030 2034
Backstop

2038

Home Production

Pipeline Imports

LNG Imports

1.5

1

0.5

0 2002 2006 2010 2014 2018 2022 2026 2030 2034 2038

Russia (Nahodka) to South Korea LNG Imports

North (Seoul) Backstop South (Pusan) Backstop

Internal North to South Pipeline

16

No N Korea pipes: NE Asia Sources of Supply
UNIVERSITY RICE
10 9 8 7 6 5 4 3 2 1 0 -1 2002 2006 2010 2014 2018 2022
LNG Imports Backstop

China
China imports more via pipeline and exports to South Korea

Japan
5 4.5 4 3.5
2026 2030
Exports

Japan takes more pipeline gas, less LNG

2034

2038

Home production Pipeline Imports

3 2.5 2 1.5 1

3

Korea
LNG and China pipeline share South Korean market

2.5

0.5 0 2002 2006 2010 2014 2018 2022 2026 2030 2034
Backstop

2

2038

Home Production

Pipeline Imports

LNG Imports

1.5

1

0.5

0 2002 2006 2010 2014 2018 2022 2026 2030 2034 2038

LNG Imports China to South Korea (undersea)

North (Seoul) Backstop South (Pusan) Backstop

Internal North to South Pipeline

17

No N Korea pipes: Select Price Changes
UNIVERSITY
$1.5

RICE

$1.0

$0.5

$0.0

$-0.5
Price zigzag results from earlier construction of a pipeline (2006 instead of 2008). The price is lowered below the reference case price in 2006, and no longer falls in 2008 as the reference price does.

$-1.0

$-1.5 2002 2006 2010 2014 2018 Tokyo Zeebrugge 2022 2026 2030 2034 Seoul 2038

Henry Hub Beijing

Buenos Aires Delhi

18

No N Korea pipes: Supply Changes
UNIVERSITY RICE
Net supply changes most years are small Russian output declines most, but increased Russian exports to Europe also displace some Middle East pipeline exports
2.5 2 1.5 1 0.5 0 -0.5 -1 -1.5 -2 2002
Venezuela

2006
Iran

2010

2014
Russia

2018

2022
Greenland

2026
Brazil

2030

2034

2038
Remaining Asia/Pacific

Remaining Middle East

Turkmenistan

Australia

Iraq

China

Canada

Remaining Europe & FSU

Remaining Americas

Azerbaijan

United States

Saudi Arabia

Africa

Norway

19

No N Korea pipes: Demand Changes
UNIVERSITY RICE
Higher prices stifle South Korean, and to some extent Chinese, demand
Demand reductions in other countries are more than compensated in other years

Largest net demand increases are in Russia, Central Asia and Europe
0.8

0.6

0.4

0.2

0

-0.2

-0.4 2002 2006 2010
Russia Uzbekistan Brazil

2014

2018

2022

2026

2030

2034

2038

United States Europe Middle East

South Korea China

Remaining Asia/Pacific Rest of world

20

No N Korea pipes: LNG Supply Changes
UNIVERSITY RICE
Sakhalin LNG exports after 2030 displace some Australian and Saudi exports Exports from Russian Atlantic and Greenland decline, and from Venezuela and Iranian expand, after 2030 Australian and Southeast Asian exports expand up to 2020
2

1.5

1

0.5

0

-0.5

-1

-1.5 2002 2006 2010
Russia Pacific Venezuela Iran Papua New Guinea

2014

2018
Brunei Indonesia Egypt Qatar

2022

2026

2030
Nigeria Malaysia

2034

2038

Philippines Russia Atlantic Australia Greenland

Saudia Arabia Remaining

21

No N Korea pipes: LNG Demand Changes
UNIVERSITY RICE
Korea is the only country to have a sustained increase in demand for LNG Other Pacific Basin importers – especially Japan, and to a lesser extent the US Pacific and China, experience reduced demand In most other countries, demand shifts are intertemporal
1.5

1

0.5

0

-0.5

-1 2002 2006
South Korea Spain Thailand

2010
India Japan

2014

2018
China UK

2022
Italy US Atlantic

2026

2030
Mexico Atlantic Canada

2034

2038
Argentina Remaining

US Pacific

22

No China or Korea pipes: NE Asia Supply
UNIVERSITY RICE China
9 8 7 6 5 4 3 2 1 0 -1 2002 2006 2010 2014 2018 2022 2026
Exports

China exports are now zero and China takes more LNG

Japan
5 4.5 4 3.5
2030 2034
Backstop

Japan takes even more pipeline gas and less LNG

2038

3
Home production LNG Imports

3

Korea
LNG beats a potential Sakhalin-Japan-Korea pipe

2.5 2 1.5

2.5

1 0.5 0 2002 2006 2010 2014 2018 2022 2026 2030 2034
Backstop

2

2038

1.5

Home Production

Pipeline Imports

LNG Imports

1

0.5

0 2002 2006 2010 2014 2018 2022 2026 2030 2034 2038
LNG Imports North (Seoul) Backstop South (Pusan) Backstop Internal North to South Pipeline

23

UNIVERSITY
$1.5

RICE

No China or Korea pipes: Select Price Changes

$1.0

$0.5

$0.0

$-0.5

$-1.0 2002 2006 2010 2014 2018 Tokyo Zeebrugge 2022 2026 2030 2034 Seoul 2038

Henry Hub Beijing

Buenos Aires Delhi

24

No China or Korea pipes: Supply Changes
UNIVERSITY RICE
Higher overall LNG demand leads to sustained expansions in the main LNG exporters (Australia, Iran, Qatar, Indonesia) US, Central Asian and Chinese domestic supply expand Largest reduction is in Russia, but outputs in Saudi Arabia, Nigeria, Greenland and Canada also fall in most years
2 1.5 1 0.5 0 -0.5 -1 -1.5 -2 -2.5 -3 2002 2006
Australia Iran Azerbaijan

2010
Qatar

2014

2018
China Russia Saudi Arabia

2022

2026

2030
Canada Norway Turkmenistan

2034
Venezuela Remaining

2038

Greenland Nigeria Iraq

United States Indonesia

25

No China or Korea pipes: Demand Changes
UNIVERSITY RICE
Now Chinese demand decline exceeds the Korean one; remaining Pacific Basin importers are again adversely affected Main demand increases occur in Russia, Europe and the Middle East
0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 -1 -1.2 -1.4 2002 2006 2010
Russia Remaining FSU Middle East

2014
China

2018

2022

2026
United States

2030

2034
Europe South America

2038

South Korea Remaining Asia Pacific

Canada & Mexico Africa

26

No China or Korea pipes: LNG Supply Changes
UNIVERSITY RICE
Again, Sakhalin LNG exports rise, but now the overall LNG market expands enough that few (only Greenland and Nigeria) LNG exports decline

3.5 3 2.5 2 1.5 1 0.5 0 -0.5 -1 -1.5 2002 2006 2010
Russia Pacific Australia Remaining Pacific Basin Iran

2014

2018
Qatar

2022

2026
Nigeria

2030
Russia Atlantic Venezuela

2034

2038

Saudia Arabia

Remaining Middle East Greenland

Remaining Atlantic Basin

27

No China or Korea pipes: LNG Demand Changes
UNIVERSITY RICE
Increased Korea and China LNG demand displaces other Pacific Basin importers North America Atlantic imports increase (mainly via Canada) Mexico shifts to pipeline imports from northern South America
4

3

2

1

0

-1

-2 2002 2006 2010
South Korea China Canada

2014
Japan

2018

2022

2026

2030

2034
Europe

2038

Mexico Atlantic US Atlantic

US Pacific Remaining Asia Pacific

Remaining Atlantic

28

UNIVERSITY

RICE

Comments on Russia/Asia Scenarios
In the coming worldwide market for natural gas, political disturbances in one area have global effects The results illustrate the key role Russia will play in the future world gas market
Russia not only has a lot of gas It also is strategically placed to ship gas either east or west and hence in a position to arbitrage between European and Asian markets Toward the end of the horizon, Russia also becomes a significant exporter of LNG, thus helping to solidify the link between LNG prices and pipeline gas prices around the world

North America and the Middle East also link Pacific and Atlantic gas markets
Middle East producers can export LNG east or west, and also can ship gas via pipeline to Europe or the Indian sub-continent In North America, if Pacific Basin gas prices rise, more Atlantic Basin LNG is imported and the arbitrage point moves toward the west coast

Japan is a close competitor to South Korea for Pacific Basin LNG and increased Korean demand raises Japanese prices
29

UNIVERSITY

RICE

A “Gas-OPEC” Scenario
In May 2001, the Gas Exporting Countries Forum (GECF) held its first ministerial meeting in Tehran, with subsequent meetings in Algiers (2002), Doha (2003) and Cairo (2004)…
Expanded participation since the first meeting includes Algeria, Brunei, Egypt, Indonesia, Iran, Libya, Malaysia, Nigeria, Oman, Qatar, Russia, Trinidad and Tobago, the United Arab Emirates, Venezuela and one observer, Norway

Case assumptions
Current members of OPEC form a cartel in natural gas. Cartel exerts control on production by requiring a higher rate of return on all export projects.

Caveats to be addressed in future scenarios
Russia is not a cartel member. There are no political barriers to infrastructure developments. Lifting of access restrictions in the US.
30

UNIVERSITY

RICE

Could there be a “Gas-OPEC”?
Likelihood and potential for market power… could it be like oil?
Cartel members control large share of market
Gas resources are distributed in a similar manner as oil

For effective collusion, members must…
agree to production or capacity controls prevent cheating prevent new entry

This is facilitated by…
low elasticity of supply of non-members fewer members inelastic product demand

Long term versus short term
Producers must balance risk of consumers moving to alternatives with rents that could be earned Consumers must balance the cost of using alternative technologies with paying more than marginal cost All depletable resources are, in the end, “transition” fuels
31

Supply concentration vs. elasticity of demand

Prices
UNIVERSITY RICE
2000$/MMBTU $8.00 $7.00 $6.00

The Reference Case indicates convergence in prices as markets become increasingly connected.

$5.00 $4.00 $3.00 $2.00 $1.00 $2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040

Henry Hub
2000$/MMBTU $0.25

Tokyo

Zeebrugge

$0.20

$0.15

$0.10

$0.05

$Henry Hub 2002-2020 Tokyo 2022-2040 Zeebrugge

Higher required returns in OPEC member nations results in higher prices. The effect is substantial post-2020… except in Europe where increased flow of Russian pipeline gas serves to keep prices from rising as much as in other regions.

32

LNG Liquefaction
UNIVERSITY RICE
tcf 50.0 40.0

Reference Case shows strong growth in LNG exports from both OPEC and Non-OPEC nations.

30.0 20.0 10.0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040

OPEC Other
tcf 2.5 2.0 1.5 1.0 0.5 (0.5) (1.0) (1.5) (2.0) (2.5) (3.0) 2002-2020 OPEC Other 2021-2030 OPEC M iddle East Non-OPEC 2031-2040

OPEC M iddle East

Non-OPEC

Higher required returns in OPEC member nations reduces total exports of LNG, but Non-OPEC LNG exports increase in response (Australia and Peru account for about half of the increase). The Non-OPEC response wanes in later time periods however as earlier exploitation raises long term costs in Non-OPEC regions.

33

LNG Regasification
UNIVERSITY RICE
tcf 50.0 40.0

Reference Case shows strong growth in LNG imports, with the strongest growth in North America.

30.0 20.0 10.0 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040

North America
tcf 0.4 0.2 (0.2) (0.4) (0.6) (0.8) (1.0) (1.2) 2002-2020 North America 2021-2030 Europe Asia Other 2031-2040

Europe

Asia

Other

Higher required returns in OPEC member nations reduces LNG imports everywhere. The largest reductions occur in Europe and North America.

34

Natural Gas Production
UNIVERSITY RICE
tcf 200.0

150.0

Reference Case shows strong growth in total production from OPEC and the FSU. North American production declines post-2008.

100.0

50.0

2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040

North America

FSU

Other Non-OPEC

OPEC - M iddle East

OPEC - Other

tcf 6.0 4.0 2.0 (2.0) (4.0) (6.0) (8.0) 2002-2020 North America FSU 2021-2030 Other Non-OPEC 2031-2040 OPEC - Other

Higher required returns in OPEC member nations reduces total production from OPEC nations. However, supply response in NonOPEC nations almost completely offsets the reduction.

OPEC - M iddle East

35

UNIVERSITY

RICE

Comments on Gas-OPEC scenario
There is little ability to raise prices in the near term (through 2020)
Australia, and others, emerge to fill the void in LNG shipments. Russian production increases to serve markets both east and west.

Longer term, resource constraints place limits on the fringe’s ability limit the cartel’s effect on gas markets.
Prices rise post-2020 and production from OPEC nations increases.

Further work will investigate other scenarios...
Russia joins the cartel
Places over 80% of the resource base in the hands of a cartel.

Pipelines also are not constructed from Russia to Asian markets
Greater rents to Pacific basin suppliers – including the Middle East

Access restrictions to drilling are lifted in the US.
Possible limitations on cartel supplies to the Atlantic Basin. Outcome will be highly sensitive to US development costs.

Facility location issues, particularly in the US, are difficult to resolve. Can consuming nations reduce potential market power?
Promote competition among energy sources by Improve energy efficiency
Liberalizing domestic energy sectors Develop technologies that facilitate fuel switching

36

UNIVERSITY

RICE

Revisions to the model in progress
Risk-adjusted rates of return differences across regions
Current: static through time Future: modify to allow risk premiums to converge as economies converge

Contracted volumes
Current: assume swaps will occur. Future: enforce existing contracts through explicit diversion costs. Note, a lack of future knowledge of contract structures means this will only dictate flows in the short to medium term.

Natural Gas resources
Current: No distinction between associated and non-associated natural gas Future: Allow oil market to dictate associated gas production and use (enhanced recovery vs. monetization through export/use)
37